Aqueous Synthesis of 14-15-Membered Crown Ethers with Mixed O, N and S Heteroatoms: Experimental and Theoretical Binding Studies with Platinum-Group Metals

Article abstract of DOI:10.1002/cplu.201800541

A series of 14-15-membered O, N, and S-containing crown ethers (CEs) was synthesized by cyclization of bis-epoxides with aryl-N or S dinucleophiles using triethylamine as a catalyst and LiCl as a metal template in water. The catalyst dosage, and metal template type and dosage were critical in achieving yields of 56–93 %. Liquid-liquid extraction (LLE) was performed to evaluate the CE complexation with Pd²⁺ and Pt²⁺. Among the CEs, a dioxa-dithia dibenzo CE exhibited the highest Pd²⁺ selectivity even in the presence of other platinum-group metals (PGMs). Complementary DFT studies reveal that this CE has the most compatible cavity dimension (?_CE=1.58 ?) with Pd²⁺ (?_Pd²⁺=1.56 ?) forming a square-planar S₄ geometry. Binding-energy calculations showed the Pd²⁺ complex has the least energy requirement for structural reorientation during complexation. Overall results highlight the importance of CE cavity dimension and presence of S heteroatoms for the structural design of CEs selective towards PGMs such as Pd²⁺.

Full text of DOI:10.1002/cplu.201800541

Accepted Article
Title: Synthesis of 14-15 membered crown ethers with mixed O, N and
S heteroatoms in water: Experimental and theoretical binding
studies with platinum-group metals
Authors: Rey Eliseo Castillo Torrejos, Grace Masbate Nisola, Sang
Hoon Min, Jeong Woo Han, Sangho Koo, Khino Jonota
Parohinog, Seong-Poong Lee, Hern Kim, and Wook-Jin
Chung
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemPlusChem 10.1002/cplu.201800541
Link to VoR: http://dx.doi.org/10.1002/cplu.201800541
A Journal of
www.chempluschem.org

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
Aqueous Synthesis of 14-15 Membered Crown Ethers with Mixed
O, N and S Heteroatoms: Experimental and Theoretical Binding
Studies with Platinum-Group Metals
Rey Eliseo C. Torrejos,[a] Grace M. Nisola,[a] Sang Hoon Min,[b] Jeong Woo Han,[c] Sangho Koo,[a],[d]
Khino J. Parohinog,[a] Seongpoong Lee,[a] Hern Kim,* [a] Wook-Jin Chung,*[a
will improve PGM recovery especially from secondary sources
Abstract: A series of 14-15-membered O, N, and S-containing
crown ethers (CEs) was synthesized by cyclization of bis-
like spent industrial catalysts and electronic wastes. [3, 4, 6]
Crown ethers (CEs) are cyclic ligands known for their high
epoxides with aryl-N or S dinucleophiles using triethylamine as a
affinity and specificity towards certain cations according to cavity
catalyst and LiCl as a metal template in water. The catalyst
size-match selectivity making them ideal PGM sequestrants. [7, 8]
dosage, and metal template type and dosage were critical in
Moreover, CEs with mixed oxygen, nitrogen, and sulfur (O, N, S)
achieving yields of 56–93%. Liquid–liquid extraction (LLE) was
heteroatoms are known to possess strong affinities towards
performed to evaluate the CE complexation with Pd2+ and Pt2+.
PGMs. [9-14] Oxathia (O, S), oxaaza (O, N), and oxaazathia (O, N,
Among the CEs, a dioxa-dithia dibenzo CE exhibited the highest
S) CEs which contain soft heteroatoms showed specific
Pd2+ selectivity even in the presence of other platinum group
coordination towards soft metals such as Pd2+ or Pt2+ according
metals (PGMs). Complementary DFT studies reveal that this CE
[12, 15, 16]
to the hard soft-acid base (HSAB) principle.
But despite
has the most compatible cavity dimension (ØCE=1.58 Å) with
Pd2+ (ØPd2+ = 1.56 Å) forming a square-planar S4 geometry.
Binding energy calculations showed the Pd2+ complex has the
least energy requirement for structural reorientation during
complexation. Overall results highlight the importance of CE
cavity dimension and presence of S heteroatoms for the structural
design of CEs selective towards PGMs such as Pd2+.
their structural tunability for improved selectivity, syntheses of
these CEs require multiple reaction steps using toxic and
hazardous solvents and afford low overall reaction yields. [9, 13-18]
From a previous study, 14 – 15 membered di-hydroxy Li+-
selective CEs have been efficiently synthesized via intermolecular
cyclization of bulky bis-epoxides with a nucleophile (i.e. catechol).
[8] Cyclization reactions involving bis-epoxides and catechol have
been typically performed in high boiling point solvents like tert-
Butyl alcohol (t-BuOH), dimethyl sulfoxide (DMSO),
dimethylformamide (DMF), and Tetrahydrofuran (THF)-H2O. To
afford O, N, S-bearing CEs with specificity towards PGMs, similar
reactions can be carried out by using ortho nitrogen- or sulfur-
substituted aromatic di-nucleophiles in lieu of catechol.
Introduction
Platinum-group metals (PGMs) are extremely scarce precious
elements due to their low natural abundance and highly complex
process required for their extraction and refining. [1-5] With their
high catalytic oxidation/reduction property, superior resistivity,
high thermal and electrical stability, and resistance to tarnish,
palladium (Pd) and platinum (Pt) are the most extensively used
Previous reports have successfully performed aminolysis or
thiolysis of epoxides in water with aryl amines or aryl thiols,
[19]
respectively.
Tertiary amines such as 1,4-diaza-
bicyclic[2,2,2]octane (DABCO) or triethylamine (Et3N) have
catalyzed the aqueous synthesis of -amino alcohols and -
aminothioethers in excellent yields under mild reaction conditions.
[20, 21] Through these routes, reactions between bis-epoxides and
aromatic diamines, dithiols or aminothiols to afford O, N, S-
bearing CEs could simplify and upgrade the cyclization steps with
the use of water as an environmentally benign reaction medium.
PGMs as catalysts in automotive, chemical, electronic and
[3-5]
biomedical industries.
Currently, PGMs are separated and
purified using commercial extractants which have low selectivities
and require multi-stage extraction process to achieve the desired
purity. Therefore, development of new highly selective extractants
[a]
Dr. R.E.C. Torrejos, Dr. G.M. Nisola, Prof. S. Koo, C.P. Lawagon,
K.J. Parohinog, Dr. S.P. Lee, Prof. H. Kim,* Prof. W.-J. Chung*
Energy and Environment Fusion Technology Center (E2FTC),
Department of Energy Science and Technology (DEST), Myongji
University Yongin Science Campus, South Korea (17058)
* Corresponding authors Email: wjc0828@gmail.com;
hernkim@mju.ac.kr
[b]
[c]
[d]
S.H. Min
Department of Chemical Engineering, University of Seoul, South
Korea (02504)
Prof. J.W. Han
Department of Chemical Engineering, Pohang University of Science
and Technology (POSTECH), Pohang, South Korea (37673).
Prof. S. Koo
Department of Chemistry, Myongji University Yongin Science
Campus, South Korea (17058)
Scheme 1. Cyclization of CEs with different O, N, S heteroatoms in water. i) 10-
Supporting information (SI) for this article is given via a link at the
end of the document.
30 mol % Et3N with 2 eq. LiCl in water at 80oC.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
Herein, 14 - 15 membered O, N, S-bearing CEs were effectively
synthesized in water via intramolecular cyclization of O-bearing
bis-epoxides with aromatic N, S, or N-S di-nucleophiles in the
presence of Et3N catalyst and lithium chloride template (Scheme
1). Three types of bis-epoxides (1a-1c) with bulky or rigid sub-
units were recruited. Previously, the bulky sub-unit of 1b was
proved to be effective in thwarting the complexation of a 14-
membered CE with unwanted cations while preferring Li+. [8] Thus,
incorporation of 1b in O, N, S-bearing CEs could enhance their
selectivity towards PGMs. Meanwhile bis-epoxide 1c was also
employed to see the effects of rigidified ring structures for PGM
selectivity and lastly, 1a to observe ring size effects as it can
afford 15-membered O, N, S-bearing CEs. The overall design of
the PGM-selective CEs was supported empirically by liquid-liquid
extractions (LLE) and theoretically by Density Functional Theory
(DFT) studies. These techniques were performed not only to
elucidate the effects of cavity size (ØCE) and structures but also to
observe the influence of O, N, and S heteroatoms on the
complexation specificity of O, N, S-bearing CEs towards PGMs
Pd2+ and Pt2+.
the reactants by controlling their solubilities in the reaction solvent.
Selection of appropriate solvents is the next sensible explorative
step to achieve CE cyclization.
Table 1. Isolated yield of 2a in different solvent dilution, catalyst and template
amount effects.
a in H2O,
mmol L-1
Isolated
yield (%)
Entry
Catalyst
Template
None
None
6 [a], [b]
14
1
2
100
100
100
100
50
None
1 eq. NaCl
None
3
1 mol% Et3N
10 mol% Et3N
10 mol% Et3N
10 mol% Et3N
10 mol% Et3N
10 mol% Et3N
10 mol% Et3N
10 mol% Et3N
50 mol% Et3N
100 mol% Et3N
10 mol% Et3N
10 mol% Et3N
12
4
1 eq. NaCl
1 eq. NaCl
1 eq. NaCl
1 eq. LiCl
1 eq. MgCl2
1 eq. KCl
1 eq. CaCl2
1 eq. MgCl2
1 eq. MgCl2
2 eq. LiCl
5 eq. LiCl
51
5
68
6
33
63
7
25
71
8
25
75
Results and Discussion
8 [a]
9
25
[a]
10
11
12
13
14
25
-
Synthesis of O, N, S-bearing CEs
25
Optimization of cyclization reactions were carried out using 2a as
model compound due to the abundant commercial supply of
reactants o-phenylenediamine a and bis-epoxide neopentyl
diglycidyl ether 1a. Initial cyclization reactions were done in
different solvents, catalysts and templates (SI Table S1). From
earlier report, intermolecular cyclization of 1a with catechol
conveniently afforded di-hydroxy 14-15 membered CE derivatives
in high yields (74-92%) using LiOH as metal template and basic
27
15 [a]
78
25
25
25
65
Reactions were done using 5 mmol a with 5 mmol 1a in water under N2 at
80oC for 24 h. Linear products detected. [a] high amount of unreacted 1a. [b]
[8]
catalyst in t-BuOH. However, using the same solvent-catalyst
system did not produce the desired CE 2a (SI Table S1, Entry 1)
nor the replacement of metal templates to NaOH, Mg(OH)2 or
Ca(OH)2 (SI Table S1, Entries 2 – 5). Similarly, cyclization of 2a
was unsuccessful in highly polar solvents such as DMSO and
DMF (SI Table S1, Entries 6 – 7). Alternatively, a different
approach using various Lewis acid perchlorates in Et2O or DCM
(SI Table S1, Entries 8 – 12) also did not produce the target CE
2a. These reactions which involve Lewis acid-catalyzed epoxide
In THF, THF-H2O and H2O solvents, a was soluble while 1a had
limited dissolution. However, reactions in these solvent systems
(SI Table S1, Entries 13 – 15) did not produce 2a. From previous
reports, excellent yields of epoxide ring opening with aromatic
amines in water have been achieved using weak Lewis base such
[20, 21]
tertiary amines.
Hence, aqueous reactions with Et3N as
catalyst combined with an appropriate metal chloride template
was carried out, which afforded 15 – 45% of 2a (SI Table S1,
entries 16 – 18). Further optimization studies were carried out
using Et3N as catalyst in H2O with different metal chlorides as
templates to improve the yield of 2a (Table 1). The critical roles
of the Et3N catalyst and templates were confirmed by the poor
reaction yield in the absence of both components (Table 1, Entry
1). The absence of either component also showed inferior 2a
production (Table 1, Entries 2 – 3). The effects of dilution, metal
template size and dosage, as well as catalyst loading were next
examined. Using NaCl as template, decreasing the concentration
ring opening reactions with different nucleophiles have been
[22]
previously reported.
Majority of the products (SI Table S1,
Entries 1 – 7, 12) were linear forms of a and 1a with 2:1
composition and occasionally, with the presence of unreacted bis-
epoxides (SI Table S1, Entries 8 – 11). These reactions did not
produce the desired cyclic 2a due to the high solubility of both a
and 1a in polar solvents such as t-BuOH, DMSO and DMF.
Excessive reactivities of both reactants promoted the formation of
[8]
linear by-products. Previously, CE cyclization of catechol with
1a was achieved by limiting the solubility of one of the reactants.
During reaction, catechol was dispersed in t-BuOH given its
limited solubility while 1a was completely dissolved in the solvent.
The slow release of catechol in the reaction medium controlled
the reaction kinetics, which limited the formation of linear products.
Thus, cyclization can be achieved under regulated reactivities of
of the reactant a by dilution (i.e. from 100 down to 50 mmol L-1
a
by increasing H2O volume) showed temporary increase in 2a yield
from 51% to 68% (Table 1, Entries 4 – 5), but slightly declined to
63% at 33 mmol L-1 a (Table 1, Entry 6). Nonetheless, changing
the template to LiCl and MgCl2 further improved the yields to 71 –
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
75% (Table 1, Entries 7 – 8) due to dilution effects (25 mmol L-1
a). The increase in yields can be ascribed to the size compatibility
of the templates with the cavity dimension of 2a. [23] This was
further confirmed when the reactions were carried out using
bigger metal templates like KCl or CaCl2 (Table 1, Entries 9 – 10).
Analysis of the crude product for CaCl2 template through high
resolution ESI-MS showed that majority of the products were
linear by-products of a and 1a with 2:1 or 2:2 composition (SI
Figure S1). Increasing the Et3N catalyst loading from 10 mol% to
100 mol% with MgCl2 template was detrimental to 2a (Table 1,
For a broader reaction scope, Et3N dosage was varied from 0 –
70 mol% (at 2 eq. LiCl) and tested not only for the cyclization of
1a with a but also for other di-nucleophiles b and c to produce 2a,
and its derivatives 2d and 2g. For 2a, the highest cyclization yield
was 82% using 10 mol% Et3N whereas that of 2d was 56% at 25
mol% Et3N and that of 2g was 86% at 30 mol% Et3N (Figure 1A).
The amount of Et3N required for optimized cyclization with 1a
follows the order a<b<c. The most acidic S containing di-
nucleophile
c required the highest Et3N concentration for
cyclization while the basic a required the least. In fact, cyclization
of a with 1a still occurred albeit at lower yields (Table 1, Entries 1
– 3), even without the presence of Et3N and/or metal template.
Further evidence (Figure 1B) is shown by the initial pH of solution
a (pH = 8.12) which is already basic even without the addition of
Et3N in contrast to b and c. Thus, cyclization reactions for 2d and
2g required higher dosages of Et3N.
Using the respective optimal Et3N dosages for each di-
nucleophile (a, b, c), additional reactions were carried out with
other bis-epoxides such as bulky 1b or rigid 1c to produce other
O, N, S-bearing CEs (Figure 2). The CE cyclization yields (Table
2) reveal that for the same di-nucleophile reactant, higher CE
productions were obtained when reacted with 1b than with 1c or
1a. This is likely due to the relatively higher solubility of 1b than
1c and 1a in water. Nonetheless, the cyclization yields of CEs 2a-
2i starting from the diol intermediates gave good to excellent
yields (56% – 93%) in comparison to other reported values for
CEs with mixed heteroatoms. [13, 24] More importantly, the resulting
CEs are dihydroxy-terminated which can be further functionalized
for extraction or covalently attached on solid supports as PGM
adsorbent materials. [8, 25]
Entries 11 – 12). Excessive presence of Et3N generates sufficient
[21]
OH- which promotes metal hydroxide precipitation.
Thus,
further increase in MgCl2 dosage was not carried out as it easily
precipitated as Mg(OH)2 in the presence of Et3N. As the next best
template, LiCl dosage was then varied. Increasing its amount
slightly improved the yield (Table 1, Entry 13) but not further to 5
eq. (Table 1, Entry 14). Therefore, 2 eq. of LiCl was necessary to
give excellent yields of 2a when using Et3N as catalyst. However,
there is a trade-off between the catalyst and metal template
loadings. Hence, further reaction optimization was implemented
at different Et3N amounts using LiCl as template at 2 eq.
Figure 1. Isolated yields of 2a, 2d and 2g using different Et3N catalyst mol% (A)
and measured initial pH of the reaction mixture at different Et3N catalyst mol%
before addition of bis-epoxide (B). Reaction conditions: 10 mmol of di-
nucleophile (a, b or c) with 10 mmol neopentyldiglycidyl ether in 200 mL H2O
solvent, 2 eq. LiCl at 80oC for 24 h.
Figure 2. Structures of synthesized O, N, S-bearing CEs.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
Table 2. Isolated yield of CE with mixed heteroatoms using optimized
reaction conditions. [a]
Entry
CE
Aryl di-nucleophile
+ bis-epoxide [b]
Isolated
yield (%)
Catalyst
Et3N, 10 mol%
Et3N, 10 mol%
Et3N, 10 mol%
Et3N, 25 mol%
Et3N, 25 mol%
Et3N, 25 mol%
Et3N, 30 mol%
Et3N, 30 mol%
Et3N, 30 mol%
1
2
3
4
5
6
7
8
9
2a
2b
2c
2d
2e
2f
a + 1a
a + 1b
a + 1c
b + 1a
b + 1b
b + 1c
c + 1a
c + 1b
c + 1c
82
85
78
56
63
58
86
93
90
2g
2h
2i
Figure 3. Distribution coefficients (D) of Pd2+ vs. other competing metals in CEs
with mixed heteroatoms (CHCl3 medium) and water.
Reactions were done using 10 mmol di-nucleophile (a-c) and 10 mmol bis-
epoxide (1a-1c) with 2 eq. LiCl in 200 mL H2O under N2 at 80oC for 24 h. [a]
Bis-epoxides: (1a) neopentyl diglycidyl ether, (1b) (2,2’-(2,3-dimethyl-
butane-2,3-diyl) bis (methylene) bis(oxirane), (1c) phenyldioxy-bis(2,3-
epoxypropane). [b]
Liquid-liquid extraction studies
Table 3. Liquid-liquid extraction of Pd2+ and Pt2+ using CEs with mixed
heteroatoms as extractant. [a]
2+
2+
2+ 2+
βPd
/Pt
CE
2a
2b
2c
2d
2e
2f
Log DPd
Log DPt
0.10±0.02
0.04±0.00
0.01±0.00
0.98±0.03
1.51±0.06
1.40±0.01
0.97±0.05
1.81±0.03
2.38±0.06
0.06±0.02
0.18±0.02
0.36±0.04
0.06±0.01
0.07±0.02
0.07±0.05
0.07±0.00
0.04±0.02
0.03±0.00
2
5 [b]
35 [b]
16
2+
Figure 4. Selectivity (Pd n+) of CEs with mixed heteroatoms towards Pd2+ in
/M
the presence of other competing metals.
22
concentrations of Pd and Pt to simulate these conditions and to
determine the applicability of the synthesized O, N, S-bearing CEs
as LLE extractants. Under this condition, Pd and Pt are in the form
25
14
2g
2h
2i
2-
of metal-chloro complexes (PdCl42- or PdCl3- for Pd2+ and PtCl6
or PtCl3- for Pt4+/2+) where majority are Pd2+ and Pt4+.[2-4] However,
thiacrown ethers form homoleptic complexes with Pd2+ and Pt2+.
53
93
[28-29]
Hence, it is likely that the LLE results reflect the
sequestration of Pd2+ and Pt2+, despite the latter being in minor
portion vs Pt4+. Results of LLE for binary Pd/Pt solutions (Table
3) reveal the preference of all CEs towards Pd2+ than Pt2+ except
for 2b and 2c. CE 2c showed the highest selectivity towards Pt2+
(βPt/Pd = 35) while 2i showed the highest preference towards Pd2+
(βPd/Pt = 93). Although, CE 2c showed selectivity towards Pt2+, the
stability of its CE-Pt2+ complex in the organic phase is very low
(Log DPt2+ = 0.36) due to the strong solvation of Pt2+ in acidic
medium. Higher ordered complexes between CE and Pt2+ might
be necessary in order to extract Pt2+ from the aqueous phase. [12]
Experimental conditions: Organic phase (5 mL CHCl3 containing [CE] = 1.5
× 10-3 M); Aqueous phase (5 mL 2 M HCl containing [Pd] = [Pt] = 1.5 × 10-4
2+ 2+ [b]
M). [a] Selectivity in terms of βPt
/Pd
.
A simple LLE experiment in CHCl3 was done to evaluate the
performance of the synthesized CEs to separate the PGMs.
Typical secondary PGM waste containing Pd and Pt as in the
[2, 4, 26, 27]
case of automotive catalyst leachates are very acidic.
Therefore, experiments were done in 2 M HCl acid with equimolar
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
On the other hand, remarkably higher distribution coefficients
carried out for a gas phase system where structures considered
in the study correspond to the local minima. Different
configurations (i.e., planar, folded, and cis- and trans-
configuration of the bulky (-R-) and rigid (-Ar) moieties) were
were observed for the Pd2+-selective CEs 2d-2i (Log DPd2+ = 0.97-
2.38). CEs containing S2 heteroatoms are shown to be efficient
[9, 10, 12]
extractants for Pd2+ in the presence of Pt2+.
Further
extraction studies were done in the presence of other PGMs (Ru3+, initially tested and showed that the planar configuration has the
Rh3+ and Ir3+) and metals (Sn4+, Te4+, Hf4+, and Au3+) (Figure 3).
Despite the addition of other competing metals, CE 2i still
lowest energy due to the minimum repulsive forces among the S,
N, and O atoms. Structure optimization of CE-Mn+ complexes
reveals that they form either of the two configurations: (1) Mn+ is
situated in the center of the CE or (2) outside the CE cavity.
Similar planar configurations of the free and CE-Mn+ complex
have also been reported in literature. [12, 13] Hence, the free CE
structure was optimized using the planar configuration, and the
lowest energy CE conformation was selected and reported. In
general, the structures of the most stable conformers of each CE
showed unique individual spatial geometry despite similar
structures except for the replacement of heteroatoms.
exhibited the highest distribution (DPd2+=1999 or Log DPd2+ = 3.30,
2+
SI Table S2) with selectivities as high as βPd n+=107 – 33,634
/M
(Figure 4). The selectivity of the synthesized CEs can be
attributed to their structures and cavity size relationships. [15, 28 – 31]
Theoretical studies
Density functional theory calculations were employed to
investigate how the structures of the synthesized CEs affected
their selectivity towards Pd2+ or Pt2+. Figure 5 shows the
optimized structures of the free CE using the ab initio Jaguar
program B3LYP with LACVP** basis set. All calculations were
Figure 5. Lowest energy configuration of free CEs using B3LYP with LACVP** basis set. Atom color scheme: grey = carbon, white = hydrogen, red = oxygen,
blue = nitrogen, yellow = sulfur. Dimensions are in Angstrom (Ǻ).
The cavity dimension was estimated based on the lowest
energy geometry configuration of the free CEs. In most cases, CE
cavity size is reported using Van der Waals distance of two
opposite oxygen heteroatoms estimated from the space-filling
CPK model. [7, 31, 32] However, a more reliable estimation can be
done using inter-atomic distances calculated using DFT optimized
structures of CEs. [8, 33] From previous report, the cavity size can
be defined as the free space at the center of the CE which can be
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
estimated from the distance of two opposite ethereal O atoms
subtracted by the effective radius of oxygen. [8] As far as it is
known, cavity size determination for O, N, S-bearing CEs has
never been reported in literature. Most of the reports on O-bearing
CE cavity sizes containing mixed heteroatoms have relied on the
CEs, 2i is the only CE that can accommodate Pd2+ (1.56 Å) while
the rest have incompatibly smaller sizes (Figure 6).
The cavity – size compatibility of 2i entraps Pd2+ more
effectively than other oxathia CEs (2g and 2h), preventing other
metals from binding. But despite their smaller cavities, minor
perching complex of Pd2+ with other CEs was still possible as
indicated by the LLE experiments (SI Table S2). Known
complexes of Pd2+ and Pt2+ form square planar coordination that
requires two or four S, respectively. Since O and N have no
coordination with Pd2+ and Pt2+, they could not form 1:1
complexation with CEs which do not contain S or has only one S
as in the case of oxo-aza and oxa-aza-thia CEs (2a-2f). Their
CE backbone available in the literature neglecting the effect of the
[34]
heteroatoms.
Therefore, following previous calculations, the
cavity sizes of the CEs with mixed heteroatoms were quantified
based on the distance between two opposite heteroatoms
subtracted by their effective radius of O (1.32 Å), N (1.39 Å) or S
(1.77 Å) atoms depending on the heteroatom combination in the
O, N, S-bearing CEs. [35]
interaction with 2a-2f is likely through multidendate sandwich type
[9, 12, 15]
complexation.
In case of dithia-CE 2g – 2i, 1:1 for CE-
Pd2+ complexes are easily formed which make them suitable Pd2+
extractants. The cavity size – match between 2i and Pd2+ might
be one of the reasons for its enhanced Pd2+ selectivity. Moreover,
supporting NMR results (SI Figure S2) reveal the significant
proton shifts of Hβ (Δδ=0.04-0.12) and ArHβ (Δδ=0.39-0.61),
which indicate coordination of S heteroatoms towards PdCl2 (SI
Table S4). Similar 1H NMR shifts were also observed in literature
for benzodithia-12-crown-4 ether with PdCl2. [12] New peaks were
also observed for protons near the S heteroatoms due to
conformational changes upon Pd2+ complex. Meanwhile, protons
near the O heteroatoms (Hτ and Hα) showed no changes in the
chemical shifts suggesting that O did not participate in Pd2+
coordination mainly due to its hard acid character. HOH proton
showed broadening of the peak due to hydrogen bonding with
absorbed water. The results affirm the HSAB principle wherein S
containing CEs like 2i, which contain soft acids, form strong
coordination with soft metals like Pd2+ while purely O-rich chelates
are hard acids that form strong complexations with hard alkali
metals. Thus, apart from cavity size match relationships, the
presence of S heteroatoms in 2i explain its effective and selective
coordination with Pd2+.
Figure 6. Estimated cavity size of free CE with mixed heteroatoms. Lowest
energy configuration of free CEs using B3LYP with LACVP** basis set. Cavity
sizes of CEs containing ethereal O are based on previous report. [8]
The estimated cavity size based on the DFT optimized
structures of the free CEs are summarized in SI Table S3. In
principle, 15 – membered CEs should have bigger cavity size in
comparison to 14 – membered rings. This holds true for CEs
containing –O–R–O– heteroatoms reported previously [8], wherein
a 15 – membered benzo-neopentyl- has the biggest cavity size in
comparison to 14-membered CEs, benzo-tetramethyl- and
dibenzo- CE derivatives. Their smaller cavity size is due to the
shorter bond distances of aromatic dibenzo ring in comparison to
bulky alkyl backbone such as in benzo-tetramethyl derivative. [8]
In the current work, oxaaza CEs (O, N-bearing) showed similar
Binding energy of CE-Mn+ complexes
The binding energy (ΔEMn+) calculations were limited to 1:1 CE-
Pd2+ or -Pt2+ complexes, excluding the effects of anions (Table
S3, SI Figure S2, and Figure 7). The structures of the free CEs
showed over-all changes in inter-atomic distances (C-C, C=C, C-
N, C-S and C-O bonds) during metal complexation following the
order, dioxa-diaza CEs (2a-2c), RSD 1.68% > dioxa-thia-aza (2d-
2f), RSD 1.21% > dioxa-dithia (2g-2i), RSD 0.88%. The negligible
changes for CEs 2g-2i structure upon complexation is due to their
relative bigger CE cavity size in comparison to the other CEs,
which shows a relative increased in X-O (X=N or S) distances.
Moreover, the ΔEMn+of the 2g-2i also showed the least energy
(Table S3). Although higher ordered complexes are possible,
especially for Pt2+ complexes, they were not included in the
calculations for simplicity. The optimized geometry of CE 2g-2i in
complex with Pd2+ and Pt2+ (Figure 7) showed structural changes
in comparison to their free forms (Figure 5). CE 2g-2i form a
square planar S4 geometry with Pd2+ involving two S along with
behavior but their cavity sizes (2a – 2c) are remarkably smaller
[8]
than previously reported O-bearing CEs.
Meanwhile,
oxaazathia CEs (O, N, S-bearing) showed almost similar cavity
sizes for both 14 – and 15 – membered rings (2d – 2f) although
2f was slightly larger than 2d and 2e. On the contrary, opposite
cavity size trend was observed for oxathia CEs (O, S-bearing),
wherein it increases in the following order: 2i > 2h > 2g. Despite
being a 14-membered ring, 2i still showed the largest cavity size
(1.58 Å) in comparison to other oxathia CEs, especially 2g. The
two chloride anions as described in previous studies on
cavity size enlargement of oxathia CEs may be attributed to the S
[12, 28, 29]
dithiacrown-Pd2+ complexation.
Hence, it is likely that
[15, 28]
– S repulsion and the longer S – C bonds.
Among all the
dithia-CE 2g-2i would require a sandwich type 2:1 CE-Pt2+
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
complex to satisfy their coordination requirement similar to other
dithiacrown ethers reported in literature. [28] On the other hand,
tetrathia-CE forms hexadentate complexes with Pt2+ with four S
atoms involved in square planar S4 coordination along with axial
S2 coordination of two chloride anions, resulting in a distorted
octahedron geometry. [9, 28, 36]
Among the dioxa-dithia-CEs, 2h showed the most remarkable
structural change upon complexation with Pd2+ or Pt2+ followed by
2g whereas 2i showed the least. This suggests that during Mn+
complexation, 2h and 2g required more energy to stabilize the
complex in comparison to 2i. Moreover, because of the slightly
smaller cavity size of 2h (1.29 – 1.44 Å) and 2g (1.19 – 1.44 Å)
than the ionic size of Pd2+ and Pt2+, the CEs pre-organized their
structures to encapsulate the metal ions and achieve stable
complexes. On the other hand, 2i which has a suitable cavity size
for Pd2+ or Pt2+ showed marginal structural change. Optimized
(Mn+– S and Mn+–O) distances of CE – Mn+ for selected CE – Mn+
complexes (Table 4) affirm the validity of the DFT method used
as the Pt2+–S (2.29-2.32 Å) and Pd2+–S (2.31-2.35 Å) distances
from single X-ray crystal structure of typical dithiacrown –Pt2+ or –
Pd2+ complexes are within ± 5% RSD. [9, 36] In contrast to Pd2+– or
Pt2+– complexed 2g and 2h (i.e. both have shorter Mn+– O than
their Mn+–S), complexed 2i-Pd2+ and 2i–Pt2+ have relatively longer
Mn+–O (2.76 Å – 2.92 Å) than their respective Mn+–S (2.28 Å –
2.32 Å) distances (Table 4). This affirms that S heteroatoms play
a critical role during complexation of 2i than the O heteroatoms
as also revealed in the 1H NMR complexation studies (SI Figure
S2). From Table 4, the calculated binding energy (ΔEMn+) of the
CE – Pd2+ complexes follow the order 2i–Pd2+ > 2g–Pd2+ > 2h-
Pd2+ while those of CE – Pt2+ complexes showed the trend 2h–
2+
Pt2+ > 2g– Pt2+ > 2i– Pt2+. CE 2i showed the lowest ΔEPd and
the highest ΔEPt2+, suggesting its stronger preference towards
Pd2+ than Pt2+. The ΔE result indicates its high affinity towards
Pd2+ than Pt2+ as also observed in LLE results (Table 3), which
2+
showed the highest selectivity (βPd 2+ = 93) of 2i towards Pd2+
/Pt
against Pt2+.
Figure 7. DFT optimized configuration of CE-Pd2+ (2g-Pd2+, 2h-Pd2+, and 2i-Pd2+) and CE-Pt2+ (2g-Pt2+, 2h-Pt2+, and 2i-Pt2+) complexes using B3LYP with
LACVP** basis set. Atom color scheme: grey = carbon, white = hydrogen, red = oxygen, yellow = sulfur, blue-green = Pd2+ and cream = Pt2+
.
The LLE results of 2i in the presence of other competing metals
still showed preference towards Pd2+ (Figure 3) despite its almost
similar ionic sizes with Pt2+, Rh3+, Ru3+ and Au3+ ØCE/ØMn+ = 0.98
Table 4. Selected DFT optimized Mn+– S and Mn+– O distances of CE – Mn+
complexes.
n+
CE-Mn+
complex
ΔEM
–
1.09 (Table 5). Typical coordination of thiacrown-Mn+
S – O (Å)
Mn+– O (Å)
Mn+– S (Å)
(eV)
complexes with other PGMs such as Rh3+ and Ru3+ favor a
distorted octahedral geometry involving two dithia-CEs similar to
CE-Pt2+ complex as discussed previously. [37] Meanwhile, Au3+
forms a distorted square planar S4 coordination similar to typical
CE-Pd2+ complex but interacts with four S heteroatoms. Therefore,
2g-Pd2+
2h-Pd2+
2i-Pd2+
2g-Pt2+
2h-Pt2+
2i-Pt2+
-19.32
-19.96
-21.23
-20.14
-19.96
-19.95
4.42-4.43
4.29-4.37
3.99-4.00
4.46-4.47
4.37-4.41
4.12-4.46
2.12±0.01
2.09±0.01
2.76±0.00
2.15±0.01
2.10±0.03
2.92±0.00
2.32±0.01
2.30±0.01
2.32±0.00
2.33±0.01
2.30±0.00
2.28±0.00
Au3+ will likely form a 2:1 sandwich type complex with dithia-CE
derivatives to complete its coordination requirement.
[38]
DFT
calculations of CE-Mn+ complexes were done to simulate the
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
structural changes in the CEs. The optimized geometry of
complexed 2i–Pt2+, 2i–Rh3+ and 2i–Ru3+ showed nesting complex
except for 2i–Au3+, which exhibited a perching complex (SI Figure
S4). This suggests that the Mn+ can fit inside the cavity of the CE
2i. Moreover, the ΔEMn+ (SI Table S5) especially for Rh3+ and Ru3+
Conclusions
Synthesis of crown ether derivatives with mixed heteroatoms (2a
– 2i) was successfully performed in water using bis-epoxide-ring
opening cyclization with o-phenylenediamine, 2-aminothiophenol
or 1,2-benzenedithiol as dinucleophiles under mild basic
2+
are relatively lower compared to ΔEPd of 2i, suggesting
spontaneity of the complex. However, LLE showed contrasting
results due to the nature of the Mn+ species in the acidic aqueous
solutions. To account for solvation changes in the binding
energies, a Poisson Boltzmann Finite element method was
implemented to calculate for the solvation binding energy
ΔES[CE-Mn+] for the 2g-Mn+ complexes. The calculated ΔES are
conditions. Highest cyclization yields (56
– 93%) for CE
derivatives were accomplished using 10, 25, and 30 mol% Et3N
in water when reacted with o-phenylenediamine, 2-
aminothiophenol or 1,2-benzenedithiol, respectively, using LiCl as
template. Experimental liquid-liquid extraction and DFT studies
revealed that among the CEs synthesized, 2i having ØCE=1.58 Å
showed the highest selectivity towards Pd2+ (ØPd2+=1.56 Å) even
in the presence of other competing PGMs (Pt2+, Ru3+, Rh3+ and
Ir3+) or metals (Sn4+, Te4+, Hf4+ and Au3+). The cavity size match
relationship, where the CE cavity size is the same as the metal
ionic diameter, as well as the presence of S heteroatoms play
major roles in synthesizing the PGM selective O,N,S-bearing CEs.
n+
significantly higher than the ΔEM due to the non-electrostatic
dispersion interaction of the Mn+ and CE-Mn+ with the aqueous
n+
solvent. Similar discrepancies in the ΔEM and ΔES have been
reported in the literature due to the effect of solvation of solutes
with the solvent. [39, 40] Despite the large difference, the calculated
n+
ΔES of CEs 2g-2i still follows the same trend of calculated ΔEM
(Table S6).
In order to quantitatively account the effects of the nature of the
Mn+, a simple Mn+ ligand exchange reaction can be applied as [CE
– Pd2+] + [Mn+– H2O] → [CE – Mn+] + [Pd2+– H2O]. The
endothermicity or exothermicity of the ligand exchange reaction
with respect to the CE – Pd2+ complex can be quantified using the
following equation: ΔH= ΔH[CE-Mn+] + ΔH[Pd2+-H2O] – ΔH[CE-
Pd2+] – ΔH[Mn+-H2O], where ΔH[CE-Mn+] is related to the DFT
calculated ΔES[CE-Mn+] with thermal corrections and ΔH[Mn+-
H2O] is the of the Mn+ chloride species. [8, 41]
Relative to 2i-Pd2+, positive ΔH values (Table 5) for all
competing Mn+ confirm the selectivity of 2i towards Pd2+ despite
their similar Mn+ ionic sizes. The theoretical selectivity of 2i based
on the enthalpy exchange reaction follows the order
Pd2+>Rh3+>Pt2+>Au3+>>Ru3+. Comparison with the experimental
LLE results in Figure 3 shows the order Pd2+ > Pt2+ > Au3+ >> Rh3+
> Ru3+. The disparity maybe attributed to the formation of higher
ordered complex (2:1 CE-Mn+) especially for Pt2+, which requires
tetra-dentate coordinating S heteroatoms. [9, 36] Nevertheless, the
theoretical binding studies and experimental LLE results both
demonstrated the high selectivity of 2i towards Pd2+ even in the
presence of other competing PGMs and metals.
Experimental Section
General reagents
Reaction solvents (tertiary butanol, t-BuOH; dimethyl sulfoxide, DMSO;
dimethylformamide, DMF; tetrahydrofuran, THF; diethyl ether, Et2O;
dichloromethane, DCM; and 1,4-dioxane), aromatic dinucleophiles (a o-
phenylenediamine, b 2-aminothiophenol, c 1,2-benzenedithiol), metal
hydroxides (LiOH, NaOH, KOH, Mg(OH)2 and Ca(OH)2), template metal
chlorides (LiCl, NaCl, KCl, MgCl2·6H2O, and CaCl2), catalysts
triethylamine (Et3N) and perchlorates (LiClO4, NaClO4, MgClO4,
Fe(ClO4)3), and PGM chloride salts (PtCl4 and PdCl2) were from either
Sigma-Aldrich (Mo, USA) or Fischer Scientific (Korea). Bis-epoxide 1a
(neopentyl glycol diglycidyl) is commercially available procured from
Sigma-Aldrich (Mo, USA) whereas 1b (2,2’-(2,3-dimethylbutane-2,3-
diyl)bis(methylene)bis(oxirane))
and
1c
(phenyldioxy-bis(2,3-
epoxypropane)) were synthesized from previously reported protocols as
described. [8]
Eluents (n-hexane; ethyl acetate, EtOAc; diethyl ether, Et2O; methanol,
MeOH; chloroform, CHCl3 and DCM) used for purification were HPLC
grade from Daejung Chemicals and Metals Co., Ltd (Korea). Simulated
solutions were dissolved in milli-Q deionized water (18.2 MΩ DI water).
RHM grade of HCl and HNO3 were procured from Daejung Chemicals and
Metals Co., Ltd (Korea). Prior reactions, THF and t-BuOH were dried over
4Å molecular sieves overnight, NaH (60% dispersion in oil) was washed
with n-pentane to remove the protecting oil, and m-CPBA (70-75%) was
dissolved in Et2O then washed with phosphate buffer (NaH2PO4/NaOH,
pH 7.4) to remove the stabilizer. All other reagents were used without
further purification.
Table 5. Enthalpy exchange reaction: [41]
[CE-Pd2+] + [Mn+-H2O] → [CE-Mn+] + [Pd2+-H2O];
ΔH= ΔH[CE-Mn+] + ΔH[Pd2+-H2O] – ΔH[CE-Pd2+] – ΔH[Mn+-H2O]
[a]
CE-Mn+
complex
ØCE
ØM
/
Mn+
specie
ΔH [a]
ΔES
ΔH [b]
ΔH [a]
n+
n+
n+
n+
M
-H2O
CE-M
CE-M
2-
2i-Pd2+
2i-Pt2+
2i-Rh3+
2i-Ru3+
2i-Au3+
1.04
1.09
1.00
0.98
0.98
-34.15 [42]
-81.58 -80.99
-37.65 -37.06
-106.67 -106.08
-138.05 -137.46
-31.38 -30.78
0
PdCl
4
2-
-40.28 [42]
-66.74 [42]
-219.1 [43]
-37.80 [44]
+50.06
+7.50
+128.5
+53.85
PtCl
6
General reaction monitoring and characterization
3-
RhCl
The reactions were monitored using thin layer chromatography, TLC
(Merck, silica gel 60 F254) and the products were purified using alumina
(Sigma, acidic Al2O3) and silica gel column chromatography (Merck, 0.035-
0.070 mm, 60Å). Structures of the compounds were confirmed by FTIR-
ATR (Varian Scimitar 2000, USA) as well as 1H and 13C NMR (400 MHz
Varian, USA) using deuterated chloroform (CDCl3) or dimethyl sulfoxide
(DMSO-d6). Molecular masses of the compounds were confirmed via HR-
6
3-
RuCl
6
3-
AuCl
6
n+
Unit: (Kcal/mol); ØM (Å) [35]: Pd2+=1.56, Pt2+=1.48, Rh3+=1.61, Ru3+=1.64,
Au3+=1.64. [a] ΔH[CE-Mn+] = ΔES[CE-Mn+] + ΔnRT. [b]
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
ESI-MS using Synapt G2 HDMS quadrupole time-of-flight (TOF) mass
spectrometer equipped with an electrospray ion source (ESI Waters,
Milford, USA) operated in positive ion mode (Korea Basic Science Institute,
Ochang Headquarter, Division of Bioconvergence Analysis, Korea).
Synthesis of 2a: (7,7-dimethyl-2,3,4,6,7,8,10,11,12,13-decahydro-1H-
benzo [1,5,9,12] dioxadiazacyclo-pentadecine-3,11-diol). Reaction of 1.08
g (10 mmol) a with 2.16 g (10 mmol) 1a using optimized catalyst/template
reaction conditions afforded 2.59 g (82%) 2a, a viscous brownish liquid
purified using column chromatography SiO2-Al2O3 (acidic) (10:1 v/v) eluted
by 1:15 (MeOH:CHCl3) Rf=0.70. IR (ATR) vmax/cm-1: 3524, 3339, 3052,
2953, 2865, 1598, 1512, 1453, 1359, 1257, 1095, 1051, 935, 847, 744,
663. 1H NMR (DMSO-d6, 400 MHz): (δ, ppm) 0.73-0.85 (6 H, m, –CH3);
2.87-3.27 (8 H, m, -CH2-O-, –CH2-NH-), 3.43-3.82 (6 H, m, -CH2-O-, -CH-
OH), 4.30-4.38 (2 H, m, -NH), 4.80-4.90 (2 H, m, -OH), 6.44-6.64 (4 H, m,
-ArH). 13C NMR (DMSO-d6, 100 MHz): (δ, ppm) 22.11, 22.21, 22.39, 22.53,
22.62, 36.82, 36.25, 36.66, 37.09, 44.89, 47.69, 49.79, 67.09, 67.71, 68.25,
70.76, 74.28, 75.71, 76.17, 77.47, 79.61, 110.28, 112.77, 114.79, 117.69,
118.99, 124.97, 137.52, 137.87 (SI Figure S5). HRMS (ESI-TOF), m/z
calculated for C17H28N2O4Na+: 347.1947 [M+Na]+; found: 347.1939 (SI
Figure S6).
Synthesis of bis-epoxide intermediates
Bis-epoxide 1a (neopentyl glycol diglycidyl) was commercially available.
Bis-epoxides 1b and 1c were synthesized as reported previously. [8] Briefly,
pinacol/THF solution (5.9 g, 50 mmol) was slowly added in a 200 mL
NaH/THF (150 mmol) suspension with agitation under N2 gas at room
temperature. Allyl bromide (15.1 g, 125 mmol) was then added slowly and
the mixture was refluxed for 24 h. The solution was quenched with DI water
and the crude product was extracted with EtOAc. The organic layer was
dried over MgSO4, filtered, concentrated, and purified using SiO2 gel
column chromatography eluted by n-hexane/EtOAc to isolate the diene
intermediate 2,3-bis(allyloxy)-2,3-dimethylbutane (8.4 g product at 85%
yield). The 20 mL diene/DCM (4.0 g, 20 mmol) solution was slowly added
to a stirred solution of chilled 80 mL m-CPBA/DCM (8.6 g, 50 mmol)
solution. After 24 h, the mixture was filtered and washed with 10% aqueous
Na2S2O3 then with saturated aqueous NaHCO3 solution. The organic
phase was dried over MgSO4, filtered, concentrated, and purified using
SiO2 gel column chromatography eluted by CHCl3 to afford 1b (4.2 g
product at 90% yield).
Synthesis of 2b: (6,6,7,7-tetramethyl-1,2,3,4,6,7,9,10, 11,12-deca-
hydrobenzo [1,4,8,11] dioxadiazacyclotetradecine-3,10-diol). Reaction of
1.08
g (10 mmol) a with 2.30 g (10 mmol) 1b using optimized
catalyst/template reaction conditions afforded 2.87 g (85%) 2b, a viscous
brownish liquid purified using column chromatography SiO2-Al2O3 (acidic)
(10:1 v/v) eluted by 1:15 (MeOH:CHCl3) Rf=0.73. IR (ATR) vmax/cm-1: 3525,
3343, 3029, 2985, 2909, 2870, 1600, 1536, 1496, 1455, 1367, 1387, 1263,
1218, 1143, 1050, 1006, 943, 851, 816, 743, 663, 578. 1H NMR (DMSO-
d6, 400 MHz): (δ, ppm) 1.09-1.14 (12 H, m, –CH3); 2.87-2.94 (2 H, m, –
CH2-NH-), 3.23-3.28 (2 H, m, –CH2-NH-), 3.34-3.41 (2 H, m, -CH2-O-),
3.54-3.61 (2 H, m, -CH2-O-), 3.61-3.80 (2 H, m, -CH-OH), 4.41-4.47 (2 H,
m, -NH), 4.89-4.93 (2 H, m, -OH), 6.36-6.42 (2 H, m, -ArH), 6.49-6.61 (2
H, m, -ArH). 13C NMR (DMSO-d6, 100 MHz): (δ, ppm) 19.31, 19.87, 20.17,
20.59, 50.82, 50.95, 67.62, 67.79, 68.03, 68.16, 79.60, 80.66, 80.89,
109.74, 109.96, 117.81, 136.50, 136.56. HRMS (ESI-TOF), m/z calculated
for C18H30N2O4Na+: 361.2103 [M+Na]+; found: 361.2097.
Catechol (5.5 g, 50 mmol), epichlorohydrin (18.5 g, 200 mmol) and 0.2
mL 10 N NaOH were stirred under N2 at 40oC for 48 h. The reaction was
cooled to room temperature and added with 21 mL of aqueous 5 N
NaOH/saturated N2CO3 solution and stirred vigorously for 20 h. The
aqueous layer was extracted with CHCl3 and the organic layer was dried,
concentrated and purified using SiO2 gel column chromatography eluted
by CHCl3 to afford 1c (9.4 g product at 85% yield).
Synthesis of CEs with (O, N, S) heteroatoms
Initial cyclization of 2a as model O, N,S-bearing CE
Synthesis of 2c: 6,7,8,9,14,15,16,17-octahydrodibenzo [1,4,8,11] di-
oxadiazacyclotetradecine-7,16-diol. Reaction of 1.08 g (10 mmol) a with
2.22 g (10 mmol) 1c using optimized catalyst/template reaction conditions
afforded 2.58 g (78%) 2c, a feathery light brown solid purified using column
chromatography SiO2-Al2O3 (acidic) (10:1 v/v) eluted by 1:15
(MeOH:CHCl3) Rf=0.67. IR (KBr) vmax/cm-1: 3526, 3311, 3006, 2931, 2868,
2831, 1594, 1501, 1451, 1319, 1252, 1216, 1116, 1037, 930, 835, 741,
651, 606, 578. 1H NMR (DMSO-d6, 400 MHz): (δ, ppm) 3.01-3.26 (4 H, m,
–CH2-NH-); 3.60-4.17 (6 H, m, -CH2-O-, -CH-OH); 4.54 (2 H, br s, -NH-);
5.08 (2 H, br s, -OH); 6.40-6.57 (4 H, m, -ArH); 6.79-6.96 (4 H, m, -ArH).
13C NMR (DMSO-d6, 100 MHz): (δ, ppm) 47.37, 48.58, 50.29, 50.62, 66.53,
66.71, 67.71, 69.15, 71.74, 73.12, 73.53, 79.62, 10920, 109.60, 111.47,
111. 76, 111.93, 114.51, 118.05, 121.56, 121.99, 136.99, 147.58, 148.96.
HRMS (ESI-TOF), m/z calculated for C18H22N2O4Na+: 353.1477 [M+Na]+;
found: 353.1469.
Initial cyclization reactions (SI Table S1) were done using CE 2a as model
compound in different solvent/catalyst/template systems. All reactions
were performed using 5 mmol a with 5 mmol 1a in 50 mL solvent under N2.
After reactions, two purification methods were adapted depending on the
type of reaction solvents used: (1) volatile solvents (t-BuOH, THF and
CH2Cl2, Et2O) were reduced in vacuum and the residues were re-dissolved
in CHCl3. The solutions were transferred in a separatory funnel and the
organic phase containing the crude product was washed with DI water.
The organic phase was dried over MgSO4, filtered, and reduced; (2) high
boiling point solvents (DMSO, DMF and 1,4-dioxane) including H2O were
added with CHCl3 and then washed several times with DI water. The
resulting organic phase was treated similarly to method (1). All crude
products were purified using column chromatography SiO2-Al2O3 (acidic)
(10:1 v/v) eluted by 1:15 (MeOH:CHCl3) Rf = 0.70 to afford 2a. Additional
optimization reactions in H2O as solvent were carried out by varying the
catalyst-template systems as detailed in Table 1.
Synthesis of 2d: 7,7-dimethyl-2,3,4,6,7,8,10,11,12,13-decahydro-
benzo [1,5,9,12] dioxathiaazacyclopentadecine-3,11-diol. Reaction of 1.25
g (10 mmol) b with 2.16 g (10 mmol) 1a using optimized catalyst/template
reaction conditions afforded 1.91 g (56%) 2d, a viscous yellow liquid
purified using column chromatography SiO2-Al2O3 (acidic) (10:1 v/v) eluted
by 1:15 (MeOH:CHCl3) Rf=0.35. IR (ATR) vmax/cm-1: 3617, 3393, 2955,
2866, 1568, 1471, 1443, 1358, 1251, 1221, 1108, 1040, 939, 832, 748,
662, 594. 1H NMR (DMSO-d6, 400 MHz): (δ, ppm) 0.72-0.81 (6 H, m, –
CH3); 2.62-2.66 (1 H, m, -CH2-NH-); 2.80-2.85 (1 H, m, -CH2-NH-); 2.97-
Aqueous cyclization of CEs with (O, N, S) heteroatoms
Reactions were done by slow addition of 10 mmol bis-epoxide (1a-1c) in
a stirred solution of 10 mmol dinucleophile (a-c) in 200 mL H2O under
nitrogen at 80oC for 24 h. The amounts of Et3N catalyst and metal template
were based on the optimized reaction conditions. Work-up method prior
purification is similar to 2a method (2). The CE derivatives was purified
using SiO2 or combination of SiO2-Al2O3 (acidic) column chromatography.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
3.15 (6 H, m, -CH2-O-, -CH2-S-); 3.20-3.61 (6 H, m, -CH2-O-, -CH-OH);
4.32 (1 H, br s, -NH-); 4.94-4.99 (2 H, m, -OH); 6.44-6.53 (1 H, m, -ArH);
6.56-6.67 (1 H, m, -ArH); 6.69-7.10 (1 H, m, -ArH); 7.20-7.30 (1 H, m, -
ArH). 13C NMR (DMSO-d6, 100 MHz): (δ, ppm) 22.11, 22.53, 36.36, 37.07,
38.84, 67.53, 68.68, 74.13, 74.51, 77.25, 77.37, 79.61, 110.38, 114.71,
116.98, 129.40, 135.06, 135.26, 149.14, 149.64 (SI Figure S7). HRMS
(ESI-TOF), m/z calculated for C17H27NSO4Na+: 364.1558 [M+Na]+; found:
364.1548 (SI Figure S8).
reaction conditions afforded 3.45 g (93%) 2h, a viscous clear liquid purified
using column chromatography SiO2 eluted by 1:20 (MeOH:CHCl3) Rf=0.79.
IR (ATR) vmax/cm-1: 3561, 3060, 2980, 2920, 2879, 1606, 1477, 1448, 1366,
1304, 1248, 1220, 1145, 1071, 1030, 934, 848, 809, 748, 665, 591. 1H
NMR (DMSO-d6, 400 MHz): (δ, ppm) 1.01-1.07 (12 H, m, –CH3); 2.81-3.07
(4 H, m, -CH2-S-); 3.11-3.61 (6 H, m, -CH2-O-, -CH-OH); 4.91-4.96 (2 H,
m, -OH); 7.06-7.21 (2 H, m, -ArH); 7.35-7.46 (2 H, m, -ArH). 13C NMR
(DMSO-d6, 100 MHz): (δ, ppm) 19.67, 20.34, 21.72, 23.41, 36.81, 38.34,
64.25, 66.34, 67.00, 68.61, 69.33, 70.18, 79.19, 79.61, 80.58, 124.06,
126.91, 127.41, 127.55, 129.68, 129.88, 131.94, 132.27, 134.49, 134.61,
136.96, 137.80 (SI Figure S11). HRMS (ESI-TOF), m/z calculated for
C18H28S2O4Na+: 395.1327 [M+Na]+; found: 395.1314 (SI Figure S12).
Synthesis of 2i: 6,7,8,15,16,17-hexahydrodibenzo [1,4,8,11] dioxa-
dithiacyclotetra decine-7,16-diol. Reaction of 1.42 g (10 mmol) c with 2.22
g (10 mmol) 1c using optimized catalyst/template reaction conditions
afforded 3.27 g (90%) 2i, a glassy clear solid purified using column
chromatography SiO2 eluted by 1:20 (MeOH:CHCl3) Rf=0.65. IR (KBr)
vmax/cm-1: 3619, 3342, 3058, 3007, 2932,2870, 1587, 1499, 1450, 1313,
1252, 1213, 1116, 1073, 1030, 931, 836, 742, 665, 652, 592. 1H NMR
(DMSO-d6, 400 MHz): (δ, ppm) 2.97-3.07 (2 H, m, –CH2-S-), 3.18-3.21 (2
H, m, –CH2-S-), 3.96 (6 H, b, –CH2-O-, -CH-OH), 5.33-5.36 (2 H, m, -OH),
6.82-6.84 (2 H, m, -ArH), 6.91-6.92 (2 H, m, -ArH), 7.04-7.11 (2 H, m, -
ArH), 7.27-7.33 (2 H, m, -ArH). 13C NMR (DMSO-d6, 100 MHz): (δ, ppm)
38.66, 49.05, 68.11, 72.11, 79.61, 114.73, 121.73, 126.37, 127.93, 136.39,
148.82 (SI Figure S13). HRMS (ESI-TOF), m/z calculated for
C18H20S2O4Na+: 387.0701 [M+Na]+; found: 387.0692 (SI Figure S14).
Synthesis of 2e: 6,6,7,7-tetramethyl-3,4,6,7,9,10, 11,12-octahydro-2H
benzo [1,4,8,11] dioxathiaaza cyclotetra-decine-3,10-diol. Reaction of 1.25
g (10 mmol) b with 2.30 g (10 mmol) 1b using optimized catalyst/template
reaction conditions afforded 2.23 g (63%) 2e, a viscous yellow liquid
purified using column chromatography SiO2-Al2O3 (acidic) (10:1 v/v) eluted
by 1:15 (MeOH:CHCl3) Rf=0.42. IR (ATR) vmax/cm-1: 3525, 3396, 3052,
2982, 2922, 2871, 1567, 1442, 1367, 1219, 1146, 1069, 1030, 932, 854,
749, 662, 598. 1H NMR (DMSO-d6, 400 MHz): (δ, ppm) 0.96-1.06 (6 H, m,
2 –CH3); 2.61-2.67 (2 H, m, -CH2-NH-); 2.83-2.89 (2 H, m, -CH2-S-); 3.21-
3.43 (4 H, m, -CH2-O-); 3.49-3.69 (2 H, m, -CH-OH); 4.61 (1 H, br s, -NH-);
5.11 (2 H, m, -OH); 6.44-6.46 (1 H, m, -ArH); 6.65-6.67 (1 H, m, -ArH);
6.94-6.98 (1 H, m, -ArH); 7.19-7.22 (1 H, m, -ArH). 13C NMR (DMSO-d6,
100 MHz): (δ, ppm) 20.27, 20.33, 39.03, 64.89, 69.51, 79.07, 79.61,
114.89, 116.85, 117.14, 129.33, 134.92, 149.58. HRMS (ESI-TOF), m/z
calculated for C18H29NSO4Na+: 378.1715 [M+Na]+; found: 378.1693.
Synthesis of 2f: 7,8,14,15,16,17-hexahydro-6H-dibenzo [1,4,8,11]
dioxathiaaza cyclotetradecine-7,16-diol. Reaction of 1.25 g (10 mmol) b
with 2.22 g (10 mmol) 1c using optimized catalyst/template reaction
conditions afforded 2.01 g (58%) 2f, a yellow solid purified using column
chromatography SiO2-Al2O3 (acidic) (10:1 v/v) eluted by 1:15
(MeOH:CHCl3) Rf=0.29. IR (KBr) vmax/cm-1: 3618, 3394, 3055, 3001, 2922,
2871, 1590, 1501, 1447, 1328, 1252, 1215, 1116, 1073, 1026, 917, 835,
742, 662, 596. 1H NMR (DMSO-d6, 400 MHz): (δ, ppm) 2.72-2.95 (2 H, m,
-CH2-NH-); 2.97-3.42 (2 H, m, -CH2-S-); 3.62-4.02 (4 H, m, -CH2-O-, -CH-
OH); 4.60-4.63 (1 H, m, -NH-); 5.11-5.21 (2 H, m, -OH); 6.50-6.53 (1 H, m,
-ArH); 6.66-6.70 (1 H, m, -ArH); 6.81-7.16 (5 H, m, -ArH); 7.17-7.36 (1 H,
m, -ArH). 13C NMR (DMSO-d6, 100 MHz): (δ, ppm) 38.17, 38.66, 46.63,
51.61, 68.06, 68.35, 70.48, 71.27, 72.27, 79.62, 114.57, 115.47, 116.65,
116.94, 117.34, 119.10, 121.77, 127.52, 129.49, 129.97, 131.70, 135.11,
148.97. HRMS (ESI-TOF), m/z calculated for C18H21NSO4Na+: 370.1089
[M+Na]+; found: 370.1079.
Liquid-liquid extraction procedure
All vessels used were acid-washed in 5% HNO3 for 24 h followed by
intensive washing with DI water. Before experiments, CHCl3 was
equilibrated with DI water to ensure constant extractant phase volume
before and after LLE. Simulated Pd and Pt solutions or multi-component
ICP standard solutions were diluted in 2M HCl as the aqueous phase. For
the LLE experiments, Pt or Pd solutions with volume Vaq were contacted
with the CE-containing extractants with volume Vorg at Vaq/Vorg = 1 ratio.
Metal analyses were performed using 1 mL aqueous samples collected
before (Ci) and after (Cf at equilibrium) LLE. The samples were digested
using MARS-5 microwave unit (CEM, USA) and analyzed using inductively
coupled plasma – mass spectrometry (ICP-MS Agilent 7500 series, USA).
Extraction results were calculated using Eqs. (1) and (2) as metal
distribution coefficients (DMn+) and metal selectivity (β), respectively.
Synthesis of 2g: 7,7-dimethyl-3,4,6,7,8,10,11,12-octahydro-2H-benzo
[1,5,9,12] dioxadithiacyclopentadecine-3,11-diol. Reaction of 1.42 g (10
mmol) c with 2.16 g (10 mmol) 1a using optimized catalyst/template
reaction conditions afforded 3.08 g (86%) 2g, a viscous clear liquid purified
using column chromatography SiO2 eluted by 1:20 (MeOH:CHCl3) Rf=0.76.
IR (ATR) vmax/cm-1: 3523, 3057, 2954, 2867, 1588, 1502, 1478, 1450, 1359,
1309, 1283, 1220, 1105, 1042, 748, 648,663, 579. 1H NMR (DMSO-d6,
400 MHz): (δ, ppm) 0.67-0.72 (6 H, m, –CH3); 2.86-3.11 (8 H, m, –CH2-S-,
-CH2-O-), 3.22-3.34 (6 H, m, –CH2-O-), 3.71-3.75 (2 H, m, -CH-OH), 5.02
(2 H, br s, -OH), 7.14-7.20 (2 H, m, -ArH), 7.38-7.45 (2 H, m, -ArH). 13C
NMR (DMSO-d6, 100 MHz): (δ, ppm) 22.44, 22.47, 36.20, 37.01, 37.82,
67.66, 68.96, 73.94, 74.28, 75.27, 75.38, 79.61, 127.38, 127.44, 131.06,
131.25, 137.36, 137.86 (SI Figure S9). HRMS (ESI-TOF), m/z calculated
for C17H26S2O4Na+: 381.1170 [M+Na]+; found: 381.1162 (SI Figure S10).
Synthesis of 2h: 6,6,7,7-tetramethyl-2,3,4,6,7,9,10,11-octahydro-
benzo[i] [1,4,8,11] dioxadithiacyclotetradecine-3,10-diol. Reaction of 1.42
g (10 mmol) c with 2.30 g (10 mmol) 1b using optimized catalyst/template





Eq. (1)
Eq. (2)
Ci  Cf Vaq

DM n


Cf
Vorg

D
2
2
Pd _or_ Pt
( Pd 2

_or_ Pt 2 )/ M n
D
n
M
Quantum chemical calculation details
All calculations were performed using Schrodinger Materials Science Suite
version 2015-4 with Jaguar ab initio electronic structure program
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
[18] V. Guyon, A. Guy, J. Foos, M. Lemaire, M. Draye, Tetrahedron, 1995,
51, 4065-4074.
[19] S. Bonollo, D. Lanari, L. Vaccaro, Eur. J. Org. Chem., 2011, 14, 2587–
[45, 46]
(Schrodinger Inc., Portland, OR, USA).
Geometric optimizations
were performed for multiple initial configurations of the free CEs. The
B3LYP hybrid density functional was used with the LACVP** basis set in
which the non-ECP atoms (H-Ar) were described with 6-31g** basis
function while the LACVP effective core potential was used for heavier
atoms. The binding energy (ΔECE-Mn+) was calculated using Eq. (3), where
2598.
[20] J. Wu, H.G. Xia, Green Chem., 2005, 7, 708-710.
[21] S. Bonollo, F. Fringuelli, F. Pizzo, L. Vaccaro, Green Chem., 2006, 8,
960-964.
[22] N. Azizi, B. Mirmashhori, M.R. Saidi, Catalysis Commun., 2007, 8, 2198-
2203.
ECE-Mn+, ECE, and EM are the total energies of crown ether-cation
n+
[23] R.E.C. Torrejos, G.M. Nisola, A. B. Beltran, M.J. Park, B. R. Patil, S.P.
Lee, J.G. Seo, W.J. Chung, Lett. Org. Chem., 2014, 11, 109-115.
[24] S.N. Dmitrieva, N.I. Sidorenko, A.I. Vedernikov, N.A. Kurchavov, L.G.
Kuz’mina, T.M. Buslaeva, S.S. Basok, A.K. Buryak, J.A. Howard, S.P.
Gromov, Mendeleev Commun., 2009, 19, 21-23.
[25] L.A. Limjuco, G.M. Nisola, R.E.C. Torrejos, J.W. Han, H.S. Song, K.J.
Parohinog, S. Koo, S.P. Lee, W.J. Chung, ACS Appl. Mater. Inter., 2017,
9, 42862-42874.
complexes, crown ether, and cation, respectively. To account for the
solvation,
a Poisson Boltzmann Finite (PBF) element method was
implemented to calculate for ΔEs[CE-Mn+] for selected CE-Mn+ complexes.
The calculation method is appropriate to describe the model according to
previous reports. [47-49]
[26] B.S. Kim, J.C. Lee, S.P. Seo, Y.K. Park, H.Y. Sohn, J. Mater. Metall. Eng.,
2004, 56, 55-58.
[27] Y.S. Kwon, J.C. Lee, D.Y. Shin, S.H. Yi, H.J. Kim, Y.G. Choi, J. Korean
Inst. Resources Recycl., 2014, 23, 3-16.
[28] G.J. Grant, D.F. Galas, M.W. Jones, K.D. Loveday, W.T. Pennington,
G.L. Schimek, C.T. Eagle, D.G. VanDerveer, Inorg. Chem., 1998, 37,
5299-5305.
ECEM n   ECEM n   ECE  EM n 
Eq. (3)
[29] G.J. Grant, N.J. Spangler, W.N. Setzer, D.G. VanDerveer, L.F. Mehne,
Inorg. Chim. Acta, 1996, 246, 31-40.
[30] S.H. Hausner, C.A.F. Striley, J.A. Krause-Bauer, H. Zimmer, J. Org.
Chem., 2005, 70, 5804-5817.
Acknowledgements
[31] U. Olsher, R.M. Izatt, J.S. Bradshaw, Chem. Rev., 1991, 91, 137-164.
[32] J.W. Steed, Coord. Chem. Rev., 2001, 215, 171-221.
[33] A. Boda, Sk. M. Ali, M.R.K. Shenoi, H. Rao, S.K. Ghosh, J. Mol. Model,
2011, 17, 1091-1108.
[34] D. Siswanta, K. Nagatsuka, H. Yamada, K. Kumakura, H. Hisamoto, Y.
Shichi, K. Toshima, K. Suzuki, Anal. Chem., 1996, 68, 4166-4172.
[35] R.D. Shannon, Acta Cryst., 1976, A32, 751-767.
[36] G.J. Grant, W. Chen, A.M. Goforth, C.L. Baucom, K. Patel, P. Repovic,
D.G. VanDerveer, W.T. Pennington, Eur. J. Inorg. Chem., 2005, 3, 479-
485.
This research was supported by the National Research
Foundation of Korea (NRF) under the Ministry of Education
(No. 2017R1D1A1B03028102 and No. 2009-0093816) and the
Ministry of Science and ICT (No. 2017R1A2B2002109).
Conflict of Interest
[37] N. Goswami, R. Alberto, C.L. Barnes, S. Jurisson, Inorg. Chem., 1996,
There are no conflicts to declare.
35, 7546-75555.
[38] D. Huang, X. Zhang, E.J.L. McInnes, J. MacMaster, A.J. Blake, E. S.
Davies, J. Wolowska, C. Wilson, M.Schroder, Inorg. Chem., 2008, 47,
9919-9929.
[39] J. Cooper, T. Ziegler, Inorg. Chem., 2002, 41, 6614-6622.
[40] S. del Piero, R. Fedele, A. Melchior, R. Portanova, M. Tolazzi, E.
Zangrando, Inorg. Chem., 2007, 46, 4683-4691.
Keywords: binding • cyclization reactions • crown ethers •
epoxide ring opening • platinum-group metals
[41] E.D. Glendening, D. Feller, M.A. Thompson, J. Am. Chem. Soc., 1994,
11, 10657-10669.
[42] A. Lienke, G. Klatt, D.J. Robinson, K.R. Koch, K.J. Naidoo, Inorg. Chem.,
2001, 40, 2352-2357.
[43] R. Ayala, M. Sprik, J. Chem. Theory Comput., 2006, 2, 1403-1415.
[44] M. Hargittai, A. Schulz, B Reffy, M. Kolonits, J. Am. Chem. Soc., 2001,
123, 1449-1458.
[45] A. D. Bochevarov, E. Harder, T. F. Hughes, J. R. Greenwood, D. A.
Braden, D. M. Philipp, D. Rinaldo, M. D. Halls, J. Zhang, R. A. Friesner,
Int. J. Quantum Chem., 2013, 113, 2110-2142.
[46] Jaguar, Schrödinger, LLC, New York, NY, 2015.
[47] C. Liu, D. Walter, D. Neuhauser, R. Baer, J. Am. Chem. Soc., 2003, 125,
13936–13937.
[48] J. C. Slater, J.C. Philipps, Physics Today, 1974, 27, 49.
[49] F. Zhuravlev, J.A. Gladysz, Chem. Euro. J., 2004, 10, 6510–6522.
[1]
[2]
[3]
[4]
R.M. Izatt, S.R. Izatt, N. E. Izatt, K. Krakowiak, R.L. Bruening, L. Navarro,
Green Chem., 2015, 17, 2236-2245.
F. L. Bernardis, R.A. Grant, D. C. Sherrington, React. Funct. Polym.,
2005, 65, 205-217.
M. Yamada, M. R. Gandhi, Y. Kaneta, N. K. Kimura, H. Katagiri , Ind.
Eng. Chem. Res., 2018, 57, 1361-1369.
M. K. Jha, J.C. Lee, M.S. Kim, J. Jeong, B.S. Kim, V. Kumar, J. Jeong,
Hydrometallurgy, 2014, 142, 23-32.
B. Pospiech, Physicochem. Probl. Miner. Process, 2012, 48, 239-246.
C.H. Kim, S.I. Woo, S.H. Jeon, Ind. Eng. Chem. Res., 2000, 39, 1185-
1192.
C.J. Pedersen, J. Am. Chem. Soc., 1967, 89, 7017-7036.
R.E.C. Torrejos, G.M., Nisola, H.S. Song, L.A. Limjuco, C.P. Lawagon,
K.J. Parohinog, S. Koo, J.W. Han, W.J. Chung, Chem. Eng. J., 2017,
326, 921-933.
[5]
[6]
[7]
[8]
[9]
A.J. Blake, G. Reid, M. Schroder, J. Chem. Soc. Dalton Trans, 1990,
3363-3373.
[10] R.L. Bruening, B.J. Tarbet, Anal. Chem., 1991, 63, 1014-107.
[11] D. Waknine, M.J. Heeg, J.F. Endicott, L.A., Inorg. Chem., 1991, 30,
3691-3700.
[12] S.N. Dmitrieva, N.I. Sidorenko, N.A. Kurchavov, A.I. Vedernikov, A.Y.
Freidzon, L.G. Kuz’mina, A.K. Buryak, T.M. Buslaeva, A.A.
Bagatur’yants, Y. A. Strelenko, J.A. Howard, S.P. Gromov, Inorg. Chem.
2011, 50, 7500-7510.
[13] H. Nagai, Y. Suzaki, K. Osakada, Eur. J. Inorg. Chem., 2014, 26, 4376-
4384.
[14] J.W. Sibert, P.B. Forshee, V. Lynch, Inorg. Chem., 2005, 44, 8602-8609.
[15] R.E. Wolf Jr., J.A.R. Hartman, J.M.E. Storey, B.M. Foxman, S.R. Cooper,
J. Am. Chem. Soc., 1987, 109, 4328-4335.
[16] H.J. Kuppers, K. Wieghardt, B. Nuber, J. Weiss, E. Bill, A. Trautwein,
Inorg, Chem., 1987, 26, 3762-3769.
[17] M.W. Glenny, L.G.A. van de Water, J.M. Vere, A.J. Blake, C. Wilson, W.L.
Driessen, J. Reedijk, M. Schroder, Polyhedron, 2006, 25, 599-612.
This article is protected by copyright. All rights reserved.

10.1002/cplu.201800541
ChemPlusChem
FULL PAPER
FULL PAPER
Aqueous crown ether synthesis:
Novel CEs with mixed O, N, S
heteroatoms were successfully
prepared in water catalysed by
triethylamine with LiCl template.
Certain CEs were highly selective
towards certain PGMs like Pd2+ as
elucidated by experimental LLE and
DFT studies.
Rey Eliseo C. Torrejos,[a] Grace M.
Nisola,[a] Sang Hoon Min,[b] Jeong Woo
Han,[c] Sangho Koo,[a],[d] Chosel P.
Lawagon,[a] Khino J. Parohinog,[a]
Seongpoong Lee,[a] Hern Kim,* [a] Wook-
Jin Chung,*[a]
Page XX – Page XX
Aqueous Synthesis of 14-15
Membered Crown Ethers with Mixed
O, N and S Heteroatoms:
Experimental and Theoretical Binding
Studies with Platinum-Group Metals
This article is protected by copyright. All rights reserved.