Group Title: Citrus Station mimeo report - Florida Experiment Station ; 56-1
Title: Oxodized flavors in frozen citrus concentrates
Full Citation
Permanent Link:
 Material Information
Title: Oxodized flavors in frozen citrus concentrates
Series Title: Citrus Station mimeo report
Physical Description: 7 leaves : ; 28 cm.
Language: English
Creator: Olsen, R. W
Citrus Experiment Station (Lake Alfred, Fla.)
Florida Citrus Commission
Publisher: Florida Citrus Experiment Station :
Florida Citrus Commission
Place of Publication: Lake Alfred FL
Publication Date: 1955
Subject: Citrus juices -- Quality -- Florida   ( lcsh )
Frozen concentrated fruit juice industry -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references (leaf 7).
Statement of Responsibility: R.W. Olsen ... et al..
General Note: Caption title.
General Note: "October 4, 1955."
 Record Information
Bibliographic ID: UF00072370
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 74727919

Full Text

Citrus Station kimeo Report 56-1
October 4, 1955

Oxodized Flavors in Frozen Citrus Concentrates 1
R. W. Olsen, E. L. Moore, F. W. Wenzel and R. L. Huggart

1. Introduction

The quality of concentrated citrus juices depends to a great extent upon
their flavor. The occurrence or development of off-flavors of any kind in these
products results in lower quality and less consumer acceptance. Therefore, the
recognition and prevention of any type of undesirable flavor in citrus concen-
trates is of importance to the industry.

During the 1952-53 season an off-flavor, usually described as either a card-
board or castor oil flavor, occurred from time to time in commercial frozen
citrus concentrates so that it became a problem of some significance to the in-
dustry. During that season and subsequently, concentrators have detected this
type of off-flavor in sizable amounts of both 420Brix frozen citrus concentrates
and 550 to 600Brix concentrates packed by them either in small containers or in
bulk. Since this undesirable flavor has been found in products from many plants
and continues to occur from season to season, it is essential that its cause be
determined so that, if possible, it may be eliminated and future trouble and loss
of product prevented.

Information obtained initially from processors indicated that (a) this off-
flavor is apparently not present when the concentrate is packed, but develops
during storage at OOF. or below; also that the off-flavor sometimes disappeared
after prolonged storage, (b) this undesirable flavor had been found in both orange
and grapefruit concentrates and (c) the off-flavor was often detected in products
having a low peel oil content.

Prior to the occurrence of this problem in the concentrate industry, an off-
flavor, designated as having a castor oil character, was noted by personnel of
the Florida Citrus Commission on January 8, 1946 as occurring in experimental
packs of Lue Gim Gong frozen concentrated orange juice and a frozen concentrate
blend of Parson Brown orange and tangerine juices; these packs were processed at
the U. S. Citrus Products Station, Winter Haven, on October 17 and November 20,
1945, respectively. Also a similar oily off-flavor was detected in two experi-
mental samples of frozen tangerine concentrate processed at the Citrus Experiment
Station, Lake Alfred, on December 6, 1951. A report (13) which briefly reviewed
this problem was distributed at a meeting of citrus processors on May 12, 1953.

Investigations were begun at the Station during the 1952-53 season to deter-
mine the cause and methods of prevention in frozen citrus concentrates of off-
flavors, such as cardboard or castor oil. All studies on this problem have been
based on the hypothesis that the detectable off-flavors are produced when sub-
stances in citrus concentrates are oxidized during frozen storage. Substances

1 L.
American Can Company cooperating through the establishment of a grant-in-aid.--

Florida Citrus Experiment Station o
and Florida Citrus Commission, G- 10
Lake Alfred, Florida.


in citrus concentrates that may be oxidized are lipids, phospholipids, proteins,
as well as various constituents that are in citrus peel oil and seed oil. The
exact nature of the substances oxidized and the mechanism of the oxidation still
remain to be determined and substantiated by experimental data. Oxidizing enzymes,
such as peroxidase, may be involved. Tauber (10) and Sumner and Myrback (6) pre-
sent excellent summaries of information on oxidizing enzymes. Off-flavor develop-
ment in frozen vegetables during storage at 00F, or below is caused by enzymes
that catalyze oxidation reactions as indicated by Joslyn (4). Davis (2) and
Willmott and Wokes (14) have reported the presence of peroxidase in various
parts of citrus fruits. Davis (2) emphasized the high peroxidase content of
citrus seeds, especially those from tangerines. Nebesky et al. (5) have presented
a discussion of the thermal destruction and stability of peroxidase in acid foods.

It is believed that this problem is similar to that of the occurrence of
oxidized flavors which frequently occur in such dairy products as milk and ice
cream. Trout and Sharp (11) reported that many terms were used to described oxi-
dized flavor in milk, including cardboard, oily, tallowy, cappy, metallic and
astringent. Turnbow, Tracy and Raffetto (12) state that oxidized flavors which
occur in ice cream during storage are also described as tallowy, stale, or metal-
lic; also that ice cream containing small amounts of such fruits as strawberry
and pineapple develop oxidized flavor sooner than does plain vanilla ice cream.
Bissett (1) reported off-flavor development in tangerine sherbet a few days after
it was prepared from good quality tangerine puree. Holm (3) described four types
of deterioration of milk fat as the cause of most off-flavor development in
dairy products, namely, (a) rancidity resulting from hydrolysis of glycerides,
(b) tallowiness caused by oxidation of unsaturated fatty acids, (c) "oxidized"
flavors perhaps due to oxidation of phospholipids and (d) fishiness as a result
of both oxidation and hydrolysis. Changes similar to these may be responsible
for the development of oxidized flavors in frozen citrus concentrates. Swift et
al. (7) (8) (9) have studied extensively the constituents of the lipids found
in citrus juices.

This report includes results from experimental studies concerning the pro-
duction and detection of oxidized flavors in frozen citrus concentrates. The ex--
pression, COF2, has been suggested for designating this group of off-flavors and
will be used in this report. Also discussed are some of the factors that may
cause the development of these undesirable flavors. Results are also reported
on the occurrence of COF in commercial frozen orange concentrates that were packed
during the 1953-54 season.

Suggested by Dr. J. S. Blair, American Can Company, to represent either card-
board off-flavor, castor oil flavor or citrus oxidized flavor.

Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake Alfred, Florida
628a 9/22/55-FWW


2. The Production and Detection of Oxidized Off-Flavors in Frozen Concentrated
Citrus Juices

Since this investigation was started, a total of 175 experimental packs of
frozen citrus concentrates have been processed, chiefly during the 1953-54 and
1954-55 seasons. Thirty-four lots of different varieties of citrus fruits were
used, including Hamlin, Pineapple and Valencia oranges, Dancy tangerines, and
Duncan (seedy) and Marsh (seedless) grapefruit.

Results obtained from the examination of 37 packs of frozen citrus concen-
trates, made from 10 different lots of fruit, are presented in Tables 1 and 2.
Citrus oxidized flavor (COF) was not initially present in any of these packs but,
as indicated in Table 1, developed during storage of the products at -80F. The
time necessary for the off-flavors to develop varied from 1 day to 22 weeks. The
intensity of COF, noted by the tasters, was indicated as either none, slight,
moderate or extreme; the terms, moderate or extreme, indicated an intensity that
was considered to be of consumer significance. The observations reported in
Tables 1 and 2 definitely show that the incorporation of air into orange or grape-
fruit concentrates prior to storage at -80F. resulted in the development of oxi-
dized flavors during storage, whereas such off-flavors did not develop in control
packs into which air was not nixed. Either a small or large Waring blendor was
used to whip air into these products. By using this method of air incorporation,
COF was produced during frozen storage in concentrates prepared from 12 lots of
citrus fruits out of a total of 34 lots used; observations made using products
obtained from 10 of these lots of fruit are included in Table 1.

The rate of development of COF in four frozen concentrated Hamlin orange
juices, prepared from the same batch of juice, is shown in Table 2. It is evi-
dent that the effect of air incorporation was very pronounced and also that a
greater detectable intensity of COF occurred more rapidly in the 550Brix concen-
trate than that in the 420Brix product, which contained cut-back juice. It
should also be noted that the intensity of the oxidized flavors increased and
then decreased during the storage period.

Examination of the packs, other than those listed in Tables 1 and 2 showed
that (a) the intensity of the oxidized flavors developed in many packs to only
a very slight degree and therefore were difficult to detect with certainty; also
in many of the packs COF did not develop at all, this being especially true with
the 420Brix products; (b) oxidized flavors developed very infrequently in frozen
concentrated Valencia orange juices; and (c) organoleptic examinations of packs
should be made regularly and continuously, otherwise COF may occur and disappear
in a product between periods of examination. Because of these observations, in
future investigations of this problem, 550Brix concentrates will be chiefly used,
air will be incorporated into the products, packs will be prepared principally
from Hamlin and Pineapple oranges or grapefruit, and all packs will be examined
weekly for one month, then every two weeks for another month, and then monthly
for at least six more months.

Previous observations by commercial plant personnel that these off-flavors
occurred in some commercial packs and sometimes either decreased in intensity or
disappeared during storage at OOF. or below have been confirmed experimentally

Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake Alfred, Florida.

Table 1
Development of Oxidized Flavors in Frozen Concentrated Citrus Juices

Date Fruit i Brix of Time to Maximum intensity I
packed varietyconcentrateProcessing variables
packed I variety concentrate develop j finally attained

lencia orange

it It
Ift It
f1 ft *

tn it


It It I

n I

Ift I
tdy grapefruit

Vacuum pack
Air pack
Air incorporated
Albedo added air pack
air incorporated
Rag added air pack
i" air incorporated

Vacuum pack
Air incorporated
Vacuum pack -
Air incorporated-

sugar added
it it








22 weeks

22 weeks




6/ 8/54












SThe terms, moderate or extreme,
consumer significance.

used to indicate the intensity of the oxidized flavors were considered to be of

Florida Citrus Experiment Station and Florida Citrus Commission,
Lake Alfred, Florida. 622a-8/11/55-FWW

Vacuum pack
Air incorporated
Vacuum pack
Air incorporated

Vacuum pack
(Air incorporated
SVacuum pack
!Air incorporated



6 weeks

14 weeks

- -

Table 1 continued

Date Fruit Brix of Time to Maximum intensity
packed variety ;concentrate Processng variaes develop finally attained
_, devlop i, inaly tAine






3/ 8/55



4/28, 55
5/ 5/55

Air pack
Air incorporated
SAlbedo added air
iRag -
i Air pack
Air incorporated

Pineapple orange
11 n
a n
tn t

Hamlin orange

I nIt

Pineapple orange
it Ii
It It

Pineapple orange
fn i

Valencia orange
n n

Valencia orange
ft It









- heated at
ft t n

175 F.
205 F.

- heated at 1750F.
- It 2050F.

incorporated low pressure extraction
high i

Air incorporated low pressure extraction
i i -high "i

Air incorporated
Stored fruit from 4/28 Air incorporated

3 weeks
3 "
3 "

22 weeks

1 week
1 "
1 "

4 weeks
9 "
3 "

3 weeks

1 week

11 weeks
9 weeks







Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake Alfred, Florida.
622b 8/11/55-FW






Table 2
Occurrence during Storage of Oxidized Flavors in
Frozen Concentrated Hamlin Orange Juices

Intensity of Oxidized Flavors
Time after No air Air
packing incorporated incorporated
420Brix 550Brix 420Brix 550Brix
1 day None None None Slight
2 days a" Slight "
5 "t n n None "
6 i" Slight Moderate
7 t t it n I
8 n I 1 t n
9 i i, Extreme
12 Moderate I
13 n a t 1 ft
1, "i "
+ It It ft It
3 weeks Extreme "
4 In It It II
5 It It It 11 it
6 Moderate Moderate
7 In t I In It
8 II iI It It It
9 t n" Slight
10 t I" "
11 it n It

Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake Alfred, Florida.

in 14 packs of frozen citrus concentrates. In these products COF developed to a
detectable intensity, which then gradually decreased and in 9 packs finally dis-
appeared. Results are also available from the examination of samples of commer-
cial frozen concentrated orange juices, packed during the 1953-54 season, which
indicate that the oxidized flavors either (a) disappear upon prolonged storage
at -80F. or (b) change in character of flavor that is detectable and described
as stale, metallic or bitter. The disappearance of COF has been observed most
frequently in products in which the off-flavors were detectable but of only
slight intensity. Changes in the nature of the off-flavors after prolonged stor-
age have been noted chiefly in orange concentrates in which COF was at one time
extremely intense.

During the 1953-54 season, 221 samples of commercial frozen concentrated
orange juices were collected. When examined by a taste panel, 13 of these samples
(5.9%) from 9 different plants were found to have oxidized flavors to such a de-
gree that these products were graded to be of poor flavor quality. These off-
flavors occurred only in concentrates packed during either early or mid-season
and such undesirable flavors were not found in products made when late season
fruit was available. Thus in the industry as a whole the occurrence of COF in
frozen orange concentrate percentage-wise is not exceedingly great; however, to
the concentrator who has 10,000 cases of hip product with these off-flavors, it
is a major problem. Also, the occurrence of COF of low intensity in frozen citrus
concentrates will lower their flavor quality, even though such products may still
be acceptable to many consumers.

Table 3

Factors that May Affect the Development of Oxidized Flavors in Frozen
Citrus Concentrates.

A. Fruit factors (1) variety, (2) maturity, (3) spray practices,

especially the use of lead arsenate, and (4) use of freeze-

damaged fruit.

B. Fruit constituents (1) albedo, (2) rag, (3) seed oil, (4)

citrus peroxidase, and (5) peel oil.

C. Processing factors (1) holding of fruit before processing,

(2) extraction procedures, (3) finishing procedures, (4)

time for evaporation, (5) heat treatment, (6) blending

procedures and (7) air incorporation during processing.

Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake Alfred, Florida.
632- 9/20/55-FWW


3. Factors Affecting the Development of Undesirable Oxidized Flavors in Frozen
Citrus Concentrates.

Many factors may be involved in the occurrence of oxidized flavors in frozen
citrus concentrates. A large number of experimental packs have been processed in
an attempt to obtain information on many of these factors, which are listed in
Table 3,

As indicated in Table 1, COF has been experimentally produced and detected
in Hamlin, Pineapple and Valencia orange concentrates and in grapefruit concen-
trates. The grapefruit products packed on June 8, 1954 were made from a mixture
of seedless and seedy grapefruit. The latest date of processing a pack in which
oxidized flavors developed during storage in a frozen Hamlin orange concentrate
was Feb. 17, 1955; in a Pineapple orange concentrate packed on March 22, 1955; in
a Valencia orange concentrate packed on May 27, 1954; and in a seedy grapefruit
concentrate prepared on June 8, 1954. An "oily" off-flavor was noted in a frozen
tangerine concentrate processed on Dec. 6, 1951; however, COF has not developed,
as yet, in frozen tangerine concentrates packed during the 1954-55 season. It
has been previously mentioned that COF was not found in commercial frozen orange
concentrate samples, collected during the 1953-54 season, that were packed .after
SMarch 15 thus indicating that oxidized flavors occurred chiefly in products
packed when early or mid-season fruit were available.

Although data have been obtained concerning some of the other factors listed
in Table 3, there is not sufficient evidence to warrant definite conclusions at
this time about the effect of these factors on the development of COF. However,
some brief comments will be made about some of these factors, but in considering
these comments it should be kept in mind that they are based on a small number of
observations and definitely will have to be confirmed by obtaining additional
experimental data. Some of the data now available are included in Table 1.

The rate of development of COF of slight intensity was slightly greater in a
frozen Valencia orange concentrate made from fruit held at room temperature for
one week than that made from the same lot of fruit that was processed immediately
after picking.

The effect of extraction pressure was investigated using a Citro-Mat extract-
or. A slight COF developed more rapidly in a Pineapple orange concentrate when
low extraction pressure was used than when a high pressure was used. A Valencia
orange concentrate made from juice extracted with low pressure developed a moder-
ate off-flavor, whereas no off-flavor developed in a concentrate from the same
lot of fruit extracted with high pressure.

In July, 1952, packs of Valencia orange concentrate wete processed to deter-
mine if pulp was a contributing factor to the occurrence of oxidized flavors.
These concentrates were processed so that packs were obtained that had low or
high pulp content, as well as coarse and fine pulp. Air was not incorporated in
any of these products. COF did not develop in any of these Valencia concentrates
during storage for one year at -80F.

Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake Alfred, Florida.

To determine if heat treatment would prevent the development of COF by the
inactivation of oxidizing enzymes in citrus juices, packs of concentrates were
made from juices that were heated to 1750 and 2050F., using a Walker-Wallace plate
type heat exchanger. When examined, the concentrates prepared from the heat
treated juices could be distinguished from those made from unheated juices be-
cause of flavor differences; however, as shown in Table 1, oxidized flavors de-
veloped in some of the packs made from Hamlin or Pineapple oranges regardless of
whether the juices were heated or unheated prior to evaporation. It should also
be noted that in the products made from heated juices, the off-flavor could best
be described as a heated flavor together with the characteristics of COF.

Experimental results showed no differences in a few packs of frozen citrus
concentrates when closed either with or without vacuum in the headspace. Air was
not incorporated into any of these packs. Also, there was no correlation between
the vacuum or pressure found in 13 packs of commercial frozen orange concentrate
in which oxidized flavors developed during storage.

The intensity of the oxidized flavors and the length of time for development
were sometimes reduced when cut-back juice was added to 550Brix products. The
addition of peel oil definitely decreases the detectable intensity of these off-
flavors, especially if the COF that develops is only of slight or moderate inten-

The addition of albedo or rag to orange concentrates had little effect upon
COF development. In a Pineapple orange concentrate no difference was found in
the time for the development of COF or in the maximum intensity attained when
albedo or rag were added to the concentrate; however, in Valencia concentrates to
which these parts of the fruit were added, COF developed after the same time of
storage, but the off-flavor intensity was less in the products containing albedo
or rag.

An extract of citrus peroxidase was made from dried grapefruit seeds and
then added to a grapefruit concentrate. Oxidized flavors did not occur in the
control pack which contained neither air nor the peroxidase extract. COF of ex-
treme intensity was found after storage at -80F. for 16 weeks in the product in
which only air was incorporated and in the concentrate which contained only the
peroxidase, extreme COF developed after 20 weeks of storage.

To obtain data that would indicate that COF results from an oxidation-re-
duction reaction, attempts were made to stimulate the development of these off-
flavors by the addition of small amounts of copper sulfate to concentrates; also
to prevent the occurrence of oxidized flavors by the addition of antioxidants,
using ascorbic acid, ascorbyl palmitate, propyl gallate, NDGA, ethyl caffeate,
and dl-a-tocopherol. Results were not obtained because of the failure of COF to
develop in any of these packs containing either CuSO4 or the antioxidants; also,
COF did not occur in the control packs.

Since incorporation of air into citrus concentrates resulted in many instan-
ces, in the development of COF, this indicates oxidation as a cause of these
flavor defects. However, the nature of the substances in citrus concentrates
that may be oxidized and the mechanism of the oxidation-reduction reactions in-
volved remain to be determined. Infra-red absorption procedures are being

Florida Citrus Experiment Station and Florida Citrus Commission,
Lake Alfred, Florida.


investigated to determine the types of compounds involved in the chemical changes
responsible for these oxidized flavors.

In summary, the following comments may be made based upon information on
this problem available at this time. Oxidized flavors will occur during frozen
storage most frequently in orange concentrates that are produced chiefly from
early or mid-season oranges; also in grapefruit concentrates. The adjusting of
the peel oil content of a concentrate to the maximum acceptable level will re-
duce the possibility of the detection of COF should such off-flavors develop to
only a slight amount. Incorporation of air into juice or concentrate at any time
during processing and canning should be avoided. Before other means of prevent-
ion may be suggested, additional data and information are needed concerning the
causes of the development of these flavor defects.

Literature Cited
1. Bissett, Owen W. Frozen purees from Florida citrus fruits. Proc. Flak
State Hort. Soc., 62, 163-165 (1949).
2. Davis, W. B. The distribution and preparation of citrus peroxidase.
Amer. J. Bot., 29, 252-254 (1942).
3. Holm, G. E. Proc. 20th Ann. State College of Washington Institute of
Dairying, March 1951. Cited by Blanck, Fred C., Handbook of Food and Agricul-
ture. 1955, Reinhold Publishing Corp., New York.
4. Joslyn, M. A. Enzyme activity index of quality in frozen vegetables.
Food Inds., 18, 1204-1210, 1334 (1946).
5. Nebesky, Edward A., Esselen, William B., Jr., Kaplan, A. M., and Fellers,
Carl R. Thermal destruction and stability of peroxidase in acid foods. Food
Research 1, 114-124 (1950).
6. Sumner, James B. and Myrback, Karl. The Enzymes. Vol. II, Part 1. 1951,
Academic Press, Inc., New York.
7. Swift, Lyle James. The determination of crude lipid in citrus juices.
J.A.O.A.C., 29, 389-395 (1946).
8. Swift, Lyle James, and Veldhuis, M. K. Constitution of the juice lipids
of the Florida Valencia orange (Citrus sinensis L.). Food Research 16, 142-146
9. Swift, Lyle James. Fatty acids of the lipids from freshly canned
Florida Valencia orange juice. Food Research i2, 8-14 (1952).
10. Tauber, Henry. The Chemistry and Technology of Enzymes. 1949, John
Wiley and Sons, Inc., New York.
11. Trout, G. Malcolm and Sharp, Paul F. The reliability of flavor judgments,
with special reference to the oxidized flavor of milk. Cornell University Agr.
Expt. Sta., Memoir 204, (1937).
12. Turnbow, Grover Dean, Tracy, Paul Hubert, and Raffetto, Lloyd Andrew.
The Ice Cream Industry. 2nd. Ed., 1947, John Wiley and Sons, Inc., New York.
13. Wenzel, F. W., Moore, E. L. and Olsen, R. W. The problem of a "castor
oil" or "cardboard" flavor in concentrated citrus juices. Mimeo report. Fla.
Citrus Expt. Sta., Lake Alfred. May 12, 1953.
14. Willmott, S. G. and Wokes, F. Oxidizing enzymes in the peel of citrus
fruits. Biochem. J. 20, 1008-1012 (1926).
Florida Citrus Experiment Station
and Florida Citrus Commission,
Lake "-,H- Florida

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs