The analysis of fuel oxygenates in soil and groundwater has not been an issue until recently. Volatile organic compounds such as chlorinated hydrocarbons (EDC, Chlorobenzene, TCE, PCE, etc.) as well as benzene, toluene, ethylbenzene and the xylene isomers (BTEX) constituted the compounds-of-concern for the majority of remediation projects. Methodologies for these 'normal' compounds have been around and modified over the years for the purpose of providing defensible, quantifiable data in environmental samples. The addition of fuel oxygenates to the mixture will add another level of complexity to an already complex issue, yet this addition should not prohibit attaining the goal - Defensible, quantifiable data in environmental samples. This paper will concentrate on tertiary butyl alcohol (TBA) and methyl tertiary butyl ether (MTBE) representing the two major groups of oxygenates - alcohols and ethers.

Regulatory laws and limits for oxygenates are as complex as the sample matrices analyzed for oxygenates. For instance, the clean-up levels for MTBE at remediation sites vary nation-wide from 200,000 ppb to 13 ppb to 'we do not have one yet.' The criteria for drinking water standards varies just as much. This state/site specificity of the clean-up levels for fuel oxygenates requires methodologies that can detect wide ranges of oxygenate concentrations in matrices from potable water to non-aqueous phase liquid (NAPL).

Currently, several methods are being utilized for the measurement of fuel oxygenates in environmental matrices. The two most widely available are SW-846 methods 8021 and 8260. Method 8021 is a gas chromatograph based analysis using a photo ionization detector (PID) at times coupled with a flame ionization detector (FID). Method 8260 uses the gas chromatograph with a mass spectrometer (GC/MS) as the detector. Sample introduction for both methods for the purposes of the paper will be purge and trap.

Methodology Overview

As stated above, the only difference between the two methods is the detector. Purge and trap (method 5030) is used to liberate the volatile compounds from the matrix by passing an inert gas through the aqueous sample or soil extract. The volatile compounds are then trapped on a bed of adsorbant media. This media is then quickly heated and the sample (now in gaseous form) is swept to the gas chromatograph. Now that the compounds have been separated from the matrix, the gas chromatography (GC) is used to separate the sample components from each other. The intent of chromatography is to separate compounds by specific properties or physical characteristics (i.e. boiling points, polarity, size, and shape) over time. Once separated, each analyte can be identified and measured.

Method 8021 utilizes a photo ionization detector. The PID is a non-destructive detector and is most sensitive to compounds with double bonds. This specificity in the detector provides little response for the group of alcohol oxygenates. The identification of a compound is solely based on retention time.

Method 8260 utilizes a mass spectrometer (MS) detector with electron impact ionization. Upon ionization, the MS generates mass spectra, which is generally unique for any given compound. This mass spectra coupled with an associated retention time is a far more dependable means of identifying compounds in complex environmental matrices. The mass spectra enables the analyst to ‘see inside’ a chromatographic peak and distinguish between compounds that would appear as one in a method 8021 analysis.

Now that a general understanding of the methods has been discussed, there are many things can be done to enhance the response for oxygenates, thus generate lower detection limits and better data.

IDENTIFY THE PROBLEM

In order to measure oxygenates, or any other compound for that matter, you need to know something about that compound. Compound solubility, molecular weight, reactivity and polarity are just some of the basic chemical characteristics which will aid in the accuracy of your measurement.

Next, you need to liberate the compound from the matrix. For ether based oxygenates, normal purge and trap (P&T) conditions generally provide sufficient response to quantify MtBE down to 1.0 – 2.0 ppb. For alcohols (e.g. TBA and ethanol) under the same conditions, responses may yield detection limits in the 100-250 ppb range. However, placing the sparge tube on the P&T in a water bath at 40-80 C will increase the purge efficiency of these compounds 2-5 fold. MtBE detection limits will approach the 0.1– 0.2 ppb range and alcohol detection limits should range from 25-100 ppb . Another recommendation is increasing the purge flow rate. To remove highly soluble alcohols, the more vigorous the agitation during the purge cycle the more compound will be removed from the matrix – generating a larger response at the detector. Sample size may also be increased from 5mL (normally) to 10mL-25mL, which will decrease the detection limits another 2-5 fold. Again, the larger sample size will contain a larger amount of the compound to be measured and increase the response at the detector. Some measurement systems may not respond as well to the conditions specified above. The key is to try each modification and see which combination works best with your measurement system.

Now that all has been done to get as much of the oxygenate compounds out of a given matrix, it is necessary to separate the compounds. Gas chromatographs over the last decade have made huge advances. The improved functionality and reproducibility of the GC has allowed the analyst to generate complex pressure, flow and temperature programs to separate and isolate many fuel oxygenates from other fuel hydrocarbons. There are advances in chromatography products including specialty columns designed for use specifically for BTEX and oxygenate analyses.

As the gaseous sample stream enters the detector, identification and quantification of the various compounds in a sample is where the real challenge begins and where detector selection becomes crucial. While all the previous steps were used to improve the response of the oxygenate compounds being analyzed for, they also increase the amount and number of interferences being detected.

8021 - Pros and Cons

The most common analytical method for BTEX, method 8021 is very widely used for the determination of all fuel components and is also the most economical ($30-$50 per sample). In clean matrices- clean defined as < 1 ppm of petroleum hydrocarbons present in the sample- the method is generally very useful for the quantification of the ether oxygenates. The photo-ionization detector’s sensitivity and reproducibility are excellent in these ‘clean’ samples with the exception of the alcohol oxygenates – which deliver poor response with the PID. However, when analyzing samples in a moderately contaminated matrix, know that the interference due to co-eluding peaks from other hydrocarbons will greatly reduce accuracy and may produce false positive results. Once petroleum hydrocarbons reach 1 ppm, the incidence of false positives rises dramatically and the accuracy of those components correctly identified suffers greatly. The separation techniques described above will not always isolate all the compounds of interest. Being able to see ‘inside’ these petroleum hydrocarbon peaks and filtering out the interference is where the mass spectra generated by method 8260 becomes vital to generating accurate data.

8260 - Pros and Cons

Although more expensive than method 8021, method 8260 (generally $100 - $150 per sample) can quantify more compounds in more matrices and provide better data to the end user. The mass spectrometer will detect alcohols, ethers, hydrocarbons and halogenated compounds as well. There are a number of modifications that can be made to the acquisition parameters of the mass spectrometer that will further filter interference, increase response and confirm the presence of oxygenates in almost all matrices. The verification of a compound’s unique mass spectra adds further confirmation to the method 8260 identification and measurement processes.

Single ion monitoring (SIM) is a method of mass spectral acquisition by which only certain distinguishable pieces or masses of a compound are monitored. This is also the most sensitive method of acquisition - enabling the analyst to detect MtBE at quantities less than 0.05 ppb.

Other Tools

Another very useful and more expensive tool using method 8260 is called isotopic dilution. This analysis technique replaces the common internal standards used for quantitation with a labeled version of the compound of interest. For instance, if analyzing for benzene – whose quantitation ion is 78, one would fortify the sample with a known amount of its deuterium labeled counterpart, benzene-d6– whose quantitation ion is 84. The analyst then calculates the concentration of the natural compound by comparing the response of benzene versus labeled benzene-d6. The isotope follows the entire analytical pathway and allows the analyst to correct for the measurement system’s response to the analyte. This method is only to be used with a mass spectrometer detector. All common environmental remediation type compounds have either a deuterium or C13 isotope. For more information on this technique, see Method 1624 in the Code of Federal Regulations (40 CFR Part 136 Appendix A).

A promising new technology is ASTM method D4815. This method is moderately expensive (> $100) and not many labs offer this analysis. Method D4815 is a very specialized analysis using two-dimensional gas chromatography and a flame ionization detector. The difference in polarity between oxygenates and the aromatic BTEX constituents of a fuel-contaminated matrix are used to separate the compounds. No BTEX data is generated. However all fuel oxygenates, including ethanol and methanol, can be quantified.

A new extraction technique called Solid Phase Micro Extraction (SPME) is being tested for its use on all oxygenates. A small fiber coated with a special polymer is allowed to soak in the aqueous matrix, removing the compounds. The fiber, having extracted the compounds of interest, is then analyzed via GC/MS. Researchers at the University of Nebraska developed this technique for the measurement of ethanol with detection limits in the 15ppb range.

QC, QC, QC

Many different techniques and methodologies have been presented in this paper. All the method enhancements are only as good as the analyst that employs them. The detection limits listed are based on a clean matrix or reagent water. Specific detection limits for fuel-contaminated matrices should be analyzed and will be generally higher in all methods depending on the level of petroleum hydrocarbons present.

The QC program for these methods needs to be stringent. Each procedure modified for the determination of fuel oxygenates in soil and groundwater needs to be proven in each lab by their chemists using their instrumentation. Matrix specific QC ranges for spikes, surrogate recoveries and laboratory control samples need to be in place. The quantification of fuel oxygenates in an aqueous or solid matrix is becoming a big issue as the demand for lower and lower levels of detection grows. Utilizing these analytical tools along with a stringent QC program will allow you to attaining the goal - Defensible, quantifiable data in environmental samples.
 

Top