PCB Source and Dechlorination Fingerprinting
Stephen Emsbo-Mattingly, NewFields Environmental Forensics Practice LLC, Rockland, MA

Forensic Analysis and PCB Fingerprinting in Sediments Using Congener Data and Multiple Statistical Evaluation Methods   
Noémi Barabás, Limno-Tech, Inc., Ann Arbor, MI

Fact or Fiction: The Source of Perchloroethylene Contamination in Groundwater is a Manufacturing Impurity in Chlorinated Solvents
Valerie Lane, GeoTrans, Inc., Harvard, MA

Thoughts on Manufacturing Changes in the US Petroleum Industry:  Implications for Age-dating, Calculating Weathering Indices and Hydrocarbon Fingerprinting
Dr. Michael Wade, Wade Research, Inc., Marshfield, MA 

Age-Dating Diesel Fuel:  A Case Study Averse to the Christensen and Larsen Method
Scott A. Stout, NewFields, Rockland, MA 

Identification of Natural Gas Sources Using Geochemical Forensic Tools

Paul Boehm, Ph.D., Exponent, Inc., 3 Clock Tower Place, Suite 205, Maynard, MA 01754, Tel: 978-461-1220, Fax: 978-461-1223
Tarek Saba, Ph.D., Exponent, Inc., 3 Clock Tower Place, Suite 205, Maynard, MA 01754, Tel: 978-461-1233, Fax: 978-461-1223
Laurie Benton, Ph.D., 15375 SE 30th Place, Suite 250, Bellevue, WA 98007, Tel: 425-519-8754, Fax: 425-519-8799

Processed natural gas (storage gas) from a number of states has been transported to the permitted storage facility and injected in an underground formation 900 ft below ground surface. The formation is surrounded by geologic divides initially thought to be preventing gas from escaping the storage formation (storage field) boundaries. Recently, pressure monitoring in observation wells around the storage field indicated that the southern divide is leaking storage gas. In addition, property owners south of the field have begun producing natural gas and claim that the natural gas they are extracting is native gas from the area and not storage gas escaping the field. The objective of this work was to apply knowledge of oil and natural gas geochemistry and geochemical fingerprinting techniques to determine if storage gas is escaping the southern divide; to determine the extent of storage gas migration beyond the divide; and to determine if the other companies are extracting storage or native gas. A field investigation was conducted where gas samples were collected from the storage field and from observation wells located outside (south of) the field. Samples were analyzed for gas hydrocarbons and nonhydrocarbons, and stable isotope compositions (carbon for methane and ethane). Using double ratio analysis (hydrocarbons represented by C1/C2+), tracers (helium), and carbon isotope ratio of ethane, the different fingerprints of native and storage gases were identified. The analysis determined that gas in the observation wells across the southern divide is storage gas indicating that the divide is in fact leaking. This work showed that storage gas is migrating beyond the storage field boundary and that property owners adjacent to the field are extracting storage gas escaping the field.

PCB Source and Dechlorination Fingerprinting

Stephen Emsbo-Mattingly, M.S., NewFields Environmental Forensics Practice, LLC, 100 Ledgewood Place, Suite 302, Rockland, MA 02370, Tel: 781-681-5040
Victor Magar, Ph.D., ENVIRON International Corporation - Chicago, IL, 123 North Wacker Drive, Suite 250, Chicago, IL 60606, Tel: 312-853-9430
Marc Mills, Ph.D., USEPA ORD NRMRL,26 W. Martin Luther King Dr., MS 420, Cincinnati, OH 45268, Tel: 513-569 7322
Richard Brenner, Ph.D., USEPA ORD NRMRL, 26 W. Martin Luther King Dr., MS 420, Cincinnati, OH 45268, Tel: 513-569 7322

Detailed chemical forensic analyses were conducted to characterize source and chemical alteration congener compositions in lake sediments.  At the Lake Hartwell Superfund Site ( Pickens County , SC ), surface sediments resembled a 50/50 mixture of Aroclors 1248 and 1254.  Congeners became increasingly dominated by lower chlorinated congeners with sediment depth and corresponding age, resulting in a relative accumulation of ortho chlorines and loss of meta and para chlorines; ortho chlorines were highly conserved. 

This study presents a quantitative approach used to measure dechlorination processes in situ.  The conservation of ortho chlorines with depth allowed the use of ortho chlorines to serve as a unique fingerprint of the original source material and dechlorination byproducts.  Toxic equivalencies were calculated using Toxic Equivalency Factors listed by the World Health Organization (WHO).  WHO congeners were preferentially removed under natural conditions resulting in an apparent reduction in toxicity relative to total PCB mass. 

Forensic Analysis and PCB Fingerprinting in Sediments Using Congener Data and Multiple Statistical Evaluation Methods

Noémi Barabás, Ph.D., Limno-Tech, Inc., 501 Avis Drive, Ann Arbor, MI  48108, Tel: 734-332-1200, Fax: 734-332-1212, Email: nbarabas@limno.com
Carrie Graff, Ph.D, Limno-Tech, Inc, 1705 DeSales St., NW, Washington, D.C. 20036, Tel: 202-833-9140, Fax: 202-833-9094, Email: cgraff@limno.com        
Rich Galloway, Honeywell International, 101 Columbia Road , Morristown, NJ 07962, Tel: 973-455-4640, Email: rich.galloway@honeywell.com 
Daniel Herrema, P.E., Limno-Tech, Inc., 501 Avis Drive, Ann Arbor, MI 48108, Tel: 734-332-1200, Fax: 734-332-1212, Email: dherrema@limno.com
Timothy J. Dekker, P.E., Ph. D., Limno-Tech, Inc., 501 Avis Drive, Ann Arbor, MI  48108, Tel: 734-332-1200, Fax: 734-332-1212, Email: tdekker@limno.com

A study was performed to determine the PCB sources and fate processes affecting congener compositions in an estuarine contaminated sediment site in New Jersey .  PCB contamination of sediments can originate from different sources and Aroclor mixtures, and fate processes can modify congener signatures.  Such signatures can help identify Aroclor mixtures, their sources and relevant fate processes.  Multivariate statistical evaluation of PCB congener patterns in sediment samples, compared to known signatures, allows identification of Aroclor fingerprints, chemical alteration processes, spatial distribution of Aroclors and alteration patterns, and a refined understanding of processes responsible for transporting solids/contaminants across the site.  Multivariate analysis depends on incomplete mixing within the sediment environment, preserving various degrees of statistically detectable ‘signals’ of the original mixtures (Aroclors) in each sample.  Three levels of statistical analysis were used to identify Aroclor fingerprints, and formulate hypotheses about source identities and fate processes.  The congener composition of individual samples, the spatial pattern of sample compositions, and the correlation patterns of congeners indicated the presence of more than one distinctive pattern at the site.  Principal component analysis (PCA) confirmed this interpretation, and polytopic vector analysis (PVA) was performed to determine their compositions and relative importance.  PVA “unmixes” congener distributions within samples and predicts congener distributions of candidate source fingerprints. PVA is based on traditional PCA and subsequent rotation of principal component axes until all PC compositions and their sample loadings are positive. Results point to both on-site and potential off-site sources of PCB impacts.  PCB contamination is dominated by Aroclor 1248, and congener patterns in buried sediments have been altered by dechlorination processes over time.  Results also allowed for evaluation of Aroclor concentration gradients across the system, and clarified the major sediment transport processes operative at the site.

Fact or Fiction:  The Source of Perchloroethylene Contamination in Groundwater is a Manufacturing Impurity in Chlorinated Solvents

Valerie A. Lane, GeoTrans, Inc., 6 Lancaster County Road, Harvard, MA, 01451
James S. Smith, Ph.D., CPC, Trillium, Inc., 8 Grace's Drive, Coatesville, PA 19320-1206  

Manufacturing impurities in chlorinated solvents have been considered to be sources of contamination in groundwater. Chlorinated solvents are manufactured in a variety of grades; the technical grade is used at many industrial and manufacturing facilities. Compounds present as manufacturing impurities in technical grade chlorinated solvents vary, and their quantity is extremely small or not measurable, because chlorinated solvents historically have been manufactured to a high degree of purity. The purity of currently manufactured TCE ranges from 99.9% for reagent grade to 98.0% for the technical grade. Impurities in  technical grade 1,2-dichloroethane, also known as ethylene dichloride (EDC), manufactured within the last 10 years with purities of 99.9991% and 99.9955% included PCE between about 0.0001% and 0.0006%, respectively.           

In a number of litigation cases where TCE released from a vapor degreaser is the major contaminant in groundwater, there is an accompanying minor concentration of PCE. The assumption made in these cases is that the PCE impurity in technical grades of TCE is the source of the PCE in groundwater. This assumption is based on the release of these two compounds together from a distillation bottom residue where PCE is concentrated, relative to the TCE, because of its much higher boiling point at atmospheric pressure.

There is no peer reviewed literature that can be cited to prove that PCE is a significant impurity in either the distillation bottoms produced from a vapor degreaser or technical grade TCE. PCE, if present as a manufacturing impurity, is present in such small amounts that significant concentrations would not be generated in groundwater. When PCE and TCE are present together in groundwater, the source of the PCE is likely not an impurity in the manufactured TCE.

Thoughts on Manufacturing Changes in the US Petroleum Industry:  Implications for Age-dating, Calculating Weathering Indices and Hydrocarbon Fingerprinting

Dr. Michael Wade, Wade Research, Inc., 110 Holly Road, Marshfield, MA 02050, Tel: 781-837-5504, Email: mjwade@waderesearch.com

Techniques for the aging of petroleum contamination in environmental matrices are undergoing major revision in the environmental forensic community at the present time. Techniques that were reliable in the past to age-date gasoline may not be that reliable in the reality of today's modern refinery practices. Age-dating of distillate fuels is undergoing a major debate in the forensics community at the present time. Forces working to expand the use of hydrocarbon degradation models from soils to other matrices may have pushed the most widely-used weathering approach too far, creating a two-sided debate which has not been settled in the peer reviewed literature as of now. Problems with numerous approaches to age-date gasoline and distillate fuel contamination will be outlined, and discussed. Both the positive and negative aspects of numerous common approaches will be discussed. Recommendations for cautiously age-dating petroleum in the environment will be presented.

Age-Dating Diesel Fuel:  A Case Study Averse to the Christensen and Larsen Method

Scott A. Stout, NewFields, 100 Ledgewood Place, Suite 302, Rockland, MA 02370, Tel: 781- 681-5040, Fax: 781- 681-5048, Email: sstout@newfields.com
Gregory S. Douglas, NewFields, 100 Ledgewood Place, Suite 302, Rockland, MA 02370, Tel: 781- 681-5040, Fax: 781- 681-5048, Email: gdouglas@newfields.com

Age dating petroleum products is a valuable forensic tool because it can be used to determine the time of release (i.e., ownership) and subsequent liability for site clean up costs.  Diesel fuels necessarily contain abundant normal alkanes, i.e., the class of hydrocarbons that most easily ignite under compression, and numerous iso-paraffins within the same boiling ranges as the normal alkanes.   It has been long recognized that comparably boiling normal alkanes are more susceptible to biodegradation in the environment than iso-paraffins.  

Investigators have used the ratio(s) between selected normal alkanes (e.g., n-C17) and iso-paraffins (e.g., pristine) to compare the extent of biodegradation among diesel fuel residues in environmental samples.  Christensen and Larsen (1993) reviewed data for soils from several European sites with documented release histories and concluded that the ratio of n-C17/Pr in soil extracts could be used to estimate the length of time (up to 20 years ± 2.1 years) that diesel fuel had been in the environment, so long as certain conditions were met.  Although the universal validity of the Christensen and Larsen Method has been debated, one particular condition – a single release of diesel fuel at a particular point in time – is extremely rare and/or difficult to prove at most petroleum handling facilities.  Consequently, subsequent testing of the Christensen and Larsen Method has been extremely rare.

In this case study, a catastrophic release of approximately 6000 gallons of on-road diesel fuel #2 occurred when an underground storage tank (UST) was accidentally punctured during drilling in February 1991.   When “fresh” NAPL appeared in 2002, the issue of its “age” was of paramount importance since the property had changed ownership.  In 2003, NAPLs and impacted soils throughout the study area were collected and analyzed using detailed chemical fingerprinting as a means of assessing the “age” of the contamination.  The NAPLs were minimally biodegraded (n-C17/Pr ~ 1.6), exhibited a consistent “genetic’ character (e.g., sesquiterpane and methyl-phenanthene patterns and relative S-PAH abundance), and contained 1870 to 2500 ppm total sulfur, which according to the Federal On-Road Diesel Fuel Sulfur Reduction Act of 1993 indicated the NAPLs’ parent fuel was used/released before November 1993.  The soils collected in 2003 met the aforementioned conditions of the Christensen and Larsen Model, yet despite exhibiting consistent genetic features with each other and with the NAPLs, exhibited a wide range of n-C17/Pr ratios (0.0 to 1.8) which would imply “ages” between 20+ and ~4.5 years.   These results argue that despite a one-time release of a particular type of diesel fuel in 1991, the rates of normal alkane biodegradation within this site’s soils were sufficiently variable to confound the application of Christensen and Larsen Model.   The high sulfur contents of the NAPLs and the “genetic” comparability to the soils provided a defensible basis to attribute both the NAPL and soil contamination to the 1991 catastrophic release – and not a new release(s).   The “sudden” appearance of NAPL in 2002, which promulgated this study, occurred after a marked drop in groundwater elevation due to drought  – and not a new release(s).   This case study serves to emphasize the imprecision of the Christensen and Larsen Method for age-dating diesel fuel residues in soils and, certainly its inappropriateness for age-dating NAPLs.

Top