5
Capillary Columns in Series for GC Analysis of Volatile Organic Pollutants in Atmospheric and Alveolar Air Pascal Clair*, Mireille Tua, and Herve Simian Laboratoire de Toxicologie Analytique, Division Facteurs Humains, Centre d ’ h d e s et de RecherchesTechniques Sous-Marines, DCAN, B.P. 77, 83800 Toulon Naval, France Key Words: Gas chromatography Capillary columns in series Volatile organic compounds Ambient air Alveolar air Summary Analysis of volatile organic compounds in air samples requires high resolution capillary gas chromatography.When the sample contains both polar and non-polar compounds, use of only one type of stationary phase can be unsuitable if it leads to the preferential separation of one kind of component having the same polarity at the expense of the separation of other classes of component. This paper describes the coupling of fused silica capillary columns of different polarity and length in order to achieve the separation of such complex mixtures. The combination is evaluated with a 42 component standard mixture and then applied to various atmospheric air samples and alveolar air of exposed subjects to demonstrate the capabilities of the com- plete sampling and separation technique. step [3] in which the pollutants are collected on 60-80 mesh Tenax-TA, a polymeric adsorbent, contained in a glass tube. During sampling 100 cm3 of air is pumped through the adsorbent trap (conditioned and supplied by Chrompack) with a Drager model 21/31 manual pump. After sampling, the glass ends of the tubes may be sealed and the traps preserved for a few weeks, or even months, before analysis. 2. 1.2 Alveolar Air Sampling The device we use to collect exhaled gas was first developed for carboxyhemoglobin determination 141. The system consists of two one liter plastic bags joined with a “T” connector, one directly (bag 1) and the second through a restrictor (bag 2). The third arm of the “T” connection is attached to a mouthpiece. After deep inhalation and 20 secondes apnea, the subject expires through the mouthpiece. The restrictor provides a resistance which enables the two bags to be filled successively: gas in bag 1 will be a mixture of alveolar air and gas coming from physiological dead space and bag 2 will contain exhaled gas in equilibrium with blood gases, and thus suitable for analysis. From bag 2 a gas volume is collected on a Tenax-TA trap in the same way as was used for atmospheric sampling. 1 Introduction Qualitative and quantitative analyses of volatile Organic com- pounds (VOCs) are commonly carried out by gas phase chroma- tography on capillary columns, sometimes coupled with mass spectrometry [ 11. With very complicated mixtures containing many constituents of a different nature present in different amounts, separation remains a critical step and often requires several GC analyses on different stationary phases. Because the technique is easy to perform, atmospheric and alveolar air can be sampled “in situ” simultaneously. We report here an optimized separation of a complex standard mixture performed with two capillary columns, having different lengths and stationary phase polarities, coupled in series. The combination has also been applied to monitoring VOCs in contaminated atmospheres and exposed humans’ alveolar air [Z]. 2 Experimental 2.1 Sampling Technique 2.1.1 Atmospheric Sampling Because most of volatile organic pollutants in air samples are present at very low concentrations we use a preconcentration 2.2 Analytical Apparatus 2.2.1 Injector Thermal desorption is the most suitable way of recovering VOCs from a solid adsorbent. The “Thermodesorption and Cold Trap- ping injector” (TCT, Chrompack) used in our laboratory enables the whole sample to be transferred from a Tenax-TA trap to the analytical column. Two parts may be distinguished: the first is an oven where the trap is heated and the second is a fused silica capillary tube, cooled by liquid nitrogen (-130 “C), for cryofocus- ing the desorbed compounds. The cryotrap is flash-heated to 270 “C to transfer the components into the GC column. The TCT injector operating conditions are listed in Table 1. 0 1991 Dr. Alfred Huethig Publishers Journal of High Resolution Chromatography 383

Capillary columns in series for GC analysis of volatile organic pollutants in atmospheric and alveolar air

Embed Size (px)

Citation preview

Page 1: Capillary columns in series for GC analysis of volatile organic pollutants in atmospheric and alveolar air

Capillary Columns in Series for GC Analysis of Volatile Organic Pollutants in Atmospheric and Alveolar Air Pascal Clair*, Mireille Tua, and Herve Simian Laboratoire de Toxicologie Analytique, Division Facteurs Humains, Centre d ’ h d e s et de Recherches Techniques Sous-Marines, DCAN, B.P. 77, 83800 Toulon Naval, France

Key Words: Gas chromatography Capillary columns in series Volatile organic compounds Ambient air Alveolar air

Summary Analysis of volatile organic compounds in air samples requires high resolution capillary gas chromatography. When the sample contains both polar and non-polar compounds, use of only one type of stationary phase can be unsuitable if it leads to the preferential separation of one kind of component having the same polarity at the expense of the separation of other classes of component.

This paper describes the coupling of fused silica capillary columns of different polarity and length in order to achieve the separation of such complex mixtures. The combination is evaluated with a 42 component standard mixture and then applied to various atmospheric air samples and alveolar air of exposed subjects to demonstrate the capabilities of the com- plete sampling and separation technique.

step [3] in which the pollutants are collected on 60-80 mesh Tenax-TA, a polymeric adsorbent, contained in a glass tube. During sampling 100 cm3 of air is pumped through the adsorbent trap (conditioned and supplied by Chrompack) with a Drager model 21/31 manual pump. After sampling, the glass ends of the tubes may be sealed and the traps preserved for a few weeks, or even months, before analysis.

2. 1.2 Alveolar Air Sampling

The device we use to collect exhaled gas was first developed for carboxyhemoglobin determination 141. The system consists of two one liter plastic bags joined with a “T” connector, one directly (bag 1) and the second through a restrictor (bag 2). The third arm of the “T” connection is attached to a mouthpiece.

After deep inhalation and 20 secondes apnea, the subject expires through the mouthpiece. The restrictor provides a resistance which enables the two bags to be filled successively: gas in bag 1 will be a mixture of alveolar air and gas coming from physiological dead space and bag 2 will contain exhaled gas in equilibrium with blood gases, and thus suitable for analysis. From bag 2 a gas volume is collected on a Tenax-TA trap in the same way as was used for atmospheric sampling.

1 Introduction Qualitative and quantitative analyses of volatile Organic com- pounds (VOCs) are commonly carried out by gas phase chroma- tography on capillary columns, sometimes coupled with mass spectrometry [ 11. With very complicated mixtures containing many constituents of a different nature present in different amounts, separation remains a critical step and often requires several GC analyses on different stationary phases.

Because the technique is easy to perform, atmospheric and alveolar air can be sampled “in situ” simultaneously.

We report here an optimized separation of a complex standard mixture performed with two capillary columns, having different lengths and stationary phase polarities, coupled in series. The combination has also been applied to monitoring VOCs in contaminated atmospheres and exposed humans’ alveolar air [Z] .

2 Experimental

2.1 Sampling Technique

2.1.1 Atmospheric Sampling

Because most of volatile organic pollutants in air samples are present at very low concentrations we use a preconcentration

2.2 Analytical Apparatus

2.2.1 Injector

Thermal desorption is the most suitable way of recovering VOCs from a solid adsorbent. The “Thermodesorption and Cold Trap- ping injector” (TCT, Chrompack) used in our laboratory enables the whole sample to be transferred from a Tenax-TA trap to the analytical column. Two parts may be distinguished: the first is an oven where the trap is heated and the second is a fused silica capillary tube, cooled by liquid nitrogen (-130 “C), for cryofocus- ing the desorbed compounds. The cryotrap is flash-heated to 270 “C to transfer the components into the GC column.

The TCT injector operating conditions are listed in Table 1.

0 1991 Dr. Alfred Huethig Publishers Journal of High Resolution Chromatography 383

Page 2: Capillary columns in series for GC analysis of volatile organic pollutants in atmospheric and alveolar air

Capillary GC Analysis of Atmospheric and Alveolar Air

The capillary cryotrap is a critical component of the TCT injector. Despite its short length, the quality of the subsequent separation can be profoundly affected by the characteristics of the cryotrap. Previous studies have led us to use an uncoated and deactivated capillary tube with 0.52 mm inner diameter [5].

2.2.2 Analytical Columns

The fused silica capillary columns used in this work had an inner diameter of 0.32 mm and 1.2 pm films of immobilized stationary phase (Chrompack).

A 14 m polar column (CP-Wax-57-CB) was coupled with a 50 m non-polar column (CP-Sil-8-CB) by means of a zero dead-volume, single ferrule connector. The advantage of this type of fitting is that the risk of breakage or leakage in GC-MS coupling is small.

2.2.3 GC and GC-MS

The TCT injector was mounted on a Chrompack-Packard 437A gas chromatograph equipped with a flame ionization detector, the output of which was connected to a Minichrom chromato- graphic data acquisition system (VG Instruments).

The conditions used for GC are presented in Table 2; the total GC run time was 40 min.

Qualitative analysis and optimization of GC conditions were confirmed by use of complementary GC-FID and GC-MS coupling (Hewlett-Packard GC 5890A-MSD 5970B coupling). The similarity of results obtained for the analysis of volatile organic atmospheric pollutants using each of these techniques has been reported in a previous paper I l l .

2.3 Standard Mixture

2.3.1 Composition

Because of the diversity of VOCs which can be detected in industrial atmospheres and, therefore, in human exhaled air, constituents have been selected which represent a wide range of compounds, including alcohols, aldehydes, ketones, esters, ethers, and aliphatic, aromatic, and halogenated hydrocarbons.

The composition of the standard mixture is shown inTable 3. The compounds are arranged in order of decreasing toxicological significance; the numbering is as used on the chromatograms.

2.3.2 Preparation

A gaseous mixture of the standard compounds was prepared by injecting 10 microliters of each component through a septum into a 40 liter vessel filled with pure air at room temperature. The

Table 1

TCT injector conditions.

Cryofocusing : Precool time: Desorption time: Desorption flow: Desorption temperature: Injection temperature: Injection time:

-130 "C 2 min 20 min 14 ml/min 200 "C 250 "C 2 min

Table 2

GC conditions.

Carrier gas: Helium, 150 kPa (2.6 ml/min at 45 "C)

Oven temperature: 45 "C (6 min) to 200 "C (9 min) at 6"/min

Flame ionization detector Temperature: 250 "C Hydrogen flow: 30 ml/min Air flow: 250 ml/min Make-up flow: 31 ml/min

mixture was stirred continuously to prevent condensation of the components. Traps containing known amounts of the standards were then prepared by injection of the gaseous mixture through Tenax-TA traps by means of a gas syringe.

3 Results and Discussion

3.1 Performance of Coupled Columns

Figures 1 A and 1B depict incomplete separations of the standard mixture on a 50 m polyethylene glycol column (CP- WAX-57-CB) and a 50m 5 % phenyl methyl polysiloxane column (CP-SIL-8-CB). In spite of its greater length, better separation is achieved in quite a short time with the serially connected columns (Figure 1 C) .

The advantage offered by coupled columns is apparent from a comparison of the chromatograms and this is confirmed by peak resolution (R) calculations. For each type of column, Table 4 lists

Table 3

Standard mixture composition.

Compound No.a) Compound

1 2 3 4 5 6 7 8 9

10 11 12 13 14 15 16 17 18 19 20 21

Acrolein Benzene Tetrachloromethane 1,l-Dichloroethylene Trichloroethylene 1,2,3-Trimethylbenzene 1,2,4-Trimethylbenzene lI3,5-Trimethylbenzene Methanol Chloroform Acetonitrile Hexane 4-Methyl-2-pentanone Ethanal Tetrachloroethylene 1 -Butanol Ethylbenzene m-Xylene p-Xylene o-Xylene 2-Butanone

22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Methyl acetate Toluene Nonane Dichloromethane 1-Propanol Cyclohexane Octane Ethyl acetate Heptane 1.1,l-Trichloroethane Diethyl ether Acetone Ethanol Trichlorotrifluoroethane Trichlorofluoromethane Isoprene 5-Methyl-3-heptanone 2-Ethoxyethyl acetate Decane Undecane Dodecane

a) Compound number as used for chromatographic peak assignment in all figures.

384 VOL. 14, JUNE 1991 Journal of High Resolution Chromatography

Page 3: Capillary columns in series for GC analysis of volatile organic pollutants in atmospheric and alveolar air

Figure 1

Separation of 42 components standard mixture on CP-Wax-57-CB (A), CP-Sil-8-CB (B) and serially connected columns (C).

Table 4

Identities of compound pairs which are unresolved, poorly separated, or just sufficiently separated on polar, apolar and coupled columns.

R = O 0 . 5 < R < 1 1 < R < 1.5

Single polar column (Cp Wax 57)

Trichlorofluoromethane Trichlorotrifluoroethane 1,l-Dichloroethylene C yclohexane Tetrachloromethane 1,1,l-Trichloroethane Ethyl acetate Nonane Methanol Dichloromethane Trichloroethylene Decane

Hexane Trichlorofluoromethane + trichlorotrifluoroethane Trichlorofluoromethane + trichlorotrifluoroethane Diethyl ether

Diethyl ether Isoprene

Ethanal Heptane Chloroforme 4-Methyl-2-pentanone

Acetone Methyl acetate Dodecane o-Xylene o-Xylene 5-Methyl 3-heptanone 1,2,4-Trimethylbenzene 2-Ethoxyethyl acetate

Single apolar column (Cp Sil 8)

Ethanal Methanol

Trichlorofluoromethane Acrolein

Acetonitrile Acetone

2-Butanone Hexane

Tetrachloromethane Benzene C yclohexane o-Xylene Nonane m-Xylene p-Xylene

Diethyl ether Isoprene

Trichlorotrifluoroethane Methyl acetate 1-Butanol Tetrachloromethane + benzene + cyclohexane

Serially connected columns

m-Xylene p-Xylene

Ethanol Dichloromethane

Decane 2-Ethoxyethyl acetate

Diethyl ether Isoprene

Acetone Acrolein 2-Butanone Ethyl acetate

Journal of High Resolution Chromatography VOL. 14, JUNE 1991 385

Page 4: Capillary columns in series for GC analysis of volatile organic pollutants in atmospheric and alveolar air

Capillary GC Analysis of Atmospheric and Alveolar Air

3 33 I

21' ' ' 11' ' ' 16' ' ' h' ' ' )o' ' ' 12. ' ' 1 4 '

Figure 2 A

Chromatogram of an alveolar air sample exhaled by a normal, healthy, non-smoking human subject.

those pairs of compounds which are unresolved (R = O ) , poorly separated (0.5 < R < 1) or just sufficiently separated to enable correct quantification (1 < R < 1.5) Pairs of compounds which are fully separated (resolution higher than 1.5) are not listed.

Although it provides good separation of most of the toxic polar compounds, a single polar column is not satisfactory when large numbers of hydrocarbons are present in sample. A single apolar column is, on the other hand, well suited for the analysis of hydrocarbons but unable to separate polar products.

By combining columns in the order polar-apolar, adequate separation is easily obtamed for both polar and apolar compounds even if they are present over a wide range of concentrations. As with a single column, this system requires only one GC oven and eliminates the need for complex multidimentional gas chromatog- raphy or multiple detection.

3.2 Applications

The particular column characteristics and overall system capabil- ity enable the analysis of samples of, for example, industrial atmospheres and alveolar air of subjects exposed to such atmo- spheres: expired air is expected to provide a useful means of comparing the pollutants present in an organism with those present in the atmosphere 161.

Flgure 2 shows the chromatographic profile of alveolar air expired by a normal, healthy, non-smoking human subject. In addition to many compounds present at low concentrations, the air also contains several major components. These were identified as ethanal, isoprene, acetone, hexane, ethanol, acetonitrile, 1-propanol, and toluene 17, 81.

Figure 3 illustrates chromatograms obtained from: an ambient air sample (Figure 3 A); a sample of alveolar air from an exposed subject, sampled during exposure (Figure 3 B); and a sample of alveolar air from the same person, sampled 3 days after the end of exposure (Figure 3 C).

On the first chromatogram there are no polar products but a petroleum fraction eluting between the C9 and Clz normal hydrocarbons.

The Cg-Ctz hydrocarbons are clearly apparent in the second chromatogram, in addition to the usual constituents of alveolar air. We also found a peak from dichloromethane and large amounts of acetonitrile, both missing from the ambient air chromatogram. The compounds are present because the subject, a chemist, had handled these two solvents before collection of the ambient air sample. Three days after exposure (Figure 3 C ) , the hydrocarbons are present only at trace levels and the peaks from the solvents are much reduced. The residual amount of aceto- nitrile may be a consequence of normal human metabolism.

Such progressive elimination of pollutants is confirmed by analy- sis of alveolar air from a subject steadily exposed to vapors of isopropyl ether (Figure 4 A). The corresponding peak (no. 43) is high during exposure yet the concentration of the compound in alveolar air sampled 16 hours after exposure is clearly reduced (Figure 4 B). This technique can also be used to study long term exposure in more polluted atmospheres. The industrial atmosphere generat- ing the chromatogram shown in Figure 5 A consisted of low molecular weight aliphatic hydrocarbons (from 2-methylpentane to methylcyclohexane), a petroleum fraction eluting between the CB and Clz hydrocarbons, and a few polar pollutants (acetone, ethanol, and 4-methyl-2-pentanone).

Figure 3

Chromatograms of ambient air (A) and alveolar air sampled during exposure (B) and 3 days after the exposure (C).

A

Figure 4

Chromatograms of alveolar air sampled during exposure to isopropyl ether (A) and 16 hours after the end of exposure (B). Peak no 43 = isopropyl ether.

386 VOL. 14, JUNE 1991 Journal of High Resolution Chromatography

Page 5: Capillary columns in series for GC analysis of volatile organic pollutants in atmospheric and alveolar air

Capillary GC Analysis of Atmospheric and Alveolar Air

34 40

24 r Figure 5

Chromatograms of an ambient air sample (A) and alveolar air sampled 24 hours (8) and 48 hours (C) after the end of exposure.

Figure 5 B was obtained from alveolar air sampled 24 hours after the end of exposure. It contains most of the atmospheric pollutants listed above (in addition to the normal alveolar air com- ponents).

In alveolar air sampled 48 hours after exposure (Figure 5 C ) the Cg-Clz hydrocarbons are still present at levels higher than the normal metabolic products, the levels of which have decreased during the same time period.

It is apparent that the rate of elimination of apolar compounds is low, probably because these compounds tend to become dis- solved in the organism's fatty tissues.

4 Conclusion Because of the effective separation of both polar and apolar compounds with a wide range of volatility, coupling of two fused silica capillary columns of different polarity has enabled us to perform analysis of ambient atmospheres and alveolar air using the same procedure.

The complete sampling and analytical technique seems to be an efficient tool for the evaluation of the toxicological effects of long term exposure to industrial atmospheres, both in terms of the absorbed doses and the kinetics of pollutant elimination.

References

111 H. Simian, P. Clair, M. Bourdin, M. Tua. and C. Colas, Analusis 16(1988)

"4 P. Clair, C. Colas, H. Simian, C. Abran, M. Bourdin. and J. Heein: L. J Brasser and W. C. Mulder (Eds) Man and His Ecosystem. Proc. 8th World Clean Air Congress, The Hague, The Netherlands, Sep 11-15, 1989, Volume 1, Elsevier Science Publishers B.V., Amsterdam (1989).

9-10,

131 M. Bourdin, R. Badre, and C. Dumas, Analusis 3 (1975) 34-38

[4] E. Radziszewski, Arch. Mal. Prof. 51 (1990) 245-249.

[5] H. Simian and P. Clair, Second Seminar "Quo Vadis Chromatographia?"

161 A. Bernard and R. Lauwerys, Arch. Mal. Prof. 50 (1989) 101-107

171 B. Krotoszynski, G. Gabriel, andH. O'Neill., J. Chromatog. Sci. 15(1977)

181 J . P. Conkle, B. J. Camp, and B. E. Welch, Arch. Environ. Health 30

Versailles, Dec 6-7, 1988.

239-244.

(1975) 290-295.

Ms received: September 14, 1990 Accepted: April 4, 1991

Journal of High Resolution Chromatography VOL. 14, JUNE 1991 387