Preconcentration of ultra-trace arsenic with nanometre-sized TiO₂ colloid and determination by AFS with slurry sampling

Abstract

This study describes a united method for the analysis of ultra-trace arsenic in environmental water samples by preconcentration of TiO₂ colloid and determination of Atomic Fluorescence Spectrometry (AFS). When the pH values of solution are between 6.0 and 7.0, the adsorption for arsenic is above 98.0% with the dosage of nanometre-sized TiO₂ colloid (0.2%) being 6.0 mL. Without desorbing the concentrated arsenic, the precipitated TiO₂ is directly inverted to colloid after centrifugation. Arsenic is then determined by AFS with slurry sampling. This united method is simple, fast and sensitive. The detection limit (3σ) and the relative standard deviation (R.S.D) of this method are 10.6 ng L⁻¹ and 7.04% (n = 6), respectively. When the method is applied to the determination of ultra-trace arsenic in environmental standard samples, the results are in good agreement with the standard values. And when it comes to environmental samples, satisfactory results are obtained.

---

1. Introduction

Arsenic is an ubiquitous element, and its abundance ranks 20th, 14th and 12th in the earth’s crust, seawater and human body, respectively.¹ Arsenic can exist in four oxidation states: As(−III), As(0), As(III), and As(V). Environmental forms include arsenites, arsenates, methylarsenic acid, dimethylarsinic acid, arsine, etc. The predominant form of inorganic arsenic in aqueous, aerobic environments is arsenate, whereas arsenite is more prevalent in anoxic environments.² Arsenic was one of the first chemicals recognized as a cause of cancer.³ The WHO guideline value for As in drinking water was provisionally reduced in 1993 from 50 to 10 μg L⁻¹. The new recommended value was based on the increasing awareness of the toxicity of arsenic, particularly its carcinogenicity, and on the ability to measure it quantitatively.⁴ Arsenic is known for its physiological and toxicological effects.⁵ For some waters that have not been polluted or subsurface water, the background value of arsenic can be very low.^6,7 It is difficult to determine directly. Therefore, the preconcentration and separation as well as the determination of arsenic are of great importance and necessity.

Nanoparticles, the building blocks of nanotechnology, have been broadly defined as having at least one dimension at 100 nm or less.⁸ Nanomaterials are made of nanoparticles. The sizes, surface structures and interparticle interactions of nanomaterials determine their unique properties and the improved performances, and make their potential application in many areas.⁹ The atomic number, the area, energy, and chemical combination energy of its surface all rapidly increase as the particle radius decreases.¹⁰ The atoms on the surface of the nanoparticles are unsaturated and can easily bind with other atoms. The high chemical activity they possess makes their combination easier. Consequently, compared with general adsorbents, nanomaterials have a much higher adsorption capacity for many metal ions and are an ideal kind of adsorbent. Nanometer-size TiO₂ powder shows great potential and advantages as solid-phase extractant for the separation and preconcentration of trace amounts of many metals. Some research on the adsorption of nanometre-size TiO₂ powder for heavy metals has been done for years, and methods of concentration and separation for these elements by using nanometre-size TiO₂ powder have been established. Relevant research has shown that nanometre-size TiO₂ powder is effective in the adsorption for trace metals.^11–17

Nanometer-size TiO₂ colloid was first applied to the analyses of trace elements by our seminar in 2006, and the methods have been continuously improved.¹⁸ Compared with nanometre-size TiO₂ powder, its colloid is more advantageous as adsorbent for the concentration and separation of trace elements. Smaller particle radius, larger specific surface and better dispersal without agglomeration are the advantages that nanometre-size TiO₂ colloid has over its powder. Compared with the conventional adsorption materials, TiO₂ colloid has better adsorption efficiency. With slurry sampling, the complicated processes of filtration and desorption were omitted, so that the time of analysis was shortened and the possibility of contamination was reduced.

A simple, fast, sensitive united-method for the determination of ultra-trace arsenic is developed in this study. Nanometer-size TiO₂ colloid, which is applied as adsorbent, shows its full advantages. The concentration of arsenic is determined by atomic fluorescence spectrometry (AFS). This method is applied to the determination of arsenic in environmental water samples with great success.

2. Experimental

2.1 Apparatus and reagents

The PF6-2 non-dispersive atomic fluorescence spectrometer (Beijing Purkinje General Instruments Co., Ltd., Beijing, China), equipped with the HAF-2 arsenic hollow cathode lamp, was used for arsenic determination. The pH values were measured with the Mettler-Toledo 320-S pH meter (Mettler-Toledo, Greifensee, Switzerland). The LD5-2A centrifuge (Beijing Medical Centrifuge Factory, Beijing, China) was applied to separate the suspensions, and the AY-120 electronic balance (Shimadzu, Tokyo, Japan) was used in this experiment. The instrumental conditions are displayed in Table 1.

Table 1 Experiment conditions

Negative high voltage/V	Flow rate of Ar/mL min^−1	Lamp current/mA		Atomization T/°C
265	500	Main	Auxiliary	170
		30	30

---

Colloid nanometre-size TiO₂ (2.0% w/w) was supplied by Infrared Academy of Wuhan University and was diluted 10 times to 0.2% w/w for use. Thiourea-ascorbic acid mixed solution (5% w/w) was used as reductant. Solution (1.5%) of KBH₄ and hydrochloric acid (2% v/v) were involved in the procedure of determination. Ar (99.99%) was applied as carrier gas. As₂O₃ (G.R.) was used to prepare standard stock solution (1.000 g L⁻¹) of As(III) and Na₂HAsO₄·7H₂O (G.R.) was used when preparing standard stock solution (1.000 g L⁻¹) of As(V). A series of working standard solutions were prepared by diluting standard stock solution. HNO₃ and NaOH aqueous solutions were used to adjust pH. Water samples were from Donghu Lake (Wuhan, Hubei Province, China), Yangtze River and tap water (in the laboratory). All solutions were prepared with ultra pure water.

2.2 Methods

2.2.1 Principle. In this method, the reaction of KBH₄ and HCl generates hydrogen. AsH₃ is generated by the reaction of As(III) in the solution and hydrogen. It is carried to the atomizer by argon. At the temperature of 170 °C, AsH₃ is broken down into hydrogen and arsenic atom. The concentration of arsine is determined by AFS. Under the condition of this experiment, only As(III) can be accurately determined by the apparatus. Thiourea-ascorbic acid mixed solution was added. Thiourea can make As(V) totally invert to As(III). Ascorbic acid makes the system more stable and helps with determining the results accurately.

2.2.2 Determination of adsorption rate for arsenic. 1.00 mL working solution (1.000 mg L⁻¹) of As(III) and about 80 mL ultra pure water was added into a centrifuge. When the solution was mixed well, a certain dosage of nanometre-size TiO₂ was also added. A certain pH value of the mixed solution was adjusted by using HNO₃ and NaOH aqueous solutions. The solution was centrifuged for 20 min with 4000 rpm. 10.00 mL thiourea-ascorbic acid mixed solution and 2.00 mL concentrated hydrochloric acid was added to the supernatant, then the mixture was diluted to 100.00 mL with ultra pure water in a volumetric flask. The As(III) content of the supernatant was determined by AFS and compared to the same concentration of working solution (10.00 μg L⁻¹) without adsorption to calculate the adsorption rate of TiO₂ colloid for As III). The adsorption rate for As(V) was determined by following the steps above.

2.2.3 Enrichment and determination of arsenic in water samples. After air pumping filtration by using a 0.45 μm filtration membrane, 100.00 mL of water samples was collected in the centrifuge tube. 6.00 mL of 0.2% Nanometer-size TiO₂ colloid was then added into the centrifuge tube. The pH value of the solution was adjusted to a certain value within a range from 6.0 to 7.0. In order to prepare the determined sample, the sequence of mixing well, centrifugation at 4000 rpm for 20 min, removal of the supernatant fluid was required. Then, a small amount of HCl (2%) was added to make TiO₂ into a colloid. After another addition of 0.50 mL thiourea–ascorbic acid mixed solution, the solution was diluted to 5.00 mL by HCl (2%). The colloid solution mentioned above was directly determined by AFS. (This way of sampling was usually called slurry sampling.) Finally, the content of dissolved inorganic arsenic in the water samples was calculated.

3. Results and discussion

3.1 Effect of pH

The particles of Nanometer-size TiO₂ colloid were well dispersed when the pH value was lower than 4.0 (the isoelectric point of TiO₂), and they could not be effectively separated. In order to separate the Nanometer-size TiO₂ colloid with adsorption of As from solution, the pH value should be higher than 4.0 theoretically. And experiments showed that satisfactory centrifugal effect could be attained with a pH value higher than 5.0 and a centrifugal speed of 4000 rpm.

The adsorption rates of different pH values were determined with the amount of Nanometer-size TiO₂ colloid being 6.00 mL. The results were shown in Fig. 1.

Fig. 1 Effect of pH on adsorption.

It can be seen that the adsorption rate could reach 98.0% or higher when the pH value is between 6.0 and 7.0. The pH was shown to have a significant influence on the adsorption of metal ions. Different results were reported for the adsorption of As(III) and As(V) on the metal oxide surface.^19–21 Generally, the adsorption amount of As(III) and As(V) was smaller in alkaline solution than that of acidic solution. At acidic pH the surface has a net positive charge that would attract arsenite and arsenate negative ions. At basic pH the surfaces has a net negative charge that would cause an electrostatic repulsion between the surface and arsenite and arsenate negative ions. However, Dutta et al. found the adsorption of As(III) onto TiO₂ particles increased with increases in pH, which was opposite to the adsorption behavior found for As(V).²⁰ Actually, for the adsorption of As(V), the adsorption between pH 5.0 and 6.0, was also negligible. When use ferrihydrite as the adsorbent, there was a trend of increasing adsorption with increasing pH for As(III), with maximum adsorption at pH 8.2–10. As for As(V), a different trend was observed, with a broad adsorption maximum at pH 3.8–7.4.²¹ Therefore, we think it was difficult to explain the mechanism at this moment and further work has already been carried out in our lab.

3.2 Effect of the amount of nanometre-size TiO₂ colloid

The adsorption rates of different volumes of nanometre-size TiO₂ colloid were determined. The adsorption rate reached 98.0% when the volume of nanometre-size TiO₂ was 6.00 mL. Therefore, 6.00 mL of nanometre-size TiO₂ colloid (0.2%) was adopted in this experiment. The results were showed in Fig. 2.

Fig. 2 Effect of nanometre-size TiO₂ colloid dosage on adsorption.

3.3 Adsorption capacity

A series of concentration of standard solution of As(III) were prepared with addition of 6.00 mL nanometre-size TiO₂ colloid (0.2%) respectively. Several adsorption rates were calculated and a curve of adsorption amounts of As(III) with initial concentration was attained. The adsorption capacity of nanometre-size TiO₂ colloid (0.2%) for As(III) was 111.9 μg mL⁻¹. The results were showed in Fig. 3.

Fig. 3 Adsorption capacity of nanometre-size TiO₂ colloid on As(III).

3.4 Effect of coexistent ions

A certain amount of different ions were added to 100.00 mL solution containing 0.10 μg L⁻¹ (the content of arsenic after being concentrated for 20 times was 2.00) arsenic. The solution then contained K⁺ (4.2 mg), Na⁺ (110.4 mg), Ca²⁺ (4.4 mg), Mg²⁺ (13.2 mg), Sr²⁺ (81.0 μg), Zn²⁺ (24.0 μg), Fe³⁺ (60.0 μg), Al³⁺ (38.0 μg), Mn²⁺ (20.0 μg), Cl⁻ (198.8 mg), SO₄²⁻ (26.6 mg), NO₃⁻ (1242 μg), PO₄³⁻ (260 μg), Br⁻ (673 μg) and F⁻ (13.0 μg). The fluorescence was determined under the procedure of enrichment experiment. The recovery of the samples was above 97.5%. It showed that there was almost no influence on the determination of arsenic.

In the control group, 2.00 μg L⁻¹ of arsenic with coexistent ions was directly determined without concentration. The recovery of the samples was about 30%. The coexistent ions showed significant effect on the detection and the obtained result was not accurately.

By comparing the data of these two groups, this method of concentration and separation can accurately determine the content of As in the composition of complex samples. This indicates that the method has strong anti-ion interference ability.

3.5 Detection limit and relative standard deviation

Under the optimum experimental conditions, 100.00 mL of standard solution (0.05 μg L⁻¹) of As(III) was enriched. Its fluorescence then were determined by AFS (n = 6). The detection limit (3σ) and the relative standard deviation (R.S.D) of this method were 10.6 ng L⁻¹ and 7.04%, respectively.

3.6 Analysis of standard reference material

To study the accuracy of this method, an environmental water sample certified reference material (GSBZ 50004-88) with an arsenic content of 5.03 ± 0.34 μg L⁻¹, was analyzed. The obtained arsenic content value 4.89 ± 0.11 μg L⁻¹, based on the average of four replicates, was in good agreement with the certified value. The standard curve was obtained with the samples determined in suspended state with slurry sampling, with an R² of 0.9990.

3.7 Environmental water samples analysis and standard adding recovery experiments

The above-mentioned method was adopted to determine arsenic in several water samples, including Donghu Lake, Yangtze River and tap water. The results were listed in Table 2.

Table 2 The results for tests of addition recovery for dissolved inorganic As determination in water samples (sample volume: 100 mL, final volume: 5 mL (n = 3))

sample	As added/μg L^−1	As found/μg L^−1	Recovery (%)
Donghu Lake	0	0.167	—
	0.20	0.362	97.5
Yangtze River	0	0.063	—
	0.10	0.159	96.0
Tap water	0	0.0268	—
	0.04	0.0645	95.3

---

4. Conclusion

This method adopted nanometre-size TiO₂ colloid as adsorbent for separation and concentration, and used AFS with slurry sampling for determination of arsenic in water samples. Short-time separation and enrichment, low detection limit, free of interference, simple operation and fast analysis are the biggest advantages of this method. Compared with the traditional method which needs a desorption process, the method introduced in this paper appears to be more advantaged. Consequently, it can be applied to determine ultra-trace amount of arsenic in environmental water samples with great success.

Acknowledgements

This research was financially supported by the National Natural Science Foundation of China (Grant No. 20877059) and National Key Technology R&D Program in the 11th Five year Plan of China (Grant NO.2006BAJ08B07).

References

1. B. K. Mandal and K. T. Suzuki, Talanta, 2002, 58, 201–235.
2. R. S. Oremland and J. F. Stolz, Science, 2003, 300, 939–944.
3. A. H. Smith, P. A. Lopipero, M. N. Bates and C. M. Steinmaus, Science, 2002, 296, 2145–2146.
4. P. L. Smedley and D. G. Kinniburgh, Appl. Geochem., 2002, 17, 517–568.
5. E. Muñoz and S. Palmero, Talanta, 2005, 65, 613–620.
6. X. Chen, J. Z. Li and Z. F. He, Shanghai Geol., 2006, 1, 24–27, in Chinese.
7. X. G. Pang and J. C. Zhan, Shandong Land Resour., 2005, 5, 1–5, in Chinese.
8. S. T. Stern and S. E. McNeil, Toxicol. Sci., 2007, 101, 4–21.
9. P. Liang, J. Cao, R. Liu and Y. Liu, Microchim. Acta, 2007, 159, 35–40.
10. M. Uehara, B. Barbara, B. Dieny and P. C. E. Stamp, Phys. Lett. A, 1986, 114, 23–26.
11. E. Vassileva and K. Hadjiivanov, Fresenius J. Anal. Chem., 1997, 357, 881–885.
12. P. Liang, Y. C. Qin, B. Hu, C. X. Li, T. Y. Peng and Z. C. Jiang, Fresenius J. Anal. Chem., 2000, 368, 638–640; P. Liang, Y. C. Qin, B. Hu, T. Y. Peng and Z. C. Jiang, Anal. Chim. Acta, 2001, 440, 207–213.
13. S. H. Qian, Y. Luo, S. B. Mo and X. Weng, Wuhan Univ. J. Nat. Sci., 2006, 11, 437–440.
14. F. Y. Zheng, S. H. Qian, S. X. Li, X. Q. Huang and L. X. Lin, Anal. Sci., 2006, 22, 1319–1322.
15. R. Liu and P. Liang, J. Hazard. Mater., 2008, 152, 166–171.
16. P. Matúš, I. Hagarová, M. Bujdoš, P. Diviš and J. Kubová, J. Inorg. Biochem., 2009, 103, 1473–1479.
17. L. Zhang, Y. N. Wang, X. J. Guo, Z. Yuan and Z. Y. Zhao, Hydrometallurgy, 2009, 95, 92–95.
18. S. H. Qian, X. Q. Li, M. Xiao, H. B. Deng and L. J. Xiang, Chinese. Chem. Lett., 2006, 17, 933; S. H. Qian, S. J. Zhang, Z. Huang, M. Xiao and F. Huang, Microchim. Acta, 2009, 166, 251–254.
19. H. Jézéquel and K. H. Chu, Environ. Chem. Lett., 2005, 3, 132–135.
20. P. K. Dutta, A. K. Raya, V. K. Sharma and F. J. Millero, J. Colloid Interface Sci., 2004, 278, 270–275.
21. K. P. Raven, A. Jain and R. H. Loeppert, Environ. Sci. Technol., 1998, 32, 344–349.

---