• The second quantum revolution is building the technology of the futureRead More
  • EditorialRead More
  • Green silver nanoparticles: Synthesis, characterization and usesRead More
  • Arsenite removal with Fe3O4, MnFe2O4 and CoFe2O4 NanoparticlesRead More
  • SENTINELRead More

Magnetic nanoparticles in the removal of arsenic from water

Technical feasibility of using magnetic nanoparticles obtained from metallic wool for arsenite (As(III)) removal from aqueous solutions

María Teresa Alarcón

Miriam López

It is necessary to develop new and economic alternatives for arsenic (As) removal by adsorbents. The aim of this study was to investigate the technical feasibility of using magnetic nanoparticles obtained from metallic wool for arsenite (As(III)) removal from aqueous solutions. The obtained adsorbent was characterized by different methods (TEM, BET, X-Ray diffraction, VMS). The adsorbent was identified as a microporous nanomaterial, composed mainly by ferric oxyhydroxide (γ-FeO(OH)) commonly known as lepidocrocite. The surface area was 88.30 m2/g. A removal efficiency of 100% was achieved in seven minutes with 0.55 g/L of γ-FeO(OH) under local environmental conditions (Water pH=7.8±0.2 and 23±3 oC). The adsorption experimental results fitted well (R2 = 0.99) with Freundlich, Langmuir, and Dubinin-Radushkevich isotherms. The adsorption capacity was 2.2 mg/g. The optimum magnetic field strength for magnetic filtration was 0.24 Teslas. Since magnetic nanoparticles with the adsorbed arsenic can be separated from water, this constitutes a promising process for water treatment.

Globally one of the biggest problems that affect the quality of drinking water is the presence of arsenic in drinking water sources. Worldwide, it is estimated that more than 300 million people worldwide [1]  drink water with arsenic levels exceeding the permissible limit (10 µg/L), and 50 million people would be drinking water with levels greater than 50  µg/L of arsenic [2]. Approximately 14 million people are at a risk situation by drink water with high levels of arsenic in South America [3].

Several countries reported the severity of the risk inherent in the ingestion of water containing arsenic, among these countries are: Argentina, Mexico, The United States, Hungary, India, Italy, China, Pakistan, Taiwan, Bangladesh, Vietnam, and Chile [4, 5, 6]. Chronic exposure to high arsenic concentrations might cause arsenicosis or HACRE (Hydroarsenicism Regional Chronic Endemic), a disease with a high occurrence in Asia and Latin America [7].

The World Health Organization (WHO) has classified inorganic arsenic as a Group A, human carcinogen [8]. Both WHO and EPA recommend10 µg/L as the maximum concentration of total arsenic in drinking water. In Mexico, the NOM-127-SSA1-1994 has established a maximum concentration of arsenic in drinking water of 25 µg/L [9]. To approach the problem, there are different technologies such as the conventional water treatments (oxidation, ion exchange, coagulation-precipitation, reverse osmosis, adsorption) and the so-called emerging technologies (phytoremediation, electrocoagulation, use of nanoma-terials as adsorbents) [10]. However all of these have advantages and limitations, therefore the search for new technologies is required.

Recently, there has been a growing interest in the development and application of nanomaterials for water treatment. Considering that micro and macro iron oxide particles are effective adsorbent mediums (99.95% of As+5 and 98% of As+3), and economically affordable for arsenic removal [11], the present research investigates the technical feasibility of using iron nanoparticles produced from a recycled material for arsenic removal from water, taking advantage of the magnetic properties of the material for their separation from the aqueous medium by using a magnetic field. Magnetic nanoparticles composed of iron oxyhydroxide (FeO-OH) are a promising adsorbent for removing As (III) from water [12].

Preparation of Arsenic solutions

Arsenic (III) stock solutions were prepared from NaAsO2 reagent grade (Fisher Scientific Laboratories). In order to work under conditions closer to reality, solutions were prepared with groundwater containing arsenic. The final arsenic concentrations of the prepared working solutions were 113, 313, 513, 613, 713 and 913 µg/L, respectively.

Adsorbent preparation

In the present study, steel wool was used to produce nanoparticles with high iron content and good magnetic properties. The material was subjected to an extensive washing process, and then was periodically moistened with water to induce its oxidation. The oxidized particles were collected and screened to remove coarse material (# 400 mesh). Following this, a suspension was prepared with screened material using deionized water, which was dark. This slurry was allowed to precipitate for 12 hours. After that, the supernatant was recovered and placed in a separating funnel for a period of 48 hours. The sedimentation of particles was accelerated by the use of permanent magnet plates. Finally, the settled material was dried for a period of seven days, under local environmental conditions.


Different amounts of nanoparticles (0.2 to 1.2 g/L) were placed in contact with arsenite solutions for 0.5 to 10 minutes. To improve the interaction between the ions of arsenic and the adsorbent, an agitator with an ultrasonic frequency of 40 kHz was used (Branson 2510 Ultrasonic Cleaner). The As (III)-solution was then fed through the magnetic filtration process. The flow rate was controlled at a low flow rate of 1 mL/min using a Masterflex peristaltic pump. The filtered samples were prepared for quantification for arsenic. The adsorption process and magnetic filtration were performed at room temperature (23 ± 3 °C), and with a natural water pH of (pH 7.8 ± 0.2).

Magnetic Filtration

The separation process or magnetic filtration was carried out using an electromagnet, from the Electromagnetic Testing Laboratory of the Research Center in Advanced Materials, at different magnetic field strengths. The electromagnet with a maximum magnetic capacity of 0.4 teslas and an 8 mm distance between their cylindrical polar caps was calibrated with a Walker Scientific MG-3DP Gaussmeter. The voltage ranged between 0 and 25 volts and the current varied from 0 to 0.16 ampers). The magnetic field range was decided to be lower than reported (0.3 Teslas) (10.Cafer T. Yavuz, et al., 2006) in order to decrease the energy requirement of the process, and to evaluate the possibility of replacing the electromagnet system by permanent magnets.

The column used for magnetic filtration was filled with stainless steel fine wool (magnetic grade) commercialized by SoBo Distribution Inc. The glass column was 3.6 cm high and had a diameter of 6 mm. The mass of fine wool was 0.7 g.

Characterization of nanoparticles

The surface area of the nanaoparticles was measured by the Brunauer-Emmett-Teller (BET) method using a Quantachrome (Nova Corporation Series 1000). The samples were degassed under vacuum at 250 °C. The crystalline structure of the adsorbent was determined by a PANalytical X-ray diffraction equipment (X’Pert PRO model) equipped with an X’Celerator detector. The step was 0.05° and the angle 2θ ranged between 10º and 80º. The detection limit of the equipment was 0.1%.

The elemental analysis and the particle size were determined using a transmission electron microscope (TEM) JEOL, JEM-2200FS model, with STEM+Cs corrector. The resolution of the device was 0.187 nm.

The magnetic properties of nanoparticles and that of the fine stainless steel wool (hysteresis curve, coercive field, saturation magnetization) were determined with a vibrating sample magnetometer (VSM) LDJ brand, model 9600.

Analytical determination

Before the arsenic analytical determinations, samples were digested in a microwave Marsx CEM Corporation, Model 3100. Total arsenic concentrations were determined using a GBC atomic absorption spectrophotometer (AAS) (model Avanta Sigma) coupled to a hydride generator (HG). This equipment was calibrated with reference standard solutions (High Purity Control Standards) traceable by The U.S. National Institute of Standards and Technology (NIST). The average percentage of analyte recovery was 99.8% ± 1.8%. The lower limit of detection of the equipment was 5 µg/L. The total iron concentration was determined with a plasma emission spectrometer (Thermo Electron) with a mass detector (ICP-MS). The pH, temperature and conductivity of the solutions were measured by Orion pH-meter, model 1260.

Table 1. Characterization of the adsorbent



BET surface area (m2/g)


Average particle size range (nm)

5 -150

Density (g/cm3), aproximately


Porosity (mainly)


Pore size (nm)

< 2.0

pH pzc



γ-FeO(OH), Fe2O3, MnMoO4.H2O


The physicochemical characteristics of the nanoparticles are shown in Table 1. The specific surface area was 88.30 m2/g. According to the BET isotherm obtained and according to the isotherm classification of Brunauer, Deming, Deming, and Teller, the adsorption curve is very similar to the type IV. The hysteresis loop of the isotherm indicates the presence of pores in the material. The pore size distribution obtained by the Dubinin-Astakhov and BJH (Barrett-Joyner-Halenda) methods suggests a mainly microporous material. According to the IUPAC classification, the pore diameter should be less than 2 nm. X ray diffraction analysis shows that the adsorbent is composed mainly of ferric oxyhydroxide known as lepidocrocite (γ-FeO(OH)) and Fe2O3 (Fig.1). The Inorganic composition of the adsorbent was Fe (50.7%), O (42.5%), Na (4.3%), Cl (0.9%) and Mn (1.6%).


Figure  1.  X-ray diffraction of the γ-FeO(OH) nanoparticles, (lepidocrocite).

The micrographs (Figure 2) show particles similar to flattened flakes and slightly elongated with pointed ends. It is observed that crystals aggregates form groups with feathery form. The size of lepidocrocite particles were between 100 to 150 nm long, and about 5 to 20 nm wide.


Figure 2. γ-FeO(OH)  nanoparticles micrographs.

The obtained hysteresis curve (Fig. 3) is characteristic of a ferromagnetic material. The material does not reach the saturation magnetization; this condition would indicate the presence of a paramagnetic component [13]. The remnant magnetization was 45.4 emu while the coercivity was 76.3 Oersted. These parameters revealed the remnant magnetism and the field intensity required to demagnetize the material.


Figure 3.  a) The hysteresis curve of γ-FeO(OH)  nanoparticles b) Samples, before and after magnetic filtration.

Calibration of the electromagnet

The electromagnet was calibrated over a range from 0 to 0.16 ampere (0 to 25 volts). The magnetic field was measured at the center of the dipoles (Fig. 4). A linear relationship was found between the current and the magnetic field. For a range greater than 0.02 A, the value of the coefficient of determination, R2, was close to 1. The electromagnet had a remnant magnetization of 0.024 T.


Figure 4. Relation between working current and the magnetic field of an electromagnet.

Effect of the Adsorbent Amount on As(III) Adsorption

The results are shown in Figure 5, which demonstrate the increase of arsenic removal efficiency in relation to time. At dosages higher than 0.7 g/L, 100% removal efficiency was achieved in less than four minutes. For dosages from 0.4 to 0.6 g/L, the adsorption rate was moderate. This may be due to the covering of the adsorbent surface by the arsenic molecules until it reached the equilibrium state. Equilibration time varied according to the amount of adsorbent used (higher amounts of adsorbent, need more time to reach equilibrium). At the dosage of 0.35 g/L and 4 minutes of contact time, the treated water met the water quality required by the international standards (<10 µg As/L).


Figure 5. As (III) removal with different amounts of nanoparticles (Co= 113 µg/L,  pH= 7.8±0.2, T °C= 23±3, Maximum Time= 10 min).

The arsenic removal was slower at low dosages (0.2 g/L); however, the removal efficiency continued to increase with the increase in retention time. Table 2 shows the removal efficiencies reported by other authors and those obtained in this study. Similar studies using Fe3O4 nanocrystals to remove As (III) and As (V) from water, obtained up to 99.2% of As (III) removal with 0.5 g/L of Fe3O4 in 24 hours of stirring at a pH of 8 [14]. Other researchers [15] used magnetite and maghemite nanoparticles to remove As (III), As (V), and Cr (VI) from water, reaching 96% removal efficiency of As (III). However the pH in the experiment was extremely acidic (pH=2) and a long detention time (24 h). The amounts of adsorbent used by the authors are similar to those used in this study; however, their operating time makes the process technically unfeasible for a practical water treatment process.

Table 2. As (III) removal conditions reported for other authors.



Dose (g/L)

Time (hour)



J.T Mayo et al. (2007)

Magnetite nanoparticles (Fe3O4)




99.2% As (III)

98.4% As (V)

Saidur Rahman et al. (2010)

nanoparticles of magnetita- maghemita




96% As (III)

Sen Lin et al. (2012)

Magnetic γFe2O3 nanoparticles




61.2% As (III)

Kyungsun Song et al. (2013)

nanoparticles of iron oxyde




2.9 mg/g As(III)

Present Research

Nanoparticles of  γFeO(OH)




100% As (III)

Figure 6 shows the result of the statistical analysis performed using Minitab 16. The graph shows six ranges of As(III) removal efficiency. It is possible to select different dosages and contact time. Drinking water was obtained at concentrations below 10 µg (As+3)/L, with an efficiency of 91%. In this study, the optimal treatment condition was achieved when 0.35 g/L of adsorbent was used for 4 minutes of contact time.


Figure 6.   Contour Plot of As (III) Removal Efficiency (%) vs. Adsorbent and Time (Minitab 16 statistical software English).

Effect of the initial concentration on As (III) adsorption

Figure 7 shows the removal efficiency when varying the initial concentration (Co) of arsenic for different contact times. It is observed that approximately between 10 to 20 minutes of contact time, for initial concentration (Co) ranging from 313 to 713 µg/L, dynamic equilibrium was reached between the concentration of solute remaining in solution and the concentration of solute adsorbed on the solid surface. For an initial concentration of 913 µg/L, the required time for equilibrium was greater than 30 minutes.


Figure 7. As(III) removal at different initial concentrations (Adsorbent: 0.5g/L, pH: 7.8±0.2; Contact time: 30 min, T: 23±3 oC).

The Effect of Initial pH on As(III) Adsorption

In order to determine the optimum pH for As(III) removal through adsorption, the effects of pH (4-9) were evaluated. To increase and decrease the pH, NaOH (0.1 Molar) and HCl (0.1, 1.0 Molar) were used. Between pH 6 and 7, the adsorption process was less effective in arsenite removal, while removal efficiencies higher than 93% were obtained between pH 8 and 9 (Fig. 8).


Figure 8. pH effect on the As(III) removal efficiency (Co: 113 µg/L; pH: 7.8±0.2; Adsorbent: 0.5 and 1.0 g/L, Time: 10 min, T: 23±3 oC).

According to Eh-pH diagram, the predominant arsenic species at pH 7 is H3AsO3; this neutral form would be another possible reason for the decrease in removal efficiency at this pH [18]. The effect of pH was not significant for the As(III) adsorption with lepidocrocite synthesized under different thermal treatments [19].

Arsenic adsorption on clay minerals shows that the As (V) adsorption decreased at pH 7.5, while As (III) adsorption increased [20]. That difference could be attributed to the de-protonation of H3AsO3 and H3AsO4. This phenomenon could also have happened in the current study.

Adsorption Isotherms

Adsorption isotherms allow correlating the involved variables when the adsorption process is in the state of equilibrium. The adsorption isotherms are a function of the variation in the degree of adsorption; and also depend on the concentration of the adsorbate in the solution at a constant temperature [21].

The adsorption parameters in this study were determined applying the isotherms of Langmuir, Freundlich, Dubinin Radushkevich and BET. The Langmuir model is useful for studying the physical adsorption monolayer [22], the Freundlich and BET serve  the study of multilayer adsorption and Dubinin is for the characteristic of adsorption energy and the maximum amount of pollutant adsorbed in microporous materials [23]. The constants of the isotherm models applied are presented in Table 2. A Coefficient of Determination (R2) of 0.99 was obtained for Freundlich, Langmuir and Dubinin models, respectively  while the Coefficient of Determination (R2) for the BET model was 0.89.

Table 2.  Langmuir,  Freundlich,  BET and Dubinnin adsorption parameters.
































Dubinin Radushkevich









Influence of the Magnetic Field

Figure 13 shows the influence of the magnetic field strength on the As (III) removal efficiency. It was observed that at higher magnetic field intensity, the efficiency of As (III) removal was also higher. It is possible that at the strongest field there is an increased retention of nanomaterial inside the magnetic column. Under the conditions carried out in this study, it was possible to obtain treated water that met the quality required by the international standards (As <10 µg/L) at 0.24 T.


Figure 13. As(III) removal efficiency at different magnetic field (Co: 113 µg/L, pH: 7.8±0.2, Agitation time: 10 min, Magnetic filtration time: 40 min, T: 23±3 oC).

Also, with a higher amount of adsorbent, the removal efficiency was higher. Efficiency above 93% was observed for 0.24 T, and doses of 0.4, 0.5 and 1.0 g/L, respectively.

Iron in Treated Water

The aim of this portion of the study was to enhance the potential migration of iron ions from the adsorbent into the aqueous medium and then retain the iron particles through magnetic filtration. Figure 14 shows that the variation of the iron concentration in the treated water did not depend on the amount of adsorbent used under a varying magnetic field. The Mexican standard states that the maximum allowable concentration of iron in drinking water should not be more than 0.30 mg/L. The effluent quality obtained in this study (quantified iron 0.028 mg/L) can easily meet this standard.


Figure 14. Iron concentration in treated water (Co: 113 µg/L, pH: 7.8±0.2, Agitation time: 10 min, Magnetic filtration time: 40 min, T: 23±3 oC).


This study investigated the feasibility of using magnetic nanoparticles obtained from metal wool for As (III) removal from water. The obtained material did not require any new products or further chemical processing for its production, and therefore, did not generate secondary pollutants. Arsenic was separated efficiently from treated water by applying an external magnetic field.

The surface area of the microporous material identified as lepidocrocite (γ-FeO (OH)) was 88.3 m2/g. The As (III) removal was complete using 0.55 g/L of γ-FeO(OH) in eight minutes. The experimental results showed a good fit (R2= 0.99) with the Freundlich, Langmuir, and Dubinin-Radushkevich models.

The maximum adsorption capacity of the material was 2.2 mg/g. Under the experimental conditions, at 0.24 Tesla, treated water met the quality required by the international standards (As<10 µg/L).


1] Quansah R., Ato Armah F., Essumang D.K., Luginaah I, Clarke E., Marfoh K, et al. Environ Health Perspect. 123(5), 412-21 (2015).

[2] Ravenscroft P., Brammer H., Richards K.. Arsenic Pollution: A Global Synthesis, John Wiley & Sons Ltd. Publication, United Kingdom 2009.

[3] Litter M.I., Alarcón-Herrera M.T., Arenas M.J., Armienta M.A., Avilés M., Cáceres R.E., Nery Cipriani H., Cornejo L., Dias L.E., Fernández Cirelli  A., Farfán E.M., Garrido S., Lorenzo L., Morgada M.E., Olmos-Márquez M.A., Pérez-Carrera A… Small-scale and household methods to remove arsenic from water for drinking purposes in Latin America.  Science of the Total Environment   429, 107–122 (2012).

[4] Ortega-Guerrero M.A.. Presencia, distribución, hidrogeoquímica y origen de arsénico, fluoruro y otros elementos traza disueltos en agua subterránea, a escala de cuenca hidrológica tributaria de Lerma-Chapala, México. Revista Mexicana de Ciencias Geológicas  26 (1), 143-161 (2019).

[5] Aliota, P; Celis, M; Juarez, D; Merli, G; Ricciuti, N; Salinas, N; Siles, A; Stoklas, C; Suquele, C. Potabilización de aguas subterráneas: Remoción de flúor. Seminario Agua. Editorial de la Universidad Tecnológica Nacional – U.T.N. Argentina, 2008.

[6] Fernández Turiel J. L., Galindo Griselda, Parada M. A., D. Gimeno, M. García Valles, J. Saavedra. Estado actual del conocimiento sobre el arsénico en el agua de Argentina y Chile: Origen, Movilidad y Tratamiento. IV Congreso Hidrogeológico Argentino y II Seminario Hispano Latinoamericano sobre temas actuales en hidrológica., Río Cuarto, Argentina 2005.

[7] Morgada Maria E., Levy Ivana K., Salomone Vanesa, Silvia S. Farias, Litter Marta I., Lopez Gerardo. Arsenic (V) removal with nanoparticulate zerovalent iron: Effect of UV light and humic acids. Catalysis Today  143, 261-268 (2012).

[8] World Health Organization (WHO), 2004. Guidelines for Drinking-Water Quality. In: Recommendations, third ed., vol. 1. WHO, Geneva, Switzerland. (Arsenico y fluor).

[9] Official Mexican Standard NOM-127-SSA1-1994. Environmental Health, water for human use and consumption – permissible limits of quality and treatments to which water must be submitted for its drinkability.

[10] Prasenjit Mondal, Bikash Mohanty, Chandrajit Balo Majumder. Removal of Arsenic from Simulated Groundwater Using GAC-Ca in Batch Reactor: Kinetics and Equilibrium Studies. Clean – Soil, Air, Water, 40(5), 506–514 (2012).

[11] Maiti Abhijit, Das Gupta Sunando, Kumar Basu Jayant, Sirshendu De. Adsorption of arsenite using natural laterite as adsorbent. Separation and Purification Technology 55, 350–359.

[12] Xiaolei Qu, Pedro J.J. Alvarez, Qilin Li. Applications of nanotechnology in water and wastewater treatment. Water Research  47, 3931-3946 (2013).

[13] Saux C., RenziniM.S., Bercoff P.G., Bertorello H.R, Pierella L.B. Study on the influence of the metal cation incorporation in the catalytic activity and magnetic behavior of zsm-5 zeolites. Avances en Ciencias e Ingeniería ACI, 2(2), 1-10 (2011).

[14] Mayo J.T., Yavuz C., Yean S., Cong L., Shipley H., Falkner W. Yu, J., Kan A., Tomson M., Colvin V.L. The effect of nanocrystalline magnetite size on arsenic removal. Science and Technology of Advanced Materials 8, 71–75 (2007).

[15] Saidur Rahman Chowdhury, Ernest K. Yanful. Arsenic and chromium removal by mixed magnetiteemaghemite nanoparticles and the effect of phosphate on removal. Journal of Environmental Management   91,  2238-2247 (2010).

[16] Sen Lin, Diannan Lu, Zheng Liu. Removal of arsenic contaminants with magnetic γ-Fe2O3 nanoparticles. Chemical Engineering Journal 211–212, 46–52 (2012).

[17] Kyungsun Song, Wonbaek Kim, Chang-Yul Suh, Dongbok Shin, Kyung-Seok Ko, Kyoochul Ha. Magnetic iron oxide nanoparticles prepared by electrical wire explosion for arsenic removal. Powder Technology   246, 572–574 (2013).

[18] Christopher T. Parsons, Raoul-Marie Couture, Enoma O. Omoregie, Fabrizio Bardelli, Jean-Marc Greneche, Gabriela Roman-Ross, Laurent Charlet. The impact of oscillating redox conditions: Arsenic immobilisation in contaminated calcareous floodplain soils Environmental Pollution 178, 254-263 (2013).

[19] Eveliina Repo, Marko Makinen, Selvaraj Rengaraj, Gomathi Natarajan, Amit Bhatnagar, Mika Sillanpa. Lepidocrocite and its heat-treated forms as effective arsenic adsorbents in aqueous medium. Chemical Engineering Journal 180, 159– 169 (2012).

[20] Kevin R. Henke. Arsenic: Environmental chemistry, health, threats and waste treatment. 1st edition. Jhon Wiley and Sons Ltda. Great Britain, England, 2009.

[21] Walter J. Weber Jr. Pysicochemical processes for quality control. John Wiley and Sons. New York, Chichester, Brisbne, Toronto y Singapore.  pp. 199-250, 1972.

[22] Züleyha Özlem Kocabas-Ataklı, Yuda Yürüm. Synthesis and characterization of anatase nanoadsorbent and application in removal of lead, copper and arsenic from water. Chemical Engineering Journal  225,  625–635 (2013).

[23] Yohanna Seminovski Pérez, Giselle Autie Castro,* Rafael López Cordero y Miguel Autie Pérez. Estudio de la microporosidad estrecha de carbones activados obtenidos de semilla de palma por activación química con KOH 2008.


María Teresa Alarcón Ph.D.

Centro de Investigación en Materiales Avanzados, Victoria 147 Nte., Zona Centro, 34000 Durango, México. E-mail: teresa.alarcon@cimav.edu.mxI

Myrian López Ph.D.

Centro de Investigación en Materiales Avanzados, Victoria 147 Nte., Zona Centro, 34000 Durango, México.

You may also like

Arsenite removal with Fe3O4, MnFe2O4 and CoFe2O4 Nanoparticles

Arsenite removal with Fe3O4, M..

This work evaluates the ability of nanoparticles (NPs) to remove arsenic (As+3) from water.

read more
Eco-Friendly alternative for   water treatment  from nanotechnology

Eco-Friendly alternative for..

To cite this article  use:  Roksana, M. Eco-Friendly alternative for water treatment from nanotechnology.  J. […]

read more
1 Minute for whole life

1 Minute for whole life

To cite this article  use:  Londoño, S., Berrio, S., Cajavilca, R., Obregón, M., Parra, G. […]

read more

Leave a Reply

Your email address will not be published. Required fields are marked *