Recovered volatiles from citrus juices and their importance in citrus processing

Material Information

Recovered volatiles from citrus juices and their importance in citrus processing
Series Title:
Citrus Station mimeo report
Wolford, R. W
Citrus Experiment Station (Lake Alfred, Fla.)
Florida Citrus Commission
Place of Publication:
Lake Alfred FL
Citrus Experiment Station :
Florida Citrus Commission
Publication Date:
Physical Description:
7 leaves : ; 28 cm.


Subjects / Keywords:
Citrus juices -- Processing -- Florida ( lcsh )
Station Lake ( local )
Flavors ( jstor )
Juices ( jstor )
Orange juice ( jstor )
government publication (state, provincial, terriorial, dependent) ( marcgt )
bibliography ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaf 7).
General Note:
Caption title.
General Note:
Statement of Responsibility:
R.W. Wolford ... et al..

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
76758603 ( OCLC )


This item has the following downloads:

Full Text


Florida Citrus Commission and
Citrus Experiment Station CES 66-4
Lake Alfred, Florida-400-10/12/65-RWW

Recovered Volatiles from Citrus Juices and Their Importance in
Citrus Processing

R. W. Wolford, C. D. Atkins, J. A. Attaway, and M. H. Dougherty
Florida Citrus Commission
Lake Alfred, Florida

Flavor enhancement of concentrated citrus juices is essential to the pro-
duction of a flavorful product. This is accomplished by addition to the
evaporator pump-out of materials containing fruit substances which contribute
to the fresh aroma and flavor of extracted juices. "Flavor enhancement" is
used here to include any form of flavor add-back that would impart the desir-
able orange flavor characteristics to juice products. The choice of several
types of flavor enhancement materials, such as coldpressed oil, cutback juice,
essences, emulsions, and juice oil can be made. This decision should be based
on subjective flavor evaluation and flavor stability along with consideration
of overall product quality. Intensive efforts to develop correlative objective
tests to supplement the subjective flavor evaluations are being continued.

It is the intent of this paper to present only a limited discussion of the
physical and chemical requirements of adding particular flavor materials to con-
centrated orange juices. Of prime concern will be the volatile chemical compon-
ents contributing to the flavor and their importance to the quality of the final

The early production of frozen concentrated orange juices using low process-
ing temperatures provided conditions very suitable for a high degree of flavor
retention. The concentrates from evaporators probably contained less adverse
factors to affect the quality and stability, initially, of the concentrated pro-
duct. Also, lower production permitted better selection of fruit from fewer
varieties of oranges and lower yields of juice were obtained than currently.
With the use of cutback juice and selected coldpressed Valencia orange oil a
quality product exemplifying much of the freshness of the original juice was
nearly always the result.

Since the 1945-46 season when frozen concentrate was marketed as a new pro-
duct, the concentrate industry has utilized increasingly higher percentages of
the crop of oranges each season. With increasing competition from other fruit
juice products it became necessary to produce a stable frozen concentrated orange
juice at a lower cost. As a consequence, the yield of juice from oranges had to
be maintained as high as possible without causing noticeable flavor deterioration
in the product. In addition, higher processing temperatures in existing and newly
designed evaporators having a shorter time cycle came into usage to keep pace with
juice production. With the continued growth of concentrate more varieties of fruit
were utilized. Other processing techniques were introduced for increased stabil-
ity of the product as well as increased yields of concentrate per box of fruit.
Fruit selection, blending, and processing techniques favorable to some of the ob-
jectives were in turn unfavorable to others. The effect on flavor maintenance,
for example, has produced some very significant problems.

If we consider the ideal conditions for present day processing of frozen
concentrated orange juice, what might we expect in the way of residual flavor in

the finished product? The following example may serve to show the approximate
amount of water-soluble volatile flavor material contained in a cutback product
as determined by the COD (chemical oxygen demand) method (1). Also, an indi-
cation of the possible amount of loss in volatile water-soluble substances
during evaporation of orange juice is shown.

The case for consideration is a 420 Brix product containing approximately
6 percent cutback juice on a reconstituted basis. Using 106 gallons for ease of
presentation, where 6 gallons consist of cutback juice and 100 gallons is the
amount of juice carried through the process, we then have a theoretical 106
gallons of reconstituted juice containing 5.66% cutback juice. The juice used
for cutback had a measured essence value of 2100 ppm. Similarly the 100 gallons
of juice to the evaporator would contain 2100 ppm essence. Since very little of
the essence components are retained during evaporation of orange juice at temp-
eratures of 800F. or above, we may normally consider an outright loss of the
essence material in the 100 gallons of juice to the evaporator. Then we end up
with 5.66% of the 2100 ppm essence in the original 106 gallons or a final
volatile flavor value of 119 ppm in the reconstituted product.

With the relatively small amount of flavor material remaining after concen-
tration there is plenty of room for addition of flavor enhancement materials with
good aroma characteristics to some satisfactory sensory perception level. The
flavor enhancement may depend largely upon juice oil or coldpressed oil, but the
full aroma of orange juice also requires that portion contributed by the water-
soluble essence materials in fresh juice.

The COD method was used to indicate the water-soluble volatile content in
both commercial concentrates and those prepared in the Experiment Station pilot
plant. These products were known to contain flavor enhancement materials in-
cluding fresh cutback juice and coldpressed oil, freeze concentrated cutback
plus oil, orange juice emulsions, and recovered juice volatiles (essences) in
combination with cutback juice and coldpressed oil. The volatile content values
for some selected commercial concentrates containing the various flavor enhance-
ment materials are shown in summary form in Table 1. The volatile content of
each is shown as a range from low to high on the several samples analyzed. For
comparison purposes, the range of values for fresh orange juices is included.

It should be pointed out that the values in Tables 1 and 2 are in parts per
million based on a 300 ml. sample or aliquot in 300 ml. The ppm COD readings
were taken directly from the standard curve which appeared in the 1964 Citrus
Processors' Meeting mimeographed reports. The volatile content values reported
previously were calculated on a milliliter basis for the convenience of using low
magnitude numbers in calculations for addition of essences to concentrates. Those
values should be converted to parts per million by multiplying each by a factor
of 300.
Volatile Content
Product Enhancing Material (ppm)
Fresh juice (Hamlin) - 540-1880
(Pineapple) - 780-2112
(Valencia) - 2240-3060

Concentrate Fresh juice + oil 84- 350
Freeze conc. juice + oil 420- 790
Juice emulsion 160- 300
Essence, cutback + oil 320- 680

These results show that in only two cases did the volatile content of a
commercial concentrate get as high as that of a fresh juice and in both cases the
maximum value for the concentrate barely exceeded the minimum value for fresh
juices. The sample calculations made earlier in this paper showed that value to
fall within the range for a fresh juice (cutback) and oil product. The value for
fresh juice used in that calculation was the highest value shown for Pineapple
orange juice.

Reference to the range of values for the volatile content of the juices from
three varieties of oranges used in processing will show that the juice with the
highest volatile content will normally provide a higher value in the final cut-
back concentrate product. Any addition of flavor enhancement material will only
add to that value to provide a higher intensity of aroma and flavor.

In Table 2 are shown some selected values from a large number of results
obtained on fresh juices of Hamlin, Pineapple, and Valencia oranges. These
values show a trend in increasing volatile content with advancing maturity.

Volatile Content (ppm)

Date Hamlin Date Pineapple Date Valencia
10/27/64 540 12/10/64 780 4/ 9/65 2240
12/ 3/64 900 12/14/64 1280 4/14/65 2480
1/ 7/65 1090 1/14/65 1350 4/22/65 2660
1/11/65 1160 1/27/65 1610 5/21/65 3060
1/28/65 1760 2/ 4/65 1790 5/25/65 2430
2/ 1/65 1880 2/19/65 2120 5/26/65 2760

In the case of juices from the Hamlin and Pineapple varieties the increase in
volatile content is continuous with the last value reported being the last juice
analyzed. The maximum value obtained for the Valencia juice occurred prior to
the last samples analyzed. However, there was only a small interval between the
last three Valencia samples analyzed. These values are for juices from fruit
grown in different locations and under different growing conditions. It is
thought that the fertilization, spray, and irrigation program used, plus the
natural variations due to location of the groves, can effect the volatile con-
tent of the juices. This might account for some of the fluctuations observed.

If it may be assumed that the volatile content of a fresh citrus juice is
directly related to the flavor and aroma response produced by that juice, then
none of these concentrates analyzed would have the flavor and aroma intensity of
fresh juice. Some recent tests made on a commercial 420 Brix frozen orange con-
centrate gave a volatile water-soluble content of 450 ppm, indicating it probably
contained some added essence plus the usual amount of cutback juice. This juice
was rated by the taste panel as having a "fair" flavor with 6 as a numerical
value. Recovered Valencia orange essence was added to this concentrate to in-
crease the volatile content in the reconstituted juice to 1800 ppm or equal to
the highest value for a high ratio Hamlin juice (Table 1). This raised the
average flavor grade to the "good" category with a numerical value of 7.

In a comparison test at this Station between this same commercial 420 Brix
frozen orange concentrate and a "standard" 42 Brix frozen concentrate prepared
for use at the World's Fair, the latter received a flavor grade of "good" with
a numerical value of 7. When the volatile content of this "standard" product

was increased from approximately 250 ppm to 1800 ppm, the average grade was
shifted in the "good" category from a numerical grade of 7 at the initial
flavor level without enhancement to an 8 at the flavor enhanced level.

Other tests were conducted to determine a threshold in flavor and odor
perception for both of the above 420 Brix products, as well as the possibility
of establishing some criteria for addition of flavor in relation to the quality
of the other components (pumpout and cutback) in the final product. Recovered
orange essence was added to each of the concentrates to establish four levels
of essence as follows: 900 ppm, 1800 ppm, 5000 ppm, and 10,000 ppm. Flavor
evaluation on the commercial product, reconstituted in all cases to 11.80 Brix,
showed a gradual decrease in preference from the lowest to the highest amounts
of flavor enhancement, with nearly equal preference for the 900 and 1800 ppm
levels. Beyond this point the concentrates with 5,000 and 10,000 ppm flavor
levels received decreasing average grades of 6+ and 6, respectively, with the
5000 ppm level being slightly better than the control and the 10,000 ppm level
showing an average flavor grade less than the control or commercial concentrate.
A numerical grade of 7 was the highest grade obtained. A somewhat different
preference pattern was shown for the "standard" product with added essence at
the four levels. There was no difference in preference for the 900, 1800, and
5,000 ppm volatile water-soluble flavor levels since all received grades of 8 in
the "good" category. The 10,000 ppm level was graded equivalent to that of the
"standard" with a numerical value of 7 and was noted as having some orange aroma
but somewhat perfumey, indicating too much added essence.

No immediate explanation can be made for the differences between these two
concentrates in their ability to accept more or less of the flavor enhancement
material. The available data on some characteristics of the commercial concen-
trate are compared with those for the "standard" concentrate in Table 3.


Commercial "Standard"
o Brix 42 42

Ratio 17 14
Top pulp 6.3 9
(gm./6 oz.)
Recoverable oil 0.050 0.0375
(ml./10g. Conc.)

As an integral part of these applied investigations, a continuing effort is
being made to relate certain physical and chemical differences between concen-
trates with their ability to require different amounts of flavor enhancement for
quality improvement. It can be stated in a general way that it is difficult to
cause significant flavor improvement in a poor concentrate, a fair concentrate
may be improved to a good category, and a good concentrate may be supplied with
the characteristics of aroma and flavor more reminiscent of fresh orange juice.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-RWW

The employment of flavor enhancement in the production of frozen concentrated
orange juice in no way reduces the need for quality pump-out concentrate and
cutback juice and if added in the proper proportion both the aroma and flavor
intensities can be increased to a desirable sensory perception level.

The production of good quality concentrates may well involve the following
three basic measures: (1) special care must be taken in the production of higher
Brix concentrates which should contain some minimum amount of coldpressed oil to
avoid oxidative changes; (2) storage and subsequent thawing should be carried out
so as to assure the minimum amount of deterioration in flavor that could carry-
over into the final product; and (3) carefully prepared quality cutback juice
must be employed with addition of a recovered concentrated flavor enhancement
material to provide a balanced aroma and flavor.

Tests made to date indicate that the amount of flavor material required is
dependent upon the quality of the concentrate withdrawn from the evaporator,
with the better quality requiring less enhancement to achieve the desired in-
tensity of aroma and flavor.

It is well recognized that proper handling is required in using fresh cut-
back juice. Likewise, the recovery of other more concentrated flavor enhance-
ment materials should in no-way be considered as a so-called by-product operation,
although it may be termed as such under present operations in some commercial
plants. Such flavor material has been largely missing in the consumer package
except for that small amount of single-fold essence present in fresh juice used
as cutback. Using the recovery of essences as an example, each step in the
separation of the essence and oil from the juices, in the concentration of these
components, and their final removal from the system must meet with the highest
standards required in handling the freshly extracted juices for manufacture of
frozen concentrated orange juice.

Recovery of both orange and grapefruit essences with equipment used at the
Citrus Experiment Station gave products with the aroma characteristics of fresh
juices. Removal of these essences from freshly extracted juices is a very exact-
ing procedure carried out under the best sanitary conditions in equipment con-
structed of 304 stainless or its equivalent. The five-stage vacuum system re-
moves volatile components from the single-strength juices following the juice
extraction and finishing operations. Temperatures for vaporization of the
volatile components are adjusted to a suitable level for the pressure needed to
obtain the most efficient removal and quality in the recovered essence. Both
separation and concentration of the volatile water- and oil-soluble components
are effected simultaneously in this system to yield the appropriate concentration
for add-back to concentrated or single-strength juice products. Refrigerated
condensers permit only a limited loss of volatiles from the vertical condensing
and concentrating heat exchangers. Removal and concentration would appear to
require less than 20 seconds to complete both the juice and volatile cycles.
Contact of the juice with heat through the 5-stage system would be only a
fraction of the time required for the cycle. However, when the juice is intended
for concentration to some higher Brix of 55 to 72 degrees the cycle is not com-
pleted back to the original fresh juice supply, but goes directly to the evap-
orator at an appropriate temperature for multi-stage of multi-effect units.
Ninety-percent removal of volatiles from the juice is easily obtained in one pass

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-RWW

through the 5-stage unit. Almost complete removal of volatiles from the most
desirable source such as Valencia oranges, may be more economical than removing
a percentage from all fruit processed. However, this feature is somewhat argu-
mentative. It is necessary to obtain a supply of essence under the most favor-
able conditions at the lowest possible cost. "Essence" from such a system would
appear to complement the cut-back addition to concentrated products prior to
packaging and would tend to satisfy the problem of returning some of the original
aroma to the processed products.

The information and discussion presented to this point has dealt entirely
with applied aspects of this flavor research program directed toward the one
goal, quality improvement of frozen concentrated orange juice and other processed
citrus products. Any program of investigation of this sort also requires a
workable knowledge of the types of chemical compounds contributing to the flavor
of the juice. In other words, what is this material and how can it best be
handled? This is necessary information whether it be for controlling their re-
tention in the product during processing or for recovery of the components as a
usable material for subsequent addition to the product.

A number of papers have been presented at previous Citrus Processors'
Meetings and elsewhere dealing largely with the recovery, isolation, and identi-
fication of many of the chemical compounds contributing to the flavor of orange
juice and its processed products. Table 4 provides an updated listing of some
of the compounds in the volatile water-soluble fraction from orange juice which
have been identified in this laboratory. A recent compilation of identified
flavor compounds in orange juice and coldpressed orange oils, taken from pub-
lished work by all research efforts dealing with the flavor of orange, has shown
approximately 100 identified compounds in the total flavor spectrum of orange
juice. In addition to the 52 identified and 11 tentative identifications shown
in Table 4, some 38 compounds in coldpressed oil identified as aldehydes, ketones,
alcohols, esters, terpene hydrocarbons, and an impressive list of sesquiterpenes,
including valencene, have been reported in the literature.

Recent results obtained on selected recovered orange essences, coldpressed
oil, and terpeneless coldpressed oil using an extremely high sensitivity hydrogen
flame ionization chromatograph, Perkin-Elmer Model 226, fitted:with a Golay col-
umn* 300' x 0.01" I.D., coated with liquid phase "Q", have vividly shown the
extreme complexity of these flavor materials which contribute to the aroma and
flavor of fresh orange juice. Well in excess of 200 chemical components have
been shown in these flavor extracts from recovered orange essences.

Using the information obtained with this new gas chromatographic system, it
has been possible to blend recovered orange essences of slightly different chemi-
cal characteristics to obtain the desired aroma and flavor characteristics for
use in flavor enhancement studies. One of these blended essences was used in
the flavor evaluation tests outlined earlier. With the high separation efficiency
using a capillary column an improved quantitative relationship between individual
components in the complex flavor mixture has been evidenced. Work is now under-
way to develop a more precise quantitative ratio between certain flavor compounds

U. S. Patent No. 2,930,478 assigned to the Perkin-Elmer Corporation.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-RWW


Volatile Flavor Components in Recovered Orange Essence
(Confirmed Identifications Unless Noted )

Aldehydes and Ketones

Methyl heptenone
alpha-Ethyl butyraldehyde*


Methyl heptenol*



Terpene Hydrocarbons



Ethyl butyrate
Ethyl caproate
Ethyl caprylate
Linalyl acetate
Terpinyl format
Citronellyl butyrate
Ethyl propionate*


cis-Linalool oxide
trans-Linalool oxide
cis-Limonene oxide
trans-Limonene oxide

Strong tentative identification lacks confirmation.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-RWW

believed to influence the overall composition and aroma characteristics of the
recovered essence. These studies are being conducted gas chromatographically
and by flavor evaluation of the product to which the essence is added.

Systematic analyses of these recovered orange essences are being continued
to increase the knowledge of the chemical composition for use in essence re-
covery studies, in evaluation of flavor retention in frozen concentrated orange
juices, and in studies on off-flavors as they may result from mishandling or
abuse of the frozen concentrated product.

In conclusion an attempt has been made to bring to light consideration of
some means for improving the quality of frozen concentrated orange juices.
Flavor enhancement can be beneficial to quality improvement but the employment
of these natural recovered flavor materials in no way reduces the need for
quality pumpout concentrate and cutback juice. It was indicated that when re-
covered volatiles are added in the proper proportion to quality concentrate
both the aroma and flavor intensities can be increased to a satisfactory sensory
perception level, to provide a consistently good product. Further, that the
amount of flavor enhancement required is directly related to the quality of the
product to which it is being added. The flavor of orange juice is supported by
an extremely complex and delicate chemical system requiring the best conditions
for recovery, handling, and employment to achieve the best results for aroma and
flavor improvement in frozen concentrated orange juices.


The authors wish to thank the following: the staff at Minute Maid Research
and Development Laboratories, Plymouth, Fla. for their generous contribution in
the evaluation of the Perkin-Elmer Model 226 for the analysis of recovered orange
essences; Mr. J. W. Kesterson for supplying the sample of terpeneless orange oil
used in analyses for this presentation; and all those who have participated in
flavor evaluations connected with these studies during the past year.


1. Dougherty, M. H. Method for Measuring the Water-Soluble Constituents
of Citrus Juices and Products. Citrus Station Mimeo Report CES 65-7, October 6,

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 400-10/12/65-RWW