The Recovery Of Pt(IV) From Aqueous Solutions By APDEMS ... · varioELcube V4.0.13 (Elementar, Germany) was used for elemental analysis. Textural properties (surface area, pore size

1111Wolkersdorfer, Ch.; Sartz, L.; Weber, A.; Burgess, J.; Tremblay, G. (Editors)

The recovery of Pt(IV) from aqueous solutions by APDEMS-functionalised zeolite

Alseno K. Mosai, Luke Chimuka, Ewa M. Cukrowska, Izak A. Kotźe and Hlanganani Tutu

Molecular Sciences Institute, School of Chemistry, University of the Witwatersrand, Private Bag X3, WITS, 2050, South Africa

Abstract� e recovery of Pt(IV) from aqueous solutions by zeolite functionalised with 3-aminopropyl(diethoxy)methylsilane (APDEMS) was studied at various adsorbent dosage, pH, time, concentration and competing ions was studied. � e adsorbent was highly e� cient at pH 2 and the maximum adsorption was reached at 180 min. � e adsorption of Pt(IV) was not a� ected by the presence of competing ions instead, other precious metals and trace elements were adsorbed. � e strong Pt-adsorbent interaction concluded from kinetic and isotherms models makes the adsorbent suitable for the re-covery of Pt(IV) and other metals from mining wastewater.Keywords: Pt(IV), recovery, platinum group elements, APDEMS-functionalised zeolite

Introduction Large amount of water is used during min-eral processing and the resulting wastewater is discharged into the environment (Lotter-moser 2003). � e discharged water normally contains signi� cant concentrations of met-als. Studies have shown that platinum group metals (PGMs) are discharged along with wastewater as by-products of nickel and cop-per mining (Xiao and Laplante 2004). More-over, PGMs mining plants lose signi� cant amounts of PGMs during processing (Hunt et al. 2013). � us, the accumulation of PGMs could be quite signi� cant in the environ-ment and their recovery could add value to the economy of many countries dependant on them. PGMs are widely used in jewellery, catalytic converters, laboratory equipment and medicine (Chassari et al 2005; Xiao and Laplante 2004). � e production of PGMs has increased over the years due to their increas-ing demand and since only a few countries (South Africa, Canada and Russia) have large reserves, there is a danger of the scarcity of their minerals worldwide and thus a need for recovery from wasteawter solutions (Alonso et al 2009).

Zeolites are aluminisilicate clay minerals which are among the cheapest naturally avail-able adsorbents. � ey consist of SiO4 and AlO4

frameworks which are linked through oxygen atoms in a three-dimentional array. � e over-all charge of zeolite is negative due to the re-placement of Si4+ with Al3+ and is balanced by cations such as Ca, Mg, Na and K in the cavi-ties of the zeolite (Malamis and Katsou 2013). Zeolites have been found to be good adsor-bents for trace metals due to their high cation exchange capacity and molecular sieve prop-erties (Bakatula et al. 2015). Moreover, stud-ies have shown that the adsorption capacities of zeolites can be improved by modi� cation where the zeolite acts as a support (Ren et al. 2016). Platinum has been found to interact strongly with nitrogen and thus, the aim of the study to use zeolite functionalised with 3-aminopropyl(diethoxy)methylsilane (AP-DEMS) which has a primary amine group to recover Pt(IV) from aqueous solutions.

Materials and methods Chemicals � e chemicals used in this study were of ana-lytical reagent grade and were obtained from Sigma Aldrich, South Africa. � e 10 mg L-1

platinum stock solution was prepared from Pt(IV) standard and deionised water in 1 L volumetric � ask and was stored in the refrig-erator (4°C) until required. Working solu-tions with speci� c concentrations were pre-

Topic 17.indb 1111 2018/09/03 07:27

11th ICARD | IMWA | MWD Conference – “Risk to Opportunity”

1112 Wolkersdorfer, Ch.; Sartz, L.; Weber, A.; Burgess, J.; Tremblay, G. (Editors)

pared daily from the stock solution through serial dilutions. � e pH of the solutions was controlled with 0.01 mol L-1 NaOH and 0.01 mol L-1 HCl. � e stock solution of the mixed metals was prepared in the same way.

Functionalisation of zeolite and charac-terisation� e zeolite was purchased from Sigma Aldrich (Steinheim, Germany). � e chemical func-tionalization was achieved by mixing 5g zeo-lite with 10 mL of 3-aminopropyl(diethoxy)methylsilane (bp: 85 – 88˚C) in a glass vial and the contents were interacted at 180°C for 30 min in an automated Microwave-assist-ed Extraction monowave 450 (Anton Paar, South Africa). � e contents were cooled to room temperature and � ltered using 0.45um cellulose natrate � ltech � lter paper using suc-tion and rinsed with ethanol. � e function-alised zeolite was then dried at 100°C for 24 h.

� e natural and functionalised zeolite were characterised with fourier transform infrared (FTIR) spectroscopy in the range of 4000 – 400 cm-1 for the determination of functional groups, the mineralogy was deter-mined with powder X-ray di� raction (PXRD) D2 Phaser (Bruker, Germany), X-ray � uores-cence (XRF) (PAnalytical, Netherlands) was used to determine the chemical composition, varioELcube V4.0.13 (Elementar, Germany) was used for elemental analysis. Textural properties (surface area, pore size and pore volume) were analysed with Micromeritics Flow Prep 060 instrument (Aachen, Ger-many). � e point of zero charge (PZC) was also determined using KCl for ionic strength, HNO3 and NaOH to control pH.

Batch adsorption studies� e study was carried out by placing the ad-sorbent and 10 mL Pt(IV) solution at various conditions in 50 mL polypropylene contain-ers which were shaken at room temperature with an elliptical benchtop shaker (Labcon, USA) to attain equilibrium and the contents were � ltered with 0.45 μm cellulose nitrate Filtech � lter paper and the � ltrate was anal-ysed with inductively coupled plasma optical emission spectroscopy (ICP-OES) (Spectro genesis, Germany).

E� ect of adsorbent mass (50 – 500 mg) on the adsorption of Pt(IV) by APDEMS-functionalised zeolite was determined at pH 2 and room temperature (25°C). E� ect of pH (2 – 9), concentration of Pt(IV) (0.5 – 5 mg L-1) and contact time (0 – 540 min) on the adsorption of Pt(IV) was determined. Com-peting ions (Rh(III), Pd(II), Ir(III), Ru(III), Os(III), Fe(III), Ca(II), Mg(II), K(I), Co(II), Ni(II), Hf(IV), Zn(II) and Au(I)) were pre-pared with Pt(IV) and exposed to APDEMS-functionalised zeolite to determine their ef-fect on adsorption or recovery of Pt(IV).

Data treatment� e adsorption capacity, qe (mg g-1) and the adsorption e� ciency were calculated using equations 1 and 2, respectively.

(1)

(2)

where Co and Ce are initial and equilibrium concentrations of metals (mg L-1), V is the volume of the solution used (L) and M is the mass of the adsorbent (g).Adsorption isotherms� e type of adsorption occurring between Pt(IV) and APDEMS-functionalised zeolite was determined with Langmuir, Freundlich and Dubinin Radushkevich (D-R) adsorp-tion isotherms. � e Langmuir isotherm de-scribes the monolayer adsorption between adsorbents and substances (Mushtaq et al. 2016). � e Freundlich isotherm describes the non-ideal and reversible adsorption of ele-ments onto the heterogeneous surface (Chen 2015; Mushtaq et al. 2016). � e Dubinin-Ra-dushkevich (D-R) isotherm describes the ad-sorption mechanism with a Gaussian energy distribution onto a heterogeneous surface (Chen 2015). � e Langmuir, Freundlich and Dubinin Radushkevich (D-R) adsorption can be expressed as, respectively:

(3)

(4)

(5)


(1)

(2)

(3)

(4)

(5)

2

Functionalisation ofzeolite and characterisation

The zeolite was purchased from Sigma Aldrich (Steinheim, Germany). The chemicalfunctionalization was achieved by mixing 5g zeolite with 10 mL of 3-aminopropyl(diethoxy)methylsilane (bp: 85 – 88˚C) in a glass vial and the contents wereinteractedat 180°Cfor 30min in an automated Microwave-assisted Extraction monowave 450(Anton Paar, South Africa). The contents were cooled to room temperature and 2iltered using0.45um cellulose natrate 1iltech 1ilter paper using suction and rinsed with ethanol. Thefunctionalisedzeolite was thendriedat 100°Cfor 24h.

Thenaturalandfunctionalisedzeolitewerecharacterisedwithfouriertransforminfrared(FTIR)spectroscopy in the range of 4000– 400 cm-1 for the determination of functional groups, themineralogy was determined with powder X-ray diffraction (PXRD) D2 Phaser (Bruker,Germany), X-ray %luorescence (XRF) (PAnalytical, Netherlands) was used to determine thechemical composition, varioELcube V4.0.13 (Elementar, Germany) was used for elementalanalysis. Textural properties (surface area, pore size and pore volume) were analysed withMicromeritics Flow Prep 060 instrument (Aachen, Germany). The point of zero charge (PZC)wasalsodeterminedusingKClforionicstrength,HNO3 andNaOH to control pH.

Batchadsorption studies

The study was carried out by placing the adsorbent and 10 mL Pt(IV) solution at variousconditions in50mLpolypropylenecontainers which wereshaken atroom temperature with anelliptical benchtop shaker (Labcon, USA) to attain equilibrium and the contents were +ilteredwith 0.45μm cellulose nitrate Filtech ,ilter paper and the (iltrate was analysed with inductivelycoupledplasma optical emissionspectroscopy (ICP-OES) (Spectrogenesis, Germany).

Effect of adsorbent mass(50 - 500mg) onthe adsorptionof Pt(IV) by APDEMS-functionalisedzeolitewasdeterminedatpH2androomtemperature(25°C). Effect ofpH (2 – 9), concentrationof Pt(IV) (0.5 - 5 mg L-1) and contact time (0 - 540 min) on the adsorption of Pt(IV) wasdetermined. Competing ions (Rh(III), Pd(II), Ir(III), Ru(III), Os(III), Fe(III), Ca(II), Mg(II), K(I),Co(II), Ni(II), Hf(IV), Zn(II) and Au(I)) were prepared with Pt(IV) and exposed to APDEMS-functionalisedzeolite to determine their effect onadsorption or recoveryof Pt(IV).

Data treatment

The adsorption capacity, qe (mg g-1) and the adsorption ef/iciency were calculated using

𝑞𝑞𝑞𝑞! =!!!!! !

!(1)

𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 = !!!!!!!

𝑥𝑥𝑥𝑥 100 (2)

whereCo and Ce are initial andequilibrium concentrations of metals (mg L-1),Visthevolumeofthe solution used(L) andM is the mass of the adsorbent (g).

Adsorptionisotherms

The type of adsorption occurring between Pt(IV) and APDEMS-functionalised zeolite wasdetermined with Langmuir, Freundlichand DubininRadushkevich(D-R) adsorption isotherms.The Langmuir isotherm describes the monolayer adsorption between adsorbents andsubstances (Mushtaq et al. 2016). The Freundlich isotherm describes the non-ideal andreversible adsorption of elements onto theheterogeneous surface (Chen 2015; Mushtaq et al.2016). The Dubinin-Radushkevich (D-R) isotherm describes the adsorptionmechanism with aGaussian energy distribution onto a heterogeneous surface (Chen 2015). The Langmuir,FreundlichandDubininRadushkevich(D-R) adsorption can beexpressed as, respectively:

⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=

mLe

me

e

qKC

qqC 11 (3)

2







Data treatment

The adsorption capacity, qe (mg g-1) and the adsorption ef/iciency were calculated usingequations 1 and 2, respectively.

𝑞𝑞𝑞𝑞! =!!!!! !

!(1)


𝑥𝑥𝑥𝑥 100 (2)


Adsorptionisotherms


⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=

mLe

me

e

qKC

qqC 11 (3)

2







Data treatment


𝑞𝑞𝑞𝑞! =!!!!! !

!(1)


𝑥𝑥𝑥𝑥 100 (2)


Adsorptionisotherms


⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=

mLe

me

e

qKC

qqC 11 (3)

3

nefe CKq1

= (4)

𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑞𝑞𝑞𝑞! = 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑋𝑋𝑋𝑋! − 𝐾𝐾𝐾𝐾!"𝜀𝜀𝜀𝜀! (5)

where qe (mg g-1) is the quantity of the adsorbate adsorbed per gram of adsorbent atequilibrium, qm (mg g-1) is themaximum monolayer coveragecapacityand KL is theLangmuirisotherm constant (Lmol-1), Kf ((mg g-1)/(mol L-1)1/n) is the Freundlichisotherm constant andnis the adsorptionintensity, Kad (mol2 (kJ2)-1) is the Dubinin-Radushkevichisotherm constant, Xm(mgg-1) is theoretical isotherm saturation capacity and ɛis thePolanyi constant.

Kinetic models

The adsorption process controlling mechanism between Pt(IV) and APDEMS-functionalisedzeolite was determined with pseudo 2irst- and second order kinetic models represented byequations 6 and7, respectively (Bakatula et al 2015). The correlation coef,icient (R2)and thecloseness of the experimental adsorption capacity, qexp to the calculated, qcal were used todeterminethemodel that best decribes thekinetic data.

log 𝑞𝑞𝑞𝑞! − 𝑞𝑞𝑞𝑞! = 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑞𝑞𝑞𝑞! −!!!!.!"!

(6)!!!= !

!!!!!+ !

!!t (7)

where qt (mg g-1) and qe (mg g-1) are the adsorption capacities at time t and equilibrium,respectively,k1 (min-1)and k2 (gmg-1min-1)are the rate constants.

Results anddiscussion

Characterisation

The results from XRFshowedthat silica (SiO2)(68.49%) and alumina(Al2O3)(11.48%) are maincomponents of the zeolite. Other chemical components are K2O (3.79%), Na2O (1.47%), CaO(1.22%), Fe2O3 (1.09%) and loss on ignitionof 11.03%. ThePXRDresultsare shown inFig.1.Thezeolitewas mainlycomposed of clinoptilolite, quartz and sanidinewerethemain mineralsof the zeolite. From these results, it was clear that the natural zeolite can be classi&ied asclinoptilolite. The FTIR spectra of the raw and functionalised zeolite are shown in Fig. 2. Thepeaks at 695 and 964 cm-1which arepresent in both theraw and functionalised zeoliteare dueto the Si-O and Al-O-Si and Si-OH groups (Guayaet al. 2015).The successof functionalizationwas con(irmedby the peaks at 790and765cm-1 which aredueto Si-O-Si bonds of the zeoliteand APDEMS. The peak at 1644 cm-1 canbeattributedtothedeformation vibration of water. Thepeaks beyond 3300 cm-1 are due to the hydroxyl groups of the zeolite. The –CH2 and -CHstretchingvibrations of APDEMSare representedbythe peaks at 2925and2872cm-1. Thepeaksat 1182 and 1257 cm-1 on the functionalised zeolite can be attributed to Si-CH2 and Si-CH3,respectively. Thepresenceof the nitrogen groups is represented bythepeaks at 1575 cm-1 (-NHvibration) and 3366 cm-1 (primary amines). Moreover, the elemental analysis showedthat theCand H percentages were high on the functionalised zeolite and that of N was observed onAPDEMS-functionalised zeolite whichwas not present on the raw zeolite.These results showthat the functionalization of raw zeolite was successful. The surface area of the raw zeolitedecreased from 16.28 to 2.17 m2 g-1 due to functionalization with APDEMS and this alsocon$irmedthe success of the additionof the ligandonzeolite. This reductioncanbe attributedtothe smotheringof the pores byAPDEMSas the poresizeand porevolumealsodecreased from12.97to4.50nm and0.052to0.0045cm3 , respectively.

Effect of adsorbent mass

Theeffect of adsorbent mass on theadsorption or recoveryof Pt(IV) was studied and theresultsareshowninFig.3.Theadsorption ef(iciency increased with increasing adsorbent mass but theadsorption capacitydecreased. For subsequent experiments, 10 gL-1 adsorbent mass wasusedas the adsorption capacity and the adsorption ef0iciency were signi0icant.

3

nefe CKq1

= (4)



Kinetic models



(6)!!!= !

!!!!!+ !

!!t (7)



Characterisation





(1)

(2)

(3)

(4)

(5)

2







Data treatment

The adsorption capacity, qe (mg g-1) and the adsorption ef/iciency were calculated using

𝑞𝑞𝑞𝑞! =!!!!! !

!(1)


𝑥𝑥𝑥𝑥 100 (2)


Adsorptionisotherms


⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=

mLe

me

e

qKC

qqC 11 (3)

2







Data treatment


𝑞𝑞𝑞𝑞! =!!!!! !

!(1)


𝑥𝑥𝑥𝑥 100 (2)


Adsorptionisotherms


⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=

mLe

me

e

qKC

qqC 11 (3)

2







Data treatment


𝑞𝑞𝑞𝑞! =!!!!! !

!(1)


𝑥𝑥𝑥𝑥 100 (2)


Adsorptionisotherms


⎟⎟⎠

⎞⎜⎜⎝

⎛+⎟⎟

⎠

⎞⎜⎜⎝

⎛=

mLe

me

e

qKC

qqC 11 (3)

3

nefe CKq1

= (4)



Kinetic models



(6)!!!= !

!!!!!+ !

!!t (7)



Characterisation




3

nefe CKq1

= (4)



Kinetic models



(6)!!!= !

!!!!!+ !

!!t (7)



Characterisation




Topic 17.indb 1112 2018/09/03 07:27



where qe (mg g-1) is the quantity of the ad-sorbate adsorbed per gram of adsorbent at equilibrium, qm (mg g-1) is the maximum monolayer coverage capacity and KL is the Langmuir isotherm constant (L mol-1), Kf ((mg g-1)/(mol L-1)1/n) is the Freundlich iso-therm constant and n is the adsorption in-tensity, Kad (mol2 (kJ2)-1) is the Dubinin-Ra-dushkevich isotherm constant, Xm (mg g-1) is theoretical isotherm saturation capacity and is the Polanyi constant. Kinetic models� e adsorption process controlling mecha-nism between Pt(IV) and APDEMS-func-tionalised zeolite was determined with pseu-do � rst- and second order kinetic models represented by equations 6 and 7, respectively (Bakatula et al 2015). � e correlation coe� -cient (R2) and the closeness of the experimen-tal adsorption capacity, qexp to the calculated, qcal were used to determine the model that best decribes the kinetic data.

(6)

(7)

where qt (mg g-1) and qe (mg g-1) are the ad-sorption capacities at time t and equilibrium, respectively, k1 (min-1)and k2 (g mg-1 min-1) are the rate constants.

Results and discussionCharacterisation� e results from XRF showed that silica (SiO2) (68.49%) and alumina (Al2O3) (11.48%) are main components of the zeolite. Other

3

nefe CKq1

= (4)

𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑞𝑞𝑞𝑞! = 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑋𝑋𝑋𝑋! − 𝐾𝐾𝐾𝐾!"𝜀𝜀𝜀𝜀!(5)

where qe (mg g-1) is the quantity of the adsorbate adsorbed per gram of adsorbent atequilibrium, qm(mgg-1) is themaximummonolayercoveragecapacityandKL is theLangmuirisothermconstant(Lmol-1),Kf((mgg-1)/(mol L-1)1/n)istheFreundlichisothermconstantandnistheadsorptionintensity,Kad(mol2(kJ2)-1)is the Dubinin-Radushkevichisothermconstant,Xm(mgg-1)is theoretical isotherm saturation capacityandɛisthePolanyiconstant.

Kinetic models

The adsorption process controlling mechanism between Pt(IV) and APDEMS-functionalisedzeolite was determined with pseudo 2irst- and second order kinetic models represented byequations6 and7, respectively (Bakatula et al2015). The correlation coef,icient (R2)and thecloseness of the experimental adsorption capacity, qexp to the calculated, qcal were used todeterminethemodelthatbestdecribesthekineticdata.


(6)!!!= !

!!!!!+ !

!!t (7)

where qt (mg g-1) and qe (mg g-1) are the adsorption capacities at time t and equilibrium,respectively,k1(min-1)andk2(gmg-1min-1)aretherateconstants.

Resultsanddiscussion

Characterisation

TheresultsfromXRFshowedthatsilica(SiO2)(68.49%) andalumina(Al2O3)(11.48%)aremaincomponents of the zeolite. Other chemical components are K2O (3.79%), Na2O (1.47%), CaO(1.22%), Fe2O3 (1.09%)and losson ignitionof11.03%.ThePXRDresultsare shown inFig.1.Thezeolitewasmainlycomposedofclinoptilolite,quartzandsanidinewerethemainmineralsof the zeolite. From these results, it was clear that the natural zeolite can be classi&ied asclinoptilolite. TheFTIR spectra of the raw and functionalised zeolite are shown in Fig. 2. Thepeaks at 695 and 964 cm-1whicharepresentinboththerawandfunctionalisedzeoliteare dueto the Si-O and Al-O-Si and Si-OHgroups (Guayaet al.2015).The successof functionalizationwascon(irmedbythepeaksat790and765cm-1whichareduetoSi-O-Si bonds of the zeoliteand APDEMS. The peak at 1644 cm-1canbeattributedtothedeformation vibration of water. Thepeaks beyond 3300 cm-1 are due to the hydroxyl groups of the zeolite. The –CH2 and -CHstretchingvibrationsofAPDEMSarerepresentedbythepeaksat2925and2872cm-1.Thepeaksat 1182 and 1257 cm-1 on the functionalised zeolite can be attributed to Si-CH2 and Si-CH3,respectively.Thepresenceofthe nitrogen groupsisrepresentedbythepeaksat 1575 cm-1(-NHvibration) and 3366 cm-1(primary amines). Moreover,theelementalanalysisshowedthattheCand H percentages were high on the functionalised zeolite and that of N was observed onAPDEMS-functionalised zeolitewhichwas notpresent on the raw zeolite.These results showthat the functionalization of raw zeolite was successful. The surface area of the raw zeolitedecreased from 16.28 to 2.17 m2 g-1 due to functionalization with APDEMS and this alsocon$irmedthesuccessoftheadditionoftheligandonzeolite.ThisreductioncanbeattributedtothesmotheringoftheporesbyAPDEMSastheporesizeandporevolumealsodecreasedfrom12.97to4.50nmand0.052to0.0045cm3, respectively.


TheeffectofadsorbentmassontheadsorptionorrecoveryofPt(IV)wasstudiedandtheresultsareshowninFig.3.Theadsorptionef(iciencyincreasedwithincreasingadsorbentmassbuttheadsorptioncapacitydecreased.Forsubsequentexperiments,10gL-1adsorbent mass wasusedas the adsorption capacity and the adsorption ef0iciency were signi0icant.


3

nefe CKq1

= (4)

𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑞𝑞𝑞𝑞! = 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑋𝑋𝑋𝑋! − 𝐾𝐾𝐾𝐾 (5)


Kinetic models



(6)!!!= !

!!!!!+ !

!!t (7)



Characterisation




3

nefe CKq1

= (4)



Kinetic models

The adsorption process controlling mechanism between Pt(IV) and APDEMS-functionalisedzeolite was determined with pseudo 2irst- and second order kinetic models represented byequations 6 and7, respectively (Bakatula et al 2015). The correlation coef,icient (R2)and thecloseness of the experimental adsorption capacity, qexp to the calculated, qcal were used to

ekinetic data.


(6)!!!= !

!!!!!+ !

!!t (7)

t e the adsorption capacities at time t and equilibrium,respectively,k1 (min-1)and k2 (gmg-1min-1)are the rate constants.


Characterisation




chemical components are K2O (3.79%), Na2O (1.47%), CaO (1.22%), Fe2O3 (1.09%) and loss on ignition of 11.03%. � e PXRD results are shown in Fig. 1. � e zeolite was mainly com-posed of clinoptilolite, quartz and sanidine were the main minerals of the zeolite. From these results, it was clear that the natural zeolite can be classi� ed as clinoptilolite. � e FTIR spectra of the raw and functionalised zeolite are shown in Fig. 2. � e peaks at 695 and 964 cm-1 which are present in both the raw and functionalised zeolite are due to the Si-O and Al-O-Si and Si-OH groups (Guaya et al. 2015). � e success of functionalization was con� rmed by the peaks at 790 and 765 cm-1 which are due to Si-O-Si bonds of the ze-olite and APDEMS. � e peak at 1644 cm-1 can be attributed to the deformation vibration of water. � e peaks beyond 3300 cm-1 are due to the hydroxyl groups of the zeolite. � e –CH2 and -CH stretching vibrations of APDEMS are represented by the peaks at 2925 and 2872 cm-1. � e peaks at 1182 and 1257 cm-1 on the functionalised zeolite can be attributed to Si-CH2 and Si-CH3, respectively. � e pres-ence of the nitrogen groups is represented by the peaks at 1575 cm-1 (-NH vibration) and 3366 cm-1 (primary amines). Moreover, the elemental analysis showed that the C and H percentages were high on the functionalised zeolite and that of N was observed on AP-DEMS-functionalised zeolite which was not present on the raw zeolite. � ese results show that the functionalization of raw zeolite was successful. � e surface area of the raw zeo-lite decreased from 16.28 to 2.17 m2 g-1 due to functionalization with APDEMS and this also con� rmed the success of the addition of

Fig. 1 PXRD pattern of zeolite Fig. 2 FTIR spectra of raw zeolite (a) ctionalised zeolite (b)

Topic 17.indb 1113 2018/09/03 07:27



the ligand on zeolite. � is reduction can be attributed to the smothering of the pores by APDEMS as the pore size and pore volume also decreased from 12.97 to 4.50 nm and 0.052 to 0.0045 cm3 , respectively.

E� ect of adsorbent mass� e e� ect of adsorbent mass on the adsorp-tion or recovery of Pt(IV) was studied and the results are shown in Fig. 3. � e adsorption ef-� ciency increased with increasing adsorbent mass but the adsorption capacity decreased. For subsequent experiments, 10 g L-1 adsor-bent mass was used as the adsorption capac-ity and the adsorption e� ciency were signi� -cant.

E� ect of pHStudies have shown that pH can a� ect the be-haviour and hence the adsorption e� ciency of adsorbents (Bakatula et al. 2015; Ren et al. 2016). � e adsorption capacity decreased with increasing solution pH and this could

be attributed to the speciation of Pt(IV) and the surface charge (Fig. 4). At acidic pH, the surface groups were protonated and thus, –NH3+ which had high a� nity for the anionic Pt(IV) species (PtCl6

2-) resulted. � e anionic chloride species were more prevalent at acid-ic pH due to the addition of HCl. During the processing of PGMs, concentrated acids are used for extraction and thus, the discharged water is highly acidic (Nguyen et al. 2016). � us, the adsorbent will work well and signif-icantly recover Pt(IV) from mining wastewa-ters which are highly acidic. PZC was found to be 5.3. � e adsorption capacity decreased signi� cantly beyond pH 2 and this can be at-tributed to the low amount of chloride spe-cies favourable for adsorption as well as elec-trostatic repulsion between the Pt species and the surface groups.

E� ect of concentration� e adsorption of Pt(IV) at pH 2 was ob-served to increase with increasing Pt(IV)

Fig. 3 E� ect of mass on the adsorption of Pt(IV) by APDEMS-functionalised zeolite

0 100 200 300 400 500

0,05

0,10

0,15

0,20

0,25

0,30

Adsorption capacity Adsorption efficiency

Mass (mg)

Adso

rptio

n ca

pacit

y (m

g g-1

)

45

50

55

60

65

70

75

80

Ads

orpt

ion

effic

ienc

y (%

)

1 2 3 4 5 6 7 8 9 10

0,15

0,16

0,17

0,18

0,19

0,20

0,21

Adso

rptio

n ca

pacit

y (m

g g-1

)

pH

Adsorption capacity

0 1 2 3 4 50,02

0,04

0,06

0,08

0,10

0,12

0,14

0,16

0,18

0,20

0,22

Adso

rptio

n ca

pacit

y (m

g g-1

)

Concentration (mg L-1)

Adsorption capacity

Fig. 4 E� ect of pH on Pt(IV) adsorption Fig. 5 E� ect of concentration on Pt(IV) adsorption

Topic 17.indb 1114 2018/09/03 07:27



concentration from 0.5 to 2 mg L-1 (Fig. 5). � e increase in adsorption capacity can be attributed to the availability of binding sites (-NH3+). � e adsorption e� ciencies were 75.6, 79.7, 78, 3, 57.7 and 40.4% at 0.5, 1, 2, 3 and 5 mg L-1, respectively. � e ability of the adsorbent to signi� cantly adsorb very low concentrations of Pt(IV) makes it a suitable adsorbent for PGMs wastewater which con-tains very low concentrations of PGMs (Hunt and Clark 2013, Hunt et al 2015). � is also means that there is a strong interaction be-tween Pt(IV) and the adsorbent. Beyond 2 mg L-1, no increase in adsorption was observed. � e adsorption isotherm results (Table 1) indicated that the interaction of Pt(IV) onto APDEMS-functionalised zeolite is through monolayer adsorption since the experimental data was described better by Langmuir iso-therm (R2 >0.98). � e free energy change as determined from D-R isotherm was found to 12.42 kJ mol-1 indicating that the adsorption process is through ion exchange (Bakatula et al. 2015).

E� ect of contact time� e e� ect of contact time on the adsorption of Pt(IV) by APDEMS-functionalised zeolite was investigated and the results are shown in Fig. 6. A rapid increase in adsorption was observed between 0 and 30 min a� er which, a slow uptake was observed until 180 min. Equilibrium was reached at 180 min where the adsorption was controlled by di� usion. � us, 180 min was chosen as the su� cient time for the recovery of Pt(IV) by the adsor-bent. � e kinetic data was best described by pseudo second-order indicating that chemi-sorption was the rate controlling mechanism (R2 >0.99) (Table 2). � is was also supported by the calculated adsorption capacity (qcal) (0.102 mg g-1) which was very similar to the experimental adsorption capacity (0.103 mg g-1). � e experimental adsorption capac-ity determined with pseudo � rst-order was found to be 0.0061 mg g-1 which is very dif-ferent from the calculated capacity, further explaining that there is no good correlation with pseudo � rst-order.

Langmuir Freundlich Dubinin-Radushkevich (D-R)

qm (mg g-1) KL (L mol-1) R2 Kf ((mg g-1)/(mol L-1)1/n) n R2 Xm (mg g-1) Es (kJ mol-1) R2

0.117 3.5x105 0.98 32.29 1.99 0.91 1.37 12.42 0.92

Table 1 Adsorption isotherm constants

0 100 200 300 400 500 600

0,00

0,05

0,10

0,15

0,20

Adso

rptio

n ca

paci

ty (m

g g-1

)

Time (min)

Adsorption capacity

Ru Ru Pt Ir Os Pd F e Co Ni Ca Mg K Z n Hf Au0,00

0,05

0,10

0,15

0,20

0,25

0,30

Adso

rptio

n ca

pacit

y (m

g g-1

)

Metals

Adsorption capacity

Fig. 6 E� ect of time on Pt(IV) adsorption Fig. 7 E� ect of competing ions on Pt(IV) adsorption

Table 2 Kinetic parameters for Pt(IV) adsorption by APDEMS-functionalised zeolite

Pseudo � rst-order Pseudo second-order

K1 (min-1) qexp (mg g-1) R2 K2 (g mg-1 min-1) qexp (mg g-1) R2

0.0088 0.0061 0.71 6.71 0.103 0.99

Topic 17.indb 1115 2018/09/03 07:27



E� ect of competing ionsA mixture of 2 mg L-1 Pt(IV) and other ions at pH 2 was exposed to APDEMS-function-alised zeolite to determine the e� ciency of the adsorbent on the recovery of Pt(IV) in the presence of other ions. � e adsorption capac-ity of Pt(IV) was not a� ected by the presence of other ions, some of which are PGMs (Fig. 7). Moreover, these PGMs and other cations except Ca and Mg were adsorbed by the ad-sorbent making APDEMS-functionalised zeolite a dual e� cacy adsorbent. � e ability to signi� cantly adsorb Pt(IV) in the presence of other ions makes the APDEMS-function-alised zeolite a suitable adsorbent to be used in mining wastewater.

Conclusion In this study, APDEMS-functionalised zeolite was synthesised and used to recover Pt(IV) from aqueous solutions. � e characterisa-tion results showed that the functionaliza-tion was successful. � e experimental studies were carried out under the optimised condi-tions (100 mg adsorbent, pH 2, 2 mg L-1 and 180 min) and indicated that 0.207 mg g-1 of Pt(IV) was removed from aqueous solutions by APDEMS-functionalised zeolite. � e ad-sorbent was signi� cantly adsorbed Pt(IV) at highly acidic pH (pH 2) which makes the adsorbent very suitable for PGM min-ing wastewater which is highly acidic. � e kinetic (pseudo second-order) and isotherm (Langmuir) models as well as the energy from D-R isotherm indicated that there was strong chermisorptive Pt-adsorbent interaction. � e presence of competing ions did not a� ect the adsorption of Pt(IV). � e adsorbent can be applied in retaining ponds into which pro-cessing solutions are discharged and fed to the processing units together with fresh ore a� er recovery.

Acknowledgements� e authors would like to thank the National Research Foundation (NRF), the University of the Witwatersrand and the Water Research Commission (WRC project no. K5/2589) for � nancial support.

References Alonso E, Field FR, Roth R, Kirchain RE (2009)

Strategies to address risks of platinum scarcity for supply chain downstream � rms. In Sustain-able Systems and Technology, 2009. ISSST’09. IEEE International Symposium on (pp. 1-6). IEEE.

Bakatula EN, Mosai AK, Tutu H (2015) Removal of Uranium from Aqueous Solutions using Ammonium-modi� ed Zeolite, S Afr J Chem, 68:165–171

Chassary P, Vincent T, Marcano JS, Macaskie LE, Guibal E (2005) Palladium and platinum recov-ery from bicomponent mixtures using chitosan derivatives, Hydrometallurgy 76(1-2):131-147

Chen X (2015) Modeling of experimental adsorp-tion isotherm data, Information 6(1):14-22.

Guaya D, Valderrama C, Farran A, Armijos C, Cortina JL (2015) Simultaneous phosphate and ammonium removal from aqueous solution by a hydrated aluminum oxide modi� ed natural zeo-lite, Chem Eng J 271:204-213

Hunt AJ, Matharu AS, King AH, Clark JH (2015) � e importance of elemental sustainabil-ity and critical element recovery, Green Chem, 17(4):1949-1950

Lottermoser B (2003) Mine Water. In Mine wastes, Springer, Berlin, Heidelberg, pp. 83-141

Malamis S, Katsou E (2013) A review on zinc and nickel adsorption on natural and modi� ed zeo-lite, bentonite and vermiculite: Examination of process parameters, kinetics and isotherms, J Hazard Mater 252:428-461

Mushtaq M, Bhatti HN, Iqbal M, Noreen S (2016) Eriobotrya japonica seed biocomposite e� cien-cy for copper adsorption: isotherms, kinetics, thermodynamic and desorption studies, J Envi-ron Manage 176:21-33

Nguyen TH, Sonu CH, Lee MS (2016) Separation of Pt (IV), Pd (II), Rh (III) and Ir (IV) from con-centrated hydrochloric acid solutions by solvent extraction. Hydrometallurgy, 164:71-77

Ren H, Jiang J, Wu D, Gao Z, Sun Y, Luo C (2016) Selective adsorption of Pb (II) and Cr (VI) by surfactant-modi� ed and unmodi� ed natural zeolites: a comparative study on kinetics, equi-librium, and mechanism, Water Air Soil Pollut 227(4):101

Xiao Z, Laplante AR (2004) Characterizing and recovering the platinum group minerals––a re-view. Miner Eng, 17(9-10):961-979

Topic 17.indb 1116 2018/09/03 07:27

Documents

The Recovery Of Pt(IV) From Aqueous Solutions By APDEMS ... · varioELcube V4.0.13 (Elementar, Germany) was used for elemental analysis. Textural properties (surface area, pore size