Recovery of Platinum from Chloride Leaching Solutions of Spent Reforming Catalysts by Ion Exchange

Recovery of Platinum from Chloride Leaching Solutions

of Spent Reforming Catalysts by Ion Exchange

Pan-Pan Sun

, Tae-Young Kim, Byoung-Jun Min,

Hyoung-Il Song and Sung-Yong Cho

Department of Environment and Energy Engineering, Chonnam National University, Gwangju 500-757, Republic of Korea

Platinum and base metals (Al, Fe, Si) are present in the chloride leaching solutions of spent reforming catalysts. In order to develop a process to recover platinum from such leaching solutions, batch and continuous ion exchange experiments were performed using a strong anionic resin (DiaionSA10AP). The adsorption equilibrium data of Pt are described by the Langmuir adsorption isotherm. Batch experiments indicate that it is possible to separate Pt and Fe from other metal ions in the present leaching solution. Fe was selectively removed from the loaded resin using diluted HCl, while Pt was eluted using thiourea after Fe removal. The results of column experiments verified the feasibility of the separation and recovery of platinum from leaching solutions using the present process. The elution efficiency exceeded 99%for both Pt and Fe. The recovery percentage of Pt from the leaching solution using this method was 99.3%. [doi:10.2320/matertrans.M2015027]

(Received January 16, 2015; Accepted February 19, 2015; Published April 3, 2015)

Keywords: platinum, iron, hydrochloric acid, ion exchange, recovery

1. Introduction

Spent catalysts from the automotive, chemical and petroleum industries are valuable secondary resources for

platinum group metals (PGMs).1­5)Recovery of PGMs from

these resources is an attractive way to manufacture materials due to economical, as well as environmental

considera-tions.3,6) _{In hydrometallurgical treatment processes, the}

valuable metals are first dissolved by a leaching process

using various lixiviants.2,4,7­9) _{In order to produce the}

metallic compounds of platinum, Pt in the leaching solution

mustfirst be separated from the base metal ions.

Ion exchange and solvent extraction are generally

em-ployed to separate metal ions from the leaching solution.10­20)

Solvent extraction has been used for the recovery and separation of PGMs from aqueous solutions, however a very

high concentration of PGMs (over hundreds of mg/dm3) in

the solution is required for this process to work efficiently.21)

Ion exchange was reported to be a powerful method to selectively recover low concentrations of PGM ions from solutions containing high amounts of base metals, particularly

aluminum.22) _{Moreover, the operation of ion exchange is}

simple and the production of high purity Pt is possible through this method. Therefore, when the concentration of Pt in the hydrochloric acid solution is not high, ion exchange resins can be selected to separate Pt from other metal ions.

In our previous study, the spent reforming catalyst (Pt/

Al2O3) was roasted at 800°C to eliminate volatile substances.

Platinum and part of the base metals (Al, Fe, Si) in the roasted spent catalyst were dissolved using a mixture of

20 vol%HCl and 3 times the stoichiometric ratio of H2O2.23)

According to the distribution diagram of Pt in aqueous

solutions, the hexachloroplatinate ion (PtCl6¹)

predom-inates in strong hydrochloric acid solutions (HCl²3

kmol/m3_).22,24) _{In the case of iron, a variety of complexes,}

such as FeCl2+_{, FeCl}

2+, FeCl3aq, FeCl4¹, exist in the chloride

solution.24,25)_{The mole fraction of each complex depends on}

the concentration of chloride ions. Employing anionic resins

would lead to loading of anions (PtCl6¹ and FeCl4¹) and

result in their separation from the cations. Furthermore, during an extraction process, the removal of Fe from the loaded organic (Aliquat336) is much easier than the removal

of Pt.23) _{Thus it is of value to investigate the possibility of}

using anionic resins to separate Pt from the leaching solution containing iron and other cations.

Diaion SA10AP, a strongly basic anionic resin, is widely

used for water purification. Little information has been found

regarding not only the separation efficiency of Pt from real

leaching solutions containing Al, Fe, and Si with Diaion SA10AP resin, but also the effective elution of Pt from the loaded resin.

In the present work, Diaion SA10AP resin was employed to separate platinum from the chloride leaching solution of spent reforming catalyst. The adsorption equilibrium data of

Pt on Diaion SA10AP resin was first investigated using

the Langmuir, Freundlich and Sips isotherm models. The conditions for separation of Pt and Fe from the leaching solution were obtained in batch experiments. Column experiments were employed to construct the breakthrough curves for Pt and Fe loading onto Diaion SA10AP resin. The recovery percentage and purity of Pt from the chloride leaching solution after column experiments is reported.

2. Experimental

2.1 Chemicals and resin

The chloride leaching solution was obtained by dissolving

a spent reforming monometallic catalyst (Pt/Al2O3) under

optimum leaching conditions, which were reported in our

previous study.23) The compositions of the spent catalyst

and the leaching solution are listed in Tables 1 and 2, respectively. The concentration of HCl in the leaching

solution was 5.9 kmol/m3_{. The elution solutions were}

prepared by dissolving NaCl, HNO3, Na2CO3, thiourea,

Na2S2O3·5H2O and NaOH in double distilled water. All the

chemicals used in this study were analytical pure grade. +_{Corresponding author, E-mail: spp1227@jnu.ac.kr}

Diaion SA10AP resin (Mitsubishi chemical) was em-ployed to adsorb platinum from the leaching solution. It is a commercial and strong basic anion exchanger with quater-nary ammonium functional groups. Its physicochemical characteristics are presented in Table 3. In the present study, the resin was used as received, without any treatment.

2.2 Ion exchange procedure

Batch experiments were carried out in a shaking incubator

(VS-8480SF, Vision Scientific Co., LTD.) using a 100 cm3

screwed cap bottle at 25°C. Twenty mL of leaching solution was bottle rolled for 24 hours, together with resin in the

concentration range of 0.5³100 g/dm3. After the solution

was separated byfiltration, the concentration of metal ions in

the solution was measured by using ICPS-7500 (Shimadzu). The concentration of metals loaded onto the resin was obtained by mass balance. The elution experiments were carried out by mixing the loaded resin with eluent in the same manner.

In continuous experiments, a glass column (250©10 mm)

was used. One or five grams of resin were poured into the

column to obtain a packed column. Throughout the entire operation process, the temperature of the column was

controlled at 25°C using a water circulator bath (Scientific

Co., Vs-1902WF). The flow direction in the column is

downward and the flow rate of the feed solution/elution

solution was controlled at 1.5 mL/min using a pump (QG20

lab pump, FMI). The effluent was fractionated into portions

of desired volume, and the concentration of metal ions in each fraction was measured using an ICPS-7500 (Shimadzu).

3. Results and Discussions

3.1 Batch experiments

3.1.1 Adsorption of platinum from platinum chloride solution

In order to investigate the adsorption behavior of Pt on the Diaion SA10AP resin, loading experiments were carried out

using a synthetic PtCl4solution. The concentrations of Pt and

HCl in the feed solution were 124 mg/dm3 _{and 5 kmol/m}3_,

respectively. The concentration of resin was varied from 0.5

to 150 g/dm3_{. Langmuir, Freundlich and Sips adsorption}

isotherms werefitted to describe the adsorption isotherm of

Pt on DiaionSA10AP resin. These isotherms are frequently used for analyzing experimental adsorption equilibrium data

of metal ions during ion exchange.26)The equilibrium models

often provide some insight into the adsorption mechanism,

the surface properties and the affinity of the adsorbent. These

isotherm equations are as follows,

Langmuir isotherm; q¼qmbce=ð1þbceÞ ð1Þ

where ce is the supernatant concentration at equilibrium

(mg/dm3_{), b is the Langmuir affinity constant and q}

m is

the maximum adsorption capacity of the material (mg/g),

assuming a monolayer of adsorbate was taken up by the adsorbent.

Freundlich isotherm; q¼kce1=n ð2Þ

where kis the Freundlich constant related to the adsorption

capacity andnis the Freundlich exponent.

Sips isotherm; q¼ ðqmbCe1=nÞ=ð1þbCe1=nÞ ð3Þ

where b is the Sips constant related to the affinity constant

and qmis the Sips maximum adsorption capacity (mg/g).

The linear least squares method and a pattern search algorithm were used to obtain the parameters for each adsorption isotherm. The value of the mean percentage error

was used as a test criterion for thefit of the correlations. The

mean percentage deviation between the experimental and predicted values was obtained using eq. (4),

error ð%Þ ¼ ð100=NÞXN

k¼1

½jqexp,kqcal;kj=qexp,k ð4Þ

whereqcal,kis each value of qpredicted by thefitted model,

qexp,k represents each value of q measured experimentally,

andNis the number of experiments performed. The obtained

parameters and average percentage difference between the measured and calculated values for the adsorption of Pt on

Diaion resin in 5 mol/dm3of HCl are given in Table 4. The

fittings of adsorption isotherms to the adsorption equilibrium

data of Pt on Diaion SA10AP resin in 5 mol/dm3_{of HCl are}

presented in Fig. 1. The results show that Langmuir and Sips

model fit the adsorption data well with little difference

(Fig. 1). Since Sips model has one more adjustable parame-ter, the error value is smaller than that of Langmuir model.

Thus, Langmuir model is sufficient to describe the adsorption

of Pt on Diaion resin under the present conditions. The

corresponding loading capacity of Pt was 128.1 mg/g. This

value suggests that Diaion SA10AP resin is a good adsorbent for Pt uptake from aqueous solutions.

3.1.2 Adsorption of platinum and iron from the leaching solution

In order to determine the effect of resin concentration on the loading of metals from leaching solutions, batch experiments were carried out by varying the concentration

of resin from 0.5 to 100 g/dm3_{. The chloride leaching}

solution containing Pt, Fe, Al and Si was used as the feed solution.

Table 2 Chemical composition of the leaching solution (Unit: mg/dm3_).

Pt Fe Al Si

124 23.7 4870 5.4

Table 3 Physicochemical characteristics of Diaion SA10AP resin used in this study.

Resin Ionic forms Particle size

(µm)

Density (g/dm3₎

Moisture (%)

Diaion

SA10AP Chloride 300­1180 685 47

Table 1 Chemical analysis of the spent catalyst by XRF and ICP (Unit: mass%).

Pt Fe Al Si

0.25 0.07 45.78 0.08

The results in Fig. 2 show that Diaion SA10AP resin has good selectivity of both Pt and Fe, compared with the other ions (Al, Si) in the leaching solution. The loading percentage of both Pt and Fe increased when the concentration of resin was increased. When the concentration of resin was higher

than 25 g/dm3_{, most of the Pt and Fe were loaded onto the}

resin simultaneously, leaving Al and Si in the effluent.

Therefore, it is possible to separate Pt and Fe from other metals ions in the leaching solution by varying the concentration of resin.

The adsorption of Pt and Fe can be explained by the nature of the predominant species of each element (as explained in the introduction). Since the predominant species of platinum

in the leaching solution is PtCl6¹, and the adsorbable iron

species is FeCl4¹, the general reaction for the adsorption of Pt

and Fe can be represented as follows,

2RClþPtCl2₆¼R2PtCl6þ2Cl ð5Þ

RClþFeCl₄ ¼RFeCl4þCl ð6Þ

where RCl represents the chloride form of Diaion SA10AP resin.

3.1.3 Removal of Fe

In ion exchange processes, elution of metals from the loaded resin is an important step to obtain the metal solution. In the reported literature, Fe was easily removed from various loaded organics, such as TBP and Aliquat336, using distilled

water and diluted acid, respectively.16,25) _{In this study, in}

order to remove iron from the loaded resin, 0.0001 to

0.01 mol/dm3 of HCl, and 1 mol/dm3 of HNO3were tested

as eluents. The loaded resin was obtained from the batch

experiments by loading the leaching solution with 40 g/dm3

of resin. The results (Table 5) suggest that when an equal

volume of 0.0001 mol/dm3of HCl or 1 mol/dm3of HNO3is

employed as eluent, iron is completely eluted from the loaded resin, while the elution percentage of iron decreases slightly

to 83.2% with an increase in the concentration of HCl from

0.0001 to 0.01 mol/dm3_{. Under these elution conditions, the}

elution percentage of Pt was zero. Therefore, it can be concluded that Fe can be selectively removed from the loaded

resin using 0.0001 mol/dm3_{of HCl.}

3.1.4 Elution of Pt

It is reported that acidic solutions can elute platinum from some loaded resins, depending on the experimental

con-ditions.27)Na2CO3, NaCl, NaOH, thiourea and Na2S2O3have

each shown the ability to elute platinum from several basic

anion exchange resins.15,28)In this study, elution of platinum

from the loaded Diaion SA10AP resin, after removal of iron, was tested using all of the above agents. Loading experiments were carried out by loading the leaching solution with

40 g/dm3 _{of resin. Co-loaded iron was removed using}

0.0001 mol/dm3 _{of HCl before the elution test of Pt. The}

results are summarized in Table 6. Under the present

experimental conditions, NaCl, NaOH, and Na2S2O3 could

not elute Pt from the loaded Diaion SA10AP resin. The

Supernatant concentration, Ce/mg.dm-3

0 20 40 60 80 100 120 140 160 180

Adsor

ption capacity

,

qm

/ mg

.g

-1

0 20 40 60 80 100 120 140

Sips eq. Freundlich eq. Langmuir eq.

Fig. 1 Adsorption equilibrium isotherm of platinum on Diaion SA10AP resin in 5 mol/dm3_{of HCl.}

0 10 20 30 40 50 60 70 80 90 100 110 0

20 40 60 80 100

Adsorption percentage(%)

Resin concentration, C/ g⋅dm-3 Pt Al Fe Si

Fig. 2 Effect of Diaion SA10AP resin concentration on the adsorption of metals.

Table 5 Removal of Fe from loaded Diaion SA10AP resin.

Elution reagent Fe loading (%) Fe elution (%)

0.01 mol/dm3_{HCl 98.4 83.2}

0.001 mol/dm3_{HCl 98.5 89.8}

0.0001 mol/dm3_{HCl 98.9 99.9}

1 mol/dm3_HNO

3 98.5 99.9

Table 4 Adsorption equilibrium constants of Pt on DiaionSA10AP resin (25°C).

Isotherm type Parameters Value

Langmuir

qm(mg/g) 128.1

b 0.030

Error (%) 7.356

Freundlich

k 9.992

n 2.020

Error (%) 13.42

Sips

qm(mg/g) 128.1

b 0.026

n 0.959

elution percentage of Pt was lower than 5%when HNO3and

Na2CO3were used as eluents. However, when 0.1 mol/dm3

of thiourea was used as eluent, the Pt that was loaded onto the Diaion SA10AP resin was completely eluted. This fact could be explained by the hard and soft acid and base theory

(HSAB theory).29)_{Thiourea is a soft ligand, which was also}

reported to be effective in stripping Pt/Pd from a loaded

organic phase, such as DEHTPA and Alamine336.28,30) The

elution reaction might be explained by the following reaction:

R2PtCl6þ2Tu¼PtðTuÞ₂4þþ4Clþ2RCl ð7Þ

where RCl and Tu represent the chloride form of Diaion-SA10AP resin and thiourea, respectively.

3.2 Column experiments

3.2.1 Breakthrough curves of Pt and Fe

In order to obtain the breakthrough curves for Pt and Fe, column experiments were carried out by passing the leaching solution through a column containing 1 g of resin. The

effluent was fractionated into portions of 4.2 cm3 _{(2 bed}

volume) for thefirst 42 cm3_{and 50.4 cm}3_{(24 bed volume) for}

the remainder. The obtained results are represented as a plot

of concentration fraction (C/C0, a ratio of the concentration

of metal ions in the effluent to that in the feed solution) to bed

volume in Fig. 3. The change in concentration of Al and Si

were zero before and after effluent was passed through the

column, indicating that the loading of Al and Si onto the resin is negligible. The breakthrough curves of Pt and Fe were obtained. These breakthrough curves can be utilized to calculate the height of a practical exchange plate (HPEP) for design intentions and validation of mathematical models for

the behavior of ion exchange beds.31)

3.2.2 Separation of Pt and Fe from leaching solution

Since the results of the batch experiments indicate that it is possible to separate Pt from Fe, Al and Si in leaching solutions using a process of adsorption and selective elution, a series of continuous column experiments was conducted to

obtain the recovery efficiency of Pt from the leaching

solution. According to Fig. 2, when the concentration of resin

is higher than 25 g/dm3, most of the Pt and Fe can be loaded

onto the resin. In column experiments, 200 cm3 of leaching

solution was poured into a column containing 5 g of resin.

Each bed volume of effluent was collected and the

concentration of metal ions was measured. The results (not

given as afigure) suggest that 99.7%of Pt and 99.6%of Fe

were loaded onto the resin simultaneously, while the differences in both Si and Al concentrations before and after passing through the column were negligible. Thus, Pt and

Fe can be separated from other metal ions in the chloride leaching solution through ion exchange using Diaion SA10AP resin under the present conditions. The loaded resin was used for the following stripping experiments.

3.2.3 Removal of Fe from the loaded resin

Based on the batch experiment results, 0.0001 mol/dm3_of

HCl was used to remove Fe from the loaded resin. Then

0.0001 mol/dm3 _{of HCl was poured into the column}

containing the loaded resin. The concentration of Fe in the

effluent was measured. The results are represented as a

function of cumulative elution percentage of metal to bed volume in Fig. 4. This indicates that the elution of Fe with

0.0001 mol/dm3_{of HCl was easy and fast; more than 80%of}

loaded Fe was removed after 1 bed volume was collected.

The amount of Fe in the effluent increased with an increase in

the bed volume. Fe was completely eluted from the loaded resin after13 bed volumes. The concentration of Pt was

negligible in each bed volume effluent.

3.2.4 Elution of Pt from the loaded resin

In order to recover Pt from the loaded resin, 0.1 mol/dm3

of thiourea was used for the elution of Pt from the loaded resin after removal of Fe. As shown in Fig. 5, the cumulative elution percentage of Pt increased with an increase in the

volume of thiourea passed through the column. 90% of Pt

Table 6 Elution of Pt from loaded Diaion SA10AP resin after removal of Fe.

Elution reagent Pt loading (%) Pt elution (%)

1 mol/dm3_{NaCl 99.6 nil}

1 mol/dm3_HNO

3 99.8 4.7

2 mol/dm3_Na

2CO3 99.2 5

0.1 mol/dm3_{Thiourea 99.3 >99.9%}

1 mol/dm3_Na

2S2O3 99.2 1

1 mol/dm3_{NaOH 99.6 nil}

0 200 400 600 800 1000

0.0 0.2 0.4 0.6 0.8 1.0

C/C

0 Pt

Fe Si Al

Amount of bed volume, n

Fig. 3 Breakthrough curves for Pt and Fe from the leaching solution on the Diaion SA10AP resin.

0 2 4 6 8 10 12 14 16

0 20 40 60 80 100

Cumulati

v

e elution of Fe (%)

Amount of bed volume, n

was eluted after 10 bed volumes of effluent were collected.

When 20 bed volumes of effluent were collected, 99.6% of

Pt was eluted from the loaded resin. The concentration of Pt

in the effluent that was further collected was lower than

0.5 mg/dm3_.

The entire process of separating and recovering Pt from the chloride leaching solution is summarized in Fig. 6. By using

this method, 99.3% of the Pt with a purity of 99.6% was

separated and recovered from the leaching solution.

4. Conclusions

Platinum was recovered from the chloride leaching solution of spent reforming catalysts using Diaion SA10AP resin. The adsorption of Pt on DiaionSA10AP resin from hydrochloric acid solution can be described using Langmuir adsorption isotherm. Under the present experimental con-ditions, platinum and iron were adsorbed onto the resin simultaneously, leaving Al and Si in the solution. The relatively low selectivity of platinum was overcome by

selective elution using different eluents. Iron was easily

removed using 0.0001 mol/dm3_{of HCl, while platinum was}

eluted using 0.1 mol/dm3 _{of thiourea. This method can be}

utilized to recover/separate platinum from chloride leaching

solutions containing Al, Si, and Fe.

Acknowledgment

This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No: 2014R1A1A2007063). The authors would like to thank them

for thefinancial support.

REFERENCES

1) P. Liu, G. Liu, D. Chen, S. Cheng and N. Tang:Trans. Nonferrous Met. Soc. China19(2009) 1509­1513.

2) M. Baghalha, Gh. H. Khosravian and H. R. Mortaheb: Hydro-metallurgy95(2009) 247­253.

3) K. Shams, M. R. Beiggy and A. G. Shirazi:Appl. Catal. A258(2004) 227­234.

4) S. Harjanto, Y. Cao, A. Shibayama, I. Naitoh, T. Nanami, K. Kasahara, Y. Okumura, K. J. Liu and T. Fujita:Mater. Trans.47(2006) 129­135. 5) D. R. Tyson and R. G. Bautista:Sep. Sci. Technol.22(1987) 1149­

1167.

6) W. Mulak, B. Miazga and A. Szymczycha:Int. J. Miner. Process.77

(2005) 231­235.

7) R. H. Matjie, M. S. Scurrell and J. Bunt:Miner. Eng.18(2005) 801­ 810.

8) D. Jafarifar, M. R. Daryanavard and S. Sheibani:Hydrometallurgy78

(2005) 166­171.

9) M. A. Barakat and M. H. H. Mahmoud:Hydrometallurgy72(2004) 179­184.

10) F. L. Bernardis, R. A. Grant and D. C. Sherrington: React. Funct. Polym.65(2005) 205­217.

11) C. S. Kedari, M. T. Coll, A. Fortuny, E. Goralska and A. Sastre:Sep. Sci. Technol.40(2005) 1927­1946.

12) J. R. Cui and L. F. Zhang:J. Hazard. Mater.158(2008) 228­256. 13) J. R. Kumar, H. I. Lee, J. Y. Lee, J. S. Kim and J. S. Sohn:Sep. Purif.

Technol.63(2008) 184­190.

14) J. Y. Lee, J. R. Kumar, J. S. Kim, D. J. Kim and H. S. Yoon:J. Ind. End. Chem.15(2009) 359­364.

15) R. S. Marinho, J. C. Afonso and J. W. S. D. Cunha:J. Hazard. Mater.

179(2010) 488­494.

16) B. R. Reddy, B. Raju, J. Y. Lee and H. K. Park:J. Hazard. Mater.180

(2010) 253­258.

17) B. Swain, J. K. Jeong, S. K. Kim and J. C. Lee:Hydrometallurgy104

(2010) 1­7.

18) A. Jaree and N. Khunphakdee:J. Ind. Eng. Chem.17(2011) 243­247. 19) B. Raju, J. R. Kumar, J. Y. Lee, H. S. Kwonc, M. L. Kantam and B. R.

Reddy:J. Hazard. Mater.227­228(2012) 142­147.

20) B. Gupta and I. Singh:Hydrometallurgy134­135(2013) 11­18. 21) A. Wolowicz and Z. Hubicki:Hydrometallurgy98(2009) 206­212. 22) Z. Hubicki and G. Wójcik:J. Hazard. Mater.136(2006) 770­775. 23) P. P. Sun and M. S. Lee:Korean J. Met. Mater.51(2013) 873­881. 24) M. S. Lee:Metall. Mater. Trans. B37B(2006) 173­179.

25) B. R. Reddy and P. V. R. Bhaskara Sarma:Hydrometallurgy43(1996) 299­306.

26) Z. Zainol and M. Nicol:Hydrometallurgy99(2009) 175­180. 27) A. Parodi, T. Vincent, M. Pilsniak, A. W. Trochimczuk and E. Guibal:

Hydrometallurgy92(2008) 1­10.

28) P. P. Sun, J. Y. Lee and M. S. Lee:Hydrometallurgy113­114(2012) 200­204.

29) G. R. Choppin:Solvent Extraction Principles and Practice, (Marcel Dekker, Inc., New York, 2004) pp. 97­98.

30) M. Rovira, J. L. Cortina and A. M. Satre:Solvent Extr. Ion Exch.17

(1999) 333­349.

31) K. Shams and F. Goodarzi:J. Hazard. Mater.131(2006) 229­237.

0 5 10 15 20 25

0 20 40 60 80 100

Amount of bed volume, n

Cumulati

v

e elution of Pt (%)

Fig. 5 Elution of Pt from the loaded resin using 0.1 mol/dm3_{of thiourea.}

Adsorption of Fe and Pt Solution containing Si and Al

Elution of Pt

0.1 mol/dm3_{of thiourea}

Leaching of spent catalyst

20vol%HCl+ 3 times the

stoichiometric ratio of H2O2

Roasting of spent catalyst

800°C, 5 h, 10°C/min

Leached catalyst

Leaching solution Roasted catalyst

Removal of Fe

0.0001 mol/dm3_{of HCl} Effluent containing Fe

Effluent containing Pt

Resin Loaded resin