SITE INFORMATION AND CORRESPONDENCE

Home Browse Search

LLISTLine Bvtletin 42 • October 2002 ■Analytical Methods from page 3 the commercial laboratories detected to 80°C during sample preparation. overall oxygenate recovery com- TBA in 10 wells;the%RSD was 21.4. However,heating creates a problem pared to the DB-Wax capillary col- The method detection limits for TBA with conventional practice for pre- umn.In addition,BTEX interferences in the EPA laboratory and the com- serving groundwater samples. did not adversely affect the chro- mercial laboratory were 2.4 and 5 If the sample is heated, the acid matographic separation, quantita- tig/L respectively.The reported con- commonly added to preserve the tion,and recovery of oxygenates. centrations of TBA ranged from 6 to sample can actually cause the For samples with high concentra- 154µg/L.The other commercial lab- hydrolysis of ether bonds.As a con- tions of hydrocarbons andoxy- oratory had a minimum reporting sequence,ether concentrations origi- genates, the samples will have to be limit of 100 jig/L and did not detect nally present in the sample may be diluted so that they are within the TBA in any of the water samples ana- underestimated, and the concentra- operating range of the instrument.As lyzed.The agreement between analy- tion of the hydrolysis products may a general rule, analysts dilute and ses of MTBE,TAME,ETBE, and TBA be overestimated (e.g., TBA formed rerun samples when the concentration was good.The other oxygenates were from the MTBE hydrolysis). of any analyte exceeds 0.5 mg/L when not present in these plumes at con- These analytical errors can cause using Method 8260 (MS detector) or centrations that made it possible to errors in risk assessment,can lead to exceeds 4 mg/L when using Method make a comparison. the implementation of a remedial 8015(FID).If the concentration of one technology that is not necessary,and of the BTEX compounds or oxy- Ether Hydrolysis can bias an evaluation of monitored genates is much higher than the other Under normal environmental condi- natural attenuation (MNA). For analytes,then multiple runs will have tions ethers do not undergo hydroly- example, the alcohol that corre- to be made using diluted samples. sis at significant rates without sponds to the ether is often the first These methods must only be enzyme catalysis; even in acidified product of biotransformation of the used by, or under the supervision of, (pH<2)groundwater samples,ethers ether. Higher concentrations of the analysts experienced in the use of gas are generally stable (Church et al., alcohol and lower levels of the ether chromatography for measurement of 1999). However, Wade (1998) may be interpreted erroneously as organic compounds at low concentra- reported evidence of decreasing evidence for natural biodegradation tions (i.e., µg/L) and skilled in the MTBE concentrations in 91 acidified in the plume. Consequently, the time interpretation of gas chromatograms groundwater samples collected over required for MNA to achieve cleanup and/or mass spectra. Each analyst a two-year period from a site known goals may be significantly underpre- must demonstrate the ability to gen- to have experienced a release of gaso- dicted. erate acceptable results with these line that contained MTBE.He postu- Recently, O'Reilly et al. (2001) methods.This should be no different lated that acid-catalyzed hydrolysis Published rate constants that can be than current good laboratory practice. of MTBE during sample storage used to calculate the effect of temper- To demonstrate that these meth- could explain these observations. ature on the rate of acid hydrolysis of ods work as well with real field sam- Most protocols for the preserva- MTBE in samples of groundwater. ples as they do with laboratory- tion of groundwater samples call for They measured the rate of MTBE prepared samples, EPA recently par- the addition of a sufficient volume of hydrolysis at 26'and 37°C. As dis- ticipated in an interlaboratory com- hydrochloric acid to adjust the pH of cussed above,a temperature of 80°C parison of the performance of the sample to<2.As a practical mat- is necessary to promote efficient methods for the BTEX compounds ter,more acid is added than is needed transfer of alcohols to the gas phase and the fuel oxygenates using static to preserve the samples. One stan- for sampling. If the rates published headspace as the sample preparation dard drop of concentrated hydrocMo- by O'Reilly et al.are extrapolated to method.Water samples from moni- ric acid will adjust distilled water in a 80°C, they predict that MTBE should toring wells in two fuel plumes on standard 40 mL VOA vial to pH=1.8. be rapidly hydrolyzed to TBA during Long Island were sent to an Most field technicians add two or analysis. EPA/ORD lab and two commercial three drops of acid to each 40 mL EPA/ORD measured the rate of labs.The agreement in the reported VOA vial. Typically, it takes seven MTBE hydrolysis at 80°C at pH =1 concentrations between the three lab- drops of acid to adjust a 40 mL VOA and pH=2;the results are presented oratories was expressed as the per- vial to pH=1.The majority of ground- in Table 1.The water samples in the cent relative standard deviation water samples that have been pre- heated headspace sampler are typi- (%RSD)of the samples. served with acid probably have a pH cally heated for 30 minutes before All three laboratories reported of between 1 and 2. they are analyzed.After 30 minutes concentrations of MTBE above their As discussed in the preceding of incubation,6 percent of the MTBE detection limit in water from 23 of the section, if purge-and-trap is used as was hydrolyzed to TBA at a pH of 2, 50 wells that were sampled. The the sample preparation procedure for and 57 percent of MTBE was %RSD for MTBE was 12.9.All three TBA and the other alcohols, then it hydrolyzed at a pH of 1. laboratories detected ETBE in water must be modified to increase method Data documenting the hydrolysis samples from six wells; the %RSD sensitivity, or an alternate high- of MTBE during analysis of ground- was 12.3. All three laboratories temperature sample preparative water samples from an MTBE plume detected TAME in 12 wells; the procedure must be used. One in California are presented in Table 2. %RSD was 5.3. straightforward approach to increase The samples were preserved in the The EPA laboratory and one of sensitivity is to heat the water sample field with hydrochloric acid to pH<2 4