ARSENIC CARBONATE COMPLEXES IN GROUNDWATER

J.S. Lee, J.O. Nriagu*. Department of Environmental Health Sciences. School of Public Health, University of Michigan, Ann Arbor, Michigan 48109.

* jnriagu@umich.edu

 

ABSTRACT

 

Exploratory research involving new arsenic carbonate complexes has been conducted, and results suggest the likelihood of their existence. Attempt was made to form the complexes by bubbling CO2 gas in arsenic trioxide solution for 24 hours. Ion chromatography of the solution showed shifts and peak broadening among the As(III) peak and the emergence of new peaks not found in solutions containing only CO2 or As(III) in Milli-Q water. Corresponding peaks obtained from atomic absorption spectroscopy showed similar elution times and increased peak areas. Likely arsenic carbonate complexes are believed to be As(CO3)2- , As(OH)( CO3)22-, As(CO3)2(OH)2-,  and As(CO3)+.  The implications of arsenic carbonate complexes on arsenic chemistry and in natural water ecosystems remain to be evaluated.

 

INTRODUCTION

 

Many communities around the world are exposed to elevated levels of arsenic in their drinking water. Exposure to arsenic is associated with cancer of the skin, bladder, liver, lung, and kidney, and hyperpigmentation and hyperkeratosis. Groundwaters that are mostly reducing and contain high concentrations of dissolved iron and manganese are associated with elevated levels of arsenic. Anoxic wells in Bangladesh show arsenic concentrations that correlate with concentrations of dissolved iron and bicarbonate, as well as increasing arsenic concentrations with increasing depth of wells (Nickson 1997).

It has been reported that arsenic in groundwater evolves from the oxidation of arsenic-rich pyrite during the water extraction process (Das et al and Saha et al 1995), but results from a previous study suggest the carbonation of arsenite minerals to be an important process in leaching arsenic into groundwater under anaerobic conditions (Kim 1999).

However, no known abiotic mechanism exists for the abstraction of arsenic from the aquifer rocks under reducing conditions. Kim has postulated that the formation of stable arseno-carbonate complexes are responsible for leaching arsenic from host rocks under anaerobic conditions.  

The notion that carbonate forms some kind of complex with arsenite was the basis of the old pharmaceutical practice Fowler’s solution, which is an aqueous solution of As2O3 in K2CO3 (Schaufelberger 1994).  Pentavalent arsenic can also form complexes with carbonate, as can be seen when CO2 increases the solubility of hydrometallurgical arsenate precipitates (Gonzalez and Monhemius 1988). Since carbonate exists naturally in the environment and is commonly found in groundwater, it is possible for arsenic carbonate complexes to form in groundwaters that contain arsenic. However, as Schaufelberger points out, apart from a note in Ephraim in 1920 on the rhodanide-arsenite complexes, the existence of intermediate arsenic carbonate complexes has never been thoroughly investigated nor reported.

 

METHODS

 

            Arsenic carbonate complexes were prepared by continuous bubbling of arsenic trioxide solution with CO2 for 24 hours. As(V) solution was prepared by dissolving sodium metaaarsenate in deoxygenated water, and used to check the quality of As(III) containing solutions. Arsenic trioxide and sodium metaarsenate were obtained from Aldrich and all solutions were prepared in Milli-Q water. Solutions were analyzed by Alltech’s Odyssey High Performance Ion Chromatography (IC) System using a Waters IC-PAC Anion Column (4.6 x 150mm) and a Wescan Cation Column (4.6 x 50mm) to detect negatively and positively charged species respectively. Ion chromatograph for anions was performed under the following conditions: mobile phase 0.85mM NaHCO3/0.9 mM Na2CO3, flow rate 1.4ml/min, column temperature 25°C, and injection volume 0.5ml. The following conditions applied to cation analysis: mobile phase 1.25mM Ethylenediamine/1.67mM Tartaric Acid, flow rate 0.4ml/min, column temperature 25°C, and injection volume 0.5ml. 

Total arsenic was analyzed using a Perkin Elmer Zeeman 1100ZL atomic absorption spectrometer equipped with a graphite furnace.  

 

RESULTS AND DISCUSSION

 

Figure 1 shows the chromatogram obtained with an anion column for (a) arsenic trioxide solution saturated with CO2, (b) arsenic trioxide solution with no CO2, (c) Milli-Q water saturated with CO2, and (d) As(V) solution. Figure 1 shows the arsenic trioxide solution saturated with CO2 to contain distinct peaks not found in the other solutions. The two solutions containing As(III) show three identical peaks with elution times of 4-10 minutes. The presence of impurities found in Milli-Q water and/or As2O3 could account for these minor peaks. For example, when the IC traces for these As(III) containing solutions were compared to an IC trace for the anion standard, one of the small peaks corresponded to chlorine (Cl-). It is common to find trace amounts of chlorine in highly purified water and chemicals. The same reasoning can be applied to explain the occurrence of three similar peaks also found in the As(V) solution. The two peaks with elution time of 10-12 minutes found in the As(III) solutions can be attributed to H3AsO3 since there is no equivalent peak in the As(V) solution, and H3AsO3 is expected to dominate when the pH is less than 8. The three distinct peaks with elution times of 16-25 minutes are only found in the CO2-saturated As(III) solution, and we are tentatively attributing these peaks to the following species: As(CO3)2- , As(OH)( CO3)22-, and As(CO3)2(OH)2-. The two peaks with elution times greater than 30 minutes are found in both CO2-saturated solutions, and are most likely not related to the formation of arsenic complexes.

            Since the cation column was shorter in length compared to the anion column, the arsenic was eluted in less than five minutes and isolation of arsenic peaks from other peaks was difficult. Therefore, eluents from the IC were collected sequentially and analyzed for total arsenic using GFAAS. Figure 2 shows the cation analysis results for As(III) solutions with and without CO2. The As(III) solution saturated with CO2 shows one peak with a short retention time and another peak with an elution time of about 3 minutes (Figure 2). The first peak is most likely H3AsO3 because this species is neutral

 

Figure 1. Chromatogram from IC showing peaks for As(V) solution, As2O3 solution saturated with CO2, Milli-Q water saturated with CO2, and As2O3 solution without CO2.

 

and should elute out rapidly, and the second peak is attributed to As(CO3)+. We observed that the presence of CO2 significantly increases the concentration of arsenic in solution. After bubbling the arsenic trioxide solution for 24 hours with CO2, the peak area for As(III) increased from 0.917 to 2.896 (Figure 3). Enhanced solubility of As2O3 can be explained by the formation of stable arseno-complexes.

 


Figure 2. Concentrations of arsenic in sequentially collected samples from an IC equipped with a cation column. Graphs for As2O3 solutions saturated with CO2 and without CO2 are compared. 

 

 



Figure 3. As(III) concentrations for As2O3 solutions saturated with and without CO2 for 24 hours.

 


The preliminary studies show the formation of stable arseno-carbonate complexes in solutions containing As(III) and carbonate ions. Current speciation methods are based on ion exchange methods that separate arsenic species into a neutral fraction (As(III)) and a negatively charged fraction (As(V)). This method would not be adequate for  groundwaters that have high carbonate concentrations. Since these arseno-carbonate complexes are believed to be fairly stable, they may actually contribute significantly to the amounts of arsenic found in well water. The presence of charged arsenic carbonate complexes also challenges the current belief that As(III) diffuses passively into cells in its neutral form (H3AsO3).

                  

References

 

1.      Das, D., Basu, G., Chowdhury, T.R. & Chakraborty, D. (1995) in Proc. Int. Conf. Arsenic in Groundwater (Calcutta).

2.      Gonzalez, VLE., & Monhemius, AJ. Arsenic Metallurgy, Fundamentals and Applications. Reddy, RG, Hendrix, JL, and Queneau, PB. (ed.), Metallurgical Society, PA, 1988.    

3.      Kim, MJ (1999). Arsenic Distribution and Speciation in Groundwater of Southeast Michigan. PhD Thesis, University of Michigan.

4.      Nickson, R. Thesis, University college Londin (1997).

5.      Saha, A.K. & Chakrabarti, C. (1995) in Proc. Int. Conf. Arsenic in Groundwater (Calcutta).

6.      Schaufelberger, FA. Arsenic in the Environment Part I: Cycling and Characterization. Nriagu, JO. (ed.), John Wiley & Sons, Inc., NY, 1994.