Lab Science News

Ultra-trace speciation of arsenic and other metals is performed using a variety of techniques tailored to the specific combination of species, matrix, and detection limits required. Currently, two analytical systems are applied: Ion Chromatography-Inductively Coupled Mass Spectrometry (IC-ICP-MS) and Hydride Generation-Cryogenic Gas Chromatography-Flame Atomic Absorption Spectrometry (HG-CGC-AAS). Each of these techniques has particular strengths that can be exploited depending upon the scientific question being asked. These techniques are combined with specific extraction techniques in order to maximize speciation integrity and data quality.

Ion Chromatography-Inductively Coupled Mass Spectrometry. IC-ICP-MS is a more generalized analytical technique allowing, through optimization of the ion chromatography parameters, the separation of a wide range of inorganic and organic arsenic species in aqueous, geological, and biological media. This technique allows quantification of As(III), As(V), CH₃As²⁺, (CH₃)₂As⁺ (cacodyllic acid), as well as the major forms of arsenic in aquatic organisms: arsenobetaine and arsenocholine at levels less than 0.2 µg/L (ppb) in aqueous solution. After applying gentle, ultra-clean extraction techniques (typically mixtures of aqueous phosphate buffers and methanol) to sediments, soils, or biota, detection limits below 2 ng/g in most solids may be obtained and even lower in biological fluids. Many other arsenic compounds are accessible to analysis by this technique, with the limitation generally being the difficulty of obtaining standards and/or the chemical stability of the compounds. Currently, optimization of this technique has begun for several arsenic compounds of interest to the environmental and food sciences community, including the poultry additives 4-aminobenzene arsenic acid (Roxarsone™) and 4-hydroxy 3-nitrobenzene arsenic acid (para-arsenillic acid), and a natural arsenic metabolic intermediate, tetramethyl arsonium. Additional species, including natural inorganic arsenosulfur compounds and arsenosugars, are also analytes quantifiable by this technique.

Hydride Generation-Cryogenic Gas Chromatography-Atomic Absorption Spectrometry. When lower detection limits are required, HG-CGC-AAS is the method of choice, as sample volumes of up to 100 mL can be concentrated prior to analysis, and high levels of dissolved salts are no problem. Although limited to those species that are inherently volatile or form volatile hydrides, detection limits of <2 ng/L (ppt) are achievable for As(III), total inorganic arsenic (TIAs), CH₃As²⁺, (CH₃)₂As⁺, and (CH₃)₃As (trimethyl arsine, which is difficult to separate at all using IC-ICP-MS). This method is likely most applicable to the low-level quantification of the war gas, Lewisite (2-chloro-ethynl dichloro arsine), which is also inherently volatile at room temperature. Although As(V) is not directly measured using HG-CGC-AAS, it is unambiguously determined as the difference between TIAs and As(III). Thus, it can generally be well quantified in samples where the As(V):As(III) ratio is 0.1 or higher. - Detection limits down to 0.4 ng/g (ppb) are quantifiable in biological and geological solids. In addition, through the use of ultra-clean UV photo-oxidation, you can use this method to determine total arsenic in solution at levels 1-2 orders of magnitude lower than can be obtained using ICP-MS, and with very few interferences.

Speciation of Geological Solids. A variety of sequential selective extraction (SSE) techniques are routinely applied to sediments, soils, and air particulate matter to help further the available speciation information. Comparison of field samples to an array of pure arsenic minerals suspended in kaolin, SSE combined with the speciation methods described above can provide additional insights into solid phase speciation and mineralogy of As in samples such as mine tailings and fly ash. Through a number of collaborative analytical research laboratories throughout the world, more specific solid phase arsenic speciation is available using synchrotron techniques such as EXAFS and XANES, as well as electron microprobe spectrometry (EMPS).

Sampling and Preservation for Arsenic Speciation. As with all trace metals speciation problems, the accurate assessment of arsenic speciation is critically dependant upon the use of the appropriate sampling and preservation procedures. In general, it is important to avoid the use of glass containers (unless inherently volatile species such as (CH₃)₃Hg are being sought), minimize or eliminate contact of the sample with oxygen and other oxidizers, and analyze the sample or preserve it using low-temperature freezing as soon as possible after sample collection. In general, soil and biota samples should be placed in a to-fit trace-clean plastic container and be frozen until analysis. Aqueous samples are best if they are quick frozen in liquid nitrogen or dry-ice plus methanol in appropriate-sized trace-metal clean polyethylene containers provided by our laboratory, and then stored at below -80 ^oC until thawed for analysis. Otherwise, it is best to send most water samples by overnight delivery, unpreserved in completely filled trace-clean plastic containers. Never use containers that have been cleaned with HNO₃ or BrCl, or preserve samples with these reagents. Oxic aqueous samples (surface waters) may be preserved for As(III), As(V), CH₃As²⁺ and (CH₃)₂As⁺ using 0.2% chlorine-free HCl (pre-purged with inert gas) if they will be analyzed using HG-CGC-AAS. Samples of sediment, sediment pore water, groundwater, and/or anoxic water samples have very specific sampling and preservation requirements that should be discussed prior to attempting. It is often best for speciation data quality for your laboratory to provide sampling equipment and personnel for these specialized procedures.

Introduction: Historical mining practices in the Coeur d’Alene River Basin (Idaho) have resulted in heavy metal contamination of soil, sediment, surface water, and groundwater. Canyon Creek, located in the upper basin, has elevated levels of dissolved zinc (average concentration ~ 3,000 μg/L), dissolved cadmium (average concentration ~ 22 μg/ L), and total lead (average concentration ~ 174 μg/L). Heavy metal loading near the mouth of Canyon Creek is influenced by surface water/groundwater interactions. Dissolved zinc concentrations in the groundwater have been detected in the 100,000 μg/L range while dissolved cadmium and lead have been detected in the hundreds to thousands μg/L ranges, respectively.

EPA’s consultant, URS Corporation (URS), developed a multi-phase treatability study to obtain quantitative information on a treatment process to effectively remove metals from the water of Canyon Creek. The treatment process incorporated different combinations of pH adjustment, chemical coagulation and coprecipitation, polymer flocculent additions, and additions of ballasted micro-sand to improve sludge settling. The results of the study will be used to help evaluate potential treatment technologies for surface water and/or groundwater at Canyon Creek. These data will also be used to help develop the pilotscale treatability study for Phase II of the study.

URS contracted Columbia Analytical Services, Inc.’s (CAS) Redding Lab to perform an in-depth, bench-scale treatability study. CAS performed the study in accordance with the procedure described in Canyon Creek Treatability Study Plan, Coeur d’Alene Basin, RAC, EPA Region 10, April 2004 and its Appendix A, Laboratory Scope of Work (SOW).

Treatability Study: The study involved performing a seven step process (see Table 1) involving sixteen jar and nineteen settling rate tests. Varying combinations of lime stabilization (adjusting the pH up to 11) with a calcium hydroxide slurry and coagulation / coprecipitation with ferric chloride followed by polymer flocculation (anion, cationic, and nonionic forms) and micro-sand additions (used as a flocculation aid) were used. The jar and settling rate tests were performed using a 6-place paddle stirrer or jar tester.

The bulk surface water feedstock collected for the treatability study was slightly acidic with a pH of 5.0 and low in conductivity (80 umhos/cm) and hardness (20 mg/L as CaCO3). The bulk sample was received in a Teflon lined 55-gallon drum by over-night courier. To help maintain sample stability the bulk sample was stored in a large walk-in cooler at 4°C for the duration of the study. Sub-samples (each approximately 8-L in volume) were taken for each jar or settling test by mixing the contents of the bulk sample using an acid washed, plastic paddle and then filling a 20-L cubitainer with the sample using an allplastic siphon pump. In addition to the treatability study, CAS analyzed 9 surface water and 4 groundwater samples for ICPMS and ICP metals, general chemistry, physical analyses, and neutralization curve analyses to evaluate site characteristics at the time of feedstock water collection.

Over two hundred samples were generated by the treatment study. CAS immediately analyzed the samples for pH, conductivity, turbidity, and zinc by the Zincon colorimetric procedure. The Zincon method was used so CAS and URS could quickly determine which combination of reagents provided the best reduction in zinc in order to quickly select the optimum combination of reagents for the next jar test. Selected samples were also analyzed for cadmium, lead, and zinc by ICPMS, calcium and magnesium by ICP, and a variety of general chemistry and physical measurements. After Step 3, all zinc analyses were performed by ICPMS because the treatment process was reducing the zinc concentration to below the Zincon detection limit of 20 μg/L.

Study Results: Overall the project was very complicated as each jar test required different combinations of lime, ferric chloride, and different types of flocculating agents depending on the results from the previous jar test. Due to the unknown stability of the water matrix CAS personnel worked through the initial weekend. The testing was completed over a period of 14 days. The bulk sample proved to be stable over this period based on daily analyses of dissolved metals concentrations and conventional parameters.

URS and CAS personnel worked closely with daily communications to insure a smooth progression and successful project outcome. Frequent communications between URS and CAS were essential, as minor modifications to the work-plan were required as the study progressed.

The first phase of the Canyon Creek Treatability Study provided quantitative information on the effectiveness of a variety of treatment approaches for removing dissolved metals from surface water collected from the mouth of Canyon Creek. The treatment process, using the optimum combinations of reagents, resulted in significant reduction of metals plus the formation of a sludge that has very good settling rates and ease of filtering (see Table 2).

- - -

Learn more about testing for mines…

- - -

Mercury is responsible for over three-quarters of all contaminant-related advisories for threats to human health. During the 1990’s, the number of mercury related fish consumption advisories more than doubled, despite significant decreases in the total mercury emissions over the last 20 years. The increase in advisories is probably the result of more testing rather than more contamination.

While the contamination is showing up in lakes and fish, most mercury does not come from effluent, rather is derived from atmospheric deposition. Atmospheric transport and subsequent bioaccumulation of mercury can affect aquatic ecosystems far from mercury sources. According to EPA estimates, emissions from coal-fired utilities account for 13 to 26 percent of the total (natural plus anthropogenic) airborne emissions of mercury in the United States. Thus, the EPA has begun to regulate emissions from power plant boilers and process heaters.

The impending MACT (Maximum Achievable Control Technologies) regulation (due out in 2004) is prompting affected manufacturers to assess the chloride, mercury and selected metals content of their fuels. The best way to meet the regulations is to burn fuels that are low in chloride, mercury and other metals. Early testing shows that mercury is more of a problem than chloride, particulate matter or selected metals.

Method Summary
Low level mercury measurements are conducted by EPA Method 1631. The analytical technique is very sensitive. In this method, an aqueous sample is oxidized with bromine monochloride and sparged with nitrogen onto a gold trap. The mercury is thermally desorbed from the gold trap into a cold vapor atomic fluorescence spectrometer. CAS can achieve a method detection limit (MDL) of 0.06 ng/L (ppt), which is three orders of magnitude less than the conventional cold vapor mercury method.

While the original method was designed for aqueous samples, CAS has implemented the “Appendix to Method 1631: Total Mercury in Tissue, Sludge, Sediment and Soil by Acid Digestion and BrCl Oxidation.” CAS has achieved an MDL of 0.3 ug/Kg in solid samples.

Sampling
The clean sampling techniques described in EPA Method 1669 should be used for sampling. Since most mercury contamination comes from the atmosphere, it is very easy to contaminate water samples. Other sources of contamination are metal containers, talc powdered gloves, improperly cleaned and stored equipment, and dust and dirt. By following these clean sampling procedures, it has been shown that much of the historical data for mercury in seawater was erroneously high because of contamination from sampling.

When it is necessary to measure dissolved mercury, field filtering with an in-line filter can be performed as long as care is taken to insure that the filter is clean and free of contamination. It may be more efficient to send an unpreserved sample to the laboratory for filtering under clean conditions. This is particularly true if an in-line filtering device is not available.

Quality Assurance
Quality assurance is performed at CAS in accordance with the EPA method, ensuring scientifically valid and legally defensible results. Matrix spikes should be at the compliance level of 1 to 5 times the background level, so it is helpful to know the compliance level or the approximate amount of mercury expected in the samples. Besides matrix spikes, the method calls for measuring standards at 5 ng/L from two sources and analyzing many different blanks to assess potential contamination originating either in the field or in the laboratory. The blanks include: equipment, field, bottle, bubbler and reagent blanks, in addition to the usual method blank.

Why choose CAS?
The CAS Kelso laboratory has been performing this method for over six years in our clean laboratory. In addition to a variety of water matrices, CAS has experience analyzing hundreds of samples on several types of solid matrices: fish, coal, oil, sawdust and bark. We have a documented history of freedom from contamination and excellent recovery of ongoing precision and recovery standards and MS/MSDs within the QC acceptance criteria of Method 1631.

References:
Federal Register, Vol. 68, No. 8, January 13, 2003, pg 1666.
Krabbenhoft & Schuster, USGS Fact Sheet FS-051-02, June 2002.

Columbia Analytical Services’ Kelso laboratory has continued the development of analytical techniques for pre-concentration and chemical separation of trace metals in saline aqueous samples using reductive precipitation followed by inductively coupled plasma-mass spectroscopy (ICP-MS). This technique replaces conventional techniques for the analysis of trace metals in complex aqueous matrices such as sea water where matrix problems prevent the ability to achieve desired low levels of detection. Detection limits for various metals of interest using these techniques are displayed in the table to the right. Details of this procedure were presented at the recent National Environmental Monitoring Conference (NEMC) held July 21-24 in Arlington, Virginia. The full abstract follows.

A procedure for analyzing a relatively wide range of trace metals in samples containing elevated levels of dissolved solids is discussed. The procedure incorporates a chemical separation to remove interfering matrix components so final analysis can be performed using inductively coupled plasma-mass spectroscopy (ICP-MS). The separation utilizes reduction of certain target analytes to the elemental state and precipitation of others as the boride, depending on reduction potentials and/or boride solubility. The precipitation is facilitated using elemental palladium plus iron boride as carriers. Once separated from the seawater matrix, the precipitate is dissolved and analyzed using ICP-MS. A number of modifications to the the procedure have been made over the past ten years to improve performance. The method meets general U.S.E.P.A. performance criteria. A general outline of the procedure is included in the most recent version of EPA Method 1640. However, variations are presented that allow a wider range of elements to be tested. Additional modification eliminates filtration and filter manipulation, which is a source of significant potential contamination. Considerations are given to the introduction of excessive chloride via the acid mixture used for dissolution of the precipitate. Thus, arsenic and chromium are validated as part of the multi-element suite of target analytes. Recovery data for low level determinations is reported and demonstrates elements suitable for this procedure, as well as elements that do not conform to the reaction mechanism(s). Detection limits are presented that show this technique is a viable approach for many elements when a procedure is needed to measure trace metal concentrations at or near ambient levels in various sample types, including open-ocean seawater. The procedure has also been adapted to the analysis of industrial chemicals such as concentrated sodium hydroxide used by municipal drinking water suppliers for pH adjustment.