Mononuclear copper(I) complexes of O-t-butyl-1,1-dithiooxalate and of O-t-butyl-1-perthio-1-thiooxalate

Article abstract of DOI:10.1016/j.ica.2011.02.037

Described are the syntheses and structures of a phosphonium salt of the anionic ligand O-t-butyl-1,1-dithiooxalate, [PPh₃Bz][i-dto ^tBu] ([PPh₃Bz][1]), and of two Cu(I) complexes of this anion, Cu(PPh₃)₂

Full text of DOI:10.1016/j.ica.2011.02.037

Inorganica Chimica Acta 374 (2011) 261–268
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Mononuclear copper(I) complexes of O-t-butyl-1,1-dithiooxalate
and of O-t-butyl-1-perthio-1-thiooxalate
David F. Zigler a, Elisa Tordin b, Guang Wu a, Alexei Iretskii c, Elena Cariati b, Peter C. Ford a,
⇑
a Department of Chemistry and Biochemistry, University of California-Santa Barbara, Santa Barbara, CA 93106-9510, USA
b Dipatimento di Chimica Inorganica Metallorganica e Analitica ‘‘Lamberto Malatesta’’, Università di Milano and Centro CNR-ISTM, Via Venezian 21, 20133 Milano, Italy
c Department of Chemistry and Environmental Sciences, Lake Superior State University, Sault Ste. Marie, MI 49783, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 19 February 2011
Described are the syntheses and structures of a phosphonium salt of the anionic ligand O-t-butyl-1,1-
dithiooxalate, [PPh3Bz][i-dtotBu] ([PPh3Bz][1]), and of two Cu(I) complexes of this anion, Cu(PPh3)2(g2
-
i-dtotBu) (2) and Cu(dmp)(PPh3)( 1-i-dtotBu) (3, dmp = 2,9-dimethyl-1,10-phenanthroline). In addition,
g
Dedicated to Prof. W. Kaim.
it was found that the reaction of CuBr2 with i-dtotBuꢀ gives a O-t-butyl-1-perthio-1-thiooxalato complex
of copper(I), [BzPh3P][Cu(Br)(S-i-dtotBu)] ([BzPh3P][4]), where [S-i-dtotBu]ꢀ is a disulfide-containing
anionic ligand. The electronic structure and absorption spectrum of this species were investigated by
time dependent DFT methods.
Keywords:
Copper complexes
X-ray crystal structure
Thiol ligands
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
needed over 2 h. Precipitated [NHEt3]Cl was separated by vacuum
filtration and rinsed with 2ꢁ 5 mL of DMF. The filtrate was added
Interest in exploring the reactions of photochemically non-
innocent ligands [1] has led us to initiate studies of complexes
based on derivatives of 1,1-dithiooxalate [2–7], specifically O-t-
butyl-1,1-dithiooxalate, [i-dtotBu]ꢀ (1) (Fig. 1). Described here
are the preparations and structures of two copper(I) complexes
to 11.7 g of [BzPh3P]Cl (30.0 mmol) dissolved in 30 mL of MeOH.
Crude [BzPh3P][1] was precipitated by slow addition of 80 mL of
water over ca. 50 min. Following filtration, excess DMF and water
were removed by suspending the crude plug of [BzPh3P][1] in
200 mL of diethyl ether and filtering. The salt [BzPh3P][1], recrystal-
lized twice from CH2Cl2/i-PrOH, was isolated as 5.24 g of dark
orange crystals (9.87 mmol, 33% yield). [BzPh3P][1] is soluble in
CH2Cl2, and CHCl3 and is somewhat soluble in MeOH, EtOH and
in mixed solvents (e.g., 1:2 CH2Cl2/EtOH). 1H NMR (CD2Cl2):
of 1, Cu(PPh3)2(i-dtotBu) (2) and Cu(dmp)(PPh3)( 1-i-dtotBu) (3,
g
dmp = 2,9-dimethyl-1,10-phenanthroline). In the course of prepar-
ing such complexes, we found that the reaction of 1 with CuBr2
gives a complex with a bidentate perthiocarboxylato ligand, and
the structure and electronic properties of this new complex salt
[BzPh3P][Cu(Br)(S-i-dtotBu)] ([BzPh3P][4], S-i-dtotBuꢀ = O-t-butyl-
1-perthio-1-thiooxalato) are described.
2
d = 6.98–7.82 (20H, m, Ar), 4.89 (2H, d, JP–H = 14.2), 1.41 (9H, s)
ppm. 13C NMR (CD2Cl2): d = 246.3 (CS2), 170.4 (CO2), 127–136 (mul-
ti. CAr), 117.6 (P–CH2–Ph, 1JCP = 88 Hz), 80.1 (C Bu), 28.1 (Me Bu) ppm.
UV–Vis (MeOH): kmax = 343 (14 600 Mꢀ1 cmꢀ1), 267 nm (4280
t
t
Mꢀ1 cmꢀ1). FTIR (KBr):
v = 2986(w), 2921(w), 2881(w), 3051(w),
2. Experimental
3017(w), 1691(C@O, vs), 1586(m), 1483(m), 1455(m), 1438(s),
1366(m), 1267(m), 1253(m), 1168(s), 1154(m), 1110(s), 1043(vs),
1001(m), 995(m), 850(m), 835(m), 803(m), 780(m), 752(m),
747(s), 718(m), 697(s), 689(s), 578(m), 512(s), 505(s), 495(s)
cmꢀ1. See Supporting information for key spectra.
X-ray quality single crystals were grown by slow diffusion of
i-PrOH into a concentrated CH2Cl2 solution of [BzPh3P][i-dtotBu]
at ꢀ20 °C.
2.1. [BzPh3P][i-dtotBu] ([BzPh3P][1])
The anionic ester O-t-butyl-1,1-dithiooxalate (i-dtotBuꢀ, 1) was
synthesized as its benzyltriphenylphosphonium salt using a proce-
dure adapted from Strauch et al. [2]. Sulfur (1.92 g, 60.0 mmol) and
12.5 mL of triethylamine (90.0 mmol) was stirred in 30 mL of deox-
ygenated dimethylformamide for ca. 50 min. O-t-Butyl-2-chloro-
acetate (4.52 g, 30.0 mmol) was added in one portion to the dark
green solution, turning it deep orange. The temperature of the reac-
tion was kept at 20–30 °C using a water bath and adding ice as
2.2. Cu(PPh3)2(i-dtotBu) (2)
⇑
This was prepared following procedures adapted from Strauch
et al. [2]. In a 100 mL beaker, 0.89 g of CuCl(PPh3)3 (1.0 mmol)
Corresponding author.
E-mail address: ford@chem.ucsb.edu (P.C. Ford).
0020-1693/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2011.02.037

262
D.F. Zigler et al. / Inorganica Chimica Acta 374 (2011) 261–268
minimal amount of water (ca. 3 mL). The aqueous phase was
removed after 30 min of stirring and the organic layer was concen-
trated in vacuo. The resulting violet solid was washed with 3ꢁ 1 mL
of THF and gravity filtered through a glass wool packed Pasture pip-
ette to extract the complex salt. Narrow red plates obtained from
slow diffusion of diethyl ether into the combined THF washings
were filtered, washed with Et2O and recrystallized to give 0.16 g
of [BzPh3P][4]ꢂ0.5THF (0.21 mmol, 60% yield). Calc. for CuC33H35
-
BrO2.5PS3: C, 53.4; H, 4.75. Found: C, 53.5; H, 4.45%. ESI(ꢀ)-MS
(CH3CN or THF): m/z (rel. peak height) = 350.81(77), 351.83(9.6),
352.81(100), 353.81(13), 354.81(51), 355.83(5.7), 356.82(7.1); the-
oretical for CuC6H9BrO2S3: m/z (rel. peak height) = 350.82(64),
351.83(4.3), 352.82(100), 353.83(6.1), 354.82(41), 355.82(3.6),
356.82(4.1). FTIR (KBr):
v = 2980(m), 2918(m), 2836 (m), 3056(w),
1717(b,s), 1586(m), 1492(m), 1482(m), 1454(m), 1436(s), 1392(m),
1368(m), 1255(b,s), 1151(b,s), 1111(s), 1074(b,s), 1031(w), 996(m),
921(w), 866(w), 836(m), 787(m), 743(m), 717(m), 700(m), 688(s),
582(m), 558(m), 510(s), 499(m), 446(w). UV–Vis (THF): kmax
(e
) = 358 (3730 Mꢀ1 cmꢀ1), 512 (3420 Mꢀ1 cmꢀ1), 720 (shoulder,
Fig. 1. Ligand and complexes described in this study.
27 Mꢀ1 cmꢀ1) nm. See Supporting information for key spectra.
X-ray quality crystals were obtained from slow vapor diffusion
of Et2O into a THF solution of [BzPh3P][4] at room temperature.
was dissolved in 20 mL of CH2Cl2 and 0.53 g of [BzPh3P][1]
(1.0 mmol) in 30 mL of CH2Cl2 was added dropwise over 5 min.
The dark reaction mixture was covered with a watch glass and stir-
red in the dark for 1 h. The solvent was removed under reduced
pressure and the residue was recrystallized twice from CH2Cl2/i-
PrOH at ꢀ20 °C. Pale brown needles of 2 crumbled upon drying
to give 0.64 g (0.84 mmol, 84%) of brown powder. 1H NMR
(CD2Cl2): d = 7.1–7.4 (30H, m), 1.43 ppm (9H, s). FTIR (KBr):
2.5. Instruments
Electronic absorption spectra were measured in spectroscopic
grade solvent using a Varian-Cary UV-2401PC double beam UV–
Vis spectrophotometer. Infrared spectra were recorded for samples
in KBr pellets using a Mattson FTIR at 2 cmꢀ1 resolution and aver-
aging 32 scans. NMR spectra (1H- and 13C) were measured in deu-
terated solvent on either a 200 MHz Varian MERCURY Vx or a
500 MHz Varian UNITY INNOVA instrument. Mass spectrometry
was performed either with a Micromass Q-ToF using a standard
electrospray source (acetonitrile) or with a modified Q-ToF instru-
ment using a nanospray source (THF) [8]. Source, cone and impact
chamber potentials were optimized to decrease ion fragmentation.
Theoretical mass spectra were determined for specific ions using
the isotopic distribution calculator in the ChemDraw Ultra Suite
of programs.
v
= 2982(w), 2935(w), 3054(w), 1703(C@O, vs), 1478(s), 1432(vs),
1389(m), 1365(m), 1253(b,vs), 1156(b,vs), 1092(s), 1077(s),
1052(s), 1018(s), 997(m), 843(m), 811(m), 744(vs), 694(vs),
527(m), 516(s), 501(s) cmꢀ1. UV–Vis (CH2Cl2): kmax
(e) = 400 (sh,
3000 Mꢀ1 cmꢀ1) 364 (5230 Mꢀ1 cmꢀ1), 265 nm (23 100 Mꢀ1 cmꢀ1).
See Supporting information for key spectra.
X-ray quality crystals of Cu(PPh3)2(i-dtotBu) were isolated after
slow diffusion of i-PrOH into a concentrated solution of 2 in CH2Cl2
at room temperature.
2.3. Cu(dmp)(PPh3)(g
1-i-dtotBu) (3)
A deoxygenated solution of 2 (0.20 g, 0.26 mmol) in minimal
CH2Cl2 was charged with 0.055 g of 2,9-dimethyl-1,10-phenanthro-
line (0.26 mmol), also in a minimal amount of deaerated CH2Cl2.
Following 3 days at 5 °C, most of the solvent was removed under
flowing argon. Diethyl ether was layered on the reaction concen-
trate to give small orange crystals after 3 days at ꢀ20 °C. Following
filtration and drying, the crystals crumbled giving 0.11 g of air-
sensitive, bright yellow microcrystals (0.16 mmol, 62% yield). Calc.
for CuC38H36N2O2PS2: C, 64.2; H, 5.10; N, 3.94. Found: C, 63.6; H,
2.6. X-ray crystallography
The solid state structures of the new series of molecules were
determined using standard procedures. A Bruker 3-axis platform
diffractometer was used to measure reflections for crystals of
[BzPh3P][i-dtotBu], Cu(PPh3)2(i-dtotBu), Cu(dmp)(PPh3)(
g
1-i-dtot-
radiation se-
graphite monochromator (k = 0.71070 Å) and
Bu) and [BzPh3P][Cu(Br)(S-i-dtotBu] using Mo K
lected with
a
a
5.13; N, 4.03%. FTIR (KBr):
v = 2981(w), 2931(w), 3056(w),
detected via SMART 1000 CCD. With the exception of [BzPh3P][1]
(293 2 K), structures were measured at 150 2 K.
3005(w), 1695(C@O, s), 1652(m), 1616(m), 1588(m), 1558(m),
1538(m), 1506(m), 1498(m), 1478(w), 1456(m), 1434(s), 1366(s),
1275(b,s), 1254(b,s), 1165(b,s), 1151(m), 1092(m), 1066(s),
1038(m), 988(s), 860(m), 842(m), 742(s), 731(m), 695(s), 522(s),
503(m), 491(m) cmꢀ1. See Supporting information for key spectra.
Single crystals of the new species were obtained as follows:
[BzPh3P][1] crystallized from CH2Cl2/i-PrOH as orange blocks in
the orthorhombic space group Fdd2. Orange plates of 2 in the tri-
ꢀ
1-i-dtotBu)ꢂCH2Cl2
clinic space group P1 were obtained by layering i-PrOH on a con-
X-ray quality crystals of Cu(dmp)(PPh3)(
g
centrated CH2Cl2 solution and allowing the CH2Cl2 to evaporate
slowly at room temperature. 3ꢂCH2Cl2 crystallized upon slow diffu-
sion of Et2O into a CH2Cl2 solution as yellow blocks in the mono-
clinic space group P21/c. [BzPh3P][4]ꢂ0.5THF was obtained by
vapor diffusion of Et2O into a room temperature solution in THF
to give red plates in the monoclinic space group P21k crystal struc-
tures were solved with SHELXS-97 using experimental parameters
listed in Table 1 and refined using SHELXL-97 [9]. Hydrogens were
empirically placed using a constrained geometric arrangement
based on energy minimized distances.
were isolated after slow diffusion of Et2O into the reaction mixture
at ꢀ20 °C. The crystals readily lost the CH2Cl2 of crystallization, so
care was made to maintain the crystals at low temperature
(150 2 K).
2.4. [BzPh3P][Cu(Br)(S-i-dtotBu)]ꢂ0.5THF ([BzPh3P][4]ꢂ0.5THF)
Under argon flow a solution of 0.40 g of [BzPh3P][1] (0.75 mmol,
2.1 equiv) in ca. 10 mL dichloromethane was added slowly to a
stirring deaerated solution of 0.080 g of CuBr2 (0.36 mmol) in a

D.F. Zigler et al. / Inorganica Chimica Acta 374 (2011) 261–268
263
Table 1
Experimental parameters and crystallographic results for the series of O-t-butyl-1,1-dithiooxalate species.
[BzPh3P][1]
31H31O2PS2
2
3ꢂCH2Cl2
C39H38Cl2CuN2O2PS2
796.24
monoclinic
yellow block
0.2 ꢁ 0.2 ꢁ 0.1
16.712(5)
19.210(6)
11.621(4)
90.00
[BzPh3P][4]ꢂ0.5THF
C33H35 BrCuO2.50 PS3
742.21
monoclinic
red plate
0.35 ꢁ 0.1 ꢁ 0.07
10.108(2)
9.2736(19)
38.513(8)
90.00
Formula
Fw
Symmetry
Morphology
C
C42H39CuO2P2S2
765.33
triclinic
530.65
orthorhombic
orange block
0.3 ꢁ 0.3 ꢁ 0.2
52.853(3)
13.0849(8)
16.6948(10)
90
orange plate
0.3 ꢁ 0.15 ꢁ 0.08
10.262(2)
13.096(3)
14.625(3)
88.430(3)
76.123(3)
76.584(3)
1855.2(6)
150(2)
Crystal dimension (mm)
a (Å)
b (Å)
c (Å)
a
(°)
b (°)
(°)
90
90
91.136(5)
90.00
3730(2)
150(2)
95.567(3)
90.00
3592.9(13)
150(2)
c
Unit-cell volume (Å3)
11545.8(12)
293(2)
Fdd2
T (K)
Space group
ꢀ
P21/i
P21/c
P1
Z
l
16
2.65
2
8.23
4
9.21
4
(cmꢀ1
)
19.67
Total reflections
Unique reflections
16 494
5490
13 778
7026
30 074
7570
28 745
6097
Reflections [I > 2
r(I)]
4392
4777
1.029b
0.0476b
0.1000b
5484
4548
Sa
1.18b
1.496b
1.587c
Ra
0.0460b
0.1036b
0.0524b
0.1335b
0.0854c
R(w)a
0.2032c
a
Refinement of F2 against all reflections. The weighted R-factor R(w) and goodness of fit S are based on F2, conventional R-factors R are based on F, with F set to zero for
negative F2. The threshold expression of F2 > 2
r
(F2) is used only for calculating R-factors, etc. and is not relevant to the choice of reflections for refinement.
Weighting factor (w) = [s2(Fo2) + (0.0500P)2 + 0.0000P]ꢀ1, where P = ((Fo2 þ 2F2c )/3.
b
Weighting factor (w) = [s2((Fo2) + (0.0686P)2 + 0.0000P]ꢀ1, where P = (Fo2 þ 2F2c )/3.
c
2.7. DFT calculations
coming from the excess 1. Scheme 1 suggests a possible pathway
by which this might occur. Interestingly, attempts to prepare 4
using CuSO4 or Cu(OAc)2 did not result in analogous products sug-
gesting that the bromide plays a role, perhaps by stabilizing the
product.
Density functional theory calculations were performed on the
anion Cu(Br)(S-i-dtotBu)ꢀ in an effort to understand the observed
electronic properties. For comparisons, geometries and Cu-local
ionization energies (ELIP) were calculated using the B3LYP func-
tional with the 6-31G⁄ basis set for 4 and several other tricoordi-
nate models. The 12 lowest energy electronic excited states of 4
were calculated with SPARTAN ‘08 using time-dependent (TD) DFT,
also at the B3LYP/6-31G⁄ level [10].
3.2. Structures determined by X-ray crystallography
The crystal structure of [BzPh3P][i-dtotBu)] (Fig. 2) is the first
reported for an uncoordinated O-alkyl-1,1-dithiooxalate anion.
3. Results and discussion
S
O
2
S
3.1. Syntheses
O
+ [O]
The anionic ligand O-t-butyl-1,1-dithiooxalate (1, i-dtotBuꢀ)
was prepared by the reaction of sulfur with t-butyl-2-chloroacetate
following procedures adapted from Strauch et al. [2] and isolated
as the benzyltriphenylphosphonium salt. This salt was used as
the source of the anionic ligand 1 in the subsequent reactions.
Reaction of [BzPh3P][1] with CuCl(PPh3)3 gave Cu(PPh3)2(i-dtotBu)
(2) with a bidentate dithiocarboxylate group in agreement with the
earlier synthesis [2] of the analogous methyl complex Cu(PPh3)2-
(i-dtoMe). Reaction of 2 with 1 equiv of 1,10-dimethyl-2,9-phenan-
throline led to displacement of one PPh3 and formation of the
S
O
O
O
S
S
O
S
I
+ Cu Br
I
Cu Br
S
O
O
S
S
monodentate dithiolcarboxylato complex Cu(dmp)(PPh3)(g
1-i-
S
O
dtotBu) (3).
O
OH
The room temperature biphasic reaction under argon of aque-
ous CuBr2 with 2.1 equiv of [BzPh3P][i-dtotBu] in CH2Cl2 led to
formation not of a complex of 1 but to a complex of the unexpected
perthiocarboxylato ligand S-i-dtotBuꢀ. The crude [BzPh3P][Cu(Br)-
(S-i-dtotBu)] salt ([BzPh3P][4]) was isolated and then recrystallized
by slow diffusion of diethyl ether into a THF solution and separat-
ing the dark red needles from a white amorphous material, proba-
bly [BzPh3P]Br.
Br
I
O
S
Cu
O
HO
S
+
S
O
S
O
The formation of 4 involves oxidation of the ligand 1 and reduc-
tion of Cu(II) to Cu(I) with the extra sulfur of the perthioligand
Scheme 1. Speculative pathway for the formation of 4 from CuBr2 and 1.

264
D.F. Zigler et al. / Inorganica Chimica Acta 374 (2011) 261–268
Table 3
Selected structural details for Cu(PPh3)2(i-dtotBu).
Atom
Bond length (Å)a
Atoms
Bond angle (°)a
Cu–P1
Cu–P2
Cu–S1
Cu–S2
C1–S1
C1–S2
C1–C2
C2–O1
C2–O2
O2–C3
2.2727(10)
2.2685(10)
2.4928(11)
2.3694(11)
1.682(4)
1.663(4)
1.528(5)
1.213(4)
1.337(4)
S1–Cu–S2
P1–Cu–P2
S2–Cu–P2
S1–Cu–P2
S1–C1–S2
C2–C1–S2
C1–C2–O2
73.75(3)
125.98(4)
111.86(4)
107.00(4)
121.6(3)
117.5(3)
111.5(3)
Torsion angle (°)a
10.9(5)
1.485(4)
O1–C2–C1–S2
a
Value in parentheses is error in last significant digit.
Fig. 2. Crystal structure of [BzPh3P][1] showing 50% thermal ellipsoids, T = 293 K.
Hydrogens omitted for clarity.
The CꢀS bond distances of 1.659(3) and 1.666(3) Å in 1 are slightly
shorter than those reported for the 1,1-dithiooxalate dianion in
Cs2[i-dto]ꢂCsClꢂH2O, C–S = 1.68(1) Å [12]. As expected, the C–O
bond lengths within the ester, 1.331(4) and 1.203(4) Å, are consis-
tent with respective single and double bonds. The C–C distance of
1.519(4) Å also is consistent with a single bond. The torsion angle
between S–C–S and O–C–O planes is 89.4(3)°. The S–C–S angle is
129.11(18)°, and is close the value 128.6(8)° reported for the i-
dto2ꢀ anion [12]. Other selected properties are summarized in
Table 2.
The structure of Cu(PPh3)2(i-dtotBu) (2) (Fig. 3), is similar to that
of the previously reported methyl ester analog Cu(PPh3)2(i-dtoMe)
with nearly the same C–S distances, 1.682(4) and 1.663(4) Å [2].
The Cu–S distances, 2.4928(11) and 2.3694(11) Å, are longer than
for less sterically encumbered systems [13]. Contrary to
Table 2
Fig. 4. Crystal structure of 3ꢂCH2Cl2 showing 50% thermal ellipsoids (T = 150 K).
Selected interatomic distances for [BzPh3P][i-dtotBu].
Hydrogens and phenyl rings omitted for clarity.
Atom
Bond length (Å)a
Atoms
Bond angle (°)a
C1–S1
C1–S2
C1–C2
C2–O1
C2–O2
O1–C3
C3–C4
1.666(3)
1.659(3)
1.519(4)
1.331(4)
1.203(4)
1.478(3)
1.512(5)
S1–C1–S2
S1–C1–C2
O2–C2–C1
O2–C2–O1
129.11(18)
116.2(2)
123.7(3)
125.5(3)
Table 4
Selected interatomic distances for Cu(dmp)(PPh3)(
g
1-i-dtotBu)]ꢂCH2Cl2.
Atom
Bond length (Å)a
Atoms
Bond angle (°)a
Torsion angle (°)a
89.4(3)
Cu–P
2.2253(11)
2.3096(11)
3.771
N1–Cu–N2
S2–Cu–P
80.03(10)
S2–C1–C2–O2
Cu–S2
Cuꢂ ꢂ ꢂS1
Cu–N1
Cu–N2
C1–S1
C1–S2
C1–C2
C2–O1
C2–O2
O2–C3
122.58(4)
111.97(8)
128.9(2)
113.00(12)
114.6(2)
111.6(3)
a
S2–Cu–N2
S1–C1–S2
C1–S2–Cu
C2–C1–S2
C1–C2–O2
Value in parentheses is error in last significant digit.
2.133(3)
2.084(3)
1.656(3)
1.686(3)
1.518(5)
1.203(4)
1.329(4)
1.505(4)
Torsion angle (°)a
61.28(14)
106.0(3)
C1–S2–Cu–P
O1–C2–C1–S2
a
Value in parentheses is error in last significant digit.
[BzPh3P][1] and the methyl analog, the 1,1-dithiooxalate backbone
is nearly coplanar, with a torsion angle of only 10.9(5)°. The C–C
distance of 1.528(5) Å is again consistent with a single bond. These
and other data are summarized in Table 3.
The structure of Cu(dmp)(PPh3)(g
1-i-dtotBu) (3) clearly shows
the monodentate coordination of the i-dtotBuꢀ anion (Fig. 4 and
Table 4). Surprisingly, the C–S distances 1.656(3) and 1.686(3) Å
for the C@S and C–S, respectively, are only modestly different than
those seen for the g2 precursor 2. However, the Cu–S distance at
2.3096(11) Å is considerably shorter than in 2 perhaps as the result
of backbonding from Cu(I) into the dmp acceptor that would make
Fig. 3. Crystal structure of 2 (50% thermal ellipsoids) at 150 K. Hydrogens and
phenyl rings omitted for clarity.

D.F. Zigler et al. / Inorganica Chimica Acta 374 (2011) 261–268
265
displays a 3-coordinate Cu(I). Around the CuSSC@S ring, the mea-
sured bond distances are, respectively, 2.216(2), 1.985(3),
1.700(8) and 1.657(8) Å with a 2.212(2) Å Cu–S(S@C) bond closing
the ring. The C–C distance within the dithiooxalate backbone,
though slightly shorter than the preceding systems at 1.494(12) Å,
is reasonable for a single bond. The torsion angle between CS2
and CO2 moieties was measured at 11.6(13)°. The CuSSC@S ring,
bromide and ester carbon are all coplanar, deviating by no more
than 3.5° from planarity. The S–Cu–S angle is 99.33(9)° with the
bromide centered between the bound S atoms; S–Cu–Br ꢃ 130°.
The distorted trigonal planar coordination environment of Cu(I) is
evident even in the packing diagram (Fig. 6) which shows no other
ligands, including cations and solvent, within a normal distance for
coordination. The shortest distance between any two copper atoms
was 8.785 Å and the closest non-bonded atoms, phenyl hydrogens,
are 2.96–3.90 Å distant.
Fig. 5. Crystal structure of [BzPh3P][4]ꢂ0.5THF showing 50% thermal ellipsoids,
T = 150 K. Hydrogens omitted for clarity.
The structure of the chelating perthiolcarboxylato ligand in
[BzPh3P][4] differs from other metal-perthio-thionyl systems as
indicated by representative structures summarized in Table 6
[3,4,7,15–23]. As might be expected for metals with differing radii,
the M–S bond lengths vary from 2.13 to 2.35 Å. For most of the
complexes the M–S(S@C) and M–S(S–S) bond lengths are similar, sug-
gesting equivalent metal/sulfur interactions. Octahedral systems
such as [Fe(S3C-PhMe)(S2CPhMe)2], [15] [M(S3CAr)2(S2CAr)]
(M = Tc, Re) [16,17] and clusters [21–23] have longer relative M–
S(S@C) bonds. On par with other planar systems, [BzPh3P][4] has
nearly identical Cu–S bonds, 2.212(2) and 2.216(2) Å. These short
CuꢀS bonds are comparable to [CuCl(dptu)2], dptu = N,N0-diphen-
ylthiourea (CuꢀS = 2.22 Å) [13], and support a strong covalent
interaction between copper and sulfur. These Cu–S bonds are con-
siderably shorter than seen for 2 (2.3694(11) and 2.4928 Å) or 3
(2.3096(11) Å). The C–S bond lengths vary widely between
MSSC@S ring systems, with little indication of the impact of metal
d-electron count or thionyl carbon substituent effects. Generally
the C@S bond is shorter than the carbon bond to the disulfide sul-
fur, aligning with a difference in bond order. The d10 complexes
[Zn(S3CPh-iPr)2] [20] and [BzPh3P][4] exhibit the largest disparity
in C–S bonds with 1.6641(6), 1.711(6) Å and 1.657(8), 1.700(8) Å,
respectively.
Table 5
Selected interatomic distances for [BzPh3P][4]ꢂ0.5THF.
Atom
Bond length (Å)a
Atoms
Bond angle (°)a
Cu–Br
Cu–S1
Cu–S3
S1–S2
C1–S2
C1–S3
C1–C2
C2–O1
C2–O2
Cuꢂ ꢂ ꢂCu
(nearest)
2.2869(14)
2.216(2)
2.212(2)
1.985(3)
1.700(8)
1.657(8)
1.494(12)
1.195(10)
1.304(10)
8.785
S1–Cu–S3
S1–Cu–Br
S3–Cu–Br
S3–C1–S2
C1–S2–S1
C2–C1–S2
C2–C1–S3
99.33(9)
129.66(8)
130.96(8)
126.9(5)
108.6(3)
109.5(6)
123.6(6)
Torsion angle (°)a
3.7(3)
11.6(13)
C1–S2–S1–Cu
O1–C2–C1–S2
a
Value in parentheses is error in last significant digit.
the copper center more electro-positive. The Cu–P distance is also
shorter, 2.2253(11) Å. The torsion angle about the 1,1-dithiooxa-
late C–C backbone is 106.0(3)° and the bond distance of
1.518(5) Å indicates this to be a C–C single bond.
The crystal structure of the mononuclear perthiocarboxylate
complex [BzPh3P][4]ꢂ0.5THF is presented in Fig. 5, with selected
structural properties listed in Table 5. Remarkably, this structure
The unusually short disulfide bond length in 4 can be rational-
ized by considering the electronic influences of the thionyl carbon
and Cu(I) center in this cyclic structure. Unhindered disulfides
Fig. 6. Crystal packing illustration of [BzPh3P][4]ꢂ0.5THF showing the distances between the central copper and its closest four neighbors (Å). Cation, solvent and hydrogens
have been omitted for clarity.

266
D.F. Zigler et al. / Inorganica Chimica Acta 374 (2011) 261–268
Table 6
Selected structural values from five-membered rings formed by a transition metal and chelated perthioligands.
Molecule perthioliganda
Atom distances (Å)b
Ref.
C@S
C–S
S–S
M–S@
M–S–
C–R
Organic analogs
MDCT
MePhC(S)SSC(S)PhMe
1.627c
1.61
1.747
1.78
2.047
2.023(6)
[14]
[11]
1.52
Monometallic complexes
d
Fe(S-dtt)(dtt)2
1.692(2)
1.647(7),
1.656(7)
1.705(3)
1.689(6),
1.676(7)
2.086(8)
2.102(3),
2.068(3)
2.240(2)
2.320e
2.183(4)
2.199e
1.468(1)
1.457(9), 1.467(8)
[15]
[16]
Re(S-dtb)2(dtb)
Tc(S-dtb)2(dtb)
1.667, 1.672
1.820(21)
1.686, 1.703
1.816(21)
2.075, 2.098
2.061(9)
2.354, 2.348
2.285(6)
2.227, 2.230
2.274(6)
1.478, 1.472
1.519(30),
1.549(33)
1.49(3)
[17]
[18]
ReS(S4)(S3CMe2)i
[Ph4P][Ni(S-i-dto-OMe) (i-dto-
OMeacac)]
1.67(2)
1.67(2)
2.013(8)
2.124(6)
2.148(6)
[3]
[Ph4P][Ni(S-i-dto-OMe)(S2CC(CN)2)]
Pt(S3CMe)2
Ni(S3CMe)2
1.689(8)
1.667(7)
1.667(3)
1.681(6)
1.678(9)
1.664(6)
1.657(8)
1.669(10)
1.695(6)
1.668(3)
1.638(6)
1.699(8)
1.711(6)
1.700(8)
2.011(4)
2.045(2)
2.0322(12)
2.027(2)
2.016(4)
2.008(4)
1.985(3)
2.129(3)
2.133(3)
1.467
[4]
[7]
2.2715(14)
2.1579(8)
2.1331(18)
2.138(5)
2.327(2)
2.212(2)
2.2959(16)
2.1623(4)
2.1225(18)
2.130(3)
2.316(3)
2.216(2)
1.498(9)
1.496(4)
1.506(8)
1.385(6)
1.48(1)
[19]
[7]
[20]
[20]
t.w.
Ni(S3CMe)(S2CMe)
Ni(S2CPh-iPr)(S3CPh-iPr)
Zn(S3CPh-iPr)2
Cu(Br)(S-i-dto-OtBu)
1.494(12)
Bridged multimetallic complexesf
[Os2(
l
-S-dtcEt2)2(dtcEt2)3] [BPh4]g
1.71(4), 1.76(3)
1.66(4), 1.76(3)
2.05(2), 2.13(1)
2.39(1),
2.36(1)
2.29(1),
1.32(5)
[21]
2.29(1)h
2.211(3)h
2.239(2)h
K[Ph4P][Cu4(S-tBuDED)3]g
[Bu4N]6[Cu6(S3CC(CN)2)6]
1.716(9)
1.692(6)
1.766(10)
1.748(8)
2.076(4)
2.053(2)
2.272(3)h
2.254(2)
1.392(13)
1.408(8)
[22]
[23]
a
b
c
d
e
f
MDCT = 4-methyl-1,2-dithia-4-cyclopentene-3-thione; dtt = ꢀS(S)CPhMe; dtb = ꢀS(S)CPh.
From the cyclic arrangement of M–S–S–C@S; C–R is atom nearest to the five-membered ring, nr = not reported.
Exocyclic.
Fe–S–S–C@S ring is folded.
Average of like values (DBD < 0.003 Å < RMS).
Situations with bridging perthioligands, only five-member ring dimensions are reported.
Metal–metal bonded clusters.
g
h
i
Sulfur bridges multiple metals.
[S3CMe2]2ꢀ is derived from acetone with an sp3 central carbon and does not have C@S.
typically have a ca. 2.05 Å bond length with a 90° torsion angle
around the S–S bond to minimize lone pair repulsive effects
[24,25]. Covalently constraining the torsion angle in a ring, such
as with trimethylene-1,1-disulfane, generally causes an elongation
of the S–S bond to about 2.10 Å and pushes the disulfide p⁄ to high-
er energy [25]. Shorter S–S distances are observed with electron
withdrawing substituents, for example, the S–S bond of FSSF is
only 1.88 Å [26]. An adjacent C@S group may have a similar effect.
The S–S bond in the organic model 4-methyl-1,2-dithia-4-cyclo-
pentene-3-thione (MDCT), which has an endocyclic disulfide and
an adjacent thionyl carbons is shorter (2.047 Å) than an aliphatic
1,2-dithiolane [14,27]. Acyclic disulfides with adjacent thionyls,
RC(@S)SSC(@S)R, also have short S–S bonds (ꢄ2.01 Å) [11,28].
While the adjacent thionyl in 4 might have a similar effect on
the S–S bond length of the endocyclic disulfide, this should not
be sufficient, thus the presence of the Cu(I) center must be the dif-
3.3. DFT calculations
Density functional theory computations using the B3LYP func-
tional and the 6-31G⁄ Pople basis set, were employed to probe
the electronic properties of 4. Fig. 8 displays the calculated frontier
orbitals for this anion. The HOMO (ꢀ4.72 eV) and the LUMO (3.14)
are primarily ligand p⁄ in character with regard to the perthiocarb-
oxylato ligand with the HOMO indicating some contribution from
the bromide lone pair perpendicular to the plane of the ring. The
HOMO-1 and -2 are largely bromide lone pairs while HOMO-3,
-4 and -5 are largely metal ligand bonding orbitals. However, the
calculated MO’s do not appear to confirm the simple picture pro-
posed in Fig. 7 for Cu(Br)(S-i-dtotBu)ꢀ.
3.4. Electronic spectra
ference. Although Cu(I) is d10 with no empty d orbitals to stabilize
The electronic absorption spectra of [BzPh3P][1],
2 and
p
the repulsive effects of sulfur lone pairs, the trigonal planar Cu(I)
coordination presents a lower energy 4p orbital perpendicular to
the CuSSC@S ring with the correct symmetry to interact with the
filled disulfide p⁄ orbital (Fig. 7).
[BzPh3P][4] are presented in Fig. 9. The solution spectrum for 3
suggested ligand scrambling upon dissolution in CH2Cl2 or THF,
consistent with observations made for similar CuX(phen)(PPh3)
complexes (phen = 1,10-phenanthroline; X = halide or nitrate)
[29–31], thus, is not reported. The spectrum of [BzPh3P][i-dtotBu]
in methanol is characterized by
a
band at kmax = 343 nm
(
e
= 14 600 Mꢀ1 cmꢀ1
)
consistent with the n ? p⁄ of other
ꢀ
R—CS2 chromophores [32] and a structured band at ꢄ267 nm
(4280 Mꢀ1 cmꢀ1) assigned to
p ?
p⁄ transitions of the aromatic
S
RO2C
C
Cu Br
groups on [BzPh3P]+. The spectrum of Cu(PPh3)2(i-dtotBu) in CH2Cl2
S
S
2p
p⁄ PPh3 centered bands at ꢄ265 nm
3p
displays strong
p ?
4p
(23 100 Mꢀ1 cmꢀ1) as well as bands at 364 nm (5230 Mꢀ1 cmꢀ1
)
3p
3p
and at 400 nm (shoulder, 3000 Mꢀ1 cmꢀ1) and a low energy tail
consistent with the literature [2]. Lastly, the THF solution of
Fig. 7. Illustration of possible p-orbital mixing in 4 .

D.F. Zigler et al. / Inorganica Chimica Acta 374 (2011) 261–268
267
Fig. 8. Pictorial representations of the LUMO, HOMO and HOMO-1 through HOMO-7 for 4 as calculated by DFT at the B3LYP/6-31G⁄ level of theory.
Notably, the spectra of [BzPh3P][4] in different solvents exhibits
solvatochromic behavior for the stronger bands as summarized
in Table 7.
The lowest lying singlet electronic excited states of 4 and its
theoretical electronic absorption spectrum were calculated using
time-dependent DFT at the B3LYP/6-31G⁄ level. This TD-DFT calcu-
lation predicted the absorption bands at 646 nm [(HOMO-1,
83%) ? LUMO, or XLCT (where X is Br and L is the perthiocarboxy-
late ligand, and CT is charge transfer)]; at 499 nm [(HOMO-5,
89%) ? LUMO, or largely MLCT]; overlapping transitions at
ꢅ
426 nm [(HOMO-3, 65%) ? LUMO, largely rCuBr to pL CT] and at
ꢅ
409 nm [(HOMO-6, 89%) ? LUMO, or pL
?
pL ligand localized];
and 356 nm [(HOMO-7, 68%) ? LUMO, or largely dCu,rL ?
pLꢅ].
Fig. 10 is the DFT predicted electronic absorption spectrum of
[BzPh3P][4] (in vacuum) overlaid with the experimental spectrum
measured in CH2Cl2. The strongest transitions in the electronic
absorption spectrum are predicted to be at about 356 and ca.
426 nm, respectively (in vacuum) and the calculated oscillator
strengths (fcalc) of these are less than factor of two larger than
the experimental fexp values calculated from the product of emax
and the full width at half maximum (in cmꢀ1).
Fig. 9. Electronic absorption spectra: [BzPh3P][1] in MeOH (—), 2 in CH2Cl2 (- - -)
and [BzPh3P][4] in THF ( ).
Several mechanisms can be considered for the observed solva-
tochromism for 4 described in Table 7 [33–35]. Experimentally,
UV
[BzPh3P][Cu(Br)(S-i-dtotBu)] displays two broad and prominent
the relative position of the two prominent bands
ð
mmax —
3
mmax ꢄ 8.5 ꢁ 10 cmꢀ1) is similar in each solvent so this agues
Vis
bands at 358 nm (3730 Mꢀ1 cmꢀ1) and 512 nm (3420 Mꢀ1 cmꢀ1
)
with a low energy tailing at 720 nm (shoulder, 27 Mꢀ1 cmꢀ1).
against any major structural change from one medium to another.
Table 7
Effect of solvent on the electronic absorption spectrum of [BzPh3P][Cu(Br)(S-i-dtotBu)].
a
Solvent
kmax (nm)
kmax (nm)
D
vmax (cmꢀ1
)
Rel. Absb
ec
p⁄d
DNe
THF
513
508
505
470
358
357
351
336
8400
8300
8700
8500
1.26
1.18
1.06
0.89
7.5
21
9.1
37.5
0.58
0.71
0.82
0.75
20
17
0
Acetone
CH2Cl2
MeCN
14.1
a
b
c
The difference in maximal frequency between ultraviolet and visible bands in each solvent.
UV/AVis
Solvent dielectric constant.
Kamlet solvent polarity factor [34].
Solvent donor number from Linert et al. [33].
A
.
d
e

268
D.F. Zigler et al. / Inorganica Chimica Acta 374 (2011) 261–268
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Spectroscopic properties of complexes (Figs. S1–S9), Packing
diagram for [BzPh3P][CuBr(S-i-dto-OtBu]ꢂ0.5THF, Results of DFT
calculations (two tables and one figure).
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ica.2011.02.037.
References
[1] P.C. Ford, Acc. Chem. Res. 41 (2008) 190.
[2] P. Strauch, B. Dempe, R. Kempe, W. Dietzsch, J. Sieler, E. Hoyer, Z. Anorg. Allg.
Chem. 620 (1994) 498.
[3] J. Piotraschke, P. Strauch, G. Zahn, E. Hoyer, Z. Anorg. Allg. Chem. 620 (1994)
505.
[4] J. Piotraschke, P. Strauch, S. Abram, U. Abram, Z. Anorg. Allg. Chem. 620 (1994)
1659.
[5] J. Piotraschke, P. Strauch, R. Kimse, Z. Anorg. Allg. Chem. 621 (1995) 159.
[6] J. Piotraschke, P. Strauch, Inorg. Chem. Acta 236 (1995) 177.
[7] R. Mas-Ballesté, A. Guijarro, R. González-Prieto, Ó. Castillo, P.J. Sanz Miguel, F.
Zamora, J. Chem. Soc., Dalton Trans. 39 (2010) 1511.
Fig. 10. TD-DFT calculated (B3LYP/6-31⁄, in vacuum, —) and experimental (in
CH2Cl2, —) electronic absorption spectrum of 4. The spectra are artificially
normalized based on the appropriate vertical scale (f = oscillator strength for the
calculated spectrum and
e = experimental extinction coefficient).
[8] S.D. Pringle, K. Giles, J.L. Wildgoose, J.P. Williams, S.E. Slade, K. Thalassinos, R.H.
Bateman, M.T. Bowers, J.H. Scrivens, Int. J. Mass Spectrom. 261 (2007) 1.
[9] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[10] Spartan ’08, Wavefunction, Inc., Irvine, CA, Methods documented in: Y. Shao,
L.F. Molnar, Y. Jung, J. Kussmann, C. Ochsenfeld, S.T. Brown, A.T.B. Gilbert, L.V.
Slipchenko, S.V. Levchenko, D.P. O’Neill, R.A. DiStasio Jr., R.C. Lochan, T. Wang,
G.J.O. Beran, N.A. Besley, J.M. Herbert, C.Y. Lin, T. Van Voorhis, S.H. Chien, A.
Sodt, R.P. Steele, V.A. Rassolov, P.E. Maslen, P.P. Korambath, R.D. Adamson, B.
Austin, J. Baker, E.F.C. Byrd, H. Dachsel, R.J. Doerksen, A. Dreuw, B.D. Dunietz,
A.D. Dutoi, T.R. Furlani, S.R. Gwaltney, A. Heyden, S. Hirata, C.-P. Hsu, G.
Kedziora, R.Z. Khalliulin, P. Klunzinger, A.M. Lee, M.S. Lee, W.Z. Liang, I. Lotan,
N. Nair, B. Peters, E.I. Proynov, P.A. Pieniazek, Y.M. Rhee, J. Ritchie, E. Rosta, C.D.
Sherrill, A.C. Simmonett, J.E. Subotnik, H.L. Woodcock III, W. Zhang, A.T. Bell,
A.K. Chakraborty, D.M. Chipman, F.J. Keil, A. Warshel, W.J. Hehre, H.F. Schaefer,
J. Kong, A.I. Krylov, P.M.W. Gill, M. Head-Gordon, Phys. Chem. Chem. Phys. 8
(2006) 3172.
Furthermore the shifts did not correlate with the solvent donor
numbers, so it appears that the anion 4 may remain 3-coordinate
in solution. In this context ESI-MS experiments in both THF or
MeCN gave [Cu(Br)(S-i-dtotBu)]ꢀ as the major ion with no evidence
of solvent adducts. From the DFT calculations we should expect the
two strongest transitions to have similar character, both having
significant MLCT contributions, thus similar dipolar changes be-
tween the ground and excited states.
In summary, we describe here the synthesis, structures and
some properties of the ligand O-t-butyl-1,1-dithiooxalate, i-
dtotBuꢀ and of bidentate and a monodentate complexes of this li-
gand with copper(I). In an attempt prepare a photochemically ac-
tive Cu(II) complex of this ligand, we found that the reaction of 1
[11] O. Niyomura, Y. Kitoh, Sulfur Lett. 22 (1999) 195.
[12] R. Mattes, W. Meschede, Chem. Ber. 109 (1976) 1832.
[13] G.A. Bowmaker, N. Chaichit, J.V. Hanna, C. Pakawatchi, B.W. Skelton, A.H.
White, J. Chem. Soc., Dalton Trans. 38 (2009) 8308.
[14] G.A. Jeffrey, R. Shiono, Acta Crystallogr. 12 (1959) 447.
[15] D. Coucouvanis, S.J. Lippard, J. Am. Chem. Soc. 91 (1969) 307.
[16] F. Mévellec, A. Roucoux, N. Noiret, H. Patin, F. Tisato, G. Bandoli, Inorg. Chem.
Commun. 2 (1999) 230.
[17] F. Mévellec, F. Tisato, F. Refosco, A. Roucoux, N. Noiret, H. Patin, G. Bandoli,
Inorg. Chem. 41 (2002) 598.
[18] A. Müller, M. Lemke, E. Krickemeyer, H. Bögge, M. Penk, Monatsh. Chem. 124
(1993) 857.
[19] A.H. Mahmoudkahni, V. Langer, Acta Cyrstallogr., Sect. E 57 (2001) m92.
[20] J.P. Fackler Jr., J.A. Fetchin, D.C. Fries, J. Am. Chem. Soc. 94 (1972) 7323.
[21] L.J. Maheu, L.H. Pignolet, Inorg. Chem. 18 (1979) 3626.
[22] D. Coucouvanis, S. Kanodia, D. Swenson, S.-J. Chen, T. Stüdemann, N.C.
Baenziger, R. Pedelty, M. Chu, J. Am. Chem. Soc. 115 (1993) 11271.
[23] C.W. Liu, R.J. Staples, J.P. Fackler, Coord. Chem. Rev. 174 (1998) 147.
[24] G.W. Kutney, K. Turnbull, Chem. Rev. 82 (1982) 333.
with CuBr2 gave
a deep red, crystalline Cu(I) compound,
[BzPh3P][4], instead of the expected Cu(i-dtotBu)2. The structure
of 4 showed the copper to be 3-coordinate and planar with a
bidentate perthio-1-thiooxalato ligand S-i-dtotBuꢀ, apparently
the product of the disproportionation of 1 as has been described
previously [4]. The SꢀS bond of the CuSSC@S cycle proved to be
quite short, suggesting delocalization of the disulfide sulfur lone
pairs. Data from mass spectrometry and the electronic absorption
spectroscopy in various solutions of 4, indicate that solvent is not
strongly bound (if at all).
Acknowledgements
[25] R. Steudel, Angew. Chem., Int. Ed. 14 (1975) 655.
[26] R.L. Kuczkowski, J. Am. Chem. Soc. 86 (1964) 3617.
[27] R. Steudel, Y. Drozdova, K. Miaskiewicz, R.H. Hertwig, W. Koch, J. Am. Chem.
Soc. 119 (1997) 1990.
The authors thank Dr. James Pavlovich (UCSB) for helpful dis-
cussions on mass spectrometry techniques and Dr. Megan Grab-
enauer (UCSB) for performing the nanospray MS experiments.
This work was partially supported by a grant from the U.S. National
Science Foundation (NSF – CHE-0749524).
[28] R. Cea-Olivares, V. García-Montalvo, S. Hernández-Ortega, C. Rodríguez-
Narvaez, P. García y García, M. López-Cardoso, P. de March, L. González,
L. Elias, M. Figueredo, J. Font, Tetrahedron: Asymmetry 10 (1999) 3337–3340.
[29] Q.-H. Jin, X.-L. Xin, Acta Crystallogr., Sect. C 55 (1999). IUC9900107.
[30] C. Pettinari, G.G. Lobbia, G. Sclavi, D. Leonesi, Polyhedron 14 (1995) 1709.
[31] F.H. Jardine, A.G. Vohra, F.J. Young, J. Inorg. Nucl. Chem. 33 (1971) 2941.
[32] M.A. Bernard, M.-M. Borel, G. Dupriez, Rev. Chem. Miner. 12 (1975) 181.
[33] W. Linert, Y. Fukuda, A. Camard, Coord. Chem. Rev. 218 (2001) 113.
[34] M.J. Kamlet, J.-L.M. Abboud, M.H. Abraham, R.W. Taft, J. Org. Chem. 48 (1983)
2877.
Appendix A. Supplementary material
CCDC 807531–807534 contains the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
[35] C. Reichardt, Chem. Rev. 94 (1994) 2319.