Field-Based Application of Developed Solid Phase Extraction with Inductively Coupled Plasma Mass Spectrometry for Vanadium Speciation Analysis of Groundwaters from Argentina.

W. A. Al Rawahi,a,b and N. I. Warda

a ICP-MS Facility, Department of Chemistry, University of Surrey, Guildford, United Kingdom, GU2 7XH.

b Higher College of Technology, Applied Sciences Department, PO Box 74, Al-Khuwair, Postal Code 133, Al Khuwair Al Janubyyah St, Muscat, Oman.

Abstract

High levels of vanadium have been reported in groundwater (< 0.05 5300 g L-1 V) from different parts of Argentina, yet no detailed study of vanadium speciation has been performed. A highly selective strong anion exchange solid phase extraction (SAX-SPE) method was used (in-situ) for vanadium speciation analysis of groundwater samples from La Pampa - LP (General San Martin and Eduardo Castex) and Buenos Aires - BA (San German) provinces in Argentina. In this method both vanadyl (VIV) and vanadate (VV) were trapped by the complexation with disodium ethylenediaminetetraacetic acid on a pre-conditioned SAX cartridge. In the laboratory, vanadium species were separated at different eluent pH levels. VIV was eluted at pH 4 using methanol and tetrabutylammonium hydroxide. VV was eluted at pH 8 using dihydrogen ammonium phosphate. The eluted species were analysed by inductively coupled plasma mass spectrometry (ICP-MS). This method was validated using an inter-analytical method comparison with HPLC-ICP-MS. A Paired t-test revealed that there was no significant difference (probability, P < 0.05) between the two methods. VV was found to be predominate species in both sample collection areas (LP: 69 100%, BA: 33 89 % of species) over the range of 158.0 4748.0 g L-1 in LP and 88.5 504.0 g L-1 in BA. VIV was found at higher levels (29.0 - 301.0 g L-1) in Buenos Aires compared to General San Martin groundwater (4.4 161.0 g L-1). The results enhance the potential knowledge of the speciation of vanadium in terms of water quality and human health.

Keywords: vanadium, speciation, SPE, ICP-MS, groundwater, Argentina

Introduction

Vanadium is a transition element listed in the fourth row (VB group) of the periodic table with an atomic number 23; an electronic configuration of [Ar]3d34s2 and mass number of 51 amu [1]. The abundance of vanadium is 0.0136 % of the Earth crust which ranks it as the 5th among all transitional elements and the 20th most abundant element in the periodic table [2].

Vanadium can be found in different oxidation states from -1 to +5 [1]. Some of these forms are stable (3, 4 and 5) while others are not. For example; VII is unstable in the environment because it is easily oxidised to the more stable VIII species [3]. However, VIII can be gradually oxidised by air or dissolved oxygen to VIV [3, 4]. In water, VIV exists as vanadyl cations which are VO2+ and VO(OH)+ whilst VV is commonly found as vanadate oxyanions, namely, H2VO4- and HVO42- [5]. The stability of each species is dependent on the redox status of the water, e.g. VV is stable in an oxidising environment whilst VIV is more stable in a moderately reducing environment [6]. Oxidation from VIV to VV depends on several factors, such as pH, the redox potential, the concentration of vanadium, the ionic strength of the aquatic system and the dissolved oxygen, iron, ammonium and sulphide concentrations [1, 3]. The solubility of vanadium species increases with increasing valence [3]. For example, the concentration of VIV increases with a decrease in the total concentration of vanadium in a more reducing environment. However, in an oxidising environment the solubility of VO2+ increases as a result of complexation with organic matter [7].

The natural release of vanadium into the environment is through the weathering of soil, rocks, dust, marine aerosol and volcanic activities [8]. Vanadium is also released from anthropogenic activities, such as, the processing and burning of crude oil (Middle East, Venezuela and Canada, and only Japan and USA); the production of iron-steel, ceramics, pigments, batteries and mining activities, domestic waste recycling, fertilisers, pesticide application and through nuclear applications [1].

Vanadium speciation analysis of an aquatic environment is essential as the most predominant species in natural water (VIV and VV) have different roles from nutritional and toxicological viewpoints, in which VV is the most toxic form [8]. It has been reported that vanadium plays a role in regulating different enzymes, such as, Na+/K+ exchanging ATPase, phosphoryl-transfer enzymes, adenylate cyclase and protein kinases [9]. Vanadium has been shown to increase glucose transport and metabolism. As a result it has been used as an alternative treatment for diabetes [10, 11]. On the other hand, it has been reported that chronic exposure of mammals to high levels of vanadium can cause haemopoietin changes, such as, nephrotoxicity, reproductive and developmental toxicity [8, 10]. Vanadate, VV has been shown to be more toxic than vanadyl, VIV [8, 10]. Moreover, the inhalation and oral ingestion of vanadium pentoxide particulates can cause carcinogenic effects on humans [12]. Vanadium can also act as a genotoxin and can cause irritation of the respiratory tract [13]. Furthermore, another report has shown that some vanadium compounds hold properties that can prevent cancer [14]. From the above published studies it is clear that a better understanding of the vanadium therapeutic and toxic properties requires more investigation, especially linked with vanadium speciation analysis [1, 8].

Reported vanadium levels in surface and groundwaters range from undetectable to hundreds of g L-1 depending on the physiochemical parameters and geological features of the study area [15]. For this reason, different pre-concentration and separation techniques have been developed for vanadium total and speciation analysis. These include solid phase extraction (SPE) [4, 6]; capillary electrophoresis (CE) [16] and different methods of liquid chromatography [17]. Coupling of these separation techniques to sensitive and highly selective detectors, such as, inductively coupled plasma mass spectrometry (ICP-MS) [17], inductively coupled plasma atomic emission spectrometry (ICP-AES) [18], graphite furnace atomic absorption spectroscopy (GFAAS) [19], neutron activation analysis [20] and some UV-vis spectroscopy methods [21] have been used for the determination of vanadium in water samples [22]. Inductively coupled plasma mass spectrometry (ICP-MS) is a successful technique for the determination of trace elements at ultra-trace levels (typically < 0.1 g L-1) in water [23]. ICP-MS has low limits of detection, large elemental linear dynamic ranges (typically 6- 8 orders of magnitude) and is capable of being used for isotopic ratio measurements [17]. It should be noted that vanadium determination using ICP-MS does have some problems, such as polyatomic interferences (35Cl16O+,37Cl14N+, 40Ar11B+, 102Ru2+ and 34S16OH+) occurring at the same m/z as the most abundance isotope of vanadium 51V (99.75%). This can be corrected by using a reaction collision system (RCS) [24].

The determination of vanadium in water require pre-concentration and separation treatment, such as, solvent extraction [25], co-precipitation [26] and ion exchange [4]. In contrast to the above separation methods, solid phase extraction is simple to handle, has a greater level of efficiency and a higher enrichment factor which enables the differentiation between vanadium species [21]. Different types of vanadium pre-concentration and speciation methods using packed micro columns have been reported [4, 21]. Two different approaches were used to determine the main vanadium species; (i) VIV or VV whilst the other species are complexed or determined directly from the difference after calculating the total vanadium level [4, 27]; (ii) both vanadium species are eluted sequentially [6], and (iii) the use of two different columns to selectively adsorb and pre-concentrate each species [28].

Even though different papers have reported high levels of vanadium in groundwater over the range of < 0.05 to 5300 g L-1 V in different parts of Argentina, no detailed study of vanadium speciation of groundwater samples has been published [15, 29-31]. A few papers have presented data on vanadium species in surface and tap water samples from Argentina [32]. For example, a room temperature ionic liquid-based micro-extraction for the separation of vanadium species and determination by electrothermal atomic absorption spectrometry (ETAAS) has been reported [32]. This method was used for the vanadium speciation analysis of tap and river waters (n =10). The levels of VIV and VV were 0.38 2.00 g L-1 VIV and 0 3.12 g L-1 VV in the tap water and 0 1.25 g L-1 VIV and 0.98 2.84 g L-1 VIV in the river water. Another study reported the use of multi-walled carbon nanotubes for the on-line pre-concentration, speciation and determination of vanadium by ETAAS in tap and river water [33]. The reported levels of VIV and VV of three natural water samples were in the range of 0.13 1.32 g L-1 VIV and 0.26 1.45 g L-1 VV, respectively. In addition, the on-line determination of vanadium species in tap and Carolina river water, San Luis province, Argentina has been reported. The levels were 4.31 4.99 g L-1 V and 1.33 1.80 g L-1 V for tap and river water samples, respectively [13].

In this research, we report the development of a strong anion exchange solid phase extraction (SAX-SPE) method for vanadium speciation analysis of Argentine groundwater from various regions, namely, southeast La Pampa (General San Martin), central La Pampa (Eduardo Castex) and southwest Buenos Aires provinces (San German). Moreover, Ro Negro groundwaters were used as control samples. The vanadium speciation data will provide a better understanding of the relationship of vanadium and other elements (such as, As, Fe, Mn) in water and thereby help provide a better understanding of the effect of vanadium in relation to human or animal health.

Material and Methods

Instrumentation and quality control

Inductively coupled plasma mass spectrometry (ICP-MS; Agilent Technologies, 7700x, University of Surrey, Guildford, UK) was used for the analysis of trace elements in water samples. The Agilent 7700x instrument has a concentric mist nebuliser, a quartz Peltier-cooled torch and a Scott-type double-pass spray chamber. The removal of possible spectral polyatomic interferences (e.g. 35Cl16O+) was undertaken using a helium collision cell. The ICP-MS instrument was optimised daily using a 1 g L-1 tuning solution containing 7Li, 89Y, 140Ce and 205Tl in 2% HNO3 (Agilent Technologies, UK). A 100 g L-1 internal standard (IS) (BDH, Aristar, UK) in 1% HNO3 (Fisher Scientific, Trace Analysis Grade) of scandium was used to correct for any instrumental drift in signal intensity. The calibration curve was established over a range of 1 1500 g L-1 of 51V; prepared by appropriate dilution from stock standards of 1000 g L-1 (BDH, Aristar, UK) in 1% (v/v) HNO3 acid (Aristar, Fisher Scientific Ltd, UK). The elemental limits of detection (LOD) for the ICP-MS Agilent 7700x instrument were determined using a series of continuous runs (n = 20) of 1% v/v HNO3, trace metal grade, (Aristar, Fisher Scientific, UK). The LOD was calculated using the concentration of the blank added to three times the standard deviation of the blank solution (LOD: 0.05 g L-1 V). The level of accuracy was determined by replicate analysis (n = 4) of two water certified reference materials (CRMs), namely, the SRM 1640a (National Institute of Standards and Technology, Maryland, USA) and TMDA-54.2 (National Water Research Institute, Ontario, Canada) (NIST SRM 1640a: certified value 15.05 0.25 g L-1 V; against a measured value 14.7 0.1 g L-1 V; TMDA 542: certified value 349 25 g L-1 V, measured value 347.4 2.0 g L-1 V).

Vanadium speciation analysis using the developed pre-conditioned SPE method was validated by inter-analytical method comparison with high performance liquid chromatography - HPLC (1220 Infinity HPLC, Agilent Technologies, UK). The instrument components were a dual-channel gradient pump with degasser (G4299A, Agilent Technologies, UK), manual injector, column (G3154-65001, 4.6 mm x 150 mm i.d.) and a double-beam photometer detector (1024-element diode array detector, deuterium and tungsten lamps). The HPLC system was fitted with an anion-exchange column and a guard column (IEC, Agilent G3154-65001,65002, 4.6 mm x 10 mm i.d., Agilent Technologies, UK) that were used for the separation of the vanadium species in water samples. The data were collected using Lab Advisor Software (Agilent Technologies, UK). A Student t-test showed that there was no statistical difference between the total speciation analyses of both methodologies SAX-SPE-ICP-MS and HPLC-ICP-MS (tcalc = 1.37 < tcrit = 2.01 at a probability level of P 3 - 9 to form very stable complexes with EDTA in the forms of [VOY]2- for vanadyl (logKIV = 18.8) and [VO2Y]3- for vanadate (logKV = 15.5), in which Y is C10H14O8N2 [34]. The SPE cartridges (SAX-SPE) were pre-conditioned prior to sample collection to enhance the successful interaction between the species and the solid phase in the cartridge. This was undertaking by using 2 mL of 99.9% methanol, 2 mL of DDW, 10 mL of 0.2 M Na2EDTA and 20 mL DDW. After each step the solution was eluted completely by applying a syringe full of air. A volume of 1 - 2 mL of DDW was left to keep the cartridge moist until the sampling takes place. The pre-conditioning steps were done following that previously reported [4]; however, development of the V species elution and subsequent analysis steps were evaluated in this study.

Loaded volume

Initially, the pre-conditioned SAX - SPE cartridges were investigated for the ability to retain vanadium standards prepared in DDW. Vanadium levels in the non-retained fractions were all below the limit of detection of the vanadium species (< 0.2 g L-1 V), which shows the ability of the cartridge to adsorb both chemical forms. In addition, the amount of sample loaded onto the pre-conditioned cartridges were tested by loading different volumes and concentrations of the two vanadium species. Different volumes (10 30 mL) of VIV, VV and VIV + V (each of 500 g L-1) were prepared in 5 mM Na2EDTA (to ensure the stability of the species) and loaded onto the pre-conditioned SAX-SPE cartridges. The non-retained fraction was eluted from the cartridge and analysed by ICP-MS. For both vanadium species the loading of 10 mL of each standard to the cartridge was found to be the best volume so as to avoid any loss of vanadium species during sample collection (see Fig. 1S, supplementary material). In addition, increasing the volume can lead to over-loading of the cartridges (as undertaken by applying > 10 mL 500 g L-1 of both species). This can cause non-retention of the vanadium species.

Eluent selection

Vanadyl (VIV) was eluted using an aqueous solution (eluent 1) of 4% methanol, 5 mM Na2EDTA and 2 mM TBA+OH-. TBA+ is a strong ion pairing chemical that can interact with [VOY]2- (pKa = 3.22) at low pH which leads to less retention on the SPE surface [4, 17]. In addition, the presence of 5 mM EDTA (pKa1 = 1.99, pKa2 = 2.67, pKa3 = 6.16, pKa4 = 10.26) in eluent 1, can act as a displacing agent at low pH so as to elute VIV [17]. Two vanadyl (VIV) standards were prepared in DDW and 5 mM EDTA and eluted with eluent 1 at pH 3 or 4. It was found that the vanadium signal in the fractions with pH controlled at 4 were higher than the fraction at pH 3. Eluent 1 was then controlled to pH 4 to elute VIV from the SAX-SPE cartridge for all samples. The volume of eluent 1 that is required to elute the 10 mL of 500 g L-1 VIV that was originally loaded onto the cartridge was investigated. This was done by loading a volume of 10 mL of the prepared VIV standard onto five different SPE cartridges (n = 5). Then different volumes of eluent 1 (3.5 15 mL) at pH 4 were used to evaluate the elution of this species from the cartridges (Fig. 1a). A 100% recovery of VIV was achieved by using 10 15 mL of eluent 1.

Vanadate (VV) was eluted using an aqueous solution prepared from ammonium dihydrogen phosphate (NH4H2PO4) at pH 8. Phosphate ion with pKa1 = 2.12, pKa2 = 7.21 and pKa3 = 12.32 can act as a displacing agent of VV [V2OY]3 with pKa = 3.6 [17]. At pH 8 phosphate ions are converted to 2 charged anions that can displace the VV - EDTA complex. Different volumes (3.5 -15 mL) of NH4H2PO4 were used to elute 10 mL of 500 g L-1 VV loaded onto different SPE cartridges (n = 5) (Fig. 1b). A 100% recovery was achieved when 15 mL of NH4H2PO4 was used for elution of this species. In addition, different concentrations of NH4H2PO4 and vanadate (VV) were used to find the optimum levels to elute all of the vanadate from the SPE cartridge. This was investigated by loading 10 mL of 500 g L-1 VV onto different SPE cartridges (n = 5). In addition, 15 mL of NH4H2PO4 at different levels (30 80 mM) was examined in terms of the ability to elute vanadate from the SPE cartridge. A 100 % recovery was achieved using 80 mM NH4H2PO4 as the eluent to remove vanadate from the cartridge (Fig. 2S).

Optimised vanadium SPE methodology

Fig. 2 shows a schematic of the three fundamental steps for the developed SPE methodology, covering the conditioning, loading and elution stages. The cartridges (SAX-SPE) were pre-conditioned according to the developed conditions highlighted above prior to the collection of samples in Argentina. In addition, for field sampling the cartridge was connected to a 0.45 m filter (Millex - GP, Millipore, Hertfordshire, UK). Then, a known volume of water sample (10 mL) was loaded through a fully labelled SPE kit using a clean or rinsed disposable 20 mL syringe (BD Plastipak, Oxford, UK). The effluent was collected in 15 mL polypropylene centrifuge tubes (Sarstedt, Leicester, UK). After passing all of the total sample volume through the SAX cartridge, a syringe full of air was pushed through to remove all the water. The SPE kits were then stored for transportation to the University of Surrey.

In the laboratory the two vanadium species were recovered used the elution steps highlighted above. The three solutions, namely, the non-separated solution collected in the field (for total vanadium and other trace element analysis), and the vanadyl (VIV) and vanadate (VV) solutions were analysed by ICP-MS.

The effect of pH on vanadium speciation

The pH level is an important factor in order to study the stability and speciation of vanadium in water. According to the Eh-pH diagram of vanadium species in water [35], vanadate (VV) exists as an anionic species at pH levels 4 8. Furthermore, at low pH levels, vanadate exists as a relatively stable dioxo-VV ion (VO2+) [21]. In addition, vanadyl (VIV) is stable only at very low pH levels (typically 2) whilst it can be oxidised completely to VV at pH 5.5 6 [36]. Vanadium speciation analysis of water is a challenge especially due to the possible changes of the species resulting from a change in pH or the redox potential. To prevent conversion of VIV to VV as a result of the loss of carbon dioxide or through exposure to atmospheric oxygen, the SAX-SPE cartridges were conditioned with EDTA.

It was deemed important to undertake a study to evaluate the stability of the vanadium species throughout the preparation stages using DDW or 5 mM EDTA. In addition, this study also looked at the effect of pH change on the stability of the vanadium species. A standard of 400 g L-1 VIV was prepared in DDW. Then the pH level was changed to cover the range of 3 8 (reflecting the range reported for natural waters). For each pH level, 10 mL of standard solution was loaded onto the pre-conditioned SAX-SPE cartridge (n = 4). Then the vanadium species were eluted following the procedures in Fig. 2 and each fraction was analysed using ICP-MS. Fig. 3 shows the conversion of vanadyl into vanadate at pH levels between 5 and 6. This is in agreement with reported data [4]. The same experiment was repeated using separate solutions of 400 g L-1 VIV and 250 g L-1 VV prepared in 5 mM EDTA at the various pH levels over the range of 3 - 8. The two species prepared in 5 mM Na2EDTA were found to be stable in relation to an increase in pH (Fig. 3b). This confirms an important feature of the developed methodology using EDTA as it is vital to preserve the species in the pre-conditioned cartridges from the time of field sampling until recovery and analysis in the laboratory (which may be weeks or months a part).

Method validation using an interanalytical method comparison with HPLC

A certified reference material for vanadium species in water was not found. The validation of the vanadium SAX-SPE method was done by an inter-analytical comparison using HPLC-UV and SPE-ICP-MS. Different vanadium standards (500 2000 g L-1) containing both vanadium species (VIV +V) were prepared in 5 mM EDTA and loaded onto a pre-conditioned SAX-SPE cartridge. Then the standard solution (before loading) and the eluted fractions were analysed by both techniques. The SAX-SPE method was evaluated by running a total vanadium standard (1000 g L-1 VIV +V) onto an anion exchange HPLC column (Fig. 4a, n = 3). The HPLC chromatogram shows the presence of both vanadium species; the vanadyl (VIV) peak at a retention time (tR) of 9.8 minutes and the vanadate (VV) peak at tR = 17.1 minutes. The peak at tR = 5.3 minutes is for the blank (5 mM Na2EDTA). Moreover, after eluting VIV from the SAX-SPE cartridge using eluent 1 the eluted fraction was analysed by HPLC as shown in Fig. 4b. The VV peak appeared at tR = 17.0 minutes and no other peaks were observed after the Na2EDTA peak (Fig. 4c). This confirms the elution of only vanadate from the SAX-SPE cartridge. The above experiment was repeated with different concentrations of the two species and a mixture of both (VIV+V) (n = 3). The vanadium levels for the standard solution and eluted fractions were analysed using HPLC and ICP-MS. The percentage recovery of the vanadium using SPE- HPLC (95.5 100.4 %) was in very close agreement with the data using SPE-ICP-MS (94.5 101.2 %). High levels of vanadium were used > 250 g L-1 V in this experiment as the HPLC has a higher limit of detection (100 g L-1) and therefore normal vanadium levels would result in the non-detection of the vanadium peak in the HPLC chromatogram.

Stability study using natural water samples from Argentina

Solid phase extraction is a superior method for conserving elemental species. A field-based experiment in Argentina was conducted to investigate the stability of the vanadium species. This important experiment was deemed necessary to evaluate whether any possible changes occur in the levels of the two main vanadium species between sample collection, transportation and analysis at the laboratory. Six samples were loaded onto the cartridges at the time of sample collection in Argentina. In addition, the samples from the sample site were collected into 50 mL non-acidified polypropylene tubes without loading them onto SPE cartridges (non-treated) and transported to the laboratory with the loaded SPE cartridge samples. The non-treated water samples were loaded onto the pre-conditioned cartridges as soon as they reached the laboratory (a typical time period of 15 days). Table 1S shows the levels of the eluted vanadyl (VIV) and vanadate (VV) species associated with (1) field sampling, and (2) in the laboratory. This data clearly shows that there is a transformation of the vanadyl species during the transportation of the samples to the laboratory. On the other hand, there is an increase in the levels of the vanadate species in all the fractions eluted after the return to the laboratory. This confirms that the conversion of vanadyl into vanadate occurs between sample collection and laboratory analysis. This may be due to possible changes in the physiochemical properties of the sample during this period. This is a very important findings as many published studies assume the stabiliity of elemental species in collected samples, which may only be analysed some weeks or months after collection using HPLC-ICP-MS [37].

Vanadium Speciation of Natural Waters from Argentina

This study presents for the first time the determination of vanadium species (VIV and VV) separated in-situ in Argentina using the developed SPE cartridge methodology. The water samples were collected from three different study areas: Ro Negro, southeast La Pampa (General San Martin), central La Pampa (Eduardo Castex) and southwest Buenos Aires (San German) provinces in Argentina.

Ro Negro

The developed vanadium speciation method was applied to water samples collected from the Ro Negro province (control site) which was selected based on previously published data of low levels of vanadium (typically 0.72 28.20 g L-1) [31]. In order to be able to detect such levels 20 mL of sample was loaded onto the SPE cartridges for water samples from this province. Table 1 reports the species levels of vanadium for twenty ground and surface waters. The total vanadium levels were over a range of 0.9 to 113.0 g L-1 V. The predominant vanadium species for all water samples was vanadate (0.9 101.0 g L-1 VV) whilst the range of the vanadyl, VIV species was 0.4 18.2 g L-1 VIV.

Southeast La Pampa (General San Martin, GSM)

The vanadium levels in groundwater samples (n = 11) collected from General San Martin in the southeast of La Pampa province ranged over 211.0 to 1192.0 g L-1 V and in relation to speciation analysis were largely found to contain the vanadate species (Table 1). The levels were found to range over 158.0 1128.0 g L-1, contributing to 69 -100% of the total vanadium level. The vanadyl levels ranged from 4.4 to 161.0 g L-1, equating to 0 31 % of the total vanadium level. The pH of these samples was found to be over 7.0 8.2 which confirms the predicted pH-Eh predominance of VV at pH levels close to neutral [37].

Central La Pampa (Eduardo Castex, EC)

Nineteen groundwaters were collected from wells in Eduardo Castex in the central region of the La Pampa province. The total vanadium levels ranged over 243.0 to 4889.0 g L-1 V. Table 1 shows that both species are at higher levels compared to that for water samples from the southeast of La Pampa (General San Martin) and southwest Buenos Aires (San German) provinces. The results revealed that the predominant species was vanadate. This suggests that the groundwater system is predominantly oxidising (Eh range 3 - 41 mV) and under alkaline pH (7.5 9.4), thereby promoting the oxidation of VIV to VV. The vanadyl levels ranged from 12.8 581.0 g L-1 and for vanadate 225.0 to 4748.0 g L-1.

Southwest Buenos Aires province (San German, BA)

Vanadium speciation analysis was applied to 24 groundwater samples collected from the community of San German in southwestern Buenos Aires province. The total vanadium levels ranged over 182 to 592 g L-1 V. Vanadate (VV) was found to be the predominant species (88.5 - 504.4 g L-1), contributing to 33 - 89 %. Vanadyl (VIV) was found at higher levels (range 29.0 - 301.0 g L-1). The proportion of both species may be explained by the higher levels of total dissolved solids or TDS (553 1365 mg L-1), higher redox potential, Eh range (62 146 mV) and higher conductivity (1072 -2706 S cm-1) of the water samples from this area compared with those from the southeast of La Pampa Eh (40 118 mV) and TDS (284 388 mg L-1) [38].

Therefore, in summary vanadate (VV) was found to be the predominant species in all samples from La Pampa. In addition, vanadate was the dominant species in 92 % of the Buenos Aires groundwater samples. Research studies have shown that vandate is the most preominant species in the oxidation environment [3, 8, 39]. Moreover, the solubilty of vanadate was noted to increase by forming complexes with dissolved organic matter [40]. However, it has been shown that these complexes will be reduced to vanadyl species at pH < 6 [41]. Moreover, this species is found at low pH and moderately reducing environment [3, 8, 39]. The solubility of the species can be enhanced by the the prescence of dissolved organic matter in alkaline and oxic conditions [37, 39]. In this study, the pH levels were between ~ 7 to > 9 in which no adsorption of vanadium to organic matter should take place [39]. Moreover, Ro Negro samples were mostly of higher redox potential compared to the other provinces in which vanadate is more stable and the dominant species. On the other hand, La Pampa and Buenos Aires provinces showed lower redox potentials with a range between - 62 and 146 mV. This can explain the presence of both species in these provinces.

Conclusions

A solid phase extraction method based on using strong anionic exchange cartridges (SAX-SPE) for the speciation analysis of vanadium in water has been presented. The method was validated by an inter-analytical method comparison with HPLC. To the best of our knowledge, the developed SAX-SPE method was applied for the first time to the speciation analysis of Argentine ground and surface waters (southeast La Pampa, central La Pampa (Eduardo Castex) and southwest Buenos Aires province). This study has focused on vanadium levels in water as in these regions of Argentina there are numerous health problems, some relating to arsenic [42], but in part may be due to the high levels of vanadium in local drinking waters. Moreover, for the first time vanadium speciation data is available based on the application of the developed field-based SPE methodology. In addition, a major feacture of this research is the stability of the vanadium species from the point of sampling to laboratory analysis.Therefore, it is now possible to undertake further studies linking the water quality of vanadium in this region and other countries, with human and animal health.

Acknowledgements

The authors wish to thank the Ministry of Manpower, Sultanate of Oman for the financial support. In addition, a special thanks to Dr. Dan Driscoll at the University of Surrey for his support during HPLC training.

References

[1] M. Imtiaz, M.S. Rizwan, S. Xiong, H. Li, M. Ashraf, S.M. Shahzad, M. Shahzad, M. Rizwan, S. Tu, Vanadium, recent advancements and research prospects: a review, Environ Int. 80 (2015) 79-88.

[2] F.A.C. Amorim, B. Welz, A.C.S. Costa, F.G. Lepri, M.G.R. Vale, S.L.C. Ferreira, Determination of vanadium in petroleum and petroleum products using atomic spectrometric techniques, Talanta 72(2) (2007) 349-359.

[3] O. Pourret, A. Dia, G. Gruau, M. Davranche, M. Bouhnik-Le Coz, Assessment of vanadium distribution in shallow groundwaters, Chem. Geol. 294295(0) (2012) 89-102.

[4] L. Minelli, E. Veschetti, S. Giammanco, G. Mancini, M. Ottaviani, Vanadium in Italian waters monitoring and speciation of V(IV) and V(V), Microchem. J. 67 (2000) 83-90.

[5] R.B. Wanty, M.B. Goldhaber, Thermodynamics and kinetics of reactions involving vanadium in natural systems: Accumulation of vanadium in sedimentary rocks, Geochim. Cosmochim. Acta 56(4) (1992) 1471-1483.

[6] D. Wang, S.A. Saudo-Wilhelmy, Development of an analytical protocol for the determination of V (IV) and V (V) in seawater: Application to coastal environments, Mar. Chem. 112(12) (2008) 72-80.

[7] X.J. Zhou, C. Wei, M.T. Li, S.A. Qiu, X.B. Li, Thermodynamics of vanadium-sulfur-water systems at 298 K, Hydrometallurgy 106(1-2) (2011) 104-112.

[8] J. Costa Pessoa, Thirty years through vanadium chemistry, J. Inorg. Biochem. 147 (2015) 4-24.

[9] M. Costigan, R. Cary, S. Dobson, Concise International Chemical Assessment Document 29 Vanadium pentoxide and other inorganic vanadium compounds, WHO, Geneva (2010).

[10] WHO, Trace elements in human nutrition and health, (1996).

[11] A.S. Tracy, G.R. Willsky, E.S. Takeuchi, Vanadium : Chemistry, Biochemistry, Pharmacology, and Practical Applications, First Edition ed., CRC Press Taylor & Frsncis Group, LLC, United States of America, 2007.

[12] P. Leopardi, P. Villani, E. Cordelli, E. Siniscalchi, E. Veschetti, R. Crebelli, Assessment of the in vivo genotoxicity of vanadate: Analysis of micronuclei and DNA damage induced in mice by oral exposure, Toxicol. Lett. 158(1) (2005) 39-49.

[13] M. Costigan, R. Cary, S. Dobson, Vanadium Pentoxide and Other Inorganic Vanadium Compounds, (2001).

[14] A. Bishayee, A. Waghray, M.A. Patel, M. Chatterjee, Vanadium in the detection, prevention and treatment of cancer: The in vivo evidence, Cancer Lett. 294(1) (2010) 1-12.

[15] M.L. Puntoriero, A.V. Volpedo, A. Fernndez Cirelli, Arsenic, Fluoride, and Vanadium in surface water (Chasic Lake, Argentina), Front. Environ. Sci. 2 (2014) 23.

[16] Z.L. Chen, G. Owens, Trends in speciation analysis of vanadium in environmental samples and biological fluidsa review, Anal. Chim. Acta 607(1) (2008) 1-14.

[17] N. Kilibarda, S.E. Afton, J.M. Harrington, F. Yan, K.E. Levine, Rapid speciation and determination of vanadium compounds using ion-pair reversed-phase ultra-high-performance liquid chromatography inductively coupled plasma-sector field mass spectrometry, J. Chromatogr. A 1304(0) (2013) 121-126.

[18] P. Roccaro, F.G. Vagliasindi, Coprecipitation of vanadium with iron (III) in drinking water: a pilot-scale study, Desalination Water Treat (ahead-of-print) (2014) 1-11.

[19] S.K. Wadhwa, M. Tuzen, T. Gul Kazi, M. Soylak, Graphite furnace atomic absorption spectrometric detection of vanadium in water and food samples after solid phase extraction on multiwalled carbon nanotubes, Talanta 116(0) (2013) 205-209.

[20] R. Greenberg, H. Kingston, Trace element analysis of natural water samples by neutron activation analysis with chelating resin, Anal. Chem. 55(7) (1983) 1160-1165.

[21] K. Pyrzyska, T. Wierzbicki, Determination of vanadium species in environmental samples, Talanta 64(4) (2004) 823-829.

[22] R. Cornelis, J.A. Caruso, H. Crews, K.G. Heumann, Handbook of Elemental Speciation, Handbook of Elemental Speciation II: Species in the Environment, Food, Medicine and Occupational Health, John Wiley & Sons2005.

[23] M. Pfeiffer, G. Batbayar, J. Hofmann, K. Siegfried, D. Karthe, S. Hahn-Tomer, Investigating arsenic (As) occurrence and sources in ground, surface, waste and drinking water in northern Mongolia, Environ. Earth Sci. 73(2) (2015) 649-662.

[24] S. DIlio, N. Violante, C. Majorani, F. Petrucci, Dynamic reaction cell ICP-MS for determination of total As, Cr, Se and V in complex matrices: Still a challenge? A review, Anal. Chim. Acta 698(1) (2011) 6-13.

[25] L.J. Lozano, D. Juan, Solvent extraction of polyvanadates from sulphate solutions by Primene 81R. Its application to the recovery of vanadium from spent sulphuric acid catalysts leaching solutions, Solvent Extr. Ion Exch. 19(4) (2001) 659-676.

[26] S. Nakano, S. Kinoshita, M. Ikuta, T. Kawashima, Fractional determination of nanogram amounts of vanadium (IV) and vanadium (V) in natural water by a catalytic method combined with solvent extraction, Anal. Sci. 6(3) (1990) 435-438.

[27] K. Pyrzyska, T. Wierzbicki, Pre-concentration and separation of vanadium on Amberlite IRA-904 resin functionalized with porphyrin ligands, Anal. Chim. Acta 540(1) (2005) 91-94.

[28] K. Okamura, M. Sugiyama, H. Obata, M. Maruo, E. Nakayama, H. Karatani, Automated determination of vanadium(IV) and (V) in natural waters based on chelating resin separation and catalytic detection with Bindschedlers green leuco base, Anal. Chim. Acta 443(1) (2001) 143-151.

[29] M.E. Espsito, J.D. Paoloni, M.E. Sequeira, N.M. Amiotti, M.d.C. Blanco, Natural Contaminants in drinking waters (arsenic, boron, fluorine and vanadium) in the southern Pampean Plain, Argentina, J Environ Prot 2 (2011) 97-108.

[30] H.B. Nicolli, J. Bundschuh, M.d.C. Blanco, O.C. Tujchneider, H.O. Panarello, C. Dapea, J.E. Rusansky, Arsenic and associated trace-elements in groundwater from the Chaco-Pampean plain, Argentina: Results from 100 years of research, Sci. Total Environ. 429(0) (2012) 36-56.

[31] H.R. Farnfield, Arsenic (Total and Speciation) in Water from Argentina and its Impact on Human Health, Faculty of Engineering and Physical Sciences, University of Surrey, Guildford, GU2 7XH, 2012.

[32] P. Berton, E.M. Martinis, L.D. Martinez, R.G. Wuilloud, Room temperature ionic liquid-based microextraction for vanadium species separation and determination in water samples by electrothermal atomic absorption spectrometry, Analytica Chimica Acta 640(12) (2009) 40-46.

[33] R.A. Gil, S.N. Goyanes, G. Polla, P. Smichowski, R.A. Olsina, L.D. Martinez, Application of multi-walled carbon nanotubes as substrate for the on-line preconcentration, speciation and determination of vanadium by ETAAS, J. Anal. At. Spectrom. 22(10) (2007) 1290-1295.

[34] Z. Chen, M.M. Rahman, R. Naidu, Speciation of vanadium by anion-exchange chromatography with inductively coupled plasma mass spectrometry and confirmation of vanadium complex formation using electrospray mass spectrometry, J. Anal. At. Spectrom. 22(7) (2007) 811-816.

[35] B. Wehrli, W. Stumm, Vanadyl in natural waters: Adsorption and hydrolysis promote oxygenation, Geochim. Cosmochim. Acta 53(1) (1989) 69-77.

[36] P.P. Coetzee, Vanadium Speciation by Chromatographic Separation of V(IV) and V(V) in Acidic solution Followed by ICP-OES Determination, Water SA, 2007, pp. 291-291.

[37] F. Aureli, S. Ciardullo, M. Pagano, A. Raggi, F. Cubadda, Speciation of vanadium(iv) and (v) in mineral water by anion exchange liquid chromatography-inductively coupled plasma mass spectrometry after EDTA complexation, J. Anal. At. Spectrom. 23(7) (2008) 1009-1016.

[38] D. Wang, S.A. Saudo Wilhelmy, Vanadium speciation and cycling in coastal waters, Mar. Chem. 117(1-4) (2009) 52-58.

[39] M.T. Wright, K.G. Stollenwerk, K. Belitz, Assessing the solubility controls on vanadium in groundwater, northeastern San Joaquin Valley, CA, Appl. Geochem. 48(0) (2014) 41-52.

[40] W. Stumm, Chemistry of the Solid-Water Interface: Processes at the Mineral-Water and Particle-Water Interface in Natural Systems, John Wiley and Son Inc.1992.

[41] X. Lu, W.D. Johnson, J. Hook, Reaction of Vanadate with Aquatic Humic Substances: An ESR and 51V NMR Study, Environ. Sci. Technol. 32(15) (1998) 2257-2263.

[42] M. Gimnez, R. Osicka, P. Blanes, Y. Morisio, S. Faras, Heavy metal concentration in arsenic contaminated groundwater of the Chaco Province, Argentina, One Century of the Discovery of Arsenicosis in Latin America (1914-2014) As2014: Proceedings of the 5th International Congress on Arsenic in the Environment, May 11-16, 2014, Buenos Aires, Argentina, CRC Press, 2014, p. 135.