<%BANNER%>

Quality of Guava Puree by Dense Phase Carbon Dioxide Process

Permanent Link: http://ufdc.ufl.edu/UFE0041952/00001

Material Information

Title: Quality of Guava Puree by Dense Phase Carbon Dioxide Process
Physical Description: 1 online resource (205 p.)
Language: english
Creator: Plaza, Maria
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: clarified, dense, guava
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: QUALITY OF GUAVA PUREE BY DENSE PHASE CARBON DIOXIDE PROCESS Maria L. Plaza (352)328-7459 e-mail: mlplaza@ufl.edu Food Science and Human Nutrition Chair: Maurice Marshall Doctor of Philosophy August, 2010 Fruits and vegetables are important components of a healthy diet and are one of the main sources of antioxidants. The consumption of tropical fruits and their products has increased over the past few years, but it perishability limits its marketability within the US making processed products from them more viable. Guava puree is a commonly processed fruit product which can be pasteurized to extend its shelf life, but pasteurization has negative effects on sensory and nutritional quality. A non-thermal process is desirable to protect the fresh flavor and nutritional value of guava puree. This study aimed to evaluate Dense Phase Carbon Dioxide Processing (DP-CO2), to extend shelf life of guava puree with minimal effect on the quality attributes and chemical composition of the product. The results of this study suggest that DP-CO2 treatment extended shelf-life and preserved the quality of guava puree reflecting the fresh-like sensory attributes of the fruit.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Maria Plaza.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Marshall, Maurice R.
Local: Co-adviser: Rouseff, Russell L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041952:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041952/00001

Material Information

Title: Quality of Guava Puree by Dense Phase Carbon Dioxide Process
Physical Description: 1 online resource (205 p.)
Language: english
Creator: Plaza, Maria
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: clarified, dense, guava
Food Science and Human Nutrition -- Dissertations, Academic -- UF
Genre: Food Science and Human Nutrition thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: QUALITY OF GUAVA PUREE BY DENSE PHASE CARBON DIOXIDE PROCESS Maria L. Plaza (352)328-7459 e-mail: mlplaza@ufl.edu Food Science and Human Nutrition Chair: Maurice Marshall Doctor of Philosophy August, 2010 Fruits and vegetables are important components of a healthy diet and are one of the main sources of antioxidants. The consumption of tropical fruits and their products has increased over the past few years, but it perishability limits its marketability within the US making processed products from them more viable. Guava puree is a commonly processed fruit product which can be pasteurized to extend its shelf life, but pasteurization has negative effects on sensory and nutritional quality. A non-thermal process is desirable to protect the fresh flavor and nutritional value of guava puree. This study aimed to evaluate Dense Phase Carbon Dioxide Processing (DP-CO2), to extend shelf life of guava puree with minimal effect on the quality attributes and chemical composition of the product. The results of this study suggest that DP-CO2 treatment extended shelf-life and preserved the quality of guava puree reflecting the fresh-like sensory attributes of the fruit.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Maria Plaza.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Marshall, Maurice R.
Local: Co-adviser: Rouseff, Russell L.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-02-28

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041952:00001


This item has the following downloads:


Full Text





QUALITY OF GUAVA PUREE BY DENSE PHASE CARBON DIOXIDE TREATMENT


By

MARIA L. PLAZA















A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2010

































2010 Maria L. Plaza





















To GOD for giving me the most supportive parents in the world, the most beautiful kids
(Erick and Amalia), and an unbelievable strength to finish this stage of my life









ACKNOWLEDGMENTS

First, I would like to extend my special and deepest thanks to my advisors, Dr.

Maurice Marshall and Dr. Murat Balaban for their support, advice and friendship.

Thanks to Dr. Russell Rouseff for his invaluable help, assistance and time for my

research, thanks for letting me be part of your laboratory and for sharing all your

knowledge. Through their guidance, wisdom, and never-ending care; Dr. Marshall, Dr.

Balaban and Dr Rouseff; they have helped me to achieve all my goals. I sincerely

appreciate the help offered by Dr. Bala Rathinasabapathi as member of my supervising

committee. I would also like to acknowledge the unconditional support of Dr. Charles

Sims, his students and the taste panel personnel (Lorenzo and Asli).

My deepest thanks go to my parents, Wilda and Jose. I would have not be where I

am without the unconditional love, support, and advice of my parents. They have been

an inexhaustible source of love, inspiration, and encouragement throughout all my life. I

thank my sister and best friend Becky, for her advice and big sister support during all

these years. Big thanks go to my brother Pepe, Freddy and both of my sisters-in-law

(Nana and Malu) for their warm support and example of perseverance and love. I thank

Erick, for giving me the best gift on earth, my kids. I want to give my deepest and more

sincere thanks to two very special persons in my life, my kids Erick and Amalia. I thank

them for their understanding and for cheering me up when I was down. These wonderful

persons have a very special place in my heart.

I thank my lab partners and friends, Milena and Alberto, for their never-ending

assistance in my project and for making the work in the lab an enjoyable experience.

Special thanks go to Sarah, Stefan, Max, Diana and Giovanna. All of them formed part









of my student life and helped me pass through this important stage in the most

enjoyable way. Thanks for just adding happiness to my graduate school days.

I thank my band friends, Frank and Cindy, for their help and support during these

last two years. To that person that does not need to be mentioned, thanks for your

unconditional support since I started this journey and for always believing in me. Last

but not least, Daniel thanks for your support, help and being there when I needed you.

All of you made my life in Gainesville an enjoyable one filled with joy and love.









TABLE OF CONTENTS

page

ACKNOW LEDG M ENTS..................................... ........................... .. 4

L IS T O F T A B L E S ........................................................................................................ 1 0

L IS T O F F IG U R E S ...................................................................................................... 12

A B S T R A C T ....................................................... 14

CHAPTER

1 IN T R O D U C T IO N ................................................................................... 17

2 LITERATU RE R EV IEW .............................................................19

G uava (P sidium guajava L.) ................ .................................................... 19
Guava Fruit and Fruit Products: World Production................................ 19
C characteristics ......... ...................................................................... ...... ........ 20
Chem ical Com position ...................................... ..... ..............21
Guava Constituents and Health Benefits ............... .............. .... ......... .... 23
G uava Flavors ....................................................... ... .. .... ....... 29
G uava Fruit P processing ............... ...... ...... .................................................35
G uava P uree P processing ...................... .. ........... ...................... .............. 37
P h y to c h e m ic a ls ...................... .. ............. .. .............................................. 4 0
Classification .............. .............................. .............. ........... 41
A ttrib u te s ...................... .. .............. .. ................................................. 4 4
Extraction and Analysis ............... .............................. ........ ........... ... 44
Sensory Evaluation and Flavor Analysis ...................... ........... ............. 45
Sensory Evaluation ...................... ...... .......... ............................... 45
Flavor Analysis ........................................................... .. ... ... ......... ............... 47
Flavor Extraction Techniques ...................... ........................48
Volatile Identification Techniques ..................................... 50
Separation and Detection of Aroma Volatiles ......... ........ ............... 51
Beverage Processing ...................................................... ......... 52
Therm al Processing ...................... ...... ........... .............................. 53
Non-therm al Processing ............. ............... ............ ... .. ......... .. ............ 54
D ense Phase C arbon D dioxide .................. .................. .. .............. ...... ............. 54
Mechanisms of Microbial Inactivation .................. .. .............................. 54
Factors Affecting Microbial Inactivation...................................... 56
Solubility of CO2 ............................................. 57
Types of system s ......... .. ...................... ..... ........................ ...... 57
Food applications and effect on quality ............. ........................ 58
O objectives of S tud y ............. ..................... .. .. ............................ ............ 62




6









3 EFFECT OF ENZYME TREATMENT ON PHYSICOCHEMICAL AND
PHYTOCHEMICAL PROPERTIES OF GUAVA PUREE....................................64

A b stra ct ................ ..... ................................................ .................... 6 4
Introduction ................. ........ .................... .......... .......... 65
M materials and M methods ........... ....... .......... ........................................ ... 66
P re lim in a ry S tu d y ............................ ................. .......... ... ............ .................6 6
Selection of an enzyme and a temperature to clarify the guava puree...... 66
Effects of enzyme treatment at low temperature on phytochemical
levels in guava (Psidium guajava) puree. ........ ........... ................ 67
Sample preparation ....................................... ................ 67
A n a lys is ...................... ........ .. .. ....... ............... ...... ............ .. ............. 6 8
Enzyme Treatment Optimization at Low Temperatures to Produce a
Clarified Guava (Psidium guajava) Juice............................. ..... ........... 68
Sample preparation and enzyme treatment .......................... ................ 68
Physicochem ical analysis .................................. ............................ ........ 68
S ta tis tic a l A n a ly s is ............... ............................................ ................ 7 0
Results and Discussion ...................................................................... 71
Selection of Enzyme and Tem perature.............................................................. 71
Enzyme Treatment Optimization at Low Temperatures to Produce a
Clarified Guava (Psidium guajava) Juice without Affecting its
Phytochemical Composition ............................................................ ......72
Enzyme Treatment Optimization at Low Temperatures to Produce a
Clarified Guava (Psidium guajava) Juice without Affecting it
Phytochem ical C om position ................ ............................... ............ ........ 76
C o n c lu s io n s ................................................................................................. 8 3

4 INFLUENCE OF DENSE PHASE CARBON DIOXIDE AND PASTEURIZATION
TREATMENTS ON THE STORAGE QUALITY OF GUAVA PUREE..................... 94

A b s tra c t ............ ... .. ......................................................................... 9 4
Introduction ................. ........ .................... .......... .......... 94
M materials and M methods ........... ....... .......... .............................................. 97
Guava Puree ..................................... ............................ ........ 97
Model System ................... ... .......... .... ..... ......... 97
Solubility M easurem ents ............................. .... ........ ... ................... 97
Processing Equipment .................... ................ ............ ........... ..... 98
Microbial Inactivation Study .................... ........ .. ........ .. ... .......... .. 99
Storage Study and M icrobial Stability ................ .. ...... .......... ......... 100
C hem ical A analyses ................... ..... .......... .......... ...... ................. 100
R results and D discussion .............................. ............. ....... ................ 102
Solubility Experim ents .................................. .. .... ........ .. ................... 102
M icrobial Inactivation Study .................................................... ....... ......... 103
Storage Study and M icrobial Stability .............................................................. 106
Conclusions ...................................................................... .......... 111









5 PHYSICO-CHEMICAL AND PHYTOCHEMICAL CHANGES OF DENSE
PHASE CARBON DIOXIDE AND THERMALLY TREATED GUAVA PUREE
DURING REFRIGERATED STORAGE.............. ..... ........................... 122

A abstract ............ ..... ...... .......................................... ................... 12 2
In tro d u c tio n .......... ...... .... .... ... ................................................................ 12 3
M materials and M ethods ..................................................................... .............. ... .. 126
Total Phenolics, Antioxidant Capacity and Ascorbic Acid Analysis ............... 126
T itratable A cidity (T A ) ........................................................... ....... ........ .. 127
C olor A analysis ....................................................................... ......... ........ 127
High Performance Liquid Chromatography Analysis...................................... 127
Results and Discussion ........... ................................................ ....... ... 129
Total Phenolics and Antioxidant Capacity............................... 129
T itrata b le A c id ity (T A ) ............... ............... .......... ...... ......... .. ..............13 1
Ascorbic Acid (Vitamin C) Content ............... .. ........ ..................... 132
C o lo r ......................................... ......................................... 1 3 2
O rganic A cid C ontent ................ ............................................ .... ........... 134
Phenolic C om pounds ................ .... ........................................ .... .......... 134
C conclusions ....................................................................................................... 137

6 CHANGES IN AROMA COMPOUNDS AND SENSORY PERCEPTION IN
GUAVA PUREE AFTER THERMAL AND NON-THERMAL PROCESSING ....... 147

A abstract ........... ...... ...... ........................................ .......... .......... 14 7
Introduction ............. .......... ......................................... ............ 147
Materials and Methods ..................... ........ .. ............................ 149
Guava Puree .................... ............................ ........ .. ........... 149
Sensory Evaluation ................ ............... ........... ................................ .......149
Statistical A analysis ........ ......... ..... .... ... ... .............. ......... ......... 150
Extraction of Volatile Compounds using Headspace SPME........................ 150
Gas Chromatography- Olfactometry Analysis (GC-O) ...................................151
Gas Chromatography- Mass Spectrometry Analysis (GC-MS)....................... 152
Sulfur Compounds Identification using Gas Chromatography-Pulse Flame
Photom etric Detector (G C-PFPD)................ .............................. .............. 153
Identification Procedures .................................... .......... ....................... 153
Results and Discussion .................. .............. ........ ....... ................ .. 154
Sensory Analysis........... ...............................154
Flavor A analysis ................. ................. .......... ............................... 155
G uava V olatile C om position ......................................... ............... ................156
G C -M S Id e ntificatio ns .................................. ....................... .......... ......... 15 7
G C -O A rom a P rofiles ........................................................ .. .... .. ........ ..... 160
G C -P FP D A nalysis........................................................................ .. ................. 162
C conclusion ................................ ......... .. .. ........................164

7 SUMMARY AND CONCLUSIONS ........ ............. ................ ....172










APPENDIX


A ENZYME TRATMENT OF GUAVA PUREE.................... .................. 176

B DP-CO2 PROCESSING AND DATA ................ .............. ............ ... ....... ...180

C MICROBIAL INACTIVATION AND STORAGE STUDY DATA AND ANALYSIS... 185

D CHANGES IN AROMA COMPOUNDS AND SENSORY PERCEPTION IN
GUAVA PUREE AFTER THERMAL AND NON-THERMAL PROCESSING ........ 189

LIST OF REFERENCES ................... ...... .... .... ........ ..... ............. 195

BIO G RA PH ICA L SKETC H ........................................... ........ ................................ 205









LIST OF TABLES


Table page


3-1 Enzyme and concentration used during enzymatic treatment..................................84

3-2 Percent yield, ORAC value, total soluble phenolics and ascorbic acid content
of guava puree.................................................................................. 86

3-3 Color values obtained for three different guava puree before and after
c la rific a tio n ................. .............................................................................................8 6

3-4 Total soluble solids for guava puree before and after clarification for the three
d iffe re nt tre atm e nts ............... ........................................................ 86

3-5 Physicochemical results for enzymatic treatment of guava puree at three
different concentrations and three different reaction times................................ 87

3-6 Color results for enzymatic treatment of guava puree at three different
concentrations and three different reaction times ............................................ 89

4-1 pH, OBrix and titraacidity values for the treated and untreated guava puree
under different processing pressures ............... ............. .. .............. 113

4-2 Processing conditions and microbial reduction obtained during DP-CO2
process optim ization ........... ..... ........................ .............. ................ 114

4-3 Actual and predicted yeast and mold log reduction using the equation from
the surface response analysis ........... .......... ...... ............. ................ 114

4-4 Pectinesterase activity, cloud and pH measurement of guava puree before
and after DP-CO 2 treatm ent..................................................... 115

4-5 OBrix and titrabable acidity of guava puree before and after DP-CO2 treatment 116

4-6 pH, OBrix and cloud measurement for control, DP-CO2 and thermal treated
guava during 14 weeks of storage............ .. ............... ........................... 121

5-1 Antioxidant capacity and total soluble phenolics of fresh, DP-CO2 and
thermal treated guava puree during storage..................................... ........ ........ 138

5-2 L*, a* and b* values for control, DP-CO2 and thermal treated guava purees
during 14 weeks of refrigerated storage................ ........................ 141

5-3 Oxalic acid (OA), malic acid (MA), and citric acid (CA) content of control, DP-
CO2 and thermal treated guava purees during 14 weeks of refrigerated
storage at 40C ............. .............................. ............ .. ... ..... .... ............... 142









5-4 Identification of guava polyphenolics at 260 and 280 nm by HPLC based on
retention time, spectral properties, and comparison to authentic standards ..... 145

5-5 Gallic acid (GA), unknown and ellagic acid (EA) for control, DP-C02 and
thermal treated guava purees during 14 weeks of refrigerated storage at 40C. 146

5-6 Hydrobenzoic acid (GA), cinnamic acid (CA) and ellagic acid derivative
(EAD) for control, DP-C02 and thermal treated guava purees during 14
weeks of refrigerated storage at 40C ................ ......... .. ..................... 146

6-1 Difference in flavor and overall likeability between freshly thawed (reference
and hidden reference), dense phase-CO2 processed (DP-C02; 30.6 MPa, 8%
C02, 6.9 min, 35 C) and thermally treated (90 oC, 60 s) guava purees
detected by untrained panelists (n = 75) at weeks 0 .................................. 165

6-2 MS identification of guava puree volatiles. ..................................................... 168

6-3 Guava puree aroma active compounds. .................... ............................... 170

6-4 Guava puree sulfur volatile compounds ........ ........ ..... .... ................ 171

A-1 SAS software code used for the statistical analysis of repeated measurement
design and Tukey's standardized range (HSD) test............... ................... 176

A-2 SAS software output used for the statistical analysis of repeated
measurement design and Tukey's standardized range (HSD) test.................. 177

B-1 Guava Puree Solubility data ....................... .. ........... ... ................ 183

B-2 Guava Puree pH, Brix and Titraacidity measurement before and after the
CO2 solubility determ nation ... ...................... .......................... .. .... .......... 183

C-1 The average initial and final aerobic plate counts (APC) standard
deviations at 11 experimental runs from 2-factor, 3-level Central Composite
Design (CCD) .................... ......... ............................. .. ...... 185

C-2 SAS software code used for the response surface methodology (RSM)
analysis of 11 experimental runs determined by Central Composite Design .... 185

C-3 SAS software output of the response surface methodology (RSM) regression
analysis of 11 experimental-run data determined by central composite design 186

D.1 Volatile compounds, CAS number, identification method, reported Linear
Retention Indexes and references for previously reported studies............... 192









LIST OF FIGURES


Figure page

3-1 Percent yield for each enzyme treatment at three different temperatures
(Treatm ent tim e: 24 hrs). ................................................ .... .. .................. 85

3-2 Percent yield of clarified juice treated at three different enzyme
concentrations and four different reaction times up to 12 h............................... 90

3-3 Ascorbic acid content of clarified guava juices after guava puree was treated
at three different enzyme concentrations and four different reaction times up
to 12 h. ............. ................. ............ ................................. 90

3-4 Antioxidant capacity (pMol TE/L) of clarified guava juice treated with three
different enzyme concentration during 12 hours of reaction time at 300C...........91

3-5 Total soluble phenolics (GAE) of clarified guava juice treated with three
different enzyme concentrations during 12 hours of reaction time at 300C......... 91

3-6 Turbidity (% transmission at 650 nm) of clarified guava juice treated with
three different enzyme concentrations during 12 hours of reaction time at
300C ..................... ....................... ............ 92

3-7 pH of clarified guava juice treated with three different enzyme concentrations
during 12 hours of reaction time at 300C. ........ .................. .. ...... ......... 92

3-8 Total soluble solids (Brix) of clarified guava juice treated with three different
enzyme concentrations during 12 hours of reaction time at 300C.....................93

3-9 L* values of clarified guava juice treated with three different enzyme
concentrations during 12 hours of reaction time at 300C...... ............... .......... 93

4-1 Schematic diagram of Dense Phase Carbon Dioxide equipment..................112

4-2 Carbon dioxide solubility results obtained for guava puree, water and guava
puree model system at different processing pressures. .................... .......... 113

4-3 Aerobic plate count for control, DP-CO2 and thermal treated guava puree
during 14 w eeks of storage ...................................................... .................... 117

4-4 Yeast and mold plate count for control, DP-CO2 and thermal treated guava
puree during 14 weeks of storage. .............. ........................... ..... .............. 118

4-5 Pectinesterase activity for control, DP-CO2 and thermal treated guava during
14 weeks of storage ........ ......... .................... ........ .......... 119









4-6 Viscosity measurement for control, DP-C02 and thermal treated guava
during 14 weeks of storage ......... ................ .................. ..... 120

5-1 Titraacidity of control, DP-C02 and thermal treated guava purees during 14
weeks of refrigerated storage. ............. .............. ................ 139

5-2 Ascorbic acid (vitamin C) content of control, DP-C02 and thermal treated
guava purees during 14 weeks of refrigerated storage. ................. ................ 140

5-4 HPLC chromatogram of organic acids found in guava puree-1) oxalic acid, 2)
m alic acid and 3) citric acid ........................... ...... ............. ...................... .. 143

5-5 HPLC chromatogram of polyphenolic compounds found in ethyl acetate
fraction in guava puree-1) gallic acid, 2) unknown 3) ellagic acid, 4)
hydrobenzoic acid and 6) cinnam ic acid. ...................................... ................ 143

5-6 HPLC chromatogram of polyphenolic compounds found in the methanol
fraction of guava puree- 3) ellagic acid and 5) ellagic acid derivative ............... 144

6-1 Chemical composition of headspace volatiles for guava purees.. ..................... 166

6-2 Total ion chromatogram (TIC) for freshly thawed guava puree on DB-5
co lu m n ...................... ................................ ......... ........ ....... 16 7

A-1 Enzyme treated guava puree without filtration (left) and after filtration (right)... 179

B-1 Removal of insoluble solids from guava puree ....... ... .............. ............... 180

B-2 Experim ental solubility apparatus ...... ........ .............................. ........ ........... 181

B-3 Continuous DP-C02 system used during processing ................. .............. 182

C-1 DP-C02 treated samples during storage (week 4).................... ..... ........... 188

D-1 Questionnaire used for taste panel: different from control test, demographic
questions .............................. ........... .. ............. .......... 189

D-2 Questionnaire used for taste panel: different from control test, sensory
questions ............................. ............ .. ....... .. ..... ........ 190

D-3 Questionnaire used for taste panel: different from control test, comments ....... 191









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

QUALITY OF GUAVA PUREE BY NON-THERMAL DENSE PHASE CARBON
DIOXIDE PASTEURIZATION


By

Maria L. Plaza

August 2010

Chair: Maurice Marshall
Co-chair: Russell Rouseff
Major: Food Science and Human Nutrition


Guava (Psidium guajava L.) is an exotic fruit with a unique tropical flavor. It is

considered to be an excellent source of nutrients, phytochemicals and antioxidants,

especially ascorbic acid. The high perishability of the fresh fruit limits its marketability

within the US. Guava puree is a commonly processed fruit product which can be

pasteurized to extend its shelf life, but pasteurization has negative effects on sensory

and nutritional quality. A non-thermal process is desirable to protect the fresh flavor and

nutritional value of guava puree, which is used as a base for production of guava

beverages and other food products.

The objective of this research was to evaluate a new technology, Non-Thermal

Dense Phase Carbon Dioxide Pasteurization (DP-CO2), for the processing of guava

puree. The hypothesis was that the use of DP-CO2, would minimize or prevent

undesirable changes in phytochemical composition compared to traditional heat

pasteurization. In order to validate this hypothesis, measurement and comparison of the









chemical composition of the guava puree subjected to both treatments and an untreated

guava puree (freshly-thawed control) were conducted.

To facilitate processing the puree through the DP-CO2-equipment, viscosity of the

puree was modified by a commercially available enzyme, Bioguavase. Samples were

treated with the enzyme and evaluated for changes in viscosity at three hour intervals

for up to 12 hours of reaction time. The enzyme treatment increased juice yield,

produced a puree of lower viscosity, decreased the antioxidant capacity, and reduced

the total phenolic content of the product. The enzyme treatment also decreased the pH

of the juice due to the release of galacturonic acid from the pectin hydrolysis and

increased the total soluble solids content. Three hours of reaction time and 600 ppm of

enzyme were adequate to produce a clarified juice.

Microbial reduction was quantified as a function of pressure and residence time

using 8% C02 and a temperature of 35 C. Optimum DP-C02 treatment conditions for

microbial inactivation were determined to be 34.5 MPa for 6.9 min and 8% C02 at 35

C. Quality attributes, including pH, OBrix, % titratable acidity (%TA) and color of DP-

C02 treated, freshly thawed and heat pasteurized (90 C for 60 s) guava puree were

measured and compared throughout refrigerated storage (4 OC for 14 weeks). DP-CO2

treatment did not cause a change in pH or OBrix but increased the titratable acidity and

viscosity of the product. Pectinesterase enzyme (PE) was partially inactivated after DP-

C02 processing. DP-C02 treated guava puree retained organic acids similar to fresh

guava puree and served to protect polyphenolic and antioxidant levels throughout

processing and storage. DP-C02 delayed the degradation of vitamin C during storage.









Flavor and aroma compounds in guava puree were identified using GC-MS, Linear

retention index matching with databases and standards. Flavor profiles showed that

heat treated guava puree had less aroma active compounds than DP-C02 treated

guava puree. Volatile compounds analysis showed a lower total peak area for the DP-

C02 when compared to fresh and pasteurized and differences in volatile composition

were found for the three samples.

DP-C02 is an effective alternative to heat pasteurization of guava puree. It reduces

microbiological load, extends the shelf life, and preserves important sensory and

nutritional characteristics of the puree. The selection of this technique as a non-thermal

processing technology was based its effectivity and preservation of quality attributes.









CHAPTER 1
INTRODUCTION

Guava (Psidium guajava, L) is a tropical fruit rich in antioxidants and vitamin C. A

member of the Myrtaceae family, it is common to all tropical areas of America and can

be found in the West Indies, Bahamas, Bermuda and southern Florida.

Guava can be eaten raw or processed to obtain other products. In Hawaii, the

guava is boiled in slices to produce a guava juice. In Brazil, Mexico and Dominican

Republic the fruit is commonly processed to obtain a puree. In South Africa, the fruit is

trimmed, minced and mixed with a natural fungal enzyme to obtain a clear guava juice

with the ascorbic acid and other properties undamaged by the heat of pasteurization.

Guava juice and nectar are among the numerous popular canned or bottled fruit

beverages of the Caribbean area (Morton 1987). Guava fruit is rich in tannins, phenols,

triterpenes, flavonoids, carotenoids, vitamins and fiber. Most of the guava's therapeutic

activity is attributed to the high content of flavonoids, which also have antimicrobial

activity. Guava puree is normally processed by heat pasteurization to extend the shelf

life of the product (determined base on pectinesterase and microbial inactivation). The

shelf life of the puree is about one year, but the fresh taste is modified by heat

pasteurization. The use of non-thermal pasteurization has the potential to minimize the

development of undesirable characteristics, or loss of desirable characteristics, by

reducing the chemical changes that occur during heat processing.

The objective of this research was to test a new technology for the processing of

guava puree that has minimal effect on the chemical composition of the product,

particularly Vitamin C, total polyphenols, and flavor. The hypothesis was that the use of

non-thermal Dense Phase Carbon Dioxide (DP-CO2) pasteurization would minimize or









prevent undesirable changes in phytochemical composition compared to traditional heat

pasteurization. In order to validate this hypothesis, measurement and comparison of the

chemical composition of the guava puree subjected to both treatments and an untreated

guava puree (control) was conducted.









CHAPTER 2
LITERATURE REVIEW

Guava (Psidium guajava L.)

Guava (Psidium guajava, L) is a tropical fruit rich in antioxidants and vitamin C. It

is a member of the Myrtaceae family, which has more than 80 genera and 3000 species

distributed throughout the tropics and subtropics (Nakasone and Paull 1998). The

genus Psidium includes five species, Psidium guianense, P. cattleianum, P. chinense,

P. fridrichsthalianum and P. guajava. P. guajava is the most widely cultivated species of

the family Myrtaceae. The origin of the fruit is uncertain because it has been cultivated

by humans and distributed by humans and birds, but it is believed that the origin is

southern Mexico or Central America. It is common to all tropical areas of America and

can be found in the West Indies, Bahamas, Bermuda and southern Florida.

Guava Fruit and Fruit Products: World Production

The consumption trend of fresh tropical fruits and their products is increasing

steadily due to consumer's education about their exotic flavors, nutritive value, and

phytochemical content with potential health benefits (Commodity market review 2010).

World production of tropical fruits was estimated at 67.7 million tons in 2004,

representing a 2.5% increase compared to 2003. The minor tropical fruits, such as

lychees, durian, rambutan, guavas and passion fruits, recorded an output of 16 million

tons in 2004, accounting for a 3% increase (24% of the total tropical fruit production)

(Current situation and medium-outlook for tropical fruits 2010). Production of guava is

increasing in importance and in 2004 reached an estimated 4 million tons. India is the

main producer, followed by Pakistan, Mexico and Brazil. Minor production occurs in

Vietnam and Malaysia.









Fresh fruit consumption in the US is increasing (The world fresh food market

2010). Guava as an imported product is divided into four categories according to the

National Agriculture and Statistics Service (Hawaii guavas 2010) paste and puree,

preserved or prepared, jam, and dried. Brazil was the leader for guava paste, puree,

preserve or prepared imports into the US in 2008. Costa Rica was the main supplier of

guava jams and the Philippines was the main supplier of dried guava. US commercial

producers are located in Hawaii and southern Florida. Hawaii is the main grower, with

180 harvested acres and a utilized production volume of 3.5 million pounds in 2008.

This production represents a decrease in 19% from previous years due to the closure of

a large producer at the end of 2006.

Guava (Psidium guajava L.) has been catalogued as one of the most nutritious

fruits. The reason for this classification is its high content of phytochemicals, especially

because of it high ascorbic acid (Commodity market review 2010). Import of fresh guava

fruit is not possible or limited due to the tropical fruit fly (quarantine issues) and its very

short shelf life (7 to 10 days).

Characteristics

Psidium guajava is almost universally known by its common English name of

guava. In Spanish, the fruit is known as guayaba or guyava. The French call it goyave

or goyavier, the Dutch, guyaba and goeajaaba; the Surinamese, guave or goejaba; and

the Portuguese, goiaba or goaibeira. Hawaiians call it guava or kuawa. In Guam it is

abas. In Malaysia, it is generally known either as guava orjambu batu. It also has

numerous dialectal names in India, tropical Africa and the Philippines where the name,

bayabas, is often applied (Morton 1987).









The plant is a shrub or small tree reaching up to ten meters in height and is

adaptable to a wide variety of habitats. Because of its robust growth, the plant runs wild

and in some countries is considered a weed (Mitra 1997). The color of the trunk is often

mottled in appearance, with reddish-brown outer scale bark. Due to the beautiful

appearance of the trunk, in Florida it has been used in landscaping. The leaves are

oval or oblong, seven to fifteen centimeters in length with prominent veins. The flowers

consist of six white petals and numerous stamens with yellow anthers. As a crop, the

plant is mainly propagated by vegetative means, can bear fruits nine months after

planting and can continue to bear fruit for ten to twenty years. The fruit yield varies

depending on the cultural practices, but ranges from twenty-five to forty tons per hectare

per year.

The guava fruit is a berry. Morphologically, it may be round, ovoid, or pear-shaped,

with four to five protruding floral remnants (sepals) at the apex; and having a thin, light-

yellow skin, frequently blushed with pink. The fruit consists of a fleshy pericarp and seed

cavity with numerous small seeds that are hard and kidney shaped. There are seedless

varieties, but these typically contain a few seeds. The size of the fruit can range

between four and ten cm in length and four to eight centimeters in diameter. The

average weight is between 5 and 500 g (Salunkhe and Kadam 1995). The flesh color

can be white, yellowish, light- or dark-pink, or near-red. Depending on the color it is

classified into two main groups: white and red. The flesh is juicy, acidic, or sweet and

flavorful (sweet, musky and highly aromatic).

Chemical Composition

The chemical composition of the fruit varies with the stage of development, variety

and season. The titratable acidity (TA) reported as citric acid content ranges from 0.08









to 2.20% by weight. The total-soluble solids (TSS) content ranges from 8 to 19.40Brix,

the TSS/acidity ratio varies between 6.2 and 53.9 and the pH ranges from 4.1 to 5.4.

Guava fruits consist of about 20% peel, 50% flesh (pericarp) and 30% seed core

(Salunkhe and Kadam 1995). The fruit contains approximately 84% moisture, 26% dry

matter, 1.5% protein, 0.7% lipids and 1% ash. Carbohydrates are the principal

constituents of guava. Chan and Kwok (1975) reported the major carbohydrates in

guava variety Beaumont. They found 59, 34 and 5% fructose, glucose and sucrose,

respectively. The amount of these three sugars varies with variety and stage of

development. Mowlah and Itoo (1982) found that fructose is the predominant sugar

component in white and red guava, and it increased in all stages of maturation and

ripening. ElBulk and others (1997) studied the changes in chemical composition for four

cultivars of guava fruit during development and ripening. They found that for all cultivars

the sugar content increased gradually during the early stages of development and more

rapidly at the later stages of development. Among fruit types, guava is the second

highest in content of vitamin C, containing up to five times the concentration founded in

oranges (Dweck 2005). Vitamin C (ascorbic acid) is water-soluble and highly

susceptible to oxidative degradation, which often is used as an index for nutrient

stability during processing or storage (Damodaran and others 2008).

The vitamin C concentration fluctuates between 37 and 1,000 mg of ascorbic acid

per 100 g of guava fruit. The variation of vitamin C content depends on variety, stage of

development and season. Vitamin C concentration for red-fleshed guava is higher than

that of white-fleshed guava (Mowlah and Itoo 1983). Mercado-Silva and others (1998)

found that vitamin C increased with the maturation process and that guava harvested









during the autumn-winter-season had a higher content than those harvested during the

spring-summer season. Within each fruit, vitamin C concentration distribution is higher

in the skin than in the central portion of the flesh. Fruits also contain niacin (0.20 to 2.32

mg/100 g of fruit), thiamin (0.03- 0.07 mg/100 g), riboflavin (0.02 -0.04 mg/100 g), 3-

carotene (0.01 0.9 mg/100 g), calcium (10.0 30.0 mg/100 g), iron (0.60 1.39

mg/100 g and phosphorus (22.50 40.0 mg/100 g).

The composition of organic acids present in guava was studied by Chan and

others (1971). They found that citric and malic acids were predominant followed by

tartaric, glycolic and lactic acid (Chan and others 1971). Similar results were found by

Wilson and others (1982) in a study of four cultivars from Florida. They found traces of

fumaric acid, which was detected for the first time in guava fruits (Wilson and others

1982).

Guava fruits contain significant amounts of polyphenols but their concentration and

corresponding astringency decreases as the fruit matures. Examples of these

polyphernols are: gallic, ellagic and cinammic acid and others. The pigments present in

the guava fruit include but are not limited to carotene (such as lycopene, 3-carotene)

xanthophylls and chlorophyll.

Guava Constituents and Health Benefits

Fruits and vegetables are important components of a healthy diet and are one of

the main sources of antioxidants. Clinical research supports the fact that consumption of

fruits and vegetables is beneficial for prevention of cancer, heart disease and other age-

related diseases (Dietary guidelines for Americans 2010). Due to the recent increases in

obesity worldwide, the US Department of Health and Human Services reports that an

effective strategy for weight management should include increasing the consumption of









fruits and vegetables. This is based on the fact that this food group is high in water and

fiber which promotes satiety and decreases energy intake (Carlton-Tohill 2007). Most of

the research related to the relationship between fruit and vegetables consumption and

obesity focuses on macronutrients effects on satiety. The important point is that

consuming a diet rich in such plant foods will provide an abundance of dietary

antioxidants, including polyphenolics, vitamins E and C, and carotenoids, all of which

provide health benefits (Huang and others 2005). The guava fruit is rich in tannins,

phenols, triterpenes, flavonoids, carotenoids, vitamins and fiber. Most of the guava's

therapeutic activity is attributed to the high content of flavonoids, which also have

antimicrobial activity.

Vitamin C is commonly used to boost our immune system to fight colds and flu.

In addition, it works as an antioxidant, destroying free radicals that can cause cancer

and other diseases in the body. According to scientists from Cambridge University, a

boost of vitamin C intake reduces the risk of death from heart disease (E-Tropical Fruit

Net 2010). To determine the contribution of ascorbic acid (AA) to total antioxidant

capacity of guava, Leong and Shui (2002) measured the antioxidant capacity of fruit

using 2,2'azino-bis-(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) free radical

decolorizing assay and measuring the vitamin C content using HPLC. The ABTS

method measures the relative antioxidant ability of fruits to scavenge the radical

ABTS compared to a standard amount of ascorbic acid. Results are reported as AEAC

(mg of AA equivalents per 100 g homogenate). They reported an AEAC value of 270 +

18.8 mg/100 g and a vitamin C content of 131 + 18.2 mg/100 g. According to the AEAC

results, guava fruit is classified as a fruit with high antioxidant capacity. The ascorbic









acid accounts for a high percentage contribution to ABTS scavenging activity which

was 48.3% (Leong and Shui 2002).

Kondo and others (2005) studied the antioxidant activity in guava using 1-

diphenyl-2-picrylhydrazyl (DDPH)-radical scavenging activity and the effect on

superoxide. In addition, they measured the total phenolic content and quantified some

phenolics (gallic acid, catechin, epicatechin, chlorogenic acid and phloridzin). The

results showed that the superoxide scavenging activity decreased in the skin but

increased in the flesh during senescence. The superoxide scavenging activity was

higher in the skin than flesh until 42 days after full bloom, after which the flesh showed a

higher value than the skin. Phenolic concentration decreased in both skin and flesh

during senescence. The values decreased from 1322 to 915 pmol kg-1 and from 1126 to

637 pmol kg-1 on skin and flesh, respectively. The concentration of ascorbic acid

increased from 0.34 to 2.19 mmol/kg in the skin and from 0.26 to 2.05 mmol/kg in the

flesh. The only phenolic detected in the flesh was catechin while gallic acid, catechin,

epicatechin and chlorogenic acid were identified in the skin. From the four phenolic

compounds identified in the skin, catechin concentration found in the highest

concentration (45 pmol/kg of fresh weight) (Kondo and others 2005).

The health benefits of consuming a diet rich in dietary fiber (DF) have been

extensively studied (Gary 1999). The World Health Organization (WHO) and the Food

and Drug Administration (FDA) have recommended an increase in the daily intake of

dietary fiber (DF) over the past ten years. For the industry, the production of food

products rich in fiber is still a challenge. The selection of suitable sources to provide

new products with high DF and antioxidant capacity is important. The only food products









recognized, advertised and consumed as rich sources of fiber are those derived from

cereals, but the antioxidant capacities of these items are negligible. However, over the

past decade high dietary fiber materials from fruits (citrus, apple, and others) have been

introduced in the market. Fruit DF concentrates have in general a better nutritional

quality than those from cereals because of the presence of significant amounts of

associated bioactive compounds (flavonoids, carotenoids, etc.) and their balanced

composition (higher fiber content, soluble/insoluble DF ratio, water and fat holding

capacities, lower energy value, and phytic acid content) than cereal materials (Saura-

Calixto 1998).

Dietary flavonoids and other plant phenolics have been reported to have

antioxidant activity, antimicrobial and anti-inflammatory action (Huang and others 1992)

and have been associated with a reduced risk of cardiovascular diseases and cancer

(Temple 2000; Pietta 2000). Research was conducted to evaluate guava as a source of

natural antioxidant compounds and DF (Jimenez-Escrig and others 2001). The

researchers classified polyphenol compounds into two categories according to solubility:

extractable (EPP) and nonextractable polyphenols (NEPP) based on previous studies

related to physiological and nutritional properties of polyphenols associated with DF.

The basic structure of EPP are flavan-3-ol and flavan-3,4-diol, whereas that of NEPP is

condensed tannins. The basis for this classification was a study conducted by Bravo

and others (1994) which defined the properties of EPPs as related to their solubility on

the DF fraction. The properties of NEPPs had been related to the insoluble DF fraction

(Bravo and others 1994). Jimenez-Escrig and others (2001) conducted the study due

to the development and introduction of a new concept: antioxidant dietary fiber (AODF).









The main characteristics of this natural product are that they are rich in both DF and

polyphenolic compounds. They found a remarkable antioxidant capacity related to the

phenolics content. Peel and pulp of Psidium guajava fruit presented high levels of DF,

an indigestible fraction, and phenolic compounds. They concluded that guava could be

a rich source of natural antioxidants and dietary fiber.

Carotenoids have an important function as natural pigments, but some such as

provitamin A have been studied for their health benefits, such as prevention of

cardiovascular disease, immune-enhancement and inhibition of cancer (Mathews-Roth

and Krinsky 1985). Wilberg and Rodriguez-Amaya (1995) quantified the major

carotenoids present in fresh and processed guava. They found that provitamin A and

the principal pigment concentration varied within the fresh and processed fruit and was

highest in the ripe fruit. In the fresh fruit, they found a maximum concentration of 5.62

pg/g and 60.6 pg/g of 3-carotene and lycopene, respectively (Wilberg and Rodriguez-

Amaya 1995).

Other studies addressed guava fruit's carotenoid composition. Researchers

identified 16 carotenoids (3-carotene isomers, lycopene isomers, phytofluene, 3-

cryptoxanthin, rubixanthin, cryptoflavin, lutein and neochrome isomers), 13 of which

were identified for the first time (Mercadante and others 1998). Gorinstein and others

(1999) conducted comparative research of the content of total polyphenols and dietary

fiber in tropical fruits and persimmon. This research was conducted as a result of recent

studies that have shown that the consumption of dietary fiber and polyphenols of plant

products improve lipid metabolism. These authors concluded that the content of

polyphenols (4.79 5.11 mg/100 g fresh fruit), gallic acid (340.6 408.0 pg/100 g fresh









fruit), total fiber (5.14 6.04g/100 g fresh fruit) and soluble fiber (2.39 3.01g/100 g

fresh fruit) in guava were higher than the amount found in persimmon fruit (Gorinstein

and others 1999).

Recent research on flavonoids has shown that the biochemical and

pharmacological activity of these compounds include anti-oxidant, anti-allergenic, anti-

platelet, anti-inflammation effects and antithrombotic action (Cook and Samman 1996).

Miean and Mohamed (2001) studied the flavonoid content of edible tropical fruits. They

found a total flavonoid content of 1128.5 mg/kg dry weight due to the presence of

myricetin (549.5 + 0.5) and apigenin (579.0 + 0.02). These two flavonoids were found in

a methanolic extract obtained from guava and analyzed by HPLC (Miean and Mohamed

2001).

Alpha-tocopherol, the most common form of vitamin E is a lipid soluble vitamin that

protects our skin and other lipid-rich body constituents. This vitamin is found in nature

and has protective effect against oxidation of low-density lipoproteins, cell membranes

and DNA by free radicals. The Recommended Dietary Allowance (RDA) for daily

vitamin E intake for adult males and females is 15 mg (22.4 IU) (Vitamin E 2010). Ching

and Mohamed (2001) found 0.88 mg a-tocopherol per 100 g edible portion of guava

fruit.

Rahmat and others (2006) conducted a study to determine the effects of guava

(Psidium guajava) consumption on total antioxidant status and lipid profile (total

cholesterol, triglycerides, LDL-cholesterol and HDL-cholesterol) in young males. The

study was carried out over nine weeks. They found a significant increase of total

cholesterol, triglyceride and HDL-cholesterol during the treatment phase (4 weeks),









compared to the baseline (1 week) and control phases (4 weeks). The increase of HDL-

cholesterol was associated with the decreased risk of heart attack and cardiovascular

disease. The increase in total cholesterol and triglycerides during the treatment phase

was still in normal range. There was a significant increase of total antioxidants during

the treatment phase, compared to the baseline and control phases. The trends of

reduction of antioxidant enzymes (glutathione peroxidase and glutathione reductase)

were associated with decreased oxidative stress and decrease in free radical activities.

They concluded that the consumption of guava could result in improved antioxidant

status and lipid profile, reducing the risk of disease caused by free radical activities and

high cholesterol in blood (Rahmat and others 2006).

Guava Flavors

Aside from its nutritional value, the flavor of the guava is one of the most

distinguishable characteristics of this tropical fruit. Studies on guava volatile compounds

have been conducted using leaf, skin, fruit and fruit puree. Different types of extractions

and detection methods have been used. The first publication on the volatile constituents

of guava was in 1970. The compounds from the guava fruit were extracted using

distillation and the essence was analyzed by Gas Chromatography-Mass Spectrometry

(GC-MS). They identified twenty-two compounds present in the oil: eight alcohols, six

esters, three aldehydes four terpenes, one ketone and an alcohol. Hexanol and cis-3-

hexen-1-ol were found in higher concentrations, compared to the other six alcohols.

They identified methyl benzoate, 3-phenyl ethyl acetate, methyl cinnamate and

cinnamyl acetate and concluded that their combination probably contributed to the

overall aroma of the fruit (Stevens and others 1970).









Wilson and Shaw (1978) conducted research on terpene hydrocarbons in guava

using puree and a solvent extraction technique. The extract, which possessed a strong

guava aroma, was separated by TLC and only the terpene-containing fraction was

analyzed by GC-MS and GC-IR. They identified eleven terpenes from which limonene

and 3-caryophyllene comprised 95% of the total hydrocarbons present, and 3-

caryophyllene was the largest single component. In addition, they identified 3-

bisabolone and 3-copaene, which both were present in a greater quantities than that of

limonene (Wilson and Shaw 1978).

Macleod and Gonzalez de Troconis (1982) analyzed the volatile flavor component

of guava. Compounds were isolated from guava pulp using a Lickens and Nickerson

apparatus. For the essence analysis, GC-MS and a GC coupled to an olfactometry port

(GC-O) were used. Fifty-five % of the essence composition was due to esters. Other

compounds identified were two monoterpenes, and five sesquiterpene hydrocarbons,

myrcene being the major terpene. From the evaluation of odor, only eight compounds

of those identified showed significant aroma characteristics: three were esters, four

sesquiterpene hydrocarbons and myrcene. They conclude that 2-methylpropyl acetate,

myrcene, hexyl acetate, benzaldehyde, ethyl decanoate, 3-caryophyllene, a-humulene

and a-selinene were important contributors to fresh guava flavor and should be retained

as much as possible in processed products (Macleod and Gonzalez de Troconis 1982)

Indstein and Schreier (1985) extracted the volatile constituents from guava fruit

using high-vacuum distillation and liquid-liquid extraction. They fractioned the extract

using three different chromatography techniques: capillary gas chromatography with

flame ionization detector (HRGS), capillary gas chromatography-FTIR spectroscopy









(HRGC-FTIR) and capillary gas chromatography-mass spectrometry. In total, one

hundred fifty-four compounds were identified from which one hundred sixteen were

described for the first time as guava fruit constituents. The compounds found in the

highest concentration were: (E)-2-hexenal, hexanal, (Z)-3-hexenyl acetate, (Z)-3-hexen-

1-ol and 1-hexanol. Among the one hundred sixteen compounds identified for the first

time, eleven nitrogen and sulfur volatile compounds were found in the guava aroma

(Idstein and Schreier 1985).

In an attempt to establish a relationship between the proximate composition of the

fruit and the volatile constituents, Chyau and others (1992) studied the differences of

volatile and nonvolatile constituents between mature and ripe guava fruits. The isolation

of the compounds was performed using vacuum distillation followed by solvent

extraction. They found a total of thirty-four components (twelve esters, eight alcohols,

seven hydrocarbons, five carbonyls, one acid and eleven sesquiterpenes) from which,

seventeen were confirmed using pure compounds. The differences between the mature

and ripe fruit were in the quantitative analysis, not in the qualitative one. The major

constituents in mature guava were 1,8-cineole, (E)-2-hexenal and (E)-3-hexenal, while

in the ripe fruit ethyl hexanoate and (Z)-3-hexenyl acetate were the major volatiles

(Chyau and others 1992).

Paniandy and others (2000) studied the chemical composition of the essential oil

and headspace of white flesh guava fruit. They extracted the essential oil using steam

hydrodistillation and for the headspace analysis they used solid-phase microextraction

(SPME) technique. They identified sixty-four compounds in the oil and twenty-four in the

headspace. In the oil, the main contributors to the guava aroma were three









sesquiterpenes: 3-caryophyllene, a-copaene and a-humulene. Sesquiterpene alcohols

contributed 38% to the volatile composition. The products in the headspace of fresh

fruits were aldehydes, esters, lactones, myrcene and 3-caryophyllene. y-butyrolactone

was reported for the first time (Paniandy and others 2000).

Jordan and others (2003) characterized the aromatic profile in commercial guava

essence and fresh fruit puree using GC-MS and GC-O and methylene chloride as the

extraction solvent. In the commercial essence, 44 compounds were identified and

quantified including seventeen alcohols, seventeen esters, three ketones, two

aldehydes, two acids, one furan, one acetal and one terpene. In the puree, only twenty-

two compounds were identified and quantified as six alcohols, five esters, one ketone,

one aldehyde, two acids, one furan, one lactone, and five terpenic hydrocarbons.

Commercial essence was characterized by a volatile profile rich in low molecular weight

compounds such as alcohols, esters and aldehydes, whereas in the fresh fruit puree,

terpenic hydrocarbons and 3-hydroxy-3-butanone were the most abundant.

Olfactometry analysis yielded forty-three and forty-eight aroma active compounds in

commercial essence and fruit puree, respectively (Jordan and others 2003).

Mahattanatawee and others (2005) studied the volatile constituents and aroma

compounds of Florida-grown guava. They used SPME and liquid extraction to obtain the

aroma isolates. The isolates were analyzed by GC-MS and GC-O. Due to the

disadvantage of each one of the extraction methods, they used both techniques to

obtain a more complete aroma profile by GC-O. The combined data from two extraction

techniques resulted in detection of forty-eight compounds from which twenty-eight were

identified as being odor-active. The compounds in highest concentrations were hexanal









and 3-caryophyllene which have aroma activity and contribute to the green fruity and

warm floral notes of guava. Sulfur compounds like methanethiol, 2-methyl-3-furanthiol,

and mercaptomethylbutyl format were detected and contributed to the unique sulfury

note of guava aroma. There was no single character impact compound that contributed

to the aroma of the guava fruit (Mahattanatawee and others 2005).

Carasek and Pawliszyn (2006) used a commercial automated cold fiber

headspace solid-phase microextraction (CF-HS-SPME) device coupled to GC-MS to

identify the volatile compounds of guava. Thirty-three compounds (alcohols, aldehydes,

esters and terpenic compounds) were tentatively identified in the guava aroma, a large

number of them being esters and terpenoid compounds (Carasek and Pawliszyn 2006).

Steinhaus and others (2008) characterized the aroma-active compounds in guava

by application of the aroma extract dilution analysis. For the extraction of the volatile

fraction, the researchers used solvent extraction followed by solvent-assisted flavor

evaporation. The aroma-active areas in the chromatogram were screened by

application of the aroma extract dilution analysis. The distillate obtained from the

extraction represented a typical guava flavor with a green, sweet, tropical-fruit and also

grapefruit-like notes. A total of thirty-one odor-active regions were detected, whereas

two areas with caramel-like aromas, grapefruit-like odor and a black-currant aroma

quality showed the highest flavor dilution-factors (FD). The two compounds eliciting the

caramel-like aromas were identified as 4-methoxy-2,5-dimethyl-3(2H)-furanone and 4-

hydroxy-2,5-dimethyl-3(2H)-furanone. The grapefruit-like note and black-currant aroma

contributors were identified as 3-sulfanyl-l-hexanol and 3-sulfanylhexyl acetate

respectively. In addition, a seasoning-like smell, green/grassy, metallic and floral notes,









were identified as 3-hydroxy-4,5-dimethyl-2(5H)-furanone, (Z)-3-hexenal, trans-4,5-

epoxy-(E)-2-decenal and cinnamyl alcohol, respectively. Also ethyl butanoate, hexanal,

methional and cinnamyl acetate showed high odor activity. Among these compounds, 5

were identified for the first time in guava fruit (3-hydroxy-4,5-dimethyl-2(5H)-furanone, 3-

sulfanyl-l-hexanol, 3-sulfanylhexyl acetate, trans-4,5-epoxy-(E)-2-decenal and

methional) (Steinhaus and others 2008).

Steinhaus and others (2009) characterized the key aroma-active compounds in

guava by means of aroma re-engineering experiments and omission tests. Sixteen

compounds previously identified and mentioned above, in addition to acetaldehyde,

were quantified by stable isotope dilution assays. (Z)-3-hexenal, 4-hydroxy-2,5-

dimethyl-3(2H)-furanone, acetaldehyde and cinnamyl alcohol were found in the highest

amounts, whereas methional was approximately 0.001% of the amount. To estimate

the aroma potency of the individual guava odorants, their concentration was correlated

with the respective odor thresholds using odor activity value (OAV). (Z)-3-hexenal

showed the highest OAV (57000), followed by 3-sulfanyl-l-hexanol (9300), 3-

sulfanylhexyl acetate (360) and ethyl butanoate (170). Since the OAV of 4 compounds

were lower than 1, these compounds were assumed not to contribute to overall aroma.

In addition to the identification and OAV determination, an aroma reconstitution

experiment was conducted. An aqueous solution containing only 13 compounds

odorantss found to exceed their respective thresholds in the concentration determined)

was prepared and compared to a fresh puree using a sensory panel. The results from

the sensory panel showed good agreement between the aroma model solution and the

fresh puree, despite the fact that the simplified matrix did not include any nonvolatile









guava constituents in the model system. The artificial and natural aromas both were

characterized by a strong green- grassy note, a moderate grapefruit-like note, some

fruity and fresh notes, and weak sweet, flowery and metallic notes (Steinhaus and

others 2009).

Guava Fruit Processing

Guava fruit are mainly consumed fresh but also are processed and preserved as

different products. Guava is one of the easiest fruits to process, since the whole fruit

may be fed into a pulper for macerating into puree (Nunez-Rueda 2005). It is physically

and biochemically stable in relation to texture or pulp browning during processing (Brasil

and others 1995). It can be processed into a variety of forms, like puree, paste, jam,

jelly, nectar, syrup, ice cream or juice. Within the United States processing industry, it is

gaining popularity in juice blends. Guava pulp is extracted using a pulper, juicer or cloth

press. Its composition is similar to the fresh fruit, and it is further processed and utilized

in the form of puree, jam, juice and other products.

Guava puree is the most important raw material for the juice industry. The puree is

a liquid product prepared by pulping the fruit and is commonly used for the preparation

of nectars, beverages, blends, clarified juice, jams and jellies. The preparation of the

puree consists of washing the whole fruit, inspection for quality and feeding into a pulper

which removes seeds and fibrous fragments of skin. The finisher removes large

aggregates of stone cells and the residual stone cells may be ground by passing the

finished pulp through a mill. The milling operation improves mouth feel but downgrades

the color quality. An alternate method to milling is centrifugation, which improves mouth

feel and reduces precipitation in the product. The puree is preserved by freezing to -20

to 0F (-29 to -180C), canning, aseptic packaging or pasteurization. Pasteurization is









conducted between 80 and 900C (190 1940F) for 60 seconds, then the pasteurized

product is cooled and filled into containers. After pasteurization, the puree can be frozen

and stored at -180C (00F) for up to a year.

The food industry uses the puree to manufacture nectars, guava juice, juice blends

and other products. The manufacture of clear juice from guava is difficult. The colloidal

particles which cause turbidity carry flavor substances and natural antioxidants. To

obtain a clarified juice enzymes can be used to clarify juices or to change the viscosity

of the fruit juices. The enzymes that are most commonly used are cellulases or

pectinases.

There are different processes to prepare a clarified juice. One involves maceration

of fruit and mixing with enzymes. After a certain amount of time, which depends on the

enzyme and temperature used for treatment, the pulp is passed through a press to

obtain the juice. With this treatment more than 80% of juice can be obtained. Another

procedure involves the direct treatment of the puree with enzymes.

Two different methods for manufacturing clarified juice have been developed. In

the first method, whole guavas are frozen to break their cellular structure and are kept

frozen until needed. The problem with this procedure is the low yield. The second

method uses guava puree as starting material. In this method, the puree is thawed and

pressed mechanically using a press cloth, and then filtered. It is advisable to warm the

puree to 400C and add a filtering aid such as diatomaceous earth (Imungi and others

1980).

A few studies have been conducted on enzyme treatment of guava puree to obtain

a clarified juice that later can be used to produce a concentrate without affecting









desirable fruit attributes such as flavor and nutritional properties (Brekkee and others

1986; Imungi and others 1980; Hodgson and others 1990).

Brasil and others (1995) studied the physical-chemical changes during extraction

and clarification of guava juice. They treated guava pulp with 600 ppm of a pectic

enzyme at 450C. After the treatment, the pulp was pressed and the cloudy liquid was

treated with fining agents and filtered. A higher yield was obtained after 120 min of

treatment time. The viscosity of the treated product decreased by 62.9% related to the

pulp. They concluded that this clarification technique followed by treatment with a fining

agent and filtering showed good results and stability in nutritional and organoleptic

characteristics (Brasil and others 1995).

Guava Puree Processing

Guava puree is normally processed by heat pasteurization to extend its shelf life.

The heat pasteurization serves as a preserving method but is carried out at conditions

for pectinesterase inactivation which is more severe. The shelf life of the puree can be

extended to one year, but the fresh taste is modified by deteriorative reactions resulting

in decreased sensory quality.

Yen and Lin (1992) studied the changes in flavor components of guava puree

during processing and frozen storage. They pasteurized puree at 85-88 C for 24

seconds. After pasteurization, the puree was immediately cooled, packed and stored at

three different temperatures for analysis over a period of 4 months. The volatiles were

extracted using Likens-Nickerson apparatus and a GC-FID and GC-MS were used for

the identification of compounds. Initially, the volatile constituents from the pasteurized

puree were similar to the unpasteurized puree. Terpene hydrocarbons were the major

volatile components followed by aldehydes. Changes in volatile constituents were









significant during the first two months of storage at 0 C. This result was attributed to

both oxidation and enzymatic reactions since the puree showed an increase in total

plate count and existing enzyme activities (such as peroxidase, pectinecterase and

polyphenoloxidase). However, the quality of guava puree stored at -20 C after 4

months was satisfactory. The deterioration of flavor quality for guava puree during

pasteurization and frozen storage may result from the changes in certain volatile

components (Yen and others 1992).

Chan and Cavaletto (1982) studied the changes in chemical and sensory quality

during processing and storage of aseptically packaged guava puree. The puree was

acidified to pH 3.9 with citric acid and the soluble solids content was 13.5%. The heat

treatment was conducted at 93 C for 26 seconds and the product was aseptically

packed. The packed puree was stored at ambient temperature and sampled after 1, 3

and 6 months storage, while the accelerated samples (stored at 38 C) were sampled

after 1, 2 and 3 months. All samples were compared to frozen puree. After 3 months of

ambient storage, the ascorbic acid retention was 72% while those stored at 38 C

retained 62% of the ascorbic acid. There were no significant changes in pH, total acids

or oBrix. Aseptic processing appears to cause a lightening of the puree color. However,

storage at 38 C showed darkening in color, while ambient temperature storage did not

cause a significant darkening of samples. Aseptic processing also caused a decrease in

both a* and b* values which also was observed by a sensory panel as a loss of pink

color. In sensory tests, flavor was not greatly affected as color and the flavor changes

were the result of storage time and not processing (Chan and Cavaletto 1982).









Microorganisms, which can be minimized by pasteurization, play an important role

in spoilage of foods. Considerable research has focused on the identification of new

processing technologies to avoid the detrimental chemical changes caused by heat

pasteurization. High pressure processing (HPP) causes minimal changes in the "fresh"

characteristics of foods by eliminating thermal degradation. Compared to thermal

processing, HPP may provide a product with fresher taste and better appearance,

texture and nutrition. High pressure processing is an alternative to heat processing and

may have potential as a food processing method.

Research was conducted to compare the effects of a high pressure treatment with

thermal pasteurization on guava puree. The researchers concluded that high pressure

maintained the original flavor of the juice. The volatile flavor components of the

pressure-processed guava juices stored for 30 days at 4 oC were similar to those of

fresh juices (Yen and Lin 1999).

Another study compared the effect of high pressure treatment and thermal

pasteurization on the quality and shelf life of guava puree. The puree had a pH of 3.8

and 8.2 oBrix. The high pressure treatment was carried at 400 and 600 MPa for 15 min

at a temperature of 25 oC. The pasteurization process was carried out at a temperature

of 88 90 oC for 24 seconds. Samples from both treatments were stored at 4 oC over a

period of 60 days. All treatments were equally effective in reducing the microbial load of

the puree. During storage, pressurized puree at 600 MPa showed lower levels of

microorganisms. Complete inactivation of POD enzyme was achieved by heat

pasteurization, however PPO and PE activities of 16 and 4% respectively were found

after the heat treatment. Both of the pressurized treatments showed residual activity for









the three enzymes. Cloud analysis showed that loss of cloud was greater in untreated

puree than in pressurized and heated puree during storage. Color of pressurized guava

puree was similar to that of freshly extracted puree. This research indicated that high

pressure treatment of guava puree at 25 C for 15 min could maintain good quality up to

40 days of storage at 4 oC (Yen and Lin 1996).

Phytochemicals

In addition to macronutrients (nutrients that are involved in normal metabolic

activity) food contains components that may provide additional health benefits such as

phytochemicals which are derived from naturally occurring compounds (Bloch and

Thomson 1995). Phytochemicals are sometimes referred to as phytonutrients. They are

natural bioactive compounds found in plant foods that work with nutrients and dietary

fiber to protect against diseases. They also serve as protective agents for plants and

some of them provide positive health benefits. They are synthesized as secondary

metabolites in all plants as a product of nutrient intake, protein synthesis and

photosynthesis (Nunez-Rueda 2005). They are considered nonnutritive substances

because they are not needed for regular metabolism. Within this group of nonnutritive

substances are phenolic compounds, terpenoids, pigments and other natural

antioxidants. Nuts, whole grains, fruits and vegetables contain an abundance of these

substances that have been associated with protection from chronic diseases such as

cancer and heart disease (Craig 1997). Polyphenols are the most abundant antioxidants

in the diet. Their total dietary intake could be 10 times higher than vitamin C

consumption, which is much higher than that of all other classes of phytochemicals

(Scalbert and others 2005).









The term 'phenolic' or 'polyphenol' may be identified chemically as a substance

which possesses an aromatic ring attached to it one or more hydroxyl groups, and may

include functional derivatives such as esters, methyl esters, glycosides or others (Ho

and others 1992).

Phenolic compounds are responsible for major organoleptic characteristics of

plant-derived foods and beverages, particularly color and taste. Plant polyphenols

comprise a great diversity of compounds that are usually divided into two groups:

flavonoids and non-flavonoids. Non- flavonoid compounds are mostly simple molecules

such as phenolic acids and complex molecules derived from them (such as

hydroxycinnamic acid derivates). Flavonoid compounds share a common structure

consisting of two phenolic rings and oxygenated heterocycles, and they are sub-divided

into several groups, based on the oxidation state of the pyran ring (Cheynier 2005).

Phenolic plant compounds, including all aromatic molecules from phenolic acids to

condensed tannins, are products of a plant aromatic pathway, which consists of three

main sections: the shikimic acid pathway which produces the aromatic amino acids

phenylalanine, tyrosine and tryptophan that are precursors of phenolic acids; the

phenylpropanoid pathway which yields cinnamic acid derivatives that are precursors of

flavonoids and lignans; and the flavonoid pathway which produces various flavonoid

compounds (Bruyne and others 1999).

Classification

The term phenolics encompasses approximately 800 naturally occurring

compounds, all of which possess a common structural feature: a phenol unit (C6).

Current classification divides these compounds in two broad categories: polyphenols

(possess at least two phenol subunits) and simple phenols or phenolic acids.









Polyphenols can be further divided in two categories: those containing only two phenolic

units (flavonoids) and those containing more than three phenolic units (tannins).

Phenolic acids are naturally occurring compounds that contain two distinguishing

consecutive carbon frameworks: hydroxycinnamic and hydrobenzoic structures.

Although the basic skeleton remains the same, the number and position of hydroxyl

groups on the aromatic ring create the variety (Robbins 2003). Hydroxylated acid

derived from benzoic acid include gallic acid, the main phenolic unit of hydrolyzed

tannins and hydoxylated acids derived from cinnamic acid including coumaric, ferrulic

and caffeic acids. Hydroxycinnamic acid derivatives represent the major group of plant

phenolics since they form the basic constituents of lignins.

Flavonoids represent the most common and widely distributed group of plant

phenolics. Their common structure is that of diphenylpropanes (C6C3C6) or flavan

nucleus which consists of two aromatic rings linked through three carbons that usually

form an oxygenated heterocycle. They usually occur in plants as aglycones, although

they are most commonly found as glucoside derivates (Laura 1998). Flavonoids are

formed in plants from the aromatic amino acid phenylalanine and tyrosine. The various

classes of flavonoids differ in their level of oxidation and pattern of substitution of the C

ring, while individual compounds within a class differ in the pattern of substitution of the

A and B ring (Pietta 2000). The flavones (such as apigenin and luteolin), flavonols (such

as quercetin, myricetin and kaempferol) and their glycosides are the most common

compounds.

Tannins make up another group of natural polyphenols. They are classified as

hydrolysable and non-hydrolysable tannin. The hydrolysable tannins are those









compounds that can be fractioned hydrolytically into their components. This group

includes both the gallotannins and ellagitannins. Gallotanins are all those tannins in

which galloyl units are bound to another compound such as catechin. Ellagitannins are

the tannins in which at least two galloyl units are coupled together by a single carbon-

carbon bond and do not contain a catechin unit glycosidically attached. The biosynthetic

pathway to hydrolysable tannins may be divided into three routes. The first section

involves the esterification of a free gallic acid unit with glucose which undergoes further

esterification to form the end product pentagalloylglucose. Pentagalloylglucose is the

starting point for the two subsequent routes. The gallotannin route is characterized by

the addition of galloyl residues to pentagalloylglucose. The ellagitannin routes are

oxidation processes that yield carbon-carbon linkages between the galloyl groups of

pentagalloylglucose (Grundhofer and others 2001). The non-hydrolysable tannins are

also known as condensed tannin and complex tannins (Khanbabaee and Ree 2001).

Complex tannins are tannins in which a catechin unit is glycosidically bound to a

gallotannin or ellagitannin. Condensed tannins are all oligomeric and polymeric

proanthocyanidins. Procyanidins consist of chains of flavan-3-ol-units, which are

commonly esterified, mainly with gallic acid units. Flavan-3-ols are derived from a

branch of the anthocyanin and other flavonoids pathway, of which elucidation is still

unclear (Dixon and others 2005). Structural variability among proanthocyanidins

depends on hydroxylation, stereochemistry at the three chiral centers, the location and

type of interflavan linkage, and terminal unit structure. A classical assay for

proanthocyanidins consists of an acid hydrolysis, where the terminal units of the

molecules convert to colored anthocyanidins.









Attributes

Phenolic compounds have been associated with positive and negative attributes in

terms of sensory and nutritional quality. Positive attributes include their close

association with sensory and nutritional quality. Their nutritional value has been linked

to: prevention of cancer, antimicrobial properties, antimutagenicity, antioxidant potential,

reduction of coronary heart disease risk, antiviral, anti-inflammatory and antitumor

activity (Sonko and Xia 2005). Their sensory attributes are related to their contribution to

flavor, astringency and color characteristics of foods. The anti-nutritional effect of

phenolic compounds involves their reaction with proteins, carbohydrates, minerals and

vitamins lowering the bioavailability of these nutrients or their nutritional value. In

addition, phenolic compounds can adversely affect the sensory qualities of food by the

production of off-flavor, their involvement in enzymatic browning or enzymatic

discoloration, nonenzymatic discoloration and precipitation of proteins (Shahidi and

Naczk 2003).

Extraction and Analysis

Extraction of phenolic compounds in plant materials is influenced by their chemical

nature, the extraction method employed, sample particle size, storage time and

conditions, and the presence of interfering substances. The chemical nature of plant

phenolics varies from simple to highly polymerized substances that include varying

proportions of phenolic acids, phenylpropanoids, anthocyanins and tannins, among

others. They may also exist as complexes with carbohydrates, proteins and other plant

components. Therefore, phenolic extracts of plant materials are always a mixture of

different classes of phenolics that are soluble in the solvent system used. Additional

steps, such as solid phase extraction (SPE), may be required for the removal of









unwanted phenolics and non-phenolic substances. Phenolics solubility is governed by

the type of solvent used, degree of polymerization, as well as the interaction of

phenolics with other food constituents. Some of the solvents most frequently used for

phenolic compound extraction are methanol, ethanol, acetone, water, ethyl acetate and,

to a lesser extent, propanol, dimethylformamide, and their combinations. Extraction

periods usually vary from 1 min to 24 h (Naczk and Shahidi 2004). Phenolic compounds

can be quantified by spectrophotometric or chromatographic methods; in addition,

separation and quantification can be done by chromatographic methods. A number of

spectrophotometric methods for quantification of phenolic compounds have been

developed and they involve the reaction of the sample containing the phenolic

compound with a specific reagent. These assays are based on different principles and

are used to determine various structural groups present in phenolic compounds. The

Folin assay is the most widely used procedure for quantification of total phenolic content

in plant materials. The disadvantage of this method is that it can detect phenolic groups

in proteins. Chromatographic methods includes: gas chromatographic techniques

(which require a sample preparation and derivatization) and high performance liquid

chromatography (HPLC).

Sensory Evaluation and Flavor Analysis

Sensory Evaluation

Sensory evaluation is the assessment of all qualities for food products as

perceived by human senses. It is a quantitative science in which numerical data are

collected to establish specific relationships between product characteristics and human

perception. The main applications of sensory evaluation in the food industry are in

quality assurance and product development (Murano 2003).









The order of perception for attributes of food is appearance, odor/aroma/fragrance,

consistency and texture, and flavor aromaticss, chemical feelings and taste). Sensory

testing involves the use of people as measurement devices. Although a specific sensory

test cannot provide definitive answers to all questions, it is a key part of a larger

sequence of information gathering during the product development process. To obtain

meaningful data, it is important to match the test objective with the type of test used.

Many variables must be controlled if the results of a sensory test are to measure the

true product differences under investigation. It is important to group these variables

under three major categories: test controls (e.g. room and environment), product

controls (e.g. equipment used, samples temperature) and panel controls (e.g. procedure

used by panelist to evaluate sample) (Meilgaard and others 2007). The test methods

can be classified according to their primary purpose in three different classes: affective,

discrimination and descriptive.

Affective tests attempt to quantify the degree of liking and disliking of a product

over another product. This type of test is used in consumer testing because they

measure preference and acceptance. These groups of tests use untrained panelists.

Preference testing uses a hedonic scale (like or dislike). Ranking is a type of preference

test where the consumers are allowed to order a group of products base on their degree

of liking or disliking. In acceptance testing, panelist rate their liking or disliking on a

scale.

Discrimination tests attempt to investigate if any differences exist between two

types of products but the degree of difference is not determined. Examples of this type









of test are: duo -trio, difference from control, paired comparison and triangle tests.

These tests do not employ trained panelists.

Descriptive testing is used with trained panelists to describe specific product

attributes related to flavor, texture, mouth feel and other characteristics of the product

and to quantify the perceived intensities in each one of the evaluated attributes.

Panelists are highly trained in relation to the scale, and the characteristics or attributes

evaluated. Examples of this type of test include flavor profile method, quantitative

descriptive analysis and texture profiling.

Flavor Analysis

Flavor is usually divided into taste (detected in the mouth) and smell (detected in

the nose). Flavor perception depends on the combined responses of our senses and

the cognitive processing of these inputs. Taste is the combined sensations arising from

specialized test receptor cells located in the mouth (tongue and throughout the oral

cavity). Olfaction is the sensory component resulting from the interaction of volatile

components in food with olfactory receptors in the nasal cavity. The stimulus of the

aroma or odor of food can be orthonasal (odor molecules enters the olfactory region

through the nose) or retronasal (odor molecules enter the nasal cavity through the back

of the tongue). Consumers consider flavor as one of three main sensory properties they

use in their selection, acceptance and ingestion of a particular food (Fisher and Scott

1997).

Aroma is a very complex sensation (Reineccius 2006). Most of the aromas present

in food systems consist of a mixture of several aroma compounds rather than a single

aroma chemical. Flavor and/or aroma are usually created by mixing many flavor









materials at the proper concentration of each component to produce the desired flavor

characteristics and profile (Spanier 2001).

The first step in the characterization of odor-active compounds in a complex

mixture or food system is to separate them from nonvolatile compounds. When isolating

the flavor compounds, a reduction of matrix interference will occur. This separation is

accomplished through a variety of techniques, such as solvent extraction, head-space

concentration, and distillation. The extraction procedure may distort or alter the

chemical composition, because each one of these methods is selective for some

compounds and probably not for all types, so there is no perfect extraction technique.

Each of these techniques yields a concentrated essence containing the odor-active

chemicals. Subsequently, separation, detection, identification, and characterization of

individual compounds are possible with sophisticated instrumentation (such as GC-MS,

GC-S and GC-O).

Flavor Extraction Techniques

Methods that have been used for extraction and concentration of flavor

compounds include steam distillation, liquid-liquid extraction, trapping of the volatiles on

adsorbents, and combinations of these methods with other techniques. They allow

extraction and concentration of the compounds from their matrix. The main drawbacks

of steam distillation extraction are the possible generation of thermal artifacts, foaming

and gel formation. Some disadvantages of liquid-liquid extraction include the variation of

compounds extracted. This is mainly due to the fact that different compounds have

different partitioning coefficients. When using solvents, water soluble compounds are

not extracted extensively while lipid-soluble compounds will be extracted more

effectively. Another problem with solvent extraction is the increase in time needed to









extract the compounds because the solvent must be evaporated and contamination with

solvent impurities is possible.

A rapid, simple and inexpensive technique [Solid Phase Microextraction (SPME)]

was developed by Pawliszyn and co-workers in 1990 (Kataoka and others 2000). Solid

Phase Micro-Extraction (SPME) is a non-solvent sample preparation technique that

uses a fused-silica fiber coated on the outside with an appropriate stationary phase,

allowing the analyte in the sample to be directly extracted. The principle of SPME is the

absorption of analytes onto a fiber followed by desorption in the injector port of a gas

chromatograph. The SPME device consists of a fiber made of fused silica gel, coated

with a stationary phase and bonded to a stainless steel plunger and a holder that looks

like a modified syringe. Fiber coating material varies in composition, such as non-polar

polydimethylsiloxane and polar carboxen, and thickness or amount of coating. These

variations help to increase the sensitivity of the extraction depending on the analytes'

chemical characteristics. SPME has a very effective concentrating effect and leads to

good sensitivity (Supelco 2005). The amount of extracted analyte depends on the

thickness of the polymer coating and the distribution constant for the analyte rather than

sample volume. The distribution constant of each analyte depends of the equilibrium

established among the concentrations of analytes in the sample, in the headspace

above the sample and in the coating material on the fiber. The extraction time is

determined by the length of time required to obtain precise extraction for the analyte

with the highest distribution constant, which generally increases with increasing

molecular weight and boiling point of the analyte. Full equilibrium is not necessary for

high accuracy and precision, but consistent sampling time and other sampling









parameters (e.g. sampling temperature) are essential. It is also important to keep

consistent the vial size, the sample volume and depth the fiber is immersed in the

sample (Lord and Pawliszyn 2000).

Volatile Identification Techniques

After isolation of the volatiles using an extraction and concentration procedure,

samples are injected into GC for separation of individual compounds. Identification of

flavor compounds in food is performed by retention time match with standards or pure

compounds. Actual identification of flavor compounds requires at least two independent

confirmation techniques. Since retention time is dependent on a number of factors (such

as column type, column length, carrier gas flow, etc.), they cannot be used to identify a

compound based on literature values. To compare retention times from a single

analysis to literature values, the standardized retention index needs to be calculated.

Standardized retention index is a way to compare chromatographic results using

different experimental conditions because they are independent of column length,

carrier gas flow and film thickness. A retention index value gives an indication of where

the compound of interest elutes relative to straight-chain hydrocarbons. The linear

retention index, or Kovats index, expresses the number of carbon atoms, multiplied by

100, of a hypothetical normal alkane which would have an adjusted retention time

identical to that of the peak of interest when analyzed under identical conditions. With

this information, linear retention index values can be used to evaluate the elution of a

mixture of compounds on a specific column for a given set of conditions. The retention

index of an unknown measured on several different columns is also useful for identifying

the unknown by comparison with tabulated retention indexes.









Separation and Detection of Aroma Volatiles

Volatiles are separated by GC and their elution time is monitored relative to a

series of n-alkanes that are injected under identical conditions. The detection of these

compounds is achieved with the use of different detectors which are classified as either

selective or non-specific detectors. The most common detector is the flame ionization

detector (FID). It is a non-specific detector because it responds to all organic

compounds that burn or ionize in the flame (Wrolstad and others 2005). It has good

sensitivity, a wide linear range in response and is used in almost all food analyses

where a specific detector is not desired or sample destruction is acceptable.

A selective detector is used for specific analytes. An example of this type of

detector is the Flame Photometric Detector (FPD). As the compounds elute from the

column, they are burned and measured by the light emitted from the flame at specific

wavelengths. The wavelengths of light that are suitable in terms of intensity and

uniqueness are characteristics of sulfur and phosphorous. Thus, this detector gives a

greatly enhanced signal for those two elements (Nielsen, 2003). A Pulsed Flame

Photometric Detector (PFPD) is used for the identification of sulfur containing

compounds. In the PFPD, the flame is pulsed and is 10 to 100 times more sensitive

than FPD.

Other chromatography techniques are used for separation, detection and

identification of flavor compounds. The most common of these techniques is GC-MS.

This instrument consists of an MS coupled to a GC. The separation of the compounds

occurs in the GC and the MS allows the peaks to be quantitated and identified or

confirmed, and, if an unknown is present, it can be identified using a library containing

MS spectra. Another technique is a gas chromatography-olfactometry (GC-O). This









instrument is used to distinguish aroma-significant compounds from the less important

volatiles present in a sample matrix. Simultaneous with the high resolution gas

chromatographic separation of a volatile extract, the odor of individual compounds is

assessed by sniffing the effluent off the GC column in parallel with electronic detection.

GC-O is a technique by which the human, rather than the machine, responds to odor

detection. Historically, analytical instruments were utilized to detect components through

an electronic device; however, the detection capabilities were limited. Detection by the

human nose occasionally is more effective than the electronic device. This technique

enables the detection of odor-active volatiles, the determination of their odor qualities

and the relative aroma intensity.

Beverage Processing

The food industry is continuously searching for novel processing technologies that

ensure microbial destruction and extend shelf life of products without having adverse

effects on their quality attributes (Butz and Tauscher 2002). Current trends in food

marketing showed that consumers desire high-quality foods with "fresh-like"

characteristics and enhanced shelf life that require only a minimum effort and time for

preparation (Butz and Tauscher 2002). A wide variety of emerging non-thermal

processing technologies are available to process food and beverages and include high

hydrostatic pressure, irradiation, ultrasound, pulsed electric fields, light pulses, and

oscillating magnetic fields. Dense phase carbon dioxide (DP-C02) pasteurization is a

promising alternative to traditional pasteurization technologies and may lessen

detrimental effects to thermolabile phytonutrients and flavor compounds (Gomes and

Ledward, 1996; Sun and others 2002; Zabetakis and others 2000). Although the use of

DP-C02 processing has been shown to inactivate microorganisms, its effect on food









quality characteristics needs more examination. The use of non-thermal pasteurization

can minimize the development of undesirable characteristics by reducing the chemical

changes that occur during heat processing.

Thermal processing

Heat treatment is used in food processing to achieve preservation of food.

Thermal processing was developed by Nicolas Appert more than 200 years ago. The

most important heat treatments used in food industry are: pasteurization, flash

pasteurization, aseptic packaging and canning.

Pasteurization involves a low level heat treatment; below the boiling point of water.

This thermal treatment has two primary objectives. The first objective is to destroy

microorganisms known to occur in some type of food, like milk and egg products, that

could affect public health. The second objective is to extend products' shelf life.

Pasteurization does not kill all microbial flora, so pasteurized products will contain living

organisms capable of growing and limiting the storage of the product.

Flash pasteurization is a high-temperature short-time (HTST) treatment in which

pourable products are heated during 3 to 15 seconds to temperatures that destroy

pathogenic microorganisms. It is a very rapid form of aseptic processing. Aseptic

processing uses temperatures higher than flash pasteurization to treat the product. After

the treatment, hot product, clean containers and clean closures are brought together in

an environment that prevents recontamination of the product. This operation normally

takes place in a closed space under pressure with sterile air.

Canning involves the use of specific times and temperatures of heating that are

defined by thermal death time (TDT), which identifies the parameters required to

destroy the spores of Clostridium botulinum in low acid products.









Non-thermal Processing

Thermal processing is still the major technique for shelf-stable food preservation.

There is a general movement in food processing away from high-heat treatment and

deep-freezing toward milder treatments, resulting in refrigerated foods with less cooked

flavors. There is always the need to kill pathogens, but there is also a demand for

"clean" labels, meaning a preference for few, if any, chemical additives and

preservatives (Clark 2009).

Non-thermal methods allow the processing of foods below temperatures used

during thermal processing, so flavor, essential nutrients and vitamins suffer minimally or

not at all (Butz and Tauscher 2002). However, non-thermal technologies not only

improve food quality, but also promote an equivalent or preferably, an enhanced level of

safety, when compared to procedures they replace (Raso and Barbosa-Canovas,

2003). Irradiation, ultra-high pressure, pulsed electric fields, DP-C02 and pulsed

magnetic fields are non-thermal technologies attracting interest and gaining acceptance

as food processing methods.

Dense Phase Carbon Dioxide

Dense Phase Carbon Dioxide (DP-C02) is a cold pasteurization method that

affects microorganisms and enzymes through the effect of CO2 under pressure below

50 MPa without affecting the fresh-like physical, chemical and sensory qualities. Carbon

dioxide, a natural constituent of many foods, is a non-toxic, nonflammable, inexpensive

gas and has a Generally Recognized as Safe status (Damar and Balaban 2006).

Mechanisms of Microbial Inactivation

A number of hypotheses have been proposed to explain the effect of microbial

inactivation caused by DP-C02, including cytoplasmic pH decrease, explosive cell









rupture due to internal pressure, modification of cell membrane and extraction of cell

wall lipids, inactivation of key enzymes for cell metabolism and extraction of intracellular

substances.

Cytoplasmic pH decrease or acidification has been proposed as the main

mechanism for microbial inactivation. In this mechanism, C02 is dissolved in an

aqueous solution forming carbonic acid, which at a sufficient concentration is

dissociated into bicarbonate and hydrogen ions lowering the extracellular pH.

The first theory proposed for microbial inactivation was the explosive cell rupture

due to internal pressure. It was thought that during the rapid depressurization of the

sample, the C02 would have rapidly expanded through the cells so that a part of them

could have been mechanically broken. However, pictures of microbial cells after

treatment have shown that the mechanism of inactivation did not always involve cell

rupture (Spilimbergo and Bertucco 2003).

Modification of cell membranes and extraction of cell wall lipids is another

hypothesis for microbial inactivation. This mechanism is based on the lipophilic and

solvent characteristics of C02. The cell membrane consists of a double layer of

phospholipids. The C02 could easily penetrate into the membrane, leading to an

increase of is fluidity and permeability, altering the characteristics of the membrane and

destroying its essential function. This mechanism is known as anesthesia effect. The

anesthetic theory is a strong explanation for microbial inactivation since images showed

a modification of cell membrane with possible leakage of cytoplasm, together with

enlarged periplasmic space between the walls and the cytoplasmic membranes

(Spilimbergo and Bertucco 2003).









The theory of the inactivation of key enzymes for cell metabolism is based on the

interference of bicarbonate and molecular carbon dioxide on certain enzymatic and

biochemical pathways. Another proposed mechanism is precipitation of intracellular

calcium and magnesium carbonate ions from bicarbonate. This can occur since there

are some proteins sensitive to calcium and magnesium that could be precipitated by

carbonate.

Factors Affecting Microbial Inactivation

Factors that affect microbial inactivation using DPCD include water content within

the cell (more log reductions are achieved in wet cells than dry cells) and water activity

of the food (DPCD is more effective as water activity increases). Generally, any factor

that increases levels and rate of C02 solubility enhances microbial inactivation caused

by DP-C02. For example, C02 solubility increases with increasing pressure when

temperature, residence time and C02 concentration are equal. Generally, inactivation

efficiency increases with increases in pressure, temperature and residence time.

Temperature has a complex effect on microbial inactivation. Even when the CO2

solubility decreases with increasing temperature, inactivation of microbes by DPCD is

more effective at high temperature. Higher temperature increases the C02 diffusion and

fluidity of the cell membrane. Another effect of temperature is the change of CO2, since

its penetrating power is higher under supercritical conditions and there is a rapid change

in solubility and density when processing temperature is near this critical region. The

initial pH of the medium is another factor that affects effectiveness of DP-C02 microbial

inactivation. Acid medium facilitates carbonic acid penetration through the cell

membrane, allowing a higher inactivation. The final factor affecting microbial inactivation

is the cell growth phase; young cells are more sensitive than mature ones. The type of









bacteria affects the DP-CO2 effectiveness. Gram positive bacteria showed more

resistance than gram negative bacteria due to the differences in their cell membrane

composition.

Solubility of C02

002 solubility in liquid foods can be affected by pressure, temperature, and food

composition. Pressure has a direct effect on CO2 solubility: as pressure increases, CO2

solubility increases. On the other hand, as temperature increases, solubility of CO2

decreases. Food composition may increase or decrease the solubility of CO2 (Calix and

others 2008).

Types of Systems

Three different types of DP-CO2 equipment have been developed: batch, semi-

continuous and continuous. The batch system was the first system developed. In this

system, CO2 and the food to be treated are stationary in a container during treatment.

This system consists of a CO2 gas cylinder, a pressure regulator, a pressure vessel, a

water bath or heater, and a CO2 release valve. The sample is placed into the pressure

vessel and temperature is set to the desired value. CO2 is then introduced into the

vessel until the sample is saturated at the desired pressure and temperature. The

sample is left in the vessel for a period of time, after which the CO2 outlet valve is

opened to release the gas. Some systems contain an agitator to decrease the time to

saturate the sample with CO2 (Damar and Balaban 2006).

A semi-continuous system allows a continuous flow of CO2 through the chamber

while a continuous system allows flow of both CO2 and the liquid food through the

system. In a continuous flow system, the liquid CO2 and the product are pumped

through the system and are mixed before entering the high pressure pump, which









allows adjustment of the pressure to the desired processing levels. The processing

temperature is controlled in holding coils. Residence time is adjusted by setting the flow

rate of the product through the coils. At the end of the process, an expansion valve is

used to release C02 from the mixture and the treated product is collected (Damar and

Balaban 2006).

Food Applications and Effect on Quality

DP-C02 has been applied mostly to fruit juices and beverages. The application of

DP-C02 treatment to some fruits can cause tissue damage even at low pressures. DP-

C02 treatment of orange juice showed that this non-thermal technology is effective in

reducing microbial load, enzyme inactivation, cloud stabilization and maintenance of the

quality of the product.

Kincal and others (2005) used a continuous high pressure carbon dioxide (HPCD)

system for microbial reduction in orange juice. They tested the effectiveness of the

equipment in reducing the natural microflora of pulp-free Valencia orange juice at

different pressures (38, 72, and 107 MPa), for a residence time of 10 min and C02/juice

ratios between 0.1 and 1.0. To test the effectiveness on spoiled juice, juice with a load

of 2 x 106 colony forming units per mL was prepared and subjected to sub- (250C) and

super-critical C02 treatments (34.5C) at pressures of 38, 72 or 107 MPa, residence

time of 10 min and C02/juice ratio of 1.0. To study the capacity of the equipment in

pathogen inactivation, untreated sterilized juice was inoculated with Salmonella

typhymurium, E. coli 0157:H7 or Listeria monocytogenes and the system was run at

pressures of 21, 38 or 107 MPa and a residence time of 10 min. A storage study was

conducted at 1.7C with juice processed at 107 MPa, C02/juice ratio of 1.0 and

residence time of 10 min. When the variables pressure and residence time were









compared, residence time had the greater influence on microbial reduction. Continuous

high pressure C02 processing was capable of destroying the natural microflora in

orange juice and residence time and pressure had little influence on the destruction of

low levels of microorganisms. The C02/juice ratio and temperature were shown not to

be the driving forces on microbial load reduction in this system. They proved that the

system was able to achieve a 5-log reduction of the natural flora in spoiled juice (38, 72,

and 107 MPa at 25 and 34.5C, C02/juice ratio of 1.0 and residence time of 10 min),

and 5-log decrease of pathogenic Salmonella typhimurium (all three pressures and

residence time of 10 min), Escherichia coli 0157:H7 (pressure-related decrease), and

Listeria monocytogenes (destroyed after treatment with all pressures). During the

refrigerated storage study, they observed an increase in the bacterial number possibly

because of an injury/repair mechanism of some of the microorganisms or due to post-

contamination (Kincal and others 2005).

Kincal and others (2006) published work on HPCD system for cloud and quality

retention in orange juice. They used a continuous HPCD to treat pulp-free Valencia

orange juice at pressures of 38, 72, and 107 MPa, and C02/juice (w/w) ratios from 0.10

to 1 with a constant residence time of 10 min. For the storage study, they treated

orange juice at 107 MPa for 10 min, a C02/juice ratio of 1.09 and stored it at 1.70C

(Kincal and others 2006). The highest PE inactivation (56.0%) occurred at 72 MPa and

a residence time of 10 min followed by inactivation of 53% achieved at 107 MPa and a

residence time of 8.6 min. When the treatment was conducted at three different

pressures and a residence time of 10 min, the highest inactivation (46.3%) was

obtained when the pressure was 107 MPa and no heat was applied. These results









demonstrate that pressure affects PE inactivation. HPCD preserved and enhanced the

cloud of the treated orange juice in some cases. The greatest increase was found in

samples treated at 38 MPa and 1.18% of C02, and it was shown that pressure has little

effect on cloud. Treatment with continuous HPDP did not have any effect on pH and

OBrix, but titratable acidity increased slightly after treatments. During the storage study,

PE activity decreased with storage time and cloud remained 4 times higher than the

control during storage. Juice color did not change drastically during storage (Kincal and

others 2006). Sensory evaluations of DPCD-treated and untreated OJ were not

significantly different after 2 weeks of refrigerated storage at 1.70C (Balaban and others

2008).

Lim and others (2006) processed mandarin juice with DP-C02. The process

variables were temperature (25, 35 and 45 C), pressure (13.8, 27.6 and 41.4 MPa),

residence time (5, 7 and 9 minutes) and %C02 (2, 7 and12). They found that

temperature and %C02 had a significant effect in log reduction of total aerobic plate

count while pressure and residence time were not significant. The HPCD processing

reduced the total aerobic count of natural flora in mandarin juice by about three orders

of magnitude. Maximum log reduction (3.47) was observed at the conditions of 35C,

41.4 MPa, 9 min and 7 %C02. PE inactivation ranged from 6.1 to 50.7% with a

maximum inactivation achieved at 45 C, 41.4 MPa, 7 min and 7% CO2. Cloud was not

only retained but enhanced. The highest cloud increase was 38.4% at 45 C, 27.6 MPa,

7 min, and 2% CO2. Lightness and yellowness increased and redness decreased after

treatment. pH and oBrix did not change after treatment while titratable acidity of treated

samples was higher than the untreated juice (Lim and others 2006).









Beer quality after pasteurization with DP-C02 was studied. A maximum log

reduction in yeast population of 7.38 logs was predicted at 26.5 MPa, 21 C, 9.6 %C02,

and residence time of 4.77 min. The maximum haze reduction from146 nephelometric

turbidity units (NTU) to 95.3 NTU was observed at a processing pressure of 27.6 MPa.

Aroma and flavor of beer processed under the same conditions were not significantly

different when compared to a fresh beer sample in a difference from control test. Foam

capacity and stability of beer were minimally affected by the process; however these

changes were unnoticed by consumers (Dagan and Balaban 2006).

The effects of DP-C02 on microbial, physical, chemical and sensorial quality of a

coconut water beverage were evaluated by Damar and others (2009). Processing

variables were: pressure (13.8, 24.1, and 34.5 MPa), temperature (20, 30, and 40 oC)

and %C02 (7, 10, 13 g CO2/100 g beverage). A constant residence time of 6 min was

used during the experiment. DPCD-treated (at 34.5 MPa, 25 oC, 13% C02, 6 min), heat

pasteurized (74 oC, 15 s) and untreated coconut beverages were evaluated during 9

wks of storage at 4C. Results showed that pressure was not significant in microbial

reduction whereas temperature and %C02 levels were significant. Total aerobic

bacteria of DPCD and heat-treated samples decreased while that of untreated samples

increased to >105 CFU/mL after 9 wks. DP-C02 increased titratable acidity but did not

change pH and oBrix. Likeability of DPCD-treated coconut water was similar to the

untreated one (Damar and others 2009).

The effect of DP-C02 on physical and quality attributes of red grapefruit juice was

studied by Ferrentino and others (2010). A central composite design was used with

pressure (13.8, 24.1, and 34.5 MPa) and residence time (5, 7, and 9 min) as variables.









A constant temperature of 400C and C02 level of 5.7% was used to treat the juice. A

storage study was performed with the fresh juice. Brix, pH, titratable acidity (TA),

pectinesterase (PE) inactivation, cloud, color, hue and color density, total phenolics,

antioxidant capacity, and ascorbic acid were measured after treatment and during 6 wk

storage at 4 C. Five log reduction for yeasts and molds and total aerobic

microorganisms occurred at 34.5 MPa and 7 min of treatment. During storage, the

DPCD-treated juice showed no growth of total aerobic microorganisms, and yeasts and

molds. Cloud increased by 91% while PE inactivation was 69.17%. No significant (a =

0.05) differences were detected between treated and untreated samples for Brix, pH,

and TA. Treated juice had higher lightness and redness and lower yellowness. Slight

differences were detected for the ascorbic acid content and the antioxidant capacity

(Ferrentino and others 2009).

Objectives of Study

No research has been conducted on the effect of dense-phase carbon dioxide on

guava puree processing as a preservation method. The hypothesis of this research is

that DP-C02 treatment, a non-thermal process, can replace pasteurization to preserve

guava puree while maintaining the fresh like attributes of flavor, and physico-chemical

and phytochemical properties. This research will focus on four main objectives:

1. Determine the optimal time and concentration of a commercially available

enzyme to reduce the viscosity of guava puree at low temperature (300C) and

possible increase the juice yield.

2. Determine the optimal conditions (pressure, % carbon dioxide and residence

time) required to achieve a 5 log reductions in microorganisms for guava puree

processing using DP-C02.









3. Determine the changes in phytochemicals and quality attributes of thermal and

non-thermal processed guava puree during storage.

4. Conduct a comparison of sensory attributes and aroma compound changes on

thermal and non-thermal processed of guava puree.









CHAPTER 3
EFFECT OF ENZYME TREATMENT ON PHYSICOCHEMICAL AND
PHYTOCHEMICAL PROPERTIES OF GUAVA PUREE

Abstract

Fruit juices are an important part of our diet. Guava (Psidium guajava L.) fruit has

not been fully utilized as a source of processed juice due to a number of quality

limitations that occur during processing, one of which is the presence of excessive

amounts of suspended solids. Enzyme treatment is one method of enhancing the

degradation and removal of suspended solids. During enzyme treatment, increasing the

temperature may produce a well clarified juice but may also modify the phytochemical

composition and ascorbic acid content due to oxidative reactions. The objective was to

determine the optimal time and concentration of a commercially available enzyme

(Bioguavase) for treatment of guava juice for obtaining a clarified product at 30 C

without affecting the phytochemical properties. Three treatment times (12, 24 and 36

hours) and three enzyme concentrations (400, 600 and 800 ppm) were tested.

Following treatment, juice was clarified by centrifugation and analyzed for vitamin C

content (2,6-dichloroindophenol titration method), antioxidant capacity (ORAC), total

soluble phenolics (Folin assay), turbidity and color (L* a* b* values). After 12 h reaction

time, the 600 ppm treatment produced the clearest juice. Juice yield was not increased

by extending the reaction time beyond 12 h. In additional experiments, four treatment

times (3, 6, 9 and 12 hours) were compared under the same conditions described

above. All enzyme treatments reduced the antioxidant capacity (between 8 and 22%)

and increased the total soluble phenolic content (between 8 and 15%) of the juice.

Treatment of guava juice with 600 ppm Bioguavase for 3 h is most suitable for obtaining

clarified juice.









Introduction

Guava (Psidium guajava, L) is an exotic tropical fruit rich in antioxidants and

vitamin C. A member of the Myrtaceae family, it is common to all warm tropical areas of

America and can be found in the West Indies, Bahamas, Bermuda and southern

Florida. Morphologically, the fruit may be round, ovoid, or pear-shaped with thin, light-

yellow skin, frequently blushed with pink. The flesh can be white, yellowish, light- or

dark-pink or near-red, juicy, acid or sweet and flavorful. The guava can be eaten raw or

processed to obtain other products. In Hawaii, the guava is boiled in slices to produce a

guava juice. In Brazil, Mexico and the Dominican Republic, fruit is processed to obtain a

puree. In South Africa, the fruit is trimmed, minced and mixed with a natural fungal

enzyme to obtain a clear guava juice without exposure to heat that degrades ascorbic

acid and other constituents. Guava juice and nectar are among the numerous popular

canned or bottled fruit beverages of the Caribbean area (Morton, 1987). Guava puree is

normally processed by heat pasteurization to extend the shelf life of the product for up

to one year, but the fresh taste is modified. The use of non-thermal pasteurization can

minimize the development of undesirable characteristics by reducing the chemical

changes that occur during heat processing.

Dense phase CO2 (DP-CO2) technology is a non-thermal method emerging as an

alternative to traditional thermal pasteurization. It is a non-thermal pasteurization

method that does not use heat to destroy microorganisms and enzymes (Damar and

Balaban, 2006) and is a promising technology to preserve phytochemicals and to retain

the fresh-like physical, nutritional and sensory qualities of the final product compared to

traditional heat pasteurization. However, the consistency of guava puree has a

significant effect on its flow at low temperatures. One way of improving flow is by









reducing the viscosity of the puree by enzymatic treatment. This enhances the

degradation and removal of suspended solids and decreases the viscosity. In the food

industry, a combination of enzymes including pectinesterase, arabanase, hemicellulase,

tannase and cellulase are used to degrade the mesocarp of guava which contains 90%

of the total cell wall material as pulp (Kashyap and others 2001).

Bioguavase is a commercially available enzyme preparation that contains a

variety of carbohydrase enzymes derived from Aspergillus niger. It causes rapid

viscosity reduction of guava puree or fresh guava fruit through pectin hydrolysis, with a

significant increase in juice yield. Hydrolysis of pectin produces carboxylic acids and

galacturonic acid which may lead to a pH decrease. During enzyme treatment,

increasing the temperature may produce a well-clarified juice but may also modify the

phytochemical composition and reduce ascorbic acid content due to oxidation.

The objectives of this study were: to apply an enzyme (Bioguavase) treatment at

temperature bellow 35 C to obtain a product with a consistency suitable for dense

phase carbon dioxide processing with an increase in yield, and to optimize (without

affecting the phytochemical properties) the time and concentration of this commercial

enzyme preparation for treatment of guava puree in obtaining a clarified product at 30

oC.

Materials and Methods

Preliminary Study

Selection of an enzyme and a temperature to clarify the guava puree

Kleryzyme 150 (DSM, Cedex, France), Rapidase TF (ADM, Decatur, IL, U.S.A.),

Cellubrix (Novozymes, Denmark), Pectinex Ultra SP-L (Novozymes, Denmark),

Crystalzyme 200XL (Valley Research, South Bend, IN, U.S.A.), Bioguavase (BioSun,









Tampa, FL, U.S.A.), Biocranase Super (BioSun, Tampa, FL, U.S.A.) and Biocellulase

FG Concentrate (BioSun, Tampa, FL, U.S.A.) were obtained from the manufacturers.

Five different enzyme treatments (Table 3-1) and three different temperatures were

used: 12.4 C (55 F), 21.4 C (70 F) and 30 C (86 F). One hundred grams of puree

were treated with the enzyme preparation at each temperature. Sodium azide (0.002%)

was added to each sample. Each sample was incubated at one of the 3 selected

temperatures for a period of 24 h + 1 h. Samples were observed and mixed periodically.

After 24 h, the purees were removed from incubation, centrifuged in a Sorvall RC-5B

Refrigerated Superspeed Centrifuge (Dupont Instruments, Newton CT, U.S.A.) at

10,410 x g (9,500 rpm), 8 min at 4 oC. The percent yields of juice were calculated using

the following equation:

% yield = [(initial weight weight after centrifugation)/initial weight] 100

Effects of enzyme treatment at 30 oC on phytochemical levels in guava (Psidium
guajava) puree

Sample preparation

Untreated guava puree was obtained from Hawaii (Kai Guava, Kilauea

Agronomics, Kilauea, HI, U.S.A.) through a distributor in Florida, and transported frozen

to the Food Science and Human Nutrition Department in Gainesville, Florida. The puree

was thawed, divided into 2 L bottles and immediately frozen at -20 oC. Before the

clarification process, the puree was thawed overnight at 60C. Bioguavase (600 ppm)

was used to treat the puree for 24 h at 30 oC. Following treatment, the purees were

removed from incubation and placed in ice slush immediately to stop the enzyme

reaction.









Analysis

Analysis of the puree before it was exposed to the temperature (sample before

temperature treatment = no-heat, no enzyme or NHNE) was performed. A control

sample of puree exposed to 30 C for the same amount of time without enzyme (heated

no-enzyme) was included. The purees were assayed for percent juice yield, vitamin C,

antioxidant capacity (ORAC), total phenolics, color and total soluble solids (TSS).

Analytical procedures are described later.

Enzyme Treatment Optimization at Low Temperatures to Produce a Clarified
Guava (Psidium guajava) Juice

Sample preparation and enzyme treatment

Four hundred grams of puree were weighed and placed in a beaker. The amount of

enzyme added to each beaker were as follows: 160 pL Bioguavase enzyme (400 ppm),

240 pL Bioguavase enzyme (600 ppm) and 320 pL Bioguavase enzyme (800 ppm).

These samples were divided into 4 beakers (each containing 100 g) and placed at 30

C in an incubator. Every 12 h, one beaker from each enzyme concentration was

removed and placed in an ice bath to stop the enzyme reaction. Samples were kept on

ice until analyzed. The experiment was performed in duplicate. The previous procedure

was conducted again, but sampling time was reduced to every 3 h for up to 12 h. After

enzyme treatment, the purees were assayed for percent juice yield, vitamin C,

antioxidant capacity, total phenolics, turbidity, pH, total soluble solids (TSS, OBrix) and

color.

Physicochemical analysis

Percent iuice yield: Juice yield was conducted using the following as previously

stated.









Vitamin C: The 2,6-dichloroindophenol dye was obtained from Sigma-Aldrich (St.

Louis, MO, U.S.A.) and the acetic acid and m-phosphoric acid chemicals were obtained

from Fisher Scientific (FL, U.S.A.). Vitamin C of the centrifuged samples was assayed

by titration using the Official Method published by the AOAC. Two ml aliquots were used

for the titration and the vitamin C content in each sample was calculated based on the

stoichiometry of the titration (AOAC method 967.21, 1990). Vitamin C content was

expressed as mg of vitamin C per 100 g sample.

Antioxidant Capacity: AAPH (2,2'-azobis(2-methylpropionamidine

dihydrochloride)), fluorescein (free acid) and Trolox (6-hydroxy-2,5,7,8-

tetramethylchroman-2carboxilic acid) were obtained from Sigma-Aldrich (St. Louis, MO,

U.S.A.). Antioxidant capacity of hydrophilic compounds in the supernatants of

centrifuged samples was determined by the oxygen radical absorbance capacity

(ORAC) assay (Huang and others 2002). Antioxidant capacity was calculated by

integrating the area under the fluorescence decay curve in the presence of guava

phytochemicals and calibrated with a standard curve of Trolox using a SpectraMax

Gemini XPS microplate sprectrofluorometer (Molecular Devices, Sunnyvale CA, U.S.A.)

and SoftMax Pro 5.2 software (Molecular Devices, Sunnyvale, CA, U.S.A.). Results

were expressed as Trolox equivalents (TE) per mL (pmol of TE/mL). A dilution of 100 X

was used to obtain the correct area from the treated sample.

Total Phenolic Compounds: Total phenolic compounds of the centrifuged samples

were analyzed using the Folin-Ciocalteu metal reduction assay (Talcott and others

2000), using gallic acid as standard. For analysis, 100 pL of sample were used.

Absorbances at 765 nm were taken using a Spectra Max 190 spectrophotometer









(Molecular devices, Sunnyville, CA, U.S.A.). Gallic acid and Folin-Ciocalteu's reagent

were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). All samples were diluted

by 10 X in order to obtain the absorbance reading within the standard curve.

Turbidity (clarity): Sample turbidity was measured using the percent transmission

mode at 650 nm in a Beckman UV-VIS scanning spectrophotometer (Model # DU-620,

Beckman Coutler, Brea, CA, USA). Clarity of samples was determined using 1.5 mL of

sample in 1 cm plastic disposable cuvettes against water.

pH: The pH was measured using an Orion expandable ion analyzer EA 920 pH

meter (Orion Research; Boston, MA, U.S.A.) equipped with an Accumate glass

electrode (Fisher Scientific, U.S.A.).

Total soluble solids (TSS): Brix of the centrifuged juice was measured using an

electronic ABBE Mark II refractometer at room temperature (Leica Inc.; Buffalo, NY,

U.S.A.).

Color Analysis: The color of the centrifuged juice was measured using a Gardner

colorimeter (BYK-Gardner USA; Columbia, MD, U.S.A.) and expressed as L*, a* and b*.

Twenty mL of sample were used to perform the analysis and the measurements were

done in the reflectance mode. Calibration of the equipment previous to color

measurement was conducted using black and white tiles.

Statistical Analysis

SAS 9.0 software (SAS Institute Inc., Cary, N.C., U.S.A.) was utilized for all

statistical analyses. For the selection of an enzyme treatment and temperature, analysis

of variance for factorial design was employed. To study the effects of enzyme treatment

at low temperature on phytochemical levels in guava puree, one way ANOVA was

employed with mean separation by Tukey's standardized test. For both experiments









conducted on the enzyme optimization treatment for guava puree, a repeated

measurement design was used. Chemical measurement data for the treated and

untreated samples were analyzed by a Tukey's standardized range (HSD) test at a

significant level of a = 0.05. Examples of the SAS program and output are presented in

Tables A-1 and A-2.

Results and Discussion

Selection of Enzyme and Temperature

In the selection of an enzyme and a temperature, a control sample of puree

without enzyme was subjected to the same incubation temperature and centrifugation

step as the treatment. The percent yield for the control puree was 66.14 + 0.03%,

irrespective of the incubation temperature. The enzyme treatment with the highest yield

(83.72 + 1.02%) was treatment 5, or 600 ppm of Bioguavase enzyme (Figure 3-1).

As expected, all enzyme treatments resulted in higher yield at 30 C (86 OF). There

was no significant difference (a = 0.05) between the type of enzyme and the percent

yield, but there was significant difference (a = 0.05) between the temperature and

percent yield. All enzymes have a temperature range where they work at optimum

conditions. In our case, all enzymes used during the first part of this research were

active at temperatures between 10 and 55 C. Enzyme activity was dependent upon

temperature. Enzyme activity increased when temperature increased to a certain point

and then denaturation occurred. The selection of the temperature and enzyme for

further experiments (Bioguavase) was based solely on the higher yield it produced.









Enzyme Treatment Optimization at Low Temperatures to Produce a Clarified
Guava (Psidium guajava) Juice without Affecting its Phytochemical
Composition

The yield for the non-heated, no enzyme (NHNE) puree was 72.18%. The heated

no-enzyme or control (30 C for 24 h) showed a yield of 72.92% and the enzyme treated

(ET) puree (heated with enzyme) had a yield of 82.93%. Thus the increase in the yield

of the enzyme treated sample compared to controls was not due to the treatment

temperature (Table 3-2). There were significant differences (a = 0.05) between the

NHNE puree and ET puree and between the control and ET puree (based on simple

ANOVA analysis). As expected, the use of enzymes to clarify juices increases the yield

of the final product. In the food industry, enzymes are used to increase yield and clarify

juices. An important component of commercial pectinases is pectin methyl esterase,

which is specifically able to convert colloidal pectin to noncolloidal pectic acid. This

results in the sedimentation of cloud-forming particles. In apple juice production,

pectinases are added with the ultimate objective of producing a high yield clear juice

(Christen and Smith, 2000). In sparkling clear juice production, enzymes are added to

increase the juice yield during pressing and straining of the juice, and to remove

suspended matter to give sparkling clear juices (Kashyap and others 2001).

Table 3-2 presents the results obtained for the antioxidant capacity analysis.

Before the temperature treatment, the antioxidant capacity was 13.27 pmol TE/mL. The

antioxidant capacity for the control and the ET purees were 13.78 and 12.63 pmol

TE/mL, respectively. There were no significant differences (a = 0.05) between the three

samples indicating that neither the enzyme nor temperature had an effect in the

antioxidant capacity of the product. These results are comparable to a previous study

(Fender, 2005) were 10.5 pmol TE/mL was obtained in guava nectar.









Total soluble phenolics were measured by the Folin-Ciocalteu assay, which

measures the capacity of phytochemical compounds to reduce an oxidized metal ion.

Although the target compounds for this assay were polyphenolics, compounds such as

ascorbic acid, certain soluble proteins, melonoidins, and reducing sugars give a

measurable interference in the assay. The average total soluble phenolics was

expressed in Gallic Acid Equivalent (GAE) (Table 3-2). The total soluble phenolics in the

NHNE sample were 837.46 mg/L GAE. The control and the ET samples had total

soluble phenolics content of 910.21 and 892.96 mg/L GAE, respectively. There was an

increase in total phenolic content due to temperature treatment and enzyme activity.

Comparing the control to the enzyme-treated puree; there was a small decrease in total

phenolic content in the enzyme-treated sample. There were significant differences (a =

0.05) between the NHNE and the control which showed no significant difference to the

ET sample.

Guava has one of the highest concentrations of vitamin C among all fruits,

typically containing between 200-300 mg vitamin C/100 g. In a study conducted on

guava juice processing optimization (Chopda and Barrett, 2001), the authors found an

ascorbic acid content of 149.4 mg ascorbic acid/100g sample in the supernatant. Table

3-2 presents the mg of ascorbic acid/1OOg sample before (NHNE) and after temperature

treatment (enzyme and no enzyme). The ascorbic acid content for the NHNE sample

was 81.57 mg ascorbic acid/100 g sample (Table 3-2). The ascorbic acid content

decreased due to the temperature effect probably due to oxidation reactions. The

control shows an ascorbic acid content of 80.72 and the ET puree had an ascorbic acid

content of 79.81 mg/100 g. Although ascorbic acid content decreased with increasing









temperature, there were no significant difference (a = 0.05) between the three samples.

This indicated that the conditions of enzyme processing did not create an environment

leading to ascorbic acid oxidation or degradation. The ascorbic acid content was lower

than that reported in other literature. In this study, the puree did not contain the guava

skin, which is the part of the fruit with the highest ascorbic acid content. The guava

puree production date is not known so some degradation may have occurred during

processing and storage.

Table 3-3 shows the results obtained for the color analysis of the centrifuged

samples. A decrease in L* (brightness) was observed after the temperature treatment.

Both the control and ET samples presented a decrease in brightness compared to the

NHNE sample. There were significant differences (a = 0.05) between the three

samples. The NHNE puree had the higher L* value, followed by the control. The

redness (a*) of the NHNE sample was higher than the control, and this value increased

during enzyme treatment. Enzyme treatment of guava puree causes the liberation of

some compounds, such as carotenoids, which are bound to complex carbohydrates

(components of the plant cell wall) that are broken down during clarification by the

enzyme (Steven, 1998). There were significant differences (a = 0.05) in redness

between the three samples. The yellowness (b*) of the control sample was lower than

NHNE sample. The enzyme treatment increased yellowness. A color analysis was

performed on the samples without centrifugation (Table 3-3) and the analysis showed

significant differences in the L*, a* and b* value between the three samples. The

centrifugation process reduced the L*, a* and b* for all three samples but redness and

yellowness of the samples were the most affected. Centrifugation removed most of the









insoluble particles, which could have contained the compounds responsible for the pink

color, but some colloidal particles remained causing turbidity in the non-heated sample

and the control juice. The enzyme treated puree produced a clear juice after the

centrifugation process. Centrifugation of the NHNE and control samples produced a

cloudy supernatant while ET puree produced a clear juice supernatantt).

The TSS (oBrix) of the clarified guava puree was between 7 and 9. Centrifugation

decreased the soluble solids content of the samples (Table 3-4). There were significant

differences (a = 0.05) between the oBrix for the three puree samples. Before

temperature treatment, the puree had a oBrix of 7.1, while after the temperature

treatment, the control and enzyme-treated purees had 7.2 and 7.4 oBrix respectively.

After centrifugation, the NHNE sample was 5.7 oBrix while the temperature treated

control and enzyme-treated samples were 6.5 and 6.6 oBrix respectively. There were

significant differences (a = 0.05) between the oBrix for the control and the enzyme-

treated sample. As expected, oBrix increases due to temperature and enzyme

treatment. The increase in total soluble solids or oBrix is due to the breakdown of pectin

and other complex carbohydrates. In the case of pectin breakdown, a release of

galacturonic acid pecticc acid) occurs which is soluble and contributes to the increase in

oBrix measurement. Pectic substances or pectin are high molecular weight

polysaccharides found in plant cell wall middle lamellae. They are composed of

galacturonic acid units, joined by a-1,4 glycosidic linkages. Some of the acid groups

along with the acid units become methylated during the fruit ripening (Murano, 2003).

When pectinases are added to fruit juices the conversion of insoluble pectin to soluble

pectic acid occur (Christen and Smith, 2000).









Enzyme Treatment Optimization at Low Temperatures to Produce a Clarified
Guava (Psidium guajava) Juice without Affecting it Phytochemical Composition

Results for the physicochemical analysis conducted on guava puree after

treatment with 400, 600 and 800 ppm of Bioguavase enzyme over three reaction times

(12, 24 and 36 h) are shown in Table 3-5. There were significant differences due to the

reaction time, enzyme concentration and their interaction (time x concentration). There

was a significant difference between the control (zero enzyme concentration and

reaction time) and the enzyme treatments. There were no significant differences

between the 3 enzyme concentrations. After 12 h reaction time, the increase in yield

was not significant for any of the 3 enzyme concentrations. Increasing enzyme

concentration will increase the velocity of the reaction. After 12 h of reaction time, the

maximum reaction velocity was achieved that is the reason why there were no further

differences between the 3 enzyme concentrations. In an attempt to reduce the reaction

time and obtain maximum juice yield, another study was performed. In this study,

samples were taken every 3 h for up to 12 h.

There were significant differences in vitamin C content due to the reaction time

and the interaction of time with enzyme concentration. The control had lower vitamin C

when compared to all levels of enzyme (only at 12 h). At 12 h the amount of vitamin C

increased for all levels of enzyme concentration (with no significant differences)

probably because the rate of ascorbic acid degradation was lower than its liberation

from the pulp. After 12 h, the vitamin C content decreased. This indicated that after 12

h, the reaction time created an environment suitable for ascorbic acid oxidation or

degradation. Comparing the ascorbic acid content with that reported in the literature,

this study sample content was lower. Vitamin C content depends on the variety of









guava, harvested season, and processing conditions. Vitamin C in the fresh fruit is

always higher than in the puree.

Antioxidants are important in foods for their potential health benefits and their

radical scavenging abilities. They can prevent oxidation in foods and in the human body.

There were significant differences in antioxidant activity due to the enzyme

concentration, reaction time and the interaction (Table 3-5). Antioxidant capacities for

the control were higher compared to the different levels of enzyme concentration, but

the value decreased with reaction time. At 12 h, the antioxidant capacity decreased with

the amount of enzyme used, but there were no significant differences in the ORAC

value between 400 and 600 ppm. At 24 h, there were no significant differences between

the control, 400 and 600 ppm enzyme concentration. At 36 h, there was no significant

difference between all levels of enzyme concentration.

Results of the Folin-Ciocalteu assay for total phenolics revealed that there were

significant differences due to the reaction time, enzyme concentration and the

interaction. At 12 h, the total phenolics in the control were not significantly different from

the 800 ppm enzyme and there were no significant differences between the 3 levels of

enzyme concentration. At 24 and 36 h, there were significant differences between the

control and the three enzyme concentrations (400, 600 and 800 ppm). There was an

increase in total phenolics content due to the enzyme concentration, and a larger

amount was observed at 36 h reaction time. The increase in total phenolics is due to

the release of phenolic compounds from the pulp. The pulp of the puree contains

phenolic compounds which were released during the enzymatic reaction. (Jimenez-

Escrig and others 2001) tested the fiber from the pulp and peel of guava from Caracas.









They found that both fibers were potent sources of radical-scavenging compounds,

presumably from the high content of cell-wall bound polyphenolics reported for each

fiber. They found for pulp fiber, a total extractable phenol content of 26.2 g GAE/ kg dry

matter.

Turbidity is an important factor in clarified juice. The results showed that the use of

enzyme is adequate to obtain a clarified juice. There were significant differences

regarding turbidity between the reaction time, enzyme concentration and the interaction.

The control had the lowest percent transmittance (lower clarity). The juice was cloudy

and showed an increase in turbidity at 36 h. This was significantly different for the 400,

600 and 800 ppm enzyme concentration at all reaction times. At 12 and 24 h, there

were no significant differences between 400 and 800 ppm. At 36 h, there were no

differences between 400, 600 and 800 ppm. As expected, enzyme treatment produces

a clear juice due to the conversion of colloidal pectin to noncolloidal pectic acid.

Colloidal pectin is responsible for the cloudy appearance of fruit juices and production of

acid results in the sedimentation of the cloud-forming particles (Christen and Smith,

2000).

There were significant differences in pH due to the enzyme concentration,

reaction time and their interaction. There were significant differences between the

control and the three levels of enzyme concentration. During the first 12 h of reaction,

the pH decreased with increasing enzyme concentration (Table 3-5). After 24 h, the pH

remained almost constant irrespective of enzyme concentration. Enzyme treatment

decreased the pH of the product (the product is more acid) due to the release of

galacturonic acid to the medium.









The oBrix of the clarified guava juice was between 6.4 and 6.7. There were

significant differences due to enzyme concentration, reaction time and their interaction

(Table3-5). The control had lower TSS than the 3 enzyme treatments. TSS for the

control stayed the same over time, but the oBrix increased with enzyme concentration.

An enzyme concentration of 400 ppm did not cause a further increase in oBrix after 12 h

reaction time. Six hundred ppm of enzyme concentration caused a further increase in

oBrix at 36 h, this value being the highest. The increase in oBrix is related to a decrease

in pH and decrease in turbidity which are related to pectin breakdown.

Color analyses are summarized in Table 3-6. L* values decreased with enzyme

concentration; there were significant differences between the control and the 3 levels of

enzyme concentration at the initial reaction time. There were no differences between

400 and 600 ppm during the entire reaction time. The a* value increased after the

clarification process during the first 12 h, but then decreased. There were no differences

between the 3 levels of enzyme concentration during the first 24 h of reaction time. The

b* values increased after the clarification process, and at 600 and 800 ppm, this value

increased with reaction time. At 600 ppm, there was no significant increase during the

first 24 h of the reaction but the value increased further after 24 h. At 12 and 24 h, there

were significant differences at all levels of enzyme concentration, but at 36 h, the b*

values for the 3 enzyme concentrations were not significantly different.

The food industry uses enzymes to increase juice yield. Figure 3-2 shows the

percent yield obtained for the control and each of the enzyme concentrations (400, 600

and 800 ppm) over reduced reaction times (3, 6, 9 and 12 h). Enzyme treatment

significantly increased juice yield. Four hundred ppm slowly increased juice yield but









after 9 h there were no significant differences between 400 and 600 ppm enzyme

concentration. This slow increase in yield may be due to the composition of the enzyme.

The enzyme system used a variety of carbohydrases. Different enzymes function

together to lower the viscosity of the puree while increasing the juice yield. Imungi and

others (1980) clarified guava puree using 400 ppm of Pectinex superconcentrate at a

temperature between 40 and 50 C. They found that the yield increases with treatment

time and 90 min was adequate to achieve maximum yield without decreasing vitamin C

content (Imungi and others 1980). Sandhu and Bhatia (1985) also observed an

increase in juice yield with enzyme treatment due to a considerable reduction of pectin.

Brasil and others (1995) observed an increase in juice yield with increasing treatment

time after treating guava puree with 600 ppm Clarex-L super-concentrate at 45 C for up

to 150 min.

There were no significant differences in the vitamin C content between the control

and the puree clarified with 600 ppm enzyme concentration (Figure 3-3). Enzyme

concentration (400 and 800 ppm) significantly decreased the vitamin C content during

the first 3 h reaction time. After 3 h, there was an increase in vitamin C content for 400

ppm enzyme concentration, and after 6 h reaction time, there were no significant

differences between the vitamin C content for the control, 400 and 600 ppm samples.

An enzyme concentration of 800 ppm decreased the amount of vitamin C in the clarified

juice probably because the rate of ascorbic acid degradation was greater than its

liberation from the pulp. Opposite results were found by Brasil and others (1985) after

treating guava puree with pectic enzyme to obtain a cloudy juice. The reason for this

difference may be explained by the type of puree used for their research. They prepared









the puree by mashing the fruits using a fruit mill. The enzyme treatment increased the

vitamin C content of the cloudy juice due to its liberation, especially from the peel of the

fruit which is known to have more vitamin C than the flesh and center of the fruit (Brasil

and others 1995).

Results of the antioxidant capacity analysis are shown in Figure 3-4. Antioxidant

capacity for the control was high compared to enzyme treatments. The antioxidant

capacity decreased slightly over time regardless of enzyme concentration. The use of

enzyme treatment to produce a clarified juice decreased the antioxidant capacity of the

guava product. Some of the compounds that have antioxidant capacity are bound to

complex carbohydrates and removed after the enzyme treatment followed by

centrifugation. Reduction of antioxidant capacity is related to a decrease in vitamin C

(which has antioxidant capacity).

Figure 3-5 shows the average total soluble phenolic compounds expressed in

Gallic Acid Equivalent (GAE). There were significant differences due to the reaction

time, enzyme concentration and their interaction. At 3 h, the total phenolic levels in the

control were not significantly different from the 400 ppm and the amount of total

phenolic compounds at 600 and 800 ppm were lower compared to the control. The total

phenolics content decreased over time after 3 h reaction time. An enzyme concentration

of 400 ppm increased total phenolics during the first 3 h reaction time, there was a slight

reduction in total phenolics content up to 6 h reaction time, and after 6 h, there was no

significant change. There was a significant increase in total phenolics after 3 h reaction

time and 600 ppm enzyme concentration. The content stayed unchanged up to 9 h

reaction time, after which the content significantly dropped to a level similar to the









enzymatic treatment of 400 ppm. Imungi and others (1980) found an increase in total

phenolics after enzyme treatment of guava puree, but with the clarification process

(filtration) used during the research, a significant decrease in this content was observed.

The results obtained from the turbidity analysis (Figure 3-6) showed that the use of

enzyme resulted in obtaining a clarified juice. There were significant differences

between the reaction time, enzyme concentration and their interaction. The control had

the lowest percent transmittance and the juice remained cloudy.

From the results presented in Figure 3-7, the pH decreased with enzyme

concentration. After 3 h, the pH stayed almost constant irrespective of enzyme

concentration. The control showed an increase (less acid) in pH during the first 3 h, and

after this, the pH was constant. The enzyme treatment decreased the pH of the juice

due to the release of galacturonic acid from pectin hydrolysis. A decrease in pH with

enzyme treatment of guava puree was observed in the literature (Imungi and others

1980). Chopda and Barret (2001) observed a decrease in pH as enzyme concentration

and incubation time increased at an incubation temperature of 50 oC.

Total soluble solids were measured and expressed as oBrix. The results obtained

showed (Figure 3-8) that the control stayed the same over time, but the oBrix increased

with enzyme concentration. An enzyme concentration of 600 ppm did not cause a

further increase in oBrix after 3 h reaction time, producing the clarified juice with the

highest soluble solid content. After 6 h, there were no significant differences in oBrix

between 600 and 800 ppm. After 9 h reaction time, there were no significant

differences in total soluble solids between the control and 400 ppm enzyme

concentration. The reason for the increase in TSS may be explained by the release of









acid during pectin breakdown. An increase in TSS was observed by Thakur and Das

Gupta (2006) after extracting beetroot juice from beetroots using Pectinex Ultra SPL

0.15% (1500 ppm) at 45 C for 2.5 h. Chopda and Barret (2001) observed an increase

in oBrix for guava juice as enzyme concentration and incubation time increased at an

incubation temperature of 50 C (Chopda and Barrett 2001).

As observed in Figure 3-9, L* values decreased with enzyme concentration, and

there were no significant differences between the control and the 3 levels of enzyme

concentration at the initial reaction time (0 h). There were no differences between 400

and 600 ppm at 3 and 9 h reaction time. Enzyme treatments reduced L* value of the

samples. Same results were observed by Hodgson and others (1990) after treating

guava puree with Pectinex Ultra Sp-L at a concentration of 0.2% (2000 ppm) for 2 h 50

C or 16 h at 20 C. In addition, a decrease in a* value and a slight increases in b* was

observed (Hodgson and others 1990).

Conclusions

After 12 h of reaction time, there was no further increase in yield at the 3

concentrations of enzyme used. At an enzyme concentration of 400 and 800 ppm, the

ascorbic acid content decreased by 20% of its initial content during the first 3 h of

incubation time. The enzyme treatment decreased (between 0.55 to 11% of the initial

concentration) the antioxidant capacity of the samples and the antioxidant capacity was

lower regardless of the enzyme concentration. After 3 h reaction time, 600 and 800 ppm

enzyme caused a slight decrease in total phenolic compounds, between 3 and 4% of

the initial value respectively. After 3 h, 600 ppm enzyme increased this value by 17% of

the initial content. The turbidity decreased with enzyme activity, and at 3 h, 600 ppm

produced the clearest juice followed by 400 ppm. The enzyme activity affected the pH









and TSS content: the pH decreased (between 0.5 and 1.6% of the initial pH) while TSS

content increased (between 4 and 10% of the initial value). As expected, the color was

affected by the enzyme concentration since the enzyme produced a clarified juice.

Based on these results, three hours of reaction time and 600 ppm of enzyme

concentration are adequate to produce a clarified juice without affecting the nutritional

value of the juice. These conditions are suitable for producing a clarified juice by

centrifugation after enzyme treatment of the puree. The limitation of this procedure

resides on using a filtering step to obtain a juice instead of centrifugation. After filtering

the enzyme treated puree, the color of the juice was much lighter than the original

puree. Guava press cake retained all the pink color and may be used to increase the

mouth feel in formulated juices. Other uses for the guava cake, such as possible source

of fiber and natural colorant should be studied.



Table 3-1. Enzyme and concentration used during enzymatic treatment
Enzyme Treatment Enzyme or Mixture of Concentration of
code Enzyme enzyme used
1 Klerzyme 150 300 ppm
Rapidase TF 600 ppm
2 Cellubrix L 450 ppm
Pectinex Ultra SP-L 750 ppm
3 Crystalzyme 200XL 600 ppm
4 BioGuavase 600 ppm
5 Biocranase Super 600 ppm
Biocellulase FG Con. 600 ppm


















*300 ppm Klerzyme + 600 ppm Rapidase TF
0450 ppm Cellubrix L + 750 Pectinasx Ultra SP-L
0600 ppm Cristalzyme 200 XL
*600 ppm Bioguavase
*600 ppm Biocranase Super + 600 ppm Biocellulase FG
E Control


12.4


21.4
Temperature o C


Figure 3-1. Percent yield for each enzyme treatment at three different temperatures
(Treatment time: 24 hrs). Error bars represent n=4


90 -


80 -


70 -


60 -
-rs

T50


40 -


30 -


20 -


10 -


0-


I


.









Table 3-2. Percent yield, ORAC value, total soluble phenolics and ascorbic acid content
of guava puree
Total
ORAC Soluble
(pmol Phenolics Ascorbic Acid
Sample % Yield TE/mL) (GAE) (mg AA/100 g)

Non-heated, no 72.78 + 0.27 13.27 + 1.56 837.46 + 81.57+ 0.00
enzyme (NHNE) ___37.02
Heated no 72.92 + 0.33 13.78 + 3.62 910.21 + 80.72 + 0.86
enzyme (control) 59.57
Heated + enzyme 82.93 + 0.32 12.63 + 0.98 892.96 + 79.81 + 2.24
(ET) I 56.94


Table 3-3. Color values obtained for three different
clarification


Table


guava puree before and after


3-4. Total soluble solids for guava puree before and after clarification for the three
different treatments


L* value a* value b* value
Clarify Clarify Clarify
Sample Puree Puree Puree Puree Puree Puree
Non-heated,
no enzyme 48.73 + 32.83 + 22.44 + -0.98+ 14.73 + -5.61 +
(NHNE) 0.01 0.32 0.03 0.07 0.01 0.20
Heated no-
enzyme 49.66 + 29.56 + 22.84 + -1.68+ 14.97 + -5.98+
(control) 0.01 0.68 0.15 0.15 0.31 0.19
Heated +
Enzyme 48.23 + 11.33 + 23.25 + -0.77 + 14.50 + -1.95 +
(ET) 0.29 0.93 0.25 0.23 0.19 0.30


Brix
Sample Puree Clarify
Non-heated, no
enzyme (NHNE) 7.1 + 0.0 5.7 + 0.0


Heated no- 7.2 + 0.0 6.5 + 0.0
enzyme (control)
Heated + Enzyme
(ET) 7.4 + 0.0 6.6 + 0.0









Table 3-5. Physicochemical results for enzymatic treatment of guava puree at three different concentrations and three
different reaction times
Enzyme
Physicochemical Concentration
Analysis (ppm) Reaction Time (hours)
0 12 24 36
0 61.27 + 0.14 60.44 + 0.26 65.72 + 0.28 67.80 + 0.40
%Yield 400 61.27 + 0.14 76.44 + 1.44 77.43 + 1.61 77.55 + 1.95
% Y ield ....
600 61.27 + 0.14 79.21 + 0.82 79.99 + 0.84 80.19 + 1.19
800 61.27 + 0.14 79.30 + 1.32 80.58 + 0.46 79.49 + 0.24
0 71.80 + 0.00 71.12 + 0.01 68.52 + 0.00 66.01 + 0.02
Vitamin C 400 71.80 + 0.00 75.56 + 0.01 69.20 + 0.00 66.50 + 0.01
(mg ascorbic acid/
100 g sample) 600 71.80 + 0.00 74.98 + 0.01 68.90 + 0.00 66.01 + 0.02
800 71.80 + 0.00 74.98 + 0.00 70.26 + 0.00 66.21 + 0.01
Antioxidant Capacity 0 11.86 + 0.92 13.30 + 0.69 12.36 + 0.27 11.79 + 1.88
(pMol TELL) 400 11.86 + 0.92 10.91 + 0.32 11.50 + 0.61 10.03 + 0.75

600 11.86 + 0.92 10.28 + 0.82 10.98 + 1.03 9.61 +1.70
800 11.86+0.92 8.79+0.31 9.01 +1.15 10.06+ 1.79
0 889.10 + 6.80 786.81 + 28.63 757.50 + 18.45 834.90 + 18.45
Total Soluble 400 889.10 + 6.80 878.61 + 39.38 869.86 + 25.83 913.44 + 25.83
Phenolics
Compounds (GAE) 600 889.10 + 6.80 874.03 + 54.63 848.89 + 21.51 962.71 +21.51
800 889.10 + 6.80 825.28 + 61.83 823.19 + 59.53 926.67 + 59.53
0 23.33 + 0.00 25.68 + 1.57 24.42 + 0.39 29.36 + 0.41
Turbidity 400 23.33+ 0.00 72.04 + 3.00 70.16+ 5.37 80.95 + 1.07
600 23.33 + 0.00 86.29 + 3.47 80.34 +6.52 82.48 + 2.29
800 23.33 + 0.00 76.39 + 4.66 69.59 + 2.84 85.14 + 4.56









Table 3-5. Continued
Enzyme Reaction Time (hours)
Physicochemical Concentration
Analysis (ppm) 0 12 24 36
0 3.10 + 0.00 3.17 + 0.16 3.15 + 0.01 3.15 + 0.01
pH 400 3.10+0.00 3.11 +0.0 3.09+0.01 3.09+0.02

600 3.10 + 0.00 3.08 + 0.01 3.07 + 0.00 3.08 + 0.01
800 3.10 + 0.00 3.05 + 0.01 3.01 + 0.01 3.05 + 0.01
Tritable Acidity 0 0.91 + 0.01 0.92 + 0.01 0.93 + 0.02 0.92 + 0.02
(mg of citric acid / 400 0.91 + 0.01 0.98 + 0.01 0.97 + 0.02 0.97 + 0.01
100 g)
600 0.91 + 0.01 0.97 + 0.02 0.97 + 0.01 0.98 + 0.01
800 0.91 + 0.01 0.96 + 0.01 0.97 + 0.01 0.98 + 0.01
0 6.40 + 0.00 6.40 + 0.00 6.40 + 0.00 6.40 + 0.00
Total Soluble Solids 400 6.40 + 0.00 6.70 + 0.30 6.70 + 0.20 6.70 + 0.10
Total Soluble Solids ....
600 6.40 + 0.00 6.7 0+ 0.10 6.70 + 0.00 6.80 + 0.10
800 6.4 0+ 0.00 6.6 0+ 0.00 6.70 + 0.00 6.70 + 0.00









Table 3-6. Color results for enzymatic treatment of guava puree at three different concentrations and three different
reaction times
Enzyme Reaction Time (hours)
Concentration
(ppm) 0 12 24 36
0 ppm 38.77 + 0.0 42.36 + 0.96 43.32 + 0.67 37.23 + 0.13
400 ppm 38.77 + 0.0 27.29 + 1.01 26.61 + 0.49 23.17 + 0.78
600 ppm 38.77 + 0.0 24.43 + 0.81 24.86 + 1.12 20.67 + 0.12
L values 800 ppm 38.77 + 0.0 26.61 + 0.5 27.66 + 0.35 22.69 + 0.31
0 ppm -2.63 + 0.04 -1.75 + 0.10 -2.29 + 0.10 -2.92 + 0.07
400 ppm -2.63 + 0.04 -1.58 + 0.09 -2.00 + 0.05 -2.34 + 0.21
600 ppm -2.63 + 0.04 -1.91 +0.10 -2.01 +0.05 -2.37 + 0.07
a values 800 ppm -2.63 + 0.04 -1.84 + 0.07 -2.02 + 0.09 -2.04 + 0.16
0 ppm -3.46 + 0.02 -4.64 + 0.24 -3.69 + 0.35 -3.18 + 0.03
400 ppm -3.46 + 0.02 -0.90 + 0.303 0.54 + 0.25 2.12 + 0.60
600 ppm -3.46 + 0.02 1.05 + 0.48 1.38 + 0.45 1.9 + 0.07
b values 800 ppm -3.46 +0.02 -0.13+0.04 -0.162+0..1 1.98 +0.41










84


82 -
b e ...................... M
80 ----- -
/ X-b .... .f ...' iJ f
78 n "'-.':'". "
.78. .* ....... ... J

v 76 // -- Control
-- .'" ...... 400 ppm
74 74 '" -A a 600 ppm
72 .- -X-800 ppm

70a m

68
0 3 6 9 12
Reaction Time (hours)
Figure 3-2.Percent yield of clarified juice treated at three different enzyme
concentrations and four different reaction times up to 12 h. Error bars
represent standard deviation for n=9. Different letters within each reaction
time represent significant differences at a=0.5.


0.9
9a bb
0.8 b b ... h
L h

cc e g
0.5
< Control
0 0.4
.. ****400 ppm
o 0.3 A- 600 ppm
< 0.2 --K 800 ppm
0.1
0 I I
0 3 6 9 12
Reaction Time (hours)

Figure 3-3.Ascorbic acid content of clarified guava juices after guava puree was treated
at three different enzyme concentrations and four different reaction times up
to 12 h. Error bars represent standard deviation for n=9. Different letters
within each reaction time represent significant differences at a=0.5.











35

230




Sgcccg g
3 15 e
0
102




-A- 600 ppm

0 3 6 9 12









1000 -
900 --- Control
800 a -""- -"ZTcc -- .H


.> 600
uJ 500
5 400 -- -- Control
5 300 ***400 ppm
c- -A-600 ppm
0 200 ---. 800 ppm
100
0 3 6 9 12
Reaction Time (hours)
Figure 3-5. Total soluble phenolics (GAE) of guava juice treated with three different
enzyme concentrations during 12 hours of reaction time at 300C. Error bars
represent standard deviation for n=9. Different letters within each reaction
time represent significant differences at a=0.5.
time represent significant differences at a=0.5.












90

E 0 k
E 70 c .. .- ... m "
( 60 -
50 -


E 30 +-
( 20 --Control
S*...*-400 ppm
10 A -600 ppm
0 J---K. 80Q ppm
0 3 6 9 12
Reaction Time (hours)
Figure 3-6. Turbidity (% transmission at 650 nm) of guava juice treated with three
different enzyme concentrations during 12 hours of reaction time at 300C.
Error bars represent standard deviation for n=9. Different letters within each
reaction time represent significant differences at a=0.5.

3.2
ab d -g k
35 e h

3.1 ... .....- pp
3.1

a 3.05

3
n T Control
S... 400 ppm
2.95 ,- 600 ppm
-X-800 ppm
2.9 ...
0 3 6 9 12
Reaction Time (hours)
Figure 3-7. pH of clarified guava juice treated with three different enzyme
concentrations during 12 hours of reaction time at 300C. Error bars represent
standard deviation for n=9. Different letters within each reaction time
represent significant differences at a=0.5.












b
6.5 hh---- h h.
,f T
a. '" :. .. . .. .
x 6
d g
) 5.5
1)
0)
a 5 -- Control
***,,400 ppm
4.5 -*--600 ppm
--X 800 ppm

4 1
0 3 6 9 12
Reaction Time (hours)
Figure 3-8. Total soluble solids (oBrix) of clarified guava juice treated with three different
enzyme concentrations during 12 hours of reaction time at 300C. Error bars
represent standard deviation for n=9. Different letters within each reaction
time represent significant differences at a=0.5.

35
b h
30 k

25
250- f"".


15 .* m
,, d'T *--.. ....." -_r .. I

d g j
> 10
10 -- Control
S***,,.400 ppm
5 A- 600 ppm
--X 800 ppm
0
0 3 6 9 12
Reaction Time (hrs)

Figure 3-9. L* values of clarified guava juice treated with three different enzyme
concentrations during 12 hours of reaction time at 30 oC. Error bars represent
standard deviation for n=9. Different letters within each reaction time
represent significant differences at a=0.5.









CHAPTER 4
INFLUENCE OF DENSE PHASE CARBON DIOXIDE AND PASTEURIZATION
TREATMENTS ON THE STORAGE QUALITY OF GUAVA PUREE

Abstract

Guava has been identified as a good source of antioxidants and other

phytochemicals, vitamin C and dietary fiber. The traditional method for preserving guava

puree is heat pasteurization, which degrades certain constituents that are beneficial to

human health (such as vitamin C). The objective of this study was to determine the

optimal conditions (pressure, % carbon dioxide, and residence time) for guava puree

processed using Dense Phase Carbon Dioxide (DP-C02), a non-thermal pasteurization

treatment. Stone cells and insoluble solids were removed from unpasteurized guava

puree and CO2 solubility in puree was measured between 6.9 and 31.03 MPa at 350C.

A log reduction > 3.2 was achieved for yeast and mold count (Y&M) and aerobic plate

count (APC) using 34.1 MPa pressure, 8% CO2 and a residence time of 6.9 min. In all

treatments titratable acidity (TA) was significantly higher compared to the fresh

samples. There were significant differences in pectinesterase activity (PEA) and cloud

values for fresh, DP-CO2 and thermally treated samples. DP-CO2 processing can be

used as a non-thermal treatment without deteriorating the quality of the product

compared to heat pasteurization.

Introduction

One of the most common heat treatments used in the food industry to preserve

products is pasteurization, which involves a low order heat treatment (bellow 100 oC) to

destroy vegetative microorganisms that could affect human health. It extends the

products' shelf life but does not kill all microbial floras. Pasteurized products contain

living organisms capable of growth that limits the shelf life of the product.









Guava (Psidium guajava, L) puree is the raw material used to manufacture

products such as juices, nectars, jams and jellies. Puree is prepared by washing the

whole fruit, removing fruit of inferior quality, and feeding high quality fruit into a pulper

which removes seeds and fibrous fragments of skin. A finisher then removes large

aggregates of stone cells and the residual stone cells may be ground by passing the

finished pulp through a mill. The milling operation improves the mouthfeel but decreases

color quality. An alternate method to milling is centrifugation, which improves mouthfeel

and reduces the sediment of the product. Guava puree is normally processed by heat

pasteurization to extend the shelf life of the product. Heat pasteurization (between 80

and 900C for 60 s) inactivates pectinesterase (PE). The shelf life of the puree can be

extended to one year at -180C (0F) but the fresh taste is lost by deteriorative reactions

caused by heat exposure.

Dense Phase Carbon Dioxide (DP-C02) is a non-thermal pasteurization method.

Microorganisms and enzymes are altered by C02 under pressure (below 50 MPa)

without affecting important physical, chemical and sensory qualities. Carbon dioxide is a

non-toxic, nonflammable, inexpensive gas and has a Generally Recognized As Safe

(GRAS) status (Damar and Balaban, 2006).

Yen and Lin (1996) studied the effects of non-thermal pasteurization on guava

puree. They compared the effect of high pressure pasteurization treatment and thermal

pasteurization on the quality and shelf life of guava puree. Puree with a pH of 3.8 and

8.2 oBrix was subjected to either 400 or 600 MPa for 15 min at 25 oC. The heat

pasteurization process was carried out at 88 90 oC for 24 seconds. Samples from both

treatments were stored at 4 oC over a period of 60 days. All treatments were equally









effective in reducing the microbial load (> 2 log reduction) of the puree. During storage,

puree pressurized at 600 MPa showed lower levels of microorganisms when compared

to 400 MPa. Cloud reduction was greater in untreated puree than in pressurized and

heated puree during storage. Color of pressurized guava puree was similar to that of

freshly extracted puree. Results indicated that high pressure treatment of guava puree

at 25 oC for 15 min could maintain good quality up to 40 days of storage at 4 C

Dense Phase Carbon Dioxide (DP-CO2) has been used as a non-thermal

preservation technique (Dagan and Balaban 2006; Damar and others 2009; Ferrentino

and others 2009; Lim and others 2006; Del Pozo-lnsfran and others 2006). It consists of

submitting a mixture of juice and CO2 to pressures under 50 MPa. The amount of CO2

dissolved in the juice increases with increasing pressure. As CO2 solubility increases,

pH of the juice decreases. Once the pressure is released the CO2 is separated from the

juice and the pH returns to its original value.

Ferrentino and others (2009) treated red grapefruit juice with continuous DP-CO2

to inactivate total aerobic microorganisms and yeasts and molds (Y&M). They achieved

a 5 log reduction for Y&M and aerobic plate counts. Cloud increased (91%) while a

partial inactivation of PE (69.17%) was achieved. No significant (a = 0.05) differences

were detected between treated and fresh samples for oBrix, pH, TA, lightness, redness

and yellowness values.

In the present study, a continuous DP-CO2 system was used to pasteurize guava

puree. The independent parameters for the process were: pressure, temperature,

residence time and % CO2. The solubility of CO2 and therefore the exact amount of CO2









used during DP-C02 was determined using an apparatus designed in the Department of

Food Science and Human Nutrition (Calix and others 2008).

The objectives of this study were: (1) to perform CO2 solubility experiments on

guava puree; (2) to optimize DP-C02 process parameters for microbial reduction, and;

(3) to study microbial stability of DP-C02 treated puree compared to fresh and thermally

treated puree during storage.

Materials and Methods

Guava Puree

Frozen unpasteurized red guava puree doctor Rubi cultivar was obtained from the

Goya Company (San Cristobal City, Dominican Republic). The puree was held at -20 C

until the time of processing when it was thawed at 4 oC for one week. Part of the

insoluble solids and stone cells were removed by straining the thawed puree through a

200 pm nylon filter (Cole Palmer, Vernon Hills, IL, U.S.A) (Figure B-1). The removal of

the insoluble solids led to a 60% yield and improved the pumpability of the puree.

Model System

Four liters of a model system were prepared the same day of the study. For this

purpose glucose, fructose, citric acid and ascorbic acid (Fisher Scientific, Fair Lawn, NJ,

U.S.A.) were used. Total soluble solids (TSS) and pH of the model system were 6.8 and

3.78, respectively.

Solubility Measurements

C02 solubility in the guava puree, water and a model system was measured

between 6.9 and 31.03 MPa at 35 oC using an apparatus designed and built at the

University of Florida Food Science and Human Nutrition department (Gainesville, FL) as

previously described by (Ferrentino and others 2009). In this batch system (Figure B-2)









a known volume of the puree was saturated by bubbling C02 through it at the desired

experimental conditions and the dissolved CO2 was measured at atmospheric pressure.

Solubility of CO2 in water and the model system was measured for comparison.

Solubility was expressed as g of CO2/100 g liquid sample. After solubility measurement,

pH, TSS and TA were determined.

Processing Equipment

The DP-C02 system was constructed by APV (Chicago, IL) for Praxair (Chicago,

IL) and provided to the University of Florida (Gainesville, FL) (Figure B-3). The

equipment was capable of continuously treating liquid foods with CO2 at pressures up to

69 MPa. The system consisted of CO2 tanks and a CO2 pump, a product tank and a

product pump, a high pressure pump, a holding coil (79.2 m, 0.635 cm i.d.), a

decompression valve and a vacuum tank (Fig. 4-1). CO2 and product were pumped

through the system and mixed before passing through the high pressure pump, which

increased the pressure to the process levels. The product temperature was brought to

35 oC in the holding coil by heating tape. Residence time was adjusted by setting the

flow rate of the product passing through the holding coil. At the end of the process, an

expansion valve was used to release pressure and separate CO2 from the mixture, and

the juice was collected into sterile bottles as previously described (Damar 2006).

Whenever the treatment parameters were changed, sterile water was passed through

the system until the desired processing conditions were stabilized. The equipment was

cleaned after each use as described by (Damar 2006).

For emulating commercial thermal processing conditions (90 oC for 60 sec), a

MicroThermics electric UHT/HTST (Ultra High Temperature & High Temperature Short

Time) Lab Model 25 (MicroThermics Inc., Raleigh NC, U.S.A.) was used. The









equipment was capable of continuously pasteurizing a variety of products at

temperatures between 76 and 152 C (170 305 OF). The system consisted of two

tubular product heaters, holding tubes, two independent tubular product coolers and a

variable speed positive displacement product pump. Equipment was taken to stable

processing conditions using water. After processing conditions were reached, product

processing was conducted. The product was pumped through the system and pre-

heated (90 + 0.5 oC) before passing through the holding tube. The product temperature

(90 + 0.3 oC) was brought to the desired processing temperature in the holding tube.

Residence time was adjusted by setting the flow rate (400 mL/min) of the product using

the product pump. After the pasteurization process, product exiting the holding tube was

chilled in a two step process using tubular coolers to 3 oC. Cold product was collected in

sterile bottles and placed in a cooler with ice. Processed product was transported to the

laboratory for analysis. The equipment was cleaned before and after each use as

described by the manufacturer, using Sani-T-10 desinfectant/sanitizing solution

(Spartan Chemical Company, Maumes, OH, U.S.A.).

Microbial Inactivation Study

Preliminary investigations were conducted to determine the DP-CO2 parameters

that could result in a required 5 log reduction of APC and Y&M using response surface

methodology. Microbial counts (Y&M and APC) were used as the dependent variable in

the experimental design. The study required 11 experiments with 4 factorial points, 4

star points and 3 center points for replications. A high initial microbial load in the juice

(1.78 x 107 cfu/mL for total aerobic microorganisms, 1.97 x 105 cfu/mL for yeasts/molds)

was achieved after incubating the sample for 4 days at 210C. Microbial log reduction

was determined for each experimental run as: log (initial number of cfu/mL) log









(number of cfu/mL after treatment). Microbial inactivation was evaluated immediately

after processing. Microbial counts were made from triplicate samples of each

processing treatment serially diluted (1 x 10-1 to 1 x 10-6) in duplicate by mixing 10 mL of

each puree with 90 mL of sterile Butterfield's buffer. Total plate counts were determined

by aerobic count Petrifilms and yeast/mold Petrifilms (3M Petrifilm Microbiology

Products, St. Paul, MN, U.S.A.) by plating 1 mL of the dilutions onto the Petrifilms in

duplicate and enumeration after 48 h at 35 C and 72 h at 24 C, respectively, according

to the manufacturers guidelines. Experimental data were analyzed by regression

analysis using SAS 9.0 software (SAS Institute Inc., Cary, NC, U.S.A.), fit to quadratic

polynomial equations, and the results used to select the optimal DP-C02 processing

conditions (34.5 MPa, 8%CO2 and 6.9 min) for assessment of microbial stability.

Storage Study and Microbial Stability

Fresh guava puree (not spoiled) was treated at 34.5 MPa, 8% C02 and 6.9 min

residence time (parameters were based on DP-C02 optimization results) and stored

under refrigerated conditions (4 C). Three 1 L bottles of puree were analyzed by the

methods described below on a weekly basis for the first 6 wks of storage followed by 2

wks intervals up to 14 wks of storage.

Chemical Analyses

PE activity: PE activity was measured by the method of Rouse and Atkins (1955).

Ten grams of guava puree were placed in a beaker with 40 mL of 1% pectin solution.

The sample was warmed in a water bath to 300C. While stirring and maintaining

constant temperature, 2 N NaOH was added until the pH was stable at 7. NaOH (0.05

N) was added until the pH was between 7.6 and 7.8 and the exact pH was recorded.


100









One hundred pL of 0.05 N NaOH were added and the time required for the solution pH

to recover between 7.6 and 7.8 was recorded and used in the following equation.


PE units/g = (mL NaOH) (Normality of NaOH)(1000)
(time)(g sample)


Cloud: Cloud was measured as described by Kincal and others (2006). Guava

puree (1.50 mL) was poured into a 1.5 mL centrifuge tube, placed in an Eppendorf

centrifuge (Model 5415; Brinkman Instruments Inc., Westbury, N.Y., U.S.A.), and

centrifuged at 320g for 10 min at room temperature. The supernatant (300 [tL) was

placed in a spectrophotometer (SpectraMax 190, Molecular Devices, Sunnyvale, CA)

and absorbance at 660 nm was recorded as the cloud value with distilled water serving

as blank (Versteeg and others 1980).

pH: The pH of treated and untreated guava puree samples was measured using a

digital pH meter (Accumet Basic AB15, Fisher Scientific, Fair Lawn, NJ, U.S.A.), after

calibration with commercial buffer solutions at pH 7.0 and 4.0. A sample (40 mL) was

placed in a 50 mL beaker with a magnetic stir bar, the pH electrode was inserted, and

pH was recorded after stabilization.

oBrix: A digital refractometer with temperature correction (Abbe Mark II; Reichert

Scientific Instruments, Buffalo, N.Y., U.S.A.) was used (Redd and others 1986)

Viscosity: Viscosity measurements were obtained using a Brookfield DV-E Digital

Viscometer (Brookfield Engineering Laboratories, Middleboro, MA, U.S.A.). Five

hundred mL of puree were placed in a 1 L beaker. A spindle number 1 and 50 rpm were

used to obtain the viscosity measurement, which was expressed as cP (centipoise or

mPa.sec).









Statistical Analysis: Repeated measures ANOVA and mean separation using

Tukey's test (a=0.05) was performed to evaluate the effect of treatment [fresh (Control),

thermal, and DP-C02 processed] and storage time (0, 1, 2, 3, 4, 5, 6, 8, 10, 12, and 14

weeks) on microbiological results and physicochemical parameters using SAS statistical

9.0 software (SAS Institute Inc., Cary, N.C., U.S.A.).



Results and Discussion

Solubility Experiments

The solubility of C02 in guava puree, in a model system and in water was

measured at 35 oC and pressures ranging from 7.58 to 31.03 MPa. Figure 4-2 illustrates

that by increasing the pressure, the solubility of CO2 increased. The presence of

dissolved solutes such as simple sugars, acids and other carbohydrates lowered the

amount of CO2 that dissolved in the puree, resulting in a significant difference between

the CO2 solubility in water, in the model system and in the puree. The puree showed a

significantly lower CO2 solubility compared to the model system (a= 0.05%). Similar

results were observed by (Calix and others 2008) when the CO2 solubility in orange

juice, an orange juice model system and water was studied.

Ferrentino and others (2010) suggested that the observed decrease in CO2

solubility in sodium phosphate solutions was affected by the type of chemical compound

dissolved in the liquid matrix The solubility increased significantly from 7.58 to 10.34

MPa and remained almost constant at 4.3 g of C02/100 g puree from 10.34 to 31.03

MPa. The final solubility for the puree was not significantly different from the solubility of

the model system at 7.58 MPa.


102









Table 4-1 shows the oBrix, pH, and titratable acidity for the puree before and after

the solubility measurements. Data are shown in Tables B-1 and B-2. A decrease in pH

and an increase in titratable acidity (TA) and oBrix with increasing pressure was

observed, probably due to the residual CO2 remaining in the puree after

depressurization. As expected, an increase in TA and OBrix was observed. These

results are related to increase in solubility of CO2 when processing pressure is

increased. Similar results were observed by (Calix and others 2008) in apple and

orange juice.

Considering that the solubility value was obtained under saturation and long

contact time conditions in a static equipment, the %C02 was set to 5.3 g CO2/100 g

sample.

Microbial Inactivation Study

The effect of DP-C02 at various processing pressures and treatment times on

Y&M and APC can be observed in Table 4-2. Initial and final numbers of bacteria were

determined by taking average cfu/mL counts on Petrifilms with the cfu's less than 250

cfu/mL .The %C02 used for the experiments was set to a constant value of 8.0 g

CO2/100 g puree, although in the solubility study, the experimental value was 5.3%. The

reason for this additional increment was due to fluctuations in the CO2 flow (equipment

limitations). Results from Table 4-2 demonstrated that under identical processing

conditions, Y&M were destroyed at a higher rate than aerobic microorganisms. Similar

results were observed by Del Pozo-lnsfran and others (2006) when muscadine grape

juice was processed by DP-C02. The average initial and final aerobic plate counts

(APC) standard deviations at each experimental condition are given in Table C-1. The

RSM analysis of data was performed with SAS 9.1 statistical software program (Cary,


103









NC, U.S.A.). The SAS code and output of the analysis are given in Table C-2 and C-3,

respectively. The statistical analysis of the response surface regression model showed

that the quadratic model fit for Y&M was statistically significant (P <0.05) and there was

a satisfactory correlation between the actual and the fitted values for Y&M (R2 = 0.88).

Significance of each parameter was decided at a=0.05 level and the parameters with p

value > 0.05 were excluded from the model. Results showed that only the parameters

pressure and pressure by residence time were significant. Lecky (2005) observed little

effect by increasing residence time from 4 to 5 min with watermelon juice treated at a

pressure of 34.4 MPa, 40C and 10% CO2. Gunes and others (2005) reported that

microbial inactivation by DP-C02 is governed by the transfer rate and the penetration of

carbon dioxide into cells, which can be improved by increasing pressure, and increasing

the processing temperature

The model with the estimated coefficients gives the prediction of log microbial

reduction (log red) as a function of pressure and residence time: Y&M = 6.0605 -

0.1619*Pressure (MPa) + 0.018945*Pressure*Residence time (min). Coefficients were

determined for the coded values of each variable. The log reductions predicted at

eleven experimental runs using this equation are close to the experimental log

reductions (Table 4-3). Statistical analysis for the APC showed no significant difference

within the 11 experimental runs so processing parameters were based on Y&M

reduction. Microbial inactivation is highly dependent on the type of microorganisms

present in the food matrix due to distinct microbial cell microstructure and the diffusion

of CO2 into the microbial cell (Ballestra and others 1996). Bacterial vegetative cells and

Y&M are more pressure and CO2 sensitive than bacterial spores, e.g. higher pressures


104









are required for complete inactivation of spores. Overall, microbial reduction is related to

the direct relationship between CO2 solubility and increasing processing pressure

(Balaban and others 1991 a) which consequently affects the diffusion of CO2 into the

microbial cell as well as the explosive decompression that occurs during DP-CO2

processing. Results of this optimization study were used to determine those DP-CO2

conditions that achieved > 5 log reduction of Y&M, rather than bacterial cells. According

to the results obtained from the regression surface analysis, 8 min was enough to

achieve this log reduction under the established processing conditions.

Because of the increase in product thickness and foam formation, the residence

time was set to 6.9 min. A higher bacterial reduction would be achieved at 6.9 min

rather than 6.5 min under the same processing conditions. For the stability study, the

processing conditions used for the DP-CO2 were pressure of 34.5 MPa, 8% CO2, a

residence time of 6.9 min and a temperature of 35 oC.

Table 4-4 and 4-5 present the results obtained for PEA, cloud measurement, pH,

oBrix and TA. There were significant differences in the PEA of the treated puree when

compared to the control. An apparent increase in PEA was observed. The method used

for PEA determination is based on adjusting sample pH to the optimal pH of the enzyme

activity (7.6 to 7.8). After this, an excess of basic solution is added and the change in

pH due to enzyme activity and liberation of carboxyl groups is monitored. The increase

may be due to microbial enzyme leakage into the puree. One of the inactivation

methods for DP-CO2 is cell wall rupture, which can cause a leakage of enzymes into the

puree. Another possible explanation for this increase in PE activity is the release of PE


105









from the cell wall of plant material. PE is a cell wall-bound pectic enzyme complex that

can be released from the pulp during DP-C02 process (Tribess and Tadini 2006).

Cloud results showed significant differences between the treated sample and

control for all treatments.There was a significant cloud loss for all treatments, but cloud

loss was higher with an increase in pressure and residence time. Cloud loss was

between 59.6 and 80.9% for a processing pressure of 13.8 MPa. For a pressure of 34.5

MPa, cloud loss was between 66.11 and 90.82%. These results were not as expected

since different results were observed by Kincal and others (2006) in orange juice treated

with high pressure carbon dioxide processing.

No significant differences (a = 0.05) were found between the pH of the treated

sample and the control. At 5 and 6.5 min residence time, there were significant

differences in oBrix between the treated sample and the control. The oBrix of the treated

sample increased with increasing pressure and residence time. The increase in oBrix is

related to increase in solubility of CO2. Increasing pressure and residence time

increases the amount of CO2 that dissolves into the product. There were significant

differences in TA between the treated sample and the control. The TA increases with

the DP-C02 treatment. As pressure and residence time increases, TA increases. The

increase in TA can be related to the increase in CO2 solubility due to increasing

pressure and contact time (Lecky 2005).

Storage Study and Microbial Stability

Differences between treatment means for the storage study data were determined

by conducting analysis of variance using SAS 9.1 Software (Cary, NC, U.S.A.) at a

significance level of a = 0.05. Tukey's standardized range comparison test (a = 0.05)

was used to determine statistically different samples. The plot of APC for untreated


106









(control), heat-treated (90 C, 60 s) and DP-C02 treated (34.5 MPa, 35 C, 8% C02, 6.9

min) guava puree during 14 weeks of refrigerated storage (4 C) is given in Figure 4-3.

The thermal pasteurization process of guava puree showed the lowest count for aerobic

bacteria. This count remained almost constant during the first 4 weeks of storage. The

pasteurization process showed a characteristic microbial growth pattern in which the

three phases can be identified. The lag phase is represented during the first 4 weeks of

storage, the log phase can be observed between week 4 and 6 and the death phase

can be observed after week 7. As shown in Figure 4-3, the thermal process was more

effective in bacterial reduction than DP-C02. Aerobic counts in DP-C02 treatments

were not significantly different from the control puree during the first 8 weeks of storage

but showed a significant increase after week 10. The bacterial count on DP-C02

decreased after week 10 and was not different from control. The number of aerobic

bacteria in untreated samples remained almost unchanged during the 14 weeks of

storage. Storage of the thermal and DP-C02 treated puree (not spoiled) at 4 C showed

no significant differences in the Y&M during the first 6 weeks of storage. Both

treatments showed a significantly lower count when compared to the control (Figure 4-

4). Between week 6 and 8, there was a significant increase in the Y&M count for the

thermal pasteurized sample in which there are no differences in counts between control

and pasteurized samples. After 10 weeks of storage, the Y&M count for pasteurized

samples decreased and remained almost constant until 14 weeks of storage. DP-C02

samples showed a steady increase in Y&M counts during the first 8 weeks of storage.

After week 12, a significant decrease in Y&M counts was observed for DP-C02. Control

puree showed the highest Y&M count and these counts remained almost the same









throughout the 14 weeks of storage. Initial Y&M counts were 3.3 log and DP-C02 was

effective in decreasing Y&M count to 1.3 log but did not cause any reduction on aerobic

count, APC of DP-C02 were similar to control. The pasteurization process caused 2 and

1.3 log reduction on APC and Y&M respectively when compared to control which had

an initial APC of 3.5 log and 3.3 log for Y&M.

DP-C02 and thermal processing significantly decreased PE activity when

compared to the fresh sample. PE activity decreased 21.45% and 51% for DP-C02 and

thermally treated samples respectively. There was a continuous decrease in PE activity

for all the samples during the 14 weeks of storage. During the first 5 weeks of storage,

there were no significant differences in enzyme activity between the DP-C02 treated

and control samples (Figure 4-5). Between week 6 and 12, fresh samples had a

significantly higher PE activity when compared to the DP-C02 treated samples and

there were no significant differences between the enzyme activity of control and DP-

C02 treated samples at the end of the storage study. Similar results were shown by

Kincal and others (2006) when orange juice was treated with a continuous DP-C02

system under different pressures at constant residence time. Ferrentino and others

(2009) reported an reduction in PE activity after treating grapefruit juice with DP-C02. A

total reduction in PE activity of 47% and 52% was observed at the end of the storage

study for DP-C02 and pasteurized samples respectively. Residual PEA in thermal

pasteurized samples can be explained by the presence of PE isoforms which possess

different inactivation kinetics. Thermal stable isoform have been reported in orange juice

(Versteeg and others 1980; Randall and others 1998) and this enzyme is more heat-

resistant than common spoilage microorganisms. Versteeg and others (1980) studied


108









the juice cloud destabilizing properties and heat stability of PE in navel orange and

found that a pasteurization time of 0.8 min at 90 C would be necessary to inactivate

99% of the high molecular weight PE. This high molecular weight enzyme is the thermo-

stable isoform. PE inhibition for control sample was observed during the 14 weeks of

storage. Temperature is known to affect enzyme activity. Every enzyme has an optimal

temperature below which the enzyme activity usually decreases. Versteeg and others

(1980) suggested that a temperature dependant change in conformation causes this

decrease in PE activity at 5 C. The PE reduction noted in this study may be due to

storage temperatures at 4 oC. Cold-deactivation of enzymes can occur at low

temperatures or bellow 10 C. Low temperature reduces the strength of non-polar

forces promoting the dissociation of sub-units and compromising the enzyme activity

(Damodaran and others 2008).

Viscosity measurements for fresh, DP-C02 and thermally treated samples are

shown in Figure 4-6. DP-C02 treatment had a significantly higher viscosity compared to

fresh and thermally treated samples. The increase in viscosity was 60%. During the first

4 weeks of storage, DP-C02 showed a maximum increase in viscosity of 95% which

decreased progressively to 69% after week 5 of storage. The possible explanation for

the increase in viscosity is the formation of a reversible gel (Figure C-1). At the

beginning, formation of bridges between calcium and pectic acid was suggested as a

mechanism for the reversible gel formation. Another possible explanation is the particle

size reduction of guava puree during the decompression of the puree at the end of the

treatment. Lim and others (2006) found that during treatment of mandarin juice with a

continuous high pressure C02 system, calcium precipitation occurs. When dissolved in


109









water, C02 combines with water and forms carbonic acid that dissociates into

bicarbonate and hydrogen ions. When the pressure is released, bicarbonate is

converted to carbonate and may precipitate calcium, preventing it from acting as a

bridge between pectic chains that leads to cloud loss (Lim and others 2006).

Cloud stability was measured and results were reported as absorbance values at

660 nm. As absorbance decreased, cloud loss increased. When performing the

centrifugation, no phase separation was achieved for the DP-C02 samples. This was

mainly due to the reversible gel formation. To achieve a phase separation, all samples

were frozen until the end of the storage study. Freezing changes the molecular structure

of the gel, causing a phase separation and cloud loss. Results for cloud loss are

presented in Table 4.6. These results indicated that cloud loss for the DP-C02 treated

samples after freezing was higher than the fresh and pasteurized ones. Pasteurized

samples showed an increase in cloud during the 14 weeks of storage. This can be

related to inactivation of PE during the thermal treatment. Control samples showed

significant reduction in cloud due to PE activity during refrigerated storage. In DP-C02

processing, it is hypothesized that the depressurization of the system leads to

homogenization of the puree causing smaller particles of the puree colloids increasing

cloud stability. Even when the results are reported as cloud loss, more studies need to

be conducted to understand the mechanism of gel formation and the changes that occur

during freezing. Ferrentino and others (2009) showed cloud retention after treating

grapefruit juice with a continuous DP-C02 system, even when PE was active in the

treated juice. Kincal and others (2006) also showed that cloud increased in orange juice

treated with a continuous DP-C02 system. When treating orange juice with a static high-


110









pressure 002 system, Aerrola and others (1991) showed that cloud enhancement

occurs.

The mean pH and Brix of the control, DP-CO2 and thermal treated samples

showed some weekly fluctuations during storage (Table 4-6). pH for control and DP-

CO2 treated samples decreased during week 4 of storage, significantly increased during

week 5 and then stayed almost constant. The decreased in pH may be related to

bacterial metabolism during which carbohydrates are hydrolyzed and acid molecules

are released causing a pH reduction. Thermal treated samples had a significantly higher

pH than control and DP-CO2 treated samples (related to bacterial inactivation during

thermal processing). There were significant differences in Brix between fresh, DP-CO2

and thermal treated samples. In general, DP-CO2 showed a significantly higher Brix

than control samples during the first 6 weeks and between week 12 and 14 weeks of

storage. The apparent increase in Brix measurement on DP-CO2 samples may be

related to residual CO2 dissolved in the samples. The fluctuations in pH and Brix were

probably because of bottle-to-bottle variations.

Conclusions

DP-CO2 treatment resulted in a guava puree with reduced microbial load, causing

5 log reductions in Y&M and a maximum of 3.5 log reductions in APC. Solubility study

showed that a 4.3% of CO2 caused saturation of guava puree in a static system. Due to

equipment limitations, a value higher than the experimental saturation solubility of

%C02 was used. The experimental design and model for determining C02 solubility

minimized the excess use of CO2 during processing and can be used to economically

optimize DP-CO2 treatment. Optimal conditions for DP-CO2 to achieve a 5 log reduction

in Y&M were determined as 8% CO2, 35 C, 6.9 min of residence time and a pressure









of 34.5 MPa. These conditions maintained guava puree microbiologically stable during

14 weeks of refrigerated storage at 4 C. DP-C02 samples showed pH similar to fresh

samples but TA for the DP-C02 was higher than the untreated ones. Increases in

viscosity and cloud loss were observed for the DP-C02 treated puree. In addition, PE

was partially inactivated. A 20% reduction on PEA was initially achieved after DP-C02

processing. Further study is recommended to understand the mechanism of gel

formation and cloud loss due to DP-C02 treatment. From the storage study, it is evident

that DP-C02 processing can maintain quality attributes of fresh puree, and extend its

shelf life.




Holding Tube
Juice Pump E
E


Chiller
Heating
System
-- CO2 Pump Expansion
Expansion
r 0 PValve
Ve Treated
Juice





I-
0
r-c



Figure 4-1. Schematic diagram of Dense Phase Carbon Dioxide equipment


112















A B



D G ...- *..-.. -


G g-








-*- guava


5 10 15 20 25 30 35
Pressure (Mpa)
Figure 4-2. Carbon dioxide solubility results obtained for guava puree, water and guava
puree model system at different processing pressures. Error bars represent
standard deviation for n=3. Different letters within each pressure represent
significant differences at a=0.5.


Table 4-1. pH, OBrix and titratable acidity values for the treated and untreated guava
puree under different processing pressures
Pressure TA (g citric acid/100 g
(MPa) pH oBrix puree)
Control 3.80a 6.8a 0.55a
7.58 3.79a 6.8a 0.55a
10.34 3.75b 6.9ab 0.58b
17.24 3.75b 6.9ab 0.60b
24.13 3.74b 6.9ab 0.60b
31.03 3.72b 7.0b 0.660
* Different letters within the columns represent significant differences


113


4

-- 5
E
0
S4 -
O
0
0


. 2
- -
v,
>,




O
3 2-

o0
0 1-


0


I I I I I I I










Table 4-2. Processing conditions and microbial reduction obtained during DP-C02
process optimization
Pressure Residence Juice Flow CO2 flow log 10 log 10
Run (MPa) time (min) (g/min) (g/min) Red Red
APC Y&M

1 34.5 8 312.5 25 3.20 4.30
2 24.1 8 312.5 25 3.28 4.28
3 24.1 6.5 384.6 30.8 3.32 4.07
4 13.8 8 312.5 25 3.23 4.09
5 34.5 6.5 384.6 30.8 3.51 3.86
6 24.1 6.5 384.6 30.8 3.52 3.97
7 13.8 6.5 384.6 30.8 3.59 4.03
8 13.8 5 500 40 3.57 4.18
9 34.5 5 500 40 3.71 3.21
10 24.1 5 500 40 3.66 3.49
11 24.1 6.5 312.5 25 3.65 3.65


Table 4-3 Actual and predicted yeast and mold log reduction
the surface response analysis


using the equation from


Pressure Residence
Run (MPa) time (min) Actual Predicted Predicted
Log (significant terms
reduction (all terms) only)
1 34.5 8 4.3 4.38 5.70
2 24.1 8 4.28 4.19 5.80
3 24.1 6.5 4.08 3.90 5.12
4 13.8 8 4.09 4.10 5.92
5 34.5 6.5 3.86 3.79 4.72
6 24.1 6.5 3.97 3.90 5.12
7 13.8 6.5 4.03 4.10 5.52
8 13.8 5 4.18 4.10 5.13
9 34.5 5 3.21 3.19 3.74
10 24.1 5 3.49 3.59 4.44
11 24.1 6.5 3.65 3.90 5.12


114









Table 4-4. Pectinesterase activity, cloud and
and aftEr DP-CO2 trEatment


pH measurement of guava puree before


115


Initial Final
PEA PEA Cloud
Pressure Residence (units/g (units/g reduction Initial Final
Run (MPa) time (min) sample) sample) (%) pH pH
2.86-4 + 4.35-4 + 3.7 + 3.7 +
1 34.5 8 9.93-6 5.01-5 90.82 0.03 0.02
2.86-4 + 4.37-4 + 3.7 + 3.7 +
2 24.1 8 9.93-7 424-5 79.84 0.0 0.01
2.86-4 + 4.27-4 + 3.7 + 3.7 +
3 24.1 6.5 9.93-8 1.44-5 86.2 0.03 0.03
2.86-4 + 5.70-4 + 3.7 + 3.7 +
4 13.8 8 9.93-9 4.93-5 80.9 0.03 0.02
3.62-4 + 4.98-4 + 3.7 + 3.7 +
5 34.5 6.5 1.60-5 6.41-5 85.89 0.01 0.02
3.62-4 + 4.47-4 + 3.7 + 3.7 +
6 24.1 6.5 1.60-6 3.23- 84.23 0.01 0.01
3.62-4 + 4.58-4 + 3.7 + 3.7 +
7 13.8 6.5 1.60-7 2.31-5 59.6 0.01 0.02
3.62-4 + 4.93-4 + 3.7 + 3.7 +
8 13.8 5 1.60-8 2.21-5 77.06 0.01 0.03
4.86-4 + 4.73-4 + 3.7 + 3.7 +
9 34.5 5 2.83-5 1.97-5 66.11 0.01 0.02
4.86-4 + 4.91-4 + 3.7 + 3.7 +
10 24.1 5 2.83-6 4.43-5 78.84 0.01 0.01
4.86-4 + 4.96-4 + 3.7 + 3.7 +
11 24.1 6.5 2.83- 3.63-5 80.26 0.01 0.02
Mean + standard deviation for n=3









Table 4-5. oBrix and titrabable acidity of guava puree before and after DP-C02 treatment
Residence Initial TA Final TA
Run Pressure time (min) Initial oBrix Final oBrix (%)* (%)*
0.62 + 0.65 +
1 34.5 8 7.1 + 0.00 7.1 + 0.00 0.01 0.01
0.62 + 0.66 +
2 24.1 8 7.1 +0.00 7.2 +0.04 0.01 0.01
0.62 + 0.67 +
3 24.1 6.5 7.1 + 0.00 7.1 + 0.05 0.01 0.01
0.62 + 0.69 +
4 13.8 8 7.1 + 0.00 7.2 + 0.04 0.01 0.01
0.64 + 0.70 +
5 34.5 6.5 6.9 + 0.04 6.9 + 0.05 0.01 0.00
0.64 + 0.68 +
6 24.1 6.5 6.9 + 0.04 6.9 + 0.00 0.01 0.01
0.64 + 0.70 +
7 13.8 6.5 6.9 + 0.04 6.9 + 0.00 0.01 0.01
0.64 + 0.69 +
8 13.8 5 6.9 + 0.04 6.9 + 0.00 0.01 0.01
0.63 + 0.67 +
9 34.5 5 6.7 + 0.05 6.7 + 0.00 0.01 0.02
0.63 + 0.69 +
10 24.1 5 6.7 + 0.05 6.7 + 0.00 0.01 0.01
0.63 + 0.66 +
11 24.1 6.5 6.7 + 0.05 6.7 + 0.00 0.01 0.01
Mean + standard deviation for n=3, Expressed as g citric acid per 100 g sample


116










7000 --Control **** DPCD Pasteurized


6000 -.
-I

S4-
o 5000


S 4000 ..... *




3 2000
E

2000



1000




0 3 6 9 12
Storage time (weeks)
Figure 4-3. Aerobic plate count for control, DP-C02 and thermal treated guava puree
during 14 weeks of storage. Error bars represent standard deviation for n=6









6000

--Control **** DP-C02 Pasteurized
5000


E 4000
u,

o 3000





E
I 2000 < ..
S** "", '-..
= 1000 A
0
Sg ... .... ...*** **

0 2 4 6 8 10 12 14
Storage time (weeks)


Figure 4-4. Yeast and mold plate count for control, DP-C02 and thermal treated guava
puree during 14 weeks of storage. Error bars represent standard deviation for
n=6


118









8.00E-04


7.00E-04
S- Control
S6.00E-04 *** DP-C02

Lu Pasteurized
5.00E-04 ..

t 4.00E-04

S3.00E-04 ., *,t

*2.00E-04 -

1.00E-04

O.OOE+00 ....
0 1 2 3 4 5 6 8 10 12 14
Storage Time (weeks)
Figure 4-5. Pectinesterase activity for control, DP-C02 and thermal treated guava during
14 weeks of storage. Error bars represent standard deviation for n=3.


119












600 -


^ 500 -
u,

400 -
)

o
S300 -
0

> 200


100

0
0


6 8


10 12 14 16


Figure 4-6. Viscosity


Storage Time (weeks)
measurement for control, DP-C02 and thermal treated guava


during 14 weeks of storage. Error bars represent standard deviation for n=3


120


700


'** *- ...
"' '"...........





-*- Control
**** DP-C02
-A Pasteurized


...-.-A. --- 2. -- .A--- .A--- .A


v


I-'









Table 4-6. pH, OBrix and cloud measurement for control, DP-C02 and thermal treated guava during 14 weeks of storage
pH oBrix Cloud
Week Control DP-C02 Pasteurized Control DP-C02 Pasteurized Control DP-C02 Pasteurized
0 3.86 + 0.01 3.87 + 0.01 3.87 + 0.01 6.9 + 0.0 7.1 + 0.0 7.0 + 0.0 0.80 + 0.12 0.13 + 0.03 0.62 + 0.05
1 3.87 + 0.01 3.87 + 0.01 3.90 + 0.01 6.9 + 0.0 7.1 + 0.0 7.0 + 0.0 0.67 + 0.04 0.04 + 0.01 0.80 + 0.03
2 3.82 + 0.01 3.88 + 0.01 3.81 + 0.01 6.8 + 0.1 7.1 + 0.1 7.0 + 0.1 0.70 + 0.08 0.03 + 0.00 0.83 + 0.05
3 3.85 + 0.01 3.85 + 0.01 3.86 + 0.01 6.8 + 0.1 7.1 + 0.0 7.0 + 0.1 0.54 + 0.08 0.05 + 0.01 1.02 + 0.02
4 3.77 + 0.01 3.72 + 0.01 3.95 + 0.01 6.7 + 0.1 7.0 + 0.1 7.1 + 0.1 0.47 + 0.02 0.06 + 0.01 1.11 + 0.04
5 3.84 + 0.00 3.82 + 0.01 3.94 + 0.01 6.7 + 0.1 7.0 + 0.1 7.0 + 0.1 0.38 + 0.02 0.03 + 0.01 1.09 + 0.06
6 3.89 + 0.01 3.82 + 0.01 3.93 + 0.01 6.8 + 0.1 6.8 + 0.1 7.0 + 0.0 0.41 + 0.02 0.03 + 0.01 1.08 + 0.04
8 3.88 + 0.00 3.84 + 0.01 3.97 + 0.01 6.8 + 0.1 6.8 + 0.1 7.2 + 0.0 0.33 + 0.04 0.04 + 0.00 1.13 + 0.05
10 3.84 + 0.0 3.85 + 0.0 3.94 + 0.0 6.9 + 0.1 6.9 + 0.1 7.1 + 0.0 0.26 + 0.03 0.05 + 0.01 1.05 + 0.05
12 3.85 + 0.01 3.86 + 0.01 3.93 + 0.02 7.0 + 0.1 7.2 + 0.1 7.2 + 0.1 0.19 + 0.01 0.05 + 0.00 1.16 + 0.03
14 3.85 + 0.00 3.83 + 0.01 3.93 + 0.02 7.0 + 0.0 7.3 + 0.1 7.4 + 0.1 0.21 + 0.02 0.05 + 0.00 1.21 + 0.03
Mean + standard deviation for n=3









CHAPTER 5
PHYSICO-CHEMICAL AND PHYTOCHEMICAL CHANGES OF DENSE PHASE
CARBON DIOXIDE AND THERMALLY TREATED GUAVA PUREE DURING
REFRIGERATED STORAGE

Abstract

Guava (Psidium guajava, L) is a tropical fruit rich in phytochemicals, antioxidants.

Guava puree is normally heat pasteurized to extend its shelf life, but the fresh taste is

modified. The use of non-thermal processes can minimize the development of

undesirable characteristics that occur during heat processing. Dense phase carbon

dioxide (DP-CO2) is a non-thermal method emerging as an alternative to traditional

thermal pasteurization. The objective of this study was to determine the changes on

some physico-chemical and phytochemical qualities of DP-CO2 and thermal processing

during refrigerated storage of guava puree.

Storage stability of DP-CO2 and thermal processed guava puree was assessed

and compared to control puree during 14 weeks of storage at 40C. Physicochemical

titratablee acidity (TA), color (L*, a, b) and organic acids) and phytochemical (total

phenolics, antioxidant capacity, ascorbic acid and identification of phenolic compounds)

analyses were performed. DP-CO2 samples showed higher TA acidity when compared

to control and thermal treated ones, and TA increases during storage for all 3 samples.

DP-CO2 increases L* and a* values but no effect on a* values were observed. Organic

acid content was not affected by treatment but little change were observed during

storage. During storage, all puree samples showed no significant difference (a>0.05) in

antioxidant capacity (ORAC) and total soluble phenolics (TP). Processing guava puree

with DP-CO2 is a viable technology for the preservation of product quality.


122









Introduction

Guava (Psidium guajava, L) is a tropical fruit rich in antioxidants and vitamin C. It

is a member of the Myrtaceae family, which has more than 80 genera and 3,000

species distributed throughout the tropics and subtropics (Nakasone and Paull 1998).

The genus Psidium includes five species, with P. guajava being the most widely

cultivated species of the family Myrtaceae. The chemical composition of the fruit varies

with the stage of development, variety and season. Titratable acidity (TA), reported as

citric acid content, ranges from 0.08 to 2.20%. Guava fruits consist of about 20% peel,

50% flesh (pericarp) and 30% seed core (Salunkhe and Kadam 1995). Among fruit

types, guava is the second highest in vitamin C content, containing up to five times the

amount in oranges (Dweck 2005). Vitamin C (ascorbic acid) is water-soluble and highly

susceptible to oxidative degradation, which often is used as an index for nutrient

stability during processing or storage (Damodaran and others 2008). The vitamin C

content of the fruit fluctuates between 37 and 1,000 mg ascorbic acid per 100 g guava

fruit. Vitamin C content of red-fleshed guava is higher than that of white-fleshed guava

(Mowlah and Itoo 1983). Within each fruit, the distribution of vitamin C is higher in the

skin than in the central portion of the flesh. The composition of organic acids present in

guava was studied by Chang and others (1971). They found that citric and malic acids

were predominant followed by tartaric, glycolic and lactic acid. Similar results were

found by Wilson and others (1982) in a study of four cultivars from Florida. They found

traces of fumaric acid, which was detected for the first time in guava. Guava fruits

contain significant amounts of polyphenols but their concentration and corresponding

astringency decreases as the fruit matures.


123









Fruits and vegetables are important components of a healthy diet and are one of

the main sources of antioxidants. Clinical research supports the fact that consumption of

fruits and vegetables is beneficial for prevention of cancer, heart disease and other age-

related diseases (Dietary guidelines for Americans 2010). The guava fruit is rich in

tannins, phenols, triterpenes, flavonoids, carotenoids, vitamins and fiber. Most of the

guava's therapeutic activity is attributed to the high content of flavonoids, which also

have antimicrobial activity. Dietary flavonoids and other plant phenolics have been

reported to have antioxidant activity, antimicrobial and anti-inflammatory action (Huang

and others 1992) and have been associated with a reduced risk of cardiovascular

diseases and cancer (Temple 2000; Pietta 2000). The health benefits of consuming a

diet rich in dietary fiber (DF) have been extensively studied (Gary 1999). Jimenez-

Escrig and others (2001) evaluated guava as a source of natural antioxidant

compounds and DF (Jimenez-Escrig and others 2001). They found a remarkable

antioxidant capacity related to the phenolics content. Peel and pulp of Psidium guajava

fruit presented high levels of DF, an indigestible fraction, and phenolic compounds.

They concluded that guava could be a rich source of natural antioxidants and dietary

fiber (Jimenez-Escrig and others 2001). Gorinstein and others (1999) showed that the

content of polyphenols (4.79 5.11 mg/100 g fresh fruit), gallic acid (340.6 408.0

pg/100 g fresh fruit) total fiber (5.14 6.04g/100 g fresh fruit) and soluble fiber (2.39 -

3.01g/100 g fresh fruit) in the guava were higher than the amount found in persimmon

fruit (Gorinstein and others 1999).

Guava puree is the most important guava form for the juice industry. The puree is

preserved by freezing to -20 to 0 OF (-29 to -18 C), canning, aseptic packaging or


124









pasteurization. Pasteurization is conducted between 80 and 90 C (190 194 OF) for 60

seconds, then the pasteurized product is cooled and filled into containers. After

pasteurization, the puree can be frozen and stored at -180C (0F) for up to a year but its

fresh qualities are diminished.

Considerable research has focused on the development of non-thermal

processing technologies to avoid the detrimental chemical changes caused by heat

pasteurization. High pressure processing (HPP) causes minimal changes in the "fresh"

characteristics of foods by minimizing thermal degradation (Raso and Barbosa-Canovas

2003). Yen and Lin (1996) compared the effect of high pressure treatment and thermal

pasteurization on the quality and shelf life of guava puree. Samples from both

treatments were stored at 40C over a period of 60 days. All the treatments were equally

effective in reducing the microbial load (> 2 log reduction) of the puree. During storage

at 4 C, pressurized puree at 600 MPa showed lower levels (< 10 cfu/min) of

microorganisms compared to control samples. The color of pressurized guava puree

was similar to that of freshly extracted puree. This research indicated that high pressure

treatment (600 MPa) of guava puree at 25 C for 15 min could maintain good quality up

to 40 days storage at 4 C (Yen and Lin 1996).

Phenolic compounds have been associated with positive and negative attributes in

terms of sensory and nutritional quality. Positive attributes include their close

association with sensory and nutritional quality. Their nutritional value has been linked

to prevention of cancer, antimicrobial properties, antimutagenicity, antioxidant potential,

reduction of coronary heart disease risk, antiviral, anti-inflammatory and antitumor

activity (Sonko and Xia 2005). Their sensory attributes are related to their contribution to


125









the flavor, astringency and color characteristics of foods. The anti-nutritional effect of

phenolic compounds involves their reaction with proteins, carbohydrates, minerals and

vitamins lowering the bioavailability of these nutrients or their nutritional value. In

addition, phenolic compounds can adversely affect the sensory qualities of food by the

production of off-flavor, their involvement in enzymatic browning, nonenzymatic

browning and precipitation of proteins (Shahidi and Naczk 2003).

The objective of this research was to compare the physicochemical and

phytochemical qualities changes of guava puree after DP-CO2 and thermal processing

during refrigerated storage.

Materials and Methods

Chemicals and Standards

Commercial standards of oxalic, malic, ascorbic, citric, gallic, ellagic and cinnamic

acids were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.). AAPH (2,2'-

azobis(2-methylpropionamidine) dihydrochloride), fluorescein (free acid), Trolox (6-

hydroxy-2,5,7,8-tetramethylchroman-2carboxilic acid) and Folin-Ciocalteu's reagent

were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.).

Total Phenolics, Antioxidant Capacity and Ascorbic Acid Analysis

Total soluble phenolic levels were measured using the Folin-Ciocalteu assay

(Talcott and others 2003) and quantified as mg of gallic acid equivalents per L sample.

Ascorbic acid was measured using the 2, 6-dichloroindophenol titration method. The

ascorbic acid reduced the indicator dye, 2, 6-dichloroindophenol, to a colorless solution

through oxidation reduction reactions. At the end point, excess unreduced dye was

rose pink in acid solution. Vitamin C was titrated in the presence of metaphosphoric

acid-acetic acid solutions to maintain proper acidity for the reaction and to avoid


126









autoxidation of ascorbic acid at high pH. The final value was expressed as mg of

ascorbic acid per mL sample. The antioxidant capacity was determined using the

oxygen radical absorbance capacity (ORAC) assay (Huang and others 2002) with data

expressed in Trolox equivalents per milliliter ([[mol of TE/mL).

Titratable Acidity (TA)

A Brinkmann Instrument (Brinkmann Instruments Co., Westbury, NY) pH meter

consisting of a Metrohm 655 Disomat, Metrohm 614 Impulsomat and Metrohm 632 pH-

meter was used for titration of guava puree. Ten + 1.0 g guava puree sample was

titrated to an end point of pH 8.2 by using standardized 0.1 N NaOH and the amount of

NaOH used for titration was recorded.

Percent titratable acidity (w/w) was expressed as percent citric acid and calculated

by the following equation:

%TA= (mL NaOH used)X(normality of NaOH)X(meq citric acid)X(100)

(g sample)

The value used as meq of citric acid was 0.064047, this value represent the

molecular weight of the acid divided by the number of equivalents (3) in the reaction.

Color Analysis

Color was measured using a ColorQuest XE colorimeter (HunterLab, Reston, VA,

U.S.A.). Samples of 40 mL were placed in a 20 mm quartz cell and L*, a, and b

parameters were recorded in reflectance, specular included mode.

High Performance Liquid Chromatography Analysis

Organic acids were identified and quantified by reverse phase HPLC using

modified chromatographic conditions described by Gokmen and others (2000).

Separation was performed on a 4.6 mm x 250 mm Acclaim 120 C18 5 pm column









(Dionex, Sunnyvale, CA, U.S.A.), using 0.2 M KH2PO4, pH 2.4 (Fisher Scientific, Fair

Lawn, NJ, U.S.A) as the mobile phase at a flow rate of 1.0 mL/min. Prior to organic acid

analysis, all samples were passed thorough pre-conditioned Envi-18 Sep-Pak cartridges

(Sigma Aldrich, St Louis, MO, U.S.A.) to remove neutral polyphenolics. After discarding

the first mL, samples were collected and filtered through a 0.45pm PTFE filter (Millipore,

Bedford, MA, U.S.A.) and analyzed for organic acids. Organic acids were characterized

based on UV-VIS spectral interpretation from 210 to 360 nm and comparison to

authentic standards. Polyphenolics from guava puree were subsequently concentrated

and purified using Envi-18 Sep-Pak Vac 20 cc mini-columns (Sigma Aldrich, St Louis,

MO, U.S.A). Polar constituents were removed with acidified water (0.01% v/v HCI) and

polyphenolic compounds subsequently fractioned and eluted with methanol (0.01% v/v

HCI) followed by ethyl acetate. Solvent in both fractions was later evaporated under

reduced pressure at <40C. The resulting polyphenolic extracts were re-dissolved in 3

mL of a MeOH:H20 (60:40) solution.

Phenolics compounds were identified and quantified using a Dionex HPLC system

equipped with a degasser, a binary pump, an autosampler/injector and diode array

(PDA 100) detector (Dionex, Sunnyvale, CA). Compounds were separated on a 250 x

4.6 mm Acclaim 120 C18 5 pm (Dionex, Sunnyvale, CA, U.S.A.) column. Mobile phases

consisted of acidified water (phase A) and 60% methanol in water (phase B), both

adjusted to pH 2.4 with o-phosphoric acid (Fisher Scientific, Fair Lawn, NJ, U.S.A). A

gradient program was used, starting at 100% phase A. The solvent program ran phase

B from 0% to 60% in 20 min; 60% to 100% in 20 min; 100% for 7 min; 100% to 0% in 3


128









min and a final hold for 2 min. The flow rate was 0.8 mL/min, and detection was done at

260, 280, 320, 360 and 520 nm.

Results and Discussion

Total Phenolics and Antioxidant Capacity

Table 5-1 shows the antioxidant capacity and total soluble phenolics content of the

control, DP-C02 and thermal treated guava purees. The antioxidant capacity of the DP-

C02 and pasteurized sample showed significant differences (a = 0.05) during the first 2

weeks of storage when compared to control. After week 4, there were no significant

differences for antioxidant capacity of the control, DP-C02 and thermal treated guava

purees. These differences in antioxidant capacity are not well understood. There were

no significant differences during the first 4 week of storage but during week 6, a

significant increase in ORAC value was observed for the DP-C02 and remained

constant until week 10. After week 10, a significant decrease was observed and the

ORAC value reached the initial value. Untreated (control) guava puree showed no

significant differences in ORAC values during the first 5 weeks of storage. A maximum

ORAC value was obtained during week 8 of storage, and after week 10, a significant

decrease was observed. The changes in antioxidant capacity for the three samples may

be the result of microbial growth. During microbial growth, compounds with antioxidant

capacity may be released from the pulp of the guava increasing the antioxidant capacity

of the puree. The observed decrease may be related to the interaction of these

compounds (such as polymerization reactions) affecting radical-scavenging properties.

Thermally treated guava puree showed a significantly lower content of total soluble

phenolics when compared to control and DP-C02 treated guava purees. DP-C02

treatment caused a significant increase in total phenolic content compared to the control


129









and thermal treated guava purees during the first 2 weeks of storage. This increase

following DP-C02 processing may be explained by several factors, including the

conditions of the assay itself. Due to the broad range of compounds detected by the

assay in the guava puree, slight changes in these compounds following DP-C02 can

easily influence the ability to reduce metal ions in solution, thus affecting the assay.

Between weeks 2 and 10, there was no significant difference in total soluble phenolics

when compared to control. After week 10 of storage at 4 C, DP-C02 showed a higher

value when compared to control and thermal treated guava purees. Similar results were

observed by Ferrentino and others (2009) when red grapefruit juice was treated with

DP-C02. Furthermore, Del Pozo Insfran and others (2006) found similar results after

processing and storage of muscadine grape juice with a continuous DP-C02 system.

Control puree showed no significant differences in the total soluble phenolic content

during the first 6 weeks of storage after which a significant increase was observed. After

week 10 of storage, there was a significant decrease in total soluble solids. Even though

the thermal treatment initially decreased (by 9%) the total phenolic content compared to

the control, a significant increase (4%) for the initial content was observed during the

first 2 weeks of storage. This increase may be explained by several factors, including

the fact that certain compounds such as ascorbic acid, certain soluble proteins,

melonoidins, and reducing sugars give a measurable interference in the assay.

Additionally, the guava puree contained intact plant cells whereby the conditions of

heating helped to release cell-wall or vacuole-bound compounds with metal reducing

capabilities. Any of these compounds present or their effect on the assay during storage

could account for the increase. Polyphenolic degradation for control, DP-C02 and


130









thermal treatment during storage may be related to the loss of ascorbic acid.

Additionally, the decrease in total soluble solids observed in thermal processed guava

puree after 6 weeks of storage may be due to formation of by-products from

carbohydrate and organic acids degradation. During thermal processing and storage,

furfurals and other carbonyl compounds can form condensation products with these

types of compounds (Es-Safi and others 2002). The results suggest that DP-C02

treatment does not affect the antioxidant capacity and total soluble phenolics of guava

puree, supporting the idea that this process may have industrial application in producing

fresh like juices with good antioxidant capacity.

Titratable Acidity (TA)

TA of samples was expressed as percent citric acid (w/w) equivalents. There was

no significant (a = 0.05) difference between DP-C02, thermal treated and control

samples for TA values regardless of storage time (Figure 5-1) during the first six week

of storage. During week 8 of storage, the control and thermal treated samples showed a

significant increase in TA, which remained constant until the end of the study. The TA

values were higher for the DP-C02 treated samples compared to the control and

thermal treated samples. The increase in TA for the three samples may be related to

microbial growth. During microbial growth, microorganisms metabolize pectin and other

carbohydrates releasing acids, which increase TA. The same trend was observed for

TA by Kincal and others (2006) when orange juice was treated with DP-C02. Similar

results were also obtained by Damar and others (2009) and Ferrentino and others

(2009).









Ascorbic Acid (Vitamin C) Content

There were slight differences in the ascorbic acid content of DP-C02 treated

guava puree regardless of storage time. Similar results were observed by Del Pozo -

Isfran and others (2006) after processing muscadine grape juice with DP-C02.

Ferrentino and others (2009) found slight differences for the ascorbic acid content in red

grapefruit juice regardless of storage time. The ascorbic acid content for the three

samples (Figure 5-2) decreased over time regardless of the treatment. The ascorbic

acid content for the DP-C02 treated sample was significantly higher than thermal

treated guava puree during the first 12 weeks of storage after which there were no

significant differences between control, DP-C02 and thermal treated guava purees.

Thermal treatment decreased the ascorbic acid content of the puree, 21% of ascorbic

acid was loss after pasteurization of guava puree. Factors that influence the oxidation of

ascorbic acid include pH, oxygen, water activity, and the presence of certain metal ions

such as iron and copper, and it is generally accelerated with light and elevated

temperatures (Fennema 2000). This indicated that the conditions of processing create

an environment suitable for ascorbic acid oxidation or degradation. DP-C02 treatment

delayed the oxidation or degradation of ascorbic acid probably by oxygen exclusion,

and the ascorbic acid content was similar if not slightly better than the control puree.

Color

Color for control, DP-C02 and thermal treated guava purees change drastically

during storage at 40C. As observed in Table 5-2, an increase in L*, a* and b* value were

observed initially for the thermal treated guava puree, probably due to the loss of green

color caused by degradation of pigments (such as chlorophyll). The same results were

observed in pasteurized guava puree from Taiwan (Yen and Lin 1996). DP-C02 and


132









thermal treatments increased L* values, increasing lightness of the guava puree. L*

values for thermal treated guava puree was not significantly affected while control and

DP-C02 samples showed a significant decrease during the 14 weeks of storage. There

was a strong correlation between ascorbic acid content and lightness for control (r2=

0.86), DP-C02 (r2=0.81) and pasteurized samples (r2=0.86). These results indicate the

darkening in color was due to browning reactions. Little change in a* values for the

three samples was noted, so the variability may be due to fluctuations between bottles.

In general, thermal treatment significantly increased a* values during the first 6 weeks of

storage while DP-C02 showed a decrease during the first 2 weeks of storage, when

compared to the control. No significant differences were found in the a* values between

4 and 6 weeks of storage for control and DP-C02 treated samples. After 6 weeks of

storage, a significant decrease in a* values was observed for the control. Furthermore,

DP-C02 and thermal treated samples showed no significant differences in a* values but

a progressive reduction in the samples redness was observed. DP-C02 and thermal

treatments caused a significant increase in b* values or increase in yellowness when

compared to control. Thermal treated puree b* values were not affected during the 14

weeks of storage. Control and DP-C02 samples showed a progressive increase in b*

value during storage. Changes in a* and b* values can be attributed to changes in

carotenoids composition and there was a change in color from less red and more

yellow.

Ferrentino and others (2009) found that DP-C02 caused an increase in lightness

and redness while yellowness of red grapefruit juice was lowered. Kincal and others

(2006) showed that both the L* and the b* values for orange juice increased after DP-


133









C02 treatment, while the a* value generally decreased. Thermal pasteurization changed

all 3 color parameters, causing an increase in lightness and yellowness and a slight

decrease in redness. Similar results were reported by Del Pozo-lsfran and others

(2006).

Organic Acid Content

Oxalic, citric and malic acids were individually identified and quantified using

HPLC (Table 5-3). Figure 5-3 shows a typical chromatogram obtained from the organic

acid analysis. There were significant differences in oxalic, citric and malic acid content

during storage regardless of the treatment. Fluctuations in citric acid content were

observed but these may be attributed to the bottle to bottle variation. There were no

significant differences in malic acid content between control, DP-C02 and thermal

treated guava purees during storage. There were no significant differences in oxalic

acid content between control and DP-C02 treated guava purees during the 14 weeks of

storage at 4 oC.

Phenolic Compounds

Various solvent extraction and fractionation procedures on guava puree were

attempted for polyphenolic detection by HPLC analysis. Enzyme treatment of guava

puree was required to provide the most reproducible HPLC chromatograms with

maximal peak separation (Figure 5-5 and 5-6). In this study, HPLC analysis of

polyphenolics was used to identity overall data trends as affected by processing

treatment and storage conditions. Among the polyphenolic compounds present, 6 were

selected based on identification and adequacy for treatment differentiation (Table 5-4).

Peaks were identified based on their spectroscopic properties and comparison to

authentic standards (Table 5-4). Few studies exist that identify and quantify the


134









polyphenolics present in guava, but those that have include ellagic acid and condensed

tannins (Misra and Seshadri 1968; Fender 2005; Nunez-Rueda, 2005), and gallic acid,

catechin, epicatechin, and chlorogenic acid (Kondo and others 2005; Fender, 2005;

Nunez-Rueda, 2005).

Ellagic acid was isolated from the ethyl acetate and methanol fractions (Figures 5-

4 and 5-5). Gallic acid, hydrobenzoic acid, cinnamic acid and an unknown were isolated

from the ethyl acetate fraction while an ellagic acid derivate was identified from the

methanol fraction. Gallic acid (Peak 1), ellagic acid (Peak 3) and cinnamic acid (Peak 6)

were clearly identified by comparison to standards. Ellagic acid derivative (Peak 5) was

tentatively identified by comparison to ellagic acid spectral properties. Hydrobenzoic

acid (Peak 4) was tentatively identified, as it shared similar spectral characteristics with

those types of compounds. Only one compound (unknown) was characterized based on

retention time and spectroscopic properties but was dissimilar to any known

polyphenolic compounds. Further work will be needed to isolate and identify this

individual polyphenolic compound in guava puree.

Gallic acid (GA) (Peak 1) has been reported as an effective antioxidant due to its

structure and positioning of hydroxyl groups. DP-CO2 and thermal treatment

significantly increased the initial concentration of GA (Table 5-5); however GA content

decreased during the first 2 weeks of storage and presented no significant difference

from the control puree. Between the second and sixth week of storage, a significant

increase was observed for the control, DP-CO2 and thermally treated samples, after

which a significant decrease throughout the remaining storage was observed.


135









Unknown (Peak 2) Quantification of the unknown compound was performed

using ellagic acid. Thermal treatment significantly increased the concentration of the

compound when compared to DP-C02 and control. DP-C02 significantly increased the

concentration during the first two weeks of storage after which it remained constant until

week 6. Thermal treated samples showed no significant differences due to storage time.

Ellagic acid (Peak 3) DP-C02 treatment decreased the initial concentration of

ellagic acid (EA) (Peak 2) when compared to the control. The EA concentration (Table

5-5) decreased during the first 2 weeks of storage after which no significant differences

were observed for the DP-C02 treated samples.

Hydrobenzoic acid (Peak 4) Hydrobenzoic acid (HBA) was quantified and

reported as gallic acid equivalents. DP-C02 significantly increased initial concentration

of HBA. During the first two weeks of storage, DP-C02treated samples showed a

significant increase in HBA and then progressively decreased.

Ellagic acid derivate (Peak 5) Ellagic acid (EA) derivative eluted in the HPLC

column immediately after EA. Due to its closeness to EA in spectral properties, this

compound was tentatively classified as EA derivative (Table 5-6). There was a

significant effect due to DP-C02 treatment as compared to the control and thermal

treated samples. During storage, control and thermal treated guava puree samples

showed a significant increase during the first two weeks of storage after which no

significant differences were observed.

Cinnamic acid (Peak 6) DP-C02 treatment showed a significantly higher

concentration of cinnamic acid (CA), when compared to control. During the first two

weeks of storage, DP-C02 treated guava puree showed a significant increase in CA


136









concentration, after which it decreased. A progressive and significant increase in CA

concentration was observed for control samples over the 14 weeks of storage.

In general, DP-C02 and thermal treatment caused an initial increase in the total

content of the identified phenolic compounds of 33 and 10% respectively. Even when

there were weekly variations in the individual content of these compounds, the total

content decreased for DP-C02 and thermal treatment during the 14 weeks of storage.

The total loss for DP-C02 and thermal process was 19% and 30%, respectively from the

initial concentration. Ellagic acid, its derivative and the unknown (quantified as ellagic

acid) were the major contributors to the total phenolic content quantified by HPLC,

followed by hydrobenzoic acid.

Conclusions

From the storage study, it is evident that the DP-C02 treatment can extend shelf

life and maintain the physical and quality attributes of fresh guava puree. DP-C02

served to protect polyphenolic and antioxidant levels throughout processing and storage

without compromising physicochemical and phytochemical properties of the guava

puree. DP-C02 delayed the degradation of vitamin C content during storage, allowing

the vitamin C content to be higher than the fresh puree. DP-C02 treated guava puree

retained organic acids contents similar to fresh guava puree. Even when DP-C02

caused an initial increase in phenolic compounds of 33%, the reduction of 19% for

phenolic compounds during storage conditions was still lower than the thermal treated

sample (30%).









Table 5-1. Antioxidant capacity and total soluble phenolics of fresh, DP-C02 and
thermal treated guava puree during storage

Total Soluble Phenolics
ORAC (pmol TE/mL) (mg of GAE/ L of sample)

Week Control DP-C02 Pasteurized Control DP-C02 Pasteurized
12.2 + 12.3 + 182.2 + 196.5 +
0 0.4 0.5 11.7 + 0.3 0.9 0.8 165.6 + 0.6
12.4 + 11.1 + 182.8+ 188.5 +
2 0.4 0.4 11.7 +0.4 0.8 0.8 172.1 +0.7
12.1 + 11.2+ 182.2+ 184.8+
4 0.4 0.6 11.4+0.6 1.4 0.7 176.6+0.1
15.4 + 15.8 + 184.1 + 181.2+
6 0.7 0.6 16.1 +0.0 0.7 0.9 178.2 +0.2
16.1 + 17.0 + 186.4 + 184.9 +
8 0.0 0.0 16.2 + 0.1 0.2 0.6 177.0 + 0.6
15.0 + 13.0 + 185.7 + 184.3 +
10 0.3 0.8 12.8 + 0.9 0.7 1.1 175.0 + 0.4
12.8 + 12.9 + 179.6 + 182.9 +
12 0.2 0.8 13.3 +0.3 1.0 0.1 177.0 + 0.3
14.3 + 13.8 + 174.5 + 183.0 +
14 0.5 0.9 14.5 +0.2 1.0 0.5 175.2 + 0.4
Mean + standard deviation for n=3


138













0.65 -



S0.6 -
C,) I
0)
1 -

0 0.55

0
0.55

-

* 0.5 -


E



u-







0.3
-Q


..... .... .* --














Control
** DP-C02
A Pasteurized


5
Storage time (weeks)


Figure 5-1. Titratable acidity of control, DP-C02 and thermal treated guava purees
during 14 weeks of refrigerated storage. Error bars represent standard
deviation for n=3.


139

















16


14
*. ....
12
r T ****** ..........
E 10 **.


E 8





o 4


2-
*2 Control

0
0 2 4 6 8 10 12 14
Storage time (weeks)

Figure 5-2. Ascorbic acid (vitamin C) content of control, DP-C02 and thermal treated
guava purees during 14 weeks of refrigerated storage. Error bars represent
standard deviation for n=3.


140










Table 5-2. L*, a* and b* values for control, DP-C02 and thermal treated guava purees during 14 weeks of refrigerated
storage
Storage L* value a* value b* value
Time
(wks) Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized
45.75 + 47.69 + 48.57 + 12.74 + 12.72 + 13.26 + 10.34 + 12.06 + 13.65 +
0 0.03 0.30 0.06 0.07 0.48 0.23 0.04 0.19 0.20
45.39 + 47.00 + 48.68 + 12.81 + 12.46 + 13.53 + 10.79 + 12.41 + 13.99 +
2 0.08 0.04 0.13 0.15 0.20 0.12 0.08 0.29 0.03
45.25 + 46.18 + 48.23 + 13.15 + 13.17 + 13.42 + 11.80 + 12.80 + 14.51 +
4 0.02 0.18 0.02 0.06 0.09 0.09 0.06 0.05 0.26
45.19 + 46.16 + 48.42 + 13.22 + 13.39 + 13.55 + 12.03 + 13.03 + 14.76 +
6 0.07 0.13 0.15 0.08 0.11 0.03 0.06 0.08 0.08
44.83 + 46.00 + 47.84 + 12.72 + 13.31 + 13.13 + 11.97 + 13.20 + 14.59 +
8 0.02 0.14 0.04 0.14 0.11 0.04 0.12 0.09 0.05
43.86 + 45.58 + 47.99 + 12.09 + 12.65 + 12.79 + 11.87 + 12.94 + 14.42 +
10 0.13 0.19 0.04 0.19 0.13 0.16 0.07 0.13 0.12
43.98+ 45.62+ 48.05+ 12.11 + 12.75+ 12.95+ 11.86+ 13.02+ 14.61 +
12 0.21 0.30 0.17 0.41 0.09 0.06 0.14 0.26 0.11
43.92 + 45.60 + 47.88 + 12.19 + 12.67 + 12.81 + 11.98 + 13.16 + 14.37 +
14 0.20 0.01 0.31 0.05 0.25 0.38 0.02 0.12 0.35
Mean + standard deviation for n=3










Table 5-3. Oxalic acid (OA), malic acid (MA), and citric acid (CA) content of control, DP-C02 and thermal treated guava
purees during 14 weeks of refrigerated storage at 40C
Storage
time Oxalic Acid (mg acid/100 g sample) Malic Acid (mg acid/100 g sample) Citric Acid (mg acid/100 g sample)
(Week) Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized
0.03 + 0.038 + 0.04 + 0.09 + 0.09 + 0.09 + 0.52 + 0.52 + 0.52 +
0 0.000 .000 0.000 0.001 0.001 0.000 0.001 0.001 0.001
0.02 + 0.024 + 0.02 + 0.08 + 0.08 + 0.08 + 0.32 + 0.36 + 0.37 +
2 0.002 .001 0.000 0.003 0.002 0.002 0.031 0.003 0.011
0.02 + 0.018 + 0.01 + 0.07 + 0.07 + 0.06 + 0.29 + 0.26 + 0.25 +
4 0.002 .001 0.001 0.007 0.000 0.005 0.004 0.031 0.027
0.02 + 0.021 + 0.02 + 0.07 + 0.08 + 0.08 + 0.39 + 0.38 + 0.39 +
6 0.000 .000 0.000 0.005 0.008 0.000 0.000 0.004 0.004
0.02 + 0.019 + 0.02 + 0.10 + 0.13 + 0.12 + 0.32 + 0.33 + 0.37 +
8 0.001 .001 0.002 0.003 0.002 0.002 0.021 0.014 0.007
0.03 + 0.030 + 0.03 + 0.12 + 0.12 + 0.12 + 0.40 + 0.43 + 0.41 +
10 0.003 .001 0.001 0.001 0.001 0.001 0.008 0.012 0.005
0.04 + 0.038 + 0.04 + 0.18 + 0.13 + 0.12 + 0..40 + 0.43 + 0.42 +
12 0.000 .003 0.001 0.001 0.001 0.004 0.005 0.004 0.007
0.05 + 0.049 + 0.05 + 0.08 + 0.08 + 0.08 + 0.39 + 0.38 + 0.39 +
14 0.001 .000 0.000 0.001 0.001 0.002 0.004 0.004 0.002
Mean + standard deviation for n=3


142





















0.0 2.5 S 7.5 IQO 12.5 150 17.5 2S 225 25

Figure 5-4. HPLC chromatogram of organic acids found in guava puree-1) oxalic acid, 2) malic acid and 3) citric acid.
Identification (210 nm) was done by comparison to authentic standards and spectral properties

70 mA
4






0. 5. 10. 15. 20. 25. 30. 35. 40. 45. 52.
Figure 5-5. HPLC chromatogram of polyphenolic compounds found in ethyl acetate fraction in guava puree-1) gallic acid,
2) unknown 3) ellagic acid, 4) hydrobenzoic acid and 6) cinnamic acid. Identification (260 and 280 nm) was
done by comparison to authentic standards and spectral properties


143












25


12



30. 32. 34. 36. 38. 40. 42. 44. 46. 48. 50. 52.

Figure 5-6. HPLC chromatogram of polyphenolic compounds found in the methanol fraction of guava puree- 3) ellagic
acid and 5) ellagic acid derivative. Identification (260 nm) was done by comparison to authentic standards and
spectral properties









Table 5-4. Identification
retention time


of guava polyphenolics at 260 and 280 nm by HPLC based on
spectral properties and comparison to s


145


Peak Retention Spectral
No. time (min) Fraction Properties Compound
Ethyl
1 12.95 Acetate 216, 272 Gallic Acid
Ethyl
2 30.15 Acetate 264.2 Unknown
Ethyl
Acetate
and
3 37.56 Methanol 255, 367 Ellagic acid
Ethyl
4 39.00 Acetate 199, 254 Hydroxybenzoic acid1
5 41.75 Methanol 255, 367 Ellagic acid derivate1
Ethyl
6 42.17 Acetate 204, 217, 278 Cinnamic acid
1 = Tentative identification










Table 5-5. Gallic acid (GA), unknown and ellagic acid (EA) for control, DP-C02 and thermal treated guava purees during
14 weeks of refrigerated storage at 40C
Storage Gallic Acid (GA) Unknown Ellagic Acid (EA)
Storage
time (mg/L) (mg/L EAE1) (mg/L)
(weeks) Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized
0 0.63 +0.03 7.42 +0.05 6.55 +0.04 1.61 +0.01 1.96 +0.06 2.19 +0.03 5.93 +0.02 4.77 +0.9 5.89 +0.06
2 1.89 + 0.02 1.88 + 0.06 1.23 + 0.01 2.20 + 0.03 2.24 + 0.06 2.22 + 0.03 5.00 + 0.08 4.42 + 0.36 5.00 + 0.27
6 14.13 + 0.05 10.34 + 0.04 9.04 + 0.05 2.54 + 0.03 2.25 + 0.02 2.17 + 0.09 5.42 + 0.27 4.30 + 0.05 4.92 + 0.17
10 2.29+0.05 2.90+0.10 1.94+0.08 2.30+0.06 2.41 +0.01 2.08+0.14 4.46+0.28 4.18+0.37 4.42+0.30
14 2.27 + 0.07 1.58 + 0.02 1.83 + 0.04 2.39 + 0.00 2.02 + 0.05 2.03 + 0.08 4.54 + 0.27 3.94 + 0.35 3.96 + 0.07
Mean + standard deviation for n=3: 1= Ellaaic acid equivalents


Table 5-6. Hydrobenzoic acid (GA), cinnamic acid (CA) and ellagic acid derivative (EAD) for control, DP-C02 and thermal
treated guava purees during 14 weeks of refrigerated storage at 40C
Storage Hydrobenzoic Acid (HBA) Cinnamic Acid (CA) Ellagic Acid Derivate (EAD)
Storage I G 2
time (mg/L GAE1) (mg/L EAE2) (mg/L EAE2)
(weeks) Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized Control DP-CO2 Pasteurized
0 5.39+0.02 7.21 +0.61 9.19+0.49 1.46+0.02 1.50+0.01 1.47+0.01 1.01 +0.00 1.21 +0.04 1.64+0.01
2 9.22 + 0.26 9.66 + 0.57 9.84 + 0.17 1.49 + 0.00 1.49 + 0.00 1.53 + 0.01 1.51 +0.00 1.68 + 0.01 1.76 + 0.02
6 10.23 + 0.26 8.67 + 0.21 8.75 + 0.22 1.50 + 0.00 1.46 + 0.01 1.55 + 0.02 1.87 + 0.03 1.73 + 0.01 1.77 + 0.02
10 9.50 + 0.79 10.45 + 0.06 9.06 + 0.25 1.50 + 0.01 1.49 + 0.01 1.53 + 0.00 2.04 + 0.01 1.80 + 0.01 1.80 + 0.03
14 13.98+0.23 9.08 +0.73 9.44+1.07 1.49 +0.01 1.55 +0.01 1.53+0.00 2.64+0.01 2.10+0.04 2.00+0.03
Mean + standard deviation for n=3; 1= Gallic acid equivalents, 2=Ellagic acid equivalents


146









CHAPTER 6
CHANGES IN AROMA COMPOUNDS AND SENSORY PERCEPTION IN GUAVA
PUREE AFTER THERMAL AND NON-THERMAL PROCESSING

Abstract

The volatiles present in freshly thawed (FT), dense phase carbon dioxide (DP-

CO2) treated and pasteurized guava puree were isolated by Solid Phase

Microextraction (SPME). Isolates analyzed by Gas Chromatography (GC-MS), Gas

Chromatography-olafactometry (GC-O) and Gas chromatography coupled to a pulse

flame photometric detector (GC-PFPD) contained 76 compounds. Analysis with GC-MS

showed 58 compounds which were classified in 6 groups: aldehydes (6), acids (2),

alcohols (15), ketones (6), esters (21) and terpenes (8). Eleven compounds were

identified for the first time in guava puree: butanal, isoamyl acetate, 4-mercapto-4-

methyl-penta-2-one, phenyl acetaldehyde, nonanal, homofuraneol, methyl nonanoate,

1-p-menthene-8-thiol, 2-octane, ethyl-3-hydroxyhexanoate and ethyl nonanoate.

Hexanal was the most abundant compound in the three guava puree samples. Analysis

by GC-O showed 26 compounds responsible for the guava puree aroma, three of which

were tentatively identified: benzyl alcohol, phenyl acetaldehyde and homofuraneol. (Z)-

3-hexenyl hexanoate was the major contributor to the aroma of the fresh and DP-CO2

treated guava puree. Eleven sulfur compounds were identified by GC-PFPD in the

guava puree. Four of these compounds were identified for first time in guava puree, and

two, hydrogen sulfide and hydrogen disulfide, were previously reported in guava leaves.

Introduction

Guava (Psidium guajava L.), is a tropical fruit characterized by its appealing flavor

and aroma. It has been catalogued as one of the most nutritious fruits due to its high

content of phytochemicals, especially ascorbic acid (United States Department of









Agriculture [USDA], 2005). Because of the high perishability and limited availability of

fresh guava, most fruit destined for US markets is processed into juice, puree, jams,

jellies, and syrup. Guava puree is normally processed by heat pasteurization to extend

the shelf life up to one year and inactivate pectinesterase, but the fresh taste is modified

by thermally accelerated reactions.

The first step in the characterization of odor and flavor (volatile) compounds in a

complex mixture or food system is to separate them from nonvolatile matrix

interference. This separation is accomplished through a variety of techniques, such as

solvent extraction, head-space concentration, and distillation. Extraction procedures

may distort or alter the chemical composition. Since each method will enhance the

concentration of certain compounds and minimize others, there is no perfect extraction

system. Each technique yields a concentrated essence containing the odor-active

chemicals. Considering that it is simple, solventless and rapid, Solid Phase

Microextraction (SPME) was used as a sample extraction technique in this study.

The volatile composition of guava constituents has been previously studied and

extraction techniques and detection methods vary (Stevens and others 1970; Wilson

and Shaw 1978; Macleod and Gonzalez de Troconis 1982; Idstein and Schreier 1985;

Chyau and others 1992). Several studies have been conducted to determine the effect

of processing on the volatile composition of guava puree. Yen and others (1992) studied

the changes in flavor components of guava puree resulting from pasteurization and

frozen storage. Chan and Cavaletto (1982) studied the changes in chemical and

sensory quality during processing and storage of aseptically packaged guava puree.

Yen and Lin (1996) studied the changes in volatile flavor components of guava juice


148









with high pressure and heat processing. Dense Phase Carbon Dioxide (DP-C02)

treatment has not been tested for its potential on retaining volatile compounds

compared to traditional thermal treatments for guava puree. The objectives of this study

were (1) to determine the volatile composition of the puree as affected by DP-C02 and

thermal processing and (2) Compare the sensory perception of DP-C02 and thermal

processing with a freshly thawed guava puree.

Materials and Methods

Guava Puree

Frozen unpasteurized red guava puree was obtained from the Goya Company

(Dominican Republic). The puree was held at -20 oC and thawed at 4 oC for one week

prior to processing. Part of the insoluble solids and stone cells were removed by

straining the thawed puree through a 200 pm nylon filter (Cole Palmer, Vernon Hills, IL,

U.S.A). DP-C02 treatment was performed using freshly thawed (FT) puree, which was

treated at 34.5 MPa, 8% CO2 and 6.9 min residence time. Volatile compounds were

compared to that of thermally treated (90 oC for 60 sec) and freshly thawed guava puree

(control). Pasteurization was performed using a Microthermic lab scale pasteurizer (see

Chapter 4: Processing equipment from the Material and Methods section).

Sensory Evaluation

Flavor and overall likeability of FT and processed (DP-C02 and pasteurized)

guava puree were compared using a difference from control test at the beginning of the

storage study. A randomized complete block design was used, and differences from

control values were recorded on a line scale from 0 to 10. "No differences" (0) and

"extremely different" (10) were used as the extreme anchors of the flavor line. Panelists

compared the flavor of the reference (FT guava puree = control) with that of a hidden


149









reference (control), the thermal (HTST), and DP-CO2 treated guava purees. A 9-point

hedonic scale was also conducted in order to compare the overall likeability of FT

(hidden reference) and processed guava purees. An example of the questionary used

for the taste panel is shown in Appendix D.

Before sensory analysis all DP-CO2 treated purees were degassed in order to

have equal carbonation levels. All samples were chilled and kept in ice at a

temperature of ~4 C before serving. They were then served on a tray in numbered

plastic cups containing ~30 mL of sample. A cup of deionized water and non salted

crackers were also provided to the panelists to cleanse their palate between

evaluations. Sensory tests were performed at the University of Florida taste panel

facility using 75 untrained panelists in each test.

Statistical Analysis

Sensory data was recorded and analyzed using Compusense five (Compusense,

Guelph, Ontario, Canada). Analysis of variance (ANOVA) and mean comparisons using

t-test and Tukey's test were conducted at the 5% significance level. The statistical

analysis was conducted using the same program used to record the data.

Extraction of Volatile Compounds using Headspace SPME

Headspace volatiles were extracted and concentrated using SPME. Ten milliliter

aliquots of guava puree (freshly thawed, DP-CO2 and pasteurized) were poured into 40

mL screw cap amber glass vials and each sealed with caps containing Teflon-coated

septa. Volatiles were subsequently extracted using a pre-conditioned 1 cm 50/30 pm

Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/CAR/PDMS) fiber (Supelco,

Bellafonte, PA, U.S.A.) for 30 min at 40C. Guava purees were allowed to equilibrate for

20 min prior to SPME. The full length of the coated fiber was exposed to the headspace


150









of the samples and after 30 min, the fiber was removed from the headspace and

immediately inserted into a GC-splitless injector, where aroma compounds were

allowed to be desorbed for 5 min. Before each volatile compound extraction, the fiber

was cleaned for 5 min in the injection port (200 C) of the GC-O, GC-S or GC-MS

instruments. Adsorbed volatiles were desorbed in the injector port of a GC. Separation

and analysis was conducted as follows.

Gas Chromatography- Olfactometry Analysis (GC-O)

GC-O analysis was carried out using a HP-5890A GC (Palo Alto, CA) with a flame

ionization detector (FID) and a sniffing port. A DB-wax column (30 m x 0.32 mm i.d. x

0.25 pm film thickness) (J&W Scientific, Folsom, CA, U.S.A.) and a ZB-5ms column (30

m x 0.32 mm i.d. x 0.5 pm film thickness) (Zebron ZB-5, Phenomenex, Torrance, CA,

U.S.A.) were used during the analysis. Initial oven temperature was 40 C (no hold) and

temperature was increased at 7 C/min until reaching a final temperature of 240 C (for

DB-Wax) or 265 C (for DB-5) and holding at this temperature for 5 min. Guava puree

volatiles from the SPME fiber were desorbed in the GC injection port. A SPME injector

liner (SPME injection sleeve, 0.75 mm i.d., Supleco; Bellefonte, PA., U.S.A.) was used.

The GC column effluent was split between the FID and the olfactometer. Helium was

used as the carrier gas at a flow rate of 1.55 mL/min. The injector temperature was 200

C, and the detector temperature was 250 C. Two olfactory assessors were employed.

Samples were sniffed in duplicate by each assessor in each column.

Aroma descriptions and approximate times were recorded for every sample.

Assessors indicated the intensity of each aroma peak using a linear potentiometer with

a 0-1 V signal. Aromagrams and FID chromatograms were recorded and integrated

using Chrom- Perfect 5.5.5 (Justice Labs, Melbourne, FL, U.S.A.) data acquisition









software. A peak was considered aroma active only if at least half the sniffers found it at

the same time with a similar description. Linear retention index values were determined

for both columns using a series of alkanes (C5-C25) run under identical conditions.

Gas Chromatography- Mass Spectrometry Analysis (GC-MS)

Mass Spectrometry (GC-MS) was used to identify the odor-active volatiles

detected in the GC-O experiments. GC-MS data were collected using a Perkin Elmer

Clarus 500 quadrupole mass spectrometer equipped with Turbo Mass software (Perkin-

Elmer, Shelton, CT). Helium was used as the carrier gas in the constant flow mode of 2

mL/min. Guava puree volatiles from the SPME fiber were desorbed in the GC injection

port (splitless mode) at 200 C. The fiber was removed after 5 min exposure in the

injection port. Gas chromatography separation was performed using a DB-wax column

(30 m x 0.32 mm i.d. x 0.25 pm film thickness) from J&W Scientific (Folsom, CA, U.S.A.)

and ZB-5ms column (30 m x 0.32 mm i.d. x 0.5 pm film thickness) from Phenomenex

(Zebron ZB-5, Phenomenex, Torrance, CA, U.S.A.). Initial oven temperature was 400C

(no hold), and temperature was increased at 7 C/min until reaching a final temperature

of 240 C (for DB-Wax) or 265 C (for DB-5) and holding at this temperature for 5 min.

The mass spectrometer detector had a delay of 0.5 min, and scans were made from

m/z 25 to 300. A mass spectrum scan was performed every 0.2 seconds. The electron

ionization was carried out in the positive mode at 70 eV. Mass spectral matches were

made by comparison with NIST 05 (Scientific Instrument Services, Ringoes, NJ, U.S.A.)

and Wiley Registry of Mass Spectral data 6th edition (John Wiley and Sons Inc) mass

spectral libraries. Only those compounds with spectral fit values equal to or greater than

850 were considered positive identifications. Linear retention index values were


152









determined for both columns using a series of alkanes (C5-C25) run under identical

conditions.

Sulfur Compounds Identification using Gas Chromatography-Pulse Flame
Photometric Detector (GC-PFPD)

Volatile sulfur compounds from the fiber were separated using a HP-5890 Series II

GC from Agilent (Santa Clara, CA, U.S.A.) equipped with a 5380 PFPD detector from 01

Analytical (College Station, TX, U.S.A.). The compounds from the SPME fiber were

desorbed for 5 min in the GC injection port (splitless mode) at 200 oC. Separation of

compounds was achieved on DB-wax column (30 m x 0.32 mm i.d. x 0.25 pm film

thickness) (J&W Scientific, Folsom, CA, U.S.A.) and a ZB-5ms column (30 m x 0.32 mm

i.d. x 0.5 pm film thickness) from Phenomenex (Zebron ZB-5, Phenomenex, Torrance,

CA, U.S.A.). GC oven temperature was initially set at 40 C and then ramped at 7

oC/min to 240 oC (DB-Wax) or to 265 oC (ZB-5). The final temperature was held for 5

min in both columns. GC was operated in a constant flow mode (2 mL/min) with helium

as the carrier gas. PFPD detector was set at 250 oC and employed WinPulse32 Version

2.0 software. The PMT voltage was set at 525 V, and the sulfur gate was opened

between 5 and 24 ms. PFPD output was recorded in the square root mode.

Chromatograms were recorded and integrated using Chrom- Perfect 5.5.5 (Justice

Labs, Melbourne, FL, U.S.A.) data acquisition software

Identification Procedures

Identifications were based on the combined matching of retention indices (LRI

values) from DB-5 and DB-Wax columns, spectral matches from the NIST 05 (Scientific

Instrument Services, Ringoes, NJ, U.S.A.) and Wiley Registry of Mass Spectral data 6th

edition (John Wiley and Sons Inc) libraries, aroma descriptors, and linear retention


153









index matches from databases (Jennings and Shibamoto 1980; Flavor database 2009;

Flavor net and human odor space 2009).

Results and Discussion

Sensory Analysis

One taste panel was conducted during storage. Results showed that 51% of

panelists were females and 49% were males, 84% of males and 83% of females were

in the 18-24 age range.

Results showed (Table 6-1) that there were no significant differences (a=0.5%)

between the reference and pasteurized guava purees. However, significant differences

were detected by panelists between the reference and DP-CO2 treated guava purees.

The difference between the reference and the hidden reference was 3.05. Some

differences are expected since this was more of a consumer testing, and no training

was involved. The difference between the reference and the pasteurized sample was

3.56. This indicated that the panelists were not able to differentiate the reference and

the pasteurized sample. Although, the ranking for overall likeability for the three tested

purees were not significantly different (a= 0.05), which indicated that regardless of

treatment, panelist preferences remained the same (Table 6-1). The overall likeability

for hidden reference was 4.29, for DP-CO2 was 4.13 and for pasteurized puree was

3.79. Due to the closeness of the overall likability ranking for the hidden reference and

DP-CO2 guava purees, results indicate that panelist preferences were toward reference

and DP-CO2. Even though the pasteurized puree was not different from DP-CO2, the

ranking value was lower. Previous studies on muscadine grape juice and coconut water

showed bigger differences in flavor and overall likeability between the DP-CO2 and

thermal treated samples. They also found that the DP-CO2 sample was very similar to


154









the reference (Damar, 2006; Del Pozo-lnsfran and others 2006). This difference may

indicate that thermal processing affected more the organoleptic characteristics of the

grape juice and coconut water than in guava puree (probably the nature of the sample

has an effect on the treatment). DP-C02 differences from the other two samples can be

attributed to the fact that even when the DP-C02 puree was partially degasified before

sensory analysis was conducted, there could still be residual C02 remaining. This would

result in a carbonated puree mouth feel which may have also affected the

acidity/sweetness balance causing the panelists to perceive the beverage as less

sweet. This can be confirmed by a higher TA in the DP-C02 beverage as described

previously. Another reason for the differences may be attributed to the separation of

DP-C02 samples into two phases. Due to problems with the pasteurization unit, FT and

DP-C02 samples were frozen for three weeks to lock the processing date. After the

pasteurization process was carried out, samples were also frozen. Three days before

the taste panel, samples were thawed at 4 C. During this time, DP-C02 samples

separate into two phases.

Flavor Analysis

SPME technique was used for the isolation of the volatile compounds present in

the guava purees. This technique employs a fused-silica fiber coated with an

appropriate stationary phase, allowing the analyte in the sample to be directly extracted

(Kataoka and others 2000) by the absorption to the fiber followed by desorption in the

injector port of a gas chromatograph. Subsequently, separation, detection, identification,

and characterization of individual compounds are possible with sophisticated

instrumentation such as GC-MS, GC-S and GC-O.


155









The SPME extraction conditions were as previously described (Paniandy and

others 2000). Although SPME has a maximum sensitivity at the partition equilibrium, a

proportional relationship is obtained between the amount of analyte adsorbed by the

SPME fiber and its initial concentration in the sample matrix before reaching partition

equilibrium. Therefore, full equilibration is not necessary for quantitative analysis by

SPME (Kataoka and others 2000).

Guava Volatile Composition

A total of 76 compounds were identified using GC-MS, GC-O and/or GC-PFPD

using different columns. Table D-1 shows the named compound, its CAS number,

identification method, the reported LRI and references where previously identified. Sixty

five of these compounds were previously identified and eleven were identified for the

first time.

Guava puree volatiles were divided into six organic compound categories. A total

of 6 aldehydes, 2 acids, 15 alcohols, 6 ketones, 21 esters and 8 terpenes were

identified by GC-MS (Figure 6-1). The relative difference in total volatiles (identified by

GC-MS) in terms of peak area was normalized to total peak area of freshly thawed

guava puree which was 395. As shown in Figure 6-2, DP-C02 treatment and thermal

pasteurization caused a 46 and 19% reduction in total peak area respectively, when

compared to the total peak area of the freshly thawed guava. Although the

pasteurization process caused a small reduction in peak area, an increase of 2% on the

area of oxygenated terpenes was observed.

The total peak area reduction caused by DP-C02 treatment may be related to the

solubility of the aroma compounds in carbon dioxide. Most of the aroma compounds

are lipophilic (Reineccius 2006). Carbon dioxide is a colorless, lipophilic and non-toxic


156









liquid that behaves as a supercritical fluid above its critical temperature (31.1C) and

critical pressure (7.39 MPa). Under DP-C02 processing conditions (temperature and

pressures above C02 standard conditions), C02 in the supercritical state is dissolved

and mixed with the guava puree. After processing, the pressure is released and the CO2

changes to its gaseous state. Furthermore, some of the volatiles will be carried along

with the outlet C02 gas. For example, ethanol and ethyl acetate (compounds of low

boiling point) were not detected in DP-C02 treated guava puree. This suggests that

these compounds were lost with the outlet C02 gas, and probably can be related to the

difference in sensory perception of the DP-C02 sample related to the FT, and

pasteurized guava purees.

Composition of the three samples was similar but there were major quantitative

differences. Alcohols comprised the largest group of volatiles in freshly thawed and

pasteurized guava purees, contributing 36% and 33% of the total respectively.

Aldehydes compromised the second group of volatile constituents for freshly thawed

and pasteurized guava (27 and 24% respectively), followed by esters, terpenes,

ketones and acids. The main contributors to the total volatile peak areas of the DP-C02

treated guava puree were aldehydes (21%), followed by alcohols, esters, terpenes,

ketones and acids (18%, 13%, 7%, 6% and 1% respectively). Therefore, DP-C02

treatment influenced the volatile profile of the puree because alcohols were reduced by

50%, esters were reduced by 43% and aldehydes were reduced by 24%.

GC-MS Identifications

A total of 58 volatiles were identified using GC-MS in guava puree samples

(Table 6-1 and Figure 6-2). Aldehydes and alcohols were the main volatiles of guava

puree used in this study. These results agree with those found previously (Idstein and









Schreier 1985; Nishimura and others 1989; Stevens and others 1970). The presence of

high amount of aldehydes and alcohols may involve enzymatic oxidation and reduction

of corresponding fatty acids (Idstein and Schreier 1985).

Seven compounds were identified for first time in guava puree. They are:

butanal, nonanal, isoamyl alcohol, 2-octanone, ethyl-3-hydroxyhexanoate, methyl

nonanoate and ethyl nonanoate. The compounds reported previously in guava fruit

were identified using different extraction techniques. For example, Chan and Cavaletto

(1982) studied the changes in chemical and sensory quality during processing and

storage of aseptically packaged guava puree. The puree was acidified to pH 3.9 with

citric acid and the soluble solids content was 13.5%. The heat treatment was conducted

at 930C for 26 seconds and the product was aseptically packed. Quality of all samples

was compared to untreated frozen puree. Flavor was not as greatly affected as color

and the flavor changes were the result of storage time and not processing (Chan and

Cavaletto 1982).

Yen and others (1992) studied the changes in flavor components of guava puree

resulting from pasteurization and frozen storage. The volatiles were extracted using

Likens-Nickerson apparatus and a GC-FID and GC-MS were used for the identification

of the compounds. Initially, the volatile constituents of the pasteurized puree were

similar to the unpasteurized puree. Terpene hydrocarbons were the main volatile

components followed by aldehydes. Significant changes in volatile constituents and

deterioration of flavor quality during the first two months of storage at 0C were

attributed to oxidation and enzymatic reactions. Yen and Lin (1996) studied the changes

in volatile flavor components of guava juice with high pressure and heat processing.


158









They used purge and trap as the extraction technique and found that high pressure

processing maintained the original flavor while heat processing caused a decrease in

the majority of volatile constituents. (Z)-3-hexenal, hexanal, ethyl-2-methyl butanoate,

(Z)-3-hexenol, 3-damascenone and t-caryophillene have been identified in guava leaves

(Rouseff and others 2008). Three esters (ethyl acetate, ethyl hexanoate and 3-hexenyl

acetate), 2 aldehydes (hexanal and (Z)-3-hexenal), 2 alcohols (hexanol and (Z)-3-

hexenol) and 1 terpene (t-caryophillene) have been reported as volatile constituents of

guava fruit. Differences in volatile profiles among studies can be attributed to different

guava cultivars used and extraction methods.

Table 6-2 lists the 58 volatiles identified by MS in this study. To compare the

volatiles in the three guava samples (freshly thawed, DP-CO2 and pasteurized guava

purees), peak areas were normalized on the single largest peak found in FT guava

puree. This peak was, hexanal, and it was assigned a value of 100 and the remaining

peaks in all three samples were normalized to it.

Fifty eight volatiles were found in fresh guava puree compared to 54 volatiles

found in both DP-CO2 and pasteurized guava purees. Acetaldehyde, butanal and 2-

octanone were only founded in FT guava puree while ethyl alcohol was found in the FT

and pasteurized guava purees. (Z)-3-hexenyl hexanoate was only found in the FT and

DP-CO2 guava purees but not in the pasteurized puree. Hexanal was one of the

highest compounds found in all three samples with a relative peak area between 79 and

100%. This agrees with results found by (Idstein and Schreier, 1985; Nishimura and

others 1989; Paniandy and others 2000; Mahattanatawee and others 2005). DP-C02


159









and pasteurization treatments reduced the hexanal peak area by 21 and 5%

respectively.

Among the 15 alcohols, (Z)-3-hexenol and 1-hexanol contribute between 68 and

78% of the total peak area in all three samples. DP-CO2 and pasteurization reduced the

peak area for (Z)-3-hexenol by about 10%. (Z)-3-hexenol has been previously reported

as the main component of white guava (Paniandy and others 2000), guava from Hawaii

(Stevens and others 1970), fresh pink guava puree (Jordan and others 2003) and guava

from Brazil (Idstein and Schreier 1985). Ethyl acetate (28%), ethyl propanoate (20%)

and ethyl hexanoate (22%) were the most abundant esters present in all three samples.

However, their concentration was lower in DP-CO2 and pasteurized guava purees.

Macleod and Gonzalez de Troconis (1982) reported ethyl acetate and ethyl hexanoate

as the major esters of guava from Venezuela. Yen and Lin (1999) reported the largest

amount of ethyl acetate in guava juice. Idstein and Schreier (1985) found that ethyl and

acetate esters were predominant in guava from Brazil and they suggest that their

presence was biogenetically derived from 3-oxidation of corresponding fatty acids. a-

humulene was the predominant terpene found in guava purees used in this study.

These results differ partially from those reported by Wilson and Shaw (1978),

Mahattanatawee and others (2005) and Jordan and others (2003) who reported 3-

caryophyllene as a major terpene. Elevated levels of a-humulene and 3-bisabolene

were observed in the pasteurized sample only.

GC-O Aroma Profiles

A total of 26 aroma compounds were found in guava puree and are listed in

Table 6-3. Of these 26 compounds, 19 were confirmed with GC-MS and 4 were

confirmed with GC-PFPD. Of the 11 newly identified, 3 of these were aroma active


160









compounds and are listed in Table 6-3 in bold. Twenty-four, 22 and 13 aroma active

compounds were detected for FT, DP-C02 and pasteurized guava purees respectively.

Peak heights were normalized to the highest peak present in the FT guava puree. This

peak was (Z)-3-hexenyl hexanoate and it was assigned a value of 100 and the

remaining peaks in all three samples were normalized to it. Eleven compounds were

detected in all three samples: octen-3-ol, benzyl alcohol, furaneol, and homofuraneol,

ethyl hexanoate, phenyl acetaldehyde, ethyl octanoate and ethyl dihydrocinnamate,4-

mercapto-4-methyl-pentan-2-ol and 3-mercapto-l-hexanol and (E,Z)-2,6-nonadienal.

Six of the eleven compounds found in the 3 samples were tentatively identified based

on LRI and odor descriptors. Phenyl acetaldehyde and 1-p-menthene-8-thiol were

identified for the first time in guava puree but they have been previously identified in

other food products (Demole and others 1982; Buettner and Schieberle 1999; Janes

and others 2009).

The most intense odorant in FT guava puree was (Z)-3-hexenyl hexanoate,

followed by homofuraneol, hexanal, ethyl hexanoate and furaneol. (Z)-3-hexenyl

hexanoate was reported as an important aroma contributor to fresh guava puree from

Florida (Jordan and others 2003). The most intense odorant in DP-C02 guava puree

was 3-mercapto-l-hexanol followed by ethyl-3-phenylpropionate and (E,Z)-2,6-

nonadienal. The last 2 compounds had the same intense perception. (E,Z)-2,6-

nonadienal was the most intense peak for pasteurized guava puree (its intensity was

5% less than in DP-C02 treated puree) followed by furaneol and octen-3-ol. Ethyl 3-

phenylpropionate, phenyl acetaldehyde and furaneol were described to have guava like

aroma. Phenyl ethylacetate was identified by Stevens and others (1970) as a









contributor to the overall pleasant fruity aroma of guava from Hawaii. Ethyl acetate

(fruity, pleasant and sweet aroma), ethyl phenylacetate (described as fruity odorant) and

(Z)-3-hexenol (fruity, green and grassy notes) were detected only in FT samples. Ethyl

phenylacetate was detected at a concentration between 50 and 250 pg/kg pulp by

Idstein and Schreier (1985) in guava from Brazil and ethyl acetate and (Z)-3-hexenol

were identified as important aroma contributors to fresh guava puree from Florida

(Jordan and others 2003).Ethyl acetate was identified as a major ester contributor to the

flavor of guava from Venezuela by Macleod and Gonzalez de Troconis (1982). (E)-3-

hexen-ol (green, grassy note) was detected only in DP-C02 guava puree. A ketone,

carvone, (described as sweet, spearmint and peppermint aroma) was detected only in

DP-C02 and pasteurized guava purees. Pentene-3-one (nutty), hexanal (fatty, grassy),

methional (cooked potato), 1,8-cineole (eucaliptus, camphorous, cool), ethyl-3-

phenylpropionate (guava, fruity), p-menthene-8-thiol (medicine), 3-ionone (floral, fruity),

(Z)-3-hexenyl hexanoate (apple peel, fruit) and y-decalactone nutlikee, coconut) were

identified only in FT and DP-C02 treated guava samples. 3-ionone, which has a very

low odor threshold and an intense violet aroma was identified as the main contributor of

floral flavor of Hawaiian guava by Stevens and others (1970). Ethyl hexanoate (fruity,

sweet) was detected in all three samples and reported as an important flavor compound

by MacLeod and Gonzalez de Troconis (1982) in guava from Venezuela.

GC-PFPD Analysis

A total of 11 sulfur compounds were found in guava puree and are listed in Table

6-4. The compounds were identified by matching LRI with databases (Rouseff 2006)

and standards. Seven of these compounds have been previously reported in guava

puree and two were reported for the first time. Hydrogen sulfide and dimethyl sulfide


162









have not been previously reported in guava puree; however, they have been reported in

guava leaves (Rouseff and others 2008). Methanethiol, dimethyl disulfide and methional

have been reported in guava fruit and leaves. 1-p-Menthene-8-thiol or grapefruit

mercaptan and 4-mercapto-4-methyl-pentan-2-one were identified for the first time in

guava puree. 4-mercapto-4-methyl-pentan-2-one has been reported as a key

component of sauvignon grapes and is used in flavor manufacture to recreate the catty

note of blackcurrant (Rowe 2000). 1-p-Menthene-8-thiol has a very low threshold

(0.0001 ng/mL) and is considered an important sulfur compound of grapefruit juice

(Demole and others 1982). Sulfur-containing compounds play an important role in

natural flavor chemistry as they are not only responsible for the objectionable odors

associated with rotting vegetable matter but also contribute, and often characterize, the

desirable aroma of many plants and thermally processed foods (Reinneccius 2006). In

thermally processed food, these compounds are formed by Maillard reactions, while in

plants its formation is part of in vivo biogenesis.

Peak heights were normalized to the highest peak area in FT guava sample, which

was 1-p-menthene-8-thiol. It was assigned a value of 100 and the remaining peaks in all

three samples were normalized to it. Eleven sulfur compounds were identified in DP-

C02 treated guava samples while 10 were found in FT and pasteurized sample. In FT

guava puree, the grapefruit smelling, menthene-8-thiol (100) was the major sulfur

compound followed by methanethiol, dimethyl disulfide and hydrogen sulfide. In DP-

C02 and pasteurized guava purees, methanethiol was the major constituent. The

second major constituent in DP-C02 was 1-p-menthene-8-thiol followed by hydrogen

sulfide and dimethyl disulfide. Mercapto-hexyl acetate was the second major constituent


163









in pasteurized guava puree, followed by hydrogen sulfide and dimethyl disulfide. 4-

mercapto-4-methyl-penta-2-one was identified only in the DP-C02 guava puree. 3-

mercapto hexyl acetate passion fruit mercaptan and 3-mercapto-l-hexanol were

identified as having high FD factors by (Steinhaus and others 2008). In addition,

methional showed high odor activity. The first 2 compounds were previously identified in

passion fruit and were reported for first time in guava puree by (Steinhaus and others

2008). Hydrogen sulfide and dimethyl sulfide were reported for the first time in guava

puree but were previously identified in guava leaves (Rouseff and others 2008). The

presence of dimethyl disulfide in the puree is not surprising since it is produced as a

plant defense response.

Conclusion

A total of 76 compounds were identified in guava purees. Eleven compounds

were identified for the first time, but were previously identified in other foods. These

eleven compounds are: butanal (previously reported in delicious apple fruit), isoamyl

acetate (strawberry), 4-mercapto-4-methyl-penta-2-one (key component of sauvignon

grape, and reported in grapefruit), phenyl acetaldehyde (key odorant in honey and

found in chocolate and buckwheat), nonanal (reported in grapefruit, orange juice and

red delicious apple), homofuraneol (coffee, muskmelon and soy sauce), methyl

nonanoate (strawberry), 1-p-menthene-8-thiol (grapefruit), 2-octane, ethyl-3-

hydroxyhexanoate (yellow mombin) and ethyl nonanoate. Fresh, DP- C02 and

pasteurized purees showed similar composition, but quantitative differences were found

in the aroma active and sulfur compounds. Total peak area was highest for the fresh

sample while pasteurization caused a slight decrease in concentration. DP-C02

reduced the total peak area by about 24%. Twenty six compounds were identified as


164









guava aroma contributors, three of which were tentatively identified. Pentene3-one,

octen-3-ol, benzyl alcohol, phenyl acetaldehyde, homofuraneol, (E,Z)-2,6-nonadienal

and carvone, were not detected by GC-MS, so these compounds may play an important

role in guava flavor. Hexanal was the volatile present in highest concentration in the

three samples and (Z)-3-hexenyl hexanoate was the major odor contributor for the fresh

and DP-C02 guava purees. GC-PFPD was a useful tool in the identification of sulfur

compounds. Four new sulfur compounds were identified in guava purees, 2 of which

were previously identified in guava leaves (dimethyl sulfide and hydrogen sulfide).

Sensory analysis indicate that there were differences between the freshly thawed and

DP-C02 guava purees, as confirmed by volatile composition but the acceptability of the

DP-C02 treated guava puree was not significantly different from the freshly thawed

puree.



Table 6-1. Difference in flavor and overall likeability between freshly thawed (reference
and hidden reference), dense phase-CO2 processed (DP-C02; 30.6 MPa, 8%
C02, 6.9 min, 35 C) and thermally treated (90 oC, 60 s) guava purees
detected by untrained panelists (n = 75) at weeks 0
Sample Differences in flavor* Overall likeability

DP-C02 5.66a 4.13a

Pasteurized 3.56b 3.79a

Hidden Reference 3.05b 4.29a
(control or freshly thawed)
Difference observed when compared to given reference (difference from control test).
Values with similar letters within columns are not significantly different (Tukey's HSD,
p > 0.05).


165










120



100 -


80



60



40


20 -



0-


Fresh


Pasteurized


* Terpenes


* Esteres Ketones Alcohols m Acids Aldehydes


Figure 6-1. Chemical composition of headspace volatiles for guava purees. Total number of compounds for each class is
put in parentheses. The three samples were normalized to the total peak area of fresh guava puree


166


DP-C02




















































6
4
fLLJ


23
39
21 45
2129
16 20 25 2 3
27 32 1 46 49
17 \ 28 31 42


13_51


18.51


51
52


23.51


28.51


33.51


Figure 6.2- Total ion chromatogram (TIC) for freshly thawed guava puree on DB-5 column








167


~


I


f


S E [ mec










Table 6-2. MS identification of guava puree volatiles. Peak areas were normalized (100) to the largest peak in all three
samples.

LRI Normalized Peak Area (%)*
# Name DB5 Wax Fresh DP-C02 Pasteurized
1 acetaldehyde 731 0.37 Nd Nd
2 ethanol 511 906 59.43 Nd 55.56
3 acetic acid 553 828 2.93 2.03 3.54
4 Butanala 607 1472 5.82 Nd Nd
5 ethyl acetate 619 893 28.49 14.86 27.09
6 ethyl propanoate 714 947 20.18 15.60 17.66
7 2-Butanone, 3-hydroxy- or acetoin 723 1318 1.14 0.66 0.80
8 1-butanol, 2-methyl or amyl alcohol 731 1.24 1.24 1.24
9 1-butanol, 3-methyl or isoamyl alcohola 738 1200 0.44 0.44 1.17
10 (Z)-3-hexenal 1158 0.30 0.69 0.18
11 hexanal 807 1101 100.00 79.25 95.54
12 ethyl-2-methyl butanoate 848 1053 0.21 0.14 0.15
13 3-hexen-1-ol (E) 851 1379 0.39 0.35 0.46
14 Z-3-hexenol 862 1402 36.93 33.10 33.44
15 1-hexanol 872 1366 36.29 28.54 32.10
16 1-butanol-3-methyl acetate 879 1137 0.78 0.80 0.90
17 methyl hexanoate 916 3.00 1.15 2.41
18 methyl-3-hexenoate 934 0.23 0.12 0.19
19 octen-3-one, 1 982 15.12 11.11 1.15
20 6-methyl-5-hepten-2-one or methyl heptenone 986 1347 1.83 5.68 11.01
21 2-octanonea 994 0.43 Nd Nd
22 ethyl hexanoate 996 1248 21.77 4.76 5.92
23 3-hexenyl acetate or 3-hexen-l-ol, acetate 1008 3.03 2.44 1.58
24 acetic acid hexyl ester or hexyl acetate 1013 1267 2.17 1.46 1.46
25 d-limonene 1041 1205 6.16 3.30 6.84
26 1,8-Cineole 1047 1212 3.05 1.80 2.31
27 (E)-b-Ocimene 1055 1265 0.27 0.45 0.24
28 furaneol 1069 0.77 0.39 0.54
29 1-octanol 1072 1572 1.99 2.19 3.28
30 ethyl heptanoate 1335 0.88 0.89 0.47
31 linalool 1105 1554 0.92 0.99 0.47


168










Table 6-2. Continued
LRI Normalized Peak Area (%)
# Name DB5 Wax Fresh DP-C02 Pasteurized
32 Nonanala 1109 1420 0.52 1.01 1.00
33 methyl octanoate 1126 0.26 0.18 0.13
34 ethyl-3-hydroxyhexanoatea 1682 0.37 0.40 0.24
35 benzoic acid ethyl ester 1187 1709 0.87 1.00 0.59
36 a-terpineol 1725 0.70 0.25 0.17
37 ethyl octanoate 1198 1452 1.65 1.38 0.68
38 decanal 1212 1526 0.39 0.48 0.48
39 methyl nonanoatea 1227 0.19 0.20 0.10
40 ethyl phenylacetate or 2-phenylacetate 1786 0.67 0.96 0.53
41 benzene propanol or 3-phenylpropanol 1254 2086 0.59 0.71 0.63
42 phenylethyl acetate 1259 0.36 0.41 0.16
43 nonanoic acid 1265 2.45 2.66 0.91
44 ethyl nonanoatea 1298 1.83 1.95 1.01
45 alpha-cubebene 1339 1.01 1.01 7.34
46 ethyl-3-phenylpropionate or ethyl dihydrocinnamate 1367 0.42 0.48 0.21
47 cis-3-hexenyl hexanoate 1386 0.18 0.17 Nd
48 B-damascenone 1407 1862 1.01 1.01 0.83
49 t-caryophyllene 1446 1641 1.01 1.01 0.60
50 a-humulene 1460 1650 17.39 18.82 26.96
51 g-decalactone 1480 2130 1.37 2.08 2.25
52 b-ionone 1495 1984 1.79 1.95 2.40
53 B-bisabolone 1503 1754 1.01 1.01 2.95
54 beta or delta- cadinene 1793 0.25 0.35 0.17
55 (E)- or d-nerolidol 1577 2055 0.52 0.71 0.25
56 2-phenylethyl alcohol 1963 0.22 0.25 0.30
57 3-phenyl propyl acetate 1984 0.25 0.40 0.27
58 eugenol 2183 1.04 1.18 1.77
a,.---------------


169


Identity Tor first tim ingaapreTo o iia eUIC om lzdae aeo eaa eKaaye nD ou n










Table 6-3. Guava puree aroma active compounds. Peak heights were normalized (100) to the most intense peak in all
three samples.

LRI Normalized Peak Area (%)*
DP-
# Name Descriptor DB5 Wax Fresh C02 Pasteurized
1 ethyl acetate e fruity, pleasant, sweet 627 42.90 ND ND
2 pentene-3-one, 1a nutty 1023 42.72 56.50 ND
3 Hexanala fatty, grassy 801 89.04 73.87 ND
4 (E)-3-hexenola green, grassy 1376 ND 67.98 ND
5 (Z)-3-hexenola'c'd fruity, green, grassy 1409 55.15 ND ND
6 Methionalb cooked potato 863 1451 69.53 79.05 ND
7 octen-3-ola, mushroom 977 52.67 59.55 85.94
8 octen-3-onead mushroom 985 64.41 ND 80.23
9 ethyl hexanoatea c, d, e fruity, sweet 999 1299 82.77 75.97 87.40
10 1,8-cineolea&0 eucaliptus, camphoreous, menthol-like 1232 62.43 70.29 ND
11 benzyl alcohol floral rose 1887 52.23 70.71 81.97
12 4-mercapto-4-methyl-pentan-2-olbd flowery 1533 48.59 70.43 71.73
13 phenyl acetaldehyde guava. Green, honey-like 1044 80.47 74.78 80.97
fruity, sour, guava, caramel, cotton
14 Furaneola, d candy 1068 82.50 73.17 87.89
15 homofuraneol caramel, candy 2069 94.56 88.28 80.45
16 mercapto hexan-l-olb,d grapefruit 1126 1862 76.21 104.53 61.11
17 ethyl-3-hydroxyhexanoatea guava = fruity 1128 1672 61.03 93.11 ND
18 nonadienal, (E,Z)-,6a d green, cucumber 1163 63.43 95.53 90.25
19 ethyl octanoatea, c d sweet, fruity 1198 1434 78.07 89.43 82.26
20 ethyl phenylacetatea&c fruity 1254 50.01 ND ND
21 Carvoned spearmint, peppermint, sweet 1255 ND 95.53 73.22
22 p-menthene-8-thiolb medicine 1298 1505 53.68 59.97 ND
23 B-iononead floral, fruity 1988 60.91 81.67 ND
24 Ethyl-3-phenylpropionatead floral 1356 80.57 63.18 73.74
25 (Z)-3-hexenyl hexanoatea,'c d apple peel, fruit 1387 100.00 80.65 ND
26 g- decalactonea nutlike or cococut 2146 45.37 61.45 ND
Normalized area base on (Z)-3-hexenyl hexanoate peak identified on DB-5 column aCompounds confirmed with GC-MS, "Compounds
confirmed with GC-PFPD,GC-PFPD, Previously reported as aroma contributors by Jordan and others (2003) dPreviously reported as aroma
contributors by Mahattanatawee and others (2005), ePreviously reported as aroma contributor by Macleod and Gonzalez de Troconis (1982), bold
compounds= newly identified compounds


170









Table 6-4. Guava puree sulfur volatile compounds. Peak
o t the most intense peak in al three samp s


heights were normalized (100)


LRI Normalized Peak Area (%)*
DP-
# Name DB5 Wax Fresh C02 Pasteurized
1 hydrogen sulfide 472 73.02 55.54 25.90
2 Methanethiolab 477 679 96.19 71.91 69.20
3 dimethyl sulfide 523 731 8.89 7.21 29.34
4 dimethyl disulfideab 759 1049 82.99 37.85 10.17
5 2-methyl-3-furanthiolb 869 1338 1.46 0.95 2.90
6 Methionalab 912 1441 1.85 0.42 0.61
7 4-mercapto-4-methyl-pentan-2-one 936 1361 Nd 0.80 Nd
8 4-mercapto-4-methyl pentan-2-olb 1057 1538 2.85 3.21 3.23
9 3-mercapto hexan-1-olb 1117 1862 4.75 1.55 17.64
10 mercapto hexyl acetateb 1241 1879 20.23 20.07 30.88
11 1-p-menthene-8-thiol 1302 1499 100.00 57.54 14.37
aPreviously reported in guava leaves, "Previously reported in guava puree, Normalized area base onp-
menthene-8-thiol peak identified on DB-5 column, bold compounds= newly identified compounds









CHAPTER 7
SUMMARY AND CONCLUSIONS


Dense phase carbon dioxide (DP-CO2) was evaluated as an alternative to

pasteurization for treatment of guava puree to extend shelf life, reduce microbiological

load, and preserve sensory and nutritional characteristics of the puree. Processing of

puree with DP-CO2 was facilitated by first treating the puree with a commercially

available enzyme, Bioguavase, at low temperature (30 oC) to obtain a product with the

proper consistency for processing without affecting the physiochemical properties. The

optimial time and concentration for enzyme treatment were determined to be 600 ppm

and 3 hours of reaction time. Additional experiments addressed the optimization of DP-

CO2 treatment for microbial reduction in guava puree. Physical, chemical, microbial and

sensory quality of DP-CO2 treated guava puree was compared to freshly thawed and

heat pasteurized samples.

The first objective was to optimize the reaction time and concentration of a

commercially available enzyme to obtain a product with the proper consistency for DP-

CO2 processing. Enzyme treatment decreased (between 0.55 to 11% of the initial

concentration) the antioxidant capacity of the samples and the antioxidant capacity was

lower regardless of the enzyme concentration. The turbidity decreased with enzyme

activity. The enzyme activity affected the pH and TSS content: the pH decreased

(between 0.5 and 1.6% of the initial pH) while TSS content increased (between 4 and

10% of the initial value). Color was affected by the enzyme concentration since the

enzyme treatment produced a clarified juice. Three hours of reaction time and 600 ppm

of enzyme concentration were adequate to produce a clarified juice with a minimal


172









effect on nutritional quality, even though antioxidant capacity was decreased and total

soluble polyphenols were increased.

When conducting the DP-C02 optimization, C02 solubility in the guava puree was

measured using an apparatus designed and built at the University of Florida Food

Science and Human Nutrition Department (Gainesville, FL). This measurement showed

that 5.3% C02 was adequate to reach saturation of the product. Due to equipment

limitations, 8% C02 was used. Surface response analysis of microbial reduction showed

that the quadratic model fit for yeasts and mold (Y&M) was statistically significant (P

<0.05) and there was a satisfactory correlation between the actual and the fitted values

for Y&M (R2 = 0.88). According to the regression surface equation, 8 min was sufficient

to achieve a 5 log reduction using a processing pressure of 34.5 MPa and a

temperature of 35 C. According to the results obtained for aerobic plate count (APC) a

complete sterilization could not be obtained. Because of the increase in viscosity and

foam formation in the puree during treatment, the residence time was set to 6.9 min.

The quality attributes pH, oBrix, titratable acidity, color, APC, Y&M), and

phytochemical content were determined during 14 weeks of storage at 4 C and

compared to those of freshly thawed (fresh) and heat pasteurized samples. APC for

fresh and DP-C02 guava puree remained constant (3.5 log) during the first 7 weeks of

storage, after which there was a significant increase in microbiological load for the DP-

C02 treated guava. Pasteurized guava puree APC started at 1.5 log and increased

significantly (approximately 1.5 log) at the end of 4 weeks. DP-C02 and pasteurization

treatments of guava puree caused approximately 1.3 log reduction in Y&M counts when

compared to fresh. DP-C02 Y&M count was lower than the pasteurized treatment


173









during the 14 weeks of storage. The pH and oBrix remained between 3.72 and 3.93

and between 6.7 and 7.4, respectively for DP-C02 and pasteurized,, throughout

storage. Titratable acidity of DP-C02 treated samples was significantly higher than fresh

and pasteurized samples. DP-C02 caused partial inactivation (20%) of pectinesterase

(PE) activity and an increase (100%) in guava puree viscosity. Further study is

recommended to understand the mechanism of gel formation and cloud loss due to DP-

C02 treatment. ORAC and total soluble phenolics (TSP) values for fresh guava puree

were between 12.06 and 16.3 pMol TE/mL and 174.5 and 186.39 mg of GAE/ L of

sample, respectively. DP-C02 values were between 11.08 and 16.95 pMol TE/mL

(ORAC) and 181.15 and 196.51 mg of GAE/ L of sample (TSP). DP-C02 treatment can

protect polyphenolic and antioxidant levels throughout processing and storage without

compromising physico-chemical and phytochemical properties of the guava puree. DP-

C02 delayed the degradation of Vitamin C content during storage compared to the fresh

puree. DP-C02 treated guava puree retained organic acids contents similar to fresh

guava puree.

Flavor analysis of guava puree allowed the identification of 76 compounds. Eleven

compounds were identified for first time in guava puree, but were previously identified in

other foods. Fresh, DP- C02 and pasteurized puree showed a similar composition but

quantitative differences were found in the aroma active and sulfur compounds. Total

peak concentration was highest for the fresh sample while pasteurization caused a

slight decrease in the compounds concentration. DP-C02 reduced the total peak area in

about 24% when compared to the total peak area of the fresh guava puree. Twenty six

compounds were identified as guava aroma contributors, three of which were tentatively


174









identified. These compounds were not detected by GC-MS, so they may play an

important role in guava aroma. Hexanal was the volatile present in higher concentration

in the three samples and (Z)-3-hexenyl hexanoate was the major odor contributor for

the fresh and DP-C02 guava puree. Twenty-four compounds were detected as aroma

contributors of the fresh guava puree while only 22 and 13 compounds were identified

as aroma active compounds in DP-C02 and pasteurized guava puree respectively.

This study showed that DP-C02 treatment extended shelf-life and preserved the

quality of guava puree. Further work is needed to investigate the mechanisms and

causes for the increase in viscosity of DP-C02 treated guava puree during storage and

to understand the mechanism of gel formation and cloud loss due to DP-C02 treatment.

Sensory evaluation with trained panelists will help to better understand the difference

between fresh, DP-C02 and pasteurized puree. Further product development of guava

juice or nectars should be conducted to determine the effect of DP-C02 nutritional

quality of this product.


175










APPENDIX A
ENZYME TREATMENT OF GUAVA PUREE

Table A-1. SAS software code used for the statistical analysis of repeated measurement
design and Tukey's standardized range (HSD) test


data guava;
infile cards;
input replicate conc time brix @@;
cards;
1 0 0 6 1 400 0
2 0 0 6 2 400 0
3 0 0 6 3 400 0
4 0 0 5.9 4 400 0
5 0 0 5.9 5 400 0
6 0 0 5.9 6 400 0
1 0 3 6 1 400 3
2 0 3 6 2 400 3
3 0 3 6 3 400 3
4 0 3 6 4 400 3
5 0 3 6 5 400 3
6 0 3 6 6 400 3
1 0 6 6 1 400 6
2 0 6 6 2 400 6
3 0 6 6 3 400 6
4 0 6 6 4 400 6
5 0 6 6 5 400 6
6 0 6 6 6 400 6
1 0 9 6.1 1 400 9
2 0 9 6.1 2 400 9
3 0 9 6.1 3 400 9
4 0 9 6.1 4 400 9
5 0 9 6.1 5 400 9
6 0 9 6.1 6 400 9
1 0 12 6.1 1 400 12
2 0 12 6.1 2 400 12
3 0 12 6.1 3 400 12
4 0 12 6.1 4 400 12
5 0 12 6.1 5 400 12
6 0 12 6.1 6 400 12


proc print; run;
proc sort; by conc time; run;
proc gim; class conc time; model brix=concltime; run;
proc sort; by conc; run;
proc gim; class time; model brix = time;
means time /tukey; by conc; run;
proc sort; by time; run;
proc gim; class conc; model brix= conc; means conc/tukey; by time; run;


176










Table A-2. SAS software output used for the statistical analysis of repeated
measurement design and Tukey's standardized range (HSD) test

The GLM Procedure
Dependent Variable: brix


Source
Model
Error
Corrected


Total
R-Square
0.849836


Sum of
DF Squares Mean Square FV
19 5.43300000 0.28594737 2!
100 0.96000000 0.00960000
119 6.39300000
Coeff Var Root MSE brix Mean
1.576502 0.097980 6.215000


alue
9.79


Pr> F
<.0001


Source DF Type I SS Mean Square F Value Pr> F
conc 3 2.28300000 0.76100000 79.27 <.0001
time 4 2.18550000 0.54637500 56.91 <.0001
conc*time 12 0.96450000 0.08037500 8.37 <.0001
--------------------------------------------- conc=0------------------------------------


Dependent Variable: bri

Source
Model
Error
Corrected Total
R-Squan
0.878049
Source
time


x
Sum of
DF Squares Mean Square F Value Pr;
4 0.10800000 0.02700000 45.00 <.00
25 0.01500000 0.00060000
29 0.12300000
e Coeff Var Root MSE brix Mean
9 0.406217 0.024495 6.030000
DF Type I SS Mean Square FValue Pr> F
4 0.10800000 0.02700000 45.00 <.0001


*F
01


DF Type III SS
4 0.10800000


Mean Square
0.02700000


FValue Pr>F
45.00 <.0001


Tukey's Studentized Range (HSD) Test for brix
NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type
II error rate than REGWQ.

Alpha 0.05
Error Degrees of Freedom 25
Error Mean Square 0.0006
Critical Value of Studentized Range 4.15337
Minimum Significant Difference 0.0415
Means with the same letter are not significantly different.
Tukey Grouping Mean N time


6.10000
6.10000
6.00000
6.00000
5.95000


Source
time











------------------------- --------------- time=0 ----------------------------------------
Dependent Variable: brix
Sum of
Source DF Squares Mean Square F Value Pr> F
Model 3 0.00000000 0.00000000 0.00 1.0000
Error 20 0.06000000 0.00300000
Corrected Total 23 0.06000000

R-Square Coeff Var Root MSE brix Mean

0.000000 0.920542 0.054772 5.950000

Source DF Type I SS Mean Square F Value Pr> F

conc 3 0 0 0.00 1.0000


Source DF Type III SS Mean Square F Value Pr> F

conc 3 0 0 0.00 1.0000

Tukey's Studentized Range (HSD) Test for brix

NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type
II error rate than REGWQ.
Alpha 0.05
Error Degrees of Freedom 20
Error Mean Square 0.003
Critical Value of Studentized Range 3.95829
Minimum Significant Difference 0.0885


Means with the same letter are not significantly different.


Tukey Grouping Mean N conc

A 5.95000 6 0
A
A 5.95000 6 400
A
A 5.95000 6 600
A
A 5.95000 6 800


178






































Figure A-1. Enzyme treated guava puree without filtration (left) and after filtration (right)

This picture showed that enzyme treated puree increased juice yield but caused

discoloration of puree after filtering through cheesecloth (sample to the right). A possible

explanation for color loss is that compounds such as phenolic compounds responsible

of provide color to the puree were retained on the guava cake or stayed behind with the

filtered out particles. These compounds were removed by the filtration process so the

juice was much lighter in color. Similar result was observed by Imungi and others

(1980).


179








APPENDIX B
DP-CO2 PROCESSING AND DATA


~sb/


Figure B-1. Removal of insoluble solids from guava puree

A. Picture presents guava puree solid removal using 200 pm nylon filter

B. Collection of the guava puree

C. Mixing of guava puree

D. Collection of guava puree for DP-CO2 treatment


180











V




PG2
SI-II-I- -


PG3
5 0'..' I BPR




I I






E -


HEl


GM


Figure B-2. Experimental solubility apparatus

Experimental solubility apparatus: C02 tank, high pressure pump (HPP), heat

exchanger (HE), pressure gauge (PG), back pressure regulator (BPR), thermocouple

(T), metering valve (MV), gas meter (GM), water bath, bottle, stopper, fume hood (FH),

two way valves (V), three way valves (TV), vessels and quick connects (QC) (Calix

2008).















181










































Figure B-3. Continuous DP-C02 system used during processing


182









Table B-1. Guava Puree Solubility data
TRT's Pressure C02 Volume C02 C02
(MPa) (Sit) (grams) va Volume (ml)Solubility Mean St
Control 0 0
Control 0 0
TRT 1 6.9 3.126 6.14 162 3.79
TRT 1 6.9 3.213 6.31 162 3.90 3.84 0.074609
TRT2 10.34 3.399 6.68 168 3.98
TRT 2 10.34 3.401 6.68 166 4.03 4.00 0.035539
TRT3 17.24 3.612 7.10 168 4.22
TRT 3 17.24 3.601 7.08 168 4.21 4.22 0.009096
TRT4 24.13 3.634 7.14 168 4.25
TRT4 24.13 3.487 6.85 164 4.18 4.21 0.051231
TRT5 31.03 3.611 7.09 164 4.33
TRT 5 31.03 3.638 7.15 168 4.25 4.29 0.050504

Table B-2. Guava Puree pH, Brix and Titratable acidity measurement before and after the CO2 solubility determination
Pressure
TRT's P pH Mean St Dev Brix Mean St Dev % acid Mean St Dev
(MPa)
Control 0 3.82 6.80 0.56
Control 0 3.79 3.81 0.021 6.70 6.75 0.071 0.55 0.55 0.007
TRT 1 6.9 3.77 6.77 0.55
TRT 1 6.9 3.80 3.79 0.019 6.73 6.75 0.024 0.55 0.55 0.005
TRT 2 10.34 3.75 6.80 0.58
TRT 2 10.34 3.75 3.75 0.002 6.80 6.80 0 0.58 0.58 0.000
TRT 3 17.24 3.74 6.80 0.60
TRT 3 17.24 3.75 3.75 0.007 6.90 6.88 0.071 0.60 0.60 0.002


183









Pressure
TRT's Pre e pH Mean St Dev Brix Mean St Dev % acid Mean St Dev
(MPa)
TRT 4 24.13 3.72 6.80 0.60
TRT 4 24.13 3.75 3.74 0.016 7.0 6.90 0.141 0.59 0.60 0.003
TRT 5 31.03 3.71 7.0 0.66
TRT 5 31.03 3.73 3.72 0.014 6.9 6.95 0.071 0.65 0.66 0.006










APPENDIX C
MICROBIAL INACTIVATION AND STORAGE STUDY DATA AND ANALYSIS

Table C-1. The average initial and final aerobic plate counts (APC) standard
deviations at 11 experimental runs from 2-factor, 3-level Central Composite
Design (CCD)
Pressure Residence
Run (MPa) time (min) Initial load (cfu/mL)* Final load (cfu/mL)*
1 34.5 8 1.06 x 107 + 1.635 x 105 6625 +6.6
2 24.1 8 1.06 x 107 +1.635 x 105 5650 +9.15
3 24.1 6.5 1.06 x 107 +1.635 x 105 5130 +6.40
4 13.8 8 1.06 x 107 + 1.635x 105 6150 +6.03
5 34.5 6.5 2.35 x 107 +7.93 x 105 3430 +3.77
6 24.1 6.5 2.35 x 107 +7.93 x 105 2730 +3.86
7 13.8 6.5 2.35 x 107 +7.93 x 105 2430 +2.22
8 13.8 5 2.35 x 107 +7.93 x 105 1680 +2.5
9 34.5 5 2.17 x 107 + 10.88 x 105 5150 +12.50
10 24.1 5 2.17 x107 + 10.88 x 105 5700 +7.05
11 24.1 6.5 2.17 x107 +10.88 x105 5800 +4.69
*Averages of the plates with APC counts lower than 250 colony forming units (cfu's)


Table C-2. SAS software code used for the response surface methodology (RSM)
analysis of 11 experimental runs determined by Central Composite Design
title 'OPTIMIZATION1';
data optimization;
input TRT$ P RT APC YM;
datalines;
T1 34.5 8 3.204816837 4.297278482
T5 34.5 6.5 3.509089594 3.863291003
T9 34.5 5 3.706133727 3.207230794
T2 24.1 8 3.276150588 4.283911356
T3 24.1 6.5 3.317110445 4.075429733
T6 24.1 6.5 3.521746446 3.965684468
T11 24.1 6.5 3.651803709 3.654499463
T10 24.1 5 3.655371191 3.490671907
T4 13.8 8 3.233819901 4.087768546
T7 13.8 6.5 3.586547572 4.032734615
T8 13.8 5 3.572065724 4.175137907

proc sort;
by P RT;
run;
proc print;
run;
proc rsreg data=optimization1;
model APC YM = P RT / lackfit;
run;


185













Table C-3. SAS software output of the response surface methodology (RSM) regression

analysis of 11 experimental-run data determined by central composite design

Response Surface for Variable APC


Response Mea
Root MSE
R-Square
Coefficient


of Variation


3.475878
0.116568
0.7990
3.3536


Regression

Linear
Quadratic
Crossproduct
Total Model




Residual


Type I Sum
DF of Squares R-Square


0.247697
0.015760
0.006659
0.270116


Sum of
DF Squares


0.7327
0.0466
0.0197
0.7990


F Value Pr > F

9.11 0.0215
0.58 0.5936
0.49 0.5151
3.98 0.0780


Mean Square F Value Pr > F


Lack of Fit 3
Pure Error 2
Total Error 5




Parameter DF Estimate

Intercept 1 2.549984
P 1 0.015333
RT 1 0.373874
P*P 1 0.000045421
RT*P 1 -0.002628
RT*RT 1 -0.034298


0.011004
0.056937
0.067941


Standard
Error

1.435044
0.041329
0.433903
0.000684
0.003754
0.032550


Sum of
Factor DF Squares

P 3 0.006846
RT 3 0.269317


0.003668
0.028468
0.013588


0.13 0.9348


t Value Pr > Itl


1.78
0.37
0.86
0.07
-0.70
-1.05


0.1357
0.7258
0.4283
0.9496
0.5151
0.3403


Mean Square F Value Pr > F


0.002282
0.089772


0.17 0.9136
6.61 0.0343


186


Parameter
Estimate
from Coded
Data

3.515327
0.004601
-0.203196
0.004866
-0.040800
-0.077170



















Response Surface for Variable YM


Response Mean


Root MSE
R-Square
Coefficient


of Variation


3.921240
0.166730
0.8815
4.2520


Type I Sum
DF of Squares


0.680830
0.007173
0.346022
1.034025


Sum of
DF Squares


Parameter DF Estimate


Intercept
P
RT
P*P
RT*P
RT*RT


6.059405
-0.161934
-0.208145
0.000494
0.018945
-0.003808


0.043640
0.095354
0.138994


Standard
Error

2.052571
0.059113
0.620619
0.000978
0.005370
0.046557


R-Square F Value Pr > F


0.5804
0.0061
0.2950
0.8815


12.25
0.13
12.45
7.44


0.0118
0.8818
0.0168
0.0230


Mean Square F Value Pr > F


0.014547
0.047677
0.027799


0.31 0.8241


t Value Pr > Itl


2.95
-2.74
-0.34
0.50
3.53
-0.08


0.0318
0.0408
0.7510
0.6350
0.0168
0.9380


Parameter
Estimate
from Coded
Data

3.896719
-0.154640
0.299793
0.052901
0.294117
-0.008569


Sum of
Factor DF Squares


Mean Square F Value Pr > F


3 0.496394
3 0.883761


Regression

Linear
Quadratic
Crossproduct
Total Model




Residual

Lack of Fit
Pure Error
Total Error


0.165465
0.294587


5.95 0.0419
10.60 0.0132









Figure C-1. DP-C02 treated samples during storage (week 4)


(a) no agitation

















(b) after
agitation


188











APPENDIX D
CHANGES IN AROMA COMPOUNDS AND SENSORY PERCEPTION IN GUAVA
PUREE AFTER THERMAL AND NON-THERMAL PROCESSING



TODAY'S SAMPLE:
Guava Drink


To start the test, click on the Continue button below:
Question # 1.
Please indicate your gender.
O Male
O Female

Question # 2.
Male: Which of the following ranges includes your age?
O Under 18
O 18-20
O 21-24
O 25-29
O 30-34
O 35-39
O 40-44
O 45-49
O 50-54
0 55-59
0 60-65
0 Over 65

Question # 3.
Female: Which of the following ranges includes your age?
O Under 18
O 18-20
O 21-24
O 25-29
O 30-34
O 35-39
O 40-44
O 45-49
O 50-54
0 55-59
0 60-65
0 Over 65

Question # 4.
Have you ever consumed any guava juice products?
O Yes
0 No

Question # 5.
How often do you consume guava beverages?
O More than once a day
O Once a day
O 2-3 times a week
O Once a week
O 2-3timesa month
O Once a month
O Twice a year
O Once a year
O Less than once a vear

Figure D-1. Questionnaire used for taste panel: different from control test, demographic questions


189









Take a bite of cracker and a sip of water to rinse your mouth.
Remember to do this before you taste each sample.
WHEN ANSWERING ANY QUESTION, MAKE SURE THE NUMBER ON THE CUP
MATCHES THE NUMBER ON THE MONITOR.

Please click on the 'Continue' button below.

Question # 6.
You are being presented with a reference sample marked 000. Please taste this sample and then taste the following samples
and compare them to the reference sample. Mark how different the sample is from the reference sample using the line scale
below.

Sample <>
Not Different Very Different

I I

Sample <>
Not Different Very Different

I I

Sample <>
Not Different Very Different


Take a bite of cracker and a sip of water to rinse your mouth.
Remember to do this before you taste each sample.
WHEN ANSWERING ANY QUESTION, MAKE SURE THE NUMBER ON THE CUP
MATCHES THE NUMBER ON THE MONITOR.

Please click on the 'Continue' button below.


Question # 7.
Please indicate how much you like the OVERALL TASTE of each sample.
Overall Taste
Sample <>
Dislike Dislike Dislike Dislike Neither like Like Like Like very
extremely very much moderately slightly nor dislike slightly moderately much


Like
extremely


I II I 4 I 5 6 I 7 I 8 9


Sample <>
Dislike Dislike Dislike
extremely very much moderately


Dislike Neither like Like Like Like very Like
slightly nor dislike slightly moderately much extremely


I 1 2 3 I 4 5 6 7 I 8 I 9


Sample <>
Dislike Dislike Dislike
extremely very much moderately


Dislike Neither like Like Like Like very Like
slightly nor dislike slightly moderately much extremely


I 1 2 W3 W4 5 6 7 W8 9

Figure D-2. Questionnaire used for taste panel: different from control test, sensory questions


190


I I










Question # 8


Using the keyboard located in the tray under the counter, please describe the
differences, if any, between the samples (please be specific).
Sample <>




Sample <>




Sample <>







The test has ended.


Figure D-3. Questionnaire used for taste panel: different from control test, comments











Table D-1. Volatile compounds, CAS number, identification method,
for previously reported studies


reported Linear Retention Indexes and references


Identification
Name CAS Number method LRI-DB5 LRI-Wax Previously reported


Acetaldehyde
hydrogen sulfide*
Methanethiol*
dimethyl sulfide*
Ethanol*
butanal
acetic acid*
ethyl acetate*
pentene-3-one*
ethyl propanoate
3-Hydroxy-2-Butanone*
1-butanol, 2-methyl*
1-butanol, 3-methyl*
dimethyl disulfide*
(Z)-3-hexenal*
hexanal*
ethyl-2-methyl butanoate
(E)-3-hexenol
(Z)-3-hexenol*
1-hexanol*
2-methyl-3-furanthiol*
1-butanol-3-methyl acetate*
methyl hexanoate
methyl-3-hexenoate
Methional*
4-mercapto-4-methyl-pentan-2-one*


75-07-0
77783-06-4
74-93-1
75-18-3
64-17-5
123-72-8
64-19-7
141-78-6
1629-58-9
105-37-3
513-86-0
137-32-6
123-51-3
624-92-0
6789-80-6
66-25-1
7452-79-1
928-97-2
928-96-1
111-27-3
28588-74-1
123-92-2
106-70-7
2396-78-3
3268-49-3
1987-52-7


GC-MS
S
S
S
GC-MS
GC-MS
GC-MS
GC-MS, 0
GC-MS, 0
GC-MS
GC-MS
GC-MS
GC-MS
S
GC-MS
GC-MS, 0
GC-MS
GC-MS, 0
GC-MS, 0
GC-MS
S
GC-MS
GC-MS
GC-MS
S,O
S


<500
<500
<500
519
537
518
600
600
697
714
718
733
739
740
793
796
846
851
855
865
870
876
906


700
528
675
736
926
813
1459
890
1032
949
1281
1209
1184
1064
1151
1099
1062
1388
1399
1364
1324
1110
1177


1450
1366


1, 4, 10, 12
13
10, 13
13
7


1, 9
1, 2, 4, 5, 6, 7, 9
1, 5,
1, 5,6,9
1, 5,6,9
9
9
5, 13
1, 5,6, 8, ,11, 12, 13
1, 2, 4, 5, 6, 7,8, 9, 11, 13
13
1,2,5,6
1, 2, 4, 5, 6, 10, 13
1, 2, 5, 6, 7, 8, 9
10


4, 5, 7, 8, 9
5
11, 13











Identification
Name CAS Number method LRI-DB5 LRI-Wax Previously reported


octen-3-ol*
Octen-3-one*
6-methyl-5-hepten-2-one*
2-octanone
Ethyl hexanoate*
(Z)-3-hexen-1-ol acetate
Hexyl acetate
d-limonene*
1,8-Cineole*
(E)-b-Ocimene*
Benzyl alcohol
4-mercapto-4-methyl pentan-2-ol*
Phenyl acetaldehyde*
Furaneol*
1-octanol*
Homofuraneol
Ethyl heptanoate*
Linalool*
Nonanal*
Methyl octanoate
Ethyl-3-hydroxyhexanoate*
3-mercapto-1- hexanol*
(E,Z)-2,6-nonadienal*
Ethyl benzoate
a-terpineol*
Ethyl octanoate*
Decanal*
Methyl nonanoate


3391-86-4
4312-99-6
110-93-0
111-13-7
123-66-0
1708-82-3
142-92-7
138-86-3
470-82-6
3779-61-1
100-51-6
31539-84-1
122-78-1
3658-77-3
111-87-5
27538-10-9
106-30-9
78-70-6
124-19-6
111-11-5
2305-25-1
51755-83-0
557-48-2
93-58-3
98-55-5
106-31-2
112-31-2
1731-84-6


GC-MS, O
GC-MS, O
GC-MS
GC-MS
GC-MS, O
GC-MS
GC-MS
GC-MS
GC-MS, O
GC-MS
0
O
S,O
0
O
GC-MS, O
GC-MS
0
O
GC-MS
GC-MS
GC-MS
GC-MS
GC-MS, O
S,O
GC-MS, O
GC-MS
GC-MS
GC-MS, O
GC-MS
GC-MS


978
980
982
997
998
1007
1008
1031
1032
1037
1039
1042
1047
1062
1068
1078
1095
1100
1103
1107
1126
1126
1153
1185
1193
1193
1203
1224


1452
1315
1347
1304
1246
1313
1267
1205
1231
1265
1893
1534
1675
2058
1575
2073
1346
1557
1416
1378
1694
1862
1611
1658
1724
1450
1523
1479


9
5, 10, 11
5, 9


1, 4, 5, 6,8, 9, 10
1, 2, 4, 5, 6, 7, 8, 9
4, 5, 8, 9
2, 3, 4, 5, 6, 9
6, 7, 9
5, 6, 7, 8, 9
1, 7
10


1, 5, 10, 11, 12
1, 2, 4, 5, 9, 10


10
1, 9, 10, 11


5


10, 11, 12
5, 10
1, 5,6,9
1,2,6
1, 4, 5, 6, 9, 10
5


193











Identification
Name CAS Number method LRI-DB5 LRI-Wax Previously reported


Ethyl-2-phenylacetate
Mercapto hexyl acetate*
Carvone*
3-phenylpropanol
Phenethyl acetate
Nonanoic acid
Ethyl nonanoate
1-p-Menthene-8-thiol*
a-cubebene
Ethyl-3-phenylpropionate
Eugenol
(Z)-3-hexenyl hexanoate
B-damascenone*
t-Caryophyllene*
a-Humulene
g-Decalactone
p-lonone*
p-Bisabolene*
beta or delta- cadinene
(E)- or d-nerolidol*
2-hidroxyethyl benzene
3-phenyl propyl acetate


101-97-3
1369-54-20-6
99-49-0
122-97-4
103-45-7
112-05-0
123-29-5
71159-90-5
17699-14-8
2021-28-5
97-53-0
53398-86-0
23726-93-4
87-44-5
6753-98-6
706-14-9
79-77-6
495-61-4
483-76-1
142-50-7
60-12-8
122-72-5


GC-MS, O
S
GC-MS, O
GC-MS
GC-MS
GC-MS
GC-MS
S,O
GC-MS
GC-MS, O
GC-MS
GC-MS, O
GC-MS
GC-MS
GC-MS
GC-MS, O
GC-MS, O
GC-MS
GC-MS
GC-MS
GC-MS
GC-MS


1243
1244
1249
1252
1260
1275
1280
1295
1345
1351
1357
1381
1390
1432
1454
1472
1496
1498
1524
1564
1118
1347


1785
1725
1779
1993
1785
2110
1523
1507
1482


2187
1654
1860
1641
1650
2132
1984
1735
1794
2047
1859
1941


1,4,5,9, 12
11, 12
10
1, 5,9
1,2
5


8
10
1, 5, 10
5
10, 13
1, 2, 3, 4, 5, 6, 8, 9, 10, 13
1, 3,4,8
1, 5, 11
1,2, 10
1, 3,4, 8, 13
6, 8
1, 8
5

1, 5, 6,8


1= (Nishimura and others 1989), 2= (Stevens and others 1970), 3= (Wilson and Shaw 1978), 4= (Macleod and Gonzalez de Troconis 1982), 5=(ldstein and
Schreier 1985), 6= (Chyauand others 1992), 7=(Yen and Lin 1999), 8= (Paniandy and others 2000), 9= (Jordan and others 2003), 10= (Mahattanatawee and
others 2005), 11= (Steinhaus and others 2008), 12= (Steinhaus and others 2009) and 13= (Rouseff and others 2008). *Confirm with authentic standard, bold
compounds= identify for first time in guava puree, S= LRI from GC-PFPD, O = LRI from GC-O









LIST OF REFERENCES


American Official Analytical Chemist (AOAC). 1990. American Official Analytical
Chemist Official Method of Analysis. 9th ed. AOAC method 967.21, Washington
DC.

Arreola AG, Balaban MO, Marshall M, PeplowAJ, Wei CI & Cornell J. 1991.
Supercritical carbon dioxide effects on some quality attributes of single strength
orange juice. J Food Sci 56(4):1030-3.

Balaban MO, Arreola AG, Marshall M, Peplow A, Wei CI, Cornell J. 1991. Inactivation of
pectinesterase in orange juice by supercritical carbon dioxide. J Food Sci
56(3):743-6.

Balaban MO, Ferrentino G, Ramirez M, Plaza ML. 2008. Review of dense phase carbon
dioxide application to citrus juices. 54th Annual Citrus Engineering Conference;
2008 March 4; Ben Hill Griffin Jr. Citrus Hall Citrus Research & Education Center
Lake Alfred, Florida: American Society of Mechanical Engineers, Florida Section.

Ballestra P, DaSilva AA, Cuq JL. 1996. Inactivation of Escherichia coli by carbon
dioxide under pressure. J Food Sci 61(4):829-31.

Bloch A, Thomson CA. 1995. Position of the American Dietetic Association:
Phytochemicals and functional foods. J Am Diet Assoc 95(4):493-96.

Brasil IM, Maia GA, de Figueiredo RW. 1995. Physical-chemical changes during
extraction and clarification of guava juice. Food Chem 54(4):383-86.

Bravo L, Abia R, Saura-Calixto F. 1994. Polyphenols as dietary fiber associated
compounds. Comparative study on in vivo and in vitro properties. J Agric Food
Chem 42(7):1481-7.

Brekkee JE, Redelinghuys HP, Torline PA. 1986. Enzyme treatment and concentration
of guava pulp. Acta hortic 194:229-40.

Bruyne TD, Pieters L, Deelstra H, Vlietinck A. 1999. Condensed vegetable tannins:
Biodiversity in structure and biological activities. Biochem Syst Ecol 27(4):445-
59.

Buettner A, Schieberle P. 1999. Characterization of the most odor-active volatiles in
fresh, hand-squeezed juice of grapefruit (Citrus paradisi Macfayden). J Agric
Food Chem 47(12):5189-93.

Butz P, Tauscher B. 2002. Emerging technologies: chemical aspects. Fod Res Int 35(2-
3):279-84.


195









Calix TF, Ferrentino G, Balaban MO. 2008. Measurement of high-pressure carbon
dioxide solubility in orange juice, apple juice, and model liquid foods. J Food Sci
73(9): E439-45.

Carasek E, Pawliszyn J. 2006. Screening of tropical fruit volatile compounds using
solid-phase microextraction (SPME) fibers and internally cooled SPME fiber. J
Agric Food Chem 54(23):8688-96.

Carlton-Tohill B. 2007. Fruits and vegetables consumption and body weight
management. ISHS Acta hortic (744):39-46.

Chan HT, Cavaletto CG. 1982. Aseptically packaged papaya and guava puree: changes
in chemical and sensory quality during processing and storage. J Food Sci
47(4): 1164-9.

Chan HT, Brekke JE, Chang T. 1971. Nonvolatile organic acids in guava. J Food Sci
36(2):237-9.

Chan HT, Kwok SCM. 1975. Identification and determination of sugars in some tropical
fruit products. J Food Sci 40(2):419-20.

Cheynier V. 2005. Polyphenols in foods are more complex than often thought. Am J Clin
Nutr 81(1):223S-9.

Ching LS, Mohamed S. 2001. Alpha-tocopherol content in 62 edible tropical plants. J
Agric Food Chem 49(6):3101-5.

Chopda CA, Barrett DM. 2001. Optimization of guava juice and powder production. J
Food Process Pres 25(6):411-30.

Christen GL & Smith JS. 2000. Food chemistry: principles and applications. West
Sacramento, CA: Science Technology System.

Chyau CC, Chen SY, Wu CM. 1992. Differences of volatile and nonvolatile constituents
between mature and ripe guava (Psidium guajava Linn) fruits. J Agric Food
Chem 40(5):846-9.

Clark P. 2009. Case studies in food engineering. New York: Springer. 224 p.

Cook NC, Samman S. 1996. Flavonoids-Chemistry, metabolism, cardioprotective
effects, and dietary sources. J Nutr Biochem 7(2):66-76.

Commodity market review: 2003-2004 [Internet]. Food and Agriculture Organization;
2003 [Accessed 2010 Jan 3]. Available from: http://www.fao.org/documents/


196









Craig WJ. 1997. Phytochemicals: guardians of our health. J Am Diet Assoc 97(10,
Supplement 1):S 199-204.

Current situation and medium-outlook for tropical fruits [Internet]. Food and Agriculture
Organization; 2005 [Accessed 2010 Jan 3]. Available from:
http://www.fao.org/es/esc/common/ecg/218/en/Sit_web_e.pdf

Dagan GF, Balaban MO. 2006. Pasteurization of beer by a continuous dense-phase
CO2 system. J Food Science 71(3):E164-9.

Damar S. 2006. Processing of coconut water with high pressure carbon dioxide
technology. [DPhil dissertation]. Gainesville, FL: University of Florida. 175 p.

Damar S, Balaban MO. 2006. Review of dense phase CO2 technology: microbial and
enzyme inactivation, and effects on food quality. J Food Science 71(1):R1-11.

Damar S, Balaban MO, Sims CA. 2009. Continuous dense-phase CO2 processing of a
coconut water beverage. Int J Food Sci Tech 44(4):666-73.

Damodaran S, Parkin KL, Fennema OR. 2008. Fennema's food chemistry. 4th ed.
Boca Raton FL: CRC Press. 1144 p.

Del Pozo-lnsfran D, Balaban MO, Talcott ST. 2006. Microbial stability, phytochemical
retention, and organoleptic attributes of dense phase CO2 processed muscadine
grape juice. J Agric Food Chem 54(15):5468-73.

Demole E, Enggist P & Ohloff G. 1982. 1-p-Menthene-8-thiol: A powerful flavor impact
constituent of grapefruit juice (Citrus parodisi). Helv Chim Acta 65(6):1785-94.

Dietary guidelines for Americans: US Department of Health and Human Services.
Washington DC: US Goverment Printing Office; 2005. [Accessed 2010 Jan 3].
Available from: www.health. gov/dietaryguidelines.

Dixon RA, Xie D-Y, Sharma SB. 2005. Proanthocyanidins: a final frontier in flavonoid
research? New Phytol 165(1):9-28.

Dweck AC. 2005. A Review of guava (Psidium guajava) Personal Care Magazine. p.
33-36.

ElBulk RE, Babiker EFE, EITinay AH. 1997. Changes in chemical composition of guava
fruits during development and ripening. Food Chem 59(3):395-9.

E- Tropical Fruit Net: International Tropical Fruit Network (TFNET). 2009 [Accessed
2010 Jan 3]. Available from: http://www.itfnet.org/e-Newsletter/J uly2009/









Fender L. 2005. Phytochemical, antioxidant, and storage stability of thermally
processed guava (Psidium guajava) and guava juice blends. [MSc thesis]
Gainesville, FL: University of Florida. 79 p.

Ferrentino G, Barletta D, Balaban MO, Ferrari G, Poletto M. 2010. Measurement and
prediction of CO2 solubility in sodium phosphate monobasic solutions for food
treatment with high pressure carbon dioxide. J Super Crit Fluid 52(1): 142-50.

Ferrentino G, Plaza ML, Ramirez-Rodrigues M, Ferrari G, Balaban MO. 2009. Effects of
dense phase carbon dioxide pasteurization on the physical and quality attributes
of a red grapefruit Juice. J Food Sci 74(6):E333-41.

Flavor Database. Lake Alfred, Fl: University of Florida, Citrus Research and Education
Center, Rouseff R; c2006-2009 [Accesed 2009 Oct 20]. Available from
http://www.crec.ifas.ufl.edu/crec_websites/Rouseff/#.

Flavornet and human odor space [Internet]. Geneva, NY: Cornell University, Acree T &
Arn H; 2004 [Accessed 2009 Oct 20]. Available from: http://www.flavornet.org.

Fisher C, Scott TR. 1997. Food flavours: biology and chemistry. Cambridge, UK: The
Royal Society of Chemistry. 165 p.

Gary RB. 1999. Phytonutrients role in metabolism: effects on resistance to
degenerative processes. Nutr Rev 57(9):3-6.

Gokmen V, Kahraman N, Demir N, Acar J. 2000. Enzymatically validated liquid
chromatographic method for the determination of ascorbic and dehydroascorbic
acids in fruit and vegetables. J Chromatogr A 881(1-2):309-16.

Gomes MRA, Ledward DA. 1996. Effect of high-pressure treatment on the activity of
some polyphenoloxidases. Food Chem 56(1) 1-5.

Gorinstein S, Zemser M, Haruenkit R, Chuthakorn R, Grauer F, Martin-Belloso O,
Trakhtenberg S. 1999. Comparative content of total polyphenols and dietary fiber
in tropical fruits and persimmon. J Nutr Biochem 10(6):367-71.

Grundhofer P, Niemet R, Schilling G, Gross GG. 2001. Biosynthesis and subcellular
distribution of hydrolyzable tannins. Phytochemistry 57(6):915-27.

Gunes G, Blum LK, Hotchkiss JH. 2005. Inactivation of yeasts in grape juice using a
continuous dense phase carbon dioxide processing system. J Sci Food Agr
85(14):2362-8.


198









Hawaii guavas: National Agricultural Statistics Services (NASS) [Internet]. 2009
[Accessed 2010 Jan 3]. Available from:
http://www.nass.usda.gov/Statistics_by_State/Hawaii/Publications/Fruits_and_N
uts/guava.pdf.

Ho C-T, Lee CY, Huang M-T. 1992. Phenolic compounds in food and their effects on
health I: analysis, ocurrence & chemistry. Washington, DC: American Chemical
Society. 338 p.

Hodgson AS, Chan HT, Cavaletto CG, Perera CO. 1990. Physical-chemical
characteristics of partially clarified guava juice and concentrate. J Food Sci
55(6): 1757-8.

Huang D, Boxin O, Prior RL. 2005. The chemistry behind antioxidant capacity assays. J
Agric Food Chem 53:1841-56.

Huang M-T, Ho C-T, Lee CY. 1992. Phenolic compounds in food and their effects on
health II: antioxidants and cancer prevention. Washington, DC: American
Chemical Society. 338 p.

Idstein H, Schreier P. 1985. Volatile constituents from guava (Psidium guajava, L.) fruit.
J Agric Food Chem 33(1):138-43.

Imungi JK, Scheffeldt P, Saint-Hislare P. 1980. Physical-chemical changes during
extraction and contraction of clear guava juice. Lebensmit. Wiss Technol 13:248-
51.

Janes D, Kantar D, Kreft S, Prosen H. 2009. Identification of buckwheat (Fagopyrum
esculentum Moench) aroma compounds with GC-MS. Food Chem 112(1): 120-4.

Jennings W, Shibamoto T. 1980. Qualitative analysis of flavor and fragance volatiles by
glass capillary gas chromatography. New York, NY: Academic Press. 472 p.

Jimenez-Escrig A, Rincon M, Pulido R, Saura-Calixto F. 2001. Guava fruit (Psidium
guajava L.) as a new source of antioxidant dietary fiber. J Agric Food Chem
49(11):5489-93.

Jordan MJ, Margaria CA, Shaw PE, Goodner KL. 2003. Volatile components and aroma
active compounds in aqueous essence and fresh pink guava fruit puree (Psidium
guajava L.) by GC-MS and multidimensional GC/GC-O. J Agric Food Chem
51(5): 1421-26.

Kashyap DR, Vohra PK, Chopra S, Tewari R. 2001. Applications of pectinases in the
commercial sector: a review. Bioresource Technol 77(3):215-27.


199









Kataoka H, Lord HL, Pawliszyn J. 2000. Applications of solid-phase microextraction in
food analysis. J Chromatogr A 880(1-2):35-62.

Khanbabaee K, Ree Tv. 2001. Tannins: classification and definition. Nat Prod Rep
18:641-9.

Kincal D, Hill WS, Balaban M, Portier KM, Sims CA, Wei CI, Marshall MR. 2006. A
Continuous high-pressure carbon dioxide system for cloud and quality retention
in orange juice. J Food Science 71(6):C338-44.

Kincal D, Hill WS, Balaban MO, Portier KM, Wei CI, Marshall MR. 2005. A Continuous
high pressure carbon dioxide system for microbial reduction in orange juice. J
Food Science 70(5):M249-54.

Kondo S, Kittikorn M, Kanlayanarat S. 2005. Preharvest antioxidant activities of tropical
fruit and the effect of low temperature storage on antioxidants and jasmonates.
Postharvest Biol Tec 36(3):309-18.

Laura B. 1998. Polyphenols: chemistry, dietary sources, metabolism, and nutritional
significance. Nutr Rev 56(11):317-33.

Lecky M. 2005. Continuous high pressure carbon dioxide processing of watermelon
juice. [MSc thesis] Gainesville, FL: University of Florida. 109 p.

Leong LP, Shui G. 2002. An investigation of antioxidant capacity of fruits in Singapore
markets. Food Chem 76(1):69-75.

Lim S, Yagiz Y, Balaban MO. 2006. Continuous high pressure carbon dioxide processing
of manadarin juice. Food Sci Biotech 15(1):13-8.

Lord H & Pawliszyn J. 2000. Evolution of solid-phase microextraction technology. J
Chromatogr A 885(1-2): 153-93.

Macleod AJ & Gonzalez de Troconis N. 1982. Volatile flavour components of guava.
Phytochemistry 21(6): 1339-1342.

Mahattanatawee K, Goodner KL, Baldwin EA. 2005. Volatile constituents and character
impact compounds of selected Florida's tropical fruit Proc. Fla. State Hort. Soc
118:414-8.

Mathews-Roth MM, Krinsky NI. 1985. Carotenoid dose level and protection against UV-
B induced skin tumors. Photochem Photobiol 42(1):35-8.

Meilgaard M, Civille GV, Carr BT. 2007. Sensory evaluation techniques. 4th ed. Boca
Raton, FL: Taylor & Francis Group. 448 p.


200









Mercadante AZ, Steck A, Pfander H. 1998. Carotenoids from guava (Psidium guajava
L.): isolation and structure elucidation. J Agric Food Chem 47(1):145-51.

Mercado-Silva E, Benito-Bautista P, Garcia-Velasco MA. 1998. Fruit development,
harvest index and ripening changes of guavas produced in central Mexico.
Postharvest Biol Tec 13(2):143-50.

Miean KH, Mohamed S. 2001. Flavonoid (myricetin, quercetin, kaempferol, luteolin, and
apigenin) content of edible tropical plants. J Agric Food Chem 49(6):3106-12.
Misra K, Seshadri TR. 1968. Chemical components of the fruits of Psidium guava.
Phytochemistry 7(4):641-5.

Mitra S. 1997. Postharvest physiology and storage of tropical and subtropical fruits. 1st
ed. New York, NY: Cab International. 423 p.

Morton J. 1987. Fruits of warm climates. 1st ed. Miami, FL: Julia Morton. 48 p.

Mowlah G, Itoo S. 1982. Guava (Psidium guajava L.) sugar components and related
enzymes at stages of fruit development and ripening. Nippon Shokuhin Kogyo
Gakkaishi (J Jpn Soc Food Sci Technol) 29(8):472-6.

Mowlah G, Itoo S. 1983. Changes in pectic components, ascorbic acid, pectic enzymes
and cellulase activity in ripening and stored guava (Psidium guajava L.). Nippon
Shokuhin Kogyo Gakkaishi (J Jpn Soc Food Sci Technol) 30(8):454-61.

Murano PS. 2003. Understanding food science and technology. Belmont, CA:
Wadsworth/Thomson Learning. 449 p.

Naczk M, Shahidi F. 2004. Extraction and analysis of phenolics in food. J Chromatogr A
1054(1-2):95-111.

Nakasone HY, Paull RE. 1998. Tropical fruits. New York, NY: Cab International. 472 p.

Nielsen SS. 2003. Food analysis. 3rd ed. New York, NY: Kluwer Academic/ Plenum
Publishers. 557 p.

Nishimura O, Yamaguchi K, Mihara S & Shibamoto T. 1989. Volatile constituents of
guava fruits (Psidium guajava L.) and canned puree. J Agric Food Chem
37(1): 139-142.

Nunez-Rueda F. 2005. Guava (Psidium guajava L.) fruit phytochemicals, antioxidant
properties and overall quality as influenced by postharvest treatments. [MSc
thesis] Gainesville, FL: University of Florida. 89 p.


201









Paniandy JC, Chane-Ming J, Pieribattesti JC. 2000. Chemical composition of the
essential oil and headspace solid-phase microextraction of the guava fruit
(Psidium guajava L.). J Essen Oil Res 12(2):153-8.

Pietta P-G. 2000. Flavonoids as antioxidants. J Nat Prod 63(7):1035-42.

Rahmat A, Bakar MFA, Hambali Z. 2006. The effects of guava (Psidium guajava)
consumption on total antioxidant and lipid profile in normal male youth. Afr J
Food Agric Nut Develop 6(2):1-12.

Randall GC, Robert AB, Karel G. 1998. Multiple forms of pectinmethylesterase from
citrus peel and their effects on juice cloud stability. J Food Sci 63(2):253-6.

Raso J, Barbosa-Canovas GV. 2003. Nonthermal preservation of foods using combined
processing techniques. Crit Rev Food Sci 43(3):265 85.

Redd J, Hendrix C, Hendrix D. 1986. Quality control manual for citrus processing plants.
Safety Harbor, FL: Intercit Inc. 180 p.

Reineccius G. 2006. Flavor chemistry and technology. 2nd ed. Boca Raton, FL: Taylor
and Francis. 489 p.

Robbins RJ. 2003. Phenolic acids in foods: an overview of analytical methodology. J
Agric Food Chem 51(10):2866-87.

Rouse AH, Atkins CD. 1955. Pectinestersase and pectin in commercial orange juice as
determined by methods used at the Citrus Experiment Station. Bulletin Agric Exp
Station (Florida) 570:1-19.

Rouseff RL, Onagbola EO, Smoot JM, Stelinski LL. 2008. Sulfur volatiles in guava
(Psidium guajava L.) leaves: possible defense mechanism. J Agric Food Chem
56(19):8905-10.

Rowe D. 2000. More fizz for your buck high-impact aroma chemicals. Perfumer &
Flavorist Magazine. p. 1-19.

Salunkhe DK, Kadam SS. 1995. Handbook of fruit science and technology: production,
composition storage and processing. New York, NY: Marcel Dekker Inc. 611 p.

Sandhu K, Bhatia BS. 1985. Physico-chemical changes during preparation of fruit juice
concentrate. J Food Sci Tech 22:202-6.

Saura-Calixto F. 1998. Antioxidant dietary fiber product: a new concept and a potential
food ingredient. J Agric Food Chem 46(10):4303-6.


202









Scalbert A, Johnson IT, Saltmarsh M. 2005. Polyphenols: antioxidants and beyond. Am
J Clin Nutr 81(1):215S-7.

Shahidi F, Naczk M. 2003. Phenolics in food and nutraceuticals. Boca Raton, FL: CRC
Press. 558 p.

Sonko L, Xia W. 2005. Food phenolics, pros and cons: a review. Food Rev Int
2005(4):367-88.

Spanier AM, Shahidi F, Parliment TH, Mussinan C, Ho C-T, Contis UT. 2001. Food
flavors and chemistry: advancements of the new millennium. Cambridge, UK:
The Royal Society of Chemistry. 666 p.

Spilimbergo S, Bertucco A. 2003. Non-thermal bacterial inactivation with dense CO2.
Biotechnol Bioeng 84(6):627-38.

Steinhaus M, Sinuco D, Polster J, Osorio C, Schieberle P. 2008. Characterization of the
aroma-active compounds in pink guava (Psidium guajava, L.) by application of
the aroma extract dilution analysis. J Agric Food Chem 56(11):4120-7.

Steinhaus M, Sinuco D, Polster J, Osorio C, Schieberle P. 2009. Characterization of the
key aroma compounds in pink guava (Psidium guajava L.) by means of aroma
re-engineering experiments and omission tests. J Agric Food Chem 57(7):2882-
8.

Steven KC. 1998. Lycopene: chemistry, biology, and implications for human health and
disease. Nutr Rev 56(2):35-51.

Stevens KL, Brekke JE, Stern DJ. 1970. Volatile constituents in guava. J Agric Food
Chem 18(4):598-9.

Sun N, Lee S, Song KB. 2002. Effect of high-pressure treatment on the molecular
properties of mushroom polyphenoloxidase. Lebensmittel-Wissenschaft und-
Technologie 35(4):315-318.

Supelco. 2005. Solid phase microextraction: theory and optimization of conditions.
Supelco Bulletin 923.

Talcott ST, Howard LR, Brenes CH. 2000. Antioxidant changes and sensory properties
of carrot puree processed with and without periderm tissue. J Agric Food Chem
48(4):1315-21.

Temple NJ. 2000. Antioxidants and disease: more questions than answers. Nutr Res
20(3):449-59.


203









The world fresh food market: Kipe Sandra [Internet]. USDA Foreign Agricultural Service.
Horticultural and Tropical Products Division. 2004 [Accessed 2010 Jan 3].
Available from
http://www.fas.usda.gov/htp/Presentations/2004/World%20Fresh%20Fruit%20M
arket%20%2808-04%29. pdf.

Tribess TB, Tadini CC. 2006. Inactivation kinetics of pectin methylesterase in orange
juice as a function of pH and temperature/time process conditions. J Sci Food
Agr 86(9): 1328-35.

Vitamin E: National Institute of Health [Internet]. Office of Dietary Supplement; 2009
[Updated 2009 Dec 15; Accessed 2010 Jan 3]. Available from:
http://ods.od.nih.gov/factsheets/Vitami nE. asp.

Versteeg C, Rombouts FM, Spaansen C, PilnikW. 1980. Thermostability and orange
juice cloud destabilizing properties of multiple pectinesterases from orange. J
Food Sci 45:969-72.

Wilberg VC, Rodriguez-Amaya DB. 1995. HPLC quantitation of major carotenoids of
fresh and processed guava, mango and papaya. Lebensmittel-Wissenschaft und-
Technologie 28(5):474-80.

Wilson CW, Shaw PE. 1978. Terpene hydrocarbons from Psidium guajava.
Phytochemistry 17(8): 1435-6.

Wilson CW, Shaw PE, Campbell CW. 1982. Determination of organic-acids and sugars
in guava (Psidium-Guajava L) cultivars by high-performance liquid-
chromatography. J Sci Food Agr 33(8):777-80.

Wrolstad RE, Acree TE, Decker EA, Penner MH, Reid DS, Schwartz SJ, Shoemaker
CF, Smith DM, Sporns P. 2005. Handbook of food analytical chemistry.
Hoboken, New Jersey: John Wiley & Sons. 606 p.

Yen GC, Lin HT. 1996. Comparison of high pressure treatment and thermal
pasteurization effects on the quality and shelf life of guava puree. International J
Food Sci Tech 31(2):205-13.

Yen GC, Lin HT. 1999. Changes in volatile flavor components of guava juice with high-
pressure treatment and heat processing and during storage. J Agric Food Chem
47(5):2082-7.

Yen GC, Ling HT, Yang P. 1992. Changes in volatile flavor components of guava puree
during processing and frozen storage. J Food Sci 57(3):679-81.

Zabetakis I, Leclerc D, Kajda P. 2000. The effect of high hydrostatic pressure on the
strawberry anthocyanins. J Agric Food Chem 48(7):2749-54.


204









BIOGRAPHICAL SKETCH

Maria de L. Plaza-Delestre received her bachelor's degree from University of

Puerto Rico, Mayaguez Campus, majoring in biology with a minor in microbiology. While

finishing her undergraduate carrier, she got very interested in food science. After

obtaining her bachelor's degree, she decided to start her master's degree at the same

University. She earned a Master of Science in food science and technology in 2002.

After four years of experience as a research assistant and instructor at the Food

Science and Technology program of the University of Puerto Rico, she was offered the

opportunity to pursue her Ph.D. degree. In August 2005, she started the graduate

program in the Food Science and Human Nutrition Department at the University of

Florida. Under Dr. Maurice Marshall's supervision, Maria received her Ph.D. in food

science and human nutrition from the University of Florida in the summer 2010 and

graduated with a GPA of 4.0


205





PAGE 1

1 QUALITY OF GUAVA PUREE BY DENSE PHASE CARBON DIOXIDE TREATMENT By MARIA L. PLAZA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 Maria L. Plaza

PAGE 3

3 To GOD for giving me the most supportive parents in the world, the most beautiful kids (Erick and Amalia), and an unbelievable strength to finish this stage of my life

PAGE 4

4 ACKNOWLEDGMENTS First, I would like to extend my special and deepest thanks to my advisors, Dr. Maurice Marshall and Dr. Murat Balaban for their support, advice and friendship. Thanks to Dr. Russell Rouseff for his invaluable help, assistance and time for my research, thanks for letting me be part of your laboratory and for sharing all your knowledge. Through their guidance, wisdom, and never ending care; Dr. Marshall, Dr. Balaban and Dr Rouseff; they have helped me to achieve all my goals. I sincerely appr eciate the help offered by Dr. Bala Rathinasabapathi as member of my supervising committee. I would also like to acknowledge the unconditional support of Dr. Charles Sims, his students and the taste panel personnel (Lorenzo and Asli). My deepest thanks g o to my parents, Wilda and Jose. I would have not be where I am without the unconditional love, support, and advice of my parents. They have been an inexhaustible source of love, inspiration, and encouragement throughout all my life. I thank my sister and best friend Becky, for her advice and big sister support during all these years. Big thanks go to my brother Pepe, Freddy and both of my sisters -in law (Nana and Malu) for their warm support and example of perseverance and love. I thank Erick, for giving m e the best gift on earth, my kids. I want to give my deepest and more sincere thanks to two very special persons in my life, my kids Erick and Amalia. I thank them for their understanding and for cheering me up when I was down. These wonderful persons have a very special place in my heart. I thank my lab partners and friends, Milena and Alberto, for their never ending assistance in my project and for making the work in the lab an enjoyable experience. Special thanks go to Sarah, Stefan, Max, Diana and Giovanna. All of them formed part

PAGE 5

5 of my student life and helped me pass through this important stage in the most enjoyable way. Thanks for just adding happiness to my graduate school days. I thank my band friends, Frank and Cindy, for their help and support during these last two years. To that person that does not need to be mentioned, thanks for your unconditional support since I started this journey and for always believing in me. Last but not least, Daniel thanks for your support, help and being there when I needed you. All of you made my life in Gainesville an enjoyable one filled with joy and love.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES .............................................................................................................. 10 LIST OF FIGURES ............................................................................................................ 12 ABSTRACT ........................................................................................................................ 14 CHAPTER 1 INTRODUCTION ........................................................................................................ 17 2 LITERATU RE REVIEW .............................................................................................. 19 G uava (Psidium guajava L.) ....................................................................................... 19 G uava F ruit and Fruit P roducts: W orld P roduction .................................................... 19 C haracteristics ...................................................................................................... 20 C hemical C omposition ......................................................................................... 21 G uava Constituents and Health B enefits ............................................................ 23 Guava F lavors ...................................................................................................... 29 Guava Fruit P rocessing ....................................................................................... 35 G uava Puree P rocessing ..................................................................................... 37 Phytochemi cals ........................................................................................................... 40 Classification ........................................................................................................ 41 Attributes .............................................................................................................. 44 E xtraction and A nalysi s ........................................................................................ 44 Sensory E valuation and F lavor A nalysis ................................................................... 45 S ensory E valuation .............................................................................................. 45 F lavor A nalysi s ..................................................................................................... 47 Flavor E xtractio n T echniques .............................................................................. 48 Volatile Identification T echniques ........................................................................ 50 S eparation and D etection of Aroma V olatiles ..................................................... 51 B everage Processing ................................................................................................. 52 T hermal P rocessing ............................................................................................. 53 N on -thermal Processing ...................................................................................... 54 D ense Phase Carbon D ioxide .................................................................................... 54 M echanisms of Microbial I nactivation ................................................................. 54 F actors Affecting Microbial I nactivation ............................................................... 56 S olubility of CO2 ............................................................................................ 57 T ypes of s ystems ........................................................................................... 57 Food applications and effect on quality ........................................................ 58 O bjectives of S tudy ..................................................................................................... 62

PAGE 7

7 3 EFFECT OF ENZYME TREATMENT ON PHYSICOCHEMICAL AND PHYTOCHEMICAL PROPER TIES OF GUAVA PUREE .......................................... 64 Abstract ....................................................................................................................... 64 Introduction ................................................................................................................. 65 Materials and Methods ............................................................................................... 66 Preliminary Study ................................................................................................. 66 Selection of an enzyme and a temperature to clarify the guava puree ... 6 6 Effects of enzyme t reatment at low temperature on phytochemical levels in g uava ( Psidium guajava) p uree. ................................................. 67 Sample preparation ....................................................................................... 67 Analysis .......................................................................................................... 68 Enzyme T reatment Optimiz ation at Low Temperatures to Produce a C larified G uava ( Psidium guajava ) J uice ......................................................... 68 Sample preparation and enzyme treatment ................................................. 68 Physicochemical analysis ............................................................................. 68 Statistical A nalysis ............................................................................................... 70 Results and Discussion .............................................................................................. 71 Selection of Enzyme and T emperature ............................................................... 71 Enzyme Treatment Optimization at Low Temperatures to Produce a Clarified Guava ( Psidium guajava ) Juice without A ffecting its Phytochemical Composition ............................................................................. 72 Enzyme Treatment Optimization at Low Temperatures to Produce a Clarified Guava ( Psidium guajava ) Juice without A ffecting it Phytochemical Composition ............................................................................. 76 Conclusions ................................................................................................................ 83 4 INFLUENCE OF DENSE P HASE CARBON DIOXIDE AND PASTEURIZATION TREATMENTS ON THE ST ORAGE QUALITY OF GUA VA PUREE ....................... 94 Abstract ....................................................................................................................... 94 Introduction ................................................................................................................. 94 Materials and Methods ............................................................................................... 97 Guava Puree ........................................................................................................ 97 Model S ystem ....................................................................................................... 97 Solubility M easurements ...................................................................................... 97 Processing Equipment ......................................................................................... 98 Microbial Inactivation Study ................................................................................. 99 Storage Study and Microbial Stability ............................................................... 100 Chemical Analyses ............................................................................................. 100 Results and Discussion ............................................................................................ 102 Solubility Experiments ........................................................................................ 102 Microbial Inactivation Study ............................................................................... 103 St orage Study and Microbial S tability ............................................................... 106 Conclusions .............................................................................................................. 111

PAGE 8

8 5 PHYSICO -CHEMICAL AND PHYTOCHEMICAL CHANGES OF DENSE PHASE CARBON DIOXIDE AND THERMALLY TREATED GUAVA PUREE DURING REFRIGERATED STORAGE ................................................................... 122 Abstract ..................................................................................................................... 122 Introduction ............................................................................................................... 123 Materials and Methods ............................................................................................. 126 Total Phenolics, Antioxidant Capacity and Ascorbic Acid Analysis ................. 126 Titratable Acidity (TA) ........................................................................................ 127 Color Analysis .................................................................................................... 127 High Performance Liquid Chromatography Analysis ........................................ 127 Results and Discussion ............................................................................................ 129 Total Phenolics and Antioxidant Capacity ......................................................... 129 Titratable Acidity (TA) ........................................................................................ 131 Ascorbic Acid (Vitamin C) Content .................................................................... 132 Color ................................................................................................................... 132 Organic Acid Content ......................................................................................... 134 Phenolic C ompounds ......................................................................................... 134 Conclusions .............................................................................................................. 137 6 CHANGES IN AROMA COM POUNDS AND SENSORY PERCEPTION IN GUAVA PUREE AFTER TH ERMAL AND NON-THERMAL PROCESSING ......... 147 Abstract ..................................................................................................................... 147 Introduction ............................................................................................................... 147 Materials and Methods ............................................................................................. 149 Guava Puree ...................................................................................................... 149 Sensory Evaluation ............................................................................................ 149 Statistical Analysis ............................................................................................. 150 Extraction of Volatile C ompounds using Headspace SPME ............................ 150 Gas Chromatography Olfactometry Analysis (GC -O) ..................................... 151 Gas Chromatography Mass Spectrometry Analysis (GC MS) ........................ 152 Sulfur Compounds I dentification using Gas Chromatography Pulse Flame Photometric Detector (GC PFPD) .................................................................. 153 Identification Procedures ................................................................................... 153 Results and Discussion ............................................................................................ 154 Sensory Analysis ................................................................................................ 154 Flavor Analysis ................................................................................................... 155 Guava Volatile Composition .............................................................................. 156 GC -MS Identifications ........................................................................................ 157 GC -O Aroma Profiles ......................................................................................... 160 GC -PFPD Analysis ............................................................................................. 162 Conclusion ................................................................................................................ 164 7 SUMMARY AND CONCLUSIONS ........................................................................... 172

PAGE 9

9 APPENDIX A ENZYME TRATMENT OF G UAVA PUREE ............................................................ 176 B DP -CO2 PROCESSING AND DATA ........................................................................ 180 C MICROBIAL INACTIVATION AND STORAGE STUDY DATA AND ANALYSIS ... 185 D C HANGES IN AROMA COMPOUNDS AND SENSORY PERCEPTION IN GUAVA PUREE AFTER TH ERMAL AND NON-THERMAL PROCESSING ......... 189 LIST OF REFERENCES ................................................................................................. 195 BI OGRAPHICAL SKETCH .............................................................................................. 205

PAGE 10

10 LIST OF TABLES Table page 3 -1 Enzyme and concentration used during enzymatic treatment .............................. 84 3 -2 Percent yield, ORAC value, total soluble phenolics and ascorbic acid content of guava puree ........................................................................................................ 86 3 -3 Color values obtained for three different guava puree before and after clarification .............................................................................................................. 86 3 -4 Total soluble solids for guava puree before and after clarification for the three different treatments ................................................................................................ 86 3 -5 Physicochemical results for enzymatic treatment of guava puree at three different concentrations and three different reaction times ................................... 87 3 -6 Color results for enzymatic treatment of guava puree at three different concentrations and three different reaction times ................................................. 89 4 -1 pH, oBrix and titraacidity values for the treated and untreated guava puree under different processing pressures .................................................................. 113 4 -2 Processing conditions and microbial reduction obtained during DP -CO2 process optimization ............................................................................................ 114 4 -3 Actual and predicted yeast and mold log reduction using the equation from the surface response analysis ............................................................................. 114 4 -4 Pectinesterase activity, cloud and pH measurement of guava puree before and after DP -CO2 treatment ................................................................................. 115 4 -5 oBrix and titrabable acidity of guava puree before and after DP CO2 treatment 116 4 -6 pH, oBrix and cloud measurement for control, DP -CO2 and thermal treated guava during 14 weeks of storage ....................................................................... 121 5 -1 Antioxidant capacity and total soluble phenolics of fresh, DP -CO2 and thermal treated guava puree during storage ....................................................... 138 5 -2 L*, a* and b* values for control, DP -CO2 and thermal treated guava purees during 14 weeks of refrigerated storage .............................................................. 141 5 -3 Oxalic acid (OA), malic acid (MA), and citric acid (CA) content of control, DP CO2 and thermal treated guava purees during 14 weeks of refrigerated storage at 4oC ....................................................................................................... 142

PAGE 11

11 5 -4 Identification of guava polyphenolics at 260 and 280 nm by HPLC based on retention time, spectral properties, and comparison to authentic standards ..... 145 5 -5 Gallic acid (GA), unknown and ellagic acid (EA) for control, DP -CO2 and thermal treated guava purees during 14 weeks of refrigerated storage at 4oC 146 5 -6 Hydrobenzoic acid (GA), cinnamic acid (CA) and ellagic acid derivative (EAD) for control, DP-CO2 and thermal treated guava purees during 14 weeks of refrigerated storage at 4oC ................................................................... 146 6 -1 Difference in flavor and overall likeability between fresh ly thawed (reference and hidden reference), dense phase-CO2 processed (DP -C O2; 30.6 MPa, 8% CO2, 6.9 min, 35 C) and thermally treated ( 90 C, 60 s) guava purees detected by untrained panelists (n = 75) at weeks 0 .......................................... 165 6 -2 MS identification of guava puree volatiles. .......................................................... 168 6 -3 Gua va puree aroma active compounds ............................................................. 170 6 -4 Guava puree sulfur volatile compounds.. ............................................................ 171 A-1 SAS software code used for the statistical analysis of repeated measurement design and Tukeys standardized range (HSD) test ........................................... 176 A-2 SAS software output used for the statistical analysis of repeated measurement design and Tukeys standardized range (HSD) test .................... 177 B-1 Guava Puree Solubility data ................................................................................ 183 B-2 Guava Puree pH, Brix and Titraacidity measurement before and after the CO2 solubility determination................................................................................. 183 C -1 The average initial and final aerobic plate counts (APC) standard deviations at 11 experimental runs from 2-factor, 3 -level Central Composite Design (CCD) ....................................................................................................... 185 C -2 SAS software code used for the response surface methodology (RSM) analysis of 11 experimental runs determined by Central Composite Design .... 185 C -3 SAS software output of the response surface methodology (RSM) regression analysis of 11 experimental -run data determined by central composite design 186 D.1 Volatile compounds, CAS number, identification method, reported Linear Retention Indexes and references for previously reported studies .................... 192

PAGE 12

12 LIST OF FIGURES Figure page 3 -1 Percent yield for each enzyme treatment at three different temperatures (Treatment time: 24 hrs). ....................................................................................... 85 3 -2 Percent yield of clarified juice treated at three different enzyme concentrations and four different reaction times up to 12 h. ................................. 90 3 -3 Ascorbic acid content of clarified guava juices after guava puree was treated at t hree different enzyme concentrations and four different reaction times up to 12 h. .................................................................................................................... 90 3 -4 Antioxidant capacity ( Mol TE/L) of clarified guava juice treated with three different enzyme concentration during 12 hours of reaction time at 30oC.. ......... 91 3 -5 Total soluble phenolics (GAE) of clarified guava juice treated with three different enzyme concentrations during 12 hours of react ion time at 30oC.. ....... 91 3 -6 Turbidity (% transmission at 650 nm) of clarified guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30oC. ....................................................................................................................... 92 3 -7 pH of clarified guava juice treated with three different enzyme conc entrations during 12 hours of reaction time at 30oC. .............................................................. 92 3 -8 Total soluble solids (oBrix) of clarified guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30oC.. ...................... 93 3 -9 L* values of clarified guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30oC. ..................................... 93 4 -1 Schematic diagram of Dense Phase Carbon Diox ide equipment ...................... 112 4 -2 Carbon dioxide solubility results obtained for guava puree, water and guava puree model system at different processing pressures. ..................................... 113 4 -3 Aerobic plate count for control, DP -CO2 and thermal treated guava puree during 14 weeks of storage. ................................................................................. 1 17 4 -4 Yeast and mold plate count for control, DP -CO2 and thermal treated guava puree during 14 weeks of storage. ...................................................................... 118 4 -5 Pectinesterase activity for control, DP -CO2 and thermal treated guava during 14 weeks of storage. ............................................................................................ 119

PAGE 13

13 4 -6 Viscosity measurement for control, DP -CO2 and thermal treated guava during 14 weeks of storage. ................................................................................. 120 5 -1 Titraacidity of control, DP -CO2 and thermal treated guava purees during 14 weeks of refrigerated storage. ............................................................................. 139 5 -2 Ascorbic acid (vitamin C) content of control, DP -CO2 and thermal treated guava purees during 14 weeks of refrigerated storage. ..................................... 140 5 -4 HPLC chromatogram of organic acids found in guava puree1) oxalic acid, 2) malic acid and 3) citric acid. ................................................................................. 143 5 -5 HPLC chromatogram of polyphenolic compounds found in ethyl acetate fraction in guava puree 1) gallic acid, 2) unknown 3) ellagic acid, 4) hydrobenzoic acid and 6) cinnamic acid. ............................................................ 143 5 -6 HPLC chromatogram of polyphenolic compounds found in the methanol fraction of guava puree3) ellagic acid and 5) ellagic acid derivative. ............... 144 6 -1 Chemical composition of headspace volatiles for guava purees.. ..................... 166 6 -2 Total ion chromatogram (TIC) for freshly thawed guava puree on DB 5 column .................................................................................................................. 167 A-1 Enzyme treated guava puree without filtration (left) and after filtration (right) ... 179 B-1 Removal of insoluble solids from guava puree ................................................... 180 B-2 Experimental solubility apparatus ........................................................................ 181 B-3 Continuous DP -CO2 system used during processing ......................................... 182 C -1 DP -CO2 treated samples during storage (week 4) .............................................. 188 D -1 Questionnaire used for taste panel: different from control test, demographic questions .............................................................................................................. 189 D -2 Questionnaire used for taste panel: different from control test, sensory questions .............................................................................................................. 190 D -3 Questionnaire used for taste panel: different from control test, comments ....... 191

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy QUALITY OF GUAVA PUREE BY NON -THERMAL DENSE PHASE CARBON DIOXIDE PASTEURIZATION By Maria L. Plaza August 2010 Chair: Maurice Marshall Co -chair: Russell Rouseff Major: Food Science and Human Nutrition Guava ( Psidium guajava L.) is an exotic fruit with a unique tropical flavor. It is considered to be an excellent source of nutrients, phytochemicals and antioxidants, especially ascorbic acid. The high perishability of the fresh fruit limit s its marketability within the US. Guava puree is a commonly processed fruit product which can be pasteurized to ex tend its shelf life but pasteurization has negative effects on sensory and nutritional quality. A non -thermal process is desirable to protect the fresh flavor and nutritional value of guava puree, which is used as a base for production of guava beverages and other food products. The objective of this research was to evaluate a new technology Non -Thermal Dense Phase Carbon Dioxide Pasteurization (DP-CO2), for the processing of guava puree. The hypothesis was that the use of DP CO2, would minimize or prevent undesirable changes in phytochemical composition compared to traditional heat pasteurization. In order to validate this hypothesis, measurement and comparison of the

PAGE 15

15 chemical composition of the guava puree subjected to both treat ments and an untreated guava puree ( freshly -thawed control ) were conducted. To facilitate processing the puree through the DP CO2 equipment, viscosity of the puree was modified by a commercially available enzyme, Bioguavase. Samples were treated with the enzyme and evaluated for changes in viscosity at three hour intervals for up to 12 hours of reaction time. T he enzyme treatment increased juice yield produced a puree of lower viscosity decreased the antioxidant capacity and reduced the total phenolic content of the product. The enzyme treatment also decreased the pH of the juice due to the release of galacturonic acid from the pectin hy drol y sis and increased the total soluble solid s content. T hree hours of reaction time and 600 ppm of enzyme we re adequ ate to produce a clarified juice Microbial reduction was quantified as a function of pressure and residence time using 8% CO2 and a temperature of 35 oC. Optimum DP -CO2 treatment conditions for microbial inactivation were determined to be 34.5 MPa for 6. 9 min and 8% CO2 at 35 oC. Quality attributes including pH, oBrix, % titratable acidity (%TA) and color of DP CO2 treated, freshly thawed and heat pasteurized (90 oC for 60 s) guava puree were measured and compared throughout refrigerated storage (4 oC fo r 14 weeks). DP -CO2 treatment did not cause a change in pH or oBrix but increased the titratable acidity and viscosity of the product. Pectinesterase enzyme ( PE) was partially inactivated after DP CO2 processing. DP -CO2 treated guava puree retained organic acids similar to fresh guava puree and served to protect polyphenolic and antioxidant levels throughout processing and storage. DP -CO2 delayed the degradation of v i tamin C during storage.

PAGE 16

16 Flavor and aroma compounds in guava puree were identified u sing GC -MS, Linear retention index matching with databases and standards Flavor profiles showed that heat treated guava puree had less aroma active compounds than DP -CO2 treated guava puree. Volatile compounds analysis showed a lower total peak area for the DP CO2 when compared to fresh and pasteurized and differences in volatile composition w ere found for the three samples. DP -CO2 is an effective alternative to heat pasteurization of guava puree. It reduces microbiological load, extends the shelf life, and preserves important sensory and nutritional characteristics of the puree. The selection of this technique as a non-thermal processing technology was based its effectivity and preservation of quality attributes.

PAGE 17

17 CHAPTER 1 INTRODUCTION Guava ( Psidium guajava L) is a tropical fruit rich in antioxidants and vitamin C. A member of the Myrtaceae family, it is common to all tropical areas of America and can be found in the West Indies, Bahamas, Bermuda and southern Florida. G uava can be eaten raw or processed to obtain other products. In Hawaii the guava is boiled in slices to produce a guava juice. In Brazil, Mexico and Dominican Republic the fruit is commonly processed to obtain a puree. In South Africa, the fruit is trimmed, minced and mixed with a natural fungal enzyme to obtain a clear guav a juice with the ascorbic acid and other properties undamaged by the heat of pasteurization. Guava juice and nectar are among the numerous popular canned or bottled fruit beverages of the Caribbean area (Morton 1987) G uava fruit is rich in tannins, phenol s, triterpenes, flavonoids, carotenoids, vitamins and fiber. Most of the guavas therapeutic activity is attributed to the high content of flavonoids, which also have antimicrobial activity. Guava puree is normally processed by heat pasteurization to extend the shelf life of the product ( determined base on pectinesterase and microbial inactivation) The shelf life of the puree is about one year, but the fresh taste is modified by heat p asteurization The use of nonthermal pasteurization has the potential t o minimize the development of undesirable characteristics or loss of desirable characteristics, by reducing the chemical changes that occur during heat processing. The objective of this research was to test a new technology for the processing of guava pur ee that has minimal effect on the chemical composition of the product, particularly Vitamin C, total polyphenols, and flavor. The hypothesis was that the use of n on-t hermal Dense Phase Carbon Dioxide (DP-CO2) p asteurization would minimize or

PAGE 18

18 prevent undesirable changes in phytochemical composition compared to traditional heat pasteurization. In order to validate this hypothesis, measurement and comparison of the chem ical composition of the guava puree subjected to both treatments and an untreated guav a puree (control) was conducted.

PAGE 19

19 CHAPTER 2 LITERATURE REVIEW Guava ( Psidium guajava L.) Guava ( Psidium guajava L) is a tropical fruit rich in antioxidants and vitamin C. It is a member of the Myrtaceae family, which has more than 80 genera and 3000 species distributed throughout the tropics and subtropics (Nakasone and Paull 1998) The genus Psidium includes five species, Psidium guianense, P. cattleianum, P. chinense, P. fridrichsthalianum and P. guajava. P. guajava is the most widely cultivated species of the family Myrtaceae. The origin of the fruit is uncertain because it has been cultivated by humans and distributed by humans and birds, but it is believed that the origin is southern Mexico or Central America. It is common to all tropical area s of America and can be found in the West Indies, Bahamas, Bermuda and southern Florida. Guava Fruit and Fruit Products : World Production The consumption trend of fresh tropical fruits and their products is increasing steadily due to consumers education about their exotic flavors, nutritive value, and phytochemical content with potential health benefits (Commodity market review 2010 ). World production of tropical fruits was estimated at 67.7 million tons in 2004, representing a 2.5% increase compared to 2003. The minor tropical fruits, such as lychees, durian, rambutan, guavas and passion fruits, recorded an output of 16 million tons in 2004, accounting for a 3% increase (24% of the total tropical fruit production) (Current situation and medium outlook fo r tropical fruits 2010). Production of guava is increasing in importance and in 2004 reached an estimated 4 million tons. India is the main producer, followed by Pakistan, Mexico and Brazil. Minor production occurs in Vietnam and Malaysia.

PAGE 20

20 Fresh fruit cons umption in the US is increasing (The world fresh food market 2010) Guava as an imported product is divided into four categories according to the National Agriculture and Statistics Service (Hawaii guavas 2010) paste and puree, preserved or prepared, jam, and dried. Brazil was the leader for guava paste, puree, preserve or prepared imports into the US in 2008. Costa Rica was the main supplier of guava jams and the Philippines was the main supplier of dried guava. US commercial producers are located in Hawaii and southern Florida. Hawaii is the main grower, with 180 harvested acres and a utilized production volume of 3.5 million pounds in 2008. This production represents a decrease in 19% from previous years due to the closure of a large producer at the end of 2006. Guava ( Psidium guajava L.) has been catalogued as one of the most nutritious fruits The reason for this classification is its high content of phytochemicals e specially because of it high ascorbic aci d (Commodity market review 2010). Import of fresh guava fruit is not possible or limited due to the tropical fruit fly (quarantine issues) and its very short shelf life (7 to 10 days). Characteristics Psidium guajava is almost universally known by its com mon English name of guava. In Spanish, the fruit is known as guayaba or guyava. The French call it goyave or goyavier ; the Dutch, guyaba and goeajaaba ; the Surinamese, guave or goejaba; and the Portuguese, goiaba or goaibeira. Hawaiians call it guava or ku awa. In Guam it is abas In Malaysia, it is generally known either as guava or jambu batu. It also has numerous dialectal names in India, tropical Africa and the Philippines where the name, bayabas, is often applied (Morton 1987)

PAGE 21

21 The plant is a shrub or s mall tree reaching up to ten meters in height and is adaptable to a wide variety of habitats. Because of its robust growth, the plant runs wild and in some countries is considered a weed (Mitra 1997) The color of the trunk is often mottled in appearance, with reddishbrown outer scale bark. Due to the beautiful appearance of the trunk, in Florida it has been used in landscaping. The leaves are oval or oblong, seven to fifteen centimeters in length with prominent veins. The flowers consist of six white pet als and numerous stamens with yellow anthers. As a crop, the plant is mainly propagated by vegetative means, can bear fruits nine months after planting and can continue to bear fruit for ten to twenty years. The fruit yield varies depending on the cultural practices, but ranges from twenty -five to forty tons per hectare per year. The guava fruit is a berry. Morphologically, it may be round, ovoid, or pear -shaped, with four to five protruding floral remnants (sepals) at the apex; and having a thin, light y ellow skin, frequently blushed with pink. The fruit consists of a fleshy pericarp and seed cavity with numerous small seeds that are hard and kidney shaped. There are seedless varieties, but these typically contain a few seeds. The size of the fruit can range between four and ten cm in length and four to eight centimeters in diameter. The average weight is between 5 and 500 g (Salunkhe and Kadam 1995) The flesh color can be white, yellowish, light or dark -pink, or near -red. Depending on the color it is classified into two main groups: white and red. The flesh is juicy, acidic, or sweet and flavorful (sweet, musky and highly aromatic). Chemical Composition The chemical composition of the fruit varies with the stage of development, variety and season. The titratable acidity (TA) reported as citric acid content ranges from 0.08

PAGE 22

22 to 2.20% by weight. The total soluble solids (TSS) content ranges from 8 to 19.4oBrix, the TSS/acidity ratio varies between 6.2 and 53.9 and the pH ranges from 4.1 to 5.4. Guava frui ts consist of about 20% peel, 50% flesh (pericarp) and 30% seed core (Salunkhe and Kadam 1995) The fruit contains approximately 84% moisture, 26% dry matter, 1.5% protein, 0.7% lipids and 1% ash. Carbohydrates are the principal constituents of guava. Chan and Kwok (1975) reported the major carbohydrates in guava variety Beaumont. They found 59, 34 and 5% fructose, glucose and sucrose, respectively. The amount of these three sugars varies with variety and stage of development. Mowlah and Itoo (1982) found t hat fructose is the predominant sugar component in white and red guava, and it increased in all stages of maturation and ripening. ElBulk and others (1997) studied the changes in chemical composition for four cultivars of guava fruit during development an d ripening. They found that for all cultivars the sugar content increased gradually during the early stages of development and more rapidly at the later stages of development. Among fruit types, guava is the second highest in content of v itamin C, containi ng up to five times the concentration founded in oranges (Dweck 2005) Vitamin C (ascorbic acid) is water -soluble and highly susceptible to oxidative degradation, which often is used as an index for nutrient stability during processing or storage (Damodaran and others 2008) The vitamin C concentration fluctuates between 37 and 1,000 mg of ascorbic acid per 100 g of guava fruit. The variation of vitamin C content depends on variety, stage of development and season. Vitamin C con entration for red -fleshed gua va is higher than that of white-fleshed guava (Mowlah and Itoo 1983) Mercado -Silva and others ( 1998) found that vitamin C increased with the maturation process and that guava harvested

PAGE 23

23 during the autumnwinter -season had a higher content than those harves ted during the spring-summer season. W ithin each fruit, vitamin C conc ent ration distribution is higher in the skin than in the central portion of the flesh. Fruits also contain niacin (0.20 to 2.32 mg/100 g of fruit), thiamin (0.03 0.07 mg/100 g), ribofl avin (0.02 0.04 mg/100 g), carotene (0.01 0.9 mg/100 g), calcium (10.0 30.0 mg/100 g), iron (0.60 1.39 mg/100 g and phosphorus (22.50 40.0 mg/100 g). The composition of organic acids present in guava was studied by Chan and others (1971). They found that citric and malic acids were predominant followed by tartaric, glycolic and lactic acid (Chan and others 1971) Similar results were found by Wilson and others (1982) in a study of four cultivars from Florida. They found traces of fumaric acid, w hich was detected for the first time in guava fruits (Wilson and others 1982) Guava fruits contain significant amount s of polyphenols but their concentration and corresponding astringency decreases as the fruit matures. Examples of these polyphernols are : gallic, ellagic and cinammic acid and others. The pigments present in the guava fruit include but are not limited to carotene -carotene) xanthophylls and chlorophyll Guava Constituents and Health Benefits Fruits and vegetables are important components of a healthy diet and are one of the main sources of antioxidants. Clinical research supports the fact that consumption of fruits and vegetables is beneficial for prevention of cancer, heart disease and other agerelated diseases (Diet ary guidelines for Americans 2010). Due to the recent increases in obesity worldwide, the US Department of Health and Human Services reports that an effective strategy for weight management should include increasing the consumption of

PAGE 24

24 fruits and vegetables. This is based on the fact that this food group is high in water and fiber which promotes satiety and decreases energy intake (Carlton -Tohill 2007) Most of the research related to the relationship between fruit and vegetables consumption and obesity focuses on macronutrients effects on satiety. The important point is that consuming a diet rich in such plant foods will provide an abundance of dietary antioxidants, including polyphenolics, vitamins E and C, and carotenoids, all of which provide health benefits (Huang and others 2005) The guava fruit is rich in tannins, phenols, triterpenes, flavonoids, carotenoids, vitamins and fiber. Most of the guavas therapeutic activity is attributed to the high content of flavonoids, which also h ave antimicrobial activity. Vitamin C is commonly used to boost our immune system to fight colds and flu. In addition, it works as an antioxidant, destroying free radicals that can cause cancer and other diseases in the body. According to scientists from Cambridge University, a boost of vitamin C intake reduces the risk of death from heart disease (E-Tropical Fruit Net 2010). To determine the contribution of ascorbic acid (AA) to total antioxidant capacity of guava, Leong and Shui (2002) measured the antio xidant capacity of fruit using 2,2azino bis -(3 ethylbenzthiazoline 6 -sulfonic acid) (ABTS) free radical decolorizing assay and measuring the vitamin C content using HPLC. The ABTS method measures the relative antioxidant ability of fruits to scavenge the radical ABTS+.compared to a standard amount of ascorbic acid. Results are reported as AEAC (mg of AA equivalents per 100 g homogenate). They reported an AEAC value of 270 + 18.8 mg/100 g and a vitamin C content of 131 + 18.2 mg/100 g. According to the AEAC results, guava fruit is classified as a fruit with high antioxidant capacity. The ascorbic

PAGE 25

25 acid accounts for a high percentage contribution to ABTS+ scavenging activity which was 48.3% (Leong and Shui 2002) Kondo and others (2005) studied the antioxidant activity in guava using 1diphenyl 2 -picrylhydrazyl (DDPH) -radical scavenging activity and the effect on superoxide. In addition, they measured the total phenolic content and quantified some phenolics (gallic acid, catechin, epicatechin, chlorogenic acid and phloridzin). The results showed that the superoxide scavenging activity decreased in the skin but increased in the flesh during senescence. The superoxide scavenging activity was higher in the skin than flesh until 42 days after full bloom, after whic h the flesh showed a higher value than the skin. Phenolic concentration decreased in both skin and flesh during senescence. The values decreased from 1322 to 915 mol kg1 and from 1126 to 637 mol kg1 on skin and flesh respectively. The concentration of ascorbic acid increased from 0.34 to 2.19 mmol/kg in the skin and from 0.26 to 2.05 mmol/kg in the flesh. The only phenolic detected in the flesh was catechin while gallic acid, catechin, epicatechin and chlorogenic acid were identified in the skin. From the four phenolic compounds identified in the skin, catechin concentration found in the highest concentration (45 mol/kg of fresh weight) (Kondo and others 2005) The health benefits of consuming a diet rich in dietary fiber (DF) have been extensively studied (Gary 1999) The World Health Organization (WHO) and the Food and Drug Administration (FDA) have recommended an increase in the daily intake of dietary fiber ( DF) over the past ten years. For the industry, the production of food products rich in fiber is still a challenge. The selection of suitable sources to provide new products with high DF and antioxidant capacity is important. The only food products

PAGE 26

26 recogni zed, advertised and consumed as rich sources of fiber are those derived from cereals, but the antioxidant capacities of these items are negligible. However, over the past decade high dietary fiber materials from fruits (citrus, apple, and others) have been introduced in the market. Fruit DF concentrates have in general a better nutritional quality than those from cereals because of the presence of significant amounts of associated bioactive compounds (flavonoids, carotenoids, etc.) and their balanced compos ition (higher fiber content, soluble/insoluble DF ratio, water and fat holding capacities, lower energy value, and phytic acid content) than cereal materials (SauraCalixto 1998) Dietary flavonoids and other plant phenolics have been reported to have antioxidant activity, antimicrobial and anti -inflammatory action (Huang and others 1992) and have been associated with a reduced risk of cardiovascular diseases and cancer (Temple 2000 ; Pietta 2000) Research was conducted to evaluate guava as a source of natural antioxidant compounds and DF (Jimenez -Escrig and others 2001) The researchers classified polyphenol compounds into two categories according to solubility: extractable (EPP) an d nonextractable polyphenols (NEPP) based on previous studies related to physiological and nutritional properties of polyphenols associated with DF. The basic structure of EPP are flavan 3 ol and flavan -3,4 diol, whereas that of NEPP is condensed tannins. The basis for this classification was a study conducted by Bravo and others (1994) which defined the properties of EPPs as related to their solubility on the DF fraction. The properties of NEPPs had been related to the insoluble DF fraction (Bravo and othe rs 1994) Jimenez -Escrig and others (2001) conducted the study due to the development and introduction of a new concept: antioxidant dietary fiber (AODF).

PAGE 27

27 The main characteristics of this natural product are that they are rich in both DF and polyphenolic compounds. They found a remarkable antioxidant capacity related to the phenolics content. Peel and pulp of Psidium guajava fruit presented high levels of DF, an indigestible fraction, and phenolic compounds. They concluded that guava could be a rich sourc e of natural antioxidants and dietary fiber. Carotenoids have an important function as natural pigments, but some such as provitamin A have been studied for their health benefits, such as prevention of cardiovascular disease, immune enhancement and inhibi tion of cancer (Mathews Roth and Krinsky 1985) Wilberg and Rodriguez -Amaya (1995) quantified the major carotenoids present in fresh and processed guava. They found that provitamin A and the principal pigment concentration varied within the fresh and proc essed fruit and was highest in the ripe fruit. In the fresh fruit, they found a maximum concentration of 5.62 g/g and 60.6 g/g of -carotene and lycopene, respectively (Wilberg and Rodriguez Amaya 1995) Other studies addressed guava fruits carotenoid composition. Researchers cryptoxanthin, rubixanthin, cryptoflavin, lutein and neochrome isomers), 13 of which were identified for the first time (Mercadante and others 1998) Gorinstein and others (1999) conducted comparative research of the content of total polyphenols and dietary fiber in tropical fruits and persimmon. This research was conducted as a result of recent studies that have shown that the consumption of dietary fiber and polyphenols of plant products improve lipid metabolism. These authors concluded that the content of polyphenols (4.79 5.11 mg/100 g fresh fruit), gallic acid (340.6 408.0 g/100 g fresh

PAGE 28

28 fruit), total fiber (5.14 6.04g/100 g fresh fruit) and soluble fiber (2.39 3.01g/100 g fresh fruit) in guava were higher than the amount found in persimmon fruit (Gorinstein and others 1999) Recent research on flavonoids has shown that the biochemical and pharmacological activity of these compounds includ e anti oxidant, anti allergenic, anti platelet, anti -inflammation effects and antithrombotic action (Cook and Samman 1996) Miean and Mohamed (2001) studied the flavonoid content of edible tropical fruits. They found a total flavonoid content of 1128.5 mg/ kg dry weight due to the presence of myricetin (549.5 + 0.5) and apigenin (579.0 + 0.02). These two flavonoids were found in a methanolic extract obtained from guava and analyzed by HPLC (Miean and Mohamed 2001) Alpha tocopherol, the most common form of vitamin E is a lipid soluble vitamin that protects our skin and other lipid-rich body constituents. This vitamin is found in nature and has protective effect against oxidation of low -density lipoproteins, cell membranes and DNA by free radicals. The Recom mended Dietary Allowance (RDA) for daily vitamin E intake for adult males and females is 15 mg (22.4 IU ) ( Vitamin E 2010 ). Ching and Mohamed (2001 ) -tocopherol per 100 g edible portion of guava fruit. Rahmat and others (2006) conducted a s tudy to determine the effects of guava (Psidium guajava ) consumption on total antioxidant status and lipid profile (total cholesterol, triglycerides, LDL -cholesterol and HDL-cholesterol) in young males. The study was carried out over nine weeks. They found a significant increase of total cholesterol, triglyceride and HDL -cholesterol during the treatment phase (4 weeks),

PAGE 29

29 compared to the baseline (1 week) and control phases (4 weeks). The increase of HDLcholesterol was associated with the decreased risk of h eart attack and cardiovascular disease. The increase in total cholesterol and triglycerides during the treatment phase was still in normal range. There was a significant increase of total antioxidants during the treatment phase, compared to the baseline an d control phases. The trends of reduction of antioxidant enzymes (glutathione peroxidase and glutathione reductase) were associated with decreased oxidative stress and decrease in free radical activities. They concluded that the consumption of guava could result in improved antioxidant status and lipid profile, reducing the risk of disease caused by free radical activities and high cholesterol in blood (Rahmat and others 2006) Guava Flavors Aside from its nutritional value, the flavor of the guava is one o f the most distinguishable characteristics of this tropical fruit. Studies on guava volatile compounds have been conducted using leaf, skin, fruit and fruit puree. Different types of extractions and detection methods have been used. The first publication on the volatile constituents of guava was in 1970. The compounds from the guava fruit were extracted using distillation and the essence was analyzed by Gas Chromatography Mass Spectrometry (GC -MS). They identified twenty two compounds present in the oil: eight alcohols, six esters, three aldehydes four terpenes, one ketone and an alcohol. Hexanol and cis -3 hexen -1 ol were found in higher concentrations, compared to the other six alcohols. phenyl ethyl acetate, methyl cinn amate and cinnamyl acetate and concluded that their combination probably contributed to the overall aroma of the fruit (Stevens and others 1970)

PAGE 30

30 Wilson and Shaw (1978) conducted research on terpene hydrocarbons in guava using puree and a solvent extraction technique. The extract, which possessed a strong guava aroma, was separated by TLC and only the terpene-containing fraction was analyzed by GC M S and GC IR. They identified eleven terpenes from which limonene -copaene, which both were present in a greater quantities than that of limonene (Wilson and Shaw 1978) Macleod and Gonzalez de Troconis (1982) analyzed the volatile flavor component of guava. Compounds were isolated from guava pulp using a Lickens and Nickerson apparatus. F or the essence analysis, GC MS and a GC coupled to an olfactometry port (GC -O) were used. Fifty -five % of the essence composition was due to esters. Other compounds identified were two monoterpenes, and five sesquiterpene hydrocarbons, myrcene being the m ajor terpene. From the evaluation of odor, only eight compounds of those identified showed significant aroma characteristics: three were esters, four sesquiterpene hydrocarbons and myrcene. They conclude that 2methylpropyl acetate, myrcene, hexyl acetat humulene -selinene were important contributors to fresh guava flavor and should be retained as much as possible in processed products (M acleod and Gonzalez de Troconis 1982) Indstein and Schreier (1985) extracted the volatile constituents from guava fruit using high vacuum distillation and liquidliquid extraction. They fractioned the extract using three different chromatography techniques: capillary gas chromatography with flame ionization detector (HRGS), capillary gas chromatography -FTIR spectroscopy

PAGE 31

31 (HRGC-FTIR) and capillary gas chromatography mass spectrometry. In total, one hundred fifty -four compounds were identified from which one hundred sixteen were described for the first time as guava frui t constituents. The compounds found in the highest concentration were: (E) 2 hexenal, hexanal, (Z) 3 -hexenyl acetate, (Z) 3 -hexen1 ol and 1hexanol. Among the one hundred sixteen compounds identified for the first time, eleven nitrogen and sulfur volatile compounds were found in the guava aroma (Idstein and Schreier 1985) In an attempt to establish a relationship between the proximate composition of the fruit and the volatile constituents, Chyau and others (1992) studied the differences of volatile and nonvolatile constituents between mature and ripe guava fruits. The isolation of the compounds was performed using vacuum distillation followed by solvent extraction. They found a total of thirty -four components (twelve esters, eight alcohols, seven hydroc arbons, five carbonyls, one acid and eleven sesquiterpenes) from which, seventeen were confirmed using pure compounds. The differences between the mature and ripe fruit were in the quantitative analysis, not in the qualitative one. The major constituents i n mature guava were 1,8-cineole, (E) -2 -hexenal and (E) -3 -hexenal, while in the ripe fruit ethyl hexanoate and (Z) -3 hexenyl acetate were the major volatiles (Chyau and others 1992) Paniandy and others (2000) studied the chemical composition of the essenti al oil and headspace of white flesh guava fruit. They extracted the essential oil using steam hydrodistillation and for the headspace analysis they used solidphase microextraction (SPME) technique. They identified sixty -four compounds in the oil and twent y -four in the headspace. In the oil, the main contributors to the guava aroma were three

PAGE 32

32 -humulene. Sesquiterpene alcohols contributed 38% to the volatile composition. The products in the headspace of fresh f -butyrolactone was reported for the first time (Paniandy and others 2000) Jordan and others (2003) characterized the aromatic profile in commercial guava essence and fresh fruit puree using GC MS and GC -O and methylene chloride as the extraction solvent. In the commercial essence, 44 compounds were identified and quantified including seventeen alcohols, seventeen esters, three ketones, two aldehydes, two acids, one furan, one acetal and one terpene. In the puree, only twenty two compounds were identified and quantified as six alcohols, five esters, one ketone, one aldehyde, two acids, one furan, one lactone, and five terpenic hydrocarbons. Commercial essence was characterized by a volatil e profile rich in low molecular weight compounds such as alcohols, esters and aldehydes, whereas in the fresh fruit puree, terpenic hydrocarbons and 3-hydroxy3 -butanone were the most abundant. Olfactometry analysis yielded forty -three and forty eight arom a active compounds in commercial essence and fruit puree, respectively (Jordan and others 2003) Mahattanatawee and others (2005) studied the volatile constituents and aroma compounds of Florida -grown guava. They used SPME and liquid extraction to obt ain the aroma isolates. The isolates were analyzed by GC MS and GC -O. Due to the disadvantage of each one of the extraction methods, they used both techniques to obtain a more complete aroma profile by GC -O. The combined data from two extraction techniques resulted in detection of forty eight compounds from which twenty eight were identified as being odor active. The compounds in highest concentrations were hexanal

PAGE 33

33 -caryophyllene which have aroma activity and contribute to the green fruity and warm flo ral notes of guava. Sulfur compounds like methanethiol, 2 -methyl 3 -furanthiol, and mercaptomethylbutyl formate were detected and contributed to the unique sulfury note of guava aroma. There was no single character impact compound that contributed to the ar oma of the guava fruit (Mahattanatawee and others 2005) Carasek and Pawliszyn (2006) used a commercial automated cold fiber headspace solidphase microextraction (CF -HS SPME) device coupled to GC MS to identify the volatile compounds of guava. Thirty -thr ee compounds (alcohols, aldehydes, esters and terpenic compounds) were tentatively identified in the guava aroma, a large number of them being esters and terpenoid compounds (Carasek and Pawliszyn 2006) Steinhaus and others (2008) characterized the aromaactive compounds in guava by application of the aroma extract dilution analysis. For the extraction of the volatile fraction, the researchers used solvent extraction followed by solvent assisted flavor evaporation. The aromaactive areas in the chromatogram were screened by application of the aroma extract dilution analysis. The distillate obtained from the extraction represented a typical guava flavor with a green, sweet, tropical -fruit and also grapefruit -like notes. A total of thirty one odor active regi ons were detected, whereas two areas with caramel -like aromas, grapefruit -like odor and a black currant aroma quality showed the highest flavor dilution -factors (FD). The two compounds eliciting the caramel -like aromas were identified as 4 -methoxy -2,5dimethyl 3(2H) -furanone and 4hydroxy -2,5 dimethyl 3(2H) -furanone. The grapefruit -like note and black -currant aroma contributors were identified as 3-sulfanyl -1 hexanol and 3-sulfanylhexyl acetate respectively. In addition, a seasoning -like smell, green/grassy metallic and floral notes,

PAGE 34

34 were identified as 3hydroxy 4,5dimethyl -2(5H) -furanone, (Z) -3 hexenal, trans 4,5epoxy -(E) -2 decenal and cinnamyl alcohol, respectively. Also ethyl butanoate, hexanal, methional and cinnamyl acetate showed high odor activity. Among these compounds, 5 were identified for the first time in guava fruit (3-hydroxy -4,5dimethyl -2(5H) -furanone, 3sulfanyl 1 -hexanol, 3-sulfanylhexyl acetate, trans -4,5 epoxy -(E)-2 decenal and methional) (Steinhaus and others 2008) Steinhaus and other s (2009) characterized the key aroma active compounds in guava by means of aroma re engineering experiments and omission tests. Sixteen compounds previously identified and mentioned above, in addition to acetaldehyde, were quantified by stable isotope dilution assays. (Z) 3 hexenal, 4hydroxy2,5dimethyl 3(2H) -furanone, acetaldehyde and cinnamyl alcohol were found in the highest amounts, whereas methional was approximately 0.001% of the amount. To estimate the aroma potency of the individual guava odorant s, their concentration was correlated with the respective odor thresholds using odor activity value (OAV). (Z) -3 hexenal showed the highest OAV (57000), followed by 3-sulfanyl 1 -hexanol (9300), 3sulfanylhexyl acetate (360) and ethyl butanoate (170). Sinc e the OAV of 4 compounds were lower than 1, these compounds were assumed not to contribute to overall aroma. In addition to the identification and OAV determination, an aroma reconstitution experiment was conducted. An aqueous solution containing only 13 c ompounds (odorants found to exceed their respective thresholds in the concentration determined) was prepared and compared to a fresh puree using a sensory panel. The results from the sensory panel showed good agreement between the aroma model solution and the fresh puree, despite the fact that the simplified matrix did not include any nonvolatile

PAGE 35

35 guava constituents in the model system. The artificial and natural aromas both were characterized by a strong green grassy note, a moderate grapefruit -like note, some fruity and fresh notes, and weak sweet, flowery and metallic notes (Steinhaus and others 2009) Guava Fruit Processing Guava fruit are mainly consumed fresh but also are processed and preserved as different products. Guava is one of the easiest fruits to process, since the whole fruit may be fed into a pulper f or macerating into puree (Nunez -Rueda 2005). It is physically and biochemically stable in relation to texture or pulp browning durin g processing (Brasil and others 1995). It can be processed i nto a variety of forms, like puree, paste, jam, jelly, nectar, syrup, ice cream or juice. Within the United States processing industry, it is gaining popularity in juice blends. Guava pulp is extracted using a pulper, juicer or cloth press. Its composition is similar to the fresh fruit, and it is further processed and utilized in the form of puree, jam, juice and other products. Guava puree is the most important raw material for the juice industry. The puree is a liquid product prepared by pulping the fruit and is commonly used for the preparation of nectars, beverages, blends, clarified juice, jams and jellies. The preparation of the puree consists of washing the whole fruit, inspection for quality and feeding into a pulper which removes seeds and fibrous f ragments of skin. The finisher removes large aggregates of stone cells and the residual stone cells may be ground by passing the finished pulp through a mill. The milling operation improves mouth feel but downgrades the color quality An alternate method t o milling is centrifugation, which improves mouth feel and reduces precipitation in the product. The puree is preserved by freezing to -20 to 0oF ( -29 to 18oC), canning, aseptic packaging or pasteurization. Pasteurization is

PAGE 36

36 conducted between 80 and 90oC (190 194oF) for 60 seconds, then the pasteurized product is cooled and filled into containers. After pasteurization, the puree can be frozen and stored at -18oC (0oF) for up to a year. The food industry uses the puree to manufacture nectars, guava juic e, juice blends and other products. The manufacture of clear juice from guava is difficult. The colloidal particles which cause turbidity carry flavor substances and natural antioxidants. To obtain a clarified juice enzymes can be used to clarify juices or to change the viscosity of the fruit juices. The enzymes that are most commonly used are cellulases or pectinases. There are different processes to prepare a clarified juice. One involves maceration of fruit and mixing with enzymes. After a certain amou nt of time, which depends on the enzyme and temperature used for treatment, the pulp is passed through a press to obtain the juice. With this treatment more than 80% of juice can be obtained. Another procedure involves the direct treatment of the puree wit h enzymes. Two different methods for manufacturing clarified juice have been developed. In the first method, whole guavas are frozen to break their cellular structure and are kept frozen until needed. The problem with this procedure is the low yield. The s econd method uses guava puree as starting material. In this method, the puree is thawed and pressed mechanically using a press cloth, and then filtered. It is advisable to warm the puree to 40oC and add a filtering aid such as diatomaceous earth (Imungi an d others 1980) A few studies have been conducted on enzyme treatment of guava puree to obtain a clarified juice that later can be used to produce a concentrate without affecting

PAGE 37

37 desirable fruit attributes such as flavor and nutritional properties (Brekkee and others 1986; Imungi and others 1980 ; Hodgson and others 1990) Brasil and others (1995) studied the physical -chemical changes during extraction and clarification of guava juice. They treated guava pulp with 600 ppm of a pectic enzyme at 45oC. After the treatment, the pulp was pressed and the cloudy liquid was treated with fining agents and filtered. A higher yield was obtained after 120 min of treatment time. The viscosity of the treated product decreased by 62.9% related to the pulp. They c oncluded that this clarification technique followed by treatment with a fining agent and filtering showed good results and stability in nutritional and organoleptic characteristics (Brasil and others 1995) Guava Puree Processing Guava puree is normally processed by heat pasteurization to extend its shelf life The heat pasteurization serves as a preserving method but is carried out at conditions for pectinesterase inactivation which is more severe The shelf life of the puree can be extended to one year, but the fresh taste is modified by deteriorative reactions resulting in decreased sensory quality. Yen and Lin ( 1992 ) studied the changes in flavor components of guava puree during processing and frozen storage. They pasteurized p uree at 8588 C for 24 seconds. After pasteurization, the puree was immediately cooled, packed and stored at three different temperatures for analysis over a period of 4 months. The volatiles were extracted using Likens -Nickerson apparatus and a GC FID an d GC MS were used for the identification of compounds. Initially, the volatile constituents from the pasteurized puree were similar to the unpasteurized puree. Terpene hydrocarbons were the major volatile components followed by aldehydes. Changes in volati le constituents were

PAGE 38

38 significant during the first two months of storage at 0 C. This result w as attributed to both oxidation and enzymatic reactions since the puree showed an increase in total plate count and existing enzyme activities (such as peroxidase, pectinecterase and polyphenoloxidase) However, the quality of guava puree stored at 20 C after 4 months was satisfactory. The deterioration of flavor quality for guava puree during pasteurization and frozen storage may result from the changes in cert ain volatile components (Yen and others 1992) Chan and Cavaletto (1982) studied the changes in chemical and sensory quality during processing and storage of aseptically packaged guava puree. The puree was acidified to pH 3.9 with citric acid and the solu ble solids content was 13.5%. The heat treatment was conducted at 93 C for 26 seconds and the product was aseptically packed. The packed puree was stored at ambient temperature and sampled after 1, 3 and 6 months storage, while the accelerated samples (st ored at 38 C) were sampled after 1, 2 and 3 months. All samples were compared to frozen puree. After 3 months of ambient storage, the ascorbic acid retention was 72% while those stored at 38 C retained 62% of the ascorbic acid. There were no significant changes in pH, total acids or Brix. Aseptic processing appears to cause a lightening of the puree color. However, storage at 38 C showed darkening in color, while ambient temperature storage did not cause a significant darkening of samples. Aseptic proc essing also caused a decrease in both a* and b* values which also was observed by a sensory panel as a loss of pink color. In sensory tests, flavor was not greatly affected as color and the flavor changes were the result of storage time and not processing (Chan and Cavaletto 1982)

PAGE 39

39 Microorganisms, which can be minimiz ed by pasteurization, play an important role in spoilage of foods. Considerable research has focused on the identification of new processing technologies to avoid the detrimental chemical changes caused by heat pasteurization. High pressure processing (HPP) causes minimal changes in the fresh characteristics of foods by eliminating thermal degradation. Compared to thermal processing, HPP may provide a product with fresher taste and better appearance, texture and nutrition. High pressure processing is an alternative to heat processing and may have potential as a food processing method. Research was conducted to compare the effects of a high pressure treatment with thermal pasteurization on guava puree. The researchers concluded that high pressure maintained the original flavor of the juice. The volatile flavor components of the pressure-processed guava juices stored for 30 days at 4 C were similar to those of fresh juices (Yen and Lin 1999) Another study compared the effect of high pressure treatment and thermal pasteurization on the quality and shelf life of guava puree. The puree had a pH of 3.8 and 8.2 Brix. The high pressure treatment was carried at 400 and 600 MPa for 15 min at a temp erature of 25 C. The pasteurization process was carried out at a temperature of 88 90 C for 24 seconds. Samples from both treatments were stored at 4 C over a period of 60 days. All treatments were equally effective in reducing the microbial load of t he puree. During storage, pressurized puree at 600 MPa showed lower levels of microorganisms. Complete inactivation of POD enzyme was achieved by heat pasteurization, however PPO and PE activities of 16 and 4% respectively were found after the heat treatment. Both of the pressurized treatments showed residual activity for

PAGE 40

40 the three enzymes. Cloud analysis showed that loss of cloud was greater in untreated puree than in pressurized and heated puree during storage. Color of pressurized guava puree was similar to that of freshly extracted puree. This research indicated that high pressure treatment of guava puree at 25 C for 15 min could maintain good quality up to 40 days of storage at 4 C (Yen and Lin 1996) Phytochemicals In addition to macronutrients (nutr ients that are involved in normal metabolic activity) food contains components that may provide additional health benefits such as phytochemicals which are derived from naturally occurring compounds (Bloch and Thomson 1995) Phytochemicals are sometimes r eferred to as phytonutrients. They are natural bioactive compounds found in plant foods that work with nutrients and dietary fiber to protect against diseases. They also serve as protective agents for plants and some of them provide positive health benefit s. They are synthesized as secondary metabolites in all plants as a product of nutrient intake, protein synthesis and photosynthesis (Nunez -Rueda 2005) They are considered nonnutritive substances because they are not needed for regular metabolism. Within this group of nonnutritive substances are phenolic compounds, terpenoids, pigments and other natural antioxidants. Nuts, whole grains, fruits and vegetables contain an abundance of these substances that have been associated with protection from chronic di seases such as cancer and heart disease (Craig 1997) Polyphenols are the most abundant antioxidants in the diet. Their total dietary intake could be 10 times higher than vitamin C consumption, which is much higher than that of all other classes of phytoch emicals (Scalbert and others 2005)

PAGE 41

41 The term phenolic or polyphenol may be identified chemically as a substance which possesses an aromatic ring attached to it one or more hydroxyl groups, and may include functional derivatives such as esters, methyl e sters, glycosides or others (Ho and others 1992) Phenolic compounds are responsible for major organoleptic characteristics of plant derived foods and beverages, particularly color and taste. Plant polyphenols comprise a great diversity of compounds that are usually divided into two groups: flavonoids and non-flavonoids. Nonflavonoid compounds are mostly simple molecules such as phenolic acids and complex molecules derived from them (such as hydroxycinnamic acid derivates). Flavonoid compounds share a c ommon structure consisting of two phenolic rings and oxygenated heterocycles, and they are sub-divided into several groups, based on the oxidation state of the pyran ring (Cheynier 2005) Phenolic plant compounds, including all aromatic molecules from phen olic acids to condensed tannins, are products of a plant aromatic pathway, which consists of three main sections: the shikimic acid pathway which produces the aromatic amino acids phenylalanine, tyrosine and tryptophan that are precursors of phenolic acids ; the phenylpropanoid pathway which yields cinnamic acid derivatives that are precursors of flavonoids and lignans; and the flavonoid pathway which produces various flavonoid compounds (Bruyne and others 1999) Classification The term phenolics encompasse s approximately 800 naturally occurring compounds, all of which posses s a common structural feature: a phenol unit (C6). Current classification divides these compounds in two broad categories: polyphenols (posses s at least two phenol subunits) and simple p henols or phenolic acids.

PAGE 42

42 Polyphenols can be further divided in two categories: those containing only two phenolic units (flavonoids) and those containing more than three phenolic units (tannins). Phenolic acids are naturally occurring compounds that cont ain two distinguishing consecutive carbon frameworks: hydroxycinnamic and hydrobenzoic structures. Although the basic skeleton remains the same, the number and position of hydroxyl groups on the aromatic ring create the variety (Robbins 2003) Hydroxylated acid derived from benzoic acid include gallic acid, the main phenolic unit of hydrolyzed tannins and hydoxylated acids derived from cinnamic acid including coumaric, ferrulic and caffeic acids. Hydroxycinnamic acid derivatives represent the major group of plant phenolics since they form the basic constituents of lignins. Flavonoids represent the most common and widely distributed group of plant phenolics. Their common structure is that of diphenylpropanes (C6C3C6) or flavan nucleus which consists of two aromatic rings linked through three carbons that usually form an oxygenated heterocycle. They usually occur in plants as aglycones, although they are most commonly found as glucoside derivates (Laura 1998) Flavonoids are formed in plants from the aromatic amino acid phenylalanine and tyrosine. The various classes of flavonoids differ in their level of oxidation and pattern of substitution of the C ring, while individual compounds within a class differ in the pattern of substitution of the A and B ring (Pie tta 2000) The flavones (such as apigenin and luteolin), flavonols (such as quercetin, myricetin and kaempferol) and their glycosides are the most common compounds. Tannins make up another group of natural polyphenols. They are classified as hydrolysable and non-hydrolysable tannin. The hydrolysable tannins are those

PAGE 43

43 compounds that can be fractioned hydrolytically into their components. This group includes both the gallotannins and ellagitannins. Gallotanins are all those tannins in which galloyl units are bound to another compound such as catechin. Ellagitannins are the tannins in which at least two galloyl units are coupled together by a single carboncarbon bond and do not contain a catechin unit glycosidically attached. The biosynthetic pathway to hydro lysable tannins may be divided into three routes. The first section involves the esterification of a free gallic acid unit with glucose which undergoes further esterification to form the end product pentagalloylglucose. Pentagalloylglucose is the starting point for the two subsequent routes. The gallotannin route is characterized by the addition of galloyl residues to pentagalloylglucose. The ellagitannin routes are oxidation processes that yield carbon-carbon linkages between the galloyl groups of pentagal loylglucose (Grundhfer and others 2001) The non-hydrolysable tannins are also known as condensed tannin and complex tannins (Khanbabaee and Ree 2001) Complex tannins are tannins in which a catechin unit is glycosidically bound to a gallotannin or ellagi tannin. Condensed tannins are all oligomeric and polymeric proanthocyanidins. Procyanidins consist of chains of flavan 3 ol -units, which are commonly esterified, mainly with gallic acid units. Flavan-3 ols are derived from a branch of the anthocyanin and other flavonoids pathway, of which elucidation is still unclear (Dixon and others 2005) Structural variability among proanthocyanidins depends on hydroxylation, stereochemistry at the three chiral centers, the location and type of interflavan linkage, and terminal unit structure. A classical assay for proanthocyanidins consists of an acid hydrolysis, where the terminal units of the molecules convert to colored anthocyanidins.

PAGE 44

44 Attributes Phenolic compounds have been associated with positive and negative at tributes in terms of sensory and nutritional quality. Positive attributes include their close association with sensory and nutritional quality. Their nutritional value has been linked to: prevention of cancer, antimicrobial properties, antimutagenicity, antioxidant potential, reduction of coronary heart disease risk, antiviral, anti inflammatory and antitumor activity (Sonko and Xia 2005) Their sensory attributes are related to their contribution to flavor, astringency and color characteristics of foods. T he anti -nutritional effect of phenolic compounds involves their reaction with proteins, carbohydrates, minerals and vitamins lowering the bioavailability of these nutrients or their nutritional value. In addition, phenolic compounds can adversely affect th e sensory qualities of food by the production of off -flavor, their involvement in enzymatic browning or enzymatic discoloration, nonenzymatic discoloration and precipitation of proteins (Shahidi and Naczk 2003) Extraction and Analysis Extraction of pheno lic compounds in plant materials is influenced by their chemical nature, the extraction method employed, sample particle size, storage time and conditions, and the presence of interfering substances. The chemical nature of plant phenolics varies from simpl e to highly polymerized substances that include varying proportions of phenolic acids, phenylpropanoids, anthocyanins and tannins, among others. They may also exist as complexes with carbohydrates, proteins and other plant components. Therefore, phenolic extracts of plant materials are always a mixture of different classes of phenolics that are soluble in the solvent system used. Additional steps, such as solid phase extraction (SPE), may be required for the removal of

PAGE 45

45 unwanted phenolics and nonphenolic substances. Phenolics solubility is governed by the type of solvent used, degree of polymerization, as well as the interaction of phenolics with other food constituents. Some of the solvents most frequently used for phenolic compound extraction are methanol, ethanol, acetone, water, ethyl acetate and, to a lesser extent, propanol, dimethylformamide, and their combinations. Extraction periods usually vary from 1 min to 24 h (Naczk and Shahidi 2004) Phenolic compounds can be quantified by spectrophotometric or chromatographic method s ; in addition, separation and quantification can be done by chromatographic methods. A number of spectrophotometric methods for quantification of phenolic compounds have been developed and they involve the reaction of the sample containing the phenolic compound with a specific reagent. These assays are based on different principles and are used to determine various structural groups present in phenolic compounds. The Folin assay is the most widely used procedure for quantification of total phenolic content in plant materials. The disadvantage of this method is that it can detect phenolic groups in proteins. Chromatographic methods includes: gas chromatographic techniques (which require a sample preparation and derivatization) and high performance liquid chromatography (HPLC). Sensory Evaluation and Flavor Analysis Sensory Evaluation Sensory evaluation is the assessment of all qualities for food products as perceived by human senses. It is a quantitative science in which numerical data are collected to establish specific relationships between product characteristics and human perception. The main applications of sensory evaluation in the food industry are in quality assurance and product development (Murano 2003)

PAGE 46

46 The order of perce ption for attributes of food is appearance, odor/aroma/fragrance, consistency and texture, and flavor (aromatics, chemical feelings and taste). Sensory testing involves the use of people as measurement devices. Although a specific sensory test cannot provi de definitive answers to all questions, it is a key part of a larger sequence of information gathering during the product development process. To obtain meaningful data, it is important to match the test objective with the type of test used. Many variables must be controlled if the results of a sensory test are to measure the true product differences under investigation. It is important to group these variables under three major categories: test controls (e.g. room and environment), product controls (e.g. equipment used, samples temperature) and panel controls (e.g. procedure used by panelist to evaluate sample) (Meilgaard and others 2007) The test methods can be classified according to their primary purpose in three different classes: affective, discrimina tion and descriptive. Affective tests attempt to quantify the degree of liking and disliking of a product over another product. This type of test is used in consumer testing because they measure preference and acceptances. These groups of tests use untrain ed panelists. Preference testing uses a hedonic scale (like or dislike). Ranking is a type of preference test where the consumers are allowed to order a group of products base on their degree of liking or disliking. In acceptance testing, panelist rate their liking or disliking on a scale. Discrimination tests attempt to investigate if any differences exist between two types of products but the degree of difference is not determined. Examples of this type

PAGE 47

47 of test are: duo trio, difference from control, paired comparison and triangle tests. These tests do not employ trained panelists. Descriptive testing is used with trained panelists to describe specific product attributes related to flavor, texture, mouth feel and other characteristics of the product and to quantify the perceived intensities in each one of the evaluated attributes. Panelists are highly trained in relation to the scale, and the characteristics or attributes evaluated. Examples of this type of test include flavor profile method, quantitative descriptive analysis and texture profiling. Flavor Analysis Flavor is usually divided into taste (detected in the mouth) and smell (detected in the nose). Flavor perception depends on the combined responses of our senses and the cognitive processing of these inputs. Taste is the combined sensations arising from spe cialized test receptor cells located in the mouth (tongue and throughout the oral cavity). Olfaction is the sensory component resulting from the interaction of volatile components in food with olfactory receptors in the nasal cavity. The stimulus of the ar oma or odor of food can be orthonasal (odor molecules enters the olfactory region through the nose) or retronasal (odor molecules enter the nasal cavity through the back of the tongue). Consumers consider flavor as one of three main sensory properties they use in their selection, acceptance and ingestion of a particular food (Fisher and Scott 1997) Aroma is a very complex sensation (Reineccius 2006) Most of the aromas present in food systems consist of a mixture of several aroma compounds rather than a single aroma chemical. Flavor and/or aroma are usually created by mixing many flavor

PAGE 48

48 materials at the proper concentration of each component to produce the desired flavor characteristics and profile (Spanier 2001 ). The fir st step in the characterization of odor active compounds in a complex mixture or food system is to separate them from nonvolatile compounds. When isolating the flavor compounds, a reduction of matrix interferences will occur. This separation is accomplished through a variety of techniques, such as solvent extraction, head-space concentration, and distillation. The extraction procedure may distort or alter the chemical composition, because each one of these methods is selective for some compounds and probabl y not for all types, so there is no perfect extraction technique Each of these techniques yields a concentrated essence containing the odor active chemicals. Subsequently, separation, detection, identification, and characterization of individual compounds are possible with sophisticated instrumentation (such as GC -MS, GC -S and GC -O). Flavor Extraction Techniques Methods that have been used for extraction and concentration of flavor compounds include steam distillation, liquid liquid extraction, trapping o f the volatiles on adsorbents, and combinations of these methods with other techniques. They allow extraction and concentration of the compounds from their matrix. The main drawbacks of steam distillation extraction are the possible generation of thermal artifacts, foaming and gel formation. Some disadvantages of liquid -liquid extraction include the variation of compounds extracted. This is mainly due to the fact that different compounds have different partitioning coefficients. When using solvents, water soluble compounds are not extracted extensively while lipid-soluble compounds will be extracted more effectively. Another problem with solvent extraction is the increase in time needed to

PAGE 49

49 extract the compounds because the solvent must be evaporated and contamination with solvent impurities is possible. A rapid, simple and inexpensive technique [Solid Phase Microextraction (SPME)] was developed by Pawliszyn and coworkers in 1990 (Kataoka and others 2000) Solid Phase MicroExtraction (SPME) is a non -solven t sample preparation technique that uses a fused-silica fiber coated on the outside with an appropriate stationary phase, allowing the analyte in the sample to be directly extracted. The principle of SPME is the absorption of analytes onto a fiber followed by desorption in the injector port of a gas chromatograph. The SPME device consists of a fiber made of fused silica gel, coated with a stationary phase and bonded to a stainless steel plunger and a holder that looks like a modified syringe. Fiber coating material varies in composition, such as non-polar polydimethylsiloxane and polar carboxen, and thickness or amount of coating. These variations help to increase the sensitivity of the extraction depending on the analytes chemical characteristics. SPME has a very effective concentrating effect and leads to good sensitivity (Supelco 2005) The amount of extracted analyte depends on the thickness of the polymer coating and the distribution constant for the analyte rather than sample volume. The distribution c onstant of each analyte depends of the equilibrium established among the concentrations of analytes in the sample, in the headspace above the sample and in the coating material on the fiber. The extraction time is determined by the length of time required to obtain precise extraction for the analyte with the highest distribution constant, which generally increases with increasing molecular weight and boiling point of the analyte. Full equilibrium is not necessary for high accuracy and precision, but consist ent sampling time and other sampling

PAGE 50

50 parameters (e.g. sampling temperature) are essential. It is also important to keep consistent the vial size, the sample volume and depth the fiber is immersed in the sample (Lord and Pawliszyn 2000) Volatile Identific ation Techniques After isolation of the volatiles using an extraction and concentration procedure, samples are injected into GC for separation of individual compounds. Identification of flavor compounds in food is performed by retention time match with sta ndards or pure compounds. Actual identification of flavor compounds requires at least two independent confirmation techniques. Since retention time is dependent on a number of factors (such as column type, column length, carrier gas flow, etc.), they canno t be used to identify a compound based on literature values. To compare retention times from a single analysis to literature values, the standardized retention index needs to be calculated. Standardized retention index is a way to compare chromatographic r esults using different experimental conditions because they are independent of column length, carrier gas flow and film thickness. A retention index value gives an indication of where the compound of interest elutes relative to straight -chain hydrocarbons. The linear retention index, or Kovts index, expresses the number of carbon atoms, multiplied by 100, of a hypothetical normal alkane which would have an adjusted retention time identical to that of the peak of interest when analyzed under identical condi tions. With this information, linear retention index values can be used to evaluate the elution of a mixture of compounds on a specific column for a given set of conditions. The retention index of an unknown measured on several different columns is also us eful for identifying the unknown by comparison with tabulated retention indexes.

PAGE 51

51 Separation and Detection of Aroma Volatiles Volatiles are separated by GC and their elution time is monitored relative to a series of n alkanes that are injected under identi cal conditions. The detection of these compounds is achieved with the use of different detectors which are classified as either selective or non -specific detectors. The most common detector is the flame ionization detector (FID). It is a non -specific dete ctor because it responds to all organic compounds that burn or ionize in the flame (Wrolstad and others 2005) It has good sensitivity, a wide linear range in response and is used in almost all food analyses where a specific detector is not desired or sample destruction is acceptable. A selective detector is used for specific analytes. An example of this type of detector is the Flame Photometric Detector (FPD). As the compounds elute from the column, they are burned and measured by the light emitted from the flame at specific wavelengths. The wavelengths of light that are suitable in terms of intensity and uniqueness are characteristics of sulfur and phosphorous. Thus, this detector gives a greatly enhanced signal for those two elements (Nielsen, 2003) A Pulsed Flame Photometric Detector (PFPD) is used for the identification of sulfur containing compounds. In the PFPD, the flame is pulsed and is 10 to 100 times more sensitive than FPD. Other chromatography techniques are used for separation, detection and identification of flavor compounds. The most common of these techniques is GC MS. This instrument consists of an MS coupled to a GC. The separation of the compounds occurs in the GC and the MS allows the peaks to be quantitated and identified or confirmed and, if an unknown is present, it can be identified using a library containing MS spectra. Another technique is a gas chromatography olfactometry (GC -O). This

PAGE 52

52 instrument is used to distinguish aroma-significant compounds from the less important volatiles present in a sample matrix. Simultaneous with the high resolution gas chromatographic separation of a volatile extract, the odor of individual compounds is assessed by sniffing the effluent off the GC column in parallel with electronic detection. GC -O is a technique by which the human, rather than the machine, responds to odor detection. Historically, analytical instruments were utilized to detect components through an electronic device; however, the detection capabilities were limited. Detection by the hu man nose occasionally is more effective than the electronic device. This technique enables the detection of odor active volatiles, the determination of their odor qualities and the relative aroma intensity. Beverage Processing The food industry is continu ously searching for novel processing technologies that ensure microbial destruction and extend shelf life of products without having adverse effects on their quality attributes (Butz and Tauscher 2002) Current trends in food marketing showed that consumer s desire high quality foods with "fresh-like" characteristics and enhanced shelf life that require only a minimum effort and time for preparation (Butz and Tauscher 2002) A wide variety of emerging non-thermal processing technologies are available to process food and beverages and include high hydrostatic pressure, irradiation, ultrasound, pulsed electric fields, light pulses, and oscillating magnetic fields. Dense phase carbon dioxide (DP -CO2) pasteurization is a promising alternative to traditional p asteurization technologies and may lessen detrimental effects to thermolabile phytonutrients and flavor compounds (Gomes and Ledward, 1996; Sun and others 2002; Zabetakis and others 2000) Although the use of DP -CO2 processing has been shown to inactivate microorganisms, its effect on food

PAGE 53

53 quality characteristics needs more examination. The use of non-thermal pasteurization can minimize the development of undesirable characteristics by reducing the chemical changes that occur during heat processing. Therm al processing Heat treatment is used in food processing to achieve preservation of food. Thermal processing was developed by Nicolas Appert more than 200 years ago. The most important heat treatments used in food industry are: pasteurization, flash pasteur ization, aseptic packaging and canning. Pasteurization involves a low level heat treatment; below the boiling point of water. This thermal treatment has two primary objectives. The first objective is to destroy microorganisms known to occur in some type of food, like milk and egg products, that could affect public health. The second objective is to extend products shelf life. Pasteurization does not kill all microbial flora, so pasteurized products will contain living organisms capable of growing and limit ing the storage of the product. Flash pasteurization is a high -temperature short -time (HTST) treatment in which pourable products are heated during 3 to 15 seconds to temperatures that destroy pathogenic microorganisms. It is a very rapid form of aseptic processing. Aseptic processing uses temperatures higher than flash pasteurization to treat the product. After the treatment, hot product, clean containers and clean closures are brought together in an environment that prevents recontamination of the product. This operation normally takes place in a closed space under pressure with sterile air. Canning involves the use of specific times and temperatures of heating that are defined by thermal death time (TDT), which identifies the parameters required to des troy the spores of Clostridium botulinum in low acid products.

PAGE 54

54 Non -thermal Processing Thermal processing is still the major technique for shelf -stable food preservation. There is a general movement in food processing away from high -heat treatment and deep-freezing toward milder treatments, resulting in refrigerated foods with less cooked flavors. There is always the need to kill pathogens, but there is also a demand for c lean labels, meaning a preference for few, if any, chemical additives and preservati ves (Clark 2009) Non -thermal methods allow the processing of foods below temperatures used during thermal processing, so flavor, essential nutrients and vitamins suffer minimally or not at all (Butz and Tauscher 2002) However, non-thermal technologies no t only improve food quality, but also promote an equivalent or preferably, an enhanced level of safety, when compared to procedures they replace (Raso and Barbosa -Canovas, 2003) Irradiation, ultra high pressure, pulsed electric fields, DP -CO2 and pulsed m agnetic fields are non-thermal technologies attracting interest and gaining acceptance as food processing methods. Dense Phase Carbon Dioxide Dense Phase Carbon Dioxide (DP -CO2) is a cold pasteurization method that affects microorganisms and enzymes through the effect of CO2 under pressure below 50 MPa without affecting the fresh-like physical, chemical and sensory qualities. Carbon dioxide, a natural constituent of many foods, is a non-toxic, nonflammable, inexpensive gas and has a Generally Recognized as Safe status (Damar and Balaban 2006) Mechanisms of Microbial Inactivation A number of hypotheses have been proposed to explain the effect of microbial inactivation caused by DP -CO2, including cytoplasmic pH decrease, explosive cell

PAGE 55

55 rupture due to intern al pressure, modification of cell membrane and extraction of cell wall lipids, inactivation of key enzymes for cell metabolism and extraction of intracellular substances. Cytoplasmic pH decrease or acidification has been proposed as the main mechanism for microbial inactivation. In this mechanism, CO2 is dissolved in an aqueous solution forming carbonic acid, which at a sufficient concentration is dissociated into bicarbonate and hydrogen ions lowering the extracellular pH. The first theory proposed for mic robial inactivation was the explosive cell rupture due to internal pressure. It was thought that during the rapid depressurization of the sample, the CO2 would have rapidly expanded through the cells so that a part of them could have been mechanically brok en. However, pictures of microbial cells after treatment have shown that the mechanism of inactivation did not always involve cell rupture (Spilimbergo and Bertucco 2003) Modification of cell membranes and extraction of cell wall lipids is another hypothe sis for microbial inactivation. This mechanism is based on the lipophilic and solvent characteristics of CO2. The cell membrane consists of a double layer of phospholipids. The CO2 could easily penetrate into the membrane, leading to an increase of is flui dity and permeability, altering the characteristics of the membrane and destroying its essential function. This mechanism is known as anesthesia effect. The anesthetic theory is a strong explanation for microbial inactivation since images showed a modifica tion of cell membrane with possible leakage of cytoplasm, together with enlarged periplasmic space between the walls and the cytoplasmic membranes (Spilimbergo and Bertucco 2003)

PAGE 56

56 The theory of the inactivation of key enzymes for cell metabolism is based on the interference of bicarbonate and molecular carbon dioxide on certain enzymatic and biochemical pathways. Another proposed mechanism is precipitation of intracellular calcium and magnesium carbonate ions from bicarbonate. This can occur since there a re some proteins sensitive to calcium and magnesium that could be precipitated by carbonate. Factors Affecting Microbial Inactivation Factors that affect microbial inactivation using DPCD include water content within the cell (more log reductions are achi eved in wet cells than dry cells) and water activity of the food (DPCD is more effective as water activity increases). Generally, any factor that increases levels and rate of CO2 solubility enhances microbial inactivation caused by DP -C O2. For example, CO2 solubility increases with increasing pressure when temperature, residence time and CO2 concentration are equal. Generally, inactivation efficiency increases with increases in pressure, temperature and residence time. Temperature has a complex effect on mi crobial inactivation. Even when the CO2 solubility decreases with increasing temperature, inactivation of microbes by DPCD is more effective at high temperature. Higher temperature increases the CO2 diffusion and fluidity of the cell membrane. Another effe ct of temperature is the change of CO2, since its penetrating power is higher under supercritical conditions and there is a rapid change in solubility and density when processing temperature is near this critical region. The initial pH of the medium is another factor that affects effectiveness of DP -CO2 microbial inactivation. Acid medium facilitates carbonic acid penetration through the cell membrane, allowing a higher inactivation. The final factor affecting microbial inactivation is the cell growth phase; young cells are more sensitive than mature ones. The type of

PAGE 57

57 bacteria affects the DP -CO2 effectiveness. Gram positive bacteria showed more resistance than gram negative bacteria due to the differences in their cell membrane composition. Solubility of CO2 CO2 solubility in liquid foods can be affected by pressure, temperature, and food composition. Pressure has a direct effect on CO2 solubility: as pressure increases, CO2 solubility increases. On the other hand, as temperature increases, solubility of CO2 decreases. Food composition may increase or decrease the solubility of CO2 (Calix and others 2008) Types of Systems Three different types of DP -C O2 equipment have been developed: batch, semi continuous and continuous. The batch system was the first syst em developed. In this system, CO2 and the food to be treated are stationary in a container during treatment. This system consists of a CO2 gas cylinder, a pressure regulator, a pressure vessel, a water bath or heater, and a CO2 release valve. The sample is placed into the pressure vessel and temperature is set to the desired value. CO2 is then introduced into the vessel until the sample is saturated at the desired pressure and temperature. The sample is left in the vessel for a period of time, after which t he CO2 outlet valve is opened to release the gas. Some systems contain an agitator to decrease the time to saturate the sample with CO2 (Damar and Balaban 2006) A semi -continuous system allows a continuous flow of CO2 through the chamber while a continu ous system allows flow of both CO2 and the liquid food through the system. In a continuous flow system, the liquid CO2 and the product are pumped through the system and are mixed before entering the high pressure pump, which

PAGE 58

58 allows adjustment of the pressure to the desired processing levels. The processing temperature is controlled in holding coils. Residence time is adjusted by setting the flow rate of the product through the coils. At the end of the process, an expansion valve is used to release CO2 from the mixture and the treated product is collected (Damar and Balaban 2006) Food Applications and Effect on Quality DP -C O2 has been applied mostly to fruit juices and beverages. The application of DP -C O2 treatment to some fruits can cause tissue damage e ven at low pressures. DP C O2 treatment of orange juice showed that this nonthermal technology is effective in reducing microbial load, enzyme inactivation, cloud stabilization and maintenance of the quality of the product. Kincal and others (2005) used a continuous high pressure carbon dioxide (HPCD) system for microbial reduction in orange juice. They tested the effectiveness of the equipment in reducing the natural microflora of pulp-free Valencia orange juice at different pressures (38, 72, and 107 MPa) for a residence time of 10 min and CO2/juice ratios between 0.1 and 1.0. To test the effectiveness on spoiled juice, juice with a load of 2 x 106 colony forming units per mL was prepared and subjected to sub(25oC) and super -critical CO2 treatments (34. 5oC) at pressures of 38, 72 or 107 MPa, residence time of 10 min and CO2/juice ratio of 1.0. To study the capacity of the equipment in pathogen inactivation, untreated sterilized juice was inoculated with Salmonella typhymurium E. coli O157:H7 or Listeria monocytogenes and the system was run at pressures of 21, 38 or 107 MPa and a residence time of 10 min. A storage study was conducted at 1.7C with juice processed at 107 MPa, CO2/juice ratio of 1.0 and residence time of 10 min. When the variables pressur e and residence time were

PAGE 59

59 compared, residence time had the greater influence on microbial reduction. Continuous high pressure CO2 processing was capable of destroying the natural microflora in orange juice and residence time and pressure had little influence on the destruction of low levels of microorganisms. The CO2/juice ratio and temperature were shown not to be the driving forces on microbial load reduction in this system. They proved that the system was able to achieve a 5log reduction of the natural flora in spoiled juice (38, 72, and 107 MPa at 25 and 34.5C, CO2/juice ratio of 1.0 and residence time of 10 min), and 5 -log decrease of pathogenic Salmonella typhimurium (all three pressures and residence time of 10 min), Escherichia coli O157:H7 (pressure -related decrease), and Listeria monocytogenes (destroyed after treatment with all pressures). During the refrigerated storage study, they observed an increase in the bacterial number possibly because of an injury/repair mechanism of some of the microorg anisms or due to post contamination (Kincal and others 2005) Kincal and others (2006) published work on HPCD system for cloud and quality retention in orange juice. They used a continuous HPCD to treat pulp -free Valencia orange juice at pressures of 38, 72, and 107 MPa, and CO2/juice (w/w) ratios from 0.10 to 1 with a constant residence time of 10 min. For the storage study, they treated orange juice at 107 MPa for 10 min, a CO2/juice ratio of 1.09 and stored it at 1.7oC (Kincal and others 2006) The highest PE inactivation (56.0%) occurred at 72 MPa and a residence time of 10 min followed by inactivation of 53% achieved at 107 MPa and a residence time of 8.6 min. When the treatment was conducted at three different pressures and a residence time of 10 min the highest inactivation (46.3%) was obtained when the pressure was 107 MPa and no heat was applied. These results

PAGE 60

60 demonstrate that pressure affects PE inactivation. HPCD preserved and enhanced the cloud of the treated orange juice in some cases. The greatest increase was found in samples treated at 38 MPa and 1.18% of CO2, and it was shown that pressure has little effect on cloud. Treatment with continuous HPDP did not have any effect on pH and oBrix, but titratable acidity increased slightly after treat ments. During the storage study, PE activity decreased with storage time and cloud remained 4 times higher than the control during storage. Juice color did not change drastically during storage (Kincal and others 2006) Sensory evaluations of DPCD -treated and untreated OJ were not significantly different after 2 weeks of refrigerated storage at 1.7oC (Balaban and others 2008) Lim and others (2006) processed mandarin juice with DP -CO2. The process variables were temperature (25, 35 and 45 C), pressure (13 .8, 27.6 and 41.4 MPa), residence time (5, 7 and 9 minutes) and %CO2 (2, 7 and12). They found that temperature and %CO2 had a significant effect in log reduction of total aerobic plate count while pressure and residence time were not significant. The HPCD processing reduced the total aerobic count of natural flora in mandarin juice by about three orders of magnitude. Maximum log reduction (3.47) was observed at the conditions of 35C, 41.4 MPa, 9 min and 7 %CO2. PE inactivation ranged from 6.1 to 50.7% with a maximum inactivation achieved at 45 C, 41.4 MPa, 7 min and 7% CO2. Cloud was not only retained but enhanced. The highes t cloud increase was 38.4% at 45 C, 27.6 MPa, 7 min, and 2% CO2. Lightness and yellowness increased and redness decreased after treatment. pH and Brix did not change after treatment while titratable acidity of treated samples was higher than the untreated juice (Lim and others 2006)

PAGE 61

61 Beer quality after pasteurization with DP -C O2 was studied. A maximum log reduction in yeast population of 7.38 logs was predicted at 26.5 MPa, 21 C, 9.6 %CO2, and residence time of 4.77 min. The maximum haze reduction from146 nephelometric turbidity units (NTU) to 95.3 NTU was observed at a processi ng pressure of 27.6 MPa. Aroma and flavor of beer processed under the same conditions were not significantly different when compared to a fresh beer sample in a difference from control test. Foam capacity and stability of beer were minimally affected by th e process; however these changes were unnoticed by consumers (Dagan and Balaban 2006) The effects of DP -C O2 on microbial, physical, chemical and sensorial quality of a coconut water beverage were evaluated by Damar and others (2009). Processing variables were: pressure (13.8, 24.1, and 34.5 MPa), temperature (20, 30, and 40 C) and %CO2 (7, 10, 13 g CO2/100 g beverage). A constant residence time of 6 min was used during the experiment. DPCD -treated (at 34.5 MPa, 25 C, 13% CO2, 6 min), heat pasteurized (74 C, 15 s) and untreated coconut beverages were evaluated during 9 wks of storage at 4 C. Results showed that pressure was not significant in microbial reduction whereas temperature and %CO2 levels were significant. Total aerobic bacteria of DPCD and heat treated samples decreased while that of untreated samples increased to >105 CFU/mL after 9 wks. DP -C O2 increased titratable acidity but did not change pH and Brix. Likeability of DPCD -treated coconut water was similar to the untreated one (Damar and other s 2009) The effect of DP -C O2 on physical and quality attributes of red grapefruit juice was studied by Ferrentino and others (2010). A central composite design was used with pressure (13.8, 24.1, and 34.5 MPa) and residence time (5, 7, and 9 min) as vari ables.

PAGE 62

62 A constant temperature of 40oC and CO2 level of 5.7% was used to treat the juice. A storage study was performed with the fresh juice. Brix, pH, titratable acidity (TA), pectinesterase (PE) inactivation, cloud, color, hue and color density, total phenolics, antioxidant capacity, and ascorbic acid were measured after treatment and during 6 wk storage at 4 C. Five log reduction for yeasts and molds and total aerobic microorganisms occurred at 34.5 MPa and 7 min of treatment. During storage, the DPCD-treated juice showed no growth of total aerobic microorganisms, and yeasts and molds. Cloud increased by 91% while PE inactivation was 69.17%. No significant ( = 0.05) differences were detected between treated and untreated samples for Brix, pH, and TA. Treated juice had higher lightness and redness and lower yellowness. Slight differences were detected for the ascorbic acid content and the antioxidant capacity (Ferrentino and others 2009) Objectives of Study No research has been conducted on the effec t of dense -phase carbon dioxide on guava puree processing as a preservation method. The hypothesis of this research is that DP -CO2 treatment, a non -thermal process, can replace pasteurization to preserv e guava puree while maintaining the fresh like attribu tes of flavor and physico -chemi cal and phytochemical properties This research will focus on four main objectives: 1 Determine the optimal time and concentration of a commercially available enzyme to reduce the viscosity of guava puree at low temperature (30oC) and possible increase the juice yield 2 Determine the optimal conditions (pressure, % carbon dioxide and residence time) required to achieve a 5 log reductions in microorganisms for guava puree processing using DP -C O2.

PAGE 63

63 3 Determine the changes in phytoc hemicals and quality attributes o f thermal and non-thermal processed guava puree during storage. 4 Conduct a comparison of sensory attributes and aroma compound changes on thermal and non-th ermal processed of guava puree.

PAGE 64

64 CHAPTER 3 EFFECT OF ENZYME TREATMENT ON PHYSICO CHEMICAL AND PHYTOCHEMICAL PROPER TIES OF GUAVA PUREE Abstract Fruit juices are an important part of our diet. Guava ( Psidium guajava L.) fruit has not been fully utilized as a source of processed juice due to a number of quality limitat ions that occur during processing, one of which is the presence of excessive amounts of suspended solids Enzyme treatment is one method of enhancing the degradation and removal of suspended solids During enzyme treatment, increasing the temperature may produce a well clarified juice but may also modify the phytochemical composition and ascorbic acid content due to oxidati ve reactions The objective was to determine the optimal time and concentration of a commercially available enzyme (Bioguavase) for treatment of guava juice for obtaining a clarified product at 30 C without affecting the phytochemical properties Three treatment times (12, 24 and 36 hours) and three enzyme concentrations (400, 600 and 800 ppm) were tested. Following treatment, juice was c larified by centrifugation and analyzed for vitamin C content (2,6dic h loroindophenol titration method), antioxidant capacity (ORAC), total soluble phenolics (Folin assay), turbidity and color (L* a* b* values). After 12 h reaction time, the 600 ppm treatm ent produced the clearest juice. J uice yield was not increased by extending the reaction time beyond 12 h. In additional experiment s four treatment times (3, 6, 9 and 12 hours) were compared under the same conditions described above. All enzyme treatment s reduced the antioxidant capacity (between 8 and 22%) and increased the total soluble phenolic content (between 8 and 15%) of the juice. T reatment of guava jui ce with 600 ppm Bioguavase for 3 h is most suitable for obtaining clarified juice.

PAGE 65

65 Introduction Guava ( Psidium guajava L) is an exotic tropical fruit rich in antioxidants and vitamin C. A member of the Myrtaceae family, it is common to all warm tropical areas of America and can be found in the West Indies, Bahamas, Bermuda and southern Florida Morphologically, the fruit may be round, ovoid, or pear -shaped with thin, light yellow skin, frequently blushed with pink. The flesh can be white, yellowish, light or dark -pink or near -red, juicy, acid or sweet and flavorful. The guava can be eaten raw or processed to obtain other products. In Hawaii the guava is boiled in slices to produce a guava juice. In Brazil, Mexico and the Dominican Republic fruit is processed to obtain a puree. In South Africa, the fruit is trimmed, minced and mixed with a na tural fungal enzyme to obtain a clear guava juice wit h out exposure to heat that degrades ascorbic acid and other consti tuents Guava juice and nectar are among the numerous popular canned or bottled fruit beverages of the Caribbean area (Morton, 1987) Gua va puree is normally processed by heat pasteurization to extend the shelf life of the product for u p to one year, but the fresh taste is modified. The use of non -thermal pasteurization can minimize the development of undesirable characteristics by reducing the chemical changes that occur during heat processing. Dense phase CO2 (DP -C O2) technology is a non -thermal meth od emerging as an alternative to traditional thermal pasteurization. It is a non-thermal pasteurization method that does not use heat to dest roy microorganisms and enzymes (Damar and Balaban, 2006) and is a promising technology to preserve phytochemicals and to retain the fresh -like physical, nutritional and sensory qualities of the final product compared to traditional heat pasteurization. How ever the consistency of guava puree has a significant effect on its flow at low temperatures. One way of improving flow is by

PAGE 66

66 reducing the viscosity of the puree by enzymatic treatment. Th is enhanc es the degradation and removal of suspended solids and decreases the viscosity In the food industry, a combination of enzymes including pectinesterase, arabanase, hemicellulase, tannase and cellulase are used to degrade the mesocarp of guava which contains 90% of the total cell wall material as pulp (Kashyap and others 2001). Bioguavase is a commercially available enzyme preparation that contains a variety of carboh y drase enzymes derived from Aspergillus niger. It causes rapid viscosity reduction of guava puree or fresh guava fruit through pectin hydrolysis, with a significant increase in juice yield. Hydrolysis of pectin produces carboxylic a cids and galacturonic acid which may lead to a pH decrease. During enzyme treatment, increasing the temperature may produce a well -clarified juice but may also m odify the phytochemical composition and reduce ascorbic acid content due to oxidation. The objectives of this study w ere: to apply an enzyme (Biogua vase) treatment at temperature bellow 35 C to obtain a product with a consistency suitable for dense phase carbon dioxide processing with an increase in yield and to optimize ( without affecting the phytochemical properties ) the time and concentration of this commercial enzyme preparation for treatment of guava puree in obtaining a clarified pr oduct at 30 C Materials and Methods Preliminary Study Selection of an enzyme and a temperature to clarify the guava puree Kleryzyme 150 (DSM, Cedex, France), Rapidase TF (ADM, Decatur, IL, U.S.A.), Cellubrix (Novozymes, Denmark), Pectinex Ultra SP -L (No vozymes, Denmark), Crystalzyme 200XL (Valley Research, South Bend, IN, U.S.A.), Bioguavase (BioSun,

PAGE 67

67 Tampa, FL, U.S.A.), Biocranase Super (BioSun, Tampa, FL, U.S.A.) and Biocellulase FG Concentrate (BioSun, Tampa, FL, U.S.A.) were obtained from the manufact urers. Five different enzyme treatments (Table 3-1) and t hree different temperatures were used: 12.4 C (55 F), 21.4 C (70 F) and 30 C (86 F). One hundred grams of puree were treated with the enzyme preparation at each temperature. S odium azide (0.00 2%) was added to each sample. Each sample was incubated at one of the 3 selected temperatures for a period of 24 h + 1 h. Samples were observed and mixed periodically. After 24 h, the purees were removed from incubation, centrifuged in a Sorvall RC 5B Refrigerated Superspeed Centrifuge (Dupont Instruments, Newton CT, U.S.A.) at 10,410 x g (9,500 rpm), 8 min at 4 C T he percent yields of juice were calculated using the following equation: % yield = [(initial weight weight after centrifugation)/initial weight ] 100 Effects of enzyme treatment at 30 C on phytochemical l evels in g uava ( Psidium guajava) p uree Sample preparation Untreated guava puree was obtained from Hawaii (Kai Guava, Kilauea Agronomics, Kilauea, HI U.S.A.) through a distributor in Florida, and transported frozen to the Food Science and Human Nutrition Department in Gainesville, Florida. Th e puree was thawed, divided into 2 L bottles and immediately frozen at 20 C. Before the clarification process, the puree was thawed overnight at 6oC. Bioguavase (600 ppm ) was used to treat the puree for 24 h at 30 C. Following treatment, the purees were removed from incubation and placed in ice slush immediately to stop the enzym e reaction.

PAGE 68

68 Analysis Analysis of the puree before it was exposed to the temperature (sample before temperature treatment = no -heat, no enzyme or NHNE ) was performed. A control sample of puree exposed to 30 C for the same amount of time without enzyme (heated no enzyme ) was included The purees w e re assayed for percent juice yield, vitamin C, antioxidant capacity (ORAC), total phenolics, color and total soluble solids (TSS). Analytical procedures are described later. Enzyme Treatment Optimization at Low Temperatures to P roduce a C larified G uava (P sidium guajava ) J uice Sample preparation and enzyme treatment Four hundred grams of puree were weighed and placed in a beaker. The amount of enzyme added to each beaker w ere as follow s : 160 L Bioguavase enzyme (400 ppm), 240 L Bioguavase enzyme (600 ppm) and 320 L Bioguavase enzyme (800 ppm). These samples were divided into 4 beakers (each containing 100 g) and placed at 30 C in an incubator Every 12 h one beaker from each enzyme concentration w as rem oved and placed in an ice bath to stop the enzyme reaction. Samples were kept on ice until analyzed. The experiment was performed in duplicate. The previous procedure was conducted again, but sampling time was reduced to every 3 h for up to 12 h. After enzyme treatment, t he purees w ere assayed for percent juice yield, vitamin C, antioxidant capacity, total phenolics, turbidity, pH, total soluble solids (TSS, oBrix ) and color. Physicochemical analysis Percent juice yield : Juice yield w as conducted using the following as previouslu stated.

PAGE 69

69 Vitamin C : The 2,6dichloroindophenol dye was obtained from Sigma-Aldrich ( St. Loui s, MO U.S.A.) and the acetic acid and m phosphoric acid chemicals were obtained from Fisher Scientific (FL, U.S.A.). Vitamin C of the centrifuged sample s w as assayed by titration using the Official Method published by the AOAC. Two ml aliquots were used for the titration and t he vitamin C content in each sample was cal culated based on the stoichiometry of the titration (AOAC method 967.21, 1990). Vitamin C content was expressed as mg of vitamin C per 100 g sample. Antioxidant Capacity : AAPH (2,2 azobis(2methylpropionamidine dihydrochlo ride)), fluorescein (free acid) and Trolox (6-hydroxy -2,5,7,8tetramethylchroman -2carboxilic acid) were obtained from Sigma-Aldrich (St. Lou i s, MO U.S.A.). Antioxidant capacity of hydrophilic compounds in the supernatants of centrifuged samples was determined by the oxygen radical abs orbance capacity (ORAC) assay (Huang and others 200 2 ). Antioxidant capacity was calculated by integrating the area under the fluorescence decay curve in the presence of guava phytochemicals and calibrated with a standard curve of Trolox using a SpectraMax Gemini XPS microplate sprectrofluorometer (Mo lecular Devices, Sunnyvale CA, U.S.A.) and SoftMax Pro 5.2 software (Molecular Devices, Sunnyvale, CA U.S.A.). R esults were expressed as Trolox equivalents (TE) per mL (mol of TE/mL). A dilution of 100 X was u sed to obtain the correct area from the treated sample. Total Phenolic Compounds : Total phenolic compounds of the centrifuged samples were analyzed using the Folin-Ciocalteu metal reduction assay (Talcott and others 2000) using gallic acid as standard. Absorbances at 765 nm were taken using a Spectra Max 190 spectrophotometer

PAGE 70

70 (Molecular devices, Sunnyville, CA U.S.A. ). Gallic acid and Foli n -Ciocalteus reagent were purchased from Sigma-Aldrich (St. Loui s, MO U .S.A. ). All samples were diluted by 10 X in order to obtain the absorbance reading within the standard curve. Turbidity (clarity) : S ample turbidity w as measured using the percent transmission mode at 650 nm in a Beckman UV -VIS scanning spectrophotometer ( Model # DU 620, Beckman Coutler, Brea, CA, USA ). Clarity of samples was determined using 1.5 mL of sample in 1 cm plastic disposable cuvettes against water. pH: The pH was measured using an Orion expandable ion analyzer EA 920 pH meter (Orion Research; Bos ton, MA U.S.A.) equipped with an Accumate glass electrode (Fisher Scientific, U.S.A.) Total soluble solids (TSS): oBrix of the centrifuged juice was measured us ing an electronic ABBE Mark II refractometer at room temperature (Leica Inc.; Buffalo, NY, U.S.A.). Color Analysis: The color of the centrifuged juice was measured using a Gardner colorimeter (BYK-Gardner USA; Columbia, MD, U S. A ) and expressed as L*, a* and b* Twenty m L of sample were used to perform the analysis and the measurements were don e in the reflectance mode. Calibration of the equipment previous to color measurement was conducted using black and white tiles. Statistical A nalysis SAS 9.0 software (SAS Institute Inc., Cary, N.C., U.S.A.) was utilized for all statistical analyses. For the selection of an enzyme treatment and temperature, analysis of varianc e for factorial design was employed To study the effects of enzyme treatment at low temperature on phytochemical l evels in g uava p uree, one way ANOVA was employed with mean separation by Tukeys standardized test For both experiments

PAGE 71

71 conducted on the enzyme optimization treatment for guava puree, a repeated measurement design was used. C hemical measurement data for the treated and untreated samples were analyzed by a Tukeys st and ar d ized range (HSD) test at a significant level of = 0.05. Examples of the SAS program and output are presented in Tables A -1 and A -2. Results and Discussion Selection of Enzyme and T emperature In the selection of an enzyme and a temperature, a control sample of puree without enzyme was subjected to the same incubation temperature and centrifugation step as the treatment. The percent yield for the control puree was 66.14 + 0.03%, irrespective of the incubation temperature. The enzyme treatment with the highest yield (83.72 + 1.02 %) was treatment 5, or 600 ppm of Bioguavase enzyme (Figure 3-1). As expected, all enzyme treatments resulted in higher yield at 30 C (86 F). There was no significant difference ( = 0.05) between the type of enzyme and the percent yield, but there was significant difference ( = 0.05) between the temperature and percent yield. All enzymes have a temperature range w h ere they work at optimum conditions. In our case, all enzymes used during the first part of this research were active at temperatures between 10 and 55 C Enzyme activity wa s dependent upon temperature. Enzyme activity increased when temperature increased to a certain point and then denaturation occur red. The selection of the temperature and enzyme for further experiments (Bioguavase) was based solely on the higher yield it produced.

PAGE 72

7 2 Enzyme Treatment Optimization at Low Temperatures to Produce a Clarified Guava (Psidium guajava ) J uice without A ffecting its Phytochemical Composition The yield for the non-heated, no enzyme (NHNE) puree was 72.18% The heated no enzyme or control (30 C for 24 h ) showed a yield of 72.92% and the enzyme treated (ET) puree (heated with enzyme) had a yield of 82.93%. Thus the increase in the yield of the enzyme treated sample compared to controls was not due to the treatment temperature (Table 3 2). NHNE p uree and ET puree and between the control and ET puree (based on simple AN OVA analysis). As expected, the use of enzymes to clarify juices increases the yield of the final product. In the food industry, enzymes are used to increase yield and clarify juices An important component of commercial pectinases is pectin methyl esterase, which is specifically able to convert colloidal pectin to noncolloidal pectic acid. This results in the sedimentation of cloud-forming particles. In apple juice production, pectinases are added with the ultimate objective of producing a high yield clear juice (Christen and Smith, 2000) In sparkling clear juice production, enzymes are added to increase the juice yield during pressing and straining of the juice, and to remove suspen ded matter to give sparkling clear juices (Kashyap and others 2001) Table 3 2 presents the results obtained for the antioxidant capacity analysis. Before the temperature treatment the a ol TE/mL. The antioxidant c apacity f or the control and the ET purees w ere mol TE/mL, respectively. T samples indicating that neither the enzyme nor temperature had an effect in the antioxidant capacity of the product. These results are comparable to a previous study (Fender, 2005) were 10.5 mol TE/mL was obtained in guava nectar.

PAGE 73

73 Total soluble phenolics were measured by the Folin-Ciocalteu assay, which measures the capacity of phytochemical compounds to reduce an oxidized metal ion. Although the target compounds for this assay were polyphenolics, compounds such as ascorbic acid, certain soluble proteins melonoidins and reducing sugars give a measurable interference in the assay. The average total soluble phenolic s was expressed in Gallic Acid Equivalent (GAE) (Table 3 2). The total soluble p henolic s in the NHNE sample were 837.46 mg/L GAE. T he control and the ET samples had total soluble phenolic s content of 910.21 and 892.96 mg/L GAE, respectively. There w a s an increase in total phenolic content due to temperature treatment and enzyme activity. Comparing the control to the enzyme -t reated puree; there was a smal l decrease in total phenolic content in the enzyme0.05) between the NHNE and the control which showed no significant difference to the ET sample Guava has one of the highest concentrations of v itamin C among all fruits, typically containing betwe en 200300 mg vitamin C/100 g. In a study conducted on g uava j uice p rocessing o ptimization (Chopda and Barrett, 2001), the authors found an ascorbic acid content of 149.4 mg ascorbic acid/100g sam ple in the supernatant. Table 3 -2 presents the mg of ascorbi c acid/100g sample before (NHNE) and after temperature treatment (enzyme and no enzyme). The ascorbic acid content for the NHNE sample was 81.57 mg ascorbic acid/100 g sample (Table 3 2) T he ascorbic acid content decreased due to the temperature effect p robably due to oxidation reactions. The control shows an ascorbic acid content of 80.72 and the ET puree had an ascorbic acid content of 79.81 mg/100 g. Although ascorbic acid content decreased with increasing

PAGE 74

74 temperature, there were no significant differ three samples. This indicated that the conditions of enzyme processing did not create an environment leading to ascorbic acid oxidation or degradation. The ascorbic acid content was lower than that reported in other literature. In this study, the puree did not contain the guava skin, which is the part of the fruit with the highest ascorbic acid content. T he guava puree production date is not known so some degradation may have occurred during processing and storage. Table 3 3 shows the res ults obtained for the color analysis of the centrifuged samples. A decrease in L* (brightness) was observed after the temperature treatment. Both the control and ET samples present ed a decrease in brightness compared to the NHNE samples The NHNE puree had the higher L* value, followed by the control T he redness (a*) of the NHNE sample was higher than the control, and this value in crease d during enzyme treat ment. Enzyme treatment of guava puree causes the liberation of some compounds, such as carotenoids, which are bound to complex carbohydrates (components of the plant cell wall) that are broken down during clarification by the enzyme (Steven, 1998) There w in redness between the three samples. T he yellowness (b*) of the control sample was lower than NHNE sample. T he enzyme treatment increased yellowness A color analysis was performed on the samples without centrifugati on (Table 3 3) and the analysis show ed significant differences in the L*, a* and b* value between the three samples The centrifugation process reduced the L*, a* and b* for all three samples but redness and yellowness of the samples were the most affected Centrifugation removed most of the

PAGE 75

75 insoluble particles which could have contained the compounds responsible for the pink color but some colloidal particles remained caus ing turbidity in the nonheated sample and the control juice. The enzyme treated pu ree produced a clear juice after the centrifugation process. Centrifugation of the NHNE and control samples produced a cloudy supernatant while ET puree produced a clear juice (supernatant). The TSS (oBrix ) of the clarified guava puree was between 7 and 9 C entrifugation decreased the soluble solid s content of the samples (Table 3-4) There were significant Brix for the three puree samples. Before temperature treatment, the puree had a Brix of 7.1, while after the temper ature treatment the control and enzyme-treated purees had 7. 2 and 7. 4 Brix respectively. After centrifugation, the NHNE sample was 5.7 Brix while the temperature treat ed control and enzyme-treated samples were 6.5 and 6.6 Brix respectively. There were Brix for the control and the enzyme treated sample. As expected, Brix increases due to temperature and enzyme treatment. The increase in total soluble solids or Brix is due to the breakd own of pectin and other complex carbohydrates. In the case of pectin breakdown, a release of galacturonic acid (pectic acid) occurs which is soluble and contributes to the increase in Brix measurement. Pectic substances or pectin are high molecular weight polysaccharides found in plant cell wall middle lamellae. They are composed of galacturonic acid units, joined by 1,4 glycosidic linkages. Some of the acid groups along with the acid units become methylated during the fruit ripening (Murano, 2003) Whe n pectinases are added to fruit juices the conversion of insoluble pectin to soluble pectic acid occur (Christen and Smith, 2000)

PAGE 76

76 Enzyme Treatment Optimization at Low Temperatures to Produce a Clarified Guava (Psidium guajava ) J uice without A ffecting it Phytochemical Composition Results for the physicochemical analysis conducted on guava puree after treatment with 400, 600 and 800 ppm of Bioguavase enzyme over three reaction time s (12, 24 and 36 h) are shown in Table 3-5 There were significant differen ces due to the reaction time, enzyme concentration and their interaction (time x concentration). There was a significant difference between the control ( zero enzyme concentration and reaction time) and the enzyme treatments. There were no significant diffe rences between the 3 enzyme concentrations. After 12 h reaction time, the increase in yield was not significant for any of the 3 enzyme concentrations. Increasing enzyme concentration will increase the velocity of the reaction. After 12 h of reaction time, the maximum reaction velocity was achieved that is the reason why there were no further differences between the 3 enzyme concentrations In an attempt to reduce the reaction time and obtain maximum juice yield, another study was performed. In this study, samples were taken every 3 h for up to 12 h. T here were significant differences in vitamin C content due to the reaction time and the interaction of time with enzyme concentration. The control had lower v itamin C when compared to all levels of enzyme (only at 12 h ). At 12 h the amount of vitamin C increased for all levels of enzyme concentration (with no significant differences) probably because the rate of ascorbic acid degradation was lower than its liberation from the pulp. After 12 h the vitamin C content decreased. This indicated that after 12 h, the reaction time created an environment suitable for ascorbic acid oxidation or degradation. Comparing the ascorbic acid content with that reported in the literature, this study sample content was lower Vitamin C content depends on the variety of

PAGE 77

77 guava, harvested season, and processing conditions. Vitamin C in the fresh fruit is always higher than in the puree. Antioxidants are important in foods for their potential health benefits and their radical scavenging abilities They can prevent oxidation in foods and in the human body. T here were significant differences in antioxidant activity due to the enzyme concentration, reaction time and the interaction (Table 3 -5) Antioxidant capacities for the cont rol were high er compared to the different levels of enzyme concentration, but the value decreased with reaction time. At 12 h, the antioxidant capacity decreased with the amount of enzyme used, but there were no significant differences in the ORAC value be tween 400 and 600 ppm. At 24 h, there were no significant differences between the control, 400 and 600 ppm enzyme concentration. At 36 h, there w as no si gnificant difference between all levels of enzyme concentration. Results of the Folin-Ciocalteu assa y for total phenolics revealed that there were significant differences due to the reaction time, enzyme concentration and the interaction. At 12 h the total phenolics in the control were not significantly different from the 800 ppm enzyme and there were n o significant differences between the 3 levels of enzyme concentration. At 24 and 36 h there were significant differences between the control and the three enzyme concentrations (400, 600 and 800 ppm). There was an increase in total phenolic s content due to the enzyme concentration, and a larger amount was observed at 36 h reaction time. The increase in total phenolics is due to the release of phenolic compounds from the pulp. The pulp of the puree contains phenolic compounds which were released duri ng the enzymatic reaction. (Jimenez Escrig and others 2001) tested the fiber from the pulp and peel of guava from Caracas.

PAGE 78

78 They found that both fibers were potent sources of radical -scavenging compounds, presumably from the high content of cell wall bound polyphenolics reported for each fiber. They found for pulp fiber, a total extractable phenol content of 26.2 g GAE/ kg dry matter. Turbidity is an important factor in clarified juice. The results show ed that the use of enzyme is adequate to obtain a cla rified juice. There were significant difference s regarding turbidity between the reaction time, enzyme concentration and the interaction. The control had the lowest percent transmittance (lower clarity). T he juice was cloudy and show ed an increase in turbi dity at 36 h. This was significantly different for the 400, 600 and 800 ppm enzyme concentration at all reaction times. At 12 and 24 h, there were no signi ficant differences between 400 and 800 ppm At 36 h, there were no differences between 400, 600 and 800 ppm. As expected, enzyme treatment produces a clear juice due to the conversion of colloidal pectin to noncolloidal pectic acid. Colloidal pectin is responsible for the cloudy appearance of fruit juices and production of acid results in the sedimentation of the cloud-forming particles (Christen and Smith, 2000) There were significant differences in pH d ue to the enzyme concentration, reaction time and their in teraction. There were significant differences between the control and the three levels of enzy me concentration During the first 12 h of reaction, t he pH decreased with increasing enzyme concentration (Table 3-5) After 24 h, the pH remained almost constant irrespective of enzyme concentration. Enzyme treatment decreased the pH of the product (the product is more acid) due to the release of galacturonic acid to the medium.

PAGE 79

79 The Brix of the clarified guava juice was between 6.4 and 6.7. There were significant differences due to enzyme concentration, reaction time and th eir interaction (Table35) The control had lower TSS than the 3 enzyme treatments TSS for the control stayed the same over time, but the Brix increased with enzyme concentration. An enzyme concentration of 400 ppm did not cause a further increase in Brix after 12 h reaction time. Six hundred ppm of enzyme concentration caused a further increase in Brix at 36 h, this value being the highest. The increase in Brix is related to a decrease in pH and decrease in turbidity which are related to pectin breakdown. Color analyses are summ arized in Table 36. L v alues decreased with enzyme concentration; t he re were significant differences between the control and the 3 levels of enzyme concentration at the initial reaction time. There were no differences between 400 and 600 ppm during the entire reaction time. The a* value increased after the clarification process during the first 12 h, but then decreased There we re no differences between the 3 levels of enzyme concentration during the first 24 h of reaction time. The b values increased a fter the clarification process and at 600 and 800 ppm, this value increased with reaction time. At 600 ppm, there was no significant increase during the first 24 h of the reaction but the value increased further after 24 h. At 12 and 24 h, there were sign ificant differences at all levels of enzyme concentration, but at 36 h, the b* values for the 3 enzyme concentrations were not significantly differen t. The food industry uses enzymes to increase juice yield. Figure 3 -2 shows the percent yield obtained for the control and each of the enzyme concentrations (400, 600 and 800 ppm) over reduced reaction times ( 3, 6, 9 and 12 h). Enzyme treatment significantly increased juice yield. Four hundred ppm slowly increased juice yield but

PAGE 80

80 after 9 h there were no signifi cant differences between 400 and 600 ppm enzyme concentration. This slow increase in yield may be due to the composition of the enzyme. The enzyme system used a variety of carbohydrases. Different enzymes function together to lower the viscosity of the pur ee while increasing the juice yield. Imungi and others (1980) clarified guava puree using 400 ppm of Pectinex superconcentrate at a temperature between 40 and 50 C Th ey found that the yield increases with treatment time and 90 min was adequate to achieve maximum yield without decreasing vitamin C content (Imungi and others 1980). Sandhu and Bhatia (1985) also observed an increase in juice yield with enzyme treatment due to a considerable reduction of pectin. Brasil and others (1995) observed an increase in juice yield with increasing treatment time after treating guava puree with 600 ppm Clarex L super -concentrate at 45 C for up to 150 min. T here were no significant differences in the v itamin C content between the control and the puree clarified with 6 00 ppm enzyme concentration (Figure 3 -3) Enzyme concentration (400 and 800 ppm) significantly decreased the vitamin C content during the first 3 h reaction time. After 3 h, there was an increase in vitamin C content for 400 ppm enzyme concentration, and a fter 6 h reaction time, there were no significant differences between the vitamin C content for the control, 400 and 600 ppm samples. An enzyme concentration of 800 ppm decreased the amount of v itamin C in the clarified juice probably because the rate of ascorbic acid degradation was greater than its liberation from the pulp. Opposite results were found by Brasil and others (1985) after treating guava puree with pectic enzyme to obtain a cloudy juice. The reason for this difference may be explained by the type of puree used for their research. They prepared

PAGE 81

81 the puree by mashing the fruits using a fruit mill. The enzyme treatment increased the vitamin C content of the cloudy juice due to its liberation, especially from the peel of the fruit which is known to have more vitamin C than the flesh and center of the fruit (Brasil and others 1995). Results of the antioxidant capacity analysis are shown in Figure 34 Antioxidant capacit y for the control w as high compar ed to enzyme treatments The antioxidant capaci ty decreased slightly over time regardless of enzyme concentration. The use of enzyme treatment to produce a clarified juice decreased the antioxidant capacity of the guava product. Some of the compounds that have antioxidant capacity are bound to complex carbohydrates and removed after the enzyme treatment followed by centrifugation. Reduction of antioxidant capacity is related to a decrease in vitamin C (which has antioxidant capacity). Figure 35 shows the average total soluble phenolic compounds expres sed in Gallic Acid Equivalent (GAE). T here were significant differences due to the reaction time, enzyme concent ration and their intera ction. At 3 h the total phenolic level s in the control were not significantly different from the 400 ppm and the amount of total phenolic compounds at 600 and 800 ppm were lower compared to the control. The total phenolics content decreased over time after 3 h reaction time. An enzyme concentration of 400 ppm inc reased total phenolics during the first 3 h reaction time, there was a slight reduction in total phenolics content up to 6 h reaction time, and after 6 h, there was no significant change. There was a significant increase in total phenolics after 3 h reacti on time and 600 ppm enzyme concentration. The content stayed unchang ed up to 9 h reaction time, after which the content significantly dropped t o a level similar to the

PAGE 82

82 enzym atic treatment of 400 ppm. Imung i and others (1980) found an increase in total phen olics after enzyme treatment of guava puree, but with the clarification process (filtration) used during the research, a significant decrease in this content was observed. The results obtained from the turbidity analysis (Figure 3-6) showed that the use of enzyme resul ted in obtaining a clarified juice. There were significant differences between the reaction time, enzyme concentration and their interaction. The c ontrol had the lowest percent transmittance and the juice remained cloudy. F rom the results pre sented i n Figure 3 7 the pH decreased with enzyme concentration. After 3 h, the pH stayed almost constant irrespective of enzyme concentration. The control showed an increase (less acid) in pH during the first 3 h, and after this the pH was constant. Th e enzyme treatment decreased the pH of the juice due to the release of galacturonic acid from pectin hydrolysis A decrease in pH with enzyme treatment of guava puree was observed in the literature (Imungi and others 1980). Chopda and Barret (2001) observed a decrease in pH as enzyme concentration and incubation time increased at an incubation temperature of 50 C. Total soluble solids were measured and expressed as Brix. The results obtained showed (Figure 3 -8) that the control stayed the same over time, but the Brix increased with enzyme concentration. An enzyme concentration of 600 ppm did not cause a further increase in Brix after 3 h reaction time, producing the clari fied juice with the highest soluble solid content. After 6 h, there were no signifi cant differences in Brix between 600 and 800 ppm. After 9 h reaction time, there were no significant differences in total soluble solids between the control and 400 ppm enzyme concentration. The reason for the increase in TSS may be explained by the release of

PAGE 83

83 acid during pectin breakdown. An increase in TSS was observed by Thakur and Das Gupta (2006) after extracting beetroot juice from beetroots using Pectinex Ultra SPL 0.15% (1500 ppm) at 45 C for 2.5 h. Chopda and Barret (2001) observed an i ncrease in Brix for guava juice as enzyme concentration and incubation time increased at an incubation temperature of 50 C (Chopda and Barrett 2001). As observed in Figure 3 9 L* values decreased with enzyme concentration, and there were no significant differences between the control and the 3 levels of enzyme concentration at the initial reaction tim e (0 h) There were no differences between 400 and 600 ppm at 3 and 9 h reaction time. Enzyme treatments reduced L* value of the samples. Same results were observed by Hodgson and others (1990) after treating guava puree with Pectinex Ultra Sp-L at a concentration of 0.2% (2000 ppm) for 2 h 50 C or 16 h at 20 C. In addition, a decrease in a* value and a slight increases in b* was observed (Hodgson and others 1990). Conclusions After 12 h of reaction time, there was no further increase in yield at the 3 concentrations of enzyme used. At an enzyme concentration of 400 and 800 ppm, the ascorbic acid content decreased by 20% of its initial content during the f irst 3 h of incubation time. The enzyme treatment decreased (between 0.55 to 11% of the initial concentration) the antioxidant capacity of the samples and the antioxidant capacity was lower regardless of the enzyme concentration. After 3 h reaction time, 6 00 and 800 ppm enzyme caused a slight decrease in total phenolic compounds between 3 and 4% of the initial value respectively. A fter 3 h, 600 ppm enzyme increased this value by 17% of the initial content The turbidity decreased with enzyme activity and at 3 h, 6 00 ppm produced the clearest juice followed by 400 ppm. The enzyme activity affected the pH

PAGE 84

84 and TSS content: the pH decreased (between 0.5 and 1.6% of the initial pH) while TSS content increased (between 4 and 10% of the initial value) As expec ted, the color was affected by the enzyme concentration since the enzyme produced a clarified juice. Based on these results, three hours of reaction time and 600 ppm of enzyme concentration are adequate to produce a clarified juice without affecting the nu tritional value of the juice. These conditions are suitable for producing a clarified juice by centrifugation after enzyme treatment of the puree. The limitation of this procedure resides on using a filtering step to obtain a juice instead of centrifugati on. After filtering the enzyme treated puree, the color of the juice was much lighter than the original puree. Guava press cake retained all the pink color and may be used to increase the mouth feel in formulated juices. Other uses for the guava cake, such as possible source of fiber and natural colorant should be studied. Table 3 1. Enzyme and concentration used during enzymatic treatment Enzyme Treatment code Enzyme or Mixture of Enzyme Concentration of enzyme used 1 Klerzyme 150 Rapidase TF 300 ppm 600 ppm 2 Cellubrix L Pectinex Ultra SP L 450 ppm 750 ppm 3 Crystalzyme 200XL 600 ppm 4 BioGuavase 600 ppm 5 Biocranase Super Biocellulase FG Con. 600 ppm 600 ppm

PAGE 85

85 Figure 3 1. Percent y ield for e ach enzyme t reatment at t hree d ifferent t emperatures (Treatment time: 24 hrs). Error bars represent n=4 0 10 20 30 40 50 60 70 80 90 12.4 21.4 30 Percent YieldTemperature C 300 ppm Klerzyme + 600 ppm Rapidase TF 450 ppm Cellubrix L + 750 Pectinasx Ultra SP L 600 ppm Cristalzyme 200 XL 600 ppm Bioguavase 600 ppm Biocranase Super + 600 ppm Biocellulase FG Control

PAGE 86

86 Table 3 2. Percent yield, ORAC value, total soluble phenolics and ascorbic acid content of guava puree Sample % Yield ORAC (mol TE/mL) Total Soluble Phenolics (GAE) Ascorbic Acid (mg AA/100 g) Non heated, no enzyme (NHNE) 72.78 + 0.27 13.27 + 1.56 837.46 + 37.02 81.57 + 0.00 Heated no enzyme (control) 72.92 + 0.33 13.7 8 + 3.6 2 910.21 + 59.57 80.72 + 0.8 6 Heated + enzyme (ET) 82.93 + 0.32 12.63 + 0.98 892.96 + 56.94 79.8 1 + 2.24 Table 3 3. Color values obtained for three different guava puree before and after clarification Sample L value a* value b* value Puree Clarify Puree Puree Clarify Puree Puree Clarify Puree Non heated no enzyme (NHNE) 48.73 + 0.01 32.83 + 0.32 22.44 + 0.03 0.98 + 0.07 14.73 + 0.01 5.61 + 0.20 Heated no enzyme (control) 49.66 + 0.01 29.56 + 0.68 22.84 + 0.15 1.68 + 0.15 14.97 + 0.31 5.98 + 0.19 Heated + Enzyme (ET) 48.23 + 0.29 11.33 + 0.93 23.25 + 0.25 0.77 + 0.23 14.50 + 0.19 1.95 + 0.30 Table 3 4. Total soluble solids for guava puree before and after clarification for the three different treatments Sample Brix Puree Clarify Non heated no enzyme (NHNE) 7.1 + 0.0 5.7 + 0.0 Heated noenzyme (control) 7. 2 + 0.0 6.5 + 0.0 Heated + Enzyme (ET) 7. 4 + 0.0 6.6 + 0.0

PAGE 87

87 Table 3 5. Physicochemical results for enzymatic treatment of guava puree at three different concentrations and three different reaction times Physicochemical Analysis Enzyme Concentration (ppm) Reaction Time (hours) 0 12 24 36 % Yield 0 61.27 + 0.14 60.44 + 0.26 65.72 + 0.28 67.80 + 0.40 400 61.27 + 0.14 76.44 + 1.44 77.43 + 1.61 77.55 + 1.95 600 61.27 + 0.14 79.21 + 0.82 79.99 + 0.84 80.19 + 1.19 800 61.27 + 0.14 79.30 + 1.32 80.58 + 0.46 79.49 + 0.24 Vitamin C (mg ascorbic acid/ 100 g sample) 0 71.80 + 0.00 71.12 + 0.01 68.52 + 0.00 66.01 + 0.02 400 71.80 + 0.00 75.56 + 0.01 69.20 + 0.00 66.50 + 0.01 600 71.80 + 0.00 74.98 + 0.01 68.90 + 0.00 66.01 + 0.02 800 71.80 + 0.00 74.98 + 0.00 70.26 + 0.00 66.21 + 0.01 Antioxidant Capacity (Mol TE/ L) 0 11.86 + 0.92 13.30 + 0.69 12.36 + 0.27 11.79 + 1.88 400 11.86 + 0.92 10.91 + 0.32 11.50 + 0.61 10.03 + 0.75 600 11.86 + 0.92 10.28 + 0.82 10.98 + 1.03 9.61 + 1.70 800 11.86 + 0.92 8.79 + 0.31 9.01 + 1.15 10.06 + 1.79 Total Soluble Phenolics Compounds (GAE) 0 889.10 + 6.80 786.81 + 28.63 757.50 + 18.45 834.90 + 18.45 400 889.10 + 6.80 878.61 + 39.38 869.86 + 25.83 913.44 + 25.83 600 889.10 + 6.80 874.03 + 54.63 848.89 + 21.51 962.71 + 21.51 800 889.10 + 6.80 825.28 + 61.83 823.19 + 59.53 926.67 + 59.53 Turbidity 0 23.33 + 0.00 25.68 + 1.57 24.42 + 0.39 29.36 + 0.41 400 23.33 + 0.00 72.04 + 3.00 70.16 + 5.37 80.95 + 1.07 600 23.33 + 0.00 86.29 + 3.47 80.34 + 6.52 82.48 + 2.29 800 23.33 + 0.00 76.39 + 4.66 69.59 + 2.84 85.14 + 4.56

PAGE 88

88 Table 3 5. Continued Physicochemical Analysis Enzyme Concentration (ppm) Reaction Time (hours) 0 12 24 36 pH 0 3.10 + 0.00 3.17 + 0.16 3.15 + 0.01 3.15 + 0.01 400 3.10 + 0.00 3.11 + 0.0 3.09 + 0.01 3.09 + 0.02 600 3.10 + 0.00 3.08 + 0.01 3.07 + 0.00 3.08 + 0.01 800 3.10 + 0.00 3.05 + 0.01 3.01 + 0.01 3.05 + 0.01 Tritable Acidity (mg of citric acid / 100 g) 0 0.91 + 0.01 0.92 + 0.01 0.93 + 0.02 0.92 + 0.02 400 0.91 + 0.01 0.98 + 0.01 0.97 + 0.02 0.97 + 0.01 600 0.91 + 0.01 0.97 + 0.02 0.97 + 0.01 0.98 + 0.01 800 0.91 + 0.01 0.96 + 0.01 0.97 + 0.01 0.98 + 0.01 Total Soluble Solids 0 6.40 + 0.00 6.40 + 0.00 6.40 + 0.00 6.40 + 0.00 400 6.40 + 0.00 6.70 + 0.30 6.70 + 0.20 6.70 + 0.10 600 6.40 + 0.00 6.7 0 + 0.10 6.70 + 0.00 6.80 + 0.10 800 6.4 0 + 0.00 6.6 0 + 0.00 6.70 + 0.00 6.70 + 0.00

PAGE 89

89 Table 3 6. Color results for enzymatic treatment of guava puree at three different concentrations and three different reaction times Enzyme Concentration (ppm) Reaction Time (hours) 0 12 24 36 L values 0 ppm 38.77 + 0 .0 42.36 + 0.96 43.32 + 0.67 37.23 + 0.13 400 ppm 38.77 + 0.0 27.29 + 1.01 26.61 + 0.49 23.17 + 0.78 600 ppm 38.77 + 0.0 24.43 + 0.81 24.86 + 1.12 20.67 + 0.12 800 ppm 38.77 + 0.0 26.61 + 0.5 27.66 + 0.35 22.69 + 0.31 a values 0 ppm 2.63 + 0.04 1.75 + 0.10 2.29 + 0.10 2.92 + 0.07 400 ppm 2.63 + 0.04 1.58 + 0.09 2.00 + 0.05 2.34 + 0.21 600 ppm 2.63 + 0.04 1.91 + 0.10 2.01 + 0.05 2.37 + 0.07 800 ppm 2.63 + 0.04 1.84 + 0.07 2.02 + 0.09 2.04 + 0.16 b values 0 ppm 3.46 + 0.02 4.64 + 0.24 3.69 + 0.35 3.18 + 0.03 400 ppm 3.46 + 0.02 0.90 + 0.303 0.54 + 0.25 2.12 + 0.60 600 ppm 3.46 + 0.02 1.05 + 0.48 1.38 + 0.45 1.9 + 0.07 800 ppm 3.46 + 0.02 0.13 + 0.04 0.162 + 0..1 1.98 + 0.41

PAGE 90

90 Figure 32. Percent yield of clarified juice treated at three different enzyme concentrations and four different reaction times up to 12 h. Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. Figure 33.Ascorbic acid content of clarified guava juices after guava puree was treated at three different enzyme concentrations and four different reaction times up to 12 h. Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. d h k m c g i l b e i j l a b f j l 68 70 72 74 76 78 80 82 84 0 3 6 9 12% Yield (w/w)Reaction Time (hours) Control 400 ppm 600 ppm 800 ppm b d f h c d e f h b d f h a c e g i 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0 3 6 9 12 Ascorbic Acid (mg/mL)Reaction Time (hours) Control 400 ppm 600 ppm 800 ppm

PAGE 91

91 Figure 34. Antioxidant c apacity ( Mol TE/L) of clarified guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30oC. Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. Figure 3 5. Total soluble phenolics (GAE) of guava juice treated with three different enzyme concentrations duri ng 12 hours of reaction time at 30oC. Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. b d f h c e g i c e g i a c e g i 0 5 10 15 20 25 30 35 0 3 6 9 12 Trolox equivalent ( M ol/L)Reaction Time (hours) Control 400 ppm 600 ppm b e f i l b e h i k c d g k a c f g h j 100 200 300 400 500 600 700 800 900 1000 0 3 6 9 12 Gallic Acid Equivalent (ppm)Reaction Time (hours) Control 400 ppm 600 ppm 800 ppm

PAGE 92

92 Figure 36. Turbidity (% transmission at 650 nm) of guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30oC Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. Figure 37. pH of clarified guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30oC. Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. d h j l c f i k c e i k a c g i k 0 10 20 30 40 50 60 70 80 90 0 3 6 9 12% Transmission at 650nmReaction Time (hours) Control 400 ppm 600 ppm 800 ppm b d g k c f j l c f i l a c e h l 2.9 2.95 3 3.05 3.1 3.15 3.2 0 3 6 9 12 pHReaction Time (hours) Control 400 ppm 600 ppm 800 ppm

PAGE 93

93 Figure 38. Total soluble solids ( Brix) of clarified guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30oC. Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. Figure 39. L* values of clarified guava juice treated with three different enzyme concentrations during 12 hours of reaction time at 30 C. Error bars represent standard deviation for n=9. Different letters within each reaction time represent significant differences at =0.5. d g i k c f i k b e h j a c e h j 4 4.5 5 5.5 6 6.5 7 0 3 6 9 12 Degrees BrixReaction Time (hours) Control 400 ppm 600 ppm 800 ppm b e h k c d f j m d g j l a c f i m 0 5 10 15 20 25 30 35 0 3 6 9 12L* valuesReaction Time (hrs) Control 400 ppm 600 ppm 800 ppm

PAGE 94

94 CHAPTER 4 INFLUENCE OF DENSE P HASE CARBON DIOXIDE AND PASTEURIZATION TREATMENTS ON THE ST ORAGE QUALITY OF GUA VA PUREE Abstract Guava has been identified as a good source of antioxidants and other phytoche micals, vitamin C and dietary fiber. The traditional method for preserving guava puree is heat pasteurization which degrades certain constituents that are beneficial to human health (such as vitamin C) The objective of this study was to determine the opt imal conditions (pressure, % carbon dioxide, and residence time) for guava puree processed using Dense Phase Carbon Dioxide (DP -CO2), a nonthermal pasteurization treatment. Stone cells and insoluble solids were removed from unpasteurized guava puree and C O2 solubility in puree was measured between 6.9 and 31.03 MPa at 35C. A log reduction > 3.2 was achieved for yeast and mold count (Y&M) and a erobic plate c ount (APC) using 34.1 MPa pressure, 8% CO2 and a residence time of 6.9 min. In all treatments titratable acidity (TA) was significantly higher compared to the fresh samples. There were significant differences in pectinesterase activity (PEA) and cloud values for fresh, DP -CO2 and thermally treated samples. DP -CO2 processing can be used as a nonthermal treatment without deteriorating the quality of the product compared to heat pasteurization. Introduction One of the most common heat treatments used in the food industry to preserve products is pa steurization, which involves a low order heat treatment (bellow 100 C) to destroy vegetative microorganisms that could affect human health. It extends the products shelf life but does not kill all microbial floras. Pasteurized products contain living org anisms capable of growth that limits the shelf life of the product.

PAGE 95

95 Guava ( Psidium guajava L ) puree is the raw material used to manufacture products such as juices, nectars, jams and jellies. Puree is prepared by washing the whole fruit, removing fruit of inferior quality, and feeding high quality fruit into a pulper which removes seeds and fibrous fragments of skin. A finisher then removes large aggregates of stone cells and the residual stone cells may be ground by passing the finished pulp through a m ill. The milling operation improves the mouthfeel but decreases color quality. An alternate method to milling is centrifugation, which improves mouthfeel and reduces the sediment of the product. Guava puree is normally processed by heat pasteuri zation to extend the shelf life of the product. Heat pasteurization (between 80 and 90oC for 60 s) inactivates pectinesterase (PE). The shelf life of the puree can be extended to one year at -18oC (0oF) but the fresh taste is lost by deteriorative reactions caused b y heat exposure. Dense Phase Carbon Dioxide (DP -CO2) is a nonthermal pasteurization method. Microorganisms and enzymes are altered by CO2 under pressure (below 50 MPa) without affecting important physical, chemical and sensory qualities. Carbon dioxide is a non-toxic, nonflammable, inexpensive gas and has a Generally Recognized As Safe (GRAS) status (Damar and Balaban, 2006) Yen and Lin ( 1996) studied the effects of nonthermal pasteurization on guava puree. They compar ed the effect of high pressure pasteurization treatment and thermal pasteurization on the quality and shelf life of guava puree. Puree with a pH of 3.8 and 8.2 Brix was sub jected to either 400 or 600 MPa for 15 min at 25 C. The heat pasteurization process was carried out at 88 90 C for 24 seconds. Samples from both treatments were stored at 4 C over a period of 60 days. All treatments were equally

PAGE 96

96 effective in reducing the microbial load ( > 2 log reduction) of the puree. During storage, puree pressurized at 600 MPa showed lower levels of microorganisms when compared to 400 MPa. Cloud reduction was greater in untreated puree than in pressurized and heated puree during sto rage. Color of pressurized guava puree was similar to that of freshly extracted puree. Results indicated that high pressure treatment of guava puree at 25 C for 15 min could maintain good quality up to 40 days of storage at 4 C Dense Phase C arbon D ioxide (DP -C O2) ha s been used as a nonthermal preservation technique (Dagan and Balaban 2006; Damar and others 2009; Ferrentino and others 2009; Lim and others 2006 ; Del Pozo-Insfran and others 2006). It consists of submitting a mixture of juice and CO2 t o pressures under 50 MPa T he amount of CO2 dissolved in the juice increases with increasing pressure. As CO2 solubility increases, pH of the juice decreases. Once the pressure is released the CO2 is separated from the juice and the pH returns to its original value. Ferrentino and others ( 2009) treated r ed grapefruit juice with continuous DP -CO2 to inactivate total aerobic microorganisms and yeasts and molds (Y&M) They achieved a 5 log reduction for Y&M and aerobic plate counts Cloud increased (91%) while a partial inactivation of PE (69.17%) was achieved. No significant ( = 0.05) differences were detected between treated and fresh samples for Brix, pH, TA, lightness, redness and yellowness values In the present study, a continuous DP -CO2 system was used to pasteurize guava puree. The independent parameters for the process were: pressure, temperature, residence time and % CO2. The solubility of CO2 and therefore the exact amount of CO2

PAGE 97

97 use d during DP -CO2 was determined using an apparatus designed in the Department of Food Science and Human Nutrition (Calix and others 2008). The objectives of this study were: (1) to perform CO2 solubility experiments on guava puree; (2) to optimize DP -CO2 process parameters for microbial reduction, and; (3) to study microbial stability of DP -CO2 treated puree compared to fresh and thermally treated puree during storage. Materials and Methods Guava Puree Frozen unpasteurized red guava puree doctor Rubi cultivar was obtained from the Goya Company (San Cristobal City, Dominican Republic). The puree was held at 20 C until the time of processing when it was thawed at 4 C for one week. Part of the insoluble solids and stone cells were removed by straining the thawed puree through a 200 m nylon filter (Cole Palmer, Vernon Hills, IL, U.S.A) (Figure B 1). The removal of the insoluble solids led to a 60% yield and improved the pumpability of the puree. Model S ystem Four liters of a model system were prepared the same day of the study. For this purpose glucose, fructose, citric acid and ascorbic acid (Fisher Scientific, Fair Lawn, NJ, U.S.A .) were used Total soluble solids (TSS) and pH of the model system were 6.8 and 3.78, respectively. Solubility M easurements CO2 solubility in the guava puree, water and a model system was measured between 6.9 and 31.03 MPa at 35 C using an apparatus designed and built at the University of Florida Food Science and Human Nutrition department (Gainesville, FL) as previously described by (Ferrentino and others 2009). In this batch system (Figure B -2)

PAGE 98

98 a known volume of the puree was saturated by bubbling CO2 through it at the desired experimental conditions and the dissolved CO2 was measured at atmospheric pressure. Solubility of CO2 in water and the model system was measured for comparison. Solubility was expr essed as g of CO2/100 g liquid sample. After solubility measurement, pH, TSS and TA were determined. Processing Equipment The DP -C O2 system was constructed by APV (Chicago, IL) for Praxa ir (Chicago, IL) and provided to the University of Florida (G ainesville, FL) (Figure B -3). The equipment wa s capable of continuously treating liquid foods with CO2 at pressures up to 69 MPa. Th e system consist ed of CO2 tanks and a CO2 pump, a product tank and a product pump, a high pressure pump, a holding coil (79.2 m, 0.635 cm i.d.) a decompression valve and a vacuum tank (Fig. 4 -1). CO2 and product we re pumped through the system and mixed before passing through the high pressure pump, which increased th e pressure to the process levels The product temperature wa s brought to 35 C in the holding coil by heating tape Residence time wa s adjusted by setting the flow rate of the product passing through the holding coil At the end of the process, an expansion valve wa s used to release pressure and separate CO2 from the mixture, and the juice was collected into sterile bottles as previously described (Damar 2006 ). Whenever the treatment parameters were changed, sterile water was pass ed through the system until the desired processing conditions were stabilized. The equipment was cleaned after each use as described by (Damar 2006). For emulating commercial thermal processing conditions (90 C for 60 sec), a MicroThermics electric UHT/ HTST (Ultra High Temperature & High Temperature Short Time) Lab Model 25 (MicroThermics Inc., Raleigh N C, U.S.A.) was used. The

PAGE 99

99 equipment wa s capable of continuously pasteurizing a variety of products at temperatures between 76 and 152 C (170 305 F). T h e system consist ed of two tubular product heaters, holding tubes, two independent tubular product coolers and a variable speed positive displacement product pump. Equipment was taken to stable processing conditions using water. After processing conditions were reached, product processing was conducted. T he product was pumped through the system and preheated (90 + 0.5 C) before passing through the holding tube. T he product temperature (90 + 0.3 C) wa s brought to the desired processing temperature in the holding tube Residence time wa s adjusted by setting the flow rate (400 mL/min) of the product using the product pump. After the pasteurization process product exiting the holding tube was chilled in a two step process using tubular coolers to 3 C. Cold product was collected in sterile bottles and placed in a cooler with ice. Processed product was transported to the laboratory for analysis. The equipment was cleaned before and after each use as described by the manufacturer, using Sani -T -10 desinfectant/s anitizing solution (Spartan Chemical Company, Maumes, OH, U.S.A.). Microbial Inactivation Study Preliminary investigations were conducted to determine the DP -CO2 parameters that could result in a required 5 log reduction of APC and Y&M using response surface methodology. Microbial counts (Y&M and APC) were used as the dependent variable in the experimental design. The study required 11 experiments with 4 factorial points, 4 star points and 3 center points for replications. A high initial microbial load i n the juice (1.78 x 107 cfu/mL for total aerobic microorganisms, 1.97 x 105 cfu/mL for yeasts/molds) was achieved after incubating the sample for 4 days at 21oC. Microbial log reduction was determined for each experimental run as: log (initial number of cf u/mL) log

PAGE 100

100 (number of cfu/mL after treatment). Microbial inactivation was evaluated immediately after processing. Microbial counts were made from triplicate samples of each processing treatment serially diluted (1 x 101 to 1 x 106) in duplicate by mixing 1 0 mL of each puree with 90 mL of sterile Butterfields buffer. Total plate counts were determined by aerobic count Petrifilms and yeast/mold Petrifilms (3M Petrifilm Microbiology Products, St. Paul, MN U.S.A.) by plating 1 mL of the dilutions onto the Petrifilms in dupli cate and enumerat ion after 48 h at 35 C and 72 h at 24 C, respectively, according to the manufacturers guidelines. Experimental data were analyzed by regression analysis using SAS 9.0 software (SAS Institute Inc., Cary, N C, U.S.A. ), fit to quadratic polynomial equations, and the results used to select the optimal DP CO2 processing conditions ( 34.5 MPa, 8%CO2 and 6.9 min) for assessment of microbial stability. Storage Study and Microbial Stability Fresh guava puree (not spoiled) was treated at 34.5 MPa, 8% CO2 and 6. 9 min residence time (parameters were based on DP -CO2 optimization results) and stored u n der refrigerated conditions (4 C ). Three 1 L bottles of puree were analyzed by the methods described below on a weekly basis for the first 6 wk s of storage followed by 2 wks intervals up to 14 wks of storage. Chemical Analyses PE activity : PE activity was measured by the method of Rouse an d Atkins (1955) Ten grams of g uava puree were placed in a beaker with 40 mL of 1% pectin solution. T he sample was warmed in a water bath to 30oC While stirring and maintaining constant temperature, 2 N NaOH was added until the pH was stable at 7. NaOH ( 0.05 N ) was added until the pH was between 7.6 and 7.8 and the exact pH w as recorded.

PAGE 101

101 One hundred L of 0.05 N NaOH w ere added and the time required for the solution pH to recover between 7.6 and 7.8 was recorded and used in the following equation. PE units/g = (mL NaOH) (Normality of NaOH)(1000) (time)(g sample) Cloud : Cloud was measured as described by Kincal and others (2006). Guava puree (1.50 mL) was poured into a 1.5 mL centrifuge tube, placed in an Eppendorf centrifuge (Model 5415; Brinkman Instruments Inc., Westbury, N.Y ., U.S.A.), and centrifuged at 320g for 10 min at room temperature. The supernatant (300 L ) was placed in a spe ctrophotometer (SpectraMax 190, Molecular Device s Sunnyvale, CA) and absorbance at 660 nm was recorded as the cloud value with distilled water serving as blank ( Versteeg and others 1980) pH : Th e pH of treated and untreated guava puree samples was measured using a digital pH meter (Accumet Basic AB15, Fisher Scientific, Fair Lawn, N J U.S.A .) after calibrat ion with commercial buffer solutions a t pH 7.0 and 4.0. A sample (4 0 mL) was placed in a 50 mL beaker with a magnetic stir bar, the pH electrode was inserted, and pH was recorded after stabilization. Brix : A digital refractometer with temperature correction (Abbe Mark II; Reichert Scientific Instruments, Buffalo, N.Y. U.S.A.) was used (Redd and others 1986) Viscosity: Viscosity measurements were obtained using a Brookfield DV -E Digital Viscometer (Brookfield Engineering Laboratories, Middleboro, MA, U.S.A.). Five hundred mL of puree were plac ed in a 1 L beaker. A spindle number 1 and 50 rpm were used to obtain the viscosity measurement, which was expressed as cP (centipoise or mPa.sec).

PAGE 102

102 Statistical Analysis: Repeated measures ANOVA and mean separation using d to evaluate the effect of treatment [ fresh (C ontrol), thermal and DP -C O2 processed ] and storage time (0, 1, 2, 3, 4, 5, 6, 8, 10, 12, and 14 weeks) on microbiological results and physicochemical parameters using SAS statistical 9.0 software (SAS Institute Inc., Cary, N.C., U.S.A.). Results and Discussion Solubility E xperiments The solubility of CO2 in guava puree, in a model system and in water was measured at 35 C and pressures ranging from 7.58 to 31 .03 MPa. Figure 4 -2 illustrates that by increasing the pressure, the solubility of CO2 increased. The presence of dissolved solutes such as simple sugars acids and other carbohydrates lowered the amount of CO2 that dissolved in the puree, resulting in a significant difference between the CO2 solubility in water in the model system and in the puree. The puree showed a significantly lower CO2 results were observed by (Calix and others 2008) when the CO2 solubility in orange juice, an orange juice model system and water was studied. Ferrentino and others ( 2010) suggested that the observed decrease in CO2 solubility in sodium phosphate solutions was affected by the type of chemical compound dissolved in the liquid matrix The solubility increased significantly from 7.58 to 10.34 MPa and remained alm o st constant at 4. 3 g of CO2/100 g pure e from 10 34 to 31.03 MPa. The final solubility for the puree was not significantly different from the solubility of the model system at 7.58 MPa.

PAGE 103

103 Table 4 1 shows the Brix, pH, and titratable acidity for the puree before and after the solubility measurements. Data are shown in Tables B 1 and B 2. A decrease in pH and an increase in titratable a cidity (TA) and Brix with increasing pressure was observed, probably due to the residual CO2 remaining in the puree after depressurization. As expected, an increase in TA and oBrix was obs erved. These results are related to increase in solubility of CO2 when processing pressure is increased. Similar results were observed by (Calix and others 2008) in apple and orange juice. C onsidering that the solubility value wa s obtained under saturation and long contact time conditions in a static equipment the %CO2 was set to 5.3 g CO2/100 g sample Microbial I nactivation Study The effect of DP -CO2 at various processing pressures and treatment times on Y&M and AP C can be observed in Table 42. Initial and final numbers of bacteria were determined by taking average cfu/mL counts on P etrifilms with the cfus less than 2 5 0 cfu/mL The %CO2 used for the experiments was set to a constant value of 8.0 g CO2/100 g puree, although in the solubility study, the experimental value was 5.3% The reason for this additional increment was due to fluctuations in the CO2 flow (equipment limitations). Results from Table 42 demonstrated that under identical processing conditions, Y&M were destroyed at a higher rate than aerobic microorganisms. Similar results were observed b y Del Pozo-Insfran and others ( 2006) when muscadine grape juice was processed by DP -CO2. The average initial and final aerobic plate counts (APC) standard deviations at each experimental c ondition are given in Table C -1. The RSM analysis of data was performed with SAS 9.1 statistical software program (Cary,

PAGE 104

104 NC, U.S.A.). The SAS code and output of the analysis are given in Table C 2 and C 3, respectively. The statistical analysis of the res ponse surface regression model showed that the quadratic model fit for Y&M was statistically significant ( P < 0.05) and there was a satisfactory correlation between the actual and the fitted values for Y&M ( R2 = 0. 88 ). Significance of each parameter was dec 05 level and the parameters with p value > 0. 05 were excluded from the model. Results showed that only the parameters pressure and pressure by residence time were significant. Lecky ( 2005) observed little effect by increasing residence time from 4 to 5 min with watermel on juice treated at a pressure of 34.4 MPa, 40 C and 10% CO2. Gunes and others (2005) reported that microbial inactivation by DP CO2 is governed by the transfer rate and the p enetration of carbon dioxide into cells, which can be improved by increasing pressure, and increasing the processing temperature The model with the estimated coefficients gives the prediction of log microbial reduction (log red) as a function of pressure and residence time: Y&M = 6.0605 0.1619* Pressure (MPa) + 0.018945* Pressure Residence time (min). Coefficients were determined for the coded values of each variable. The log reductions predicted at eleven experimental runs using this equation are close t o the experimental log reductions (Table 43 ). Statistical analysis for the APC showed no significant difference within the 11 experimental runs so processing parameters were based on Y&M reduction. Microbial inactivation is highly dependent on the type of microorganisms present in the food matrix due to distinct microbial cell microstructure and the diffusion of CO2 into the microbial cell (Ballestra and others 1996). Bacterial vegetative cells and Y&M are more pressure and CO2 sensitive than bacterial s pores, e.g. higher pressures

PAGE 105

105 are required for complete inactivation of spores Overall, microbial reduction is related to the direct relationship between CO2 solubility and increasing processing pressure (Balaban and others 1991a) which consequently affect s the diffusion of CO2 into the microbial cell as well as the explosive decompression that occurs during DP -CO2 processing. Results of this optimization study were used to determine those DP CO2 conditions that achieved > 5 log reduction of Y&M, rather tha n bacterial cells. According to the results obtained from the regression surface analysis, 8 min was enough to achieve this log reduction under the established processing conditions. Because of the increase in product thickness and foam formation, the re sidence time was set to 6.9 min. A higher bacterial reduction would be achieved at 6.9 min rather than 6.5 min under the same processing conditions. For the stability study, the processing conditions used for the DP CO2 were pressure of 34.5 MPa, 8% CO2, a residence time of 6.9 min and a temperature of 35 C. Table 4 4 and 4 -5 present the results obtained for PEA, cloud meassurement, pH, Brix and TA. There were significant differences in the PEA of the treated puree when compared to the control. An apparent increase in PEA was observed. The method used for PEA determination is based on adjusting sample pH to the optimal pH of the enzyme activity (7.6 to 7.8). After this, an excess of basic solution is added and the change in pH due to enzyme activity and liberation of carboxyl groups is monitored. The increase may be due to microbial enzyme leakage into the puree. One of the inactivation methods for DP -CO2 is cell wall rupture, which can cause a leakage of enzymes into the puree. Another possible explanation for this increase in PE activity is the release of PE

PAGE 106

106 from the cell wall of plant material. PE is a cell wall -bound pectic enzyme complex that can be released from the pulp during DP -CO2 process (Tribess and Tadini 2006) Cloud results show ed si gnificant differences between the treated sample and control for all treatments .There was a significant cloud loss for all treatments, but cloud loss was hig h er with an increase in pressure and residence time. Cloud loss was between 59.6 and 80.9% for a pr ocessing pressure of 13.8 MPa. For a pressure of 34.5 MPa, cloud loss was between 66.11 and 90.82%. These results were not as expected since different results were observed by Kincal and others ( 2006) in orange juice treated with high pressure carbon dioxi de processing. sample and the control. At 5 and 6.5 min residence time, there were significant differences in Brix between the treated sample and the control. The Brix of th e treated sample increased with increasing pressure and residence time. The increase in Brix is related to increase in solubility of CO2. Increasing pressure and residence time increases the amount of CO2 that dissolves into the product. There were signif icant differences in TA between the treated sample and the control. The TA increase s with the DP -C O2 treatment As p ressure and residence time increase s TA increase s The increase in TA can be related to the increase in CO2 solubility due to increasing pressure and cont act time (Lecky 2005) Storage Study and Microbial S tability D ifference s between treatment means for the storage study data w ere determined by conducting analysis of variance using SAS 9.1 Software (Cary, NC, U S. A. ) at a significance level of = 0.05. Tukeys standardized range comparison test ( = 0.05) was used to determine statistically different samples. The plot of APC for untreated

PAGE 107

107 (control ), heat -treated (90 C, 60 s) and DP -C O2 treated (34.5 MPa, 3 5 C, 8 % CO2, 6 .9 min) guava puree during 14 weeks of refrigerated storage (4 C) is given in Figure 4 -3 The thermal pasteurization process of guava puree showed the lowest count for aerobic bacteria. This count remained almost constant duri ng the first 4 weeks of storage The pasteurization process showed a characteristic microbial growth pattern in which the three phases can be identified. The lag phase is represented during the first 4 weeks of storage, the log phase can be observed between week 4 and 6 and the death phase can be observed after week 7. As shown in Figure 43, the thermal process was more effective in bacterial reduction than DP -CO2. Aerobic counts in DP -CO2 treatments were not significantly different from the control puree during the first 8 weeks of storage but showed a significant increase after week 10. The bacterial count on DP -CO2 decreased after week 10 and was not different from control. The number of aerobic bacteria in untreated samples remained almost unchanged during the 14 weeks of storage. Storage of the thermal and DP -CO2 treated puree (not spoiled) at 4 C showed no significant differences in the Y&M during the first 6 weeks of storage. Both treatments showed a significantly lower count when compared to the control (Figure 44). Between week 6 and 8, there was a significant increase in the Y&M count for the thermal pasteurized sample in which there are no differences in counts between control and pasteurized samples. After 10 weeks of storage, the Y&M count for pasteurized sam ples decreased and remained almost constant until 14 weeks of storage. DP -CO2 samples showed a steady increase in Y&M counts during the first 8 weeks of storage. After week 12, a significant decrease in Y&M counts was observed for DP -CO2 Control puree showed the highest Y&M count and these counts remained almost the same

PAGE 108

108 throughout the 14 weeks of storage. Initial Y&M counts were 3.3 log and DP -CO2 was effective in decreasing Y&M count to 1.3 log but did not cause any reduction on aerobic count, APC of DP -CO2 were similar to control. The pasteurization process caused 2 and 1.3 log reduction on APC and Y&M respectively when compared to control which had an initial APC of 3.5 log and 3.3 log for Y&M. DP -CO2 and thermal processing significantly decreased PE activity when compared to the fresh sample. PE activity decreased 21.45% and 51% for DP CO2 and thermally treated samples respectively. There was a continuous decrease in PE activity for all the samples during the 14 weeks of storage. Dur ing the first 5 weeks of storage, there were no significant differences in enzyme activity between the DP -CO2 treated and control samples (Figure 45). Between week 6 and 12, fresh samples had a significantly higher PE activity when compared to the DP -CO2 treated samples and there were no significant differences between the enzyme activity of control and DP C O2 treated samples at the end of the storage study. Similar results were shown by Kincal and others ( 2006) when orange juice was treated with a continu ous DP -C O2 system under different pressures at constant residence time. Ferrentino and others (2009) reported an reduction in PE activi ty after treating grapefruit juice with DP -CO2. A total reduction in PE activity of 47% and 52% was observed at the end o f the storage study for DP CO2 and pasteurized samples respectively. Residual PEA in thermal pasteurized samples can be explained by the presence of PE isoforms which possess different inactivation kinetics. Thermal stable isoform have been reported in orange juice (Versteeg and others 1980; Randall and others 1998) and this enzyme is more heat resistant than common spoilage microorganisms. Versteeg and others (1980) studied

PAGE 109

109 the juice cloud destabilizing properties and heat stability of PE in navel orange and found that a paste urization time of 0.8 min at 90 C would be necessary to inactivate 99% of the high molecular weight PE. This high molecular weight enzyme is the thermo stable isoform. PE inhibition for control sample was observed during the 14 week s of storage. Temperature is known to affect enzyme activity. Every enzyme has an optimal temperature below which the enzy me activity usually decreases. Versteeg and others (1980) suggested that a tem perature dependant change in conformation causes this de crease in PE activity at 5 C. The PE reduction noted in this study may be due to storage temperatures at 4 C. Colddeactivation of enzymes can occur at low temperatures or bellow 10 C Low temperature reduces the strength of non-polar forces promoting the dissociation of subunits and compromising the enzyme activity (Damodaran and others 2008). Viscosity measurements for fresh, DP -CO2 and thermally treated samples are shown in Figure 46. DP -CO2 treatment had a significantly higher viscosity compared to fresh and thermally treated samples. The increase in viscosity was 60%. During the first 4 weeks of storage, DP -CO2 showed a maximum increase in viscosity of 95% which decreased progressively to 6 9% after week 5 of storage. The possible explanation for the increase in viscosity is the formation of a reversible gel (Figure C 1). At the beginning, formation of bridges between calcium and pectic acid was suggested as a mechanism for the reversible gel formation. Another possible explanation is the particle size reduction of guava puree during the decompression of the puree at the end of the treatment. Lim and others ( 2006) found that during treatment of mandarin juice with a continuous high pressure CO2 system, calcium precipitation occurs. When dissolved in

PAGE 110

110 water, CO2 combines with water and forms carbonic acid that dissociates into bicarbonate and hydrogen ions. When the pressure is released, bicarbonate is converted to carbonate and may precipitate c alcium, preventing it from acting as a bridge between pectic chains that leads to cloud loss (Lim and others 2006) Cloud stability was measured and results were reported as absorbance values at 660 nm. As absorbance decreased, cloud loss increased. When performing the centrifugation, no phase separation was achieved for the DP -CO2 samples. This was mainly due to the reversible gel formation. To achieve a phase separation, all samples were frozen until the end of the storage study. Freezing changes the molecular structure of the gel, causing a phase separation and cloud loss. Results for cloud loss are presented in Table 4.6. These results indicated that cloud loss for the DP -CO2 treated samples after freezing was higher than the fresh and pasteurized ones Pasteurized samples showed an increase in cloud during the 14 weeks of storage. This can be related to inactivation of PE during the thermal treatment. Control samples showed significant reduction in cloud due to PE activity during refrigerated storage. In DP -CO2 processing, i t is hypothesized that the depressurization of the system leads to homogenization of the puree causing smaller particles of the puree colloids increasing cloud stability. Even when the results are reported as cloud loss, more studies need to be conducted to understand the mechanism of gel formation and the chang es that occur during freezing. Ferrentino and others ( 2009) showed cloud retention after treating grapefruit juic e with a continuous DP -CO2 system, even when PE was active in the treated juice. Kincal and others ( 2006) also showed that cloud increased in orange juice treated with a continuous DP -C O2 system When treating orange juice with a static high-

PAGE 111

111 pressure CO2 s ystem, Aerrola and others ( 1991) showed that cloud enhancement occurs. The mean pH and Brix of the control, DP -CO2 and thermal treated samples showed some weekly fluctuations during storage (Table 4 -6). pH for control and DP CO2 treated samples decreased during week 4 of storage, significantly increased during week 5 and then stayed almost constant. The decreased in pH may be related to bacterial metabolism during which carbohydrates are hydrolyzed and acid molecules are released causing a pH reduction. T hermal treated samples had a significantly higher pH than control and DP -CO2 treated samples (related to bacterial inactivation during thermal processing). The re were significant differences in Brix between fresh, DP -CO2 and thermal treated samples. In general, DP -CO2 showed a significantly higher Brix than control samples during the first 6 weeks and between week 12 and 14 weeks of storage. The apparent increase in Brix measurement on DP -CO2 samples may be related to residual CO2 dissolved in the samples. The fluctuations in pH and oBrix were probably because of bottle -to -bottle variations Conclusions DP -C O2 treatment resulted in a guava puree with reduced microbial load, causing 5 log reductions in Y&M and a maximum of 3.5 log r eductions in APC Solubility study showed that a 4.3% of CO2 caused saturation of guava puree in a static system. Due to equipment limitations, a value higher than the experimental saturation solubility of %CO2 was used. The experimental design and model f or determining CO2 solubility minimized the excess use of CO2 during processing and can be used to economically optimize DP -CO2 treatment. Optimal conditions for DP -CO2 to achieve a 5 log reduction in Y&M were determined as 8% CO2, 35 C, 6.9 min of reside nce time and a pressure

PAGE 112

112 of 34.5 MPa. These conditions maintained guava puree microbiologically stable during 14 weeks of refrigerated storage at 4 C. DP -CO2 samples showed pH similar to fresh samples but TA for the DP -CO2 was higher than the untreated one s Increases in viscosity and cloud loss were observed for the DP -CO2 treated puree. In addition, PE was partially inactivated A 20% reduction on PEA was initially achieved after DP -C O2 processing Further study is recommended to understand the mechanism of gel formation and cloud loss due to DP -CO2 treatment. From the storage study, it is evident that DP -C O2 processing can maintain quality attributes of fresh puree, and extend its shelf life. Figure 41 Schematic diagram of Dense Phase Carbon Dioxide equipment

PAGE 113

113 Figure 42. Carbon dioxide solubility results obtained for guava puree, water and guava puree model system at different processing pressures. Error bars represent standard deviation for n=3. D ifferent letters within each pressure represent Table 4 1. pH, oBrix and titratable acidity values for the treated and untreated guava puree under different processing pressures Pressure (MPa) pH Brix TA (g citric acid/100 g puree) Control 3.80 a 6.8 a 0.55 a 7.58 3.7 9 a 6.8 a 0.55 a 10.34 3.7 5 b 6.9 ab 0.58 b 17.24 3.75 b 6.9 ab 0.60 b 24.13 3.74 b 6.9 ab 0.60 b 31.03 3.72 b 7.0 b 0.66 c Different letters within the columns represent significant differences 0 1 2 3 4 5 6 7 5 10 15 20 25 30 35CO2 Solubility (g of CO2/100 ml of liquid)Pressure (Mpa) guava A A B B C D E E E E D H H H G G

PAGE 114

114 Table 4 2. Processing conditions and microbial reduction obtained during DP -CO2 process optimization Run Pressure (MPa) Residence time (min) Juice Flow (g/min) CO 2 flow (g/min) log 10 Red log 10 Red APC Y&M 1 34.5 8 312.5 25 3.20 4. 30 2 24.1 8 312.5 25 3.2 8 4.28 3 24.1 6.5 384.6 30.8 3.3 2 4.07 4 13.8 8 312.5 25 3.23 4.0 9 5 34.5 6.5 384.6 30.8 3.5 1 3.86 6 24.1 6.5 384.6 30.8 3.52 3.9 7 7 13.8 6.5 384.6 30.8 3.5 9 4.03 8 13.8 5 500 40 3.57 4.1 8 9 34.5 5 500 40 3.7 1 3.2 1 10 24.1 5 500 40 3.6 6 3.49 11 24.1 6.5 312.5 25 3.65 3.65 Table 4 3 Actual and predicted yeast and mold log reduction using the equation from the surface response analysis Run Pressure (MPa) Residence time (min) Actual Predicted Predicted Log red uction ( all terms ) ( sign ificant terms only ) 1 34.5 8 4.3 4.38 5.7 0 2 24.1 8 4.28 4.19 5.8 0 3 24.1 6.5 4.08 3.9 0 5.12 4 13.8 8 4.09 4.1 0 5.92 5 34.5 6.5 3.86 3.79 4.72 6 24.1 6.5 3.97 3.9 0 5.12 7 13.8 6.5 4.03 4.1 0 5.52 8 13.8 5 4.18 4.1 0 5.13 9 34.5 5 3.21 3.19 3.74 10 24.1 5 3.49 3.59 4.44 11 24.1 6.5 3.65 3.9 0 5.12

PAGE 115

115 Table 4 4. Pectinesterase activity, cloud and pH measurement of guava puree before and after DP -CO2 treatment Run Pressure (MPa) Residence time (min) Initial PEA (units/g sample) Final PEA (units/g sample) Cloud reduction (%) Initial pH Final pH 1 34.5 8 2.86 4 + 9.93 6 4.35 4 + 5.01 5 90.82 3.7 + 0.03 3.7 + 0.02 2 24.1 8 2.86 4 + 9.93 7 4.37 4 + 424 5 79.84 3.7 + 0.0 3.7 + 0.01 3 24.1 6.5 2.86 4 + 9.93 8 4.27 4 + 1.44 5 86.2 3.7 + 0.03 3.7 + 0.03 4 13.8 8 2.86 4 + 9.93 9 5.70 4 + 4.93 5 80.9 3.7 + 0.03 3.7 + 0.02 5 34.5 6.5 3.62 4 + 1.6 0 5 4.98 4 + 6.41 5 85.89 3.7 + 0.01 3.7 + 0.02 6 24.1 6.5 3.62 4 + 1.60 6 4.47 4 + 3.23 5 84.23 3.7 + 0.01 3.7 + 0.01 7 13.8 6.5 3.62 4 + 1.60 7 4.58 4 + 2.31 5 59.6 3.7 + 0.01 3.7 + 0.02 8 13.8 5 3.62 4 + 1.6 0 8 4.93 4 + 2.21 5 77.06 3.7 + 0.01 3.7 + 0.03 9 34.5 5 4.86 4 + 2.83 5 4.73 4 + 1.97 5 66.11 3.7 + 0.01 3.7 + 0.02 10 24.1 5 4.86 4 + 2.83 6 4.91 4 + 4.43 5 78.84 3.7 + 0.01 3.7 + 0.01 11 24.1 6.5 4.86 4 + 2.83 7 4.96 4 + 3.63 5 80.26 3.7 + 0.01 3.7 + 0.02 Mean + standard deviation for n=3

PAGE 116

116 Table 4 5. Brix and titrabable acidity of guava puree before and after DP -CO2 treatment Run Pressure Residence time (min) Initial Brix Final Brix Initial TA (%)* Final TA (%)* 1 34.5 8 7.1 + 0.00 7.1 + 0.00 0.62 + 0.01 0.65 + 0.01 2 24.1 8 7.1 + 0.00 7.2 + 0.04 0.62 + 0.01 0.66 + 0.01 3 24.1 6.5 7.1 + 0.00 7.1 + 0.05 0.62 + 0.01 0.67 + 0.01 4 13.8 8 7.1 + 0.00 7.2 + 0.04 0.62 + 0.01 0.69 + 0.01 5 34.5 6.5 6.9 + 0.04 6.9 + 0.05 0.64 + 0.01 0.70 + 0.00 6 24.1 6.5 6.9 + 0.04 6.9 + 0.00 0.64 + 0.01 0.68 + 0.01 7 13.8 6.5 6.9 + 0.04 6.9 + 0.00 0.64 + 0.01 0.70 + 0.01 8 13.8 5 6.9 + 0.04 6.9 + 0.00 0.64 + 0.01 0.69 + 0.01 9 34.5 5 6.7 + 0.05 6.7 + 0.00 0.63 + 0.01 0.67 + 0.02 10 24.1 5 6.7 + 0.05 6.7 + 0.00 0.63 + 0.01 0.69 + 0.01 11 24.1 6.5 6.7 + 0.05 6.7 + 0.00 0.63 + 0.01 0.66 + 0.01 Mean + standard deviation for n=3, Expressed as g citric acid per 100 g sample

PAGE 117

117 Figure 43. Aerobic plate count for control, DP -CO2 and t hermal treated guava puree during 14 weeks of storage. Error bars represent standard deviation for n=6 0 1000 2000 3000 4000 5000 6000 7000 0 3 6 9 12Number of microorganisms (cfu/mL)Storage time (weeks) Control DPCD Pasteurized

PAGE 118

118 Figure 44. Yeast and mold plate count for control, DP -CO2 and t hermal treated guava puree during 14 weeks of storage. Error bar s represent standard deviation for n=6 0 1000 2000 3000 4000 5000 6000 0 2 4 6 8 10 12 14Number of microorganism (cfu/mL)Storage time (weeks) Control DP -CO2 Pasteurized

PAGE 119

119 Figure 45. Pectinesterase activity for control, DP -CO2 and t hermal treated guava during 14 weeks of storage. Error bars represent standard deviation for n=3. 0.00E+00 1.00E -04 2.00E -04 3.00E -04 4.00E -04 5.00E -04 6.00E -04 7.00E -04 8.00E -04 0 1 2 3 4 5 6 8 10 12 14 Pectinesterase Activity (PE units/g)Storage Time (weeks) Control DP CO2 Pasteurized

PAGE 120

120 Figure 46. Viscosity measurement for control, DP -CO2 and t hermal treated guava during 14 weeks of storage. Error bars represent standard deviation for n=3 0 100 200 300 400 500 600 700 0 2 4 6 8 10 12 14 16Viscosity (Centipoise) Storage Time (weeks) Control DP CO2 Pasteurized

PAGE 121

121 Table 4 6. pH, oBrix and cloud measurement for control, DP -CO2 and t hermal treated guava during 14 weeks of stor age pH Brix Cloud Week Control DP C O 2 Pasteurized Control DP C O 2 Pasteurized Control DP C O 2 Pasteurized 0 3.86 + 0.01 3.87 + 0.01 3.87 + 0.01 6.9 + 0.0 7.1 + 0.0 7.0 + 0.0 0.80 + 0.12 0.13 + 0.03 0.62 + 0.05 1 3.87 + 0.01 3.87 + 0.01 3.90 + 0.01 6.9 + 0.0 7.1 + 0.0 7.0 + 0.0 0.67 + 0.04 0.04 + 0.01 0.80 + 0.03 2 3.82 + 0.01 3.88 + 0.01 3.81 + 0.01 6.8 + 0.1 7.1 + 0.1 7.0 + 0.1 0.70 + 0.08 0.03 + 0.00 0.83 + 0.05 3 3.85 + 0.01 3.85 + 0.01 3.86 + 0.01 6.8 + 0. 1 7.1 + 0.0 7.0 + 0.1 0.54 + 0.08 0.05 + 0.01 1.02 + 0.02 4 3.77 + 0.01 3.72 + 0.01 3.95 + 0.01 6.7 + 0. 1 7.0 + 0.1 7.1 + 0.1 0.47 + 0.02 0.06 + 0.01 1.11 + 0.04 5 3.84 + 0.00 3.82 + 0.01 3.94 + 0.01 6. 7 + 0 .1 7.0 + 0.1 7.0 + 0.1 0.38 + 0.02 0.03 + 0.01 1.09 + 0.06 6 3.89 + 0.01 3.82 + 0.01 3.93 + 0.01 6.8 + 0.1 6.8 + 0.1 7.0 + 0.0 0.41 + 0.02 0.03 + 0.01 1.08 + 0.04 8 3.88 + 0.00 3.84 + 0.01 3.97 + 0.01 6.8 + 0.1 6.8 + 0.1 7.2 + 0.0 0.33 + 0.04 0.04 + 0.00 1.13 + 0.05 10 3.84 + 0.0 3.85 + 0.0 3.94 + 0.0 6.9 + 0.1 6.9 + 0.1 7.1 + 0.0 0.26 + 0.03 0.05 + 0.01 1.05 + 0.05 12 3.85 + 0.01 3.86 + 0.01 3.93 + 0.02 7.0 + 0.1 7.2 + 0.1 7.2 + 0.1 0.19 + 0.01 0.05 + 0.00 1.16 + 0.03 14 3.85 + 0.00 3.83 + 0.01 3.93 + 0.02 7.0 + 0.0 7.3 + 0.1 7.4 + 0.1 0.21 + 0.02 0.05 + 0.00 1.21 + 0.03 Mean + standard deviation for n=3

PAGE 122

122 CHAPTER 5 PHYSICO -CHEMICAL AND PHYTOCHEMICAL CHANGES OF DENSE PHASE CARBON DIOXIDE AND T HERMALLY TREATED GUA VA PUREE DURING REFRIGERATED STORAGE Abstract Guava ( Psidium guajava L) is a tropical fruit rich in phytochemicals, antioxidants Guava puree is normally heat pasteurized to extend its shelf life, but the fresh taste is modified. The use of non-thermal processes can minimize the development of undesirable characteristics that occur during heat processing. Dense phase carbon dioxide (DP -CO2) is a non-thermal me thod emerging as an alternative to traditional thermal pasteurization. The objective of this study was to determine the changes on some physico -chemical and phytochemical qualities of DP -CO2 and thermal processing during refrigerated storage of guava puree. Storage stability of DP -CO2 and thermal process ed guava puree was assessed and compared to control puree during 14 weeks of storage at 4oC. Physicochemical (titratable acidity (TA), color (L*, a*, b) and organic acids ) and phytochemical (to tal phenolics, antioxidant capacity, ascorbic acid and identification of phenolic compounds) analyses were performed. DP -CO2 samples showed higher TA acidity when compared to control and thermal treated ones, and TA increases during storage for all 3 sampl es. DP -CO2 increases L* and a* values but no effect on a* values were observed. O rganic acid content was not affected by treatment but little change were observed during storage During storage, all puree samples showed no significant difference ( >0.05) i n antioxidant capacity (ORAC) and total soluble phenolics (TP). Processing guava puree with DP -CO2 is a viable technology for the preservation of product quality

PAGE 123

123 Introduction Guava ( Psidium guajava L) is a tropical fruit rich in antioxidants and vitamin C. It is a member of the Myrtaceae family, which has more than 80 genera and 3,000 species distributed throughout the tropics and subtropics (Nakasone and Paull 1998). The genus Psidium includes five species, with P. guajava being the most widely cultivated species of the family Myrtaceae. The chemical composition of the fruit varies with the stage of development, variety and season. T itratable acidity (TA) reported as citric acid content ranges from 0.08 to 2.20%. Guava fr uits consist of about 20% peel, 50% flesh (pericarp) and 30% seed core (Salunkhe and Kadam 1995) Among fruit types, guava is the second highest in vitamin C content, containing up to five times the amount in oranges (Dweck 2005). Vitamin C (ascorbic acid) is water soluble and highly susceptible to oxidative degradation, which often is used as an index for nutrient stability during proc essing or storage (Damodaran and others 2008) The vitamin C content of the fruit fluctuates between 37 and 1,000 mg ascorbic acid per 100 g guava fruit. Vitamin C content of red-fleshed guava is higher than that of white -fleshed guava (Mowlah and Itoo 1983). Within each fruit, the distribution of vitamin C is higher in the skin than in the central portion of the flesh. The c omposition of organic acids present in guava was studied by Chang and others (1971). They found that citric and malic acids were predominant followed by tartaric, glycolic and lactic acid. Similar results were found by Wilson and others (1982) i n a study o f four cultivars from Florida. They found traces of fumaric acid, which was detected for the first time in guava. Guava fruits contain significant amounts of polyphenols but their concentration and corresponding astringency decreases as the fruit matures.

PAGE 124

124 Fruits and vegetables are important components of a healthy diet and are one of the main sourc es of antioxidants Clinical research supports the fact that consumption of fruits and vegetables is beneficial for prevention of cancer, heart disease and other agerelated diseases (Dietary guidelines for Americans 2010) The guava fruit is rich in tannins, phenols, triterpenes, flavonoids, carotenoids, vitamins and fiber. Most of the guavas therapeutic activity is attributed to the high content of flavonoids, which also have antimicrobial activity. Dietary flavonoids and other plant phenolics have been reported to have antioxidant activity, antimicrobial and anti -inflammatory action (Huang and others 1992) and have been associated with a reduced risk of cardiovascular diseases and cancer (Temple 2000; Pietta 2000). The health benefits of consuming a diet rich in dietary fiber (DF) have been extensively studied (Gary 1999). Jimenez Escrig and others (2001) evaluated guava as a source of natural antioxidant comp ounds and DF (Jimenez Escrig and others 2001) They found a remarkable antioxidant capacity related to the phenolics content Peel and pulp of Psidium guajava fruit presented high levels of DF an indigestible fraction, and phenolic compounds. They concluded that guava could be a rich source of natur al antioxidants and dietary fiber (Jimenez Escrig and others 2001). Gorinstein and others (1999) showed that the content of polyphenols (4.79 5.11 mg/100 g fresh fruit) gallic acid (340.6 408.0 g/10 0 g fresh fruit) total fiber (5.14 6.04g/100 g fresh fruit) and soluble fiber (2.39 3.01g/100 g fresh fruit) in the guava were higher than the amount found in persimmon fruit (Gorinstein and others 1999). Guava puree is the most important guava form fo r the juice industry. The puree is preserved by freezing to -20 to 0 F ( 29 to -18 C), canning, aseptic packaging or

PAGE 125

125 pasteurization. Pasteurization is conducted between 80 and 90 C (190 194 F) for 60 seconds, then the pasteurized product is cooled an d filled into containers. After pasteurization, the puree can be frozen and stored at 18oC (0oF) for up to a year but its fresh qualities are diminished. Considerable research has focused on the development of n onthermal processing technologies to avoid the detrimental chemical changes caused by heat pasteurization. High pressure processing (HPP) causes minimal changes in the fres h characteristics of foods by minimizing thermal degradation (Raso and BarbosaCanovas 2003) Yen and Lin ( 1996 ) compared t he effect of high pressure treatment and thermal pasteurization on the quality and shelf life of guava puree. Samples from both treatments were stored at 4oC over a period of 60 days. All the treatments were equally effective in reducing the microbial load ( > 2 log reduction) of the puree. During storage at 4 C pressurized puree at 600 MPa showed lower levels (< 10 cfu/min) of microorganisms compared to control samples. The color of pressurized guava puree was similar to that of freshly extracted puree. This research indicated that high pressure treatment (600 MPa) of guava puree at 25 C for 15 min could maintain good quali ty up to 40 days storage at 4 C (Yen and Lin 1996). Phenolic compounds have been associated with positive and negative attributes in terms of sensory and nutritional quality. Positive attributes include their close association with sensory and nutritional quality. Their nutritional value has been linked to prevention of cancer, antimicrobial properties, antimutagenicity, antioxidant potential, reduction of coronary heart disease risk, antiviral, anti inflammatory and antitumor activity (Sonko and Xia 2005) Their sensory attributes are related to their contribution to

PAGE 126

126 the flavor, astringency and color characteristics of foods. The anti -nutritional effect of phenolic compounds involves their reaction with proteins, carbohydrates, minerals and vitamins loweri ng the bioavailability of these nutrients or their nutritional value. In addition, phenolic compounds can adversely affect the sensory qualities of food by the production of off -flavor, their involvement in enzymatic browning, nonenzymatic browning and pre cipitation of proteins (Shahidi and Naczk 2003). The objective of this research was to compare the physicochemical and phytochemical qualities changes of guava puree after DP -CO2 and thermal processing during refrigerated storage. Materials and Methods Ch emicals and Standards Commercial standards of oxalic, malic, ascorbic, citric, gallic, ellagic and cinnamic acids were purchased from Sigma-Aldrich (St. Loui s, MO U.S.A. ). AAPH (2,2 azobis(2methylpropionamidine) dihydrochloride), fluorescein (free aci d), Trolox (6hydroxy -2,5,7,8-tetramethylchroman2carboxilic acid) and Folin-Ciocalteus reagent were purchased from Sigma-Aldrich (St. Loui s, MO U.S.A. ). Total Phenolics, Antioxidant Capacity and Ascorbic Acid Analysis Total soluble phenolic levels were measured using the Folin-Ciocalteu assay (Talcott and others 2003) and quantified as mg of gallic acid equivalents per L sample. Ascorbic acid was measured using the 2, 6 dichloroindophenol titration method. The ascorbic acid reduced the indicator dye, 2, 6 dichloroindophenol, to a colorless solution through oxidation reduction reactions. At the end point, excess unreduced dye was rose pink in acid solution. Vitamin C was titrated in the presence of metaphosphoric acid acetic acid solutions to maintain proper acidity for the reaction and to avoid

PAGE 127

127 autoxidation of ascorbic acid at high pH. The final value was expressed as mg of ascorbic acid per mL sample. The antioxidant capacity was determined using the oxygen radical absorbance capacity (ORAC) assay (Hu ang and others 2002) with data expressed in Trolox equivalents per milliliter ( mol of TE/mL). Titratable A cidity (TA) A Brinkmann Instrument (Brinkmann Instruments Co., Westbury, NY) pH meter consisting of a Metrohm 655 Disomat, Metrohm 614 Impulsomat and Metrohm 632 pH meter was used for titration of guava puree. Ten + 1.0 g guava puree sample was titrated to an end point of pH 8.2 by using standardized 0.1 N NaOH and the amount of NaOH used for titration was recorded. Percent titratable acidity (w/ w ) was expressed as percent citric acid and calculated by the following equation: %TA= (mL NaOH used)X( n ormality of NaOH)X(meq citric acid)X(100) (g sample) The value used as meq of citric acid was 0.06 4047 this value represent the molecular weight of the a cid divided by the number of equivalents (3) in the reaction. Color A nalysis Color was measured using a ColorQuest XE colorimeter (HunterLab, Reston, VA U.S.A.). Samples of 40 mL were placed in a 20 mm quartz cell and L*, a*, and b* pa rameters were recorded in reflectance, specular included mode. High Performance Liquid Chromatography Analysis Organic acids were identified and quantified by reverse phase HPLC using modified chromatographic conditions described by Gkmen and others (2000). Separation was performed on a 4.6 mm x 250 mm Acclaim 120 C18 5 m column

PAGE 128

128 (Dionex, Sunnyvale, CA U.S.A.), using 0.2 M KH2PO4, pH 2.4 (Fisher Scientific, Fair Lawn, NJ, U.S.A) as the mobile phase at a flow rate of 1.0 mL/min. Prior to organic acid analysis, all samples were passed thorough pre -conditioned Envi -18 Sep Pak cartridges (Sigma Aldrich, St Louis, MO, U.S.A.) to remove neutral polyphenolics. After discarding Bedford, MA, U.S.A.) and analyzed for organic acids. Organic acids were characterized based on UV -VIS spectral interpretation from 210 to 360 nm and comparison to authentic standards. Polyphenolics from guava puree were subsequently concentrated and purified using Envi 18 Sep-Pak Vac 20 cc mini -columns (Sigma Aldrich, St Louis, MO, U.S.A). Polar constituents were removed with acidified water (0.01% v/v HCl) and polyphenolic compounds subsequently fractioned and eluted with methanol (0.01% v/v HCl) followed by ethyl acetate. Solvent in both fractions was later evaporated under reduced pressure at <40C. The resulting polyphenolic extracts were re -dissolved in 3 mL of a MeOH:H2O (60:40) solution. Phenolics compounds were identified and quantified using a Dionex HPLC s ystem equipped with a degasser, a binary pump, an autosampler/injector and diode array (PDA 100) detector (Dionex, Sunnyvale, CA). Compounds were separated on a 250 x 4.6 mm Acclaim 120 C18 5 m (Dionex, Sunnyvale, CA U.S.A.) column. Mobile phases consist ed of acidified water (phase A) and 60% methanol in water (phase B), both adjusted to pH 2.4 with o -phosphoric acid (Fisher Scientific, Fair Lawn, NJ, U.S.A) A gradient program was used, starting at 100% phase A. The s olvent program ran phase B from 0% to 60% in 20 min; 60% to 100% in 20 min; 100% for 7 min; 100% to 0% in 3

PAGE 129

129 min and a final hold for 2 min. The flow rate was 0.8 mL/min, and detection was done at 260, 280, 320, 360 and 520 nm. Results and Discussion Total Phenolics and Antioxidant Capacity Table 5 1 shows the antioxidant capacity and total soluble phenolics content of the control, DP -CO2 and thermal treated guava purees T he antioxidant capacity of the DP CO2 and pasteurized sample showed significant differences ( = 0.05) during the first 2 weeks of storage when compared to control. After week 4, there were no significant differences for antioxidant capacity of the control, DP -CO2 and thermal treated guava purees. These differences in antioxidant capacity are not well understood. There were no significant differences during the first 4 week of storage but during week 6, a significant increase in ORAC value was observed for the DP -CO2 and remained constant until week 10. After week 10, a significant decrease was observed and the ORAC value reached the initial value. Untreated (control) guava puree showed no significant differences in ORAC values during the first 5 weeks of storage. A maximum ORAC value was obtained during week 8 of storage, and after week 10, a significant de crease was observed. The changes in antioxidant capacity for the three samples may be the result of microbial growth. During microbial growth, compounds with antioxidant capacity may be released from the pulp of the guava increasing the antioxidant capacit y of the puree. The observed decrease may be related to the interaction of these compounds (such as polymerization reactions) affecting radical -scavenging properties. Thermally treated guava puree showed a significantly lower content of total soluble phenolics when compared to control and DP -CO2 treated guava purees. DP -CO2 treatment caused a significant increase in total phenolic content compared to the control

PAGE 130

130 and thermal treated guava purees during the first 2 weeks of storage This increase following D P-CO2 processing may be explained by several factors, including the conditions of the assay itself. Due to the broad range of compounds detected by the assay in the guava puree, slight changes in these compounds following DP -CO2 can easily influence the ability to reduce metal ions in solution, thus affecting the assay. Between weeks 2 and 10, there was no significant difference in total soluble phenolics when compared to control. After week 10 of storage at 4 C, DP -CO2 showed a higher value when compared to control and thermal treated guava purees. Similar results were observed by Ferrentino and others (2009) when red grapefruit juice was treated with DP -CO2. Furthermore, Del Pozo Insfran and others (2006) found similar results after p rocessing and storage of muscadine grape juice with a continuous DP -C O2 system. Control puree showed no significant differences in the total soluble phenolic content during the first 6 weeks of storage after which a significant increase was observed. After week 10 of storage, there was a significant decrease in total soluble solids. Even though the thermal treatment initially decreased (by 9%) the total phenolic content compared to the control a significant increase (4%) for the initial content was observed during the first 2 weeks of storage. This increas e m ay be explained by several factors, including the fact that certain compounds such as ascorbic acid, certain soluble proteins, melonoidins, and reducing sugars give a measurable interference in the assay. Addit ionally the guava puree contained intact plant cells whereby the conditions of heating helped to release cell wall or vacuole bound compounds with metal reducing capabilities. Any of these compounds present or thei r effect on the assay during storage coul d account for the increase. P olyphenolic degradation for control, DP -CO2 and

PAGE 131

131 thermal treatment during storage may be related to the loss of ascorbic acid. Additionally, the decrease in total soluble solids observed in thermal processed guava puree after 6 weeks of storage may be due to formation of by products from carbohydrate and organic acids degradation D uring thermal processing and storage, furfurals and other carbonyl compounds can form condensation products with t h e s e types of compounds (Es -Safi an d others 2002). The results suggest that DP -C O2 treatment does not affect the antioxidant capacity and total soluble phenolics of guava puree, supporting the idea that this process may have industrial application in producing fresh like juices with good antioxidant capacity T itratable A cidity (TA) TA of samples was expressed as percent citric acid (w w ) equivalents. T here was no significant ( = 0.05) difference between DP -CO2, thermal treated and control samples for TA values regardless of storage time (Figure 5 1 ) during the first six week of storage. During week 8 of storage, the control and thermal treated samples showed a significant increase in TA, which remained constant until the end of the study. The TA values were higher for the DP -CO2 treated s amples compared to the control and thermal treated samples The increase in TA for the three samples may be related to microbial growth. During microbial growth, microorganisms metabolize pectin and other carbohydrates releasing acids, which increase TA. The same trend was observed for TA by Kincal and others (2006) when orange juice was treated with DP -C O2. Similar results were also obtained by Damar and others (2009) and Ferrentino and others (2009).

PAGE 132

132 Ascorbic Acid (Vitamin C) Content There were slight differences in the ascorbic acid content of DP -CO2 treated guava puree regardless of storage time. Similar results were observed by Del Pozo Isfran and others (2006) after processing muscadine grape juice with DP -CO2. Ferrentino and others (2009) found s light differences for the ascorbic acid content in red grapefruit juice regardless of storage time The ascorbic acid content for the three samples (Figure 5-2) decreased over time regardless of the treatment. The ascorbic acid content for the DP -CO2 treat ed sample was significantly higher than thermal treated guava puree during the first 12 weeks of storage after which there were no significant differences between control, DP -CO2 and thermal treated guava purees. Thermal treatment decreased the ascorbic ac id content of the puree, 21% of ascorbic acid was loss after pasteurization of guava puree. Factors that influence the oxidation of ascorbic acid include pH, oxygen, water activity, and the presence of certain metal ions such as iron and copper and it is generally acceler ated with light and elevated temperatures (Fennema 2000). This indicated that the conditions of processing create an e nvironment suitable for ascorbic acid oxidation or degradation. DP -CO2 treatment delayed the oxidation or degradation of ascorbic acid probably by oxygen exclusion, and the ascorbic acid content was similar if not slightly better than the control puree. Color Color for control, DP -CO2 and thermal treated guava purees change drastically during storage at 4oC. As observed in T able 5-2, an increase in L*, a* and b* value were observed initially for the thermal treated guava puree, probably due to the loss of green color caused by degradation of pigments (such as chlorophyll). The same results were observed in pasteurized guava puree from Taiwan (Yen and Lin 1996) DP -CO2 and

PAGE 133

133 thermal t reat ments increased L* values, increasing lightness of the guava puree. L* values for thermal treated guava puree was not significantly affected while control and DP -CO2 samples showed a significant decrease during the 14 weeks of storage. There was a strong correlation between ascorbic acid content and lightness for control (r2= 0.86), DP -CO2 (r2=0.81) and pasteurized samples (r2=0.86). These results indicate the darkening in color was due to brownin g reactions. Little change in a* values for the three samples was noted, so the variability may be due to fluctuations between bottles. In general, thermal treatment significantly increased a* values during the first 6 weeks of storage while DP -CO2 showed a decrease during the first 2 weeks of storage, when compared to the control. No significant differences were found in the a* values between 4 and 6 weeks of storage for control and DP -CO2 treated samples. After 6 weeks of storage, a significant decrease in a* values was observed for the control. Furthermore, DP -CO2 and thermal treated samples showed no significant differences in a* values but a progressive reduction in the samples redness was observed. DP -CO2 and thermal treatments caused a significant increase in b* values or increase in yellowness when compared to control. Thermal treated puree b* values were not affected during the 14 weeks of storage. Control and DP -CO2 samples showed a progressive increase in b* value during storage. Changes in a* and b* values can be attributed to changes in carotenoids composition and there was a change in color from less red and more yellow. Ferrentino and others (2009) found that DP -CO2 caused an increase in lightness and redness while yellowness of red grapefruit juice was lowered. Kincal and others (2006) showed that both the L and the b values for orange juice increased after DP -

PAGE 134

134 C O2 treatment, while the a value generally decreased. T hermal p asteurization changed all 3 color parameters causing an increase in lightness and yellowness and a slight decrease in redness. Similar results were reported by Del Pozo-Isfran and others (2006). Organic Acid Content Oxalic, citric and malic acids were individually identified and quantified using HPLC (Table 5 -3). Figure 53 shows a typical chromatogram obtained from the organic acid analysis. There were significant differences in oxalic, citric and malic acid content during storage regardless of the treatment. Fluctuations in citric acid content were observed but these may be attributed to the bottle to bottle variation. There were no significant differences in malic acid content between control, DP -CO2 and thermal treated guava purees during storage. Th ere were no significant differences in oxalic acid content between control and DP -CO2 treated guava puree s during the 14 weeks of storage at 4 C. Phenolic Compounds Various solvent extraction and fractionation procedures on guava puree were attempted for polyphenolic detection by HPLC analysis. E nzyme treatment of guava puree was required to provide the most reproducible HPLC chromatograms with maximal peak separation (Figure 55 and 5 -6). In this study, HPLC analysis of polyphenolics was used to iden t ity overall data trends as affected by processing treatment and storage conditions Among the polyphenolic compounds present, 6 were selected based on identification and adequa cy for treatment differentiation (Table 5 -4 ). Peaks were identified ba sed on their spectroscopic properties and comparison to authentic standards (Table 5 -4 ). Few studies exist that identify and quantify the

PAGE 135

135 polyphenolics present in guava, but those that have include ellagic acid and condensed tannins (Misra and Seshadri 1968 ; Fender 2005 ; Nunez -Rueda, 2005), and gallic acid, catechin, epicatechin, and chlorogenic acid (Kondo and others 2005 ; Fender, 2005 ; Nunez Rueda, 2005) Ellagic acid was isolated from the ethyl acetate and methanol fractions (Figures 5 4 and 55). Gallic acid, hydrobenzoic acid, cinnamic acid and an unknown were isolated from the ethyl acetate fraction while an ellagic acid derivate was identified from the methanol fraction. Gallic acid (Peak 1) ellagic acid (Peak 3) and cinnamic acid (Peak 6) were clear ly identified by comparison to standards. Ellagic acid derivative (Peak 5) w as tentative ly identified by comparison to ellagic acid spectral properties Hydrobenzoic acid (Peak 4) was tentatively identified, as it shared similar spectral characteristics wi th those types of compounds. Only one compound (unknown) was characterized based on retention time and spectroscopic properties but was dissimilar to any known polyphenolic compounds. Further work will be needed to isolate and identify th is individual polyphenolic compound in guava puree. Gallic acid (GA) (Peak 1) has been reported as an effective antioxidant due to its structure and positioning of hydroxyl groups. DP -CO2 and thermal treatment significantly increased the initial concentration of GA (Table 5 5); however GA content decreased during the first 2 week s of storage and presented no significant difference fr o m the control puree. Between the second and sixth week of storage, a significant increase was observed for the control DP -CO2 and thermal l y treated samples, after which a significant decrease through out the re maining storage was observed.

PAGE 136

136 Unknown (Peak 2) Quantification of the unknown compound was performed using ellagic acid. Thermal treatment significantly increased the con centration of the compound when compared to DP -CO2 and control DP -CO2 significantly increased the concentration during the first two weeks of storage after which it remained constant until week 6. Thermal treated samples showed no significant differences due to storage time. Ellagic acid (Peak 3) DP -CO2 treatment decreased the initial concentration of ellagic acid (EA) (Peak 2) when compared to the control The EA concentration (Table 5 -5) decreased during the first 2 weeks of storage after which no sign ificant differences were observed for the DP -CO2 treated samples. Hydrobenzoic acid (Peak 4) Hydrobenzoic acid (HBA) was quantified and reported as gallic acid equivalents. DP -CO2 significantly increased initial concentration of HBA. During the first tw o weeks of storage, DP -CO2 treated samples showed a significant increase in HBA and then progressively decreased. Ellagi c acid derivat e (Peak 5 ) Ellagic acid ( EA ) derivat ive eluted in the HPLC column immediately after EA. Due to its closeness to EA in spectral properties, this compound was tentatively classified as EA derivative (Table 5 -6 ). There was a significant effect due to DP -CO2 treatment as compared to the control and thermal treated samples. During storage, control and therm a l treated guava puree samples showed a significant increase during the first two week s of storage after which no significant differences were observed Cinnamic acid (Peak 6 ) DP -CO2 treatment showed a significant ly higher concentration of cinnamic acid (CA), when compared to control During the first two week s of storage, DP -CO2 treated guava puree showed a significant increase in CA

PAGE 137

137 concentration, after which it decreased. A progressive and significant increase in CA concentration was observed for contr ol samples over the 14 weeks of storage. In general, DP -CO2 and thermal treatment caused an initial increase in the total content of the identified phenolic compounds of 33 and 10% respectively. Even when there were weekly variations in the individual co ntent of these compounds, the total content decreased for DP -CO2 and thermal treatment during the 14 weeks of storage. The total loss for DP -CO2 and thermal process was 19% and 30%, respectively from the initial concentration. Ellagic acid, its derivative and the unknown (quantified as ellagic acid) were the major contributors to the total phenolic content quantified by HPLC, followed by hydrobenzoic acid. Conclusions From the storage study, it is evident that the DP -C O2 treatment can extend shelf life and maintain the physical and quality attributes of fresh guava puree. DP -CO2 served to protect polyphenolic and antioxidant levels throughout processing and storage without compr om ising physicochemical and phytochemical properties of the guava puree. DP -CO2 delayed the degradation of vitamin C content during storage, allowing the vitamin C content to be higher than the fresh puree. DP -CO2 treated guava puree retained organic acids contents similar to fresh guava puree. Even when DP -CO2 caused an initial inc rease in phenolic compounds of 33%, the reduction of 19% for phenolic compounds during storage conditions was still lower than the thermal treated sample (30%).

PAGE 138

138 Table 5 1 Antioxidant capacity and total soluble phenolics of fresh, DP -C O2 and thermal tre ated guava puree during storage ORAC ( mol TE/mL ) Total Soluble Phenolics (mg of GAE/ L of sample) Week Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized 0 12.2 + 0.4 12.3 + 0. 5 11.7 + 0.3 182.2 + 0.9 196.5 + 0. 8 165.6 + 0. 6 2 12.4 + 0.4 11.1 + 0.4 11. 7 + 0. 4 182.8 + 0. 8 188. 5 + 0.8 172.1 + 0.7 4 12.1 + 0.4 11.2 + 0.6 11.4 + 0. 6 182.2 + 1.4 184.8 + 0.7 176. 6 + 0.1 6 15.4 + 0.7 15.8 + 0. 6 16.1 + 0.0 184.1 + 0.7 181.2 + 0.9 178.2 + 0.2 8 16.1 + 0.0 17.0 + 0.0 16.2 + 0.1 186.4 + 0.2 184.9 + 0. 6 177.0 + 0. 6 10 15.0 + 0.3 13.0 + 0. 8 12.8 + 0. 9 185.7 + 0. 7 184.3 + 1. 1 175.0 + 0. 4 12 12.8 + 0.2 12.9 + 0. 8 13.3 + 0.3 179.6 + 1.0 182.9 + 0. 1 177.0 + 0.3 14 14.3 + 0. 5 13.8 + 0. 9 14.5 + 0.2 174.5 + 1.0 183.0 + 0.5 175.2 + 0.4 Mean + standard deviation for n=3

PAGE 139

139 Figure 51. Titratable acidity of control, DP -C O2 and thermal treated guava purees during 14 weeks of refrigerated storage. Error bars represent standard deviation for n=3. 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0 5 10% Titratable Acidity (mg citric acid/ 100g puree)Storage time (weeks) Control DP CO2 Pasteurized

PAGE 140

140 Figure 5 2. Ascorbic acid (vitamin C) content of control, DP -C O2 and thermal treated guava purees during 14 weeks of refrigerated storage. Error bars represent standard deviation for n=3. 0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14Ascorbic acid (mg / g of sample)Storage time (weeks) Control

PAGE 141

141 Table 5 2. L*, a* and b* values for control, DP C O2 and thermal treated guava purees during 14 weeks of refrigerated storage Storage L* value a* value b* value Time (wks) Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized 0 45.75 + 0.03 47.69 + 0.30 48.57 + 0.06 12.74 + 0.07 12.72 + 0.48 13.26 + 0.23 10.34 + 0.04 12.06 + 0.19 13.65 + 0.20 2 45.39 + 0.08 47.00 + 0.04 48.68 + 0.13 12.81 + 0.15 12.46 + 0.20 13.53 + 0.12 10.79 + 0.08 12.41 + 0.29 13.99 + 0.03 4 45.25 + 0.02 46.18 + 0.18 48.23 + 0.02 13.15 + 0.06 13.17 + 0.09 13.42 + 0.09 11.80 + 0.06 12.80 + 0.05 14.51 + 0.26 6 45.19 + 0.07 46.16 + 0.13 48.42 + 0.15 13.22 + 0.08 13.39 + 0.11 13.55 + 0.03 12.03 + 0.06 13.03 + 0.08 14.76 + 0.08 8 44.83 + 0.02 46.00 + 0.14 47.84 + 0.04 12.72 + 0.14 13.31 + 0.11 13.13 + 0.04 11.97 + 0.12 13.20 + 0.09 14.59 + 0.05 10 43.86 + 0.13 45.58 + 0.19 47.99 + 0.04 12.09 + 0.19 12.65 + 0.13 12.79 + 0.16 11.87 + 0.07 12.94 + 0.13 14.42 + 0.12 12 43.98 + 0.21 45.62 + 0.30 48.05 + 0.17 12.11 + 0.41 12.75 + 0.09 12.95 + 0.06 11.86 + 0.14 13.02 + 0.26 14.61 + 0.11 14 43.92 + 0.20 45.60 + 0.01 47.88 + 0.31 12.19 + 0.05 12.67 + 0.25 12.81 + 0.38 11.98 + 0.02 13.16 + 0.12 14.37 + 0.35 Mean + standard deviation for n=3

PAGE 142

142 Table 5 3. Oxalic acid ( O A), ma lic acid (MA) and citric acid (CA) content of control, DP -CO2 and thermal treated guava purees during 14 weeks of refrigerated storage at 4oC Storage time Oxalic Acid (mg acid/100 g sample) Malic Acid (mg acid/100 g sample) Citric Acid (mg acid/100 g sample) (Week) Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized 0 0 .03 + 0 .000 0.038 + .000 0.04 + 0.000 0.09 + 0 .001 0.09 + 0.001 0.09 + 0.000 0.52 + 0.001 0.52 + 0.001 0.52 + 0.001 2 0.02 + 0 .002 0.024 + .001 0.02 + 0.000 0.08 + 0 .003 0.0 8 + 0.002 0.08 + 0.002 0.32 + 0.031 0.36 + 0.003 0.37 + 0.011 4 0.02 + 0 .002 0.018 + .001 0.01 + 0.001 0.07 + 0 .007 0.0 7 + 0.000 0.0 6 + 0.005 0.29 + 0.004 0.26 + 0.031 0.25 + 0.027 6 0.02 + 0 .000 0.021 + .000 0.02 + 0.000 0.07 + 0 .005 0.08 + 0.008 0.08 + 0.000 0.39 + 0.000 0.38 + 0.004 0.39 + 0.004 8 0.02 + 0 .001 0.019 + .001 0.02 + 0.002 0.10 + 0.003 0.1 3 + 0.002 0.12 + 0.002 0.32 + 0.021 0.33 + 0.014 0.37 + 0.007 10 0.03 + 0 .003 0.030 + .001 0.03 + 0.001 0.12 + 0.001 0.12 + 0.001 0.12 + 0.001 0.40 + 0.008 0.4 3 + 0.012 0.41 + 0.005 12 0.04 + 0 .000 0.038 + .003 0.04 + 0.001 0.18 + 0.001 0.13 + 0.001 0.12 + 0.004 0. .40 + 0.005 0.43 + 0.004 0.42 + 0.007 14 0.05 + 0 .001 0.049 + .000 0.05 + 0.000 0.08 + 0.001 0.08 + 0.001 0.08 + 0.002 0.3 9 + 0.004 0.3 8 + 0.004 0.39 + 0.002 Mean + standard deviation for n=3

PAGE 143

143 Figure 5 4 HPLC chromatogram of organic acids found in guava puree1) oxalic acid, 2) malic acid and 3) citric acid. Identification ( 210 nm) was done by comparison to authentic standards and spectral properties Figure 5 5 HPLC chromatogram of polyphenolic compounds found in ethyl acetate fraction in guava puree1) gallic acid, 2) unknown 3) ellagic acid, 4) hydrobenzoic acid and 6) cinnamic acid. Identification (260 and 280 nm) was done by comparison to authentic standards and spectral properties 0. 5. 10. 15. 20. 25. 30. 35. 40. 45. 52. 70 mA mi 1 2 3 4 6 1 2 3

PAGE 144

144 Figure 56. HPLC chromatogram of polyphenolic compounds found in the methanol fraction of guava puree3) ellagic acid and 5) ellagic acid derivative. Identification (260 nm) was done by comparison to authentic standards and spectral properties 30. 32. 34. 36. 38. 40. 42. 44. 46. 48. 50. 52. 0 12 25 35 mA mi 3 5

PAGE 145

145 Table 5 4. I dentification of guava polyphenolics at 260 and 280 nm by HPLC based on retention time, spectral properti es, and comparison to authentic standards Peak No. Retention time (min) Fraction Spectral Properties Compound 1 12.95 Ethyl Acetate 216, 272 Ga l lic Acid 2 30.15 Ethyl Acetate 264.2 Unknown 3 37.56 Ethyl Acetate and Methanol 25 5, 367 Ellagic acid 4 39.00 Ethyl Acetate 19 9 25 4 Hydroxybenzoic acid 1 5 41.75 Methanol 255, 36 7 Ellagic acid derivate 1 6 42.17 Ethyl Acetate 20 4 21 7 278 Cinnamic acid 1 = Tentative identification

PAGE 146

146 Table 5 5 Gallic acid (GA), unknown and ellagic acid (EA) for control DP -CO2 and thermal treat ed g uava purees during 14 weeks of refrigerated storage at 4oC Storage time (weeks) Gallic Acid (GA) Unknown Ellagic Acid (EA) (mg/L) (mg/L EAE 1 ) (mg/L) Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized 0 0.63 + 0.03 7.42 + 0.05 6.55 + 0.04 1.61 + 0.01 1.96 + 0.06 2.19 + 0.03 5.93 + 0.02 4.77 + 0.9 5.89 + 0.06 2 1.89 + 0.02 1.88 + 0.06 1.23 + 0.01 2.20 + 0.03 2.24 + 0.06 2.22 + 0.03 5.00 + 0.08 4.42 + 0.36 5.00 + 0.27 6 14.13 + 0.05 10.34 + 0.04 9.04 + 0.05 2.54 + 0.03 2.25 + 0.02 2.17 + 0.09 5.42 + 0.27 4.30 + 0.05 4.92 + 0.17 10 2.29 + 0.05 2.90 + 0.10 1.94 + 0.08 2.30 + 0.06 2.41 + 0.01 2.08 + 0.14 4.46 + 0.28 4.18 + 0.37 4.42 + 0.30 14 2.27 + 0.07 1.58 + 0.02 1.83 + 0.04 2.39 + 0.00 2.02 + 0.05 2.03 + 0.08 4.54 + 0.27 3.94 + 0.35 3.96 + 0.07 Mean + standard deviation for n=3; 1= Ellagic acid equivalents Table 5 6. Hydrobenzoic acid (GA), cinnamic acid (CA) and ellagic acid derivative (EAD ) for control DP -CO2 and thermal treat ed g uava purees during 14 weeks of refrigerated storage at 4oC Storage time (weeks) Hydrobenzoic Acid (HBA) Cinnamic Acid (CA) Ellagic Acid Derivate (EAD) (mg/L GA E 1 ) (mg/L EAE 2 ) (mg/L EAE 2 ) Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized Control DP CO 2 Pasteurized 0 5.39 + 0.02 7.21 + 0.61 9.19 + 0.49 1.46 + 0.02 1.50 + 0.01 1.47 + 0.01 1.01 + 0.00 1.21 + 0.04 1.64 + 0.01 2 9.22 + 0.26 9.66 + 0.57 9.84 + 0.17 1.49 + 0.00 1.49 + 0.00 1.53 + 0.01 1.51 + 0.00 1.68 + 0.01 1.76 + 0.02 6 10.23 + 0.26 8.67 + 0.21 8.75 + 0.22 1.50 + 0.00 1.46 + 0.01 1.55 + 0.02 1.87 + 0.03 1.73 + 0.01 1.77 + 0.02 10 9.50 + 0.79 10.45 + 0.06 9.06 + 0.25 1.50 + 0.01 1.49 + 0.01 1.53 + 0.00 2.04 + 0.01 1.80 + 0.01 1.80 + 0.03 14 13.98 + 0.23 9.08 + 0.73 9.44 + 1.07 1.49 + 0.01 1.55 + 0.01 1.53 + 0.00 2.64 + 0.01 2.10 + 0.04 2.00 + 0.03 Mean + standard deviation for n=3; 1= Gallic acid equivalents, 2=Ellagic acid equivalents

PAGE 147

147 CHAPTER 6 CHANGES IN AROMA COM POUNDS AND SENSORY PERCEPTION IN GUAVA PUREE AFTER THERMAL AND NON-THERMAL PROCESSING Abstract The volatiles present in freshly thawed (FT), dense phase carbon dioxide ( DP CO2) treated and pasteurized guava puree were isolated by Solid Phase Microextraction (SPME). I solates analyzed by Gas Chromatography (GC -MS) Gas Chromatography olafactometry ( GC -O ) and Gas chromatography coupled to a pulse flame photometric detector ( GC -PFPD ) contained 76 compounds. Analysis with GC MS showed 58 compounds which were classified in 6 groups: aldehydes (6), acids (2), alcohols (15), ketones (6), esters (21) and terpenes (8). Eleven compounds were identified for the first time in guava puree: butanal, isoamyl acetate, 4 -mercapto4 methyl penta-2 one, phenyl acetaldehyde, nonanal, homofuraneol, methyl nonanoate, 1 -p menthene-8 -thiol, 2 octane, ethyl -3 hydroxyhexanoa te and ethyl nonanoate. Hexanal was the most abundant compound in the three guava puree samples. Analysis by GC -O showed 26 compounds responsible for the guava puree aroma, three of which were tentatively identif ied: benzyl alcohol, phenyl acetaldehyde and homofuraneol (Z) 3 -hexenyl hexanoate was the major contributor to the aroma of the fresh and DP -CO2 treated guava puree. Eleven sulfur compounds were identified by GC -PFPD in the guava puree. Four of these compounds were identif ied for first time in guav a puree, and tw o, hydrogen sulfide and hydrogen disulfide, were previously reported in guava leaves. Introduction Guava ( Psidium guajava L.), is a tropic al fruit characterized by its appealing flavor and aroma. It has been catalogued as one of the most nutritious fruits due to its high content of phytochemicals, especially ascorbic acid ( United States Department of

PAGE 148

148 Agriculture [USDA], 2005). Because of the high perishability and limited availability of fresh guava, m ost fruit destined for US markets is processed into juice, puree, jams, jellies, and syrup. Guava puree is normally processed by heat pasteuri zation to extend the shelf life up to one year and inactivate pectinesterase, but t he fresh taste is modified by thermally accelerated reactions. The first step in the characterization of odor and flavor (volatile) compounds in a complex mixture or food system is to separate them from nonvolatile matrix interferences This separation is accomplished through a variety of techniques, such as solvent extraction, head -space concentration, and distillation. E xtraction procedure s may distort or alter the chemical composition Since each method will enhance the concentration of c ertain compounds and minimize others there is no perfect extraction system. Each technique yields a concentrated essence containing the odor active chemicals. Considering that it is simple, solventless and rapid, Solid Phase Microextraction (SPME) was use d as a sample ext ra c tion technique in this study. The volatile composition of guava constituents has been previously studied and extraction techniques and detection methods vary (Stevens and others 1970; Wilson and Shaw 1978; Ma cleod and Gonzalez de Troco nis 1982; Idstein and Schreier 1985; Chyau and others 1992). Several studies have been conducted to determine the effect of processing on the volati le composition of guava puree. Yen and others ( 1992) studied the changes in flavor components of guava puree resulting from past eurization and frozen storage. Chan and Cavaletto ( 1982) studied the changes in chemical and sensory quality during processing and storage of aseptically packaged guava puree. Yen and Lin ( 1996) studied the changes in volatile flavor c omponents of guava juice

PAGE 149

149 with high pressure and heat processing. Dense Phase Carbon Dioxide (DP -CO2) treatment has not been tested for its potential on retaining volatile compounds compared to traditional thermal treatments for guava puree. The objectives of this study were (1) to determine the volatile composition of the puree as affected by DP -CO2 and thermal processing and (2) Compare the sensory perception of DP -CO2 and thermal processing with a freshly thawed guava puree. Materials and Methods Guava Pu ree Frozen unpasteurized red guava puree was obtained from the Goya Company (Dominican Republic). The puree was held at 20 C and thawed at 4 C for one week prior to processing. Part of the insoluble solids and stone cells were removed by straining the thawed puree through a 200 m nylon filter (Cole Palmer, Vernon Hills, IL, U.S.A). DP-CO2 treatment was performed using freshly thawed (FT) puree, which was treated at 34.5 MPa, 8% CO2 and 6. 9 min residence time Volatile compounds were compared to that o f thermally treated (90 C for 60 sec) and freshly thawed guava puree (control). Pasteurization was performed using a Microthermic lab scale pasteurizer (see Chapter 4: Processing equipment from the Material and Methods section). Sensory Evaluation Flavo r and overall likeability of FT and processed (DP-CO2 and pasteurized) guava puree were compared using a difference from control test at the beginning of the storage study A randomized complete block design was used, and differences from control values were recorded on a line scale from 0 to 10. No differences (0) and extremely different (10) were used as the extreme anchors of the flavor line. Panelists compared the flavor of the reference (FT guava puree = control ) with that of a hi dden

PAGE 150

150 reference (control ), the thermal (HTST), and DP -C O2 treated guava purees A 9 point hedonic scale was also conducted in order to compare the overall likeability of FT (hidden reference) and processed guava purees An example of the questionary used for the taste panel is shown in Appendix D. Before sensory analysis all DP -CO2 treated purees were degassed in order to have equal carbonation levels All samples were chilled and kept in ice at a temperature of ~4 C before serving. They were then served on a tray in numb ered plastic cups containing ~30 mL of sample. A cup of deionized water and non salted crackers were also provided to the panelists to cleanse their palate between evaluations. S ensory tests were performed at the University of Florida taste panel facility using 75 untrained panelists in each test. Statistical Analysis Sensory data was recorded and analyzed using Compusense five (Compusense, Guelph, Ontario, Canada). Analysis of variance (ANOVA) and mean comparisons using t -test and Tukeys test were conducted at the 5% significance level. The statistical analysis was conducted using the same program used to record the data. Extraction of V olatile C ompounds using Headspace SPME Headspace volatiles were extracted and concentrated using SPME. Ten m illiliter aliquots of guava puree (fresh ly thawed DP -CO2 and pasteurized) w ere poured into 40 mL screw cap amber glass vials and each sealed with caps containing Teflon-coated septa. Volatiles were subsequently extracted using a pre-conditioned 1 cm 50/30 m Divinylbenzene/Carboxen/Polydimethylsiloxane (DVB/ C AR/ PDMS ) fiber (Supel co, Bellafonte, PA U.S.A.) for 30 min at 40C. Guava purees were allowed to equilibrate for 20 min prior to SPME. The full length of the coated fiber was exposed to the headspace

PAGE 151

151 of the samples and after 30 min the fiber was removed from the headspace and immediately inserted into a GC -splitless injector, where aroma compounds were allowed to be desorbed for 5 min. Before each volatile compound extraction, the fiber was cleaned for 5 min in the injection port (200 C) of the GC -O, GC S or GC MS instrument s. Adsorbed volatiles were desorbed in the injector port of a GC. Separation and analysis was conducted as follows. Gas Chromatography Olfactometry Analysis (GC O) GC -O analysis was carried out using a HP 5890A GC (Palo Alto, CA) with a flame ionization detector (FID) and a sniffing port. A DB wax column (30 m x 0.32 mm i.d. x 0.25 m film thickness) (J&W Scientific, Folsom, CA U.S.A.) and a ZB -5ms column (30 m x 0.32 mm i.d. x 0.5 m film thickness) (Zebron ZB 5, Phenomenex, Torrance, CA, U.S.A.) were used during the analysis. Initial oven temperature was 40 C ( no hold ) and temperature was increased at 7 C/min until reaching a final temperature of 240 C (for DB Wax) or 265 C (for DB 5) and holding at this temperature for 5 min. Guava puree volatiles from the SPME fiber were desorbed in the GC injection port. A SPME injector liner (SPME injection sleeve, 0.75 mm i.d., Supleco; Bellefonte, PA., U.S.A.) was used. The GC column effluent was split between the FID and the olfactometer. Helium was used as the carrier gas at a flow rate of 1.55 mL/min. The injector temperature was 200 C, and the detector temperature was 250 C. Two olfactory assessors were employed. Samples were sniffed in duplicate by each assessor in each column. Aroma descriptions and approximate times were recorded for every sample. Assessors indicated the intensity of each aroma peak using a linear potentiometer with a 01 V signal. Aromagrams and FID chromatograms were recorded and integrated using Chrom Perfect 5.5.5 ( Justice Labs, Melbourne, FL, U.S.A. ) data acquisition

PAGE 152

152 software A peak was considered aroma active only if at least half the sniffers found it at the same time with a similar description. Linear retent ion index values were determined for both columns using a series of alkanes (C5 -C25) run under identical conditions. Gas Chromatography Mass Spectrometry Analysis (GC -MS) Mass Spectrometry ( GC -MS ) was used to identify the odor active volatile s detected in the GC -O experiment s GC MS data were collected using a Perkin Elmer Clarus 500 quadrupole mass spectrometer equipped with Turbo Mass software (Perkin Elmer, Shelton, CT) Helium was used as the carrier gas in the constant flow mode of 2 mL/min. Guava puree volatiles from the SPME fiber were desorbed in the GC injection port (splitless mode) at 200 C. The fiber was removed after 5 min exposure in the injection port. G as chromatography separation was performed using a DB wax column (30 m x 0.32 mm i.d. x 0.25 m film thickness) fr om J&W Scientific (Folsom, CA, U.S.A.) and ZB -5ms column (30 m x 0.32 mm i.d. x 0.5 m film thickness) from Phenomenex (Zebron ZB -5, Phenomenex, Torrance, CA, U.S.A.) Initial oven temperature was 40oC (no hold ), and temperature was increased at 7 C/min until reaching a final temperature of 240 C (for DB -Wax) or 265 C (for DB 5) and h olding at this temperature for 5 min. The mass spectrometer detector had a delay of 0.5 min, and scans were made from m/z 25 to 300. A m as s spectrum scan was performed e very 0.2 seconds. The electron ionization was carried out in the positive mode at 70 eV Mass spectral matches were made by comparison with NIST 05 (Scientific Instrument Services, Ringoes, NJ, U.S.A.) and Wiley Registry of Mass Spectral data 6th edition (John Wiley and Sons Inc) mass spectra l libraries. Only those compounds with spectral fit values equal to or greater than 850 were considered positive identifications. L inear retention index values were

PAGE 153

153 d etermined for both columns using a series of alkanes (C5 -C25) run under identical conditions. Sulfur Compounds I dentification using Gas Chromatography Pulse Flame Photometric Detector (GC PFPD) Volatile sulfur compounds from the fiber were separated using a HP 5890 Series II G C from Agilent (Santa Clara, CA U.S.A.) equipped with a 5380 PFPD detector from OI Analytical (College Station, TX U.S.A. ). The compounds from the SPME fib e r were desorbed for 5 min in the GC injection port (splitless mode) at 200 C. Separation of compo unds was achieved on DB wax column (30 m x 0.32 mm i.d. x 0.25 m film thickness) (J&W Scientific, Folsom, CA U.S.A.) and a ZB 5ms column (30 m x 0.32 mm i.d. x 0.5 m film thickness) from Phenomenex (Zebron ZB 5, Phenomenex, Torrance, CA, U.S.A.). GC oven temperature was initially set at 40 C and then ramped at 7 C/min to 240 C (DB Wax) or to 265 C (ZB-5). The final temperature was h e ld for 5 min in both columns GC was operated in a constant flow mode (2 mL/min) with helium as the carrier gas. PFPD detector was set at 250 C and employed WinPulse32 Version 2.0 software. The PMT voltage was set at 525 V, and the sul f ur gate was opened between 5 and 24 ms. PFPD output was recorded in the square root mode. Chromatograms were recorded and integrated usin g Chrom Perfect 5.5.5 (Justice Labs, Melbourne, FL, U.S.A.) data acquisition software Identification Procedures Identifications were based on the combined matching of retention indices (LRI values) from DB 5 and DB -Wax columns, spectral matches from the N IST 05 (Scientific Instrument Services, Ringoes, NJ, U.S.A.) and Wiley Registry of Mass Spectral data 6th edition (John Wiley and Sons Inc) librar ies aroma descriptors, and linear retention

PAGE 154

154 index matches from databases ( Jennings and Shibamoto 1980 ; Flavor database 2009 ; Flavor net and human odor space 200 9 ). Results and Discussion Sensory Analysis One taste panel was conducted during storage. Results showed that 51% of panelists were females and 49% were males, 84 % of males and 83% of females were in the 18 -24 age range. Results showed (Table 6-1) that there were no significant differences between the reference and pasteurized guava purees However, significant differences were detected by panelists between the reference and DP -C O2 treated guava purees. The difference between the reference and the hidden reference was 3.05. Some differences are expected since this was more of a consumer testing, and no training was involved. The difference between the reference and the pasteurized sample was 3.56. This indicated that the panelists were not able to differentiate the reference and the pasteurized sample. Although, t he ranking for overall likeability for the three tested purees were not significantly different which indicated that regardless of treatment panelist preference s remained the same (Table 6 -1). The overall likeability for hidden reference was 4.29, for DP -CO2 was 4.13 and for pasteurized puree was 3.79. Due to the closeness of the overall likability ranking for the hidden refer ence and DP -CO2 guava purees, results indicat e that panelist preference s were toward reference and DP -CO2. Even though the pasteurized puree was not different from DP -CO2, the ranking value was lower. Previous studies on muscadine grape juice and coconut w ater showed bigger differences in flavor and overall likeability between the DP -C O2 and thermal treated samples. They also found that the DP -C O2 sample was very similar to

PAGE 155

155 the reference (Damar, 2006 ; Del PozoInsfran and others 2006) This difference may i ndicate that thermal processing affected more the organoleptic characteristics of the grape juice and coconut water than in guava puree (probably the nature of the sample has an effect on the treatment). DP -C O2 differences from the other two samples can be attributed to the fact that even when the DP -CO2 puree w as partially degasified before sensory analysis was conducted, there could still be residual CO2 remaining This would result in a carbonated puree mouth feel which may have also affected the acidity/sweetness balance causing the panelists to perceive the beverage as less sweet. This can be confirmed by a higher TA in the DP -CO2 beverage as described previously. Another reason for the differences may be attributed to the separation of DP -CO2 samples into two phases. Due to problems with the pasteurization unit, FT and DP -CO2 samples were frozen for three weeks to lock the processing date. After the pasteurization process was carried out, samples were also frozen. Three days before the taste panel, samples were thawed at 4 C. During this time, DP -CO2 samples separate into two phases. Flavor Analysis SPME technique was used for the isolation of the volatile compounds present in the guava purees. This t echnique employs a fused-silica fiber coated with an appropriate stationary phase, allowing the analyte in the sample to be directly extracted (Kataoka and others 2000) by the absorption to the fiber followed by desorption in the injector port of a gas chr omatograph. Subsequently, separation, detection, identification, and characterization of individual compounds are possible with sophisticated instrumentation such as GC MS, GC -S and GC -O.

PAGE 156

156 The SPME extraction conditions were as previous ly described (Paniandy and others 2000) Although SPME has a maximum sensitivity at the partition equilibrium, a proportional relationship is obtained between the amount of analyte adsorbed by the SPME fiber and its initial concentration in the sample matrix before reaching p artition equilibrium. Therefore, full equilibration is not necessary for quantitative analysis by SPME (Kataoka and others 2000) Guava Volatile Composition A total of 76 compounds were identified using GC MS, GC -O and/or GC -PFPD using different columns Table D 1 shows the named compound, its CAS number, identification method, the reported LRI and references where previously identified. Sixty five of these compounds were previously identified and eleven were identified for the first time. Guava puree vol atiles were divided into six organic compound categories. A total of 6 aldehydes, 2 acids, 15 alcohols, 6 ketones, 21 esters and 8 terpenes were identified by GC MS (Figure 6 -1). The relative difference in total volatiles (identified by GC -MS) in terms of peak area was normalized to total peak area of freshly thawed guava puree which was 395. As shown in Figure 62, DP -CO2 treatment and thermal pasteurization caused a 46 and 19% reduction in total peak area respectively, when compared to the total peak area of the freshly thawed guava. Although the pasteurization process caused a small reduction in peak area, an increase of 2% on the area of oxygenated terpenes was observed. The total peak area reduction caused by DP -CO2 treatment may be related to the so lubility of the aroma compounds in carbon dioxide. Most of the aroma compounds are lipophilic (Reineccius 2006). Carbon dioxide is a colorless, lipophilic and non toxic

PAGE 157

157 liquid that behaves as a supercritical fluid above its critical temperature (31.1C) and critical pressure ( 7.39 MPa) Under DP -CO2 processing conditions (temperature and pressures above CO2 standard conditions), C O2 in the supercritical state is dissolved and mixed with the guava puree. After processing, the pressure is released and the CO2 changes to its gaseous state. Furthermore, some of the volatiles will be carried along with the outlet CO2 gas. For example, ethanol and ethyl acetate (compounds of low boiling point) were not detected in DP -CO2 treated guava puree. This suggests that t hese compounds were lost with the outlet CO2 gas, and probably can be related to the difference in sensory perception of the DP -CO2 sample related to the FT, and pasteurized guava purees. Composition of the three samples was similar but there were major quantitative differences. Alcohols comprised the largest group of volatiles in freshly thawed and pasteurized guava purees, contributing 36% and 33% of the total respectively. Aldehydes compromised the second group of volatile constituents for freshly thaw ed and pasteurized guava (27 and 24% respectively), followed by esters, terpenes, ketones and acids. The main contributors to the total volatile peak areas of the DP -CO2 treated guava puree were aldehydes (21%), followed by alcohols, esters, terpenes, keto nes and acids (18%, 13%, 7%, 6% and 1% respectively). Therefore, DP -CO2 treatment influenced the volatile profile of the puree because alcohols were reduced by 50%, esters were reduced by 43% and aldehydes were reduced by 24%. GC -MS Identifications A tot al of 58 volatiles were identified using GC MS in guava puree samples (Table 61 and Figure 62). Aldehydes and alcohols were the main volatiles of guava puree used in this study. These results agree with those found previously (Idstein and

PAGE 158

158 Schreier 1985; Nishimura and others 1989; Stevens and others 1970). The presence of high amount of aldehydes and alcohols may involve enzymatic oxidation and reduction of corresponding fatty acids (Idstein and Schreier 1985). Seven compounds were identified for first t ime in guava puree. They are: butanal, nonanal, isoamyl alcohol, 2octanone, ethyl 3 -hydroxyhexanoate, methyl nonanoate and ethyl nonanoate. The compounds reported previously in guava fruit were identified using different extraction techniques. For example, Chan and Cavaletto (1982) studied the changes in chemical and sensory quality during processing and storage of aseptically packaged guava puree. The puree was acidified to pH 3.9 with citric acid and the soluble solids content was 13.5%. The heat treatment was conducted at 93oC for 26 seconds and the product was aseptically packed. Quality of all samples was compared to untreated frozen puree. Flavor was not as greatly affected as color and the flavor changes were the result of storage time and not proce ssing (Chan and Cavaletto 1982). Yen and others ( 1992) studied the changes in flavor components of guava puree resulting from pasteurization and frozen storage. The volatiles were extracted using Likens -Nickerson apparatus and a GC -FID and GC MS were used for the identification of the compounds. Initially, the volatile constituents of the pasteurized puree were similar to the unpasteurized puree. Terpene hydrocarbons were the main volatile components followed by aldehydes. Significant changes in volatile c onstituents and deterioration of flavor quality during the first two months of storage at 0oC were attributed to oxid ation and enzymatic reactions. Yen and Lin ( 1996) studied the changes in volatile flavor components of guava juice with high pressure and h eat processing.

PAGE 159

159 They used purge and trap as the extraction technique and found that high pressure processing maintained the original flavor while heat processing caused a decrease in the majority of volatile constituents. (Z) -3 hexenal, hexanal, ethyl -2 m ethyl butanoate, (Z) -3 -damascenone and t -caryophillene have been identified in guava leaves (Rouseff and others 2008). Three esters (ethyl acetate, ethyl hexanoate and 3hexenyl acetate), 2 aldehydes (hexanal and (Z) -3 hexenal), 2 alcohols (hexanol and (Z) -3 hexenol) and 1 terpene (t -caryophillene) have been reported as volatile constituents of guava fruit. Differences in volatile profiles among studies can be attributed to different guava cultivars used and extraction methods. Table 6 2 lists t he 58 volatiles identified by MS in this study. To compare the volatiles in the three guava samples (freshly thawed, DP -CO2 and pasteurized guava purees), p eak areas were normalized on the single largest peak found in FT guava puree. This peak was, hexanal and it was assigned a value of 100 and the remaining peaks in all three samples were normalized to it Fifty eight volatiles were found in fresh guava puree compared to 54 volatiles found in both DP -CO2 and pasteurized guava purees. Acetaldehyde, butanal and 2octanone were only founded in FT guava puree while ethyl alcohol was found in the FT and pasteurized guava purees. (Z) -3 hexenyl hexanoate was only found in the FT and DP -CO2 guava purees but not in the pasteurized puree. Hexanal was one of the hi ghest compounds found in all three samples with a relative peak area between 79 and 100%. This agrees with results found by (Idstein and Schreier, 1985; Nishimura and others 1989; Paniandy and others 2000 ; Mahattanatawee and others 2005). DP -CO2

PAGE 160

160 and pasteurization treatments reduced the hexanal peak area by 21 and 5% respectively. Among the 15 alcohols, (Z) 3 -hexenol and 1 hexanol contribute between 68 and 78% of the total peak area in all three samples. DP -CO2 and pasteurization reduced the peak area for (Z) 3 -hexenol by about 10%. (Z) -3 -hexenol has been previously reported as the main component of white guava (Paniandy and others 2000), guava from Hawaii (Stevens and others 1970), fresh pink guava puree (Jordan and others 2003) and guava from Br azil (Idstein and Schreier 1985). Ethyl acetate (28%), ethyl propanoate (20%) and ethyl hexanoate (22%) were the most abundant esters present in all three samples. However, their concentration was lower in DP -CO2 and pasteurized guava purees. Ma cleod and Gonzalez de Troconis ( 1982) reported ethyl acetate and ethyl hexanoate as the major esters of guava from Venezuela. Yen and Lin ( 1999) reported the largest amount of ethyl acetate in guava juice. Idstein and Schreier (1985) found that ethyl and acetate est ers were predominant in guava from Brazil and they suggest that their humulene was the predominant terpene found in guava purees used in this study. These results differ p artially from those reported by Wilson and Shaw (1978), Mahattanatawee and others (2005) and Jordan and others ( caryophyllene as a major terpene. Elevated levels of -bisabolene were observed in the pasteurized sample only. GC -O Aroma Profiles A total of 26 aroma compounds were found in guava puree and are listed in Table 6 3. Of these 26 compounds, 19 were confirmed with GC MS and 4 were confirmed with GC -PFPD. Of the 11 newly identified, 3 of these were aroma active

PAGE 161

161 com pounds and are listed in Table 63 in bold. Twenty -four 22 and 1 3 aroma active compounds were detected for FT, DP -CO2 and pasteurized guava purees respectively. Peak heights were normalized to the highest peak present in the FT guava puree. This peak was (Z) -3 hexenyl hexanoate and it was assigned a value of 100 and the remaining peaks in all three samples were normalized to it. Eleven compounds were detected in all three samples: octen3 ol, benzyl alcohol, furaneol, and homofuraneol ethyl hexanoate, phenyl acetaldehyde, ethyl octanoate and ethyl dihydrocinnamate,4mercapto4 -methyl -pentan-2 ol and 3mercapto1 -hexanol and (E,Z) 2,6nonadienal. Six of the eleven compounds found in the 3 samples were tentatively identified based on LRI and odor descriptor s. Phenyl acetaldehyde and 1p -menthene-8 -thiol were identified for the first time in guava puree but they have been previously identified in other food products (Demole and others 1982; Buettner and Schieberle 1999 ; Janes and others 2009). The most intense odorant in FT guava puree was (Z) 3 -hexenyl hexanoate, followed by homofuraneol, hexanal, ethyl hexanoate and furaneol. (Z) -3 -hexenyl hexanoate was reported as an important aroma contributor to fresh guava puree from Fl orida (Jordan and others 2003). T he most intense odorant in DP -CO2 guava puree was 3 mercapto1 -hexanol followed by ethyl 3 -phenylpropionate and (E,Z) 2,6 nonadienal. The last 2 compounds had the same int ense perception. (E,Z) 2,6 nona d i enal was the most intense peak for pasteurized guava puree (its intensity was 5% less than in DP -CO2 treated puree) followed by furaneol and octen-3 ol. Ethyl 3 phenylpropionate, phenyl acetaldehyde and furaneol were described to have guava like aroma. Phenyl ethylacetate was identified by Stevens and other s ( 1970) as a

PAGE 162

162 contributor to the overall pleasant fruity aroma of guava from Hawaii. Ethyl acetate (fruity, pleasant and sweet aroma), ethyl phenylacetate (described as fruity odorant) and (Z) -3 hexenol (fruity, green and grassy notes) were detected only in FT samples. Ethyl phenylacetate was detected at a concentration between 50 and 250 g/kg pulp by Idstein and Schreier (1985) in guava from Brazil and ethyl acetate and (Z) 3 hexenol were identified as important aroma contributors to fresh guava puree fr om Florida (J ordan and others 2003).Ethyl acetate was identified as a major ester contributor to the fla vor of guava from Venezuela by Ma cleod and Gonzalez de Troconis ( 1982). (E) 3 hexen ol (green, grassy note) was detected only in DP -CO2 guava puree. A k etone, carvone, (described as sweet, spearmint and peppermint aroma) was detected only in DP -CO2 and pasteurized guava purees. Pentene3 one (nutty), hexanal (fatty, grassy), methional (cooked potato), 1,8-cineole (eucaliptus, camphorous, cool), ethyl 3 ph enylpropionate (guava, fruity), pmenthene 8 -ionone (floral, fruity), (Z) -3 -decalactone (nutlike, coconut) were identified only in FT and DP -CO2 -ionone, which has a very low odor threshold and an intense violet aroma was identified as the main contributor of flor al flavor of Hawaiian guava by Stevens and others (1970). Ethyl hexanoate (fruity, sweet) was detected in all three samples and reported as an important flavor compound by MacLeod and Gonzalez de Troconis (1982) in guava from Venezuela. GC -PFPD Analysis A total of 11 sulfur compounds were found in guava p uree and are listed in Table 6 -4. The compounds were identified by matching LRI with databases (Rouseff 2006) and standards. Seven of these compounds have been previously reported in guava puree and two were reported for the first time. Hydrogen sulfide and dimethyl sulfide

PAGE 163

163 have not been previously reported in guava puree; however, they have been reported in guava leaves (Rouseff and others 2008). Methanethiol, dimethyl disulfide and methional have been reported in guava fruit and leaves. 1 p -Menthene 8 -t hiol or grapefruit mercaptan and 4mercapto4 methyl pentan2 one were identified for the first time in guava puree. 4mercapto 4 methyl pentan2 one has been reported as a key component of sauvignon grapes and is used in flavor manufacture to recreate the catty note of blackcurrant (Rowe 2000). 1 -p -Menthene -8 -thiol has a very low threshold (0.0001 ng/mL) and is considered an important sulfur compound of grapefruit juice (Demole and others 1982). Sulfur -containing compounds play an important role in natural flavor chemistry as they are not only responsible for the objectionable odors associated with rotting vegetable matter but also contribute, and often characterize, the desirable aroma of many plants and thermally processed foods (Reinneccius 2006). In the rmally processed food, these compounds are formed by Maillard reactions, while in plants its formation is part of in vivo biogenesis. Peak heights were normalized to the highest peak area in FT guava sample, which was 1 p menthene8 -thiol. It was assigned a value of 100 and the remaining peaks in all three samples were normalized to it. Eleven sulfur compounds were identified in DP CO2 treated guava samples while 10 were found in FT and pasteurized sample. In FT guava puree, the grapefruit smelling, menth ene-8 -thiol (100) was the major sulfur compound followed by methanethiol, dimethyl disulfide and hydrogen sulfide. In DP CO2 and pasteurized guava purees, methanethiol was the major constituent. The second major constituent in DP -CO2 was 1 p menthene8 -th iol followed by hydrogen sulfide and dimethyl disulfide. Mercapto-hexyl acetate was the second major constituent

PAGE 164

164 in pasteurized guava puree, followed by hydrogen sulfide and dimethyl disulfide. 4mercapto4 -methyl -penta 2 one was identified only in the DP -CO2 guava puree. 3mercapto hexyl acetate passion fruit mercaptan and 3mercapto1 -hexanol were identified as having high FD factors by (Steinhaus and others 2008). In addition, methional showed high odor activity. The first 2 compounds were previously identified in passion fruit and were reported for first time in guava puree by (Steinhaus and others 2008). Hydrogen sulfide and dimethyl sulfide were reported for the first time in guava puree but were previously identified in guava leaves (Rouseff and other s 2008). The presence of dimethyl disulfide in the puree is not surprising since it is produced as a plant defense response. Conclusion A total of 76 compounds were identified in guava purees. Eleven compounds were identified for the first time, but were previously identified in other foods. These eleven compounds are: butanal (previously reported in delicious apple fruit), isoamyl acetate (strawberry), 4mercapto-4 methyl penta-2 one (key component of sauvignon grape, and reported in grapefruit), phenyl acetaldehyde (key odorant in honey and found in chocolate and buckwheat), nonanal (reported in grapefruit, orange juice and red delicious apple), homofuraneol (coffee, muskmelon and soy sauce), methyl nonanoate (strawberry), 1-p menthene-8 thiol (grapefruit), 2 octane, ethyl 3 hydroxyhexanoate (yellow mombin) and ethyl nonanoate. Fresh, DP CO2 and pasteurized purees showed similar composition, but quantitative differences were found in the aroma active and sulfur compounds. Total peak area was highest for the fresh sample while pasteurization caused a slight decrease in concentration. DP CO2 reduced the total peak area by about 24%. Twenty six compounds were identified as

PAGE 165

165 guava aroma contributors, three of which were tentatively identified. Pentene3one, oc ten -3 ol, benzyl alcohol, phenyl acetaldehyde, homofuraneol, (E,Z) -2,6 -nonadienal and carvone, were not detected by GC -MS, so these compounds may play an important role in guava flavor. Hexanal was the volatile present in highest concentration in the three samples and (Z) 3 -hexenyl hexanoate was the major odor contributor for the fresh and DP -CO2 guava purees. GC PFPD was a useful tool in the identification of sulfur compounds. Four new sulfur compounds were identified in guava purees, 2 of which were previ ously identified in guava leaves (dimethyl sulfide and hydrogen sulfide). Sensory analysis indicate that there were differences between the freshly thawed and DP -CO2 guava purees, as confirmed by volatile composition but the acceptability of the DP -CO2 tr eated guava puree was not significantly different from the freshly thawed puree. Table 6 1. Difference in flavor and overall likeability between freshly thawed (reference and hidden reference), dense phase-CO2 processed (DP -C O2; 30.6 MPa, 8% CO2, 6.9 min, 35 C) and thermally treated ( 90 C, 60 s) guava purees detected by untrained panelists (n = 75) at weeks 0 Sample Differences in flavor* Overall likeability DP CO 2 5.66 a ** 4.13 a Pasteurized 3.56 b 3.79 a Hidden Reference (control or freshly thawed ) 3.05 b 4.29 a Difference observed when compared to given reference (difference from control test). ** Values with similar letters within columns are not significantly different (Tukeys HSD, p > 0.05).

PAGE 166

166 Figure 61. Chemical composition of headspace volatiles for guava purees. Total number of compounds for each class is put in parentheses. The three samples were normalized to the total peak area of fresh guava puree

PAGE 167

167 Figure 6.2Total ion chromatogram (TIC) for freshly thawed guava puree on DB -5 column

PAGE 168

168 Table 6 2. MS identification of guava puree volatiles. Peak areas were normalized (100) to the largest peak in all three samples. LRI Normalized Peak Area (%) # Name DB5 Wax Fresh DP CO2 Pasteurized 1 acetaldehyde 731 0.37 Nd Nd 2 ethanol 511 906 59.43 Nd 55.56 3 acetic acid 553 828 2.93 2.03 3.54 4 Butanal a 607 1472 5.82 Nd Nd 5 ethyl acetate 619 893 28.49 14.86 27.09 6 ethyl propanoate 714 947 20.18 15.60 17.66 7 2 Butanone, 3 hydroxy or acetoin 723 1318 1.14 0.66 0.80 8 1 butanol, 2 methyl or amyl alcohol 731 1.24 1.24 1.24 9 1 butanol, 3 methyl or isoamyl alcohol a 738 12 0 0 0.44 0.44 1.17 10 (Z) 3 hexenal 1158 0.30 0.69 0.18 11 hexanal 807 1101 100.00 79.25 95.54 12 ethyl 2 methyl butanoate 848 1053 0.21 0.14 0.15 13 3 hexen 1 ol (E) 851 1379 0.39 0.35 0.46 14 Z 3 hexenol 862 1402 36.93 33.10 33.44 15 1 hexanol 872 1366 36.29 28.54 32.10 16 1 butanol 3 methyl acetate 879 1137 0.78 0.80 0.90 17 methyl hexanoate 91 6 3.00 1.15 2.41 18 methyl 3 hexenoate 934 0.23 0.12 0.19 19 octen 3 one, 1 982 15.12 11.11 1.15 20 6 methyl 5 hepten 2 one or methyl heptenone 986 1347 1.83 5.68 11.01 21 2 octanone a 994 0.43 Nd Nd 22 ethyl hexanoate 996 1248 21.77 4.76 5.92 23 3 hexenyl acetate or 3 hexen 1 ol, acetate 1008 3.03 2.44 1.58 24 acetic acid hexyl ester or hexyl acetate 1013 1267 2.17 1.46 1.46 25 d limonene 1041 1205 6.16 3.30 6.84 26 1,8 Cineole 1047 1212 3.05 1.80 2.31 27 (E) b Ocimene 1055 1265 0.27 0.45 0.24 28 furaneol 1069 0.77 0.39 0.54 29 1 octanol 1072 1572 1.99 2.19 3.28 30 ethyl heptanoate 1335 0.88 0.89 0.47 31 linalool 1105 1554 0.92 0.99 0.47

PAGE 169

169 Table 6 2. Continued LRI Normalized Peak Area (%) # Name DB5 Wax Fresh DP CO2 Pasteurized 32 Nonanal a 1109 1420 0.52 1.01 1.00 33 methyl octanoate 1126 0.26 0.18 0.13 34 ethyl 3 hydroxyhexanoate a 1682 0.37 0.40 0.24 35 benzoic acid ethyl ester 1187 1709 0.87 1.00 0.59 36 a terpineol 1725 0.70 0.25 0.17 37 ethyl octanoate 1198 1452 1.65 1.38 0.68 38 decanal 1212 1526 0.39 0.48 0.48 39 methyl nonanoate a 1227 0.19 0.20 0.10 40 ethyl phenylacetate or 2 phenylacetate 1786 0.67 0.96 0.53 41 benzene propanol or 3 phenylpropanol 1254 2086 0.59 0.71 0.63 42 phenylethyl acetate 1259 0.36 0.41 0.16 43 nonanoic acid 1265 2.45 2.66 0.91 44 ethyl nonano ate a 1298 1.83 1.95 1.01 45 alpha cubebene 1339 1.01 1.01 7.34 46 ethyl 3 phenylpropionate or ethyl dihydrocinnamate 1367 0.42 0.48 0.21 47 cis 3 hexenyl hexanoate 1386 0.18 0.17 Nd 48 B damascenone 1407 1862 1.01 1.01 0.83 49 t caryophyllene 1446 1641 1.01 1.01 0.60 50 a humulene 1460 1650 17.39 18.82 26.96 51 g decalactone 1480 2130 1.37 2.08 2.25 52 b ionone 1495 1984 1.79 1.95 2.40 53 B bisabolone 1503 1754 1.01 1.01 2.95 54 beta or delta cadinene 1793 0.25 0.35 0.17 55 (E) or d nerolidol 1577 2055 0.52 0.71 0.25 56 2 phenylethyl alcohol 1963 0.22 0.25 0.30 57 3 phenyl propyl acetate 1984 0.25 0.40 0.27 58 eugenol 2183 1.04 1.18 1.77 aIdentify for first time in guava puree from Dominican Republic, Normalized area base on hexanal peak analyzed on DB 5 column

PAGE 170

170 Table 6 3. Guava puree aroma active compounds. Peak heights were normalized (100) to the most intense peak in all three samples. LRI Normalized Peak Area (%) # Name Descriptor DB5 Wax Fresh DP CO2 Pasteurized 1 ethyl acetate a e fruity, pleasant, sweet 627 42.90 ND ND 2 pentene 3 one, 1 a nutty 1023 42.72 56.50 ND 3 Hexanal a fatty, grassy 801 89.04 73.87 ND 4 (E) 3 hexenol a green, grassy 1376 ND 67.98 ND 5 (Z)3 hexenol a, c d fruity, green, grassy 1409 55.15 ND ND 6 Methional b cooked potato 863 1451 69.53 79.05 ND 7 octen 3 ol a, c mushroom 977 52.67 59.55 85.94 8 octen 3 one a, d mushroom 985 64.41 ND 80.23 9 ethyl hexanoate a, c, d, e fruity, sweet 999 1299 82.77 75.97 87.40 10 1,8 cineole a,c eucaliptus, camphoreous, menthol like 1232 62.43 70.29 ND 11 benzyl alcohol floral rose 1887 52.23 70.71 81.97 12 4 mercapto 4 methyl pentan 2 ol b d flowery 1533 48.59 70.43 71.73 13 phenyl acetaldehyde guava. Green, honey like 1044 80.47 74.78 80.97 14 Furaneol a, c, d fruity, sour, guava, caramel, cotton candy 1068 82.50 73.17 87.89 15 homofuraneol caramel, candy 2069 94.56 88.28 80.45 16 mercapto hexan 1 ol b d grapefruit 1126 1862 76.21 104.53 61.11 17 ethyl 3 hydroxyhexanoate a guava = fruity 1128 1672 61.03 93.11 ND 18 nonadienal, (E,Z) ,6 a, d green, cucumber 1163 63.43 95.53 90.25 19 ethyl octanoate a, c d sweet, fruity 1198 1434 78.07 89.43 82.26 20 ethyl phenylacetate a,c fruity 1254 50.01 ND ND 21 Carvone d spearmint, peppermint, sweet 1255 ND 95.53 73.22 22 p menthene 8 thiol b medicine 1298 1505 53.68 59.97 ND 23 B ionone a, d floral, fruity 1988 60.91 81.67 ND 24 Ethyl 3 phenylpropionate a, d floral 1356 80.57 63.18 73.74 25 (Z) 3 hexenyl hexanoate a,c d apple peel, fruit 1387 100.00 80.65 ND 26 g decalactone a nutlike or cococut 2146 45.37 61.45 ND Normalize d area base on (Z) 3 hexenyl hexanoate peak identified on DB 5 column aCompounds confirmed with GC MS, bCompounds confirmad with GC PFPD,GC PFPD, cPreviously reported as aroma contributors by Jordan and others ( 2003) dPreviously reported as aroma contributors by Mahattanatawee and others ( 2005) ePreviously reported as aroma contributor by Macleod and Gonzalez de Troconis ( 1982) bold compounds= newly identified compounds

PAGE 171

171 Table 6 4. Guava puree sulfur volatile compounds. Peak heights were normalized (100) to the most intense peak in all three samples. LRI Normalized Peak Area (%) # Name DB5 Wax Fresh DP CO2 Pasteurized 1 hydrogen sulfide a 472 73.02 55.54 25.90 2 Methanethiol a,b 477 679 96.19 71.91 69.20 3 dimethyl sulfide a 523 731 8.89 7.21 29.34 4 dimethyl disulfide a,b 759 1049 82.99 37.85 10.17 5 2 methyl 3 furanthiol b 869 1338 1.46 0.95 2.90 6 Methional a, b 912 1441 1.85 0.42 0.61 7 4 mercapto 4 methyl pentan 2 one 936 1361 Nd 0.80 Nd 8 4 mercapto 4 methyl pentan 2 ol b 1057 1538 2.85 3.21 3.23 9 3 mercapto hexan 1 ol b 1117 1 862 4.75 1.55 17.64 10 mercapto hexyl acetate b 1241 1879 20.23 20.07 30.88 11 1 p menthene 8 thiol 1302 1499 100.00 57.54 14.37 aPreviously reported in guava leaves bPreviously reported in guava puree, Normalized area base onp menthene 8 thiol peak identified on DB 5 column, bold compounds= newly identified compounds

PAGE 172

172 CHAPTER 7 SUMMARY AND CONCLUSI ONS D ense phase carbon dioxide (DP-CO2) was evaluated as an alternative to pasteurization for treatment of guava puree to extend shelf life, reduce microbiological load, and preserve sensory and nutritional characteristics of the puree. Processing of puree with DP -CO2 was facilitated by first treating the puree with a commercially available enzyme, Bioguavase at low temperature (30 C) to obtain a product with the proper consistency for processing without affecting the physiochemical properties T he optimial time and concentration for enzyme treatment were determined to be 600 ppm and 3 hours of reaction time. Additional experiments addressed the optimization of DP CO2 treatment for microbial reduction in guava puree. P hysical, chemical, microbial and sensory quality of DP -C O2 treated guava puree was compared to freshly thawed and heat pasteurized samples The first objective was to optimize the reaction time and concentration of a commercially available enzyme to obtain a product with the proper consistency for DP CO2 processing. E nzyme treatment decreased (between 0.55 to 11% of the initial concentration) t he antioxidant capacity of the samples and the antioxidant capacity was lower regardless of the enzyme concentration. The turbidity decreased with enzyme activity The enzyme activity affected the pH and TSS content: the pH decreased (between 0.5 and 1.6% of the initial pH) while TSS content increased (between 4 and 10% of the initial value). C olor was affected by the enzyme concentration since the enzyme treatment produced a clarified juice. T hree hours of reaction time and 600 ppm of enzyme concentration we re adequate to produce a clarified juice with a minimal

PAGE 173

173 effect on n utritional quality, even though antioxidant capacity was decreased and total soluble polyphenols were increased. When conducting the DP -CO2 optimization, CO2 solubility in the guava puree was measured using an apparatus designed and built at the University of Florida Food Science and Human Nutrition D epartment (Gainesville, FL) This measurement showed that 5.3% CO2 was adequate to reach saturation of the product. Due to equipment limitations, 8% CO2 was used. Surface response analysis of microbial reduction showed that the quadratic model fit for yeasts and mold (Y&M) was statistically significant ( P < 0.05) and there was a satisfactory correlation between the actual and the fitt ed values for Y&M (R2 = 0. 88 ). According to the regression surface equation, 8 min was sufficient to achieve a 5 log reduction using a processing pressure of 34.5 MPa and a temperature of 35 C. According to the results obtained for aerobic plate count (A PC) a complete sterilization could not be obtained. Because of the increase in viscosity and foam formation in the puree during treatment, the residence time was set to 6.9 min. The quality attributes pH, Brix, titratable acidity, color, APC, Y&M), and ph ytochemical content were determined during 14 weeks of storage at 4 C and compared to those of freshly thawed (fresh) and heat pasteurized samples. APC for fresh and DP -CO2 guava puree remained constant (3.5 log) during the first 7 weeks of storage, after which there was a significant increase in microbiological load for the DP CO2 treated guava. Pasteurized guava puree APC started at 1.5 log and increased significantly (approximately 1.5 log) at the end of 4 weeks DP -CO2 and pasteurization treatments of guava puree caused approximately 1.3 log reduction in Y&M counts when compared to fresh. DP -CO2 Y&M count was lower than the pasteurized treatment

PAGE 174

174 during the 14 weeks of st orage. The pH and Brix remained between 3.72 and 3.93 and between 6.7 and 7.4, respectively for DP -CO2 and pasteurized, throughout storage. Titratable acidity of DP -C O2 treated samples was significantly higher than fre sh and pasteurized samples DP CO2 caused partial inactivation (20%) of pectinesterase (PE) activity and an increase (100%) in guava puree viscosity. Further study is recommended to understand the mechanism of gel formation and cloud loss due to DP CO2 treatment. ORAC and total soluble ph enolics (TSP) values for fresh guava puree were between 12.06 and 16.3 Mol TE/mL and 174.5 and 186.39 mg of GAE/ L of sample respectively. DP -C O2 values were between 11.08 and 16.95 Mol TE/mL (ORAC) and 181.15 and 196.51 mg of GAE/ L of sample (TSP). DPCO2 treatment can protect polyphenolic and antioxidant levels throughout processing and storage without compr om ising physico -chemical and phytochemical properties of the guava puree. DP CO2 delayed the degradation of Vitamin C content during sto rage compared to the fresh puree. DP -CO2 treated guava puree retained organic acids contents similar to fresh guava puree. Flavor analysis of guava puree allowed the identification of 76 compounds. Eleven compounds were identified for first time in guava puree, but were previously identified in other foods. Fresh, DP CO2 and pasteurized puree showed a similar composition but quantitative differences were found in the aroma active and sulfur compounds. Total peak concentration was highest for the fresh sam ple while pasteurization caused a slight decrease in the compounds concentration. DP -CO2 reduced the total peak area in about 24% when compared to the total peak area of the fresh guava puree. Twenty six compounds were identified as guava aroma contributor s, three of which were tentatively

PAGE 175

175 identified. These compounds were not detected by GC MS, so they may play an important role in guava aroma. Hexanal was the volatile present in higher concentration in the three samples and (Z) -3 -hexenyl hexanoate was the major odor contributor for the fresh and DP -CO2 guava puree. Twenty -four compounds were detected as aroma contributors of the fresh guava puree while only 22 and 13 compounds were identified as aroma active compounds in DP -CO2 and pasteurized guava puree r espectively. This study showed that DP -C O2 treatment extended shelf -life and preserved the quality of guava puree. F u rther work is needed to investigate the mechanisms and causes for the increase in viscosity of DP -CO2 treated guava puree during storage and to understand the mechanism of gel formation and cloud loss due to DP -CO2 treatment Sensory evaluation with trained panelists will help to better understand the difference between fresh, DP -CO2 and pasteurized puree. Further product development of guava juice or nectars should be conducted to determine the effect of DP CO2 nutritional quality of this product.

PAGE 176

176 APPENDIX A ENZYME TRATMENT OF G UAVA PUREE Table A -1. SAS software code used for the statistical analysis of repeated measurement design a n d Tukeys st an d ard ized range (HSD) test data guava; infile cards; input replicate conc time brix @@; cards ; 1 0 0 6 1 400 0 6 1 600 0 6 1 800 0 6 2 0 0 6 2 400 0 6 2 600 0 6 2 800 0 6 3 0 0 6 3 400 0 6 3 600 0 6 3 800 0 6 4 0 0 5.9 4 400 0 5.9 4 600 0 5.9 4 800 0 5.9 5 0 0 5.9 5 400 0 5.9 5 600 0 5.9 5 800 0 5.9 6 0 0 5.9 6 400 0 5.9 6 600 0 5.9 6 800 0 5.9 1 0 3 6 1 400 3 6 1 600 3 6.5 1 800 3 6.2 2 0 3 6 2 400 3 6 2 600 3 6.5 2 800 3 6.2 3 0 3 6 3 400 3 6 3 600 3 6.5 3 800 3 6.2 4 0 3 6 4 400 3 6.4 4 600 3 6.6 4 800 3 6.2 5 0 3 6 5 400 3 6.4 5 600 3 6.6 5 800 3 6.2 6 0 3 6 6 400 3 6.4 6 600 3 6.6 6 800 3 6.2 1 0 6 6 1 400 6 6 1 600 6 6.5 1 800 6 6.5 2 0 6 6 2 400 6 6 2 600 6 6.5 2 800 6 6.5 3 0 6 6 3 400 6 6 3 600 6 6.5 3 800 6 6.5 4 0 6 6 4 400 6 6.4 4 600 6 6.5 4 800 6 6.5 5 0 6 6 5 400 6 6.4 5 600 6 6.5 5 800 6 6.5 6 0 6 6 6 400 6 6.4 6 600 6 6.5 6 800 6 6.5 1 0 9 6.1 1 400 9 6.3 1 600 9 6.4 1 800 9 6.4 2 0 9 6.1 2 400 9 6.3 2 600 9 6.4 2 800 9 6.4 3 0 9 6.1 3 400 9 6.3 3 600 9 6.4 3 800 9 6.4 4 0 9 6.1 4 400 9 6 4 600 9 6.5 4 800 9 6.4 5 0 9 6.1 5 400 9 6 5 600 9 6.5 5 800 9 6.4 6 0 9 6.1 6 400 9 6 6 600 9 6.5 6 800 9 6.4 1 0 12 6.1 1 400 12 6 1 600 12 6.5 1 800 12 6.5 2 0 12 6.1 2 400 12 6 2 600 12 6.5 2 800 12 6.5 3 0 12 6.1 3 400 12 6 3 600 12 6.5 3 800 12 6.5 4 0 12 6.1 4 400 12 6.4 4 600 12 6.4 4 800 12 6.5 5 0 12 6.1 5 400 12 6.4 5 600 12 6.4 5 800 12 6.5 6 0 12 6.1 6 400 12 6.4 6 600 12 6.4 6 800 12 6.5 ; proc print ; run ; proc sort ; by conc time; run; proc glm ; class conc time; model brix=conc|time; run; proc sort ; by conc; run ; proc glm ; class time; model brix = time; means time / tukey ; by conc; run; proc sort ; by time; run ; proc glm ; class conc; model brix = conc; means conc / tukey ; by time; run;

PAGE 177

177 Table A -2. SAS software output used for the statistical analysis of repeated measurement design a nd Tukeys st an d ard ized range (HSD) test The GLM Procedure Dependent Variable: brix Sum of Source DF Squares Mean Square F Value Pr > F Model 19 5.43300000 0.28594737 29.79 <.0001 Error 100 0.96000000 0.00960000 Corrected Total 119 6.39300000 R Square Coeff Var Root MSE brix Mean 0.849836 1.576502 0.097980 6.215000 Source DF Type I SS Mean Square F Value Pr > F conc 3 2.28300000 0.76100000 79.27 <.0001 time 4 2.18550000 0.54637500 56.91 <.0001 conc*time 12 0.96450000 0.08037500 8.37 <.0001 --------------------------------------------conc=0 -------------------------------------------Dependent Variable: brix Sum of Source DF Squares Mean Square F Value Pr > F Model 4 0.10800000 0.02700000 45.00 <.0001 Error 25 0.01500000 0.00060000 Corrected Total 29 0.12300000 R Square Coeff Var Root MSE brix Mean 0.878049 0.406217 0.024495 6.030000 Source DF Type I SS Mean Square F Value Pr > F time 4 0.10800000 0.02700000 45.00 <.0001 Source DF Type III SS Mean Square F Value Pr > F time 4 0.10800000 0.02700000 45.00 <.0001 Tukey's Studentized Range (HSD) Test for brix NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Ty pe II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 25 Error Mean Square 0.0006 Critical Value of Studentized Range 4.15337 Minimum Significant Difference 0.0415 Means with the same l etter are not significantly different. Tukey Grouping Mean N time A 6.10000 6 12 A 6.10000 6 9 B 6.00000 6 6 B 6.00000 6 3 C 5.95000 6 0

PAGE 178

178 --------------------------------------------time=0 ------------------------------------------Dependent Variable: brix Sum of Source DF Squares Mean Square F Value Pr > F Model 3 0.00000000 0.00000000 0. 00 1.0000 Error 20 0.06000000 0.00300000 Corrected Total 23 0.06000000 R Square Coeff Var Root MSE brix Mean 0.000000 0.920542 0.054772 5.950000 Source DF Type I SS Mean Square F Value Pr > F conc 3 0 0 0.00 1.0000 Source DF Type III SS Mean Square F Value Pr > F conc 3 0 0 0.00 1.0000 Tukey's Studentized Range (HSD) Test for brix NOTE: This test controls the Type I experimentwise error rate, but it generally has a higher Type II error rate than REGWQ. Alpha 0.05 Error Degrees of Freedom 20 Error Mean Square 0.003 Critical Value of Studentized Range 3.95829 Minimum Significant Difference 0.0885 Means with the same letter are not significantly different. Tukey Grouping Mean N conc A 5.95000 6 0 A A 5.95000 6 400 A A 5.95000 6 600 A A 5.95000 6 800

PAGE 179

179 Figure A 1. Enzyme treated guava puree without filtration (left) and after filtration (right) This picture showed that enzyme treated puree increased juice yield but caused discoloration of puree after filtering through cheesecloth (sample to the right). A possible explanation for color loss is that compounds such as phenolic compounds responsible of provide color to the puree were retained on the guava cake or stayed behind with the filtered out particles. These compounds were removed by the filtration process so the juice was much lighter in color. Similar result was observed by Imungi and others (1980).

PAGE 180

180 APPENDIX B DP -CO2 PROCESSING AND DATA A. B. C. D. Figure B 1. Removal of insoluble solids from guava puree A. Picture presents guava puree solid removal using 200 m nylon filter B. Collection of the guava puree C Mixing of guava puree D Collection of guava puree for DP -CO2 treatment

PAGE 181

181 Figure B 2. Experimental solubility apparatus Experimental solubility apparatus: CO2 tank, high pressure pump (HPP), heat exchanger (HE), pressure gauge (PG), back pressure regulator (BPR), thermocouple (T), metering valve (MV), gas meter (GM), water bath, bottle, stopper, fume hood (FH), two way valves (V), three way valves (TV), vessels and quick connects (QC) (Calix 2008)

PAGE 182

182 Figure B 3. Continuous DP -CO2 system used during processing

PAGE 183

183 Table B -1. Guava Puree Solubility data TRTs Pressure (MPa ) CO2 Volume (Slt) CO2 (grams) Guava Volume (ml) CO2 Solubility Mean St Dev Control 0 0 Control 0 0 TRT 1 6.9 3.126 6.14 162 3.79 TRT 1 6.9 3.213 6.31 162 3.90 3.84 0.074609 TRT 2 10.34 3.399 6.68 168 3.98 TRT 2 10.34 3.401 6.68 166 4.03 4.00 0.035539 TRT 3 17.24 3.612 7.10 168 4.22 TRT 3 17.24 3.601 7.08 168 4.21 4.22 0.009096 TRT 4 24.13 3.634 7.14 168 4.25 TRT 4 24.13 3.487 6.85 164 4.18 4.21 0.051231 TRT 5 31.03 3.611 7.09 164 4.33 TRT 5 31.03 3.638 7.15 168 4.25 4.29 0.050504 Table B -2. Guava Puree pH, Brix and Titratable acidity measurement before and after the CO2 solubility determination TRT's Pressure (MPa ) pH Mean St Dev Brix Mean St Dev % acid Mean St Dev Control 0 3.82 6.8 0 0.5 6 Control 0 3.79 3.8 1 0.021 6.7 0 6.75 0.071 0.5 5 0.55 0.007 TRT 1 6.9 3.77 6.7 7 0.5 5 TRT 1 6.9 3.8 0 3.7 9 0.01 9 6.73 6.75 0.02 4 0.55 0.55 0.00 5 TRT 2 10.34 3.75 6.8 0 0.58 TRT 2 10.34 3.75 3.75 0.002 6.8 0 6.8 0 0 0.58 0.58 0.000 TRT 3 17.24 3.74 6.8 0 0. 60 TRT 3 17.24 3.75 3.7 5 0.007 6.9 0 6.88 0.071 0. 60 0. 60 0.002

PAGE 184

184 TRT's Pressure (MPa ) pH Mean St Dev Brix Mean St Dev % acid Mean St Dev TRT 4 24.13 3.72 6.8 0 0. 60 TRT 4 24.13 3.7 5 3.7 4 0.016 7 .0 6.9 0 0.141 0.59 0. 60 0.003 TRT 5 31.03 3.71 7 .0 0.6 6 TRT 5 31.03 3.73 3.72 0.01 4 6.9 6.95 0.07 1 0.65 0.66 0.006

PAGE 185

185 APPENDIX C MICROBIAL INACTIVATION AND STORAGE STUDY DATA AND ANALYSIS Table C -1. The average initial and final aerobi c plate counts (APC) standard deviations at 11 experimental runs from 2 -factor, 3 -level Central Composite D esign (CCD) Run Pressure (MPa) Residence time (min) I nitial load (cfu/mL)* Final load (cfu/mL)* 1 34.5 8 1.06 x 10 7 + 1.635 x 10 5 6625 + 6. 6 2 24.1 8 1.06 x 10 7 + 1.635 x 10 5 5650 + 9.1 5 3 24.1 6.5 1.06 x 10 7 + 1.635 x 10 5 5130 + 6. 40 4 13.8 8 1.06 x 10 7 + 1.635 x 10 5 6150 + 6.03 5 34.5 6.5 2.35 x 10 7 + 7.93 x 10 5 3430 + 3.77 6 24.1 6.5 2.35 x 10 7 + 7.93 x 10 5 2730 + 3.86 7 13.8 6.5 2.35 x 10 7 + 7.93 x 10 5 2430 + 2.2 2 8 13.8 5 2.35 x 10 7 + 7.93 x 10 5 1680 + 2.5 9 34.5 5 2.17 x 10 7 + 10.8 8 x 10 5 5150 + 12.50 10 24.1 5 2.17 x 10 7 + 10.8 8 x 10 5 5700 + 7.0 5 11 24.1 6.5 2.17 x 10 7 + 10.8 8 x 10 5 5800 + 4.69 *Averages of the plates with APC counts lower than 2 50 colony forming units (cfus) Table C -2. SAS software code used for the response surface methodology (RSM) analysis of 11 experimental runs determined by Central Composite Design title 'OPTIMIZATION1' ; data optimization1; input TRT$ P RT APC YM; datalines ; T1 34.5 8 3.204816837 4.297278482 T5 34.5 6.5 3.509089594 3.863291003 T9 34.5 5 3.706133727 3.207230794 T2 24.1 8 3.276150588 4.283911356 T3 24.1 6.5 3.317110445 4.075429733 T6 24.1 6.5 3.521746446 3.965684468 T11 24.1 6.5 3.651803709 3.654499463 T10 24.1 5 3.655371191 3.490671907 T4 13.8 8 3.233819901 4.087768546 T7 13.8 6.5 3.586547572 4.032734615 T8 13.8 5 3.572065724 4.175137907 ; proc sort ; by P RT; run; proc print ; run; proc rsreg data=optimization1; model APC YM = P RT / lackfit ; run;

PAGE 186

186 Table C -3. SAS software output of the response surface methodology (RSM) regression analysis of 1 1 experimental -run data determined by central composite design Response Surface for Variable APC Response Mean 3.475878 Root MSE 0.116568 R Square 0.7990 Coefficient of Variation 3.3536 Type I Sum Regression DF of Squares R Square F Value Pr > F Linear 2 0.247697 0.7327 9.11 0.0215 Quadratic 2 0.015760 0.0466 0.58 0.5936 Crossproduct 1 0.006659 0.0197 0.49 0.5151 Total Model 5 0.270116 0.7990 3.98 0.0780 Sum of Residual DF Squares Mean Square F Value Pr > F Lack of Fit 3 0.011004 0.003668 0.13 0.9348 Pure Error 2 0.056937 0.028468 Total Error 5 0 .067941 0.013588 Parameter Estimate Standard from Coded Parameter DF Estimate Error t Value Pr > |t| Data Intercept 1 2.549984 1.435044 1.78 0.1357 3.515327 P 1 0.015333 0.041329 0.37 0.7258 0.004601 RT 1 0.373874 0.433903 0.86 0.4283 0.203196 P*P 1 0.000045421 0.000684 0.07 0.9496 0.004866 RT*P 1 0.002628 0.003754 0.70 0.5151 0.040800 RT*RT 1 0.034298 0.032550 1.05 0.3403 0.077170 Sum of Factor DF Squares Mean Square F Value Pr > F P 3 0.006846 0.002282 0.17 0.9136 RT 3 0.269317 0.089772 6.61 0.0343

PAGE 187

187 Response Surface for Variable YM Response Mean 3.921240 Root MSE 0.166730 R Square 0.8815 Coefficient of Variation 4.2520 Type I Sum Regression DF of Squares R Square F Value Pr > F Linear 2 0.680830 0.5804 12.25 0.0118 Quadratic 2 0.007173 0.0061 0.13 0.8818 Crossproduct 1 0.346022 0.2950 12.45 0.0168 Total Model 5 1.034025 0.8815 7.44 0.0230 Sum of Residual DF Squares Mean Square F Value Pr > F Lack of Fit 3 0.043640 0.014547 0.31 0.8241 Pure Error 2 0.095354 0.047677 Total Error 5 0.138994 0.027799 Parameter Estimate Standard from Coded Parameter DF Estimate Error t Value Pr > |t| Data Intercept 1 6.059405 2.052571 2.95 0.0318 3.896719 P 1 0.161934 0.059113 2.74 0.0408 0.154640 RT 1 0.208145 0.620619 0.34 0.7510 0.299793 P*P 1 0.000494 0.000978 0.50 0.6350 0.052901 RT*P 1 0.018945 0.005370 3.53 0.0168 0.294117 RT*RT 1 0.003808 0.046557 0.08 0.9380 0.008569 Sum of Factor DF Squares Mean Square F Value Pr > F P 3 0.496394 0.165465 5.95 0.0419 RT 3 0.883761 0.294587 10.60 0.0132

PAGE 188

188 Figure C -1. DP -CO2 treated samples during storage (week 4) (a) no agitation (b) after agitation

PAGE 189

189 APPENDIX D CHANGES IN AROMA COM POUNDS AND SENSORY PERCEPTION IN GUAVA PUREE AFTER THERMAL AND NON-THERMAL PROCESSING Figure D 1. Questionnaire used for taste panel: different from control test demographic questions TODAY'S SAMPLE: Guava Drink To start the test, click on the Continue button below: Question # 1. Please indicate your gender. Male Female Question # 2. Male: Which of the following ranges includes your age ? Under 18 18 20 21 24 25 29 30 34 35 39 40 44 45 49 50 54 55 59 60 65 Over 65 Question # 3. Female: Which of the following ranges includes your age ? Under 18 18 20 21 24 25 29 30 34 35 39 40 44 45 49 50 54 55 59 60 65 Over 65 Question # 4. Have you ever consumed any guava juice products? Yes No Question # 5. How often do you consume guava beverages? More than once a day Once a day 2 3 times a week Once a week 2 3 times a month Once a month Twice a year Once a year Less than once a year

PAGE 190

190 Figure D 2 Questionnaire used for taste panel: different from control test sensory questions Take a bite of cracker and a sip of water to rinse your mouth. Remember to do this before you taste each sample. WHEN ANSWERING ANY QUESTION, MAKE SURE THE NUMBER ON THE CUP MATCHES THE NUMBER ON THE MONITOR. Please click on the 'Continue' button below. Question # 6. You are being presented with a reference sample marked 000. Please taste this sample and then taste the following samples and compare them to the reference sample. Mark how different the sample is from the reference sample using the line scale below. Sample <> Not Different Very Different Sample <> Not Different Very Different Sample <> Not Different Very Different T ake a bite of cracker and a sip of water to rinse your mouth. Remember to do this before you taste each sample. WHEN ANSWERING ANY QUESTION, MAKE SURE THE NUMBER ON THE CUP MATCHES THE NUMBER ON THE MONITOR. Please click on the 'Continue' button below. Question # 7. Please indicate how much you like the OVERALL TASTE of each sample. Overall Taste Sample <> Dislike extremely Dislike very much Dislike moderately Dislike slightly Neither like nor dislike Like slightly Like moderately Like very much Like extremely 1 2 3 4 5 6 7 8 9 Sample <> Dislike extremely Dislike very much Dislike moderately Dislike slightly Neither like nor dislike Like slightly Like moderately Like very much Like extremely 1 2 3 4 5 6 7 8 9 Sample <> Dislike extremely Dislike very much Dislike moderately Dislike slightly Neither like nor dislike Like slightly Like moderately Like very much Like extremely 1 2 3 4 5 6 7 8 9

PAGE 191

191 Figure D 3 Questionnaire used for taste panel: different from control test comments Question # 8 Using the keyboard located in the tray under the counter, please describe the differences if any, between the samples (please be specific) Sample <> _________________________________________________________________________________ _________________________________________________________________________________ _________________________________________________________________________________ Sample <> _________________________________________________________________________________ _________________________________________________________________________________ _________________________________________________________________________________ Sample <> _________________________________________________________________________________ _________________________________________________________________________________ _______________________________________________________________________ __________ The test has ended.

PAGE 192

192 Table D -1 Volatile compounds, CAS number, identification method, reported Linear Retention Indexes and references for previously reported studies Name CAS Number Identification method LRI DB5 LRI Wax Previously reported 1 Acetaldehyde 75 07 0 GC MS < 500 700 1, 4, 10, 12 2 hydrogen sulfide* 77783 06 4 S <500 528 13 3 Methanethiol* 74 93 1 S <500 675 10, 13 4 dimethyl sulfide* 75 18 3 S 519 736 13 5 Ethanol 64 17 5 GC MS 537 926 7 6 butanal 123 72 8 GC MS 518 813 7 acetic acid 64 19 7 GC MS 600 1459 1, 9 8 ethyl acetate 141 78 6 GC MS, O 600 890 1, 2, 4, 5, 6, 7, 9 9 pentene 3 one 1629 58 9 GC MS, O 697 1032 1, 5, 10 ethyl propanoate 105 37 3 GC MS 714 949 1, 5, 6, 9 11 3 Hydroxy 2 Butanone 513 86 0 GC MS 718 1281 1, 5, 6, 9 12 1 butanol, 2 methyl 137 32 6 GC MS 733 1209 9 13 1 butanol, 3 methyl 123 51 3 GC MS 739 1184 9 14 dimethyl disulfide* 624 92 0 S 740 1064 5, 13 15 (Z) 3 hexenal 6789 80 6 GC MS 793 1151 1, 5, 6, 8, 11, 12, 13 16 hexanal 66 25 1 GC MS, O 796 1099 1, 2, 4, 5, 6, 7,8, 9, 11, 13 17 ethyl 2 methyl butanoate 7452 79 1 GC MS 846 1062 13 18 (E) 3 hexenol 928 97 2 GC MS, O 851 1388 1, 2, 5, 6 19 (Z) 3 hexenol 928 96 1 GC MS, O 855 1399 1, 2, 4, 5, 6, 10, 13 20 1 hexanol 111 27 3 GC MS 865 1364 1, 2, 5, 6, 7, 8, 9 21 2 methyl 3 furanthiol* 28588 74 1 S 870 1324 10 22 1 butanol 3 methyl acetate 123 92 2 GC MS 876 1110 23 methyl hexanoate 106 70 7 GC MS 906 1177 4, 5, 7, 8, 9 24 methyl 3 hexenoate 2396 78 3 GC MS 5 25 Methional* 3268 49 3 S, O 909 1450 11, 13 26 4 mercapto 4 methyl pentan 2 one* 1987 52 7 S 942 1366

PAGE 193

193 Name CAS Number Identification method LRI DB5 LRI Wax Previously reported 27 octen 3 ol 3391 86 4 GC MS, O 978 1452 9 28 Octen 3 one 4312 99 6 GC MS, O 980 1315 5, 10, 11 29 6 methyl 5 hepten 2 one 110 93 0 GC MS 982 1347 5, 9 30 2 octanone 111 13 7 GC MS 997 1304 31 Ethyl hexanoate 123 66 0 GC MS, O 998 1246 1, 4, 5, 6,8, 9, 10 32 (Z) 3 hexen 1 ol acetate 1708 82 3 GC MS 1007 1313 1, 2, 4, 5, 6, 7, 8, 9 33 Hexyl acetate 142 92 7 GC MS 1008 1267 4, 5, 8, 9 34 d limonene 138 86 3 GC MS 1031 1205 2, 3, 4, 5, 6, 9 35 1,8 Cineole 470 82 6 GC MS, O 1032 1231 6, 7, 9 36 (E) b Ocimene 3779 61 1 GC MS 1037 1265 5, 6, 7, 8, 9 37 Benzyl alcohol 100 51 6 O 1039 1893 1, 7 38 4 mercapto 4 methyl pentan 2 ol* 31539 84 1 S, O 1042 1534 10 39 Phenyl acetaldehyde 122 78 1 O 1047 1675 40 Furaneol 3658 77 3 GC MS, O 1062 2058 1, 5, 10, 11, 12 41 1 octanol 111 87 5 GC MS 1068 1575 1, 2, 4, 5, 9, 10 42 Homofuraneol 27538 10 9 O 1078 2073 43 Ethyl heptanoate 106 30 9 GC MS 1095 1346 10 44 Linalool 78 70 6 GC MS 1100 1557 1, 9, 10, 11 45 Nonanal 124 19 6 GC MS 1103 1416 46 Methyl octanoate 111 11 5 GC MS 1107 1378 5 47 Ethyl 3 hydroxyhexanoate 2305 25 1 GC MS, O 1126 1694 48 3 mercapto 1 hexanol* 51755 83 0 S, O 1126 1862 10, 11, 12 49 (E,Z) 2,6 nonadienal 557 48 2 GC MS, O 1153 1611 5, 10 50 Ethyl benzoate 93 58 3 GC MS 1185 1658 1, 5, 6, 9 51 terpineol 98 55 5 GC MS 1193 1724 1, 2, 6 52 Ethyl octanoate 106 31 2 GC MS, O 1193 1450 1, 4, 5, 6, 9, 10 53 Decanal 112 31 2 GC MS 1203 1523 5 54 Methyl nonanoate 1731 84 6 GC MS 1224 1479

PAGE 194

194 Name CAS Number Identification method LRI DB5 LRI Wax Previously reported 55 Ethyl 2 phenylacetate 101 97 3 GC MS, O 1243 1785 1, 4, 5, 9, 12 56 Mercapto hexyl acetate* 1369 54 20 6 S 1244 1725 11, 12 57 Carvone 99 49 0 GC MS, O 1249 1779 10 58 3 phenylpropanol 122 97 4 GC MS 1252 1993 1, 5, 9 59 Phenethyl acetate 103 45 7 GC MS 1260 1785 1, 2 60 Nonanoic acid 112 05 0 GC MS 1275 2110 5 61 Ethyl nonanoate 123 29 5 GC MS 1280 1523 62 1 p Menthene 8 thiol* 71159 90 5 S, O 1295 1507 63 cubebene 17699 14 8 GC MS 1345 1482 8 64 Ethyl 3 phenylpropionate 2021 28 5 GC MS, O 1351 10 65 Eugenol 97 53 0 GC MS 1357 2187 1, 5, 10 66 (Z) 3 hexenyl hexanoate 53398 86 0 GC MS, O 1381 1654 5 67 B damascenone 23726 93 4 GC MS 1390 1860 10, 13 68 t Caryophyllene 87 44 5 GC MS 1432 1641 1, 2, 3, 4, 5, 6, 8, 9, 10, 13 69 Humulene 6753 98 6 GC MS 1454 1650 1, 3, 4, 8 70 g Decalactone 706 14 9 GC MS, O 1472 2132 1, 5, 11 71 Ionone 79 77 6 GC MS, O 1496 1984 1, 2, 10 72 Bisabolene 495 61 4 GC MS 1498 1735 1, 3, 4, 8, 13 73 beta or delta cadinene 483 76 1 GC MS 1524 1794 6, 8 74 (E) or d nerolidol 142 50 7 GC MS 1564 2047 1, 8 75 2 hidroxyethyl benzene 60 12 8 GC MS 1118 1859 5 76 3 phenyl propyl acetate 122 72 5 GC MS 1347 1941 1, 5, 6, 8 1= (Nishimura and others 1989) 2= (Stevens and others 1970) 3= (Wilson and Shaw 1978) 4= (Macleod and Gonzalez de Troconis 1982) 5= (Idstein and Schreier 1985) 6= (Chyauand others 1992) 7= (Yen and Lin 1999) 8= (Paniandy and others 2000) 9= (Jordan and others 2003) 10= (Mahattanatawee and others 2005) 11= (Steinhaus and others 2008) 12= (Steinhaus and others 2009) and 13= (Rousef f and others 2008) Confirm with authentic standard, bold compounds= identify for first time in guava puree, S= LRI from GC PFPD, O = LRI from GC O

PAGE 195

195 LIST OF REFERENCES American Official Analytical Chemist (AOAC). 1990. American Official Analyt ical Chemist Official Method of Analysis. 9th ed. AOAC method 967.21, Washington DC. Arreola AG, Balaban MO, Marshall M, Peplow AJ, Wei CI & Cornell J. 1991. Supercritical carbon dioxide effects on some quality attributes of single strength orange juice. J Food Sci 56(4):10303. Balaban MO, Arreola AG, Marshall M, Peplow A, Wei CI Cor nell J. 1991. Inactivation of pectinesterase in o range juice by supercritical c arbon d ioxide. J Food Sci 56(3):743-6. Balaban MO, Ferrentino G, Ramirez M, Plaza ML. 2008. Review of dense phase carbon dioxide application to citrus juices 54th Annual Citrus Engineering Conference ; 2008 March 4; Ben Hill Griffin Jr. Citrus Hall Citrus Research & Education Center Lake Alfred, Florida: American Society of Mechanical Engineers, Florida Section. Ballestra P, DaSilva AA, Cuq JL. 1996. Inactivation of Escherichia coli by carbon dioxide under pressure. J Food Sci 61(4):82931. Bloch A, Thomson CA. 1995. Position of the American Dietetic Association: Phytochemicals and functional foods. J Am Diet Assoc 95(4):49396. Brasil IM, Maia GA, de Figueiredo RW. 1995. Physical -chemical changes during extraction and clarification of guava juice. Food Chem 54(4):383-86. Br avo L, Abia R, Saura-Calixto F. 1994. Polyphenols as dietary fiber asso ciated compounds. Comparative study on in vivo and in vitro properties. J Agric Food Chem 42(7):14817. Brekkee JE, Redelinghuys HP, Torline PA. 1986. Enzyme treatment and concentration of guava pulp. Acta hortic 194:229-40. Bru yne TD, Pieters L, Deelstr a H, Vlietinck A. 1999. Condensed vegetable tannins: Biodiversity in structure and biological activities. Biochem Syst Ecol 27(4):4 4559. Buettner A, Schieberle P. 1999. Characterization of the most odor active volatiles in fresh, hand-squeezed juice of grapefruit ( Citrus paradisi Macfayden). J Agric Food Chem 47(12):518993. Butz P, Tauscher B. 2002. Emerging technologies: chemical aspects. Fod Res Int 35(2 3):279 -84.

PAGE 196

196 Calix TF, Ferrentino G, Balaban MO. 2008. Measurement of high-pressure carbon dioxide solubility in orange juice, apple juice, and model liquid foods. J Food Sci 73(9):E43945. Carasek E, Pawliszyn J. 2006. Screening of tropical fruit volatile compounds using solid-phase microextraction (SPME) fibers and internally cooled SPME fiber. J Agr ic Food Chem 54(23):868896. Carlton -Tohill B. 2007. Fruits and vegetables consumption and bod y weight management. ISHS Acta h ortic (744):39 46. Chan HT, Cavaletto CG. 1982. Aseptically packaged papaya and guava puree: changes in chemical and sensory quality during processing and storage. J Food Sci 47(4):11649. Chan HT, Brekke JE Chang T. 1971. Nonvolatile organics acids in guava. J Food Sci 36(2):237-9. Chan HT, Kwok SCM. 1975. Identification and determination of sugars in some tropical fruit produc ts. J Food Sci 40(2):419-20. Cheynier V. 2005. Polyphenols in foods are more complex than often thought. Am J Clin Nutr 81(1):223S 9. Ching LS, Mohamed S. 2001. Alpha-tocopherol content in 62 edible tropical plants. J Agric Food Chem 49(6):3101-5. Chopd a CA, Barrett DM. 2001. Optimization of guava juice and powder production. J Food Process Pres 25(6):41130. Christen GL & Smith JS. 2000. Food chemistry: principles and applications. West Sacramento, CA: Science Technology System. Chyau CC, Chen SY, Wu CM. 1992. Differences of volatile and nonvolatile constituents between mature and ripe guava ( Psidium guajava Linn) fruits. J Agric Food Chem 40(5): 846-9. Clark P. 2009. Case studies in food engineering. New York: Spr inger. 224 p. Cook NC, Samman S. 1996. Flavonoids -Chemistry, metabolism, cardioprotective effects, and dietary sources. J Nutr Biochem 7(2):66-76. Commodity market review: 2003 2004 [Internet]. Food and Agriculture Organization; 2003 [Accessed 2010 Jan 3] Available from: http://www.fao.org/documents/

PAGE 197

197 Craig WJ. 1997. Phytochemicals: guardians of our health. J Am Diet Assoc 97(10, Supplement 1):S199-204. Current s ituation and me dium outlook for tropical fruits [Internet] Food and Agriculture Organization; 2005 [Accessed 2010 J an 3] Available from: http://www.fao.org/es/esc/common/ecg/218/en/Sit_web_e.pdf Dagan GF, Balaban MO. 2006. Pasteurization of beer by a continuous dense -phase CO2 system. J Food Science 71(3):E164 -9. Damar S. 2006. Processing of coconut water with high pressure carbon dioxide technology. [DPhil dissertation]. Gainesville FL : University of Florida. 175 p Damar S, Balaban MO. 2006. Review of dense phase CO2 technology: microbial and enzyme inactivation, and effects on food quality. J Food Science 71(1):R1 -11. Damar S, Balaban MO, Sims CA. 2009. Continuous dense phase CO2 processing of a coconut water beverage. I nt J F ood Sci T ech 44(4):66673. Damodaran S, Parkin KL, Fennema OR. 2008. Fennema's food chemistry. 4th ed. Boca Raton FL: CRC Press 1 144 p Del Pozo Insfran D, Balaban MO, Talcott ST. 2006. Microbial stability, phytochemical retention, and organoleptic attributes of dense phase CO2 processed muscadine grape juice. J Agric Food Chem 54(15):5468-73. Demole E, Enggist P & Ohloff G. 1982. 1 p Menthene8 -thiol: A powerful flavor impact constituent of grapefruit juice ( Citrus parodisi ). Helv Chim Acta 65(6):178594. Dietary guidelines for Americans: US Department of Health and Human Services Washington DC: US Goverment Printing Office ; 200 5. [Accessed 2010 Jan 3]. Available from : www.health.gov/dietaryguidelines Dixon RA, Xie D Y, Sharma SB. 2005. Proanthocyanidins: a final frontier in flavonoid research? New Phytol 165(1):9 -28. Dweck AC. 2005. A Review of guava ( Psidium guajava) Personal Care Magazine. p. 3336. ElBulk RE, Babiker EFE, ElTinay AH. 1997. Changes in chemical composition of guava fruits during development and ripening. Food Chem 59(3):395 -9. ETropical Fruit Net : International Tropical Fruit Network (TFNET). 2009 [Accessed 2010 Jan 3] Available from: http://www.itfnet.org/e -Newsletter/July2009/

PAGE 198

198 Fender L. 2005. Phytochemical, antioxidant, and storage stability of thermally processed guava (Psidium guajava) and guava juice blends. [MSc thesis] Gainesville FL : Univ ersity of Florida. 79 p. Ferrentino G, Barl etta D, Balaban MO, Ferrari G, Poletto M. 2010. Measurement and prediction of CO2 solubility in sodium phosphate monobasic solutions for food treatment with high pressure carbon dioxide. J Super Crit Fluid 52(1): 142-50. Ferrentino G, Plaza ML, Ramirez -Rodrigues M, Ferrari G, Balaban MO. 2009. Effects of dense phase carbon dioxide pasteurization on the physical and quality attributes of a red grapefruit Juice. J Food Sci 74(6):E33341. Flavor Database. Lake Alfred, Fl: University of Florida, Citrus Research and Education Center, Rouseff R ; c2 0062009 [Accesed 2009 Oct 20]. Available from http://www.crec.ifas.ufl.edu/crec_websites/Rouseff/#. Flavornet and human odor space [Internet]. Geneva, NY: Cornell University Acree T & Arn H ; 2004 [Accessed 2009 Oct 20]. Available from: http://www.flavornet.org Fisher C, Scott TR. 1997. Food flavours: biology and chemistry. Cambridge, UK: The Royal Society of Chemistry. 165 p. Gary RB. 1999. Phytonutrients role in me tabolism: effects on resistance to degenerative processes. Nutr Rev 57(9):3-6. Gkmen V, Kahraman N, Demir N, Acar J. 2000. Enzymatically validated liquid chromatographic method for the determination of ascorbic and dehydroascorbic acids in fruit and veg etables. J Chromatogr A 881(1-2):30916. Gomes MRA, Ledward DA. 1996. Effect of highpressure treatment on the activity of some polyphenoloxidases. Food Chem 56(1):15. Gorinstein S, Zemser M, Haruenkit R, Chuthakor n R, Grauer F, Martin-Belloso O, Trakhtenberg S. 1999. Comparative content of total polyphenols and dietary fiber in tropical fruits and persimmon. J Nutr Biochem 10(6):36771. Grundh fer P, Niemet R, Schilling G, Gross GG. 2001. Biosynthesis and subcellular distribution of hydrolyzable tan nins. Phytochemistry 57(6):915-2 7. Gunes G, Blum LK, Hotchkiss JH. 2005. Inactivation of yeasts in grape juice using a continuous dense phase carbon dioxide processing system. J Sci Food Agr 85(14):2362-8.

PAGE 199

199 Hawaii guavas : National Agricultural Stati stics Services (NASS) [Internet]. 2009 [Accessed 2010 Jan 3]. Available from: http://www.nass.usda.gov/Statistics_by_State/Hawaii/Publications/Fruits_and_N uts/guava.pdf Ho C -T, Lee CY, Huang M -T. 1992. Phenolic compounds in food and their effects on heal th I: analysis, ocurrence & chemistry. Washington, DC: American Chemical Society. 338 p. Hodgson AS, Chan HT, C avaletto CG, Perera CO. 1990. Physical -chemical characteristics of partially clarified guava juice and concentrate. J Food Sci 55(6):17578. H uang D, Boxin O, Prior RL. 2005. The chemistry behind antioxidant capacity assays. J Agric Food Chem 53:184156. Huang M -T, Ho C -T, Lee CY. 1992. Phenolic compounds in food and their effects on health II: antioxidants and cancer prevention. Washington, DC : American Chemical Society. 338 p. Idstein H, Schreier P. 1985. Volatile constituents from guava ( Psidium guajava L.) fruit. J Agric Food Chem 33(1):13843. Imungi JK, Scheffeldt P, Saint -Hislare P. 1980. Physical-chemical changes during extraction and contraction of clear guava juice. Lebensmit. Wiss Technol 13:24851. Janes D, Kantar D, Kreft S, Prosen H. 2009. Identification of buckwheat (Fagopyrum esculentum Moench) aroma compounds with GC -MS. Food Chem 112(1):1204. Jennings W, Shibamoto T. 1980. Qualitative analysis of flavor and fragance volatiles by glass capillary gas chromatography. New York, NY : Academic Press. 472 p. Jimenez -Escrig A, Rincon M, Pulido R, Saura-Calixto F. 2001. Guava fruit ( Psidium guajava L.) as a new source of antioxidant dietary fiber. J Agric Food Chem 49(11):5489-93. Jordan MJ, Margaria CA, Shaw PE, Goodner KL. 2003. Volatile components and aroma active compounds in aqueous essence and fresh pink guava fruit puree ( Psidium guajava L.) by GC -MS and multidimensional GC/GC -O. J Agric Food Chem 51(5):142126. K ashyap DR, Vohra PK, Chopra S, Tewari R. 2001. Applications of pectinases in the commercial sector: a review. Bioresource Technol 77(3):21527.

PAGE 200

200 Kataoka H, Lord HL, Pawliszyn J. 200 0. Applications of solidphase microextraction in food analysis. J Chromatogr A 880(12):3562. Khanbabaee K, Ree Tv. 2001. Tannins: classification and definition. Nat Prod Rep 18:6419. Kincal D, Hill WS, Balaban M Portier KM, Sims CA, Wei CI, Marshal l MR. 2006. A Continuous high pressure carbon dioxide system for cloud and quality retention in orange juice. J Food Science 71(6):C33844. Kincal D, Hill WS, Balaban MO, Portier KM Wei CI, Marshall MR. 2005. A Continuous high pressure carbon dioxide sys tem for microbial reduction in orange juice. J Food Science 70(5):M249 -54. Kondo S, Kittikorn M, Kanlayanarat S. 2005. Preharvest antioxidant activities of tropical fruit and the effect of low temperature storage on antioxidants and jasmonates. Postharves t Biol Tec 36(3):309 -18. Laura B. 1998. Polyphenols: chemistry, dietary sources, metabolism, and nutritional significance. Nutr Rev 56(11):317-33. Lecky M. 2005. Continuous high pressure carbon dioxide processing of watermelon juice [MSc thesis] Gainesv ille FL : University of Florida. 109 p. Leong LP, Shui G. 2002. An investigation of antioxidant capacity of fruits in Singapore markets. Food Chem 76(1):69-75. Lim S, Yagiz Y, Balaban MO. 2006. Continuous high pessure carbon dioxide processing of manadar in juice. Food Sci Biotech 15(1):138. Lord H & Pawliszyn J. 2000. Evolution of solidphase microextraction technology. J Chromatogr A 885(12):153 -93. Macleod AJ & Gonzalez de Troconis N. 1982. Volatile flavour components of guava. Phytochemistry 21(6):13391342. Mahattanatawee K, Goodner KL, Baldwin EA. 2005. Volatile constituents and character impact compounds of selected Florida's tropical fruit Proc. Fla. State Hort. Soc 118:414 -8. Mathews Roth MM, Krinsky NI. 1985. Carotenoid dose level and protection against UV B induced skin tumors Photochem Photobiol 42(1):35-8. Meilgaard M, Civille GV, Carr BT. 2007. Sensory evaluation techniques 4 th ed. Boca Raton, FL: Taylor & Francis Group. 448 p.

PAGE 201

201 Mercadante AZ, Steck A, Pfander H. 1998. Carotenoids from guava ( Psidium guajava L.): isolation and structure elucidation. J Agric Food Chem 47(1):14551. Mercado -Silva E, Benito -Bautista P, Garca -Velasco M A. 1998. Fruit development, harvest index and ripening changes of guavas produced in central Mexico. Postharvest Biol Tec 13(2):143-50. Miean KH, Mohamed S. 2001. Flavonoid (myricetin, quercetin, kaempferol, luteolin, and apigenin) content of edible tropical plants. J Agric Food Chem 49(6):310612. Misra K, Seshadri TR. 1968. Chemical components of the fruits of Psidium guava Phytochemistry 7(4):6415. Mitra S. 1997. Postharvest physiology and storage of tropical and subtropical fruits. 1st ed. New York, NY: C ab International 423 p. Morton J. 1987. Fruits of warm climates. 1st ed. Miami, FL: Julia Morton. 48 p. Mowlah G, Itoo S. 1982. Guava ( Psidium guajava L.) sugar components and related enzymes at stages of fruit development and ripening. Nippon Shokuhin Kogyo Gakkaishi (J Jpn Soc Food Sci Technol) 29(8):472-6. Mowlah G, Itoo S. 1983. Changes in pectic components, ascorbic acid, pectic enzymes and cellulase activity in ripening and stored guava ( Psidium guajava L.). Nippon Shokuhin Kogyo Gakkaishi (J Jpn Soc Food Sci Technol) 30(8):45461. Murano PS. 2003. Understanding food science and technology. Belmont, CA: Wadsworth/Thomson Learning. 449 p. Naczk M, Shahidi F. 2004. Extraction and analysis of phenolics in food. J Chromatogr A 1054(1-2):95 -111. Nakasone HY, Paull RE. 1998. Tropical fruits. New York, NY : C ab I nternational 472 p. Nielsen SS. 2003. Food analysis. 3 rd ed. New York, NY: Klu wer Academic/ Plenum Publishers. 557 p. Nishimura O, Yamaguchi K Mihara S & Shibamoto T. 1989. Volatile constituents of guava fruits (Psidium guajava L.) and canned puree. J Agric Food Chem 37(1):139-142. Nunez Rueda F. 2005. Guava ( Psidium guajava L.) fruit phytochemicals, antioxidant properties and overall quality as influenced by postharvest treatments [MSc thesis] Gainesville FL : University of Florida. 89 p.

PAGE 202

202 Paniandy JC, Chane Ming J, Pieribattesti JC. 2000. Chemical composition of the essential oil and headspace solidphase microextraction of the guava fruit (Psidium guajava L.). J Essen Oil Res 12(2):1538. Pietta P -G. 2000. Flavonoids as antioxidants. J Nat Prod 63(7):103542. Rahmat A, Bakar MFA, Hambali Z. 2006. The effects of guava ( Psidium guajava ) consumption on total antioxidant and lipid profile in normal male youth. Afr J Food Agric Nut Develop 6(2):112. Randall GC, Robert AB, Karel G. 1998. Mul tiple forms of pectinmethylesterase from citrus peel and their effects on juice cloud stability. J Food Sci 63(2):2536. Raso J, BarbosaCanovas GV. 2003. Nonthermal preservation of foods using combined processing techniques. Crit Rev Food Sci 43(3):265 85. Redd J, Hendrix C, Hendrix D. 1986. Quality control manual for citrus processing plants. Safety Harbor, FL: Intercit Inc. 180 p. Reineccius G. 2006. Flavor chemistry and technology 2nd ed. Boca Raton, FL: Taylor and Francis. 489 p. Robbins RJ. 20 03. Phenolic acids in foods: an overview of analytical methodology. J Agric Food Chem 51(10):286687. Rouse AH, Atkins CD. 1955. Pectinestersase and pectin in commercial orange juice as determined by methods used at the Citrus Experiment Station. Bulletin Agric Exp Station (Florida) 570:1-19. Rous eff RL, Onagbola EO, Smoot JM, Stelinski LL. 2008. Sulfur volatiles in guava (Psidium guajava L.) leaves: possible defense mechanism. J Agric Food Chem 56(19):8905-10. Rowe D. 2000. More fizz for your buck highimpact aroma chemicals. Perfumer & Flavorist Magazine. p. 1-19. Salunkhe DK, Kadam SS. 1995. Handbook of fruit science and technology: production, composition storage and processing. New York, NY : Marcel Dekker Inc 611 p. Sandhu K, Bhatia BS. 1985. Physico -chemical changes during preparation of fruit juice concentrate. J Food Sci Tech 22:202-6. Saura -Calixto F. 1998. Antioxidant dietary fiber product: a new concept and a potential food ingredient. J Agric Food Chem 46(10):43036.

PAGE 203

203 Scalbert A, Johnson IT, Saltmarsh M. 2005. Polyphenols: antioxidants and beyond. Am J Clin Nutr 81(1):215S -7. Shahidi F, Naczk M. 2003. Phenolics in food and nutraceuticals. Boca Raton, FL: CRC Press. 558 p. Sonko L, Xia W. 2005. Food phenolics, pros and cons: a rev iew. Food Rev Int 2005(4):367-88. Spanier AM, Shahidi F, Parliment TH, Mussinan C, Ho C -T, Contis UT. 2001. Food flavors and chemistry: advancements of the new millennium. Cambridge, UK: The Royal Society of Chemistry. 666 p. Spilimbergo S, Bertucco A. 2 003. Nonthermal bacterial inactivation with dense CO2. Biotech nol Bioeng 84(6):627-38. Steinhaus M, Sinuco D, Polster J, Osorio C, Schieberle P. 2008. Characterization of the aromaactive compounds in pink guava ( Psidium guajava L.) by application of th e aroma extract dilution analysis. J Agric Food Chem 56(11):41207. Steinhaus M, Sinuco D, Polster J, Osorio C, Schieberle P. 2009. Characterization of the key aroma compounds in pink guava ( Psidium guajava L.) by means of aroma re engineering experiments and omission tests. J Agric Food Chem 57(7):28828. Steven KC. 1998. Lycopene: chemistry, biology, and implications for human health and disease. Nutr Rev 56(2):3551. Stevens KL, Brekke JE, Stern DJ. 1970. Volatile constituents in guava. J Agric Food C hem 18(4):598-9. Sun N, Lee S, Song KB. 2002. Effect of highpressure treatment on the molecular properties of mushroom polyphenoloxidase. Lebensmittel Wissenschaft und Technologie 35(4):315 -318. Supelco. 2005. Solid phase microextraction: theory and optimization of conditions. Supelco Bulletin 923. Talcott ST, Howard LR, Brenes CH. 2000. Antioxidant changes and sensory properties of carrot puree processed with and without periderm tissue. J Agric Food Chem 48(4):131521. Temple NJ. 2000. Antioxidants and disease: more questions than answers. Nutr Res 20(3):449-59.

PAGE 204

204 The world fresh food market : Kipe Sandra [Internet] USDA Foreign Agricultural Service. Horticultural and Tropical Products Division. 2004 [Accessed 2010 Jan 3]. Available from http://www.fas.usda.gov/htp/Presentations/2004/World%20Fresh%20Fruit%20M arket%20%2808-04%29.pdf Tribess TB, Tadini CC. 2006. Inactivation kinetics of pectin methylesterase in orange juice as a function of pH an d temperature/time process conditions. J Sci Food Agr 86(9):132835. Vitamin E: National Institute of Health [Internet]. Office of Dietary Supplement ; 2009 [Updated 2009 Dec 15; Accessed 2010 Jan 3]. Available from: http://ods.od.nih.gov/factsheets/Vitami nE.asp Verste eg C, Rombouts FM, Spaansen C, Pilnik W. 1980. Thermostability and orange juice cloud destabilizing properties of multiple pectinesterases from orange. J Food Sci 45:969-72. Wilberg VC, Rodriguez -Amaya DB. 1995. HPLC quantitation of major c arotenoids of fresh and processed guava, mango and papaya. Lebensmittel Wissenschaft und Technologie 28(5):474 -80. Wilson CW, Shaw PE. 1978. Terpene hydrocarbons from Psidium guajava Phytochemistry 17(8):14356. Wilson CW, Shaw PE, Campbell CW. 1982. Determination of organic acids and sugars in guava ( Psidium -Guajava L) cultivars by highperformance liquidchromatography. J Sci Food Agr 33(8):777-80. Wrolstad RE, Acree TE, Decker EA, Penner MH, Reid DS, Schwa rtz SJ, Shoemaker CF, Smith DM, Sporns P. 2005. Handbook of food analytical chemistry Hoboken, New Jersey: John Wiley & Sons. 606 p. Yen GC, Lin H T. 1996. Comparison of high pressure treatment and thermal pasteurization effects on the quality and shelf life of guava puree. Internation al J Food Sci Tech 31(2):205-13. Yen GC, Lin H T. 1999. Changes in volatile flavor components of guava juice with high pressure treatment and heat processing and during storage. J Agric Food Chem 47(5):20827. Yen GC, Ling HT, Yang P. 1992. Changes in volatile flavor components of guava puree during processing and frozen storage. J Food Sci 57(3):67981. Zabetakis I, Leclerc D, Kajda P. 2000. The effect of h igh hydrostatic pressure on the strawberry anthocyanins. J Agric Food C hem 48(7):2749 -54.

PAGE 205

205 BIOGRAPHICAL SKETCH Maria de L. Plaza Delestre received h er b achelors degree from University of Puerto Rico, Mayaguez Campus majoring in b iology with a minor in m icrobiology While finishing her undergraduate carri er, she got very interested in f ood s cience. After obtaining her bachelor s de gree, she decided to start her m aster s d egree at the same University. She earned a Master of Science in food science and technology in 2002. After four years of experience as a research assistant and instructor at the Food Science and Technology program of the University of Puerto Rico, she was offered the opportunity to pursue her P h.D. d egree. In August 200 5 s he started the graduate program in the Food Science and Human Nutrition Department at the University of Florida. Under Dr. M a urice Marshall s supervision, Maria receive d her Ph.D. in food science and human n utrition from the University of Florida in the summer 2010 and graduated with a GPA of 4.0