908-788-5550

Note 27: Analysis of Volatile Organics In Soils By Automated Headspace GC

Home
By John J. Manura

INTRODUCTION

A new GC Cryo-Trap was interfaced to a Gas Chromatograph with an automated GC Headspace system in order to develop an automated GC headspace method for the analysis of volatile organic compounds in contaminated soil samples utilizing capillary GC columns. The LEAP Headspace Auto Injector uses a heated syringe concept to directly inject 0.10 to 2.5 milliliter gas samples from the heated and agitated headspace sample vial into the GC injection port for subsequent trapping on a GC Cryo-Trap at the front of the GC column. The GC Cryo-Trap consists of a small heating/cooling chamber which surrounds the front 4 inches of the capillary or guard column. It is installed inside the GC oven, just under the GC injection port to permit the trapping of volatile organics at the front of the column. Liquid Nitrogen is utilized to permit the trapping of volatiles down to -180 degrees C.

The purpose of this study was to evaluate the usefulness of the GC Cryo-Trap in conjunction with the headspace injection technique for the analysis of volatile organics in soil, water and other samples utilizing capillary GC columns. The GC Cryo-Trap permits the injection of large gas sample volumes via techniques, as Headspace and Thermal Desorption. It eliminates the need to cool the entire GC oven for cryo-focusing samples on the column. During the entire injection process, the volatiles in the gas sample injected into the GC injection port are concentrated in a narrow band at the front of the GC capillary column. When the GC Cryo-Trap is subsequently heated, the trapped volatiles are released in a narrow band to be chromatographed via the GC capillary column.

Figure 1

Figure 1 - Static Headspace System Components

Experimental

The LEAP Model CTC HS500 Headspace Autosampler was attached to the injection port of a HP Model 5890 Series II GC with electronic pressure control (Figure # 1). This headspace system uses a heated syringe concept to directly inject the headspace volatiles from a heated sample into the GC injection port (Figure #2). No long transfer lines or rotary valves are used in this system. The system is automated to permit the automatic injection and analysis of up to 50 samples unattended by the operator. Samples to be analyzed are placed into 10 ml glass vials with crimp tops which are sequentially inserted into a heated oven which can be agitated during the timed equilibrium step. Headspace volumes between 0.1 ml and 2.5 ml can be accurately removed from the sample into a heated syringe for injection into the GC. The HS500 Headspace Autosampler is mounted directly over the injection port of the GC in order to permit the direct injection of the sample into the GC injection port where the volatiles are subsequently trapped in the GC Cryo-Trap at the head of the GC column.

Figure 2

Figure 2 - Theory of Operation Of Headspace Sampler and GC Cryo-Trap

An HP 5890 Series II GC with electronic pressure control and a split/splitless injector was used for the following experiments. The HP Engine mass spectrometer was used as the detector in the EI mode. A J&W DB5-MS capillary column, 60 meters long by 0.32 mm I.D. x 0.25 micron film thickness was used in the GC oven.

The SIS GC-Cryo-Trap was attached to the bottom of the GC injection port inside the GC oven. A short 0.53 mm uncoated deactivated fused silica guard column was used from the GC injection port and through the GC Cryo-Trap. At the exit of the guard column, an SGE zero-dead volume union was used to join the uncoated guard column to the DB5-MS capillary column. The GC injection port was maintained at 200 degrees C and the GC Cryo-Trap was set for -160 degrees C during the trapping phase and 200 degrees C during the sample release phase and GC run time. The GC oven was temperature programmed from an initial 30 degrees C during the injection phase and to 200 degrees C at 4 degrees /min. The HP Engine (HP Model 5989 Mass Spec) was used for the detection and analysis of all compounds.

The LEAP Headspace Autosampler was used in the automatic mode. In fact, the entire process including sample equilibration, injection, cryo-trapping, release of volatiles and chromatography of the compounds was fully automated. The samples were placed into 10 ml headspace vials, sealed with PTFE lined septa, heated to 90 degrees C with agitation for 15 minutes in the sample block after which 1 to 2.0 ml gas samples were injected into the GC injection port. The samples were injected very slowly (25 ul/sec or 1.5 ml/min). This slow injection assures the full delivery of the analytes to the capillary column. Faster injection would result in sample splitting due to the backpressure design of the HP injection port. After injection into the cooled GC Cryo-Trap, the trap was maintained at the cryo-cooled temperature for at least another 3.0 minutes before the volatiles were released and the GC program begun. This cryo cooling equilibration time, after the injection is complete, permits the entire contents of the GC injection port to be passed onto the Cryo-Trap guard column and flushes the injection port of any remaining sample, thereby eliminating peak tailing and broadening. When the cryo-trap is rapidly heated, the released volatiles are eluted in a sharp band which produces highly resolved chromatography peaks.

Results and Discussion

Volatiles In Soil

Figure 3

Figure 3 - Volatiles In Sand At 40ng/gr - Total Ion Scan Mass Spec Analysis

A mixture of 20 volatile organics was spiked into headspace vials containing 5.0 grams of clean sand plus 5.0 ml of purge and trap quality water at levels from 800 ng/gram of sand (ppb) down to 0.05 ng/gram (ppb). Two milliliters of the heated headspace volume was injected into the GC. The higher concentrations of samples (800 ng/gram down to 10 ng/gram) were analyzed via a full mass spec scan from 25 to 250 daltons. A typical chromatogram is shown in Figure #3. A SIM method was developed using the two most intense mass spec ions of each volatile organic to run the weaker samples (100 ng/gram down to 0.05 ng/gram) as demonstrated in Figure #4.

Figure 4

Figure 4 - Volatiles In Sand At 1.0ng/gr - SIM Mass Spec Analysis Both sets of data were individually analyzed to develop calibration curves via the HP ChemStation Software.

The results of the analysis of all the volatile organics via the two techniques are summarized in Figure #5. The calibration curves for all the volatiles via both techniques were linear with correlation coefficients close to 1.000 for all the compounds. This high degree of linearity for all the volatiles over more than 4 decades of sensitivity is extremely good considering no internal standard was used for any of the samples analyzed. The limits of detection of the analytes via the total ion scan is about 10 ng/gram of sand for most of the analytes. Via the SIM method the limits of detection of the analytes are about 0.1 ng/gram of sand. This 100 fold increase in sensitivity is in line with the theoretical expectations of increases in sensitivity of the SIM method over the total ion scan mode for mass spectrometers.


Headspace/Cryotrapping of Volatiles in Sand
Figure 5

Figure 5 - Quantitative Analysis Via Total Ion Scan and SIM Methods To Evaluate Linearity Of the Analysis

Conditions For Determination Of Data Above

Headspace samples consisted of 5.0 grams of sand plus 5.0 ml of water plus volatiles. Analyze at levels from 800 ng/gr to 0.1 ng/gr. Equilibrate 25 min at 90 degrees C, then inject 2.0 ml of headspace volume into GC at 25 ul/sec (1.5 ml/min). Cryotrap at -160 degrees for 5.0 min, heat to 200 degrees to release volatiles and GC program from 30 degrees C to 100 at 4/min. Column 60 meter DB5-MS, 0.32 mm x 0.25 u film thickness. MDL is the amount of volatiles that can be accurately quantitated and with a signal to noise ratio of at least 10:1 in the chromatogram. A minimum of 10 points of quantification levels were used to set up the calibration curves and to determine the correlation coefficients.


Figure 6

Figure 6 - Volatiles In Sand At 2.5 ppb Level - Multiple 2.0 ml Headspace Injections

2.5 ppb levels of volatiles in 5.0 grams of sand plus 2.5 ml water. Headspace Injection of 2.0 ml volumes onto blank capillary column and trap at 150 degrees C for 5.0 min. Then heat to 200 degrees C to release and chromatograph.


Multiple Injections

The cryo-focusing of volatiles at the front of the capillary column, enables the multiple injection of samples into the GC injection port. This is shown in Figure # 6. The same mixture of 20 volatile organics used above was spiked at a level of 2.5 ng/gram of sand into 5.0 grams of sand plus 5.0 ml of water. A 2.0 ml sample of the headspace volume was injected into the GC injection port at increasing numbers of repetitive injections. With a single injection only benzene (peak 8) is barely detectable. With increased numbers of injections, the analytes are more easily detected. Very little increase was seen between 3 and 4 injections because the injections originated from the same headspace sample vial. The limits of repetitive injection occurred at 5 injections where an ice plug developed in the GC Cryo-Trap. The multiple injection technique can be used to expand the usefulness of the headspace injection technique and increase the sensitivity of the analysis of volatile organics.

Reproducibility

A mixture of 10 volatile organics at a concentration of 500 ng/gram (ppb) in 5.0 grams of sand plus 5.0 ml of water were placed into the headspace sample vials and the headspace volume was injected into the GC and analyzed. Separate samples were analyzed 10 times to determine the reproducibility of the analysis technique and the results summarized in Figure #7. The GC retention times were repetitive within 0.01 minutes for all the volatiles and the Relative Standard Deviation for the peak areas was better than 5.0 % for all of the volatiles analyzed. This is very good, especially since no internal standard was used for the analysis. This data along with the data above demonstrates the linearity and reproducibility of this technique for the quantitative analysis of volatile organics. The headspace technique in combination with cryo-focusing can readily be incorporated into a scientifically acceptable method for the quantitative analysis of volatile organics in liquid and solid samples.


Figure 7

Figure 7 - Reproducibility of Headspace / GC Cryo-Trap

2.0 ml of Headspace gas at 90 degrees C, Cryo-Trap at -120 deg C for 5.0 min. Chromatograph on 0.32 mm x 60 meter x .25u DB5-MS Capillary Column.


Effect of Cryo-Trapping Temperature

The GC Cryo-Trap technique traps compounds based on their melting point when using an uncoated capillary guard column. The range of volatiles that are trapped can be varied by adjusting the GC Cryo-Trap temperature. For this study, 0.5 ul of gasoline was injected into 5.0 ml of water in a headspace vial. The headspace vial was heated to 70 degrees C and 1.0 ml of the headspace gas was injected into the GC injection port and then cryo-trapped at various temperature between -60 degrees C and -180 degrees C (Figure #8). At -60 degrees C volatile hydrocarbons down to hexane are trapped. The lighter volatiles pass through the GC Cryo-Trap and capillary column during the injection and equilibrium steps of the method. If the mass spec or GC detector were left on during this sampling period, these lighter compounds which are not trapped would appear as broad low intensity GC peaks. Lower trapping temperatures will trap increasingly lower melting point volatiles. Hydrocarbons down to ethane can be trapped at temperatures of -180 degrees C. This ability to control the melting range of volatiles that can be trapped can be a useful analytical technique and permit the modification of the operational parameters to analyze or not analyze a range of volatile organics.

Figure 8

Figure 8 - Headspace Analysis of Gasoline In Water Onto a Capillary GC Column

0.5 ul gasoline in 5.0 ml of water, Heat to 70 degrees C., Inject 1.0 ml into 0.32 mm DB-5 MS Capillary column and trap at various temperatures.

Analysis of Gasoline in Soil Samples

Figure #9 displays the quantitative results of a gasoline sample in various sample matrices. For this study, 0.5 ul of gasoline was injected into headspace vials containing water, sand plus water and finally top soil plus water. As above the headspace vial was heated to 70 degrees C and 1.0 ml of the headspace, gas was injected into the GC injection port. All of the volatiles were easily detected in the water sample. In the pure sand plus water sample, the results were near identical. Sand has little affinity or binding effect on the volatile organics. However, in the top soil plus water sample, the volatiles above toluene and octane are not readily detected. Even many of the lower boiling volatile organics such as benzene and toluene exhibit lower peaks heights in the chromatogram. These results have been observed in other studies in which soils, especially top soils and clay soils, have a strong binding effect on many organic compounds. Increased sensitivity of the volatiles in the soils have been achieved by heating the headspace sample vial to a higher temperature. However, due to the high levels of water, this temperature is limited to 95 degrees C.

Figure 9

Figure 9 - Headspace Yields On Gasoline Spiked Into Sand and Top Soil

0.5 ul ofgasoline in 5.0 ml of water - NO SAND

Conclusion

The use of the GC Cryo-Trap in conjunction with the automated headspace system has proven to be a powerful technique. This combination of hardware permits the quantitative analysis of volatiles in various matrix samples over more than 4 decades of sensitivity utilizing capillary GC columns. Most previous GC headspace methods have been done using packed GC columns and megabore capillary columns. However, it is now possible to inject large gas volumes (0.5 ml to +100 ml) into a microbore capillary column by concentrating the analytes in a narrow band at the front of the GC column over a time period from 1.0 minute to >30 minutes, then rapidly releasing the analytes for chromatography. The relatively slow injection of large sample volumes at injection rates of more than 1.0 ml/min eliminates the normal splitting of samples which would occur with faster sample injection. All of the volatiles are passed onto the capillary column and cryo-trapped and concentrated in a narrow band at the front of the GC column or guard column inside the GC Cryo-Trap.

The resulting GC peaks are highly resolved, the retention times extremely reproducible and the areas of the GC peaks reproducible and linear over more than 4 decades of sensitivity. The techniques permit the adjustment of a number of variables such as the sample temperature, the gas volume injected, the trapping temperature and the release temperature. This will permit the development of experimental conditions to fit the needs of the chemist. This technique can be used for applications such as the analysis of volatiles in soils, residual solvents in pharmaceuticals, blood alcohols and other headspace techniques that could profit from the analysis of compounds on a capillary column or which could be enhanced by the injection of larger gas sample volumes or repetitive injections onto the GC column. The use of the GC Cryo-Trap can expand the range of usefulness of the GC headspace technique.