M K Mohammad Ziaul Hyder1, Bungo Ochiai1. 1. Department of Chemistry and Chemical Engineering, Faculty of Engineering, Yamagata University, Jonan 4-3-16, Yonezawa, Yamagata 992-8510, Japan.
Abstract
We report the synthesis and practical application of a novel scavenger for precious metals. The scavenger was prepared from cellulose filter paper with grafted chains of poly(glycidyl methacrylate) modified with a novel ligand group of N-methyl-2-hydroxyethylcarbamothioate moieties, introduced by the reaction with O-1-mercapto-3-phenoxypropan-2-yl N-methyl-2-hydroxyethylcarbamothioate. Batch experiments were performed to evaluate the capability of the scavenger in ranges of pH and acid concentration as well as to determine the kinetics and isotherm models. The scavenger was found to adsorb only Ag(I), Pd(II), and Au(III) from an aqueous media in the presence of coexisting ions of different bases and precious metals at wide ranges of pH and acid concentration. The adsorption rates fit a pseudo-second-order kinetic equation, and the adsorption reached equilibrium within 60 min. The isotherm studies indicated that the obtained data were a good fit with the Langmuir model. The maximum adsorption capacities of Ag(I), Pd(II), and Au(III) were 126.95, 124.67, and 230.67 mg g-1, respectively. Regeneration experiments indicated that the adsorbent maintained 97% of its initial efficiency even after five adsorption/desorption cycles. The scavenger was effectively utilized to recover Ag(I), Pd(II), and Au(III) from an aqua regia solution of waste printed circuit boards.
We report the synthesis and practical application of a novel scavenger for precious metals. The scavenger was prepared from cellulose filter paper with grafted chains of poly(glycidyl methacrylate) modified with a novel ligand group of N-methyl-2-hydroxyethylcarbamothioate moieties, introduced by the reaction with O-1-mercapto-3-phenoxypropan-2-yl N-methyl-2-hydroxyethylcarbamothioate. Batch experiments were performed to evaluate the capability of the scavenger in ranges of pH and acid concentration as well as to determine the kinetics and isotherm models. The scavenger was found to adsorb only Ag(I), Pd(II), and Au(III) from an aqueous media in the presence of coexisting ions of different bases and precious metals at wide ranges of pH and acid concentration. The adsorption rates fit a pseudo-second-order kinetic equation, and the adsorption reached equilibrium within 60 min. The isotherm studies indicated that the obtained data were a good fit with the Langmuir model. The maximum adsorption capacities of Ag(I), Pd(II), and Au(III) were 126.95, 124.67, and 230.67 mg g-1, respectively. Regeneration experiments indicated that the adsorbent maintained 97% of its initial efficiency even after five adsorption/desorption cycles. The scavenger was effectively utilized to recover Ag(I), Pd(II), and Au(III) from an aqua regia solution of waste printed circuit boards.
Precious
metals are widely used in diverse applications such as
catalysis, electronics, automobiles, jewelry, and so on. In addition,
the recent advance of technology is widening the field of their application
more and more. However, the natural deposits and occurrences of the
precious metals are limited. The recent technological advancements
of electrical and electronic equipment are rapidly changing our lifestyles,
resulting in a discharge of a substantial amount of waste electrical
and electronic equipment at a quick pace.[1−4] The quick production is accelerating
the consumption of valuable metals such as gold, silver, and palladium.
Consequently, waste electrical and electronic equipment
is an essential resource for recovering valuable metals.[5,6] Printed circuit boards are the key components of electrical and
electronic equipment, where large amounts of valuable metals are used,
and hence, these are considered as a greater source of precious metals
than natural high-grade ores.[7−10] Further, the amounts of these treasured metals are
diminishing and in many mines are already exhausted. As a result,
viable recuperation of these valuable metals from secondary assets
such as waste printed circuit boards (WPCBs) is quite crucial from
economic and environmental points of view.[11,12] The precious metals in circuit boards coexist with other base metals,
and their ratios and the coexisting metals are widely different. The
aqueous chemistry of metal ions sometimes differs whether they are
in individual solutions or in mixtures with other metal ions. Hence,
the recovery of precious metals from multielemental mixtures is quite
difficult. Therefore, materials for the recovery of precious metals
must have no affinity with other coexisting metal ions such as copper,
iron, and zinc, which are abundantly used in electronics.[13]The recovery of precious metals is carried
out by various methods
such as a hydrometallurgical process, membrane filtration, extraction,
and ion exchange, but these processes are often accompanied by problems
in selectivity, rate, cost, and difficulty of operation.[14−17] Thus, it is crucial to develop economically practical and sufficiently
selective methods and materials for the recovery of precious metals.[18] Low-cost adsorbents with excellent selectivity
toward the target precious metals are expected to provide a beneficial
solution to this problem.Modified cellulose materials have
been attracting significant attention
as efficient and cost-effective bio-based adsorbents.[19−21] Cellulose is the most abundant biopolymer with outstanding physical
and chemical properties. Cellulose can be easily modified by oxidation,
etherification, esterification, and graft copolymerization due to
the active hydroxyl groups of the cellulose at the 2-, 3-, and 6-positions.[22−24] Various chemically functionalized celluloses have been designed
for the efficient recovery of heavy metals[25,26] and precious metals.[19,22,27,28] Incorporating sulfur functional groups on
polymers is a promising approach for the design of adsorbents for
soft metals. Adsorption characters of adsorbents, such as selectivity
and adsorption capacity, depend on the structures of the sulfur functional
groups bound to the adsorbents.[29] While
sulfur functional groups have affinities with soft metals, most sulfur
functional groups, such as thiourea[30,31] and dithiocarbamate
anions,[32−35] are not effective for selective attraction due to their affinities
with wide ranges of metals. The selective collection of precious metals
such as Ag(I), Pd(II), and Au(III) requires appropriate designs of
the ligands.[36−38] We have focused on soft thiocarbonyl moieties with
a lower contribution of harder tautomeric thiol groups.[29,39] For the support of the sulfur ligand, the authors have focused on
cellulose filter paper (CFP) that is advantageous due to its abundance,
inexpensiveness, accessibility, mechanical strength, and high surface
area.[40,41] In our previous work, we synthesized the
effective adsorbent CFP-g-PHCTMA from O-1-mercapto-3-phenoxypropan-2-yl-2-hydroxyethylcarbamothioate (HCT)
and CFP grafted with poly(glycidyl methacrylate) (CFP-g-PGMA). CFP-g-PHCTMA showed good and selective adsorption
capability toward the precious metals Ag(I), Pd(II), and Au(III).[42] The adsorption reached equilibrium within 60
min for both Au(III) and Pd(II), whereas it reached equilibrium at
a slower rate for Ag(I) of 120 min. The Langmuir monolayer adsorption
capacities of the Ag(I), Pd(II), and Au(III) ion species were found
to be 78.26, 92.35, and 152.38 mg g–1, respectively.Encouraged by this result, we designed the novel adsorbent CFP-g-PMHCTMA bearing the novel selective ligand N-methyl-2-hydroxyethylcarbamothioate by the reaction of CFP-g-PGMA and O-1-mercapto-3-phenoxypropan-2-yl N-methyl-2-hydroxyethylcarbamothioate (MHCT). We synthesized
the MHCT ligand, the N-methyl analogue of the HCT
ligand,[42] for systematic studies of the
adsorption properties of precious-metal ions. The adsorption manners
of CFP-g-PMHCTMA toward precious metals were investigated
on the basis of selectivity, adsorption isotherm, and kinetics for
a better understanding of the mechanism of adsorption of precious
metals. The empirical recuperation of Ag(I), Pd(II), and Au(III) from
WPCB leachate in aqua regia utilizing CFP-g-PMHCTMA
was also demonstrated in this study.
Synthesis
and Characterization of the Novel
Ligand
Synthesis of MHCT
MHCT was prepared
by the previously reported method for HCT.[42] DTC (2.26 g, 10.0 mmol) and N-methylaminoethanol
(9.01 g, 12.0 mmol) was stirred for 3 h in THF (5 mL) at room temperature
under an N2 atmosphere. Volatile substances were removed
under reduced pressure. MHCT was obtained by silica gel column chromatography
with CHCl3/MeOH as eluent (v/v = 1/0.05). Yield: 2.03 g,
6.75 mmol, 67.5%. The product was analyzed only by 1H and 13C NMR spectroscopy due to the low stability of the thiol
group.1H NMR (400 MHz, CDCl3): 2.99 (−CH2SH), 3.07–3.44 (cis- or trans-NCH3), 3.59–3.69
(−CH2CH2OH), 3.84–3.92
(−CH2OH), 4.25–4.28 (C6H5OCH2−) 5.72–5.78
(−CH2CHOCH2−),
6.92 (C6H5−), 7.25 (−C6H5) ppm (Figure S1).13C NMR (100 MHz, CDCl3):
24.73 (−CH2SH), 37.61 and 42.47
(cis or trans-NCH3−),
53.13 and 56.82 (−CH2CH2OH), 60.62 and 60.99 (C6H5OCH2−), 66.43 (−CH2OH), 79.35 and 79.72 (−CH2CHOCH2−), 114.79, 121.49, 129.63, 158.34 (−C6H5), 188.56 (−OC=S) ppm (Figures S2 and S3).
Acylation of MHCT
MHCT was acetylated
to protect the −OH and −SH groups for further confirmation
of the structure of MHCT. DTC (0.226 g, 1.00 mmol) and N-methylaminoethanol (0.090 g, 1.2 mmol) were reacted in THF (5 mL)
for 3 h at room temperature under an N2 atmosphere. Then,
acetic anhydride (1.0 g, 10 mmol) and triethylamine (0.020 g, 2.0
mmol) were added to the reaction mixture, and the reaction was allowed
to continue for 18 h. The acetylated product was obtained just by
washing once with a brine solution and twice with distilled water
and characterized by 1H NMR (400 MHz, CDCl3)
(Figure S4) and 13C NMR (100
MHz, CDCl3) (Figure S5).
Synthesis of the Novel Scavenger CFP-g-PMHCTMA
CFP-g-PMHCTMA was synthesized
as shown in Scheme by following the previously reported method for CFP-g-PHCTMA, the NH analogue of CFP-g-PMHCTMA.[42] CFP-g-PGMA (1.0 g) was added
to a mixture of MHCT (2.0 equiv with respect to the GMA unit) and
K2CO3 (2.0 equiv with respect to the GMA unit)
in 10 mL of acetone, and the mixture was stirred at 60 °C for
3 h under a nitrogen atmosphere. CFP-g-PMHCTMA was
consecutively washed with water, methanol, and acetone and was dried
at 60 °C under vacuum for 24 h.
Scheme 1
Synthesis of MHCT
and CFP-g-PMHCTMA
Result and Discussion
Synthesis
and Characterization of CFP-g-PMHCTMA
The N-methylthiourethane
skeleton in CFP-g-PMHCTMA was fabricated by the reaction
of N-methylaminoethanol and a cyclic dithiocarbonate.
The N-methylated mercaptothiourethane was used for
modification of CFP-g-PGMA by the nucleophilic ring-opening
addition of the thiol moieties to the epoxy rings under basic conditions.CFP-g-PMHCTMA was characterized by FTIR and EDX
spectroscopic analyses. The FTIR spectra of CFP, CFP-g-PGMA, and CFP-g-PMHCTMA are represented in Figure . The characteristic
absorption band of the carbonyl moieties at 1728 cm–1 and those of the epoxy moieties at 849 and 907 cm–1 appeared in the FTIR spectrum of CFP-g-PGMA (Figure b), which was not
found in the spectrum of CFP (Figure a). The characteristic peaks of the C=S moieties,
the NH moieties, and the aromatic ring of the MHCT ligand at 1243,
1587, and 1598 cm–1, respectively, were observed
in the FTIR spectrum of CFP-g-PMHCTMA (Figure c), which were not observable
in the spectrum of CFP-g-PGMA or in the spectrum
of CFP. Moreover, the absorptions of the epoxy moieties observed at
849 and 907 cm–1 for CFP-g-PGMA
are not observable in the spectrum of CFP-g-PMHCTMA,
strongly indicating the effective incorporation of the MHCT ligand.
The introduction of the sulfur-containing structure was further confirmed
by the EDX spectrum of CFP-g-PMHCTMA (Figure S6c) showing a characteristic peak of
sulfur at 2.3 keV in addition to the original carbon and oxygen peaks
observed in the EDX spectra of CFP-g-PGMA (Figure S6b) and CFP (Figure S6a).
Figure 1
FTIR spectra of (a) CFP, (b) CFP-g-PGMA,
and (c)
CFP-g-PMHCTMA.
FTIR spectra of (a) CFP, (b) CFP-g-PGMA,
and (c)
CFP-g-PMHCTMA.
Adsorption Behavior of CFP-g-PMHCTMA
toward Various Metal Ions
The adsorption behaviors
of CFP-g-PMHCTMA toward various metal ions—namely
Cu(II), Ag(I), Ni(II), Os(IV), Pd(II), Pt(IV), Au(III), Re(VII), and
Rh(III)—from their individual solutions were investigated to
understand the selectivity of the adsorbent (Figure ). CFP-g-PMHCTMA was highly
selective toward Ag(I), Pd(II), and Au(III) with an excellent adsorption
capacity exceeding that of CFP-g-PHCTMA and showed
very weak affinity with Cu(II) and Pt(IV) metal ions. Other precious
and base metals were not adsorbed at all, similarly to the case for
CFP-g-PHCTMA. This selective adsorption ability of
CFP-g-PHCTMA agrees with Pearson’s hard–soft
acid–base concept,[44] where the very
soft Lewis basicity of the thione functional groups on CFP-g-PMHCTMA results in high affinity with the very soft Lewis
acids Ag(I), Pd(II), and Au(III). Losev et al.[45] reported that an adsorbent bearing dialkylthiourea moieties
adsorbs Ag(I), Pd(II), Au(III), and Pt(IV) due to their labile properties
and has negligible affinity toward Rh(III), Ru(IV), Ir(IV), and Os(IV)
due to their harder characteristics and that an adsorbent bearing
disulfide moieties shows higher selectivity. Hence, the higher selectivity
of the MHCT ligand toward Ag(I), Pd(II), and Au(III) can be attributed
to the absence of the contribution of the iminothiol tautomeric structure
with harder basicity.
Figure 2
Adsorption of single metal ions from individual solutions
of Cu(II),
Ag(I), Ni(II), Os(IV), Pd(II), Pt(IV), Au(III), Re(VII), and Rh(III)
by CFP-g-PMHCTMA. Conditions: concentrations of Cu(II),
Ag(I), Ni(II), Os(IV), Pd(II), Pt(IV), Re(VII), and Rh(III), 200 mg
L–1; concentration of Au(III), 280 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.1;
25 °C; 24 h.
Adsorption of single metal ions from individual solutions
of Cu(II),
Ag(I), Ni(II), Os(IV), Pd(II), Pt(IV), Au(III), Re(VII), and Rh(III)
by CFP-g-PMHCTMA. Conditions: concentrations of Cu(II),
Ag(I), Ni(II), Os(IV), Pd(II), Pt(IV), Re(VII), and Rh(III), 200 mg
L–1; concentration of Au(III), 280 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.1;
25 °C; 24 h.The adsorbents after
the adsorption of Ag(I), Pd(II), and Au(III)
were analyzed by EDX. The clear peaks of Ag(I), Pd(II), and Au(III)
that appeared in the EDX spectra indicate the adsorption of these
metal ions onto CFP-g-PMHCTMA (Figure S7).
Effect of pH on Adsorption
of Ag(I), Pd(II),
and Au(III) Ions
The effect of pH on the adsorption of metal
ions on CFP-g-PMHCTMA was studied over the range
of 1.20 ≤ pH ≤ 5.95, and the results are presented in Figure a. The experiments
at varying pH were conducted with 5 mL amounts of 200 mg L–1 aqueous solutions of Ag(I) and Pd(II) and 300 mg L–1 aqueous solutions of Au(III) on contact with 5 mg of CFP-g-PMHCTMA. The pH was adjusted with aqueous solutions of
HCl and NaOH. The suspensions were agitated with a mechanical shaker
for 24 h at 25 °C. The uptakes of Au(III) and Ag(I) remained
almost constant from pH 1.20 to 3.10 and pH 1.20 to 4.43, respectively,
and dropped at the higher pH region. In contrast, the uptakes of Pd(II)
were almost identical in the examined range of pH. The adsorption
capacity became the highest for Ag(I), Pd(II), and Au(III) at pH 3.1,
pH 4.30, and pH 4.43, respectively. Earlier studies on the uptake
of these precious metals indicated the suitable pH was from 1.0 to
4.0.[46,47] The constant uptake of the metal ions over
the wide range of pH probably originates from the negligible competition
of hard H+ toward the target precious metals on interaction
with the soft thiocarbamate ligand even under highly acidic conditions.
The most plausible structure of the complexes is based on the coordination
of the free lone pair electron of the sulfur atom to the metal ions.
The decrease in uptakes observed for Ag(I) and Au(III) at pHs higher
than 4.0 probably originated from the hydrolysis of the metal ions
to form hydroxide complexes.[48,49]
Figure 3
(a) Effect of pH on the
adsorption of Ag(I), Pd(II), and Au(III)
ions. (b) Effect of [H+] concentration on the adsorption
of Ag(I), Pd(II), and Au(III) onto CFP-g-PMHCTMA.
Conditions; initial concentration of Ag(I) and Pd(II), 200 mg L–1, initial concentration of Au(III), 300 mg L–1, CFP-g-PMHCTMA, 1 g L–1; 24 h;
25 °C.
(a) Effect of pH on the
adsorption of Ag(I), Pd(II), and Au(III)
ions. (b) Effect of [H+] concentration on the adsorption
of Ag(I), Pd(II), and Au(III) onto CFP-g-PMHCTMA.
Conditions; initial concentration of Ag(I) and Pd(II), 200 mg L–1, initial concentration of Au(III), 300 mg L–1, CFP-g-PMHCTMA, 1 g L–1; 24 h;
25 °C.
Effect
of H+ Ion Concentration
on Adsorption Procedure
This is an important requirement
for metal scavengers for the selective adsorption of precious metals
in strong acid with a view to determining empirical hydrometallurgical
collections of precious metals, as leaching of a metal is often carried
out with concentrated acids. Hence, adsorption isotherm studies under
highly acidic conditions were carried out for Pd(II) and Au(III) in
aqueous HCl and for Ag(I) in aqueous HNO3 (Figure b). Quantitative amounts of
Ag(I), Pd(II), and Au(III) were adsorbed by CFP-g-PMHCTMA even at 6 M acid concentration. The outstanding adsorption
capacity of CFP-g-PMHCTMA at a higher acid concentration
places this adsorbent as a viable candidate for the empirical recovery
of Ag(I), Pd(II), and Au(III) from waste streams.The adsorption
capacity for Ag(I), Pd(II), and Au(III) decreased upon an increase
in the acid concentration, while the capacity is higher than that
of CFP-g-PHCTMA[42] even
at a higher acid concentration. The reduced adsorption capacities
at higher acid concentrations can be illustrated by the marginal protonation
of sulfur atoms in the thione functional groups of CFP-g-PMHCTMA under the very acidic conditions (Figure ), as described in the adsorption of Au(III)
and Pd(II) by phosphine sulfide type chelating polymers[50] and by the competition between protons and Ag(I),
Pd(II), and Au(III) for the adsorption positions on the CFP-g-PMHCTMA.[21]
Figure 4
Scheme of the protonation
of the sulfur atom on functional groups.
Scheme of the protonation
of the sulfur atom on functional groups.
Adsorption Isotherm
Adsorption isotherms
provide important information on adsorption mechanisms in the design
of sorption systems. Figure represents the isotherms for batch adsorption of Ag(I), Pd(II),
and Au(III) by CFP-g-PMHCTMA at pH 3.0. The weight-based
adsorption isotherms shown in Figure a confirm that the uptake capacities for Ag(I), Pd(II),
and Au(III) at the equilibria increase with an increase in the concentration
of the metal solutions. An adsorption capacity exceeding 100 mg g–1 is high among selective adsorbents. Typical adsorbents
with capacities above 100 mg g–1 employ weakly selective
adsorption groups such as ethylenediamine and amino acid or contain
nonspecifically binding substrates such as chitosan and graphene oxide.[51] The high capacity along with the selectivity
described below is the significant advantage of CFP-g-PMHCTMA. The improved adsorption capacities probably originated
from the excellent affinity of the N-methylthiocarbamothioate
ligand. For a better explanation of the relationship between the adsorption
capacity and the equilibrium concentration of the metal ions, the
Langmuir isotherm (eq ) (Figure b) and
Freundlich isotherm (eq ) were investigated by fitting the experimental data for the batch
adsorption isothermwhere qe (mg g–1) is the amount of
metal ion adsorbed at equilibrium, qm (mg
g–1) is the maximum
adsorption in a monolayered adsorption system, Ce (mg L–1) is the equilibrium concentration
of metal ions in solution, KL (L g–1) is the Langmuir adsorption equilibrium constant
related to adsorption energy, KF (L g–1) is the Freundlich adsorption equilibrium constant
related to adsorption capacity, and n is an empirical
parameter related to the adsorption intensity.
Figure 5
Isotherm of (a) adsorption
capacity of Ag(I), Pd(II), and Au(III)
ions in milligrams. (b) Linear plots of the Langmuir isotherms of
adsorption of Ag(I), Pd(II), and Au(III) on CFP-g-PMHCTMA. Conditions: initial concentration of Ag(I), Pd(II), and
Au(III) ions, 20–300 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.0; 25 °C;
24 h.
Isotherm of (a) adsorption
capacity of Ag(I), Pd(II), and Au(III)
ions in milligrams. (b) Linear plots of the Langmuir isotherms of
adsorption of Ag(I), Pd(II), and Au(III) on CFP-g-PMHCTMA. Conditions: initial concentration of Ag(I), Pd(II), and
Au(III) ions, 20–300 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.0; 25 °C;
24 h.The Langmuir adsorption isotherm
model was presumably utilized
on the basis of adsorbing one metal ion by each C=S moiety.
The adsorbed quantities of Ag(I), Pd(II), and Au(III) ions at the
equilibria (qe) was found to increase
with an increase in the initial metal ion concentrations, and plateau
regions were noted at high concentrations of the metal ions. This
implies the monolayer adsorption of this adsorption system, which
corresponds to Langmuir-type adsorption.[52] Moreover, the linear correlation between the metal ion concentration
at equilibrium (Ce) and Ce/qe reveals that the experimental
data have a better fit to the Langmuir isotherm model in comparison
to the Freundlich isotherm model (Table S1 and eqs and 2) in the same manner as that for the S-modified cellulose
adsorbent previously reported[42] and similar
to those of other S-modified celluloses,[53,54] indicating that the active adsorption sites contributed equally
to the adsorption process. The adsorbed amounts of Ag(I), Pd(II),
and Au(III) calculated from the Langmuir isotherm studies were determined
to be 126.95, 124.67, and 230.67 mg g–1, respectively.
These figures are near the molar amount of C=S in CFP-g-PMHCTMA (1.32 mmol g–1) and the almost
equimolar adsorption of Ag(I) (1.17 mmol g–1), Pd(II)
(1.19 mmol g–1), and Au(III) (1.17 mmol g–1) on CFP-g-PMHCTMA, indicating that most of the
adsorption sites on the adsorbent are active. The total results denote
the complexation between the metal ions and the thiocarbamate moieties
in a 1:1 stoichiometry in the adsorbent with excellent efficiency.Furthermore, these capacities are significantly higher than those
of the previously reported CFP-g-PHCTMA,[42] for which the adsorption capacities toward Ag(I),
Pd(II), and Au(III) were 78.26, 92.35, and 152.38 mg g–1, respectively. The higher adsorption capacity of CFP-g-PMHCTMA toward Ag(I), Pd(II), and Au(III) is ascribed to the incorporation
of a greater amount of C=S contents. The introduction ratio
of the ligand structure in CFP-g-PMHCTMA being higher
than that of CFP-g-PHCTMA probably originates from
the absence of the thiourea proton with a higher acidity in comparison
to the thiol group, which was included in the previous ligand. This
functionalization proceeds through the nucleophilic addition of thiolate
moieties produced by a hydrogen abstraction with triethylamine, and
the presence of active protons with acidities higher than that of
the thiol group delays the reaction. The higher acidity of the thiourea
proton can be confirmed by its 1H NMR signal appearing
at 6.95 ppm, being at a lower magnetic field in comparison to the
signal of the thiol proton appearing at 2.98 ppm.[42]As is evident from this isotherm study, the adsorption
proceeded
on the homogeneous C=S sites binding metal ions in the 1:1
stoichiometry. A plausible structure of the binding site in CFP-g-PMHCTMA is illustrated in Figure . The most important binding site is the
C=S moieties, but the sulfide moieties might assist the ligation.
Figure 6
Plausible
mechanism of adsorption of metal ions on the functional
groups of CFP-g-PMHCTMA.
Plausible
mechanism of adsorption of metal ions on the functional
groups of CFP-g-PMHCTMA.
Effect of Contact Time and Studies of Adsorption
Kinetics
Fast uptake of precious metals is an important feature
of scavengers for precious metals from waste streams. To confirm the
rates of the adsorption, adsorption experiments were carried out for
10 mg L–1 of Ag(I), Pd(II), and Au(III) solution
using 1 g L–1 adsorbent (Figure ). The uptake exceeded 90% within 30 min,
and the adsorption reached equilibrium within 60 min.
Figure 7
Effect of contact time
of adsorption of Ag(I), Pd(II), Au(III)
onto CFP-g-PMHCTMA. Conditions: initial concentration
of each Ag(I), Pd(II), and Au(III) ions, 10 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.1;
25 °C.
Effect of contact time
of adsorption of Ag(I), Pd(II), Au(III)
onto CFP-g-PMHCTMA. Conditions: initial concentration
of each Ag(I), Pd(II), and Au(III) ions, 10 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.1;
25 °C.The adsorption rates of Ag(I),
Pd(II), and Au(III) on CFP-g-PMHCTMA were measured
at 25 °C to understand the
kinetic mechanism of the adsorption process. The kinetics of adsorption
was investigated with a pseudo-first-order kinetic model (eq ) and its linear form (eq ) and with a pseudo-second-order
model (eq ) and its
linear form (eq ).where qe (mg g–1) and q (mg g–1) are the mass of adsorbed metal
ions per
unit mass of the adsorbent at equilibria and time (t), respectively, and k1 (g mg–1 min–1) and k2 (dm–3 mg–1 min–1) are
the rate constants of pseudo-first-order and pseudo-second-order models,
respectively. The kinetic results of adsorption were obtained by evaluating Figure .
Figure 8
Adsorption kinetics of
Ag(I), Pd(II), and Au(III) on CFP-g-PMHCTMA. Conditions:
initial concentrations of Ag(I),
Pd(II), and Au(III) ions, 200 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.0–3.3;
25 °C.
Adsorption kinetics of
Ag(I), Pd(II), and Au(III) on CFP-g-PMHCTMA. Conditions:
initial concentrations of Ag(I),
Pd(II), and Au(III) ions, 200 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.0–3.3;
25 °C.The linearity in the pseudo-second-order
plots (Figure a) is
better than that of pseudo-first-order
plots (Figure b),
as proved by the higher R2 values exceeding
0.997 (Table S2). This pseudo-second-order
kinetics indicates that the rate-determining step of this adsorption
is the chemical reaction between the metal ions and the adsorbent.
Figure 9
Linear
plots of (a) pseudo-second-order kinetics and (b) pseudo-first-order
kinetics of adsorption of Ag(I), Pd(II), and Au(III) ions on CFP-g-PMHCTMA. Conditions: initial concentrations of Ag(I),
Pd(II), and Au(III) ions, 200 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.0–3.3;
25 °C.
Linear
plots of (a) pseudo-second-order kinetics and (b) pseudo-first-order
kinetics of adsorption of Ag(I), Pd(II), and Au(III) ions on CFP-g-PMHCTMA. Conditions: initial concentrations of Ag(I),
Pd(II), and Au(III) ions, 200 mg L–1; CFP-g-PMHCTMA, 1 g L–1; pH 3.0–3.3;
25 °C.
Selective
Adsorption of Ag(I), Pd(II), and
Au(III) from Multielemental Solutions
The recovery of precious
metal ions has been conducted for multielemental solutions comprising
a range of base metals such as nickel, copper, and zinc at high acid
concentrations. In order to confirm the selective adsorption ability
of CFP-g-PMHCTMA toward precious-metal ions, adsorption
experiments were performed using two types of multielemental solutions
with various acid concentrations (Figure S8). Figure S8a represents the selective
capture of Ag(I) from a multielemental solution (1) comprising
Cu(II), Ag(I), Zn(II), Ni(II), Cd(II), V(V), Cr(III), Mn(II), Fe(III),
and Co(III) in aqueous HNO3 in an acid concentration range
from 1 to 6 M. This result shows that CFP-g-PMHCTMA
captures only Ag(I) from the mixture of different base-metal ions
without coadsorption of other metals in the highly acidic media. Figure S8b shows the metal adsorption of CFP-g-PMHCTMA from a multielemental solution (2) comprising Ru(III), Rh(III), Au(III), Ir(III), Pd(II), Pt(IV),
and Os(IV) in aqueous HCl in an acid concentration range from 1 to
6 M. Pd(II) and Au(III) were predominantly bound to the adsorbent.
No other precious metals were adsorbed at all by CFP-g-PMHCTMA at higher acid concentrations, whereas very trace amounts
of Pt(IV) and Os(IV) were found to be adsorbed at lower acid concentrations.
These data indicate that specific precious metals can be selectively
recovered from a mixture of various precious-metal ions even in highly
acidic media. A possible factor for the trace adsorption of Os(IV)
and no affinity toward Cu(II) from multielemental solutions by CFP-g-PMHCTMA is the effect of coexisting ions. Adsorption behaviors
in multielemental solutions are often affected by the coexisting ions,
and an analysis of the effects is very difficult due to the complexity,
as reviewed by Neris et al.[55]CFP-g-PMHCTMA was employed for the study of adsorption from
a mixture of 1 in HNO3 media and 2 in HCl media (total acid concentration 1.53 M) (Figure S9). CFP-g-PMHCTMA was found to adsorb
Ag(I), Pd(II), and Au(III) selectively and efficiently even from a
mixture of the two multielemental solutions as well as the individual
and the original multimetal solutions. The photo images of the state
of CFP-g-PMHCTMA before and after adsorption from
the mixtures of 1 and 2 are shown in Figure S10, indicating the coloring of the adsorbent
after the adsorption.
Selective Recovery of Ag(I),
Pd(II), and Au(III)
from a WPCB Leachate
The metal adsorption behavior of CFP-g-PMHCTMA from a WPCB metal leachate in aqua regia (Figure S11) was investigated, and the result
is shown in Figure S12. Pd(II) and Au(III)
were absolutely captured onto the adsorbent CFP-g-PMHCTMA, while Cu(II), Ni(II), V(V), Zn(II), and Fe(II) were negligibly
captured. The adsorption capacity of Ag(I) was to be found lower than
that of Au(III) and Pd(II), probably due to the formation of a stable
complex in aqua regia.[56]Figure S13 shows that the color of CFP-g-PMHCTMA
changed from white to yellow in a similar fashion with an adsorption
from the mixture of the multielemental solutions shown in Figure S10. These results evidently demonstrate
that the precious metals Ag(I), Pd(II), and Au(III) were effectively
retrieved by CFP-g-PMHCTMA from the metal leachate
from WPCBs containing a significantly higher concentration of various
metals and unidentified compounds originating from WPCBs. Specifically,
CFP-g-PMHCTMA is a potential and promising candidate
for the practical recovery of precious metals.
Desorption
of Precious Metals from Metal-Loaded
CFP-g-PMHCTMA and Its Reuse Ability
The
regeneration and reusability of an adsorbent is a vital feature from
a consideration of practical and economical purposes. The desorption
efficiency of the adsorbed precious metal (50 mg L–1) onto 1 g L–1 CFP-g-PMHCTMA was
determined using various concentrations of HNO3, thiourea-HNO3, and thiourea solutions. The effect of the eluent on the
desorption of the precious metals and the adsorption capacity of the
regenerated adsorbent are presented in Table . HNO3 solutions were not effective
for desorption, as the degrees of desorption are less than 10% regardless
of the concentrations. The 0.5 M thiourea solution performed better,
while it was not effective enough. In contrast, quantitative desorption
was attained using the eluent 0.1 M HNO3 and 0.5 M thiourea,
while lower concentrations of thiourea resulted in lower desorption
efficiencies. The adsorbent after the desorption can be recycled. Table shows the adsorption
capacity during the five repeated adsorption–desorption processes
relative to the initial capacity. The relative adsorption capacity
exceeded 97% even at the fifth cycle, indicating the excellent durability
of CFP-g-PMHCTMA.
Table 1
Desorption of Ag(I),
Pd(II), and Au(III)
from CFP-g-PMHCTMA after Adsorption of Metal Ions
with Different Concentrations of HNO3, Thiourea, and Their
Mixturesa
degree
of desorption of metals (%)
eluent
Ag(I)
Pd(II)
Au(III)
0.5 M HNO3
0.3
1.5
0.6
1 M HNO3
0.4
4.4
0.9
2 M HNO3
0.8
10.6
1.0
0.1 M HNO3 + 0.1
M thiourea
61.9
53.2
76.0
0.1 M HNO3 + 0.3 M thiourea
83.2
75.9
100.0
0.1 M HNO3 + 0.5 M Thiourea
100.0
99.1
100.0
0.5 M thiourea
40.8
49.7
65.0
Conditions: Ag(I), Pd(II), and
Au(III), 50 mg L–1; CFP-g-PMHCTMA,
1 g L–1; 25 °C; desorption time, 3 h.
Table 2
Repeated Adsorption
of Ag(I), Pd(II),
and Au(III) Ions by CFP-g-PMHCTMAa
relative
adsorption capacity (%)
cycle no.
Ag(I)
Pd(II)
Au(III)
1
100.0
100.0
100.0
2
100.0
99.4
100.0
3
99.7
97.6
99.1
4
98.8
97.3
99.1
5
98.6
96.7
98.7
Conditions:
Ag(I), Pd(II), and
Au(III), 50 mg L–1; CFP-g-PMHCTMA,
1 g L–1; 25 °C; contact time for each adsorption–desorption
cycle, 3 h.
Conditions: Ag(I), Pd(II), and
Au(III), 50 mg L–1; CFP-g-PMHCTMA,
1 g L–1; 25 °C; desorption time, 3 h.Conditions:
Ag(I), Pd(II), and
Au(III), 50 mg L–1; CFP-g-PMHCTMA,
1 g L–1; 25 °C; contact time for each adsorption–desorption
cycle, 3 h.
Conclusions
This study has shown that CFP-g-PMHCTMA is a promising
adsorbent for the precious metals Ag(I), Pd(II), and Au(III) from
the solutions containing various kinds of metals under a wide range
of pH conditions, including high acid concentrations. CFP-g-PMHCTMA demonstrated an outstanding selectivity of adsorption
toward Ag(I), Pd(II), and Au(III) from multielemental solutions containing
various base and noble metals. These precious metals were selectively
adsorbed with negligible adsorption of coexisting metal ions present
in excess. The efficiency of CFP-g-PMHCTMA relies
on the N-methyl-2-hydroxyethylcarbamothioate ligand
designed to make the Lewis basicity of the sulfur ligand softer by
eliminating the contribution of the harder iminothiol tautomer. The
adsorption capacity is higher than that of the −NH analogue
previously reported, probably due to the absence of active hydrogens
competitively abstracted during the nucleophilic modification of the
epoxy ring. The experimental data exactly fitted the Langmuir isotherm
model and pseudo-second-order kinetics. The loaded metals from CFP-g-PMHCTMA can be effectively desorbed by 0.5 M thiourea
in 0.1 M HNO3. CFP-g-PMHCTMA retained
its original efficacy even after five times of reuse. CFP-g-PMHCTMA was shown to be an excellent scavenger for the
selective recovery of Au(III), Pd(II), and Ag(I) from WPCB leachates
in aqua regia. This investigation establishes that this cellulose-based
CFP-g-PMHCTMA is an eco-friendly and economically
viable selective adsorbent for precious metals due to its recyclability
and high selectivity even in the presence of multiple ions.
Experimental Section
Chemicals
All
the chemicals used
were of analytical reagent grade. CFP (Advantec5C; Toyo Roshi, Tokyo,
Japan) was used for modification. Glycidyl methacrylate (GMA), potassium
carbonate, sodium hydroxide, HCl, and N-methyl-2-aminoethanol
were purchased from Kanto Chemical (Tokyo, Japan). K2H2PO4-NaOH-7.2 buffer solution was purchased from
Tokyo Chemical Industry (Tokyo, Japan). Diammonium ceric(IV) nitrate
(CAN), nitric acid, and Cu(II) (Cu(NO3)2) and
Ni(II) (Ni(NO3)2) standards for ICP (1000 mg
L–1) were purchased from Wako Chemicals (Tokyo,
Japan). Pd(II) (Pd(NO3)2 + HCl), Au(III) (HAuCl4), Ag(I) (AgNO3), Rh(III) (HRhCl4),
and Pt(IV) (H2PtCl6) standards for ICP (1000
mg L–1), transition metal mix 1 (Cu(II), V(V), Ni(II),
Co(II), Fe(III), Mn(II), Cr(III), Ag(I), Zn(II) and Cd(II)), and transition
metal mix 2 (Pd(II), Au(III), Pt(IV), Ru(III), Rd(III), Os(IV), and
Ir(III)) for ICP (100 mg/L) were purchased from Sigma-Aldrich (St.
Louis, Missouri, USA). Re(VII) (NH4ReO4 in H2O) and Os(IV) ((NH4)2OsCl6 in 7% HCl) were purchased from Merck (Darmstadt, Germany). Water
was purified with a MINIPURE TW-300RU apparatus (Nomura Micro Science,
Kanagawa, Japan). 4-(Phenoxymethyl)-1,3-oxathiolane-2-thione (DTC)
was prepared according to the literature.[43] CFP-g-PGMA with a grafting percentage of 156% was
prepared as described in a previous work.[42] In brief, grafting from the polymerization of GMA (1.20 g, 8.57
mmol) was carried out with CFP (0.60 g) in 20 mL of water for 30 min
in the presence of a freshly prepared 0.005 M solution of CAN in 10
mL of 0.1 M HNO3.
Measurements
1H and 13C NMR spectra were recorded on a JEOL ECX
400 NMR spectrometer
(400 MHz for 1H and 100 MHz for 13C). FTIR spectra
were recorded on a JASCO FT/IR-460 plus Fourier transform infrared
spectrophotometer in a KBr matrix. Energy-dispersive X-ray (EDX) spectra
were measured by a JEOL JSM-6510A scanning electron microscope equipped
with a JEOL JED 2300 EDX spectrometer. The concentration of metal
ions was determined by a PerkinElmer ELAN DRC II inductively coupled
plasma (ICP) mass spectrophotometer.All of the adsorption experiments
were carried out in batch mode. For a determination of the adsorption
isotherm, CFP-g-PMHCTMA (5 mg) was placed in small
plastic bottles containing 5 mL of the metal ion solution (10–280
mg L–1) at pH 3 and a temperature of 25 °C.
The bottles were equilibrated on a thermostatic shaker. The adsorption
amounts were calculated from the residual amounts of the metal ions
in the solutions.Kinetic experiments were conducted using 5
mg of CFP-g-PMHCTMA with 5 mL of 200 mg L–1 aqueous solutions
of metal ions. The samples were collected at different time intervals
to determine the equilibrium point of adsorption.The percent
of metal adsorption in solution was evaluated using eq C (mg L–1) = initial metal ion concentration; and C (mg L–1) = equilibrated metal ion concentration.The adsorption amounts
were calculated from residual amounts of
metal ions in solutions according to eq .where qe (mg g–1) is the adsorption capacity; V (L)
is the volume of the metal solution, and W (g) is
the mass of dry adsorbent.
Leaching of Metals from
Waste Printed Circuit
Board (WPCB) Using Aqua Regia
The metals in WPCBs were leached
by aqua regia. In a typical procedure, WPCBs collected from waste
computers were cut into 3 × 4 cm sizes. Then, 12 pieces of cut
WPCBs were put into freshly prepared 30 mL portions of aqua regia
and soaked for 30 min for complete leaching of metals into aqua regia.
The resulting mixture was filtered, and the filtrate was analyzed
by ICP-MS after proper dilution. The detected metal ions are V(V),
Fe(III), Ni(II), Cu(II), Mn(II), Cr(III), Co(II), Zn(II), Pd(II),
Au(III), and Ag(I), and the concentrations are indicated in Table S3.