N-heterocyclic carbene ligands as mimics of imidazoles/histidine for the stabilization of Di-and trinitrosyl iron complexes

Article abstract of DOI:10.1021/ic201138f

N-heterocyclic carbenes (NHCs) are shown to be reasonable mimics of imidazole ligands in dinitrosyl iron complexes determined through the synthesis and characterization of a series of {Fe(NO)₂}¹⁰ and {Fe(NO)₂}⁹ (Enemark-Feltham notation) complexes. Monocarbene complexes (NHC-iPr)(CO)Fe(NO)₂ (1) and (NHC-Me)(CO)Fe(NO)₂ (2) (NHC-iPr = 1,3-diisopropylimidazol-2-ylidene and NHC-Me = 1,3-dimethylimidazol-2-ylidene) are formed from CO/L exchange with Fe(CO)₂(NO)₂. An additional equivalent of NHC results in the bis-carbene complexes (NHC-iPr)₂Fe(NO)₂ (3) and (NHC-Me)₂Fe(NO)₂ (4), which can be oxidized to form the {Fe(NO)₂}⁹ bis-carbene complexes 3⁺ and 4 ⁺. Treatment of complexes 1 and 2 with [NO]BF₄ results in the formation of uncommon trinitrosyl iron complexes, (NHC-iPr)Fe(NO) ₃⁺ (5⁺) and (NHC-Me)Fe(NO)₃ ⁺ (6⁺), respectively. Cleavage of the Roussins Red "ester" (μ-SPh)₂[Fe(NO)₂]₂ with either NHC or imidazole results in the formation of (NHC-iPr)(PhS)Fe(NO) ₂ (7) and (Imid-iPr)(PhS)Fe(NO)₂ (10) (Imid-iPr = 2-isopropylimidazole). The solid-state molecular structures of complexes 1, 2, 3, 4, 5⁺, and 7 show that they all have pseudotetrahedral geometry. Infrared spectroscopic data suggest that NHCs are slightly better electron donors than imidazoles; electrochemical data are also consistent with what is expected for typical donor/acceptor abilities of the spectator ligands bound to the Fe(NO)₂ unit. Although the monoimidazole complex (Imid-iPr)(CO)Fe(NO)₂ (8) was observed via IR spectroscopy, attempts to isolate this complex resulted in the formation of a tetrameric {Fe(NO) ₂}⁹ species, [(Imid-iPr)Fe(NO)₂]₄ (9), a molecular square analogous to the unsubstituted imidazole reported by Li and Wang et al. Preliminary NO-transfer studies demonstrate that the {Fe(NO)₂}⁹ bis-carbene complexes can serve as a source of NO to a target complex, whereas the {Fe(NO)₂}¹⁰ bis-carbenes are unreactive in the presence of a NO-trapping agent.

Full text of DOI:10.1021/ic201138f

ARTICLE
pubs.acs.org/IC
N-Heterocyclic Carbene Ligands as Mimics of Imidazoles/Histidine for
the Stabilization of Di- and Trinitrosyl Iron Complexes
Jennifer L. Hess, Chung-Hung Hsieh, Joseph H. Reibenspies, and Marcetta Y. Darensbourg*
Department of Chemistry, Texas A&M University, College Station, Texas 77843, United States
S Supporting Information
b
ABSTRACT: N-heterocyclic carbenes (NHCs) are shown to
be reasonable mimics of imidazole ligands in dinitrosyl iron
complexes determined through the synthesis and characteriza-
tion of a series of {Fe(NO)2}10 and {Fe(NO)2}9 (Enemarkꢀ
Feltham notation) complexes. Monocarbene complexes
(NHC-iPr)(CO)Fe(NO)2 (1) and (NHC-Me)(CO)Fe(NO)2
(2) (NHC-iPr = 1,3-diisopropylimidazol-2-ylidene and NHC-
Me = 1,3-dimethylimidazol-2-ylidene) are formed from CO/L
exchange with Fe(CO)2(NO)2. An additional equivalent of
NHC results in the bis-carbene complexes (NHC-iPr)2Fe(NO)2
(3) and (NHC-Me)2Fe(NO)2 (4), which can be oxidized to form
the {Fe(NO)2}9 bis-carbene complexes 3+ and 4+. Treatment of
complexes 1 and 2 with [NO]BF4 results in the formation of
+
+
uncommon trinitrosyl iron complexes, (NHC-iPr)Fe(NO)3 (5+) and (NHC-Me)Fe(NO)3 (6+), respectively. Cleavage of the
Roussin’sRed“ester”(μ-SPh)2[Fe(NO)2]2 with either NHC or imidazole results in the formation of (NHC-iPr)(PhS)Fe(NO)2 (7) and
(Imid-iPr)(PhS)Fe(NO)2 (10) (Imid-iPr = 2-isopropylimidazole). The solid-state molecular structures of complexes 1, 2, 3, 4, 5+, and 7
show that they all have pseudotetrahedral geometry. Infrared spectroscopic data suggest that NHCs are slightly better electron donors than
imidazoles; electrochemical data are also consistent with what is expected for typical donor/acceptor abilities of the spectator ligands
bound to the Fe(NO)2 unit. Although the monoimidazole complex (Imid-iPr)(CO)Fe(NO)2 (8) was observed via IR spectroscopy,
attempts to isolate this complex resulted in the formation of a tetrameric {Fe(NO)2}9 species, [(Imid-iPr)Fe(NO)2]4 (9), a molecular
square analogous to the unsubstituted imidazole reported by Li and Wang et al. Preliminary NO-transfer studies demonstrate that the
{Fe(NO)2}9 bis-carbene complexes can serve as a source of NO to a target complex, whereas the {Fe(NO)2}10 bis-carbenes are
unreactive in the presence of a NO-trapping agent.
’ INTRODUCTION
of the active site and the sulfur from glutathione that would
normally be expected to reside in the active site.7 While the
presence of the DNIC guest, carried in with the glutathione,
within this site correlates with diminished activity of GST, its
identification encourages conjectures regarding protein chaper-
ones or transport agents.
To date, most potential models of biological DNICs have
relied on thiolate ligands that give rise to monomeric
[(RS)2Fe(NO)2]ꢀ as well as dimeric, Roussin’s red-ester type
complex forms, that is, (μ-RS)2[Fe(NO)2]2.8ꢀ12 In both, the
Fe(NO)2 is in its oxidized form of {Fe(NO)2}9 electron config-
uration according to EnemarkꢀFeltham notation.13 Phosphine
DNICs are known (stabilizing the reduced {Fe(NO)2}10 redox
level) as well as DNICS containing nitrogen donors, particularly
those mimicking the coordination of the histidine amino acid
residue, frequently found as an adjunct to cysteine in
metallobiomolecules.14ꢀ16 A recent study by Tinberg and Lip-
pard et al. looked for evidence of both (Cys-S)2Fe(NO)2ꢀ and
Dinitrosyl iron complexes (DNICs) are of fundamental interest
to chemists for their unique and often ambiguous redox states that
have a sensitive response to spectator ligand fields. Following their
discovery during animal studies some 45 years ago, it has been
suggested that DNICs might function in vivo as nitric oxide
storage and transfer agents.1,2 In fact, the established degradation
of ironꢀsulfur clusters by excess NO gives cause to question
whether DNICs exist only as reporters of the deleterious effects of
NO overload.3 Regardless of their in vivo function, the readily
accessible chemistry of these pseudo-organometallic units has
inspired chemists to pursue the possibility that DNICs might be
developed as NO-delivery pharmaceuticals.4,5
Throughout the past two decades, studies of DNICs have sug-
gested that, although cysteine and glutathione are the major thiol
components of cellular DNICs, N- and O-donor ligands may also
trap the dinitrosyl iron moiety.6 Only recently has a protein-bound
DNIC been observed by X-ray crystallography; the Fe(NO)2 unit,
introduced exogenously to human glutathione transferase (GST P1-1)
as bis-glutathionyl dinitrosyl iron, is found to bind within the
active site of the enzyme through the phenolate oxygen of Tyr7
Received: May 27, 2011
Published: August 08, 2011
r
2011 American Chemical Society
8541
dx.doi.org/10.1021/ic201138f Inorg. Chem. 2011, 50, 8541–8552
|

Inorganic Chemistry
ARTICLE
(His-N)2Fe(NO)2 that might be derived from a Rieske-type
ferredoxin protein.17 Excess NO (or a NO-donor molecule) was
reacted with the toluene/o-xylene monooxygenase component C
(ToMOC) from Pseudomonas sp. OX1), containing a [2Fe-2S]
cluster with two histidines and two cysteine ligands on individual
irons. The observation of an EPR signal characteristic of a
{Fe(NO)2}9 unit supported the expectation of bis-cysteine DNIC
as seen with common 4Fe-4S clusters,18 or of one of the possible
forms of bis-histidine within an oxidized DNIC. As EPR spectros-
copy is the major tool for detecting DNICs, the authors stated that
the formation of the histidine-bound {Fe(NO)2}10 product was not
verified but cannot be ruled out.17
In contrast to cysteine, histidine binding to metals offers
additional complexities, including the basic nitrogen as a neutral
donor and, on deprotonation of the weakly acidic NꢀH, a
1ꢀ donor is obtained. Consequently, there are three possibil-
ities for (His)2Fe(NO)2 complexes: a neutral EPR inactive
{Fe(NO)2}10 complex, a neutral EPR active {Fe(NO)2}9 com-
plex, or an anionic EPR active {Fe(NO)2}9 complex. It is
expected that the bioactivity of histidine-containing DNICs
could critically depend on differences derived from these various
redox/protonation levels.
Li and co-workers have approached the difficult task of model-
ing histidine-containing DNICs and have ν(NO) infrared values
for several neutral (imidazole)2Fe(NO)2 complexes mostly char-
acterized in situ.19 In one case, a neutral (Imid-Me)2Fe(NO)2
(Imid-Me = 1-methylimidazole) complex of {Fe(NO)2}10 elec-
tron configuration was isolated and structurally characterized
(Figure 1a).14 The extreme air sensitivity of this complex and all
other members of the series led to detection of the oxidized
{Fe(NO)2}9 species by EPR spectroscopy, presumably arisinꢀg
from adventitious O2 as an oxidant.14,19 As in the (RS)2Fe(NO)2
complexes, the characteristic EPR signal is ∼2.03.8ꢀ12
Figure 1. Model complexes of histidine-containing {Fe(NO)2}10
DNIC: (a) (Imid-Me)2Fe(NO)2 (Imid-Me = 1-methylimidazole),14
(b) (NHC-Me)2Fe(NO)2 (NHC-Me = dimethyl N-heterocyclic carbene),
+
and (c) (NHC-Mes)Fe(NO)3 (NHC-Mes = 1,3-bis(2,4,6-trimethyl-
phenyl) N-heterocyclic carbene).20
N-bound to C-bound form of histidine in several metal deriva-
tives of imidazoles (eq 1). They concluded that N-binding is
more favorable for first-row elements, whereas C-binding is
preferred by second- and third-row elements. Additionally,
C- versus N-binding could be influenced by an imposed hydro-
gen-bonding network around the histidine.24 It should also be
noted that the general preparation of NHCs involves the facile
alkylation of imidazoles (eq 2).25 Furthermore, histidinium salts
have been known to react with AgI, PdII, or RhI sources to form
the respective transition-metal NHC complex.26 Such intercon-
versions between imidazole and NHC are intriguing possibilities
for biological systems, as yet undiscovered.
The N-heterocyclic carbene (NHC) ligands have gained
widespread use in organometallic chemistry for their strong
σ-donating characteristics and ability to bind to and stabilize a
range of transition-metal complexes.21,22 An example appropriate
to Fe-NO chemistry is the observation of a trinitrosyl iron complex
(TNIC) ligated by 1,3-bis(2,4,6-trimethylphenyl)imidazol-2-yli-
dene (NHC-Mes, Figure 1c) that can be readily isolated and
manipulated in solution under ambient conditions; the analogous
(R3P)Fe(NO)3+ (R = p-tolyl) immediately decomposes in solu-
tion at room temperature (22 ꢀC).20
To further explore the analogy described above, as well as to
develop the synthetic chemistry relating to Fe(NO)2 as NO releasing
agents, we present a series herein of DNICs containing NHCs and
imidazoles in both the reduced {Fe(NO)2}10 and the oxidized
{Fe(NO)2}9 forms as mimics of histidine-containing DNICs.
The planar NHC ligands mimic ligands, such as imidazoles
and pyridine; steric and electronic properties of the NHCs are
largely influenced by the substituents on the N atoms of the
NHC ring. A structural comparison of three imidazoles with
the dimethyl NHC complexed to square-planar Ni(N2S), Ni-
(mmp-dach) (mmp-dach = 1-(2-mercapto-2-methyl-propyl-1,
4-diazacycloheptane), found in all cases the plane of the ligands
to be perpendicular to the NiN2S plane with no significant differ-
ences in the NiꢀNImid and NiꢀCNHC distances (1.89ꢀ1.90 Å),
as well as the NiꢀN or NiꢀS spectator ligand distances
(Figure 2).23 The barrier to rotation about the NiꢀN bond of
the imidazoles was experimentally determined by VT 1H NMR
studies, finding that, for nonsterically hindered methylimidazole,
ΔG+ = 8.97 kcal/mol, a value that was corroborated by DFT
calculations. In contrast, the NHC derivative was sterically
prohibited from such intramolecular dynamics over the accessible
temperature range.23
’ EXPERIMENTAL SECTION
General Methods and Materials. All solvents were reagent
grade and were purified and degassed by a Bruker solvent purification
system and stored over molecular sieves. Reagents, including nitroso-
nium tetrafluoroborate, 1,3-diisopropylimidazolium tetrafluoroborate,
sodium tert-butoxide, sodium thiophenolate, 2-isopropylimidazole, and
5,10,15,20-tetraphenyl-21H,23H-porphine cobalt(II), Co(TPP), were
purchased from Sigma-Aldrich Chemical Co. and were used as received.
Standard Schlenk-line techniques (N2 atmosphere) and an Ar-filled
glovebox were used to maintain anaerobic conditions during prepara-
tion, isolation, and product storage. Fe(CO)2(NO)2,27 [Na-18-crown-
6-ether][Fe(CO)3(NO)],28 and 1,3-dimethylimidazolium iodide25 were
prepared according to published procedures.
Physical Measurements. Infrared spectra were recorded on a
Bruker Tensor 27 FTIR spectrometer in CaF2 solution cells with a path
length of 0.1 mm. Mass spectrometry (ESI-MS) was performed by the
Laboratory for Biological Mass Spectrometry at Texas A&M University.
Crabtree and Eisenstein have described DFT computational
results relating to the possibility of tautomerization from the
8542
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
Figure 2. Structures of [LNi(N2S)]+ complexes where the plane of L is perpendicular to the NiN2S plane. L = Imid (imidazole), Imid-Me
(1-methylimidazole), Imid-iPr (2-isopropylimidazole), and NHC-Me (dimethyl N-heterocyclic carbene).23
Nanoelectrospray ionization in positive mode was performed using an
Applied Biosystems QSTAR Pulsar (Concord, ON, Canada) equipped
with a nanoelectrospray ion source. Solution was flowed at 700 nL/min
through a 50 μm ID fused-silica capillary that was tapered at the tip.
Electrospray needle voltage was held at 1900 V.
m/z = 309 [M ꢀ CO + CH3CN]; 268 [M ꢀ CO]; 238 [M ꢀ CO ꢀ NO].
IR (THF): ν(CO) 1986 (m), ν(NO) 1738 (m), 1696(s) cmꢀ1. Anal.
Calcd (found): C, 38.2 (38.7); H, 5.74 (6.31); N, 17.8 (15.1).
(NHC-Me)(CO)Fe(NO)2, Complex 2. In a similar manner to that
described above, Fe(CO)2(NO)2 was freshly prepared and added to a
mixture of 0.49 g (2.19 mol) of 1,3-dimethylimidazolium iodide and 0.21
g (2.19 mmol) of NaOtBu in 5 mL of THF to ultimately produce 0.23 g
(44.3%) of a red-brown crystalline solid. X-ray quality crystals were
obtained by evaporation of pentane from the product mixture at 0 ꢀC.
Elemental analyses of crystalline samples were determined by Atlantic
Microlab, Inc., Norcross, GA. The poor match of found to calculated
values of several samples is likely due to their sensitivity and resulting
instability when exposed to O2. EPR spectra were typically recorded in
frozen THF using a Bruker ESP 300 equipped with an Oxford ER910
cryostat operating at 10 K. The WinEPR Simfonia program was used to
simulate spectral parameters.29 Cyclicvoltammogramswererecordedona
BAS-100A electrochemical analyzer. All experiments were performed at
room temperature under an Ar blanket in THF solution containing 0.1 M
[t-Bu4N][BF4] as the electrolyte, with a 3.0 mm glassy carbon working
electrode, a Ag/AgNO3 reference electrode, and a Pt coil counter
electrode. All values have been internally referenced to Cp2Fe/Cp2Fe+.
The mass magnetic susceptibility, χg, of a crystalline sample of complex 9
+
FeC6H8N4O3 (MW = 240 g/mol) ESI-MS: m/z = 287 [M ꢀ CO +
CH3CN] IR (THF): ν(CO) 1988 (m), ν(NO) 1740 (m), 1697 (s) cmꢀ1
.
Anal. Calcd (found): C, 30.0 (29.6); H, 3.34 (3.71); N, 23.4 (21.6).
(NHC-iPr)2Fe(NO)2, Complex 3. A pale yellow solution of NHC-iPr
(prepared from 0.19 g (0.81 mmol) of 1,3-diisopropylimidazolium
tetrafluoroborate and 0.078 g (0.81 mmol) in 5 mL of THF) was
transferred via a cannula to a flask charged with 0.24 g (0.80 mmol) of
complex 1 in 10 mL of THF. The reaction mixture was stirred overnight
and then dried in vacuo. A 40 mL portion of hexanes was added, the
mixture was stirred for 10 min, and the supernatant was separated from
the precipitate via a “football” cannula. The precipitate was dissolved in
THF and filtered through Celite. The solution was concentrated in
vacuo, transferred to several degassed test tubes, and layered with
hexanes to produce X-ray quality green-brown crystals at 0 ꢀC. Isolation
of the crystals afforded 0.13 g (38.0%) of analytically pure product.
FeC18H32N6O2 (MW = 419 g/mol) +ESI-MS: m/z = 420 [M + H]+. IR
(THF): ν(NO) 1664 (m), 1619 (s) cmꢀ1. Anal. Calcd (found): C, 49.3
(49.3); H, 7.77 (7.62); N, 19.2 (18.5).
(NHC-Me)2Fe(NO)2, Complex 4. In a similar manner to that described
above, a mixture of 0.14 g (0.60 mmol) of 1,3-dimethylimidazolium
iodide and 0.058 g (0.60 mmol) of NaOtBu dissolved in 5 mL of THF
was added to a flask charged with 0.14 g (0.60 mmol) of complex 2 in
10 mL of THF, ultimately producing 0.034 g (18.3%) of a brown
crystalline solid. A THF solution of 4 was layered with hexanes at 0 ꢀC to
obtainX-rayqualitycrystals.+ESI-MS of FeC10H16N6O2 (FW=307g/mol):
m/z = 308 [M + H]+, 278 [M ꢀ NO]. IR (THF): ν(NO) 1667 (m),
1624 (s) cmꢀ1. Anal. Calcd for FeC10H16N6O2 (found): C, 39.0 (37.6);
H, 5.23 (5.12); N, 27.3 (25.5).
was measured (Gouy balance at 299 K) as 0.057 ꢁ 10ꢀ4 erg Gꢀ2 cmꢀ3
.
3
X-ray Crystallography. Crystal samples were coated in mineral
oil, affixed to a Nylon loop, and placed under streaming N2 (110 K) in a
Bruker SMART 1000 CCD or single-crystal APEXii CCD diffract-
ometer. X-ray diffraction data were collected by covering a hemisphere
of space upon combination of three sets of exposures. The structures
were solved by direct methods. H atoms were placed at idealized
positions and refined with fixed isotropic displacement parameters,
and anisotropic displacement parameters were employed for all non-
hydrogen atoms. The following programs were used: for data collection
and cell refinement: SMART WNT/2000, version 5.63230 or APEX2;31
data reductions, SAINTPLUS, version 6.63;32 absorption correction,
SADABS;33 structure solutions, SHELXS-97;34 and structure refine-
ment, SHELXL-97.35 Structure plots were generated in Mercury,
version 2.3.36
Preparation of Compounds. (NHC-iPr)(CO)Fe(NO)2, Complex
1. In a 100 mL Schlenk flask, 1.00 g (2.20 mmol) of [Na-18-crown-
6-ether][Fe(CO)3(NO)] and 0.26 g (2.23 mmol) of [NO]BF4 were
dissolved in 15 mL of THF and stirred for at least 10 min to produce
Fe(CO)2(NO)2, which was vacuum-transferred to a flask immersed in
liquid N2. A separate flask loaded with 0.52 g (2.18 mmol) of 1,
3-diisopropylimidazolium tetrafluoroborate and 0.21 g (2.21 mmol) of
NaOtBu was dissolved in 5 mL of THF and stirred for at least 30 min.
This pale yellow solution was transferred via a cannula to the flask of
Fe(CO)2(NO)2 in THF, and the mixture was stirred for at least an hour,
resulting in a brown opaque mixture. The solvent was removed in vacuo,
and a minimum amount of THF was added (∼2 mL), followed by
30ꢀ40 mL of hexanes. This mixture was filtered through Celite, and the
filtrate was dried in vacuo and then dissolved in a minimum amount
of pentane. The pentane solution was transferred via a “football”
cannula (5.5 cm filter paper folded around and affixed to one end of
the cannula with Teflon tape) to several stoppered and degassed test
tubes, which were stored at 0 ꢀC. Within several days, X-ray quality red-
brown crystals were obtained. Isolation of the crystals afforded 0.24 g
(37% yield) of product. FeC10H16N4O3 (MW = 296 g/mol) +ESI-MS:
[(NHC-iPr)Fe(NO)3][BF4], Complex 5+. A 0.056 g (0.19 mmol)
sample of complex 1 was dissolved in 10 mL of THF and transferred
via a cannula to a Schlenk flask containing 0.024 g (0.21 mmol) of
[NO]BF4. The mixture was stirred for 1 h, resulting in a green
precipitate, which was filtered and washed with THF and ether to afford
0.026 g (35.7%) of a green powder. X-ray quality crystals were obtained
by layering a CH2Cl2 solution of the product with hexanes at 0 ꢀC.
IR (CH2Cl2): ν(NO) 1915 (m), 1826 (sh), 1810 (s) cmꢀ1. Anal. Calcd
for C9H17B1F4Fe1N5O3 (found): C, 28.1 (28.6); H, 4.19 (4.12); N,
18.2 (17.8).
[(NHC-Me)Fe(NO)3][BF4], Complex 6+. In a similar manner to that
described above, 0.097 g (0.34 mmol) of complex 2 was dissolved in
10 mL of THF and transferred via a cannula to a flask containing 0.037 g
(0.32 mmol) of [NO]BF4. Isolation of the precipitate afforded 0.050 g
(48.1%) of a green solid. IR (CH2Cl2): ν(NO) 1915 (m), 1826 (sh),
1814 (s) cmꢀ1. The green powder was not stable, even under an
8543
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
anaerobic environment, resulting in decomposition to an insoluble tan
solid and loss of ν(NO) IR bands.
8 were unsuccessful and led to the isolation of a tetramer. The optimized
procedure for complex 9 is as follows: Fresh Fe(CO)2(NO)2 in CH2Cl2
was prepared following the procedure above using 0.75 g (1.64 mmol) of
[Na-18-crown-6-ether][Fe(CO)3(NO)] and 0.20 g (1.71 mmol) of
[NO]BF4. A 0.36 g (3.3 mmol) portion of Imid-iPr in 15 mL of CH2Cl2
was added via a cannula, and the mixture was stirred overnight. [Note:
Shorter reaction times could be achieved on deliberate addition of
oxygen to the reaction vessel; however, excess oxygen degrades the
product. Hence, best results were obtained as described above with
adventitious oxygen presumably serving as an oxidant.] The solvent was
removed in vacuo from the green-brown solution, and 30ꢀ40 mL of
hexanes was added and filtered through Celite. The solvent was removed
in vacuo to afford 0.072 g (19.5%) of the red-brown solid product. X-ray
quality crystals were grown via evaporative removal of solvent from an
[(NHC-iPr)2Fe(NO)2][BF4], Complex 3+. A 0.024 g (0.10 mmol)
sample of 1,3-diisopropylimidazolium tetrafluoroborate and 0.010 g
(0.10 mmol) of NatOBu were dissolved in 10 mL of CH2Cl2 and stirred
for at least 30 min prior to transfer to a Schlenk flask containing 0.044 g
(0.11 mmol) of complex 5+ in 5 mL of CH2Cl2. The solution was stirred
for 30 min, and then the solvent was reduced in vacuo. About 20ꢀ30 mL
of hexanes was added to the mixture to precipitate out a brown powder
(0.015 g, 29.6%). +ESI-MS forFeC18H32N6O2(BF4) (MW = 420 g/mol):
m/z = 420 [M]+; 390 [M ꢀ NO]. IR (THF): ν(NO) 1789 (m), 1733
(s) cmꢀ1. The powder slowly degrades over the course of several days at
22ꢀ, even under an anaerobic environment, resulting in decomposition
to an insoluble orange-brown solid and loss of ν(NO) IR bands. In
solution under N2, complex 3+ as the BF4ꢀ (as well as the PF6ꢀ) salt is
largely stable over the course of 8 h; however, within 24 h, there is major
decomposition.
ether solution at 0 ꢀC. IR (THF): ν(NO) 1794 (m), 1726 (s) cmꢀ1
.
EPR: broad rhombic signal at 2.03. Mass spectrometry (Nano-(+)-ESI
MS): low intensity parent ion (Fe4N16O8C24H36) isotope bundle
centered at m/z = 900, base (100%) peak centered at 538
(Fe3N9O5C12H18, representing Fe3(NO)5(Imid-iPr)2), other intense
bundles centered at m/z 478.8 represent loss of 2NO from the base peak
and at 647.9 representing addition of one Imid-iPr (+110 mass units) to
the base peak. The full mass spectrum is in the Supporting Information.
Complex 9 reacted with [Na-18-crown-6-ether]+ imidazolate-iPrꢀ
(formed by deprotonation of Imid-iPr in THF solvent by Na0 in the
presence of 18-crown-6-ether) to yield a compound of IR and EPR
parameters (see Tables 3 and 5) similar to that of [Na-18-crown-6-
ether][(Imid)2Fe(NO)2].15
[(NHC-Me)2Fe(NO)2][BF4], Complex 4+. In a manner similar to that
above, 0.014 g (0.063 mmol) of 1,3-dimethylimidazolium iodide and
0.006 g (0.062 mmol) of NatOBu were dissolved in 5 mL of CH2Cl2 and
stirred for 30 min. This mixture was transferred to a Schlenk flask
containing 0.020 g (0.061 mmol) of complex 6+ in 5 mL of CH2Cl2 to
yield a yellow-brown solution. The unstable product was detected in
solution by IR (THF): ν(NO) 1791 (m), 1723 (s) cmꢀ1 and mass
spectrometry FeC10H16N6O2(BF4) (MW = 308 g/mol) +ESI-MS:
m/z = 308 [M]+. Over the course of an hour, the color of the solution
bleaches, ν(NO) IR bands disappear, and an insoluble orange-brown
precipitate forms.
NO-Trapping Experiments. Yellow-brown complex 3+ was
formed in situ according to the procedure described above and transferred
via a cannula to a Schlenk flask containing Co(TPP) (0.013 g, 0.019 mmol
in 10 mL of THF, Co(TPP) = 5,10,15,20-tetraphenyl-21H,23H-porphine
cobalt(II)). Within minutes, an absorption band at 1683 cmꢀ1 assignable
to(NO)Co(TPP) appearedand continued togrow inoverthe courseof6
h, along with a distinct color change of the solution from dark brown-red
to red. The reaction mixture was monitoredbyIR spectroscopy for 24 h. A
decrease of complex 3+ IR bands at 1789 and 1733 cmꢀ1 and an increase
of the band at 1683 cmꢀ1 were taken as an indication of NO transfer.37
The same procedure was performed for complex 3 (0.010 g, 0.024 mmol)
with Co(TPP) (0.016 g 0.024 mmol); however, the IR bands associated
with complex 3 did not decrease and the band at 1683 cmꢀ1 characteristic
of (NO)Co(TPP) was not observed.
(NHC-iPr)(PhS)Fe(NO)2, Complex 7. Method A. A 0.027 g (0.11 mmol)
sample of 1,3-diisopropylimidazolium tetrafluoroborate and 0.011
(0.11 mmol) g of NaOtBu were dissolved in 10 mL of THF and stirred
for 30 min prior to transfer to a Schlenk flask containing 0.025 g
38
(0.056 mmol) of the Roussin’s Red ester (μ-SPh)2[Fe(NO)2]2 in
5 mL of THF. The solution was stirred for 30 min, and then the solvent
was removed in vacuo. The resulting dark purple-red residue was
dissolved in a minimum amount of pentane (∼10 mL) and portioned
out to three degassed and stoppered test tubes. This process was
repeated several times until the entire product was dissolved in pentane
and transferred to test tubes. The test tubes were put in the freezer at
0 ꢀC overnight to afford 0.030 g (71.4%) of dark red crystalline material.
X-ray quality crystals were obtained via slow evaporation from ether
solution at 0 ꢀC. FeC15H21N4O2S (MW = 377 g/mol) +ESI-MS: m/z =
378 [M + H]+. IR (THF): ν(NO) 1757 (m), 1712 (s) cmꢀ1. Anal.
Calcd for FeC15H21N4O2S (found): C, 47.8 (47.9); H, 5.57 (5.68); N,
14.9 (14.7).
’ RESULTS AND DISCUSSION
Synthesis, Isolation, and Physical Properties. Synthetic
access to reduced {Fe(NO)2}10 DNICs relies on the freshly
prepared Fe(CO)2(NO)2 precursor, with CO/L exchange read-
ily occurring for both L = N-heterocyclic carbenes (NHC) and
imidazoles. As noted in Scheme 1, on addition of 1 equiv of
NHC, the mono-NHC complexes 1 and 2 were obtained, and
with 2 equiv of NHC, disubstituted complexes 3 and 4 were
formed. Complex 8 was observed with reaction of the Fe(CO)2-
(NO)2 precursor with 1 equiv of imidazole; however, excesses of
imidazole led to the formation of a reddish brown tetrameric
{Fe(NO)2}9 species, complex 9. This product, presumed to
require adventitious oxygen for oxidation, could be obtained in
greater yield in the presence of excess imidazole. Its structure is
described below.
Method B. A 0.036 g (0.094 mmol) sample of complex 5+ was
dissolved in 10 mL of CH2Cl2 and transferred to a Schlenk flask loaded
with excess NaSPh (0.022 g, 0.17 mmol). The mixture was stirred for 30
min, resulting in a red solution. The solvent was removed in vacuo, and
the residue was extracted with 20 mL of CH2Cl2 and filtered through
Celite. Removal of solvent resulted in 0.021 g (74.7%) of a dark red solid.
Product from this route and that of method A above had identical
properties. Method A is, however, preferred for simplicity.
(Imid-iPr)(CO)Fe(NO)2, Complex 8. Fresh Fe(CO)2(NO)2 in 10 mL
of THF was prepared following the procedure described above for the
preparation of complex 1 using 0.50 g (1.10 mmol) of [Na-18-crown-6-
ether][Fe(CO)3(NO)] and 0.13 g (1.11 mmol) of [NO]BF4. A 0.12 g
(1.1 mmol) sample of 2-isopropylimidazole (Imid-iPr) in 10 mL of THF
was added via a cannula and stirred for 1 h. The IR spectrum (ν(CO)
THF solution: 1992 (s); ν(NO) 1744 (m), 1698 (s) cmꢀ1) indicated
formation of expected product; however, conversion to a tetramer (see
the following for complex 9) occurred upon attempts to isolate product.
[(Imid-iPr)Fe(NO)2]4, Complex 9. Attempts to form and isolate
(Imid-iPr)2Fe(NO)2 with addition of 1 equiv of Imid-iPr to complex
The NHC products 1ꢀ4 were isolated as air sensitive,
thermally stable, crystalline solids and subjected to X-ray diffrac-
tion analysis, vide infra. Exposure of the {Fe(NO)2}10 NHC
complexes 1ꢀ4 to air either as solids or in solution led
to eventual degradation under ambient conditions. The air
8544
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
Scheme 1. Reactions of NHC and Imidazole with
Scheme 3. Comparison of NHC and Imidazole Reacting with
{Fe(NO)2}10 Precursor
{Fe(NO)2}9 Precursor
Scheme 2. Synthetic Routes to Oxidized bis-NHC DNICs
with O2 or NO+ as oxidant leads to a color change from brown to
yellow-brown, with ν(NO) band shifts to higher wavenumbers
consistent with the formation of the 3+ and 4+ {Fe(NO)2}9 prod-
ucts (Scheme 2). In contrast, addition of NO+ to complexes 1 and 2
resulted in CO/NO+ ligand exchange and formation of the green
trinitrosyl complexes 5+ and 6+, respectively. Treatment of TNICs
5+ and 6+ with the appropriate NHC, proceeding with release of
NO radicals (identified by capture of NO in a separate solution
containing a NO-trapping reagent, that is, a double tube arrange-
ment as described in ref 20), resulted in the formation of complexes
3+ and 4+, respectively, with the same color and ν(NO) bands as
observed with direct oxidation of complexes 3 and 4. This latter
approach to 5+ and 6+ to the DNICs 3+ and 4+ appears to give
cleaner compounds. Because of instability, as evidenced by the
bleaching of color and loss of ν(NO) IR bands, the cationic
complexes 3+ and 4+ could not be isolated as pure and crystalline
materials; however, mass spectral results are consistent with the
formulations given.
A common precursor to neutral {Fe(NO)2}9 DNICs is the
Roussin’s Red “ester”, RRE (μ-SPh)2[Fe(NO)2]2 (Scheme 3).38
Cleavage of this dimeric complex by NHCs or imidazoles leads to
analogous DNICs containing one NHC or imidazole and one SPhꢀ
(complexes 7 and 10). Complete conversion of the RRE dimer
required excess (>14 equiv) imidazole, whereas only 2 equiv of NHC
was needed for complete conversion of the dimer to (NHC-iPr)-
(PhS)Fe(NO)2 (7), which was isolated and structurally character-
ized by X-ray diffraction. Subsequent reaction of complex 7, with
additional NHC-iPr, resulted in bimolecular reductive elimination of
PhSSPh, with formation of the reduced {Fe(NO)2}10 complex 3.
The imidazole complex 10 is unreactive with excess imidazole.
sensitivity in solution was found to be dependent on the NHC
substituent, requiring ca. 24 h for degradation of the NHC-iPr
DNICs 1 and 3; within 1 h, the NHC-Me complexes 2 and 4 are
oxidized and degraded.
Nevertheless, intermediates in the oxidation process, especially of
complexes 3 and 4, can be observed. Reaction of complexes 3 and 4
8545
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
Figure 5. Solid-state molecular structure of (NHC-iPr)(PhS)Fe(NO)2,
complex 7, in ball-and-stick form. Thermal ellipsoid plots are given in the
Supporting Information.
inward toward each other, as was observed for the {Fe(NO)2}9
complex, (NHC-Mes)(SPh)Fe(NO)2.20 Allinall, theL2Fe(NO)2
complex structures are similar to each other and to structures of
LXFe(NO)2 complexes, even when the spectator ligands are within
bidentate frameworks.18,19
Figure 3. As derived from X-ray diffraction analysis, ball-and-stick
structures of the {Fe(NO)2}10 complexes: 1 = (NHC-iPr)(CO)Fe-
(NO)2, 2 = (NHC-Me)(CO)Fe(NO)2, 3 = (NHC-iPr)2Fe(NO)2, and
4 = (NHC-Me)2Fe(NO)2. Hydrogen atoms have been removed for
clarity. Thermal ellipsoid plots are given in the Supporting Information.
Selected metric data are in Table 1.
The {Fe(NO)3}10 complex 5+ is an uncommon, trinitrosylir-
on complex, TNIC, stabilized by the NHC-iPr (Figure 4a). The
(NHC-Mes)Fe(NO)3+ complex is a precise analogue of 5+.20 An
overlay of the molecular structures of TNIC 5+ and (NHC-Mes)-
Fe(NO)3+ is given in Figure 4b. The average —NNOꢀFeꢀNNO of
112.3ꢀ of the TNICs reflects a narrow range of angles; that is, the
bulk of the NHC does not influence one NO position over another.
The steric bulk of the pendant N-substituents on the NHC rings
undoubtedly contributes to the stability of these TNICs. In support
of this conclusion, the (NHC-Me)Fe(NO)3+ cation, complex 6+,
did not lend itself to crystallization because of instability in solution.
This is similar to other reported TNICs, such as those derived from
phosphines.20,39
The molecular structure of complex 9 was also determined by
X-ray crystallographic analysis and is shown in Figure 6. This
tetrameric structure is similar to the unsubstituted imidazole-
containing tetramer ([(Im-H)Fe(NO)2]4) reported by Wang
and Li et al.40 In both structures, the irons are the corners of a
square plane (average deviation from the best plane is 0.0175 Å for
complex 9 and 0.0094 Å for [(Im-H)Fe(NO)2]4) with deproto-
nated imidazole units as bridges along the edges and nitrosyl
ligands capping the corners. Select metric parameters of both
complexes are given in Table 2. A major difference in the two is the
orientation of the methenyl or ethenyl groups with respect to the
Fe4 plane. In complex 9, alternating up/down orientations mini-
mize the steric interaction of the i-Pr groups. In the simpler
unsubstituted imidazole, the methylene units are all positioned on
one side while the opposite ethenyl units produce a bowl-like
arrangement in which an acetone molecule resides as a solvent of
crystallization. The imidazole planes intersect the Fe4 plane at an
average 71ꢀ for complex 9 and 53ꢀ for [(Im-H)Fe(NO)2]4. The
average Fe-to-Fe distances are 6.24 Å in complex 9 (5.97 Å in the
unsubstituted imidazole analogue). The overall dimensions of the
cyclic tetramer of 8.76 Å ꢁ 8.90 Å, defined as the distances
between opposite irons in the metallocycle, (Fe1 Fe3 ꢁ Fe2
Figure 4. (a) Ball-and-stick structure of the trinitrosyl (NHC-iPr)
+
ꢀ
Fe(NO)3 BF4 salt, complex 5+. (b) Overlay of the (NHC-Mes)Fe-
(NO)3+20 with TNIC 5+.
Molecular Structures. The molecular structures of complexes
1, 2, 3, 4, 5+, and 7 were determined by X-ray diffraction analysis
and are represented in ball-and-stick form in Figures 3ꢀ5. Full
structure reports are available in the Supporting Information, as
are thermal ellipsoid plots. Selected metric data for these com-
plexes are presented in Table 1. The monomeric DNIC com-
plexes are pseudotetrahedral, with average CNHCꢀFeꢀNNO
bond angles in the range of 107ꢀ109ꢀ. The NNOꢀFeꢀNNO
angles in complexes 3 and 4 are 119ꢀ and 122ꢀ, and the
CNHCꢀFeꢀCNHC bond angles of complexes 3 and 4 are 90ꢀ
and 96ꢀ, respectively. This distortion from standard tetrahedral
angles is similar to that of other neutral L2Fe(NO)2 complexes of
{Fe(NO)2}10 composition, including Wang et al.’s (Imid-Me)2Fe-
(NO)2 complex, in which the —NimidꢀFeꢀNimid is 91.2ꢀ.14 The
FeꢀNꢀO bond angles are substantially linear for the {Fe(NO)2}10
complexes with an average of 174ꢀ. The bis-imidazole complex
(Imid-Me)2Fe(NO)2 has somewhat bent FeꢀNꢀO angles of 168ꢀ.
Likewise for complex 7, Figure 5, a {Fe(NO)2}9 complex, the
—FeꢀNꢀO averages to 167ꢀ, with the NꢀO ligands oriented
3 3 3
3
Fe4), are larger than those of [(Im-H)Fe(NO)2]4 (8.18 Å ꢁ
3 3
8.70 Å), likely to accommodate the additional steric bulk of the
isopropyl groups on the 2-position of the imidazole ring.
8546
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
Table 1. Selected Bond Distances (Å) and Angles (ꢀ) for Molecular Structures 1ꢀ5+ and 7
1
2
3
4
5+
7
bond distances (Å)
2.015(3)
FeꢀCNHC
FeꢀNO avg
FeꢀCO
2.005(2)
1.675(2)
1.784(2)
2.025(6)
1.727(6)
1.729(6)
1.973(6)
1.659(6)
2.004(5)
1.692(5)
2.062(2)
1.668(3)
1.642(3)
FeꢀS
2.256(11)
1.172(3)
NꢀO avg
1.181(2)
1.176(7)
1.204(3)
1.202(6)
1.151(6)
bond angles (ꢀ)
119.48(19)
173.8(2)
NNOꢀFeꢀNNO
FeꢀNꢀO avg
119.07(10)
175.8(2)
107.9(8)
111.3(2)
121.9(3)
112.3(3)a
172.8(4)
106.5(2)
113.80(13)
166.8(3)
173.3(5)
107.6(3)
174.0(5)
109.7(2)
96.4(2)
CNHCꢀFeꢀNNO avg
CNHCꢀFeꢀCNHC
CNHCꢀFeꢀS
109.39(11)
89.72(17)
108.93(13)
110.79(9)
CNHCꢀFeꢀCO
99.04(9)
103.6(2)
a Average of angles.
Table 2. Selected Bond Distances (Å) and Angles (ꢀ) in
Complex 9 and Imidazole Analogue40
9
[(Im-H)Fe(NO)2]4
avg
FeꢀNImid
NꢀO
bond distances (Å)
2.036(11)
1.194(13)
1.712(13)
6.241
2.005(5)
1.166(6)
1.694(5)
5.971
FeꢀNNO
Fe Fe (adj)
3 3 3
Fe Fe (cross)
8.827
8.440
3 3 3
bond angles (ꢀ)
110.72(5)
Figure 6. Left: from X-ray diffraction analysis, a ball-and-stick rendering
of the molecular structure of complex 9 (isopropyl groups have been
removed for clarity). Right: the analogous ChemDraw representation
of 9.
NImidꢀFeꢀNImid
NNOꢀFeꢀNNO
FeꢀNꢀO
109.24(17)
114.8(2)
166.8(6)
110.39(5)
164.02(11)
Infrared Spectral Data. Table 3 lists the diatomic ligand
stretching frequencies for the DNIC complexes of this study as
well as other complexes selected for appropriate comparisons.
Typically, DNICs of the {Fe(NO)2}10 configuration have ν(NO)
stretching frequencies in the 1620ꢀ1740 cmꢀ1 range, whereas
those of the oxidized {Fe(NO)2}9 form are shifted positively into
the 1700ꢀ1800 cmꢀ1 regime. Cationic {Fe(NO)2}9 derivatives
have higher ν(NO) than neutral analogues, demonstrated in this
study by the differences in values for complex 3+ and complex 7.
Notably, the difference between ν(NO) values of neutral complex
3, of the {Fe(NO)2}10 configuration, and cationic complex 3+, is
over 100 cmꢀ1, whereas the difference between the neutral
{Fe(NO)2}9 complex, (NHC-iPr)(PhS)Fe(NO)2, and the anio-
transition-metal complexes has been established. Thus, in the
case of the tetrahedral DNIC complexes, both steric properties
and electron-donor properties of the NHC ligands should make
them suitable mimics of imidazole ligands.
Table 4 lists ν(NO) infrared results for the cationic {Fe(NO)3}10
complexes with NHC and phosphine ligands. The higher stretching
frequencies of these complexes speak to the replacement of CO by
the isoelectronic NO+ ligand in, for example, complexes 1 and 2,
yielding a typical pattern of pseudo C3v symmetry.
Electron Paramagnetic Resonance Spectral Data and
Magnetic Susceptibility of Complex 9. The EPR spectra for
the following paramagnetic {Fe(NO)2}9 complexes prepared in
this study were recorded at 10 and 295 K in THF solution: 3+, 4+,
7, 9, and (Imid-iPrꢀ)2Fe(NO)2ꢀ (vide infra). At room tempera-
ture, complexes 3+, 4+, and 7 show isotropic signals at g = 2.028,
2.057, and 2.026, respectively. These are listed and compared
with analogous complexes in Table 5. Example spectra are
displayed in Figure 7. The EPR spectra for the anionic, bis-
nic {Fe(NO)2}9 complex, (PhS)2Fe(NO)2ꢀ, is only 20 cmꢀ1 10
.
This phenomenon is consistent with the differences of ν(NO)
stretchingꢀfrequencies between(sparteine)Fe(NO)20/+ and (RS)2-
Fe(NO)2 versus (Imid)(SR)Fe(NO)2 (R = Ph-o-NH-CO-
Ph).10,41 As the structural differences in FeꢀNꢀO distances
and angles is minimal, the source of these major discrepancies
awaits computational delineation.
imidazolate DNIC complexes show rhombicity with g values, for
example, of 2.038, 2.027, and 2.008 for (Imid-iPrꢀ)2Fe(NO)2
.
ꢀ
According to the report of Wang and Li et al., the EPR
spectrum of a frozen THF solution of tetrameric [(Im-H)Fe-
(NO)2]4 (based on the unsubstituted imidazolate) shows a nine-
line spectrum with N-14 hyperfine coupling superimposed on a
Where comparisons of NHC and imidazole complexes are
appropriate, both ν(NO) and ν(CO) values suggest that the
former is the (slightly) better donor. The better donating ability
of NHC ligands as reported by ν(CO) values in NHC/CO
8547
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
Table 3. ν(NO) Values for Selected DNICs Containing S, N, and C Donorse
LL0Fe(NO)2
ν(CO) cmꢀ1 (THF)
ν(NO) cmꢀ1 (THF)
complex
L
L0
sym, asym
ref
neutral {Fe(NO)2}10
CO
CO
CO
CO
CO
2089(m), 2038(s)
1988(m)
1807(m), 1762(s)
1738(m), 1696(s)
1740(m), 1697(s)
1744(m), 1698(s)
1664(m), 1619(s)
1667(m), 1624(s)
1673(m), 1616(s)
1679(m), 1622(s)
1627(m), 1567(s)d
1791(m), 1723(s)
1789(m), 1733(s)
1808(m), 1739(s)
1757(m), 1712(s)
1763(m), 1715(s)
1767(m), 1715(s)
1794(m), 1726(s)
1761(m), 1709(s)d
1765(m), 1699(s)
1774(m), 1712(s)
1737(m), 1693(s)
27
a
1
2
8
3
4
NHC-iPr
NHC-Me
Imid-iPr
a
a
a
a
1986(m)
1992(m)
NHC-iPr
NHC-Me
Imid-Me
sparteineb
Ar-nacnacc
NHC-iPr
NHC-Me
sparteine
NHC-iPr
NHC-Mes
Imid-iPr
Imid-iPrꢀ
Ar-nacnac
Imid-iPrꢀ
Imidꢀ
NHC-iPr
NHC-Me
Imid-Me
14
10
anionic {Fe(NO)2}10
cationic {Fe(NO)2}9
42
a
3+
4+
NHC-iPr
a
NHC-Me
10
a
neutral {Fe(NO)2}9
7
SPh
SPh
20
a
10
9
SPh
a
Imid-iPr
42
a
anionic {Fe(NO)2}9
Imid-iPrꢀ
Imidꢀ
SPh
15
10
SPh
a This work. b Sparteine = (6R,8S,10R,12S)-7,15-diazatetracyclo[7.7.1.02,7.010,15]heptadecane. c Ar-nacnac = [(2,6-diisopropylphenyl)NC(Me)]2CH.
d Benzene-d6 solution measurements. e THF solution measurements except where noted.
Table 4. ν(NO) Values for TNICs Containing P, N, and C
Table 5. EPR-Derived g Values of Selected DNIC in THF
Solution at Low Temperatures As Indicated
Donorse
L
ν(NO) cmꢀ1
ref
complex
g value
temp (K) ref
a
a
a
cationic {Fe(NO)3}10 5+ NHC-iPr
6+ NHC-Me
1915(m), 1826(sh), 1810(s)
1915(m), 1825(sh), 1814(s)
3+
4+
2.028
2.057
2.015
2.026
10
a
10
NHC-Mes 1932(s), 1831(s), 1804(vs)b 20
P(p-Tolyl)3 1917(s), 1838(vs), 1813(vs)c 20
(Imid-Me)2Fe(NO)2
7
240
10
14
a
+
P(CH2OH)3 1927(w), 1833(vs)d
39
(NHC-Mes)(SPh)Fe(NO)2
2.049, 2.029, 2.013
2.09, 2.06, 2.05
2.038, 2.027, 2.008
2.040, 2.022, 2.013
2.055, 2.029, 2.012
2.031
77
20
EtCN 1939(w), 1836(s)
43
(Ar-nacnac)Fe(NO)2
(Imid-iPrꢀ)2Fe(NO)2
77
42
a
a This work. b In THF. c Solid, ATR-FTIR. d In nitromethane. e CH2Cl2
solution measurements.
10
ꢀ
(Imidꢀ)2Fe(NO)2
77
15
a
ꢀ
9
10
rhombic signal.40 Preliminary EPR spectral data taken on a
solution of the analogous complex 9 show a broad rhombic
signal with g values of 2.055, 2.029, and 2.012 at 10 K that appears
to be an unresolved rhombic envelope of the 9-line spectrum
reported by Wang and Li et al.40 As of this report, further
resolution of the spectrum of 9 into hyperfine features has not
been achieved. At 295 K, the signal becomes isotropic with a g
value of 2.033. Interestingly, the imidazolateꢀ-cleaved tetramer of
Wang and Li et al., the (Imidꢀ)2Fe(NO)2 anion, also shows
hyperfine coupling.15 This feature is not observed for the bis-
Imid-iPr anion described above.
[(Im-H)Fe(NO)2]4
170
298
40
41
(Imid)(SPh-o-NHꢀCO-Ph)Fe(NO)2 2.031
a This work.
spinꢀspin coupling is not expected. This is confirmed for complex 9
by the room-temperature (Gouy balance) magnetic susceptibility
measurement yielding an effective magnetic moment per Fe atom,
μeff, of 1.75 μB (theoretical μSO value for one unpaired electron per
Fe is 1.73 μB).
Rather than attribute the hyperfine observed in the EPR
spectrum of [(Im-H)Fe(NO)2]4 to N-14 coupling within the
intact tetramer, Wang and Li et al. have suggested that the simpler
complex dissociates in THF solvent, producing an [(Imidꢀ)-
(THF)Fe(NO)2] monomeric DNIC with N-14 coupling from
nitrogens of two nitrosyls and one imidazolate ligand, accounting
for the 9-line spectrum.40 In support of this conclusion, Tsai and
It is well known that spinꢀspin coupling between {Fe(NO)2}9
units occurs in clusters, such as Roussin’s Red ester (FeꢀFe
separation = 2.5ꢀ2.6 Å), resulting in EPR silent complexes. In a
dimer designed to have an FeꢀFe separation of ca. 4 Å, an EPR
signal is observed.6 Hence, as the separation between {Fe(NO)2}9
units in the imidazolate tetramers under discussion here is ca. 6 Å,
8548
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
Figure 7. X-band EPR spectra taken at 10 K in a THF solution of (a) complex 3+ with a frequencyꢀat 9.468 GHz, (b) complex 7 with a frequency at 9.482
GHz, (c) complex 9 with a frequency at 9.468 GHz, and (d) complex (Imid-iPrꢀ)2Fe(NO)2 as a Na-18-crown-6-ether salt with a frequency at
9.473 GHz.
Table 6. Electrochemical Potentials of Selected DNICs and
TNICsa
measured relative to a Ag/AgNO3 electrode using a glassy carbon
working electrode and are referenced to Cp2Fe/Cp2Fe+. The redox
potentials for each complex are given in Table 6, and selected CVs
are shown in Figure 8. For complexes 1 and 2, an irreversible
oxidation event occurs at 0.14 and 0.12 V, respectively, whereas for
complexes 3 and 4, a reversible redox couple assigned to the
{Fe(NO)2}9/10 couple is centered at ꢀ0.76 and ꢀ0.77 V, respec-
tively (Figure 8b). Compared to the previously reported neutral-
{Fe(NO)2}10, (bipy)Fe(NO)2 (bipy = 2,20-bipyridine), which has
a reversible redox event at ꢀ0.48 V,44 the neutral, bis-NHC-
containing DNICs are oxidized at more negative values, ca. ꢀ0.76
V, which is consistent with the stronger electron-donating char-
acteristics of the NHC ligands. Despite the reversibility of the
cationic complex 5+ couple at 0.90 V (recorded in CH2Cl2 due to
insolubility in THF), the TNIC rapidly decomposed with repeated
scans. Note that the reversible reduction of the analogous NHC-
Mes TNIC is more negative by a volt (E1/2 = ꢀ0.29 V in THF and
ꢀ0.39 V in CH2Cl2) than that of 5+.20 This major discrepancy is not
understood.
Epc (V)
E1/2 (V)
ref
b
b
1
0.14
0.12
2
(CO)(NHC-Mes)Fe(NO)2
ꢀ0.34
20
b
3
ꢀ0.76
b
4
ꢀ0.77
(bipy)Fe(NO)2
ꢀ0.48
44
b
5+
0.90c
ꢀ0.29 (ꢀ0.395)c
20
b
+
(NHC-Mes)Fe(NO)3
7
ꢀ1.33
(NHC-Mes)(SPh)Fe(NO)2
ꢀ1.48
20
42
(Ar-nacnac)Fe(NO)2
ꢀ1.34
a In THF solution (0.1 M Bu4NBF4). All experiments were recorded
using a glassy carbon working electrode and Pt counter electrode and
referenced to Cp2Fe/Cp2Fe+ at a scan rate of 100 mV/s. b This work. c In
CH2Cl2 solution.
t
Complex 7 has a reversible redox event at ꢀ1.33 V (measured
in THF), and similar to the previously reported (NHC-Mes)-
(PhS)Fe(NO)2 complex, the reversible process is scan-rate-
Liaw et al. report a monomeric {Fe(NO)2}9 complex, (Imid)-
(SR)Fe(NO)2 (R = Ph-o-NH-CO-Ph), with two nitrosyls and one
imidazole, which also exhibits a well-resolved 9-line EPR signal
centered at g = 2.031 with hyperfine coupling constants of 2.4 and
4.1 G at 298 K;41 these are among the few observations of such
hyperfine coupling in paramagnetic DNICs. We note that the ν(NO)
IR spectra of complex 9 in THF solution and in the solid state are
identical. Hence, at this stage of investigation, there is no compelling
evidence for breakup of the tetramer in solution; however, the lack
of resolvable features in the EPR spectrum of complex 9 deserves
further inquiry.
dependent.20 It is notable that the (Ar-nacnac)Fe(NO)2
0/1ꢀ
complex couple is at ꢀ1.34 V in THF solution,42 closely
matching the reversible {Fe(NO)2}9/10 couple of complexes 7
and (NHC-Mes)(PhS)Fe(NO)2, but almost a volt more nega-
tive than the couple for (bipy)Fe(NO)2.
Overall, these data affirm that the redox properties of the
Fe(NO)2 unit respond to the ligand environment in a manner
reflecting the typical donor/acceptor abilities of the spectator
ligands.
Reactivity Studies. As noted above, imidazoles cleave the
RRE (μ-SPh)2[Fe(NO)2]2 to form the (Imid-R)(SPh)Fe(NO)2
complex. The imidazole ligand can, subsequently, be displaced by
addition of NHC-iPr to form complex 7; addition of imidazole, even
in large excesses, does not result in NHC displacement. Likewise,
in the {Fe(NO)2}10 cases, addition of imidazole to Fe(CO)2(NO)2
Cyclic Voltammetry. Cyclic voltammograms of complexes 1,
2, 3, 4, and 7 were recorded in 2 mM THF solution, and complex
5+ was recorded in 2 mM CH2Cl2 solution with 100 mM
[t-Bu4N][BF4] as the supporting electrolyte. All potentials were
8549
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
Figure 8. Cyclic voltammograms of complexes (a) 1, (b) 3, (c) 7 in 2 mM THF solution, and (d) 5+ in 2 mM CH2Cl2 solution. All are referenced to
Cp2Fe/Cp2Fe+.
Scheme 4. Reactivity of Complex 3+ with NaSPh in THF Solvent
results in the formation of (Imid-R)(CO)Fe(NO)2. Again, the
imidazole is readily (within minutes) displaced upon addition of
NHC to form complexes 1 and 2, in an irreversible reaction
(Scheme 1). Addition of excess NHC-iPr to complex 7 results in
the formation of complex 3 with bimolecular reductive elimination
of diphenyl disulfide; however, addition of excess imidazole to
(Imid)(SPh)Fe(NO)2 had no effect, as reported in Scheme 3. No
reaction occurred on addition of NaSPh to reduced complex 3;
however, 1 equiv of NaSPh displaced NHC-iPr from oxidized
oxidized {Fe(NO)2}9 unit, as demonstrated via IR and EPR
spectroscopies. Consistent with the results of Chen et al. working
with the unsubstituted [(Im-H)Fe(NO)2]4 cluster, complex 9
can also be cleaved by deprotonated phꢀenyl thiolate to form the
mononuclear, (Imidꢀ)(SPh)Fe(NO)2 DNIC.16 The forma-
tion of clusters is highly relevant to thiolate-containing DNICs.
Vanin et al. have reported that this type of reactivity, under
biological conditions using cysteine and glutathione, can be
regulated by the pH, where a decrease to pH = 5 results in
formation of dinuclear {Fe(NO)2}9 DNICs and an increase to a
pH of 9ꢀ10 results in the formation of mononuclear
{Fe(NO)2}9 DNICs.45 Additionally, it has been established that,
in biological systems in the absence of excess thiol, diamagnetic,
spin-coupled dinuclear DNICs are formed rather than the
paramagnetic mononuclear DNICs.46 Thus, from our studies
with imidazole-containing DNICs, it is reasonable that similar
reactivity modulated by pH and the presence or absence of
histidine/imidazole could control formation of polynuclear
versus mononuclear DNICs.
NO-Transfer Studies. To probe the ability of the NHC-
containing DNICs to release or transfer NO, the DNICs were
combined with the NO-trapping reagent Co(TPP) in THF
solution. Coupled with an almost immediate color change, an
IR band that grew in at 1683 cmꢀ1 concomitant with loss of
bands from the DNIC, indicated (NO)Co(TPP) formation.37
This reactivity of oxidized {Fe(NO)2}9 complexes 3+ and 4+ was
in great contrast to the reduced {Fe(NO)2}10 complexes 3 and 4,
complex 3+ to form complex 7 (Scheme 4). Further addition of an
ꢀ
excess of NaSPh to 3 results in the formation of (SPh)2Fe(NO)2
.
In the neutral {Fe(NO)2}9 complexes, the replacement of imidazole
by the stronger donating NHC likely results in a stabilized oxidized
iron unit. Nevertheless, anionic thiolates are even better stabilizers of
the oxidized {Fe(NO)2}9 unit, as exemplified by the instability of the
+
cationic (NHC)2Fe(NO)2 complexes. The neutral, paramagnetic
(NHC)(PhS)Fe(NO)2 are perhaps the most stable of the complexes
explored in this study, whereas the neutral, diamagnetic
(NHC)2Fe(NO)2 are the second most stable. Again, the remarkable
ability of the N-heterocyclic carbenes to stabilize two redox levels is
noted. Interestingly, the PhSꢀ anions did not appear to extract NO,
forming PhSNO, in a decomposition side reaction, in any of the
reactions explored.
Imidazole-Containing Analogues. The tetrameric complex
9, resulting from attempts to prepare monomeric complex 8
(Scheme 1), can be cleaved with deprotonated Imid-iPr to form
ꢀ
the anionic (Imid-iPrꢀ)2Fe(NO)2 complex, containing the
8550
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
which showed no change upon mixing with THF solutions of
Co(TPP). This preliminary study suggests that DNICs in the
reduced {Fe(NO)2}10 oxidation level are inert to NO loss/
transfer, whereas DNICs in the oxidized {Fe(NO)2}9 oxidation
state are sources of NO in the presence of a suitable NO-trapping
agent. This conclusion is supported by the work of Chiang et al.
and Tonzetich et al.37,42 Such assenting results encourage future
studies to develop a biologically compatible DNIC that can be
“turned on” by oxidation to deliver NO to a specific target.
Nevertheless, the mechanism of NO transfer is unknown, neither
is the fate of the DNIC following loss of NO. From “double-tube”
experiments, we have shown that the cationic NHCꢀTNIC
complex stabilized by the bulky NHC-Mes releases NO as free
NO radical/gas, which can diffuse into a solution containing a
NO-trapping reagent.20 Such a design to explore NO release
from 3+ or 4+ was not successful.
’ ACKNOWLEDGMENT
We appreciate financial support from the National Science
Foundation (CHE-0910679) and the R. A. Welch Foundation
(A-0924). We thank Dr. Timothy Hughbanks (TAMU) and
Dr. Andrey Prosvirin (TAMU) for their helpful discussions
regarding the EPR data. We also thank Jason Denny (TAMU) for
assistance with the X-ray studies, and Dr. Yohannes Rezenom for
the mass spectrometry results. Kelsey Shogren, an NSF-REU
student, the summer of 2010, contributed to the syntheses of
complexes 1ꢀ4.
’ REFERENCES
(1) (a) Vithayathil, A. J.; Ternberg, J. L.; Commoner, B. Nature
1965, 207, 1246–1249. (b) Woolum, J. C.; Commoner, B. Biochim.
Biophys. Acta 1970, 201, 131–140.
(2) Vanin, A. F. Nitric Oxide 2009, 21, 1–13.
(3) Foster, M. W.;Cowan, J. A. J. Am. Chem. Soc. 1999, 121, 4093–4100.
(4) Ostrowski, A. D.; Ford, P. C. Dalton Trans. 2009, 10660–10669.
(5) Chen, Y.-J.; Ku, W.-C.; Feng, L.-T.; Tsai, M.-L.; Hsieh, C.-H.;
Hsu, W.-H.; Liaw, W.-F.; Hung, C.-H.; Chen, Y.-J. J. Am. Chem. Soc.
2008, 130, 10929–10938.
’ SUMMARY AND COMMENTS
This work demonstrates that N-heterocyclic carbene ligands
(NHCs) are suitable as mimics of imidazole and histidine. From
the above demonstrated syntheses and reactivity, and chemical
precedent, it can be concluded that the {Fe(NO)2}10 unit
coexists with neutral ligands, such as CO, phosphines, NHCs,
and imidazoles. The {Fe(NO)2}9 unit is stable in the presence of
anionic ligands, such as SRꢀ, deprotonated imidazoles, or mixed
anionic/neutral ligand sets, neutral ligands being NHCs. Addi-
tionally, the precursors to {Fe(NO)2}10 and {Fe(NO)2}9, Fe-
(CO)2(NO)2, and (μ-RS)2[Fe(NO)2]2, respectively, are both
susceptible to reaction with imidazoles as well as NHCs. In the
latter case, trinitrosyl iron complexes may be derived. N-hetero-
cyclic carbene ligands also stabilize neutral {Fe(NO)2}10, neutral
{Fe(NO)2}9, and cationic {Fe(NO)2}9 DNICs. The NO-releas-
ing order in these derivatives is as follows:
(6) Tsai, M.-L.; Hsieh, C.-H.; Liaw, W.-F. Inorg. Chem. 2007, 46,
5110–5117 and references therein.
(7) Cesareo, E.; Parker, L. J.; Pedersen, J. Z.; Nuccetelli, M.; Mazzetti,
A. P.; Pastore, A.; Federici, G.; Caccuri, A. M.; Ricci, G.; Adams, J. J.; Parker,
M. W.; Lo Bello, M. J. Biol. Chem. 2005, 280, 42172–42180.
(8) Chiang, C.-Y.; Miller, M. L.; Reibenspies, J. H.; Darensbourg,
M. Y. J. Am. Chem. Soc. 2004, 126, 10867–10874.
(9) Tsai, F.-T.; Chiou, S.-J.; Tsai, M.-C.; Tsai, M.-L.; Huang, H.-W.;
Chiang, M.-H.; Liaw, W.-F. Inorg. Chem. 2005, 44, 5872–5881.
(10) Hung, M.-C.; Tsai, M.-C.; Lee, G.-H.; Liaw, W.-F. Inorg. Chem.
2006, 45, 6041–6047.
(11) Chiou, S.-J.; Wang, C.-C.; Chang, C.-M. J. Organomet. Chem. 2008,
693, 3582–3586.
(12) Lu, T.-T.; Tsou, C.-C.; Huang, H.-W.; Hsu, I.-J.; Chen, J.-M.;
Kuo, T.-S.; Wang, Y.; Liaw, W.-F. Inorg. Chem. 2008, 47, 6040–6050.
(13) Enemark, J. H.; Feltham, R. D. Coord. Chem. Rev. 1974, 13,
339–406.
10
9
fFeðNOÞ3g TNICs > fFeðNOÞ2g DNICs .
10
fFeðNOÞ2g DNICs
(14) Reginato, N.; McCrory, C. T. C.; Pervitsky, D.; Li, L. J. Am.
Chem. Soc. 1999, 121, 10217–10218.
When deprotonated, the distal nitrogen functionality in the
imidazole ligands of [(Imid)2Fe(NO)2]ꢀ imidazolates can lead
to aggregation through bridging, illustrated here in the new tetra-
meric complex 9. Such interesting molecular squares are composed
of {Fe(NO)2}9 units. Magnetic susceptibility data demonstrate
that each Fe within the tetramer can be considered as noncoupled,
S = 1/2, units at room temperature. Temperature-dependent
magnetic properties of 9 and analogues and the effect of sterically
encumbered imidazoles on structure and aggregation are currently
under investigation.
(15) Huang, H.-W.; Tsou, C.-C.; Kuo, T.-S.; Liaw, W.-F. Inorg.
Chem. 2008, 47, 2196–2204.
(16) Chen, C.-H.; Ho, Y.-C.; Lee, G.-H. J. Organomet. Chem. 2009,
694, 3395–3400.
(17) Tinberg, C. E.; Tonzetich, Z. J.; Wang, H.; Do, L. H.; Yoda, Y.;
Cramer, S. P.; Lippard, S. J. J. Am. Chem. Soc. 2010, 132, 18168–18176.
(18) Harrop, T. C.; Tonzetich, Z. J.; Reisner, E.; Lippard, S. J. J. Am.
Chem. Soc. 2008, 130, 15602–15610.
(19) Li, L. Comments Inorg. Chem. 2002, 23, 335–353.
(20) Hsieh, C.-H.; Darensbourg, M. Y. J. Am. Chem. Soc. 2010,
132, 14118–14125.
(21) Cornils, B.; Herrmann, W. A. Aqueous-Phase Organometallic
Catalysis, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2004.
(22) de Frꢀemont, P.; Marion, N.; Nolan, S. P. Coord. Chem. Rev.
2009, 253, 862–892.
(23) Jenkins, R. M.; Singleton, M. L.; Leamer, L. A.; Reibenspies,
J. H.; Darensbourg, M. Y. Inorg. Chem. 2010, 49, 5503–5514.
(24) Sini, G.; Eisenstein, O.; Crabtree, R. H. Inorg. Chem. 2002, 41,
602–604.
’ ASSOCIATED CONTENT
S
Supporting Information. X-ray crystallographic data (CIF)
b
from the structure determinations, TEPs, and full lists of metric
parameters for complexes 1ꢀ4, 5+, 7, and 9 and mass spectral scans
for complex 9. This material is available free of charge via the
Internet at http://pubs.acs.org.
(25) Schaub, T.; Backes, M.; Radius, U. Organometallics 2006, 25,
4196–4206.
(26) Hannig, F.; Kehr, G.; Fr€ohlich, R.; Erker, G. J. Organomet.
Chem. 2005, 690, 5959–5972.
(27) McBride, D. W.; Stafford, S. L.; Stone, F. G. A. Inorg. Chem.
1962, 1, 386–388.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: marcetta@chem.tamu.edu.
8551
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552

Inorganic Chemistry
ARTICLE
(28) Connelly, N. G.; Gardner, C. J. Chem. Soc., Dalton Trans.
1976, 1525–1527.
(29) WINEPR Simfonia, version 1.25; Bruker Analytische Messtechnik
GmbH: Berlin, 1996.
(30) SMART: Program for Data Collection on Area Detectors, version
5.632; Bruker AXS Inc.: Madison, WI, 2005.
(31) APEX2, version 2009.7ꢀ0; Bruker AXS Inc.: Madison, WI,
2007.
(32) SAINTPLUS: Program for Reduction of Area Detector Data,
version 6.63; Bruker AXS Inc.: Madison, WI, 2007.
(33) Sheldrick, G. M. SADABS: Program for Absorption Correction of
Area Detector Frames; Bruker AXS Inc.: Madison, WI, 2001.
(34) Sheldrick, G. M. SHELXS-97: Program for Crystal Structure
Solution; Universit€at G€ottingen: G€ottingen, Germany, 1997.
(35) Sheldrick, G. M. SHELXL-97: Program for Crystal Structure
Refinement; Universit€at G€ottingen: G€ottingen, Germany, 1997.
(36) Macrae, C. F.; Edgington, P. R.; McCabe, P.; Pidcock, E.;
Shields, G. P.; Taylor, R.; Towler, M.; van de Streek, J. J. Appl. Crytsallogr.
2006, 39, 453–457.
(37) Chiang, C.-Y.; Darensbourg, M. Y. J. Biol. Inorg. Chem. 2006,
11, 359–370.
(38) Rauchfuss, T. B.;Weatherill, T. D. Inorg. Chem. 1982, 21, 827–830.
(39) Dillinger, S. A. T.; Schmalle, H. W.; Fox, T.; Berke, H. Dalton
Trans. 2007, 3562–3571.
(40) Wang, X.; Sundberg, E. B.; Li, L.; Kantardjieff, K. A.; Herron,
S. R.; Lim, M.; Ford, P. C. Chem. Commun. 2005, 477–479.
(41) Tsai, M.-L.; Liaw, W.-F. Inorg. Chem. 2006, 45, 6583–6585.
(42) Tonzetich, Z. J.; Hꢀeroguel, F.; Do, L. H.; Lippard, S. J. Inorg.
Chem. 2011, 50, 1570–1579.
(43) Hayton, T. W.; McNeil, W. S.; Patrick, B. O.; Legzdins, P. J. Am.
Chem. Soc. 2003, 125, 12935–12944.
(44) Wang, R.; Wang, X.; Sundberg, E. B.; Nguyen, P.; Grant,
G. P. G.; Sheth, C.; Zhao, Q.; Herron, S.; Kantardjieff, K. A.; Li, L.
Inorg. Chem. 2009, 48, 9779–9785.
(45) Vanin, A. F.; Poltorakov, A. P.; Mikoyan, V. D.; Kubrina, L. N.;
Burbaev, D. S. Nitric Oxide 2010, 23, 136–149.
(46) McDonald, C. C.; Phillips, W. D.; Mower, H. F. J. Am. Chem.
Soc. 1965, 87, 3319–3326.
8552
dx.doi.org/10.1021/ic201138f |Inorg. Chem. 2011, 50, 8541–8552