Rapid determination of five arsenic

species in polished rice using

HPLC-ICP-MS

Application note
Food safety
Authors

Bing Yue
China National Center For Food
Safety Risk Assessment, Beijing,
China

Juane Song, Ke Xie, Zhao-jun Yun

and Jian-qiu Mi
Agilent Technologies, Beijing, China

Introduction
Arsenic (As) is a well-known toxic element that can be present in the
environment and foodstuffs. Consequently it is a highly regulated element
that is widely monitored. However, As speciation analysis rather than total
As analysis is important because its biological toxicity is highly dependent
on its chemical form. For example, if a seaweed sample is known to contain
a high level of As, but mainly in the form of AsB, then there is no potential
risk from its consumption as AsB is non-toxic. Among the five major species
of arsenic, As(V), MMA, As(III) DMA and AsB, only the two inorganic forms
(As(III) and As(V)) are toxic. They are also carcinogenic to humans, so need
to be measured, especially in foodstuffs. Arsenic in rice is of special concern
in countries where it is consumed daily. Rice plants absorb As from soil and
water. Since rice requires a lot of water to grow compared to other plants,
the potential for arsenic accumulation is increased.

In this study, we present a rapid method for the determination of the toxic
As species in rice using an Agilent 1260 HPLC system equipped with an
Agilent ZORBAX SB-Aq column coupled to an Agilent 7900 ICP-MS.
Experimental Chemicals and reagents
Stock standards of As(V), MMA, As(III), DMA and
Instrumentation AsB were purchased from NIMC (National Institute of
An Agilent 1260 HPLC system comprising a quaternary Metrology China, Beijing, China). High purity citric acid
pump, autosampler and vacuum degasser was coupled (≥99.5%), sodium hexane-sulfonate (≥98%) and sodium
to an Agilent 7900 CP-MS. The column exit was simply hydroxide (≥99.99%) were purchased from Sigma
connected to the nebulizer of the ICP-MS using PFA Aldrich Chemicals (UK). Ultra-pure grade HNO3 (Sigma
tubing. The Agilent ZORBAX SB-Aq (part number Aldrich, UK) was used for the sample preparation.
880975-914, 4.6 mm id x 250 mm, 5 μm) reversed
phase column was maintained at room temperature Sample preparation
throughout the analysis. The mobile phase was 20 mM The polished rice samples were obtained from different
citric acid and 5 mM sodium hexane sulfonate, adjusted locations in the following provinces of China: Jiangsu,
to pH = 4.3 with sodium hydroxide. Two injection Fujian, Liaoning, Chongqing and Jiangxi.
volumes, 5 μL or 100 μL, were evaluated. HPLC and
ICP‑MS parameters are given in Table 1. The polished rice samples were crushed and sieved
through a 30 mesh strainer to make rice flour. 1.00 g of
The Agilent 7900 ICP-MS with the Octopole Reaction the rice flour sample was weighed into a polypropylene
System (ORS4) collision/reaction cell (CRC) provides centrifuge tube (50 mL). 15 mL of 0.15 M HNO3 was
sensitive and specific analysis of As in the presence added and the sample was vortexed for 30 seconds. The
of multiple interferences. In some regulatory guidance, tightly capped tube was left to stand overnight at room
food samples must be extracted using HCl to simulate temperature; it was then placed in an oven at 90 °C
gastric conditions. In this case, the formation of for 2.5 hours where it was agitated for 30 seconds,
polyatomic ions based on ArCl and CaCl may cause every 30 minutes. After the extraction had finished, the
spectral interferences on the sole isotope of As at sample was cooled in a refrigerator at 4 °C. The cooled
m/z 75. However, operating the 7900 ICP-MS ORS4 in rice flour suspension was centrifuged at 8000 rpm
helium mode effectively removes these matrix-based for 15 minutes and then the supernatant was passed
polyatomic interferences on As. through a 0.22 μm polyether sulphide (PES) filter. The
filtrate was stored at 4 °C and analyzed within 24 hours
Table 1. Operating conditions of the Agilent 1260 HPLC and 7900 ICP-MS to minimize any species inter-conversion.
ICP-MS parameters
RF power 1550 W Results and discussion
Sampling depth 8 mm
Carrier gas 0.6 L/min Separation and quantification
Makeup gas 0.5 L/min
Calibration standard solutions containing the five
Cell gas He at 4 mL/min
As species at different concentrations (1 ppb, 5 ppb,
KED 3V
10 ppb, 50 ppb, 100 ppb and 200 ppb) were prepared and
Isotopes monitored 75
As, 35Cl
analyzed. The chromatogram for the 10 ppb calibration
standard (5 μL injection) shows the five As species are
Dwell time 1.0 s/75As; 0.1 s/35Cl
well separated in 4 minutes (Figure 1). The results were
HPLC parameters used to create calibration curves for the As species, as
Isocratic, 20 mM C6H8O7 and 5 mM shown in Figure 2.
Mobile phase C6H13NaO3S, adjusted to pH 4.3 with
NaOH
Column Agilent ZORBAX SB-Aq
Flow rate 1.2 mL/min
Temperature Ambient
Injection volume 5 μL or 100 μL
Run time 4 min

2
 Table 2. Detection limits (ppb) for As(V), MMA, As(III), DMA and AsB with
5μL and 100 μL injections

Injection
volume As(V) MMA As(III) DMA AsB
(μL)
5 0.106 0.121 0.187 0.267 0.667
100 0.010 0.012 0.025 0.028 0.042

Figure 3 shows the chromatograph of a 50 ppt

As species standard mixture (100 μL injection)
demonstrating the ultra-trace analysis capability of the
Figure 1. Chromatogram of a 10 ppb As (conc. as As element) species method.
mixture standard (5 μL injection)

Figure 3. Chromatogram of 50 ppt As species mixture (100 μL injection)

Certified Reference Material (CRM) analysis

The method was applied to the analysis of two rice
flour CRMs; NIST 1568a and NIST 1568b. Though only
total As content is certified for NIST 1568a, (290 ±
30 ppb), the CRM has been well-characterized in the
literature. The two CRMs were extracted following
the aforementioned sample preparation method and
analyzed for the five As species.

Tables 3 and 4 summarize the quantified results of

the inorganic As species, MMA, DMA and the total
As content in the two CRMs, the certified values or
reported values in the literature. For this study, total As
Figure 2. Calibration graphs for As(V), MMA, As(III),DMA and AsB showing was calculated from the sum of inorganic As, MMA,
excellent linearity. DMA and AsB. As shown, the quantified inorganic As
Detection limits (iAs) in NIST 1568b and total As in both CRMs are in
Detection limits (DLs) for each arsenic species were good agreement with the certified values and within
calculated as the concentration equivalent to three the standard deviation. The quantified results for each
times the peak-to-peak baseline noise (S/N) of the species are also in good agreement with reported values
standard chromatogram. The results are summarized in in the literature [1-4].
Table 2. As shown, the DL of the five As species were
between 0.106 ppb and 0.667 ppb with a 5 μL injection
and 0.010 ppb and 0.042 ppb with a 100 μL injection.

3
Table 3. Quantification of As species in NIST 1568b Rice Floura Precision
As(V) MMA As(III) DMA iAs b
Total As Precision of the method was checked by repeatedly
(µg/kg) (µg/kg) (µg/kg) (µg/kg) (µg/kg) (µg/kg) analyzing a rice sample, with ten separate injections.
This
62.0 ± 5.1 10.6 ± 1.3 36.8 ± 5.8 179 ± 7 98.8 ± 3.6 288.4 Table 6 shows the concentration results for all the As
study
species, indicating the excellent reproducibility for the
Certified
Value
11.6 ± 3.5 180 ± 12 92 ± 10 285±14 ten separate analyses. The %RSDs for all As species
were less than 4%, and less than 2% for inorganic As.
a
value is mean ± SD of ten injections, binorganic arsenic: sum of As(III) and
As(V) Table 6. Results of reproducibility of rice # 4 (n = 10) with 5 μL injection
Table 4. Comparison of quantified As species in NIST 1568a Rice Flour Sample As(V) MMA As(III) DMA AsB iAs
determined in this study compared to values from the literaturea. The certified (µg/kg) (µg/kg) (µg/kg) (µg/kg) (µg/kg) (µg/kg)
total As content of NIST 1568a is 290 ± 30 ppb.
Rice #
56.10 N.D. 107.32 22.29 N.D. 163.42
As(V) MMA As(III) DMA iAsb Total As 4-1
(µg/kg) (µg/kg) (µg/kg) (µg/kg) (µg/kg) (µg/kg) Rice #
55.23 N.D. 110.39 24.74 N.D. 165.62
Ref. [1] 44±2 12±0.8 52±1 173±2 100 281±2 4-2

Ref. [2] 50.3±2.9 14.9±3.9 63.4±3.5 144±4.5 113.7 272.8±9.9 Rice #

56.97 N.D. 107.30 23.10 N.D. 164.27
4-3
Ref. [3] 53.7±3.3 14.8±1.8 54.7±1.4 165±8 108.4 288.2
Rice #
57.68 N.D. 108.51 24.19 N.D. 166.18
Ref. [4] 13.6±2.5 168.0±2.2 104.3±4.4 286.1±3.8 4-4
This Rice #
64±5 14±2 57±4 140±4 121 275±4 57.16 N.D. 108.46 23.64 N.D. 165.62
study 4-5
a
value is mean ± SD of ten injections, binorganic arsenic: sum of As(III) and Rice #
56.10 N.D. 108.52 24.82 N.D. 164.62
As(V) 4-6
Rice #
56.43 N.D. 108.89 23.96 N.D. 165.32
Spiked sample recovery 4-7
Spiked sample recovery was checked for the five As Rice #
57.27 N.D. 107.90 22.37 N.D. 165.17
species. A mixture of As(V), MMA, As(III), DMA and 4-8

AsB was spiked into three rice sample extracts at 1 Rice #

56.18 N.D. 109.56 22.70 N.D. 165.73
4-9
ppb, 5 ppb and 10 ppb. Three replicates were prepared
Rice #
(spike 1, 2 and 3). As shown in Table 5, good recoveries 4-10
56.71 N.D. 108.76 23.35 N.D. 165.47
ranging from 85% to 110% were achieved for all species, Average 56.58 – 108.56 23.52 – 165.14
at all concentrations, in each of the spiked solutions. % RSD 1.3% – 0.9% 3.9% – 0.5%

Table 5. Results of spiked sample recovery with 5 μL injection

Sample
Spike Conc. As(V) MMA As(III) DMA AsB
name
Conc.(ppb) Rec.(%) Conc.(ppb) Rec.(%) Conc.(ppb) Rec.(%) Conc.(ppb) Rec.(%) Conc.(ppb) Rec.(%)
Rice # 1 0.48 – N.D. – 4.60 – 0.48 – N.D. –
Spike 1 1.38 89.2% 1.02 101.5% 5.65 105.6% 1.54 106.1% 1.03 103.2%
Spike 2 1 ppb 1.43 94.5% 0.96 96.2% 5.51 90.8% 1.42 93.9% 1.07 106.5%
Spike 3 1.47 98.1% 0.95 94.3% 5.52 92.2% 1.45 97.4% 0.98 98.4%
Rice # 2 1.14 – N.D. – 5.82 – 1.32 – N.D. –
Spike 1 5.52 87.6% 4.96 99.2% 10.98 103.2% 6.23 98.3% 4.82 96.3%
Spike 2 5 ppb 5.65 90.1% 4.77 95.3% 10.88 101.3% 6.45 102.6% 4.97 99.4%
Spike 3 5.77 92.5% 4.34 86.7% 11.35 110.6% 6.62 106.1% 5.13 102. %
Rice # 3 0.28 – N.D. – 3.44 – 0.73 – N.D. –
Spike 1 9.65 93.7% 9.68 96.8% 13.35 99.1% 9.72 89.9% 8.91 89.1%
Spike 2 10 ppb 9.89 96.1% 10.08 100.8% 12.97 95.3% 10.29 95.6% 9.54 95.4%
Spike 3 9.53 92.5% 10.62 106.2% 13.10 96.6% 10.65 99.2% 10.48 104.8%

4
Commercial rice sample analysis Conclusions
The As concentration was determined in the 20
different commercial rice samples and the results are The separation of five arsenic species in polished
summarized in Table 7. As shown, As(V), As(III) and rice was carried out using an Agilent 1260 Infinity LC
DMA were found in all 20 rice samples. The inorganic coupled to an Agilent 7900 ICP-MS. The method is rapid
As and DMA content varied significantly among the rice - all five species including the toxicologically relevant
samples, likely reflecting the As concentration in the inorganic forms As(III) and As(V) were determined
growing environment of the original plant. Although the in 4 minutes, with excellent sensitivity, accuracy and
concentration of toxic inorganic As varied among the precision. Any potential interferences on As, such
samples, all of the rice samples contained lower limit as ArCl+, were removed by the ORS4 cell operating in
than the upper level of 0.2 mg/kg recommended by helium mode.
Codex Alimentarius Commission.
Inorganic arsenic and DMA species were detected in all
Table 7. Quantitative results (μg/kg) for all five As species in 20 rice samples
of the twenty rice samples at various concentrations.
Sample As(V) MMA As(III) DMA AsB iAs However, none of the samples contained more than
name Codex Alimentarius Commission’s recommended limit of
Rice # 1 7.27 N.D. 68.97 7.16 N.D. 76.24 0.2 mg/kg.
Rice # 2 17.15 N.D. 87.23 19.77 N.D. 104.38
Rice # 3 4.26 N.D. 51.57 10.93 N.D. 55.83 The validated methodology outlined here is suitable for
Rice # 4 56.10 N.D. 107.32 22.29 N.D. 163.42 the rapid determination of As(V), MMA, As(III), DMA
Rice # 5 17.31 N.D. 147.2 26.38 N.D. 164.73 and AsB at trace levels in rice and other food samples.
Rice # 6 12.52 N.D. 75.82 8.79 N.D. 88.35
Rice # 7 7.14 N.D. 53.84 2.67 N.D. 68.94 References
Rice # 8 14.4 N.D. 106.14 27.27 N.D. 120.54
Rice # 9 14.49 N.D. 72.24 29.51 N.D. 86.73 1. Narukawa T., Inagaki K., Kuroiwa T., Chiba K, ‘The
Rice # 10 13.73 N.D. 101.68 39.41 N.D. 115.41 extraction and speciation of arsenic in rice flour by
Rice # 11 20.50 N.D. 87.04 285.21 N.D. 107.55 HPLC-ICP-MS’ [J], Talanta 77, 2008, 427-432
Rice # 12 22.78 N.D. 70.94 160.77 N.D. 93.72
2. Huang J H, Ilgen G, ‘Blank value, adsorption,
Rice # 13 23.74 N.D. 122.19 34.53 N.D. 145.94
preconcentration, and preservation for arsenic
Rice # 14 13.85 N.D. 59.79 361.82 N.D. 73.64 speciation of the environmental water samples’ [J].
Rice # 15 16.38 N.D. 91.34 83.42 N.D. 125.38 Analytica Chimica Acta, 2004, 512:1-10.
Rice # 16 18.48 N.D. 110.58 32.25 N.D. 129.06
Rice # 17 10.83 N.D. 61.56 6.69 N.D. 72.39
3. Caruso J.A, B’Hymer C, Heitkemper D.T., ‘An
evaluation of extraction techniques for arsenic species
Rice # 18 16.96 N.D. 105.14 209.49 N.D. 122.1
from freeze-dried apple sample’ [J]. Analyst, 2001, 126:
Rice # 19 36.04 N.D. 97.17 176.12 N.D. 133.22 136-140.
Rice # 20 16.35 N.D. 74.55 587.49 N.D. 90.9
4. Toni LIorente-Mirandes et al., ‘A fully validated
method for the determination of arsenic species in rice
and infant cereal products’, Pure Appl. Chem., 2012, vol.
84, No. 2, 225-238

5
 www.agilent.com
Agilent shall not be liable for errors contained herein or for incidental or consequential
damages in connection with the furnishing, performance or use of this material.

Information, descriptions, and specifications in this publication are subject to change

without notice.

© Agilent Technologies, Inc. 2015

Published July 15, 2015
Publication number: 5991-5933EN