Citation
Storage stability of aseptically packaged single strength orange juice and orange drinks

Material Information

Title:
Storage stability of aseptically packaged single strength orange juice and orange drinks
Alternate title:
Orange drinks
Creator:
Kacem, Bechir, 1948-
Publication Date:
Language:
English
Physical Description:
xiv, 143 leaves : ill. ; 28 cm.

Subjects

Subjects / Keywords:
Amino acids ( jstor )
Flavors ( jstor )
Food ( jstor )
Juices ( jstor )
Orange fruits ( jstor )
Orange juice ( jstor )
Oxidation ( jstor )
Oxygen ( jstor )
Pouches ( jstor )
Sugars ( jstor )
Dissertations, Academic -- Food Science and Human Nutrition -- UF
Food Science and Human Nutrition thesis Ph. D
Orange juice -- Packaging ( lcsh )
Orange juice -- Storage ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1986.
Bibliography:
Includes bibliographical references (leaves 131-142).
Additional Physical Form:
Also available online.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Bechir Kacem.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
029557835 ( ALEPH )
15376335 ( OCLC )

Downloads

This item has the following downloads:


Full Text









STORAGE STABILITY OF ASEPTICALLY PACKAGED
SINGLE STREfTH ORANGE JUICE AND ORAT~ DRINKS






BY

BECHIR KACEM


























A DISSERTATION PRESENI'ED O THE GRADUATE SCIECL
OF THE UNrIVE~SITY' OF FLORIFA IN
PARTIAL FTUFITIE>'Tf OF THE REQUIRE~-''S
FOR THE D 1PJEE OF DCiTOR OF PE ILSOPHY



URnIVERSITY OF FLORIDA

1986































IN M~FORY OF MY FATHER














ACCOWLEDGEMENS



The author wishes to express his sincere thanks to Dr. R.F.

Matthews, his major professor, for support and guidance which made

possible the completion of this work. Special gratitude is also

extended to Dr. M.R. Marshall, co-chairman, Dr. P.G. Crandall, Dr.

J.F. Gregory and Dr. R.B. Shireman, members of his supervisory

committee, for their continual encouragement and counsel throughout

the work. Special thanks go to Dr. J.A. Cornell, also a member of this

committee, for his helpful advice and assistance concerning the

statistical evaluation of the experimental results.

Appreciation is also extended to Mr. P. West for his technical

assistance throughout the months of laboratory work, to Ms. Virginia

Wily for her help with the amino acids analysis and to Ms. Robin

Adkins for printing the manuscript.

The author also acknowledges partial financial support for this

research from the Tunisia Agricultural Technology Transfer Project.

Last but not least, special gratitude must be credited to his

wife, Saida, for her love, dependability and understanding throughout

this work. To his mother, the author extends his deepest appreciation

for her faith and her patience during his absence from home.





iii














TABLE OF C3NTEiTS

PAGE

AC2 WLEDGC~EENTS .............................................

LIST OF TABL S .............................. ........ .. ..... vi

LIST OF FIGURES ............................................... ix

LIST OF ABBREV ATIONS ......................................... xii

ABSTRACT ........................... ............ ........... . xii

I TRODUCTION .............................................. 1

LITERATURE REVIEW ........ ....... ....... .. .. .... 3

A Brief History of Citrus Distribution .................. 3
Citrus Production in Florida ............................ 3
Citrus Demand ........................................... 9
Citrus Processing ....................................... 10
Nation l ............................................... 10
Florida ........................................ ...... 10
Juice Packaging ................. ............ ..... ... 11
Aseptic Processing ............. .......... ...... ... ... 14
Nonenzymatic Browning ................................. 15
Maillard Reaction ...................................... 17
Active-Aldehyde Theory ............................... 21
Ascorbic Acid Theory ................................. 23
Factors Affecting the Browning of Packaged Orange Juice.. 24
Role of Ascorbic Acid ............................... 24
Role of Nitrogenous Compounds ........................ 41
Role of Sugars .. ....... ..... ... ...... ............. 43
Role of Container Type ............................... 45

i AR SUDY ......................... ................... L

SZSJLTA S ANALYSIS OF ASCCRBIC AND DEkTYDRASCOCBIC ACIDS BY
CHIGH PERFORMANCE LIQUID *C-ROMATCGRAPHY WITH POST-COLUMN
D:EPVATTzATICN AD Uv ABSORc r .............................. 49

Introduction ........................................... 49
Materials and Methods ............. .................... .... 51
Reagents ....................... ........... ........ 51
Apparatus ..... ........ .. .... .................... 52


iv










Post-column Derivatization .................. ........ 52
Sample Preparation .................................. 54
Recovery Study ....................................... 54
Calibration Curves ................................... 55
Results and Discussion .................................. 55

EFFECT CF ASCORBIC ACID AR) AMrIO ACIDS CCETRATIS CN
OQALITY OF ASEPTICALLY PACKAGED OPRAGE DRINTKS ................ 65

Introductin ........................... ..... ................. 65
Materials and Methods ................................... 66
Reagents .............................................66
Orange Drinks Composition ............................ 66
Preparation of the Mixtures ......................... 68
Method of Analysis ...................................... 70
Ascorbic and Dehydroascorbic Acids Deterrmination ..... 70
Browning Measurement ................................. 70
Statistical Treatments .............................. 71
Results and Discussion .................................. 7
Ascorbic Acid Retention .............................. 71
Dehydroascorbic Acid Production .................... 74
Browning as Influenced by Ascorbic Acid Concentration 80
Browning as Influenced by Amino Acids Concentration .. 81
Statistical Analysis ............. ............... 90
Conclusion .............................................. 93

EFFECT OF AMqNO ACIDS CONCENTRATION, PROCESSING, AND STORAGE
CONDITIONS ON THE QUALITY OF ASEPTICALLY PACKAGED ORANGE JUICE
AND ORANGE DRINKS ............................................. 94

Introduction ............................................ 94
Materials and Methods ....................... ............ 94
Reagents .................................... ...... 94
Orange Drinks Composition ................. ........ 95
Preparation of the Samples ........................... 96
Methods of Analysis ...................... ............... 99
Ascorbic and Dehydroascorbic Acids ................... 99
Browning ......................... ... ....... .... .... 99
Amino Acids Analysis ................................. 99
Sensory Evaluation ................. ...... ..... 100
Statistical Analysis ................................. 100
Results and Discussion ............................ 100
Orange Juice ............... ....................... 00
Orange Drinks ........................................ 107
Conclusion ................................. ......... .. 115

SUMMARY ...................... .... ......................... 116

APPENDIX ............................... .. ... ................ '.8


V












P FERENE S .................................................... 131

BIOGRAPHICAL SKETCH ........................................... 143

















































vi










LIST OF TABLES

PAGE

1 Principal Citrus Fruits: Production for h United
States and Florida, Crop Years 1964-65 trough 1983-84.. 7

2 Oranges: Production for the United States and Florida.
Crop Years 1960-61 through 1983-P ................... 7

3 U.S. Citrus Per Caita Consumotion ...................... 9

4 Florida Oranges: Production, Utilization and Value
for Crop Years 1954-65 through 1983-84 .................. 12

5 Permeability of the Packaging Film ...................... 47

6 Comparison of the Fluorometric and the Dye Titration
Procedures for Ascoric Acid Determination .............. 48

7 Ascorbic Acid and Dehydroascorbic Acid Content
of Various Foods .......................................... 56

8 Recovery of Ascorbic Acid and Dehydroascorbic
Acid from Spiked Samples ................................ 59

9 Composition of the Orange Drink Mixtures ................. 68

10 Initial Ascorbic Acid and Dehydroascorbic Acid Levels for
Orange Juice and Orange Drink Products in Retort Pouch
and Polyethylene Pouch ................................... 72

11 Three-Factor Analysis of Variance of the Browning Data
at 8 Weeks ...................... ...................... 90

12 Composition of Orange Drinks .............................. 95

13 Processing and Storage Conditions of Orange Juice and
Orange Drinks ........................................... 98

14 Rate Constants (Weeks-") for Ascorbic Acid Loss as a
Function of Amino Acid Content and Scoraae Conditions
in Orange Drinks ................ ...... ..... ... ... 110

15 Changes in AmJont of A.--no Acids in Orange Drinks M2 an
M3 as Influenced by S torac Time ....................... 11.3

A-L Absorbance at 420 nm as a Function of Storage Time
(samles stored in retort pouch) ......................... 121



vii









A-2 Absorbance at 420 nm as a Function of Storage Time
(samples stored in polyethylene pouch) ................... 122

A-3 Ascorbic Acid Concentration as a Function of Storage Time
(samples stored in retort pouch) ....................... 123

A-4 Ascorbic Acid Concentration as a Function of Storage Time
(samples stored in polyethylene pouch) ................... 124

A-5 Dehyroascorbic Acid Concentration as a Function of Storage
Tine (samples stored in retort pouch) ................ 125

A-6 Dehydroascorbic Acid Concentration as a Function of Storage
Time (samples stored in polyethylene pouch) ............. 126

A-7 Absorbance at L20 nm as a Function of Storage Time
(samples stored in Tetra Pak carton) ..................... 127

A-8 Ascorbic Acid Retention as a Function of Storage Time
(samples stored in Tetra Pak carton) ................... 128

A-9 Flavor Score as a Function of Storage Time
(samples stored in Tetra Pak carton) .................... 129

A-10 Sensory Evaluation Form ............................... 130


























viii














LIST OF FIGURES

PAGE

1 Florida Citrus Production ......... .. ........................ 4

2 Principal Citrus Fruits: Production for United States
and Florida, Crop Years 1960-61 through 1983-84................ 6

3 Oranges: Production for the United States and Florida,
Crop Years 1960-61 through 1983-34 ............................ 8

4 Initial Stage of Carbonyl-Amine Reactions .................... 18

5 Amadori Rearrangement ....................................... 19

6 Major Pathway for Carbonyl-Amine Reactions ................... 20

7 Minor Pathway for Carbonyl-Amine Reactions ................... 21

8 Initial Stage for Degradation of Ascorbic Acid ............... 25

9 Degradation Pathway for Diketogulonic Acid ................... 26

10 HPLC System with Post-Column Derivatization and Tandem
Ultraviolet and Fluorometric Detection ....................... 53

11 Typical HPLC Chromatograms of Orange Juice and Parsley,
Monitored by Tandem Ultraviolet (uv, 254 nm) and
Fluorometric Detection ..................................... 57

12 Typical HPLC Chromatograms of Tomato and Strawberry
Monitored by Tandem Ultraviolet (uv,254 nm) and
Fluorometric Detection ..................................... 58

13 Resolution of AA, DHAA, and DKGA by HPLC ..................... 62

14 Ascorbic and Dehvyroascorbic Acids Monitored
at 210 and 254 ........................................ 53

15 Absorption Specr-ru for AA, DEAA, and DKGA ................... 6

16 Effect of Anino Acid Concentration on Ascorbic Acid Retention
(initial level of ascorbic acid 71.8 mg/100 ml, sazples
stored in polyethylene pouh) ................... ............. 76



ix










17 Effect of Amino Acid Concentration on Ascorbic Acid Retention
(initial level of ascorbic acid 71.8 mg/100 ml, sanples
stored in retort pouch) ................................... 77

18 Effect of Amino Acid Concentration on Ascorbic Acid Retention
(initial level of ascorbic acid 38.0 mg/100 ml, samples
stored in retort pouch) ..................................... 7

19 Effect of Amino Acid Concentration on Ascorbic Acid Retention
(initial level of ascorbic acid 4.2 mg/100 ml, saples
stored in retort pouch) ...................................... 79

20 Effect of Ascorbic Acid Concentration on Browning
of Orange Drinks with 0.06% Amino Acids ...................... 82

21 Effect of Ascorbic Acid Concentration on Browning
of Orange Drinks with 0.66% Amino Acids ..................... 83

22 Effect of Ascorbic Acid Concentration on Browning
of Orange Drinks with 1.267% Amino Acids ...................... 84

23 Effect of Amino Acid Concentration on Browning of
Orange Drinks (4.2 mg/100 ml AA) ............................. 86

24 Effect of Amino Acid Concentration on Browning of
Orange Drinks (38.0 mg/100 ml AA) ............................ 88

25 Effect of Amino Acid Concentration on Browning of
Orange Drinks at 71.8 mg/100 ml AA .......................... 89

26 Effect of Ascorbic Acid and Amino Acid Concentration on
Browning of Orange Drinks at 8 Week Storage ................. 92

27 Ascorbic Acid and Dehydroascorbic Acid Concentration in Orange
Juice as Influenced by Processing and Storage Conditions .... 101

28 Ascorbic Acid Retention (Log Scale) in Orange Juice as
Influenced by Processing and Storage Conditions ............. 102

29 Browning Formation in Orange Juice as Influenced by
Processing and Storage Conditions ........................... 05

30 Flavor of Orange Juice as Influenced by Processing
and Storage Conditions .................................... 106

31 Ascorbic Acid and Dehydroascorbic Acid Concentration of Orange
Drinks as Influenced by Amino Acid Content and Storage
Conditions ................ ... ................... . . . . .... 108

32 Ascorbic Acid Retention (Log Scale) in Orange Drinks as
Influenced by Amino Acid Content and Storage Conditions ..... 109

x










33 Browning Formation in Orange Drinks as Influenced by Amino
Acid Content and Storage Conditions ........................ 112

34 Flavor of Orange Drinks as Influenced by Amino
Acid Content and Storage Conditions ............... ... 114

A-1 Effect of Amino Acid Concentration on Ascorbic Acid Retention
(initial level of ascorbic acid 30.0 mg/100 ml, samples
stored in polyethylene pouch .............................. 119

A-2 Effect of Amino Acid Concentration on Ascorbic Acid Retention
(initial level of ascorbic acid 4.2 mg/100 ml, samples
stored in polyethylene pouch .................. ............ 120






































xi














LIST OF A3BBOEIAC S



AA ascorbic acid

AER aerobic

ANA anaerobic

ANOVA analysis of variance

aw water activity

Cpd compound

DHAA dehydroascorbic acid

DKGA diketogulonic acid

DP 3-deoxy-L-pentosone

FA furoic acid

FDA Food and Drugs Administration

HFCS high fructose corn syrup

HPLC high performance liquid chromatography

MMT million metric tons

MT million tons

OJD orange juice deaerated

OJND orange juice nondeaerated

OPDA orthophenylenediamine

RDA recoamended daily allowances

SD standard deviaticn

SSOJ single strength orange juice



xii












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

STOPAGE STABILITY OF ASEPTICALLY PACKAGED
SINGLE STREGTIH ORAIGE JUICE AND ORAINE DRINKS

By

Bechir Kacem

August 1986

Chairman: R.F. Matthews
Cochairman: M.R. Marshall
Major Department: Food Science and Human Nutrition

Nonenzymatic browning of single strength orange juice, and

synthetic orange drinks containing 10% orange juice and various

combinations of ascorbic acid, amino acids (aspartic acid, arginine,

and 4-aminobutyric acid), sugars, citric acid and potassium citrate

has been studied under aerobic and anaerobic conditions. The juice

and the drinks were aseptically packaged and stored at 750F for up to

20 weeks. In the presence of oxygen, ascorbic acid was found to be

the most reactive constituent in the darkening of orange drinks. The

presence of amino acids at the high level (1.26%) increased

significantly the rate of ascorbic acid degradation and the rate and

extent of browning pigments formation. However, reducing the amino

acids evel from 0.66 to 0.06% had no significant effect on the

browning of crange drinks stored under anaerobic conditions, but did

significantly affect browning when samples were stored aerobically.

The study also showed that the storage of aseptically packaged orange



xiii










juice in anaerobic jars, as compared to aerobic storage, resulted in

higher ascorbic acid retention. However, there was no significant

difference in the sensor, evaluation, nor in t)e amino acids

concentrations after 16 week storage at 75cF.

In addition, a high-perfornance liquid chromatography (:TLC)

procedure has been developed for the rapid and simultaneous estimation

of ascorbic and dehydroasccrbic acids in fresh fruits and vegetables.

Isocratic separation of these components was accomplished by

anion-exchange chromatography using acetonitrile:0.05M KH2P04 (75:25,

v/v) as eluant. The concentration of ascorbic acid was determined by

monitoring its absorbance at 254 nm, while dehydroascorbic acid

detection was achieved by fluorescence as a result of post-column

derivatization involving the condensation of dehydroascorbic acid with

o-phenylenediamine, forming a highly fluorescent quinoxaline

derivative. The procedure allows detection of both forms of vitamin C

at levels well below those usually found in orange juice, and was used

to follow the rate of change of ascorbic acid into dehydroascorbic

acid in orange juice and orange drinks during storage.
















xiv














INIRODUJCTION



Citrus products during processing and storage at room temperature

are susceptible to a number of deteriorative reactions, which result

in the development of off-flavor. Such off-flavor is generally

accompanied by other changes, in particular, browning of the product

and loss of nutritional value. This type of discoloration called

nonenzymatic browning is one of the most disturbing problems for

citrus industries. It is the main reason for the reduction in

commercial value and consumer rejection of citrus products, and has

been the subject of research for many years. Many different types of

reactions lead to the discoloration of the product at moderate

temperatures. This change in color may occur through formation of

dark pigments by the breaking down of certain constituents such as

ascorbic acid, by reaction between some constituents present in the

juice product or by reaction between some constituents of the juice

with oxygen in the air.

Factors which can influence the nature of the degradation

mechanism include temperature, oxygen, amino acids, metal catalysts,

pH, and sugar concentration. Various hypotheses have been developed

to explain the mechanis4 of browing in citrs juices. Arong these,

ascorbic acid degradation is thought to be the major pathway

responsible for browning in citrus prod.ucts.



1






2



However, in spite of the many reviews of the subject, there has been

no comprehensive organization of the reactions involved. The work

reported is ccntradictory in nature, and there is still a lack of

unerstanding concernig the runamental factors involved in the

deterioration of packaged orange juice.

It is not known whether decomposition of ascorbic acid alone or

decomposition in the presence of sugars and amino acids of orange

juice is more important than the Maillard reaction in browning. A

clear mowledge of the reactions involved in this deterioration and

the roles played by the various constituents of the juice is necessary

before an entirely satisfactory method for controlling nonenzymatic

browning in citrus can be realized.

In view of this lack of understanding it was decided to carry out

this research project to study the roles of ascorbic acid

concentration, amino acids concentration, juice concentration, oxygen

concentration, storage and processing conditions, and storage time in

the browning of aseptically packaged orange juice and orange drinks

containing various combinations of ascorbic acid, amino acids, sugars,

citric acid and potassium citrate. In addition, the development of an

analytical procedure to follow the rate of change of ascorbic acid

into deh~ydroascorbic acid in orange juice and orange drinks during

stor.ae usin the H-LC -techni-ue s ivestigated.











LITERATURE PTJRVI



A Brief History of Citrus Distribution

Citrus trees first appeared in Chinese gardens during the 13th to

16th centuries. From this area the Arab traders propagated the fruits

progressively with their conquests of the Middle East, Near East,

North Africa and Spain. The fruits were introduced into North America

from the Canary Islands by Christopher Columbus during his second

voyage in 1493 (Cooper and Chapot, 1977). The exact date of the

introduction of citrus trees into Florida is not known. Ziegler and

Wolfe (1961) pointed out that oranges were likely brought into Florida

at the time the colony at St. Augustine was established in 1565.

Seeds were probably then scattered throughout the State by Indians.

Citrus Production in Florida

The first recorded citrus crop was noted in a report by the

Governor of St. Augustine in 1579, but commercial production and trade

did not begin to develop on a large scale until after the Civil War.

In 1886-87, the U.S.Department of Agriculture reported a total citrus

production of 1.26 million boxes (Figure 1). The "Great Freeze" of

1894-95 almost destroyed the entire citrus industry of the state. It

was nct until 1903-04, that this level was reached again. Since that

time the volume has steadily increased to reach a maximum of 283.6

millicn boxes In 1979-80 (Fla. Crop and Livestock Reporting Ser.

Citrus Sur =ary, 1985).


3












(1)

( )

SI () )
0. )I













(-) 4 J





( )



'U
i -.


(I)



I J






(H)
I II





) 0 (i (( (I

.-4 -. I-1






5


As the decade of the sixties began, Florida's dominance of the

nation's citrus became more apparent, and in the 1983-8-L season

Florida accounted for more than 69% of the U.S. production of citrus

fruits (Figure 2 and Table 1).
Even though the industry suffered crippling nat~ral disasters

ranging from winter freezes to canker, Florida's 1984-85 citrus crop

has been valued at a record $1.Y04 billion in a preliminary report

(Citrus Valued at Record $1.4 billion, 1985). Although the amount of

oranges harvested for 1984-85, 103.9 million boxes (4.6M), is the

smallest yield since 1967-68, higher prices pushed the value to the

record level. The total citrus harvest, including oranges,

grapefruit, lemons and limes was 158.9 million boxes (7Mff). The

1979-80 harvest held the previous record high of $ 1.03 billion,

according to the Florida Crop and Livestock Reporting Service of

Orlando. The latest report issued by the service shows a 23% increase

in value over the 1983-84 harvest of $ 849 million.

Orange production increased from about 2.5 million tons in

1963-64 to over 8 million tons in the 1975-76 season, and presented

78% of the U.S. total production (Figure 3 and Table 2).














Mi i l l 1 1 111 l



I l,


I;'. lUNITED) STATS/





3 \
F FLORIDA/








I 1lIi/6 I1 1 95/6i0 1970/71 1975/75 1J900/81 190 5/O30
SEASON


Figure 2. Principal Citrus Fruits: Production for the United States
and Florida, Crop Years 1960-61 through 1983-84






7






Table 1

Princiral Citrus Fruits: Produci-on for the United
Statcs and Florida, Crop Years 1964-65 through
1983-84 (1,000 Tons)


Year LUnited Scares Fiorida X


1964-65 7,633 5,480 71.8
1965-66 8,768 6,242 71.2
1970-71 11,919 83786 73.7
1975-76 14,788 10,943 74.0
1980-81 15,105 10,470 69.3
1982-83 13,608 8,513 62.6
1983-84 10,741 7,436 69.2

Source: Adapted from Fla. Crop and Livestock Reporting
Serv. (1985).



TABLE 2

Oranges: Production for United States and
Florida, Crop Years 1960-61 through 1983-84


Production: 1,000 tons
Crop year -------------------------------- %
United States Florida
I !
I I I
1963-64 3 3,732 2,469 66
1965-66 5 5,812 4,316 74
1970-71 8,205 6,402 78
1975-76 10,493 3 ,154 78
1980-81 ; 10,487 7,758 7
1982-83 : 9,519 6.282 65
1983-84 7,238 5,252 73

Source: Adacted from Fla. Cro and Livestoc.
Reoorting Service, (1985).















I I






FLORIDA/

",v/




i] --------------------' --------~- --- ______ ---- ----------- I -
I li ,liil I ()5/Ii( .. 1970/71 1975/75 1 l 0/01 1 i/0'. 1/
SEASHH

Figure 3. Oranges: Production for United States and Florida,
Crop Years 1960-61 through 1983-84






9



Citrus Demand

The per capita consumption has increased ,fro 22.2 pounds in 1920

to amost 118 ponds in 1980, more than a five- fld inc-ase (Table 3).



Table 3

U.S. Citrus Per Capita Consumtion.
Fresh Weight iquTialent (pounds)


It e i20 19i 0 1960 1980
Fresh. 52.1 30.7 7 26.
Processed -- 10.4 52.2 91.2
Total 22.2 62.5 82.9 117.5

a Excludes lemons and limes.
None reported.

There has been a major shift in the form of the product demand.

Per capita consumption of fresh citrus increased during the early

1900s until the mid-1940s. Since 1940, fresh per capita citrus

consumption has declined by about 50%, while processed consumption

increased nine-fold (Gunter, 1983). Introduction of canned juice in

the 1920's and frozen concentrate in the mid 40' are the major factors

contributing to the growth in processed citrus juice demand.






10



Citrus Processing

National

Before canned orange juice was introduced co rercially in 1929,

only 1% of the domestic crop was processed. By 1945-47, an average of

over 1.4 MMT (million metric tons) of oranges and tangerines were

processed with almost the entire volume going into juice related

products. This volume represented about 1/3 of the total domestic

crop which was reported at 4.6 MMff. By 1959, about 3.5 MMT of oranges

and tangerines were processed into juice products representing nearly

64% of the 5.6 MMT crop that year. The percent of processed citrus

fruits continued to increase and by 1983-84, processors used 67% of

the total citrus crop of 9.74 IMT. They used 74% of the orange

production, 53% of the grapefruit and 46% of the lemons compared with

76%, 47%, and 54% respectively, in 1982-83. (Citrus Fruits, 1984).

Florida

In Florida, citrus processing represents the largest food

processing industry, producing 73% of the total United States citrus

products (Fla. Crop and Livestock Reporting Service, 1985). More than

90% of the Florida's orange crop is processed with 80% utilized for

frozen concentrate orange juice, 12% for chilled juice, 3% for canned

single strength orange juice; 4% is sold in the fresh market and less

than 1% is used in products like sections, salads, and blends. The

equivalent of 231 million 450Brix gallons of concentrated juice were

produced during 1979-80 season (Fla. Crop and Livestock Reporting

Service, 1980). Table 4 illustrates the upward trend in processed

oranges compared to declining trends for fresh oranges in Florida.










Juice Packaging

Juice and juice drink products have been distributed in three

forms, shelf-stable products, chilled juices, and frozen

concentrates. Single strength citrus juice became available in the

early 1930' s. During the period from the early 1930's to World War

II, conercial flash pasteurization of juice was developed. This

allowed processed juices to retain soe of their natural fresh aroma

and flavor during heat processing. The availabilty of high quality

juices in the form of frozen juice concentrates occurred in the

1940's. Prior to the introduction of this new concentrating

technology to the United States, only a third of the domestic orange

crop was processed into juice. By the late 1940s, with the

availability of the improved technology for frozen concentrates,

nearly two-thirds of the orange crop was being processed into juice,

with the largest percent being used to produce high-quality, frozen

orange juice concentrate. The success of these products is attributed

to the fact they offer convenience, low cost and are available year

round.

However, until recently, the shelf-stable juices and juice drinks

have shown a lack of growth in overall consumtion compared to juices

that were distributed as frozen concentrate or in refrigerated forms.

This is because of the higher quality that these "cold" methods of

distribution could deliver to the ce and juice products.






12




Table 4

Florida Oranges: Production, Utilization and
Value for Crop Years 1964-65 through 1983-84


Utilization of Production
--------------------------- On-Tree Value
Crop year 1,000 boxes Proces- I of Production
---------------- -sed 1,000 dollars
STotal Fresh Processed '
I I I I I
1963-64 54,90 ; 11,94 42,96 78 243,935
1965-66 95,90 15,38 80,52 84 155,625
1970-71 142,30 13,96 128,34 90 208,146
1975-76 181,20 ; 11,73 169,47 94 321,449
1980-81 172,40 8,28 164,12 95 697,231
1981-82 125,80 7,62 118,18 94 538,686
1982-83 I 139,60 10,32 129,28 93 718,420
1983-84 116,70 7,64 109,06 93 578,954

Source: Adapted from Fla. Crop and Livestock Reporting
Service, (1985).


The traditional method of processing a shelf-stable juice product

involves the hot filling of the juice in cans or glass bottles. This

method of processing has some difficulties if one wishes to achieve a

high-quality product. In this type of processing and packaging

operation, juices are commonly subjected to a severe heat treatment

for relatively long periods of time. The usual hot-filling process

involves heating and filling a rigid container above 170-180oF

(76.70-82.20C), followed by a cooling period which requires 10 to 20

minutes for the juice to reach ambient tenerature. The severity of

this type of processing has often decreased the overall quality of

juices processed in this manner. This method of processir- has also

discouraged the introduction of other quality-improving technologies.






13



Any pctrial in~rovement that could be made in the quality of the

juice delivered for bottling or canning was quickly lost due to the

severe beat treatment necessary for the hot-filling operation.

Another proble- affecting the quality of juice products packed in cans

is the interaction between the juice and the metal containers.

Problems such as the "pick up" of tin by juices in unlined containers

contribute to off-flavor of the consumed product. It is also a comon

practice to select the premium juices for use in chilled and frozen

products, and consider shelf-stable juice products as unaffected by

long periods of ambient storage, hence disregarding the inherent

degradative changes that occur in many fruit juice products when

stored at ambient temperature for long periods.

Until recently the most common package style used in processing

citrus shelf-stable juice products was the metal container. The tin

container guaranteed long shelf life and package integrity. The

second most important package, the glass bottle, has the same

benefits as the tin can. Both products do not require refrigeration

in distribution but the containers added significantly to total

freight costs due to their heavy tare weight. The energy crisis of

the mid 1970's has dramatically influenced this situation. Although

both cans and glass bottles are still used, the trend of the citrus

in-ustry is now ,oving toward low cost, light-weight, lmited shelf

life paczages. The most recent develcpment is the clearance of the

"etra Brk" package for use in aseptic packaging of licuid C roducts.






14



Aseptic Processing

Aseptic processing is a processing and packaging technique by

which a commercially sterilized product is put into a presterilized,

hermetically sealed container in a sterile environment. To produce a

cornerciaily sterile product, an aseptic system oust meet three basic

requirements:

-The product must be sterile.

-The package in which the product will be placed must be sterile.

-The environment in which the product and package will be brought

together must be sterile.

Commercial sterility is a term used to denote a product that has

been processed in such a manner so that it is free of all contaminants

that cause the product to spoil or be a detriment to public health.

Aseptic processing and packaging is not a new technology. The

first commercial system in the U.S. was installed in the 1950's for

use with non-acid products such as milk. However, since the recent

(February 9, 1981) FDA regulation permitting the use of hydrogen

peroxide and heat as sterilizing agents for aseptic packaging, the

interest in aseptic packaging has increased dramatically. This

technique is now applied in several food industries to produce

microbially stable products. By using this technology it is possible

to min"ize the ti the product is subjected to the bhgh tererature

necessary to obtain conercial sterility, and produce shelrf stble






15



juices that are equal in organoleptic quality to frozen and chilled

products, at a more economical price. It is estimated that 1-liter

aseptic boxes cost about half as much as cans and 30i as much as

bottles (Paper Bottles Are Coming on Stong, 1983). Although some of

these savings are surrendered because of the more complicated filling

process, it is estimated that the cost of packaging juice concentrate

in 8oz "Tetra Brik" boxes is 18o cheaper than filling coventional

paper and metal cans (Business Week, January 16, 1984), and since

aseptic goods are subjected to a briefer heating treatment during

sterilization than canned goods, they have better flavors. In

addition a major saving results because they do not require

refrigeration during shipment or storage.

However, aseptic processing will not be of commercial importance

until chemical stability of citrus products is understood and

improved. The limiting factor in the acceptance and eventual success

of aseptically packaged citrus juices is the control of chemical

changes that are accelerated during ambient temperature distribution

and storage. Nonenzymatic browning reaction is believed to be the

quickest and most dramatic quality defect to appear during ambient

temperature storage (Clegg, 1964; Tatum et al., 1967; and Tatum et

al., 1975).

Ncnen'z-atic Browning

The nonenzymatic browniin reaction in foods during processing or

on storage has long been recognized as one of the most important

problems of fruit preservation, and has been the subject of research

for many decades. This reaction refers to the formation of brown






16



pigments in foods that cause the product to become brownish to black

in appearance, and is usually characterized by undesirable charnes in

flavor, odor, and nutritive value.

In order to study the reaction under controlled conditions, model

systems have been widely used, but full details of the physical

conditions employed have not always been reported. The work done

before 1948 has been reviewed by Stadtman (1948). Since then there

have been excellent reviews by Hodge (1953), Reynolds (1963), B-uron

and MdJeeny (1964), Shaw et al. (1977), and Handwerk and Coleman

(1986) which mainly covered the reactions between amines and sugars.

Three main theories have been advanced to explain the mechanism

of nonenzymatic browning (Stadman, 1948).

The Maillard or melanoidin condensation theory: According to

this theory (the most common) the reaction involves a condensation of

amino acids and reducing sugars and gives rise to the formation of

dark colored substances.

The active-aldehyde theory: It proposes that browning involves

the decomposition of sugars and sugar acids to furfuraldehydes or

similar compounds characterized by having an active carbonyl group,

and that these products condense with nitrogen compounds and/or

polymerize to form colored substances.

The ascorbic acid theory: According to this theory the most

important precursors to browning are asccrbic acid and related

compounds, which upon oxidation yield reactive products that may

polymerize or react with nitrogenous constituents of the food to form

brown pigments. This third theory seems the most likely to apply to






17



the conditions pertaining to an acidic product such as orange juice;

the concentration of ascorbic acid is relatively high and free amino

acids are present to combine hwith th reactive products resultrin from

the oxidation of the ascorbic acid and lead to the formation of brown

pigments.

Actually all three of the above mechanisms may be involved in the

browning of fruit products. Major research efforts have been

conducted to prove the first two hypotheses of nonenzymatic browning,

as noted in the review by Shaw et al. (1977), but relatively little

research has exlored the ascorbic acid theory. It is not known

whether decomposition of ascorbic acid alone or in the presence of

sugars and amino acids of orange juice is more important than the

sugar-amino acids reaction in browning. Let us now look at these 3

theories in more detail.

Maillard Reaction

Maillard was the first to study systematically the interaction

which occurs initially between amino acids and sugars, and to realize

some of their relations to the chemistry of natual products. He found

that simple amino acids react on warming with certain sugars, to

produce dark-brown products. This explains why the reaction has also

been commonly called the browning reaction, nonenzymatic browning,

melanoidin forma-tion or caraeli.zation; and the brown products have

been referred to as melanoidins, or huiin-li'e sstances (5Elis

1959).

The interaction of Faino acids and sugars falls into two general

types. The first is the simple controlled condensation of the









reactants; this leads to coopounds which are identifiable as

N-substituted glycosylamines or, occasionally their Amadori

rearrangener procucts. The secnd is the typical Maillard reaction,

which leads o mixture of _products of in--creasing complexity if

conditions are Eeavorable.

Reynolds (1970) has outlined the major features of the initial

stages of the carbonyl-amine reaction. At the first step, an aldose

or ketose sugar reacts wirh a primary or secondary amine to form a

glycosylamine, and the reaction is reversible (Figure 4).



R-NH
I
H-C=O H-C=N-R H-C------,

(H-COH)4 + R- NH ----- (H-OH)4 ----- (Hc-CI)3 0
H-I I I
H-CHOH 1--.-- H-CHOR- --- H-C----

H-CHCH

glucose amino acid schiffbase glycosylamine


Figure 4. Initial Stage of the Carbonyl-Amine Reactions


The role of water is iportnt in deteinin the yield of

glycosylamine. At low water content t-re is substantial formation of
this comounc, (Shallnberger and Birch, 1 75); therefore,

car cny ;-aine brownng is belie ed to ibe a significnt pathway for

browning in dried and concentrat-e foods. The probable mechanism for

the formation of glycosylamine is the addition of the amine to the

carbonyl group of the open-chain form of the sugar, followed by






19


elimination of a molecule of water and subsequent ring closure

(Hodge, 1976).

The nex- step is the Amadori rearrangement which involves the

protonaticr of the nitrogen atom at C-i (Figure 5). This

rearrangement takes place in either acid or basic solutions.


+-r +
PMH RNH2 RRi
I I II
CH---- + H- CH---+ CH


I I I
(HCO1)3 O -------- (1COH)3 0 -------
HC------' HC--- ------ (HCOH )3

HCHOH HCHOH HCHOH

-H+

H2CNHR HCNHR
I 11
C= ------- COH

Amadori (HCOH)3 ------- (HCOH)3

compound H2COH H2COH


Figure 5. Amadori Rearrangement


The secornd stage in carbonyl-amine reactions is based on the

degradation a. d dehydration of the Aoadori compound, which can occur

thIough to miajcr parLwa-T The major branch leads from the

1,2-ene.e--'ol cf the Amadori coound to hydroxymethyl furfuraldehyde

(Figure 6). The amino acid may be retained in some molecules

throughout the dehydration reactions of this pathway.





20



This pa~tay is the major pathway for the production of brown color in

foods. The minor branch which probably represents less than 57. of the

total su-ar decomosition (Hodge and Osnan, 1976) begins with the

2,3-enedii of te- Amnadori copotund; then the amino acid is totally

eliminated (Figure 7). The degradation compounds formed seem to be

i~portant in the production of flavor.




+
H2C-N, HC-N HC=NI HC=O

C=O COH COH C=0
I I I I
(HCCH)3 -----. (HCOi)3 OH- (HCOH)3 + H20 Hf
I--- I
H2COHI <----- H2lI ------* (HCCH)2 -------- (HCOH)2
1 i
lmadori cpd. 1,2 eneaminol H2Cti H2CCUI




-H20


5 hydroxymethyl- HC=O

2- uraldehyde -H20 c=O

CH

a ^mi ne


melanoidins -H20 H2CCH



Figure 6. Major Pathway for Carbonyl-Amine Reactions






21





H2C-NN H2C-NN CH2

C=0 C CC C=0

-,C/3 --C---- -- ----- .C=0
4H9CCH (HOcH)-2 KCCE)2 -

H2COH H2 i VCGUI

Amadori cpd 2,3-enediol





melanoidins C

C=0

COH

COH

c-methyl reductones HCOH

and a-dicarbonyls H2CCH



Figure 7. Minor Pathway for Carbonyl-Amine Reactions



Active-Aldehyde Theory

When the nonenzymatic browning reactions occur in the absence of

nitrogenous compounds, they are described as caramelization

reactions. Under .an-drous conditions uoon the anpliz tion of heat,

or at high concentraton in dilute acid solutions, thea iitial stages

of the caranelizazicn reaction are characterized by the formation of

anhydro sugars. Glucose has been reported to yield glucosan and

levoglucosan. The two anhydroglucose comoounds are readly






22



distinguished by their specific rotation (Shallenberger and Birch,

1975). When sucrose is heated at about 200C, simultaneous hydrolysis

and dehydration occur, pparently followed by rapid dimerization of

the products, so that a series of compounds characterized by

isosacchrosan (corresponding to sucrose minus one molecule of water)

is formed (Hodge, 1976). Isosacchrosan has no sweetness, but does

have a mildly bitter taste.

In dilute solutions of reducirng sugars, the initial stages of the

caramelization, reactions are a series of events involving

enolization, dehydration and fragmentation reactions. Subsequently,

polymerization reactions occur, which lead to the formation of

pigments similar to those formed in caramelization reactions at either

higher temerature or in more concentrated solution.

The classic caramelizatin reaction is the phenomenon exibited by

sucrose when subjected to heat, and the fundamental reactions are

Inversion of sucrose to D-glucose and D-fructose

Equilibration of anomeric and ring forms

Condensation, intermolecular

Condensation, intramolecular

Isomerization of aldoses to ketoses

Fra-mentation reactions

7BrowinS

The caramelization reaction is autocatalytic and increasing

temperatures not only increases the reaccion rate but also alters the

qualitative nature of th Dpigr cs (Doss and Ghosh, 1949).






23



The effect of pH is dramatic and the reaction rate at pH 8 is ten

times greater than that at pH 5.9 (Ledon and Lananeta, 1950).

Caramelization reactions can be use ei fr coloring purpose,

or for flavoring purpose. For flavor, sucrose is caramelized i

concentrated syrups. The sugar fragmentation reactions are prcmoted

by alkaline or neutral medium, whereby formation of bhmic substances

is limited to avoid bitter, astringent tastes. Caramel for coloring

use is produced in acid medium. Glucose syrup is treated with dilute

sulfuric acid, partially neutralized with aconira, then heated in the

presence of a sulfite at a pH value of about 4 (Hodge and Osman, 1976).

Ascorbic Acid Theory

The exact route of ascorbic acid degradation is highly variable

and dependent upon the particular system (Tannenbaum, 1976). Three

different mechanisms of degradation of ascorbic acid have been proposed

Catalyzed aerobic degradation

Uncatalyzed aerobic degradation

Anaerobic degradation

When oxygen is present in the system, ascorbic acid is degraded

primarily via its monoanion (HA-) to dehydroascorbic acid (DHAA). The

exact pathway and overall rate is a function of the concentration of

metal catalysts (M') in the system. In the absence of oxygen the

degradation pathway has not been yet verified. Following the

suggestion of Kurata and Sakirai (1967), ascorbic acid is srhrwn to

react via its keto tautomer. HA-keto. The tautomer is in equil briu

with its anion, HA- keto, wh ich undergoes delactonization to

diketogulonic acid (DKGA), as shown in Figure 8. Further degradation





24



beyond EKGA is closely related to nonenzymatic browning in some food

products. The mechanism of these reactions has been studied by Kurata

and Skurai (1967) under oxidative and non-oxidative conditions. One

of the characteristic differences between the two pathways is that

furfural is much more easily produced through the non-_xidative

reaction. Under non-oxidative conditions, ascorbic acid in an acid

solution was degraded to furfural with the formation of

3-deoxy-L-pentosone as an intermediate. This acid-catalyzed

degradation took place under the storage or the cooking conditions of

foodstuffs. It was shown that aldcpentoses and 2-keto-L-gulonic acid

themselves were not intermediates of the reaction. On the basis of

their data, they assumed that the first step of the non-oxidative

degradation of ascorbic acid in an acid condition is hydrolysis of the

lactone ring followed by decarboxylation and dehydrations forming

3-deoxy-L-pentosone and furfural.

Factors Affecting the Browning of Packaged Orange Juice

Role of Ascorbic Acid

The primary oxidation product of L-ascorbic acid (AA) is

dehydro-L-ascorbic acid (DHAA). The latter is not a stable compound;

it undergoes spontaneous hydrolysis to a second product which has been

well characterized as 2,3-diketo-L-gulonic acid (DKGA), formed by the

opening of the lactone ring of dehydroascorbic acid. The first stage

o oxidation to EAAA is reversible and the biological activity is

-re- ied; .iwever, the oxidation of DHAA to DKGA is not reversible and

th bological activity is lost.






25





HO OH HO 0



(.A-) H :-i. --C / 0 <. p. =0
H 2OC 0

anaerobic

I\ \)
I ... OH? (HA-)

C H (D A A)
I/ \/
... ...r o 0 o



-R R








aerobic=0

S-- 0 C0








Iatal Sze for Degradation of Ascor
aerobicd ro T -urn 976)
Acid Fron Tsannenbatum, 1976)






26




0

C-CCOH




HOCH
I



HOCH2 O 2 0 0
SII 11
C---C
/ \
0 0 -CO HO HC H
i l \ / \ii
C---C HC OH
/ \ I
HCH H (DP) HCH Xylosone

H--C OH

HO CH2CH


C Amino acids

C=0

0 OH V1

= Reductones

(FA) C=0


Furfural

Brown pigments


gure 9. Degradation Pathway for Diketogulonic Acid
(From Tannenbaum, 1976)






27



Further degradation of EKGA yields active products (such as furfural)

hich u-nergo polymerization, or react with nitrogenous constituents

yieldi brown pigments (Figure 9). These reactions may occur under

aerobic as well as anaerobic conditions. For many years a

relationship between browning in orange juice and AA loss has been

known (Landen and Harrisi, 1950; Joslyn and Marsh, 1935; Joslyn et

al., 1934; Loeffer, 1941; Moore et al., 1942; Stephens et al., 1942;

Joslyn, 1941; Curl et al., 1946; Curl, 1948; Clegg and Morton, 1965;

and Mdceeny and Burton, 1963). However, the exact role of AA in the

discoloration of fruit products is still not well understood. It has

been shown that when a citrus juice darkens carbon dioxide is evolved

and AA is lost (Hamburger and Joslyn, 1941; Joslyn, 1957; Joslyn et

al., 1934; Wilson, 1928; and Curl et al., 1946). This observed loss

in AA during browning has led to the suggestion that AA is involved in

the browning reaction in one of two ways: (a) It may serve as an

antioxidant, being oxidized in preference to other substances present

which upon oxidation yield compounds or precursors of dark compounds.

(b) It may be oxidized along with other reducing constituents, such as

the flavonols, as suggested by Szent Gyorgyi (1937), to yield

different copounds that are the actual precursors of dark pigments.

Moore t al. (1942) showed that addition of AA. to orange juice

resulted in a marked increase in the rate of browninr when juice is

stored in the presence o of oxen. Siilar results were obtained by

3eattie et al. (1943) when they added AA (5 Cg/100,Il) to strawberry

juice. These data show convincing edence that AA is involved in the

browning of orange juice, and it is effective not as an antioxidant as






23



reported by Hamburger and Joslyn (1941), but rather as an intermediate

in the browning reaction; otherwise, the addition of AA should retard

rather than accelerate browning.

Wilson (1928) repored accumulation of carbon dioxide durin t

storage of sterile orange concentrate, and suggested that it may arise

from the decomposition of AA. Hoever, Eddy (1936) detected no carbon

dioxide formation in his experiments on the oxygen absorption of

unprocessed orange juice and AA solutions. Curl (1947), w oring on

storage of pasteurized, concentrated orange juiceui at 26.70C and above,

noticed the formation of gas, which was shown to be essentially carron

dioxide. This gas does not result from fermentation. A number of

possible sources of the carbon dioxide have been suggested, including

the Maillard reaction between amino acids and sugars, decomposition or

oxidation of AA, and a chemical breakdown of sugars. Later Curl and

Veldhuis (1948) reviewed this subject and presented data showing that

the formation of gas is accompanied by almost total losses of AA,

significant losses of total sugars, and considerable darkening.

Loeffer (1941) found that the quantity of carbon dioxide produced in

canned orange juice during five month's storage at 350C was almost

equivalent to the quantity of AA lost. Lahikainen et al. (1958)

showed, in model systems containing AA, glycine, and citrate buffers,

and at concentrations corresponding to those found in frit juices'

that the carbon dioxide evolved during browning was not produced b-

decarboxylation of the glycine. This finding indicated lthat glcine

had not undergone a Strecker degradation, although AA has been

regarded as an active agent (Schonberg and Moubaker, 1952).






29



Lahikainen et al. (1958) reported that the rate of production of

carbon dioxide was linearly related to the rate of browning at 370C,

but not at 500C. This is in agreement with the finding of Joslyn,

(1957) that a change in reaction mechanism may occur in AAi systems

between 30 and 500C, and was later confirmed by Nagy and Smoot (1977)

and Kanner et al. (1982). It is possible that pigment may be formed

at higher tenperatures by a separate mechanism or course of reaction

than at the lower temperatures, or a second pigment producing reaction

may be activated at the higher temperature. The total carbon dioxide

produced was in excess of that required by the mono-decarboxylation of

ascorbic acid.

Earlier, Lamden and Harris (1950) found that in solutions of AA

and citric acid there was no loss or breakdown of citric acid during

the browning of the mixture when heated in closed tubes at 1000C.

They found, also, that the quantity of carbon dioxide evolved from a

2.5% AA, 50% citric acid solution in five hour at 600C was only 13.5%

of the destroyed ascorbic acid. Jackson et al. (1960), using AA in

acetate buffer, showed less than 4% of evolved carbon dioxide was

derived from the acetate buffer. These findings indicated that

evolution of carbon dioxide was due primarily to the multiple

decarboxylation of ascorbic acid.

Joslyn (1957) reviewed earlier work on browning in model systems

containing AA, and on the part pla-ed by AA in browning of orange

juice. He reported that AA was the ocst reactive component in

browning in systems containing AA. amino acids, and sugars. He

observed that the rate of browning in orange juice was significantly






30



reduced when the anionic constituents were removed. However, the

removal of the cationic constituents had less effect.

The rate of AA destruction in citrus products depends on several

factors, including o:rgen. storage temperature, pH value, mezal

catalyst and concentration of ascorbic acid itself.

Effect of oxyen. It is a generally accepted fact that oxygen has

a pronounced effect on the rate of AA degradation and browning in

orange juice. The processing operations involved in the citrs

industry, such as extraction, screening, rmxing ard blending, increase

the dissolved oxygen concentration of the products. The presence of

oxygen is known to accelerate the AA destruction and the browning

reactions. However, it has also been reported that the removal of

molecular oxygen did not prevent loss of AA and darkening completly;

and it has been shown that the AA degradation can proceed either

aerobically or anaerobically. (Boyed and Peterson, 1945; Kefford et

al., 1959; Nagy and Smoot, 1977; Nagy, 1980). One of the

characteristic differences between these two reactions is that the

anaerobic destruction is generally believed to proceed at a slower

rate, with much easier production of furfural. An excellent review of

both types of degradation has been published by Reyolds (1965). In

early studies by Gore (1915) and McDermott (1916) it was observed tha

the color of citr jus juices was more stable n the absence of :oyn.

Matthews (1928) noted that samples of orange juice store n a

darkened more rapidly than those stored in oxygen-free gas

atmosperes. Clark (1941) discussed the effect of an excess of cxygen

in the sealed containers on color and flavor changes and reduction in






31



AA with possible darkening of the juice. Moore et al. (1942) were

able to show that the rate of browning in pasteurized bottled orange

juice was correlated with the volume of air-filled head space; as the

volTume of air was increased the rate of browning was also increased.

Joslyn and Marsh (1932, 1934, 1935) and Joslyn et al. (1934) followed

changes in vitamin C and browning in orarge juice exosed to air.

They reported that the browning of orange juice involves oxidation,

and showed a decrease in vitamin C with an increased exposure to

oxygen at room temperature. Browning was parallel to the loss of

vitamin C, suggesting a possible relationship between the two

processes.

Tressler et al. (1939) further commented that darkening was more

rapid in the presence of oxygen. It also proceeded rapidly even when

juice was deaerated to remove dissolved oxygen, and stored in vacuum

sealed containers. Joslyn and Marsh (1935) concluded that the primary

products of oxidation apparently undergo a condensation in which

secondary reactions, probably of the amino acid sugar type, occur. In

this connection, Joslyn (1941) suggested that the original, almost

momentary, contact of orange juice with air before dearation may

result in the formation of compounds (possibly peroxides) which supply

oxygen for the later deterioration of juice even in the absence of

oxygen. A number of papers have since appeared discussing the

relationship of o:xen co -qualit in crange juice. Johnson and Toledo

(1975) found that aset ly pavac-aged 550 Brix orange concentrate

lost 68% of its ascorbic acid content and was unacceptable after one

week if oxygen was left in the package headspace. Completely






32



eliminating oxygen by storing the orange concentrate in glass

containers with zero headspace considerably reduced AA degradation,

and greatly extended the shelf life, but browing and flavor

degradation still rendered the product nacceptable after ten weeks at

240C. This does not exclude the possibility that the original, aLmost

momentary contact of orange juice with air before dearation may form

peroxides which supply oxygen for the later deterioration of the juice

as previously suggested by Joslyn (1941).

The non-oxidative decomposition was studied at 300C, and 1000C,

from pH 2.2 to 6.0 (Huelin, 1953). In citrate-phosphate buffer the

reaction proceeded most rapidly at pH 3-4, and was accelerated by

fructose. Furfural and carbon dioxide were the main products of

decomposition at high temperatures or acidities. With lowering of

temperature or acidity other products became important. Cier et al.

(1959) studied the anaerobic decomposition in an unbuffered solution

of pH 2-6.6, and found L-xylose as well as carbon dioxide and

furfural. However, xylose gave only traces of furfural at this pH,

and did not appear to be the major intermediate between AA and

furfural. Kurata and Sakurai (1967) studied the decomposition of AA

at pH 2.2 and found 3-deoxy-L-pentosone and furfural, claiLngs the

former as an intermediate. Coggiola (1963) also found acid products,

and identified the major acid as 2,5 dihydro-2-furoic acid. Tat=u et

al. (1969), who used an unbuffered solution of AA and did not exiude

oxygen, obtaned 15 cCoCounds as degradation products from AA, of

which only furfural and 2,5-dihydro 2-furoic acid had been previously

shown to come from non-oxidative decomposition of AA (Kamiya, 1959).





33



They suggested the condensation of furfural probably to be responsible

for the formation of the two of the major components.

Tre kinetics of the oxidation reaction of AA have been reported

by many researchers. Joslyn and Miller (19L9a,b) reported the

oxidation of AA in sugar solutions to be essentially first order with

respect to the AA oxidation. Under conditions of limited oxygen, the

sugar solution showed reduced initial rates of oxidation. Khan and

Martel (1967) reported that the rate of the spontaneous oxidation of

AA in an aqueous model system was proportional to the concentration of

molecular oxygen down to 0.2 atm of oxygen. They also reported that

the rates of ferric and cupric ion catalyzed AA oxidation were

first-order with respect to the molecular oxygen concentration. Singh

et al. (1976), working with infant formula, concluded that when

dissolved oxygen is present in abundant supply, the reaction can be

considered to follow a first order kinetics. However, with limited

oxygen, (0.1-8.7ppm) the reaction followed second order kinetics.

Eison-Perchonok and Downes (1982) studied the autoxidation of AA by

varying temperature and AA and dissolved oxygen concentration. They

reported that AA autoxidation is dependant on the dissolved oxygen

concentration. However, Robertson and Samaniego (1986) observed no

significant effects on the rate of asorbic acid degradation and

furfural formation that could be attributed to the different initial
oxygen levels in lemon juices. In summary one can say that there is

considerable evidence that one important consequence of exposure to

oxygen is the oxidation of AA which in turn is in same way responsible

for browning of orange juice.









Effect of taeperature. One of the most important factors

influencing the rate of AA degradation is temperature. Many studies

(Brenner et l., 1-; Freed et al., 1949; Bissett and Berry, 1975;

Nagy and S-ot, 1977) ave shown losses of ascorbic acid to be related

to storage temeratures. VorLtosecke et al. (1934) reported that

packs of orange juice in glass darkened if stored at temperatures of

27C or higher, but not at 160C or below. They further reported that

the flavor of the juices in botties and citrus-enameled cans was

superior to that packed in plain tin cans. Pederson et al. (1941)

observed a relationship of the temperature of storage to changes in

flavor, color, and AA content. Loeffler (1941) stated that less than

two months'storage at hot summer temperatures could make unpalatable

the best quality of glass-packed orange juice, but that the flavor of

the freshly bottled juice could be retained almost indefinitely at

storage conditions around 40C. He further remarked that the changes

in bottled orange juice under warm storage temperature (350C) were of

the same order as those found by Vonloeseke et al. (1934) for samples

packed in enameled-tin cans. Stephens et al. (1942) found the effect

of storage temperature on the rate of darkening and stability of

flavor to be rouJly prcoortional to its effect on AA stability. This

observation was later confirmed by Curl et al. (1946) who observed a

parallel effect wi=t respect to caron dioxide production and thereby

contributed suppor to tohe suggestion that carbon dioxide formation in

these products is deri-ed frao a breakdown of AA (Joslyn et al., 1934;

Nelson and Motter, 1933). Curl (1947), working with concentrated






35



orange juice, showed that the changes occurring at 270 and 490C were

similar, but the rate of change was about 20 times as fast at the

higher temperature.

Brenner et al. (1948) and Freed et al. (1949) su;died the

retention of vitamin C in canned single strength orange juice at

21.10, 32.20 and 37.80C. These workers concluded that the logarit h

of vitamin C retention was inversely related to storage tire at these

three temperatures. No statistical treatment was applied to their

data to confirm their interpretation. Other investigators (Evenden

and March, 1948; Joslyn and Miller, 1949b; Huelin, 1953) assumed that

vitamin C degradation in orange juice was a first order reaction with

the rate of degradation proportional to concentration. These findings

differ significantly from those reported by Nagy and Smoot (1977) who

found a nonlinear relationship between log percent vitamin C retention

and time at high temperature. Their Arrhenius plot showed two

distinct temperature regions. They defined a critical region between

220 and 26.70C above which storage of juices resulted in an

accelerated rate of vitamin C breakdown. For grapefruit juice the

activation energy (Ea) was 18.2 kcal/mole, and the reaction was first

order. For orange juice, two (Eas) were determined 12.8 kcal/mole in

the temperature range 4-280C, and 24.5 kcal/mole in the range

28-50oC. The change in reaction kinetics was attributed to different

destruction mechanisms, although no explnation was offered. Kanner

et al. (1982), studying the storage stablit of range juice

concentrate packaged aseptically, showed that degradation of ascorbic

acid follows first order reaction kinetics at temperatures of 250C and






36



below. 360C the degradation of ascorbic acid did not follow a

first orcer reaction. These data are in good agreement with the

results cf Nagy and Smoot (1977) on AA degradation in stored canned

sile s5trength orange juice, but differ from those of others (Brenner

et al.,1948; Huelin, 1953), who found a first order reaction of AA

degradation until 40C or higher temperatures. Apparently, results

differ because of the long storage time. During this period many

breakdown products develop from juice constituents, which seem to

affect and accelerate the degradation of AA (Clegg, 1954; 1966).

Role of metal catalysts. Metal ions can affect the deterioration

of citrus fruits in two different ways: (a) metal ions may affect the

browning pigment formation. Joslyn and March (1935) studied the

effect of metal catalysts on the browning of orange juice, and

reported that ferrous ions increased browning, stannous ions decreased

it, while other metallic salts (including ferric, stannic, and copper

salts) had no effect. Curl (1948) has shown that browning of a

sugar-ascorbic acid system was increased by the presence of trace

elements. Jackson et al. (1960) concluded that the addition of iron

or copper decreased the rate of browning of aerated AA solutions

buffered at pH 7, although the rate of loss of AA was increased. In

the absence of heavy metals, DHAA and DKGA browned at a slightly

faster rate than AA. (b) Metal ions may affect the ascorbic acid

degradation. The catalytic properties of mheavy mets on the

oxi datio of AA, notably copper and iron have been e-xtensivel

discussed in the literature. Weissber"er and LuValle (1944) reported

that only the monoanionic AA species was susceptible to copper






37



catalysis. Later, studies by Khan and Martell (1967) showed the

oxidation of AA in solution to be linearly dependant on the

concentration of copper and iron ions. Shtan and Shurlator (1974),

and Jameson and Blackburn (1975) have also resorted the catl-ric

properties of copper ions in the degradation of A in solution. More

recently, Dennison and Kirk (1982) stdied the influence of trace

mineral fortification on the storage stability of ;AA in a low moisture

model food system as a function of water activity at a constant

storage temperature of 300C. They reported that Ai degradation

increased with increasing copper and iron levels. This concentration

dependance is in accord with the results of Khan and Martel (1967) and

Ogata et al. (1968).

It is somehow surprising that metal ions such as copper decrease

browning, and at the same time increase the conversion of the AA

present to DHAA and DKGA. This negative effect of copper placed some

doubt on the correlation of the darkening of orange juice with loss of

AA. Jackson et al. (1960) reported that metal ions in the presence of

oxygen increase the conversion of AA to DHAA and to DKGA. If the

formation of these compounds were the only limiting reactions in the

browning of AA, their accelerated formation by metallic ions would

indeed accelerate the rate of browning. Such is not the case,

however, and in fact their results showed a negative catalyis

Conversely, the rate of browning in AA buffer systems devoid of

metallic ions is increased when the available AA is converted to

DHAA. These results would indicate that in a non-metallic system one

of the first reactions in the presence of oxyen is a conversion of






38



the AA to its dehydro form and this conversion step may be partially

rate-limiting. In the presence of metal, however, a reaction leading

to the formation of polymeric compounds must be sensitive to retallic

ions and becomes the rate-liniting reaction. It is possible that some

intermediates beyond DKGA would chelate metallic ions, and if this

were the case, the ability of these materials to form long chains

capable of light absorption would be deleteriously affected, Hodges

(1953).

The mechanism of metal catalyzed oxidation of AA in aqueous

solutions is varied. Kihan and Martell (1967) postulated an

ascorbate-metal-oxygen complex involving a one electron transfer to

oxygen. Jameson and Blackburn (1975) proposed the formation of a

metal-metal dinuclear ascorbate-oxygen intermediate with a two

electron transfer to oxygen. However, the exact mechanism of action

of metal ions on the AA oxidation is still uncertain.

Effect of water activity. The nonenzymatic browning in foodstuff

takes place over a wide range of water activity. In most foods a

maximum browning reaction occurs at a certain value of aw depending on

the type of food product. It is generally agreed that the rate of

browning in fruit juices and concentrates is increased as the water

content in the product decreases (concentration of solids is

increased), (Reynolds, 1969). Thus, ncnenzymatic brownian should

occur more rapidly in dehydrated orange juice (1-3% water) than in

frozen concentrated orange juice (55% water) or in single strength

orange juice (85% water). It may be that the detrimental effect of






39



moisture on browning, so often observed, is actually due to an

increase in the rate of AA destruction by oxygen uptake.

Karel and Nickerson (1964), Jensen (1967), Vojnovich and Pfeifer

(1970) and Lee and Labuza (1975) have studied the stability of AA in

various low and intermediate moisture dehydrated foods and model

systems as a function of moisture content and water activity. Results

reported by these investigators showed that the rate of destruction of

AA in dehydrated foods increased as the total moisture content and aw

increased.

Kinetic data generated by Jensen, (1967), Vojnovich and Pfeifer

(1970) and reported by Lee and Labuza (1975) have shown the energy of

activation required for the destruction of AA to increase with

moisture content in some foods but the opposite effect occured in the

other foods. This could mean that a change in mechanism was occuring,

but only limited data were collected. Lee and Labuza (1975)

attributed the increased destruction rate for AA as a function of

water activity to the decreased viscosity of the aqueous phase,

resulting in increased mobility of reactant and catalysts.

Kirk et al. (1977) further noted that the destruction of AA could

be described by first order kinetics under all storage conditions.

The greatest stability of total ascorbic acid was observed at low

storage temperature and water activity, and was shown to result from

AA stability at these conditions. Due to availaility of water for

hydrolysis of DHAA, the stability of D-AA decreased at storaze

temeratures above 200C and water activities above monola-ver 0.24.






40



As previously mentioned, there is a certain value of a, at which a

maximum browning reaction occurs. By mainizaini a.- level, either

above or below the point of onai br-e~ ;ning, scoe increase in storage

life ould be obtained. nfortunal here is very little

experimental evidence in the literature to define the range of aw at

which this maxi~r u occurs in orange juice.

Effect of tH. It ias been shown that the rates of browning in

acidic focds such as citrus juices and in model systems are strongly

pH dependent. Berry et al. (1970) studied the storage stability of

dehydrated orange e uie at pH values that ranged from 3.3 to 6.5 and

found storage stability to decrease with increasing pH. They

concluded that the greater storage stability of instant grapefruit

juice (pH 3.3) over that of instant orange juice (pH 3.7) may be due

to the greater acidity of the former.

Wolfrom et al. (1974) found the rate of browning in 1:1

glucose:glycine model system to increase as the pH was increased from

6.0 to 7.5, and the rate of browning in that pH range to be much

greater than that between pH 3 and 4, the normal range for orange

juice.

Braverman (1963) rorted that the AA-induced browning in citrus

juices and concentrates which TiitialT involves its decomoosition to

furfural ad~ subsecuent -oloerination react=in with amino coouncds

is dependent on pH, and w~i t- the p ran: e of 2.0 to 3.5 the extent

of browning is inversely proportional to pE.










Horton and Dickman (1977) reported that AA is much less stable in

phosphate buffer than in orange juice at the same pH. They suggested

that factors other than pH must protect AA in orange juice against

oxidation. Possibly citrate, a known chelator of heavy metal ions,

inhibits the catalytic oxidation of AA. At higher pH values and at

room te~merature, all three reactions, (a) oxidation of AA to DHAA,

(b) hydrolysis of EHAA to KIGA, and (c) the decarboxylation and

further degradation of DKGA to a mixture of products, proceed much

more rapidly in phosphate buffer.

Role of Nitrogenous Compounds

The nitrogenous constituents of citrus juices have long been

suspected of being involved in the nonenzymatic browning of orange

juice. Hall (1927) and Wilson (1928) were probably the first to

suggest the possibility that the Maillard reaction was responsible for

the darkening in citrus products. According to Hall it was definitely

established that the amino nitrogen content of orange concentrates

steadily decreases in storage and may drop to zero. Wilson found a

reduction in amino nitrogen and reducing sugars, and suggested that

darkening may be due to a Maillard reaction between sugars and amino

acids. However, Nelson and Motter (1933) could detect no changes in

the concentration of the nitrogen bases present in orange juice, and

thus had reason to doubt Wilson's theory.

Joslyn and Marsh (1935) followed changes in anino nitrogen by

means of the formol titra:icn and also by the Van Slyke method. They

observed that the a-ino-nitrogen level remained practically constant

during the course of browning of Valencia and Navel juices, even after






42



storage for 126 days at room temperature when the juice had become

very dark brown in color. They postulated the possible hydrolysis of

soluble polypeptides during scorage to account for the aintainance of

constant amino nitrogen coenent duing browning. Th isolation of an

alkali soluble protein from the cnhroatophores of the orange by Smith

(1925), and later by Sinclair et al. (1935), lent support to this

hypothesis. However, Stadtman (194S) doubted this explanation and

stated that it would seem rather unlikely that the rate of amino

nitrogen formation by such a process would exactly parallel the loss

of amino acids through darkening. Nelson and Motter (1933) deter-ined

the distribution of nitrogen in various fractions of darkened and

undarkened, filtered orange juice by the use of various precipitating

agents. The darkened juice contained about twice as much soluble

nitrogen as the fresh juice. No nitrogen constituents were found in

the darkened juice which had not been identified in the fresh juice.

There was, however, a considerable increase in the arginine content

with darkening of the juice. Their results do not rule out the

possibility that nitrogen compounds are involved in the browning of

orange juice, but they do indicate that browning can occur without the

transformation of large amounts of these substances. Loeffler (1941)

also observed lower amino acid content in juices stored at 0C than

those stored at ,igher temperature.

The addition of certain amino acids to various fruit products has

been studied. Thus, Matthews (1928) added asparagine in amounts

varying from 0.01 to O.lg/100 ml. to orange juice, with and without

additions of 0.5g of glucose and citric acid. After a storage period






43



of one year, in air and nitrogen at 26.7-32.20C, darkening in treated

and untreated juice occured at the same rate and to the sane extent.

Joslyn and March (1935) found that aniline, tryptophan, and other

aromatic amines markedly increased the browning of orange juice. The

effect varied greatly with the different compounds. Richert (1930)

showed that free ammonia or ammonium ions greatly hastened the

darkening of both sugar syrup and grape juice concentrates. Richert

found glycine to be more effective than alanine but less so than

ammonium tartrate. It is also common experience in the dried fruit

industry that traces of ammonia, such as from leaks in refrigeration

coils in cold storage, rapidly enhance darkening.

The confusion in the literature on changes in nitrogenous

constituents during the browning of orange juice, may be due in part

to failure to include the reaction of reducing sugars with proteins

such as that found by Lea and Hannan (1950), and to the possibility

that relatively small chemical changes are required to produce brown

pigments of intense color. If this is the case, then the changes in

reducing sugars, or amino nitrogen, necessary to produce large changes

in color might be so small as not to be detectable by the methods

ordinarily used.

Role of Sugars

The major sugars in citrus juices are sucrose, fructose, and

gluccse. Curl and _elduis (i197) reported the sugar composition of

orange juice to be 7 sucrose, 2.57o fructose and 2.5% glucose. In

order to see if reducing sugars were involved in browning of orange

juice, Joslyn and Marsh (1935) studied the effect of their removal by





44



fermentation. The results showed that all samples (fermented and

unfermented juice) darkened at about the same rate. However, how

completly fermentation removed the sugars is not evident from the

data. Stadia.n et al. (1946) a-most cocpetely removed the reducing

sugars from apricot syrups, yet the rate of browning was decrease to

only about one half the rate in unfermenred samples. The addition of

fructose and glucose to fermented syrups in amounts equal to the sugar

lost by fermentation, resulted in a restoration of the nor al browning

rate. These results indicate that sugar may not be involved in the

darkening of orange juice, and that probably only part of the browning

in apricots involves sugar reactions. Hall (1927), in a summary of

work done on darkening in orange concentrates, stated that slight

decreases in reducing sugar during storage have been observed. Curl

et al., (1946) have verified this conclusion and showed that these

losses in reducing value are roughly parallel to changes in color.

Following the suggestion that the Maillard reaction was

responsible for browning (Wilson, 1928) a number of investigators have

attempted to correlate browning with changes in reducing sugars. The

initial rate of the Maillard reaction between a reducing sugar and an

amino compound is directly related to the conformational stability of

the favored cyclic structure of the sugar (Burton and Md eeny, 1963).

Browning reaction in amino acid-sugar systems has also been shown to

depend on the type of sugar (Cole, 1967; Spark, 1969) it follows that

pentoses are more reactive than hexoses whnich are more reactive than

disaccharides.






45



Wolfrom et al. (1974) studied the influence of different sugars

on the browning reaction, and compared a 1:5 D-glucose-glycine system

with similar systems in which D-glucose was replaced by D-fructose and

sucrose. In the unbuffered syste used, D-fructose showed a somewhat

higher initial rate of browning than D-glucose. However, the authors

stated that this difference might be reversed in the buffered media

generally characteristic of foodstuffs.

Role of Container Type

Many studies have been made on the relationships between

container, browning, and ascorbic acid retention in citrus products.

Most early studies of single strength orange juice in metal containers

or glass bottles generally showed that up to 75% or more AA was

retained after one year storage at 26.70C or lower (Moore et al.,

1944). Single strength orange juice kept frozen in tin-lined cans at

-17.80C showed no change in AA concentration after one year (Nelson

and Mottern, 1933).

In continuation of this work, Moore et al. (1944) compared the

changes in color, flavor, and ascorbic acid content occurring during

storage of the bottled and canned orange and grapefruit juices. At

the end of six months' storage at room temperature, the orange juices

in glass and tin were off-flavor, with the bottle juice slightly

better in taste than the canned juice; the results would indicate that

plain tin was found superior for packing crange juice with the

exception that at room temperature the bctzle orange juice retained a

slightly better flavor during storage than the canned juice.

Pasteurized single strength orange juice stored in glass bottles for 1






46



year retained 87% AA at 4.4iC but only 68% at 26.70C (Curl and

Veldhuis, 1947). AA retention in frozen concentrated orange juice in

tin cans was 9 or greater aftr 1 year, at 4.40C or below,

regardless of headspace amosphere, concentration of product, or

preli'=iary h-eat treatments (Curl et al.,1946).

Curl (1947) also reported that loss of AA increased with temperature

and concentration from 4.4 to 48.9C and from 13 to 710 Brix,

respectively. Retention at 4.40C ranred from 99% for SSOJ to 93% for

710Brix concentrate and at 26.70C was reduced to 707 and 60%,

respectively, after 1 year. DuBois and Kew, (1951) found frozen

concentrated orange juice stored in tin-lined cans for 11 months at

-28.9 to 23.90C had very high (95% or more) retention of AA.

Variation in frozen storage temperature had little effect on AA

retention (McColloch et al., 1957) and samples stored from -12.20 to

15.60C for 1 year, to simulate warehouse conditions, retained 95% of

their AA.

Bissett et al. (1975) studied the AA retention in orange juice as

related to container type. They found that single strength orange

juice packed in glass retained about 907/ of initial AA for over 4

months and 87% fcr 1 vear at ,.4oc. AA retention was progressively

less at 10 and 15.60c (S~3 and 79% res-ectively).













PRELIMINARY STUiJ



In a preliminary study, single strength orange juice was

aseptically filled into pouches of 3 types of flexible film: retort

pouch (American Can Company, Desmoines, Iowa.), vinylidene cryovac

(W.R. Grace & Co., Cryovac Div., Duncan, SC.), and polyethylene

Whirl-Pak (Nasco West, Modesto, CA.). These packaging materials

posess different permeabilities to oxygen as indicated in Table 5.

Table 5

Permeability of the Packaging Film


Oxygen permeability
Packaging film cc 02/100 sq.in./24hr.
720F., Imil.

Retort pouch 0a
Vinylidene cryovac 0.1-0.2b
Polyethylene Whirl-Pak 5000

-Adams (1982)
b-Sacharow and Griffin (1970)


The pouches were overfilled and sealed through the juice to

prevent the inclusion of air. The pouches were stored at room

tenperature (22 + 302) and analyzed at 0, 2, and 4 week intervals.

Ascorbic acid and dehi-roasccrbic acid concentrations were determined

using the autated filurmetric -ethod of Roy et al., (1976). The

Autoanalyzer teclniue called segmented flow analysis, where the

saple and reagent streans are segmented by air bubbles, and

continuously pumped with automated sampling of sample solutions was



47






48



used. The orocezdue is based on the oxidation of ascorbic acid to

dehydroasccobic acid by N-bromo succinimide followed by condensation

of the dehroasccrbic acid with o-phenylenediamine to form a

flucroohore. The fluorescence is then measured using a fluorometer

with a flow- crough sanoie cell. The dehvdroascorbic acid can be

determined by omitting the oxidation step of ascorbic acid.

This procedure was found accurate only when used with fresh samples of

orane juice. The accuracy of the method was determined by comparing

the analytical results obtained by the continuous flow procedure with

the titration method with 2,6-dichlorophenolindcphenol (AOAC, 1975).

Table 5 shows that the two procedures were significantly different

after 2 weeks storage. At this time some of the degradation products

in the orange juice samples interfered with the fluorescence

measurement. Therefore, an HPLC procedure was investigated which

allowed the simultaneous measurement of ascorbic and dehydroascorbic

acids in fresh samples as well as in browned samples of orange juice

using a single injection analysis procedure.


Table 6

Comparison of the Fluorometric and the Dye Titration
Procedures for Ascorbic Acid Determunation




a b a b a

ret ocuzc 26.5 27.7 21.4 .7 18i 25.0
C-ovac rack 26.7 27.0 10.0 17.4 8.3 16.9
Pol etnvlene 22.0 27.5 0.3 0.4 0.2 0.3

a = fluorometric procedure
b = dye titration procedure.














SIMULTLECOUS ANALYSIS OF ASCO RC A2 DEH DRASCOR1IC ACIDS
BY HIGH PE1RFCRIANCE LIQUID CEpaVArCnr A WIT7H POST C'LUM
DERIVATIZATIDN AD UVN ABSORBAX


Introductionr

Fruits and vegetables constitute the major sources of vi --i C

for human diets. The total vitamin C consists of the sum of ascorbic

acid and its oxidized form, dehydroascorbic acid. Both zfms rhave

equal antiscorbutic activity (Tannenbaum, 1974).

Numerous methods for the analysis of vitamin C activity have been

described. The most conmonly used are the 2,6 dichlorophenolindrphenol

visual titration (AOAC, 1975), the spectrophotometric method with

dinitrophenylhydrazine derivatization of DHAA (Roe et al., 1948), and

the microfluorometric method by condensation of DIIAA with OPDA (AOAC.

1975). However, these methods are not specific and are often limited

by the number of interfering substances present in foods. In

addition, it is difficult to determine visually the end point when

these methods are used with colored solutions. Pelletier and Brassard

(1977) described an improved photometric method based on

2,4-dinitrophenylhydrazine for the AA and DHAA determination in foods.

Though their method eliminated interference from other coMpotds, it

was time consuming and requires special samole preparation.

Recently, due to the develon of cocmercial HPLC syst-s,

quantitative measurement of AA and DEAA in various substances has been

reported by many investigators. Procedures vary in the type of

49






50



column, elution conditions, detection systems and the extraction

technique used to stabilize AA and DHAA. AA can be determined easily

by HPLC, with UV detection, but the determination of DHAA is

compicated by its et renely lew UV absorptivity. A procedure using

two reversed-phase HPLC columns in series for the separation of AA and

DHAA was reported by Finley and Duang (1981). They used water with a

counter-ion reagent (tri-n-butylamine) as a mobile phase. AA and DHAA

were detected at 254 nm and 210 nm respectively. Rose and Nahrwold

(1981) and Wimalasiri and Wills (1983) used a similar detection system

with a single ion-exchange column arnd a mobile phase of

acetonitrile-water containing 2.5 mM potassium dihydrogen phosphate.

Doner and Hicks (1981) reported a separation of AA and DHAA by HPLC on

a Zorbax-NE2 column. The AA was monitored at 268 nm, while refractive

index (PJ) detection allowed the detection of DHAA. However, neither

RI nor low wavelength (210 nm) can detect small amounts of DHAA such

as that present in foodstuffs. In addition, measurement of DHAA at

low wavelengths introduces instrumental noise from solvent impurity.

Therfore, most of the HPLC analytical procedures used are based

on either the reduction of DHAA to AA and detection of the total

ascorbic acid (TAA) by UV or oxidation of AA to DHAA. The TAA is

determined by fluorometry after condensation of DHAA with OPDA.

Dennison et al. (1981) described an HPLC ethod for the analysis of

total -vietarn C in bever by -u ceasure-ent of AA after reduction

of DHAA with homocystG: ne. Keatin and :Hddad (1982) reported the

simultaneous determination of AA and sDRAA using precolumn

derivatization. DHAA was converted to a fluorophore using OPDA.






51



The detection was made at 290 nm for AA and 348 nm for the

fluorophore. Speek et al. (1984) developed an HPLC method for the

simultaneous determination of total vita-in C based on precoir'n

erzvatic o.idation of AA to D AA. The latter is condesed with 0? '

and detected fluoroeietrically. DIHAA can be deter:lned with omission

of the oxidacion step.

While these methods give increased sensitivity for the

estimation of EHAA, the addition of the derivatization step increases

the complexity and adds another variable to the analysis. Also,

problems were encountered with the stab lity of the derivative.

Recently, Vanderslics and Higgs (1984) proposed an HPLC method with

fluorometric detection and post-column derivatization involving

oxidation of AA to DHAA followed by reaction with OPDA to form a

fluorescent product.

In this study we have examined the system proposed by Vanderslice

and Higgs (1984) and modified it to obtain an estimation that includes

AA and DHAA as a separate value using a single injection. The step

for oxidation of ascorbic acid was also omitted.

Materials and Methods

PReaents

Ascorbic and dehydroascorbic acids (Aldrich Chenical Copany,

Inc. Milwaukee, Wis.), c-phenylenedinine (OPA) (Eas-; a n Kodak

Copany, Rochester, N.Y.), metaplhospmric acid, pocassium p-snhos te

monobasic, and H~LC grade acetonitrile (Fisher Scientific Cosany,

Fair Lawn, N.J.) were used as received. Double distilled deionized

water was used to prepare solutions.






52



Apparatus

Liquid chromatograoh. High performance liquid chromatography was

performed using a system incorporating a Waters Associates Model 6000A

pup, a Waters Model U6K injector, a Spectra Physics, Model 8440

variable-wavelength ultraviolet detector (Spectra-Physics, San Jose,

CA.) set at 254 nm, and a Fisher Recordall Series 5000 recorder.

Separation of AA from DHAA was achieved by use of an amine column

(Alltech Associates, Inc., Applied Science Labs, Deerfield, IL.) in

the weak anion exchange mode. The mobile phase was 75% acetonitrile

in 0.05M monobasic potassium phosphate (pH=5.9). The eluant was

filtered through a 0.45 um Millipore filter (Gelman Sciences. Inc.,

Ann Arbor.Mich.), and subsequently degassed under vacuum. The flow

rate was 1.5 ml/min.

Post-column Derivatization

The system used for the post-column derivatization was similar to

that described by Vanderslice and Higgs (1984). After separation of

AA and DHAA on the analytical column, the exit stream from the

UV-detector was mixed with a second stream containing the

derivatization reagent (OPDA) in a mixing "PTFE Tee" (Rainin, Woburn,

MA., Catalog No. 45-1003). The final eluant was passed through a

heating coil then into a cooling coil before entering a fluorometric

detector A. nco Fluor~cmonitor) equipped with an ultraviolet mercury

light source (t-ye GE no.F4T4/BL. w att), a Corning 7-51 exitation

filter, a Wratten 2A emission filter and a 70uL flow cell, whose

output was sent to the recorder, Fig 10. All post-column tubing was

0.40 mm i.d. Teflonr. The reaction path length was 20 meters and was






53














Inj ec orA
___ njectr Peristaltic \
S iPump Pumo O PDA



Waste
/ \Flow Column

Mobile Phase Fluoro-
monitor --
1 *-- --' I

UV
-- Detector -
|o -PTFE, Tee



Heating Cooling
Bath Bath_

Dual Recorder





Fz'ze 10. TL C t~ IY, s i-~- ?cst-Col~ t Derivatization and
Tane- Ultr -iolet and Fluorocetric Detection






54



maintained at constant temperature (530C), while the cooling coil

(220C) was 2 meters. The fluorogenic reagent consisted of 0.057 (w/v)

OPDA. in distilled water and was puped using a Gilson Minipuls 2

peristaltic pump at a flow rate of 0.5 rL/ain.

Sample Preparation

Fresh fruits and vegetables were purchased from a local maret

and homogenized in a domestic blender. A sample (20g) was then blended

with 37 (w/v) metaphosphoric acid solution (50mL) for 2 min and

diluted to volume (100 or 200 ml) with extracting solution. The

resulting solution was filtered through paper (Whatman 541), and a

portion of the filtrate was purified by percolation through a C18

Sep-Pak (Waters Associates, Milford, MA), a short plastic column

containing uBondapak C18 as described by Wimalasiri and Wills (1983).

The C18 Sep-Pak was placed on the Luer tip of the syringe barrel and

the column preconditioned with 4mL of methanol followed by 0lmL of

water. The sample (4mL) was then passed through the Sep-Pak. The first

3mL were discarded and the remaining lmL was collected for analysis.

The Sep-Pak C18 could be reused up to eight times provided it was

washed with methanol and water between samples. Frozen orange juice

was first diluted according to package directions. The resulting

solution was filtered and purified as above. The injection volume was

20 uL.

Recovery Study

Proper amounts of ascorbic acid and dehydroascorbic acid

standards were added as solutions in metaphosphoric acid to the

various fruits and vegetables during extraction so that the AA and






55



HAA. content of the spiked samples approximately doubled that of the

unspiked. The AA and DHAA of sp ':ed samples were then determined as

described previously and percent recovery was calculated.

Calibration Curves

Samples of reagent grade AL and DHAA in the mobile phase were

combined to contain 2.0 AA + 0.5 EH A, 4.0 AA + 1.0 DHAA, 6.0 AA + 2.0

DHAA, 8.0 AA + 3.0 DHAA and 10.0 AA + 4.0 DHAA s/100ml.

The standard mixtures had to be prepared daily. Aliquots (20ul) of the

combined solutions were injected into the chromatographic system, and

the resulting peak heights were plotted against concentrations for the

calibration curve.

Results and Discussion

Using the HPLC procedure described, linear calibration curves

were obtained for DHAA in the range 0-4mg/100ml and for AA in the

range 0-10mg/100ml. Correlation coefficients of the linear regression

equations were 0.9994 for AA and 0.9999 for DHAA, and the limits of

detection were 0.05ug for AA and 0.01ug for DHAA per 20 ul injected.

The signal to noise ratio was S/N = 8.

Typical chromatograms of orange juice, parsley, tomato, and

strawberry are shown in Fig 11 and Fig 12. These illustrate the

ii of tande~ ultraviolet-fluorometry detection to determine

sim-umircusly AA and DEAA. In ali s-a: pes, the AA and DEAA peaks were

well res ved with no interference. The DHAA pe had a shoulder in

some of the samples. The procedure was successful ly aplied to the

analysis of vitain C in different food products and the results are

presented in Table 6. Although direct comparison was not made with





56



other methods of analysis such as the dye titration or the

fluorometric method, the levels of AA and DHAA were found to be

similar to those reported in the literature (Ashoor et al., 1984;

Wills et al., 1983; Wimalasiri and Wills, 1983).


Table 7

Ascorbic Acid and Dehydroascorbic Acid Content
of Various Foods.


Concentration (mg/100g)

Samplea AA DHAA TAA
Broccoli, fresh 81.5 Z 2.5 6.2 : 0.9 87.7 1.6
Orange juice 42.7 1.1 2.9 0.2 45.6 1.0
Fruit punch 51.0 1.9 3.9 0.2 54.9 2.1
Orange drink 45.7 1.6 1.0 0.1 46.7 1.5
Parsley 148.7 7.4 9.7 1.0 158.5 7.0
Tomato fresh 9.1 0.4 1.1 0.2 10.2 0.5
Strawberry 51.0 1.9 6.1 0.2 57.1 1.8
Banana 7.7 0.6 3.0 0.1 10.7 0.6

aNumber of samples = 4. Mean SD






57
































IParsley, Mnitored b- ancemo
Slecion
I j!











"cr'- ,v Orcnce deice






UV (2-4nm ;


SI




Figure !i Typiica! C" rc*tc s of ra~e Juice a
Parsley, Monitored by Tandem t aviolet
(UV, 254 ns) anid Fluconoetric Detection































c.. i 1-1 -









AA



ICmrc:o StfrcNt r ry








i j










Figure 12. Typical HPLC ChroaS; t-gras of Toato and
Strawberry Monitored by Tandem Ultraviolet
(V, 254 rnm) and Fluoroz.tric Detection






59




Table 8

Recovery of Ascorbic and Dehydroascorbic
Acids from Spiked Samples


Recoveries (%)

Sarmlea AA DHAA i

Broccoli, 97.7 2.9 110.7 + 4.6 104.2 .
Orange juice 99.5- 2.7 101.7 T 0.8 100.6 1.8
0 i 99.8 e 0.9
Fruit punch 97.0 0.7 102.5 1.1 99. 0.
Orange drink 96.7 1.5 101.7 2.2 99.2 1.8
Parsley 91.0 1.9 111.7 2.5 101.4 2.2
Tonato 96.0 1.2 104.7 1.5 100.4 1.3
Strawberry 96.5 1.1 105.7 1.5 101.0 1.3
Banana 91.2 1.9 102.5 1.1 96.9 1.5
a';ber of saroles = 4. Mean SD.



The data in Table 7. show that both AA and 1~AA are completely

recovered from the samples examined. The recoveries ranged from 91 to

99.5% for AA and from 101 to 112% for DHAA. The slightly higher

recoveries for DHAA could be due to the oxidation of some of the AA to

HAA during extraction and sample preparation. For samples that

require homogenization and extraction such as banana and parsley etc.,

conducting the extraction at 30C should be helpful in preventing

potential conversion of AA to DHAA during the extraction procedure.

This IPLC prccedure provides a relatively fast and sensitive

tech?2niue for the si-ularneous determination of AA and EHAA in

fcdstuffs and beverages. The ethod is simple and requires a m-nirnu

of saIple preparation during the simultaneous determination of AA and

DHAA. Further, this 'PILC method measured AA and DHAA directly, which






60



eliminates the need for the odxdation of AA to DHAA or the reduction

of DHAA to AA prior to the anal-sis. The procedure was also found to

be very useful for reasure.en: of A and DA in browned samples of

or-ane e juice w-ere moan intereing co uds limited the use of the

dye tiration method and the icronfluoroetric method.

Finally, an attempt was made to include the diketcgulonic acid

(DKGA) deter-mination in our assay. The iDGA was prepared according to

the method of Dcner and Kevin (1981) as follow: EHAA was readily

prepared from AA by air oxidation of an ethanolic solution containing

activated charcoal. After fltration and remcval of ethanol, pure

syrup of RAA was obtained, as determined by HPLC analysis. ELHAA (an

aqueous solution of 10mg/ml) was then converted to DKGA by gradual

titration in an ice bath over a period of 1 hour with an aqueous

solution of 10mg/ml with 0.5N sodium hydroxide until the pH remained

constant at 7.0. Standard mixtures of AA, DHAA, and DKGA were

prepared just prior to analysis by HPLC (UV detection). The

acquisition of a new Hewlett Packard 1090 Liquid -Cromatograph with

HP-85B Personal Computer and DPU multichannel integrator, permitted

the simultaneous monitoring of the different compounds at different

wavelengths. Figure 13 presents a chromatcgran of the standard

mixture (AA, EHAA, and K-GA) with their absorption spectra. This

orocecdr .rovided an zzel'lrt resoution of fhe three cci-ounds with

retention tiaes of 7.9, 3.7, and 10.7 rinuts respectively. However,

the UV detection of DHAA and DiSA was znot sufficiently sensitive to

detect these two copounds at the levels actually present in orange

juice, even when a low wavelength (210 nm) was used for analysis.






61



Figure 14 shows two chromatograms of AA (40 ug/ml) and DHAA (1 mg/ml)

at two different wavelengths (210 and 254 nmn). At the level used DHAA

was only slightly detected at 210 nm. This demonstrates the advantage

of the fluorometric detection procedure w era thee ensit iit was

significantly improved over UV detection. Measurement of THAA and

DKGA has been made by others using UV detection (Finley and Duang,

1982). We found the maximum absorption for DHAA occurs at a

wavelength of 227 nm, and for DKGA 200 nm, figure A-6. Many other

compounds which occur in biological samples also absorb at these low

wavelengths. Therefore, determination of DHAA and DKGA by UV

detection was not a suitable procedure for the low levels in orange

juice.






62












A-
-- \ i ,D > --

7i i \ i \
/ \










AA




DHAA DKGA
















Figure 13. Resolution of AA. DHAA, anrd DiA








Iffl 1 ine plot from DPU memory
i n l13 Anno i t I I
S I ,bLJ 21 ,20 254 ,20
R r,]<: 50 [C C
7(;r 3 -, 5





























Figure 14. Ascorbic and. Dehydroascorl7 c A cidl roni tord at
210 and 254 nm







S ,,I*,i: :AIl (ill 1), 56. (264/ 2) 34. 2 (228/ 2) 60. 2 (.I (/ ;)

[miAl I I\


\ DDIIAA AA


N\












l I I I -- --. .. ... ..... .......




S1 II. pc fo A r i 11n i




Figure 15. Absorption Spectrum for AA, DHIA, and XKGA














EFECT CF ASCCRIIC ACID DAD AM-O ACIDS CaCE-TPTICTS
ON H QUAITY OF ASEPTCALLY PACAGED OPAFQE CDRIC


Introduc tion

Nonenzymatic browning is one of the main reasons for the

reduction in commercial value of citrus products. is the cuickest

and most dramatic quality defect to appear during ab:ient tp3erature

storage. Knowledge of the runda mntal factors and the mechanism of

reactions which these factors can undergo under different storage

conditions is critical to the understanding and subsequent control of

nonenzymatic browning in citrus. While some progress has been made in

the study of the changes responsible for darkening, the chemistry of

many of these changes is not well understood, and the nature of all

the constituents involved is not known.

In view of this lack of understanding, it was decided to carry

out the present investigation to study the reactions taking place

during the nonenzymatic browning of aseptically packaged orange

drinks. Our main objectives were to determine the loss of ascorbic

acid and the devel"opment of browning as affected by ascorbic acid and

amino acid content, and oxygen permeablity of nackair Tarial.

A 323x2 factorial experiment was desig-ed to measure te sai

effect of ascorbic acid, amino acids, and ogen permeabilit aln as

well as all permutations of any two or three factors'interaction on

the browning of orange drinks aseptically packaged and stored at 750F.


65






66



Materials and Methods

7pj^ents

Ascoric acid (Aldrich Chemical Company, Inc. Milwaulee, Wis.),

arginine, aspartic acid, citric acid, potassium citrate, frctose, and
glucose (Fisher Scientific Company, Fair Lawn, N.J.), 4-aminobutyric

acid (East=an Kodak Company, Rochester, N.Y.) and sucrose (local

market) were obtained and used as purchased. The juice used in this

study was reconstituted from a high quality Florida commercial

concentrate. The frozen concentrated orange juice was first diluted

according to package directions with distilled water which was boiled

then cooled to room temperature to remove any dissolved oxygen. The

final degree Brix was 11.8.

Orange Drinks Composition

Nine orange drink mixtures containing 10% (w/w) orange juice and

various compositions of ascorbic acid, amino acids, sugars, ciric

acid, and potassium citrate were prepared, aseptically packaged, and

stored at controlled room temperature (750F, 23.90C). The

compositions of the mixtures are given in Table 9. Analyses of the

mixtures were conducted over a period of 20 weeks.

The mixture of sugars used (5% sucrose, 2.5% glucose, and 2.5%

fructose) was sirilar to that which occurs in orange juice, (Curl and

Veldu4is, 1947). For amino acids, a mixt-re of 0.2% each of

L-aspartic acid, L-arir-ine, and 4-a- butyric acid was used. Since

these are the most abund.ant amino acids i orange j-Tuice and in many

other fruit juices (Winston, 1961), it was considered that these amino

acids may play an important role in the deterioration of orange juice







67



on storage. Wolfrom et al. (1974) reported L-argnine and

4-aminobutyric acid to give the most intense and rapid color

formation, and vwre cqantitatvely much more efecti've than glycine or

any other of the 9 amino acids e:-amined. According to the literature

orange juice contains 0.5%(w/w) of -crde proteins (Chatfield and

Adams, 1940; and aatt and Merrill, 1963), hence this quantity of the

miead amino acid ws ws used in mixtures, 2, 5, and 8. To mixture 1, 4,

and 7 no amino acids were added, and the only amount present in the

mixture is that contributed by the 10% orange juice. This amount was

assumed to be 0.06%. To mixture 3, 6, and 9 amino acids were added to

the level of 1.26% (w/w).

The quantity of ascorbic acid used in the orange drinks 4, 5, and

6 was 38.0 mg/ 100ml. To mixtures 1, 2, and 3, no ascorbic acid was

added, and the mixture content was 4.2 mg/ 100ml, provided by the 10%

orange juice. Mixture 7, 8, and 9 contained 71.8 mg/ 100ml of

ascorbic acid. Each mixture contained 1% (w/w) of citric acid and

0.7% (w/w) of potassium citrate (both as hydrates) to buffer the

solutions to a pH of about 3.8, which falls within the normal range of

pH of orange juice. These compounds are part of the buffer system of

orange juice. The mixtures were filled aseptically at room

terrerature.

With mixture 1, 4, and 7 t-e ascor-ic acid level was increased

(4.2, 38.0, and 71.3 g/100 nL), thoe object-ie was to determine the

effect of the concentration of ascorbic acid on the rate of ascorbic

acid loss and the rate of browning without addition of amino acids.

The amino acid level (0.06%) was that naturally provided by the 10%







68






Table 9

Cocposition of the Orange Drink Mixtures



Mixture Ascorbic Acid Amino acidsa Other constituents
1 I
(mg/lOOm100l) o (w/w) i % (w/w)


M 1 4.2 0 0.06 I All mixtures
M 2 4.2 0.66 contained
M 3 1 4.2 : 1.26 1 % citric acid
0.7 % potassium
M 4 38.0 ; 0.06 citrate
M 5 38.0 0.66 5 % sucrose
M 6 I 38.0 1.26 2.5 % glucose
2.5 % fructose
M 7 I 71.8 0.06
M 8 71.8 0.66
M 9 1 71.8 1.26

MO 0 42.0 I 0.60 S.S.O.J.


a A mixture of equal amounts of aspartic acid, 4-aminobutyric
acid, and arginine was used in each case with the indicated
percentages.


orange juice. With mixture 2, 5, 8, and 3, 6, 9 the objective was to

determine the effect of increased levels of amino acids (0.66 and

1.26%) on browning and ascorbic acid degradation.

Preparation of the Mixtures

Using te-~ fcilities at the Food Science and Hum;an Nutritio

Depa-men (Urniversity of Florida, Gainesville), the different

ingredients were mixed with single strength orange juice and distilled

water to make an orange drink containing 10% orange juice, the final

pH was 3.8. Three mixtures were prepared each day. The filling of







69



the pouches was done on the day following the preparation of the

mixtures. The mixtures were maintained overnizt at a -C before

processing and packaging.

The juice and the drinks were processed using a No-Bac Uni~t-rm

IV Processing syste. (Cherr-urrll Corporation, Cedar Rapids,

Iowa). It is a complete unitized systea for sterilizing fluid

products at a rate of 22.5 to A5 gallons per hour. It consists of two

surge tanks, one with agitator, spply pump, high pressure pump, heat

exchangers, aseptic remote h-mogeniing valve, valve manifold and

control panel. The high pressure pu-p has a 3000 PSI maximum pressure

limitation. The 1/4" tubular heat exchangers have a 150 PSI maximum

steam pressure limitation. The system was first sterilized with

circulating water at 2850F for 20 minutes. The product was pumped

from an agited supply tank and was heated to 205F for 14 seconds,

with the flow rate of the product kept at 31.5 gallons per hour. The

sterile product was then immediately cooled to 800F. From this point

on every precaution was taken to make certain the product did not

become recontaminated.

The products were then packaged into 7 by 10 cm pouches of two

types of flexible films: retort pouch (zero perr-ability to ocygen)

and polyethylene pack (high perseabilicy to c:0o n). The retort pouch

composition from inside was pol7etnylene teraohtha~iat s/al ,min

f Cil/oly propylene. They w'er-- sea sterlifd r hour-7 s. The

polye ehylene packs were Vnirl-Pak type, conCercial sterile, and were

used without further treatment. The pouches were aseptically

overfilled and sealed through the liquid to prevent the inclusion of







70



air. The filling was made in a sterile environment in a laminar flow

hood (The Baker Co., Inc. Sanford Airport, Sanford, Me.). The hood was

equipped with a blower that provided an average air velocir of 99.6

fpm., a prefilter-Scottfoam. (washable), a final filter (Zero Probed

HEPA, 99.997 efficient on all particules 0.3 micron by D.O.P. taest

and an UV light to maintain the sterile conditions. Thir-y t-

pouches were prepared from each mixture and each packaging material.

Eight pouches from each mixture (4 retort pouches and 4 hirl-Pak)

were opened after each storage period and analyzed for ascorbic acid,

dehydroascorbic acid, and browning over a period of 20 weeks.

Method of Analysis

Ascorbic and Dehydroascorbic Acids Determination

Ascorbic and dehydroascorbic acids were determined as soon as

possible after the pouches were opened using the HPLC procedure

previously described. The zero-time analysis was made on the day

following the filling of the pouches. The pouches were allowed to

stand overnight at room temperature. Thereafter, samples were

analyzed at 2-week intervals for the first 8 weeks, and then at 12, 16

and 20 weeks. All measurements were made on 4 packages from each

mixture.

Brownirng Measurement

Browning expressed as absorbance at 420 nm, was measured

according to the method dev.eloped by Mey dav a., (1977). In this

method the pulp and ser~m were separated cen=rifug;all (2000 rpm for

20 min). The supernatant was then diluted 1:1 (v/v) with ethyl

alcohol (95, v/v) to cause floculation of the cloud and was then







71



filtered through a Whatrn nc.l2 filter paper to obtain a fully

clarified e:xtract. The clear solution was checked for its absorbance

at 20 nr, in T3 4050 U1ltr ec. (l,3. Biochrom LTD Science Park

Cambridge, Engilard).

Statistical reatments

Regression etlhod were used for the calculation of factor

effects, and for the analysis of variance (ANOVA).

Results arn Discussion

Ascorbic Acid Retention

Effect of processing. From a nutritional point of view, the

retention of AA is an important factor for citrus products. Table 10

shows the estimated initial and zero time AA and DIAA levels for

orange juice and orange drinks products in both types of packaging

materials. There was considerable discrepancy of AA between the

initial estimated values, and the measured value at zero time. This

loss occurred in all mixtures and in both types of packaging

materials, although it was more pronounced in the Whirl-Paks. This

decrease in AA level represents the combined loss during processing,

storage of the sample prior to the first analysis, and preparation of

the drinks. The incorporation of oxygen into the mixtures prior to

theal processng (e. dil with oxygen containing water, mixing

in open kettle to dissolve the ingredients) may have contributed to a

arge ex t to th oxidatin of AA to DHAA and the conversion of DHAA

to DKGA. This is evidenced by the high level of DIIAA four:d in all

mixtures at the zero time. On average the percent loss for total

ascorbic acid (TAA) ranged from 12 to 21% in retort pouches and 20 to

26% in the Whirl-Pak.









Table 10

Initial AA and DIIAA Levels for Orange Juice and
Orange Drink Products in Retort Pouch and Polyethylene Pouch

ni tial estimated Zero time (mg/100 nL)
values
(rm/100 mL) Retort pouch Polyethylene pouch
I'dL AA DIIAA TAA AA DIIAA TAA AA DItAA TAA

f) 7i 4T.4 3.0 45.4 31.5. 6 4.1+.1 35.6 30-_6+-5.3-.5 35.9

11 4.2 0.3 4.5 1.0F.2 2.3.2 3.3 .3" .0 2.1 .2 2.7
M2 4.2 0.3 4.5 .9.3 2.4.2 3.3 I.I .1 2.1f .l 2 .
M3 4.2 0.3 4.5 1.4.l 2.3.l 3.7 1.3 .2 2.31 .1 3.5

1i1 38.0 0.3 38.3 19.6.3 13.6.4 33.2 12.2 .9 15.41 .9 27.6
115 38.0 0.3 38.3 20.1.8 11.2.4 31.3 15.91 .4 13.0: .8 28.'
M16 38.0 0.3 38.3 16.1.9 11.91.4 28.0 13.5 .5 14.1r .4 27.6

117 71.8 0.3 72.1 51.6.7 12.5.l 64.1 44.01i1.7 16.5i-1.1 60.5
118 71.8 0.3 72.1 50.6.9 13.81.9 63.8 32.2-1:2.4 24.11.9 56 3
19 71.8 0.3 72.1 46.0k.7 15.6.7 61.6 35.81-1.2 19.51.0 55.2


aNumilxr of samples = 4. Mean + SD.







73



Asc.ric -cid reten.tion as affected by amino acid concentration.

The effect of a rno acid level on ascorbic acid retention and the

concentra:ion of its ox'vdation product, dehydroascorbic acid, are

presented in Figures 16 to 19. Figure 16 shows the loss of ascorbic

acid in sales scored in Whirl-Paks initially fortified with the

highest concentration of ascorbic acid (71.8 mg/100 ml) and increasing

levels of amino acids. In addition to the initial rapid loss of AA,

these samples continued to lose AA more rapidly in storage than did

those in retort pouches and by the second week of storage both AA and

DHAA had almost completly disappeared. Similar results were obtained

at the 38.0 and 4.2 mg/100 ml ascorbic acid levels (data in Appendix

Figure A-i and A-2). Packaging of drinks in polyethylene film (oxygen

permeable) resulted in extremely rapid destruction of the AA and

DHAA. Experimental data and the results reported from a study done by

Dennison and Kirk, (1978) indicated that oxygen must be evaluated as a

reactant in the stability of AA. Figure 17 shows the effect of

increasing amino acids concentration at the highest AA content (71.8

mg/100 ml) in samples stored in retort pouches. There was a rapid

decrease in AA concentration with the first 2 to 4 weeks storage,

followed by a slower decrease for the remainder of the 20 weeks

storage priod. Such an initial rapid loss zay be due to oxidation by

residual cgen in the drinks reacting wth AA. After this initial

peri~_ d, A wa as roablD y degraded anaerobically, at rate lower w tiz -

aerobic process. Considerable loss of AdA occured in the presence of

the highest level of amino acids (1.26%). This suggests that amino

acids may have some effect on the AA degradation.







74



In samples containing 1.26% amino acids, AA retention was reduced to

56% of its initial level after 20 weeks.

Samples containing 0.66 and 0.06% anno acids retained 61 and 63%

respectively, after the same period of time. Figure 18 shows the sane

effect but with a fi_-ed concentration of 38.0 =m/100 i of AA. It is

interesting to note that this level of AA is similar to that found in

single strength orange juice. At this level of AA the increase of

amino acids from 0.06 to 0.66% did not affect the loss of ascorbic

acid. The per cent losses in single strength orange juice stored in

retort pouches were close to those in orarge drinks 4, 5, 7, and 8

which contained the low levels of amino acids. Curl et al., (1949)

working with ascorbic acid-amino acids-sugar systems, reported similar

results and suggested that, since the losses in orange juice were of

same order of magnitude as the model systems, it is possible that the

reactants involved are the same or similar.

Under similar conditions, but with an initial concentration of

4.2 mg/ 100 ml ascorbic acid, and increasing amino acids (Figure 19),

there was no significant effect of amino acid levels on AA retention.

The small amount of AA present (4.2 mg/100 nL) in these drinks made

any difference difficult to detect.

Dehydroascorbic Acid Production

All AA initially added to te orange drinks was in the reduced

form. However, cur results shoved :hat significantly hih Lels C

DHAA were found in all mixtures at the zero time analysis. These

unusually high levels of DHAA are the result of incorporation of air

and oxidation of a large amount of AA during preparation of the drinks.






75



During the first 2 weeks the levels of DHAA decreased rapidly in

all mixtures. This decrease reflects either the conversion of DHAA to

DKGA, or its reaction with amino acids, we were unable to measured

DKGA with the HPLC procedure used. It has been reported that DHAA

reacts very rapidly with alpha amino acids to produce strongly colored

(reddish to brown) complexes (Koppanyi et al., 1945). Once DHAA is

degraded, its antiscorbutic activity is lost. In samples stored in

retort pouches (Figure 17 and 18) the amount of DHAA gradually

increased after 8 weeks and remained relatively constant at this

level. These results are in agreement with those reported by Moore et

al. (1944) who showed that the levels of DHAA in canned and bottled

orange and grapefruit juices stored for six months at 4 and 270C

remained constant at about 1-2% of the total ascorbic acid, neither

container type nor storage temperature appeared to influence the DHAA

level. Smoot and Nagy (1980) reported that with single strength

grapefruit juice stored at different temperatures the DHAA and DKGA

contents remained virtually unchanged during a 12 week period. Our

results showed that DHAA, following an initial drop in concentration,

remained relatively constant throughout the 20 week storage period.

No accumulation of DHAA occurred in the different orange drinks and

once it is formed it is further transformed to DKGA or reacts with

amino acids. Although AA-induced browning in citrus products, which

involves its conversion to furfural, is well known (Braverman, 1963),

discoloration involving the reaction of amino acids with DHAA or DKGA

has not been widely investigated. Dulkin and Friedemann (1956)

studied the role of DHAA in the browning reaction and reported that







iG/ l[() Ill.
e nchl point: -is the avernge of 4i piacl t
solid 1 ines = ascorlbic acidl
(Inshed .:ines = debyd ronscorbic ncldl
1 ) !, l,; 1,\ !l II .\( 1, |;I
I I











Iii






1 8 12 10

STORAGE TITE (WqEEKS)









Figure 16. Effect of Amino Acid Concentration on Ascorbic Acid
Retention. (initital level of ascorbic acid 71.8 mg/100 ml,
samples stored in polyethylene pouch)







il)oo HIl each point: is the vcerage of 4 packs
fO) solid l:ines = ascorbic acid1 .l,,; .... AMI I A( I
dlnshed lines = dehydrons corci c acid

' i *i j. O) i i", ; \l.l liil A II





-- -- --- ..-.. -
~~ - .. ....
i (i





I D H A A .... .. ...- -....-- --.- .. .






sI IfI\ARE TIII-IE (w I,







Fiigure 17. Effect of Amino Acid Concentration on Ascorbic Acid
Retention. (initial level of ascorbic acid 71.8 mg/100 ml,
samples stored in retort pouch)
samplesi stored in retort Iouch)







ii- U each po:fnt Is the average oC li pi :l.;
solid lines = ascorbic acid
i dashed lines = dchyd roascorb tc ic d l.l,: III 1 l







n
























Figure 18. Effect of Amino Acid Concentration on Ascorbic Acid
S!-----























Retention. (initial level of ascorbic acid 38.0 mg/100 ml,
sam s stored in retort h)
D H A A ,




i) A t 1 ?- ifi ;'l
STor1AGE TIM1E (IWIEEKS)








Figure 18. Effect of Amino Acid Concentration on Ascorblic Acid
Retention. (ini.tial level of ascorbic acid 38.0 ng/100 ml,
sanples stored in retort pouch)






(); ( 1i1 ench point is the averagne of i pn;ckk
solid 1:ines = acorh:ic nc:id








i ;
hed les eydrocorc i A H il

















TORAGE TIHE (W1EHKS)








FIgure 19. Effect of Amino Acid Concentration on the Ascorhic Acid
Retention. (initial level of ascorbic aicid 4.2 inm/100 in].
samples stored in retort pouch)
\


: "..^" .---. ^-_r _

,I H 1;] IR ?).

STOnAGE TIlde (WI-EElS)








} 'itre 19. Effect of Ami-no Acid Coneentration on the Ascorbic Acid
Retention. (initial level of ascorbic acid 4.2 ing/100 ml
samples stored in retort pouch)






80



oxidation of AA to DHAA is a prerequisite to browning. They also

observed that the rate of browning was not increased by the addition

of an amino acid (tryptophan) and suggested that browning in their

experiment was not identical to the Maillard-type of reaction, and

that DHAA must first undergo an irreversible transformation to DKGA.

This transformation was first postulated by Herbert et al. (1933) and

was substantiated by the work of Penney and Zilva (1943).

Browning As Influenced By Ascorbic Acid Concentration

Since AA appeard to be the most reactive constituent in the

orange drinks investigated, and since browning in citrus juices is

considered to be mainly due to AA degradation, it is therefore

reasonable to expect the trend in browning to correspond to that of AA

loss. Browning in the different orange drinks, as measured by

absorbance and as affected by ascorbic acid concentration is presented

in Figures 20, 21, and 22. In all cases, the diffusion of atmospheric

oxygen through the polyethylene film was found to decrease

significantly the ascorbic acid retention and to increase the

development of brown pigments of non-enzymatic origin. It is

therefore reasonable to conclude that the brown pigments originate

from the oxidation of ascorbic acid. Orange drinks containing an

initial concentration of 0.067 amino acids and with increasing level

of AA (Figure 20) showed little browning in samples stored in retort

pouches. However, in the presence of oxygen there was significant

darkening. The effect of increasing AA was more significant in the

presence of oxygen than in its absence. Under similar conditions, but

with a initial concentration of 0.66% amino acids and increasing level






81



of AA (Figure 21) there was a significant effect of AA concentration

only in the presence of oxygen. High AA concentrations yielded a

darker colored product th the Iow AA level. It is intestLg

note that si 0.66. an.o acids is the concentration ra-ne four in

orange juice then an increase of the AA concentration in the absence

of oxygen would not increase the browrning significantlv whereas, in

the presence of oxywen, there would be significant darkening. This

result indicates that oxdation of ascorbic acid is a major factor in

the formation of brown pigmnts.

Orange drinks containing 1.26% (w/w-) a-ino acids and increasing

level of AA (Figure 22) showed significant differences in browning.

In both packaging materials AA had an effect on the browning. In the

polyethylene pouch, with high oxygen permeability increases in

browning followed increases in the ascorbic acid content, whereas

there was not a linear relationship in the retort pouch. In retort

poch high levels of ascorbic acid (30.8 72.8 mg/100 ml) increased

the browning. This is in agreement with the earlier finding that only

when the level of amino acids was high (1.26%) in retort pouch was the

ascorbic acid retention decreased. This indicates that high amino

acid levels accelerated ascorbic acid degradation which resulted in an

increase in browning.

Browninz As Influenced B' Ami'no Acids Concentration

Orange drinks containing 4.2 mg/100 ml of ascorbic acid and

increasing levels of amino acids (0.06, 0.66, 1.20 5 ) browndc cnly

slightly even in the po-letnilene pouch (Figure 23). Increasing the

concentration of amino acids at the low level of AA had no significant









If',[ A I' .1-0 nm
Sench point is the avernge of / pliacks
sol id l.:ines = st:ored inlirel:ori: pouch 71.
dan'shed .:lnes = ,toLred in pol y tLliync po)(iicl 11

30.0 n iir/100 11 mll



A."2 mn,/100nn fil
I


I .. .. .

A- .. ----------........ A.............----......








I .. ....... ... ... .

STORAGE TIMTE (WEEKl






Figure 20. Effect of Ascorbic Acid Concentration oln Browning
of Orange Drinks with 0.06% Amino Acids








A II S !'!I:- A F 120 nr
1e. / ench point Is tlhe averrge of 4 pnIcks 71 .8 i /lIon I
solid 1 ines = st.ore in relort: pouch
daslhed Unes = stlorSc in poI yet liyl ne pouch -
1H). (0 i 1,/ (MC! II


_.I C/Al i0 I





























of Orange Drinks with 0.667 Amino Acids
/ I I


t) /]



----- ...---- .. -o---------- ;, .. ...


C) I-C 4"_-__'__- -...-..-' _+,0 -Z .. .. 0 .. I ...
,11 n3 12 I! ;'i)

STORAGE TIME f('WEEKS)








IF'ii:e 2. I Effect of Ascorbic Acid Concentration ornt Browving
of Orange Drinks with 0.667% Amino Acid.s







A['3 11 / ll I T I1 0) nm
i eicih po-lit in the avern.fe or 4 pricks 71., I f)0 mi
Sso'l d l tnes = Sti (I In retor p] ucli
a d slied 1 ines = tlorc( In polycvth:ylone po <'hli

(t I I. I O) l) in i
H fil" / 100 Jill


l0 \





/!. -; - --- --A
0 91\ 0(
ii ,i i .- -. -. ----.- -. . .





1I (,









STORAGF TIME (W KS)









Figure 22. Effect Of Ascorbic Acid Concentration on Browni ing
of Orange Drinks with 1.26% Amino Acids






85



effect on the browning of the product. This indicates that the

bro win Ln this case is probably different from the Maillard

reactn. Furzhrmore, it is well known the conditions which would

favor Che velco ment of the aillard reaction, near neutral pH or

slightly alkaline pH, are absent in orange juice. Therefore, it is

unlikely that this mechanism is the maior contributor to the browning

of a highly acid product such as orange juice at pH 3.8. Under

similar conditions but with a higher level of AA (Figures 24 and 25),

there was an appreciable increase in the darkening only at 1.267% amino

acid concentration. This effect was more significant in the presence

of oxygen than in its absence. It is interesting to note that when

samples were stored in retort pouches an increase of amino acids from

0.06 to 0.66% had no significant effect on the browning regardless of

the levels of ascorbic acid. These results indicate, that in orange

juice where the amino acid concentration is below 0.66%, the amino

acids will not significantly affect the browning of the product.

Joslyn and March (1935) reported that amino acids play a minor role in

the oxidative nonenzymatic browning of orange juice; this has been

confirmed for orange juice and for a model system composed of ascorbic

acid ar-d l-cine and other amino acids (Dulkin and Friedemann, 1956;

and Jos--ly 1957). Our study further substantiated this finding. The

accelerated breakdown of ascorbic acid in the presence of high levels

o ne acids thas been reported in the literature (Josl n, 1957

Clegs, 1964; and Seck and Crouzet, 1981). On the basis of :he

ch=Mical structure of ascorbic acid, it has been postulated that in

citrus luice, ascorbic acid reacts with amino acids in much the same







Aj ;mii | 1A lrL A, r i; nirim
Seach point is thl average of /i piccls I i ;'.L Ul \f !'i
sol.id 11ines = stored :in ret:ort Ipouch
daslhedl :lines = stored in poly tliyl(nc' piich





Ii;




I




I ) L -... 0:1-1
!- .. . . .....--.--



,I [ 1 \ Cii .r !

SIO[1AG[E TIME (W1TFKS)










Figure 23. Effect of Amino Acid Concentration on Browning
of Orange Drinks (4.2 mg/100 ml AA)




Full Text

PAGE 1

STORAGE STABILITY OF ASEPTICALLY PACKAGED SINGLE STRENGTH ORANGE JUICE AS! ORANGE DRINKS BY BECHIR KACEM A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY' OF FLORIDA IN PARTIAL FULFILLMENT OF THE REOJJIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1986

PAGE 2

IN MEMORY OF MY FATHER

PAGE 3

ACKNOWLEDGEMENTS The author wishes to express his sincere thanks to Dr. R.F. Matthews, his major professor, for support and guidance which made possible the completion of this work. Special gratitude is also extended to Dr. M.R. Marshall, co-chairman, Dr. P.G. Crandall, Dr. J.F. Gregory and Dr. R.B. Shireman, members of his supervisory committee, for their continual encouragement and counsel throughout the work. Special thanks go to Dr. J. A. Cornell, also a member of this committee, for his helpful advice and assistance concerning the statistical evaluation of the experimental results. Appreciation is also extended to Mr. P. West for his technical assistance throughout the months of laboratory work, to Ms. Virginia Wily for her help with the amino acids analysis and to Ms. Robin Adkins for printing the manuscript. The author also acknowledges partial financial support for this research from the Tunisia Agricultural Technology Transfer Project. Last but not least, special gratitude must be credited to his wife, Saida, for her love, dependability and understanding throughout this work. To his mother, the author extends his deepest appreciation for her faith and her patience during his absence from home. • • • 111

PAGE 4

TABLE OF CONTENTS PAGE ACKNCWLEDGEMENTS lit LIST OF TABLES vii LIST OF FIGURES 5.x LIST OF ABBREVIATIONS xli ABSTRACT xLii INTRODUCTION 1 LITERATURE REVIEW 3 A Brief History of Citrus Distribution 3 Citrus Production in Florida 3 Citrus Demand 9 Citrus Processing 10 National 10 Florida 10 Juice Packaging 11 Aseptic Processing 14 Nonenzymatic Browning 15 Maillard Reaction 17 Active-Aldehyde Theory 21 Ascorbic Acid Theory 23 Factors Affecting the Browning of Packaged Orange Juice. 24 Role of Ascorbic Acid 24 Role of Nitrogenous Conpounds 41 Role of Sugars 43 Role of Container Type 45 PRELIMINARY STUDY Ui SIMULTANEOUS ANALYSIS OF ASCCR3IC AND DEHYIK0ASCGR3IC ACIDS BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY WITH POST-COLUMN DERIVATIZATICN AND UV A3SCRBANCE 49 Introduction 49 Materials and Methods 51 Reagents 51 Apparatus 5? iv

PAGE 5

Postcolumn Derivatization 52 Sample Preparation 54 Recovery Study 54 Calibration Curves 55 Results and Discussion 55 EFFECT OF ASCORBIC ACID AND AMINO ACIDS CONCENTRATIONS ON THE QUALITY OF ASEPTICALLY PACKAGED GRANGE DRINKS 65 Introduction 65 Materials and Methods 66 Reagents 66 Orange Drinks Composition 66 Preparation of the Mixtures 68 Method of Analysis 7 0 Ascorbic and Dehydroascorbic Acids Determination 70 Browning Measurement 70 Statistical Treatments 71 Results and Discussion 71 Ascorbic Acid Retention 71 Dehydroascorbic Acid Production 74 Browning as Influenced by Ascorbic Acid Concentration 80 Browning as Influenced by Amino Acids Concentration 81 Statistical Analysis 90 Conclusion 93 EFFECT OF AMINO ACIDS CONCENTRATION, PROCESSING, AND STORAGE CONDITIONS ON THE QUALITY OF ASEPTICALLY PACKAGED ORANGE JUICE AND ORANGE DRINKS 94 Introduction 94 Materials and Methods 94 Reagents 94 Orange Drinks Composition 95 Preparation of the Samples 96 Methods of Analysis 99 Ascorbic and Dehydroascorbic Acids 99 Browning 99 Amino Acids Analysis 99 Sensory Evaluation 100 Statistical Analysis 100 Results and Discussion 100 Orange Juice 100 Orange Drinks 107 Conclusion 115 SUMMARY 116 APPENDIX 118 v

PAGE 6

REFERENCES 131 BIOGRAPHICAL SKETCH 143 vi

PAGE 7

LIST OF TABLES PAGE 1 Principal Citrus Fruits: Production for the United States' and Florida, Crop Years 1964-65 through 1983-84... 7 2 Oranges: Production for the United States and Florida. Crop Years 1960-61 through 1983-84 '. 7 I 3 U.S. Citrus Per Capita Consumption 9 4 Florida Oranges: Production, Utilization and Value for Crop Years 1964-65 through 1983-84 12 5 Permeability of the Packaging Film 4" 6 Corxiarison of the Fluorometric and the Dye Titration Procedures for Ascorbic Acid Determination 43 7 Ascorbic Acid and Dehydroascorbic Acid Content of Various Foods 56 8 Recovery of Ascorbic Acid and Dehydroascorbic Acid from Spiked Samples 59 9 Composition of the Orange Drink Mixtures 68 10 Initial Ascorbic Acid and Dehydroascorbic Acid Levels for Orange Juice and Orange Drink Products in Retort Pouch and Polyethylene Pouch 72 11 Three-Factor Analysis of Variance of the Browning Data at 8 Weeks 90 12 Composition of Orange Drinks 9S 13 Processing and Storage Conditions of Orange Juice and Orange Drinks 98 14 Rate Constants (Weeks *-) for Ascorbic Acid Loss as a Function of Amino Acid Content and Storage 'Conditions in Orange Drinks 110 15 Changes in Amount of Amino Acids in Orange Drinks M2 and >G as Influenced by Storage Time 113 A-l Absorbance at 420 in as a Function of Storage Time (samples stored in retort pouch) 121 • • VIA

PAGE 8

A-2 Absorbance at 420 ran as a Function of Storage Time (samples stored in polyethylene pouch) 122 A-3 Ascorbic Acid Concentration as a Function of Storage Time (samples stored in retort pouch) 123 A-4 Ascorbic Acid Concentration as a Function of Storage Time (samples stored in polyethylene pouch) 124 A5 Dehydroascorbic Acid Concentration as a Function of Storage Time (samples stored in retort pouch) 125 A6 Dehydroascorbic Acid Concentration as a Function of Storage Time (samples stored in polyethylene pouch) 126 A7 Absorbance at 420 nm as a Function of Storage Time (samples stored in Terra Pak carton) 127 A-8 Ascorbic Acid Retention as a Function of Storage Time (samples stored in Tetra Pak carton) 128 A9 Flavor Score as a Function of Storage Time (samples stored in Tetra Pak carton) 129 A10 Sensory Evaluation Form 130 viii

PAGE 9

LIST OF FIGURES PAGE 1 Florida Citrus Production 4 2 Principal Citrus Fruits: Production for United States and Florida, Crop Years 1960-61 through 1983-84 6 3 Oranges: Production for the United States and Florida, Crop Years 1960-61 through 1983-34 8 4 Initial Stage of Carbonyl-Amine Reactions 18 5 Amadori Rearrangement 19 6 Major Pathway for Carbonyl-Amine Reactions 20 7 Minor Pathway for Carbonyl-Amine Reactions 21 8 Initial Stage for Degradation of Ascorbic Acid 25 9 Degradation Pathway for Diketogulonic Acid 26 10 HPLC System with Post-Column Derivatization and Tandem Ultraviolet and Fluorometric Detection 53 11 Typical HPLC Chromatograms of Orange Juice and Parsley, Monitored by Tandem Ultraviolet (uv, 254 nm) and Fluorometric Detection 57 12 Typical HPLC Chromatograms of Tomato and Strawberry Monitored by Tandem Ultraviolet (uv,254 nm) and Fluorometric Detection 58 13 Resolution of AA, DKAA, and DXGA by HPLC 62 14 Ascorbic and Dehvdroascorbic Acids Monitored at 210 and 254 nm 63 15 Absorption Spectrum for AA, DHAA, and DKGA 64 16 Effect of Amin o Acid Concentration on Ascorbic Acid Retention (initial level of ascorbic acid 71.8 mg/100 ml, samples stored in polyethylene pouch) 76 ix

PAGE 10

17 Effect of Amino Acid Concentration on Ascorbic Acid Retention (initial level of ascorbic acid 71.8 mg/100 ml, samples stored in retort pouch) 77 18 Effect of Amino Acid Concentration on Ascorbic Acid Pretention (initial level of ascorbic acid 38.0 mg/100 ml, samples stored in retort pouch) 78 19 Effect of Amino Acid Concentration on Ascorbic Acid Retention (initial level of ascorbic acid 4.2 mg/100 ml, samples stored in retort pouch) 79 20 Effect of Ascorbic Acid Concentration on Browning of Orange Drinks with 0.06% Amino Acids 82 21 Effect of Ascorbic Acid Concentration on Browning of Orange Drinks with 0.66% Amino Acids 83 22 Effect of Ascorbic Acid Concentration on Browning of Orange Drinks with 1.26% Amino Acids 84 23 Effect of Ami no Acid Concentration on Browning of Orange Drinks (4.2 mg/100 ml AA) 86 24 Effect of Amino Acid Concentration on Browning of Orange Drinks (38.0 mg/100 ml AA) 88 25 Effect of Amino Acid Concentration on Browning of Orange Drinks at 71.8 mg/100 ml AA 89 26 Effect of Ascorbic Acid and Amino Acid Concentration on Browning of Orange Drinks at 8 Week Storage 92 27 Ascorbic Acid and Dehydroascorbic Acid Concentration in Orange Juice as Influenced by Processing and Storage Conditions .... 101 28 Ascorbic Acid Retention (Log Scale) in Orange Juice as Influenced by Processing and Storage Conditions 102 29 Browning Formation in Orange Juice as Influenced by Processing and Storage Conditions 105 30 Flavor of Orange Juice as Influenced by Processing and Storage Conditions 106 31 Ascorbic Acid and Dehydroascorbic Acid Concentration of Orange Drinks as Influenced by Amino Acid Content and Storage Conditions 108 32 Ascorbic Acid Retention (Log Scale) in Orange Drinks as Influenced by Amino Acid Content and Storage Conditions 109 x

PAGE 11

33 Browning Formation in Orange Drinks as Influenced by Amino Acid Content and Storage Conditions 112 34 Flavor of Orange Drinks as Influenced by Amino Acid Content and Storage Conditions 114 A-l Effect of Amino Acid Concentration on Ascorbic Acid Retention (initial level of ascorbic acid 30.0 mg/lOO ml, samples stored in polyethylene pouch 119 A-2 Effect of Amino Acid Concentration on Ascorbic Acid Retention (initial level of ascorbic acid 4.2 mg/100 ml, samples stored in polyethylene pouch 120 xi

PAGE 12

LIST OF A3HREVTATICNS AA ascorbic acid AER aerobic ANA anaerobic ANOVA analysis of variance a w water activity Cpd compound DHAA dehydroascorbic acid DKGA diketogulonic acid DP 3-deoxy-L-pentosone FA furoic acid FDA Food and Drugs Administration HFCS high fructose corn syrup HPLC high performance liquid chromatography MMT million metric tons MT million tons OJD orange juice deaerated OJND orange juice nondeaerated OPDA nrthnphpTrl pnpH i prr.i np RDA recommended daily allowances SD standard deviation SSOJ single strength orange juice xLi

PAGE 13

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 STORAGE STABILITY OF ASEPTICALLY PACKAGED SINGLE STRENGTH ORANGE JUICE AND ORANGE DRINKS By Bechir Kacem August 1936 Cnairman: R.F. Matthews Cochairman: M.R. Marshall Major Department: Food Science and Human Nutrition Nonenzymatic browning of single strength orange juice, and synthetic orange drinks containing 10% orange juice and various combinations of ascorbic acid, amino acids (aspartic acid, arginine, and 4-aminobutyric acid) sugars citric acid and potassium citrate has been studied under aerobic and anaerobic conditions. The juice and the drinks were aseptically packaged and stored at 75F for up to 20 weeks. In the presence of oxygen, ascorbic acid was found to be the most reactive constituent in the darkening of orange drinks. The presence of amino acids at the high level (1.261) increased significantly the rate of ascorbic acid degradation and the rate and extent of browning pigments formation. However, reducing the amino acids level from 0.66 to 0.06% had no significant effect on the browning of orange drinks stored under anaerobic conditions, but did significantly affect browning when samples were stored aerobically. The study also showed that the storage of aseptically packaged orange xiii

PAGE 14

juice in anaerobic jars, as compared to aerobic storage, resulted in higher ascorbic acid retention. However, there was no significant difference in the sensory evaluation, nor in the amino acids concentrations after 16 week storage at 75F. In addition, a high-performance liquid chromatography (HPLC) procedure has been developed for the rapid and simultaneous estimation of ascorbic and dehydroascorbic acids in fresh fruits and vegetables. Isocratic separation of these components was accomplished by anion-exchange chromatography using acetonitrile : 0 05M KH2FO4 (75:25, v/v) as eluant. The concentration of ascorbic acid was determined by monitoring its absorbance at 254 ran, while dehydroascorbic acid detection was achieved by fluorescence as a result of post-column derivatization involving the condensation of dehydroascorbic acid with o-phenylenediamine, forming a highly fluorescent quinoxaline derivative. The procedure allows detection of both forms of vitamin C at levels well below those usually found in orange juice, and was used to follow the rate of change of ascorbic acid into dehydroascorbic acid in orange juice and orange drinks during storage. xiv

PAGE 15

INTRDDUCTICN Citrus products during processing and storage at room temperature are susceptible to a number of deteriorative reactions, which result in the development of off-flavor, Such off-flavor is generally acconpanied by other changes, in particular, browning of the product and loss of nutritional value. This type of discoloration called nonenzymatic browning is one of the most disturbing problems for citrus industries. It is the main reason for the reduction in commercial value and consumer rejection of citrus products, and has been the subject of research for many years. Many different types of reactions lead to the discoloration of the product at moderate temperatures. This change in color may occur through formation of dark pigments by the breaking down of certain constituents such as ascorbic acid, by reaction between some constituents present in the juice product or by reaction between some constituents of the juice with oxygen in the air. Factors which can influence the nature of the degradation mecnanism include temperature, oxygen, amino acids, metal catalysts, pH, and sugar concentration. Various hypotheses have been developed to explain the mechanism of browning in citrus juices. Among these, ascorbic acid degradation is thought to be the major pathway responsible for browning in citrus products. 1

PAGE 16

2 However, in spite of the many reviews of the subject, there has been no comprehensive organization of the reactions involved. The work reported is contradictory in nature, and there is still a lack of understanding concerning the fundamental factors involved in the deterioration of packaged orange juice. It is not known whether decomposition of ascorbic acid alone or decomposition in the presence of sugars and amino acids of orange juice is more important than the Maillard reaction in browning. A clear knowledge of the reactions involved in this deterioration and the roles played by the various constituents of the juice is necessary before an entirely satisfactory method for controlling nonen2ymatic browning in citrus can be realized. In view of this lack of understanding it was decided to carry out this research project to study the roles of ascorbic acid concentration, amino acids concentration, juice concentration, oxygen concentration, storage and processing conditions, and storage time in the browning of aseptically packaged orange juice and orange drinks containing various combinations of ascorbic acid, amino acids, sugars, citric acid and potassium citrate. In addition, the development of an analytical procedure to follow the rate of change of ascorbic acid into dehydroascorbic acid in orange juice and orange drinks during storage using the HPLC technique was investigated.

PAGE 17

LITERATURE REVIEW A Brief Historv of Citrus Distribution Citrus trees first appeared in Chinese gardens during the 13th to 16th centuries. From this area the Arab traders propagated the fruits progressively with their conquests of the Middle East, Near East, North Africa and Spain. The fruits were introduced into North America from the Canary Islands by Christopher Columbus during his second voyage in 1493 (Cooper and Chapot, 1977) The exact date of the introduction of citrus trees into Florida is not known. Ziegler and Wolfe (1961) pointed out that oranges were likely brought into Florida at the time the colony at St. Augustine was established in 1565. Seeds were probably then scattered throughout the State by Indians. Citrus Production in Florida The first recorded citrus crop was noted in a report by the Governor of St. Augustine in 1579, but commercial production and trade did not begin to develop on a large scale until after the Civil War. In 1886-87, the U.S. Department of Agriculture reported a total citrus production of 1.26 million boxes (Figure 1). The "Great Freeze" of 1894-95 almost destroyed the entire citrus industry of the state. It was not until 1903-04, that this level was reached again. Since that time the volume has steadily increased to reach a maximum of 283.6 million boxes In 1979-80 (Fla. Crop and Livestock Reporting Ser. Citrus Summary, 1985). 3

PAGE 18

4

PAGE 19

5 As the decade of the sixties began, Florida's dominance of the nation's citrus became more apparent, and in the 1983-8^ season Florida accounted for more than 69% of the U.S. production of citrus fruits (Figure 2 and Table 1) Even though the industry suffered crippling natural disasters ranging from winter freezes to canker, Florida's 1984-85 citrus crop has been valued at a record $1.04 billion in a preliminary report (Citrus Valued at Record $1.4 billion, 1985). Although the amount of oranges harvested for 1984-85, 103.9 million boxes (4.6MT), is the smallest yield since 1967-68, higher prices pushed the value to the record level. The total citrus harvest, including oranges, grapefruit, lemons and limes was 158.9 million boxes (7MT) The 1979-80 harvest held the previous record high of $ 1.03 billion, according to the Florida Crop and Livestock Reporting Service of Orlando. The latest report issued by the service shows a 23% increase in value over the 1983-84 harvest of $ 849 million. Orange production increased from about 2.5 million tons in 1963-64 to over 8 million tons in the 1975-76 season, and presented 78% of the U.S. total production (Figure 3 and Table 2).

PAGE 20

6

PAGE 21

7 Table 1 Principal Citrus Fruits: Production for the United States and Florida, Crop Years 1964-65 through 1983-84 (1,000 Tons) Year United states riorxca 7= 196465 7,633 5,480 71.8 196566 8,768 6,242 71.2 1970-71 11,919 8,786 73.7 1975-76 14,788 10,943 74.0 1980-81 15,105 10,470 69.3 198283 13,608 8,513 62.6 198384 10,741 7,436 69.2 Source: Adapted from Fla. Crop and Livestock Reporting Serv. (1935). TABLE 2 Oranges: Production for United States and Florida, Crop Years 1960-61 through 1983-84 Production: 1,000 tons \ Crop year J 0 United States \ Florida I 1963-64 3,732 2,469 66 1965-66 5,812 4,316 1 74 1970-71 8,205 6,402 78 1975-76 10,493 8,154 78 1980-81 10,487 1 7,758 74 1932-83 9,519 6,282 I 66 1983-34 I 7,238 5,252 \ 73 Source: Adapted from Fla. Crop and Livestock Reporting Service, (1985).

PAGE 22

Z2 in \ IT) > CO O zr. I CD CD \ 1 s 00 < 9 S 1 DO (U U tc -i C Cu m o o u CO i

PAGE 23

9 Citrus Demand The per capita consumption has increased from 22.2 pounds in 1920 to almost 118 ponds in 1980, more than a fivefold increase (Table 3). Table 3 U.S. Citrus Per Capita Consumption. Fresh Weight Equivalent (pounds) 1940 1960 1980 b re sn" ,7 T'T" r T 52. 1 30.7 •)£ 1 Processed _. — ^ 10.4 52.2 91.2 Total a „ in o 62.5 82.9 117.5 Excludes lemons and limes. None reported. There has been a major shift in the form of the product demand. Per capita consumption of fresh citrus increased during the early 1900s until the mid-1940s. Since 1940, fresh per capita citrus consumption has declined by about 50%, while processed consumption increased ninefold (Gunter, 1983). Introduction of canned juice in the 1920 's and frozen concentrate in the mid 40' are the major factors contributing to the growth in processed citrus juice demand.

PAGE 24

10 Citrus Processing National Before canned orange juice was introduced commercially in 1929, only 1% of the domestic crop was processed. By 1945-47, an average of over 1.4 MKT (million metric tons) of oranges and tangerines were processed with almost the entire volume going into juice related products. This volume represented about 1/3 of the total domestic crop which was reported at 4.6 MMT. By 1959, about 3.5 MMT of oranges and tangerines were processed into juice products representing nearly 64% of the 5.6 MMT crop that year. The percent of processed citrus fruits continued to increase and by 1983-84, processors used 67% of the total citrus crop of 9.74 MMT. They used 74% of the orange production, 53% of the grapefruit and 46% of the lemons compared with 76%, 47%, and 54% respectively, in 1982-83. (Citrus Fruits, 1984). Florida In Florida, citrus processing represents the largest food processing industry, producing 73% of the total United States citrus products (Fla. Crop and Livestock Reporting Service, 1985) More than 90% of the Florida's orange crop is processed with 80% utilized for frozen concentrate orange juice, 12% for chilled juice, 3% for canned single strength orange juice; 4% is sold in the fresh market and less than 1% is used in products like sections, salads, and blends. The equivalent of 231 million 45Brix gallons of concentrated juice were produced during 1979-80 season (Fla. Crop and Livestock Reporting Service, 1980) Table 4 illustrates the upward trend in processed oranges compared to declining trends for fresh oranges in Florida.

PAGE 25

11 Juice Packaging Juice and juice drink products have been distributed in three forms, she If -stable products, chilled juices, and frozen concentrates. Single strength citrus juice became available in the early 1930 s During the period from the early 1930 1 s to World War II, commercial flash pasteurization of juice was developed. This allowed processed juices to retain some of their natural fresh aroma and flavor during heat processing. The availabilty of high quality juices in the form of frozen juice concentrates occurred in the 1940' s. Prior to the introduction of this new concentrating technology to the United States, only a third of the domestic orange crop was processed into juice. By the late 1940s, with the availability of the improved technology for frozen concentrates, nearly two-thirds of the orange crop was being processed into juice, with the largest percent being used to produce high-quality, frozen orange juice concentrate. The success of these products is attributed to the fact they offer convenience, low cost and are available year round. However, until recently, the shelf-stable juices and juice drinks have shown a lack of growth in overall consumption compared to juices that were distributed as frozen concentrate or in refrigerated forms. This is because of the higher quality that these "cold" methods of distribution could deliver to the juice and juice products.

PAGE 26

12 Table 4 Florida Oranges: Production, Utilization and Value for Crop Years 1964-65 through 1983-84 Crop year 1963-64 1965-66 1970-71 1975-76 198081 198182 198283 198384 Utilization ot Production 1,000 hoses Total Fresh Processed 54,90 95,90 142,30 181,20 172,40 125,80 139,60 116,70 11,94 15,38 13,96 11,73 8,28 7,62 10,32 7,64 42,96 80,52 128,34 169,47 164,12 118,18 129,28 109,06 7 SO Prcces-sed 78 84 90 94 95 94 93 93 OnTree Value of Production 1,000 dollars 243,935 155,625 208,146 321,449 697.231 538 ,'686 718.420 578^54 Source: Adapted from Fla. Crop and Livestock Reporting Service, (1985). The traditional method of processing a shelf-stable juice product involves the hot filling of the juice in cans or glass bottles. This method of processing has some difficulties if one wishes to achieve a high-quality product. In this type of processing and packaging operation, juices are commonly subjected to a severe heat treatment for relatively long periods of time. The usual hotfilling process involves heating and filling a rigid container above 170-l80oF (76.7-82.2C) followed by a cooling period which requires 10 to 20 minutes for the juice to reach ambient temperature. The severity of this type of processing has often decreased the overall quality of juices processed in this manner. This method of processing has also discouraged the introduction of other qualityimproving technologies.

PAGE 27

13 .Any potential improvement that could be made in the quality of the juice delivered for bottling or canning was quickly lost due to the severe heat treatment necessary for the hotfilling operation. Another problem affecting the quality of juice products packed in cans is the interaction between the juice and the metal containers. Problems such as the "pick up" of tin by juices in unlined containers contribute to offflavor of the consumed product. It is also a common practice to select the premium juices for use in chilled and frozen products, and consider shelf-stable juice products as unaffected by long periods of ambient storage, hence disregarding the inherent degradative changes that occur in many fruit juice products when stored at ambient temperature for long periods. Until recently the most common package style used in processing citrus shelf-stable juice products was the metal container. The tin container guaranteed long shelf life and package integrity. The second most important package, the glass bottle, has the same benefits as the tin can. Both products do not require refrigeration in distribution but the containers added significantly to total freight costs due to their heavy tare weight. The energy crisis of the mid 1970' s has dramatically influenced this situation. Although both cans and glass bottles are still used, the trend of the citrus industry is now moving toward low cost, light-weight, limited shelf life packages. The most recent development is the clearance of the "Terra Brik" package for use in aseptic packaging of liquid products.

PAGE 28

14 Aseptic Processing Aseptic processing is a processing and packaging technique by which a commercially sterilized product is put into a pre sterilized, hermetically sealed container in a sterile environment. To produce a commercially sterile product, an aseptic system must meet three basic requirements : -The product must be sterile. -The package in which the product will be placed must be sterile. -The environment in which the product and package will be brought together must be sterile. Commercial sterility is a term used to denote a product that has been processed in such a manner so that it is free of all contaminants that cause the product to spoil or be a detriment to public health. Aseptic processing and packaging is not a new technology. The first commercial system in the U.S. was installed in the 1950' s for use with non-acid products such as milk. However, since the recent (February 9, 1981) FTA regulation permitting the use of hydrogen peroxide and heat as sterilizing agents for aseptic packaging, the interest in aseptic packaging has increased dramatically. This technique is now applied in several food industries to produce microbially stable products. 3y using this technology it is possible to minimize the time the product is subjected to the high temperature necessary to obtain commercial sterility, and produce shelf stable

PAGE 29

15 juices that are equal j_n organoleptic quality to frozen and chilled products, at a more economical price. It is estimated that 1-liter aseptic boxes cost about half as much as cans and 30% as much as bottles (Paper Bottles Are Coming on Stong, 1933). Although some of these savings are surrendered because of the more complicated filling process, it is estimated that the cost of packaging juice concentrate in 8oz "Tetra Brik" boxes is 18% cheaper than filling coventional paper and metal cans (Business Week, January 16, 1984), and since aseptic goods are subjected to a briefer heating treatment during sterilization than canned goods, they have better flavors. In addition a major saving results because they do not require refrigeration during shipment or storage. However, aseptic processing will not be of commercial importance until chemical stability of citrus products is understood and improved. Hie limiting factor in the acceptance and eventual success of aseptically packaged citrus juices is the control of chemical changes that are accelerated during ambient temperature distribution and storage. Nonenzymatic browning reaction is believed to be the quickest and most dramatic quality defect to appear during ambient temperature storage (Clegg, 1964; Tatum et al., 1967; and Tatum et al., 1975). Nonenzymatic Browning The nonenzymatic browning reaction in foods during processing or on storage has long been recognized as one of the most inportant problems of fruit preservation, and has been the subject of research for many decades. This reaction refers to the formation of brown

PAGE 30

16 pigments in foods that cause the product to become brownish to black in appearance, and is usually characterized by undesirable charges in flavor, odor, and nutritive value. In order to study the reaction under controlled conditions, model systems have been widely used, but fall details of the physical conditions employed have not always been reported. The work done before 1948 has been reviewed by Stadtman (1948) Since then there have been excellent reviews by Hodge (1953), Reynolds (1963), Burton and McWeeny (1964), Shaw et al. (1977), and Handwerk and Coleman (1986) which mainly covered the reactions between amines and sugars. Three main theories have been advanced to explain the mechanism of nonenzymatic browning (Stadman, 1948) The Maillard or melanoidin condensation theory: According to this theory (the most common) the reaction involves a condensation of amino acids and reducing sugars and gives rise to the formation of dark colored substances. The activealdehyde theory: It proposes that browning involves the decomposition of sugars and sugar acids to furfuraldehydes or similar compounds characterized by having an active carbonyl group, and that these products condense with nitrogen compounds and/or polymerize to form colored substances. The ascorbic acid theory: According to this theory the most important precursors to browning are ascorbic acid and related compounds, which upon oxidation yield reactive products that may polymerize or react with nitrogenous constituents of the food to form brown pigments. This third theory seems the most likely to apply to

PAGE 31

17 the conditions pertaining to an acidic product such as orange juice; the concentration of ascorbic acid is relatively high and free amino acids are present to combine with the reactive products resulting from the oxidation of the ascorbic acid and lead to the formation of brown pigments. Actually all three of the above mechanisms may be involved in the browning of fruit products. Major research efforts have been conducted to prove the first two hypotheses of nonenzymatic browning, as noted in the review by Shaw et al. (1977), but relatively little research has explored the ascorbic acid theory. It is not known whether decomposition of ascorbic acid alone or m the presence of sugars and amino acids of orange juice is more important than the sugar-amino acids reaction in browning. Let us now look at these 3 theories in more detail. Maillard Reaction Maillard was the first to study systematically the interaction which occurs initially between ami no acids and sugars, and to realize some of their relations to the chemistry of natual products. He found that simple amino acids react on warming with certain sugars, to produce dark-brown products. This explains why the reaction has also been commonly called the browning reaction, nonenzymatic browning, melanoidin formation, or caramelization; and the brown products have been referred to as melanoidins, or humin-like substances (Ellis, 1959) The interaction of amino acids and sugars falls into two general types. The first is the simple controlled condensation of the

PAGE 32

IS reactants; this leads to compounds which are identifiable as Nsubstituted glycosylamines or, occasionally their Amadori rearrangement products. The second is the typical Maillard reaction, which leads to a mixture of products of increasing complexity if conditions are favorable. Reynolds (1970) has outlined the major features of the initial stages of the carbonyl-ainine reaction. At the first step, an aldose or ketose sugar reacts wirh a primary or secondary amine to form a glycosylamine and the reaction is reversible (Figure 4) H-C=0 (H-C0H)4 + R-NH 2 H-CHCH glucose amino acid Figure 4. Initial Stage of the Carbonyl -Amine Reactions The role of water is important in determining the yield of glycosylamine. At low water content there is substantial formation o this compound, (Shallenberger and 3irch, 1975); therefore, carbonylamine browning is believed to be a significant pathway for browning in dried and concentrated foods. The probable mechanism for the formation of glycosylamine is the addition of the amine to the carbonyl group of the open-chain form of the sugar, followed by H-C=N-R I (H-C0H)4 H-CHOHschiffbase R-NH I H-C-* (H-C0H)3 0 H-C 1 I H-CHH glycosylamine

PAGE 33

19 elimination of a molecule of water and subsequent ring closure (Hodge, 1976). The next step is the Amador i rearrangement which involves the protonation of the nitrogen atom at C-l (Figure 5) This rearrangement takes place in either acid or basic solutions. RNH I CH I (HGCH)3 0 I HC I HCHOH + H' Amadori compound enh2 I CH+ ENH II •i CH i I > (KCCH)3 0 > HCCH I I HC 1 (HCCH)3 I I HCHOH HCHOH -H+ HqCNHR hcnhr I II CO > CCH I I (HCOH) 3 < (HCCH) 3 H2COH H2COH Figure 5. Amadori Rearrangement The second stage in carbonylamine reactions is based on the degradation and dehydration of the Amadori compound, which can occur through two major pathways. The major branch leads from the 1,2-eneaminol of the Amadori compound to hydroxymethyl furfuraldehyde (Figure 6) The amino acid may be retained in some molecules throughout the dehydration reactions of this pathway.

PAGE 34

20 This pathway is the major pathway for the production of brown color in foods. The minor branch which probably represents less than 5? 0 of the total sugar decomposition (Hodge and Osman, 1976) begins with the 2,3-enediol of the Amadori compound; then the amino acid is totally eliminated (Figure 7) The degradation compounds formed seem to be important in the production of flavor. -r H?C-N > \ c=o I (HC0H) 3 H2COH <— Amadori cpd. KC-N, COH HON, I COH > (HCai)3 OH" (HOT) 3 + H 2 0 ?C0H racm\ n H 2 C 1,2 eneaminol (HC0H) 2 H2COH HC=0 I CO I HCH (HC0H)2 H2COH -H 2 0 5 hydrojymethyl2-furaldehyde -f amine melanoidins Figure 6. Major Pathway for Carbonyl -Amine Reactions

PAGE 35

21 HoC-N N H?C-Nx CH? CH3 I "I r I CO OOH CCH C=0 I II I I CHC0H)3 ^ CCH C=0 ? 0=0 1 H2CGH (HOCK) 2 (HOCK) 2 (HCCH)2 H2CCH H 2 OCH H2CCH Amadori cpd 2,3-enediol 'f melanoidins CH3 cmethyl reduc tones HCCH I and a-dicarbonyls H2CCH Figure 7. Minor Pathway for Carbonyl -Amine Reactions ActiveAldehyde Theory '•/hen the nonenzymatic browning reactions occur in the absence of nitrogenous compounds, they are described as caramel ization reactions. Under anhydrous conditions upon the application of heat, or at high concentration in dilute acid solutions, the initial stages of the caramelization reaction are characterized by the formation of anhydro sugars. Glucose has been reported to yield glucosan and levoglucosan. The two anhydroglucose compounds are readly

PAGE 36

22 distinguished by their specific rotation (Sliallenberger and Birch, 1975) When sucrose is heated at about 200C, simultaneous hydrolysis and dehydration occur, apparently followed by rapid dimerization of the products, so that a series of compounds characterized by isosacchrosan (corresponding to sucrose minus one molecule of water) is formed (Hodge, 1976). Isosacchrosan has no sweetness, but does have a mildly bitter taste. In dilute solutions of reducing sugars, the initial stages of the caramelization, reactions are a series of events involving enolization, dehydration and fragmentation reactions. Subsequently, polymerization reactions occur, which lead to the formation of pigments similar to those formed in caramelization reactions at either higher temperature or in more concentrated solution. The classic caramelizatin reaction is the phenomenon exibited by sucrose when subjected to heat, and the fundamental reactions are Inversion of sucrose to D-glucose and Dfructose Equilibration of anomeric and ring forms Condensation, intermolecular Condensation, intramolecular Iscmerization of aldoses to ketoses Fragmentation reactions Browning The caramelization reaction is autocatalytic and increasing temperatures net only increases the reaction rate but also alters the qualitative nature of the pigments (Doss and Ghosh, 1949).

PAGE 37

23 The effect of pH is dramatic and the reaction rate at pH 8 is ten times greater than that at pH 5.9 (Ledon and Lananeta, 1950) Caramelization reactions can be used either for coloring purpose, or for flavoring purpose. For flavor, sucrose is caramelized in concentrated syrups. The sugar fragmentation reactions are promoted by alkaline or neutral medium, whereby formation of humic substances is limited to avoid bitter, astringent tastes. Caramel for coloring use is produced in acid medium. Glucose syrup is treated with dilute sulfuric acid, partially neutralized with amGonia, then heated in the presence of a sulfite at a pH value of about 4 (Hodge and Osman, 1976) Ascorbic Acid Theory The exact route of ascorbic acid degradation is highly variable and dependent upon the particular system (Tannenbaum, 1976) Three different mechanisms of degradation of ascorbic acid have been proposed Catalyzed aerobic degradation Uncatalyzed aerobic degradation Anaerobic degradation When oxygen is present in the system, ascorbic acid is degraded primarily via its monoanion (HA~) to dehydroascorbic acid (DHAA) The exact pathway and overall rate is a function of the concentration of metal catalysts (M 1 ***) in the system. In the absence of oxygen the degradation pathway has not been yet verified. Following the suggestion of Kurata and Sakurai (1967), ascorbic acid is shown to react via its keto tautomer, K2A-keto. The tautomer is in equilibrium with its anion, HA" keto, which undergoes delactonization to diketogulonic acid (EKGA) as shown in Figure 8. Further degradation

PAGE 38

24 beyond EKGA is closely related to nonenzymatic browning in some food products. The mechanism of these reactions has been studied by Kurata and Skurai (1967) under oxidative and non-oxLdative conditions. One of the characteristic differences between the two pathways is that f ur f u r a l is much more easily produced through the nonoxidative reaction. Under non-oxidative conditions, ascorbic acid in an acid solution was degraded to furfural with the formation of 3-deoxy-L-pentosone as an intermediate. This acidcatalyzed degradation took place under the storage or the cooking conditions of foodstuffs. It was shown that aldopentoses and 2-keto-L-gulonic acid themselves were not intermediates of the reaction. On the basis of their data, they assumed that the first step of the non-oxidative degradation of ascorbic acid in an acid condition is hydrolysis of the lactone ring followed by decarboxylation and dehydrations forming 3-deoxy-L-pentosone and furfural. Factors Affecting the Browning of Packaged Orange Juice Role of Ascorbic Acid The primary oxidation product of L-ascorbic acid (AA) is dehydro-L-ascorbic acid (EHAA) The latter is not a stable compound; it undergoes spontaneous hydrolysis to a second product which has been well characterized as 2,3-diketo-L-gulonic acid (EKGA), formed by the opening of the lactone ring of dehydroascorbic acid. The first stage of oxidation to DRAA is reversible and the biological activity is retained; however, the oxidation of DHAA to EKGA is not reversible and the biological activity is lost.

PAGE 39

25 Figure 8. Initial Stage for Degradation of Ascorbic Acid (From Tannenbaum, 1976)

PAGE 40

26 Figure 9. Degradation Pathway for Diketogulonic Acid (From Tannenbaum, 1976)

PAGE 41

26 Figure 9. Degradation Pathway for Diketogulonic Acid (From Tannenbaum, 1976)

PAGE 42

27 Further degradation of DK.GA yields active products (such as furfural) which undergo polymerization, or react with nitrogenous constituents yielding brown pigments (Figure 9) These reactions may occur under aerobic as well as anaerobic conditions. For many years a relationship between browning in orange juice and AA loss has been known (Lamden and Harrisl, 1950; Joslyn and Marsh, 1935; Joslyn et al., 1934; Loeffer, 1941; Moore et ai. 1942; Stephens et al., 1942; Joslyn, 1941; Curl et al., 1946; Curl, 1948; Clegg and Morton, 1965; and McWeeny and Burton, 1963) However, the exact role of AA in the discoloration of fruit products is still not well understood. It has been shown that when a citrus juice darkens carbon dioxide is evolved and AA is lost (Hamburger and Joslyn, 1941; Joslyn, 1957; Joslyn et al., 1934; Wilson, 1928; and Curl et al., 1946). This observed loss in AA during browning has led to the suggestion that AA is involved in the browning reaction in one of two ways: (a) It may serve as an antioxidant, being oxidized in preference to other substances present which upon oxidation yield compounds or precursors of dark compounds, (b) It may be oxidized along with other reducing constituents, such as the flavonols, as suggested by Szent Gyorgyi (1937), to yield different compounds that are the actual precursors of dark pigments. Moore et al. (1942) showed that addition of AA to orange juice resulted in a marked increase in the rate of browning when juice is stored in the presence of oxygen. Similar results were obtained by Beattie et al. (1943) when they added AA (50mg/lCOmi) to strawberry juice. These data show convincing evidence that AA is involved in the browning of orange juice, and it is effective not ioxidant as

PAGE 43

23 reported by Hamburger and Joslyn (1941) but rather as an intermediate in the browning reaction; otherwise, the addition of AA should retard rather than accelerate browning. Wilson (1928) reported accumulation of carbon dioxide during the storage of sterile orange concentrate, and suggested that it may arise from the decomposition of AA. However, Eddy (1936) detected no carbon dioxide formation in his experiments on the oxygen absorption of unprocessed orange juice and AA solutions. Curl (1947), working on storage of pasteurized, concentrated orange juice at 26.7C and above, noticed the formation of gas, which was shown to be essentially carbon dioxide. This gas does not result from fermentation. A number of possible sources of the carbon dioxide have been suggested, including the Maillard reaction between amino acids and sugars, decomposition or oxidation of AA, and a chemical breakdown of sugars. Later Curl and Veldhuis (1948) reviewed this subject and presented data showing that the formation of gas is accompanied by almost total losses of AA, significant losses of total sugars, and considerable darkening. Loeffer (1941) found that the quantity of carbon dioxide produced in canned orange juice during five month's storage at 35C was almost equivalent to the quantity of AA lost. Lahikainen et al. (1958) showed, in model systems containing AA., glycine, and citrate buffers, and at concentrations corresponding to those found in fruit juices, that the carbon dioxide evolved during browning was not produced bv decarboxylation of the glycine. This finding indicated that glycine had not undergone a Strecker degradation, although AA has been regarded as an active agent (Schonberg and Moubaker, 1952)

PAGE 44

29 Lahikainen et al. (1958) reported that the rate of production of carbon dioxide was linearly related to the rate of browning at 37C, but not at 50C. This is in agreement with the finding of Joslyn, (1957) that a change in reaction mechanism may occur in AA systems between 30 and 50C, and was later confirmed by Nagy and Smoot (1977) and Kanner et al. (1982). It is possible that pigment may be formed at higher temperatures by a separate mechanism or course of reaction than at the lower temperatures, or a second pigment producing reaction may be activated at the higher temperature. The total carbon dioxide produced was in excess of that required by the mono-decarboxylation of ascorbic acid. Earlier, Laraden and Harris (1950) found that in solutions of AA and citric acid there was no loss or breakdown of citric acid during the browning of the mixture when heated in closed tubes at 100C. They found, also, that the quantity of carbon dioxide evolved from a 2.57o AA, 507=, citric acid solution in five hour at 60C was only 13.57o of the destroyed ascorbic acid. Jackson et al. (I960), using AA in acetate buffer, showed less than 4% of evolved carbon dioxide was derived from the acetate buffer. These findings indicated that evolution of carbon dioxide was due primarily to the multiple decarboxylation of ascorbic acid. Joslyn (1957) reviewed earlier work on browning in model systems containing AA, and on the parr played by AA in browning of orange juice. He reported that AA was the most reactive component in browning in systems containing AA, amino acids, and sugars. He observed that the rate of browning in orange juice was significantly

PAGE 45

30 reduced when the anionic constituents were removed. However, the removal of the cationic constituents had less effect. The rate of AA destruction in citrus products depends on several factors, including oxygen, "storage temperature, pH value, metal catalyst and concentration of ascorbic acid itself. Effect of oxygen It is a generally accepted fact that oxygen has a pronounced effect on the rate of AA degradation and browning in orange juice. The processing operations involved in the citrus industry, such as extraction, screening, mixing and blending, increase the dissolved oxygen concentration of the products. The presence of oxygen is known to accelerate the AA destruction and the browning reactions. However, it has also been reported that the removal of molecular oxygen did not prevent loss of AA and darkening completly; and it has been shown that the AA degradation can proceed either aerobically or anaerobically. (Boyed and Peterson, 1945; Kef ford et al., 1959; Nagy and Smoot, 1977; Nagy, 1980). One of the characteristic differences between these two reactions is that the anaerobic destruction is generally believed to proceed at a slower rate, with much easier production of furfural. An excellent review of both types of degradation has been published by Reynolds (1965) In early studies by Gore (1915) and McDemott (1916) it was observed that the color of citrus juices was more stable in the absence of oxygen. Matthews (1928) noted that samples of orange juice stored in air darkened more rapidly than those stored in oxygenfree gas atmospheres. Clark (1941) discussed the effect of an excess of oxygen in the sealed containers on color and flavor changes and reduction in

PAGE 46

31 AA with possible darkening of the juice. Moore et al. (1942) were able to show that the rate of browning in pasteurized bottled orange juice was correlated with the volume of airfilled head space; as the volume of air was increased the rate of browning was also increased. Joslyn and Marsh (1932, 1934, 1935) and Joslyn et al. (1934) followed changes in vitamin C and browning in orange juice exposed to air. Tney reported that the browning of orange juice involves oxidation, and showed a decrease in vitamin C with an increased exposure to oxygen at room temperature. Browning was parallel to the loss of vitamin C, suggesting a possible relationship between the two processes. Tressler et al. (1939) further commented that darkening was more rapid in the presence of oxygen. It also proceeded rapidly even when juice was deaerated to remove dissolved oxygen, and stored in vacuum sealed containers. Joslyn and Marsh (1935) concluded that the primary products of oxidation apparently undergo a condensation in which secondary reactions, probably of the amino acid sugar type, occur. In this connection, Joslyn (1941) suggested that the original, almost momentary, contact of orange juice with air before dearation may result in the formation of compounds (possibly peroxides) which supply oxygen for the later deterioration of juice even in the absence of oxygen. A number of papers have since appeared discussing the relationship of oxygen to quality in orange juice. Johnson and Toledo (1975) found that aseptically packaged 55 Brix orange concentrate lost 68% of its ascorbic acid content and was unacceptable after one week if oxygen was left in the package headspace. Completely

PAGE 47

32 eliminating oxygen by storing the orange concentrate in glass containers with zero headspace considerably reduced AA degradation, and greatly extended the shelf life, but browning and flavor degradation still rendered the product unacceptable after ten weeks at 24C. This does not exclude the possibility that the original, almost momentary contact of orange juice with air before dearation may form peroxides which supply oxygen for the later deterioration of the juice as previously suggested by Joslyn (1941) The non-oxidative decomposition was studied at 30C, and 100C, from pH 2.2 to 6.0 (Huelin, 1953). In citrate-phosphate buffer the reaction proceeded most rapidly at pH 3-4, and was accelerated by fructose. Furfural and carbon dioxide were the main products of decomposition at high temperatures or acidities. With lowering of temperature or acidity other products became important. Cier et al. (1959) studied the anaerobic decomposition in an unbuffered solution of pH 2-6.6, and found L-xylose as well as carbon dioxide and furfural. However, xylose gave only traces of furfural at this pH, and did not appear to be the major intermediate between AA and furfural. Kurata and Sakurai (1967) studied the decomposition of AA at pH 2.2 and found 3-deoxy-L-pentosone and furfural, claiming the former as an intermediate. Coggiola (1963) also found acid products, and identified the major acid as 2,5 dihydro-2-furoic acid. Tatum et al. (1969), who used an unbuffered solution of AA and did not exlude oxygen, obtained 15 compounds as degradation products from AA, of which only furfural and 2,5-dihydro 2-furoic acid had been previously shown to come from non-oxidative decomposition of AA (Kamiya, 1959)

PAGE 48

33 They suggested the condensation of furfural probably to be responsible for the formation of the two of the major components. The kinetics of the oxidation reaction of AA have been reported by many researchers Joslyn and Miller (1949a, b) reported the oxidation of AA. in sugar solutions to be essentially first order with respect to the AA oxidation. Under conditions of limited oxygen, the sugar solution showed reduced initial rates of oxidation. Khan and Mart el (1967) reported that the rate of the spontaneous oxidation of AA in an aqueous model system was proportional to the concentration of molecular oxygen down to 0.2 atm of oxygen. They also reported that the rates of ferric and cupric ion catalyzed AA oxidation were first-order with respect to the molecular oxygen concentration. Singh et al. (1976), working with infant formula, concluded that when dissolved oxygen is present in abundant supply, the reaction can be considered to follow a first order kinetics. However, with limited oxygen, (0.1-8.7ppm) the reaction followed second order kinetics. Eison-Percbonok and Downes (1982) studied the autoxidation of AA by varying temperature and AA and dissolved oxygen concentration. They reported that AA autoxidation is dependant on the dissolved oxygen concentration. However, Robertson and Samaniego (1986) observed no significant effects on the rate of asorbic acid degradation and furfural formation that could be attributed to the different initial oxygen levels in lemon juices. In summary one can say that there is considerable evidence that one important consequence of exposure to oxygen is the oxidation of AA which in turn is In some way responsible for browning of orange juice.

PAGE 49

34 Effect of temperature One of the most important factors influencing the rate of AA degradation is temperature. Many studies (Brenner et al., 1948; Freed et al. 1949; Bissett and Berry, 1975; Nagy and Smcot, 1977) have shown losses of ascorbic acid to be related to storage temperatures. VcriLoesecke et al. (1934) reported that packs of orange juice in glass darkened if stored at temperatures of 27C or higher, but not at 16C or below. They further reported that the flavor of the juices in bottles and citrus-enameled cans was superior to that packed in plain tin cans. Pederson et al. (1941) observed a relationship of the temperature of storage to changes in flavor, color, and AA content. Loeffler (1941) stated that less than two months storage at hot summer temperatures could make unpalatable the best quality of glass-packed orange juice, but that the flavor of the freshly bottled juice could be retained almost indefinitely at storage conditions around 4C. He further remarked that the changes in bottled orange juice under warm storage temperature (35C) were of the same order as those found by Vonloeseke et al. (1934) for samples packed in enameledtin cans. Stephens et al. (1942) found the effect of storage temperature on the rate of darkening and stability of flavor to be roughly proportional to its effect on AA stability. This observation was later confirmed by Curl et al. (1946) who observed a parallel effect with respect to carbon dioxide production and thereby contributed support to the suggestion that carbon dioxide formation in these products is derived from a breakdown of AA (Joslyn et al., 1934; Nelson and Hotter, 1933). Curl (1947), working with concentrated

PAGE 50

35 orange juice, snowed that the changes occurring at 27 and 49C were similar, but the rate of change was about 20 times as fast at the higher tenperature. 3renner et al. (1948) and Freed et al. (1949) studied the retention of vitamin C in canned single strength orange juice at 21. 1, 32.2 and 37.8C. These workers concluded that the logarithm of vitamin C retention was inversely related to storage time at these three temperatures. No statistical treatment was applied to their data to confirm their interpretation. Other investigators (Evenden and March, 1948; Joslyn and Miller, 1949b; Huelin, 1953) assumed that vitamin C degradation in orange juice was a first order reaction with the rate of degradation proportional to concentration. These findings differ significantly from those reported by Nagy and Smoot (1977) who found a nonlinear relationship between log percent vitamin C retention and time at high tenperature. Their Arrhenius plot showed two distinct temperature regions. They defined a critical region between 22 and 26.7C above which storage of juices resulted in an accelerated rate of vitamin C breakdown. For grapefruit juice the activation energy (Ea) was 18.2 kcal/mole, and the reaction was first order. For orange juice, two (Eas) were determined 12.8 kcal/mole in the temperature range 4-28C, and 24.5 kcal/mole in the range 28-50C. The change in reaction kinetics was attributed to different destruction mechanisms, although no explanation was offered. Kanner et al. (1982), studying the storage stability of orange juice concentrate packaged aseptically, showed that degradation of ascorbic acid follows first order reaction kinetics at temperatures of 25C and

PAGE 51

36 below. At 36C the degradation of ascorbic acid did not follow a first order reaction. These data are in good agreement with the results cf Nagy and Smoot (1977) on AA degradation in stored canned single strength orange juice, but differ from those of others (Brenner et al.,1948; Huelin, 1953), who found a first order reaction of AA degradation until 40C or higher temperatures. Apparently, results differ because of the long storage time. During this period many breakdown products develop from juice constituents, which seem to affect and accelerate the degradation of AA (CI egg, 1954; 1966). Role of metal catalysts Metal ions can affect the deterioration of citrus fruits in two different ways: (a) metal ions may affect the browning pigment formation. Joslyn and March (1935) studied the effect of metal catalysts on the browning of orange juice, and reported that ferrous ions increased browning, stannous ions decreased it, while other metallic salts (including ferric, stannic, and copper salts) had no effect. Curl (1948) has shown that browning of a sugar-ascorbic acid system was increased by the presence of trace elements. Jackson et al. (1960) concluded that the addition of iron or copper decreased the rate of browning of aerated AA solutions buffered at pH 7, although the rate of loss of AA was increased. In the absence of heavy metals, DHAA and DKGA browned at a slightly faster rate than AA. (b) Metal ions may affect the ascorbic acid degradation. The catalytic properties of heavy metals on the oxidation of AA., notably copper and iron have been extensively discussed in the literature. Weissberger and LuValle (1944) reported that only the monoanionic AA species was susceptible to copper

PAGE 52

37 catalysis. Later, studies by Khan and Martell (1967) showed the oxidation of AA in solution to be linearly dependant on the concentration of copper and iron ions. Shtamn and Shuriator (1974), and Jameson and Blackburn (1975) have also reported the catlytic properties of copper ions in the degradation of AA in solution. More recently, Dennison and Kirk (1982) studied the Influence of trace mineral fortification on the storage stability of AA in a low moisture model food system as a function of water activity at a constant storage temperature of 30C. They reported that AA degradation increased with increasing copper and iron levels. This concentration dependance is in accord with the results of Khan and Martel (1967) and Ogata et al. (1968). It is somehow surprising that metal ions such as copper decrease browning, and at the same time increase the conversion of the AA present to DHAA and DKGA. This negative effect of copper placed some doubt on the correlation of the darlcening of orange juice with loss of AA. Jackson et al. (1960) reported that metal ions in the presence of oxygen increase the conversion of AA to DHAA and to DKGA. If the formation of these compounds were the only limiting reactions in the browning of AA, their accelerated formation by metallic ions would indeed accelerate the rate of browning. Such is not the case, however, and in fact their results showed a negative catalysis. Conversely, the rate of browning in AA buffer systems devoid of metallic ions is increased when the available AA is converted to DHAA. These results would indicate that in a non-metallic system one of the first reactions in the presence of oxygen is a conversion of

PAGE 53

38 the AA to its dehydro form and this conversion step may be partially rate-limiting. In the presence of metal, however, a reaction leading to the formation of polymeric compounds must be sensitive to metallic ions and becomes the rate-limiting reaction. It is -oossible that some intermediates beyond EKGA would chelate metallic ions, and if this were the case, the ability of these materials to form long chains capable of light absorption would be deleteriously affected, Hodges (1953) The mechanism of metal catalyzed oxidation of AA in aqueous solutions is varied. Khan and Martell (1967) postulated an ascorbate-metal-oxygen complex involving a one electron transfer to oxygen. Jameson and Blaclcburn (1975) proposed the formation of a metal-metal dinuclear ascorbate-oxygen intermediate with a two electron transfer to oxygen. However, the exact mechanism of action of metal ions on the AA oxidation is still uncertain. Effect of water activity The nonenzymatic browning in foodstuff takes place over a wide range of water activity. In most foods a maximum browning reaction occurs at a certain value of depending on the type of food product. It is generally agreed that the rate of browning in fruit juices and concentrates is increased as the water content in the product decreases (concentration of solids is increased), (Reynolds, 1969). Thus, nonenzymatic browning should occur more rapidly in dehydrated orange juice (1-31 water) than in frozen concentrated orange juice (557o water) or in single strength orange juice (857 0 water) It may be that the detrimental effect of

PAGE 54

39 moisture on browning, so often observed, is actually due to an increase in the rate of AA destruction by oxygen uptake. Karal and Nickerson (1964) Jensen (1967) Vojnovich and Pfeifer (1970) and Lee and Labuza (1975) have studied the stability of AA in various low and intermediate moisture dehydrated foods and model systems as a function of moisture content and water activity. Results reported by these investigators showed that the rate of destruction of AA in dehydrated foods increased as the total moisture content and increased. Kinetic data generated by Jensen, (1967), Vojnovich and Pfeifer (1970) and reported by Lee and Labuza (1975) have shown the energy of activation required for the destruction of AA to increase with moisture content in some foods but the opposite effect occured in the other foods. This could mean that a change in mechanism was occuring, but only limited data were collected. Lee and Labuza (1975) attributed the increased destruction rate for AA as a function of water activity to the decreased viscosity of the aqueous phase, resulting in increased mobility of reactant and catalysts. Kirk et al. (1977) further noted that the destruction of AA could be described by first order kinetics under all storage conditions. The greatest stability of total ascorbic acid was observed at low storage temperature and water activity, and was shown to result from EHAA stability at these conditions. Due to availability of water for hydrolysis of DHAA, the stability of DHAA decreased at storage temperatures above 20C and water activities above monolayer 0.24.

PAGE 55

40 As previously mentioned, there is a certain value of a^ at which a maximum browning reaction occurs. By maintaining a w level, either above or below the point of maximum browning, some increase in storage life could be obtained. Unfortunatly there is very little experimental evidence in the literature to define the range of a w at which this maximum occurs in orange juice. Effect of pH It has been shown that the rates of browning in acidic foods such as citrus juices and in model systems are strongly pH dependent. Berry et al. (1970) studied the storage stability of dehydrated orange juice at pH values that ranged from 3.3 to 6.5 and found storage stability to decrease with increasing pH. They concluded that the greater storage stability of instant grapefruit juice (pH 3.3) over that of instant orange juice (pH 3.7) may be due to the greater acidity of the former. Wolfrom et al. (1974) found the rate of browning in 1:1 glucose: glycine model system to increase as the pH was increased from 6.0 to 7.5, and the rate of browning in that pH range to be much greater than that between pH 3 and 4, the normal range for orange juice. Braverman (1963) reported that the AAinduced browning in citrus juices and concentrates which initially involves its decomposition to furfural and subsequent polymerization reaction with amin o compounds is dependent on pH, and within the pH range of 2.0 to 3.5 the extent of browning is inversely proportional to pH.

PAGE 56

41 Horton and Dickman (1977) reported that AA is much less stable in phosphate buffer than in orange juice at the same pH. They suggested that factors other than pH must protect AA in orange juice against oxidation. Possibly citrate, a known chelator of heavy metal ions, inhibits the catalytic oxidation of AA. At higher pH values and at room temperature, all three reactions, (a) oxidation of AA to DHAA, (b) hydrolysis of DHAA to DKGA, and (c) the decarboxylation and further degradation of DKGA to a mixture of products, proceed much more rapidly in phosphate buffer. Role of Nitrogenous Compounds The nitrogenous constituents of citrus juices have long been suspected of being involved in the nonenzyaiatic browning of orange juice. Hall (1927) and Wilson (1928) were probably the first to suggest the possibility that the Maillard reaction was responsible for the darkening in citrus products. According to Hall it was definitely established that the amino nitrogen content of orange concentrates steadily decreases in storage and may drop to zero. Wilson found a reduction in amino nitrogen and reducing sugars, and suggested that darkening may be due to a Maillard reaction between sugars and amino acids. However, Nelson and Hotter (1933) could detect no changes in the concentration of the nitrogen bases present in orange juice, and thus had reason to doubt Wilson' s theory. Joslyn and Marsh (1935) followed changes in amino nitrogen by means of the formal titration and also by the Van Slyke method. They observed that the aminonitrogen level remained practically constant during the course of browning of Valencia and Navel juices, even after

PAGE 57

42 storage for 126 days at room temperature when the juice had become very dark brown in color. They postulated the possible hydrolysis of soluble polypeptides during storage to account for the maintainance of constant amino nitrogen content: during browning. The isolation of an alkali soluble protein from the chrcmatophores of the orange by Smith (1925), and later by Sinclair et al. (1935), lent support to this hypothesis. However, Stadtaan (194S) doubted this explanation and stated that it would seem rather unlike ly that the rate of amino nitrogen formation by such a process would exactly parallel the loss of amino acids through darkening. Nelson and Motter (1933) determined the distribution of nitrogen in various fractions of darkened and undarkened, filtered orange juice by the use of various precipitating agents. The darkened juice contained about twice as much soluble nitrogen as the fresh juice. No nitrogen constituents were found in the darkened juice which had not been identified in the fresh juice. There was, however, a considerable increase in the arginine content with darkening of the juice. Their results do not rule out the possibility that nitrogen compounds are involved in the browning of orange juice, but they do indicate that browning can occur without the transformation of large amounts of these substances. Lceffler (1941) also observed lower amino acid content in juices stored at 0G than those stored at higher temperature. The addition of certain amino acids to various fruit products has been studied. Thus, Matthews (1928) added asparagine in amounts varying from 0.01 to 0.1g/l00 ml. to orange juice, with and without additions of 0.5g of glucose and citric acid. After a storage period

PAGE 58

43 of one year, in air and nitrogen at 26.7-32.20C, darkening in treated and untreated juice occured at the same rate and to the same extent. Joslyn and March (1935) found that aniline, tryptophan, and other aromatic amines markedly increased the browning of orange juice. The effect varied greatly with the different compounds. Richert (1930) showed that free ammonia or ammonium ions greatly hastened the darkening of both sugar syrup and grape juice concentrates. Richert found glycine to be more effective than alanine but less so than ammonium tartrate. It is also common e.^erience in the dried fruit industry that traces of ammonia, such as from leaks in refrigeration coils in cold storage, rapidly enhance darkening. The confusion in the literature on changes in nitrogenous constituents during the browning of orange juice, may be due in part to failure to include the reaction of reducing sugars with proteins such as that found by Lea and Hannan (1950) and to the possibility that relatively small chemical changes are required to produce brown pigments of intense color. If this is the case, then the changes in reducing sugars, or amino nitrogen, necessary to produce large changes in color might be so small as not to be detectable by the methods ordinarily used. Role of Sugars The major sugars in citrus juices are sucrose, fructose, and glucose. Curl and Veldhuis (1947) reported the sugar composition of orange juice to be 57= sucrose, 2.57 0 fructose and 2.5% glucose. In order to see if reducing sugars were involved in browning of orange juice, Joslyn and Marsh (1935) studied the effect of their removal by

PAGE 59

44 fermentation. Hie results shoved that all samples (fermented and unfermented juice) darkened at about the same rate. However, how ccmpletly fermentation removed the sugars is not evident from the data. Stadtman et al. (1946) almost completely removed the reducing sugars from apricot syrups, yet the rate of browning was decreased to only about one half the rate in unfermented samples. The addition of fructose and glucose to fermented syrups in amounts equal to the sugar lost by fermentation, resulted in a restoration of the normal browning rate. These results indicate that sugar may not be involved in the darkening of orange juice, and that probably only part of the browning in apricots involves sugar reactions. Hall (1927), in a summary of work done on darkening in orange concentrates, stated that slight decreases in reducing sugar during storage have been observed. Curl et al., (1946) have verified this conclusion and showed that these losses in reducing value are roughly parallel to changes in color. Following the suggestion that the Maillard reaction was responsible for browning (Wilson, 1928) a number of investigators have attempted to correlate browning with changes in reducing sugars. The initial rate of the Maillard reaction between a reducing sugar and an amino compound is directly related to the conformational stability of the favored cyclic structure of the sugar (Burton and McWeeny, 1963) Browning reaction in amino acidsugar systems has also been shown to depend on the type of sugar (Cole, 1967; Spark, 1969) it follows that pentoses are more reactive than hexoses which are more reactive than disaccharides.

PAGE 60

45 • Wolfrom et al. (1974) studied the influence of different sugars on the browning reaction, and compared a 1:5 D-glucose-glycine system with similar systems in which D-glucose was replaced by Dfructose and sucrose. In the unbuffered system used, Dfructose showed a somewhat higher initial rate of browning than I>glucose. However, the authors stated that this difference might be reversed in the buffered media generally characteristic of foodstuffs. Role of Container Type Many studies have been made on the relationships between container, browning, and ascorbic acid retention in citrus products. Most early studies of single strength orange juice in metal containers or glass bottles generally showed that up to 75% or more AA was retained after one year storage at 26.7C or lower (Moore et al., 1944). Single strength orange juice kept frozen in tin-lined cans at -17.8C showed no change in AA concentration after one year (Nelson and Mottern, 1933) In continuation of this work, Moore et al. (1944) compared the changes in color, flavor, and ascorbic acid content occurring during storage of the bottled and canned orange and grapefruit juices. At the end of six months' storage at room temperature, the orange juices in glass and tin were offflavor, with the bottle juice slightly better in taste than the canned juice; the results would indicate that plain tin was found superior for packing orange juice with the exception that at room temperature the bottled orange juice retained a slightly better flavor during storage than the canned juice. Pasteurized single strength orange juice stored in glass bottles for 1

PAGE 61

46 year retained 87% AA at 4.4C but only 68% at 26.7C (Curl and Veldhuis, 1947). AA retention In frozen concentrated orange juice in tin cans was 90% or greater after 1 year, at 4.4C or below, regardless of headspace atmosphere, concentration of product, or preliminary heat treatments (Curl et al.,1946). Curl (1947) also reported that loss of AA increased with temperature and concentration from 4.4 to 48.9C and from 13 to 71 Brix, respectively. Retention at 4.4C ranged from 99% for SSOJ to 93% for 713rix concentrate and at 26.7C was reduced to 7C% and 60%, respectively, after 1 year. Du3ois and Kew, (1951) found frozen concentrated orange juice stored in tin-lined cans for 11 months at -28.9 to 23.9C had very high (95% or more) retention of AA. Variation in frozen storage temperature had little effect on AA retention (McColloch et al., 1957) and samples stored from -12.2 to 15.6C for 1 year, to simulate warehouse conditions, retained 95% of their AA. Bissett et al. (1975) studied the AA retention in orange juice as related to container type. They found that single strength orange juice packed in glass retained about 90% of initial AA for over 4 months and 87% fcr 1 year at 4.4C. AA retention was progressively less at 10 and 15.6^0 (84 and 79% respectively).

PAGE 62

PRELIMEWRy STUDY In a preliminary study, single strength orange juice was aseptically filled into pouches of 3 types of flexible film: retort pouch (American Can Company, Desmoines, Iowa.), vinylidene cryovac (W.R. Grace Sc Co. Cryovac Div. Duncan, SC.), and polyethylene Whirl-Pak (Nasco West, Modesto, CA.). These packaging materials posess different permeabilities to oxygen as indicated in Table 5. Table 5 Permeability of the Packaging Film Oxygen permeability Packaging film cc 02/100 sq.in./24hr. 72of., lmil. Retort pouch 0* Vinylidene cryovac 0.1-0. 2b Polyethylene Whirl-Pak 500b b -Sacharow and Griffin (1970) The pouches were overfilled and sealed through the juice to prevent the inclusion of air. Trie pouches were stored at room temperature (22 43oC) and analyzed at 0, 2, and 4 week intervals. Ascorbic acid and dehycroasccrbic acid concentrations were determined using the automated fluorometric method of Roy et al. (1976) The Autoanalyzer technique called segmented flow analysis, where the sample and reagent streams are segmented by air bubbles, and continuously pumped with automated sampling of sample solutions was

PAGE 63

48 used. The procedure is based on the oxidation of ascorbic acid to dehydroasccorbic acid by N-bromo succinimide followed by condensation of the dehydroascorbic acid with o-phenylenediamine to form a fluorophore. The fluorescence is then measured using a fluorometer with a flowthrough sample cell. The dehydroascorbic acid can be determined by omitting the oxidation step of ascorbic acid. This procedure was found accurate only when used with fresh samples of orange juice. The accuracy of the method was determined by comparing the analytical results obtained by the continuous flow procedure with the titration method with 2,6-dichlorophenoiindophenol (AOAC, 1975). Table 5 shows that the two procedures were significantly different after 2 weeks storage. At this time some of the degradation products in the orange juice samples interfered with the fluorescence measurement. Therefore, an HPLC procedure was investigated which allowed the simultaneous measurement of ascorbic and dehydroascorbic acids in fresh samples as well as in browned samples of orange juice using a single injection analysis procedure. Table 6 Comparison of the Fluorometric and the Dye Titration Procedures for Ascorbic Acid Determunation U" 2-WEEK MS £ b a b a b Retort pouch Crvovac pack Polyethylene 16.5 26.7 22.0 27.7 27.0 27.5 21.4 10.0 0.3 26 7 17 '.4 0.4 18.6 8.3 0.2 26.0 16.9 0.3 a = fluorometric procedure b = dye titration procedure.

PAGE 64

SIMULTANEOUS ANALYSIS OF ASCORBIC AND DFJIYDROASCCRBIC ACIDS BY HIGH PERFORMANCE LIQUID CHROMATOGRAPHY WITH POST COLUMN DERIVATIZATICN AND UV ABSORBANCE Introduction Fruits and vegetables constitute the major sources of vitamin C for human diets. The total vitamin C consists of the sum of ascorbic acid and its oxidized form, dehydroascorbic acid. Both forms have equal antiscorbutic activity (Tannenbaum, 1974) Numerous methods for the analysis of vitamin C activity have been described. The most commonly used are the 2,6 dichlorophenolindophenol visual titration (AOAC, 1975), the spec trophotome trie method with dinitrophenylhydrazine derivatization of DHAA (Roe et al 1948) and the microfluorometric method by condensation of DHAA with OPDA (AOAC. 1975) However, these methods are not specific and are often limited by the number of interfering substances present in foods. In addition, it is difficult to determine visually the end point when these methods are used with colored solutions. Pelletier and Brassard (1977) described an improved photometric method based on 2,4-dinitrcphenylhydrazine for the AA and DHAA determination in foods. Though their method eliminated interference from other compounds, it was time consuming and requires special sample preparation. Recentlj^, due to the development of commercial HPLC systems, quantitative measurement of AA and DKAA in various substances has been reported by many investigators. Procedures vary in the type of 49

PAGE 65

50 column, elution conditions, detection systems and the extraction technique used to stabilize AA and DHAA. AA can be determined easily by HPLC, with UV detection, but the determination of DHAA is complicated by its extremely low UV absorptivity. A procedure using two reversed-phase HPLC columns in series for the separation of AA and DHAA was reported by Finley and Duang (1981) They used water with a counterion reagent (tri-n-butylamine) as a mobile phase. AA and DHAA were detected at 254 run and 210 nm respectively. Rose and Nahrwold (1981) and Wimalasiri and Wills (1983) used a similar detection system with a single ion-exchange column and a mobile phase of acetonitrile-water containing 2.5 mM potassium dihydrogen phosphate. Doner and Hicks (1981) reported a separation of AA and DHAA by HPLC on a Zorbax-NH2 column. The AA was monitored at 268 nm, while refractive index (PJL) detection allowed the detection of DHAA. However, neither RI nor low wavelength (210 nm) can detect small amounts of DHAA such as that present in foodstuffs. In addition, measurement of DHAA at low wavelengths introduces instrumental noise from solvent impurity. Therfore, most of the HPLC analytical procedures used are based on either the reduction of DHAA to AA and detection of the total ascorbic acid (TAA) by UV or oxidation of AA to DHAA. The TAA is determined by fluorometry after condensation of DHAA with OPDA. Dermison et al. (1981) described an HPLC method for the analysis of total vitamin C in beverages by UV measurement of AA after reduction of DHAA with homocysteine. Keating and Haddad (1982) reported the simultaneous determination of AA and DHAA using precolumn derivatization. DHAA was converted to a fluorophore using OPDA.

PAGE 66

51 The detection was made at 290 ran for AA and 348 ran for the fluorophore. Speek et al. (1984) developed an HPLC method for the simultaneous determination of total vitamin C based on precoiumn enzymatic oxidation of AA to DHAA. The latter is condensed with OFLA and detected fluorometrically. DHAA can be determined with omission of the oxidation step. While these methods give increased sensitivity for the estimation of EHAA, the addition of the derivatization step increases the complexity and adds another variable to the analysis. Also, problems were encountered with the stability of the derivative. Recently, Vanderslics and Higgs (1984) proposed an HPL£ method with fluorometric detection and post-column derivatization involving oxidation of AA to DHAA followed by reaction with OPDA to form a fluorescent product. In this study we have examined the system proposed by Vanderslice and Higgs (1984) and modified it to obtain an estimation that includes AA and DHAA as a separate value using a single injection. The step for oxidation of ascorbic acid was also omitted. Materials and Methods Re agents Ascorbic and dehydroascorbic acids (Aldrich Chemical Company, Inc. Milwaukee, Wis.), o-phen3, T lenediaml ne (GPDA) (Eastman Kodak Company, Rochester, N.Y.), metaphosphoric acid, potassium phosphate monobasic, and HPLC grace acetcnitrile (Fisher Scientific Cocpany, Fair Lawn, N.J.) were used as received. Double distilled deionized water was used to prepare solutions.

PAGE 67

52 Apparatus Liquid chromatography High performance liquid chromatography was performed using a system incorporating a Waters Associates Model 6000A pump, a Waters Model U6K. injector, a Spectra Physics, Model 8440 variablewave length ultraviolet detector (Spectra-Physics, San Jose, CA.) set at 254 ran, and a Fisher Recordall Series 5000 recorder. Separation of AA from DHAA was achieved by use of an amine column (Alltech Associates, Inc., Applied Science Labs, Deerfield, IL.) in the weak anion exchange mode. The mobile phase was 75% acetonitrile in 0.05M monobasic potassium phosphate (pH=5.9). The eluant was filtered through a 0.45 um Millipore filter (Gelman Sciences. Inc. Ann Arbor. Mich.) and subsequently degassed under vacuum. The flow rate was 1.5 ml/min. Post-column Derivatization The system used for the post-column derivatization was similar to that described by Vanderslice and Higgs (1984) After separation of AA and DHAA on the analytical column, the exit stream from the UV-detector was mixed with a second stream containing the derivatization reagent (CPDA) in a mixing "PTFE Tee" (Rainin, Wobum, MA. Catalog No. 45-1003) The final eluant was passed through a heating coil then into a cooling coil before entering a fluorometric detector (Aminco Fluorcocnitor) equipped with an ultraviolet mercury light source (type GE no.F4T4/BL.4watt) a Corning 7-51 exitation filter, a Wratten 2A emission filter and a 70uL flow cell, whose output was sent to the recorder, Fig 10. All postcolumn tubing was 0.40 mm i.d. Teflon^. The reaction path length was 20 meters and was

PAGE 68

53 Injector Flow Mobile Phase Column Peristaltic j Pump OPDA UV Detector Dual Recorder PTFE, Tee Heating Bath *= Waste Cooling _Ba_th_ Figure 10. HPLC System with Post-Column Derivatization and Tandem Ultraviolet and Fluorccetric Detection

PAGE 69

54 maintained at constant temperature (53C) while the cooling coil (22C) was 2 meters. The fluorogenic reagent consisted of 0.05% (w/v) OPDA in distilled water and was Dumoed usin^ a Gilson Minipuls 2 peristaltic pump at a flow rate of 0.5 mL/min. Sample Preparation Fresh fmits and vegetables were purchased from a local market and homogenized in a domestic blender. A sample (20g) was then blended with 3% (w/v) metaphosphoric acid solution (50mL) for 2 min and diluted to volume (100 or 200 ml) with extracting solution. The resulting solution was filtered through paper (Whatman 541) and a portion of the filtrate was purified by percolation through a C^g Sep-Pak (Waters Associates, Milford, M), a short plastic column containing uBondapak C^8 as described by Wimalasiri and Wills (1983). The (43 Sep-Pak was placed on the Luer tip of the syringe barrel and the column preconditioned with 4mL of methanol followed by lOmL of water. The sample (4mL) was then passed through the Sep-Pak. The first 3mL were discarded and the remaining lmL was collected for analysis. The Sep-Pak Cig could be reused up to eight times provided it was washed with methanol and water between samples. Frozen orange juice was first diluted according to package directions. The resulting solution was filtered and purified as above. The injection volume was 20 uL. Recovery Study Proper amounts of ascorbic acid and dehydroascorbic acid standards were added as solutions in metaphosphoric acid to the various fruits and vegetables during extraction so that the AA and

PAGE 70

55 DHAA content of the spiked samples approximately doubled that of the unspiked. The AA and DHAA of spiked samples were then determined as described previously and percent recovery was calculated. Calibration Curves Samples of reagent grade AA and DHAA in the mobile phase were combined to contain 2.0 AA + 0.5 DHAA, 4.0 AA + 1.0 DHAA, 6.0 AA + 2.0 DHAA, 8.0 AA + 3.0 DHAA and 10.0 AA + 4.0 DHAA, ng/lOOml. The standard mixtures had to be prepared daily. Aliquots (20ul) of the combined solutions were injected into the chromatographic system, and the resulting peak heights were plotted against concentrations for the calibration curve. Results and Discussion Using the HPLC procedure described, linear calibration curves were obtained for DHAA in the range 0-4mg/l00ml and for AA in the range 0-l0mg/100ml. Correlation coefficients of the linear regression equations were 0.9994 for AA and 0.9999 for DHAA, and the limits of detection were 0.05ug for AA and O.Olug for DHAA per 20 ul injected. The signal to noise ratio was S/N = 8. Typical chromatcgrams of orange juice, parsley, tomato, and strawberry are shown in Fig 11 and Fig 12. These illustrate the ability of tandem ultravioletfluorometry detection to determine simultaneously AA and DHAA. In all samples, the AA and DHAA peaks were well resolved with no interference. The DHAA peak had a shoulder in some of the samples. The procedure was successfully applied to the analysis of vitamin C in different food products and the results are presented in Table 6. Although direct comparison was not made with

PAGE 71

56 other methods of analysis such as the dye titration or the fluorometric method, the levels of AA and DHAA were found to be similar to those reported in the literature (Ashoor et al., 1984; Wills et al., 1983; Wimalasiri and Wills, 1983). Table 7 Ascorbic Acid and Dehydroascorbic Acid Content of Various Foods. Concentration (mg/lOOg) Sample c AA DHAA TAA Broccoli, fresh Orange juice Fruit punch Orange drink Parsley Tomato fresh Strawberry Banana 81.5 2.5 42.7 1.1 51.0 1.9 45.7 1.6 148.7 7.4 9.1 0.4 51.0 1.9 7.7 0.6 6.2 0.9 2.9 0.2 3.9 0.2 1.0 0.1 9.7 1.0 1.1 0.2 6.1 0.2 3.0 0.1 87.7 1.6 45.6 1.0 54.9 2.1 46.7 1.5 158.5 7.0 10.2 0.5 57.1 1.8 10.7 0.6 sr Number of samples = 4. Mean — SD

PAGE 72

57 rDHAA Farsiey Orcnge Juics UV (254nm) C Figure 11. Typical HPLC Chromatcgrans of Orange Juice and Parsley, Monitored by Tandem Ultraviolet (UV, 254 nn) and Fluoroinetric Detection

PAGE 73

53 i i iomaro Strawberry UV (254nm] 1 J Lv/kJ Figure 12. Typical HPLC Chromatograns of Tomato and Strawberry Monitored by Tandem Ultraviolet (UV, 254 mn) and Fluorosetric Detection

PAGE 74

59 Table 8 Recovery of Ascorbic and Dehydroascorbic Acids from Spiked Samples Recoveries (%) AA DHAA TAA Broccoli, Orange juice Fruit punch Orange drink Parsley Tomato Strawberry Banana 97.7 2.9 99.5 i 2.7 97.0 J 0.7 96.7 t 1.5 91.0 J 1.9 96.0 1.2 96.5 J 1.1 91.2 1.9 104.2 3.5 100.6 1.8 99.8 0.9 99.2 1.8 101.4 2.2 100.4 1.3 101.0 1.3 96.9 1.5 ilumber of samples = 4. Mean ~~ SD. The data in Table 7. show that both AA and DHAA are completely recovered from the samples examined. The recoveries ranged from 91 to 99.5% for AA and from 101 to 1121 for DHAA. The slightly higher recoveries for DHAA could be due to the oxidation of some of the AA to DHAA during extraction and sample preparation. For samples that require homogenization and extraction such as banana and parsley etc., conducting the extraction at 3C should be helpful in preventing potential conversion of AA to DHAA during the extraction procedure. This HPLC procedure provides a relatively fast and sensitive technique for the simultaneous determination of AA and DHAA in foodstuffs and beverages. The method is simple and requires a minimum of sample preparation during the simultaneous determination of AA and DHAA. Further, this HPLC method measured AA and DHAA directly, which

PAGE 75

60 eliminates the need for the oxidation of AA to DHAA or the reduction of DHAA to AA prior to the analysis. The procedure was also found to be very useful for measurement of AA and DHAA in browned samples of orange juice where many interfering compounds limited the use of the dye titration method and the microfluorometric method. Finally, an attempt was made to include the diketogulcriic acid (DKGA) determination in our assay. The DKGA was prepared according to the method of Doner and Kevin (1981) as follow: DHAA was readily prepared frcm AA by air oxidation of an ethanoiic solution containing activated charcoal. After filtration and removal of ethanol, pure syrup of DHAA was obtained, as determined by HPLC analysis. DHAA (an aqueous solution of lOmg/ml) was then converted to DKGA by gradual titration in an ice bath over a period of 1 hour with an aqueous solution of lOmg/ml with 0.5N sodium hydroxide until the pH remained constant at 7.0. Standard mixtures of AA, DHAA, and DKGA were prepared just prior to analysis by HPLC (UV detection) The acquisition of a new Hewlett Packard 1090 Liquid Ghromatograph with HP-85B Personal Computer and DPU multichannel integrator, permitted the simultaneous monitoring of the different compounds at different wavelengths. Figure 13 presents a chromatogram of the standard mixture (AA, DHAA, and DKGA) with their absorption spectra. This procedure provided an excellent resolution of the three compounds with retention times of 7.9, 3.7, and 10.7 minutes respectively. However, the UV detection of DHAA and DKGA was not sufficiently sensitive to detect these two compounds at the levels actually present in orange juice, even when a low wavelength (210 nm) was used for analysis.

PAGE 76

61 Figure 14 shows two chromatograms of AA (40 ug/ml) and DHAA (1 mg/ml) at two different wavelengths (210 and 254 ran) At the level used DHAA was only slightly detected at 210 nm. This demonstrates the advantage of the fluorometric detection procedure where the sensitivity was significantly imp roved oyer UV detection. Measurement of DHAA and DKGA has been made by others using UV detection (Finiey and Duang 1982) We found the maximum absorption for DHAA occurs at a wavelength of 227 nm, and for DKGA 200 nm, figure A-6. Many other compounds which occur in biological samples also absorb at these low wavelengths. Therefore, determination of DHAA and DKGA by UV detection was not a suitable procedure for the low levels in orange juice.

PAGE 77

62 Figure 13. Resolution of AA, DHAA, and DKGA

PAGE 78

63 c c J p r— CD — O < LT3 S C CM LTJ LT3 c u C3 e I t_ o E s Ol S 0 L
PAGE 79

64

PAGE 80

EFFECT CF ASCORBIC ACID AND AMINO ACIDS CaTCENTRATICKS CM THE QUALITY OF ASEPTTCALLY PACKAGED ORANSE DRINKS Introduction Nonenzymatic browning is one of the main reasons for the reduction in comae rcial value of citrus products. It is the quickest and most dramatic quality defect to appear during ambient temperature storage. Knowledge of the fundamental factors and the mechanism of reactions which these factors can undergo under different storage conditions is critical to the understanding and subsequent control of nonenzymatic browning in citrus. While some progress has been made in the study of the changes responsible for darlcening, the chemistry of many of these changes is not well understood, and the nature of all the constituents involved is not known. In view of this lack of understanding, it was decided to carry out the present investigation to study the reactions taking place during the nonenzymatic browning of aseptically packaged orange drinks. Our main objectives were to determine the loss of ascorbic acid and the development of browning as affected by ascorbic acid and amino acid content, and oxygen permeability of packaging materia] A 3x3x2 factorial experiment was designed to measure the main effect of ascorbic acid, amino acids, and oxygen permeability alone as well as all permutations of any two or three factors' interaction on the browning of orange drinks aseptically packaged and stored at 75F. 65

PAGE 81

66 Materials and Methods Reagents Ascorbic acid (Aldrich Chemical Company, Inc. Milwaukee, Wis.), arginine, aspartic acid, citric acid, potassium citrate, fructose, and glucose (Fisher Scientific Company, Fair Lawn, N.J.), 4-aminobutyric acid (Eastman Kodak Company, Rochester, N.Y.) and sucrose (local market) were obtained and used as purchased. The juice used in this study was reconstituted from a high quality Florida commercial concentrate. The frozen concentrated orange juice was first diluted according to package directions with distilled water which was boiled then cooled to room temperature to remove any dissolved oxygen. The final degree Brix was 11.8. Orange Drinks Composition Nine orange drink mixtures containing 10% (w/w) orange juice and various compositions of ascorbic acid, amino acids, sugars, ciric acid, and potassium citrate were prepared, aseptically packaged, and stored at controlled room temperature (75F, 23.9C). The compositions of the mixtures are given in Table 9. Analyses of the mixtures were conducted over a period of 20 weeks. The mixture of sugars used (5% sucrose, 2.5% glucose, and 2.5% fructose) was similar to that which occurs in orange juice, (Curl and Veldhuis, 1947). For amino acids, a mixture of 0.2% each of L-aspartic acid, L-arginine, and 4-aminobutyric acid was used. Since these are the most abundant amino acids in orange juice and in many other fruit juices (Winston, 1961) it was considered that these amino acids may play an important role in the deterioration of orange juice

PAGE 82

67 on storage. Wolfroo et al. (1974) reported L-arginine and 4aminobutyric acid to give the most intense and rapid color formation, and -were quantitatively much more effective than glycine or any other of the 9 amino acids examined. According to the literature orange juice contains 0.6% (w/w) of crude proteins (Chatfield and Adams, 1940; and watt and Merrill, 1963), hence this quantity of the mixed amino acids was used in mixtures, 2, 5, and 8. To mixture 1, 4, and 7 no amino acids were added, and the only amount present in the mixture is that contributed by the 10% orange juice. This amount was assumed to be 0.06%. To mixture 3, 6, and 9 amino acids were added to the level of 1.267 0 (w/w) The quantity of ascorbic acid used in the orange drinks 4, 5, and 6 was 38.0 mg/ 100ml. To mixtures 1, 2, and 3, no ascorbic acid was added, and the mixture content was 4.2 mg/ 100ml, provided by the 107 o orange juice. Mixture 7, 8, and 9 contained 71.8 mg/ 100ml of ascorbic acid. Each mixture contained 1% (w/w) of citric acid and 0.7% (w/w) of potassium citrate (both as hydrates) to buffer the solutions to a pH of about 3.8, which falls within the normal range of pH of orange juice. These compounds are part of the buffer system of orange juice. The mixtures were filled aseptically at room temperature. With mixture 1, 4, and 7 the ascorbic acid level was increased (4.2, 38.0, and 71.8 mg/100 mL), the objective was to determine the effect of the concentration of ascorbic acid on the rate of ascorbic acid loss and the rate of browning without addition of amino acids. The amino acid level (0.067,) was that naturally provided by the 10%

PAGE 83

63 Table 9 Composition of the Orange Drink Mixtures Mixture Ascorbic Acid > Amino acids a Other constituents (mg/lOOml) [ % (w/w) i % (w/w) M 1 4.2 i 0.06 I All mixtures M 2 4.2 1 0.66 : contained M 3 4.2 1.26 1 % citric acid 0.7 % potassium M 4 38.0 1 0.06 citrate M 5 38.0 I • 0.66 5 7 sucrose M 6 38.0 • 1.26 2.5 % glucose 2.5 % fructose M 7 71.8 t • 0.06 M 8 71.8 I 0.66 M 9 71.8 1 • 1.26 M 0 42.0 1 0.60 S. S.O.J. A mixture of equal amounts of aspartic acid, 4-aminobutyric acid> and arginine was used in each case with the indicated percentages. orange juice. With mixture 2, 5, 8, and 3, 6, 9 the objective was to determiae the effect of increased levels of amino acids (0.66 and 1.26%) on browning and ascorbic acid degradation. Preparation of the Mixtures Using the facilities at the Food Science and Human Nutrition Department (University of Florida, Gainesville) the different ingredients were mixed with single strength orange juice and distilled water to make an orange drink containing 10% orange juice, the final pH was 3.8. Three mixtures were prepared each day. The filling of

PAGE 84

69 the pouches was done on the day following the preparation of the mixtures. The mixtures were maintained overnight at 4C before processing and packaging. The juice and the drinks were processed using a No-Bac Uni therm IV Processing system. (Cherry Burrell Corporation, Cedar Rapids, Iowa) It is a complete unitized system for sterilizing fluid products at a rate of 22.5 to 45 gallons per hour. It consists of two surge tanks, one with agitator, supply pump, high pressure pump, heat exchangers, aseptic remote homogenizing valve, valve manifold and control panel. The high pressure pump has a 3000 PSI maximum pressure limitation. The 1/4" tubular heat exchangers have a 150 PSI maximum steam pressure limitation. The system was first sterilized with circulating water at 285F for 20 minutes. The product was pumped from an agited supply tank and was heated to 205F for 14 seconds, with the flow rate of the product kept at 31.5 gallons per hour. The sterile product was then immediately cooled to 80F. From this point on every precaution was taken to make certain the product did not become recontaminated. The products were then packaged into 7 by 10 cm pouches of two types of flexible films: retort pouch (zero permeability to oxygen) and polyethylene pack (high permeability to oxygen) The retort pouch composition from inside was polyethylene terephthalates/aluminum foil/polypropylene. They were steam sterilized for 8 hours. The polyethylene packs were Whirl -Pak type, commercially sterile, and were used without farther treatment. The pouches were aseptically overfilled and sealed through die liquid to prevent the inclusion of

PAGE 85

70 air. Toe filling was made in a sterile environment in a laminar flow hood (The Baker Co., Inc. Sanford Airport, Sanford, Me.). The hood was equipped with a blower that provided an average air velocity of 99.6 fpm. a pref ilter-Scottfoam (washable) a final filter (Zero Probed HEPA, 99.99% efficient on all particules 0.3 micron by B.O.P. test) and an UV light to maintain the sterile conditions. Thirty two pouches were prepared from each mixture and each packaging material. Eight pouches from each mixture (4 retort pouches and 4 Whirl -Pak) were opened after each storage period and analyzed for ascorbic acid, dehydroascorbic acid, and browning over a period of 20 weeks. Method of Analysis Ascorbic and Dehydroascorbic Acids Determination Ascorbic and dehydroascorbic acids were determined as scon as possible after the pouches were opened using the HPLC procedure previously described. The zerotime analysis was made on the day following the filling of the pouches. The pouches were allowed to stand overnight at room temperature. Thereafter, samples were analyzed at 2-week intervals for the first 8 weeks, and then at 12, 16 and 20 weeks. All measurements were made on 4 packages from each mixture. Browning Measurement Browning expressed as abscrbance at 420 nm, was measured according to the method developed by Meydav et al. (1977). In this method the pulp and serjm were separated centrifugally (2000 rpm for 20 min) The supernatant was then diluted 1 : 1 (v/v) with ethyl alcohol (957 0 v/v) to cause floculation of the cloud and was then

PAGE 86

71 filtered through a Whatman no. 42 filter paper to obtain a fully clarified extract. The clear solution was checked for its absorbance at ^-20 run in an LK3 4050 Ultrospec. (LxB. Biochrcm LTD Science Park Cambridge, England) Statistical Treatments Regression methods were used for the calculation of factor effects, and for the analysis of variance (ANOVA) Results and Discussion Ascorbic Acid Retention Effect of processing From a nutritional point of view, the retention of AA is an important factor for citrus products. Table 10 shows the estimated initial and zero time AA and DHAA levels for orange juice and orange drinks products in both types of packaging materials. There was considerable discrepancy of AA between the initial estimated values, and the measured value at zero time. This loss occurred in all mixtures and in both types of packaging materials, although it was more pronounced in the Whirl-Paks. This decrease in AA level represents the combined loss during processing, storage of the sample prior to the first analysis, and preparation of the drinks. The incorporation of oxygen into the mixtures prior to thermal processing (ie. dilution with oxygen containing water, mixing in open kettle to dissolve the ingredients) may have contributed to a large extent to the oxidation of AA to DHAA and the conversion of DHAA to DK.GA. This is evidenced by the high level of DHAA found in all mixtures at the zero time. On average the percent loss for total ascorbic acid (TAA) ranged from 12 to 21% in retort pouches and 20 to 26% in the Whirl -Pak.

PAGE 87

72 C r-i 3) H s I -g c C Q) CC rH 0) X CJ 4-1 •H Q) 3 O CJ C BP, r-l Sj U O 3 4-1 O CO r-l 4J > o c in o o if) CO fcj bj r-l CO • in I +I+I+J it —J CO CN CM CN! -! OJ +J+'+i CO r— I CO co co rco co co CN! CM t— I "ti+iti CO t— I CN rH CO CO CO 00 00 CO CO CO CO CN CM CM •m O CM ao cm m co -dcm m r-l r-J r-l -CT CO CO CM CO O r-i CO cm co m 1*^ G"n +•++! • • • r-i O m m
PAGE 88

73 Ascorbic acid retention as affected by amino acid concentration The effect of amino acid level on ascorbic acid retention and the concentration of its oxydation product, dehydroascorbic acid, are presented in Figures 16 to 19. Figure 16 shows the loss of ascorbic acid in samples stored in Whirl -Paks initially fortified with the highest concentration of ascorbic acid (71.8 mg/100 ml) and increasing levels of amino acids. In addition to the initial rapid loss of AA, these samples continued to lose AA more rapidly in storage than did those in retort pouches and by the second week of storage both AA and DHAA had almost completly disappeared. Similar results were obtained at the 38.0 and 4.2 mg/100 ml ascorbic acid levels (data in Appendix Figure A-l and A-2) Packaging of drinks in polyethylene film (oxygen permeable) resulted in extremely rapid destruction of the AA and DHAA. Experimental data and the results reported from a study done by Dennison and Kirk, (1978) indicated that oxygen must be evaluated as a reactant in the stability of AA. Figure 17 shows the effect of increasing am i no acids concentration at the highest AA content (71.8 mg/100 ml) in samples stored in retort pouches. There was a rapid decrease in AA concentration with the first 2 to 4 weeks storage, followed by a slower decrease for the remainder of the 20 weeks storage period. Such an initial rapid loss may be due to oxidation by residual oxygen in the drinks reacting with AA. After this initial period, AA. was probably degraded anaerobically, at rate lower than by aerobic process. Considerable loss of AA occured in the presence of the highest level of and no acids (1.26%). This suggests that amino acids may have some effect on the AA degradation.

PAGE 89

74 In samples containing 1.26% amino acids, AA retention was reduced to 56%, of its initial level after 20 weeks. Samples containing 0.66 and 0.067, amino acids retained 61 and 63% respectively, after the same period of time. Figure 18 shows the same effect but with a fired concentration of 38.0 mg/lOO mL of AA. It is interesting to note that this level of AA is similar to that found in single strength orange juice. At this level of AA the increase of amino acids from 0.06 to 0.66% did not affect the loss of ascorbic acid. The per cent losses in single strength orange juice stored in retort pouches were close to those in orange drinks 4, 5, 7, and 8 which contained the low levels of amino acids. Curl et al. (1949) working with ascorbic acid-amino acids-sugar systems, reported similar results and suggested that, since the losses in orange juice were of same order of magnitude as the model systems, it is possible that the reactants involved are the same or similar. Under similar conditions, but with an initial concentration of 4.2 mg/ 100 ml ascorbic acid, and increasing ami no acids (Figure 19), there was no significant effect of amino acid levels on AA retention. The small amount of AA present (4.2 mg/100 mL) in these drinks made any difference difficult to detect. Dehydroascorbic Acid Production All AA initially added to the orange drinks was in the reduced form. However, our results showed that significantly high levels of DHAA were found in all mixtures at the zero time analysis. These unusually high levels of EHAA are the result of incorporation of air and oxidation of a large amount of AA during preparation of the drinks.

PAGE 90

75 During the first 2 weeks the levels of DHAA decreased rapidly in all mixtures. This decrease reflects either the conversion of DHAA to DKGA, or its reaction with amino acids, we were unable to measured DKGA with the HPLC procedure used. It has been reported that DHAA reacts very rapidly with alpha amino acids to produce strongly colored (reddish to brown) complexes (Koppanyi et al., 1945). Once DHAA is degraded, its antiscorbutic activity is lost. In samples stored in retort pouches (Figure 17 and 18) the amount of DHAA gradually increased after 8 weeks and remained relatively constant at this level. These results are in agreement with those reported by Moore et al. (1944) who showed that the levels of DHAA in canned and bottled orange and grapefruit juices stored for six months at 4 and 27C remained constant at about 1-2% of the total ascorbic acid, neither container type nor storage temperature appeared to influence the DHAA level. Smoot and Nagy (1980) reported that with single strength grapefruit juice stored at different temperatures the DHAA and DKGA contents remained virtually unchanged during a 12 week period. Our results showed that DHAA, following an initial drop in concentration, remained relatively constant throughout the 20 week storage period. No accumulation of DHAA occurred in the different orange drinks and once it is formed it is further transformed to DKGA or reacts with amino acids. Although AAinduced browning in citrus products, which involves its conversion to furfural, is well known (Braverman, 1963), discoloration involving the reaction of amino acids with DHAA or DKGA has not been widely investigated. Dulkin and Friedemann (1956) studied the role of DHAA in the browning reaction and reported that

PAGE 91

76 en CH U_' ~ C — c a CJ tc 03 C3 U o c SI !_ > T0 c o cn JJ ii co li H 03 03 U 4-1 c c c C H c CUJ *— — 03 c a 03 o 2 UJ I — UJ CD -a cd < 3 i i Ij I' / / / 3 1 o < cc y 1-3 "§ 3 O i-i CJ O w eg < O C H /-s C J2 J=. COD O CJ o JJ al C CJ jj o 3 c V o; i — i >s c >t> Q C g) 2 ^ •i-i CD < H c s o c C tI "C H CD a o JJ DO c CO <3J JJ JJ CJ C i— i CD CD C, iw a) 53 W pi w

PAGE 92

77 in — cn — < r < — • ; 2 i < CO c £ 9 .[J L p. t5 77 c c o cu tc — ll en II T-i -. c p P c — •H •H i — i C l— i P7 — •H u cn C a 1 1 1 I 1 1 11 \" 5 1 I 1 1 II cm .'' I < < •CO — m LU UJ LU < e c — // / 1' I 3 eP < 00 o r-3 O v-l O O W aj < O S t! 8 8 — I cc •U 3 CC U O S H O 0) § § t — I •r-i (3 < ^ CtI'O I • 4J W C CO O O •r4 CO 4J 4J a U C l-H aj i 0 i bp

PAGE 93

c — c r a >> o u U — CO to II !! — co CC CJ U c c — c ^_ — ~ CJ •H u T— i 03 0 01 1 I 1 I l' 1 1 1 1 1 ill i" i 1 l \ 1 11 n < < <3 6 I en c/i UJ < DC CJ CC —1 C") e* C -i-i y W 03 < O 6 8 T-i W cd S-j U-i cj •u o c CO rH O 0) C > -U O O c TJ r-l •H cC s C .5 o 4J CJ c i CU C CC c co c 4J C d)
PAGE 94

79 c u — cu d u ca to K c 0 — •> C3 >'. c u JJ M ll CO II •H CO CO QJ 4J G C P *H — o rH n •H H CO to c o CO ; I! II '£ Mi 'I Ml II; 1 ll I 1 1 1A *1 I 1 3 I ill i 91 C3 i j i Li. 1 M UJ < — cj c < \ CJ CO H £ jj CM o • CJ c c a. 5 g rH •H C3 o >H ._ .H 9 H — C •<-! J & u CO o : CO c c >t-l u 4J CJ W pi DC fa

PAGE 95

80 oxidation of AA to DHAA is a prerequisite to browning. They also observed that the rate of browning was not increased by the addition of an amino acid (tryptophan) and suggested that browning in their experiment was not identical to the Maillard-type of reaction, and that EHAA must first undergo an irreversible transformation to DKGA. This transformation was first postulated by Herbert et al. (1933) and was substantiated by the work of Penney and Zilva (1943) Browning As Influenced By Ascorbic Acid Concentration Since AA appeard to be the most reactive constituent in the orange drinks investigated, and since browning in citrus juices is considered to be mainly due to AA degradation, it is therefore reasonable to expect the trend in browning to correspond to that of AA loss. Browning in the different orange drinks, as measured by absorbance and as affected by ascorbic acid concentration is presented in Figures 20, 21, and 22. In all cases, the diffusion of atmospheric oxygen through the polyethylene film was found to decrease significantly the ascorbic acid retention and to increase the development of brown pigments of non-enzymatic origin. It is therefore reasonable to conclude that the brown pigments originate from the oxidation of ascorbic acid. Orange drinks containing an initial concentration of 0.06% amino acids and with increasing level of AA (Figure 20) showed little browning in samples stored in retort pouches. However, in the presence of oxygen there was significant darkening. The effect of increasing AA was more significant in the presence of oxygen than in its absence. Under similar conditions, but with a initial concentration of 0.66% amino acids and increasing level

PAGE 96

81 of AA (Figure 21) there was a significant effect of AA concentration only in the presence of oxygen. High AA concentrations yielded a darker colored product than the low AA level. It is interesting to note that since 0.66% amino acids is the concentration range found in orange juice then an increase of the AA concentration in the absence of oxygen would not increase the browning significantly whereas, in the presence of oxygen, there would be significant darkening. This result indicates that oxidation of ascorbic acid is a major factor in the formation of brown pigments. Orange drinks containing 1.26% (w/w) amino acids and increasing level of AA (Figure 22) showed significant differences in browning. In both packaging materials AA had an effect on the browning. In the polyethylene pouch, with high oxygen permeability increases in browning followed increases in the ascorbic acid content, whereas there was not a linear relationship in the retort pouch. In retort poch high levels of ascorbic acid (30.8 72.8 mg/lOO ml) increased the browning. This is in agreement with the earlier finding that only when the level of amino acids was high (1.26%) in retort pouch was the ascorbic acid retention decreased. This indicates that high amino acid levels accelerated ascorbic acid degradation which resulted in an increase in browning. Browning As Influenced By Amino Acids Concentration Orange drinks containing 4.2 mg/lOO ml of ascorbic acid and increasing levels of amino acids (0.06, 0.66, 1.26 %) browned only slightly even in the polyethylene pouch (Figure 23) Increasing the concentration of amino acids at the low level of AA had no significant

PAGE 97

82 C 1-1 1 E i (J > 1 — c c c !sc = Z > c 3 : SJ ij tn ][ K n ••tn c — a c c c — *— — C — t i — c c 01 -a CJ LJ L3 o i— CT! 2 pa c c C CJ x o Sj c C E cu < u C 8-3 Q O O O •H O C U "H W SJ < C C a u Ed O

PAGE 98

83 u 0 to v z >, c JJ c -~ C 0 a stc > a L c 03 II 05 II X V c P P •H c ^ — 5 c ra c o 01 *0 I \ < I ill sc e lli : i 3C ESDG 111 1 UJ < c C c C .9 c c g •r4J 2 o o
PAGE 99

34 60 c e& c c G t G StC C c — — G — > rz G RS O ^ (-1 c O C 4J K jj or I! 03 II •H CD 03 G JJ G C C H •a g u h "5 2 co -r i B i : i i i l < i ; i i O 3 i EG I \ 1 i GO \ i \ en LLJ LU 13 < ~ a :— cn C PQ C O co •a 8*3 o c < ~ 4J C E u C 6-3 C CJ CN •a ih <2 O 3 -2 CO Ti ,2 o c U •T-l CO S-i < c c 4J U 60 C 2 £ "4-1 4-1 W O C CN CN 21 c CD — I

PAGE 100

35 effect on the brov/ning of the product. This indicates that the browning in this case is probably different from the Maillard reaction. Furthermore, it is well known the conditions which would favor the development of the Maillard reaction, near neutral pH or slightly alkaline pH, are absent in orange juice. Therefore, it is unlikely that this mechanism is the major contributor to the browning of a highly acid product such as orange juice at pH 3.8. Under similar conditions but with a higher level of AA (Figures 24 and 25) there was an appreciable increase in the darkening only at 1.26% amino acid concentration. This effect was more significant in the presence of oxygen than in its absence. It is interesting to note that when samples were stored in retort pouches an increase of amino acids from 0.06 to 0.66% had no significant effect on the browning regardless of the levels of ascorbic acid. These results indicate, that in orange juice where the amino acid concentration is below 0.667=,, the amino acids will not significantly affect the browning of the product. Joslyn and March (1935) reported that amino acids play a minor role in the oxidative nonenzymatic browning of orange juice; this has been confirmed for orange juice and for a model system composed of ascorbic acid and glycine and other and no acids (Dulkin and Friedemann, 1956; and Joslyn 1957). Our study further substantiated this finding. The accelerated breakdown of ascorbic acid in the presence of high levels of amino acids has been reported in the literature (Joslyn, 1957; Clegg, 1964; and Seek and Crouzet, 1981). On the basis of the chemical structure of ascorbic acid, it has been postulated that in citrus juice, ascorbic acid reacts with amino acids in much the same

PAGE 101

8.6 II' 03 CJ cL — 1 CU cu u c o u J_J tr. JJ CO II co *— ii CO K CJ a p a •H z H p, £ — CJ H 'to 0 CO
PAGE 102

87 way as sugars react in the Maillard browning system (Joslyn, 1957) The fact that no significant browning occurred between 0.06 and 0.66% amino acid concentration in orange drinks, in the absence of oxygen, suggests that ascorbic acid must first be oxidized to dehydroascorbic acid. Hodge (1953), Clegg (1964), and Kurata et al. (1973) reported that in the presence of amino acids, it is dehydroascorbic acid which is the reactive intermediate in the pathway to furfural and brown pigment production, probably through the Strecker degradation of amino acids as suggested by Reynolds (1965) However, it is possible also that some of the active groups resulting from oxidation of ascorbic acid were polymerizing and not involving amino acids. This is demonstrated by the similar degree of browning in samples containing the same levels of ascorbic acid but different levels of amino acids. It is interesting to note that the intensity of browning under aerobic conditions increased to a maximum and then decreased. The more rapid the initial pigment production was, the more definite was this decrease. Reaction in the retort pouches proceeded so slowly that no maximum was reached during the time the reaction was followed. Our results are in agreement with those of Seaver and Kertesz (1946) who found maximum color production when AA was heated in the presence of glycine, and with Lahikainen et al. (1958) who reported the intensity of browning under aerobic condition to increase to a maximum then decreased. Joslyn (1957) reported that the concentration of AA in browning systems determined whether or not the color production went through a maximum. At low concentration of AA the color increased continuously

PAGE 103

83 'S. i — i< L2 cvi c c— c 5 c > r" c XJ a Vc T — 0 c p. CJ tc p CD a. > c> ai c c CO — 11 03 1! — w W a 01 C — 3 p. 5 r-i O a c w T3 co <; i ; i j i GO I I C I U T i LiJ :— CI < 2 8^ •H r-l 3 6 ^ O U C C r-l 55 8 If o C C T3 CO •H CO < 0 q E < 4-1 CJ O 60 4-1 Q O *j 3 4-4 4-1 W O O nj CM •t— i

PAGE 104

8^ S. — E c p. c >i >. c i — c c a tc c ra u a > c a '0 c w & iJ w w re n re II re re a dJ •H c U — 73 re c a re 13 ; / ( I I I I I I I I I c <3 I P £ ; / ; / i i E < I e< i! c i 6 ED c o o I', 1 \ E < i ^ f r\ 1— i < — I — en CO c •H C M PQ 5 ^ o u o C r-i c if cS CO TJ rH o ^ < o i Si is o co u § O k £ W o m CN M &

PAGE 105

90 with time. The same relation was observed at high concentrations of AA but at intermediate levels the color production went through a maxim um. This bleaching phenomena nay be due to further changes in the decree of polymerization or/and to the precipitation of the brown pigments formed. With regard to the second possibility, considerable dark precipitate was observed in samples stored in polyethylene pouch Statistical Analysis Because of the possible interaction between the different factors involved in the browning of the orange drinks, a statistical analysis of the data was conducted to determine the significance of these interactions at different storage times. Table 11 shows the results Table 11 Three-Factor Analysis of Variance of the Browning Data at 8 weeks. Source df SS MS F Oxygen 1 0.1881 0.1881 62.7 AA 2 0.2514 0.1057 35.2 aa 2 0.0636 0.0411 13.7 Oxy*AA 2 0.1687 0.0963 32.1 Oxy*aa 2 0.0075 0.0125 4.1 AA*aa 4 0.0108 0.0040 1.3 Error 4 0.0091 0.0030 Samples within 54 0.0063 0.0001 Total 71 0.7057 dr= degrees of freedom; SS= sum of squares; MS= mean square F= MS factor/ MS error; AA= ascorbic acid; aa= amino acids. of the 3factor analysis of variance for browning data at 8 weeks. An examination of the F values indicates that the main effects of oxygen, ascorbic acid, and amino acids are statistically significant.

PAGE 106

91 Another important observation drawn from this table is the significance of the twofactor interaction (Oxygen*AA) Therefore, the analysis of variance for the polynomial models for AA was conducted for each type of packaging material. Considering, there are two quantitative factors (ascorbic acid and amino acids) with equally spaced levels, each f actors effect can be partitioned into 2 polynomial effects, namely a linear effect and a quadratic effect. Figure 26 summarizes the effects of ascorbic acid and amino acids on the browning of orange drinks at 8 week storage. Since the Interaction AA*aa was not significant, the effect of ascorbic acid at the 3 levels of amino acids were combined and are presented at each level of oxygen. For amino acids, the effect of both types of packaging and at the different ascorbic acid concentrations were combined. The results indicate that in the retort pouch an increase in ascorbic acid concentration did not affect the browning of orange drinks, whereas in the polyethylene pouch browning increased linearly with increasing ascorbic acid concentration. The effect of amino acids was stricly linear, and was less pronounced than that of ascorbic acid.

PAGE 107

92 £d d d CSJ d £ y c •H > C cc 4J ~ r. 13 •H u S < M •e c 5 u O oq -3 U-l 0 o 8? 4J O c £ 1 '— Ed P2 CM I E

PAGE 108

93 Conclusion In conclusion, this study on orange drinks aseptically packaged and stored at room temperature has yielded additional data on nonenzymatic browning of the product. There were indications that: 1The presence of oxygen greatly accelerated the loss of ascorbic acid and increased considerably the formation of browning pigments. 2Considerable loss of ascorbic acid occurred during preparation and processing of the samples, due to incorporation of air. 3 High levels of dehydroascorbic acid were found in all samples at the zero time, indicating that there was considerable oxidation of ascorbic acid during preparation and processing of the samples. 4Considerable loss of ascorbic acid occurred in the presence of amino acids, indicating that amino acids may have some effect on the degradation of ascorbic acid. 5When samples were stored in retort pouches, the level of dehydroascorbic acid, decreased rapidly during the initial stage of storage, and remained relatively constant thereafter. This is probably due to the reaction of EHAA with amino acids to produce browning pigments. 6Ascorbic acid was the most reactive constituent of orange drinks, with respect to the formation of browning pigments. Its effect on the browning reaction was much more intense in the presence of oxygen. 7Amino acids have an effect on the browning of orange drinks. This effect was stricly linear within the concentration range used, and was more pronounced in the presence of high levels of ascorbic acid.

PAGE 109

EFFECT OF AMINO ACID (XNCE^TRATICN PROCESSING AND STORAGE CONDITIONS ON THE QUALITY OF ASEPTICALLY PACKAGED ORANGE JUICE AND ORANGE DRINKS. Introduction To obtain additional information on the relative reactivity of the major constituents believed to be involved in browning, a second study was conducted. The objective of this study was to investigate the effect of ami n o acid concentration and deaeration vs. no deaeration on the quality of aseptically packaged single strength orange juice and orange drinks, when the packages are stored under aerobic and anaerobic conditions. Material and Methods Reagent Frozen concentrate (Citrus World Inc., Lake Vales, Fla.), high fructose corn syrup (Pacific Gateway Co., San Francisco, CA.), ascorbic acid (Eastman Kodak Company, Rochester, N.Y.), amino acids (Ajinomoto U.S.A. Inc., Englewood Cliffs, N.J.), citric acid and potassium citrate (Pfizer Chemical, Inc., New York, N.Y.), sucrose (local market) FD&C Yellow if 5, FD&C Yellow it 6, and Redd natural Orange Juice Drink Flavor 375316U (Redd Citrus Specialities) were used to make the orange drinks. 94

PAGE 110

95 Orange Drinks Composition Three orange drink mixtures containing 10% orange juice, 38.0 mg/100 ml ascorbic acid and various amounts of amino acids were prepared, aseptically packaged, stored at controlled temperature set to 75F, and examined periodically over a period of 16 weeks. The composition of the orange drinks is given in table 12. In addition, the orange drink mixtures contained 10% orange juice, 5% sucrose, 5% HFCS, 0.01% Redd natural orange juice drink flavor (375316U0) 0.0015% FT&C yellow // 5 and 0.0006% FDUZ yellow // 6. Glucose and fructose were provided by the KFCS which contained 71% solids, 42% fructose, 50% glucose, 5% disaccharides, and 1.5% maltose. A mixture containing equal amounts of arginine, aspartic acid, and 4 aminobj tyric acid was used to make the drinks. The final degree Brix of the orange drinks was 12. Table 12 Composition of Orange Drinks Products Ascorbic acid Amino acids other mg/100 ml added w/w constituents Ml 38.0 0.0% All mixtures M 2 38.0 0.4% contained 0.6% M 3 38.0 0.8% citric acid and 0.2% potassium M 0 38.0 S. S.O.J. citrate

PAGE 111

96 Preparation of the samples The juices and drinks were aseptically filled into flexible packages, using the Tetra Pak, Inc. research pilot plant facilities in Irving, Texas. The packaging material is a laminate of: (from outside to food contat surface) polyethylene/printed paper/polyethylene/ aluminum foil/polyethylene/polyethylene. The aluminum foil gives the package the excellent barrier properties to both light and oxygen. The product contact surface of polyethylene provides an inert lining for the package and also is necessary for forming the hermetic under the proper conditions of temperature and pressure. The mechanics of the Brik Pak system can be divided into four functions: 1Sterilization of the product 2Sterilization of the packaging material 3Forming and filling in sterile surrounding 4Sealing of packs to prevent recontami nation In the Ultra High Temperature (UHT) processing the product was rapidly heated to 195F and held for 10 seconds. The juice was then cooled to room temperature and conveyed in a closed system to a vertical formfill and seal machine, Model AB3. This machine as described by Russo and Bannar (1981) starts with a reel of printed, laminated material web containing 2,000 to 22,000 packages which is loaded into a cassette in the back c£ the machine. After attachment to the machine, the web is drawn upwards to the top of the unit where it is stamped with the date and code number. A plastic longitudinal strip is applied to one edge of the paper. This strip has two functions: to reinforce tine longitudinal seam, and to

PAGE 112

97 prevent the product from coming in contact with paper edge. After application of this strip, the material passes through a peroxide bath where it is coated with a thin film of 35% hydrogen peroxide for chemical sterilization. A pair of pressure rollers removes surplus hydrogen peroxide which runs back into the sterile bath. Passing a bending roller on the very top of the machine the packaging material starts its way down towards the front of the machine through forming rings which shape the material into a longitudinally sealed tube. The remaining hydrogen peroxide is evaporated. The product to be packed is introduced through a stainless steel filling tube which is jacketed by a second tube through which sterile hot air can be blown into the paper tube. As an initial stage in the sealing of the longitudinal seam, one edge of the packaging material passes through an element which is heated by hot sterile air. The longitudinal seam is sealed in this forming ring where both edges of the packaging material are pressed together. The tube heater consists of a coil-shaped electric element which heats the inside of the packaging material with radiant heat. This radiant heat sterilizes the packaging material and at the same time a sterile atmosphere is created above the liquid level. The flow of product into the tube is modulated by a butterfly valve which in turn is controlled by a float. The level of product in the tube is approximatly two feet above the bottom transversal seal and is regulated mechanically by a float so that it is always higher than the mouth of the filling tube. By this means, frothing is avoided. Transversal seams are done at regular intervals below the level of the

PAGE 113

98 product. In order to seal transversely, the product has to be squeezed away from the sealing zone. This is done by closing the sealing jaws, applying pressure and then heat. Individual units are cut at a rate of about one package per second. The pouches thus obtained are fed into the final folder, where they receive their brik like shape by sealing the flaps down onto the sides and the bottom of the package respectively. To evaluate the effect of deaeration on the quality of aseptically packaged orange juice two types of products were obtained; one where the juice obtained was deaerated by vacuum. In the second the deaerator was turned off and the juice obtained was nondeaerated. All orange drinks were deaerated. The samples were then stored at 75F under aerobic or anaerobic conditions as indicated in Table 13. For practical reasons, sensory evaluation was not performed on orange drinks M 2 and M 3 stored anaerobically. Table 13 Processing and Storage Conditions of Orange Juice and Orange Drinks Orange juice Orange drink deaerated Time wks OJD ANA. AER CJND AER M 1 M 2 M 3 ANA. AER ANA AER ANA AER 0 4 /. 4 4 2 4 2 4 4 4 4 4 4 2 4 2 4 8 4 4 4 4 4 2 4 2 4 12 4 4 4 4 2 4 2 4 16 4 4 4 4 4 2 4 2 4

PAGE 114

99 Immediately after processing samples to be stored anaerobically were placed into BBL GasPak jars (150 ram) where the anaerobic atmosphere (H2 and CO2) was provided by BBL GasPak anaerobic systems (Becton Dickinson and Co. Cockeysville, MD) Tne samples were then transported to the University of Florida, Gainesville for storage and analysis. One day after arrival the anaerobic samples were transferred to 2 large glass containers where air was removed using a vacuum pump and was replaced with a nitrogen atmosphere. The anaerobic conditions were controlled through a GasPak Disposable Anaerobic Indicator, and nitrogen was continously flushed as needed. All analytical measurements were made on 3 replicate packages and the coefficient of variation was small. Methods of Analysis Ascorbic and Dehydroascorbic Acids Ascorbic and dehydroascorbic acids were determined using the procedure of Kacem et al., (1986). Browning Production of browning pigment expressed as absorbance at 420 nm, was measured spectrophotometrically using the method of Meydav et al. (1977) Amino acids analysis Amino acids were measured using a Beckman amino acid analyzer Model 119 CL, equipped with a 6x460 mm column and W3 resin with a bed height of 220 mm. The buffer and ninhydrin flow rates were 44 and 22 ml/h respectively. The recorder chart speed was 6 in/h. The amino acid analyzer was connected to a 3390 A Hewlett-Packard integrator.

PAGE 115

100 Sensory evaluation Sensory evaluation (color and flavor) of samples was conducted at various tines during storage by at least 15 inexperienced taste panelists from among department personnel, using a 1 to 9 hedonlc scale (9 being the best) to establish general acceptability of the products. Statistical analysis Regression methods were used for the calculation of factor effects, and for the analysis of variance (ANOVA) Results and Discussion Orange Juice Ascorbic acid and dehydroascorbic acid concentrations Figure 27 and Figure 28 show ascorbic acid and dehydroascorbic acid retentions as a function of storage time. An examination of these plots reveals that there were some differences among the 3 juices. The nondeaerated juice showed the largest loss of ascorbic acid. The anaerobically stored juice, had better ascorbic acid retention than the aerobically stored juice. At the end of the 16-week storage period, the retention of ascorbic acid in orange juice deaerated (OJD) anaerobically stored, OJD aerobically stored, and orange juice nondeaerated (OJND) were, 97, 91, and 86% respectively. The fact that the nondeaerated juice showed the highest drop in ascorbic acid content during the first 4 weeks of storage indicates the importance of oxygen in ascorbic acid degradation. After this period, retention of ascorbic acid in both types of juices under the same storage conditions were practically parallel.

PAGE 116

101 > CJ 1! n = M c co < < it : ii i ii | iiJ ii : ii i :i I II i 1 1
PAGE 117

102 5J c n 0) to (3 05 H iJ C •H c p. a cc Ixl < IX — CO HJ 9J 'J •rl — OJ DO c o — •u II £ C3 t— i-l tfl o -8 8 1 M .5 w QQ OJ 2 8 oj £ p; £ •H ,Q u 8 o o •H C s O rH ^5 00 8 to lt;

PAGE 118

103 Deaerated and anaerobically stored juice exhibited the highest ascorbic acid retention value. This result suggests that some oxygen was transmitted through the package seal area under aerobic conditions. The heat seal area, general ly polyethylene or polypropylene, may permit some gas exchange. No significant differences in the concentration of dehydroascorbic acid were noted between the juices. The concentration of DHAA remained practically constant during the storage period except for the nondeaerated juice which showed a slightly higher concentration during the first 4 weeks. Figure 28 illustrates the loss of ascorbic acid potency (log scale) as a function of time. The retention of ascorbic acid followed a first order kinetic model since the data fit a straight line. It is possible that the reaction is pseudo first order. Studies have shown ascorbic acid degradation to be first order (Huelien, 1953; Waletzko and Labuza, 1976; Saguy et al., 1978; Passy and Mannheim, 1979). From this plot the rate constants of ascorbic acid destruction (K) were determined for OdD anaerobically stored, OJD aerobically stored, and OJND and were 0.84, 2.32, and 3.49 weeks~l, respectively. Browning Changes in color during storage are presented in Figure 29. No significant difference in browning could be observed in the three types of juices. The limited browning could be due to the fact that insufficient ascorbic acid was degraded (86-97X retention) Curl and Talburt (1961) found that browning in citrus juices could be observed only after a 10 to 15% decrease in ascorbic acid content had taken place. The absorbance increased with storage time, but this

PAGE 119

104 increase took place only after an initial lag period (Joslyn, 1957; Karel and Nikerson, 1964) During this lag period, colorless conpounds are probably formed which do not contribute to an increase in absorbance. Joslyn and Marsh, (1935) and Joslyn et al.j (1934) found in Valencia orange juice that storage at elevated temnerature does increase the rate of browning, and darkening began to occur when the iodine reducing value of the juice had decreased from 26.0 to 12.9. They believed that reducing substances must be almost completly destroyed before browning can begin. Hamburger and Joslyn (1941) suggested that the darkening occurs after the ascorbic acid is in the dehydroascorbic acid form and when no readily oxidizable substances are left in the juice. Sensory evaluation Sensory evaluation of the juices was performed using the hedonic scale. As flavor was judged by taste, it cannot be presented as a sharply defined analytical result. Figure 30 shows the mean sensory scores for the three juices as a function of storage time. In general, all samples of the juices showed slight change in flavor. The mean value for all samples at zero time was 7.2 (like moderately), and. at 16 weeks was 5.9 (like slightly). Each 4 week time period was subjected to an analysis of variance and there was no significant difference at any time period among the three juices at a 951 probability level; and ac the end of the 16 weeks storage period, the different orange juices were considered equal in flavor, with mean scores of 5.6, 6.0, and 5.4, respectively for anaerobically, aerobically stored and nondeaerated orange juices.

PAGE 120

105 <3 o to c o ca C3 < i — CD 5 c H W 4

PAGE 122

107 Orange Drinks Ascorbic and dehydroascorbic acids The degradation of ascorbic and dehydroascorbic acids in aseptically packaged orange drinks, stored under different conditions was studied as a function of and no acid concentration and storage time. The experimental results for the loss of ascorbic acid are presented in Figure 31 and Figure 32. Each ascorbic acid and dehydroascorbic acid value is the average for three packages. These data (Figure 32) conformed to the first-order function d(C) = K(C) dt where (C) = molar concentration of ascorbic acid t = time (weeks) K = first-order rate constant (weeks~l) The experimentally determined rate constants for ascorbic acid degradation in the different orange drinks are presented in Table 14. Examination of these data showed a significant increase in the rate constants of ascorbic acid loss with the increase of amino acids level from 0.4% to 0.8%. When the amino acids level was increased from 0.0 to 0.4 the rate constant did not increase significantly. These results are in agreement with our previous experiment when amino acids significantly affected ascorbic acid degradation only when added at the 1.26% level.

PAGE 123

108 < i ^ ™ j mi] i i i i I I 1 < I 1 I I i i i o sail 60 tS <-<5 ; 0 I I I I I il I il i i! 1 !' r-i 03 C C < HI L3 =: c I— tn 6 D-l o c c C CS AJ 4J CO C S-j CO 4-' AJ c c c O O f-l u ^< o o
PAGE 124

109 S U C — c to X M 0 II II a a c c T5 a c CO 'C < — in 03 03 O 5 C — 8*8 c tb 0! CJ I— I CJ W CJ C 03 CD s a 4J c o 0 c •U H 0) s PS < •t-i ,0 O < T3 O o o •H C CD Tj 3 O r-l CJ M-l 03 C CM CO 12 5c

PAGE 125

110 Table 14 Rate Constants (weeks 1) for Ascorbic Acid Loss as Function of Amino Acid Content and Storage Conditions in Orange Drinks Amino added acids (%) Storage ccnditi oris Anaerobic Aerobic 0.0 1.49 3.45 0.4 1.35 3.85 0.8 2.56 4.67 In general, ascorbic acid losses in samples stored under aerobic conditions were greater than in the samples stored under anaerobic conditions. At the end of the 16 weeks storage period the retention of ascorbic acid in orange drinks Ml, M2, and M3 was 94.4, 94.9 and 89% for samples stored anaerobically and 86, 84, and 81% for samples stored aerobically. Diffusion of oxygen through the package seal and the high sensitivity of ascorbic acid to oxygen appear to be the main reasons for this difference. Considerable loss of ascorbic acid occurred in the presence of the highest level of amino acids (0.8%) but no difference occurred when the amino acid level was increased from 0 to 0.4%. This is in agreement with our previous results, and indicates that high levels of amino acids affect the ascorbic acid degradation. Browning One of the main reasons for the reduction in the commercial value of citrus products is the nonenzymatic browning. The data for the browning expressed as absorbance at 420 nm and as a

PAGE 126

Ill function of amino acid content are given in Figure 33. No detectable browning occurred in any of the orange drinks. The absorbance measurement r emain ed almost constant for the 16 week storage period. This delay in color changes, also observed by Karel and Nickers on (1964) was referred to as the tim e during which colorless intermediates of the browning reaction were formed (Clegg and Morton, 1965) The zero time difference in absorbance between the orange drinks is probably due to experimental error in the initial drinks composition. Amino acids Amino acid content in the orange drinks and as a function of storage time are given in Table 15 and Table 16. No detectable difference was observed during the 16 week storage. The amino acid concentration remained almost constant during the entire storage time. This result is in agreement with that of Joslyn and Marsh (1935) who observed that the amino-nitrogen level remained practically constant during the course of browning of Valencia and Navel juices, even after 126 days of storage at room temperature, even though the juice had become very dark brown in color. Flavor Flavor scores were not significantly different among the different orange drinks. At 16 weeks the mean flavor scores were 5.7 and 6.1 for orange drink Ml stored anaerohically and aerobically respectively. For orange drinks M2, and M3 the mean flavor scores were 5 and 5.2 respectively, and were not significantly different.

PAGE 127

112 i, 3 i i i i i i i i i i i c£m i i i i i i i i jo o j-j JJ CO h a CM LLl c c K T3 bJ IS < c i— CO id 5 I C Cj BPS e W ^ CO C C •r-i CO U E a c c CO m I

PAGE 128

113 Table 15 Changes in Amount of Amino Acids in Orange Drink M2 and M3 as Influenced by Storage Time (mg/lOOml) Amino acid Storage conditions 0 stor 4 age ti 8 me (.weeks) 12 16 M2 Aspartic acid Ana 129 105 122 127 120 Aer 127 107 122 122 123 Arginine Ana 172 170 173 180 177 Aer 172 175 175 183 170 4-anri nobutyric Ana 140 138 135 139 140 acid Aer 137 140 129 138 138 M3 Aspartic acid Ana 191 168 203 210 190 Aer 209 165 208 195 195 Arginine Ana 257 241 253 268 245 Aer 263 256 251 265 252 4-ami nobutyric Ana 241 240 242 247 240 acid rter 252 246 233 O/ A 9A5

PAGE 129

114

PAGE 130

115 Conclusion This study on single strength orange juice and synthetic orange drinks containing 10% orange juice, resulted in the following findings 1Deaeration of orange juice resulted in increased retention of ascorbic acid, and had no effect on sensory quality of the juices. 2Anaerobically stored juice had greater ascorbic acid retention than aerobically stored juice. However, there was no significant difference in sensory quality and browning between anaerobic and aerobic storage. 3There was no significant difference in dehydroascorbic acid in orange juice and orange drinks due to storage conditions or deaeration. 4There was no effect of amino acid concentration, anaerobic or aerobic storage on flavor by sensory evaluation orange juice or orange drinks. 5The amino acid concentration of orange drinks remained constant during the entire storage period. 6Increasing the amino acid concentration of orange drinks from 0.4 to 0.8% increased the ascorbic acid loss.

PAGE 131

SUMMARY The study on single strength orange juice and synthetic orange drinks containing 10%, orange juice evaluated the effect of ascorbic acid, amino acids and oxygen permeable versus non permeable packaging. Tne effects of deaeration and anaerobic storage versus aerobic storage for orange juice aseptically packaged in commercial Tetra Pak flexible carton and stored at 75F were also evaluated. Oxygen and ascorbic acid were found to be the most critical factors. Amino acids played an important role only in the presence of oxygen. The diffusion of atmospheric oxygen through the polyethylene film increased significantly the ascorbic acid degradation and the brown pigments formation. It is therefore, reasonable to conclude that the brown pigments in orange drinks originate from the oxidation of ascorbic acid. High level of amino acids (1.26%) increased ascorbic acid loss in samples stored in the retort pouch (zero permeability to oxygen), and resulted in an increase in browning. The results indicate the possibility of at least two mechanisms for the formation of browning pigments depending on whether or not oxygen is present. Under the actual storage practice used in the aseptic packaging and because of the high acidity (pH = 3.8), it appears that of the three possible modes of browning, the ascorbic acid theory seems to be likely to occur in orange juice and orange drinks. The Maillard reaction plays 116

PAGE 132

117 a minor role. This is in agre&ment with Joslyn and Marsh, (1935) who showed that the removal of sugars from orange juice had no effect on the browning of the product. However, the reviews of browning reactions have, in general, emphasized the conplexLty of the subject and the lack of specific knowledge on the chemical reactions and intermediates involved. Browning, of whatever type, is caused by the formation of unsaturated, colored polymers of varying composition. It should be possible to analyze the browning complex by identifying the individual reactions and studying each in model systems, and determine which of these routes are operative and to what extent one affects the other in a given food, under a given set of conditions. The isotonic tracer techniques seems a promising outlook for an acceptable solution. Unfortunately, even though the use of this technique was an original objective of this study, it had to be omitted for financial reasons.

PAGE 133

APPENDIX

PAGE 134

cn Ul UJ Ul < a — •H M 0) o a >^ *c o CJ C3 c c •3 a) jj UJ C DO o c T-i W 4J JJ CJ O C t-J oj cj a, 14J 4J B UJ CJ CO K pi to < to •1*1

PAGE 135

to LU CO -< cr C ii CJ CM •e* O TD CJ i-l CO CJ < s ,£ O g c5 CJ DO a 0 cant 0 S ^ n U M-l Q) C O r— f cu C CD +3 O > CU U 0) >> c 33 4J O "H C CO •H T-i Q)
PAGE 136

121 Q q\ on co on i w% CM r— t-H r— f 3§£ in cc £\ CO Q K o s \£> CM CM ON O O CO SJ 0 £ 4J C K O W CM 0) CQ DO |CM (!) 5 5) 5 O CC cm CM in — i in CM in 4b CM CM CM CM CM + 1 in c CM cm r — in CM VO fi+l-H C <0 CM C CM CO r 1 V s P) o c o CN CM CO O I-! r-H cM one CN I OHH HHrn CC CO ON coo CM CO in on co -i-i — • —i to> co co r— I CM CC _jn SJ"_7^ t r~r r~ -r'-H -r< CM CO -3" C r-< ON CO CM m o o CM r— ON On CO vO ON CO o O O !-< CO • co co oc ON On CO +1+1
PAGE 137

122 o ZM C r-l CM ^ & n 0 +1+1 + i co co in CC' CO o cm co
PAGE 138

123 co I < — f co U_i r; O "0 9 § a c c c co i-i CO C 0) o u •H O 4-> 4J co co U AJ CO C CD O H CJ c c 6, •r0 < u •r-l 1 3 CO CN 'J = u cfl M c — CO CN 4J ri c r : 4-1 to CN CN CN CN I I LP) co co c i i i ceo • • T en CO CO CN CO OC O" 1 OC LO oo -"-14-1-4-1 to vC 0% CO "0 CO CO CO CM C^. OC U". +•+1+1 +•++ •f'-h'-r 1 HOC KhO ocOnC CN CO i— I CO -Cj" CO CO CCO VO CM CM +i+t-y H H C + •+1+1 "co"5"£n CN CM rr-~ cm 1— 1 .— 1 H vj" CN UO on
PAGE 139

124 c I I I I I I H c co y u-i 5 o (L c a c c o >. >> C5 i— i O w CC H o •i-l 2 & J co s c cu O rH C5 & •H W CJ w — 1 cj r. Si 4J 0! 5 u c AJ CN cc C +i c CN C *i +1 # C r-, I I I coo c c c -M+i+i o c o o o ec c +(+!+( i t i r-> r-l O COO £c' t^o i— i — I CN COO I I I I I I o c o c o c c'c'c' tJ I I I c c c c o o 0> CN O c c c coo ++!+! +J+I+1 cn r1 ro COO occn r-<
PAGE 140

125 .1 H CD I CO o 1-1 CJ g c 5 O 1-i •H 4J 22 U O C 4J cu ca u c co T3 & •r-l CJ CJ CO 1 U CO CO CD B c r-i r-l CO r- t— m • • • o *i CM CM • 1 1 i -M+l-f! m in co o> co r • • • cc r • i & r-t -V \ r>co ~cn 6o"tt • CM • • CM* CO* CM • • • vi- 1 — i l-H 1—1 cm co m HHH S-l — j CM CO r" T ^" fc-t *-t <£ in £ IS S tee 0^* r*i1

PAGE 141

126 o CM I I I I I I I I I M 1 o x o C 3 o o O 0) £ c a a; 03 >•> C c c H 4J TJ C
PAGE 142

127 Table A-7 Absorbance at 420 ran as a Function of Storage Time (samples stored in Tetra Pak carton) Storage Time Pdt a 0 4 8 12 16 OJDAE .200.003 .234+. 005 .201+. 003 .217+. 005 .237+. 003 OJMN .224+. 005 .210+. 006 .188+. 004 .208+. 005 .234+. 009 OJND .215+. 007 .224+. 010 .221+. 005 .225.009 .250t.012 ODLAE .439t.012 .429+.005 .424+.009 .408+. 005 .414+.002 0D1AN .436+.023 .424+. 009 .426.O07 .408+.008 .410+. 007 0D2AE .544+ 007 .523+-. 012 .536.008 .540+. 004 .545.004 0D2AN .545+, 019 .534+. 015 .531.006 .540.001 .540+. 009 0D3AE .501+. 026 .508.023 .483+.004 .489t.0O4 .494t.008 0D3AN .492+.031 .487+.018 .484t.003 .486+. 009 .485+. 005 a + Number of pouches = 4. Mean — SD OJDAE = orange juice deaerated aerobically stored 0JDAN = orange juice deaerated anaerobically stored OJND = orange juice nondeaerated 0D1AE = orange drink with no amino acids added, aerobically stored 0D1AN = orange drink with no amino acids added, anaerobically stored 0D2AE = orange drink with 0.4% amino acids, aerobically stored 0D2AN — orange drink with 0.4% amino acids, anaerobically stored 0D3AE = orange drink with 0.8% amino acids, aerobically stored 0D3AN = orange drink with 0.8% amino acids, anaerobically stored

PAGE 143

128 Table A-8 Ascorbic Acid Retention (%) as a Function of Storage Tine (samples stored in Tetra Pak carton) itorase Tune Pdt 4 T2~ T5~ OJDAE 100 95.3 9^.0 90.7 90.7 OJDAN 100 99.7 99.4 97.1 97.1 OJND 100 93.8 91.3 86.7 86.7 0D1AE 100 96.7 92.5 89.8 86.2 0D1AN 100 100.0 99.3 96.0 94.4 0D2AE 100 95.7 91.8 89.0 84.1 0D2AN 100 • 100.0 99.4 96.4 94.9 0D3AE 100 94.2 91.1 86.1 80.7 0D3AN 100 97.1 93.6 92.9 89.3 OJDAE = orange juice deaerated aerobically stored 0JDAN = orange juice deaerated anaerobically stored QJND = orange juice nondeaerated 0D1AE = orange drink with no amino acids added, aerobically stored 0D1AN = orange drink with no amino acids added, anaerobically stored 0D2AE = orange drink with 0.4% amino acids, aerobically stored 0D2AN = orange drink with 0.4% amino acids, anaerobically stored 0D3AE = orange drink with 0.8% amino acids, aerobically stored 0D3AN = orange drink with 0.8% amino acids, anaerobically stored

PAGE 144

129 Table A-9 Flavor Score as a Function of Storage Time (samples stored in Tetra Pak carton) Storage Time Pdt a 0 4" S 12 Id" OJDAS 7.1+1.8 6.4+1.4 5.42.1 6.0+1.5 6.41.4 OJmN 7.3+1.6 7.1+1.8 6.1+1.5 6.4+1.4 5.61.3 OJND 7.31.2 6.91.3 6.3+1.6 6.5+2.0 5.4+1.9 0D1AE 6.5+2.5 6.6+1.5 6.7+1.8 5.8+2.1 6.2+1.3 0D1AN 6.6+1.0 6.31.5 5.92.0 6.2+1.1 6.0+1.5 0D2AE 6.8+1.5 6.4+1.2 5.52.3 5.8+1.5 5.1+1.8 OD3AE 5.9+2.0 6.1+1.2 5.9+1.8 6.0+1.5 5.4+1.6 Number of score = 15. Mean SD 0JDAE = orange juice deaerated aerobically stored OJDAN = orange juice deaerated anaerobically stored OJND = orange juice nondeaerated 0D1AE = orange drink with no amino acids added, aerobically stored 0D1AN = orange drink with no amino acids added, anaerobically stored 0D2AE = orange drink with 0.4% amino acids, aerobically stored 0D2AN = orange drink with 0.4% amino acids, anaerobically stored 0D3AE = orange drink with 0.8% amino acids, aerobically stored 0D3AN = orange drink with 0.8% amino acids, anaerobically stored

PAGE 145

130 Table A-10 Sensory Evaluation Form Name: Date Please evaluate these samples for flavor. Taste test each one. Use the appropriate scale to show your evaluation and check the point that best describes your feeling about the flavor of the sample. Code Code .Like extremely .Like very much .Like moderatly .Like slightly .Neither like nor dislike .Dislike slightly .Dislike moderately .Dislike very much .Dislike extremely .Like extremely .Like very much .jjike moderatly .Like slightly .Neither like nor dislike .Dislike slightly .Dislike moderately .Dislike very much •Dislike extremelv Comments :

PAGE 146

REFERENCES Adams, J. P. 1982. Orange juice quality as a function of oxygen concentration during packaging and storage. Proceeding of the 22 nd Annual Schort Course for the Food Industry, p. 245. University of Florida. Gainesville. Ashoor, S.K., Monte, W.C., and Wei ty, J. 1984. Liquid chromatographic determination of ascorbic acid in foods. J. Assoc. Off. Anal. Chem. 67: 78. ACAC. 1975. "Official Methods of Analysis," 12th ed. Association of Official Analytical Chemists, Washington, DC. Beattie, H.G., Wheeler, K.A. and Pederson, C.S. 1943. Changes occuring in fruit juice during storage. Food Res. 8: 395. Berry, R.E., Wagner, C.J.,Jr., and Bissett, O.W. 1970. Storage stability of foam-mat instant orange juice as related to pH. Proc. Fla. State Hortic. Soc. 84: 193. Bissett, O.W. and Berry, R.E. 1975. Ascorbic acid retention in orange juice as related to container type. J. Food Sci. 40: 178. Blair, J.S. 1964. Thiofurfural as an important factor in the development of offflavor in canned orange juice during storage. Univ. of Florida Citrus Expt. Sta. Mimeo Rpt. CES. 56:4. Blair, J.S., Godar, E.M., Masters, J.E., and Riester, D.W. 1952. Exploratory experiments to identify chemical reactions causing flavor deterioration during storage of canned orange juice. Food Res. 17: 235. Bchardt, G.S. and Carson, J.F. 1955. Effect of trace metals, oxygen, and light on the glucose-glycine browning reaction. ITatjre. 175: 470. Boyec, J.M. and Peterson, G.T. 1945. Oualitv of canned orange iuice. Ind. Eng. Chem. 37: 370. Braverman, J.B.S. 1963. "Introduction to the Biochemistrv of Foods," Elsevier Publishing Co. Amsterdam. Brenner, S., Wodicka, V.O., and Dunlop, S.G. 1943. Effect of high temperature storage on the retention of nutrients in canned foods. Food Technol. 2: 207. 131

PAGE 147

132 Burton, H.S. and McWeeny, D.J. 1964. Nonenzymatic browning route to the production of melanoidins from aldoses and aminocompounds. J. Chem. and Ind. 1964: 462. Burton, H.S., McWeeny, D.J., and Biltcliffe, D.O. 1963. Nonenzvmatic browning development of chromophores in the glucose-glvcine and sucrose-glycine systems. J. Food Sci. 28: 631. Chatfield, C. and Adams, G. 1940. Proximate composition of American food materials. Circ. No. 549, U.S. Dept. of Agr. Washington, DC. Chichester, CO. Stadtman, F.H., and Mackinney, G. 1952. On the products of the Maillard reaction. J. Am."Chem. Soc. 74: 3418. Cier, A., Nofre, C, Drevon, B. and Lefier, A. 1959. Etude de la degradation de l'acide ascorbique sous atmosphere inerte. Bull. Soc. Chim. Fr. 13: 74. Citrus Fruits, Production, Use, and Value. 1984. US. Dept. of Agriculture, Crop Reporting Board, Statistical Reporting Service. Washington, DC. Citrus Valued at Record $1.04 Billion, 1985. Gainesville Sun October 6: 1. Clark, B.S. 1941. Technology of canned juices. Fruit Products J. 9: 265. Clegg, K.M. 1964. Non-enzymic browning of lemon juice. J. Sci. Food Agric. 15: 878. Clegg, K.M. 1966. Citric acid and the browning of solutions containing ascorbic acid. J. Sci. Food Agric. 17: 546. Clegg, K.M. and Morton, A.D. 1965. Carbonyl compounds and the nonen2ymatic browning of lemon juice. J. Sci. Food Agric. 16: 191. Coggiola, I.M. 1963. 2,5-dihydro-2furoic acid: a product of the anaerobic decomposition of ascorbic acid. Nature, Lond. 200: 954. Cole, S.J. 1967. The Maillard reaction in food products carbon dioxide production. J. Feed Sci. 32: 245. Cooper, W.C. and Chapot, H. 1977. Fruit production with special emphasis on fruit for processing. Ch. 1. In "Citrus Science and lechnology, vol. 2. S. Naggy, P.E. Shaw, and M.K. Veldhuis (Eds.), AVI Pub. Co., Westport, Conn. Curl, A.L. 1947. Concentrated orange juice storage studies. The effects of degree of concentration and of temperature of storage. Canner. 105: 14-16, 38, 40, 42. ^

PAGE 148

133 Curl, A.L. 1948. Gas formation in concentrated orange juices and analogous synthetic mixtures. Food Res. 13: 381. CU^1 ^G6^V1 1 ?nc^^ ^ corbic a ? id losses and darkening on storage at w c C120CF) Or syntnetic mixtures analogous to oranse iuice Food Res. 14: 9. J Curl, A.L., Moore, E.L., Wiederhold, E. and Veldhuis, M.K. 1946 Concentrated orange juice storage studies with particular rererence to the deveicr-ment of swells. Fruit Products J 26: 101. Curl A.L and Talburt, W.F. 1961. Deterioration in storage. Ch. 14. in Fruit and Vegetable Juice Processing Technology D K Tressler and M.A. Joslyn (Eds.), p. 410. AVI Publishing* Co. Westport, CT. Curl A.L and Veldhuis, M.K. 1947. The origine of off-flavor which develops in processed orange juice. Fruit Products J. 26: 329. CUrl '^ ,L : J an 5 Y eldhuis > M.K. 1948. The composition of the sugars in Honda Valencia orange juice. Fruit Products J. 11: 342-343, 361. Dennisor^ D.B., Brawley, T.G., and Hunter, G.L.K. 1981. Rap id high performance liquid chromatography determination of ascorbic acid and combined ascorbic acid dehydroascorbic acid in beve-ases J. Agri. Food Chem. 29: 927. ~ 5 Dennison, D.B and Kirk, R.J. 1978. Oxygen effect on the degradation 43S 609 a ld 3 dehydrated food system. J. Food Sci. Dennison, D.B. and Kirk, R.J. 1982. Effect of trace mineral fortification on the storage stability of ascorbic acid in a dehydrated model food system. J. Food Sci. 47: 1198. Dinsmore, H.L. and Nagy S. 1971. A rapid gas chromatograDhic method tor studying volatile carbonyl coroounds from orange" iuice and their changes during storage J. Agr. Food Chem. 19: 517. Dormer L.W. and Hicks, K.B. 1981. High performance liquid chromatography separation of ascorbic acid erythorbic acids dehydroascorbic acid, diketogulonic acid, and dike tccluconic acid. Anal. Biochem. 115: 225. ~ Doss, K.S.G. and Ghosh, S.K. 1949. Kinetics of color development in invert sugar solutions, Part II. Proc. Sugar Technol. Assoc. \jsi£cl) 16: lb. Du Bois C.W.and Kew, T.J. 1951. Storage temperature effects in rrozen citrus concentrates. Refrig. En?in. 59: 772.

PAGE 149

134 Um w S ; I i;,r d Fl ^ £d f aana T E 1956. The role of dehydroascorbic £ nfTlt^ 0 aCldS in bTOWnin S reaction^ 1C Einson-Perchcnok H.M. and Downes, W.T. 1982. Kinetics of ascorbic ^^^J^Jt^i^ W concentration E111S Che^strJ 59 i2' n l3 Maillard reaCti n ******* Carbohydrate EVSRd S !;^;l.^sh G.L 1943. Effect of storage te^erature on die retention of ascorbic acid in orange juice. Food Res. 13: 244. Flnle J'i V Du3I ^> E 1981 Resolution of ascorbic de^ycroascoroic and diketcgulonic acids bv paired-ion reversed phase chromatography. J. Chrom. 207: 449 reversed Florida Corp and Livestock Reporting Service, Florida Agricultural Statistics. 1980. "Citras Summary," Orlando, Florida. ^^sLSSi? WsT^^ 0 ^ 8 Se ^ lc f' Fl0rida Agricultural statistics. 1985. Citrus Summary," Orlando, Florida. ^'s^l^of 1 ^^^ f diSS 1Ved OXy 3 en milk on the stability of some vitamins towards heating and during subseouent exposure to sunlight. J. Dairy Res. 34: 239. suosequent Freed M., f Brenner, S and Wodicka, V.O. 1949. Prediction of te^.TSs? acid stability 1X1 stored ^ f ds ^soluHAn^iS^' E S 5* ?Iarris R 3 1948 Stability of Iclence lSf: P S6. aSC rClC ^ f dehydroascorbic' acid. Gmt 1rcceed4s 3 'JS% ^-tpn and Processing Directions. ^Ttn^r^itv^fn^JTr 3hort .?? urse f <* the Food Industry, p. Jiij.versiry or fionda, Gainesville. HaaS *TlI; A A^ a ? d E.R. 1949. Deterioration of dried fruits bse of ion excnange resins to identifv types of comoouncS involved in browning. Inc. Eng. Chem/41893 COmpOUndS

PAGE 150

135 Kali, J. A. 1927. Summary of the orange juice problem. Unpublished report of research lab. Calif. Fruit Growers Exchange, Los Angeles. Hamburger, J.J. and Joslyn, M.A. 1941. Autooxidation of filtered citrus juices. Food Res. 6: 599. Handwerk, R.L. and Coleman, R.L. In press. Approches to the Citrus browning problem. A review. J. Agric. Food Chem. Herbert, R.W. Hirst, E.L., Percival, E.G.V., Reynolds, R.J.W., and Smith, F. 1933. The constitution of ascorbic acid. J. Chem. See., London. 2: 1270. Hodge, J.E. 1953. Chemistry of browning reaction in model systems. J. Agric. Food. Chem. 1: 928. Hodge, J.E. 1976. Carbohydrates. Ch.3. In "Principles of Food Science, Part I, Food Chemistry." O.R. Fennema (Ed.), p. 41. Marcel Dekker, Inc. New York, NY. Hodge, J.E. and Osman, E.M. 1976. Lipids. Ch. 4. In "Principle of Food Science, Part I, Food Chemistry." O.R. Fennema (Ed.), p. 139. Marcel Dekker, Inc., New York, 'NY. Horowitz, I., Fabry, E.M., and Gerson, CD. 1976. Bioavailability of ascorbic acid in orange juice. J. Am. Med. Assoc. 235: 2625. Horton, B.P. and Dickman, R.S. 1977. Stability of ascorbic acid in reconstituted frozen orange juice. J. Food Protection. 40: 584. Hough, L., Jones, J.K.N. and Richards, E.L. 1953. The reaction of aminocompounds with sugars II. The action of ammonia on glucose, maltose, and lactose. J. chem. Soc. 1953: 2005. Huelin, F.E. 1953. Studies on the anaerobic decomposition of ascorbic acid. Food Res. 18: 633. Hume, H.H. 1926. "The Cultivation of Citrus Fruits. Data for 1886-87 through 1908-09." The Macmillan Co., New York, NY. Jackson, S.F., Chichester, CO., and Joslyn, M.A. 1960. The browning of ascorbic acid. Food Res. 25: 484.' Jameson, R.F. and Blackburn, N.J. 1975. The role of Cu-Cu dinuclear complexes in the oxidation of ascorbic acid bv oxwen. J. Inor^. Nucl. Chem. 37: 809. Jensen, A. 1967. Tocopherol content of seaweed and seameal. Influence of processing and storage on the content of tocopherol, carotenoids and ascorbic acid in seaweed meal. J. Sci. Food Agric. 20: 622.

PAGE 151

136 Johnson, L.R. and Toledo, T.R. 1975. Storage stability of 55 Brix orange juice concentrate aseptically packaged in plastic and glass containers. J. Food Sci. 40: 433. Joslyn, M. A. 1941. Color retention in fruit product. Ind. Eng. Chen. 33: 308. Joslyn, M.A. 1957. Role of amino acids in the browning of orange juice. Food Res. 22: 1. Joslyn, M.A. and Marsh, G.L. 1932. The relation of deterioration of orange juice to its iodine reducing value. Science. 76: 82. Joslyn, M.A. and Marsh, G.L. 1934. Iodine reducing value of orange juice. Effect of sodium benzoate and heat. Inc. Eng. Chem. 26: 857. Joslyn, M.A. and Marsh, G.L. 1935. Browning of orange juice Ind. Eng. Chem. 27: 186. Joslyn, M.A. Marsh, G.L., and Morgan, A.F. 1934. The relation of reducing value and extent of browning to the vitamin C content of orange juice exposed to air. J. Biol. Chem. 105: 17. Joslyn, M.A. and Miller, J. 1949a. Effects of sugars on the oxidation of ascorbic acid. 1Kinetics of autooxidation of ascorbic acid. Food Res. 14: 325. Joslyn, M.A. and Miller, J. 1949b. Effects of sugars on the oxidation of ascorbic acid. General and specific effects. Food Res. 14: 340 Kacem, B., Marshall, M.R., Matthews, R.F., and Gregory, J.F. 1986. Simultaneous analysis of ascorbic and dehydroascorbic acids by high-performance liquid chromatography with postcolumn derivatization and u.v absorbance. J. Agric. Food Chem. 34: 271. Kamiya, S. 1959. Mechanism of ascorbic acid degradation. J. Agri. Chem. Soc. Japan. 33: 398. Kanner, J., Fishbein, J., Shalom, P., Harel, S., and Ben-Gera, I. 1982. Storage stability of orange juice concentrate packaged aseptically. J. Food Sci. 47: 429. Karel, M. and Nickerson, J. T.R. 1964. Effect of relative humidity, air and vacuum on browning of dehydrated oranse juice. Food Technol. 18: 104. Keating, R.W. and Haddad, P.R. 1982. Simultaneous determination of ascorbic acid and dehydroascorbic acid by reversed-phase ion-pair high performance liquid chromatography with precolumn derivatization. J. Girom. 245: 249/

PAGE 152

137 Kefford, J.F., Mckenzie, H.A., and Thompson, P. CO. 1959. Effect of oxygen on quality and ascorbic acid retention in canned and frozen orange juice. J. Sci. Food Agric. 10: 51. Khan, M.M.T. and Martel, A.E. 1967. Metal ion and metal chelate catalyzed oxidation of ascorbic acid by molecular oxygen. 1Cupric and ferric catalvzed oxidation. J. Am. Chem. Sec. 89: 4176. Kirchner, J.S. and Miller, J.M. 1957. Volatile water-soluble and oil constituents of Valencia orange juice. J. Agr. Food Chem. 5: 283. Kirk, J., Dennison, D., Kokoczka, P., and Heldman, D. 1977. Degradation of ascorbic acid in a dehvdrated food system. J. Food Sci. 42: 1274. Koppanyi, T., Vivino, A.E., and Veitch, F.P., Jr. 1945. A reaction of ascorbic acid with alpha amino acids. Science, 101: 541 Kurata, T., FujmakL, M. and Skurai, Y. 1973. Red pigment produced by the reaction of dehydro-L-ascorbic acid with alpha amino acid. Agr. Biol. Chem. 37: 1471. Kurata, T. and Sakurai, Y. 1967. Degradation of L-ascorbic and mechanism of nonenzymatic browning. Agr. Biol. Chem 31: 170. Lahikainen, T., Joslyn, M.A. and Chichester, CO. 1958. Mechnism of browning of ascorbic acid-citric acid-glycine systems. J. Agr. Food Chem. 6: 135 Lamden, M.P. and Harris, R.S. 1950. Browning of ascorbic acid in pure solutions. Food Res. 15: 79. Lea, C.H. and Hannan, R.S. 1950. Biochemical and nutritional significance of the reaction between proteins and reducing sugars. Nature. 165: 438 Ledon, A.C. and Lananeta, P. 1950. Spec tropho tome trie comparison of the factors that affect the formation of colored substances during manufacture of susar. Biol. Ofic. Asoc. Technicos Azucar (Cuba). 8: 547. Lee, S.H. and Labuza, T.P. 1975. Destruction of ascorbic acid as a function of water activity. J. Food Sci. 40: 370. Loeffler, H.J. 1941. Processing of orange iuice. Ind. Eng. Chem. 33: 1308. Lund, A.F. and Lawler, F.K. 1966. Cold packs juice aseptically packed. J. Food Eng. 38.

PAGE 153

138 Matthews, A. 1928. Effect of oxygen on the storage of orange juice. M.S. Thesis, Library, Univ. of Calif. McColloch, R.J., Rice, R.G., BandurskL, M.B., and Gentili, B. 1957. The timetemperature tolerance of frozen foods and frozen concentrate orange juice. Food Technol. 11: 444. McDermott, F.A. 1916. The utilization of cull Florida citrus fruit. Ind. Eng. Chem. 8: 136. McWeeny, D.J. and Burton, H.S. 1963. Some possible glucose/glycine browning intermediates and their reactions with sulfites. J. Scie. Food Agric. 14: 291. Meydav, S., Saguy, I., and Kcpe loan, J.I. 1977. Browning determination in citrus products. J. Agric. Food Chem. 25: 602. Mills, B.M., Damron, M.C., and Roe, H.J. 1949. Ascorbic acid, dehydroascorbic acid, and diketogulonic acid in fresh and processed foods. Analytical Chemistry. 21: 707. Moore, E.L., Esselen, W.B., and Fellers, C.R. 1942. Factors responsible for the darkening of packaged orange juice. Fruit Prod. J. 22: 100 Moore, E.L., Weiderhold, E. and Atkins, CD. 1944. Changes occuring in orange and grapefruit juices during commercial processing and subsequent storage of the glass and tin packed products. Fruit Products J. 23: 270 Mottern, H.H. and Von Loesecke, H.W. 1933. Dearation and flash pasteurization of orange and grapefruit juices. Fruit Pro. J. 12: 325. Nagy, S. 1980. Vitamin C contents of citrus fruit and their products: A review. J. Agric. Food Chem. 28: 8. Nagy, S. and Smoot, M.J. 1977. Temperature and storage effects on percent retention and percent U.S. Recommended Dietary Allowance of vitamin C in canned single strength orange juice. J. Agri. Food Chem. 25: 135. Nelson, E.M. and Motter, H.H. 1933. Vitamin C content of frozen orange juice. Ind. Eng. Chem. 25: 216. Ogata, Y. Kosugi, Y., and Morimato, T. 1968. Kinetics of cuoric salt catalyzed autoxidation of L-ascorbic acid in acueous solutions. Tetrahedron 24: 4057. Paper Bottles Are Coming on Strong, 1983. Business Week, January 16, p. 25.

PAGE 154

139 Passy, N. and Mannheim, C.H. 1979. "Tropical Foods: Chemistry and Nutrition." Vol. 1, G.S. Inglett and G. Charalambous (Eds.), p. 141. Academic Press, New York, NY. Pederson, C.S., Beattie, H.G., and Beavens, E.A. 1941. Processing and storage of fruit juices. Int. Food Technol. Proc. 75: 83. Pelletier, 0. and Brassard, R. 1977. Determination of vitamin C (L-ascorbic acid and cehydroascorbic acid in food by manual and automated photometric methods. J. Food Sci. 42: 1471. Penney, J.R. and Zilva, S.S. 1943. The chemical behavior of dehydro1 -ascorbic acid in vitro and in vivo. Biochem. J. 37: 403. Reynolds, T.M. 1963. Chemistry of nonenzvmatic browning. Adv. Food Res. 12: 1. Reynolds, T.M. 1965. Chemistry of nonenzymic browning II. Adv. Food Res. 14: 168. Reynolds, T.M. 1969. Nonenzymatic browning sugaramine interactions. Ch. 12. In "Symposium on Foods: Carbohydrates and Their Roles," H.W. Schultz, R'.F. Cain, and R.W. Wrolstad (Eds.), p. 219. AVI Publishing Co., Westport, CT. Reynolds, T.M. 1970. Flavors from nonenzymic browning reactions. Food Technol. Australia. 1970: 610. Richert, P.H. 1930. Darkening and other grape products problems. Fruit Products J. 10: 36. Robe, K. 1981. Aseptic-pack fruits retain color, flavor, save 30% processing energy. Food Processing. 42: 86. Robertson, G.L. and Samaniego, L.M.C. 1986. Effect of initial dissolved oxygen levels on the degradation of ascorbic acid and the browning of lemon juice during storage. J. rooc Scien. 51: 184. Roe, J.H., Mills, M.B., Oesterling, M.H., and Damron, CM. 1948. The determination of diketo-l-gulonic acid, dehydrol-ascorbic acid, and 1-asccrbic acid in the same tissue extract by the 2,4, -dLnitrophenylbydrazine method. J. Biol. Chem. 174: 201. Rose, R.C. and Nahrwold, D.L. 1981. Quantitative analysis of ascorbic acid and cehydroascorbic acid by high performance liquid chromatography. Anal. Biochem. 114: 140. Roy, R.B., Conetta, A., and Salpeter, J. 1976. Automated fluorometric methods for the determination of total vitamin C in food products. J. of the AGAC. 59: 1244

PAGE 155

140 Rymal, K.S., Wolfcrd, R.W., Ahmed, E.M., and Dennison, R.A. 1968. Changes in volatile flavor constituents of canned single strength orange juice as influenced by storage temperature Food Techno 1. 22: 1592. SacharoWj S. and Griffin, R.C. 1970. The evaluation of food packaging. Ch. 1. In "Food Packaging." p. 42. AVI Publishing Company, Inc. Westport, CT. Saguy, I., Kopelman, I.J., and Mizrahi, S. 1978a. Simulation of ascorbic acid stability during heat processing and concentration of grapefruit juice. J. Food Proc. Eng. 2: 213. Saguy. I., Kopelnan, I.J., and Mizrahi, S. 1978b. Extent of nonenzymatic browning in grapefruit juice during thermal and concentration process: Kinetics and prediction. J. Food Proc. Preservation. 2: 175. Schoriberg, A. and Moubacker, R. 1952. The Strecker degradation of alphaamino acids. Chem. Rev. 50: 261. Scott, H.C. 1974. Commercially sterile cold filling of citrus juices in glass containers. Food Product and Management. 1974: 12. Seaver, J.L. and Kertesz, Z.I. 1946. The browning (Maillard) reaction in heated solutions of uronic acids. J. Am. Chem. Soc. 68: 2178 Seek, S. and Crouzet, J. 1981. Formation of volatile compunds in sugar-phenylalanine and ascorbic acidphenylanine model ysystems during heat treatment. J. Food Sci. 46: 790 Shallenberger, R. and Birch, G. 1975. In "Sugar Chemistry." AVI Publishing Co., Westport, CT. Shaw, P.E., Tatum, J.H., and Berry, R.E. 1977. Ch. 6. In "Developments in Food Carbohydrate," Vol. 1, G.G. Birch and R.S. Schallenberger (Eds.), p. 91. Applied Science Publishers, Ltd., London, Shaw, P., Tatum, H.J., Kew, J.T., Wagner, J.C., and 3erry, E.R. 1970. Taste evaluation of nonenzymatic browning compounds from orange powder and use of inhibitors. J. Agr. Food Chem. 18: 343. Shtamm, E.V. and Shurlator, Y.I. 1974. Catalysis of the oxidation of ascorbic acid by copper II ions. 1 -Kinetics of the oxidation of ascorbic acid in the copper II-ascorbic acid-mclecular oxygen system. Zbur. Fiz. Khin. 45: 1857. Sinclair, W.H. Bartholomew, E.L., and Nedvidek, R.D. 1935. The isolation and distribution of nitrogen in dilute alkali soluble proteins of heaty Valencia and Washington Navel orange fruit. J. Agr. Res. 50: 173.

PAGE 156

141 Singh, R.P., Heldman, R.D., and Kirk, R.J., 1976. Kinetics of quality degradation. Ascorbic acid oxidation in infant formula during storage. J. Food Sci. 41: 304. Smith, A.H. 1925. A protein in the edible portion of the orange juice. J. Biol. Ctaem. 63: 71. Smoot, J.M. and Nagy. S. 1980. Effect of storage temperature and duration on the total vitamin C content of canned single st r eng th grapefruit juice. J. Agri. Food Cham. 28: 417. Spark, A. A. 1969. Role of amino acids in non-enzymatic browning. J. Sci. Food Agric. 20: 308. Speek, A.J., Schrijver, J., and Schreurs, W.H.P. 1984. Fluorometric determination of total vitamin C and total isovitamin C in Foodstuffs and beverages by high performance liquid chromatograDhy wiyh precolumn derivatizarion. J. Agric. Food Chem. 32: 352. Stadtman, E.R. 1948. Nonenzymatic browning in fruit products Adv. Food Res. 1: 325. Stadtman, E.R. Haas, V.A. Mackinney, G., and Temmer, 0. 1946. Storage of dried fruit: Influence of temperature on deterioration of apricots. Ind. Eng. Chem. 38: 541. Stephens, J.W. Shipston, G.T., and Wilson, CP. 1942. Value and uses of concentrated citrus juices, unpublished data, Calif. Orange Growers Exchange, Los Angeles. Szent-Gyorgyi, A.V. 1937. "Studies on Biological Oxidation and Some of Its Catalysts." Leipzig, J. A. Barth. Tannenbaum, S.R. 1976. Vitamins and Minerals. Ch. 7. In Principle of Food Science, Part I. Food Chemistry." O.R. Fennema (Ed.), p. 347. Marcel Dekker, Inc., New York, NY. Tatum,_ J .H. Nagy, S., and Berry, R.E. 1975. Degradation products formed in canned single strength orange iuice during storage. J. Food Sci. 40: 707. Tatum, J.H., Shaw, P.E., and Berry, R.E. 1967. Some compounds formed during nonenzymic browning of orange powder. J. Agr. Food 'Chem. 15: 773. Tatum, J.H., Shaw, P.E., and Berry, R.E. 1969. Degradation products from ascorbic acid. J. Agr. Food Chem. 17: 38. Townsley, P.M., Joslyn, M.A., and Smit, C.J.B. 1953. The amino acids in various tissues of citrus fruits and in orange protein. Food Res. 18: 522

PAGE 157

142 Tressler, D.K., Joslyn, M.A., and Marsh, G.L. 1939. "Fruit and Vegetable Juices." AVT Publishing Co., Westport, CT. Vojnovich, C. and Pfeifer, V.F. 1970. Stability of ascorbic acid blends with wheat flour, CSM and infant cereals. Cereal Sci. Today. 19: 317. Vanderslice, J.T. and Higgs, D.J. 1934. High performance liquid chromatography analysis with fluorometric detection of vitamin C in food samples. J. Chrom. Sci. 22: 435. VonLoesecke, H.W., Mottern, H.H., and Pulley, G.W. 1934. Preservation of orange juice by deaeration and flash pasteurization. Ind. Eng. Chem. 25: 771. Waletzko, P. and Labjza, T.P. 1976. Accelerated shelf-life testing of intermediate moisture food in air and in oxygenfree atmosphere. Food Sci. 41: 1333 Watt, B. and Merrill, A.L. 1963. Agriculture Handbook No. 8. U.S. Dept. of Agriculture, Washington, DC. Weissberger, A. and LuValle, J.E. 1944. Oxidation processes 17. The autoxidation of ascorbic acid in the presence of copper. J. Amer. Chem. Soc. 66: 700. Wills, R.H., Wimalasiri, P., and Greenfield, H. 1983. liquid chromatography, microfluorometry, and dye titration determination of vitamin C in fresh fruit and vegetables J. Assoc. Off. Anal. Chem. 66: 1377. Wilson, CP. 1928. Relation of chemistry to the citrus products industry. Ind. Eng. Chem. 20: 1302. Wimalasiri, P. and Wills, R.B.H. 1983. Simultaneous analysis of ascorbic acid and dehydroascorbic acid in fruit and vegetables by high performance liquid chromatography. J. Chroma. 255: 368. Winston, W.J. 1961. Environmental and cultural factors influencing the chemical composition and physical characters. Ch. 2. In "The Orange: Its Biochemistry and Physiology," W.B. Sinclair (Ed.), p. 25. Univ. of California Press, Riverside. Wclfrom, M.L., Kashimura, N. and Horton, D. 1974. Factors affecting the Maillards browning reaction between sugars and amino acids. Studies on the nonenzymatic browning of dehydrated orange iuice J. Agric. Food Chem. 22: 796. Zlegler, L.W. and Wolfe, H.S. 1961. "Citrus Growing in Florida." University of Florida Press, Gainesville, Florida.

PAGE 158

BIOGRAPHICAL SKETCH Bechir Kacem was born on December 24, 1943, in Ksar Hellal, Tunisia. After graduation from the Lycee Technique de Sousse, he attended the Faculte des Science at the University of Tunis, receiving a Bachelor of Science degree in chemistry in September 1973. He began his master's program in human nutrition in September 1974 at Harvard University. After graduation in June 1976, he went back to Tunisia and worked at the National Institute of Nutrition and Food Technology as an Assistant Professor for six years. He was admitted to the University Of Florida, Gainesville, in August 1982 to pursue graduate studies in the Food Science and Human Nutrition Department. He expects to receive the degree of Doctor of Philosophy in August, 1986. 143

PAGE 159

I certify that I have read this study and that in cy opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. R. r. Matthews, Chairman Professor of Food Science and Human Nutrition I certify that I have read this study and that in ray opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. ft M/iR. Marshall, Cochaii Associate Professor of Food Science and Human Nutrition I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. p. F. Gregory, /J ) 'Associate Professor o'f^Tood Science and Human Nutrition I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Associate Professor of Food Science and Human Nutrition

PAGE 160

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Associate Professor of Food Science and Human Nutrition I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Ak>Cor^\l J. A. Cornell Professor of Statistics This dissertation was submitted to the Graduate Faculty of the College of Agriculture and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 1986 Dean, Graduate School


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ET3K4ISZK_BDK4QO INGEST_TIME 2015-01-16T18:28:03Z PACKAGE AA00026611_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES