Green Chemistry Dynamic Article Links

Cite this: Green Chem., 2012, 14, 1917

www.rsc.org/greenchem PAPER

Selective recovery of gold using some cross-linked polysaccharide gels

Bimala Pangeni,a Hari Paudyal,a Minoru Abe,a Katsutoshi Inoue,*a Hidetaka Kawakita,a Keisuke Ohto,a

Birendra Babu Adhikaria and Shafiq Alamb

Received 5th March 2012, Accepted 10th April 2012DOI: 10.1039/c2gc35321k

Effective recovery of gold from aqueous acidic chloride media was achieved with different kinds ofpolysaccharides with different structures by treating with concentrated sulfuric acid. The gels were foundto be highly selective for Au(III) over other precious and base metals at varying hydrochloric acidsolution. The adsorption isotherms of Au(III) followed the Langmuir type of adsorption and the maximumadsorption capacities for Au(III) was evaluated as high as 7.57, 7.20, 5.64 and 4.80 mmol g−1 for the gelsof cellulose, dextran, alginic acid and pectic acid, respectively. Analysis of kinetic data shows that theuptake of gold(III) on celulose gel at the initial stage follows pseudo-first-order kinetics. Crystallinestructure of the cellulose before and after the treatment with concentrated sulfuric acid was measured byXRD analysis. Furthermore, Au(III) was found to be reduced to elemental form by all types ofpolysaccharides gels, which was confirmed by the formation of clearly visible elemental gold particlesand by means of the XRD-spectra of the adsorbents after adsorption. Infrared spectrum studies providedsupporting evidence for the reduction of Au(III) to Au(0) with a suitable mechanism of Au(III) adsorptionfollowed by reduction using various polysaccharide gels.

Introduction

Nowadays, precious metals are extensively used in many fields,such as catalysts in various chemical processes, in electrical andelectronic industries, in medicine and in jewelry.1 The increasingamount of electronic and electrical waste (e-waste) is causingharm to the environment and also causes serious problems withregard to the supply of raw materials. Recycling and reuse ofmetals, in particular, is very crucial for preventing the practice ofexcessive mining and supply of primary metals. In most casesthe wastes generated from these products contain appreciableamounts of valuable metals like gold. For example, the contentof gold in these wastes discarded in this form is much higherthan that in its ore form.2 Taking into consideration the decreas-ing resources of precious metals like gold with its ever increasingapplications, it is important to develop the process for its recov-ery and recycling from spent appliances.3 Consequently, themost suitable recovery and reuse technologies can contribute toreduce the dependency on primary metal resources.4

Various conventional methods for the recovery of gold fromaqueous solutions including precipitation, ion exchange5 andsolvent extraction6 are in practice. However, these recovery pro-cesses are uneconomical and also energy and time consuming.Furthermore, a large amount of secondary wastes are generated

resulting from the use of a large amount of chemical reagents forthese processes. Therefore, it is very important to find an appro-priate method which must be cost-effective and environmentallybenign for the recovery of gold from aqueous solutions from theview point of environment protection and full utilization of goldresources. Adsorptive recovery of precious metals has recentlyemerged as a potentially attractive and environmentally benignalternative to traditional reclaiming treatments.7 Compared withconventional methods, adsorption offers distinct advantages formetal ion recovery, including high efficiency, low operating costand minimal volume of sludge, etc.8

In the last decade, biopolymer based products have beenincreasingly studied with the focus on polysaccharides. Mainlycellulose, the most important renewable resource, has beenstudied as a starting material for chemical modification. Cellu-lose and its derivatives attract significant attention from research-ers due to their outstanding physical and chemical behavior9 aswell as its abundance in nature, renewability, low cost andavailability.

Native cellulose exhibits only poor adsorption for the majorityof metal ions because of the crystalline structure and absence ofparticular functional groups with high affinity for metal ions.However, proper modification by chemical methods can developadsorption capacity and structural stability of native cellulose. Inour previous study, we developed dimethylamine-immobilized(DMA) paper gel for recovering precious metals,10 which exhib-ited a maximum adsorption capacity for Au(III) as high as4.6 mmol g−1. However, it takes a lot of chemical reagents andhigh cost to prepare DMA-paper gel from spent paper, a cellulo-sic material, though the feed material is cheap. In the previous

aDepartment of Applied Chemistry, Faculty of Science and Engineering,Saga University, Honjo 1, Saga 840-8502, Japan. E-mail:inoue@elechem.chem.saga-u.ac.jp; Fax: +81952288669bFaculty of Engineering and Applied Science, Memorial University,St. John’s, NL A1B 3X5, Canada

This journal is © The Royal Society of Chemistry 2012 Green Chem., 2012, 14, 1917–1927 | 1917

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online / Journal Homepage / Table of Contents for this issue

paper, we found that cotton cellulose can be converted into aninteresting adsorption gel which exhibits high selectivity andloading capacity for gold by simple treatment with concentratedsulfuric acid. In order to examine whether such phenomenon isspecific only to cotton cellulose or common to all kinds of poly-saccharides, in the present work, we prepared similar gels fromtypical polysaccharides like cellulose, dextran, alginic acid andpectic acid by the same method to observe their adsorption beha-viors for metal ions, especially for Au(III). That is, we investi-gated the adsorption of Au(III) on cellulose gel in details andqualitatively compared with those of other polysaccharide gels.

Experimental

Materials

Analytical grade chloride salts of copper (Katayama Chemical,Japan), zinc (Sigma Aldrich), iron and palladium (Wako, Japan)were used to prepare test solutions of respective metals. Analyti-cal grade HAuCl4·4H2O and H2PtCl6·6H2O (Wako, Japan) wereused to prepare gold and platinum solutions, respectively.All other reagents were of analytical grade and were usedwithout further purification.

Preparation of cross-linked polysaccharide gels

Crystalline cellulose powder (102 330 cellulose) marketed byMerck, Germany, for thin layer chromatography was used as thefeed material of cellulose gel. Dextran, produced by Leuconostocmesenteroides, strain No. B-512, average molecular weight ofwhich is around 200 000 (Sigma Aldrich), was employed as thefeed material for dextran gel. Sodium alginate, 500 cps (SigmaAldrich) and polygalacturonic acid, Nacalai Tesque, Inc., Kyoto,Japan, were used for the feed materials for alginic acid andpectic acid gels, respectively. The adsorption gels were preparedas follows. For the preparation of cross-linked polysaccharidesgels, (as shown in Scheme 1), for example, 10 g of cellulosepowder was mixed with 50 cm3 of concentrated H2SO4 andstirred for 24 h at 373 K using reflux condenser for cross-linkingcondensation reaction. Then, after the mixture was cooled, it wasneutralized with NaHCO3, and washed several times with dis-tilled water until neutral pH. The black product obtained wasdried in a convection oven for 24 h at 343 K. Then, dried massof the product was ground to regulate the uniform particle sizeof 75–100 μm. Thus prepared gel was termed as cross-linkedcellulose (CLC). The gels of dextran, alginic acid and pectic acidwere prepared also by the same method; these were termed as

Scheme 1 Inferred example of synthetic route of cross-linked cellulose and its adsorption–reduction mechanism.

1918 | Green Chem., 2012, 14, 1917–1927 This journal is © The Royal Society of Chemistry 2012

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

cross-linked dextran (CLD), cross-linked alginic acid (CLAA)and cross-linked pectic acid (CLPA), respectively.

Measurement and analysis

The metal concentrations in the adsorption tests were measuredby using Shimadzu model ICPS-8100 ICP/AES spectrometerand Shimadzu model AAS-6800 atomic absorption spectropho-tometer. Visual observation was recorded using KEYENCEmodel VHX/VH series optical microscope. Spectroscopicstudies were performed by JASCO model FT/IR-410 Fouriertransform infrared spectrometer. The generation of elementalgold (Au0) on the gels after the adsorption of Au(III) was eluci-dated by means of X-ray diffraction spectrum by usingShimadzu model, XRD-7000, X-ray diffractometer. The surfacemorphology of the adsorbents before and after gold recoveryprocess was recorded using a JEOL (Tokyo, Japan) modelJCM-5100 scanning electron microscope (SEM) at accelerationenergy of 20 kV.

Batch adsorption test

In the present investigation, the adsorption behaviors of gold andother metal ions were tested individually in batch mode of oper-ation. In a typical experimental run, each 10 mg of dried CLC,CLD, CLAA and CLPA gels were shaken together with 10 cm3

of test solution containing 0.1 mM (M = mol dm−3) of eachmetal ion in varying concentration of hydrochloric acid at 303 Kfor 48 h. After equilibrium, the mixture was filtered and thefiltrate was analyzed for remaining metal ion concentrations.

For adsorption isotherm studies, a series of Au(III) solution(100 cm3) were prepared in 0.1 M hydrochloric acid in whichthe initial concentration of Au(III) was varied in the range of0.5–16 mM. Ten milligrams of dried gel was added into the10 cm3 of Au(III) solutions to be shaken at four different temp-eratures (293, 303, 313 and 323 K) by CLC gel and at only303 K by CLD, CLAA and CLPA gels for 150 h to attain com-plete equilibrium. Percentage adsorption for each metal ion wascalculated according to eqn (1), where Ci and Ce (mmol dm−3)are the initial and equilibrium concentrations of metal ion,respectively. The equilibrium adsorption capacity was calculatedby the mass balance of Au(III) before and after the adsorptionaccording to eqn (2).

%Adsorption ¼ðCi � CeÞCi

� 100 ð1Þ

and,

Qe¼ðCi � CeÞW

� V ð2Þ

where Qe (mmol g−1) represents the equilibrium adsorptioncapacity V (cm3) is the volume of the test solution used andW (g) is the dry weight of the adsorbent.

The adsorption kinetics of Au(III) were measured in detail forCLC gel as follows. Two hundred milligrams of dried CLC gelwas shaken together with 200 cm3 of 2 mM of Au(III) solution in0.1 M hydrochloric acid at four different temperatures (293, 303,313 and 323 K). Suspended mixture (3.5 cm3) was immediately

sampled at definite time intervals from the start of the operationto measure the time variation of the residual Au(III) concentrationafter filtration.

Measurement of the degree of crystallinity of cellulose

For the sample of cellulose employed in the present work, thecrystallinity index before and after cross-linking was measuredby means of X-ray diffraction. The peak height method wasemployed to calculate the crystallinity index (CrI). Awide rangeof reported values for native cellulose by X-ray diffraction, inthe range 62–87.6% using peak height method and from 39 to75.3% using other methods are found in the literature.11 X-Raydiffraction patterns of cellulose samples were measured by usingShimadzu model X-ray diffractometer (XRD-7000), at roomtemperature from 5 to 40° using Cu/Kα1 irradiation (1.54 Å) at40 kV and 30 mA. The scanning speed was 2° min−1 and thedata were collected in continuous mode. Based on the methodproposed by Segal et al.,12 the crystallinity index was determinedaccording to the following equation.

CrI ¼ðI002 � IamÞI002

� 100 ð3Þ

where, I002 is the intensity of the peak at 2θ = 22.5° and Iam isthe minimum intensity corresponding to the amorphous contentat 2θ = 18°.

Results and discussion

Adsorption behavior of metal ions

Fig. 1 shows the effect of hydrochloric acid concentration on theadsorption of Au(III), Pt(IV), Pd(II), Cu(II) and Zn(II) ions on CLCfrom their individual solutions. As seen from this figure, nometal ions are adsorbed at all except for Au(III), which is nearlyquantitatively adsorbed over the whole concentration region ofhydrochloric acid tested, suggesting that CLC gel exhibits very

Fig. 1 Effect of hydrochloric acid concentration for the adsorption ofvarious metal ions by CLC gel. Conditions: volume of solution =10 cm3, concentration of metals = 0.1 mM, dry weight of gel = 10 mg,shaking time = 48 h, temp. = 303 K.

This journal is © The Royal Society of Chemistry 2012 Green Chem., 2012, 14, 1917–1927 | 1919

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

high affinity for Au(III) over other precious and base metal ionstested. Similar phenomena were observed also for CLD, CLAAand CLPA. From these results, it may be expected that cheapadsorption gels with excellent selectivity to Au(III) can be pre-pared from a variety of polysaccharides by a simple treatmentusing concentrated sulfuric acid.

Surface analysis of the adsorbents

The surface morphology of the adsorbents (cellulose anddextran) after treatment with concentrated sulfuric acid, is shownin Fig. 2. The figure shows that the dry gel surfaces before goldrecovery were smooth and lacked any cracks or holes. The SEMimages of cellulose and dextran gels after gold recovery are alsoshown in Fig. 2(b, d). The elemental gold particles are clearlyseen with some rough surfaces compared to Fig. 2(a, c). Becauseboth the gels are non-porous, significant physical adsorption isnot likely to occur. From a practical point of view, such types ofsurface structures are considered to be beneficial for the recoveryof gold in its elemental form.

Adsorption isotherm studies

Since all types of modified polysaccharide gels developed in thepresent work were observed to be distinctively selective for gold,subsequent experiments were carried out only for Au(III) ionsusing the polysaccharide gels mentioned earlier. Fig. 3(a) showsthe adsorption isotherms of Au(III) on CLC, CLD, CLAA andCLPA gels at 303 K for comparison. In all cases, the amount ofadsorption increases with increasing concentration of gold in thelow concentration region whereas it tends to approach constantvalues corresponding to each gel in the high concentrationregion, suggesting the monolayer adsorption according to theLangmuir isotherm model. Consequently, the adsorption iso-therms data were analyzed on the basis of the Langmuir adsorp-tion model expressed by eqn (4), in which it is assumed that theadsorption takes place on a homogenous monolayer surface withidentical adsorption sites, and there is no lateral interactionbetween the adsorbed sites.

Ce

Qe¼ 1

Qmaxbþ Ce

Qmaxð4Þ

Fig. 2 SEM micrographs of (a) cross-linked cellulose, (b) cellulose after gold adsorption, (c) cross-linked dextran, and (d) dextran after gold adsorp-tion taken at 20 kVacceleration energy, at 1500× magnification, scale = 5 μm, respectively.

1920 | Green Chem., 2012, 14, 1917–1927 This journal is © The Royal Society of Chemistry 2012

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

where, Qe (mmol g−1) represents the equilibrium adsorptioncapacity, Qmax (mmol g−1) represents the maximum adsorptioncapacity, b (dm3 mmol−1) is the Langmuir constant related to theadsorption energy.

The plot of Ce/Qe versus Ce is presented in Fig. 3(b). Asexpected from the Langmuir model, a linear relationship withhigh correlation coefficient values was observed for each type ofgel studied. The values of maximum adsorption capacities(Qmax) and Langmuir constant (b) for Au(III) for each gel werecalculated from the slope and intercept of these straight lines asshown in Table 1. As seen from this table, the maximum adsorp-tion capacities at 303 K are in the following order; CLC > CLD> CLAA > CLPA.

The maximum adsorption capacity for Au(III) at 303 K werecompared to various adsorbents reported in the literature as listedin Table 2. As seen from Table 2, among all the reported adsor-bents, the maximum adsorption capacity of CLC gel was found

to be the highest value, which suggests the possibility of recover-ing Au(III) from industrial effluent using this adsorbent. Suchhigh adsorption capacity of the cross-linked cellulose may beattributable to the change in the polymeric structure and decreasein crystallinity index after acid treatment, which will be dis-cussed in later sections.

Since the highest adsorption of Au(III) was observed for CLCgel, detailed works were conducted only for CLC gel in the sub-sequent work.

Effect of temperature on the adsorption of Au(III) on CLC gel

Fig. 4(a) shows the adsorption isotherms of Au(III) on CLC gelat four different temperatures viz. 293, 303, 313 and 323 K. Asseen from Fig. 4(a), temperature greatly affects the adsorption ofAu(III) on CLC gel; i.e. the increase of temperature enhanced theadsorption of Au(III) on CLC gel. Similar to the precedingsection, the results were re-plotted on the basis of the Langmuiradsorption model as shown in Fig. 4(b). The maximum adsorp-tion capacity and Langmuir constant were evaluated from theslopes and intercepts of the straight lines in this figure at eachtemperature studied as listed in Table 3. From this table, it is

Fig. 3 Adsorption isotherms of gold(III) on cross-linked cellulose,cross-linked dextran, cross-linked alginic acid, and cross-linked pecticacid, gels (a) Experimental plots and (b) Langmuir plots. Conditions:volume of the feed solution = 10 cm3, weight the dried gel added =10 mg, shaking time = 150 h, HCl = 0.1 mol dm−3.

Table 1 Langmuir isotherm parameters for Au(III) on various cross-linked polysaccharides

Adsorbents Qmax (mmol g−1) b (dm3 mmol−1) R2

Cross-linked cellulose 7.57 7.03 0.99Cross-linked dextran 7.20 10.36 0.99Cross-linked alginic acid 5.64 27.86 0.99Cross-linked pectic acid 4.80 52.05 0.99

Table 2 Comparison of maximum adsorption capacities for gold(III) ondifferent adsorbents

Adsorbents

Maximumuptake capacity(mmol g−1) Conditions Reference

Cross-linked cellulose gel 7.57 0.1 M HCl Presentstudy

Cross-linked dextran gel 7.20 0.1 M HCl Presentstudy

Cross-linked alginic acid gel 5.64 0.1 M HCl Presentstudy

Cross-linked pectic acid gel 4.80 0.1 M HCl Presentstudy

Glycine modified cross-linked chitosan

0.86 pH 2 1

Dimethyamine-paper 4.6 1 M HCl 10p-Aminobenzoic acid-paper 5.1 1 M HCl 10Persimmon waste 4.95 0.1 M HCl 15Cross-linked lignophenol 1.92 0.5 M HCl 16Primary amine-cross-linkedlignophenol

1.98 0.5 M HCl 16

Ethylene diamine cross-linked lignophenol

3.08 0.5 M HCl 16

Dimethyamine cross-linkedlignophenol

7.2 0.5 M HCl 17

Chemically modifiedchitosan with magnetic

3.43 pH 0.5 18

Cross-linked microalgae 3.25 0.1 M HCl 19Cross-linked cotton gel 6.21 0.1 M HCl 20Cross-linked paper gel 5.05 0.1 M HCl 21

This journal is © The Royal Society of Chemistry 2012 Green Chem., 2012, 14, 1917–1927 | 1921

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

obvious that the adsorbed amount of gold increases with increas-ing temperature, which suggests that the adsorption of Au(III) onCLC gel is an endothermic process.

From the temperature dependency of the Langmuir constant,or adsorption equilibrium constant (b), the thermodynamic par-ameters such as changes in the free energy (ΔG°), enthalpy(ΔH°) and entropy (ΔS°) associated to the adsorption process ofAu(III) on CLC gel were evaluated according to eqn (5) and (6).

ΔG° ¼ �RT lnb ð5Þ

lnb ¼ �ΔG°=RT ¼ �ΔH°=RT þ ΔS°=R ð6ÞThe plot lnb as a function of 1/T yields a straight line as

depicted in Fig. 4(c) for CLC gel, from which values of ΔH°

and ΔS° were calculated from the slope and intercept, respec-tively. The negative values of Gibbs free energy (ΔG°) at alltemperatures studied confirm the feasibility of the process andspontaneous nature of adsorption. The positive value of enthalpy(ΔH°) demonstrates the endothermic nature of the adsorption.Entropy, as the measure of freedom of the system, also indicatesthe positive values, suggesting the increased randomness atsolid-solution interface during the adsorption of Au(III) onCLC gel.

Adsorption kinetics of Au(III) on CLC gel

Fig. 5(a) shows the time variation of the amount of adsorbedAu(III) on CLC gel at various temperatures. As seen from these

Fig. 4 Effect of temperature for the adsorption of Au(III) by CLC gel (a) Experimental plot, (b) Langmuir plot and (c) Van’t Hoff plot. Conditions:dry weight of the gel = 10 mg, shaking time = 150 h, volume of the test solution = 10 cm3, HCl = 0.1 mol dm−3.

Table 3 Thermodynamic parameters for Au(III) by cross linked cellulose gel

Temperature (K) b (dm3 mmol−1) Qmax (mmol g−1) ΔG° (kJ mol−1) ΔH° (kJ mol−1) ΔS° (J mol−1 K−1) R2

293 1.66 3.89 −1.23 0.93303 7.33 7.58 −5.02 102.17 353.26 0.99313 26 9.62 −8.48 0.99323 82 12.20 −11.84 0.99

1922 | Green Chem., 2012, 14, 1917–1927 This journal is © The Royal Society of Chemistry 2012

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

results, the rate of adsorption rapidly increases at the initial stageof adsorption and tends to approach a constant value or equili-brium value after a certain period of shaking. It is evident fromthese results that although it takes a very long time to attain equi-librium at low temperature like 293 and 303 K, such slow kin-etics can be improved by increasing the temperature of thesystem.

The rate of adsorption of the Au(III) onto the CLC gel at theinitial stage was analyzed on the basis of a pseudo-first-orderkinetic model expressed by eqn (7) as shown in Fig. 5(b), inwhich all the plots cluster on the straight lines passing throughthe origin corresponding to different temperatures.

lnðCt=CiÞ ¼ �kt ð7Þ

where, Ct and Ci stand for the concentration of Au(III) at time t,and initial concentration, respectively; k is the pseudo-first-orderkinetic rate constant (h−1) and t is the time (h).

From the slopes of these straight lines, the pseudo-first-orderrate constants were evaluated for four different temperatures.From the rate constants, k, evaluated at different temperatures the

activation energy of the adsorption of Au(III) on CLC gel wasevaluated according to the Arrhenius equation expressed byeqn (8).

lnk ¼ lnA� Ea=R ð1=TÞ ð8Þwhere R is the universal gas constant, Ea is the activation energy,A is the Arrhenius constant and T is the absolute temperature inKelvin. Fig. 5(c) shows the plot of lnk versus 1/T according toeqn (8). From the slope of this straight line, the value of the acti-vation energy (Ea) was evaluated as 71.8 kJ mol−1. This valuesuggests that the adsorption of Au(III) onto CLC gel is a chemi-cal adsorption, which is in accordance with the magnitude ofactivation energy for chemical adsorption usually being between8.4 to 83.7 kJ mol−1.13 The positive value of Ea suggests that anincrease in temperature favors the adsorption.

Crystalline structure of cellulose

Fig. 6 shows the X-ray diffraction patterns of raw cellulose andCLC gel. The value of CrI of the former was evaluated as 83%

Fig. 5 Adsorption kinetics of Au(III) by CLC gel at different temperatures (a) Experimental plot, (b) Pseudo-first-order plot, and (c) Arrhenius plot.Conditions: Weight of the dry gel added = 200 mg, volume of the feed solution = 200 cm3, concentration of Au(III) = 2 mM, HCl concentration =0.1 mol dm−3.

This journal is © The Royal Society of Chemistry 2012 Green Chem., 2012, 14, 1917–1927 | 1923

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

while it was found decrease to 31% after treatment by concen-trated sulfuric acid. The higher the crystalline structure, the lessis the chemical activity in the cellulose. Because the majority ofreagents in aqueous solution can only penetrate into the amor-phous parts of cellulose, these domains are also termed theaccessible regions of cellulose, and crystallinity and accessibilityare closely related.14 A highly crystalline cellulose sample has atight structure with cellulose chains closely bound to each other,leaving too little space for chemical reactions anywhere withinthe cellulose crystal. It is obvious from Fig. 6 that the treatmentby concentrated sulfuric acid disrupts the crystalline structure ofcellulose and turns to an amorphous structure which leads to thevery high and selective adsorption of Au(III).

Solid state analysis of cross-linked polysaccharide gels afterAu(III) adsorption

The instrumental analysis of the gels after the adsorption ofAu(III) was carried out in order to ascertain the formation ofelemental gold during adsorption for all types of gels tested.That is, the X-ray diffraction analyses and visual observations ofCLC, CLD, CLAA and CLPA gels were performed after theadsorption of Au(III). Almost the same patterns of peaks andgold aggregations were observed for all types of adsorbents. TheXRD spectra observed for CLC, CLD, CLAA and CLPA gelsare shown in Fig. 7. The sharp peaks observed at 2θ values ofaround 38, 44, 64 and 77 degrees in this figure belong to thecrystal structure of elemental gold, verifying the generation ofAu(0) on the polysaccharide gels after the adsorption, whichsuggests that Au(III) underwent subsequent reduction to Au(0) onthe surface of the polysaccharides after treatment by concentratedsulfuric acid. The formation of elemental gold was furtherconfirmed from the optical microscope photographs as shown inFig. 8, as these gels show very fine and clear aggregated particlesof elemental gold with various shapes, and are distinctly visibleafter long period of shaking at high concentration of Au(III) sol-ution. On the other hand, for short shaking period at low concen-tration of Au(III) for example, 0.2 mmol dm−3, only microscopic

gold particles are observed to be float on the surface of samplesolution. Both the results from XRD and observation analysesconfirmed the reductive adsorption of Au(III) on different gelsprepared from simple chemical modification using various poly-saccharides. Extremely high selectivity and loading capacity ofthese gels for Au(III) may be attributed to the reduction of Au(III)to Au(0) during adsorption. From the results of Fig. 8, it can beinferred that gold particles formed by the reduction process havea certain sort of tendency to be detached from the gel surfaceforming clear aggregates. Similar observations had been alsoreported in our previous study in the adsorption on chemicallymodified spent paper gel.10

Measurement of Fourier transform infrared spectra

The structural analysis of the polysaccharide gels before andafter cross-linking, and after the adsorption of Au(III) on eachtype of gel was carried out by means of the observation of IR-spectra in order to infer the binding mechanism of Au(III) ionsonto the surface of the gels. Fig. 9(A) and 9(B) show the infrared

Fig. 6 X-Ray diffraction pattern of cellulose (multiple peaks) andamorphous cellulose (single smooth peak) generated with concentratedsulfuric acid (96%) treatment. X-axis: Bragg angle (2θ). I002 representsthe maximum intensity at 2θ = 22.5°. Iam shows the minimum intensityat 2θ = 18° used to calculate crystallinity in the peak height method.

Fig. 7 XRD patterns of gold loaded cross-linked cellulose, cross-linked dextran, cross-linked alginic acid and cross-linked pectic acid.

1924 | Green Chem., 2012, 14, 1917–1927 This journal is © The Royal Society of Chemistry 2012

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

Fig. 8 Optical microscope photographs of gold loaded cross-linked (a) cellulose (b) dextran (c) alginic acid and (d) pectic acid, gels.

Fig. 9 FT-IR spectra of (a) crude, (b) cross-linked, and (c) gold loaded gels of (A) cellulose, and (B) dextran.

This journal is © The Royal Society of Chemistry 2012 Green Chem., 2012, 14, 1917–1927 | 1925

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

spectra for cellulose and dextran, respectively. In the spectra forcrude materials (a), in Fig. 9(A), (B) and also in alginic acid andpectic acid gels (figures not included), a broad band centered ataround 3450 cm−1 due to O–H stretching, a sharp band at round3050 cm−1 is considered to be attributable to the C–H stretching,another sharp band at 1720 cm−1 due to CvO stretching and abroad band centered at 1100 cm−1 were observed.

The peaks observed at around 1780 cm−1 and 1690 cm−1 afteracid treatment in Fig. 9(A), (B) and also in the alginic acid andpectic acid (figures not included) are ascribed to the carbonyl orcarboxylate stretching frequencies. Thus, it can be inferred that,during the acid treatment, long polymeric chains of polysacchar-ides were dissociated into shorter chains which were then sub-jected to condensation cross-linking reaction with the treatmentof concentrated sulfuric acid that can be confirmed by theappearance of C–O–C stretching at around 1200 cm−1 observedin all kinds of polysaccharide gels.

After Au(III) adsorption as shown in spectra c of Fig. 9(A),(B) and for gold loaded gels of pectic acid and alginic acid(figures not included), the absorption band due to hydroxylgroups have flattened and shifted to a lower frequency region(centered at around 3410 cm−1). Since coordination with metalion weakens the O–H bond resulting in absorption at a lower fre-quency, shifting of the O–H stretching band to a lower frequencyregion is attributed to the coordination of hydroxyl groups withloaded Au(III) ion. In addition, the intensity of the absorptionband attributed to CvO stretching of carbonyl groups at around1690 cm−1 increases after Au(III) loading as shown in spectra cof Fig. 9(A), (B) and also for other studied gels.

Adsorption–reduction reaction mechanism

Based on the spectroscopic analysis mentioned in the precedingsection, the mechanism of adsorption of Au(III) was inferred asfollows. The dissociated chain of polysaccharide in presence ofconcentrated sulfuric acid undergoes the condensation cross-linking reaction as shown in Scheme 1 in the case of cellulose,for example, with the release of water molecules. Thus, preparedgel is inferred to contain new ether bond linkages of C–O–C asobserved in IR spectrum mentioned earlier. The oxygen atoms ofthese new C–O–C bonds play important roles as the coordinationsites for Au(III) together with oxygen atoms of un-reactedhydroxyl groups and those of C–O–C bonds existed before thecondensation reaction to form stable 5-membered chelate rings.In acidic chloride media, majority of Au(III) exists in the form oftetra chlorocomplex, AuCl4

−. This auric chloride complex isstable in acidic and oxidizing media. When contacted with thecross-linked polysaccharide gels, Au(III)–chloro complex is co-ordinated with the above-mentioned many oxygen atoms of theacid treated polysaccharides and oxygen atoms of ether bond.Furthermore, during this adsorption process some hydroxylgroups are oxidized to carbonyl groups, and the hydroxyl groupsplay an important role for the reduction of Au(III) to Au(0).Besides this, as we carried out the experimental works in acidicchloride media, some of the chloride ions also responsible forthe donation of electron for reducing the Au(III) to Au(0).

It is well known that Au(III) is easily reduced because it has ahigh redox potential (1.0 V) in comparison to other metal ionslike Pd2+ (0.68 V) and Pt4+ (0.62 V). Furthermore, the change in

the polysaccharide structures after acid treatment played a charac-teristic role in the improvement of the stability of the gels. Thus,it is inferred that these are the driving force for the formation ofelemental gold particles, resulting in the extraordinary highselectivity and loading capacity. Also in the case of other poly-saccharide gels treated with concentrated sulfuric acid like CLD,CLAA and CLPA, Au(III) ions are presumed to be adsorbed ontothe gel surface and further reduced to metallic gold particles inthe similar mechanism as shown below.

From the results of FT-IR spectra, hydroxyl groups of poly-saccharide gels are oxidized to carbonyl groups, which isexpressed as:

R–OH ! RvOþ Hþ þ e� ð9ÞThe net reaction takes place as follows:

AuðiiiÞ þ 3e� ! Au0ðsÞ ð10ÞBased on the eqn (9) and (10), the overall reaction is

expressed as follows:

AuðiiiÞðaqÞ þ R–OH ! Auð0ÞðsÞ þ RvOþ Hþ

Conclusions

The results obtained in the present work suggest that differentkinds of polysaccharides can be used as effective adsorbents forAu(III) by simple treatment with concentrated sulfuric acid due totheir high affinity and selectivity towards Au(III) in acidic chlor-ide media at ambient temperature. The Langmuir adsorption iso-therm was found to be best fit to the experimental data withmaximum adsorption capacities of 7.57, 7.20, 5.64 and4.80 mmol g−1 for cellulose, dextran, alginic acid and pecticacid gels, respectively, at 303 K. Analysis of kinetic datasuggests that the uptake of Au(III) on cellulose gel followspseudo-first-order kinetics. The activation energy evaluated fromthe Arrhenius equation was found to be 71.8 kJ mol−1 for cellu-lose gel. Thermodynamic studies show that the adsorption ofgold(III) onto cross-linked cellulose gel is a spontaneously favor-able, endothermic and increased disorder process. The resultsobtained from the solid state analysis (XRD spectrum andoptical microscope images) for all gels after the adsorption ofgold(III) reveal the noticeable observation of elemental gold par-ticles from ionic form i.e. gold(III). Based on the observation ofinfrared spectra the mechanism of Au(III) adsorption was dis-cussed, and it was inferred to involve an apparent redox reaction;the ionic gold species is reduced to the metallic form usingchloroauric complex of gold.

Thus, the present study demonstrated that gold recovery canbe successfully achieved using various polysaccharides aftersimple treatment with concentrated sulfuric acid as an alternativeto traditional adsorbents employed in industrial gold recovery.

Acknowledgements

The present work was financially supported in part by a Grant-in-Aid for Scientific Research about Establishing a SoundMaterial-Cycle Society (K2131) from the Ministry of Environ-ment of Japanese Government.

1926 | Green Chem., 2012, 14, 1917–1927 This journal is © The Royal Society of Chemistry 2012

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online

References

1 A. Ramesh, H. Hasegawa, W. Sugimoto, T. Maki and K. Ueda, Bioresour.Technol., 2008, 99, 3801.

2 X. Huang, Y. Wang, X. Liao and B. Shi, J. Hazard. Mater., 2010, 183, 793.3 D. Parajuli, H. Kawakita, K. Inoue, K. Ohto and K. Kajiyama, Hydro-metallurgy, 2007, 87, 133.

4 A. Manhart, J. Ind. Ecol., 2010, 15, 13.5 T. Ogata and Y. Nakano, Water Res., 2005, 39, 4281.6 J. L. Cortina, E. Meinhardt, O. Roijals and V. Marti, React. Funct.Polym., 1998, 36, 149.

7 K. Deplanche and L. E. Macaskie, Biotechnol. Bioeng., 2008, 99, 1055.8 C. Mack, B. Wilhelmi, J. R. Dunkan and J. E. Burgess, Biotechnol. Adv.,2007, 25, 264.

9 C. Sena, M. H. Godinho, C. L. P. Oliveira and A. M. Figueiredo Neto,Cellulose, 2011, 18, 1151.

10 C. R. Adhikari, D. Parajuli, H. Kawakita, K. Inoue, K. Ohto andH. Harada, Environ. Sci. Technol., 2008, 42, 5486; C. R. Adhikari,D. Parajuli, K. Inoue, K. Ohto, H. Kawakita and H. Harada, NewJ. Chem., 2008, 32, 1634.

11 F. M. Gama, J. A. Teixeira and M. Mota, Biotechnol. Bioeng., 1994, 43,381.

12 L. Segal, J. J. Creely, A. E. Martin and C. M. Conrad, Text. Res. J., 1959,29, 786.

13 Q. Li, L. Chai, Z. Yang and Q. Wang, Appl. Surf. Sci., 2009, 255, 4298.14 M. Hall, P. Bansal, J. H. L. Matthew, J. Realff and A. S. Bommarius,

FASEB J., 2010, 277, 1571.15 Y. Xiong, C. R. Adhikari, H. Kawakita, K. Ohto, K. Inoue and

H. Harada, J. Hazard. Mater., 2009, 100, 4083.16 D. Parajuli, C. R. Adhikari, M. Kuriyama, H. Kawakita, K. Ohto,

K. Inoue and M. Funaoka, Ind. Eng. Chem. Res., 2006, 45, 8; D. Parajuli,H. Kawakita, K. Inoue and M. Funaoka, Ind. Eng. Chem. Res., 2006, 45,6405.

17 D. Parajuli, K. Khunathai, C. R. Adhikari, K. Inoue, K. Ohto,H. Kawakita, M. Funaoka and K. Hirota, Miner. Eng., 2009, 22, 1173.

18 A. M. Donia, A. A. Atia and K. Z. Elwakeel, Hydrometallurgy, 2007, 87,197.

19 K. Khunathai, Y. Xiong, B. K. Biswas, B. B. Adhikari, H. Kawakita,K. Ohto, K. Inoue, H. Kato, M. Kurata and K. Atsumi, J. Chem. Technol.Biotechnol., 2011, 87, 393.

20 B. Pangeni, H. Paudyal, K. Inoue, H. Kawakita, K. Ohto and S. Alam,Cellulose, 2012, 19, 381.

21 B. Pangeni, H. Paudyal, K. Inoue, H. Kawakita, K. Ohto and S. Alam,J. Chem. Eng. Data, 2012, 57, 796.

This journal is © The Royal Society of Chemistry 2012 Green Chem., 2012, 14, 1917–1927 | 1927

Publ

ishe

d on

13

Apr

il 20

12. D

ownl

oade

d by

Uni

vers

ity o

f N

ew H

amps

hire

on

02/1

0/20

13 0

8:34

:02.

View Article Online