Grantee Research Project Results
Final Report: Development of Chemical Methods to Assess the Bioavailability of Arsenic in Contaminated Media
EPA Grant Number: R825410Title: Development of Chemical Methods to Assess the Bioavailability of Arsenic in Contaminated Media
Investigators: Basta, Nicholas T. , Rodriguez, Robin R. , Casteel, Stan W.
Institution: Oklahoma State University , University of Missouri - Columbia
EPA Project Officer: Hahn, Intaek
Project Period: November 1, 1996 through October 31, 1999 (Extended to October 31, 2000)
Project Amount: $431,677
RFA: Environmental Fate and Treatment of Toxics and Hazardous Wastes (1996) RFA Text | Recipients Lists
Research Category: Hazardous Waste/Remediation , Land and Waste Management , Safer Chemicals
Objective:
The objectives of the proposed research were to provide answers to the following questions:1. Can chemical fractionation and in vitro gastrointestinal (IVG) laboratory methods provide accurate assessments of bioavailable arsenic in contaminated media?
2. Can an IVG method that simulates gastrointestinal (GI) absorption provide a more accurate estimate of arsenic bioavailability than an IVG method that does not simulate GI absorption?
3. Is arsenic bioavailability a function of arsenic concentration where soil may serve as a sink and decrease arsenic bioavailability?
To answer these questions, arsenic measured by chemical extraction methods (soil chemical extraction methods based on fractionation and in vitro gastrointestinal methods) were compared with arsenic uptake by immature pigs for contaminated media (soil and slag) collected from smelter sites.
Summary/Accomplishments (Outputs/Outcomes):
Soil ingestion from incidental hand-to-mouth activity by children is an important issue in assessing public health risks associated with exposure to arsenic-contaminated soils and media. Risk from enteric bioavailability of arsenic is difficult to assess because arsenic exists in many geochemical forms (e.g., oxides, sulfides) and physical forms (e.g., flue dust, slag, tailings, waste ore) at hazardous waste sites contaminated by mining or smelting of ore. Dosing studies using animal models, such as immature swine, have been used to determine contaminant (i.e., Pb, As) bioavailability in contaminated soil and media. However, several disadvantages in conducting animal studies include expense, specialized facilities and personnel requirements, and time with several months required to measure contaminant bioavailability. Chemical methods (soil chemical fractionation, in vitro gastrointestinal) that provide a reasonable estimate of the bioavailable arsenic in contaminated media may be capable of providing rapid and inexpensive information needed to characterize risk at Superfund sites.The two in vitro gastrointestinal methods evaluated were the in vitro gastrointestinal method [with (IVG-AB) and without (IVG) adsorbent] developed in this study (Rodriguez, et al., 1999) and the published Physiologically Based Extraction Test or PBET method (Ruby, et al., 1996). In vitro results were compared with in vivo relative bioavailable arsenic determined from dosing trials using immature swine. Fifteen contaminated soils collected from mining/smelter sites ranging from 401 to 17,500 mg As kg-1 were used in this study. In the IVG and PBET methods, arsenic is sequentially extracted from contaminated soil with simulated gastric and intestinal solutions. Extracted arsenic was measured by inductively coupled plasma atomic emission spectroscopy with introduction by using hydride generation (ICP-HG). Only the IVG in vitro methods (gastric, intestinal, and IVG-AB intestinal) were equivalent with the in vivo method (P < 0.05) across all media. Evaluating the contaminated media separately, the noncalcinated waste materials (i.e., slags and soils) tested by the IVG methods were statistically equivalent to the in vivo method. However, the in vitro methods underestimated bioavailable arsenic in calcine materials. However, when only the noncalcinated materials were evaluated (all media except the calcines), the close agreement between the IVG and PBET gastric methods and in vivo bioavailable arsenic was obtained. The IVG methods, but not the PBET method, provided accurate estimates of bioavailable arsenic across all materials. The IVG methods and the PBET gastric phase were accurate estimators of bioavailable arsenic for noncalcinated slags and soils. Few statistical differences between the IVG stomach phase, IVG intestinal phase, and IVG-AB intestinal phase were found for most groups of material. In other words, extending the in vitro method beyond the gastric phase did not improve the ability of the method to measure bioavailable arsenic. Results of the IVG stomach phase were linearly correlated r = 0.92) with in vivo bioavailable arsenic (P < 0.01). Statistical analysis of linear regression parameters showed the slope of 0.89 was not different than one and the intercept of -3.08 was not different than zero; IVG stomach phase results were statistically the same as in vivo results. The IVG intestinal phase also was linearly correlated with in vivo arsenic with an r value of 0.91 (P < 0.01). Statistical analysis of linear regression parameters showed that the slope of 0.77 was not different than one and the intercept of ?2.42 was not different than zero; IVG intestinal phase results were statistically the same as in vivo results. The PBET stomach phase results were linearly correlated with in vivo arsenic (P < 0.05), but the PBET intestinal phase was more strongly correlated with an r of 0.88 (P < 0.01). Slopes of the PBET vs in vivo method were less than one showing that the PBET method underestimated bioavailable arsenic.
Results from five soil chemical extraction methods were compared with in vivo relative bioavailable arsenic determined from dosing trials using immature swine to answer the first question and objective of the research study: "to determine the degree of correlation between soil chemical extraction methods and in vivo methods and the ability of soil chemical extraction methods to estimate bioavailable arsenic in contaminated media." The same 15 arsenic-contaminated soils evaluated by the in vitro methods were extracted with the following five soil chemical extraction methods: (1) extraction with deionized water to measure water soluble arsenic; (2) extraction with 1 M sodium acetate (pH 5.0) to measure weakly adsorbed and acid labile forms of arsenic; (3) extraction with 1 M sodium phosphate (pH 7.0) to measure water soluble and adsorbed arsenic; (4) extraction with acidified hydroxylamine hydrochloride to measure arsenic in amorphous Fe and Mn oxide; and (5) extraction with acidic ammonium oxalate to measure arsenic in amorphous and crystalline Fe and some Al oxides.
The mean and median percent (mean, median) of total arsenic extracted from soil by soil chemical extractants followed the trend ammonium oxalate (58.7 percent, 66.0 percent) ? hydroxylamine hydrochloride (53.7 percent, 56.2 percent) > phosphate (10.5 percent, 5.39 percent), acetate (7.16 percent, 6.88 percent) > water (0.15 percent, 0.03 percent). These results suggest most of the arsenic in these soils is in the occluded form and can only be released when Fe oxides and other minerals that occlude the arsenic are dissolved by the chemical extractant (i.e., oxalate and hydroxylamine hydrochloride). Much lesser amounts of surficially adsorbed arsenic were extracted by phosphate and acetate than reagents that dissolved mineral forms occluding arsenic (oxalate, hydroxylamine hydrochloride). Chemical extraction results showed the amount of arsenic in chemical pools were occluded > surficially adsorbed > water soluble.
Three chemical extractant methods that showed significant linear regression coefficients, r values, were water (r = 0.68, P < 0.01), acetate (r = 0.57, P < 0.05), and hydroxylamine hydrochloride (r = 0.88, P < 0.01). The strongest relationship between arsenic measured by a chemical extraction method and in vivo bioavailable arsenic was found for hydroxylamine hydrochloride. None of the chemical extraction methods extracted the same amount of bioavailable arsenic measured by immature swine (P < 0.05). The chemical extractants that dissolve surficially complexed arsenic (phosphate, acetate) extracted less arsenic than the bioavailable arsenic fraction. Hydroxylamine hydrochloride and ammonium oxalate extracted occluded and surficial arsenic, which was greater than in vivo bioavailable arsenic. Apparently, the fraction of arsenic that is bioavailable in contaminated soils and solid wastes is comprised of arsenic fractions between the surficially complexed (desorbable) arsenic and occluded and strongly adsorbed arsenic associated with iron oxides. Because the stomach environment is acidic (pH 1.8) and anaerobic, it is likely that most of the arsenic surficially adsorbed and some of the arsenic occluded in iron oxide are dissolved. The amounts of extracted arsenic between surficially adsorbed (phosphate or acetate) and surficially adsorbed plus occluded (oxalate, hydroxylamine hydrochloride) in iron and manganese oxide fractions may be similar to the chemical forms of arsenic dissolved in the stomach.
We incorporated an adsorbent in a permeable membrane to act as a sink for dissolved arsenic "to determine if in vitro methods that mimic gastrointestinal absorption are more accurate than current in vitro methods that do not address gastrointestinal absorption to assess the bioavailability of arsenic in contaminated media," the second question posed and objective of this study. The in vitro procedure with absorption simulation (IVG-AB) is the same as the IVG method described above with the exception of adding freshly prepared amorphous iron hydroxide gel during the intestinal phase as an adsorbent. The total amount of arsenic measured by the IVG-AB method is the summed mass of arsenic in the intestinal phase solution and the arsenic dissolved from the iron hydroxide gel.
A strong linear relationship (r = 0.88, P < 0.01) was found between arsenic measured by the IVG-AB method and in vivo arsenic determined from immature swine dosing trials. However, incorporation of the adsorbing Fe gel to the in vitro solution of the IVG-AB method had little effect on the improvement of r value of 0.91 from the linear regression of the IVG method without adsorbent vs. in vivo bioavailable arsenic. Statistical analysis of linear regression parameters showed that the slope of 0.79 was not different than one and the intercept of ?1.34 was not different than zero; IVG-AB intestinal phase results were statistically the same as in vivo results. In other words, it was difficult and unnecessary to improve upon the strong correlation between IVG arsenic without adsorbent and in vivo arsenic by adding adsorbent to the intestinal phase. Comparison of mean relative percent arsenic of 17.3 percent extracted by IVG intestinal phase and mean relative percent arsenic of 18.3 percent extracted by IVG-AB methods suggests small but not significant (P < 0.05) increases in arsenic extracted by the adsorbent in the IVG-AB method. Results show that dissolution of arsenic solid phases in studied contaminated media by simulated gastrointestinal solutions appears to be the rate-limiting step rather than the subsequent Fe gel adsorption step in controlling dissolved arsenic in the IVG method. Therefore, adding a step to the IVG method that mimics gastrointestinal absorption does not increase the accuracy of the IVG method to assess the bioavailability of arsenic in contaminated media.
Immature swine were dosed with the same weight of five contaminated media that varied in arsenic concentration to "determine if the bioavailability of arsenic in contaminated media is a function of arsenic concentration (arsenic loading rate)," the third question posed and objective of this study. Two types of contaminated media investigated were calcine waste material and slag waste material. Arsenic content of the five calcine wastes were 6250, 11300, 11500, 13500, and 17500 mg kg-1. Arsenic content of the five slag waste materials, a combination of weathered slag and soil, were 405, 450, 1180, 4650, and 5020 mg kg-1. Immature swine were dosed with 100 mgL/d of contaminated media. Relative bioavailability of arsenic in contaminated media was determined by urine analysis.
Relative bioavailable arsenic ranged from 4.08 to 19.7 percent for calcinated waste and from 21.9 to 43.0 percent for slag waste. Bioavailable arsenic was not constant for a contaminated media (same matrix). Simple correlation coefficients (r values) of 0.69 and 0.47 showed poor correlation between total arsenic content and RBA arsenic. These results reject the hypothesis that arsenic bioavailability will increase with arsenic content.
The hypothesis of Objective 3 is based on increased solubility of arsenic with increased "loading" of contaminated material with arsenic. Adsorption of arsenic to contaminant surfaces over a range of surface coverage would be consistent with increasing solubility as a function of arsenic loading. The results from this study suggest chemical adsorption processes do not control arsenic solubility in the gastrointestinal tract or bioavailable arsenic in the studied contaminated media. Most arsenic in the calcine materials is associated with Fe oxides (63 to 86 percent with a mean value of 75 percent) or as an arsenic jarosite analog. Arsenic in the three analyzed slag materials was associated with Fe oxides (36 to 72 percent with a mean value of 57 percent) and Pb oxides (18 to 49 percent with mean value of 33 percent). Results suggest arsenic bioavailability in slag and maybe calcine materials are controlled by mineral dissolution and not desorption processes required to test the hypothesis of Objective 3.
In summary, results from this research project provided the following answers to the three questions posed and hypotheses studied. The IVG in vitro gastrointestinal methods provided an accurate assessment of the bioavailability of arsenic in contaminated media. It was not necessary to simulate gastrointestinal absorption processes to improve the IVG method. Measuring arsenic dissolved in the gastric phase by the IVG method provided a very accurate estimate of bioavailable arsenic. It was unnecessary to simulate intestinal conditions to obtain an accurate estimate of available arsenic by the IVG method.
Soil extraction using hydroxylamine hydrochloride, as modified by Amacher and Kotuby-Amacher (1994), was correlated with bioavailable arsenic. However, this soil extraction method overestimated bioavailable arsenic. Refinement of extracting conditions (extractant concentration, time, pH) is necessary for this extractant to provide an accurate estimate of bioavailable arsenic. Arsenic bioavailability was not a function of arsenic concentration (arsenic loading rate) in the contaminated media studied. However, this result may be a function of the media selected for study. Arsenic bioavailability may be a function of total arsenic content when adsorption processes control arsenic solubility. It is unlikely for slag wastes and unclear for calcine wastes that adsorption processes controlled solubility of arsenic. Further studies that use uncontaminated materials and controlled arsenic additions are necessary to test the hypothesis of Objective 3.
The IVG method is an inexpensive method to obtain site-specific arsenic availability information, thereby lowering the degree of uncertainty in risk assessment. The IVG method can provide scientifically derived data at these sites to aid in the selection of appropriate remedies that are cost-effective and protective of human health and the environment. However, the IVG may not provide highly accurate estimates of arsenic for all contaminated media. The human digestive system is too complex and dynamic to simulate in the laboratory. A more reasonable approach may be by the use of the IVG and other in vitro methods as screening tools to provide rapid estimates of arsenic bioavailability at a contaminated site. An estimate of variability of arsenic bioavailability in many samples collected from heterogeneous contaminated sites can be obtained rapidly and inexpensively by the IVG method. Bioavailable "hot spots" can readily be identified and aid in the design and cost effectiveness of remedial strategies at contaminated sites.
References:
Amacher MC, Kotuby-Amacher J. Selective extraction of arsenic from mine spoils, soils, and sediments. American Society of Agronomy, p. 256 (abstract).
Rodriguez RR, Basta NT, Casteel SW, Pace LW. An in vitro gastrointestinal method to assess bioavailable arsenic in contaminated soils and solid media. Environmental Science and Technology 1999;33(4):642-649.
Ruby MV, Davis A, Schoof R, Eberle S, Sellstone CM. Estimation of lead and arsenic bioavailability using a physiologically based extraction test. Environmental Science and Technology 1996;30:422-430.
Journal Articles:
No journal articles submitted with this report: View all 20 publications for this projectSupplemental Keywords:
soil, sediments, absorption, exposure, risk, risk assessment, human health, bioavailability, dose-response, carcinogen, children, chemicals, toxics, heavy metals, remediation, cleanup, environmental chemistry, biology, analytical, measurement methods, cleanup levels, remediation endpoints, site-specific bioavailability, oral bioavailability, health effects., RFA, Health, Scientific Discipline, Waste, Water, Ecosystem Protection/Environmental Exposure & Risk, Bioavailability, Contaminated Sediments, Environmental Chemistry, Chemistry, Epidemiology, Arsenic, Risk Assessments, Fate & Transport, Susceptibility/Sensitive Population/Genetic Susceptibility, Children's Health, genetic susceptability, fate and transport, health effects, risk assessment, fate, contaminated mines, ingestion, soil ingestion, Superfund sites, contaminant transport, risk characterization, soil sediment, human health effects, contaminated sediment, dermal contact, sediment transport, transport contaminants, chemical speciation, adverse human health affects, chemical contaminants, kinetic studies, soils, hazardous waste, children, toxicity, human exposure, superfund site, chronic health effects, environmental toxicant, mobility, arsenic mobility, animal research, ecology assessment models, arsenic exposure, hazardous waste sites, exposure assessmentRelevant Websites:
http://www.agr.okstate.edu/soilchem
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.