Documents

Nanoparticulas de Hierro Remocion de Arsenico Ingles

Description
nanoparticulas de hierro
Categories
Published
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Arsenic removal by modi 󿬁 ed activated carbons with iron hydro(oxide)nanoparticles Alma Veronica Vitela-Rodriguez, Jose Rene Rangel-Mendez * Division of Environmental Sciences, Instituto Potosino de Investigación Cientí   󿬁 ca y Tecnológica, A.C. Camino a la Presa San José 2055, Col. Lomas 4ta. Sección,C.P. 78216 San Luis Potosí, S.L.P., Mexico a r t i c l e i n f o  Article history: Received 21 March 2012Received in revised form6 September 2012Accepted 3 October 2012Available online 10 November 2012 Keywords: Activated carbonIron hydro(oxide)ArsenicAdsorptionKinetics a b s t r a c t Different activated carbons modi 󿬁 ed with iron hydro(oxide) nanoparticles were tested for their ability toadsorb arsenic from water. Adsorption isotherms were determined at As (V) concentrations  <  1 ppm,with varying pH (6, 7, 8) and temperature (25 and 35   C). Also, competition effect of anions on the As (V)adsorption capacity was evaluated using groundwater. The surface areas of the modi 󿬁 ed activatedcarbons ranged from 632 m 2 g  1 to 1101 m 2 g  1 , and their maximum arsenic adsorption capacity variedfrom 370  m g g  1 to 1250  m g g  1 . Temperature had no signi 󿬁 cant effect on arsenic adsorption; however,arsenic adsorption decreased 32% when the solution pH increased from 6 to 8. In addition, whengroundwater was used in the experiments, the arsenic adsorption considerably decreased due to thepresence of competing anions (mainly SO 42  , Cl  and F  ) for active sites. The data from kinetic studies 󿬁 tted well to the pseudo-second-order model ( r  2 ¼  0.98 e 0.99). The results indicated that sample CAZ-Mhad faster kinetics than the other two materials in the  󿬁 rst 10 min. However, sample F400-M was only5.5% slower than CAZ-M. The results of this study show that iron modi 󿬁 ed activated carbons are ef  󿬁 cientadsorbents for arsenic at concentrations lower than 300  m g L   1 .   2012 Elsevier Ltd. All rights reserved. 1. Introduction Contamination of groundwater with arsenic (As), a highly toxicelement,isaseriousproblemaroundtheworldthataffectsmillionsofpeople(SmedleyandKinniburgh,2002).Chronicexposuretothispollutant is associated with several health problems such asarsenicosis and different kinds of cancer (Hughes, 2002; Kapaj et al., 2006; Ferreccio et al., 2000). The World Health Organiza- tion (WHO) guideline value for As in drinking water is 10  m g L   1 (WHO, 2008). Nevertheless, approximately 50 million peopleworldwide are exposed to higher arsenic concentrations indrinking water. Well-known As contaminated countries includeBangladesh, India, China, Hungary, Chile, Argentina, Mexico, andVietnam (Smedley and Kinniburgh, 2002). In Mexico, As contami-nated aquifers exceeding the maximum permissible level of 25  m g L   1 have been found in Coahuila, Chihuahua, Durango,Guanajuato, San Luis Potosi and Zacatecas (CAN, 1999; Del Razo et al., 1990; Camacho et al., 2011; Armienta and Segovia, 2008). Several technologies have been reported to remove arsenic fromdrinking water, including coagulation- 󿬂 occulation, ion exchange,reverse osmosis, membrane  󿬁 ltration, and adsorption processes.Fromthese,adsorptionhasbeenwidelyusedtoremovearsenicdueto its low cost and high ef  󿬁 ciency (Han et al., 2002; DeMarco et al., 2003; USEPA, 2007; Mohan and Pittman, 2007; Vaclavikov et al., 2008). Iron oxides, such as hematite and goethite, have showngood performance as arsenic adsorbents due to their high selec-tivity for this element. However, iron oxides have low mechanicalresistance, which does not permit their application in  󿬁 xed bedcolumns (Guo et al., 2007; Solozhenkin et al., 2003; Raven et al., 1998). Modi 󿬁 ed adsorbents, such as activated carbon with ironoxides, have been reported in the literature to improve adsorptioncapacities and mechanical properties, indicating that arsenicadsorption processes are becoming more ef  󿬁 cient (Chen et al.,2007; Fierro et al., 2009; Jang et al., 2008). It is known that arsenic is adsorbed on iron hydro(oxides) by inner-sphere surfacecomplexes (Goldberg and Johnston, 2001; Vaughan and Reed, 2005), so the available surface area of iron loaded on the carbo-naceous matrix determines the maximum adsorption capacity of the adsorbent, rather than the amount of iron contained in theactivated carbon. Because of this, the objective of our previouspublications (Nieto-Delgado and Rangel-Mendez, 2012; Arcibar- Orozco et al., 2012) was to determine the optimal parameter toanchor iron nanoparticles onto activated carbon. *  Corresponding author. Tel.:  þ 52 (444) 834 20 00; fax:  þ 52 (444) 934 20 10. E-mail address:  rene@ipicyt.edu.mx (J.R. Rangel-Mendez). Contents lists available at SciVerse ScienceDirect  Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman 0301-4797/$  e  see front matter    2012 Elsevier Ltd. All rights reserved.http://dx.doi.org/10.1016/j.jenvman.2012.10.004  Journal of Environmental Management 114 (2013) 225 e 231  Since a detail study of As (V) adsorption by granular activatedcarbons containing iron hydro(oxide) nanoparticles has not beenreported, the main objective of this research is to correlate surfacearea, iron content and the materials ’  pH to the As (V) adsorptioncapacity and kinetics of these iron modi 󿬁 ed activated carbons insynthetically and naturally contaminated water. 2. Materials and methods  2.1. Materials All chemicals used were reagent grade. Iron solutions werepreparedfromFeCl 3 $ 6H 2 Osalt(Fermont).Anarsenic stocksolutionof 5 mg L   1 was prepared from Na 2 HAsO 4 $ 7H 2 O salt (Sigma e Aldrich). Appropriated dilutions were prepared to conductadsorption experiments with initial concentrations of As (V)ranging from 25  m g L   1 to 1500  m g L   1 , in accordance with thosefound in contaminated waters (Smedley and Kinniburgh, 2002;CAN,1999;DelRazoetal.,1990;Camachoetal.,2011;Armientaand Segovia, 2008). Deionized water was used to prepare all solutions.The granular activated carbons used in this research were: bitu-minous Filtrasorb-400 (F400) from Calgon, CAZ and CAP, bothproduced from  Agave salmiana  bagasse and chemically activatedwith ZnCl 2  and H 3 PO 4 , respectively (Nieto-Delgado and Rangel-Mendez, 2011). Natural groundwater was collected from a deepwell located in the city of San Luis Potosi, Mexico, following theNOM-014-SSA1-1993 methodology (NOM, 1993). The characteris-tics of this water are given in Table 1.  2.2. Activated carbons modi  󿬁 cation Granular activated carbons (F400, CAZ and CAP) were modi 󿬁 edwith iron hydro(oxide) nanoparticles by forced hydrolysis of FeCl 3 solution following the established methodology by Nieto-Delgadoand Rangel-Mendez (2011): (1) 200 mg of activated carbon of U.S.meshno.100/140(148 e 105 m m)were placed incontactwith10mL of FeCl 3  solution (FeCl 3 $ 6H 2 O, 97% ACS grade from Fermont) insideasealedglasscontainer;(2)themixturewasmixedfor24hat25  Cto allow the diffusion of iron into the carbon pores; (3) thermalhydrolysis was carried out by placing the containers in a preheatedfurnace (70 e 120   C) during a selected time; (4) after heating, thesamples were rinsed with distilled water until the elimination of all the soluble iron and the iron particles that were not anchoredto the carbon surface: this was determined by analyzing thesupernatant. The activated carbons reported in this manuscriptwere treated at the optimum established conditions (Nieto-Delgado and Rangel-Mendez, 2011): F400 (96   C for 56 h), CAZ(110   C for 6.8 h) and CAP (58   C for 6.8 h).  2.3. Materials characterization 2.3.1. Surface area and porosity Surface area and pore size distribution of the activated carbonswere determined from the corresponding nitrogen adsorption e desorption isotherms at 77 K using a Micromeritics ASAP 2020system. The surface area was determined by the Brunauer, Emmettand Teller (BET) equation and the pore size distribution wascalculated by using the density functional theory (DFT) method.  2.3.2. Iron content determination The iron content was determined after acid digestion of theactivatedcarbons.40mgofsamplewereaddedto20mLofaHNO 3 :H 2 SO 4  (5:1) solution. The samples were digested for 1 h at 150   Cusing an advanced digestion system by microwave (Milestone,Ethos 1). The obtained solution was analyzed by ICP-OES (Varian730-ES) at a wavelength of 271.9 nm.  2.3.3. Slurry pH  The point of zero charge (PZC) was determined as following.100 mg of activated carbonwere added to 5 mL of deionized water,previously aerated with nitrogen. The containers were sealed andconstantly stirred for 72 h. After that, the slurry pH was measuredand considered to be equal to the material ’ s pH.  2.3.4. X-ray diffraction studies XRD patternswere obtained ina BrukerD8 diffractometerusinga CuK a  radiation ( l  ¼  1.5418   A). Iron modi 󿬁 ed samples werereduced to a mesh size  < 90 in an agate mortar, and then thepowder was placed in the XRD sample port. The patterns wereobtained with a step size of 0.02  2 q  at 10 s per step.It should be mentioned that scanning electron micrographs of these samples have been recently reported some were else (Nieto-DelgadoandRangel-Mendez,2011),however,themain 󿬁 ndingsarementioned in the discussion Section 3.1 of this manuscript.  2.4. Adsorption isotherms The As (V) adsorption capacity of modi 󿬁 ed activated carbonsand control samples (unmodi 󿬁 ed activated carbon) was deter-minedin duplicate at pH 7 and 25   C and the averagewas reported.15 mgofactivatedcarbonwere added to 20mL ofAs (V) solutionatdifferent initial concentrations. The experiments were kept underconstant stirring. The pH of each sample was adjusted daily with0.1 N NaOH and/or 0.1 N HNO 3  until the equilibrium was reached:thiswasdetermined once thesolutionpH andtheAs concentrationremained constant. The initial and the  󿬁 nal arsenic concentrationwasmeasuredbyatomicemissionspectroscopy(AES)inanICP-AES(Varian 730-ES) at a wavelength of 188.980 nm. The adsorptioncapacity of the activated carbons was calculated by using thefollowing equation: q  ¼  V  ð C  o    C  e Þ = m  (1) Where  q  is the adsorption capacity (mg g  1 ),  V   is the volume of solution (L),  C  o  is the initial As concentration (mg L   1 ),  C  e  is theequilibrium As concentration (mg L   1 ), and  m  is the mass of adsorbent (g).  2.5. Adsorption kinetics 1 g of carbon sample was added to a rotating basket that wasthen submerged in a 1 L reactor: it should be mentioned that theadsorberdesignful 󿬁 lled thecriteriaforavertical-shaftturbine.Thereactor was partially immersed in a water bath that operated at  Table 1 Physicochemical parameters of well water used in adsorption experiments.Parameter ConcentrationdetectedMaximum permissiblelimit (NOM-127-SSA1-1994)pH 7.55 6.5 e 8.5Total hardness(mg L   1 CaCo 3 )20.24 500Cl  (mg L   1 ) 14.55 250F  (mg L   1 ) 4.03 1.50PO 43  (mg L   1 ) 0.44 c NI a SO 42  (mg L   1 ) 12.70 400As ( m g L   1 ) ND b 25 a Not indicated. b Not detected. c Determined as total phosphorous.  A.V. Vitela-Rodriguez, J.R. Rangel-Mendez / Journal of Environmental Management 114 (2013) 225 e  231 226  constant temperature. The adsorption kinetics were conductedunder the following conditions: 25   C, 300 min  1 , initial pH 7, 1 L solution,andataninitialarsenicconcentrationof50 m gL   1 .5mLof solutionwere taken after 1, 3, 5, 7,10, 20, 40, 60, 80, and 100 min of the start of the experiment. The samples were analyzed for arsenicas previously described. 3. Results and discussion  3.1. Activated carbons characterization Table 2 shows that the iron content in the activated carbonsincreased almost 2% after modi 󿬁 cation. For example, for F400 thisincreased from 0.25% to 1.67%, and for CAZ and CAP it went from0.08% to 1.67% and to 1.81%, respectively. X-ray diffraction analysisof F400-M (Fig. 1) reported the presence of hematite ( a -Fe 2 O 3 ) at33  ,36  ,41  and50  andakaganeite ( b -FeOOH)at35  ,39  and68  .It should be mentioned that under oxic conditions, gohetite andhematite are thermodynamically the most stable compounds inthis iron hydro(oxide) system (Cornell and Schwertmann, 2003;Hristovski et al., 2009), therefore, it is expected that the arsenicadsorption capacity of the modi 󿬁 ed materials reported herein willremainwith time. Moreover, the surface area and pore volume alsoslightly decreased after modi 󿬁 cation with iron hydro(oxide)nanoparticles due to the obstruction of some micro and mesoporesas shown inTable 2. CAZ-M and CAP-M had a surface area decreaseof around 12% while that of F400-M reduced by only 4.5%: CAZ-Mand F400-M have the highest surface area, 1101 and 1045 m 2 g  1 respectively. The reduction in pore volume of micropores, fromabout 5 to 12% (Table 2), suggested that the iron hydro (oxides)particles anchored on activated carbons are smaller than 20   A. Themicrographs of iron modi 󿬁 ed activated carbons recently reportedby our research group (Nieto-Delgado and Rangel-Mendez, 2011)showed iron hydro(oxide) particles from 3 to 36 nm inside poresand amorphous agglomerates at the external surface of carbongrains, which support the results reported in this manuscript.  3.2. Adsorption isotherms Arsenic adsorption isotherms of activated carbons at pH 7 areshown in Fig. 2. Experimental data were  󿬁 tted with the Langmuirmodel (Table S.1, see Supplementary material) since this adjustedbetterthanthe Freundlich model. It can be observedthat the As (V)adsorptioncapacityofuntreatedCAZandCAPisverylow.However,the As (V) adsorption capacity of these carbons considerablyincreased after modi 󿬁 cation with iron hydro(oxide) nanoparticlesfrom 250  m g g  1 to 526  m g g  1 for CAZ-M, and from 167  m g g  1 to370  m g g  1 for CAP-M at 500 ppb, which is related to the approxi-mately 21 fold increase in iron content. In contrast, commercialF400 removed a considerable amount of As (V), which was attrib-uted toits 0.25% of iron.However, its adsorption capacity increasedalmost 25% after modi 󿬁 cation from 1010  m g g  1 to 1250  m g g  1 .From these results, it is clear that surface area, iron content, andthe materials ’  pH are factors that in 󿬂 uence the arsenic adsorptioncapacity of adsorbents (Table 2). Some authors (Viraghavan et al., 2001) have reported an improvement of arsenic adsorption  Table 2 Physical characteristics and iron content of unmodi 󿬁 ed and treated activatedcarbons.Sample % Fe Slurry pH S BET  (m 2 g  1 ) Pore volume (cm 3 g  1 )Micro Meso TotalF400-SM 0.25 8.62 1045 0.329 0.073 0.402F400-M 1.67 6.05 998 0.312 0.064 0.376CAZ-SM 0.08 3.28 1249 0.394 0.076 0.470CAZ-M 1.67 4.04 1101 0.346 0.071 0.417CAP-SM 0.08 2.40 726 0.211 0.185 0.396CAP-M 1.81 3.57 632 0.185 0.162 0.347 303540455055606570 2 theta    I  n   t  e  n  s   i   t  y   (  a .  u .   ) θθ **** ++++ * Hematite AkaganeiteGraphite Fig. 1.  XRD patterns of iron modi 󿬁 ed activated carbon with a step size of 0.02  2 q  at10 s per step. This spectrum shows the presence of hematite and akaganeite on thesurface of modi 󿬁 ed activated carbons. 0100200300400500600Ce ( µ g L -1 ) 020040060080010001200    Q  e   (      µ   g  g   -   1    )  F400-SM CAZ-SM CAP-SM  ____ Langmuir model A 0100200300400500600Ce ( µ g L -1 ) 020040060080010001200    Q  e   (      µ   g  g   -   1    ) F400-MCAZ-MCAP-M ___ Langmuir model B Fig. 2.  As (V) adsorption isotherms of unmodi 󿬁 ed activated carbons (A) and modi 󿬁 edactivated carbons (B) at pH 7 and 25   C. These  󿬁 gures show that the adsorptionexperimental data follow the Langmuir model at low and high concentration of arsenicthat is in agreement with the adsorption mechanism proposed.  A.V. Vitela-Rodriguez, J.R. Rangel-Mendez / Journal of Environmental Management 114 (2013) 225 e  231  227  capacity as the iron content increases. In this study, although themodi 󿬁 ed activated carbons have similar iron content, their As (V)adsorption capacity is quite different, due to the in 󿬂 uence of bothsurfaceareaandslurrypH.EventhoughCAP-Mhasthehighestironcontent (1.81%), its As (V) adsorption capacity is the lowest becauseit has the least surface area (632 m 2 g  1 ) and slurry pH (3.57). Onthe other hand, CAZ-M has 1.67% of iron, the highest surface area(1101 m 2 g  1 ) and a slurry pH of 4.04, and its As (V) adsorptioncapacity is higher than for CAP-M: this is obviously related tosurface area but also to the effect of the materials pH that isexplained in section 3.3. Moreover, F400-M has 1.67% of iron,asurfacearea(998m 2 g  1 )slightlylowerthanCAZ-M,andboththehighest slurry pH (6.05) and As (V) adsorption capacity. Theseresults suggest that the materials pH is a determinant parameter.Accordingtothespeciationdiagram(Vaclavikovetal.,2008),As(V)is present as negative species (H 2 AsO 4  and HAsO 42  ) at pH 7, atwhich the adsorption experiments were conducted. This meansthat electrostatic interactions partially contribute tothe adsorptionprocess. F400-M has a less negative surface at pH 7 and hence thearsenic anions experience less repulsion than with CAZ-M, whichhas a morenegativesurface at the same conditions. Hence, F400-Mhas a higher As (V) adsorption capacity than CAZ-M despite havinga higher surface area. Comparing these results with others previ-ously reported in the literature (Fierro et al., 2009; Badruzzaman et al., 2004; Chuang et al., 2005), ranging from 4 to 28  m g L   1 , forsimilar materials at the same experimental conditions (25   C, pH 7and300 m gL   1 atequilibrium),theAs(V)adsorptioncapacityofthemodi 󿬁 ed activatedcarbons reportedherein, 847  m g L   1 the highest,is much higher (see Table 3). Something to remark is that theanchorage of iron hydro(oxide) nanoparticles onto activatedcarbons without considerably decreasing surface area is a veryimportant factor for the removal of arsenic from water.Since F400-M had the highest As (V) adsorption capacity, it wasselected to study the effect of pH, temperature, and competinganions on adsorption capacity.  3.3. Effect of pH  Adsorption experiments were conducted at pH 6, 7 and 8(Fig. 3). Result show that the As (V) adsorption capacity of F400-Mat 300  m g L   1 at equilibrium was: 880  m g g  1 at pH 6, 810  m g g  1 atpH 7, and 590  m g g  1 at pH 8, respectively. The As (V) adsorptioncapacity decreased 32% when the pH increased from 6 to 8. Thiseffect can be attributed to the electrostatic repulsion between As(V) and activated carbon (Streat et al., 2008; Ona-nguema et al., 2005). The surface area of F400-M becomes more negative as thesolution pH increases (slurry pH 6.05), hence, the attractive forcetoward As (V) anionic species decreases. Several authors(Solozhenkin et al., 2003; Sherman and Randall, 2003; Mondal et al., 2007) have reported that adsorption of arsenate onto ironoxides occurs by inner-sphere surface complexes. However, if theelectrostatic repulsion between As (V) and the activated carbonssurfaceincreases,theinner-spheresurfacecomplexeswilldiminishand consequently the adsorption capacity decreases.  3.4. Effect of temperature As(V)adsorptionisotherms(Fig.4)wereconductedat25  Cand35   C and  󿬁 tted with the Langmuir model (Table S.2, seeSupplementarymaterial).ThemaximumAs(V)adsorptioncapacityof F400-M at 400  m g L   1 was 588  m g g  1 at 35   C and 526  m g g  1 at25   C. These results indicate that temperature did not signi 󿬁 cantlyaffect adsorption capacity, and suggest that adsorption takes placeby chemical bonds between As (V) and iron hydro(oxides). Severalauthors have reported similar results. For instance, Banerjee et al.(2008), Mondal et al. (2007) and Solozhenkin et al. (2003) re- ported that when changing the temperature from 25   C up to 60   Cduring the removal of As (V) fromwater, the adsorption capacity of iron modi 󿬁 ed carbons and/or iron hydro(oxides) does not changesigni 󿬁 cantly.  Table 3 Comparison of As (V) adsorption capacities among different adsorbents reported inthe literature at 25   C, pH 7 and 300 g L   1 at equilibrium.Adsorbent Qe a ( m g As  g  1 ) ReferenceF400-M 847 This studyCAZ-M 431 This studyCAP-M 181 This studyCAG-Fe 28 [Fierro et al.]IOCS b 18 [Viraghavan et al.]GFH c 4 [Badruzzaman et al.] a 300  m g L   1 at equilibrium. b Iron oxide coated sand. c Granular ferric hydroxide. 050100150200250300350400Ce ( µ g L -1 ) 02004006008001000    Q  e   (      µ   g  g   -   1    ) pH 6pH 7pH 8 ___ Langmuir model Fig. 3.  As (V) adsorption isotherms of F400-M at different pH and 25   C. An increase inpH slightly decreases the adsorption capacity and the Langmuir model adequatelyadjusts the experimental data suggesting that the adsorption mechanism does notchange. 020406080100120Ce ( µ g L -1 ) 0100200300400500600    Q  e   (      µ   g  g   -   1    ) 25°C35°C Fig. 4.  As (V) adsorption isotherms of F400-M at different temperature and pH 7.A change in temperature does not signi 󿬁 cantly change the arsenic adsorption capacityand the Langmuir model (line) adequately adjusts the experimental data suggestingthat the adsorption mechanism does not change.  A.V. Vitela-Rodriguez, J.R. Rangel-Mendez / Journal of Environmental Management 114 (2013) 225 e  231 228
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks