Chemical Method for Determination of Carbon Dioxide Content

in Egg Yolk and Egg Albumen1,2

K. M. Keener,*,3 J. D. LaCrosse, and J. K. Babson
*Department of Food Science, North Carolina State University, Raleigh, North Carolina 27695-7624; Department of
Chemical and Biochemical Engineering, University of Iowa, Iowa City, Iowa 52242; and Department of Biological and
Agricultural Engineering, North Carolina State University, Raleigh, North Carolina 27695-7625
liberation of CO2 from an acidified egg sample into a
standardized, dilute sodium hydroxide solution inside a
sealed jar. The egg sample and a small beaker containing
the standardized sodium hydroxide solution are placed in
a glass jar and sealed. Next, a concentrated acid phosphate
solution is injected through a rubber septum in the cap
of the jar onto the egg sample, while avoiding contact
with the sodium hydroxide solution. The sample is then
stored at 37 C for 24 h. During this storage period, the
carbon dioxide is released from the egg sample and is
absorbed into the sodium hydroxide solution. Afterwards, the dilute sodium hydroxide solution is removed
and titrated to the phenolphthalein endpoint using a dilute, standardized hydrochloric acid solution. The
amount of hydrochloric acid solution required for neutralization can be directly related to CO2 content in the
sample.

ABSTRACT The safety, quality, and shelf life of shell

eggs is a function of carbon dioxide content. A commercial
process was recently developed for rapidly cooling shell
eggs by using cryogenic CO2. The benefit of this new
process over existing cooling processes is that the CO2
addition during cryogenic cooling provides additional
safety and quality enhancements. In order for these benefits to be fully developed into a process that can be
adopted by the egg industry, and thus realized by the
consumer, the amount of CO2 absorbed by the egg during
this process needs to be quantified. Because the albumen
pH of rapidly cooled eggs was reduced to pH <6.5, existing titrametric methods were not adequate for determining CO2 content. They did not prevent CO2 loss
during neutralization. A simple and accurate method for
determining CO2 content in acidified egg albumen and
yolk samples was developed. This method involves the

(Key words: carbon dioxide, egg yolk, egg albumen)

2001 Poultry Science 80:983987

industry, and thus realized by the consumer, the amount

of CO2 absorbed by the egg during this process needs to
be quantified.
There are reference methods available to measure CO2
content in a number of foods and food products, such as
beer (Official Methods of Analysis, 1995a), baking powder
(Official Methods of Analysis, 1995b), and water (Official
Methods of Analysis, 1995c), but there is currently no
published method for CO2 measurement in egg albumen
or egg yolk. The recent application of cryogenic CO2 gas
to shell eggs for rapid cooling has necessitated the development of an economical method to monitor CO2 levels
in shell egg components. Because of the viscous nature
of egg components, direct measurement of CO2 using an
ion selective probe does not accurately measure dissolved
CO2 content. Calibrated near- infrared systems and gas
chromatography systems are adaptable for CO2 analysis;
however, these systems are expensive and require peri-

INTRODUCTION
The safety, quality, and shelf life of shell eggs is a
function of carbon dioxide content. A commercial process
was recently developed for rapidly cooling shell eggs
using cryogenic CO2. The benefit of this new process over
existing cooling processes is that CO2 addition during
cryogenic cooling provides additional safety and quality
enhancements. In order for these benefits to be fully developed into a process that can be adopted by the egg

2001 Poultry Science Association, Inc.

Received for publication August 28, 2000.
Accepted for publication February 25, 2001.
1
Journal Paper No. FSR 00-12 of the North Carolina State University
Department of Food Science Journal Series.
2
This paper was presented at the Institute of Food Technologists
Annual Meeting Paper No. 51G-6 at the IFT Annual Meeting in Dallas,
Texas, from June 10 to 14, 2000. The use of trade names in this publication
does not imply endorsement by the North Carolina Agricultural Research Service of the products named or criticism of similar ones not
mentioned.
3
To whom correspondence should be addressed: kevin_
keener@ncsu.edu.

Abbreviation Key: PT = phenolphthalein; KPH = Potassium phthalate hydroxide; mEq = milliequivalents.

983

984

KEENER ET AL.

odic calibration. Historically, titration procedures have

been used for measuring CO2 and carbonate species in
eggs and egg components (Mueller, 1958; Reinke and
Baker, 1966; Heath, 1977), and the proposed method for
CO2 analyses of acidified egg albumen and egg yolk also
requires a titration step. The main improvement of this
method is the ability to prevent CO2 volatilization during
titration. Cotterill et al. (1959) showed that directly titrating an acidified egg albumen sample to a neutral pH
resulted in a CO2 volatilization. With increased use of
carbon dioxide in food products, it is necessary to develop
an inexpensive method of CO2 analysis that can be used
on acidified samples of less than 10 g.
The procedure developed for determining the CO2 content in acidified egg albumen and egg yolk was adapted
from a method developed by Fleming et al. (1974) to
measure CO2 levels in cucumber brines. The process involves driving CO2 out of the albumen and into a standardized NaOH solution by using a high ionic strength
acid phosphate solution (pH <1). The NaOH solution is
then combined with BaCl2 and finally titrated with HCl
to determine CO2 concentration. The objective of this
study was to evaluate the proposed method for accuracy
and precision in measuring CO2 content in egg albumen
and egg yolk.

MATERIALS AND METHODS

For determining method accuracy, a 2 3 6 randomized block design experiment was setup for determining
CO2 concentration in egg albumen and egg yolk. Eighteen
white eggs, approximately 30 d old, were purchased from
a local grocery store. The eggs were broken out, and
yolks and albumens were pooled. Duplicate samples of
albumen and yolk samples consisting of approximately
0, 2, 4, 6, 8, and 10 g with additions of 0, 2, and 4 mL of
a 1.1 mg/mL CO2 solution were made and evaluated
using this chemical method. The CO2 solution was added
to cover the range of CO2 expected in cryogenically cooled
eggs. The maximum solubilities of CO2 in water at 0 and
20 C under one atmosphere (101.3 kPa) of CO2 are 0.33
and 0.17%, respectively (Brey, 1978).
For determining experiment precision, six 2-d-old eggs
from Hy-line chickens were collected and broken out.
Yolks and albumens were pooled. Five 6-g samples
were tested.
Samples were prepared by placing a 30-mL glass jar4
containing 12 mL of standardized, dilute NaOH solution
(0.15 N) into a 250-mL glass jar.4 A measured amount
of albumen was then placed into the 250-mL glass jar,
and a screw-cap lid was placed on the jar. The screw-cap
lid contained a 1.25-cm rubber septum4 and a plastisol
liner. Figure 1 shows details of the apparatus. The lid
was heated in a drying oven at 50 C for approximately
3 min, before placing on the jar, to soften the liner and

Inmark Inc., Atlanta, GA 30336-0309.

FIGURE 1. Equipment used to determine CO2 concentration in albumen and yolk samples.

provide a good seal. The CO2 solution, if added, was then

injected through the septum onto the top of the albumen
or yolk sample. Next, 15 mL of the acid phosphate
(NaH2PO4 H2O + H3PO4) solution (see below) was injected onto the albumen. The acid was not allowed to
splash into or otherwise contact the standardized NaOH
solution. When the acid phosphate solution was added
to the albumen, the carbonate (CO32) in the albumen
was converted to CO2. The high ionic strength (pH <2.5)
reduced the solubility of CO2, which was driven into the
12 mL of standardized NaOH in the 30-mL vial. The
NaOH reacted with CO2 as shown in reaction 1:
CO2 + 2 NaOH Na2CO3 + H2O.

[1]

From this reaction, it can be shown that 22 mg of CO2

reacts with each millequivalent (mEq) of NaOH. Background CO2 measurements were performed using exactly
the same procedure except no sample was added. After
24 h at 37 C, the vials were removed from the jars. For
this investigation, storage at 37 C for 24 h was found to
provide 100% CO2 recovery for the samples tested and
was selected for convenience. The earlier study on cucumber brines (Fleming et al., 1974) indicated that various
incubation times and temperature combinations could be
used to produce 100% recovery of CO2 in cucumber
brines, but these were not investigated in this study. After
24 h, the jars were removed from the incubator, and the
30-mL vial containing the standardized NaOH solution
was removed from each jar. The vials were then randomized and analyzed. For this analysis, 3 mL of 1 N BaCl2
and one drop of phenolphthalein (PT) indicator were
added to each vial and reaction 2 occurred:
Na2CO3 + BaCl2 2 NaCl + BaCO3 .

[2]

The excess BaCl2 reacted with the Na2CO3 to form NaCl

and BaCO3 precipitate. This step removed all of the
Na2CO3. This solution was then titrated to the PT end
point with a standardized HCl solution (0.15 N). The
mEq of NaOH remaining was calculated from the amount
of HCl used in the titration. Equation 3 was used to determine the quantity of CO2 in the albumen.

CHEMICAL METHOD FOR DETERMINATION OF CARBON DIOXIDE IN EGGS

mg of CO2
=
g albumen

[mEq

of NaOH in solution mEq HCl used]

mg of CO2

mEq

[3]

1
.
of alabumen

985

age.5 In addition, an analysis with SAS PROC GLM6 evaluated the statistical equivalence of grouped means based
on sample size and added CO2 solution. Interaction effects
and non-normality concerns were also evaluated using
SAS software.6 Statistical significance and mean grouping
were based on P < 0.01.

RESULTS AND DISCUSSION

Reagents
Acid Phosphate Solution. Dissolve 200 g of NaH2PO4
H2O and 30 mL of 85% H3PO4 in 600 mL of deionized
water. Slowly add 63 mL concentrated H2SO4 and dilute
to 1 L.
Indicator. Dissolve 1 g PT in 100 mL ethanol.
NaOH Solution (0.2 N). Dilute 8 g of NaOH to 1 L with
deionized H2O. Weigh a nominal amount of potassium
phthalate hydroxide (KPH FW 204.228), approximately
0.75 g, and dilute to about 10 mL with deionized H2O.
Add one drop of PT indicator to the KPH solution and
titrate with NaOH solution. Equation 4 was used to determine the normality of the NaOH.
Xg
204.228 g/gEq
1000 mEq/gEq

[4]

YmEq
Normality
ZmL

where X is the actual grams of KPH used, Y is the mEq

of KPH, and Z is the milliliters of NaOH used to reach
PT endpoint.
HCl Solution (0.15 N). Dilute 1 N HCl to produce a
0.15 N solution. Add one drop PT indicator to the HCl
solution, and titrate with standardized NaOH solution to
determine normality (Equation 5).
XmEq
ZmEq
YmL
Normality
mL
WmL

[5]

where X is the normality of NaOH solution, Y is the

amount of NaOH solution required to reach PT endpoint,
Z is the resulting mEq of the NaOH solution, and W is
the amount of HCl used in the titration. This procedure
was performed three times on each HCl solution, and
the average normality was determined. This average was
used in further calculations.
CO2 Solution (1 mg CO2/mL Solution). Dilute 0.1909
g NaHCO3 to 100 mL with 0.1 N NaOH (carbonatefree) solution.

Statistical Analyses
Means and standard errors were obtained for each sample pair using Microsoft Excel Statistical Analysis Pack-

5
Microsoft Excel, Version 97, Microsoft Corp., Redmond, WA
98052-6399.
6
SAS software, Version 6.12, SAS Institute Inc., Cary, NC 27513.

Tables 1 and 2 summarize the data for the egg albumen

and egg yolk samples with added CO2 solution, respectively. The CO2 content (mg CO2/g sample) was obtained
by taking the predicted CO2 concentration from the multiple regression equation and subtracting the overall mean
weighted average CO2 concentration (1.10 mg CO2/mL
solution) times the added CO2 solution. This result was
then divided by the sample mass. No interaction effects or
non-normality effects were observed for yolk or albumen
data. These results indicate that CO2 concentration within
the egg albumen samples, egg yolk samples, and CO2
solutions were statistically equal with an overall mean
value of 1.66 mg/g in the albumen, 0.09 mg/g in the
yolk, and 1.10 g/mL in the CO2 solutions, respectively.
These results are shown in Table 3. Pooled data comparisons found no statistical differences in CO2 concentration
for different size samples; however, the standard error
generally decreased as sample size increased.
As sample size increases (more CO2 in sample), the
influence of background CO2 levels on measurements will
decrease. In our laboratory and pilot plant area, background CO2 levels have been observed to range from 0.04
to 0.5% (5,000 ppm). CO2 is typically present in air at
approximately 0.04%. Thus, a 250-mL container (at 0.04%
CO2) should contain approximately 0.2 mg CO2. Our experience has indicated that background CO2 levels can
vary considerably, up to 2 mg CO2, and a background CO2
measurement needs to be done for each set of samples.
Background CO2 content was found to be slightly negative for data in Tables 1 and 2 and was positive for Table
4 data. Besides variations in background CO2 levels, CO2
measurements are subject to operator errors in accurately
determining titration endpoint, determining normality of
solutions, and reading burette volumes. These errors become less significant as sample size increases and, thus,
are the main factors in determining a recommended sample size.
The developed method accurately measured the
amount of CO2 present in the egg albumen and egg yolk
samples for samples of 10 g or less. In addition, the CO2
concentrations determined for the added CO2 solution
was shown to be statistically equivalent between sample
groups. It was also observed that standard error increased
as sample size decreased below 6 g; therefore, the recommended sample size would be 6-g samples. Overall, these
results show that this method will accurately measure
CO2 concentration in egg albumen and yolk samples.
A follow-up study with fresh eggs was performed to
better determine method precision for 6-g samples. Twoday-old eggs were obtained from Hy-Line chickens. The

986

KEENER ET AL.
TABLE 1. Summary of egg albumen CO2 concentration data1

Mass2 (g)

CO2 added
(mL)

Average CO2
content2 (mg)

Average albumen
CO2 concentration
(mg/g albumen)

0.00
1.99
3.99
6.04
8.05
10.04
0.00
1.97
4.03
6.04
7.99
9.9
0.00
2.12
4.01
5.99
8.02
9.99

0.00
0.00
0.00
0.00
0.00
0.00
2.00
2.00
2.00
2.00
2.00
2.00
4.00
4.00
4.00
4.00
4.00
4.00

0.82
2.57
6.14
9.37
13.53
18.40
1.69
5.12
8.57
11.18
15.06
17.89
3.56
7.27
10.27
15.26
16.79
19.56

...
1.57
1.74
1.68
1.78
1.91
...
1.68
1.78
1.62
1.71
1.65
...
1.83
1.67
1.94
1.64
1.60

0.014
0.014
0.035
0.000
0.063

0.07
0.124
0.063
0.035
0.049

0.077
0.000
0.049
0.077
0.127

0.120
0.036
0.280
0.040
0.401
3.28
0.160
0.601
0.120
0.200
0.481
0.320
1.121
0.280
0.361
1.803
0.200
0.441

0.194a
0.076a
0.016a
0.049a
0.314a

0.232a
0.029a
0.016a
0.067a
0.023a

0.068a
0.089a
0.284a
0.008a
0.023a

Letters indicate statistical equivalence at P < 0.01.

Average of two samples.
2
Means standard error.
a
1

TABLE 2. Summary of egg yolk CO2 concentration data1

Mass2 (g)

CO2 added
(mL)

Average CO2
content2 (mg)

Average yolk
CO2 concentration
(mg/g yolk)

0.00
2.05
4.05
5.97
8.00
9.97
0.00
1.99
4.00
6.01
7.99
9.94
0.00
1.99
4.03
6.02
7.97
10.04

0.00
0.00
0.00
0.00
0.00
0.00
2.00
2.00
2.00
2.00
2.00
2.00
4.00
4.00
4.00
4.00
4.00
4.00

0.76
0.19
0.34
0.50
0.67
0.50
1.89
1.83
2.14
2.77
2.51
2.26
4.10
4.61
4.78
4.35
5.37
5.15

...
0.50
0.11
0.22
0.18
0.13
...
0.23
0.19
0.23
0.14
0.089
...
0.52
0.30
0.13
0.22
0.15

0.028
0.014
0.141
0.035
0.056

0.042
0.077
0.120
0.063
0.000

0.007
0.014
0.014
0.021
0.014

0.440
0.120
0.320
0.160
0.160
0.080
0.120
0.841
0.481
0.080
0.120
0.240
0.200
0.280
1.082
0.401
0.561
0.240

0.066a
0.079a
0.028a
0.019a
0.008a

0.427a
0.123a
0.017a
0.016a
0.024a

0.138a
0.269a
0.066a
0.069a
0.024a

Letters indicate statistically equivalence at P < 0.01.

Average of two samples.
2
Means standard error.
a
1

TABLE 3. Average CO2 concentrations determined for egg albumen, egg yolk, and CO2 solution samples
CO2 Concentration1
CO2 Solution (mg/mL)
Albumen + CO2
Yolk + CO2
CO2 Solution
CO2 Solution
Literature2

1.01
1.11
1.23
1.26
...

0.14
0.11a
0.11a
0.08a

Letters indicate statistical equivalence at P < 0.01.

Means standard error.
2
Fresh egg (Mueller, 1958).
a
1

Albumen (mg/g)

Yolk (mg/g)

1.66 0.03
...
...
...
1.90

0.09 0.02
...
...
0.00

CHEMICAL METHOD FOR DETERMINATION OF CARBON DIOXIDE IN EGGS

987

TABLE 4. CO2 measurements of fresh egg samples

Albumen blank
Yolk blank
Albumen
Yolk

CO2 Content1
(mg)

CO2 Concentration1
(mg/g)

Coefficient of
variation
(%)

0.938 0.033
0.158 0.006
8.64 0.0024
0.456 0.0000132

...
...
1.44 0.0004
0.076 0.0000022

3.5
3.8
0.028
0.0029

Mean standard error.

eggs were pooled, and five sample blanks, five 6-g yolk
samples, and five 6-g albumen samples were made. These
eggs were then tested using the prescribed method, and
the results are shown in Table 4. From these results, one
can observe that the CO2 concentration in the fresh egg
albumen and egg yolk is 1.44 mg/g and 0.076 mg/g,
respectively. The CV for the albumen and yolk were 0.03
and 0.003%, respectively. These results suggest that using
five 6-g sample averages will allow measurements to be
within 0.03% precision for albumen and 0.003% precision
for yolk. In practice, this method produced CO2 measurements in albumen and yolk with CV of 5% or less for 15
6-g samples. The large increase in sample size and CV
resulted from CO2 content differences between individual eggs.
In conclusion, these results suggest that the chemical
method presented here can be accurately and precisely
used to measure CO2 concentrations in egg albumen and
egg yolk samples. Emphasis should be placed on determining the background CO2 levels because they can
vary considerably with time. In our laboratory, this
method of CO2 measurement with 15 6-g egg albumen
and egg yolk samples has consistently produced CV of
5% or less. This developed method for determining CO2
content also has possibilities in other food systems because it has been shown to work in high protein (egg
albumen) and high lipid (egg yolk) products.

ACKNOWLEDGMENTS
The authors would like to acknowledge the assistance
of Roger Thompson of the USDA Fermentation Labora-

tory (North Carolina State University, Raleigh, North Carolina) who provided details on the method developed
for determining CO2 content in cucumber brines.

REFERENCES
Brey, W. S., 1978. Physical Chemistry and Its Biological Applications. Academic Press Inc., New York, NY.
Cotterill, O. J., F. A. Gardner, F. E. Cunningham, and E. M.
Funk, 1959. Titration curves and turbidity of whole egg
white. Poultry Sci. 38:836842.
Fleming, H. P., R. L. Thompson, and J. L. Etchells, 1974. Determination of carbon dioxide in cucumber brines. Assoc. Off.
Anal. Chem. J. 57:130133.
Heath, J. L., 1977. Chemical and related changes in egg albumen
during storage. Poultry Sci. 56:822828.
Mueller, W. J., 1958. Shell porosity of chicken eggs 1. CO2 loss
and CO2 content of infertile eggs during storage. Poultry Sci.
37:437444.
Official Methods of Analysis, 1995a. AOAC Official Method
940.17Carbon Dioxide in BeerManometric Method. Section 27.1.30. 16th ed. AOAC International, Arlington, VA.
Official Methods of Analysis, 1995b. AOAC Official Method
923.02Carbon Dioxide (Total) in Baking PowdersGasometric Determination. Section 25.1.02. 16th ed. AOAC International, Arlington, VA.
Official Methods of Analysis, 1995c. AOAC Official Method
920.194Carbonate and Bicarbonate in WaterTitrametric
Method. Section 11.1.17. 16th ed. AOAC International, Arlington, VA.
Reinke, R. C., and R. C. Baker, 1966. Relationship between carbon dioxide permeability and bacterial penetration in chicken
egg shell models. Poultry Sci. 45:13271334.