Article pubs.acs.org/est
Quantitative Assessment of the Distribution of Dissolved Au, As and Sb in Groundwater Using the Diffusive Gradients in Thin Films Technique Andrew R. Lucas,*,† Nathan Reid,‡ S. Ursula Salmon,†,§ and Andrew W. Rate† †
School of Earth and Environment, University of Western Australia, 35 Stirling Highway, Crawley 6009, Western Australia, Australia CSIRO Mineral Resources Flagship, 26 Dick Perry Avenue, Kensington 6151, Western Australia, Australia § National Centre for Groundwater Research and Training, School of Earth and Environment, University of Western Australia, 35 Stirling Highway, Crawley 6009, Western Australia, Australia ‡
S Supporting Information *
ABSTRACT: The mobility of groundwater and its reactivity with subsurface lithologies makes it an ideal medium for investigating both the mineralogy of the extensive volume of the rocks and soils that it comes into contact with, including the distribution of potential commodities, and the presence of contaminants. Groundwater grab sampling is potentially an effective tool for evaluating metal and metalloid concentrations but can suffer from poor replication and high detection limits. This study evaluates the diffusive gradients in thin films (DGT) technique to detect signatures of Au mineralization in groundwater, as well as associated pathfinder and potential contaminant elements (As and Sb). The DGT technique was modified for Au by evaluating a “gel-less” configuration, with diffusion onto an activated carbon binding layer being controlled by the 0.13 mm thick filter membrane (0.45 μm porosity) only, in order to increase sensitivity in quiescent solutions. Laboratory-based measurements indicated that the diffusive boundary layer (DBL) was ∼0.40 mm in thickness in quiescent solutions. The modified DGT samplers were then deployed alongside ferrihydrite DGT devices (fitted with 0.8 mm diffusive gels) to simultaneously measure Au, As and Sb in groundwaters surrounding a known arsenopyrite-hosted Au ore body. DGT-measured Au concentrations ranged from 2.0 ng/L to 38.5 ng/L, and were within a factor of 5 of grab sample concentrations. DGT-measured concentrations of As and Sb were above the detection limits, while grab sample concentrations of As and Sb were often close to or below detection. The DGT technique demonstrated methodological improvement over grab sampling of groundwater for the investigated elements with respect to sensitivity, replication, and portability, although DGT requires further evaluation in a wider range of groundwater environments and conditions.
■
INTRODUCTION Hydrogeochemical sampling can be used to assess metal contamination of groundwaters and to determine the characteristics of regional lithologies, especially in areas where transported overburden conceals underlying geology.1−3 Geochemically, groundwater represents the extensive volume of the rocks and soils that it come into contact with, due to its mobility and reactivity with subsurface lithologies,2 and the movement of groundwater can redistribute elements including commodities such as Au and contaminants such as Asinto the surrounding rock as well as to the surface.4,5 Element mobility in groundwaters depends on a number of factors including pH and Eh,6 the speciation of elements,7 the presence of other dissolved species such as organics,8 the composition and reactivity of solid phases in contact with the solution,9 and microbial interactions.10 In arid environments where groundwater is near the surface, electrochemical processes that redistribute redox-sensitive elements may be critical.9 Studying these processes requires sampling techniques for detecting elements of interest at low concentrations. © 2014 American Chemical Society
In this study, a new analytical approach was evaluated to assess the distribution of metals in groundwater, including contamination from natural sources along with expressions of mineralization, and to examine the spatial patterns of groundwater concentrations of Au, As and Sb. The concentration of Au in groundwaters is typically below the detection limit of standard analytical methods, requiring preconcentration, for example, onto activated carbon11 or a Au-selective resin.12 This method of grab sampling of groundwaters can achieve detection limits as low as 5 ng/L or less for Au.11,13 Often, additional processing either in the field or in the laboratory is required in order to enhance sensitivity, examples of which include: the addition of a 5% Br2−HCl solution;11 the addition of aqua regia to bring the solution pH to 1.5;12 or the addition of NaCl.14 Problems Received: Revised: Accepted: Published: 12141
May 26, 2014 August 11, 2014 September 24, 2014 September 24, 2014 dx.doi.org/10.1021/es502468d | Environ. Sci. Technol. 2014, 48, 12141−12149
Environmental Science & Technology
Article
hydrogel (toward the edge of the binding gel) increases the effective area of the sampling window (c.f., eq 1) and cancels out the effect of the DBL.30,31 The sensitivity of the technique can be enhanced by prolonging the deployment period or reducing the thickness of the applied diffusion layer, thereby allowing more metal to accumulate onto the binding layer. In this study, the diffusive hydrogel was removed entirely; diffusion-controlled uptake by the binding gels was dependent only on the 0.13 mm-thick filter membranes (pore size 0.45 μm), placed directly onto the binding gels, and the DBL. Earlier studies of the performance of membrane-only DGT devices in well-stirred solutions showed that the diffusion coefficient through filter membranes is not statistically different from diffusion through a diffusive hydrogel.32,33 Laboratory evaluation was undertaken in this study to confirm whether the absence of the diffusive hydrogel affected DGT uptake. DGT devices fitted with carbon gels and filter membranes, with and without 0.8 mm diffusive hydrogels, were placed in 2 L solutions of 10 μg/L Au(III)Cl3(aq) for 4 h with stirring. In order to ensure the DGT cap was properly fitted in the configuration without the diffusive hydrogel, a 0.8 mm agarose gel was placed between the DGT piston and the binding gel. Values for DAu were obtained from Lucas et al.,22 and values for DAs and DSb were obtained from Luo et al.,26 with temperature, pH, and ionic strength accounted for. In the absence of the diffusive hydrogel, it was assumed that the DBL needed to be accounted for, as lateral diffusion in the DGT may be less than otherwise; under the well-stirred conditions in this experiment, the DBL thickness was assumed to be 0.23 mm.30 An unknown factor in using DGT in groundwater is whether conditions in the bore are relatively quiescent and therefore lead to an increase in the thickness of the DBL, which is reported to be as much as 1.5 mm in unstirred solutions.30 The efficacy of DGT in quiescent conditions for Co, Ni, Cu, Cd, and Pb has been demonstrated (with diffusive hydrogels present) when the increase in the DBL is accounted for.30 On the other hand, it has been found that even under extreme low flow conditions (
