Note 24: Selection of GC Guard Columns For Use With the GC Cryo-Trap

INTRODUCTION

The GC Cryo-Trap consists of a small heating/cooling chamber which surrounds the first 4 inches of the GC capillary or guard column. The GC Cryo-Trap is installed inside the GC oven just under the GC injection port to permit the trapping of volatiles and semi-volatiles at the front of the capillary column (Figure #I). Liquid C02 is normally used as the cooling gas to permit the trapping of volatiles at temperatures down to -70 deg. C. Liquid nitrogen can be used to permit trapping volatiles down to - 180 deg. C. A separate digital dual temperature range controller permits the accurate temperature setting and the regulation of both the cooling and the heating temperatures of the GC Cryo-Trap. After trapping the volatiles, the GC Cryo-Trap can be rapidly heated to a selected temperature up to 400 deg. C at a heating ramp rate in excess of 800 deg./min. Applications of the GC Cryo-Trap include thermal desorption and headspace GC techniques for the cryo-focusing and subsequent analysis of the volatile and semi-volatile organics.

Various guard columns can be utilized with the GC Cryo-Trap depending on the applications of the user. This paper studies the improvement in the retention range of volatiles that can be trapped as a function of the type of guard column used inside the GC Cryo-Trap. A variety of guard columns from deactivated fused silica with no liquid phase up through thick film megabore guard columns were studied using a mixture of hydrocarbons from ethane (C2) through nonane (C9). The purpose of this study was to determine the optimum guard column to use with the GC Cryo-Trap to permit the analysis of the widest range of volatile organics.

Fig. #1 -Theory of Operation of the GC Cryo-Trap

Experimental

All samples were analyzed on an H.P. 5890 Series II GC with electronic pressure control at the injection port and were detected on an H.P. Engine (Model 5989A) mass spectrometer in the El mode. A wide variety of megabore guard columns were tested as listed in Table I.

Table I - Guard Columns Tested With the GC Cryo-Trap

Column #    I.D.    Liquid Phase     Film Thickness

--------  --------  ------------     ---------------

  A.      0.53 mm        None            0.00 ul

  B.      O.53 mm       DB5-MS           0.25 ul

  C.      0.53 mm       DB5-MS           1.50 ul

  D.      0.45 mm       DB5-VRX          2.50 ul

  E.      0.53 mm       DB5 5            5.00 ul

  F.      0.53 mm       GS-0 PLOT         ---

Megabore columns were selected as guard columns due to their larger inside diameter and larger surface area which will minimize ice plugs and due to their larger dynamic loading capabilities. Each of the guard columns was attached to the GC injection port utilizing a graphitized vespel® ferrule, passed through the GC Cryo-Trap and then at its exit from the trap was mated to a 0.32 mm by 0.25 um film thickness by 60 meter DB5-MS capillary column utilizing an SGE capillary union with graphitized Vespel® ferrules. A 0.5 ul syringe was loaded with 0.1 ul of a neat hydrocarbon mix of the hydrocarbons ethane (C2) through nonane (C9). The syringe was emptied until no sample volume remained. The syringe was then injected into the GC injection port utilizing the residual volume of sample retained by the syringe needle as the sample for analysis. The injection port was maintained at 250 deg. C. The sample was split in the injection port at about 5:1 and then trapped on the GC Cryo-Trap set to selected temperatures from - 180 deg. C up to 0 deg. C in increments of 20 degrees. After trapping and holding the sample at the indicated temperature for 5.0 minutes, the GC Cryo-Trap was rapidly heated to 200 deg. C to release the hydrocarbons to the front of the GC column for subsequent analysis. The GC column was held at 30 deg. C for 5.0 minutes, temperature ramped to 120 deg. C at 10 deg./min, temperature ramped to 240 deg. C at 20 deg./min and held at 240 deg. C for 1.0 minute. The mass spec was used in the El mode and scanned from 14 to 200 daltons. The purpose of this series of tests was to determine the trapping efficiency of the various guard columns at different temperatures in order to compare and select a guard column which will give the best results for routine sample analysis and permit the trapping and analysis of the widest range of volatile organics for both thermal desorption as well as headspace sample analysis.

A second study was conducted using a 0.5 ul sample of gasoline injected into 5.0 ml of water in a headspace sample vial. The vial was inserted into a CTC Headspace Sampler (LEAP Technologies), heated to 70 deg. C for 5.0 minutes and then 1.0 ml of the headspace volume was injected into the GC injection port and cryo-trapped at the front of a 1.5 micron film thickness megabore (0.53 mm I.D.) guard column at a series of temperatures from -60 deg. C to - 180 deg. C. The purpose of this study was to investigate the range of volatiles that could be trapped at the various temperatures and to compare the effect of the trapping temperature on the resolution of the organics in a mixture of more than 100 compounds.

Table 11 - Temperatures of Efficient Trapping of Hydrocarbons On Guard Columns

Compound  Melting Pt  Efficient Trapping Temperatures of Hydrocarbons on Column 

                       Col. A        Col. B      Col. C      Col. D     Col. E    Col. F

                     Fused Silica  .25u DB5MS  1.5u DB5MS  2.5u DBVRX  5.0u DB5  GSQ-PLOT

--------  --------   ------------  ----------  ----------  ----------  --------  --------

Ethane     -182        ----            -180       -180        -180        -180      -60

Propane    -189        -180            -160       -160        -160        -160      -20

Butane     -138        -140            -140       -120        -120        -100       -0

Pentane    -129        -120            -100        -80         -80         -60       -0

Hexane      -95        -100             -80        -60         -60         -40       

Heptane     -91         -80             -60        -40         -40         -20

Octane      -56         -60             -40        -20         -20           0

Nonane      -51         -40             -40          0           0           0

Results and Discussion

Figure # 2 - Trapping Efficiency of Hydrocarbons On Deactivated Fused Silica At Various Cryo-Cooling Temperatures

The first guard column studied was the deactivated fused silica column with no liquid phase. This guard column would be expected to trap volatile organics based strictly on their melting points. The results of the trapping of the hydrocarbons ethane through nonane are shown in Figure #2 and the trapping temperatures versus melting points are listed in Table II. At the higher temperatures (>-60 deg. C), shifted retention times and broad peaks are due to the hydrocarbons chromatographing through the capillary column at 30 deg C which were not trapped in the GC Cryo-Trap.

Figure # 3 - Trapping Efficiency of Hydrocarbons On .53 mm DB5 (1.5u film thickens)

In likewise manner, the other guard columns were tested and the effective trapping temperatures are listed in Table II. Note that as the column phase thickness increases, the effective trapping temperature increases for the particular hydrocarbon. The results of the 1.5 um film thickness DB5-MS guard column are shown in Figure #3. By adding the liquid phase to the guard column, the trapping efficiency of the guard column increases for the lower boiling hydrocarbons. By adding the liquid phase coating, the trapping temperature is about 40 degrees higher for the hydrocarbons pentane through nonane on the 1.5 um DB5-MS guard column as compared to the deactivated fused silica guard column. In addition, the 1.5 u film thickness guard column was able to trap ethane at -180 degrees C. Therefore, by using thicker film columns, higher cryo-trapping temperatures can be utilized to obtain the same results as using the lower temperatures with deactivated fused silica guard columns.

Figure # 4 - Trapping Efficiency of Hydrocarbons on 0.53 GS-Q Plot Column

The results of the GS-Q PLOT column are show in Figure #4. This guard column effectively traps ethane at -60 degrees C and can prove very useful for the trapping of formaldehyde, ethylene oxide and other very low boilers. However, note that the higher boilers are not effectively released from the guard column at 200 degrees C. The maximum temperature for this column is 250 degrees C, and this higher temperature would improve the results shown. However, we feel that this guard column, while very practical for the trapping of the low boilers, should not be used for hydrocarbons above hexane.

For the selection of which guard column to use for a particular application the nature of the volatiles that need to be trapped and the type of cooling liquid used in the GC Cryo-Trap must be considered. Liquid C02 will cool down to -70 degrees C in the GC Cryo-Trap while liquid nitrogen will cool down to - 180 degrees. In many cases the user may not want to trap the very low boilers, and therefore a higher trapping temperature or a non phase coated guard column can be used.

Figure # 5 - Trapping Efficiency of Hydrocarbons at -60 degrees C

The comparison of the guard columns at -60 degrees C using the GC Cryo-Trap with liquid C02 is shown in Figure #5. Based on this comparison, we recommend the 1.5u film thickness DB5-MS guard column for most applications. This guard column will provide the widest range of volatile trapping and release of the volatiles. Hydrocarbons down to hexane can be effectively trapped on this column at -60 degrees C. This bonded phase column was developed for mass spec applications and will produce very low column bleed. Going to thicker film guard columns does not improve the trapping efficiency greatly and the column bleed will increase as the liquid phase thickness increases.

Figure # 6 - Trapping Efficiency of Hydrocarbons At -180 degrees C Using Liquid Nitrogen

The comparison of the guard columns using liquid nitrogen at a temperature of -180 degrees C is shown in Figure #6. Note that the selection of the guard column is not as critical at this temperature. This is probably due to the loss of activity of the liquid phase permitted the trapping of ethane while ethane was not trapped on the deactivated fused silica column. On the very thick film guard columns, peak broadening was quite apparent and therefore these guard columns should not be used. For use with liquid nitrogen, we recommend either a deactivated fused silica guard column or a thin film guard column for optimum results.

Figure # 7 - Headspace Analysis of Gasoline In Water By Cryo-Trapping At Various Cryo-Trap Temperatures

Figure #7 demonstrates the results of the headspace analysis of a gasoline sample containing more than 100 compounds using the 1.5u DB5-MS guard column in the GC Cryo-Trap as a function of temperature. As expected, at -60 deg. C the lower boiling hydrocarbons C3 through C5 are not trapped. While at -180 deg. C, propane, isobutane, butane, and C02 were effectively trapped. However, note that as the temperature of the GC Cryo-Trap decreased below -120 degrees C, the resolution of the resulting chromatogram decreased. This is probably due to the loss of surface activity of the liquid phases at temperatures below -60 degrees C as noted above. While the trapping temperature of - 180 degrees C provides for the optimum trapping of volatiles, the loss of resolution may not be acceptable. We do not recommend using the GC Cryo-Trap below -140 degrees C when using the 1.5 um film thickness DB5-MS guard column.

The reproducibility of sample analysis utilizing the GC Cryo-Trap in conjunction with the CTC Headspace Sampler (LEAP Technologies) is demonstrated with the data in Table III. A mixture of 9 aromatics at a concentration of 500 ppb in sand were prepared. Five grams of sand plus 5.0 ml of water were added to the headspace vial and then 2.0 ml of the headspace were injected over a 70 second time frame into the system as described above using the 1.5u film thickness DB5-MS guard column at the Cryo-Trap temperature of -120 degrees C. The results produced a relative standard deviation of 4.8% or better for all of the volatiles analyzed. These results were considered quite good, especially since no internal standard was used and this error represents the total error including pipetting, injecting, injection port splitting and other losses. This also shows the advantage of this technique to slowly inject a large volume of gas onto a capillary over an extended period of time to inject the maximum sample onto the GC column. A quick injection on the H.P. split/splitless injector which uses a back pressure regulator, would result in a major splitting of the sample thereby reducing the sensitivity of the analysis. This splitting is minimized by using the slow (1.0 ml/min) injection speed.

Table III- Reproducibility of Static Headspace/GC Cryo-Trap

Volatile                 RT (min)   Peak Area    S.D.     % RSD

----------------        ---------   ---------   -----   -----------

Methyl-tert-butylEther     4.15       367         14       3.7%

Benzene                    5.22       608         26       4.3

Toluene                    6.97       480         35       4.8%

Ethyl Benzene              8.71       740         25       3.4%

m/p xylene                 8.90      1377         46       3.3%

o-xylene                   9.30       634         22       3.4%

1,3,5-Trimethyl Benzene   10.55       754         32       4.2%

1,2,4-Trimethyl Benzene   10.92       687         16       2.3%

Naphthalene               13.24       157          4       2.5% 

Applications of Coated Guard Columns In the GC Cryo-Trap

We have used the GC Cryo-Trap with various guard columns in conjunction with our thermal desorption system for more than 2 years. To date, most of our work has been with liquid C02 at a temperature of -70 degrees C and a 1.5 um film thickness DB5-MS guard column. In the last six months, we have been using liquid nitrogen at a cryo-trap temperature of - 120'C. The technique has been used with both purge and trap (dynamic headspace) as well as the direct thermal extraction techniques with the thermal desorption system for the analysis of a wide variety of samples including flavors in foods, packaging materials, manufactured products and environmental samples. In the last few months, we have been working in cooperation with LEAP Technologies and their CTC Headspace Sampler in order to develop the GC Cryo-Trap as an accessory for their static headspace analysis. The GC Cryo-Trap has been interfaced to this automated headspace sampler to develop a completely automated static headspace sampler for the unattended injection, cryo-trapping and analysis of samples. This has been successfully accomplished in our lab.

Figure # 8 - Comparison of Cryo-Trap Temperatures For Black Tea

Figure #8 demonstrates the volatiles from a sample of black tea utilizing the purge and trap thermal desorption technique. Two hundred milligrams of black tea in 5.0 ml of water at 80 degrees C was purged with 450 ml of helium and trapped on a Tenax® TA desorption trap. The sample was then desorbed at 250 degrees C into the GC injection port, trapped at the indicated temperatures on a 1.5 um DB5-MS guard column and subsequently heated to 250 degrees C to release the volatiles for chromatography. Two samples were analyzed at two different trapping temperatures in the GC Cryo-Trap. At a trapping temperature of 0 degrees C, volatiles below 2-pentenal are not effectively trapped. However, at a trap temperature of -70 degrees C, many lower volatiles including acetone are quantitatively trapped. Depending on the applications, different groups of volatiles may be trapped and analyzed. Through proper selection of desorption temperature and Cryo-Trap temperature, this combination of Thermal Desorption and GC Cryo-Trapping approaches results obtained with multidimensional GC.

Figure # 9 - GC Headspace Analysis of Tobacco Smoke Via the GC Cryo-Trap

In addition to the static headspace gasoline samples in water and soil that were analyzed in the guard column evaluation procedures described above a large number of other headspace applications are currently being evaluated. A headspace sample of 2.0 ml air sample from air containing tobacco smoke (Figure 9) was injected slowly (over 60 seconds) into the GC injection port and cryo-trapped at -70 degrees C on the 1.5 m DB5-MS guard column. The sample was held on the trap for 2.0 minutes and then heated to 250 degrees C to release the volatiles into the GC for analysis. The major peak detected was nicotine; however, many lower volatiles including acetone, benzene and toluene were also trapped and identified.

Conclusion

Coated capillary guard columns used in the GC Cryo-Trap increase the efficiency of trapping of lower boiling volatiles over uncoated guard columns. This effect is most noticeable at trapping temperatures greater than -60 degrees C. While trapping temperatures less than -150 degrees C effectively trap the lowest range of volatiles, some loss in resolution is observed at these lower temperatures. Based on this study, a 1.5 um film thickness DB5-MS megabore guard column was selected as the optimum column for use in the GC Cryo-Trap. It provides greater efficiency in trapping volatiles than the uncoated guard column. In addition, this film thickness and type provides for minimal phase bleed from the guard column. The megabore column was selected due to its larger inside diameter which increases the dynamic loading capabilities and minimizes the chance of ice plugs when samples containing significant amounts of water are analyzed. When utilizing C02 as the cryo-cooling gas, a trap temperature of -70 degrees C is recommended. When using liquid nitrogen as the cryo-cooling gas, a temperature of - 120 degrees C is recommended unless the very low boilers (ethane and propane) must be analyzed. This methodology can readily be utilized for the analysis of thermal desorption as well as headspace GC samples for a wide range of volatile organics. For the trapping of very low boilers such as ethane, formaldehyde and ethylene oxide, the PLOT columns such as the GS-Q PLOT column from J&W is recommended for use with the GC Cryo-Trap. However, this column may present problems if less volatile compounds are analyzed in that they may not be eluted off the guard column. Depending on the range of volatiles that need to be analyzed, a wide range of guard columns and cryo-trapping temperatures can be selected for the optimum results when using the GC Cryo-Trap with either static headspace or thermal desorption applications.