Journal of Chromatographic Science, Vol.
34, October 1996
Gas Chromatographic Analysis of Oxygen and Argon at
Room Temperature
S.S. Raj R.K. Sumangala, K.B. Lal*, and P.K. Panicker
Downloaded from https://academic.oup.com/chromsci/article/34/10/465/356994 by guest on 15 February 2024
;
Centralized Waste Management Facility, Nuclear Waste Management Group, Bhabha Atomic Research Centre,
Kalpakkam, PIN 603 102, India
• A MS 5A column for the elution of argon and oxygen as a
Abstract
composite peak at 100°C. O is subsequently removed with
2
A simple and accurate method for the analysis of oxygen and argon the use of a reducing agent to obtain the argon peak only (2)
at room temperature with a dual-column gas chromatographic • A column of fire bricks loaded with 8 wt% of blood at 20°C
system is described. The system consists of a 5A molecular sieve (for the separation of oxygen and argon) (2)
(MS 5A) column and a combination column (MS 5A and palladium • A molecular sieve column at temperatures higher than 300°C
alumina catalyst). The MS 5A column is used to separate the air under a helium flow (3)
sample into two main peaks at room temperature: an argon and
• A molecular sieve column heated in an argon purge at a
oxygen combined peak and a nitrogen peak. The combination
humidity of 100 ppm at 450°C-500°C for 30 h followed by slow
column is used to separate the argon and oxygen peaks; the oxygen
cooling for 8 h. A 2.2-m column packed with MS 5A (70-80
is eliminated due to the catalytic reaction with hydrogen (hydrogen
is used as the carrier gas), so only the argon peak is eluted. The
mesh) treated as above is used to separate argon and oxygen
concentrations of oxygen and argon can be calculated by at 23°C. Helium is used as the carrier gas at a flow rate of
comparing the output of both columns. 90 mL/min (4).
• Porous polymer column custom made from high-purity
divinylbenzene (11 m × 1.04-mm i.d.) (for the separation of
N , O , Ar, and CO). Helium is used as the carrier gas at a flow
2 2
Introduction rate of 15 mL/min and a temperature of 22°C (5).
The gas chromatographic (GC) separation of oxygen and Procedure
argon with commercially available columns is difficult at room As all the aforementioned techniques are either tedious or
temperature. In the process of developing a technique for the require extreme conditions, we developed a technique that uses
separation of argon from air that involved adsorption at low a dual-column system (GC 9A with a thermal conductivity
temperature, it was necessary for us to analyze samples of air detector from Shimadzu [Duisburg, Germany]).
for oxygen, argon, and nitrogen content with a GC method to The dual-column system consisted of a molecular sieve 5A
find the efficiency and breakthrough capacity for argon. A (MS 5A) column and a combination column of MS 5A and pal
method for combining O with H had already been developed
2 2
ladium alumina (Pd/alumina) catalyst. Specifications are given
in our laboratory and was suitably adopted for the GC analysis in Table 1(1). The MS 5A column was used to separate the air
(1). CO and H O did not interfere in this method of analysis, as
2 2
they were not detected along with O or Ar. Instead, they were
2
Table I. Specifications for the Palladium Alumina Column
retained in the column under the operating conditions used.
The response of air in the working range of 0.2-1 mL was Table 1. Specifications for the Palladium
Alumina source
Alumina Column
M/S Associated Cement Co.,
linear. This technique can also be used to determine the purity Thane, Bombay
of O -Ar cylinders.
2
Size supplied 3-5-mm spheres
Size made in our lab 50-60 BSS
Palladium loading 2 wt%
Separation methods 2
Surface area 150 m /g
The following columns and conditions are generally used 2
Crushing strength 15 kg/cm
for the separation of air samples:
pH Alkaline
• A molecular sieve 5A column at -80°C (2) Porosity 40%
Range of oxygen in outlet of catalyst bed 6-10 ppm
* Author to whom correspondence should be addressed.
Reproduction (photocopying) of editorial content of this journal is prohibited without publisher's permission. 465
Journal of Chromatographic Science, Vol. 34, October 1996
sample at room temperature into two main peaks (an argon and rier gas. By comparing the area of peaks of both the columns,
oxygen combined peak and a nitrogen peak). Only argon was the concentrations of O and Ar can be calculated. The GC pa
2
eluted from the combination column because O was elimi 2 rameters under which the analysis was carried out are given in
nated by a catalytic reaction with H , which was used as the car-
2 Table II. The arrangement of columns in the gas chromato
graph is shown in Figure 1, and sample chromatograms from
Table II. GC Operating Conditions both columns are shown in Figure 2.
The analysis can be carried out without the use of subam-
Gas chromatograph Shimadzu GC 9A bient temperatures and with a maximum column length of
Column oven temperature 30°C-80°C only 3 m. These columns do not require long hours of pre-
Injection port 30°C-50°C
treatment before analysis.
Detector oven temperature 70°C-100°C
Detector current 100 mA
Carrier gas inlet pressure 5 kg/cm 2
2
Column 1 inlet pressure 3.25 kg/cm
Downloaded from https://academic.oup.com/chromsci/article/34/10/465/356994 by guest on 15 February 2024
Experimental
Column 2 inlet pressure 2 kg/cm 2
Carrier flow 60 cc/min
The combination column in the dual-column system was
prepared in our laboratory. It was a 3-m stain
less steel column (3-mm o.d.) of which 1 m
was packed with palladium alumina catalyst
(50-60 BSS) and the remaining length of 2 m
was packed with MS-5A (60-80 BSS; M/S
Chromatopak, Bombay, India). Palladium
acted as a catalytic site for hydrogen-oxygen
reaction. Room air was taken as a standard
homogeneous mixture of oxygen and argon
for all the experiments that were performed.
Calibration was performed with the use of
pure gases, and the linearity in the working
range of the experiments was observed.
Figure 2 shows a sample chromatogram
from which the areas under various peaks
were computed with the Chromatopak CR
3A. Typical values for the peak areas of the
pure gases that were used for calibration are
given in Table HI.
Analyses were carried out at three dif
Figure 1 . Schematic of a dual-column system.
ferent temperatures (30°C, 50°C, and 80°C)
to determine the effect of temperature on
the analytical results. Sample volumes of
0.2, 0.5, and 1.0 mL were injected to deter
mine the effect of sample volume on analyt
ical results.
To prepare calibration standards with dif
ferent concentrations of Ar-O in air, a sam
2
pling tube of calibrated volume was filled with
air and known amounts of argon, nitrogen,
or helium. Analyses were carried out at three
different temperatures to determine the range
of temperature for which this technique could
be used for various types of applications.
Results and Discussion
An amount of sample in the range of 0.2-
Figure 2. Sample chromatograms of argon, oxygen, and nitrogen. 1 mL did not affect the performance of the
column (Table IV). Table V shows the tem-
466
Journal of Chromatographic Science, Vol. 34, October 1996
The same procedure was used in the calculation of the detec
Table III. Typical Peak Areas of Pure Gases Used for
tion limits for oxygen (0.41 μg).
Calibration
Calibration Area Under the Peak Deviation
Sample (counts) (%)
Conclusion
0.2 mL N 2 174,984 0.05
0.2 mL Ar 178,027 0.05
0.2 mL O 2 163,036 0.06 The method is simple, and the technique has been used for
the analysis of more than 500 samples of air at 60°C. The re
sults were found to be reproducible within ± 2%.
Table IV. Variation in Measured Concentration with
Sample Volume
Downloaded from https://academic.oup.com/chromsci/article/34/10/465/356994 by guest on 15 February 2024
Composition (%) Amount of Sample Acknowledgments
Ar O2 (mL)
0.9536 20.946 0.2 The authors acknowledge Shri M. Cheralathan for assis
0.9537 20.946 0.5 tance in preparation of the manuscript, evaluation of the
0.9537 20.946 1.0 columns, and collection of data. We also acknowledge Shri
P.T. Hariharan for procurement of empty columns and packing
materials.
Table V. Variation in Measured Concentration at
Different Column Oven Temperatures
Oven Temperature Composition (%)
References
Ar O 2
30°C 0.954 20.946 1. A.A. Khan, S.S. Raj, and P. Prabhakaran. Air Cleaning Engi
50°C 0.945 20.99 neering Research Section, Waste Management Division, Bhabha
80°C 0.921 20.99 Atomic Research Centre, Bombay. Removal of Hydrogen and its
isotopes from air-gas stream. Proceedings of CATSYMPO 80, The
Fifth National Catalysis Symposium, December 11-13,1980.
perature at which O , which is present in the sample, combined
2 2. J.A.J. Walker. Chromatographic separation of argon and oxygen
completely with the H , which was the carrier gas. Once this
2
using molecular sieves. Nature 209:197 (1966).
reaction was complete, the concentration of argon in air was 3. B.M. Karlsson. Heat treatment of molecular sieves for direct sep
aration of Ar and O at room temperature by gas chromatography.
close to the actual concentration. Our results suggest that a 2
Anal. Chem. 38: 668-69 (1966).
temperature close to 60°C is suitable for an accurate analysis. 4. G.E. Baulescu. Method for the determination of O and Ar mix
2
At a signal-to-noise ratio of 5:1, the detection limits of tures at room temperatures. Stationary Phases in Gas Chro
oxygen and argon were 0.41 μg and 0.51 μg, respectively. The matography. Pergamon Press, New York, NY, 1975, ρ 321.
noise level peak area corresponded to 50 units. Hence, 5 times 5. G.E. Pollock, D.O. Hara, and O.L. Hollis. Gas chromatographic
separation of N , O , Ar and C O using custom made porous
the noise level will be 250 units, which is equal to 0.284 μL 2 2
polymers from high purity divinylbenzene.J. Chromatogr. Sci. 22:
argon and 0.51 μg argon. This was the limit of detection. If it 343_47 (1984).
is assumed that the same quantity is present in 0.2 mL of a
sample, the concentration can be calculated to be 15 ppm. Manuscript accepted February 26,1996.
467
