A biomaterial based approach for arsenic removal from water

Abstract

We demonstrate that the non-living, dried roots of the water hyacinth plant [Eichhornia crassipes (Mart.) Solms] can rapidly remove arsenic from water. Atomic absorption spectrometry was used to demonstrate that more than 93% of arsenite (As(III)) and 95% of arsenate (As(V)) were removed from a solution containing 200 μg As l⁻¹ within 60 minutes of exposure to a powder produced from dried roots. No difference in removal efficiency was observed between the two oxidation states of As studied. The amount of arsenic remaining in solution was found to be less than 10 μg l⁻¹ which is the WHO guideline limit value for As in drinking water. The presence of arsenic in drinking water in a number of countries in the developing world has been found to be much higher than the WHO level, affecting the health of millions of people. In this paper, we show that a biomaterial produced from dried water hyacinth roots, a plant that is found in abundant supply in many parts of the world, can provide a simple, effective and yet cheap method for removing arsenic from contaminated water.

---

Introduction

High doses of arsenic are known to be toxic to humans, but chronic, long term exposure to inorganic arsenic can also be detrimental to health. Exposure to low concentrations of As can affect the gastrointestinal tract, respiratory tract, skin, liver, cardiovascular, hematopoietic and nervous systems. Some of these human health effects are currently being observed in populations in Argentina, Bangladesh, China, India, Mexico and Taiwan where the drinking water is contaminated by arsenic.¹ The WHO guideline² value for arsenic in drinking water is 10 μg l⁻¹. The US EPA recently approved new guidelines,³ which will phase in a maximum contaminant level of 10 μg l⁻¹. The EU As in drinking water limit value⁴ is also 10 μg l⁻¹. In Bangladesh alone, more than 60% of the groundwater contains arsenic concentrations greater than the WHO guideline value⁵ and this is estimated to affect the health of millions of people.

A range of different methods can be used for the removal of arsenic from drinking water and patents protect a number of these technologies. A selection of different approaches to this problem includes the use of: natural zeolites;⁶ ion exchange resins;⁷ natural products, such as chitosan and chitin;⁸ fly-ash;⁹ and activated alumina and other oxides.^10,11 However, cost effective, straightforward methods are necessary for the people in developing countries, particularly technologies utilising materials that are widely available. Nikolaidis et al.¹² suggested a simple filter using sand and iron filings, which was based on the precipitation of arsenic with iron. Other potential approaches would include the use of phytoremediation, whereby green plants are used to remove pollutants from the environment and render them non-available¹³ or rhizofiltration, defined as the use of plant roots to adsorb, absorb, concentrate and precipitate metals from solution.¹⁴ Several plant species are known to accumulate metals from the environment,^15–18 including water hyacinth (Eichhornia crassipes), which grows in waterways in many parts of the world and has been used as a model system for studying the uptake of metal ions by aquatic plants.^19–23 These studies employed living plants, which were not shown to be highly effective in removing arsenic.²⁴ In contrast, there are two studies, which report that living plants can be effective in removing arsenic from water.^25,26 The use of biomaterial derived from non-living dried water hyacinth roots may be a more convenient and effective method for the removal of toxic metal(loid)s. Although some studies have been reported on the use of dried plant material as a potential industrial tool for metal removal,^27,28 these studies have not focused on the removal of arsenic from aqueous solutions.

An appropriate technology for arsenic removal, particularly in the context of less developed countries, should address the following criteria: (1) the technology must be of simple design and easily produced; (2) it must be low cost; (3) it must use local, easily accessible materials; (4) it must have a rural focus.¹⁶ This study presents data on the potential of using a biomaterial derived from dried water hyacinth roots as a way to remove arsenite and arsenate from aqueous solution. This could be a simple yet effective means to reduce the As content of water in many developing countries, like Bangladesh and India, where this plant is available in abundant supply. Arsenic in drinking and irrigation water is a major problem in both of these latter countries. Indeed, we have recently published a paper which reported presence of high arsenic content in vegetables grown in Bangladesh.²⁹ The source of arsenic in the vegetables is most likely due to use of arsenic contaminated irrigation water. The arsenic removal method presented here could be readily applied to remove arsenic from such irrigation water, thereby reducing the entry of arsenic into the food chain.

Materials and methods

Reagents

A 100 mg l⁻¹ arsenic(III) stock solution was prepared from sodium arsenite Na₃AsO₃ (BDH, Dorset, UK) and a 100 μg l⁻¹ arsenic(V) stock solution was prepared from a 1000 ± 3 μg l⁻¹ standard As(V) solution (CPI, International, USA). The arsenic solutions used in the experiments were prepared by appropriate dilution of the stock solutions with deionised water. All the other reagents used were analytical or general reagent grade.

Sample preparation

Water hyacinth (Eichhornia crassipes) plants were collected from a pond in Dhaka, Bangladesh. The plants were thoroughly washed with deionised water before being left to air dry. The dried roots were subsequently pulverized in a laboratory blender to obtain a fine powder.

Instrumentation

An atomic absorption spectrometer (Varian model 220-Z) was used in conjunction with a graphite furnace atomizer (GTA-110) and Zeeman background correction. The 193.7 nm wavelength generated by a single element hollow cathode lamp (Cathodeon Ltd, Nuffield Road, Cambridge) was used with a current of 12.0 mA and a slit width of 0.2 nm. The peak height measurement mode was used along with Zeeman background correction, for all the standard and sample measurements, which were performed in triplicate. The temperature programme used is given in Table 1. The palladium modifier used was prepared by dissolving Pd(II) acetate in 1% nitric acid (v/v) with gentle heating for 4 h, followed by a volumetric adjustment to a final concentration of 1000 mg l⁻¹. The matrix modifier was added to the graphite furnace by the autosampler, at the same time as the sample. The instrumental parameters and temperature programme used for the GF-AAS analysis were those supplied in the instrumental software.

Table 1 Graphite furnace AAS temperature programme used for the measurement of total As

Step	Temperature/°C	Time/s	Gas type	Gas flow rate/l min^−1	Read
1	85	5.0	Nitrogen	3.0	No
2	95	40.0	Nitrogen	3.0	No
3	120	10.0	Nitrogen	3.0	No
4	1400	5.0	Nitrogen	3.0	No
5	1400	1.0	Nitrogen	3.0	No
6	2600	2.0	Nitrogen	0.0	No
7	2600	0.6	Nitrogen	0.0	Yes
8	2600	2.0	Nitrogen	0.0	Yes
9	2600	2.0	Nitrogen	3.0	No

---

Quality assurance and quality control

The method was validated by analysis of a seaweed CRM (CRM 9, NIES, Japan) which produced a quantitative recovery (mean of 6 acid digests 119.8 mg kg⁻¹ std. dev. 8.9, certified value for As 115.0 mg kg⁻¹ std. dev. 9.0). The recovery of 25 μg l⁻¹ As(V) spiked into tap water was also quantitative. Calibration of the instrument was optimized between 3.0 and 26.0 μg l⁻¹ and calibration curves were fitted by the computer software with correlation coefficients better than 0.999 in all cases. The limit of detection was determined by analysis of ten replicate tap water blank samples and determined to be 0.25 μg l⁻¹ (LOD = 3 × std. dev. of blank concentration n = 6). During the sample analysis tap water blanks were analysed along with a check standard (10 μg l⁻¹) which was run every 6th sample and recalibration was not required during any of the sample analysis runs.

Experimental procedure

A specific weight (750 mg) of water hyacinth root powder was suspended in 25 ml of tap water spiked with 200 μg l⁻¹ of arsenite [As(III)] or arsenate [As(V)] and adjusted to pH 6. The appropriate pH of the solution was obtained by addition of either nitric acid (10% v/v) or ammonium hydroxide (10% v/v). The mixture was shaken for 120 minutes using a flask shaker on a medium setting, at room temperature. At different time intervals, the roots were removed by filtration (Millipore 0.45 μm pore size membrane filter) to obtain the supernatant solution, which was analysed by GF-AAS.

In this preliminary study the dried water hyacinth roots were evaluated using a static batch system rather than using columns packed with the material. This helps to eliminate any problems related to column blockage, non-equilibrium conditions or uncertainties associated with non-uniform flow in a dynamic system. To determine the most appropriate mass-to-solution volume ratio to use in the experiments a range of different test solutions were prepared, containing a fixed volume of arsenate (200 μg l⁻¹), but with increasing mass (0.2–4.5% w/w) of root material.

Results and discussion

The results presented in Fig. 1 indicate that quantitative removal of arsenate from a solution at pH 6.0 was achieved in 60 minutes, using 30 to 45 mg ml⁻¹ of dried root material. A mass-to-volume ratio of 30 mg roots ml⁻¹ solution was therefore adopted in the further batch experiments used to evaluate the material as a bioadsorbent. The jump in As removal between 25 and 30 mg ml⁻¹ does not fit the expected removal characteristics of solid absorbents reported previously for As⁶ or transition elements.¹⁴ In these cases metal removal was modeled using simple linear or curvilinear plots that were asymptotic in nature. These characteristics are present in Fig. 1, but the sudden increase noted between 25 to 30 mg ml⁻¹ could be attributable to a number of different effects. One possibility is that these solutions contain markedly different particle size distributions in comparison to the other solutions, which would affect the uptake of As because of the differences in surface area and therefore exposed absorption sites.

Fig. 1 The effect of varying the mass of water hyacinth roots on removal of arsenic as As(V) from a 200 μg As l⁻¹ solution at pH 6.0 after 1 h exposure. The error bars represent the standard deviation of three replicate measurements.

The removal of arsenite and arsenate from aqueous solution using dried, non-living, water hyacinth root powder as a function of time is shown in Fig. 2. Over 80% of the arsenic was taken up within 30 minutes exposure and 96% of both chemical species of arsenic were removed within one hour, leaving solutions containing 9.50 μg l⁻¹ As(III) and 7.85 μg l⁻¹ As(V), which are less than the WHO guideline value of 10 μg l⁻¹. The removal rate was rapid within the first 30 minutes, slowing down between 30–40 minutes and then gradually approaching equilibrium after 40 minutes. This could be due to two different sorption processes, a fast ion exchange process followed by chemisorption, similar to that which has been previously suggested for sorption of lead and cadmium by non-living water hyacinth roots.²⁸

Fig. 2 Arsenic(III) and arsenic(V) removal by dried water hyacinth roots as a function of time (30 mg roots ml⁻¹ at pH = 6.0 and 200 μg As l⁻¹ as the initial metalloid concentration in solution). The error bars represent the standard deviation of three replicate measurements.

Arsenic removal varied with the pH of the aqueous solution (Fig. 3) and was optimal between pH 2.5 and 8.0, which covers the normal pH range for water. The low percentage removal at pH 1.0 cannot be readily explained because of the limited characterization of the material that was undertaken during this initial study. However, it has been reported²⁸ that water hyacinth roots are negatively charged due to the presence of carboxylate, sulfate, amino and other groups on the root surface. At low pH neutralisation of the negative groups on the root surface, as a result of interaction with H⁺ ions, will reduce the number of negatively charged groups available for interaction with positively charged metal ions. This could result in desorption of the cations present on the surface which will then be available to complex with the oxyanionic arsenic species present in solution. The exact chemical speciation of the soluble inorganic arsenic will also be dependent on the prevailing pH (and pε) conditions. Under the redox conditions present in this experiment both arsenic oxidation states will be present as the corresponding oxyanion species and these will vary in protonation with pH. At lower pH values both species will be fully protonated and exist as the non-ionised acid.

Fig. 3 The effect of pH on arsenic removal as As(V) by dried water hyacinth roots, 30 mg roots ml⁻¹, using a constant concentration of 200 μg As l⁻¹, for a total contact time of 4 h. The error bars represent the standard deviation of three replicate measurements.

The need for materials to remove arsenic from waters used for drinking and irrigation of crops has been recognized for a number of years and many different materials have been developed for this purpose. Some natural zeolites from Mexico and Hungary were investigated for removing arsenate (As(V)) from aqueous solution.⁶ After 5 days contact time with different spiked water samples both zeolites removed 75% of the 200 μg l⁻¹ added arsenate. A chitosan and chitin mixture was shown to remove As(V) from groundwater⁸ and the capacity of the mixture at pH 7 was recorded to be 10 μg of arsenic per gram of mixture. Precipitation with iron oxides and hydroxides is another processes commonly used for treating water in some parts of the world affected by As. Recently, Katsoyiannis et al.³⁰ have reported the use of a combination of iron and oxidizing bacteria to remove 65% of As(III) from solutions containing initial arsenic concentrations over 150 μg l⁻¹. In comparison to these approaches the biomaterial reported herein achieved better results than the zeolites in a shorter time period, displaying 5-fold better removal efficiency than the chitosan/chitin mixture and a higher efficiency compared to the iron precipitation method.

Water hyacinth is considered to be an environmental “plague” throughout many tropical and subtropical parts of the world and in Bangladesh, and India, the plant grows abundantly in ponds, lakes and rivers. It has been reported²⁸ that water hyacinth biomass can be produced at a rate of 160 to 1000 kg hectare⁻¹ day⁻¹ depending on the nutrient and temperature conditions. In this study, the water hyacinth roots exposed to 1500 μg As l⁻¹ (Fig. 4) removed up to 50 μg As g⁻¹ of roots. Assuming an arsenic solution concentration of 1500 μg l⁻¹, approximately 30 grams of dried roots would be required to remove 1500 μg of arsenic from 1 litre of water over a 24 hour time period. Therefore, to treat 1000 litre of water 30 kg of roots would be required. This would be sufficient to provide the drinking water needs of a village consisting of hundreds of people. The dried powder could be used to produce cheap filters similar to the simple three-pitcher filtration method reported by Khan et al.³¹ However, instead of using iron chips, necessary for the method developed by the latter workers, people could use dried water hyacinth roots. This approach can be particularly attractive to people who do not have access to more expensive alternatives. Production of this biomaterial is inexpensive and the raw material is widely available. Bearing in mind the poor infrastructure of most developing countries, it is essential to find low-cost techniques suitable for the removal of toxic elements from drinking and irrigation waters in rural areas. The use of dried water hyacinth roots may be one way to achieve this goal.

Fig. 4 Arsenic removal by dried water hyacinth roots as a function of arsenic concentration at a constant mass of water hyacinth roots (30 mg roots ml⁻¹, at pH 6.0 and after 24 h of exposure). The error bars represent the standard deviation of three replicate measurements. Symbols as shown in Fig. 2.

Notes and references

1. M. E. Cebrian, A. Albores, G. Garcia-Vargas and L. M. Del Razo, in Arsenic in the Environmental Part II: Human Health and Ecosystem Effects, ed. J. O. Nriagu, Wiley, New York, 1994, p. 93.
2. WHO Arsenic compounds, Environmental Health Criteria 224, World Health Organization, Geneva, 2001.
3. US Environmental Protection Agency, Arsenic in Drinking Water Rule: Economic Analysis, EPA document 815-R-00-026, Washington, DC, USA, 2000.
4. Council of the European Union, Council Directive 98/83/EC of November 1998 on the quality of water intended for human consumption, Off. J. Eur. Communities, May 12, 1998, pp. L330/32–L330/52.
5. K. Tanabe, H. Yokota, H. Hironaka, S. Tsushima and Y. Kubota, Appl. Organomet. Chem., 2001, 15, 241–251.
6. M. P. Elizalde-Gonzalez, J. Mattusch and R. Wennrich, J. Environ. Monit., 2001, 3, 22–26.
7. W. H. Ficklin, Talanta, 1983, 30, 371–373.
8. C. M. Elson, D. H. Davies and E. R. Hayes, Water Res., 1980, 14, 1307–1311.
9. D. Goswami and A. K. Das, Int. J. Water, 2000, 1, 61–70.
10. D. Clifford, Environ. Sci. Technol., 1986, 20, 1072–1080.
11. K. P. Raven, A. Jain and R. H. Loeppert, Environ. Sci. Technol., 1998, 32, 344–349.
12. N. P. Nikolaidis and J. Lackovic, Arsenic remediation technology – AsRT, presented at International Conference on Arsenic Pollution of Groundwater in Bangladesh: Causes, Effects and Remedies, Dhaka, Bangladesh, 8–12 February 1998.
13. D. E. Salt, R. D. Smith and I. Raskin, Ann. Rev. Plant Physiol. Plant Mol. Biol., 1998, 49, 643–668.
14. V. Dushenkov, N. B. P. A. Kumar, H. Motto and I. Raskin, Environ. Sci. Technol., 1995, 29, 1239–1245.
15. D. E. Salt, M. Blaylock, N. B. P. A. Kumar, V. Dushenkov, B. Ensley, I. Chet and I. Raskin, Biotechnology, 1995, 13, 468–474.
16. D. E. Salt, I. J. Pickering, R. C. Prince, D. Gleba, S. Dushenkov, R. D. Smith and I. Raskin, Environ. Sci. Technol., 1997, 31, 1636–1644.
17. S. L. Brown, R. L. Chaney, J. S. Angle and A. J. M. Baker, J. Environ. Qual., 1994, 23, 1151–1157.
18. A. M. A. Mazen and OMO El Maghraby, Biol. Plant., 1998, 40, 411–417.
19. C. Kelley, R. M. Mielke, D. Dimaquibo, A. J. Curtis and J. G. Dewitt, Environ. Sci. Technol., 1999, 33, 1439–1443.
20. O. Saltabas and G. Akcin, Toxicol. Environ. Chem., 1994, 41, 131–134.
21. S. Sinha, R. Saxena and S. Singh, Environ. Monit. Assess., 2002, 80, 17–31.
22. M. E. Soltan and M. N. Rashed, Adv. Environ. Res., 2003, 7, 321–334.
23. K. S. Low, C. K. Lee and C. H. Tai, J. Environ. Sci. Health, 1994, 29, 171–188.
24. F. E. Chigbo, R. W. Smith and F. L. Shore, Environ. Pollut., Ser. A, 1982, 27, 31–36.
25. Y.-L. Zhu, A. M. Zayed, J.-H. Qian, M. de Souza and N. Terry, J. Environ. Qual., 1999, 28, 339–344.
26. M. I. R. Misbahuddin and A. T. M. Fariduddin, Arch. Environ. Health, 2002, 57, 516–518.
27. Y. Hao, A. L. Roach and G. L. Ramelow, J. Environ. Sci. Health, 1993, 28, 2333–2343.
28. G. X. Wang, M. C. Fuerstenau and R. W. Smith, Miner. Metall. Process., 1999, 16, 1–47.
29. S. W. Al Rmalli, P. I. Haris, C. F. Harrington and M. Ayub, Sci. Total Environ., 2005, 337, 23–30.
30. I. A. Katsoyiannis and A. I. Zouboulis, Water Res., 2004, 38, 17–26.
31. A. H. Khan, S. B. Rasul, A. K. M. Munir, M. Habibuddowla, M. Alauddin, S. S. Newaz and A. Hussam, J. Environ. Sci. Health, 2000, 35, 1021–1041.

---