The toxicity and environmental persistence of anthropogenic per- and poly-fluoroalkyl substances (PFAS) are of global concern. To address legacy PFAS concerns in the United States, industry developed numerous replacement PFAS that commonly are treated as confidential information. To investigate the distribution of PFAS in New Jersey, soils collected from across the state were subjected to nontargeted mass-spectral analyses. Ten chloroperfluoropolyether carboxylates were tentatively identified, with at least three congeners in all samples. Nine congeners are ≥(CF2)7. Distinct chemical formulas and structures, as well as geographic distribution, suggest airborne transport from an industrial source. Lighter congeners dispersed more widely than heavier congeners, with the most widely dispersed detected in an in-stock New Hampshire sample. Additional data were used to develop a legacy-PFAS fingerprint for historical PFAS sources in New Jersey.

The determination of the animal feed additive roxarsone (3-nitro-4-hydroxyphenylarsonic acid) and six of its possible transformation products (arsenite, arsenate, monomethylarsonate, dimethylarsinate, 3-amino-4-hydroxyphenylarsonic acid, and 4-hydroxyphenylarsonic acid) in chicken manure was investigated using capillary electrophoresis-inductively coupled plasma-mass spectrometry (CE-ICP-MS). Initial method development was conducted using ultraviolet (UV) detection for ruggedness and time efficiency. Separation of these seven arsenic species was effected using a 20 mM phosphate buffer at pH 5.7. The CE-ICP-MS limits of detection in terms of As for each of the species was in the low microg.L(-1) range, corresponding to absolute detection limits in the range 20-70 fg As (based on a 23 nL injection). Overall, the method developed in this study provides high selectivity and low limits of detection (1-3 microg.L(-1) or low-ppb, based on As), uses small sample volume (low nL), and produces minimal wastes.

Agricultural pesticides are being transported by air large distances to remote mountain areas and have been implicated as a cause for recent population declines of several amphibian species in such locations. Largely unmeasured, however, are the magnitude and temporal variation of pesticide concentrations in these areas, and the relationship between pesticide use and pesticide appearance in the montane environment. We addressed these topics in the southern Sierra Nevada mountains, California, by sampling water weekly or monthly from four alpine lakes from mid-June to mid-October 2003. The lakes were 46-83 km from the nearest pesticide sources in the intensively cultivated San Joaquin Valley. Four of 41 target pesticide analytes were evaluated for temporal patterns: endosulfan, propargite, dacthal, and simazine. Concentrations were very low, approximately 1 ng/L or less, at all times. The temporal patterns in concentrations differed among the four pesticides, whereas the temporal pattern for each pesticide was similar among the four lakes. For the two pesticides applied abundantly in the San Joaquin Valley during the sampling period, endosulfan and propargite, temporal variation in concentrations corresponded strikingly with application rates in the Valley with lag times of 1-2 weeks. A finer-scale analysis suggests that a large fraction of these two pesticides reaching the lakes originated in localized upwind areas within the Valley.

An analytical method using solid phase extraction (SPE) and analysis by gas chromatography/mass spectrometry (GC-MS) was developed to determine trace levels of a variety of 41 agricultural pesticides and selected transformation products in high-elevation surface waters. Large-volume water sampling (up to 100L) was employed because it was anticipated that pesticide contamination, if present, would be at very low levels. The target compounds comprise pesticides (and selected oxygen transformation products) known to have been extensively used in agriculture in the San Joaquin Valley, CA, USA. Solid phase extraction using the polymeric resin Abselut Nexus was optimized to extract the pesticide analytes from water samples. A single determinative method using GC-MS with electron ionization was used for all the analytes. Recoveries from 100L of reagent water at 100pg/L and 1ng/L concentrations were generally greater than 75%, although dimethoate, disulfoton, and phorate were not recovered. Analysis of the extracts without cleanup yielded detection limits for the remaining 38 analytes between 0.1 and 30ng/L. A silica cleanup with separate analysis of 3 eluant fractions improved detection limits for 37 of the compounds to between 6 and 600pg/L in high-elevation surface waters.

The development, verification, and comparison study between LC-MS libraries for two manufacturers' instruments and a verified protocol are discussed. Compounds in the libraries are among those considered by the U.S. EPA Office of Water as threats to drinking water including pesticides, drugs of abuse, and pharmaceuticals. The LC-MS library protocol was verified through an inter-laboratory study that involved Federal, State, and private laboratories. The results demonstrated that the libraries are transferable between the same manufacturer's product line, and have applicability between manufacturers. Although ion abundance ratios within mass spectra were shown to be different between the manufacturers' instruments, the NIST search engine match probability was at 96% or greater for 64 out of 67 compounds evaluated.