Group Title: Citrus Station mimeo report - Florida Citrus Experiment Station ; CES 68-8A
Title: Measurement of color of frozen concentrated orange juice using the Hunter citrus colorimeter
CITATION PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00072449/00001
 Material Information
Title: Measurement of color of frozen concentrated orange juice using the Hunter citrus colorimeter
Series Title: Citrus Station mimeo report
Physical Description: 4 leaves : ; 28 cm.
Language: English
Creator: Huggart, R. L
Barron, R. W
Wenzel, F. 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: 1967
 Subjects
Subject: Frozen concentrated orange juice -- Color -- Florida   ( lcsh )
Genre: government publication (state, provincial, terriorial, dependent)   ( marcgt )
non-fiction   ( marcgt )
 Notes
Statement of Responsibility: R.L. Huggart and R.W. Barron and F.W. Wenzel.
General Note: Caption title.
General Note: "450-10/12/67-RLH."
 Record Information
Bibliographic ID: UF00072449
Volume ID: VID00001
Source Institution: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 76241582

Full Text



Florida Citrus Commission and
Citrus Experiment Station CES 68-8A
Lake Alfred, Florida. 450-10/12/67-RLH


MEASUREMENT OF COLOR OF FROZEN CONCENTRATED ORANGE JUICE USING
THE HUNTER CITRUS COLORIMETER

R. L. Huggart and R. W..Barron
Florida Citrus Commission
and
F. W. Wenzel
Florida Citrus Experiment Station
Lake Alfred, Florida


The use and evaluation of the Hunterlab Citrus Colorimeter, Model D45
(Food Technology 21, 100-105, 1967) in the quality control laboratories of 3
commercial Florida processing plants was undertaken during the 1965-66 citrus
season. A report (CES 67-5) was made at the last Processors' Meeting, Oct. 4,
1966, on procedures used to determine subjective visual USDA color scores and
objective color values, using the citrus colorimeter for samples of reconsti-
tuted frozen concentrated orange juice. A summary of the data obtained during
the 1965-66 season was also presented and discussed.

Since it was considered necessary that data be collected during 2 citrus
seasons prior to a final evaluation of this colorimeter, this investigation was
duplicated during 1966-67. Information in this report should bring the industry
up to date on the subject being discussed so that a decision may be made, as
soon as possible, as to the feasibility of using the Hunterlab Citrus Colorimeter
to objectively measure, both accurately and precisely, the visual color of orange
juice so that USDA color scores can be based upon such measurements.

Procedures

Three Hunterlab Citrus Colorimeters, purchased by the Florida Citrus
Commission in 1966, were allotted to the quality control laboratories of 3 pro-
cessing plants, which were designated by the Advisory Committee of the Florida
Canners Association. These plants are located in 3 major citrus districts of
Florida. The cooperating companies were Libby, McNeill and Libby, Ocala; Florida
Citrus Canners Cooperative, Lake Wales; and TreeSweet Products Company, Fort
Pierce. The colorimeters were used by USDA inspectors and quality control per-
sonnel of these companies.

Samples of frozen concentrated orange juice were taken at hourly intervals
on a 24 hr. basis during operation. The sampling period started January 28 and
continued through June 29, 1966; also, during the second citrus season on
November 2, 1966 through June 26, 1967.

Visual color scores for the reconstituted juices were determined1 using
USDA plastic color comparator tubes OJ2, OJ3, OJ4, and OJ5 which represent
visual color score points ranging from 36 (OJ5) to 40 (OJ2). The juices, in

1 U.S. Standards for grades of frozen concentrated orange juice. USDA,
Agr. Marketing Service, Washington, D. C. 1964.








1-inch-OD screw-cap culture tubes, and the USDA plastic comparator tubes were
viewed together in a Macbeth Examolite, daylight Model EBA-220, with a rated
color temperature of 74000 Kelvin. Instrument scale readings for Citrus
Redness (CR) and Citrus Yellowness (CY) for the orange juices were obtained
using the citrus colorimeter.

Results and Discussion

Color data obtained during the 2 seasons from the examination of over 10,000
samples of reconstituted frozen concentrated orange juices are summarized in
Tables 1-4.

The number of samples, examined at each of the 3 plants, which were given a
USDA visual color score corresponding to each of the color score points, ranging
from 36 through 40, are shown in Tables 1 and 2. The number of samples examined

Table 1. Frequency distribution of USDA color scores for samplesI of
reconstituted commercial frozen concentrated orange juice scored at three
Florida plants during the 1965-66 citrus season starting January 28 and
ending June 29.
Plant USDA OJ score points Total plant % of total
code 36 37 38 39 40 samples samples
B none 17 323 1842 14 2196 49.5
C 1 337 376 390 none 1104 24.9
D 2 282 170 680 1 1135 25.6

Total group
samples 3 636 869 2912 15 4435
% of samples 0.1 14.3 19.6 65.7 0.3 100.0

Samples include both midseason and late season products; 3, 648 samples
packed in 6 oz. or 12 oz. cans and 787 in bulk containers.



Table 2. Frequency distribution of USDA color scores for samples of
reconstituted commercial frozen concentrated orange juice scored at three
Florida plants during the 1966-67 citrus season starting November 2 and
ending June 26.
Plant USDA OJ score points Total plant % of total
code 36 37 38 39 40 samples samples
B 809 221 472 992 none 2494 43.7
C 548 378 208 331 none 1465 25.7
D 256 466 66 960 none 1748 30.6
Total group
samples 1613 1065 746 2282 none 5707
% of samples 28.2 18.7 13.1 40.0 100.0
1 Samples include both midseason and late season products packed in either 6
oz. or 12 oz. cans. Data for 374 samples of concentrates packed in bulk
containers were reported but are not included in this Table.

Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 450-10/12/67-RLH








in each of the 3 plants was in approximately the same ratio for both seasons,
with plants B, C, and D reporting on about 47, 25 and 28% of the total samples,
respectively.

On the basis of the USDA visual color scores, 14.4% of the 1965-66 reconsti-
tuted juices were scored either 36 or 37 points and 85.3% 38 or 39 points, where-
as the corresponding percentages were 46.9 and 53.1% for the 1966-67 samples.
Thus, these data indicate that the color of the frozen concentrated orange juices
packed during the 1965-66 season was better than that packed during 1966-67.

Minimums, maximums, and means (averages) of Citrus Red (CR) and Citrus
Yellow (CY) color values, measured with the citrus colorimeter, for the recon-
stituted orange juices in each group of the USDA color scores are shown in
Tables 3 and 4.

The range of the CR values for all samples packed during both the 1965-66
and 1966-67 seasons was from 26 to 57, corresponding to a difference of 31
instrument scale units; and of the CY values from 76 to 90 or a difference of
14 scale units.
The means of the CR and CY values increased stepwise as the color scores
of the reconstituted orange juices increased from 36 to 39 in both the 1965-66
and 1966-67 seasons. The CR values increased 11.1 (42.1-31.0) and 9.5 (40.4-
30.9) scale units and the CY values 6.7 (87.0-80.3) and 4.5 (85.0-80.5) units
in the 2 seasons, respectively. Thus, the redness of the samples, included in
the above mentioned 4 color score groups, was slightly more in 1965-66 than in
1966-67 and the yellowness increased definitely more in 1965-66. Also, the
redness in all of the samples for both seasons increased more than the yellowness.


Precision and Accuracy of Instrumental Method

For the evaluation of an instrument, such as the Hunterlab Citrus Colori-
meter, the precision and accuracy of the method for using it must be determined.
Precision is the ability to duplicate results with one instrument or with
different individual replicate instruments. Accuracy is the degree to which it
measures that which it purports to measure and refers to the degree to which the
instrument or procedure reflects human evaluation Many factors may affect the
precision and accuracy of an instrument or method. Some of these are variations
within and between instruments, within and between operators, as well as differ-
ences in cans of concentrate from the same batch, and in reconstituted juices
falling within a single USDA color score.

To obtain data for statistical analysis to determine the precision of the
method for obtaining color scores for reconstituted juices, using the citrus
colorimeter, the following procedure was carried out. Five packs of frozen
concentrated orange juice the visual color of which ranged from scores of 36
to 40 points were obtained from a commercial plant. Coded cans representing
each of the 5 color scores, in quintuplet, were submitted in a random manner to
each of 2 USDA inspectors in each of the 3 plants where the citrus colorimeters
were being used. Each of the six inspectors determined independently, visual
color scores and the CR and CY colorimeter values for 5 sets or a total of 25
1 Kramer, Amihud and Bernard, A. Twigg. "Fundamentals of Quality Control
for the Food Industry", pp. 512, Avi Publishing Company, Westport, Conn., 1962.
2 These packs were prepared under the supervision of Mr. Morris W. Ratcliff,
Florida Citrus Canners Cooperative, Lake Wales, Florida.









Table 3. Ranges and means of Citrus Red (CR) color values,measured with the citrus
colorimeter, grouped on the basis of their USDA visual color scores.
Citrus Red (CR) color values


1965-66 seasoni


USDA % of total Range of values
color samples
score (4435) Min. Max.
36 0.1 30 32
37 14.3 28 42
38 19.6 30 48
39 65.7 35 53
40 0.3 42 57
100.0


Mean
31.0
34.1
38.7
42.1
48.8


1966-67 season1
% of total Range of values
samples
(5707) Min. Max.
28.2 26 38
18.7 27 43
13.1 33 43
40.0 33 50

100.0


For all samples


57 40.3


50 35.9


1 See footnotes on Tables 1 and 2.




Table 4. Ranges and means of Citrus Yellow (CY) color values, measured with the citrus
colorimeter, grouped on the basis of their USDA visual color scores.
Citrus Yellow (CY) color values
1965-66 season1 1966-67 season1
USDA % of total Range of values % of total Range of values
color samples samples
score (4435) Min. Max. Mean (5707) Min. Max. Mean
36 0.1 79 82 80.3 28.2 76 85 80.5
37 14.3 79 90 83.1 18.7 76 87 81.7
38 19.6 80 92 86.0 13.1 80 89 84.2
39 65.7 82 90 87.0 40.0 80 89 85.0
40 0.3 88 90 88.9 -
100.0 100.0
For all samples 79 90 86.3 76 89 83.0
1 See footnotes on Tables 1 and 2.


Mean
30.9
33.1
37.2
40.4









randomly selected coded samples. Reports showing the codes and USDA color
scores, together with the CR and CY colorimeter readings for the samples ex-
amined were returned by the inspectors to the senior author for summarization
and analysis. The summarized data were also submitted by Mr. J. W. McAllister,
USDA Consumer and Marketing Service, Winter Haven, to Richard P. Bartlett,
Director, Statistical Division, C & MS, Washington, D. C.

Statistical data, calculated by the authors, relative to the visual color
scores and the colorimeter readings are presented in Tables 5 and 6. The
means of the visual color scores, based on the data reported by the 6 inspectors,
corresponded well with those for the coded samples. The range of CY values
(Table 5) was 1.7 for all of the reconstituted juices except that at the 36
point color level, which was 2.2. This indicated that the colorimeter responded
well over the range of yellowness measured. This was not true for the CR values
(Table 6). For some reason, not yet determined, the inspectors were in closer
agreement on CR values for OJ 37 and 38, with ranges of 2.3 and 2.5 units and
with variance standard deviation squared of 0.50 and 0.43, than on those for
juices having color levels of 36, 39, and 40. The largest variance was 1.41 at
the 40 point level with a range of 4.1 units. Application of the data in Tables
5 and 6, to illustrate the reproducibilityy" of results, indicate that if 2
inspectors at different locations and at different dates, using 2 different
colorimeters, measured the CR and CY values of 2 cans of concentrate from the
same lot, their results should agree within 3.7 CR and 1.4 CY units. These
limits were derived by multiplying the largest standard deviations (Tables 5
and 6) for CR and CY values by 3. This would be at the 99% level of confidence
and cover 99% of the measurements reported for the 150 coded samples. As an
example of "repeatability" or precision of the colorimeter, other calculations
indicated that a careful operator, using the same colorimeter to measure differ
ences between readings on 2 samples from the same batch of concentrate, would
obtain readings within 1.4 CR and 1.0 CY units.

Using the experimental design or procedure mentioned above, estimates were
obtained of the variation associated with the process of determining color
scores with the citrus colorimeter under practical plant conditions. However,
to determine various components or factors contributing to the estimates of
variation discussed, additional measurements will have to be made and such data
statistically analyzed.

The degree of correlation obtained between panel scores, such as USDA color
scores, and the instrumental method, such as the use of the citrus colorimeter,
indicate the accuracy of the proposed instrumental procedure. Simple linear and
multiple correlation coefficients have been calculated using data available on
USDA color scores of reconstituted orange juices and CR and CY colorimeter read-
ings. On the basis of such correlation coefficients, high or excellent corre-
lation have been found between USDA visual color scores and the CR and CY values.
Also, the CR (redness) and the CY yellownesss) instrument values are highly
correlated, indicating that optimum accuracy may be obtained by using both CR
and CY values to predict color scores.

Acknowledgments

This report was made possible by the assistance and cooperation of the
Florida Citrus Canners Cooperative, Lake Wales; Libby, McNeill & Libby, Ocala;
TreeSweet Products Company, Fort Pierce; and the USDA Consumer and Marketing
Service, Fruit and Vegetable Division Processed Food Inspection, Winter Haven,
Florida. The authors appreciated and express their thanks to all personnel of
these companies and the USDA who obtained and reported all of the data on which
this report is based.











Table 5. Some statistical data relative to USDA visual color scores
and Citrus Yellow (CY) color values measured with the citrus colorimeter.

OJ designation of the coded samples
36 37 38 39 40
Mean visual scores 34.9 36.5 37.6 39.2 40.0

Mean CR values 73.7 77.9 80.5 84.0 87.7

Range of CR values 2.2, 1.7 1.7 1.7 1.7

Standard deviation-s 0.470 0.421 0.363 0.455 0.459

Variance s2 0.22 0.18 0.13 0.21 0.21


Table 6. Some statistical data relative to USDA visual color scores
and Citrus Red (CR) color values measured with the citrus colorimeter.

OJ designation of the coded samples
36 -7. 38 40
Mean visual scores 34. 36.5 3.6 3 4U.U

Mean CR values 25.6 32.8 37.3 44.3 54.9

Range of CR values 3.4 2.3 2.5 4.1 4.1

Standard deviation-s 0.768 0.707 0.656 1.004 1.236

Variance s2 0.59 0.50 0.43 1.00 1.51


Florida Citrus Commission and
Florida Citrus Experiment Station,
Lake Alfred, Florida. 450-10/12/67-RLH




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