Iron(III) bound by hydrosulfide anion ligands: NO-promoted stabilization of the [FeIII-SH] motif

Article abstract of DOI:10.1021/ja503683y

Spontaneous transformation of the thermally stable [HS]^--bound {Fe(NO)₂}⁹ dinitrosyl iron complex (DNIC) [(HS) ₂Fe(NO)₂]^- (1) into [(NO)₂Fe(μ-S) ]₂^2- (Rouss

Full text of DOI:10.1021/ja503683y

Article
pubs.acs.org/JACS
Iron(III) Bound by Hydrosulfide Anion Ligands: NO-Promoted
Stabilization of the [FeIII−SH] Motif
Chih-Chin Tsou,*,† Wei-Chun Chiu,† Chun-Hung Ke,† Jia-Chun Tsai,‡ Yun-Ming Wang,‡
Ming-Hsi Chiang,§ and Wen-Feng Liaw*,†
†Department of Chemistry and Frontier Research Center on Fundamental and Applied Sciences of Matters, National Tsing Hua
University, Hsinchu 30013, Taiwan
‡Department of Biological Science and Technology, National Chiao Tung University, Hsinchu 300, Taiwan
§Institute of Chemistry, Academia Sinica, Nankang, Taipei 115, Taiwan
S
* Supporting Information
ABSTRACT: Spontaneous transformation of the thermally stable
[HS]−-bound {Fe(NO)2}9 dinitrosyl iron complex (DNIC)
2−
[(HS)2Fe(NO)2]− (1) into [(NO)2Fe(μ-S)]2 (Roussin’s red
salt (RRS)) along with release of H2S, probed by NBD-SCN (NBD
= nitrobenzofurazan), was observed when DNIC 1 was dissolved in
water at ambient temperature. The reversible transformation of RRS
into DNIC 1 (RRS → DNIC 1) in the presence of H2S was
demonstrated. In contrast, the thermally unstable hydrosulfide-
containing mononitrosyl iron complex (MNIC) [(HS)3FeIII(NO)]−
(3) and [FeIII(SH)4]− (5) in THF/DMF spontaneously dimerized
into the first structurally characterized FeIII−hydrosulfide complexes
2−
[(NO)(SH)Fe(μ-S)]2 (4) with two {Fe(NO)}7 motifs anti-
2−
ferromagnetically coupled and [(SH)2Fe(μ-S)]2 (6) resulting
from two FeIII (S = 5/2) centers antiferromagnetically coupled to
yield an S = 0 ground state with thermal occupancy of higher spin states, respectively. That is, the greater the number of NO
ligands bound to [2Fe2S], the larger the antiferromagnetic coupling constant. On the basis of DFT computation and the
experimental (and calculated) reduction potential (E1/2) of complexes 1, 3, and 5, the NO-coordinate ligand(s) of complexes 1
and 3 serves as the stronger electron-donating ligand, compared to thiolate, to reduce the effective nuclear charge (Zeff) of the
iron center and prevent DNIC 1 from dimerization in an organic solvent (MeCN).
S-nitrosothiol derived from the hydrosulfide anion ([SH]−) and
INTRODUCTION
■
a NO+ agent (RSNO, NO2 ) or the hydrosulfide radical
−
Similarly to nitric oxide (NO) and carbon monoxide (CO),
([SH•]) and NO, is proposed as an effective molecule to
hydrogen sulfide (H2S) has recently drawn a lot of attention as
regulate these distinct functions.2,4 Iron metal may participate
an endogenously produced biological signaling molecule in the
cardiovascular and nervous systems and acts as a regulator
in the interaction of H2S and NO. In chemistry, N-bound [L−
FeII−HSNO] (L = porphyrin or (CN)5) adducts have recently
been proposed as intermediates in the reaction of {Fe(NO)}6
mononitrosyl iron complexes (MNICs) and [SH]−.5
against antibiotics in the defense systems of bacteria.1 Complex
biological effects of H2S (contraction vs relaxation of blood
vessels and neurodegenerative vs neuroprotective, apoptotic vs
antiapoptotic, pro-inflammatory vs anti-inflammatory, and
hyperalgesic vs analgesic activities) have been reported.1b
These diversities in physiological functions in health and
disease might be ascribed to different concentrations of H2S
and, presumably, interplay of H2S with NO (perhaps CO).1f,2
H2S−NO cross-talk is still unclear in chemical/biological
systems and warrants more work in this developing area. The
cooperative interaction of H2S and NO in the regulation of
angiogenesis and endothelium-dependent vasorelaxation has
recently been reported.3 The reaction products between H2S
and NO display distinct biological functions compared to those
regulated by H2S and NO individually. Although H2S cannot
react with NO directly, thionitrous acid (HSNO), the smallest
The development of a H2S-releasing agent, as a NO-releasing
agent, for physiological use is significant. To our best
knowledge, rare inorganic complexes acting as H2S donors
are known,6 in spite of the iron-based complex [LFeII−SH]+ (L
= (2-CH2Py)(Me)NCH2CH2N(2-CH2Py)(2-CH2Py-6-NC-
(tBu)O−); Py = pyridine) synthesized and identified to release
H2S.7 The synthesis of the metal−SH motif is challenging due
to its high propensity to produce multinuclear complexes, in
particular, for iron.8 It is noticed that all of the fully
characterized mononuclear iron−hydrosulfide complexes re-
Received: April 13, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
ported are in the iron(II) state because [SH]− has a strong
ability to reduce iron(III) to iron(II).7,8b,9 The complex
[FeIII(OEP)(SH)] (OEP = octaethylporphyrin) with a 1H
NMR chemical shift of −50.0 ppm (s br, SH), quickly reduced
to [FeII(OEP)], was observed.9f Although the isolation of
[FeIII(T-p-OMePP)(SH)] (T-p-OMePP = meso-tetrakis(p-
methoxyphenyl)porphyrin) was reported,10 the identification
and reproducibility of this [FeIII−SH] complex remain to be
elucidated.9b,11
temperature, leading to the formation of [(EtS)2Fe(NO)2]−,14c
the addition of 2 equiv of [SH]− to [(PhS)2Fe(NO)2]− in
THF, showing the formation of the known polynuclear
complexes [Fe4S3(NO)7]− (Roussin’s black salt (RBS) and
[(μ-S)Fe(NO)2]22− (Roussin’s red salt (RRS)),16 indicates that
analogous [SH]− displacement does not occur. [K-18-crown-6-
ether][(HS)2Fe(NO)2] (1) (or [PPN][(HS)2Fe(NO)2] (1−
PPN)) was synthesized from the reaction of [K-18-crown-6-
ether][(EtS)2Fe(NO)2] (or [PPN][(tBuS)2Fe(NO)2])14b,c,17
and H2S in THF at room temperature (Scheme 1a). Complexes
1 and 1−PPN are thermally stable in THF and the solid state
and are characterized by IR, UV−vis, EPR spectroscopy, X-ray
absorption spectroscopy (XAS), and single-crystal X-ray
diffraction (XRD) (Figures 1 and 2 and Supporting
The dinitrosyl iron complex (DNIC) is known to be a
possible form for storage and transport of NO in biological
systems.12 NO has been demonstrated to react with [Fe−S]
clusters of several proteins, including mitochondrial aconitase,
cytosolic aconitase, HiPIP, endonuclease III, mammalian
ferrochelatase, and SoxR, to form the EPR-active DNIC with
a characteristic signal at g ≈ 2.03.13 The reactivity between H2S
and DNIC [(RS)2Fe(NO)2]− (and its precursors MNIC
[(RS)3Fe(NO)]− and [Fe(SR)4]−)14 may merit further
exploration to build the chemistry of NO−H2S−Fe cross-talk.
In this study, the first iron(III)−hydrosulfide complexes
[(HS)2Fe(NO)2]− (1) and [(NO)(HS)Fe(μ-S)]2
(4)
2−
deriving from thermal conversion of [(HS)3Fe(NO)]− (3)
and [(HS)2Fe(μ-S)]2 (6) generated from spontaneous
2−
transformation of [Fe(SH)4]− (5) were isolated and charac-
terized. Of importance, the {Fe(NO)2}9 DNIC 1, stable as a
solid and in THF solution at room temperature, is
demonstrated to act as a H2S donor in aqueous solution,
characterized by the H2S probe NBD-SCN (NBD = nitro-
benzofurazan). DFT computation and experimental (calcu-
lated) reduction potentials (E1/2) of complexes 1−6 show that
triplet NO-coordinate ligand(s) reduce the effective nuclear
charge (Zeff) of iron to prevent [(HS)nFe(NO)m]− (n = 1, 2; m
= 3, 2) motifs from autoreduction and reduce the tendency of
self-rearrangement of [(HS)nFe(NO)m]− into bis(sulfide)-
Figure 1. ORTEP drawing and labeling scheme of [K-18-crown-6-
ether] [(HS)2Fe(NO)2] (1) with the thermal ellipsoid drawn at 50%
probability. Selected bond lengths (Å) and angles (deg): Fe(1)−N(1),
1.676(4); N(1)−O(1), 1.177(5); Fe(1)−N(2), 1.680(4); N(2)−
O(2), 1.184(5); Fe(1)−S(1), 2.2955(13); Fe(1)−S(2), 2.2987(13);
S(1)−K(1), 3.4935(15); S(2)−K(1), 3.4333(15); Fe(1)−N(1)−
O(1), 171.4(6); Fe(1)−N(2)−O(2), 170.5(4); N(1)−Fe(1)−N(2),
119.3(2); S(1)−Fe(1)−S(2), 109.11(5).
2−
bridged dimer [(L)2Fe(μ-S)]2 (L = SH, NO). In addition,
the larger the number of NO ligands bound to the [2Fe2S]
2−
core of complexes [(L)2Fe(μ-S)]2 (L = SH, NO), the larger
Information Figures S1−S4). The local geometry of complex
1 exhibits a distorted tetrahedron. The average Fe−N(O) and
N−O bond lengths of DNIC 1 (Fe−N(O) = 1.678(4) Å and
N−O = 1.181(5) Å) fall in the ranges of 1.661(4)−1.700(3) Å
(Fe−N(O)) and 1.160(6)−1.186(3) Å (N−O) observed in the
{Fe(NO)2}9 DNICs (Figure 1).18 The IR νNO stretching
frequencies 1735 and 1686 cm−1 (THF) of DNIC 1 lie
between νNO = 1715 and 1674 cm−1 for [(EtS)2Fe(NO)2]− and
νNO = 1737 and 1693 cm−1 for [(PhS)2Fe(NO)2]− (Figure
S2),14c,19 suggesting the electron-donating ability of [SEt]−,
[SH]−, and [SPh]− toward the [Fe(NO)2] motif follows the
series [SEt]− > [SH]− > [SPh]−. Consistent with the
characteristic g value of DNICs, complex 1 (or complex 1−
PPN) exhibits an isotropic EPR spectrum with a signal at gav =
2.028 (or gav = 2.029 and aN = 2.8 G for 1−PPN) at 298 K
(Figure 2), indicating an S = 1/2 ground state. In contrast to
the EPR spectrum of complex 1−PPN showing a well-resolved
five-line hyperfine splitting (Figure 2), the EPR spectrum of
complex 1 displaying an isotropic signal indicates the lesser spin
contribution of NO in the SOMO of complex 1. This may be
attributed to the interactions between SH/NO ligands and the
[18-crown-6-ether-K] cation, which decrease the spin con-
tribution of the NO motifs, as shown in the crystal structure.
The Fe K-edge pre-edge energy of XAS is 7113.5 eV for DNIC
1, within the range of 7113.4−7113.8 eV for {Fe(NO)2}9
DNICs (Supporting Information Figure S5) (vs 7113.1−7113.3
eV for {Fe(NO)2}10 DNICs).20 In combination with EPR,
the antiferromagnetic coupling constant between the two Fe
cores.
RESULTS AND DISCUSSION
■
In the previous study, the formation pathways of the
{Fe(NO)2}9 DNIC [(RS)2Fe(NO)2]− from nitrosylation of
the biomimetic oxidized/reduced-form rubredoxin [Fe-
(SR)4]1−/2− or the MNIC [(RS)3Fe(NO)]− (R = Ph, Et,
tBu), an intermediate in the conversion of [Fe(SR)4]1−/2− into
DNICs in the presence of NO(g), were elucidated (Scheme
1).14b,c,15 In contrast to treatment of the THF solution of
[(PhS)2Fe(NO)2]− with [SEt]− in a 1:2 molar ratio at ambient
Scheme 1
B
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 2. X-band EPR spectra of (a) DNIC 1 in THF at 298 K with gav = 2.028 and (b) DNIC 1−PPN in THF at 298 K with gav = 2.029 and aN =
2.8 G.
XAS, and XRD study, the dominate electronic structure of
complex 1 is consistent with a high-spin FeIII (SFe = 5/2)
antiferromagnetically coupled with two NO− anions (SNO = 1),
as observed in the previous study on DNICs.18a,21
Of importance, H2S release along with the formation of the
known RRS [Fe2S2(NO)4]2− (2)16b,22 was observed when
DNIC 1 was dissolved in a protic solvent (H2O or MeOH) at
room temperature (Scheme 1b). To quantitate the amount of
the released H2S derived from DNIC 1 dissolved in water,
NBD-SCN, analogous to the H2S quantification probe NBD-Cl
reported recently,23 was developed as a H2S probe. The
maximum percentage of H2S release generated from DNIC 1
dissolved in HEPES buffer solution (pH 7.4) was determined
to be 55% on the basis of a standard curve obtained from a
MeOH solution (0.2 mL) of NBD-SCN and [Na][SH] in 9.8
mL of HEPES buffer (40 mM, pH 7.4) (Supporting
Information Figure S6). Here, we propose that the reaction
of NBD-SCN and H2S displays the same reactivity, producing
NBD-SH (530 nm absorption band of NBD-SH), as the known
reaction of NBD-Cl and H2S, also yielding NBD-SH (Scheme
2).23 Reversibly, bridged sulfide cleavage of RRS triggered by
excess H2S occurs to convert RRS back into DNIC 1,
characterized by IR and UV−vis spectroscopies (Scheme 1c).
Figure 3. X-band EPR spectrum of complex 3 in THF at 4 K with g⊥ =
3.90 and g|| = 2.011. The asterisk indicates the signal (g = 2.025) of
[(HS)2Fe(NO)2]− (1), which derives from the reaction of a small
amount of [(EtS)2Fe(NO)2]− and H2S.
1731 cm−1 (THF); EPR g values at 3.76 and 2.012 (CH2Cl2/
toluene)).14b,c As learned from the EPR spectrum, the
dominate electronic structure of MNIC 3 (S = 3/2,
{Fe(NO)}7) is the result of strong antiferromagnetic coupling
between a high-spin FeIII (SFe = 5/2) and a NO-coordinate
ligand (NO−, SNO = 1).24 MNIC 3 in THF is thermally
unstable and spontaneously dimerizes into [(NO)(HS)Fe(μ-
S)]22− (4), characterized by IR (νNO = 1683 sh and 1668 cm−1
Scheme 2. Proposed Reaction Scheme for the Reaction of
NBD-SCN and H2S
1
(MeCN)), UV−vis, H NMR spectroscopy (δ 3.49 ppm (s br,
SH)), and XRD (Figure 4 and Supporting Information Figures
S7b−S10), along with release of H2S probed by NBD-SCN at
room temperature overnight (Scheme 1e). Complex 4 is the
first dinuclear MNIC with the local geometry of iron in a
distorted tetrahedron. The mean Fe−N (1.690 Å) and N−O
(1.165 Å) bond distances of complex 4 fall in the range of Fe−
N(O) (1.659(6)−1.779(9) Å) and N−O (1.002(9) to
1.193(4) Å) bond lengths observed in the {Fe(NO)}7
MNICs.25 The two IR νNO stretching frequencies (1683 sh
and 1668 s cm−1 (MeCN) and 1671 s and 1651 vs cm−1
(KBr)) of complex 4 may arise from trans and cis isomers, albeit
To explore the reactivity of DNIC’s precursors with H2S,
MNIC [(EtS)3Fe(NO)]− and [Fe(SEt)4]− were exposed to
excess H2S individually. As shown in Scheme 1d, treatment of
[(EtS)3Fe(NO)]− with excess H2S in THF led to the formation
of mononuclear MNIC [(HS)3Fe(NO)]− (3), characterized by
IR (νNO = 1728 cm−1 (THF)) and EPR (g⊥ = 3.90 and g|| =
2.011 (THF)) spectroscopies (Figure 3 and Supporting
Information Figure S7a), compared to those of [(tBuS)3Fe-
(NO)]− (IR νNO = 1697 cm−1 (THF); EPR g values at 3.84 and
1
2.01 (CH2Cl2/toluene)) and [(PhS)3Fe(NO)]− (IR νNO
=
only the trans isomer is observed in the X-ray structure. H
C
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
contrast to the thermal stability of [Fe(SEt)4]−,29 complex 5 is
thermally unstable in DMF even at −20 °C. Attempts to isolate
complex 5 were unsuccessful. Quantitative transformation of
2−
complex 5 into [(HS)2Fe(μ-S)]2 (6), characterized by UV−
1
vis, H NMR, superconducting quantum interference device
(SQUID) measurements, and XRD (Figure 6 and Supporting
Figure 4. ORTEP drawing and labeling scheme of [PPN]2[(μ-
S)Fe(SH)(NO)]2 (4) with the thermal ellipsoid drawn at 50%
probability. The unit cell contains two sets of disordered [(μ-
S)Fe(SH)(NO)]22− cores. Selected bond lengths (Å) and angles (deg)
of one (51.7% (a) and 48.3% (b)) of them: (a) Fe(1)−Fe(1)#1,
2.6833(6); Fe(1)−S(1), 2.2047(7); Fe(1)−S(1)#1, 2.2117(7);
Fe(1)−S(2), 2.317(2); Fe(1)−N(4), 1.685(6); N(4)−O(2),
1.169(10); Fe(1)−S(1)−Fe(1)#1, 74.83(2); S(1)−Fe(1)−S(1)#1,
105.17(2); S(2)−Fe(1)−N(4), 107.30(17); Fe(1)−N(4)−O(2),
169.0(6); (b) Fe(1)−S(3), 2.323(3); Fe(1)−N(3), 1.695(8); N(3)−
O(1), 1.160(11); S(3)−Fe(1)−N(3), 104.8(2); Fe(1)−N(3)−O(1),
166.4(8). The structure of the other set is depicted in Supporting
Information Figure S10.
Figure 6. ORTEP drawing and labeling scheme of [PPN]2[(μ-
S)Fe(SH)2]2 (6) with the thermal ellipsoid drawn at 50% probability.
Selected bond lengths (Å) and angles (deg): (a) Fe(1)−Fe(1)#,1
2.7266(8); Fe(1)−S(1), 2.3378(9); Fe(1)−S(2), 2.2046(8); Fe(1)−
S(2)#1, 2.2122(8); Fe(1)−S(3), 2.3072(9); Fe(1)−S(2)−Fe(1)#1,
76.24(3); S(2)−Fe(1)−S(2)#1, 103.76(3); S(3)−Fe(1)−S(1),
107.09(3); (b) Fe(2)−Fe(2)#2, 2.6950(7); Fe(2)−S(4), 2.3147(7);
Fe(2)−S(5), 2.3165(8); Fe(2)−S(6), 2.1959(8); Fe(2)−S(6)#2,
2.2031(8); Fe(1)−S(2)−Fe(1)#1, 76.24(3); S(6)−Fe(2)−S(6)#2,
104.44(3); S(4)−Fe(2)−S(5), 110.65(3).
NMR of complex 4 shows a broad peak of [SH] appearing at δ
3.49 ppm (SH, d7-DMF) at room temperature (Supporting
Information Figure S9). Compared to the known electronic
structure of RRS 2 (Stotal = 0) with two {Fe(NO)2}9 motifs
antiferromagnetically coupled,21b,26 complex 4 may be
described as two {Fe(NO)}7 motifs antiferromagnetically
coupled. The assigned {FeIII(NO−)}7 electronic structure of
complex 4 is further supported by the broken-symmetry DFT
calculation (see below). In particular, isolation of complex 4
may support the existence of the intermediate [(μ-SR)Fe-
Information Figures S12 and S13), was demonstrated in 10 min
at 0 °C (Scheme 1g). The corresponding 1H NMR spectrum of
the [SH]-coordinate ligands (δ 36.5 ppm (s br, SH)) displaying
an isotropic shift, consistent with that of the oxidized [2FeIII2S]
clusters,8a suggests that the electronic structure of the [Fe(μ-
S)2Fe] core of complex 6 is best described as two FeIII (S = 5/
2) centers antiferromagnetically coupled to yield an S = 0
ground state with thermal occupancy of higher spin states
(Figure S13a).30 Magnetic susceptibility data of a microcrystal-
line sample of complex 6 were collected in the temperature
range of 2−300 K (Figure 7). The molar magnetic
2−
(NO)(SR)]2 in nitrosylation of the [2Fe2S] cluster, yielding
2−
DNIC (that is, [2Fe2S] cluster → [(μ-SR)Fe(NO)(SR)]2
→
DNIC)27 and in the repair of DNIC back into the [2Fe2S]
2−
cluster (that is, DNIC → RRS → [(μ-S)Fe(NO)(SR)]2
→
[2Fe2S] cluster).28
As shown in Scheme 1f, upon bubbling excess H2S into a
DMF solution of [Fe(SEt)4]− at 25 °C, the EPR spectrum with
g values of 9.30 and 4.29 (DMF) at 4 K suggests the formation
of [Fe(SH)4]− (5) with a high-spin FeIII (S = 5/2) (Figure 5),
compared to that of [Fe(SEt)4]− (EPR g values of 7.20 and
4.25 (DMF) at 4 K (Supporting Information Figure S11)). In
Figure 7. χM (open circles) and χMT (open squares) versus
temperature for complex 6. The solid line is the best fit of the
experimental data to the theoretical model.
susceptibility increases from 1.253 × 10−3 cm3 mol−1 at 300
K to a value of 0.010 cm3 mol−1 at 2 K. The corresponding χMT
value decreases as a function of the temperature to 0.036 cm3 K
mol−1 at 95 K. The value remains practically constant until T is
below 5 K and reaches 0.02 cm3 K mol−1 at 2 K. The magnetic
behavior of complex 6 is consistent with the observation of the
oxidized [2FeIII2S] clusters in the literature.8a The experimental
Figure 5. X-band EPR spectrum of complex 5 (g = 9.30 and 4.29) in
DMF at 4 K.
D
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
data were fitted to an expression of the isotropic Hamiltonian,
H = JS1S2, where J is the magnetic exchange integral between
two FeIII (S = 5/2) units. The stability issue of complex 6 was
taken into consideration via inclusion of an FeIII impurity in the
theoretical expression. The least-squares fitting to the
theoretical model (χobsd = (1 − p)χFeFe + pχimp) gives g =
gain more information about the stability of [FeIII−SH]-bound
complexes 1−6, density functional theory (DFT) computation
with the BP86 functional33 and a mixed basis set of SDD ECP34
35
on Fe, BSXLANL on S, and 6-311++G(d,p)36 on all other
atoms was employed on complexes 1, 3, and 5 and their
reduced forms. This method has been determined by Hall and
co-workers to exhibit accurate results for optimized geometries
and NO vibrational frequencies in the [Fe−NO] system.37 For
singlet complexes 2, 4, and 6, the broken-symmetry (BS) DFT
method was employed. This method provides an effective
approach to estimate coupling constants in spin-coupled [Fe−
NO] systems and more appropriate Gibbs free energies
compared to the closed-shell method.26,38 The comparisons
of experimental parameters (complexes 1, 2, 4, and 6) and
computational parameters (conformations A, B, C, D, E, F, G,
H, and I, as shown in Chart 1) are summarized in Tables 1−3
2.019, J = 426 cm−1, p = 4.9 × 10−3, and R = 1.7 × 10−3. The
calcd
agreement factor (R) is defined as ∑(χM
− χMobsd)2/
∑(χMobsd)2. The results suggest that a singlet-spin ground state
is achieved for complex 6.
Transformation of complex 6 into the known
1
[Fe4S4(SH)4]2− was monitored by UV−vis and H NMR (δ
47.0 ppm (s br, SH) (d7-DMF)) spectroscopies when complex
6 was dissolved in DMF or MeCN at ambient temperature
overnight (Supporting Information Figure S13b). The shift of
309, 397, 433, 482 sh, and 585 sh nm into 368 nm confirmed
the formation of [Fe4S4(SH)4]2−.31 As shown in Table 1, the
Chart 1
Table 1. Selected Metric Data for [iPr4N]2[(NO)2Fe(μ-S)]2
(2),22 [PPN]2[(NO)(HS)Fe(μ-S)]2 (4), and
[PPN]2[(HS)2Fe(μ-S)]2 (6)
c
c
2
4
6
Fe···Fe (Å)
2.704(2)
2.225(2)
2.6833(6)
2.2082(7)
2.320(3)
1.690(8)
1.165(11)
3.508
2.7266(8)
2.2084(8)
2.3225(9)
a
Fe−(μ-S) (Å)
a
Fe−S (Å)
a
Fe−N (Å)
1.653(7)
1.178(8)
3.535
a
N−O (Å)
(μ-S)···(μ-S) (Å)
3.474
Fe−(μ-S)−Fe (deg)
74.82(7)
105.2(1)
111.5(3)
163.2(6)
74.83(2)
105.17(2)
106.1(2)
167.7(8)
180
76.24(3)
103.76(3)
107.09(3)
(μ-S)−Fe−(μ-S) (deg)
N(or S)−Fe−N(or S) (deg)
and Supporting Information Tables S1−S3. The geometric
parameters of conformations [(HS)2Fe(NO)2]− (A),
[(NO)2Fe(μ-S)]2]2− (G), [(NO)(HS)Fe(μ-S)]2]2− (H), and
[(HS)2Fe(μ-S)]2]2− (I) compare well to those of complexes 1,
2, 4, and 6, respectively, except a bent dihedral angle (159.3°)
in [Fe(μ-S)2Fe] of conformation G. The structures of
computationally derived conformations [(HS)2Fe(NO)2]2−
(B), [(HS)3Fe(NO)]− (C), [Fe(SH)4]− (E), and [Fe(SH)4]2−
(F) are also similar to those of [(EtS)2Fe(NO)2]2−, [(EtS)3Fe-
(NO)]−, [Fe(SEt)4]−, and [Fe(SEt)4]2−.14a,c,20b,39 As shown in
Tables 2, S1, and S2, the features of antiferromagnetic coupling
between Fe and NO in A, B, C, D, and H were observed. The
low spin densities of Fe and NO are ascribed to strong
antiferromagnetic coupling between Fe and NO within the
[Fe(NO)x] motif, which leads to partial cancellation of spin
a
Fe−N−O (deg)
b
dihedral angle (deg)
180
180
a
b
Average bond distance or angle. Defined by the intersection of the
c
two Fe2(μ-S) planes. Selected metric data for complexes 4 and 6 are
copied from Figures 4 and 6a.
Fe···Fe bond distances of 2.704(2), 2.6833(6), and 2.7266(8) Å
for complexes 2, 4, and 6, respectively, suggest a weak Fe···Fe
interaction. The stabilization of complexes 2 and 4 promoted
by NO-coordinate ligands, compared to the instability of
complex 6 slowly converting into the [(FeII)2(FeIII)2S4(SH)4]2−
cluster, may implicate that the [Fe(μ-S)2Fe] cores of complexes
2 and 4 are tailored to preserve the [FeIII(μ-S)2FeIII] oxidation
state, modulated by the NO-coordinate ligands. In brief, the
noninnocent NO-coordinate ligand(s) bound to the [(HS)-
FeIII(μ-S)2FeIII(SH)] motif serves as an effective regulator to
prevent the reductive elimination of the FeIII−SH bond of
complex 4. Measurements of the XAS S K-edge of complexes 2,
4, and 6 are done to rationalize the roles of NO and sulfide
ligands in stabilizing complexes 2 and 4.
It is noteworthy that the thermal stability of mononuclear
complexes 1, 3, and 5 containing the [FeIII−SH] motif is in the
order of 1 > 3 > 5. That is, the [FeIII−SH] motif coordinated
by more NO groups exhibits a reluctance to undergo self-
rearrangement into the bis(sulfide)-bridged dimer [(L)2Fe(μ-
S)2Fe(L)2]2− (L = NO, SH) along with release of H2S. The
recent report shows that bridged sulfide donates more electrons
to the iron center than the terminal thiolate ligand does,
according to S K-edge studies of iron−sulfur clusters.32 It
rationalizes that the mononuclear complexes 1, 3, and 5 tend to
dimerize into dinuclear complexes 2, 4, and 6, respectively. To
Table 2. Comparisons of Experimental and Computational
Reduction Potentials
complex pair in MeCN
E1/2 (V vs Fc+/Fc)
ref
[(EtS)2Fe(NO)2]1−/2−
−1.69
−1.50
−1.46
−1.55
−1.67
−1.62
−1.17
−0.41 to −0.56
19
a
[(EtS)3Fe(NO)]1−/2−
this work
29a
b
[Fe(SEt)4]1−/2−
[(HS)2Fe(NO)2]1−/2−
[(HS)2Fe(NO)2]1−/2− (A/B) (calcd)
[(HS)3Fe(NO)]1−/2− (C/D) (calcd)
[Fe(SH)4]1−/2− (E/F) (calcd)
[Por−Fe−SR]0/− (MeOH)
c
this work
this work
this work
this work
41
a
b
Quasi-reversible (Supporting Information Figure S14). Calculated
from [Fe(SEt)4]1−/2− with −1.08 V (vs SCE), and the Fc+/Fc couple
occurs at +0.38 V (vs SCE) in MeCN. See Supporting Information
c
Figure S15.
E
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
density.26,38b As shown in Table 2 and Figure 8, the spin
density of conformation H in the Ms = 0 broken-symmetry
Table 3. Selected Computational Geometry Parameters,
Natural Charge, Mulliken Spin Density, NO Vibrational
Frequencies, Heisenberg J Coupling Constant, and Reaction
Free Energy for [(NO)2Fe(μ-S)2Fe(NO)2]2− (G),
[(NO)(HS)Fe(μ-S)2Fe(SH)(NO)]2− (H), and [(HS)2Fe(μ-
S)2Fe(SH)2]2− (I)
G
H
I
Fe···Fe (Å)
2.68
2.724
2.238
2.347
1.685
1.194
3.553
75
2.721
2.231
2.349
a
Fe−(μ-S) (Å)
2.245
a
Fe−S (Å)
a
Fe−N (Å)
1.647
1.203
3.569
73.3
a
Figure 8. Broken-symmetry Ms = 0 spin density (isovalue 0.01) of
conformation H.
N−O (Å)
(μ-S)···(μ-S) (Å)
Fe−(μ-S)−Fe (deg)
(μ-S)−Fe−(μ-S) (deg)
3.535
75.2
a
a
105.3
112.4
163.7
159.3
105
104.8
105.3
state shows oppositely aligned spin populations on irons (2.56
for Fe1 and −2.56 for Fe2) and NO ligands (−0.51 for N1O1
and 0.51 for N2O2), and the plot of NO spin density displays a
cylindrical symmetry. On the basis of the previous assignment
of {FeIII(NO−)} determined from the cylindrical shape of NO
spin density and opposite spin densities for Fe (2.95−3.55) and
NO (−0.67 ∼ −0.91) in the {Fe(NO)}7 unit,24,26 the dominate
electronic structure of {Fe(NO)}7 in complex 4 may be best
described as a high-spin FeIII (SFe = 5/2) antiferromagnetically
coupled with a triplet NO-coordinate ligand (NO−) (SNO = 1).
The Heisenberg J coupling constants (H = JSASB) (4322
cm−1 for G, 813 cm−1 for H, and 415 cm−1 for I) were
calculated from the Yamaguchi equation [J = 2(EHS − EBS)/
(⟨S2⟩HS − ⟨S2⟩BS)].40 That is, a greater number of NO ligands
bound to the [2Fe2S] core leads to a higher antiferromagnetic
coupling constant. The larger J value is attributed to the smaller
a
N(or S)−Fe−N(or S) (deg)
103.9
169.9
180
a
Fe−N−O (deg)
b
dihedral angle (deg)
180
d
Mulliken spin density of Fe1
2.47
2.56
3.37
e
e
Mulliken spin density of N1−O1
−0.36, −0.32
−0.29,-
0.22
Mulliken spin density of N3−O3
Mulliken spin density of Fe2
Mulliken spin density of N2−O2
−0.36, −0.33
−2.47
0.36, 0.32
e
−2.56
0.29,
0.22
−3.37
e
e
Mulliken spin density of N4−O4
IR νNO (cm−1) (MeCN)
0.36, 0.33
1695, 1655, 1561, 1630
1542
e
⟨S2⟩HS
⟨S2⟩BS
5.02
2.87
12.42
2.84
813
30.02
4.23
415
e
c
e
Heisenberg J coupling constant
4322
(cm−1
)
denominator in the Yamaguchi equation (⟨S2⟩HS − ⟨S2⟩BS
=
d
reaction free energy ΔG
−0.6
−6.8
−15.1
2.15 for G, 9.58 for H, and 25.79 for I), whereas the energy
differences between high-spin and low spin-states (EHS − EBS)
are quite similar.26 This rationalizes why complex 2 exhibits
pure singlet ground states whereas complex 6 shows a singlet
ground state with thermal occupancy of higher spin states. As
shown in Table 3 and Supporting Information Figures S14 and
S15, the calculated reduction potentials (E1/2) of A/B (−1.67
V), C/D (−1.62 V), and E/F (−1.17 V) are consistent with the
trend of the experimental E1/2 of [(EtS)2Fe(NO)2]1−/2− (−1.69
V), [(HS)2Fe(NO)2]1−/2− (−1.55 V), [(EtS)3Fe(NO)]1−/2−
(−1.50 V), and [Fe(SEt)4]1−/2− (−1.46 V). This result
unambiguously shows that triplet NO− serves as a stronger
electron-donating ligand, compared to thiolate, in mononuclear
iron complexes. Compared to the FeIII/II E1/2 = −0.41 to −0.56
V (vs Fc+/Fc) observed in octahedral the [porphyrin−Fe−SR]
complex,41 the more negative E1/2 of tetrahedral [FeIII−SH]
complexes also explains the intrinsic difficulty of reduction of
FeIII in the tetrahedral system to prevent the [FeIII−SH]
complex from autoreduction. The trend of increasing natural
charge of iron from A (0.49) to C (0.72) to E (0.82) indicates
that the triplet NO-coordinate ligand somewhat reduces the Zeff
of iron (Supporting Information Tables S1−S3). This result is
supported by the previous S K-edge study on [(PhS)2Fe-
(NO)2]−, [(PhS)3Fe(NO)]−, and [Fe(SPh)4]− showing the
oxidation state of Fe, estimated from S K-edge data, is 2.75 for
[(PhS)3Fe(NO)]− and 2.59 for [(PhS)2Fe(NO)2]−.21a Pre-
sumably, the Zeff of iron in A, C, and E reflects the stability of
complexes 1, 3, and 5. Specifically, the higher Zeff of iron shows
a preference to rearrange into a dimer with strong electron-
donating bridged sulfide ligands. Moreover, the reaction free
energy values (−0.6 kcal/mol for A to G, −6.8 kcal/mol for C
(kcal/mol)
a
b
Average bond distance, angle, charge, or spin density. Defined by
c
the intersection of the two Fe2(μ-S) planes. J value (H = JSASB)
calculated from the Yamaguchi equation: J = 2(EHS − EBS)/(⟨S2⟩HS
−
⟨S2⟩BS).40 SA = SB = 1/2 for G. SA = SB = 3/2 for H. SA = SB = 5/2 for I.
d
Reaction free energy derived from the dimerization reaction
2[(L)2Fe(SH)2]− → 2H2S + [(L)2Fe(μ-S)2Fe(L)2]2− (L = NO,
SH). Due to the very strong antiferromagnetic coupling in
e
conformation G, the BP86 wave function (pure functional) collapses
to a spin-restricted solution, similarly to the OLYP wave function
(pure functional) in the previous study.26 The values of the spin
density and coupling constant are adopted from the broken-symmetry
calculation with the hybrid functional B3LYP/6-311G(d,p)/COSMO
electronic structure at the OLYP/TZP/COSMO geometry.26 Pure
functionals have been shown to result in a lesser degree of spin
polarization than hybrid functionals.26
to H, and −15.1 kcal/mol for E to I, respectively) calculated
from [2[(L)2Fe(SH)2]− (A/C/E) → 2H2S + [(L)2Fe(μ-
S)2Fe(L)2]2− (G/H/I) (L = NO, SH)] also reflect that
complex 5 prefers to dimerize into complex 6 via a large
exothermic pathway, compared to small energy differences
between MNIC 3 and dinuclear MNIC 4 (and between DNIC
1 and RRS 2). That is, the preference of thermodynamic
dimerization for complexes 1, 3, and 5 is in the order 5 > 3 > 1.
CONCLUSION AND COMMENTS
■
A series of FeIII−hydrosulfide DNIC, MNIC, and [2Fe2S]
clusters (that is, complexes 1, 3, 4, 5, and 6) were synthesized
1
and characterized by IR, UV−vis, H NMR, EPR, SQUID
measurements, XAS, and single-crystal X-ray diffraction.
F
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
and stored in dried, N2-filled flasks over 4 Å molecular sieves. Nitrogen
was purged through these solvents before use. Solvent was transferred
to the reaction vessel via a stainless cannula under a positive pressure
of N2. The reagents H2S(g) (Chien-Fa), 18-crown-6-ether (TCI),
bis(triphenylphosphoranylidene)ammonium chloride ([PPN][Cl]), 2-
[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES), D2O
(Sigma-Aldrich), and DMF (J. T. Baker) were used as received.
Complexes [PPN][(tBuS)2Fe(NO)2],17 [PPN][(EtS)3Fe(NO)],14c
[PPN][(tBuS)3Fe(NO)],14b [PPN][Fe(SEt)4],29b [K-18-crown-6-
ether][(EtS)2Fe(NO)2],14c and NBD-SCN43 were synthesized by
published procedures. Infrared spectra of the νNO stretching
frequencies were recorded on a PerkinElmer model Spectrum One
B spectrometer with sealed solution cells (0.1 mm, CaF2 windows) or
KBr. UV−vis spectra were recorded on a Jasco V-570 spectrometer.
1H NMR spectra were acquired on a Varian UNITY INOVA 500
Consistent with the previous S K-edge study on [(PhS)2Fe-
(NO)2]−, [(PhS)3Fe(NO)]−, and [Fe(SPh)4]− showing the
2.75 oxidation state of Fe for [(PhS)3Fe(NO)]− and 2.59 for
[(PhS)2Fe(NO)2]−,21a DFT computation and the experimental
(and calculated) reduction potential (E1/2) of DNIC 1, MNIC
3, and complex 5 unambiguously show that triplet NO− acts as
a stronger electron-donating ligand, compared to thiolate, to
reduce the Zeff of the iron center and stabilize DNIC 1 to
prevent dimerization as well as reduction of FeIII by [HS]−-
coordinate ligands. This result indicates that the thermal
stability of mononuclear complexes 1, 3, and 5 containing the
[FeIII−SH] motif is in the order 1 > 3 > 5. Moreover, the
reaction free energies obtained from DFT computation (−0.6
kcal/mol for A to G, −6.8 kcal/mol for C to H, and −15.1
kcal/mol for E to I) also reflect that the preference of
thermodynamic dimerization for complexes 1, 3, and 5 is in the
order 5 > 3 > 1.
The thermally unstable MNIC 3 and complex 5 in THF/
DMF spontaneously dimerize into FeIII−hydrosulfide complex
4 (1H NMR chemical shift δ 3.49 ppm (SH, d7-DMF)) along
with release of H2S and complex 6 with 1H NMR chemical shift
δ 36.5 ppm (SH, d7-DMF) displaying an isotropic shift,
respectively. That is, the electronic structure of complex 6 may
be described as two FeIII (S = 5/2) centers antiferromagneti-
cally coupled to yield an S = 0 ground state with thermal
occupancy of higher spin states (the corresponding χMT value
decreases as a function of the temperature to 0.036 cm3 K
mol−1 at 95 K),30b compared to that of complex 4 described as
two {Fe(NO)}7 motifs antiferromagnetically coupled. These
results and conclusion are supported by the Heisenberg J
coupling constants (H = JSASB) derived from the Yamaguchi
equation.40 That is, a greater number of NO ligands bound to
the [2Fe2S] core leads to a larger antiferromagnetic coupling
constant (4322 cm−1 for G, 813 cm−1 for H, and 415 cm−1 for
I). In contrast to complex 6 spontaneously converting into the
[Fe4S4(SH)4]2− cluster in DMF/MeCN at ambient temper-
ature, the stabilization of complex 4 may also implicate that the
noninnocent NO-coordinate ligand(s) bound to the [(HS)Fe-
(μ-S)2Fe(SH)] motif serves as an effective regulator to reduce
the Zeff of iron and to prevent the reductive elimination of the
FeIII−SH bond.
NMR spectrometer. Analyses of carbon, hydrogen, and nitrogen were
obtained with a CHN analyzer (Heraeus).
Preparation of [K-18-crown-6-ether][(HS)2Fe(NO)2] (1) and
[PPN][(HS)2Fe(NO)2] (1−PPN). The complex [K-18-crown-6-ether]-
[(EtS)2Fe(NO)2] (0.2166 g, 0.4 mmol) (or [PPN][(tBuS)2Fe(NO)2]
(0.3207 g, 0.4 mmol)) was dissolved in THF (10 mL), and dry H2S
(excess) was then injected into the THF solution of [(EtS)2Fe-
(NO)2]− with a gastight syringe. The reaction solution was stirred
under a N2 atmosphere at ambient temperature for 5 min. The color of
the reaction solution changed from brown to dark green. The IR νNO
stretching frequencies shifting from 1715 s and 1674 s to 1735 s and
1686 s cm−1 (THF) were assigned to the formation of 1 (or 1−PPN).
The dark green solution was filtered through Celite, and hexane was
then added to the filtrate to precipitate a dark green solid (yield 0.151
g (77%) for 1 and 0.235 g (85%) for 1-PPN). The cyclic
voltammogram of the complex 1−PPN in 20 mM MeCN with 0.1
M [n-Bu4N][PF6] as the supporting electrolyte at room temperature
and at a scan rate of 500 mV/s shows a quasi-reversible redox wave
with E1/2 = −1.55 V (vs Cp2Fe+/Cp2Fe) and ipa/ipc = 0.83.
Recrystallization from a THF solution of complex 1 layered with
diethyl ether−hexane at −20 °C for a week gives dark green crystals,
suitable for XRD analysis. Attempts to measure the IR νSH stretching
frequency in KBr were unsuccessful. IR νNO (cm−1): 1735 s, 1686 s
(THF); 1746 s, 1695 s (MeCN). Absorption spectrum (THF) [nm,
λmax (M−1 cm−1, ε)]: 333 (3000), 382 (2600), 732 (300). Anal. Calcd
for C36H32FeN3O2P2S2 (1−PPN) C, 60.01; H, 4.48; N, 5.83. Found:
C, 60.17; H, 4.66; N, 5.92.
Detection of the Released H2S from an Aqueous Solution of
1 by Using the H2S Probe NBD-SCN. To a flask of 1 (0.0485 g, 0.1
mmol) was added D2O (5 mL) to dissolve complex 1. The color of the
solution changed from green to dark red-brown. The IR stretching
frequencies νNO shifting to 1713 s br and 1689 s br cm−1 (D2O) were
consistent with the formation of the known [K-18-crown-6-
ether]2[(NO)2Fe(μ-S)]2 (2) (RRS).28 To prove H2S release derived
from the aqueous solution of complex 1, the H2S probe NBD-SCN
was used since NBD-SCN reacts with complex 1 to produce NBD-SH
in aprotic solvent THF at room temperature. Complex 1 (0.001
mmol) was prepared under a N2 atmosphere in a vial. The vial
containing complex 1 was then placed in a larger vial containing a
mixture of a MeOH (0.2 mL) solution of NBD-SCN (0.001 mmol)
and 9.8 mL of HEPES buffer (40 mM, pH 7.4) at room temperature.
The larger vial was then capped with a well-sealed septum. HEPES
buffer (2 mL, pH 7.4) was then added to the vial containing complex 1
with a syringe. After the solution was stirred for 4 h, NBD-SH was
produced and detected by UV−vis spectroscopy. On the basis of the
absorption intensity of NBD-SH at 530 nm obtained from NBD-SCN
(100 μM) and [Na][SH] (10, 20, 30, ..., 100 μM) (a standard curve),
the amount of released H2S derived from the aqueous solution of
complex 1 at ambient temperature was quantified as 55% (Supporting
Information Figure S6).
Upon DNIC 1 being dissolved in water at ambient
temperature, the rapid formation of RRS along with release
of H2S probed by NBD-SCN may signify DNIC 1 serves a role
in the storage and transport of H2S in the hydrophobic domain
and may release H2S in hydrophilic conditions in biological
systems. Presumably, [HS]− in biology may most frequently
bind to iron metal according to the preference dictated by the
noninnocent characteristic of the coordinated NO ligands. The
hydrosulfide-coordinated DNIC and MNIC may serve not only
a role in H2S storage/transport but also as the sulfide carrier/
source in the synthesis of the [Fe−S] cluster, although cysteine
persulfide (Cys-SSH, a source of H2S in the presence of a
reducing agent) on cysteine desulfurase (IscS) is known to
deliver its sulfane sulfur for the assembly of [Fe−S] clusters on
the scaffold protein (IscU).42
EXPERIMENTAL SECTION
■
Manipulations, reactions, and transfers were conducted under nitrogen
according to the Schlenk technique or in a glovebox (nitrogen gas).
Solvents were distilled under nitrogen from appropriate drying agents
(MeOH from I2 and Mg, MeCN from CaH2, and hexane, diethyl
ether, and tetrahydrofuran (THF) from sodium and benzophenone)
Preparation of [PPN][(HS)3Fe(NO)] (3) and Its Transforma-
tion into [PPN]2[(NO)(SH)Fe(μ-S)]2 (4). The complex [PPN]-
[(EtS)3Fe(NO)] (0.3107 g, 0.4 mmol) (or [PPN][(tBuS)3Fe(NO)]
(0.3444 g, 0.4 mmol)) was dissolved in THF (10 mL). Dry H2S
(excess) was then injected into the THF solution of the complex
G
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
[PPN][(EtS)3Fe(NO)] with a gastight syringe at ambient temper-
ature. The color of the reaction solution changed from dark red to
dark yellow. The appearance of IR νNO = 1728 s cm−1 (THF)
indicated the formation of the thermally unstable 3. After the dark
yellow solution was stirred at ambient temperature overnight, the
green solid 4 precipitated from the THF solution accompanied by
release of H2S probed by NBD-SCN. The precipitate was collected
and redissolved in MeCN. The IR νNO stretching frequencies 1683 sh
and 1668 s cm−1 (MeCN) suggest the formation of complex 4.
Recrystallization from the MeCN solution of complex 4 layered with
diethyl ether at −20 °C for a week yields dark green crystals (yield
0.15 g, 54%), suitable for XRD analysis. Attempts to measure the IR
νSH stretching frequency of complex 4 in KBr were unsuccessful. IR
νNO (cm−1): 1683 sh, 1668 s (MeCN); 1671 s, 1651 vs (KBr).
Absorption spectrum (MeCN) [nm, λmax (M−1 cm−1, ε)]: 329 (8900),
crystal Si(111) monochromator. The energy resolution ΔE/E was
estimated to be about 2 × 10−4. High harmonics were rejected by Rh-
coated mirrors. The spectra were scanned from 6.912 to 7.972 keV. A
reference Fe foil is always measured simultaneously, in which the first
inflection point at 7112.0 eV of the Fe foil spectrum is used for energy
calibration. Ion chambers used to measure the incident (I0) and
transmitted (I) beam intensities were filled with a mixture of N2 and
He gases and a mixture of N2 and Ar gases, respectively.
Crystallography. The crystals of complexes 1, 1−PPN, 4, and 6
chosen for X-ray diffraction study were measured in sizes of 0.15 ×
0.13 × 0.07, 0.36 × 0.32 × 0.21, 0.30 × 0.30 × 0.25, and 0.12 × 0.10 ×
0.07 mm, respectively. The crystals were mounted on a glass fiber.
Unit-cell parameters were obtained by least-squares refinement.
Diffraction measurements for complexes 1, 1−PPN, 4, and 6 were
carried out on a Bruker Kappa Apex II diffractometer with graphite-
monochromated Mo Kα radiation (λ = 0.7107 Å) and between 1.98°
and 25.05° for 1 (between 2.37° and 25.03° for 1−PPN, between
1.148° and 26.433° for 4, and between 1.149° and 26.537° for 6).
Least-squares refinement of the positional and anisotropic thermal
parameters of all non-hydrogen atoms and fixed hydrogen atoms was
based on F2. The SHELXTL structure refinement program was
employed.44 A summary of the crystallographic data, intensity
collection, and structure refinement parameters for complexes 1, 1−
PPN, 4, and 6 are collected in Supporting Information Table S4.
Magnetic Measurements. The magnetic data of the powdered
samples of complex 6 were recorded on a SQUID magnetometer
(SQUID-VSM, Quantum Design) under an external magnetic field
(0.1 T) in the temperature range of 2−300 K. The magnetic
susceptibility data were corrected with the ligands’ diamagnetism by
the tabulated Pascal constants.
Computational Details. All DFT calculations were performed
with the Gaussian 09 program.45 All geometry optimizations of
conformations A−F were conducted with DFT using the BP86
functional33 and a mixed basis set of SDD ECP34 on Fe, BSXLANL35 on
S, and 6-311++G(d,p)36 on all other atoms. The methodology was
chosen because it has been shown to have good accuracy for the
DNIC system.37 For solvent correction a PCM model with acetonitrile
(ε = 35.688) was applied.46 According to the EPR data of complexes 1,
3, and 5, the spin states of conformations A, C, and E were assigned as
S = 1/2, 3/2, and 5/2, respectively. Since conformations B and E are
similar to [(EtS)2Fe(NO)2]2− and [Fe(SEt)4]2−,20b,39 the spin states of
their reduced forms were assigned as S = 0 and 2, respectively. Despite
the lack of a structure for [(EtS)3Fe(NO)]2−, the spin state of
conformation D was expected to be S = 1 derived from a high-spin FeII
(SFe = 2) antiferromagnetically coupled with a nitroxyl ligand (NO−)
(SNO = 1). For singlet conformations G−I, the BS DFT method was
employed.26,38a First, high-spin (HS) states (S = 1 for G, S = 3 for H,
and S = 5 for I) were optimized. Subsequently, a flip of the spin on
selected atoms for an initial guess of the antiferromagnetically coupled
state was conducted, followed by geometry optimization to generate
the broken-symmetry Ms = 0 state.26 The “stable” keyword was used to
test the stability of the wave function. Frequency calculations were
carried out for detected stationary points to ensure that they
correspond to the true minima on the potential energy hypersurface.
The thermal correction to the Gibbs free energy was made under the
conditions of 298.15 K and 1 atm. The Heisenberg J coupling constant
(H = JSASB) was estimated by using the Yamaguchi approach: J =
2(EHS − EBS)/(⟨S2⟩HS − ⟨S2⟩BS),40 in which EHS − EBS is energy
difference between the high-spin and low-spin states, and ⟨S2⟩HS and
⟨S2⟩BS are the expectation values of the HS and BS S2 operators. Due
to the very strong antiferromagnetic coupling in conformation G, the
BP86 wave function collapses into a spin-restricted solution, similarly
to the OLYP wave function in the previous study.26 The spin density
map was plotted with the Chemissian 3.3 program.47 The natural
charge analysis was obtained from the NBO 5.9 program in the
Gaussian 09 package.48
1
407 (4500), 585 sh (1700). H NMR (500 MHz, d7-DMF, 25 °C,
TMS): δ 3.49 ppm (s br, SH). Anal. Calcd for C72H62Fe2N4O2P4S4: C,
62.70; H, 4.53; N, 4.06. Found: C, 63.02; H, 4.69; N, 3.48.
Preparation of [PPN]2[(SH)2Fe(μ-S)]2 (6). The complex [PPN]-
[Fe(SEt)4] (0.3232 g, 0.4 mmol) was loaded into a Schlenk flask (20
mL) and dissolved in DMF (10 mL) at room temperature. Dry H2S
(excess) was injected into the DMF solution of the complex
[PPN][Fe(SEt)4] to yield a dark yellow-brown solution at 25 °C.
The EPR spectrum displaying g values of 9.30 and 4.29 (DMF, 4 K) (g
values of 9.6 and 4.30 (MeCN, 4 K)) is attributed to the formation of
[PPN][Fe(SH)4] (5). The rapid transformation of the DMF solution
of complex 5 occurred over the course of 5 min to give 6 monitored by
UV−vis at ambient temperature. X-ray-quality crystals were obtained
by layering the MeCN solution of complex 6 with diethyl ether at −20
°C for 3 days. Conversion of complex 6 into the known
1
[Fe4S4(SH)4]2− identified by UV−vis and H NMR spectroscopy (δ
47.0 ppm (s br, SH) d7-DMF, 25 °C, TMS) was observed when
complex 6 was dissolved in DMF or MeCN at room temperature
overnight (Supporting Information Figure S13b). Attempts to measure
the IR νSH stretching frequency of complex 6 in KBr were
1
unsuccessful. H NMR (500 MHz, d7-DMF, 25 °C, TMS): δ 36.5
ppm (s br, SH). Absorption spectrum (MeCN) [nm, λmax (M−1 cm−1,
ε)]: 309 (6345), 397 (4185), 433 (3351), 482 sh (1932), 585 sh
(1360), 656 sh (677). Anal. Calcd for C72H64Fe2N2P4S6: C, 62.43; H,
4.66; N, 2.02. Found: C, 62.18; H, 4.42; N, 2.36.
EPR Measurements. EPR measurements were performed at the
X-band using a Bruker E580 spectrometer equipped with a Bruker
ELEXSYS super-high-sensitivity cavity. X-band EPR spectra of a THF
solution of complex 1 (1−PPN) in a 4 mm EPR tube at 298 K were
obtained with a microwave power of 15.000 (19.971) mW, frequency
of 9.658817 (9.496315) GHz, conversion time of 20.39 (20.48) ms,
receiver gain of 10, and modulation amplitude of 0.8 G at 100 kHz. X-
band EPR spectra of a THF solution of complex 1 (1−PPN) in a 4
mm EPR tube at 77 K were obtained with a microwave power of
15.000 (19.971) mW, frequency of 9.656525 (9.494421) GHz,
conversion time of 20.39 (81.92) ms, receiver gain of 10, and
modulation amplitude of 0.8 G at 100 kHz. X-band EPR spectra of
complex 3 in THF placed in a 4 mm EPR tube at 4 K were obtained
with a microwave power of 15.000 mW, frequency of 9.460709 GHz,
conversion time of 20.39 ms, receiver gain of 30, and modulation
amplitude of 1.6 G at 100 kHz. X-band EPR spectra of complex 5 in
DMF placed in a 4 mm EPR tube at 4 K were obtained with a
microwave power of 15.000 mW, frequency of 9.462448 GHz,
conversion time of 81.79 ms, receiver gain of 30, and modulation
amplitude of 1.6 G at 100 kHz.
X-ray Absorption Measurements. All X-ray absorption experi-
ments were carried out at the National Synchrotron Radiation
Research Center (NSRRC), Hsinchu, Taiwan. Fe K-edge spectra were
recorded at room temperature. The data were averaged, and a smooth
background was removed from all spectra by fitting a straight line to
the pre-edge region and subtracting this straight line from the entire
spectrum. Normalization of the data was accomplished by fitting a flat
polynomial to the postregion and normalizing the edge jump to 1.0 at
7400 eV. For Fe K-edge measurements, experiments were performed
in transmission mode at the BL17C wiggler beamline with a double-
H
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(10) English, D. R.; Hendrickson, D. N.; Suslick, K. S.; Eigenbrot, C.
W.; Scheidt, W. R. J. Am. Chem. Soc. 1984, 106, 7258−7259.
(11) Balch, A. L.; Cornman, C. R.; Safari, N. Organometallics 1990, 9,
2420−2421.
(12) (a) Szaciłowski, K.; Chmura, A.; Stasicka, Z. Coord. Chem. Rev.
2005, 249, 2408−2436. (b) Butler, A. R.; Megson, I. L. Chem. Rev.
2002, 102, 1155−1165. (c) Vanin, A. F. Nitric Oxide 2009, 21, 1−13.
(d) Tonzetich, Z. J.; McQuade, L. E.; Lippard, S. J. Inorg. Chem. 2010,
49, 6338−6348.
(13) (a) Kennedy, M. C.; Antholine, W. E.; Beinert, H. J. Biol. Chem.
1997, 272, 20340−20347. (b) Foster, M. W.; Cowan, J. A. J. Am.
Chem. Soc. 1999, 121, 4093−4100. (c) Rogers, P. A.; Eide, L.;
Klungland, A.; Ding, H. DNA Repair 2003, 2, 809−817. (d) Sellers, V.
M.; Johnson, M. K.; Dailey, H. A. Biochemistry 1996, 35, 2699−2704.
(e) Ding, H.; Demple, B. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 5146−
5150.
ASSOCIATED CONTENT
■
S
* Supporting Information
IR, UV−vis, 1H NMR, EPR, and XAS spectra, cyclic
voltammograms, crystallographic and computational details,
and CIF data. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
cctsou0327@gmail.com
wfliaw@mx.nthu.edu.tw
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
(14) (a) Koch, S. A.; Maelia, L. E. J. Am. Chem. Soc. 1983, 105,
5944−5945. (b) Harrop, T. C.; Song, D.; Lippard, S. J. J. Am. Chem.
Soc. 2006, 126, 3528−3529. (c) Lu, T.-T.; Chiou, S.-J.; Chen, C.-Y.;
Liaw, W.-F. Inorg. Chem. 2006, 45, 8799−8806.
■
We gratefully acknowledge financial support from the Ministry
of Science and Technology (Taiwan). We thank Dr. Jyh-Fu
Lee, the National Synchrotron Radiation Research Center of
Taiwan (NSRRC), and the National Center for High-
Performance Computing (NCHC) for their support of the
hardware and software applied in this work. We also thank Mr.
Ting-Shen Kuo and Mrs. Pei-Lin Chen for the single-crystal X-
ray structural determinations.
(15) Harrop, T. C.; Song, D.; Lippard, S. J. J. Inorg. Biochem. 2007,
101, 1730−1738.
(16) (a) Chu, C. T.-W.; Dahl, L. F. Inorg. Chem. 1977, 16, 3245−
3251. (b) Weatherill, T. D.; Rauchfuss, T. B.; Scott, R. A. Inorg. Chem.
1986, 25, 1466−1472.
(17) Tsou, C.-C.; Lu, T.-T.; Liaw, W.-F. J. Am. Chem. Soc. 2007, 129,
12626−12627.
(18) (a) Tsou, C. C.; Tsai, F. T.; Chen, H. Y.; Hsu, I. J.; Liaw, W. F.
Inorg. Chem. 2013, 52, 1631−1639. (b) Hung, M.-C.; Tsai, M.-C.;
Liaw, W.-F. Inorg. Chem. 2006, 45, 6041−6047.
(19) 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.
(20) (a) Tsai, M.-C.; Tsai, F.-T.; Lu, T.-T.; Tsai, M.-L.; Wei, Y.-C.;
Hsu, I.-J.; Lee, J.-F.; Liaw, W.-F. Inorg. Chem. 2009, 48, 9579−9591.
(b) Yeh, S.-W.; Lin, C.-W.; Li, Y.-W.; Hsu, I.-J.; Chen, C.-H.; Jang, L.-
Y.; Lee, J.-F.; Liaw, W.-F. Inorg. Chem. 2012, 51, 4076−4087.
(21) (a) Lu, T.-T.; Lai, S.-H.; Li, Y.-W.; Hsu, I.-J.; Jang, L.-Y.; Lee, J.-
F.; Chen, I.-C.; Liaw, W.-F. Inorg. Chem. 2011, 50, 5396−5406.
(b) Hopmann, K. H.; Conradie, J.; Ghosh, A. J. Phys. Chem. B 2009,
113, 10540−10547. (c) Ye, S.; Neese, F. J. Am. Chem. Soc. 2010, 132,
3646−3647.
REFERENCES
■
(1) (a) Szabo, C. Nat. Rev. Drug Discovery 2007, 6, 917−935. (b) Li,
L.; Moore, P. K. Trends Pharmacol. Sci. 2008, 29, 84−90. (c) Li, X.;
Bazer, F. W.; Gao, H.; Jobgen, W.; Johnson, G. A.; Li, P.; McKnight, J.
R.; Satterfield, M. C.; Spencer, T. E.; Wu, G. Amino Acids 2009, 37,
65−78. (d) Martelli, A.; Testai, L.; Breschi, M. C.; Blandizzi, C.; Virdis,
A.; Taddei, S.; Calderone, V. Med. Res. Rev. 2012, 32, 1093−1130.
(e) Kolluru, G. K.; Shen, X.; Bir, S. C.; Kevil, C. G. Nitric Oxide 2013,
35, 5−20. (f) Shatalin, K.; Shatalina, E.; Mironov, A.; Nudler, E. Science
2011, 334, 986−990.
(2) (a) Li, Q.; Lancaster, J. R., Jr. Nitric Oxide 2013, 35, 21−34.
(b) Bruce King, S. Free Radical Biol. Med. 2013, 55, 1−7.
(3) Coletta, C.; Papapetropoulos, A.; Erdelyi, K.; Olah, G.; Modis, K.;
Panopoulos, P.; Asimakopoulou, A.; Gero, D.; Sharina, I.; Martin, E.;
Szabo, C. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 9161−9166.
(4) Filipovic, M. R.; Miljkovic, J.; Nauser, T.; Royzen, M.; Klos, K.;
Shubina, T.; Koppenol, W. H.; Lippard, S. J.; Ivanovic-Burmazovic, I. J.
Am. Chem. Soc. 2012, 134, 12016−12027.
(22) (a) Thomas, J. T.; Robertson, J. H.; Cox, E. G. Acta Crystallogr.
1958, 11, 599−604. (b) Sanina, N. A.; F, O. S.; Aldoshin, S. M.;
Ovanesyana, N. S. Russ. Chem. Bull. 2000, 1109−1112.
(23) Montoya, L. A.; Pearce, T. F.; Hansen, R. J.; Zakharov, L. N.;
Pluth, M. D. J. Org. Chem. 2013, 78, 6550−6557.
(5) (a) Filipovic, M. R.; Ivanovic-Burmazovic, I. Chem.Eur. J. 2012,
18, 13538−13540. (b) Miljkovic, J. L.; Kenkel, I.; Ivanovic-
Burmazovic, I.; Filipovic, M. R. Angew. Chem., Int. Ed. 2013, 52,
12061−12064.
(24) (a) Li, M.; Bonnet, D.; Bill, E.; Neese, F.; Weyhermuller, T.;
̈
Blum, N.; Sellmann, D.; Wieghardt, K. Inorg. Chem. 2002, 41, 3444−
3456. (b) Conradie, J.; Quarless, D. A. J.; Hsu, H.-F.; Harrop, T. C.;
Lippard, S. J.; Koch, S. A.; Ghosh, A. J. Am. Chem. Soc. 2007, 129,
10446−10456.
(6) (a) Ma, E. S.; Rettig, S. J.; Patrick, B. O.; James, B. R. Inorg. Chem.
2012, 51, 5427−5434. (b) Perez-Torrente, J. J.; Jimenez, M. V.;
Hernandez-Gruel, M. A.; Fabra, M. J.; Lahoz, F. J.; Oro, L. A. Chem.
Eur. J. 2009, 15, 12212−12222. (c) Mudalige, D. C.; Ma, E. S.; Rettig,
S. J.; James, B. R.; Cullen, W. R. Inorg. Chem. 1997, 36, 5426−5427.
(d) Sellman, D.; Lechner, P.; Knoch, F.; Moll, M. Angew. Chem., Int.
Ed. Engl. 1991, 30, 552−553.
(25) (a) Tsou, C. C.; Yang, W. L.; Liaw, W. F. J. Am. Chem. Soc.
2013, 135, 18758−18761. (b) Tsai, F.-T.; Chen, P.-L.; Liaw, W.-F. J.
Am. Chem. Soc. 2010, 132, 5290−5299.
(26) Hopmann, K. H.; Noodleman, L.; Ghosh, A. Chem.Eur. J.
2010, 16, 10397−10408.
(27) (a) Harrop, T. C.; Tonzetich, Z. J.; Reisner, E.; Lippard, S. J. J.
Am. Chem. Soc. 2008, 130, 15602−15610. (b) Tennyson, A. G.; Dhar,
S.; Lippard, S. J. J. Am. Chem. Soc. 2008, 130, 15087−15098.
(28) Lu, T.-T.; Huang, H.-W.; Liaw, W.-F. Inorg. Chem. 2009, 48,
9027−9035.
(7) Galardon, E.; Roger, T.; Deschamps, P.; Roussel, P.; Tomas, A.;
Artaud, I. Inorg. Chem. 2012, 51, 10068−10070.
(8) (a) Venkateswara, P.; Holm, R. H. Chem. Rev. 2004, 104, 527−
559. (b) Kuwata, S.; Hidai, M. Coord. Chem. Rev. 2001, 213, 211−305.
(9) (a) Jove, F. A.; Pariya, C.; Scoblete, M.; Yap, G. P.; Theopold, K.
H. Chem.Eur. J. 2011, 17, 1310−1318. (b) Pavlik, J. W.; Noll, B. C.;
Oliver, A. G.; Schulz, C. E.; Scheidt, W. R. Inorg. Chem. 2010, 49,
1017−1026. (c) Dhifet, M.; Belkhiria, M. S.; Daran, J. C.; Nasri, H.
Acta Crystallogr., Sect. E 2009, 65, m967−m968. (d) Arif, A. M.;
Hefner, J. G.; Jones, R. A.; Koschmieder, S. U. J. Coord. Chem. 1991,
23, 13−19. (e) Vaira, M. D.; Midollini, S.; Sacconi, L. Inorg. Chem.
1977, 16, 1518−1524. (f) Cai, L.; Holm, R. H. J. Am. Chem. Soc. 1994,
116, 7177−7188.
(29) (a) Maelia, L. E.; Millar, M.; Koch, S. A. Inorg. Chem. 1992, 31,
4594−4600. (b) Chang, S.; Koch, S. A. J. Inorg. Biochem. 2007, 101,
1758−1759.
(30) (a) Noodleman, L.; Case, D. A.; Aizman, A. J. Am. Chem. Soc.
1988, 110, 1001−1005. (b) Gillum, W. O.; Frankel, R. B.; Foner, S.;
Holm, R. H. Inorg. Chem. 1976, 15, 1095−1100.
(31) Hoveyda, H. R.; Holm, R. H. Inorg. Chem. 1997, 36, 4571−
4578.
I
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(32) (a) Claser, T. G.; Rose, K.; Shadle, S. E.; Hedman, B.; Hodgson,
K. O.; Solomon, E. I. J. Am. Chem. Soc. 2001, 123, 442−454. (b) Rose,
K.; Shadle, S. E.; Glaser, T.; de Vries, S.; Cherepanov, A.; Canters, G.
W.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc.
1999, 121, 2353−2363.
(33) (a) Perdew, J. P.; Yang, W. Phys. Rev. B 1986, 33, 8800−8802.
(b) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100.
(34) Perdew, J. P. Phys. Chem. Chem. Phys. 1986, 33, 8822−8824.
(35) Check, C. E.; Faust, T. O.; Bailey, J. M.; Wright, B. J.; Gilbert, T.
M.; Sunderlin, L. S. J. Phys. Chem. A 2001, 105, 8111−8116.
(36) (a) Hay, P. J. J. Chem. Phys. 1977, 66, 4377−4384. (b) Krishnan,
R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72, 650−
654. (c) Raghavachari, K.; Trucks, G. W. J. Chem. Phys. 1989, 91,
1062−1065. (d) Wachters, A. J. H. J. Chem. Phys. 1970, 52, 1033−
1037.
(37) Brothers, S. M.; Darensbourg, M. Y.; Hall, M. B. Inorg. Chem.
2011, 50, 8532−8540.
(38) (a) Schmitt, E. A.; Noodleman, L.; Baerends, E. J.; Hendrickson,
D. N. J. Am. Chem. Soc. 1992, 114, 6109−6119. (b) Hopmann, K. H.;
Ghosh, A.; Noodleman, L. Inorg. Chem. 2009, 48, 9155−9165.
(c) Jaworska, M.; Stasicka, Z. New J. Chem. 2005, 29, 604−612.
(39) Hagen, K. S.; Watson, A. D.; Holm, R. H. J. Am. Chem. Soc.
1983, 105, 3905−3913.
(40) Soda, T.; Kitagawa, Y.; Onishi, T.; Takano, Y.; Shigeta, Y.;
Nagao, H.; Yoshioka, Y.; Yamaguchi, K. Chem. Phys. Lett. 2000, 319,
223−230.
(41) (a) Suzuki, N.; Higuchi, T.; Urano, Y.; Kikuchi, K.; Uekusa, H.;
Ohashi, Y.; Uchida, T.; Kitagawa, T.; Nagano, T. J. Am. Chem. Soc.
1999, 121, 11571−11572. (b) Das, P. K.; Chatterjee, S.; Samanta, S.;
Dey, A. Inorg. Chem. 2012, 51, 10704−10714.
(42) (a) Johnson, D. C.; Dean, D. R.; Smith, A. D.; Johnson, M. K.
Annu. Rev. Biochem. 2005, 74, 247−281. (b) Lill, R.; Muhlenhoff, U.
Trends Biochem. Sci. 2005, 30, 133−141. (c) Rouault, T. A.; Tong, W.-
H. Nat. Rev. Mol. Cell Biol. 2005, 6, 345−351.
(43) Chen, Y. H.; Tsai, J. C.; Cheng, T. H.; Yuan, S. S.; Wang, Y. M.
Biosens. Bioelectron. 2014, 56, 117−123.
(44) Sheldrick, G. M. SHELXTL, Program for Crystal Structure
Determination; Siemens Analytical X-ray Instruments Inc.: Madison,
WI, 1994.
(45) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.;
Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao,
O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J.
B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R.
E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.;
Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J.
J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. G.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 09, revision
A.02; Gaussian, Inc.; Wallingford, CT, 2009.
(46) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117−
129.
(47) Chemissian v3.3, http://www.chemissian.com (accessed Sep-
tember 8, 2012).
(48) (a) Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102,
7211−7218. (b) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem.
Phys. 1985, 83, 735−746. (c) Reed, A. E.; Weinhold, F. J. Chem. Phys.
1983, 78, 4066−4073.
J
dx.doi.org/10.1021/ja503683y | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX