A series of cyanide-bridged binuclear complexes

Article abstract of DOI:10.1016/j.ica.2009.05.042

A series of cyanide-bridged binuclear complexes, ('S₃')Ni-CN-M[Tp^tBu] ('S₃' = bis(2-mercaptophenyl)sulfide, Tp^tBu = hydrotris(3-tert-butylpyrazolyl)borate, M = Fe (2-Fe), Co (2-Co), Ni (2-Ni), Zn (2-Zn)) was pre

Full text of DOI:10.1016/j.ica.2009.05.042

Inorganica Chimica Acta 362 (2009) 4553–4562
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
A series of cyanide-bridged binuclear complexes
Michael T. Mock a, Matthew T. Kieber-Emmons a, Codrina V. Popescu b,
Patrick Gasda b, Glenn P.A. Yap a, Charles G. Riordan a,
*
a Department of Chemistry and Biochemistry, University of Delaware, Brown Laboratory, Newark, Delaware 19716, United States
b Department of Chemistry, Ursinus College, Collegeville, Pennsylvania 19426, United States
a r t i c l e i n f o
a b s t r a c t
A series of cyanide-bridged binuclear complexes, (‘S3’)Ni–CN–M[TptBu] (‘S3’ = bis(2-mercaptophenyl)sul-
fide, TptBu = hydrotris(3-tert-butylpyrazolyl)borate, M = Fe (2-Fe), Co (2-Co), Ni (2-Ni), Zn (2-Zn)) was
prepared by the coupling of K[(‘S3’)Ni(CN)] with [TptBu]MX. The isostructural series of complexes was
structurally and spectroscopically characterized. A similar coupling strategy was used to synthesize
the anionic copper(I) analogue, Et4N{(‘S3’)Ni–CN–Cu[TptBu]}, 2-Cu.
Article history:
Received 10 February 2009
Accepted 13 May 2009
Available online 27 May 2009
Dedicated to our friend and former
An alternative synthesis was devised for the preparation of the linkage isomers of 2-Zn, i.e. of cyanide-
bridged linkage isomers. X-ray diffraction, 13C NMR and IR spectral studies established that isomerization
to the more stable Ni–CN–Zn isomer occurs. DFT computational results buttressed the experimental
observations indicating that the cyanide-bridged isomer is ca. 5 kcal/mol more stable than its linkage
isomer.
colleague, Swiatoslav ‘Jerry’ Trofimenko.
Keywords:
Cyanide
Cyanide-bridged
Binuclear complex
Linkage isomers
Tp
Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction
is the target of cyanide poisoning in higher organisms [7]. Inhibitor
binding often alters the spectroscopic characteristics of the en-
Transition metal cyanide complexes have been subject of in-
tense research due to their interesting physical properties includ-
ing magnetism, color, conductivity and hydrogen storage capacity
[1,2]. For example, variations of one of the earliest coordination
compounds, Prussian blue, ([Fe4{Fe(CN)6}3]ꢀxH2O), has been used
to construct room temperature organometallic magnets [3].
Despite the novel physical properties that many transition
metal–cyanide complexes exhibit, simple M–CN–M0 complexes
containing only one cyanide linkage remain less uncommon. As a
consequence, the vibrational properties of such an isolated linkage
are not well understood [4] and are worthy of study. In fact, recent
reports have focused on fundamental vibrational spectra of M–CN
and M–CN–M0 type binding modes, specifically, to understand the
spectroscopic differences between CN and the isoelectronic CO li-
gand [4,5].
Recently, transition metal cyanide motifs have also been re-
vealed as integral components of metalloenzyme active sites. For
example, both Ni–Fe and Fe-only hydrogenases, enzymes that cat-
alyze the reversible oxidation of H2, contain terminal cyanide and
carbonyl ligands ligated to the iron center of the active site [6]. Fur-
thermore, cyanide often inhibits or inactivates enzyme catalysis by
binding to the enzyme active site(s) in a competitive fashion, i.e.
binding at the substrate site. For example, cytochrome c oxidase
zymes, such that these spectroscopic changes can be used as
probes providing mechanistic insight concerning the location of
inhibitor ion/substrate binding.
Of particular interest in this laboratory is a class of metalloen-
zymes found in anaerobic bacteria and archaea, carbon monoxide
dehydrogenases (COdHs) and acetyl coenzyme A synthase [8].
These metalloenzymes catalyze the interconversion of CO and
CO2 and the synthesis of acetyl coenzyme A, respectively. Crystal-
lographic studies have shown the C-cluster of COdHs from anaer-
obes contain a NiFe3S4 cubane subunit ligated to an external
single iron site by a bridging cysteine [9], l-sulfido [10] or hydroxo
[11] ligand. Cyanide, a known inhibitor of COdH [12], has been pro-
posed to inhibit by bridging between the nickel and the unique
iron [13]. In the proposed catalytic mechanism [14], which finds
support in recent crystallographic studies [11], CO binds to the
Ni site of the cubane and is attacked by a substrate hydroxide
ion that was formally bridging the Ni and unique Fe sites. Hydrox-
ide attack leads to the formation of a Ni-bound carboxylate that
subsequently dissociates as CO2 completing the catalytic cycle.
Cyanide was found to displace substrate hydroxide, leading to
the suggestion that these anions bind to the same site. This
hypothesis raises important questions regarding the role and iden-
tity of the bridging ligand in enzyme structure and catalysis. To ad-
dress these issues, we sought to prepare synthetic complexes that
contain a single cyanide bridge between iron and nickel. Whereas
* Corresponding author. Tel.: +1 302 831 1073; fax: +1 509 463 3815.
E-mail address: riordan@udel.edu (C.G. Riordan).
there are examples of polycyanide complexes containing
a
0020-1693/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2009.05.042

4554
M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
Ni–NC–Fe linkage [15], we are unaware of either a monocyanide
complex containing this bridge or any example of the linkage iso-
mer, Ni–CN–Fe1. Herein, we detail binuclear metal complexes con-
taining a single cyanide bridge aimed at modeling inhibited forms
of COdH, so as to provide structural and spectroscopic information
applicable for analysis of the more complex active site.
tion resistance was established prior to each measurement. Ferro-
cene was used as the internal standard (E1/2 = 801 mV in THF
versus NHE). Substrate concentrations were in the range of 5–
20 mM. Melting points were taken with a Melt-Temp melting point
apparatus. Elemental analyses were performed by Desert Analytics,
Inc. Tucson AZ, or Atlantic Microlabs, Atlanta, GA.
2.2. Syntheses
2. Experimental
2.2.1. K[Ni(‘S3’)CN] (K1, K1-13CN)
2.1. General procedures
Sellmann and co-workers previously reported the preparation
of the [NBu4]1 [23] and [NMe4]1 [18] salts of the anion [Ni(‘S3’)CN].
The K salt was prepared and isolated as follows: potassium cyanide
(0.145 g, 2.23 mmol) in 10 mL methanol and 1 mL H2O was added
to a stirring suspension of brown [Ni(‘S3’)]2 (0.685 g, 1.12 mmol) in
50 mL THF. The solution color changed from brown to yellow–
brown with dissolution of the solids. Following 3 h of stirring,
the solvent was removed in vacuo. The resulting solid was redis-
solved in 30 mL methanol yielding a tan precipitate that was re-
moved by filtration through a medium porosity glass frit. The
solvent was removed in vacuo and the brown solids were extracted
with acetonitrile, filtered, and dried in vacuo. Trituration of the
sticky brown solids with ethyl ether afforded a fine green powder
that was collected by vacuum filtration. Yield: 0.553 g, 66%. 1H
NMR (d6-acetone): d 6.95 (t, C6H4, 2 H), 7.05 (t, C6H4, 2 H), 7.28
(d, C6H4, 2 H), 7.78 (d, C6H4, 2 H). 13C NMR (d6-acetone): d 122.2
(C5–‘S3’), 126.0 (Ni–13CN), 128.3 (C4–‘S3’), 128.8 (C3–‘S3’), 130.1
All air and moisture sensitive reactions were performed under
N2 using standard Schlenk line techniques or carried out under
an argon or N2 atmosphere in a Vacuum Atmospheres glovebox
equipped with a gas purification system [16]. Unless described
otherwise, all reagents were purchased from commercial sources
and were used as received. Solvents were dried by passage through
activated alumina [17], sparged with N2, and tested with Na/ben-
zophenone before use. Deuterated solvents and were purchased
from Cambridge Isotope Laboratories and stored over 4 Å molecu-
lar sieves. K13CN was purchased from Cambridge Isotope Laborato-
ries. KCN and Et4NCl were purchased from Acros Chemical
Company. ‘S3’–H2 [18], Tl[TptBu] [19], NMe4[Ni(‘S3’)(CN)] [18],
[TptBu]MX (M = Fe, Co, Ni, Zn; X = Cl, Br, I) [19] [TptBu]Cu(NCCH3)
[20], [PhTttBu]Cu(NCCH3) [21], and [Ni(‘S3’)]2 [18] and were pre-
pared following the published procedures. [TptBu]Zn(CN) [22] and
its 13CN isotopomer were prepared following the procedure pub-
lished [19] for the synthesis of [TptBu]Zn(N3) replacing NaN3 with
KCN (or K13CN).
(C6–‘S3’), 133.3 (C1–‘S3’), 156.3 (C2–‘S3’). UV–Vis (THF), kmax (e,
cmꢁ1 Mꢁ1): 290 (20 025), 318 (sh), 422 (1200), 628 (57). IR
13
) = 2106 (2065) cmꢁ1. Anal. Calc. for C13H8KNNiS3: C,
1H and 13C NMR spectra were recorded on a Bruker AC-250, AM-
360, DRX-400 or AV-400 MHz NMR spectrometers. NMR spectra
were referenced to residual protio solvent signals at ambient tem-
perature unless noted otherwise. NMR abbreviations are as fol-
lows: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br,
broad. 31P NMR spectra were recorded on either a Bruker DRX-
400 or AV-400 MHz NMR spectrometer using an aqueous solution
of 85% phosphoric acid in D2O as an external reference (0 ppm). So-
lid-state magnetic moments were determined on a Johnson Mat-
they magnetic susceptibility balance. The magnetic susceptibility
of the sample was adjusted for diamagnetic contributions using
Pascal’s constants. Mössbauer spectra were collected on a spec-
trometer with closed cycle refrigeration, model CCR4K, operating
between 4.5 and 325 K, in constant acceleration mode. The spec-
trometer was fitted with a permanent 0.3 Tesla magnet. The spec-
tra were analyzed using the software package WMOSS (Dr. Thomas
Kent, SEE Co., Minnesota.) Isomer shifts are quoted with respect to
iron metal at room temperature. Perpendicular mode, X-band EPR
spectra were recorded at 9.44 GHz on a Bruker EMX spectrometer
equipped with a ER073 magnet, T103 resonator cavity, ER041MR
microwave bridge and ER081 power supply. Sample temperature
was controlled by an Oxford Systems LHe cryostat system. Infrared
spectra were recorded on a Mattson Genesis Series FTIR spectrom-
eter at ambient temperature and under a purge stream of nitrogen.
Solid-state FT-IR samples were prepared as KBr pellets. Cyclic vol-
tammetry was performed in a Vacuum Atmospheres MO–40 M
glovebox under an argon atmosphere using a BAS (Bioanalytical
Systems) CV–50 W voltammetric analyzer. A typical three-elec-
trode single compartment electrochemical cell included a glassy
carbon (r = 1 mm) working electrode, a Ag+/Ag reference electrode
and a platinum wire auxiliary electrode in 10 mL of 0.1 M [n-
Bu4N][PF6] as the supporting electrolyte. IR compensation for solu-
(KBr): mCN
(
CN
41.95; H, 2.16; N, 3.76. Found: C, 41.85; H, 2.44; N, 3.61%.
2.2.2. Et4N[Ni(‘S3’)13CN] (Et4N1–13CN)
This complex was prepared following the literature procedure
13
employing K13CN [18]. IR (KBr): (
m .
) 2067 cmꢁ1 13C NMR (d6-ace-
CN
tone): d 122.2 (C5–‘S3’), 124.7 (Ni–13CN), 128.3 (C4–‘S3’), 128.8 (C3–
‘S3’), 130.1 (C6–‘S3’), 133.3 (C1–‘S3’), 156.3 (C2–‘S3’).
2.2.3. (‘S3’)Ni–CN–Fe[TptBu](2-Fe)
K1 (0.060 g, 0.161 mmol) in 5 mL of THF was added dropwise to
a stirring solution of [TptBu]FeBr (0.084 g, 0.161 mmol) in 20 mL of
THF. The resulting green–brown solution was to stirred overnight
after which time a precipitate formed and the solution color chan-
ged to light brown. The solution was filtered through Celite and re-
duced to dryness in vacuo. The dry solid was washed with ethyl
ether and a light brown solid was collected by filtration. X-ray
quality crystals were grown by slow evaporation of concentrated
benzene or THF solution. Yield: 0.076 g, 61%. 1H NMR (C6D6): d
67.4 (s, 4-pz), 29.5 (s, 5-pz), 18.4 (br, B–H), 9.5 (s, C6H4), 9.1 (s,
C6H4), 8.3 (s, C6H4), 7.4 (s,C6H4), 3.3 (br, 3-pz-C(CH3)3). UV–Vis
(THF), kmax
412 (1900),
BH) = 2493 cmꢁ1
(e
, cmꢁ1 Mꢁ1): 262 (29 000), 290 (29 000), 320 (sh),
640
(160).
IR
(KBr):
(
m
CN) = 2120 cmꢁ1
B. Anal. Calc. for
,
(m
.
leff (solid, 294 K) = 5.4(1)
l
C34H42BFeN7NiS3: C, 53.01; H, 5.49; N, 12.72. Found: C, 52.75; H,
5.20; N, 12.36%.
2.2.4. (‘S3’)Ni–CN–Co[TptBu] (2-Co)
To a stirring solution of [TptBu]CoI (0.100 g 0.176 mmol) in
20 mL of THF was added K1 (0.066 g 0.176 mmol) in 5 mL THF.
The solution was stirred under N2 overnight during which time
the color changed from blue to dark green. The solution was fil-
tered through Celite to remove the white precipitate (KI). The
resulting deep green solution was dried in vacuo affording a green
solid. X-ray quality crystals were grown by cooling a concentrated
toluene solution to ꢁ25 °C for 48 h. Yield: 0.121 g, 89%. 1H NMR
(C6D6): d 75.6 (s, 4-pz), 34.8 (s, 5-pz), 7.5 (s, C6H4), 7.3 (s, C6H4),
1
Note added in proof: Following submission of this manuscript, a report of a Ni-
CN-Fe complex appeared: D. Huang, L. Deng, J. Sun, R.H. Holm, Inorg. Chem 48 (2009)
asap, doi:10.1021/ic900494u.

M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
4555
7.1 (br, 3-pz-C(CH3)3), 6.8 (s, C6H4), 6.3 (s, C6H4), ꢁ5.8 (br, B–H).
UV–Vis (THF), kmax
, cmꢁ1 Mꢁ1): 262 (28 000), 290 (29 000),
321 (sh), 405 (2500), 573 (970), 604 (1287), 623 (1337). IR (KBr):
CN) = 2132 cmꢁ1, ( BH) = 2494 cmꢁ1
leff (solid, 293 K) = 4.2(1)
B. Anal. Calc. for C37.50H46BCoN7NiS3: C, 54.96; H, 5.65; N, 11.96.
(0.080 g,. 0.16 mmol) in 20 mL of acetonitrile. The solution was
stirred overnight, filtered through Celite and the solvent was re-
moved in vacuo. The off-white solid was extracted with THF, fil-
tered through Celite and reduced to dryness under vacuum
affording a white solid. Crystals were grown by vapor diffusion
of pentane into a benzene solution or alternatively, by slow evap-
oration of concentrated acetonitrile solution. Crystalline Yield:
0.060 g, 75%. 1H NMR (CD3CN): d 7.31 (s, o-(C6H5)B, 4 H), 7.04 (t,
m-(C6H5), 4 H), 6.89 (t, p-(C6H5)B, 2 H), 3.13 (q, (CH2CH3)4N, 8 H),
1.87 (s, CH2B, 12 H), 1.29 (s, (C(CH3)3, 54 H), 1.18 (t, (CH2CH3)4N,
12 H). 13C NMR (CD3CN): d 132.9 ((o-C6H5)B), 127.4 ((m-C6H5)B),
124.2 ((p-C6H5)B), 53.0 ((CH2CH3)4N), 43.7 (CH2B), 30.2 (C(CH3)3),
(
e
(
l
m
m
.
Found: C, 54.40; H, 5.59; N, 12.16%.
2.2.5. (‘S3’)Ni–CN–Ni[TptBu] (2-Ni)
K1 (0.087 g, 0.236 mmol) in 10 mL of THF was added dropwise
to a stirring solution of [TptBu]NiCl (0.111 g, 0.236 mmol) in 10 mL
of THF. The resulting solution was stirred for 6 h during which time
the color turned dark red. The red solution was filtered through
Celite and reduced to dryness in vacuo. The solid was washed with
ethyl ether and a dark red product was collected by filtration. X-ray
quality crystals were grown by evaporation of a concentrated
1,3,5-trimethylbenzene solution under a slow stream of N2. Yield:
0.044 g, 25%. 1H NMR (C6D6): d 86.5 (s, 4-pz), 18.7 (s, 5-pz), ꢁ13.0
(br, B–H), 7.6 (s, C6H4), 7.4 (s, C6H4), 7.3 (s, C6H4), 7.1 (s,C6H4), 5.75
7.75 ((CH2CH3)4N). UV–Vis (THF), kmax
(e
, cmꢁ1 Mꢁ1): 220 (sh). IR
(KBr): (
m
CN) = 2121 cmꢁ1. Anal. Calc. for C51H96BCuN2S6: C, 56.80;
H, 8.97; N, 2.59. Found: C, 56.86; H, 8.68; N, 2.46%.
2.2.9. [TptBu]Zn13CN
This complex was prepared following the reported procedure
(br, 3-pz-C(CH3)3). UV–Vis (THF), kmax
289 (18 000), 320 (sh), 409 (1300), 559 (270), 809 (67). IR (KBr):
CN) = 2124 cmꢁ1, ( BH) = 2494 cmꢁ1
leff (solid, 293 K) = 3.6(2)
B. Anal. Calc. for C38.50H48BN7Ni2S3: C, 55.50; H, 5.80; N, 11.76.
(e
, cmꢁ1 Mꢁ1): 264 (14 500),
[22] employing K13CN. IR (KBr): ( BH) = 2526 cmꢁ1 13C NMR
m .
(C6D6): d 30.92 (C(CH3)3, 32.1 (C(CH3)3, 102.5 (4-pz), 135.9 (5-pz),
(m
m
.
137.7 (Zn–13CN), 165.1 (3-pz).
l
Found: C, 55.93; H, 5.80; N, 10.87%.
2.3. Density functional calculations
2.2.6. (‘S3’)Ni–CN–Zn[TptBu] (2-Zn)
Calculations were performed using GAUSSIAN03 [24]. Input atomic
coordinates were derived from crystallographic determined struc-
tures. For computational expediency, the tert-butyl moieties of
the TptBu ligand were replaced with hydrogens. Molecular struc-
tures were optimized using the BLYP functional, Ahlrich’s polarized
triple-f quality basis set (TZVP) on zinc, nickel, copper, cyanide, sul-
fur, and the metal-bound pyrazolyl-nitrogen, and polarized split va-
lence basis set on remaining atoms. Calculations using the BP86
method were also performed and yielded qualitatively similar re-
sults. Density fitting was employed and geometries were optimized
to tight convergence criteria, which necessitated the use of tight
SCF cutoffs and an ultrafine integration grid. To verify that each
optimized structure was in fact a stationary point on the potential
energy surface, analytical frequency calculations were performed,
and no imaginary frequencies were obtained. Calculations aimed
at the reliability of the wave functions with respect to open shell
instabilities were also performed, and uncovered no lower energy
solutions. Single point calculations were performed using the hy-
brid functional B3LYP with a balanced triple-f quality basis set,
6ꢁ311+g(d), on all atoms. The final energy was the sum of the elec-
tronic energy from the single point calculations and the thermal
corrections from analytical frequency calculations at 298.15 K.
K1 (0.077 g, 0.207 mmol) in 10 mL of THF was added dropwise
to a stirring solution of [TptBu]ZnCl (0.100 g, 0.207 mmol) in 20 mL
of THF. The resulting brown colored solution was stirred overnight,
then filtered through Celite and reduced to dryness in vacuo. The
solid was extracted with 1,3,5-trimethylbenzene. The soluble
material was isolated by filtration through Celite and dried in va-
cuo. X-ray quality crystals were grown by slow evaporation of a
concentrated methylene chloride solution. Yield: 0.069 g, 43%. 1H
NMR (C6D6): d 7.61 (d, C6H4, 2 H), 7.21 (d, C6H4, 2 H), 7.18 (s, 5-
pz, 3 H), 6.79 (t, C6H4, 2 H), 6.59 (t, C6H4, 2 H), 5.69 (s, 4-pz, 3 H),
1.51 (s, 3-pz-C(CH3)3, 27 H). 13C NMR (C6D6): d 165.8 (3-pz),
154.8 (C2–‘S3’), 153.6 (Ni–13CN–Zn), 136.6 (5-pz), 136.2 (C1–‘S3’),
133.2 (C6–‘S3’), 130.4 (C3–‘S3’), 129.0 (C4–‘S3’), 122.5 (C5–‘S3’),
103.0 (4-pz), 32.4 (C(CH3)3), 31.2 (C(CH3)3). UV–Vis (THF), kmax
cmꢁ1 Mꢁ1): 321 (sh), 400 (745). IR (KBr): mCN, 13CN) = 2152,
(2109) cmꢁ1 mBH 2494, (2497) cmꢁ1
Anal. Calc. for
(e,
(
,
.
C35H44BCl2N7NiS3Zn: C, 48.61; H, 5.13; N, 11.33. Found: C, 48.68;
H, 5.36; N, 11.46%.
2.2.7. [NEt4][(‘S3’)Ni–CN–Cu(TptBu)] (2-Cu)
NEt41 (0.048 g, 0.103 mmol) in a 5 mL THF/CH3CN (4:1) solu-
tion was added dropwise to a stirring solution of [TptBu]Cu(NCCH3)
(0.053 g, 0.110 mmol) in 20 mL THF. The orange solution was stir-
red for 8 h, filtered through Celite, and reduced to dryness in vacuo.
The product was recrystallized by diffusion of pentane into a THF
solution. Crystalline yield: 0.066 g, 70%. 1H NMR (d6-acetone): d
7.82 (d, C6H4, 2 H), 7.36 (s, 5-pz, 3 H), 7.33 (d, C6H4, 2 H), 7.10 (t,
C6H4, 2 H), 6.97 (t, C6H4, 2 H), 5.84 (d, 4-pz, 3 H), 3.39 (q,
(CH2CH3)4N, 8 H), 1.48 (s, 3-pz-C(CH3)3, 27 H), 1.31 (t, (CH2CH3)4N,
12 H). 13C NMR (d6-acetone): d 161.6 (3-pz), 156.5 (C2–‘S3’), 133.4
(5-pz), 133.2 (C1–‘S3’), 130.1 (C6–‘S3’), 128.9 (C3–‘S3’), 128.5
(Ni–13CN–Cu), 128.3 (C4–‘S3’), 122.2 (C5–‘S3’), 100.1 (4-pz), 52.9
((CH2CH3)4N), 32.6 (C(CH3)3), 31.3 (C(CH3)3), 7.62 ((CH2CH3)4N).
2.4. X-ray crystallography
X-ray structural analysis for 2-Feꢀ0.5(thf), and 2-Znꢀ(pentane):
Crystals were mounted using viscous oil on glass fibers and cooled
to the data collection temperature. Data were collected on a Bru-
ker-AXS APEX CCD diffractometer with graphite-monochromated
Mo-K
a
radiation (k = 0.71073 Å). Unit cell parameters were ob-
, from three different sections
tained from 60 data frames, 0.3°
x
of the Ewald sphere. No symmetry higher than triclinic was ob-
served for 2-Feꢀ0.5(thf) and solution in the centrosymmetric space
group option yielded chemically reasonable and computationally
stable results of refinement. The systematic absences in the diffrac-
tion data are consistent with Cmc21 and Cmcm for 2-Znꢀ(pentane).
Only solution in the space group Cmc21 for 2-Znꢀ(pentane) yielded
chemically reasonable and computationally stable results of refine-
ment. The absolute structure parameter refined to nil indicating
that the true hand of the data for Znꢀ(pentane) had been deter-
mined. The data sets were treated with SADABS absorption correc-
tions based on redundant multiscan data [25]. The structures were
UV–Vis (THF), kmax
(e
, cmꢁ1 Mꢁ1): 368 (sh), 419 (1681). IR (KBr):
mCN, (m
) = 2127, (2081) cmꢁ1 (weak), mBH 2430, (2432) cmꢁ1. Anal.
CN
13
Calc. for C42H62BCuN8NiS3: C, 55.54; H, 6.88; N, 12.34. Found: C,
55.59; H, 6.95; N, 12.48%.
2.2.8. Et4N{[(2-PhTttBu)Cu]2(
l-CN)} 3
[NEt4]CN (0.012 g, 0.076 mmol) was dissolved in 1 mL of aceto-
nitrile and added to a stirring solution of [PhTttBu]Cu(NCCH3)

4556
M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
solved using direct methods and refined with full-matrix, least-
squares procedures on F2. Two symmetry unique but chemically
identical compound molecules and one thf solvent molecule are lo-
cated in the asymmetric unit of 2-Feꢀ0.5(thf). The compound mol-
ecule lies on a mirror plane in 2-Znꢀ(pentane). Disordered solvent
molecules of crystallization were treated as diffused contributions
[26]. Slight disorder, caused by the rotation of the t-butyl groups,
which cannot be modeled because of the <1 Å difference in the dis-
ordered atomic positions, yielded less than optimal Ueq ranges. All
non-hydrogen atoms were refined with anisotropic displacement
parameters. All hydrogen atoms were treated as idealized contri-
butions. Structure factors and anomalous dispersion coefficients
are contained in the SHELXTL 6.12 program library [25]. The crystal-
lographic details for isostructural compounds 2-Coꢀ2(toluene), 2-
Niꢀ0.5(mesitylene), 2-Znꢀ2(CH2Cl2) and 2-Cuꢀ0.5(pentane)ꢀ(thf)
are reported in the Supplementary material.
Fig. 1. Molecular structure of 2-Fe. Thermal ellipsoids are drawn to 30% probability
and hydrogen atoms have been omitted for clarity.
3. Results and discussion
3.1. Synthesis of cyanide-bridged binuclear complexes 2-Fe, 2-Co, 2-
Ni, 2-Zn
summarized in Table 1. Plots of the isostructural complexes along
with X-ray collection and refinement data are available in the Sup-
plementary material. The general structural features of the com-
plexes are quite similar, supporting the compositions and
structures assigned based on spectroscopic evaluation, vide infra.
The [TptBu]M fragment maintains its pseudo-tetrahedral geometry
ligated to the linear cyanide bridge via the nitrogen atom. The cya-
nide is coordinated via the carbon atom to the square planar nickel
ion. The M–NC distances fall within a very narrow range, decreas-
ing across the period, from 1.980(2) Å for 2-Fe to 1.925(2)/
1.927(2) Å for 2-Zn. The M–N bond lengths are consistent with
the M2+ oxidation state assignment. The M–N–C and Ni–C–N an-
gles are all within a few degrees, at most, to linear. The Ni–CN dis-
tances at ꢂ1.850 Å, are insensitive of the identity of the second
metal, whereas, the C–N distances fall within 1.135 to 1.163 Å, void
of any apparent trends. The nickel ‘S3’ coordination sphere is com-
posed of two thiolate sulfur donors and one thioether sulfur donor
with the tridentate ligand exhibiting a butterfly pucker as noted by
Sellmann et al. [18] in which the thiolate sulfur atoms lie slightly
below the idealized square plane and the thioether sulfur atom lies
slightly above the plane.
The synthesis of a series of cyanide-bridged binuclear com-
plexes proceeds by the coupling of two independently ligated me-
tal complexes, Scheme 1. We chose the monoanion [Ni(CN)(‘S3’)]ꢁ,
1ꢁ, (‘S3’ = bis(2-mercaptophenyl)sulfide) first prepared by Sell-
mann et al. [18] due to its sulfur-rich ligation and square planar
stereochemistry, the latter characteristic ensuring a diamagnetic
complex. As emphasized by Holm is his recent studies [27], the
Ni site in COdH has approximate square planar coordination de-
spite being contained within the NiFe3S4 core. The nickel(II) ion
is ligated by the tridentate thioether-dithiolate ligand (‘S3’) and a
cyanide ligand. The second portion of the cyanide-bridged com-
plexes consists of the [TptBu]M fragment. The sterically demanding
[TptBu] ligand [19] allows for the isolation of monomeric metal ha-
lide complexes, [TptBu]MX, (M = Fe, Co, Ni, Zn; X = Cl, Br or I) as
suitable precursors to the targeted binuclear complexes.
Initial attempts to synthesize the cyanide-bridged binuclear
complexes using Et4N1 resulted only in recovery of the starting
materials. Consequently, the synthetic protocol of Sellmann was
altered to permit for isolation of K1 reasoning that the potassium
would add to the driving force of the coupling reaction via precip-
itation of KCl. Indeed, the room temperature reaction of K1 with
[TptBu]MX in THF produces the cyanide-bridged binuclear com-
plexes (‘S3’)Ni–CN–M[TptBu] (M = Fe (2-Fe), Co (2-Co), Ni (2-Ni),
Zn (2-Zn)) in variable yields. X-ray diffraction studies confirmed
the formation of the isostructural series of binuclear cyanide-
bridged complexes, vide infra.
3.3. Spectroscopy of cyanide-bridged binuclear complexes 2
The room temperature proton NMR spectra of 2-Fe, 2-Co and 2-
Ni exhibit well dispersed paramagnetically-shifted signals for pro-
tons of the [TptBu]M fragments with the number of signals indicat-
ing three-fold symmetry of the ligand. The t-butyl groups of the
[TptBu] ligands appear as broad signals downfield of the diamag-
netic ligand, d = 3.3 for 2-Fe, 7.1 for 2-Co and 5.7 for 2-Ni. The
4-pz and 5-pz protons of the pyrazoyl rings and the B–H proton ex-
hibit wide chemical shift dispersions, similar to those for analo-
gous [TptBu]MX complexes. In each complex, the ‘S3’ ligand
displays four signals consistent with the mirror symmetry. Inter-
3.2. X-ray structures of cyanide-bridged binuclear complexes 2
A representative thermal ellipsoid plot (2-Fe) of the molecular
structures of 2 is depicted in Fig. 1, and selected metric parameters
K
S
S
N
N
N
N
N
N
N
THF
CN
Ni
S
S
+
H
M
X
M
B
NC
Ni
S
N
H B
S
–KX
N
N
N
N
K 1
2
Scheme 1.

M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
4557
Table 1
The solid-state magnetic moments of solid samples of 2-Fe, 2-
Co and 2-Ni were determined using a magnetic susceptibility bal-
ance. In each case, the magnetic moment was consistent with the
Selected bond lengths (Å) and bond angles (°) for 2-Fe, 2-Co, 2-Ni, 2-Zn, and 2-Cu.
2-Fe (2 independent molecules)
Length (Å)
Angle (°)
expected ground states as follows: 2-Fe,
leff = 5.4(1) lB for S = 2;
Fe1–N1
Fe1–Npz (avg)
Ni1–C1
Ni1–S1
Ni1–S2
Ni1–S3
C1–N1
Fe2–N8
Fe2–Npz (avg)
Ni2–C35
Ni2–S4
1.980(5)
2.064
Fe1–N1–C1
Ni1–C1–N1
175.5(5)
177.5(5)
2-Co, eff = 4.2(1) lB for S = 3/2; 2-Ni, eff = 3.6(2) lB for S = 1.
l
l
1.850(6)
2.181(2)
2.110(2)
2.159(2)
1.145(7)
1.980(5)
2.057
1.850(6)
2.181(2)
2.111(2)
2.177(2)
1.135(7)
The electronic absorption spectra of 2-Fe, 2-Co, 2-Ni and 2-Zn
in THF displays two high energy Ni Sthiolate and Ni Sthioether li-
gand-to-metal charge transfer (LMCT) bands located at ca. 260 and
290 nm and a Ni CN metal-to-ligand charge transfer band
(MLCT) at ꢂ400 nm. These assignments are based on a comparison
with the spectrum of the anion, [(‘S3’)Ni(CN)]ꢁ, which displays
nearly identical bands. The square planar nickel also displays a
weak ligand field transition located at ꢂ600 nm; this feature is ob-
served in 2-Fe, 2-Ni and 2-Zn [31]. In 2-Co the Ni ligand field band
is obscured by the Co ligand field transitions at 573, 604, and
623 nm, energies typical of tetrahedral cobalt(II) complexes [32].
2-Ni also displays two absorption features, located at 559 and
809 nm, assigned as ligand field transitions for the tetrahedral
nickel site.
Fe2–N8–C35
Ni2–C35–N8
172.1(5)
178.1(6)
Ni2–S5
Ni2–S6
C35–N8
2-Co
Co–N7
Co–Npz (avg)
Ni–C34
Ni–S1
Ni–S2
Ni–S3
1.949(3)
2.014
Co–N7–C34
Ni–C34–N7
174.4(3)
179.1(3)
1.846(3)
2.159(1)
2.106(1)
2.172(1)
1.163(4)
3.3.1. Infrared spectra of 2
C34–N7
Terminal M–CN linkages typically exhibit sharp, intense mCN
modes between 2000 and 2200 cmꢁ1 [33]. The 13CN isotopically-
labeled complexes were synthesized in order confirm spectral
assignments and to address the issue of CN linkage isomerism in
the bridged complexes as detailed later in this paper.
Upon formation of the cyanide-bridged complexes, the mCN fre-
quency shifts to higher energy, Fig. 2 and Tables S3 and S4 (Supple-
mentary material). The frequency shift ranges from 14–46 cmꢁ1
depending on the identity of the metal ion with the order increas-
2-Ni
Ni–N5
1.954(7)
1.992
Ni1–N5–C14
Ni2–C14–N5
178.5(7)
177.2(8)
Ni1– Npz (avg)
Ni2–C14
Ni2–S1
Ni2–S1A
Ni2–S2
1.842(8)
2.174(2)
2.174(2)
2.132(2)
1.135(9)
C14–N5
2-Zn (2 independent molecules)
Zn1–N7
Zn1–Npz (avg)
Ni1–C34
Ni1–S1
Ni1–S2
Ni1–S3
C34–N7
Zn2–N14
Zn2–Npz (avg)
Ni2–C68
1.925(4)
2.018
Zn1–N7–C34
Ni1–C34–N7
179.0(4)
179.2(5)
ing, 2-Fe < 2-Ni < 2-Co < 2-Zn. Increases in mCN values for M–CN–M’
compared with M–CN are well documented [5,33,34]. Typically the
mCN increase has been rationalized as a consequence of the dona-
1.850(4)
2.159(1)
2.131(1)
2.179(1)
1.135(6)
1.927(4)
2.014
1.842(4)
2.164(2)
2.124(1)
2.168(2)
1.135(6)
tion of
r-electron density from the antibonding lone pair on nitro-
gen. The depopulation of the antibonding orbital strengthens the
CN triple bond, leading to an increase in mCN [1]. The kinematic ef-
fect [35], or the restraint of the CN group due to the attachment of
a second mass, has also been offered to explain the observed
Zn2–N14–C68
Ni2–C68–N14
174.8(4)
176.3(5)
Ni2–S4
Ni2–S5
Ni2–S6
C68–N14
2-Cu
Cu–N1
Cu–Npz (avg)
Ni–C13
Ni–S1
Ni–S2
Ni–S3
1.890(2)
2.100
1.856(2)
2.1685(8)
2.1108(7)
2.1814(8)
Cu–N1–C13
Ni–C13–N1
177.8(2)
178.8(2)
estingly, in 2-Fe these signals also exhibit small paramagnetic
shifts. These slight downfield shifts are attributed to a through
space dipolar shift interaction [28] caused by the paramagnetic
iron. It is unlikely that a Fermi contact interaction is at the origin
of the paramagnetic contributions to the phenyl proton chemical
shifts due to the fact that the pathway through which spin could
be delocalized is common amongst 2-Fe, 2-Co, and 2-Ni. Specifi-
cally, in each complex, the magnetic orbitals are involved in cya-
nide binding, thus a Fermi contact shift would be expected for
each complex. A dipolar shift mechanism is reasonable when con-
sidering (i) the smaller spin states of 2-Co (S = 3/2) and 2-Ni (S = 1)
compared to 2-Fe (S = 2) and (ii) the magnetic anisotropy as re-
flected in the molecular g-tensors, specifically, g||, estimated at
ꢂ2.2 for 2-Ni [29], 5.2 for 2-Co, and ꢂ8.5 for 2-Fe [30]. The 1H
NMR spectrum of diamagnetic 2-Zn is consistent with its structure
as determined by X-ray diffraction.
Fig. 2. Infrared spectra for the complexes K1 and
frequency range.
2
in the 2000–2600 cmꢁ1

4558
M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
increase of
m
CN in bridging complexes. However, a recent study has
reversible cathodic wave in the range ca. ꢁ1.0 to ꢁ1.2 V. Given the
similarity among these complexes and with the monomeric pre-
cursor, K[(‘S3’)Ni(CN)], this feature is ascribed to the Ni(II)/Ni(I)
couple of the square planar nickel site, although (‘S3’) ligand-based
reduction affords an alternative site to house the added electron.
The CV of 2-Ni displays a cathodic reduction wave at ꢁ0.96 V as-
signed as the Ni(II)/Ni(I) reduction for [TptBu]Ni. For calibration,
the reduction wave observed for [TptBu]NiCl is at ꢁ0.91 V. The
reduction potential for the tetrahedral site in 2-Ni is lower than
the square planar (‘S3’)Ni site due to the ability of the tetrahedral
Ni site to support the Ni(I) oxidation state because of the lower en-
ergy of an available metal orbital (t2-like symmetry) as compared
found that this effect has been overestimated and suggests that the
kinematic effect does not contribute significantly to the observed
mCN frequency increase [5]. In the present series of complexes, a
correlation between the mCN value and M–N bond length is noted,
wherein the complex with the shortest M–N bond length has the
highest mCN value. This trend is consistent with the analysis above,
as removal of electron density from the cyanide N lone pair
strengthens the CN bond.
A number of factors can influence CN vibrational energy includ-
ing metal oxidation state and C–N–M bond angle. Notable exam-
ples in biological systems include studies on COdH and
cytochrome c oxidase. Infrared spectroscopic studies of COdH en-
zyme incubated with cyanide demonstrated a very low energy
CN stretch at 2037 cmꢁ1 [36]. It was postulated that the cause of
the low energy vibration was due to carbon-bound cyanide coordi-
to the higher energy LUMO (dx2 ) of the square planar site.
ꢁy2
2-Co exhibits an irreversible cathodic wave at ꢁ1.73 V assigned
to the Co(II)/Co(I) reduction based on comparison with [TptBu]CoI.
3-Fe exhibits two irreversible cathodic waves at ꢁ2.02 and
ꢁ2.33 V, tentatively assigned as the Fe(II)/(I) and Fe(I)/Fe(0) reduc-
tions, respectively.
nation to a strongly p-donating metal (suggested to be Fe) in com-
bination with a bent M–CN–M’ bridge. Supporting the latter
hypothesis, similarly low mCN energies were observed in synthetic
models of an inhibited form of cytochrome c oxidase containing
a Fe(III)–CN–Cu(II) linkage, with the frequency decreasing up to
60 cmꢁ1 as the C–N–Cu bond angle decreased [34].
3.3.3. Mössbauer spectroscopy of 2-Fe
The Mössbauer spectrum of 2-Fe at 4.4 K in a magnetic field of
0.03 T exhibits a quadrupole doublet with an isomer shift,
The mBH vibrational mode is sensitive to electronic changes in
the metal coordination sphere and can be used to indicate binu-
clear complex formation. The mBH of the cyanide-bridged com-
plexes shifts to lower energy as compared to the energy for the
[TptBu]MX starting materials. Interestingly, the mBH value of ca.
2494 cmꢁ1 is invariant among the complexes within the series.
This significant change in stretching frequency upon bridge forma-
d = 0.94(3) mm/s and quadrupole splitting DEQ = 2.61(2) mm/s.
The large isomer shift and quadrupole splitting are in the ranges
corresponding to high-spin ferrous pseudo-tetrahedral species
[38]. Mössbauer parameters of the C-cluster of COdH ferrous com-
ponent II, FCII, (the unique iron) are
DEQ = 2.82 mm/s and
d = 0.82 mm/s [39], suggesting a pentacoordinate iron [40]. Inter-
estingly, upon addition of cyanide to COdH, the quadrupole split-
tion indicates that the [(‘S3’)Ni(CN)]ꢁ fragment is a weaker
r
-do-
ting changed to
remained the same.
DEQ = 2.53 mm/s, while the isomer shift
nor than the halide substituents of the [TptBu]MX precursors,
resulting in reduction of negative charge at boron [37].
3.3.4. Electron paramagnetic resonance of 2-Co
3.3.2. Electrochemical characterization of cyanide-bridged binuclear
complexes 2
Cyclic voltammetry experiments were conducted to establish
the redox characteristics of the series of binuclear cyanide-bridged
complexes. These electrochemical data are summarized in Table 2
(potentials referenced versus NHE). Each complex exhibits a quasi-
The EPR spectrum of 2-Co recorded in toluene at 5 K exhibits a
broad rhombic signal with effective g-values, g = 5.3, 4.2 and 2.2.
These broad features are typical for high-spin, S = 3/2, tetrahedral
cobalt(II) due to the 59Co (I = 7/2) nuclear spin. Similar, broad
EPR spectra have been reported for the mononuclear complex
[PhBPiPr]CoI with g ꢃ 4.8 and g ꢃ 2.2 [41]. A similar EPR signal
was reported for the cobalt substituted rubredoxin model complex
(Et4N)2[Co(SC6H4-p-Cl)4] with effective g-values, g = 4.8, 3.6 and
2.08 [42].
Table 2
Summary of CV data for K1 and the series of complexes 2.
3.4. Cyanide-bridged binuclear complex Et4N{(‘S3’)Ni–CN–Cu[TptBu]}
2-Cu
E1/2, V Ni(II)/Ni(I)
Ipa/Ipc
Epc, V
K1
2-Fe
ꢁ1.26
ꢁ1.24
0.40
0.40
ꢁ
ꢁ2.02
ꢁ2.33
ꢁ1.73
ꢁ0.96
ꢁ
Synthetic efforts to prepare the copper(II) cyanide-bridged
binuclear complex commencing with [TptBu]CuCl proved unsuc-
cessful. Thus, we turned to copper(I) precursors given the stability
and established reactivity of copper(I) nitriles. The ionic complex
Et4N{(‘S3’)Ni–CN–Cu[TptBu]}, 2-Cu, was isolated in good crystalline
yield following the addition of Et4N1 to a THF solution of [TptBu]-
Cu(NCCH3) [20] (Scheme 2).
2-Co
2-Ni
2-Zn
ꢁ1.25
ꢁ1.18
ꢁ0.91
ꢁ1.21
0.21
0.11
0.54
0.15
ꢁ
Electrochemical experiments were performed in THF at a scan rate of 200 mV/s.
Reduction potentials are referenced to internal Fc+/Fc couple at +801 mV vs. NHE.
Et4N
Et4N
S
S
N
N
N
N
N
N
THF
CN
Ni
S
S
+
Cu NCCH3
Cu
H
B
NC
Ni
S
N
H B
N
S
N
N
N
N
2-Cu
Et4N1
Scheme 2.

M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
4559
Table 3
3.4.1. X-ray structure of 2-Cu
Selected DFT optimized bond lengths and vibrational frequencies for the truncated
model of 2-Cu.
The molecular structure of the anion of 2-Cu, which is isostruc-
tural with the other, neutral derivatives in this series is provided in
the Supplementary material with selected metric parameters for
the anion contained in Table 1. The copper coordination sphere is
composed of three nitrogens from the [TptBu] ligand and a nitrogen
of the cyanide ligand. The average Cu–N bond length from the
nitrogen donors of the [TptBu] ligand is 2.100(2) Å, notably longer
than the average Cu–N bond length of 2.064(5) Å in [TpiPr2]-
Cu(NCCH3) [43]. The Cu–NC distance is 1.890(2) Å, the shortest
within the series of complexes reported here, and similar to the
Cu–N bond of coordinated acetonitrile in [TpiPr2]Cu(NCCH3) at
1.875(6) Å. The Ni–CN–Cu linkage is linear. The square planar nick-
el(II) coordination sphere is composed of two thiolate sulfur do-
nors at 2.1685(8) and 2.1814(8) Å and a thioether sulfur at
2.1108(7) Å. The fourth coordination site is occupied by the cya-
nide ligand with a Ni–C bond distance of 1.856(2) Å. The C–N bond
length of the bridging cyanide ligand in 5 was found to be
1.148(3) Å, matching the C–N bond length of 1.153 Å in Me4N1
[18].
{[(Tp)Cu]–NC–[Ni(‘S3’)]}ꢁ1
C–N (Å)
1.1760
2106.7 (3)
2448.6
mCN (cmꢁ1) ([I])
mBH (cmꢁ1
)
Ni–CCN (Å)
Cu–NCN (Å)
Cu–NPz (Å)
Ni–Strans (Å)
1.8639
1.8994
2.1437, 2.1523, 2.1579
2.1878
3.4.3. DFT analysis of 2-Cu
DFT calculations were performed on 2-Cu to clarify the unusu-
ally weak mCN intensity. In the computational model, the tert-butyl
groups of the [TptBu] ligand were replaced with hydrogens, i.e. [Tp].
The optimized structure closely resembles the experimentally
determined structure of 2-Cu (Table 3). The key DFT derived vibra-
tional bands,
m m
BH = 2449 cmꢁ1 and CN = 2107 cmꢁ1 are in good
3.4.2. Spectroscopy of 2-Cu
agreement with the experimental values, 2429 and 2127 cmꢁ1
,
The electronic absorption spectrum of diamagnetic 2-Cu in THF
solution displays the bands of the nickel fragment: two high en-
ergy Ni Sthiolate and Ni Sthioether LMCT bands at 261 and
290 nm. An absorption feature at 419 (1680 Mꢁ1 cmꢁ1) nm is as-
signed to a Ni CN MLCT as seen for the other complexes of this
series.
respectively. The weak mCN intensity is reflected in part, in the va-
lue of the dipole moment of the anion. A comparison of the calcu-
lated dipole moments of 2-Cu with 2-Zn and its linkage isomer 2-
Zn0 reveals a significantly lower dipole moment,
l
= 1.6440 D in
the former compared to
l = 9.2796 and 9.6028 D, in the latter
two complexes. The correlation between the gas-phase DFT calcu-
lations and the experimental solid-state IR measurements indi-
cates that this phenomenon is not a consequence of ion-pairing
in the solid state.
The solid state IR spectrum of 2-Cu displays a very weak mCN
band at 2127 cmꢁ1 and a mBH mode at 2429 cmꢁ1. The frequencies
of these two modes are in accord with the frequencies for the neu-
tral binuclear analogues. To confirm assignment of the unusually
weak
ple displays a weak
m
CN mode, 13CN-enriched 2-Cu–13CN was prepared. This sam-
3.5. Cyanide-bridged binuclear complex Et4N{[(2-PhTttBu)Cu]2-
m
CN mode at 2081 cmꢁ1; the shift in energy is in
(l-CN)} 3
quantitative agreement with reduced mass considerations, i.e. ca.
2082 cmꢁ1. The unusually weak mCN signal intensity indicates a
small dipole moment change associated with the vibration. As a
consequence of the overall negative charge of 2-Cu the charge dis-
tribution is such that the normally large CN dipole is significantly
diminished. This assertion is supported by DFT calculations, vide
infra.
Following the preparation of cyanide-bridged binuclear com-
plexes detailed in the preceding sections, efforts to extend this ser-
ies of complexes using the tetrahedral metal precursors,
[PhTttBu]MCl (PhTttBu = phenyl(tris(tert-butylthio)methyl)borate,
M = Fe, Co, Ni, Zn) [45] in lieu of [TptBu]MX were investigated.
These condensation reactions proved uniformly unsuccessful due,
in part to lability of the [PhTttBu] ligand under the reaction condi-
tions. Subsequent synthetic efforts focused on the analogous reac-
tion in Scheme 2, substituting the monovalent copper complex
[PhTttBu]Cu(NCCH3) for [TptBu]Cu(NCCH3). This reaction afforded
The 13C NMR spectrum of 2-Cu–13CN contains an intense
singlet at d = 128.5, ascribed to the bridging ligand. Interest-
ingly, this signal is shifted only slightly downfield of that
for the enriched starting material, Et4N1–13CN at d = 124.7. In
contrast, the 13C NMR signal for 2-Zn–13CN, d = 153.6, exhibits
a 28 ppm downfield shift indicating that in neutral 2-Zn the
cyanide carbon is significantly deshielded relative to the
unanticipated products, Et4N{[(j l-CN)}, 3, and the
2-PhTttBu)Cu]2(
known nickel dimer [Ni(‘S3’)]2. The formation of 3 suggests that
the copper(I) complex extracts cyanide from the nickel complex,
perhaps via intermediacy of undetected {(‘S3’)NiCNCu[PhTttBu]}ꢁ.
The resulting (‘S3’)Ni fragments couple yielding [Ni(‘S3’)]2. The
independent reaction of [PhTttBu]Cu(NCCH3) with half an equiva-
lent of [NEt4]CN produces 3 in 75% yield.
monomeric precursor.
A
similar downfield shift of the 13C
NMR resonance has been observed upon bridge formation in
(CO)5W–CN–Cu(PPh3)3, d = 147.8, relative to the anionic precur-
sor, Na[(CO)5W(CN)], d = 136.4 [44]. As with the analysis of
the intensity difference of the mCN bands, the charge distribu-
tion in the anion of 2-Cu impacts this 13C NMR spectral
feature.
The cyclic voltammogram of 2-Cu exhibits an irreversible
Ni(II)/Ni(I) cathodic wave at ꢁ1.27 V, similar in value to this
reduction in the neutral complexes. An irreversible anodic wave
at 0.81 V is ascribed to the Cu(II)/Cu(I) couple. The irreversible
electrochemical behavior indicates that upon oxidation to cop-
per(II) 2-Cu is unstable or undergoes a structural change. While
it is perhaps surprising that the oxidized product is electrochem-
ically unstable, this result is consistent with our inability to inde-
pendently prepare the neutral Ni–CN–Cu(II) complex. A second
irreversible anodic wave is located at 1.19 V is tentatively ascribed
to (‘S3’) ligand oxidation.
3.5.1. Spectroscopy of 3
The proton NMR spectrum of 3 in d3-acetonitrile exhibited sin-
gle tert-butyl and methylene proton resonances, d = 1.29 and 1.87,
respectively, indicating magnetic equivalence of the thioether
groups consistent with either symmetric ligation of the [PhTttBu] li-
gand or that the thioether arms are in fast exchange on the NMR
timescale. Such phenomena have been previously noted, for exam-
ple in [
Compound 3 exhibits a medium intensity
j g
2-PhTttBu]Ni( 3-allyl) [46].
m
CN = 2121 cmꢁ1 for
the bridging CN ligand. This band is similar in energy to the weak
CN vibrational mode in the anionic 2-Cu located at 2127 cmꢁ1, but
with an intensity more characteristic of the other cyanide contain-
ing complexes.

4560
M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
S
Ni
S
S
Ni
S
N
N
N
N
N
N
S
S
Ni
S
THF
0.5
+
Zn CN
Zn
H
B
CN
N
H B
N
S
S
N
N
N
N
2-Zn
'
Scheme 3. Synthetic scheme for the (attempted) preparation of the CN linkage isomer, 2-Zn0.
Fig. 3. Molecular structures of the proposed 2-Zn0 refined alternatively as the linkage isomers 2-Zn (left) and 2-Zn0 (right) with thermal ellipsoids at 30%. Note the differences
in the thermal ellipsoids of the cyanide ligands.
3.6. Cyanide linkage isomers
Fig. 3. The 2-Zn model passed the Hirshfeld Rigid-Bond Test [47],
which is a test for the agreement of thermal parameters for adja-
cent, bonded atoms. The 2-Zn0 model failed this test because the
cyanide atoms (C200 and N50) were apparently misidentified. None-
theless, while these data support isomerization to the 2-Zn form,
detailed spectroscopic studies were undertaken in effort to confirm
the identity of this isomer.
Here we describe efforts to assemble the cyanide-bridged link-
age isomer of 2-Zn, denoted as 2-Zn0, for the purpose of establish-
ing its structural and spectroscopic characteristics in comparison
with 2-Zn. The strategy for the synthesis of the series of cyanide-
bridged binuclear complexes described above, highlights the ap-
proach of coupling of two metal complexes in which the cyanide
ligand is affixed to one of the precursors. Implicit in this strategy
is the assumption that under kinetic control, bridge formation pro-
ceeds without cyanide isomerization. With this consideration in
mind, reaction of [TptBu]MCN with [Ni(‘S3’)]2 should yield the cor-
responding isomer of 2 (at least under kinetic control). It is noted
that the related trimer, [Ni(‘S3’)]3 undergoes facile bridge rupture
in reactions with nucleophiles including KCN [23]. Very few four-
coordinate tetrahedral metal complexes with terminal cyanide li-
gands have been reported and attempted syntheses in our labora-
tory have been largely unsuccessful. This reality limits the scope of
this investigation. However, the application of known [TptBu]ZnCN
[22] permits us to test the validity of this approach (Scheme 3). As
the NiZn binuclear complexes are diamagnetic, a direct comparison
of the reaction products is amenable to NMR interrogation.
3.6.2. Spectroscopic analyses
The proton NMR spectra of the products derived for the two
independent synthetic routes are indistinguishable and do not
warrant further description. Table 4 contains 13C NMR and infrared
spectroscopic data for the 13CN labeled precursors and the prod-
ucts. The two products, 2-Zn and proposed 2-Zn0, exhibit identical
13CN chemical shifts, d = 153.3. This result clearly supports the
assertion that the products are identical. Further, the lack of a sec-
ond cyanide signal rules out the possibility of compositional disor-
Table 4
IR and 13C NMR spectroscopic data for K1, [TptBu]Zn13CN, 2-Zn and the material
proposed as 2-Zn0.
Experimental
CN, cmꢁ1
Calculateda
m
13CN, cmꢁ1
m
BH, cmꢁ1
d
13CN
m
3.6.1. X-ray diffraction analyses
Addition of [TptBu]ZnCN to suspension of [Ni(‘S3’)]2 in THF pro-
duced upon workup a brown solid in 48% yield. X-ray quality crys-
tals were grown by slow diffusion of pentane into a concentrated
1,3,5-trimethylbenzene solution. The proposed 2-Zn0 material
was recrystallized by pentane diffusion into concentrated mesity-
lene solution which yielded unsolvated crystals different from 2-
Znꢀ2CH2Cl2. The molecule lies on a crystallographic mirror plane
rendering each half of the complex metrically equivalent. The data
set was alternatively modeled as the two isomers, 2-Zn and 2-Zn0,
K1
2106
2065
2214b
2152
2109
2146
2105
–
2062
–
–
2107
–
–
–
–
K1-13CN
[TptBu]Zn13CN
2-Zn
125.9
137.7
–
153.3
–
2526
2495
2497
2493
2491
2-Zn–13CN
2-Zn0
2-Zn0–13CN
2103
153.3
a
Value calculated based on reduced mass differences assuming a simple dia-
tomic oscillator.
b
Ref. [49].

M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
4561
Table 5
exhibit an intense mCN stretching band. DFT computations reveal
a significantly lower dipole moment for 2-Cu than for the neutral
2-Zn and 2-Zn0 (and by extension 2-Fe, 2-Co and 2-Ni). The lack
of a significant dipole moment in the anion presumably translates
to a small dipole moment change upon C–N stretching.
Selected DFT optimized bond lengths and vibrational frequencies.
(‘S3’)Ni–NC–Zn[Tp]
(‘S3’)Ni–CN–Zn[Tp]
C–N (Å)
1.16754
2176.0 (253)
2514.0
1.9580
1.8637
n/a
n/a
1.17742
2114.3 (1327)
2514.7
n/a
n/a
1.8309
1.9083
2.0706, 2.0706, 2.0711
2.1889
mCN (cmꢁ1) ([I])
mBH (cmꢁ1
)
Two synthetic strategies were employed for the attempted syn-
thesis of cyanide-bridged linkage isomers. The products of two dif-
ferent 13CN labeled reactions intended to yield (‘S3’)Ni–CN–
Zn[TptBu] and (‘S3’)Ni–NC–Zn[TptBu] were analyzed. The structural
refinement of the proposed (‘S3’)Ni–NC–Zn[TptBu] complex exhibits
a better structural refinement as (‘S3’)Ni–CN–Zn[TptBu]. This struc-
tural assignment was supported by the Hirshfeld Rigid-Bond Test
[47] analysis, which clearly favored assignment as 2-Zn over 2-
Zn0. Nearly identical infrared and 13C NMR spectroscopic data pro-
vide additional evidence that (‘S3’)Ni–CN–Zn[TptBu] is produced in
both reactions. DFT optimized structures of (‘S3’)Ni–CN–Zn[Tp] and
(‘S3’)Ni–NC–Zn[Tp] suggest that (‘S3’)Ni–CN–Zn[Tp] is the thermo-
dynamically preferred structure by 5 kcal/mol supporting the for-
mation of (‘S3’)Ni–CN–Zn[TptBu] by both synthetic routes.
Zn–CCN (Å)
Ni–NCN (Å)
Ni–CCN (Å)
Zn–NCN (Å)
Zn–NPz (Å)
Ni–Strans (Å)
2.0757, 2.0759, 2.0766
2.1630
der (at least at concentrations detectable by NMR). In a relevant
report of cyanide-bridged linkage isomers, Darensbourg and
co-workers prepared (CO)5W–CN–Cu(PPh3)3 and (CO)5W–NC–
Cu(PPh3)3 by two different synthetic routes, the products affording
different spectral characteristics distinguishable by 13C NMR spec-
troscopy [44]. The latter was found to undergo rearrangement in
solution forming thermodynamically favored, (CO)5W–CN–
Cu(PPh3)3. Therefore, our 13C NMR data indicate only 2-Zn is gen-
erated here.
Acknowledgements
These studies were supported by the NIH (GM59191). Compu-
tational resources used in these studies were provided in part by
the Bioinformatics Center of the University of Arkansas for Medical
Sciences (NIH Grant P20 RR-16460 from the IDeA Networks of Bio-
medical Research Excellence (INBRE) Program of the National Cen-
ter for Research Resources). The Mössbauer spectrometer was
purchased through NSF grant CHE-0421116 (to CVP).
Solid-state IR spectra of the samples of 2-Zn and proposed 2-Zn0
(and their 13CN labeled forms) exhibit
mCN bands within four wave-
numbers of one another, Table 4. The congruence of these values
further suggests that the materials are identical. However, exam-
ples from the literature indicate that IR analysis of the mCN bands
does not necessarily provide a compelling method for establishing
cyanide-bridged linkages isomers. Notably, the two isomers
(CO)5W–CN–Cu(PPh3)3 and (CO)5W–NC–Cu(PPh3)3, distinguish-
able by 13C NMR spectroscopy, exhibit identical mCN stretching
bands [44]. Alternatively, Vahrenkamp’s [Cp(CO)2Fe–CN–
Mn(CO)2Cp] and [Cp(CO)2Fe–NC–Mn(CO)2Cp] isomers display mCN
modes that differ by 50 cmꢁ1 [48]. The similar mCN frequencies
for 2-Zn and proposed 2-Zn0 when considered with the X-ray
and 13C NMR data point to formation of a single isomer, 2-Zn.
Appendix A. Supplementary material
CCDC 718982, 718983, 718983, 718984, 718985, and 718986
contains the supplementary crystallographic data for [2-
Feꢀ0.5(thf)], [2-Coꢀ2(toluene)], [2-Niꢀ0.5(mesitylene)], (2-Zn),
[2-Znꢀ(pentane)], and [2-Cuꢀ0.5(pentane)ꢀ(thf)]. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via http://www.ccdc.cam.ac.uk/data_request/cif.
3.6.3. DFT calculations of the linkage isomers, 2-Zn and 2-Zn0
To further corroborate the experimental data indicating that the
cyanide-bridged linkage isomer 2-Zn is the thermodynamically
more stable isomer, a computational study was undertaken. The
computational models used [Tp] as a truncated version of [TptBu].
The DFT optimized structures of the linkage isomers afforded over-
all slightly longer calculated bond distances compared to the X-ray
crystallographic data (Table 5). The DFT optimized frequency anal-
ysis reveals that significant differences in mCN values are expected,
with 2-Zn ca. 46 cmꢁ1 lower in energy. These results support the
conclusion that a single product is formed from the two indepen-
dent synthetic routes. Further, the 2-Zn isomer is calculated to
be 5 kcal/mol lower in energy.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.ica.2009.05.042.
References
[1] K.R. Dunbar, R.A. Heintz, Prog. Inorg. Chem. 45 (1997) 283.
[2] G.N. Richardson, U. Brand, H. Vahrenkamp, Inorg. Chem. 38 (1999) 3070;
S.S. Kaye, J.R. Long, Catal. Today 120 (2007) 311.
[3] S. Ferlay, T. Mallah, R. Ouahes, P. Veillet, M. Verdaguer, Nature 378 (1995) 701.
[4] S.F.A. Kettle, G.L. Aschero, E. Diana, R. Rossetti, P.L. Stanghellini, Inorg. Chem.
45 (2006) 4928.
[5] S.F.A. Kettle, E. Diana, E. Boccaleri, P.L. Stanghellini, Inorg. Chem. 46 (2007)
2409.
[6] A. Volbeda, E. Garcia, C. Piras, A.L. Delacey, V.M. Fernandez, E.C. Hatchikian, M.
Frey, J.C. Fontecilla-Camps, J. Am. Chem. Soc. 118 (1996) 12989;
J.W. Peters, W.N. Lanzilotta, B.J. Lemon, L.C. Seefeldt, Science 282 (1998) 1853.
[7] C.E. Cooper, G.C. Brown, J. Bioenerg. Biomembr. 40 (2008) 533.
[8] S.W. Ragsdale, J. Inorg. Biochem. 101 (2007) 1657;
C.G. Riordan, J. Biol. Inorg. Chem. 9 (2004) 542.
[9] C.L. Drennan, J.Y. Heo, M.D. Sintchak, E. Schreiter, P.W. Ludden, Proc. Natl.
Acad. Sci. USA 98 (2001) 11973.
[10] H. Dobbek, V. Svetlitchnyi, L. Gremer, R. Huber, O. Meyer, Science 293 (2001)
1281.
[11] J.H. Joeung, H. Dobbek, Science 318 (2007) 1461.
[12] S.A. Ensign, M.R. Hyman, P.W. Ludden, Biochemistry 28 (1989) 4973;
M.E. Anderson, P.A. Lindahl, Biochemistry 33 (1994) 8702.
[13] J. Feng, P.A. Lindahl, J. Am. Chem. Soc. 126 (2004) 9094.
[14] V.J. DeRose, J. Telser, M.E. Anderson, P.A. Lindahl, B.M. Hoffman, J. Am. Chem.
Soc. 120 (1998) 8767.
[15] D. Li, R. Clérac, G. Wang, G.T. Yee, S.M. Holmes, Eur. J. Inorg. Chem. (2007)
1341;
J.Z. Gu, L. Jiang, M.Y. Tan, T.B. Lu, J. Mol. Struct. 890 (2008) 24.
[16] D.F. Shriver, M.A. Drezdon, The Manipulation of Air Sensitive Compounds,
second ed., Wiley, New York, 1986.
4. Summary
A binuclear Ni–CN–Fe complex, 2-Fe, designed to provide in-
sight into the fundamental characteristics of the cyanide-inhibited
form of the NiFeS enzyme COdH was spectroscopically and struc-
turally characterized. The linear Ni–CN–Fe bridge displays a mCN va-
lue ꢂ80 cmꢁ1 higher than that found in the inhibited form of COdH
supporting the hypothesis that the bridge in the latter is bent. The
scope of these studies was extended to include a series of isostruc-
tural cyanide-bridged binuclear complexes, 2-Co, 2-Ni and 2-Zn. A
similar coupling scheme was used to synthesize the copper(I) ana-
logue, 2-Cu, yielding the anionic complex. The copper analogue is
isostructural with the neutral complexes as elucidated by spectro-
scopic and X-ray diffraction analysis. Interestingly, 2-Cu did not

4562
M.T. Mock et al. / Inorganica Chimica Acta 362 (2009) 4553–4562
[17] A.B. Pangborn, M.A. Giardello, R.H. Grubbs, R.K. Rosen, F.J. Timmers,
Organometallics 15 (1996) 1518.
[32] R.Y. Han, A. Looney, K. McNeill, G. Parkin, A.L. Rheingold, B.S. Haggerty, J. Inorg.
Biochem. 49 (1993) 105.
[18] D. Sellmann, D. Häußinger, F.W. Heinemann, Eur. J. Inorg. Chem. (1999) 1715.
[19] S. Trofimenko, J.C. Calabrese, J.S. Thompson, Inorg. Chem. 26 (1987) 1507.
[20] S.M. Carrier, C.E. Ruggiero, W.B. Tolman, J. Am. Chem. Soc. 114 (1992) 4407.
[21] C. Ohrenberg, C.G. Riordan, L.M. Liable-Sands, A.L. Rheingold, Inorg. Chem. 40
(2001) 4276.
[33] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, fifth ed., John Wiley
& Sons Inc., New York, 1997. Vol. B:
Applications in Coordination, Organometallic and Bioinorganic Chemistry.
[34] M.J. Scott, R.H. Holm, J. Am. Chem. Soc. 116 (1994) 11357.
[35] D.A. Dows, A. Haim, W.K. Wilmarth, J. Inorg. Nucl. Chem. 21 (1961) 33.
[36] D. Qiu, M. Kumar, S.W. Ragsdale, T.G. Spiro, J. Am. Chem. Soc 118 (1996)
10429.
[37] S. Senda, Y. Ohki, T. Hirayama, D. Toda, J.-L. Chen, T. Matsumoto, H. Kawaguchi,
K. Tatsumi, Inorg. Chem. 45 (2006) 9914.
[38] For recent examples, see: N. Muresan, C.N. Lu, M. Ghosh, J.C. Peters, M. Abe,
L.M. Henling, T. Weyermoller, E. Bill, K. Wieghardt, Inorg. Chem. 47 (2008)
4579.
[39] Z. Hu, N.J. Spangler, M.E. Anderson, J. Xia, P.W. Ludden, P.A. Lindahl, E. Münck,
J. Am. Chem. Soc. 118 (1996) 830.
[40] S. Ciurli, M. Carrie, J.A. Weigel, M.J. Carney, T.D.P. Stack, G.C. Papaefthymiou,
R.H. Holm, J. Am. Chem. Soc. 112 (1990) 2654.
[41] T.A. Betley, J.C. Peters, Inorg. Chem. 42 (2003) 5074.
[42] K. Fukui, H. Ohya-Nishiguchi, N. Hirota, Bull. Chem. Soc. Jpn. 64 (1991) 1205.
[43] K. Fujisawa, T. Ono, Y. Ishikawa, N. Amir, Y. Miyashita, K. Okamoto, N. Lehnert,
Inorg. Chem. 45 (2006) 1698.
[44] D.J. Darensbourg, J.C. Yoder, M.W. Holtcamp, K.K. Klausmeyer, J.H. Reibenspies,
Inorg. Chem. 35 (1996) 4764.
[22] K. Yoon, G. Parkin, Inorg. Chem. 31 (1992) 1656.
[23] D. Sellmann, F. Geipel, F.W. Heinemann, Chem. Eur. J. 6 (2000) 4279.
[24] R.C. Gaussian 03, M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A.
Robb, J.R. Cheeseman, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant,
J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G.
Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M.
Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J.
Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J.
Dannenberg, V.G. Zakrzewski, S. Dapprich, A.D. Daniels, M.C. Strain, O. Farkas,
D.K. Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G.
Baboul, S. Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A.
Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong,
C. Gonzalez, J.A. Pople, Gaussian, Inc., Wallingford CT, 2004.
[25] G. Sheldrick, Bruker-AXS Inc., Madison, WI, 2001.
[26] A.L. Spek, J. Appl. Cryst. 36 (2003) 7.
[45] P.J. Schebler, C.G. Riordan, I. Guzei, A.L. Rheingold, Inorg. Chem. 37 (1998)
4754;
[27] J. Sun, C. Tessier, R.H. Holm, Inorg. Chem. 46 (2007) 2691.
[28] L.-J. Ming, Nuclear magnetic resonance of paramagnetic metal centers in
proteins, in: J. Larry Que (Ed.), Physical Methods in Bioinorganic Chemistry:
Spectroscopy and Magnetism, University Science Books, Sausalito, CA, 2000,
pp. 375–464.
S.-J. Chiou, J. Innocent, K.-C. Lam, C.G. Riordan, L. Liable-Sands, A.L. Rheingold,
Inorg. Chem. 39 (2000) 4347;
M.T. Mock, C.V. Popescu, G.P.A. Yap, W.G. Dougherty, C.G. Riordan, Inorg.
Chem. 47 (2008) 1889.
[29] P.J. Desrochers, J. Telser, S.A. Zvyagin, A. Ozarowski, J. Krzystek, D.A. Vicic,
Inorg. Chem. 45 (2006) 8930.
[30] E.G. Pavel, N. Kitajima, E.I. Solomon, J. Am. Chem. Soc. 120 (1998)
3949.
[31] F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochmann, Advanced Inorganic
Chemistry, sixth ed., John Wiley & Sons Inc., New York, 1999. p. 835.
[46] J.A. DuPont, M.B. Coxey, P.J. Schebler, C.D. Incarvito, W.G. Dougherty, G.P.A.
Yap, A.L. Rheingold, C.G. Riordan, Organometallics 26 (2007) 971.
[47] F.L. Hirshfeld, Acta Crystallogr. A 32 (1976) 239.
[48] N.Y. Zhu, H. Vahrenkamp, Angew. Chem., Int. Ed. Engl. 33 (1994) 2090.
[49] A. Looney, R. Han, I.B. Gorrell, M. Cornebise, K. Yoon, G. Parkin, A.L. Rheingold,
Organometallics 14 (1995) 274.