Bioresource Technology 129 (2013) 108–117

Contents lists available at SciVerse ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

N-aminoguanidine modified persimmon tannin: A new sustainable materialfor selective adsorption, preconcentration and recovery of precious metals fromacidic chloride solution

Manju Gurung a, Birendra Babu Adhikari b, Shintaro Morisada a, Hidetaka Kawakita a, Keisuke Ohto a,Katsutoshi Inoue a,⇑, Shafiq Alam b

a Department of Chemistry and Applied Chemistry, Faculty of Science and Engineering, Saga University, 840-8502 Honjo-1, Saga, Japanb Faculty of Engineering and Applied Science, Memorial University of Newfoundland, St. John’s, NL, Canada A1B 3X5

h i g h l i g h t s

" Preparation of new adsorption material by the introduction of N-aminoguanidine ligands on the surface of persimmon tannin." High selectivity of the material towards Au(III), Pd(II) and Pt(IV) over base metals from acidic chloride media." Reduction of adsorbed Au(III) to elemental gold and very high loading capacity of the material." Highly efficient sorption material having high selectivity and adsorption efficiency than that of commercial adsorbents.

a r t i c l e i n f o

Article history:Received 5 July 2012Received in revised form 30 October 2012Accepted 1 November 2012Available online 10 November 2012

Keywords:Persimmon tanninAminoguanidinePrecious metalsAdsorptionLeach liquor of e-wastes

0960-8524/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.biortech.2012.11.012

⇑ Corresponding author. Tel./fax: +81 952 28 8548.E-mail address: inoue@elechem.chem.saga-u.ac.jp

a b s t r a c t

A new adsorption gel has been developed by immobilizing N-aminoguanidine (AG), a chelating ligand, onpersimmon tannin extract through consecutive reactions. Adsorption behavior of the gel was investigatedfor the adsorptive separation and recovery of precious metal ions from varying concentration of HCl med-ium. The adsorption isotherms of precious metal ions on the gel were described by the typical monolayertype of Langmuir model and the maximum adsorption capacities were evaluated as 8.90 mol kg�1 forAu(III), 2.01 mol kg�1 for Pd(II) and 1.01 mol kg�1 for Pt(IV). Real time applicability of the gel was exam-ined for the recovery of precious metals from actual leach liquor of e-waste leached with chlorine con-taining hydrochloric acid. The gel was found to be highly efficient and selective for the uptake oftargeted metal ions in the presence of excess base metal ions and also exhibited superior selectivity overcommercially available anion exchange resins.

� 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Development of a sustainable environment for future genera-tion requires a holistic approach to recycling of precious and valu-able materials. Recovery of valuable metals from a wide variety ofsources such as primary sources (ore) and from secondary sourceslike electronic scraps and industrial wastes has been the subject ofgreat interest and also has been an imperative issue (Hagelükenand Meskers, 2009; Li et al., 2007). Since the waste material con-tains various coexisting metals, selectivity towards target speciesplays a vital role. Among the numerous techniques for preconcen-tration and separation of precious metals, adsorption has beenextensively employed and proven to be more effective comparedto other techniques (Anticó et al., 1994; Hubicki and Wołowicz,

ll rights reserved.

(K. Inoue).

2009; Myasoedova et al., 1985; Shah and Devi, 1997). In recentyears, increasing efforts have been devoted to the developmentof new resins containing selective functional groups, which havebeen applied to hydrometallurgical processes for metal ion adsorp-tion (Anticó et al., 1994; Myasoedova et al., 1985; Shah and Devi,1997). However, challenges for the implementation of these adsor-bents as commercial adsorbents for metal recovery face variousdifficulties such as low metal binding capacity, poor selectivity,costly operation, generation of secondary sludge materials, andso on (Cortina et al., 1998; Wołowicz and Hubicki, 2009; Zuo andMuhammed, 1995). Consequently, it is necessary to explore newseparation materials and methods for these purposes.

In recent years, waste biomass-based low costs adsorbents havebeen received increasing attention as a versatile class of adsorbent(Volesky and Holan, 1995). One of the most striking advantages ofusing waste biomass materials is that they are easily available inhuge quantity at cheap cost. Further, since they are the natural

M. Gurung et al. / Bioresource Technology 129 (2013) 108–117 109

polyfunctional materials, the biomass can be modified in a numberof ways with ligating atoms/groups which exhibit high selectivitytowards the target species (Yun et al., 2001). Thus, taking advan-tage of their easiness for chemical modification, it would be a goodstrategy to fabricate waste biomass materials as adsorbents forselective uptake of concerned metal ions.

Guanidine and its derivatives are very interesting organic com-pounds. Guanidine and aminoguanidine (AG) are highly basicamines with pKa values 13.5 and 11.04, respectively (Gulkoet al., 1971; Kolarz et al., 2001). It was reported that solventextracting reagent containing guanidine functional groups is effec-tive for the recovery of Au(I) from cyanide solution at pH 9–10 oreven at pH 13 (Kordosky et al., 1992). To date, however, no litera-ture has been reported for selective adsorption of precious metalsover base metals from acidic chloride solutions by using waste bio-products modified with such functional groups.

Persimmon extract, produced from the juice of young astringentpersimmons, is marketed as materials for tanning of leather, tradi-tional natural dyes and paints, and so on in East Asian countries,such as China and Japan. It is also marketed as dry powder afterspray drying. Being rich in tannin, persimmon extract containsmultiple adjacent phenolic groups and is known to exhibit specificaffinity for several metal ions (Matsuo and Ito, 1978). Because of itsspecific affinity towards metal ions, persimmon tannin has beenused as an efficient adsorbent for selective recovery of metals (Gur-ung et al., 2011; Nakajima and Sakaguchi, 1993, 2000; Nakajimaand Baba, 2004; Ogata and Nakano, 2005; Yurtsever and S�engil,2009). On the other hand, ligands with amine functionality appearto be very promising for adsorption of precious metals from acidicchloride media because these ligands are protonated in acidic solu-tions and form stable bindings with anionic chloro complexes ofprecious metals (Adhikari et al., 2008; Won et al., 2011). Accord-ingly, aminoguanidine appears to be an effective functional groupfor adsorptive preconcentration of anionic species of precious met-als from acidic chloride media. In addition, persimmon tannin hasdemonstrated strong tendency to reduce the adsorbed Au(III) tometallic gold (Gurung et al., 2011). Taking this into consideration,we have immobilized aminoguanidine onto powder persimmontannin extract to develop a low cost and environmentally ‘‘green’’adsorption gel and tested its potentiality for adsorptive preconcen-tration and recovery of precious metal ions from acidic chloridemedia. In this paper, we report the adsorption behavior of amino-guanidine modified persimmon tannin gel towards precious met-als such as Au(III), Pd(II) and Pt(IV) from model solutions ofhydrochloric acid, investigated by batch wise as well as by contin-uous column method. Effectiveness of the prepared adsorbent forthe recovery of precious metals from actual leach liquor of e-wastehas also been tested to verify its potential application to theindustry.

2. Experimental

2.1. Materials and chemicals

A sample of dry persimmon tannin powder (abbreviated as PTpowder hereafter), the feed material extracted from astringent per-simmon, was kindly donated by Persimmon-Kaki TechnologyDevelopment Co. Ltd., Jincheng, China. The stock solutions of met-als were prepared individually by dissolving the respective analyt-ical grade metal chloride salts (HAuCl4�4H2O, PdCl2, H2PtCl6�6H2O,CuCl2�2H2O, NiCl2�6H2O, FeCl3�6H2O and ZnCl2�6H2O) in 0.1 M HCl(M = mol L�1). The stock solutions were further diluted to the de-sired concentrations with HCl solutions of different concentrationsto prepare the test solutions. Aminoguanidine hydrochloride (CH6-

N4�HCl) and all other reagents used in the present study were pur-chased from Wako Pure Chemical Industries Ltd., Japan and were

used without further purifications. The ion-exchange resins; DIA-ION WA21 and DIAION SA11A were kindly donated by MitsubishiChemical Corp., Tokyo, Japan. The real industrial leach liquor ofPCBs of electronic scraps was kindly provided by Tanaka KikinzokuKogyo K.K., Hiratsuka, Japan and was used as received without fur-ther treatment.

2.2. Instrumentation

Metal ion concentrations in the aqueous solution before andafter adsorption were quantitatively measured by using inductivelycoupled plasma atomic emission spectrometer (ICP-AES, Shimadzumodel 8100 ICPS) and/or atomic absorption spectrophotometer(AAS, Shimadzu 6800). The FT-IR spectra of the feed material as wellas adsorption gel (before and after adsorption of metals) were re-corded by a Jasco model FTIR-410 spectrophotometer using KBrpellet method. The X-ray diffraction (XRD) and digital photographof elemental gold particles formed on the surface of adsorptiongel was recorded by using Rigaku L-094 X-ray diffractometer andKEYENCE VHX-1000 microscope, respectively.

2.3. Preparation of aminoguanidine modified persimmon tannin gel

The sample of crude PT powder is partly water soluble. Hence, itis necessary to render it water insoluble and strengthen its stabilitybefore it is applied as an adsorbent. This can be achieved throughchemical crosslinking (Gurung et al., 2011; Nakano et al., 2001).In the present case, the polyphenolic functional groups of persim-mon tannin were first crosslinked and then AG functional groupswere introduced onto the surface of crosslinked matrix by two-step chemical modification. At first, gelated tannin powder wasprepared by adopting a gelation procedure proposed by Nakanoet al. (2001). In a typical run, 14.0 g of tannin powder was dissolvedin 50 cm3 of 0.225 M NaOH solution at room temperature. Then,6 cm3 of 37% aqueous formaldehyde solution was added and themixture was stirred for 12 h at 353 K. The gelated product was fil-tered, washed repeatedly with distilled water until the filtrate wasneutral towards pH indicators and dried in a convection oven at353 K for 24 h. The chemical modification of gelated PT powderwith AG functional group was accomplished as shown in Scheme 1.As shown in the scheme, the polymer matrices of gelated PT werefirst chloromethylated by stirring the mixture of 8.0 g gelated tan-nin powder together with 8.0 g paraformaldheyde and 10 cm3 con-centrated HCl at 353 K for 24 h. After the crude product wasdiluted with distilled water, the solid was separated by filtrationand washed with distilled water several times until the filtrate be-came neutral. After drying in a convection oven overnight at 333 K,6.0 g of chloromethylated persimmon tannin was obtained. Then,the slurry of 6.0 g of chloromethylated product, 3.0 g of powderedaminoguanidine hydrochloride and 2.0 g of sodium carbonate in 50cm3 DMF was stirred at 363 K for 24 h. Thus obtained aminoguani-dine modified product was filtered and successively washed withdistilled water until neutral pH of the filtrate and then dried in aconvection oven at 333 K for 24 h. Finally, the dried product(4.5 g) was crushed and sieved into uniform particles of less than75 mesh size and named as AG–PT gel.

2.4. Batch wise adsorption tests

In order to investigate the adsorption behavior of the preparedgel towards Au(III), Pt(IV) and Pd(II) as well as some base metals,batch wise adsorption tests were first carried out using syntheticsolutions prepared in varying concentration of HCl ranging from0.1 to 5.0 M. In typical experiments, 0.01 g dry AG–PT gel was sha-ken together with 10 cm3 individual metal solution using a ther-mostatic shaker at 303 K for 24 h, which, as will be discussed

Scheme 1. Preparation route for AG–PT gel. (Hypothetical structure is shown where P stands for persimmon tannin matrix).

110 M. Gurung et al. / Bioresource Technology 129 (2013) 108–117

later, is sufficient to attain equilibrium. The heterogeneous mixturewas separated by filtration and the filtrate was collected for anal-ysis. From the measured initial and equilibrium concentrations ofmetal ions in aqueous solutions, percentage of each metal ion ad-sorbed on the adsorbent (%A) was calculated according to the rela-tion given in Eq. (1).

%A ¼ Ci � CCi

� 100 ð1Þ

where, Ci (mmol dm�3) and C (mmol dm�3) are initial and equilib-rium concentrations of metal ions in test solution, respectively.

Adsorption kinetics study was carried out to find the optimumtime required to attain equilibrium of adsorption of precious metalions on AG–PT gel. In this study, 0.1 g of dry gel was shaken to-gether with 100 cm3 of metal solution individually containing0.5 mM of Au(III), Pd(II) and Pt(IV) in 0.1 M HCl at 303 K. Definitevolume (5 cm3) of each mixture was sampled at different timeintervals for the analysis of residual metal concentration in thesolution. From the measured initial concentration and the concen-tration at any time, t, the amount of metal adsorbed, qt, was calcu-lated according to Eq. (2).

qt ¼Ci � Ct

W� V ð2Þ

where, Ci (mmol dm�3) and Ct (mmol dm�3) are the initial andremaining metal concentrations at any time t, respectively, W(mg) is the weight of dry gel and V (cm3) is the volume of testsolution.

The maximum metal uptake capacity of the gel towards pre-cious metals was evaluated from adsorption isotherm experimentsconducted by batch wise method. In this experiment, 0.01 g of drygel was shaken together with 10 cm3 of test solutions individuallycontaining varying concentrations of metal ions prepared in 0.1 MHCl solution for 96 h at constant temperature of 303 K in a thermo-static shaker. After equilibration, the metal ion concentrations inthe initial as well as final solutions were measured and the amountof adsorption (q (mol kg�1)) was calculated according to Eq. (3).

q ¼ Ci � CW

� V ð3Þ

where, V (cm3) and W(mg) are volume of test solution and weight ofthe dry gel, respectively.

The precious metals recovery from actual acidic leach liquor ofe-waste using the AG–PT gel was also examined by batch wiseadsorption test at different solid/liquid ratio, the ratio of the dryweight of the gel added to the volume of test solution, from 1 to20 g dm�3 keeping other parameters constant. The mixture wasshaken for 96 h at 303 K in a thermostatic shaker and percentagerecovery of each metal was calculated as mentioned earlier. Similartests were carried out using commercially available strong base an-ion exchange resin containing quaternary amine functional groupsnamely DIAION SA11A and weak base anion exchange resin con-taining secondary amine functional groups namely DIAION WA21under identical conditions for comparison of separation efficiencyof newly developed adsorbent with those of commercial resins.

2.5. Mutual separation of metals using packed column and reusabilityof adsorption gel

The potentiality of the present gel for mutual separation of pre-cious metals and coexisting base metals was also investigated bydynamic adsorption test using glass column packed with AG–PTgel. A glass column of 24.5 cm height and 8 mm internal diameterwas packed with 0.2 g of gel, which was soaked with deionizedwater before packing. The packed bed of the gel was supportedat both ends by cotton and glass beads and volume of the packedbed of the gel was calculated to be 0.60 cm3. Prior to passing thesynthetic feed solution, the column was conditioned by passingdistilled water for few hours followed by passing 0.1 M HCl for an-other couple of hours. Then, the synthetic feed solution containingrelatively low concentration (20 mg dm�3) each of Pd(II) and Pt(IV)and high concentration (100 mg dm�3) of Au(III) and Cu(II) pre-pared in 0.1 M HCl was percolated through the column at a con-stant flow rate of 5 cm3 h�1 using Iwaki model PST-100Nperistaltic pump. The effluent solution was collected at hourlyintervals by using a fraction collector (Biorad Model 2110 fractioncollector) and was analyzed to determine the residual metal con-centration. Elution test of the loaded metal ions was carried outin order to recover the adsorbed metal ions and to regenerate thegel for another cycle of adsorption. For this, the metal loaded col-umn was first washed with deionized water to expel any unboundmetal ions. Then, the solution of 0.5 M thiourea in 0.5 M HCl as aneluting agent was passed through the column at the same feed rate

M. Gurung et al. / Bioresource Technology 129 (2013) 108–117 111

of 5 cm3 h�1. The eluted solution was collected at hourly intervaland analyzed as before.

The repetitive test of the adsorption–elution cycle by continu-ous column method using AG–PT gel with synthetic solution con-taining Au(III), Pt(IV), Pd(II) and Cu(II) ions was carried out in thesimilar manner up to five cycles to evaluate the reusability of theadsorbent in continuous experiment.

3. Results and discussion

3.1. Characterization of the adsorbent

FT-IR spectra studies and elemental analyses were carried outfor the characterization of crude PT powder as well as the AG–PTgel.

3.1.1. FT-IR spectra of crude PT powder and AG–PT gelIn order to identify the surface functional groups on crude PT

powder and its modified product, FT-IR spectra were recorded.According to Li et al., persimmon tannin consists of phenol, ether,ester as well as quinone functional groups (Li et al., 2010). Accord-ingly, the broad absorption band in the range 3400–3200 cm�1 ofIR spectrum of crude PT powder (Supplementary Fig. 1a) is attrib-uted to O–H stretching vibration, the weak band at 1698 cm�1 andthe sharp band at 1611 cm�1 are attributed to C@O stretchingvibrations of ester and ketone groups, the bands at 1539 and1449 cm�1 are due to ring C–C stretching vibrations, the moderatebands at 1339, 1231 and 1087–1026 cm�1 region are attributed toO–H bending and symmetrical as well as asymmetrical C–O–Cstretching vibrations. It is reported that the guanidine compoundsdisplay four characteristic peaks in IR spectrum for the followinggroup frequencies: N–H stretching at about 3300 cm�1; C@Nstretching at 1689–1650 cm�1; N–H bending at about 1640 cm�1

and C–N stretching at about 1300 cm�1 (Yumei et al., 1999). As ex-pected from corresponding group frequencies, significant changeswere observed in 1680–1600 cm�1 and 1250–1050 cm�1 regionsof IR spectrum of AG–PT gel (Supplementary Fig. 1b) which corrob-orates the introduction of AG group on PT matrix. The moderateband centered at 1616 cm�1 in the IR spectrum of AG–PT gel isattributed to the combination of N–H bending and C@N stretchingvibrations and the broad band centered at 1124 cm�1 is assigned toC–N stretching vibrations. The N–H stretching band is not obviousin the spectrum because it is superimposed with broad O–Hstretching vibrations.

0

20

40

60

80

100

0 1 2 3 4 5

% A

dsor

ptio

n

HCl concnetration/mol dm-3

Au(III)

Pd(II)

Pt(IV)

Fe(III)

Ni(II)

Zn(II)

Cu(II)

Fig. 1. Effect of HCl concentrations on adsorption of various metal ions on the AG–PT gel. Initial concentration of metal ions = 0.2 mM, weight of dry gel = 0.01 g,volume of test solution = 10 cm3, shaking time = 24 h, temperature = 303 K.

3.1.2. Elemental analysisThe extent of AG functional groups immobilized on the polymer

matrix of persimmon tannin surface was evaluated from elementalanalysis of AG–PT gel and crude PT powder. The elemental compo-sition of crude PT powder and AG–PT gel is presented in Table 1. Asevident from the Table, content of nitrogen in crude PT powder isnegligible. The increase in nitrogen content to 12.9% in the caseof AG–PT gel is a direct consequence of immobilization of nitrogencontaining functional groups on the surface of feed material. Fromthe observed elemental composition, nitrogen density in AG–PT gelwas calculated as 9.21 mol kg�1 dry gel. Accordingly, the functionalgroup density immobilized on the PT powder is roughly2.5 mol kg�1 of the dry adsorbent. The substantial nitrogen contenton AG–PT gel indicates that immobilization of AG functionalgroups on the surface of crude PT powder was successfullyachieved through consecutive chemical reactions.

3.2. Adsorption behavior and selectivity test of AG–PT gel towardsvarious metal ions

The adsorption behavior of AG–PT gel towards precious as wellas base metal ions at varying hydrochloric acid concentration isshown in Fig. 1. Selectivity of the gel towards some precious metalsappears to be in the order Au(III) >> Pd(II) > Pt(IV). It was foundthat complete adsorption of Au(III), Pd(II) and Pt(IV) took placefrom 0.1 M hydrochloric acid solution. Adsorption of Au(III) was100% up to 2.0 M HCl concentration and it decreased with increas-ing acid concentration. Although the adsorption of Pd(II) was re-duced significantly after 0.5 M HCl concentration, that of Pt(IV)was dramatically decreased even when the acid concentration in-creased from 0.1 to 0.5 M. The adsorption efficiency of AG–PT geltowards Pd(II) and Pt(IV) became almost negligible in higher HClconcentration region. It is worth noting that the base metals werenot adsorbed at all on the AG–PT gel over the whole HCl concentra-tion region, indicating that the gel exhibits strong affinity to pre-cious metal ions over base metal ions.

The outstanding selectivity of AG–PT gel for precious metal ionsover base metal ions is attributable to the function of aminegroups, which are protonated and converted to positively chargedcentre in acidic medium, while Au(III), Pt(IV) and Pd(II) are con-verted into the respective chloro-anionic species, such as AuCl4

�,PtCl6

2� and PdCl42�, in chloride media. Consequently, the nega-

tively charged chloro-complexes of precious metals are attractedonto the positively charged centre of the adsorbent by electrostaticinteraction and are adsorbed by an anion exchange reaction as willbe described in the later section. In consequence, the decrease inthe adsorption of Au(III), Pd(II) and Pt(IV) on AG–PT gel at highHCl concentration, as can be seen in Fig. 1, is attributed partly tothe competitive adsorption of chloride ions. It is not uncommonthat the metals forming stable anionic chloro complexes, such asAu(III), Pd(II) and Pt(IV), exhibit decreasing adsorption behaviorwith increasing acid concentration. Some reports indicate that io-nic interaction and/or chelation mechanism or combination of bothare responsible for the sorption of soft metals like Au(III), Pt(IV)and Pd(II) with ligands containing N donor atoms (Donia et al.,2005; Iglesias et al., 1999; Park et al., 2010; Zhou et al., 2010).

Table 1Elemental composition of crude PT powder and AG–PT gel.

Adsorbents Elemental composition(%)

N-density (mol kg�1 of dry gel)

C H N

Crude PT powder 45.8 5.83 0.33 –AG–PT gel 49.9 4.00 12.9 9.21

0

0.1

0.2

0.3

0.4

0.5

0.6

0 2 4 6 8 10 12 14 16 18 20 22 24

q t(m

ol k

g-1)

Time (h)

Au(III)

Pd(II)

Pt(IV)

Fig. 2. Adsorption kinetics of Au(III), Pd(II) and Pt(IV) on AG–PT gel. Weight of drygel = 0.1 g, volume of test solution = 100 cm3, concentration of metal ions = 0.5 mM,[HCl] = 0.1 M, temperature = 303 K, shaking speed = 150 rpm.

0

1

2

3

4

5

6

7

8

9

10

0 10 20 30 40

q [m

ol k

g-1]

Ce [mmol dm-3]

Au(III)

Pd(II)

Pt(IV)

0

1

2

3

4

5

6

7

8

0 10 20 30 40

Ce/

q [g

dm

-3]

Ce/mmol dm-3

Au(III)

Pd(II)

Pt(IV)

a

b

Fig. 3. The adsorption isotherms of Au(III), Pd(II) and Pt(IV) on AG–PT gel. (a)Experimental plots; (b) Langmuir plots. [HCl] = 0.1 M, weight of dry gel = 0.01 g,volume of test solution = 10 cm3, shaking time = 96 h, temperature = 303 K.

112 M. Gurung et al. / Bioresource Technology 129 (2013) 108–117

The decrease in adsorption of Pd(II) and Pt(IV) with increasing acidconcentration, as will be discussed later, is therefore considered tobe related to the mechanism of adsorption as well.

In contrary to the precious metal ions, base metal ions existmostly as cationic or neutral species in hydrochloric acid med-ium and they are not attracted on the surface of positivelycharged adsorbent. Although iron exists in anionic form instrong HCl region (Adhikari et al., 2008), the present adsorbentis reluctant to capture this anion probably due to competitiveadsorption of chloride ion in high HCl region. It is obvious fromthe results of Fig. 1 that the gel can be applied for adsorptiveseparation of precious metal ions from coexisting base metalions. From the viewpoint of selectivity, it is expected thatAu(III) ions, which are present in minimum concentration inreal industrial solutions along with elevated concentrations ofbase metal ions, can be selectively separated from the basemetals in wide range of HCl medium. Pd(II) and Pt(IV) ionsare also separable from base metals at low concentration regionof HCl solution.

3.3. Adsorption kinetics of metal ions on AG–PT gel

The adsorption kinetics of Au(III), Pd(II) and Pt(IV) on AG–PTgel was studied by using 0.5 mM metal solutions in 0.1 M HCl.Fig. 2 shows the amount of metal adsorbed on the adsorbent asa function of shaking time. The adsorption of Au(III) and Pd(II)ions on the adsorbent was very fast at the initial stage. Almost85% Pd(II) and 83% Au(III) were adsorbed on the gel within0.5 h whereas only 30% Pt(IV) was adsorbed in the same time.After a fast adsorption period of nearly two hours, the amountof metal adsorbed on the gel was only slightly increased for Pd(II)as well as Au(III) and equilibrium was reached after 2 h for Pd(II)and 4 h for Au(III). The amount of Pt(IV) adsorbed on the AG–PTgel increased sluggishly with contact time and reached the equi-librium after 24 h. The fast kinetics of Au(III) and Pd(II) adsorp-tion on AG–PT is attributable to the easier access of anionicspecies of these metals to the positively charged sites (quaternaryammonium groups) of the adsorbent. Since PtCl6

2� is bulkier thanAuCl4

� and PdCl42�, access of the former to the active sites of the

adsorbent is more difficult than that of the latter. Although theequilibrium was reached within a few hours for Au(III) and Pd(II)and within 24 h for Pt(IV) in the case of 0.5 mM concentration ofmetal ions, the mixtures were shaken for 96 h to ensure the equi-librium in the adsorption isotherm tests.

3.4. Adsorption isotherms of precious metal ions on AG–PT gel

The maximum Au(III), Pd(II) and Pt(IV) uptake capacity of AG–PT gel was evaluated by adsorption isotherm experiments andthe results are presented in Fig. 3a. As evident from Fig. 3a, adsorp-tion of each metal ion on AG–PT gel increased rapidly with increas-ing metal concentration of test solution at low concentration, thenslightly increased with metal concentration and was followed by aplateau corresponding to each metal species at high concentrationof test solution, indicating that the adsorption of precious metalson AG–PT gel follows typical Langmuir adsorption isotherm. Con-sequently, the results shown in Fig. 3a were replotted accordingto the Langmuir equation described by Eq. (1) as shown inFig. 3b, which exhibits good linear relationships for all metals asexpected from Eq. (4).

Ce

qe¼ 1

qbmaxþ Ce

qmaxð4Þ

where Ce = equilibrium concentration of metal ion remained in thesolution after adsorption (mmol dm�3); qe = amount of metal ionadsorbed per unit weight of the gel (mol kg�1); qmax = maximumamount of metal ion adsorbed on the gel (mol kg�1); b = Langmuirconstant related to the energy of adsorption (dm3 mmol�1).

The various parameters calculated from Langmuir isothermmodel are presented in Table 2. The higher value of Langmuir con-stant (b) of the present gel towards Au(III) compared to Pd(II) and

Table 4Standard redox potentials (E0) of tannin gel and eachprecious metal ion in aqueous chloride medium.

Equilibrium E0 (V)

AuCl4�/Au +1.00

Tannin gel +0.74PtCl6

2�/PtCl42� +0.68

PdCl42�/Pd +0.62

M. Gurung et al. / Bioresource Technology 129 (2013) 108–117 113

Pt(IV) indicates the strong affinity of the adsorbent towards Au(III).Further, the apparent exceptionally high Au(III) loading capacity ofthe present gel compared to Pd(II) and Pt(IV) is attributable to theadsorption followed by reduction of Au(III) to elemental gold,which will be discussed later in details.

The value of the maximum adsorption capacity of AG–PT gel to-wards Pt(IV) and Pd(II) is comparable to other bioadsorbents mod-ified with amine groups but it has much higher Au(III) uptakecapacity than other bioadsorbents reported in the literature aslisted in Table 3.

3.5. Reduction of adsorbed Au(III) to metallic gold

We have previously reported the reduction of adsorbed Au(III)to metallic gold and formation of gold aggregates in the case ofadsorption of Au(III) on sulphuric acid crosslinked persimmon tan-nin (CPT) gel. Also in the present case, the adsorbed Au(III) specieswere subsequently reduced to elemental gold and the metallic goldflakes were clearly visible in the heterogeneous mixture of AG–PTgel and the test solution after adsorption (Supplementary Fig. 2a).Reduction of adsorbed Au(III) to metallic gold is also evident fromthe X-ray diffraction (XRD) pattern (Supplementary Fig. 2b) of thegel taken after adsorption of Au(III). The sharp peaks which ap-peared at two-theta values 38.220, 44.480, 64.740 and 780 in theXRD spectrum of Au(III) loaded AG–PT gel corresponds to that oftypical gold peaks, confirming the reduction of adsorbed gold spe-cies to metallic form. However, no such peaks were observed in theXRD spectrum of AG–PT gel taken after adsorption of Pt(IV) andPd(II), which indicates that the adsorbed species of Pt(IV) and Pd(II)were not reduced to their metallic form.

The distinct behavior of gold adsorption with tannin-basedadsorbents can be explained taking into account the redox poten-tial values of Au(III), Pt(IV) and Pd(II). Table 4 shows the redox po-tential values of tannin and the predominant species of eachprecious metal ion in our system (Ogata et al., 2007). Because theredox potential of the AuCl4

�/Au couple is much higher than that

Table 2Various parameters obtained from the Langmuir adsorption isotherm model.

Metal ions qmax (mol kg-1) b � 10�3 (dm3 mmol�1) R2

Au(III) 8.90 27.4 0.99Pd(II) 2.01 4.76 0.97Pt(IV) 1.01 2.42 0.98

Table 3Comparison of adsorption capacities of the present adsorbent with various kinds ofother adsorbents for precious metal ions reported in the literatures.

Adsorbents Adsorption capacity(mol kg�1)

Refs.

Au(III) Pd(II) Pt(IV)

PEI-modified Corynebacteriumglutamicum

– 1.66 – Won et al.(2011)

Pollyallylamine modifiedEscherichia coli

– 2.50 – Park et al.(2010)

2-Mercaptobenzothiazole-bonded silica gel

0.02 0.17 0.03 Pu et al.(1998)

Amberlite IRC 718 0.69 0.55 0.34 Park et al.(2000)

Glycine modified cross-linkedchitosan resin

0.86 1.13 0.62 Ramesh et al.(2008)

Ethylenediamine modifiedchitosan nanoparticles

– 1.30 0.87 Zhou et al.(2010)

Dimethylamine modified wastepaper

4.60 2.10 0.90 Adhikari et al.(2008)

AG–PT gel 8.90 2.01 1.01 Present work

of the tannin, the adsorbed AuCl4� species were subsequently re-

duced to elemental gold. However, as the PtCl62�/PtCl4

2� coupleand PdCl4

2�/Pd couple have redox potential lower than that ofthe tannin gel, it is unlikely that the adsorbed species are reducedto their metallic form.

The (poly)phenols are a class of compounds which are highlysensitive to oxidation reactions. Since the abundant phenolic hy-droxyl groups in the tannin matrix have strong tendency to get oxi-dized even under mild conditions, it is reasonable to infer that aredox reaction takes place due to electron transfer from polyphe-nolic matrix of AG–PT gel to the cationic centre of AuCl4

� species.A plausible electron transfer mechanism in the present case isshown in Scheme 2. As shown in the scheme, the phenolic groupsare oxidized to quinone groups and Au(III) is simultaneously re-duced to Au(0). We have already demonstrated the oxidation ofphenol to quinone during the reduction of Au(III) to Au(0) (Gurunget al., 2011).

3.6. Inferred adsorption mechanism

As mentioned in Section 3.2, phenomena such as electrostaticinteraction, ion exchange and chelation have been proposed by dif-ferent researchers for the adsorption of precious metals on adsor-bents modified with nitrogen containing ligands from acidicchloride media (Adhikari et al., 2008; Donia et al., 2005; Iglesiaset al., 1999; Park et al., 2010; Zhou et al., 2010). Ion pairing andelectrostatic interaction is considered to be the driving forcemainly responsible for adsorption of chloro anionic species ofAu(III), Pd(II) and Pt(IV) on AG–PT gel, which takes place by ion ex-change mechanism as shown in Scheme 3. Structurally, aminogua-nidine ligand is capable of forming 5-membered chelate complexwith metals. While the monoprotonated aminoguanidine has beenreported to coordinate with platinum as a monodentate ligandbound via the N4 amino group, the electroneutral N-aminoguani-dine can form 5-membered chelate complex with Pt and Pd chlorocomplexes in which aminoguanidine is coordinated to the metal bythe N-4 amino group and the deprotonated N-1 imino group (Ait-ken et al., 2007). In the present case also, chelation through nitro-gen atoms may take place in lower acid concentration regionleading to the formation of 5-membered chelate complex as de-picted in Scheme 3.

Analyzing the results of Fig. 1 from mechanistic approach, it ap-pears that the adsorption of Au(III) on the present adsorbent takesplace predominantly by ionic interaction mechanism whereas che-lation/coordination mechanism has significant contribution foradsorption of Pd(II) and Pt(IV). The notable decrease in percentageadsorption of Pd(II) and Pt(IV) with increasing HCl concentration is

Scheme 2. Electron transfer mechanism for reduction of Au(III) with polyphenolmatrix.

Scheme 3. Mechanism of adsorption of Au(III) and Pd(II) on AG–PT gel. Adsorption of Pt(IV) also follows the same mechanism.

114 M. Gurung et al. / Bioresource Technology 129 (2013) 108–117

attributed to the protonation at coordinating nitrogen atoms sothat coordination and subsequent formation of chelate structurebecomes increasingly difficult. On the other hand, adsorption ofAu(III), as evident from the results of Fig. 1, seems to be almostindependent of HCl concentration up to 3.0 M and it can be arguedthat chelation has no predominant effect for adsorption of Au(III)on the present adsorbent. The slight decrease in Au(III) adsorptionafter 3.0 M HCl concentration is rather due to competitive sorptionof chloride ions. Hence, it is not unreasonable to propose that theadsorption of Au(III) on AG–PT gel takes place predominantlydue to anion exchange coupled with electrostatic interaction andthat of Pt(IV) and Pd(II) takes place with the aid of chelation mech-anism, the driving force being anion exchange with chloride ion.

3.7. Chromatographic separation of precious metals from base metalsusing a packed column

The performance of AG–PT gel in continuous adsorption modefor selective recovery of precious metal ions from base metal ionwas studied using a column packed with the present adsorbent.Fig. 4a shows the breakthrough profiles of the tested metal ionsfrom the column packed with AG–PT gel, where bed volume(B.V.) represents the volume ratio of the feed solution that passesthrough the column to the volume of the packed gel. As seen fromthis figure, the breakthrough of Cu(II) took place immediately justafter the start of the flow, suggesting that it passes through the col-umn without being adsorbed on the gel. As expected from the re-sults of the batch wise adsorption tests, precious metals wereselectively adsorbed on the packed bed of the AG–PT gel. Break-

through of Pt(IV) occurred at 166 B.V. (18 h) and that of Pd(II) oc-curred at 848 B.V. (92 h), which are also in accordance to the higheraffinity of the gel towards Pd(II) over Pt(IV) as found in the batchexperiments. In the case of Au(III) adsorption, however, uniquebreakthrough profile was observed. The breakthrough started at250 B.V. (27 h) but the Au(III) concentration in the effluent solutiondid not increase smoothly as in the case of Pt(IV) and Pd(II), ratherit varied intermittently and a number of ups and downs were ob-served in the breakthrough profile till complete saturation wasachieved after 1835 B.V. The unusual breakthrough profile ofAu(III) as in the present case is believed to be due to successiveadsorption of Au(III) on the packed bed of AG–PT gel. As in the caseof batch experiment, the adsorbed Au(III) in column experimentwas also subsequently reduced to metallic gold and shining goldparticles were visible in the packed bed of the column. Hence, itis not unreasonable to infer that the reduced gold particles wereseparated from the gel surface and reactive sites were regeneratedat which more Au(III) ions were subsequently adsorbed. In conse-quence, the bed exhibited increased adsorption capacity towardsAu(III) and its concentration in the effluent solution was dramati-cally reduced. Repeated occurrence of this phenomenon until thesaturation of the bed led to the appearance of zigzags in the break-through curve. The dynamic adsorption capacity of the gel calcu-lated from the breakthrough profiles were 1.12 mol kg�1

(220 mg g�1) for Au(III), 0.43 mol kg�1 (46.0 mg g�1) for Pd(II)and 0.07 mol kg�1 (13.7 mg g�1) for Pt(IV), respectively. The calcu-lated amounts of adsorbed metal ions in the column were found tobe much lower than the corresponding equilibrium adsorptioncapacities observed in the batch wise adsorption tests, which

0

0.2

0.4

0.6

0.8

1

0 200 400 600 800 1000 1200 1400 1600 1800

Ct/C

i(-

)

B.V.(-)

Au(III)

Pd(II)

Pt(IV)

Cu(II)

0

20

40

60

80

100

120

140

0 20 40 60 80 100 120 140

Ct/C

i (-)

B.V.(-)

Au(III)

Pd(II)

Pt(IV)

a

b

Fig. 4. (a) Breakthrough profiles for Au(III), Pd(II), Pt(IV) and Cu(II) with a columnpacked with the AG–PT gel. Feed concentrations (mg dm�3): Au(III) = 100,Cu(II) = 100, Pd(II) and Pt(IV) = 20 each. (b) Elution profiles for Au(III), Pd(II) andPt(IV) with 0.5 M thiourea in 0.5 M HCl. Weight of gel = 0.2 g, flow rate = 5 cm3 h�1.

0

20

40

60

80

100

0 2 4 6 8 10 12 14 16 18 20

% A

dsor

ptio

n

S/L (mg cm-3)

0

20

40

60

80

100

0 2 4 6 8 10

% A

dsor

ptio

n

S/L (mg cm-3)

0

20

40

60

80

100

0 2 4 6 8 10

% A

dsor

ptio

n

S/L (mg cm-3)

a

b

c

Au(III)

Pd(II)

Pt(IV)

Cu(II)

Ni(II)

Zn(II)

Fe(III)

Au(III)

Pd(II)

Pt(IV)

Fe(III)

Cu(II)

Ni(II)

Zn(II)

Au(III)

Pd(II)

Pt(IV)

Fe(III)

Cu(II)

Ni(II)

Zn(II)

Fig. 5. Adsorption of various metal ions from acidic leach liquor of e-waste scrapsusing; (a) AG–PT gel, (b) DIAION WA21, (c) DIAION SA11A. Metal concentrations(mg dm�3); Cu(23.8), Zn(5.62), Ni(3.04), Fe(667), Au(331), Pd(121), Pt(743), tem-perature = 303 K, shaking time = 96 h.

M. Gurung et al. / Bioresource Technology 129 (2013) 108–117 115

may be attributed to various phenomena, such as channeling flowof the feed solution, compactness of the adsorbent so that fewernumber of active sites are available for adsorption, competitiveadsorption of metal ions, and slow kinetics of adsorption or shortcontact time between metal ion and the gel, etc.

After the column was saturated, elution of loaded metals wascarried out by using mixture of 0.5 M thiourea and HCl. The elutionprofiles of the loaded precious metal ions are presented in Fig. 4b.The sharp elution profiles of these metals demonstrate that elutionis easy and efficient for all metal ions since almost 95%, 97% and98% of the adsorbed Pd(II), Pt(IV) and Au(III), respectively, wereeluted within less than 12 h. From the measured concentration ofeach metal ions in the eluted sample, preconcentration factors ofAu(III), Pt(IV) and Pd(II) were evaluated as 23.5, 29.3 and 134,respectively. Since the initial concentration of Au(III) was fivetimes higher than that of Pd(II), the preconcentration factor forAu(III) seems to be lower than that of Pd(II). From the results ofcolumn experiment, it can be concluded that AG–PT gel is promis-ing for achieving selective and effective separation of trace amountof precious metals from industrial liquors and waste streams con-taining substantial amount of base metals.

3.8. Regeneration and reutilization of the adsorbent

Regeneration of the adsorbent and its repeated use for anothercycle is essentially important from cost-effective perspective. The

practical reusability of the gel for successive cycles was examinedby adsorption/elution cycles for at least five repeated cycles bycontinuous mode and the results are presented in Table 4. In caseof the first cycle, the test solution was passed through the bed untilcomplete saturation was reached (Fig. 4a) whereas the test solu-

Table 5Performance of the AG–PT gel in consecutive adsorption/elution cycles.

Cycles Element Adsorbed (mg) Eluted (mg) % Recovery

1a Au 220 215 97.7Pd 46.0 44.5 96.7Pt 14.4 14.0 97.2

2b Au 16.1 15.7 97.5Pd 16.0 15.4 96.0Pt 12.1 11.8 97.5

3 Au 16.1 15.9 98.7Pd 16.8 16.0 95.0Pt 12.7 12.1 96.2

4 Au 17.0 16.7 98.2Pd 16.8 16.2 96.4Pt 13.9 13.5 97.4

5 Au 17.1 16.6 97.0Pd 16.7 16.0 96.1Pt 13.2 13.8 97.7

a For cycle (1), feed time = until saturation, feed concentration: 100 mg dm�3 forAu(III), 20 mg dm�3 for ((Pd(II), Pt(IV)).

b For cycle (2–5), feed time = 48 h, feed concentration: 20 mg dm�3 ((Au(III),Pd(II), Pt(IV)).

116 M. Gurung et al. / Bioresource Technology 129 (2013) 108–117

tion was passed only for 48 h in the latter cycles. For this reason, asignificant difference in the amount of metal loaded is observed inother cycles, for example in cycle 2. It is understandable from theresults of Table 5 that almost quantitative elution recoveries of me-tal ions were achieved for at least 5 repeated cycles with undimin-ished metal uptake, suggesting the possibility of repeated use ofthe AG–PT gel for the design of a continuous sorption process.

In addition to quantitative recovery of precious metals throughadsorption followed by elution, it is also mandatory that the elu-tion process should regenerate the adsorbent close to the originalcondition for effective reuse. The FT-IR spectra of Au(III) loadedand regenerated gel after elution are presented in (SupplementaryFig. 3). After metal loading, significant changes were observed inthe region of aminoguanidine group frequencies which is inferredto the participation of aminoguanidine group for adsorption ofAu(III) species on the gel. The IR spectrum of regenerated gel afterelution of the loaded metal (Supplementary Fig. 3c) is commensu-rate with the fresh adsorbent (Supplementary Fig. 3a). These re-sults indicate that the adsorbent was regenerated without anyphysical/chemical damage and can be used repeatedly for anothercycle of application.

Table 6General characteristics of the ion-exchange resins.

Properties DIAION SA11

Chemical structure

Matrix PolystyreneFunctional group Quaternary amTotal exchange capacity (meq/mL) >0.85Water content (%) 55–65Maximum temperature <80Mean bead size (nm) >0.40

3.9. Application of AG–PT gel for selective recovery of precious metalsfrom industrial solution

Based on the results of the adsorption tests using synthetic testsolutions, the applicability of the gel was examined for the recov-ery of precious metals from actual industrial leach liquor of e-waste donated by Tanaka Kikinzoku Kogyo Co. Ltd., Japan. Thisindustrial leach liquor (named as TKK solution) was produced inthe conventional leaching of spent PCBs (printed circuit boards)of waste electronic devices with chlorine containing hydrochloricacid (Cl2/HCl) followed by nitric acid leaching for the removal ofsilver and major parts of base metals. The concentrations of variousmetal ions contained in the liquor were determined quantitativelyand were found to be (mg dm�3): Cu(23.8), Zn(5.62), Ni(3.04),Fe(667), Au(331), Pd(121), Pt(743). Although the leach liquor ofwaste electronic devices usually contains elevated concentrationof base metals, this sample contained comparatively low concen-tration of base metals and, consequently, excess amount of pre-cious metals like gold, palladium and platinum over base metals.The acid concentration of this sample was determined to be0.5 M by neutralization titration. In addition, since the samplewas obtained after leaching with Cl2/HCl, it is expected that thepredominant species of all precious metal ions are anionic chlorocomplexes and this sample was used without further treatment.The adsorption tests were carried out batch wise at varying so-lid–liquid (S/L) ratio, which is the ratio of weight of dry gel tothe volume of test solution. In order to compare the efficiency ofthe gel prepared in the present study, the adsorption behavior ofcommercially available anion exchange resins, DIAION SA11Aand DIAION WA21, was also studied under identical conditions.The characteristics of these resins are presented in Table 6.

Fig. 5a–c show the comparative plots illustrating the effect ofadsorbent dose on percentage adsorption of various metal ionsfrom TKK solution on AG–PT gel, DIAION SA11A and DIAIONWA21, respectively. It is apparent from Fig. 5a that Au(III) can bequantitatively recovered even at low S/L ratio of 1 g dm�3. It isnoteworthy that mutual separation of Au(III) and Pt(IV) from thereal leach liquor is possible with AG–PT gel by adjusting the S/L ra-tio of the adsorption system. Also interesting is that Pd(II) andPt(IV) were almost quantitatively adsorbed on the gel at S/L ratioof 15 and 20 g dm�3 but the adsorption of base metal ions was neg-ligible which not only ensures the quantitative recovery of pre-cious metals from base metals but also gives the precious metalconcentrates free from base metals. As can be seen in Fig. 5b and

DIAION WA21

Polystyreneine Polyethylene polyamine

>2.040–52<100>0.40

M. Gurung et al. / Bioresource Technology 129 (2013) 108–117 117

c, the commercial resins exhibited poor separation efficiency ofprecious metals compared to AG–PT gel. Although almost quantita-tive adsorption of precious metals on the commercial resins wasachieved at around S/L ratio of 10 g dm�3, DIAION WA21 adsorbednearly 40% zinc and DIAION SA11A adsorbed more than 90% zincwhich demands additional operating steps for purification of pre-cious metals. Since selectivity towards target metal is one of themost important aspects of adsorptive separation, AG–PT gel seemsto be a more effective alternative to commercial resins such asDIAION WA21 and DIAION SA11A. These results suggest that thepresent adsorbent is a promising candidate for adsorptive precon-centration and recovery of precious metals from acidic chloridemedia containing substantial quantity of base metals.

4. Conclusions

Aminoguanidine functionalized persimmon tannin gel was pre-pared by simple chemical reactions. The gel exhibited remarkableselectivity towards Au(III) followed by Pd(II) and Pt(IV). SinceAu(III), Pd(II) and Pt(IV) form chloroanionic species in hydrochloricacid medium, adsorption of these metal ions occurred by ion pair-ing at the positively charged quaternary nitrogen atoms on the gel.The gel can be used repeatedly with undiminished metal uptakecapacity. When compared with the commercially available anionexchange resins, the functional persimmon tannin gel preparedin the present work exhibited remarkable selectivity and greatpotentiality to separate precious metal ions from base metal ions.

Acknowledgements

The authors gratefully acknowledge support by a Grant-in-Aidfor Scientific Research about Establishing a Sound material-CycleSociety (K2131) from the Ministry of Environment of JapaneseGovernment. The authors also gratefully acknowledge the supplyof crude persimmon tannin powder from Persimmon-Kaki Tech-nology Development Co. Ltd., China. The authors thank the TanakaKikinzoku Kogyo (TKK) Co. Ltd., Japan, for their kind supply of theacidic leach liquor sample of electronic waste.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.biortech.2012.11.012.

References

Adhikari, C.R., Parajuli, D., Kawakita, H., Inoue, K., Ohto, K., Harada, H., 2008.Dimethylamine-modified waste paper for the recovery of precious metals.Environ. Sci. Technol. 42, 5486–5491.

Aitken, D.J., Albinati, A., Gautier, A., Husson, H.-P., Morgant, G., Nguyen-Huy, D.,Kozelka, J., Ongeri, P.S., Rizzato Viossat, S.B., 2007. Platinum(II) andpalladium(II) complexes with N-Aminoguanidine. Eur. J. Inorg. Chem. 2007,3327–3334.

Anticó, E., Masana, A., Salvadó, V., Hidalgo, M., Valiente, M., 1994. Adsorption ofpalladium by glycolmethacrylate chelating resins. Anal. Chim. Acta 296, 325–332.

Cortina, J.L., Meinhardt, E., Roijals, O., Martí, V., 1998. Modification and preparationof polymeric adsorbents for precious-metal extraction in hydrometallurgicalprocesses. React. Funct. Polym. 36, 149–165.

Donia, A.M., Atia, A.A., Elwakeel, K.Z., 2005. Gold(III) recovery using syntheticchelating resins with amine, thio and amine/mercaptan functionalities. Sep.Purif. Technol. 42, 111–116.

Gulko, A., Feigenbaum, H., Schmuckler, G., 1971. Separation of palladium(II) andplatinum(II) chlorides by means of a guanidine resin. Anal. Chim. Acta 51, 397–402.

Gurung, M., Adhikari, B.B., Kawakita, H., Ohto, K., Inoue, K., Alam, S., 2011. Recoveryof Au(III) by using low cost adsorbent prepared from persimmon tannin extract.Chem. Eng. J. 17, 556–563.

Hagelüken, C., Meskers, C. 2009. Technology Challenges to Recover Precious andSpecial Metals from Complex Products. Available from: <http://ewasteguide.info/files/Hageluecken_2009_R09.pdf>.

Hubicki, Z., Wołowicz, A., 2009. Adsorption of palladium(II) from chloride solutionson Amberlyst A 29 and Amberlyst A 21 resins. Hydrometallurgy 96, 159–165.

Iglesias, M., Anticó, E., Salvadó, V., 1999. Recovery of palladium(II) and gold(III) fromdiluted liquors using the resin duolite GT-73. Anal. Chim. Acta 381, 61–67.

Kolarz, B.N., Jermakowicz-Bartkowiak, D., Jezierska, J., Apostoluk, W., 2001. Anionexchanger with alkyl substituted guanidyl groups gold sorption and Cu(II)coordination. React. Funct. Polym. 48, 169–179.

Kordosky, G.A., Sierakoshi, J.M., Virnig, M.J., Mattison, P.L., 1992. Gold solventextraction from typical cyanide leach solutions. Hydrometallurgy 30, 291–305.

Li, J., Lu, H., Guo, J., Xu, Z., Zhou, Y., 2007. Recycle technology for recoveringresources and products from waste printed circuit boards. Environ. Sci. Technol.411, 995–2000.

Li, C., Reed, J.D., Hagerman, A.E., 2010. Fractionation and characterization of highmolecular weight proanthocyanidin from persimmon fruit by thiolysis–HPLC,SEC, MALDI-TOF/MS and NMR. Presented at the XXVth International Conferenceon Polyphenols, Montpellier, France.

Matsuo, T., Ito, S., 1978. On mechanism of removing astringency in persimmonfruits by carbon dioxide treatment. Part(II). The chemical structure of Kaki-tannin from immature fruit of the persimmon (Diopyros Kaki L.). Agric. Biol.Chem. 42, 1637–1643.

Myasoedova, G.V., Antokol’SKAYA, I.I., Savvin, S.B., 1985. New chelating sorbents fornoble metals. Talanta 32, 1105–1112.

Nakajima, A., Baba, Y., 2004. Mechanism of hexavalent chromium adsorption bypersimmon tannin gel. Water Res. 38, 2859–2864.

Nakajima, A., Sakaguchi, T., 1993. Uptake and recovery of gold by immobilizedpersimmon tannin. J. Chem. Technol. Biotechnol. 57, 321–326.

Nakajima, A., Sakaguchi, T., 2000. Uptake and removal of iron by immobilizedpersimmon tannin. J. Chem. Technol. Biotechnol. 57, 977–982.

Nakano, Y., Takeshita, K., Tsutsumi, T., 2001. Adsorption mechanism of hexavalentchromium by redox within condensed-tannin gel. Water Res. 35, 496–500.

Ogata, T., Nakano, Y., 2005. Mechanism of gold recovery from aqueous solutionsusing a novel tannin gel adsorbent synthesized from natural condensed tannin.Water Res. 39, 4281–4286.

Ogata, T., Kim, Y.-H., Nakano, Y., 2007. Selective recovery process of gold utilizing afunctional gel derived from natural condensed tannin. J. Chem. Eng. Jpn. 40,270–274.

Park, C., Chung, J.S., Cha, K.W., 2000. Separation and preconcentration method forpalladium, platinum and gold from some heavy metals using Amberlite IRC 718chelating resins. Bull. Korean Chem. Soc. 21, 121–124.

Park, J., Won, S.W., Mao, J., Kwak, I.S., Yun, Y.-S., 2010. Recovery of Pd(II) fromhydrochloric solution using pollyallylamine hydrochloride-modified Escherichiacoli biomass. J. Hazard. Mater. 181, 794–800.

Pu, Q., Su, Z., Hu, Z., Chang, X., Yang, M., 1998. 2-Mercaptobenzothiazole-bondedsilica gel as selective adsorbent for preconcentration of gold, platinum andpalladium prior to their simultaneous inductively coupled plasma opticalemission spectrometric determination. J. Anal. At. Spectrom. 13, 249–253.

Ramesh, A., Hasegawa, H., Sugimoto, W., Maki, T., Ueda, K., 2008. Adsorption ofgold(III), platinum(IV) and palladium(II) onto glycine modified crosslinkedchitosan resin. Bioresour. Technol. 99, 3801.

Shah, R., Devi, S., 1997. Preconcentration and separation of palladium(II) andplatinum(IV) on a dithizone anchored poly(vinylpyridine)-based chelatingresin. Anal. Chim. Acta 341, 217–224.

Volesky, B., Holan, Z.R., 1995. Biosorption of heavy metals. Biotechnol. Prog. 11,235–250.

Wołowicz, A., Hubicki, Z., 2009. Palladium(II) complexes adsorption from thechloride solutions with macrocomponent addition using strongly basic anionexchange resins, type 1. Hydrometallurgy 98, 206–212.

Won, S.W., Park, J., Mao, J., Yun, Y.-S., 2011. Utilization of PEI-modifiedCorynebacterium glutamicum biomass for the recovery of Pd(II) in hydrochloricsolution. Bioresour. Technol. 102, 3888–3893.

Yumei, Z., Jianming, J., Yanmo, C., 1999. Synthesis and antimicrobial activity ofpolymers guanidine and biguanidine salts. Polymer 40, 6189–6198.

Yun, Y.-S., Park, D., Park, J.M., Volesky, B., 2001. Biosorption of trivalent chromiumon the brown seaweed biomass. Environ. Sci. Technol. 35, 4353–4358.

Yurtsever, M., S�engil, _I.A., 2009. Biosorption of Pb(II) ions by modified quebrachotannin resin. J. Hazard. Mater. 163, 58–64.

Zhou, L., Xu, J., Liang, X., Liu, Z., 2010. Adsorption of platinum(IV) and palladium(II)from aqueous solutions by magnetic crosslinking chitosan nanoparticlesmodified with ethylenediamine. J. Hazard. Mater. 182, 518–524.

Zuo, G., Muhammed, M., 1995. Thiourea-based coordinating polymers: synthesisand binding to noble metals. React. Polym. 24, 165–181.