Available online at www.sciencedirect.

com
JOURNAL OF
ENVIRONMENTAL
SCIENCES
ISSN 1001-0742
CN 11-2629/X
Journal of Environmental Sciences 2012, 24(7) 1341–1346
www.jesc.ac.cn

A novel colorimetric method for field arsenic speciation analysis

Shan Hu1 , Jinsuo Lu2 , Chuanyong Jing1,∗

1. State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences,
Chinese Academy of Sciences, Beijing 100085, China. E-mail: [email protected]
2. School of Environmental and Municipal Engineering, Xi’an University of Architecture and Technology, Xi’an 710055, China

Received 06 September 2011; revised 22 November 2011; accepted 24 November 2011

Abstract
Accurate on-site determination of arsenic (As) concentration as well as its speciation presents a great environmental challenge especially
to developing countries. To meet the need of routine field monitoring, we developed a rapid colorimetric method with a wide dynamic
detection range and high precision. The novel application of KMnO4 and CH4 N2 S as effective As(III) oxidant and As(V) reductant,
respectively, in the formation of molybdenum blue complexes enabled the differentiation of As(III) and As(V). The detection limit of
the method was 8 μg/L with a linear range (R2 = 0.998) of four orders of magnitude in total As concentrations. The As speciation in
groundwater samples determined with the colorimetric method in the field were consistent with the results using the high performance
liquid chromatography atomic fluorescence spectrometry, as evidenced by a linear correlation in paired analysis with a slope of 0.9990–
0.9997 (p < 0.0001, n = 28). The recovery of 96%–116% for total As, 85%–122% for As(III), and 88%–127% for As(V) were achieved
for groundwater samples with a total As concentration range 100–800 μg/L. The colorimetric result showed that 3.61 g/L As(III) existed
as the only As species in a real industrial wastewater, which was in good agreement with the HPLC-AFS result of 3.56 g/L As(III).
No interference with the color development was observed in the presence of sulfate, phosphate, silicate, humic acid, and heavy metals
from complex water matrix. This accurate, sensitive, and easy-to-use method is especially suitable for field As determination.

Key words: arsenic speciation; colorimetric method; groundwater; field analysis

DOI: 10.1016/S1001-0742(11)60922-4

Introduction formance liquid chromatography (HPLC) with an adequate

element-specific detector such as mass spectrometry (MS)
Naturally occurring arsenic (As) has been promulgated and atomic fluorescence spectrometry (AFS). However,
with the greatest exposure in developing countries in- these instruments are not designed for field application. It
cluding Bangladesh, India, Vietnam, and China where is an urgent need to develop effective in situ speciation
groundwater As concentrations are at as high as sever- method with affordable equipments for routine environ-
al mg/L level (Nordstrom, 2002). Until 2010, over 100 mental monitoring (Liao and Deng, 2006).
million people are still drinking untreated groundwater The colorimetric As analysis has become the most
with As concentrations far exceeding the World Health practical technique for field application. One of the widely
Organization (WHO) guideline of 0.13 μmol/L (10 μg/L) used colorimetric As analyses is based on the Gutzeit’s
(Fendorf et al., 2010) . The toxicity and mobility of As method. A recent development of the Gutzeit’s method is
depend on its oxidation state where arsenite (As(III)) is 50- the As field testing kits. However, the kits failed in a high
100 times more toxic than arsenate (As(V)). Nevertheless, fraction (up to 68%) when analyzing groundwater samples
As(III) is the dominant As species in most groundwater (Arora et al., 2009; Rahman et al., 2002). In addition,
samples and can be readily oxidized to As(V) upon expo- the Gutzeit’s method generates highly toxic arsine gas
sure to air (Meng et al., 2003). Thus, the As(III) oxidation that requires specific protective equipments to practice this
reactions may happen during the sample collection, stor- assay.
age, and transferring to the laboratory, and could lead to The As molybdenum blue method, another commonly
biased conclusions about the toxicity. Therefore, accurate used field As analysis technique, is based on the fact
rapid in situ speciation analysis is of great importance. that As(V) can form molybdenum blue complexes with
Powerful methods of detecting low concentrations of reduced molybdate while As(III) could not (Johnson,
As(III/V) are generally established with sophisticated 1971; Johnson and Pilson, 1972). Researchers optimized
laboratory instrumentation. The most popular speciation the method to reach a detection limit of 7 μg/L with a
analysis combines an effective separation using high per- mean recovery of total As of 97% ± 5% in a total As range
10–1000 μg/L (Dhar et al., 2004). However, the reducing
* Corresponding author. E-mail: [email protected]
1342 Journal of Environmental Sciences 2012, 24(7) 1341–1346 / Shan Hu et al. Vol. 24

reagent used in previous studies (14% Na2 S2 O5 + 1.4% 4°C until analysis.
Na2 S2 O3 + 10% H2 SO4 ) generates highly toxic SO2 gas The industrial wastewater sample was obtained from a
(Johnson, 1971; Johnson and Pilson, 1972; Dhar et al., copper smelting company in China. The wastewater was
2004). Tsang et al. (2007) attempted to modify the re- highly contaminated with 3.56 g/L As(III), 24 mg/L Cu, 5
ducing reagent and used sodium dithionite to reduce 7–75 mg/L Pb, 369 mg/L Cd, and 0.28 mol/L SO4 2− at pH 1.4
μg/L As(V) in neutral samples. Nevertheless, this narrow (Luo et al., 2010).
linear range and complication in operation limit its field
1.2 Procedure of As colorimetric method
application. The other limitation in previous molybdenum
blue method is the usage of KIO3 as oxidant which could Briefly, 30 mL samples were acidified with 1% HCl and 10
inhibit molybdate reduction (Dhar et al., 2004). μmol/L phosphate. Three parallel 10 mL sub-samples were
The objective of this study was to develop a rapid treated with 1 mL of an oxidizing reagent (Oxidized sub-
colorimetric method for routine analysis of As specia- sample), a reducing reagent (Reduced sub-sample), and
tion in environmental samples. In the present report, we deionized (DI) water (Untreated sub-sample), respectively.
compared and proposed the oxidants and reductants based After 30 min, 1 mL color reagent was added which is
on the molybdenum blue complexes to differentiate As a mixture of 10.8% C6 H8 O6 , 3% (NH4 )6 Mo7 O24 ·4H2 O,
species. To explore the possible interference from the com- 0.56% C8 H4 K2 O12 Sb2 ·3H2 O, and 13.98% H2 SO4 in
plex matrix of environmental samples to the colorimetric a volume ratio 2:2:1:5. The absorbance was measured
method, natural groundwater and industrial wastewater after 5 min using a portable single-beam Hach DR2800
samples were analyzed and the results were compared with spectrophotometer at 880 nm. The As(III) and As(V)
the high-performance liquid chromatography coupled with concentrations were calculated using following equations:
atomic fluorescence spectrometry (HPLC-AFS). The pro-
posed method with a portable spectrophotometer provides Total As = Asoxidized – Asreduced
a reliable and affordable approach for in situ determination
of As concentration and speciation. As(III) = Asoxidized – Asuntreated

1 Materials and methods As(V) = Asuntreated – Asreduced

1.1 Materials 1.3 Selection of oxidant

All reagents were of analytical reagent grade and used Three oxidizing reagents (NaClO, KMnO4 and KIO3 ) were
without further purification. Potassium iodate (KIO3 , selected as potential oxidant because they do not interfere
>99.0%), potassium permanganate (KMnO4 , >99.5%), the process of formation of molybdenum blue complexes
sodium hypochlorite (NaClO, >9%), monopotassium (Lenoble et al., 2003). To compare the oxidation kinet-
phosphate (KH2 PO4 , >99.5%), sodium thiosulfate ics using KMnO4 , NaClO, and KIO3 , experiments were
(Na2 SO3 , >97.0%), sodium pyrosulfite (Na2 S2 O5 , carried out by reacting 1 L of 500 μg/L As(III) with
>96.0%), hydrochloric acid (HCl, >36%), and sulphuric 0.3 mmol/L oxidant. The solution was mixed vigorously
acid (H2 SO4 , >95.0%) were obtained from Beijing and samples were taken at designated time intervals for
Chemicals Corporation (China); sodium arsenate As speciation analysis. As(III) and As(V) were separated
(Na3 AsO4 ·12H2 O, >99.0%), sodium arsenite (NaAsO2 , immediately after collection using As speciation cartridges
>99.8%), ascorbic acid (C6 H8 O6 , >99.7%), ammonium as described in our previous report (Jing et al., 2003). The
molybdate ((NH4 )6 Mo7 O24 ·4H2 O, >99.0%), antimony cartridges selectively removed As(V) but did not adsorb
potassium tartrate (C4 H4 O6 KSbO·1/2H2 O, >99.0%), As(III). The As concentration was then determined using
potassium iodide (KI, >99.0%), hydroxylamine a graphite furnace atomic absorption spectrometer (AAS-
hydrochloride (NH2 OH·HCl, >98.5%), hydrazine 800, Perkin Elmer, USA).
hydrate (N2 H4 ·H2 O, 50%), thiourea (CH4 N2 S, >99.0%),
stannous chloride (SnCl2 , >98.0%), sodium sulfide (Na2 S, 1.4 Selection of reductant
>98.0%), and sodium citrate (C6 H5 O7 Na3 ·2H2 O, >99.0%) In order to optimize the reducing reagent, 11 reductants
were from Beijing Chemical Reagents Company (China). including Johnson’s reducing agent (14% Na2 S2 O5 + 1.4%
Milli-Q water was used in all experiments. Na2 S2 O3 + 10% H2 SO4 ) (Johnson and Pilson, 1972),
Groundwater samples were collected in Shanxi, China. C6 H8 O6 , KI, NH2 OH·HCl, N2 H4 ·H2 O, CH4 N2 S, SnCl2 ,
The samples were passed through a 0.45-μm syringe filter Na2 SO3 , Na2 S2 O5 , Na2 S, and C6 H5 O7 Na3 ·2H2 O were
for soluble As analysis. Subsamples were acidified with tested and compared. Experiments were carried out by
HCl for As speciation analysis using HPLC-AFS (Jitian, reacting 10 mL 1000 μg/L As(V) sample with 1 mL
China) in the laboratory. Soluble metals (Al, Ca, Mg, different concentrations of reducing reagent. Then, one
Na, K, Si, total Fe, and Mn) and anions (F− , Cl− , NO3 − , group of the mixtures was incubated at room temperature
SO4 2− , PO4 3− ) were determined using an inductively for an hour to facilitate As(V) reduction, the other group
coupled plasma optical emission spectrometry (ICP-OES, was placed in a water bath at 80°C for 30 min. After
Perkin Elmer, USA) and an ion chromatography DX-1100 samples were cooled to room temperature, 1 mL color
(Dionex, USA), respectively. All samples were stored at reagent was added and the absorbance was measured.
No. 7 A novel colorimetric method for field arsenic speciation analysis 1343

1.5 Accuracy comparison between colorimetric method oxidation in 5 min for acidified samples and an incomplete
and HPLC-AFS oxidation in 30 min without acidification (Fig. 1b). In stark
contrast, KMnO4 required no acidification, and As(III) was
The As speciation results obtained with the proposed
completely oxidized to As(V) in 2 min. The results suggest
colorimetric method was compared with those using
that KMnO4 is a more efficient oxidant comparing to KIO3 .
HPLC-AFS. Separation of As compounds was carried out
Furthermore, KMnO4 has been used as an As(III) oxidant
in a Hamilton PRP-X100 anion exchange column, using
in groundwater treatment for its high efficiency and no
15 mmol/L phosphate buffer at pH 6 as mobile phase at
dependence on pH values (Li et al., 2007). In addition,
1.0 mL/min flow rate. Following the separation of different
no interference to colorimetric As analysis was observed
As species, online hydride generation was achieved by
using KMnO4 , which is in agreement with a previous study
reacting with 1.5 wt.% KBH4 and 7 wt.% HCl. The hydride
(Lenoble et al., 2003). Therefore, 0.3 mmol/L KMnO4 was
was atomized in a hydrogen flame and the fluorescence
used as oxidant regent in the proposed method.
signal for each of the As species was recorded using an
AFS-8130 spectrometry (Jitian, China). 2.2 Optimization of reducing reagent
Among the eleven tested reducing agents, N2 H4 ·H2 O,
2 Results and discussion Na2 SO3 , SnCl2 , Na2 S, and C6 H5 O7 Na3 ·2H2 O interfered
with the molybdenum blue color formation. Therefore,
2.1 Optimization of oxidizing reagent these five agents were excluded for further considerations.
Figure 1a shows the change of absorbance as a function Figure 2 shows the change of As(V) species as a
of KIO3 , KMnO4 , and NaClO concentrations. When con- function of concentrations of NH2 OH·HCl, KI, CH4 N2 S,
centrations of KMnO4 and KIO3 were increased to higher C6 H8 O6 , and Na2 S2 O5 . None of the reagents could com-
than 0.1 mmol/L, the absorbance reached a plateau at pletely reduce As(V) to As(III) at room temperature
approximately 0.22 and remained unchanged with further (Fig. 2a), even though the concentrations of these reducing
increasing concentrations up to 5 mmol/L. The results agents were increased up to 30% (mass ratio). Although
indicate that 0.1 mmol/L KMnO4 and KIO3 could com- Na2 S2 O5 resulted in a slightly higher reducing capacity
pletely oxidize 1000 μg/L (0.013 mmol/L) As(III) to than other agents, it was excluded for further consideration
As(V). However, NaClO resulted in a significantly lower because toxic SO2 gas was generated as a byproduct. When
absorbance than that of KMnO4 and KIO3 (Fig. 1a), which the mixed solution was heated with an alcohol burner for 1
could be attributed to the lower standard potential of min and then cured at room temperature for 1 hr, none of
◦
NaClO (E = 0.89 eV) comparing to KMnO4 (1.51 eV) the reagents could completely reduce 1000 μg/L As(V) to
and KIO3 (1.14 eV) (Speight, 2005). Therefore, KMnO4 As(III) (Fig. 2b).
and KIO3 were selected for further kinetic studies at a Because high temperature may be required for some
0.3 mmol/L concentration level to compensate for possible reagents to fulfill their reducing abilities, the reducing
interferences from reducing materials such as sulfide in step was incubated in a water bath at 80°C for 30 min.
groundwater. NH4 N2 S could completely reduce 1000 μg/L As(V) when
Figure 1b shows the kinetics of 500 μg/L As(III) oxida- its concentration was higher than 10% (Fig. 2c). Therefore,
tion by 0.3 mmol/L KMnO4 and KIO3 . KIO3 exhibited a 10% NH4 N2 S was used as reducing reagent in the present
great pH dependence as evidenced by a complete As(III) method.

Total As KMnO4 non-acidified

KIO3 KMnO4 NaClO KIO3 non-acidified KIO3 acidified
0 .24 650
a
b
550
As(III) concentration (μg/L)

450
0.20
. 0.22
Absorbance

0.20 350

0.18 250
0 .16 0.16
150
0.14
0 0.1 0.2 0.3
50

0.12 -50
-0 .2 1 .8 3 .8 5 .8 0 10 20 30
Oxidation reagent concentration (mmol/L) Time (min)
Fig. 1 Absorbance of 1000 μg/L As(III) as a function of KIO3 , KMnO4 , and NaClO concentration. Inset shows the absorbance with oxidation
concentrations less than 0.3 mmol/L (a). Changes of As(III) concentration as a function of time when oxidized by KIO3 and KMnO4 (b).
1344 Journal of Environmental Sciences 2012, 24(7) 1341–1346 / Shan Hu et al. Vol. 24

NH2OH·HCl KI CH4N2S C6 H 8 O 6 Na2S2O5

100 100 100

a b c

90 80 80
As(V) remaining (%)

As(V) remaining (%)

As(V) remainng (%)

80 60 60

70 40 40

60 20 20

50 00 0
10 0 20 30 10 20 30 0 10 20 30
Reducing reagent concentration (%) Reducing reagent concentration (%) Reducing reagent concentration (%)
Fig. 2 Percentage of remaining As(V) concentrations in an initial 1000 μg/L As(V) sample as a function of reducing reagents under three conditions:
at room temperature in an hour (a); heated with an alcohol burner for 1 min before color development at room temperature in an hour (b); and reacting
in a water bath at 80°C for 30 min (c).

Total As As(III) As(V) added in the sample, and the absorbance of the three sub-
2.0 samples in reduced, acidified, and oxidized forms were
Total As
monitored as shown in Fig. 4. The absorbance in all
R2 = 0.998 three treatment samples proportionally increased with the
1.5 increase of SO4 2− concentrations (Fig. 4a), indicating the
addition of SO4 2− would lead to an increase in absorbance.
However, the interference of SO4 2− was eliminated by the
Absorbance

1.0 subsequent subtraction calculations as shown in the ex-

As(III) perimental section. Furthermore, the absorbance difference
R2 = 0.992 remained unchanged for various As concentrations when
As(V) SO4 2− concentrations increased to 0.3 mol/L (Fig. 4b). In
0.5
R2 = 0.999 addition, no interference was observed with HA concentra-
tions up to 10 mg/L (Fig. 4c). The results suggest that the
novel colorimetric method is applicable for As analysis in
0
40000 2000
6000 8000 10000 high sulfate and DOM waters.
As concentration (μg/L)
2.5 As determination in natural waters
Fig. 3 Absorbance of total As, As(III), and As(V) as a function of As
concentration. Figure 5 shows the comparison of total As, As(III), and
As(V) concentrations measured using the colorimetric
2.3 Dynamic detection range method and HPLC-AFS for 28 groundwater samples col-
Figure 3 shows the linear dynamic detection range of lected from Shanxi Province, China. The soluble metals
total As, As(III), and As(V) using the present colorimetric in the groundwater samples are Al (0.1–0.2 mg/L), Ca
method. A linear regression (R2 = 0.9979) was achieved (3.6–219.8 mg/L), Na (40.0–946.4 mg/L), K (0.4–10.9
in four orders of magnitude concentration range (total As, mg/L), Si (7.2–11.8 mg/L), Fe (0–0.35 mg/L), and Mn
0.01–10 mg/L). The wider linear range increased from 2– (0–1.2 mg/L). The soluble anions in the groundwater
220 (Johnson and Pilson, 1972) and 10–1000 μg/L (Dhar samples are F− (0.3–5.6 mg/L), Cl− (39.0–1644.5 mg/L),
et al., 2004) to 10 mg/L enabled direct measurements NO3 − (0–20.0 mg/L), SO4 2− (0.4–973.3 mg/L), and PO4 3−
of groundwater As concentrations in several mg/L level (0.075–1.589 mg/L). The colorimetric results obtained in
(Nordstrom, 2002) without sample dilution. the field were consistent with the HPLC-AFS results with
The method detection limit was estimated to be 0.1 a slope in the range 0.998–1.026 (p < 0.0001).
μmol/L (8 μg/L) with seven repeated measurements of Table 1 shows the recovery of total As, As(III), and
reagent blank containing 10 μmol/L phosphate, which is As(V) in six groundwater samples spiked with 50 μg/L
comparable to previous reported value of 0.09 μmol/L (7 As(III) and As(V). A recovery of 96%–115% (total
μg/L) (Dhar et al., 2004). As), 85%–122% (As(III)), and 88%–127% (As(V)) was
achieved for samples with total As concentration between
2.4 Interferences of sulfate and humic acid (HA) 100–800 μg/L.
Because sulfate and dissolved organic matter (DOM) are Some common constituents such as Fe, SiO3 2− , PO4 3− ,
commonly found in elevated concentrations in industrial and SO4 2− may inhibit the application of molybdenum
and natural water samples, their possible interference to blue analysis (Geng and Zhu, 2006; Melamed, 2005).
the color formation is critical to develop an effective The substration procedure could eliminate the interference
colorimetric method. To study their effects, increasing from complex groundwater matrix as evidenced by the
amount of sulfate and HA, a representative DOM, were good agreement with results of HPLC-AFS (Fig. 5). In
No. 7 A novel colorimetric method for field arsenic speciation analysis 1345

Table 1 Recovery analysis of total As, As(III), and As(V) in groundwater samplesa

Sample Total As As(V) As(III)

Raw Spiked Recovery Raw Spiked Recovery Raw Spiked Recovery
(μg/L) (μg/L) (%)b (μg/L) (μg/L) (%) (μg/L) (μg/L) (%)

1 24 125 101 10 63 107 14 61 94

2 326 432 106 44 93 98 277 329 113
3 370 485 115 59 122 127 305 357 103
4 384 490 106 107 151 88 272 333 122
5 0 96 96 0 54 107 0 42 85
6 648 758 110 73 132 117 563 615 103
a All the samples was spiked with 50 μg/L As(III) and As(V); b Recovery = ((Spiked-Raw)/Spiked) ×100%.

Reduced Untreated Oxidized As = 0 As = 50 μg/L As = 500 μg/L As = 1000 μg/L

0.25 0.25
a b c
0.6 0.20 0.20
Absorbance difference

Absorbance difference
0.15 0.15
Absorbance

0.5
0.10 0.10

0.05 0.05
0.4
0 0

0.3 -0.05 -0.05

-0.02 0.08 0.18 0.28 -0.02 0.08 0.18 0.28 -0.5 1.5 3.5 5.5 7.5 9.5 11.5
SO42- concentration (mol/L) SO42- concentration (mol/L) HA concentration (mg/L)
Fig. 4 Absorbance of reduced, untreated, and oxidized samples of 1000 μg/L total As (As(III):As(V) = 1:1) as a function of SO4 2− concentration (a).
Absorbance difference (oxidized-reduced) of total As (As(III):As(V) = 1:1) as a function of SO4 2− (b), and HA (c) concentration.

1000 800 200

Total As As(III) As(V)
Colorimetric analysis (μg/L)
Colorimetric analysis (μg/L)

Colorimetric analysis (μg/L)

800 y = 0.9977x 600 150

R = 0.9997
600 P < 0.0001
400 100 y = 1.0283x
y = 1.0263x
400 R = 0.9995 R = 0.9995
P < 0.0001 P < 0.0001
200 50
200
a b c
0 0 0
400 600 0
800 1000 200 0 200 400 600 800 0 50 100 150 200
HPLC-AFS (μg/L) HPLC-AFS (μg/L) HPLC-AFS (μg/L)
Fig. 5 Comparison of total As (a), As(III) (b), and As(V) (c) concentrations in groundwater samples determined by the present colorimetric method
and HPLC-AFS.

addition, the great recovery further confirmed that no for the analysis of groundwater and industrial wastewater
interference occurred with high concentrations of Si, Fe, samples. Furthermore, the presence of phosphate, silicate,
and PO4 3− in our groundwater samples. sulfate, humic acid, and other heavy metals in the water
To extend the usage of the colorimetric method, it was matrix could not interfere with the colorimetric analysis. In
employed to determine As concentration and speciation in addition, the portable spectrophotometer used in the study
an industrial wastewater containing 0.28 mol/L SO4 2− at is an inexpensive and versatile instrument which is suitable
pH 1.4. The colorimetric analysis showed that 3.61 g/L for a broad spectrum of analytical applications. Therefore,
As(III) existed as the only As species in the sample, which the present method offers an excellent choice in field As
was in good agreement with the HPLC-AFS result of 3.56 determination.
g/L As(III). The highly complex matrix of this copper
smelting wastewater (Luo et al., 2010) further proved the
versatile application and consistency of our method. Acknowledgments

3 Conclusions We acknowledge the financial support of the National

Natural Science Foundation of China (No. 20977098,
The As concentration and speciation determined using the 20921063) and the National Basic Research Program (973)
colorimetic method are in line with the HPLC-AFS results of China (No. 2010CB933502).
1346 Journal of Environmental Sciences 2012, 24(7) 1341–1346 / Shan Hu et al. Vol. 24

References Journal of Environmental Sciences, 19(7): 783–786.

Liao M X, Deng T L, 2006. Arsenic species analysis in pore-
Arora M, Megharaj M, Naidu R, 2009. Arsenic testing field kits: waters and sediments using hydride generation atomic
some considerations and recommendations. Environmental fluorescence spectrometry. Journal of Environmental Sci-
Geochemistry and Health, 31(S1): 45–48. ences, 18(5): 995–999.
Dhar R K, Zheng Y, Rubenstone J, van Geen A, 2004. A rapid Luo T, Cui J L, Hu S, Jing C Y, 2010. Arsenic temoval and
colorimetric method for measuring arsenic concentrations recovery from copper smelting wastewater using TiO2 .
in groundwater. Analytica Chimica Acta, 526(2): 203–209. Environmental Science & Technology, 44(23): 9094–9098.
Fendorf S, Michael H A, van Geen A, 2010. Spatial and temporal Melamed D, 2005. Monitoring arsenic in the environment: a
variations of groundwater arsenic in south and Southeast review of science and technologies with the potential for
Asia. Science, 328(5982): 1123–1127. field measurements. Analytica Chimica Acta, 532(1): 1–13.
Geng C N, Zhu Y G, 2006. Arsenate and phosphate interaction Meng X G, Jing C Y, Korfiatis G P, 2003. Biogeochemistry
in Saccharomyces cerevisiae. Journal of Environmental of environmentally important trace elements. In: American
Sciences, 18(1): 96–100. Chemical Society Symposium Series 835 (Cai Y, Braids O
Jing C Y, Korfiatis G P, Meng X G, 2003. Immobilization C, eds.). American Chemical Society, Washington, DC. 70–
mechanisms of arsenate in iron hydroxide sludge stabilized 83.
with cement. Environmental Science & Technology, 37(21): Nordstrom D K, 2002. Public health-worldwide occurrences of
5050–5056. arsenic in ground water. Science, 296(5576): 2143–2145.
Johnson D L, 1971. Simultaneous determination of arsenate Rahman M M, Mukherjee D, Sengupta M K, Chowdhury U
and phosphate in natural waters. Environmental Science & K, Lodh D, Chanda C R et al., 2002. Effectiveness and
Technology, 5(5): 411–414. reliability of arsenic field testing kits: Are the million dollar
Johnson D L, Pilson M E Q, 1972. Spectrophotometric determi- screening projects effective or not? Environmental Science
nation of arsenite, arsenate, and phosphate in natural waters. & Technology, 36(24): 5385–5394.
Analytica Chimica Acta, 58(2): 289–299. Speight J G, 2005. Lange’s Handbook of Chemistry (16th ed).
Lenoble V, Deluchat V, Serpaud B, Bollinger J C, 2003. Arsenite McGraw-Hill. 1382–1386.
oxidation and arsenate determination by the molybdene Tsang S, Phu F, Baum M M, Poskrebyshev G A, 2007. Deter-
blue method. Talanta, 61(3): 267–276. mination of phosphate/arsenate by a modified molybdenum
Li N, Fan M H, van Leeuwen J, Saha B, Yang H Q, Huang C blue method and reduction of arsenate by S2 O4 2− . Talanta,
P, 2007. Oxidation of As(III) by potassium permanganate. 71(4): 1560–1568.