DSpace at VNU: Removal of arsenic (V) from aqueous medium using manganese oxide coated lignocellulose silica adsorbents

DSpace at VNU: Removal of arsenic (V) from aqueous medium using manganese oxide coated lignocellulose silica adsorbents...

Full Terms & Conditions of access and use can be found at

ISSN: 0277-2248 (Print) 1029-0486 (Online) Journal homepage: http://www.tandfonline.com/loi/gtec20

Removal of arsenic (V) from aqueous medium using manganese oxide-coated lignocellulose/

silica adsorbents

Igor W K Ouédraogo, Erol Pehlivan, Hien T Tran, Samuel Paré, Yvonne L.

Bonzi-Coulibaly, Dieter Zachmann & Müfit Bahadir

To cite this article: Igor W K Ouédraogo, Erol Pehlivan, Hien T Tran, Samuel Paré, Yvonne

L Bonzi-Coulibaly, Dieter Zachmann & Müfit Bahadir (2015): Removal of arsenic (V) fromaqueous medium using manganese oxide-coated lignocellulose/silica adsorbents, Toxicological

& Environmental Chemistry, DOI: 10.1080/02772248.2015.1133815

To link to this article: http://dx.doi.org/10.1080/02772248.2015.1133815

Accepted author version posted online: 28 Dec 2015.

Submit your article to this journal

Article views: 11

View related articles

View Crossmark data

Publisher: Taylor & Francis

Journal: Toxicological & Environmental Chemistry

DOI: http://dx.doi.org/10.1080/02772248.2015.1133815

Removal of arsenic (V) from aqueous medium using manganese oxide-coated

lignocellulose/silica adsorbents Igor W K Ouédraogo 1* , Erol Pehlivan 2 , Hien T Tran 3 , Samuel Paré 4 , Yvonne L Bonzi-

Coulibaly 4 , Dieter Zachmann 5 and Müfit Bahadir 5

Environmental Engineering (2iE), 01 BP 594, Ouagadougou 01, Burkina Faso

2 Department of Chemical Engineering, Selcuk University, Campus, 42031 Konya, Turkey

3 Hanoi University of Science, Hanoi, Vietnam

Université de Ouagadougou, 03 BP 7021, Ouagadougou 03, Burkina Faso

Braunschweig, Hagenring 30, 38106 Braunschweig, Germany

Abstract:

Arsenic (V) adsorption on manganese oxide-coated rice wastes was investigated in this study The

modified adsorbents were characterized by Fourier-transform infrared spectroscopy, scanning electron microscopy and pH measurements to determine the point of zero charge Batch adsorption equilibrium experiments were conducted to study the effects of pH, contact time and initial

concentration on arsenic removal efficiency The adsorption capacity of rice waste was significantly

improved after modification with permanganate The Langmuir isotherm model fitted the equilibrium data better than the Freundlich model which confirms surface homogeneity of the adsorbent

Maxima adsorption capacities are determined as 10 and 12 mg/g at pH 3 for manganese oxide-coated rice husk and straw respectively The adsorption energy indicates that the adsorption process may

be dominated by chemisorption Pseudo‒second‒order rate equation described the kinetics sorption

of arsenic with good correlation coefficients, better than a pseudo‒first‒order equation Manganese

oxide-coated rice husk and straw appear to be promising low cost adsorbents for removing arsenic from water

Keywords: Lignocellulose/silica adsorbent, Manganese oxide, Arsenic, Adsorption, Rice

I Introduction The presence of arsenic (As) in ground and surface waters at elevated concentrations creates serious problems to human’s health and other organisms (Wu et al 1989) Anthropogenic or biological activities, as geochemical conditions may entail the mobilization of As into groundwater (Malik et al 2009), and natural processes such as soil erosion, mineral leaching and weathering are responsible of introducing As into surface waters Mining activities, combustion of fossil fuels, and the use of As additives to

livestock feed may entail additional impacts (Mohan, Charles, and Pittman 2007)

High concentrations of As in groundwater were reported from Argentina, Austria,

Bangladesh, Chile, China, Ghana, India, Italy, Japan, Malaysia, Mexico, Monogolia, Nepal, Poland, South Africa, Taiwan, Vietnam, and some parts of the United States (Malik et al 2009; Mohan, Charles, and Pittman 2007) Smedley, Knudsen, and Maiga (2007) investigated

As concentrations in groundwater in the area of the city of Ouahigouya, located in the North

of Burkina Faso Most of them had As concentrations ranging between 0.5 and 1630 μg/L Other studies in Essakane (Barro-Traoré et al., 2008) showed As concentrations of 13‒212 μg/L in human urine and 69‒101 μg/L in drinking water; while an As concentration of 10 μg/L is recommended by the World Health Organization as guideline value for drinking water The presence of As in streams and lakes at elevated concentrations may lead to bioaccumulation in living organisms, potentially causing health problems to plants, animals, and humans Chronic human intake of As has been associated with increased risk of skin,

liver and lung cancer, diabetes, developmental and reproductive problems, and cardiovascular diseases (Wu et al 1989)

Co-precipitation, liquid–liquid extraction, ion-exchange, ultrafiltration and reverse osmosis

process have been so far used to remove As from aqueous media (Biswas et al 2008) Most

of them are cost-intensive and have disadvantages such as incomplete removal, require high energy, and may generate toxic sludge or waste products which are difficult to dispose

(Choonga et al 2007) Adsorption is considered to be one of the most promising technologies because it is low cost and easy to set up (Malik et al 2009) Removal of As from drinking water using low cost sorbents, simple and appropriate methods is highly desirable Rice husk (RH) and straw (RS) are two of the most abundant lignocellulose and silica waste materials in

the world, with 155.7 million hectares of land cultivated in 2008 (USDA 2009) These

natural materials are considered to be economic precursors for the production of adsorbents that are extensively used for many purposes of separation and purification

techniques (Sharma et al 2014) However, the adsorption capacities of lignocellulose for As

removal are low and slow (Al Rmalli et al 2005; Asif and Chen 2015; Haque et al 2007; Ranjan, Talat, and Hasan 2009; Shafique et al 2012) Therefore, chemical modifications have

been often used to synthesize appropriated and efficient adsorbents for As ions removal

(Anirudhan et al 2012; Ouédraogo et al 2015)

Manganese oxides is a subject of interest in various fields including molecular adsorption due

to their outstanding structural multiformity combined with novel chemical and physical properties In addition to their ability to remove a wide range of ions from wastewater (Mandal and Suzuki 2002), manganese oxides have various advantages such as low cost, eco-friendliness and abundance as well as their excellent electrochemical properties

(Subramanian, Zhu, and Wei 2008) However, the particle nanosizes agglomerate in

contact with aqueous system To overcome this difficulty, the nanoparticles were

functionalized by organic, inorganic and composite materials which provide chemical stability and better distribution of manganese oxides in aqueous system A recent investigation shows that various manganese oxides sorbents can be prepared through a simple method by KMnO4reduction (Subramanian, Zhu, and Wei 2008) Some of the reported modifications used natural clinoptilolite zeolite (Camacho, Parra, and Deng 2011), bentonite (Eren, Afsin, and Onal 2009), polystyrene resin (Lenoble et al 2004) and cellulose fiber (Maliyekkal, Lisha, and Pradeep 2010; Wang et al 2014) by forming manganese oxides on the surface, which

improves the metal adsorption capacity Though limited reports are available on cellulose

as a support for metal oxide nanoparticles, to the best of our knowledge, no studies have been reported on the synthesis of manganese oxide-coated rice waste fiber and its application as an adsorbent for As removal

This work is related to the preparation and evaluation of hybrid organic-inorganic manganese oxide-coated RH and RS adsorbents, which mainly consist of hemicelluloses, cellulose,

lignins and silica, for removal of As(V) from aqueous solutions Adsorbents were

characterized with Fourier transform infrared (FTIR) spectroscopy, scanning electron microscopy (SEM), and pH measurements to determine the point of zero charge (pH pzc )

As(V) adsorption on the adsorbents was investigated through different parameters effects :

pH, contact time and initial concentration

2 Materials and methods

2.1 Chemicals

All reagents used in this work were of analytical-grade purity A stock solution (1000 mg/L)

of arsenate was purchased from Merck (Darmstadt, Germany) Diluted standard solutions of

As(V) were prepared daily before use NH4OH and HCl, purchased from Merck, were used to

adjust the pH of solutions KI used for reduction of As(V) to As(III) was obtained from Roth

(Karlsruhe, Germany) KMnO 4 used for RH and RS modification was purchased from Fluka (Taufkirchen, Germany) NaBH4 (Fluka), NaOH (Merck) and HCl (Merck) were

used for hydride generation All glassware and plastic bottles were cleaned with 10% HNO 3 (Merck) solution and rinsed with deionized water Whatman cellulose acetate membrane (0.2 μm pore size), used for As(V) solution filtration, was purchased from Sigma-Aldrich (Taufkirchen, Germany) Water was purified using a Seralpur purification system Pro90CN (Ransbach-Baumbach, Germany)

2.2 Modification of RH and RS

RH and RS used in this work were collected from local rice fields in Burkina Faso The dried

materials were powdered in a ball mill (BLB Braunschweig, Germany) and sieved with a sieve shaker machine (Retsch, Haan, Germany) The fractions size of 0.063‒0.125 mm

were washed several times with pure water to remove dust and fines, and dried in an oven at 60°C for 24 h The adsorbents were prepared by impregnation method: 200 mL of an

aqueous solution of KMnO4 (0.1 mol/L) and 5 g of RH or RS were mixed in a 500 mL flask

under continuous stirring After 5 h, the suspension was filtered The formation of manganese oxide on the matrix is shown by the change of the adsorbents’ color from golden brown to

brownish black The manganese oxide-coated RH (MOCRH) and RS (MOCRS) were

thoroughly washed with deionized water Samples were dried in an oven at 50°C for 24

h and stored in a desiccator

2.3 Adsorbent characterization

The functional groups present in RS and MOCRS were characterized by FTIR (BX II, Perkin

Elmer, Waltham, USA) using the KBr pellet method The spectra were recorded from 4000

to 500 cm‒1 for 5 scans The pHpzc of adsorbents was determined by the following procedure

A definite amount, i.e 0.2 g of sample was added to 50 mL of deionized water at 22±2°C Next, the initial pH (pHi) values of the solution were adjusted roughly from pH 5 to 10 by

adding 0.1 mol/L of HCl or NH4OH After a shaking time of 4 h, the top solution was filtered and the final pH (pHf) value of the filtrate was measured (Ouédraogo et al 2015) The difference between pHi and pHf values (ΔpH = pHf‒pHi) was plotted versus the pHi The

pHpzc of the sorbent was determined from the point of intersection of the resulting curve, at

which ΔpH = 0 The surface morphology of samples was characterized using a microscope

(JEOL JSM‒6480, Tokyo, Japan) The surface area was characterized from N2 adsorption–

desorption isotherms at ‒196°C obtained using an adsorption instrument (Micromeritics

ASAP 2020, Georgia, USA) The specific surface areas were assessed using the BET

(Brunauer–Emmett–Teller) equation Thermal degradation characteristics were carried out

with a thermogravimetric analyzer (SETSYS Evolution TGA 16/18, Setaram

Instrumentation, Caluire-et-Cuire, France) The ramping rate was 10°C/min up to 900°C

using N2 as the carrier gas The results were reported in percentage weight (TG) and

differential weight (dTG) versus temperature Sample ash contents have been determined

using NF MO3-003/ EN 14775 / ISO 1171 standard method Ashes were digested with 37% HCl and the manganese contents were determined by atomic absorption spectrometry (AAS).

2.4 Batch sorption experiments

Batch experiments were carried out in plastic bottles (50 mL) by adding 0.2 g sorbent to 50 mL of aqueous As(V) solution The initial pH of the solution was adjusted to between 2 -10 by adding 0.1 mol/L HCl or NH4OH solution The plastic bottles were gently agitated in an electric shaker (Guwina-Hofmann, Berlin, Germany) at 150 rpm for 4 h The sorbent was removed by filtration with cellulose acetate membrane and the remaining As was determined The influence of contact time was studied at pH 3 for 0.25 to 6 h and for initial concentrations of 2 to 150 mg/L All

experiments were performed in triplicates and the average values were taken Maximum

deviation was <5%.

2.5 Analysis of As in aqueous system

A hydride generation atomic absorption spectrometry with a Zeeman correction (HGAAS-

Hitachi Z-2000 AAS, Tokyo, Japan) was used to determine As concentrations Arsine

obtained from As(III) gives the best signal at a concentration range from 1‒20 μg/L;

therefore, all As(V) in the sample should be previously reduced to As(III) and diluted with

3% HCl solution The reduction was carried out with the optimized protocol: 2.5 mL of 30%

HCl and 2.5 mL of 20% KI solutions were added to 10 mL of all samples One hour later,

the volume was adjusted to 25 mL with 3% HCl solution before analysis For hydride generation, the following solutions were used: (i) 3.7% HCl; (ii) NaBH4–NaOH solution: 10

g of NaBH4 in 1 L of deionized water by adding 4 g of NaOH

3 Results and discussion

3.1 Adsorbent characterization

Infrared spectroscopy was used to study the interaction of the metal oxide with the lignocellulosic support Fig 1 shows the FTIR spectra of both RS (a) and MOCRS (b) The peak located at around 3350 cm−1 is attributed to O–H stretching vibration The band located

at 1420 cm−1 can be assigned to the CH2 scissoring motion The spectra indicated weak peak

near 1730 cm−1 corresponding to C=O stretching from carbonyl group The band at 1620 and

1462 cm−1 can be assigned to C=C aromatic skeletal vibration The characteristic band at around 1060 cm-1 corresponding to C–O–C stretching is also observed in both spectra The features of these bands are typical to lignocellulosic material The band at 802 cm−1 was assigned to symmetric Si–O–Si stretching vibration (Shahrokh Abadi et al 2015) The bending vibration for H–O–H was observed at 1638 cm−1 (Ma, Deng, and Yao 2014) The MOCRS spectrum shows some changes mainly in the decrease of the ratio of COOH (1741

cm−1) group's intensity, which can be explained by the complexation of Mn ions with carboxylate group in the matrix A new band appears at 569 cm−1 that was attributed to the

stretching vibration of Mn–O (Maliyekkal, Lisha, and Pradeep 2010; Wang et al 2014) The

peak at 3350 cm ‒1 became broad, suggesting a strong interaction between the OH groups

of cellulose and manganese oxide (Ma, Deng, and Yao 2014)

The structure and the morphology of RS and MOCRS are showed by SEM images in Fig 2 Table 1 summarizes the characteristics properties of the samples Both RS and MOCRS have relatively uniform surface, where roughness could be seen Comparing the SEM images, RS shows a significantly wrinkled surface than MOCRS The color of fibers changes from golden brown to brownish black suggesting the formation of MnO2 phase on the RS (Shaabani,

Hezarkhani, and Shaabani 2014; Wang et al 2014) Actually, KMnO 4 reacts with cellulose

to form cellulose–nanoscale manganese oxide composite (Maliyekkal, Lisha, and Pradeep

2010) The synthesis of MOCRS composite was based on the reduction and precipitation of manganese oxide, which resulted in a homogeneous distribution of manganese oxide particles

in the lignocellulose matrix It was also found that ash and total manganese content in the RS and MOCRS were increased from 16 to 44.63% and 0.11 to 16.96%, respectively This high percentage of residual ash in MOCRS is assumed to be from manganese–lignocellulose

complex Moreover, the surface area of RS is slightly increased through KMnO 4 treatment

TG and dTG curves of RS and MOCRS are presented in Fig 3 A minor weight loss ~1.5%

was first observed around 160°C in the RS, which may be due to water and the decomposition of volatile matters such as low molecular weight sugars Following this, a

weight loss by approximately ~50% occurred from 220 to 400°C, which is hemicellulose

and cellulose decomposition (Wattanasiriwech, Wattanasiriwech, and Svasti 2010) The dTG

peaks are respectively 322 and 366°C Another important loss of ~14% due to the

decomposition of lignin occurred between 350 and 600°C The results show that the

thermal stability of RS moves to lower temperatures by modification with manganese A

weight loss by approximately 40% took place from 120 to 500°C, which should be attributed

to the decomposition of hemicelluloses, cellulose and lignins The corresponding dTG

peaks are respectively at 200, 293 and 394°C The percentage weight loss of the residue

after heating up to 900°C was ~20% higher for MOCRS (57.25%) when compared to RS

(38.26%) This result indicates the presence of manganese in MOCRS As shown in Fig

4, the pHpzc of MOCRS is ~7.6, close to the value found with manganese oxide-coated resin

(7.8) (Lenoble et al 2004) This reveals that the adsorbent had a basic nature, as pHpzc is

higher than 7

3.2 As (V) removal experiments

It is well known that the pH is an important parameter in the adsorption process The charge

of the adsorbate and the adsorbent often depends on the pH of the solution due to H+ ions

exchange The results (Fig 5) show that As(V) adsorption by MOCRS occurs at pH

below the pH pzc Optimum removal of As(V) by MOCRS was achieved at initial pH 3 with the removal of more than 60% of the As(V) In the pH range of 3–6, the adsorbent surface is positively charged, which is appropriated to adsorb the major H2AsO4– species present in

aqueous medium Under high basic conditions, i.e pH of 8–10, As adsorption by MOCRS

was minimal, with a removal of 3% at pH 10 The decrease in As adsorption with the increase

in pH is due to the increase in the negative charge density on the adsorbent surface, and to the

increase in the number of OH‒ ions in the solution, in competition with HAsO42– species

The kinetic adsorption study was carried out with 50 mg/L As(V) solution The uptake

equilibrium was achieved after 3 h and no noticeable changes were observed for longer

reaction times The experimental data of As(V) on MOCRS were analyzed by using

pseudo-first-order and pseudo-second-order kinetic models (Han et al 2013) The rate constants K 1,

K 2 , the calculated adsorption capacities q e(cal) values and the corresponding linear regression correlation coefficients R2 were summarized in Table 2 Rate constants were 0.011 min ‒1 for

first-order expression and 0.042 g/mg/min for second-order The maxima adsorption

capacities, calculated by first- and second-order models were respectively ~1.0 and 7.8 mg/g

The data showed that the R 2 value of pseudo-second-order model is more than 0.99, while the corresponding value of pseudo-first order model is less than 0.97 More

importantly, the q e(cal) obtained with the pseudo-second kinetic model, are in agreement with

experimental adsorption capacity q e(exp) The effect of As(V) concentration on the sorption by the MOCRH and MOCRS were

investigated and Fig 6 shows the equilibrium adsorption capacity q e (mg/g) versus As(V) concentrations Ce (mg/L) at equilibrium The sorption of As ions with both sorbents increased

with the increasing of ions concentration

Characterization of the adsorption process is often carried out using a number of isotherm

models These include the most common Langmuir, Freundlich and Dubinin–Radushkevich (D–R) isotherms (Han et al 2013) Langmuir model has been the most often used empirical

model, since it contains the two useful and easily imaginable parameters, i.e the affinity of

binding sites or bonding energy b and the monolayer sorption saturation capacity Q max

Freundlich model is an exponential equation that describes two physical parameters, i.e

the adsorption capacity K f and the intensity of adsorption 1/n Table 3 gives an overview

of Langmuir and Freundlich constants for MOCRH and MOCRS It was found that the R2

values of Langmuir (> 0.98) model were higher than those obtained from the linear form of Freundlich (≤ 0.98) This result indicates that Langmuir type sorption isotherm is suitable for

equilibrium studies Therefore, the adsorption of As(V) ions onto MOCRS and MOCRH is

considered forming a monolayer that takes place at the binding sites on the sorbent surface

The maximum adsorption capacity, at pH 3, for MOCRS (12.0 mg/g) was found to be higher

than for MOCRH (10.2 mg/g) Manganese oxide content in the adsorbent is a key factor

which might influence As adsorption It was reported that metal oxide amount in RS

composite is higher than RH composite (Ouédraogo et al 2015) RH consists of hard

materials, including silica and lignin, and has a more recalcitrant cell wall structure than RS,

which has lower lignin content (Boonmee 2012) A wide range of low-cost sorbents has been

studied worldwide for As(V) removal The values reported in this study are several times

higher than those of unmodified biomass adsorbents, e.g., sorghum biomass (Haque et al

2007), hyacinth root (Al Rmalli et al 2005), rice husk (Asif and Chen 2015) and rice polish

(Ranjan, Talat, and Hasan 2009) and at least three times higher than pine leaves (Shafique et

al 2012) Moreover, it was also found that As(V) adsorption by MOCRS (~70 mg As(V)/g of

Mn present in the adsorbent) is higher than the micro-/nano-structured MnO2 (~23 mg

As(V)/g of Mn) (Zhang and Sun 2013) This is due to a well distribution of manganese

oxide in the lignocellulosic matrix, which leads to a better surface contact Furthermore,

As speciation combined with the different pHzpc values highlighted these differences

In addition, adsorption data were further analyzed using D–R model The general form of the

D–R equation model is:

ln q e = ln Xm − β å2 (1)

where

9 = RT ln (1 + 1/C e ) (2)

where q e is the amount of As ions adsorbed per unit weight of sorbent (mol/g), and Ce is the

concentrations at equilibrium (mol/L) Xm is the maximum adsorption capacity (mol/g), β is

the activity coefficient (mol2/kJ2) related to sorption mean free energy (kJ/mol) and 9 is the

Polanyi potential, where R (8.32 J/mol/K) is the gas constant and T (K) is the absolute temperature The results of parameters obtained from the model are listed in Table 3 The D–

R isotherm model showed higher R2 value (≥ 0.99) than those obtained for Langmuir and Freundlich The maximum adsorption capacity Xm of both MOCRH and MOCRS calculated

from Eq (1) were 0.00032 mol/g (23.975 mg/g) and 0.00038 mol/g (28.470 mg/g),

respectively The mean adsorption energy (E) could be estimated from the β value (Table 3)

using the following equation:

E = (‒2β) ‒0.5 (3)

The E values was calculated to be 12.5 kJ/mol for As(V) on both MOCRS and MOCRH

adsorbents, which ranges between 8 and 16 kJ/mol, implying chemical adsorption by ligand

exchange

3 Conclusion

The availability of the natural materials, their simple modification without pretreatment, and

the relative good adsorption capacity of As(V) levels make the prepared MOCRH and

MOCRS materials suitable for potential practical applications In a future study, the

correlation between the amount of manganese coated adsorbents and the As adsorption performance will be investigated

Acknowledgement

This investigation was performed at the Guest Chair within the project “Exceed - Excellence Center for Development Cooperation - Sustainable Water Management in Developing Countries” at the Technische Universitaet Braunschweig; Prof Dr Pehlivan being the visiting professor, and Ms Tran and Dr Ouédraogo being the international exchange staff members The Exceed Project is granted by the German Federal Ministry for Economic Cooperation and Development (BMZ) and German Academic Exchange Service (DAAD); we gratefully acknowledge their financial support