Relative binding affinity of thiolate, imidazolate, phenoxide, and nitrite toward the (fe(no)2) motif of dinitrosyl iron complexes (dnics): the characteristic pre-edge energy of {fe(no)2}9 dnics

Article abstract of DOI:10.1021/ic901675p

The synthesis, characterization, and transformation of the anionic {Fe(NO)₂}⁹ dinitrosyl Iron complexes (DNICs) [(NO) ₂Fe(ONO)₂]^- (1), [(NO)₂Fe(OPh) ₂]^- (2), [(NO)

Full text of DOI:10.1021/ic901675p

Inorg. Chem. 2009, 48, 9579–9591 9579
DOI: 10.1021/ic901675p
Relative Binding Affinity of Thiolate, Imidazolate, Phenoxide, and Nitrite Toward the
{Fe(NO)2} Motif of Dinitrosyl Iron Complexes (DNICs): The Characteristic Pre-Edge
Energy of {Fe(NO)2}9 DNICs
Ming-Che Tsai,† Fu-Te Tsai,† Tsai-Te Lu,† Ming-Li Tsai,† Yin-Ching Wei,† I-Jui Hsu,*,‡ Jyh-Fu Lee,§ and
Wen-Feng Liaw*,†
†Department of Chemistry, National Tsing Hua University, Hsinchu 30013, Taiwan, ‡Department of Molecular
Science and Engineering, National Taipei University of Technology, Taipei 10608, Taiwan, and
§National Synchrotron Radiation Research Center, Hsinchu, Taiwan
Received August 22, 2009
The synthesis, characterization, and transformation of the anionic {Fe(NO)2}9 dinitrosyl iron complexes (DNICs)
[(NO)2Fe(ONO)2]- (1), [(NO)2Fe(OPh)2]- (2), [(NO)2Fe(OPh)(C3H3N2)]- (3) (C3H3N2=imidazolate), [(NO)2Fe-
(OPh)(-SC4H3S)]- (4), [(NO)2Fe(p-OPhF)2]- (5), and [(NO)2Fe(SPh)(ONO)]- (6) were investigated. The binding
affinity of ligands ([SPh]-, [-SC4H3S]-, [C3H3N2]-, [OPh]-, and [NO2]-) toward the {Fe(NO)2}9 motif follows the
ligand-displacement series [SPh]- ∼ [-SC4H3S]- > [C3H3N2]- > [OPh]- > [NO2]-. The findings, the pre-edge
energy derived from the 1s f 3d transition in a distorted Td environment of the Fe center falling within the range of
7113.4-7113.8 eV for the anionic {Fe(NO)2}9 DNICs, implicate that the iron metal center of DNICs is tailored to
minimize the electronic changes accompanying changes in coordinated ligands. Our results bridging the ligand-
substitution reaction study and X-ray absorption spectroscopy study of the electronic richness of the {Fe(NO)2}9 core
may point the way to understanding the reasons for nature’s choice of combinations of cysteine, histidine, and tyrosine
in protein-bound DNICs and rationalize that most DNICs characterized/proposed nowadays are bound to the proteins
almost through the thiolate groups of cysteinate/glutathione side chains in biological systems.
Introduction
quantitatively converted to paramagnetic protein-bound
DNICs from the exposure of CIP to free •NO.4d Although
cysteine and glutathione have been proposed to be the major
thiol components of cellular DNICs in vivo,5 four different
kinds of potential coordinated ligands (cysteinate, histidine,
deprotonated imidazole, and tyrosinate) in protein-bound
DNICs, on the basis of EPR spectra, were proposed in
enzymology.6 Recently, the protein-bound DNIC resulting
from the nitrosylation of human glutathione transferase
P1-1 by a dinitrosyl diglutathionyliron complex in vitro/in vivo
Dinitrosyliron complexes (DNICs), characterized by their
distinctive electron paramagnetic resonance (EPR) signals at
g=2.03 in vitro/in vivo, have been suggested as one of the
possible forms for storage and transport of NO in biological
systems.1-3 The binding of NO to [Fe-S] cluster-containing
proteins or enzymes in mitochondria yielding DNICs and
dinuclear [Fe(μ-SR)(NO)2]2 (Roussin’s red ester) have been
intensely studied.1,3,4a-4c In addition, it has been demon-
strated that the chelatable iron pool (CIP) is rapidly and
(4) (a) Ding, H.; Demple, B. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 5146–
5150. (b) 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.;
Bello, M. L. J. Biol. Chem. 2005, 280, 42172–42180. (c) Maria, F. D.; Pedersen,
J. Z.; Caccuri, A. M.; Antonini, G.; Turella, P.; Stella, L.; Bello, M. L.; Federici,
G.; Ricci, G. J. Biol. Chem. 2003, 278, 42283–42293. (d) Toledo, J. C. Jr.;
Bosworth C. A.; Hennon, S. W.; Mahtani, H. A.; Bergonia, H. A.; Lancaster, J. R. Jr.
J. Biol. Chem. 2008 283, 28926–28933.
*To whom correspondence should be addressed. E-mail: wfliaw@
mx.nthu.edu.tw (W.-F.L.).
(1) (a) Lancaster, J. R. Jr.; Hibbs J. B. Jr. Proc. Natl. Acad. Sci. U.S.A.
1990 87, 1223–1227. (b) Foster, M. W.; Cowan, J. A. J. Am. Chem. Soc. 1999,
121, 4093–4100. (c) Cooper, C. E. Biochim. Biophys. Acta 1999, 1411, 290–309.
(d) Butler, A. R.; Megson, I. L. Chem. Rev. 2002, 102, 1155–1166. (e) Ueno, T.;
Susuki, Y.; Fujii, S.; Vanin, A. F.; Yoshimura, T. Biochem. Pharmacol. 2002, 63,
(5) (a) Lewandowska, H.; Meczynska, S.; Sochanowicz, B.; Sadlo, J.;
Kruszewski, M. J. Biol. Inorg. Chem. 2007, 12, 345–352. (b) Stadler, J.;
Bergonia, H. A.; Di Silvio, M.; Sweetland, M. A.; Billiar, T. R.; Simmons, R. L.;
Lancaster, J. R. Jr. Arch. Biochem. Biophys. 1993 302, 4–11.
485–493. (f) McCleverty, J. A. Chem. Rev. 2004, 104, 403–418.
::
(2) (a) Mulsch, A.; Mordvintcev, P. I.; Vanin, A. F.; Busse, R. FEBS Lett.
1991, 294, 252–256. (b) Badorff, C.; Fichtlscherer, B.; Muelsch, A.; Zeiher,
A. M.; Dimmeler, S. Nitric Oxide 2002, 6, 305–312.
(3) (a) Stamler, J. S.; Singel, D. J.; Loscalzo, J. Science 1992, 258, 1898–
1902. (b) Stamler, J. S. Cell 1994, 78, 931–936. (c) Frederik, A. C.; Wiegant, I. Y.;
Malyshev, I. Y.; Kleschyov, A. L.; van Faassen, E.; Vanin, A. F. FEBS Lett. 1999,
455, 179–182.
(6) (a) Kennedy, M. C.; Antholine, W. E.; Beinert, H. J. Biol. Chem. 1997,
272, 20340–20347. (b) Boese, M.; Mordvintcev, P. I.; Vanin, A. F.; Busse, R.;
::
Mulsch, A. J. Biol. Chem. 1995, 270, 29244–29249. (c) Lee, M.; Arosio, P.;
Cozzi, A.; Chasteen, N. D. Biochemistry 1994, 33, 3679–3687.
r
2009 American Chemical Society
Published on Web 09/11/2009
pubs.acs.org/IC

9580 Inorganic Chemistry, Vol. 48, No. 19, 2009
Tsai et al.
was characterized by single-crystal X-ray crystallography.4b
Upon the addition of NO to the [3Fe-4S] form of
m-aconitase, new transient signals, tentatively assigned as
DNICs containing two histidyl ligands or DNICs containing
mixed histidine and cysteine ligands, appear during the early
phases of the reaction.6a
Also known in inorganic chemistry is the precedent for
low-molecular-weight DNICs at four oxidation levels, in-
cluding the EPR-active, anionic/neutral/cationic {Fe(NO)2}9
DNICs as well as the EPR-silent, neutral {Fe(NO)2}10
DNICs coordinated by CO, PPh3, and N-containing ligands.7-10
Here, the electronic state of the {Fe(NO)2} unit of DNICs is
generally designated as {Fe(NO)2}n (n is the total number of
electrons associated with the metal d and π* (NO) orbitals),
invoking the Enemark-Feltham notation.11 Recently, we
have established that the IR νNO in combination with EPR
spectra (pattern) may be employed to discriminate between
the {Fe(NO)2}9 DNICs with a variety of coordinated ligands
(thiolate, imidazole, and imidazolate).7c,12 Transformations
of [(NO)2Fe(μ-SC6H4-o-NHCOPh)]2 into the neutral
{Fe(NO)2}10 DNICs [(NO)2Fe(PPh3)2] via reductive elimi-
nation of the bridged thiolates and into the anionic
{Fe(NO)2}9 [(NO)2Fe(SC6H4-o-NCOPh)]- containing anio-
nic sulfur-amide chelating ligands were elucidated to be
controlled by the nucleophile L (L = PPh3, [OPh]-).13 In
addition, conversion of the EPR-active, anionic {Fe(NO)2}9
DNICs [(NO)2Fe(C3H3N2)2]- (C3H3N2=imidazolate) into
the anionic {Fe(NO)2}9 [(NO)2Fe(C3H3N2)(SR)]- (R=tBu,
Et, Ph) via ligand exchange was demonstrated.12
Figure 1. ORTEP drawing and labeling scheme of complex 1 with
˚
thermal ellipsoids drawn at 50% probability. Selected bond lengths (A)
and angles (deg): Fe(1)-N(4) 1.691(2), Fe(1)-N(3) 1.699(2), Fe(1)-O(5)
2.002(1), Fe(1)-O(3) 2.038(1), N(3)-O(1) 1.173(2), N(4)-O(2) 1.173(2),
N(5)-O(4) 1.236(2), N(5)-O(3) 1.271(2), N(6)-O(6) 1.214(2), N(6)-O-
(5) 1.303(2), O(1)-N(3)-Fe(1) 161.1(2), O(2)-N(4)-Fe(1) 164.3(2),
N(3)-Fe(1)-N(4) 110.3(1), O(4)-N(5)-O(3) 112.9(2), O(6)-N(6)-O-
(5) 115.0(2), O(3)-Fe(1)-O(5) 93.5(1).
antiferromagnetically coupled to two NO (S=1/2) ligands”.7a
Although some DNICs have been proposed to contain [N,
O]/[O,O]-coordinate ligands in enzymology,6 no mononuc-
lear DNICs coordinated by phenoxide, the mixed phenox-
ide-imidazolate, and the mixed phenoxide-thiolate have
been isolated and characterized in synthetic chemistry. In this
manuscript, the synthesis, reactivity, and transformation of
the EPR-active, anionic {Fe(NO)2}9 [(NO)2Fe(ONO)2]- (1),
[(NO)2Fe(OPh)2]- (2), [(NO)2Fe(OPh)(C3H3N2)]- (C3H3N2=
imidazolate) (3), [(NO)2Fe(OPh)(-SC4H3S)]- (-SC4H3S=
thienylthiolate) (4), [(NO)2Fe(p-OPhF)2]- (5), and [(NO)2-
Fe(SPh)(ONO)]- (6) are described. The binding affinity of
nitrite/phenoxide/imidazolate/thiolate toward the {Fe(NO)2}9
motif was investigated. Also, the present work was under-
taken to propose the electronic structure of the {Fe(NO)2}
core of {Fe(NO)2}9 DNICs containing the various ligation
modes [S,S]/[S,O]/[O,O]/[N,O] on the basis of crystal struct-
ure data, EPR spectroscopy, X-ray absorption spectroscopy
(XAS) of the Fe K-edge, and DFT computation.
Because of the small energy difference between transition-
metal d and NO π* orbitals, it is complicated to define the
“noninnocent” character of NO acting as NOþ, NO;, or
NO.14 Within the class of Fe(NO)2}9 DNICs, the oxidation
state of iron and NO are in controversy. Intriguingly, a recent
investigation of the electronic structure of the {Fe(NO)2}9
;
unit of complex [(NO)2FeS5]
inferred that [(NO)2-
FeS5]; is better described as an {FeI(NO)2}9 electronic
structure related to the spin configuration as “Fe (S=3/2)
(7) (a) Tsai, M. L.; Chen, C. C.; Hsu, I. J.; Ke, S. C.; Hsieh, C. H.; Chiang,
K. A.; Lee, G. H.; Wang, Y.; Liaw, W. F. Inorg. Chem. 2004, 43, 5159–5167.
(b) 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. (c) Tsai, M. L.; Liaw, W. F. Inorg.
Chem. 2006, 45, 6583–6585. (d) Hung, M. C.; Tsai, M. C.; Lee, G. H.; Liaw, W. F.
Inorg. Chem. 2006, 45, 6041–6047. (e) Lu, T. T.; Chiou, S. J.; Chen, C. .Y.; Liaw,
W. F. Inorg. Chem. 2006, 45, 8799–8806. (f) Chen, T. N.; Lo, F. C.; Tsai, M. L.;
Shih, K. N.; Chiang, M. H.; Lee, G. H.; Liaw, W. F. Inorg. Chim. Acta 2006, 359,
2525–2533. (g) Chiou, S. J.; Wang, C. C.; Chang, C. M. J. Organomet. Chem.
2008, 693, 3582–3586. (h) Hsu, I.-J.; Hsieh, C.-H.; Ke, S.-C.; Chiang, K.-A.; Lee,
J.-M.; Chen, J.-M.; Jang, L.-Y.; Lee, G.-H.; Wang, Y.; Liaw, W.-F. J. Am. Chem.
Soc. 2007, 129, 1151–1159. (i) Lee, C.-M.; Chen, C.-H.; Dutta, A.; Lee, G.-H.;
Liaw, W.-F. J. Am. Chem. Soc. 2003, 125, 11492–11493.
(8) (a) Chiang, C. Y.; Miller, M. L.; Reibenspies, J. H.; Darensbourg, M.
Y. J. Am. Chem. Soc. 2004, 126, 10867–10874. (b) Baltusis, L. M.; Karlin, K.
D.; Rabinowitz, H. N.; Dewan, J. C.; Lippard, S. J. Inorg. Chem. 1980, 19, 2627–
2632.
(9) Albano, V. G.; Araneo, A.; Bellon, P. L.; Chani, G.; Manassero, M. J.
Organomet. Chem. 1974, 67, 413–422.
Results and Discussion
Upon the addition of [PPN][NO2] into a THF solution of
complex [Fe(CO)2(NO)2][BF4] in a 2:1 stoichiometry,15,16
a
reaction ensued over the course of 5 min to yield the anionic
{Fe(NO)2}9 [PPN][(NO)2Fe(ONO)2] (1) containing two O-
bound nitrito ligands, identified by IR, UV-vis, EPR, and
single-crystal X-ray crystallography (Figure 1). Complex 1 is
thermally stable in THF solution and exhibits diagnostic IR
ν
NO stretching frequencies at 1775 s and 1705 s cm-1 (THF).
The EPR spectrum (Figure 2) of complex 1 displays an
isotropic signal with g=2.033 at 298 K and a rhombic signal
with g1=2.052, g2=2.029, and g3=2.014 at 77 K, consistent
with the characteristic g value of {Fe(NO)2}9 DNICs.7,12,13,15
In contrast to the decomposition observed in the reaction
of [Fe(CO)2(NO)2]þ and [Na][OPh] in THF at ambient
temperature, the addition of 2 equiv of [Na][OPh] to the
(10) Reginato, N.; McCrory, C. T. C.; Pervitsky, D.; Li, L. J. Am. Chem.
Soc. 1999, 121, 10217–10218.
(11) Enemark, J. H.; Feltham, R. D. Coord. Chem. Rev. 1974, 13, 339–
406.
(12) Huang, H. W.; Tsou, C. C.; Kuo, T. S.; Liaw, W. F. Inorg. Chem.
2008, 47, 2196–2204.
(13) Tsai, M. L.; Hsieh, C. H.; Liaw, W. F. Inorg. Chem. 2007, 46, 5110–
5117.
(14) (a) Franz, K. J.; Lippard, S. J. J. Am. Chem. Soc. 1998, 120, 9034–
9040. (b) Brown, C. A.; Pavlosky, M. A.; Westre, T. E.; Zhang, Y.; Hedman, B.;
Hodgson, K. O.; Solomen, E. I. J. Am. Chem. Soc. 1995, 117, 715–732.
(15) Tsai, F. T.; Kuo, T. S.; Liaw, W. F. J. Am. Chem. Soc. 2009, 131,
3426–3427.
(16) Atkinson, F. I.; Blackwell, H. E.; Brown, N. C.; Connelly, N. G.;
Crossley, J. G.; Orpen, A. G.; Rieger, A. L.; Rieger, P. H. J. Chem. Soc.,
Dalton Trans. 1996, 3491–3502.

Article
Inorganic Chemistry, Vol. 48, No. 19, 2009 9581
Figure 2. EPR spectrum of complex 1 (a) at 298 K (g=2.033) and (b) at 77 K (g1=2.052, g2=2.029, and g3=2.014).
Scheme 1
observed in the previous study that the stronger electron-
donating thiolates [R0S]-, compared to the coordinated
thiolates of [(NO)2Fe(SR)2]-, may promote thiolate-ligand
exchange to produce the stable [(NO)2Fe(SR0)2]-.7b The
conversion of complex 1 to 2 was monitored by IR (IR νNO
shifting from 1775 s and 1705 s cm-1 to 1739 m and 1674 s
cm-1) and UV-vis spectra (the intense bands at 510 and 615
nm disappeared, accompanied by the simultaneous forma-
tion of the absorption bands at 427, 502, and 657 nm). The
EPR spectra (Figure 4) exhibiting a well-resolved five-line
EPR signal with a g value of 2.025 (the hyperfine coupling
constants aN(NO) =2.54 G) at 298 K and a rhombic signal
with g1 = 2.041, g2 = 2.022, and g3 = 2.013 at 77 K are
consistent with the formation of complex 2. The magnetic
susceptibility measurement of the powder sample of complex
2 was collected in the temperature range of 2.00-300 K at
5 kG (0.5 T). The net molar magnetic susceptibility (χM)
increased from 7.47 ꢀ 10-4 cm3 mol-1 at 300 K to 8.04 ꢀ
10-2 cm3 mol-1 at 2 K. The temperature-dependent effective
magnetic moment (μeff) decreases from 2.058 μB at 300 K to
1.141 μB at 2 K (Figure S1, Supporting Information).
Irreversibly, the formation of complex 1 was not observed
upon the addition of 2 equiv of [NO2]- to the THF solution
of complex 2 (Scheme 1a).
Figure 3. ORTEP drawing and labeling scheme of complex 2 with
thermal ellipsoids drawn at 30% probability. Selected bond distances
(A) and angles (deg): Fe(1)-N(1) 1.696(3), Fe(1)-N(2) 1.683(3),
Fe(1)-O(3) 1.913(2), Fe(1)-O(4) 1.920(2), N(1)-O(1) 1.177(4), N(2)-
O(2) 1.182(4), N(2)-Fe(1)-N(1) 110.2(2), N(1)-Fe(1)-O(3) 104.8(2),
N(2)-Fe(1)-O(3) 112.4(1), N(1)-Fe(1)-O(4) 108.3(1), N(2)-Fe(1)-O-
(4) 114.9(1), O(3)-Fe(1)-O(4) 105.6(1), O(1)-N(1)-Fe(1) 156.4(4),
O(2)-N(2)-Fe(1) 164.1(3).
˚
THF solution of complex 1 led to the formation of the
anionic {Fe(NO)2}9 [PPN][(NO)2Fe(OPh)2] (2) (yield 85%)
with monodentate phenoxide ligands coordinated to the
{Fe(NO)2}9 motif (Scheme 1a), characterized by IR,
UV-vis, EPR, SQUID, and single-crystal X-ray crystallo-
graphy (Figure 3). This result seems to follow the rule
In order to elucidate the reactivity of complex 2 and
synthesize dinitrosyliron complexes with mixed imidazola-
te-phenoxide ligands, the reaction of complex 2 and imidaz-
olate was investigated. Careful injection of 1 equiv of the
THF-diluted sodium imidazolate solution ([Na][C3H3N2])

9582 Inorganic Chemistry, Vol. 48, No. 19, 2009
Tsai et al.
Figure 4. EPR spectrum of complex 2 (a) at 298 K (g=2.025) and (b) at 77 K (g1=2.041, g2=2.022, and g3=2.013) and the simulated EPR spectrum of
complex 2 with hyperfine coupling constants of aN(NO)=2.54 G.
˚
Figure 5. ORTEP drawing and labelingschemeof complex3-Na with thermal ellipsoidsdrawn at30% probability. Selectedbonddistances (A) and angles
(deg): Fe(1)-N(1) 1.692(3), Fe(1)-N(2) 1.685(3), Fe(1)-O(3) 1.884(2), Fe(1)-N(3) 1.995(2), N(1)-O(1) 1.173(3), N(2)-O(2) 1.181(3), N(2)-Fe(1)-N(1)
110.5(1), N(2)-Fe(1)-O(3) 119.2(1), N(1)-Fe(1)-O(3) 107.9(1), N(1)-Fe(1)-O(3) 107.9(1), N(3)-Fe(1)-O(3) 101.6(1), N(2)-Fe(1)-N(3) 107.4(1),
N(1)-Fe(1)-N(3) 109.8(1), O(1)-N(1)-Fe(1) 164.5(2), O(2)-N(2)-Fe(1) 160.7(3).
into the THF solution of complex 2 at 0 ꢀC led to the IR νNO
stretching frequency shift from 1739 m and 1674 s to 1755 m
and 1691 s cm-1 (THF), consistent with the ligand displace-
ment of one [OPh]- yielding [cation][(NO)2Fe(OPh)-
(C3H3N2)] (cation = PPN (3-PPN), Na-18-crown-6-ether
(3-Na); Scheme 1b). Complex 3-Na was characterized by
single-crystal X-ray crystallography (Figure 5). The UV-vis
spectrum of complex 3-Na shows absorptions at 216 and 270 nm.
The EPR spectra (Figure 6) of complex 3-Na exhibit a nine-
line spectrum with g = 2.026 (the hyperfine coupling con-
stants aN(NO)=2.42 G and aN(Im)=4.15 G) at 298 K, and a
rhombic signal with g1=2.046, g2=2.021, and g3=2.013 at 77 K.
However, the addition of 1 equiv of [OPh]- to complex
3-PPN in THF shows no spectroscopically detectable
changes in the IR and UV-vis spectra (Scheme 1b0). Upon
the addition of a second equivalent of the THF-diluted
sodium imidazolate to the THF solution of complex 3-Na,
a pronounced color change from dark-brown to red-brown
occurs at 0 ꢀC. The IR, UV-vis, EPR, and single-crystal
X-ray diffraction studies confirmed the formation of the
known anionic {Fe(NO)2}9 [(NO)2Fe(C3H3N2)2]- with two
anionic [C3H3N2]- ligands bound to the {Fe(NO)2} core
(Scheme 1c).12
and imidazolate, was determined through the addition of
[-SC4H3S]-/[SEt]- to the THF solution of complex 2 at
0 ꢀC, respectively. Upon 1 equiv of the THF-diluted sodium
thienylthiolate ([Na][-SC4H3S]) added to the THF solution
of complex 2 dropwise at 0 ꢀC, the IR νNO (1740 s, 1681 s
cm-1) suggested the formation of the anionic {Fe(NO)2}9
[PPN][(NO)2Fe(OPh)(-SC4H3S)] (4; Scheme 1d). At 298 K,
complex 4 displays a well-resolved five-line EPR signal with a
g value of 2.028 (the hyperfine splitting constant aN(NO)
=
2.58 G) and a rhombic signal with g1=2.038, g2=2.027, and
g3 =2.013 at 77 K (Figure 7). However, the addition of 1
equiv of [OPh]- to complex 4 in THF reveals no detectable
changes (Scheme 1d0). This result implicates that the
{Fe(NO)2} motif displays a preference for [-SC4H3S]- over
[OPh]-. In contrast to the reaction of complex 2 and
[Na][-SC4H3S] leading to complex 4, the addition of 1 equiv
of the THF-diluted [SEt]- to the THF solution of complex 2
resulted in the formation of the known dinuclear complex
[Fe2(μ-SEt)2(NO)4], characterized by the IR νNO (shift from
1739 m and 1674 s to 1808 w, 1774 s, and 1749 s cm-1).17 No
detectable quantities of [(NO)2Fe(OPh)(SEt)]- were observed
spectrally (Scheme 1e). Instead of the formation of the
five-coordinate intermediate [(NO)2Fe(OPh)2(-SC4H3S)]2-
The relative binding affinity of the {Fe(NO)2} motif for
thienylthiolate/ethylthiolate, compared to phenoxide, nitrite,
(17) Rauchfuss, T. B.; Weatherill, T. D. Inorg. Chem. 1982, 21, 827–830.

Article
Inorganic Chemistry, Vol. 48, No. 19, 2009 9583
Figure 6. EPR spectrumof complex 3-Na(a) with anisotropic gav=2.026 (a well-resolved nine-lineEPR signal) at 298 K and(b) with a rhombic g1=2.046,
g2=2.021, and g3=2.013 and the simulated EPR spectrum of complex 3 with hyperfine coupling constants of aN(NO)=2.42 G and aN(Im)=4.15 G.
Figure 7. EPR spectrum ofcomplex4 (a) at298 K (g = 2.028) and(b) at77K (g1 = 2.038, g2 = 2.027, and g3 = 2.013) andthe simulated EPR spectrumof
complex 4 with hyperfine coupling constants of aN(NO) = 2.58 G.
(Scheme 1f 0). As expected, the addition of 1 equiv of [SPh]-
Scheme 2
into the THF solution of complex 1 yielded [(NO)2-
Fe(SPh)(ONO)]- (6), characterized by IR, UV-vis, and
single-crystal X-ray crystallography (Scheme 1g), and the
reaction of complex 1 with 2 equiv of imidazolate led to
the formation of the known [(NO)2Fe(C3H3N2)2]-
(Scheme 1h).12
The unique ability of the {Fe(NO)2} motif to bind nitrites,
phenoxides, and imidazolates as well as thiolates provides the
opportunity to directly probe the binding preference of
{Fe(NO)2} motif of DNICs. In combination with the pre-
vious studies,7b the stronger electron-donating thiolates [R0S]
promoting thiolate ligand exchange to produce the stable
[(NO)2Fe(R0S)2]- and the more electron-donating function-
(Scheme 2a,b), it is presumed that a nucleophilic attack of the
stronger electron-donating [SEt]- on complex 2 resulted in
ality of thiolates triggering the transformation of
[(NO)2Fe(C3H3N2)2]- into complex [(NO)2Fe(C3H3N2)-
the buildup of the unstable three-coordinate intermediate
(SR)]- (R=tBu, Et, Ph) via an imidazolate-thiolate exchange
reaction,12 these results conclude that the binding preference
for a given ligand ([SPh]-, [-SC4H3S]-, [C3H3N2]-, [OPh]-,
and [NO2]-) of the {Fe(NO)2} motif follows the ligand-
displacement series [SPh]- ∼ [-SC4H3S]- > [C3H3N2]- >
[OPh]- > [NO2]- (Scheme 3). This order, presumably, is
related to the electronic structure of the {Fe(NO)2} core, an
effect that leads to a specific ligand-substitution reaction. The
ability of the coordinated ligands to tune the electronic
structure of the {Fe(NO)2} core of DNICs affords the
optimum electronic conditions to stabilize DNICs. Due to
the “homeostasis” of electron density among Fe, NO, and the
varieties of coordinating ligands, it is presumed that the
electron density surrounding the Fe center of DNICs coordinated
by the distinct electron-donating ligands may also be modulated
by the noninnocent NO ligands via adjustment of the oxida-
tion level. Although modification of the coordinated ligands
in a systematic manner to delineate the electronic structure of
the {Fe(NO)2} core is still a challenge, it may be possible to
rationalize the preferred formation of thiolate-containing
[(NO)2Fe(SEt)] accompanied by the release [OPh]-, which
subsequently dimerized to yield [Fe2(μ-SEt)2(NO)4]
(Scheme 2c,d).
In order to rationalize the ligand-exchange reactions ob-
served in the above study, the ligand-substitution chemistry
of complex 2 with 2 equiv of theanalogue [p-OPhF]- was also
conducted. Transformation of complex 2 into [PPN][(NO)2-
Fe(p-OPhF)2] (5) monitored by IR νNO spectroscopy (the
shift of the νNO stretching frequencies from 1739 m and1674 s
cm-1 to 1751 m and 1685 s cm-1) was observed when a THF
solution of complex 2 was reacted with 2 equiv of [p-OPhF]-
at ambient temperature (Scheme 1f). Complex 5 was char-
acterized by IR, UV-vis, and single-crystal X-ray crystal-
lography (Figure S2, Supporting Information). The EPR
spectra (Figure S3, Supporting Information) of complex 5
exhibits a well-resolved five-line signal with g = 2.026 (the
hyperfine splitting constants aN(NO)=2.52 G) at 298 K and
a rhombic signal with g1=2.041, g2=2.024, and g3=2.013 at
77 K. The addition of 2 equiv of [OPh]- to complex 5 in THF
shows no detectable changes in the IR and UV-vis spectra

9584 Inorganic Chemistry, Vol. 48, No. 19, 2009
Tsai et al.
Scheme 3
Figure 9. ORTEP drawing and labeling scheme of complex 6 with
thermal ellipsoids drawn at 50% probability. Selected bond distances
˚
(A) and angles (deg): Fe(1)-N(1) 1.686(6), Fe(1)-N(2) 1.641(6), Fe-
(1)-S(1) 2.231(2), Fe(1)-O(3) 2.043(6), N(1)-O(1) 1.157(6), N(2)-O(2)
1.176(6), N(3)-O(3) 1.186(7), N(3)-O(4) 1.100(1), Fe(1)-N(1)-O(1)
162.8(5), Fe(1)-N(2)-O(2) 165.7(5), O(3)-Fe(1)-S(1) 110.8(2).
Figure 8. ORTEP drawing and labeling scheme of complex 4 with
thermal ellipsoids drawn at 30% probability. Selected bond distances
(A) and angles (deg): Fe(1)-N(2) 1.686(4), Fe(1)-N(3) 1.683(4), Fe-
(1)-O(1) 1.960(3), Fe(1)-S(1) 2.286(2), N(2)-O(2) 1.144(6), N(3)-O(3)
1.185(5), N(2)-Fe(1)-N(3) 111.9(2), N(3)-Fe(1)-O(1) 115.5(1), N-
(2)-Fe(1)-O(1) 108.2(1), N(3)-Fe(1)-S(1) 110.6(1), N(2)-Fe(1)-S(1)
104.3(2), O(1)-Fe(1)-S(1) 105.6(1), O(2)-N(2)-Fe(1) 167.3(6), O-
(3)-N(3)-Fe(1) 164.3(4).
˚
Figure 5 shows the thermal ellipsoid plot of complex 3-
Na, and the selective bond distances and angles are given
in the figure captions. The structure of complex 3-Na can
be viewed in terms of the [(NO)2Fe(OPh)(C3H3N2)]-
anion and the [C4H8O-Na-18-crown-6-ether]þ cation
held together by Naþ [C3H3N2]- (30-N of the coordi-
3 3 3
DNICs via associating a ligand-substitution reaction study
with an XAS study of the electronic richness of the
{Fe(NO)2}9 core.
nated depronated imidazole of complex 3-Na) ionic inter-
action and the Naþ-O (THF) bond. The shorter
˚
Fe(1)-N(3) bond distance of 1.995(2) A, compared to
˚
Structures. The X-ray crystal structure of complex 1
consists of two crystallographically independent molecules,
and the structure of the [(NO)2Fe(ONO)2]- unit in the
[PPN]þ salt is shown in Figure 1. The local structure of iron
is a distorted tetrahedral geometry with a O(3)-Fe(1)-O-
(5) bond angle of 93.5(1)ꢀ. It is noticed that the O(3)-N(5)
the reported Fe-N(Im) bond length of 2.021 (4) A in
[(NO)2Fe(SC6H4-o-NHCOPh)(Im)] (Im = imidazole),7c
was attributed to the stronger π-donating ability of the
deprotonated imidazole [C3H3N2]- ligand. The Fe-
(1)-N(1) and Fe(1)-N(2) bond lengths of 1.692(3) and
˚
1.685(3) A, respectively, also fall in the range of 1.661-
9
˚
˚
and O(5)-N(6) bond lengths of 1.271(2) and 1.303(2) A in
(4)-1.695(3) A observed in the anionic {Fe(NO)2}
DNICs.7d Also, the nearly identical N(1)-O(1) and
the monodentate O-bound nitrito complex 1, compared
˚
˚
to those of 1.259(4) and 1.263(4) A in the six-coordinate
N(2)-O(2) bond lengths of 1.173(3) and 1.181(3) A,
chelating nitrito {Fe(NO)2}9 DNIC [(1-MeIm)2(η2-ONO)-
respectively, are within the range of 1.160(6)-1.178(3)
A observed for the anionic {Fe(NO)2} DNICs.
Fe(NO)2],15 are significantly longer than the distal N(5)-O-
9
7d
˚
The structure of the [(NO)2Fe(OPh)(-SC4H3S)]- unit
˚
(4) and N(6)-O(6) bond lengths of 1.236(2) and 1.214(2) A,
of complex 4 in [PPN]þ salt is shown in Figure 8. The
˚
respectively. The average Fe-NO bond length of 1.695(2) A
˚
˚
falls in the range of 1.661(4)-1.695(3) A observed in the
longer Fe-O(Ph) bond length (1.960(3) A) in complex 4,
as compared to those of 1.917(2) and 1.884(2) A in
anionic {Fe(NO)2}9 DNICs.7d
˚
Figure 3 displays an ORTEP plot of the anionic
{Fe(NO)2}9 complex 2, and the selected bond distances
and angles are listed in the figure caption. The two
[OPh]--coordinate ligands generate a ca. 105.6(1)ꢀ
O(3)-Fe(1)-O(4) angle. The average N-O bond length
complexes 2 and 3-Na, respectively, may be attributed
to the stronger electronic donation of the [-SC4H3S]-
ligand. These results illustrate one aspect of how the
distinct electron-donating coordinated ligands positioned
near each other in the iron site work in concert to achieve
the optimum electronic condition of the {Fe(NO)2} core
to stabilize DNICs. Figure S2 (Supporting Information)
˚
˚
of 1.180(4) A is close to the range of 1.160(6)-1.178(3) A
observed in the anionic {Fe(NO)2}9 DNICs.7d

Article
Inorganic Chemistry, Vol. 48, No. 19, 2009 9585
Table 1. Selected Bond Distances and IR Vibrational Frequencies of Complexes 1-3, [(NO)2Fe(SPh)2]-, [(NO)2Fe(C3H3N2)(SPh)]-, [(NO)2Fe(C3H3N2)2]-, and
[(NO)2Fe(SEt)2]- 7e,12
displays an ORTEP plot of the anionic complex 5, and the
selected bond distances and angles are listed in the figure
caption. Interestingly, the mean Fe-O bond length of
of [Fe(SPh)4]-. Compared to the tetrahedral Fe(III)
complex [Fe(SPh)4]-, complex [Fe(OPh)4]- coordinated
by the less covalent phenoxide ligands displaying a
higher-rising edge and pre-edge energy at 7114.2 eV
may suggest that the effective nuclear charge (Zeff) of
Fe in [Fe(OPh)4]- is higher than that of [Fe(SPh)4]-. We
noticed that the metal K-edge absorptions are strongly
perturbed by the metal oxidation state, coordination
geometry, and ligand environment.18 That is, the energies
for these features cannot be used unambiguously as a
marker of the oxidation state without an isoleptic refe-
rence complex containing an identical coordination geo-
metry. It is interesting to note that the pre-edge energy
was known to correlate with the metal oxidation state
under the presence of the same coordination geometries
and ligand types.18b Complexes 1, 2, and 5 coordinated by
O-containing ligands display the higher pre-edge energies
(7113.6, 7113.8, and 7113.7 eV; Figure 10b), compared to
those of complexes [(NO)2Fe(SEt)2]- and [(NO)2Fe-
(SPh)2]- (7113.5 and 7113.5 eV; Figure 10c). Acciden-
tally, we noticed that the pre-edge energies of complexes
[(NO)2Fe(SEt)2]-, [(NO)2Fe(μ-SEt)]2, and [(NO)2Fe-
(SPh)2]- (7113.5, 7113.8, and 7113.5 eV) lie between those
of complexes [Fe(SPh)4]2- and [Fe(SPh)4]-. Also, com-
plexes 1, 2, and 5 display the lower pre-edge energies
(7113.6, 7113.8, and 7113.7 eV), compared to that of
complex [Fe(OPh)4]-. A list of corresponding pre-edge
positions and EPR g tensors of iron-nitrosyl complexes
is displayed in Table 2, and selected bond distances/angles
and spectroscopic data for complexes 1, 2, 5, [(NO)2Fe-
(SPh)2]-, and [(NO)2Fe(SEt)2]- are given in Table 3. As
shown in Figure 11, we conclude that the monomeric
DNICs containing the varieties of ligation modes exhibit
the characteristic pre-edge energy falling in the range of
7113.4-7113.8 eV, in addition to the characteristic EPR
signal at g=2.03 (g=2.021-2.033) displayed. Presumably,
˚
1.916(2) A (Fe(1)-O(3) = 1.925(1) and Fe(1)-O(4) =
˚
1.906(2) A) in complex 5 is comparable to that of 1.917(3)
˚
˚
A (Fe(1)-O(3) 1.913(2) and Fe(1)-O(4)=1.920(2) A) in
complex 2.
Figure 9 shows the thermal ellipsoid plot of complex 6,
and the selective bond distances and angles are given in
the figures’ captions. The lengthening of the Fe-O(NO2)
˚
bond length (2.043(6) A) in complex 6, compared to the
˚
Fe-O(NO2) bond length (average 2.020(1) A) in complex
1, is presumably caused by electronic perturbation from
the stronger electron-donating phenylthiolate-coordi-
nated ligand of complex 6. Interestingly, the longer
˚
Fe-N(O) and N-O bond distances (average 1.695(2) A
and 1.173(2) A, respectively) of complex 1, compared to
˚
˚
the bond lengths Fe-N(O) (average 1.664(6) A) and
˚
N-O (average 1.167(6) A) for complex 6, also reflect
the electron deficiency induced by the less electron-donat-
ing nitrite-coordinated ligands.
In comparisons with all N-O bond distances of DNICs
containing [ONO]-, [OPh]-, [Im]-, [SEt]-, and [SPh]-
coordinated ligands, there are no significant differences
within 3σ standard deviations (Table 1). In contrast, the
Fe-N bond distances of DNICs shorten from [ONO]-,
[OPh]-, and [Im]- to [SPh]- (a significant difference of
-
˚
0.033(4) A between [(NO)2Fe(SPh)2] and [(NO)2Fe-
(OPh)2]-). This result may implicate that the highest
occupied molecular orbital is dominated by the contribu-
tions of the coordinated ligands. This assumption will be
further discussed in the MO calculation.
X-Ray Absorption Spectroscopy. The Fe K-edge spec-
tra of a series of reference compounds [PPN][FeIII(SPh)4],
[PPN]2[FeII(SPh)4], and [PPN][FeIII(OPh)4] are shown in
Figure 10a. The pre-edge transition is due to the d-p
mixing between Fe and ligand atoms in the distorted Td
local environment of the Fe center.7a Taking the apparent
peak position of the pre-edge for comparisons, the energy
of oxidized form [Fe(SPh)4]- shifted from 7113.8 to
7112.5 eV ([Fe(SPh)4]2-) upon one-electron reduction
(18) (a) DuBois, J. L.; Mukherjee, P.; Stack, T. D. P.; Hedman, B.;
Solomon, E. I.; Hodgson, K. O. J. Am. Chem. Soc. 2000, 122, 5775–5787.
(b) Westre, T. E.; Kennepohl, P.; Dewitt, J. G.; Hedman, B.; Hodgson, K. O.;
Solomon, E. I. J. Am. Chem. Soc. 1997, 119, 6297–6314.

9586 Inorganic Chemistry, Vol. 48, No. 19, 2009
Tsai et al.
Table 2. XAS Pre-edge Energies and Isotropic g-tensor of EPR for Iron Nitrosyl
Complexes 1-2, 5, [(NO)2Fe(SEt)2]- (6), [(NO)2Fe(SCH2C(dO)NHCH3)2]- (7),
[(NO)2Fe(SPh)2]- (8), [(NO)2FeS5]- (9), [(NO)2Fe(SePh)2]- (10), [(NO)Fe(SPh)3]-
(11), [(NO)2Fe(μ-SEt)]2 (12), [(NO)2Fe(μ-SEt)]2 (13), [(1-MeIm)2(η2-ΟΝΟ)Fe
-
(NO)2] (14), [(NO)Fe(S2CNEt2)2] (15), [(NO)Fe(S2C6H4)2]3 (16), [(NO)Fe(S2C6-
H4)2]3þ (17), [(NO)Fe(S,S-C6H4)2]- (18)
apparent
pre-edge
peak
1st derivative (dμ/dE = 0)
pre-edge
compound
EPR gav
1
2
5
6
7
8
9
10
11
12
13
14
15
16
17
18
7113.6
7113.8
7113.7
7113.5
7113.5
7113.5
7113.5
7113.5
7113.0
7113.8
7113.4
7113.8
7113.0
7113.4
7113.2
7113.0
7113.6
7113.7
7113.7
7113.6
7113.5
7113.6
7113.4
7113.6
7113.0
7113.8
7113.4
7113.7
7113.0
7113.4
7113.2
7113.0
2.033
2.025
2.026
2.028
2.028
2.028
2.03
2.021
3.76, 2.012
silent
1.997
2.013
2.04
1.999
silent
silent
Table 3. Selected Bond Distances/Angles and Spectroscopic Data for Complexes
1, 2, 5, [(NO)2Fe(SPh)2]-, and [(NO)2Fe(SEt)2]- 7e
[(NO)2Fe [(NO)2Fe
(SPh)2]-
complex
1
2
5
(SEt)2]-
Fe-N(O)
N-O
1.695(2) 1.690(3) 1.700(2) 1.662(4)
1.173(2) 1.180(4) 1.179(2) 1.177(5)
1.676(6)
1.186(7)
171.6(6)
122.3(3)
1715,
—Fe-N-O 162.7(2) 160.3(4) 161.7(2) 167.2(4)
—N-Fe-N 110.3(1) 110.2(2) 110.4(1) 115.1(3)
IR (cm-1
(THF)
)
1775,
1705
70
1739,
1674
65
1751,
1685
66
1737,
1693
44
1674
41
Δv (cm-1
)
Figure 10. (a) Fe K-edge spectra of complexes [Fe(OPh)4]-,
[Fe(SPh)4]-, and [Fe(SPh)4]2-. (b) Fe K-edge spectra of complexes 1, 2,
and 5, and the pre-edge absorption spectra are enlarged in the inset. (c) Fe
K-edge spectra of complexes [(NO)2Fe(SPh)2]-, [(NO)2Fe(SEt)2]-,
[(NO)2FeS5]-, and [(NO)2Fe(μ-SEt)]2, and the pre-edge absorption spec-
tra are enlarged in the inset.
Figure 11. Diagram of XAS pre-edge energy vs isotropic g tensor of
EPR for complexes 1, 2, 5, [(NO)2Fe(SEt)2]- (6), [(NO)2Fe(SCH2C-
(dO)NHCH3)2]- (7), [(NO)2Fe(SPh)2]- (8), [(NO)2FeS5]- (9), [(NO)2Fe-
(SePh)2]- (10), [(NO)Fe(SPh)3]- (11), [(NO)2Fe(μ-SEt)]2 (12), [(NO)2Fe-
in combination with the IR νNO spectrum, the
{Fe(NO)2}9 DNICs featuring the pre-edge energy within
the range of 7113.4-7113.8 eV may be adopted to probe
the formation of the EPR-silent dimeric DNICs in bio-
logical systems.
-
(μ-SEt)]2 (13), [(1-MeIm)2(η2-OΝO)Fe (NO)2] (14), [(NO)Fe(S2CN-
þ
Et2)2] (15), [(NO)Fe(S2C6H4)2]3 (16), [(NO)Fe(S2C6H4)2]3 (17), and
[(NO)Fe(S,S-C6H4)2]- (18).
the DFT calculations of [(NO)2Fe(SEt)2]-, [(NO)2Fe-
(SPh)2]-, and complex 2 were performed. The initial
coordinates of [(NO)2Fe(SEt)2]-, [(NO)2Fe(SPh)2]-, and
Computational Study of [(NO)2Fe(SEt)2]-, [(NO)2-
Fe(SPh)2]-, and Complex 2. In order to elucidate how
the {Fe(NO)2}9 core could be modulated to accommo-
date the biologically relevant coordination ligands and
rationalize the factors contributing to the binding affinity,
(19) Lu, T.-T.; Tsou, C.-C.; Huang, H.-W.; Hsu, I.-J.; Chen, J.-M.; Kuo,
T.-S.; Liaw, W.-F. Inorg. Chem. 2008, 47, 6040–6050.

Article
Inorganic Chemistry, Vol. 48, No. 19, 2009 9587
complex 2 for geometry optimization with DFT calcula-
tions were taken from the crystal structures published in
the literatures and this report.7e,19 As shown in Table 4,
the optimized structures of three complexes well repro-
duce the experimental data with slightly shorter Fe-N
B3LYP; 62%, 56%, and 66% of ligands SPh, SEt, and
OPh for BP86, respectively).
The unexpected trend of Fe-N and N-O bond dis-
tances may be interpreted via analysis of bonding inter-
actions. The representive energy diagram of Fe and NO
predominant molecular orbitals and the population of
selected antibonding molecular orbitals related to the
Fe-NNO bonding of [(NO)2Fe(SPh)2]- and complex 2
were shown in Figure 13 and Table 5, respectively, and the
corresponding antibonding orbitals of complex 2 are
shown in Figure 14. The four lowest-energy unoccupied
R spin orbitals and five unoccupied β spin orbitals of
[(NO)2Fe(SPh)2]- and complex 2 are predominated by
NO 2π* and Fe 3d characters, respectively. Bonding
characters between Fe and NO evaluated by the corresponding
Fe-NNO antibonding orbitals of both R and β spin
orbitals are well documented.20,21 Adopting the extreme
electronic structure description of {FeIII(NO-)2}9, the
unoccupied antibonding orbital analysis of the corre-
sponding Fe-NO bond might be able to discriminate
the different bonding interactions between Fe and NO.
˚
and longer N-O bond distances (deviation <0.02 A) and
Fe-N-O angles (deviation <3ꢀ). The calculated vibra-
tional frequencies νN-O reasonably reproduce the relative
trend of [(NO)2Fe(SEt)2]-, [(NO)2Fe(SPh)2]-, and com-
plex 2 measured in IR spectroscopy (Table 4). The SOMO
(singly occupied molecular orbital) of [(NO)2Fe(SPh)2]-
and complex 2 are displayed in Figure 12. The atomic
contributions of the frontier orbitals are listed in the
Supporting Information (Tables S1-S3). Both BP86
and B3LYP functionals indicate that the highest occupied
molecular oribtal is dominated by ligands’ contributions
(87%, 77%, and 86% of ligands SPh, SEt, and OPh for
Table 4. Experimental and Calculated Data of [(NO)2Fe(SEt)2]-,
[(NO)2Fe(SPh)2]-, and Complex 2
geometry
optimized
structure (A)
IR νNO
(cm-1
X-ray crystal
˚
structure (A)
˚
)
Table 5. Selected Mulliken Population of the Corresponding Antibonding
Orbitals of [(NO)2Fe(SPh)2]- and Complex 2
exptl
calcd Fe-N
N-O Fe-N N-O
[(NO)2Fe(SPh)2]-
complex 2
%3d %NO
[(NO)2Fe(SEt)2]- 1674, 1639, 1.676(6) 1.186(7) 1.660 1.198
label %3d %NO
label
1715 1689
[(NO)2Fe(SPh)2]- 1693, 1674, 1.662(4) 1.177(5) 1.660 1.191
1737 1726
1674, 1680, 1.690(3) 1.180(4) 1.675 1.194
1739 1750
R89 -πh*/dxy
R88 -πv*/dyz
10
10
15
12
59
62
56
54
71
89
90
81
76
32
28
25
43
6
R81 -πh*/dxy
R80 -πv*/dyz
R79 þπv*/dxz
R78 þπh*/dx2-y2
β81 dxy/-πh*
β80 dxz/þπv*
β79 dx2-y2/þπh*
β78 dyz/-πv*
β77 dz2/þπh*
9
8
90
92
83
86
33
21
24
40
10
a
complex 2
R87 þπh*/dx2-y2
R86 þπv/dxz
12
13
59
67
71
58
76
β89 dxy_dxz/-πh*
β88 dx2-y2/þπh*
β87 dxz/þπv*
β86 dyz/-πv*
β85 dz2/þπh*
a The combinations of (NO)2 2π* orbitals are denoted by the follow-
ing notations: “v” and “h” represent vertical and horizontal to the
Fe-N-O plane, respectively. The symbol “þ” represents two NO 2π*
orbitals possessing the symmetry plane which is perpendicular to the
N-O-O-N plane, and “-” represents two NO 2π* orbitals not
possessing the symmetry plane which is perpendicular to the
N-O-O-N plane. “-πh*/dxy” is used to denote an orbital that is a
mixture of the -πh* and dxy orbitals, and the former represents the larger
contribution to the orbital.
Figure 12. The SOMO of [(NO)2Fe(SPh)2]- and [(NO)2Fe(OPh)2]-.
Figure 13. Representive energy level diagram of Fe and NO predominant molecular orbitals for spin-unrestricted calculations of optimized
[(NO)2Fe(SEt)2]-, [(NO)2Fe(SPh)2]-, and complex 2.

9588 Inorganic Chemistry, Vol. 48, No. 19, 2009
Tsai et al.
Figure 14. The corresponding antibonding orbitals of complex 2. The coordinates of those complexes were defined as follows: the x axis was chosen as the
direction which bisects the N-Fe-N angle within the N-Fe-N plane, the y axis was chosen as the direction which roughly passes through two N atoms
within the N-Fe-N plane, and the z axis was chosen as the direction which is perpendicular to the N-Fe-N plane and orthogonal to the x and y axes.
Table 6. Selected Natural Charge of [(NO)2Fe(SEt)2]-, [(NO)2Fe(SPh)2]-, and
Complex 2
significantly more positive than that of [(NO)2-
Fe(SPh)2]- (0.98) and [(NO)2Fe(SEt)2]- (1.00), which is
consistent with the XAS results showing a more electro-
positive charge of Fe in complex 2 than that in
[(NO)2Fe(SPh)2]-.
natural charge [(NO)2Fe(SEt)2]- [(NO)2Fe(SPh)2]- complex 2
Fe
1.00
0.98
1.29
DFT calculation results lead to the proposition that the
electronic structure of the {Fe(NO)2}9 core of DNICs
might be described as a resonance hybrid of {FeI(NO)2}
(NBO charge) and {FeIII(NO-)2} in terms of the extents
of donor interactions (dominant) and back-bonding in-
teractions (minor) between Fe and NO ligands. The
coordinated phenoxide may polarize the {Fe(NO)2}9 core
of complex 2; that is, the population of the {FeIII(NO-)2}
resonance form in [(NO)2Fe(OPh)2]- is higher than that
of thiolate-coordinated [(NO)2Fe(SPh)2]-. It implicates
that the facile conversion of complex 2 to complex 4 may
be rationalized by the preferred formation of five-coordi-
nated intermediate [(NO)2Fe(OPh)2(-SC4H3S)]2- result-
ing from the nucleophilic binding of [-SC4H3S]- to the
relative electron-deficient Fe center (attributed to the
enhancement of {FeIII(NO-)2} character) of complex 2.
Presumably, the distinct electronic structures of DNICs
polarized by the nature of coordinated ligands may
elaborate the relative binding affinity of biologically
relevant ligands toward the {Fe(NO)2}9 core in the above
study.
N1
N2
O1
O2
-0.12
-0.12
-0.30
-0.30
-0.10
-0.10
-0.27
-0.27
-0.13
-0.13
-0.27
-0.26
That is, the Fe 3d population in the lowest four unoccu-
pied R spin orbitals represents the back-bonding interac-
tions between the occupied Fe 3d R spin orbitals and the
unoccupied NO 2π* R spin orbital. As shown in Table 5,
the strengths of back-bonding interactions (summation
of Fe 3d population in R corresponding to Fe-NNO
antibonding orbitals) for [(NO)2Fe(SPh)2]- are larger
than those of complex 2 (47% ([(NO)2Fe(SPh)2]-) vs
42% (complex 2)). The NO population of the correspond-
ing Fe-NNO antibonding orbitals in the β spin manifold
represents the donor interactions between the occupied
NO 2π* β spin orbitals and the unoccupied Fe 3d β spin
orbital. Since no bonding interactions between dz2 and
NO 2π* orbitals are found in the bonding orbitals, the
strength of donor interactions is estimated by the summa-
tion of the NO population in β corresponding to Fe-NNO
antibonding orbitals associated with dxy, dyz, dxz, and
dx2-y2 orbitals. Similar to the back-bonding interactions
presented in the R spin orbitals, [(NO)2Fe(SPh)2]- also
demonstrates the stronger donor interaction in compari-
son with complex 2 (128% ([(NO)2Fe(SPh)2]-) vs 118%
(complex 2)). Overall, the analysis of corresponding
Fe-NNO antibonding orbitals demonstrates that
[(NO)2Fe(SPh)2]- exhibits the stronger back-bonding
and donor interactions compared to those of complex
2. The NBO charge analysis displayed in Table 6 indicates
that the natural charge of Fe (1.29) of complex 2 is
Conclusion and Comments
Studies on the dinitrosyl iron complexes 1-6 containing
nitrite-, imidazolate-, phenoxide-, and thiolate-coordinated
ligand(s) and the transformations among complexes 1-6 and
[(NO)2Fe(C3H3N2)2]- have led to the following results,
including certain results from earlier studies.7,12,13
(1) The anionic {Fe(NO)2}9 DNICs containing the various
ligations [NO2,NO2]/ [NO2,SPh]/[OPh,OPh]/[OPh,SC4H3S]-
/[OPh,C3H3N2] were synthesized. The differences in
{Fe(NO)2} affinity for thiolates, imidazolate, phenoxide,
and nitrite, probed by ligand-substitution reactions, explain
the inclusion of the stronger electron-donating thiolate-
coordinated ligand producing the more thermally stable
{Fe(NO)2}9 DNICs (i.e., [(NO)2Fe(SPh)2]- > [(NO)2Fe-
(SPh)(C3H3N2)]- > [(NO)2Fe(C3H3N2)2]- > [(NO)2Fe-
(OPh)(C3H3N2)]- > [(NO)2Fe(OPh)2]- > [(NO)2Fe-
(ONO)2]-). The preferred formation of thiolate-containing
DNICs via a ligand-exchange reaction may be dictated in
::
(20) (a) Praneeth, V. K. K.; Nather, C.; Peters, G.; Lehnert, N. Inorg.
Chem. 2006, 45, 2795–2811. (b) Fujisawa, K.; Tateda, A.; Miyashita, Y.;
Okamoto, K.-I.; Paulat, F.; Praneeth, V. K. K.; Merkle, A.; Lehnert, N. J. Am.
Chem. Soc. 2008, 130, 1205–1213.
(21) (a) Brown, C. A.; Pavlosky, M. A.; Westre, T. E.; Zhang, Y.;
Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1995,
117, 715–732. (b) Schenk, G.; Pau, M. Y. M.; Solomon, E. I. J. Am. Chem. Soc.
2004, 126, 505–515. (c) Brown, C. D.; Neidig, M. L.; Neibergall, M. B.;
Lipscomb, J. D.; Solomon, E. I. J. Am. Chem. Soc. 2007, 129, 7427–7438.

Article
Inorganic Chemistry, Vol. 48, No. 19, 2009 9589
part by thermodynamic preferences for the softer donor and
the stronger electron-donating ligands.
(Aldrich), p-flourophenol (Acros), sodium nitrite
(Riedel-deHaen), phenol, bis(triphenylphosphoranylidene)
ammonium chloride, and 18-crown-6-ether (TCI) were used
as received. Complexes [(NO)2Fe(SEt)2]-, [(NO)2Fe-
(SCH2C(dO)NHCH3)2]-, [(NO)2Fe(SPh)2]-, [(NO)2-
FeS5]-, [(NO)2Fe(SePh)2]-, [(NO)Fe(SPh)3]-, [(NO)2Fe-
(μ-SEt)]2, [(NO)2Fe(μ-SEt)]2-, [(1-MeIm)2(η2-OΝO)-
Fe(NO)2], [(NO)Fe(S2CNEt2)2], [(NO)Fe(S2C6H4)2]3,
[(NO)Fe(S2C6H4)2]3þ, and [(NO)Fe(S,S-C6H4)2]- were
synthesized on the basis of the literature reported.7,15,17
Infrared spectra of the νNO stretching frequencies were re-
corded on a PerkinElmer model spectrum One B spectro-
meter with sealed solution cells (0.1 mm, CaF2 windows).
UV-vis spectra were recorded on a Jasco V-570 spectro-
meter. Analyses of carbon, hydrogen, and nitrogen were
obtained with a CHN analyzer (Heraeus).
(2) The {Fe(NO)2} motif coordinated by different ligands
([SPh]-, [-SC4H3S]-, [C3H3N2]-, [OPh]-, and [NO2]-) dis-
plays distinct electronic properties (structure) which are
cooperatively regulated by the noninnocent NO ligands.
That is, presumably, the iron center of DNICs is tailored to
minimize the electronic changes accompanying changes in
coordinated ligands. This point has received further support
in studies of XAS of DNICs containing different ligations,
where it was found that the pre-edge energy derived from the
1s f 3d transition in a distorted Td environment of the Fe
center of DNICs is within the range of 7113.4-7113.8 eV for
the anionic {Fe(NO)2}9 DNICs ([(NO)2FeS5]-=7113.4 eV,
[(NO)2Fe(SEt)2]-=7113.5 eV, [(NO)2Fe(SPh)2]-=7113.5 eV,
Preparation of [PPN][(NO)2Fe(ONO)2] (1). Compounds
[PPN][(CO)3Fe(NO)] (0.7070 g, 1 mmol)22 and [NO][BF4]
(0.2240 g, 2 mmol) were dissolved in THF (10 mL), and then
the mixture solution was stirred for 5 min in an ice bath. The
resulting suspension was transferred into the flask containing
[PPN][NO2] (1.1690 g, 2 mmol) by cannula under positive N2
pressure. After being stirred for 1 h at ambient temperature, the
mixture solution was filtered through Celite to remove the
insoluble solid. The addition of hexane (20 mL) to the filtrate
led to precipitation of a dark solid [PPN][(NO)2Fe(ONO)2] (1)
(28%). Diffusion of diethyl ether-hexane into the THF solution
of complex 1 led to dark crystals suitable for X-ray diffraction.
IR: 1705 s, 1775 s (νNO) cm-1 (THF); 1711, 1782 (νNO), 1420,
1073 (νΟ-N-O) cm-1 (KBr). Absorption spectrum (THF) [λmax, nm
(ε, M-1 cm-1)]: 510 (306), 615 (140). Anal. Calcd for
C36H30N5O6P2Fe: C, 57.87; H, 4.02; N, 9.38. Found: C, 58.50; H,
4.32; N, 9.15.
[(NO)2Fe(ONO)2]- = 7113.6 eV, [(NO)2Fe(p-OPhF)2]-
=
7113.7 eV, [(NO)2Fe(OPh)2]-=7113.8 eV).7a
(3) The stabilization of these {Fe(NO)2}9 DNICs
[(NO)2Fe(L)(L0)]- (L = NO2, OPh; L0 = NO2, OPh, SPh,
-SC4H3S, C3H3N2) by the particular combinations of thio-
lates, imidazolate, phenoxide, and nitrite ligation can be
attributed to the optimum electronic structures of the
{Fe(NO)2} core. The resonance hybrids of {FeI(NO)-
(NO)}9, {FeII(NO)(NO-)}9, and {FeIII(NO-)2}9 electronic
structures may provide an additional mechanism for the ad-
justment of the charge at the iron site, facilitated by the co-
ordinated ligands. Specifically, an enhancement of the
{FeIII(NO-)2} electronic structure for the O-bound DNICs
compared to S-bound DNICs, modulated by the intraelec-
tron transfer among the redox-active Fe center, noninnocent
ligand NO, and the coordinated ligands to accommodate the
[S,S]/[S,N]/[S,O]/[N,N]/[N,O]/[O,O] ligation modes, was
proposed to preserve the {Fe(NO)2}9 electronic core of
DNICs.
Preparation of [PPN][(NO)2Fe(OPh)2] (2). Complex
1
(0.3735 g, 0.5 mmol) and [Na][OPh] (0.1160 g, 1.0 mmol) were
dissolved in THF and stirred for 5 min under a N2 atmosphere at
ambient temperature. The reaction was monitored with FTIR.
The IR spectrum (1739 m, 1674 s cm-1 (νNO) (THF)) was
assigned to the formation of [PPN][(NO)2Fe(OPh)2] (2). The
resulting mixture was filtered through Celite to remove the
insoluble solid. The filtrate was concentrated under a vacuum,
and diethyl ether-hexane (10:15 mL) was added to precipitate
red-brown solid 2 (yield 0.3574 g, 85%). Diffusion of diethyl
ether-hexane into the THF solution of complex 2 at -15 ꢀC for
1 week led to red-brown crystals suitable for single-crystal X-ray
diffraction. IR νNO: 1739 m, 1674 s cm-1 (THF). Absorption
spectrum (THF) [nm, λmax(M-1 cm-1, ε)]: 427 (18512), 502
(11185), 657 (3476). Anal. Calcd for C48H40FeN3O4P2: C, 68.52;
H, 4.76; N, 4.99. Found: C, 68.97; H, 4.93; N, 5.25.
The binding affinity of ligands ([SPh]-, [-SC4H3S]-,
[C3H3N2]-, [OPh]- and [NO2]-) toward the {Fe(NO)2}9 motif
is in the order of [SPh]- ∼ [-SC4H3S]- > [C3H3N2]-
>
[OPh]- > [NO2]-. The ligand-exchange experiments used in
the synthesis of DNICs containing a variety of [S,S]/[S,N]/-
[S,O]/[N,N]/[N,O]/[O,O] coordination modes have demon-
strated that the {Fe(NO)2}9 motif shows strong preferences
for thiolates over imidazolate/phenoxide/nitrite.12 Our results
bridging the ligand-substitution reaction study and XAS
study of the electronic richness of the {Fe(NO)2}9 core may
point the way to understanding the reasons for nature’s
choice of combinations of cysteine, histidine, and tyrosine
in protein-bound DNICs1-4,6 and rationalize that most of
the DNICs characterized and proposed nowadays are bound
to the proteins almost through the thiolate groups of cystei-
nate/glutathione side chains in biological systems.5b
Preparation of [cation][(NO)2Fe(OPh)(C3H3N2)] (C3H3N2 =
imidazolate; cation=PPNþ (3-PPN), Na-18-crown-6-ether (3-Na)).
To a stirred solution of complex 2 (0.4205 g, 0.5 mmol) in THF
was added a THF solution of sodium imidazolate (0.0450 g,
0.5 mmol) dropwise by syringe at 0 ꢀC. The IR stretching
frequencies (1755 m, 1691 s cm-1 (νNO) (THF)) implicated the
formation of [PPN][(NO)2Fe(OPh)(C3H3N2)] (3-PPN). The
solution was concentrated under a vacuum, and hexane was
then added to precipitate the red-brown solid 3-PPN. Diffusion
of hexane into the THF solution of complex 3-Na (obtained
from the reaction of 3-PPN, NaBPh4, and 18-crown-6-ether)
under nitrogen at -15 ꢀC led to red-brown crystals suitable for
X-ray diffraction. Complex 3-PPN. IR νNO: 1755 m, 1691 s
cm-1 (THF). Absorptionspectrum (THF) [nm, λmax(M-1 cm-1, ε)]:
216 (18588), 270 (9097). Anal. Calcd for C21H32FeN4NaO9: C,
44.77; H, 5.68; N, 9.95. Found: C, 44.41; H, 5.87; N, 9.24.
Experimental Section
Manipulations, reactions, and transfers were conducted
under nitrogen according to Schlenk techniques or in a glove-
box (N2 gas). Solvents were distilled under nitrogen from
appropriate drying agents (diethyl ether from CaH2, acetoni-
trile from CaH2-P2O5, methylene chloride from CaH2,
methanol from Mg/I2, and hexane and tetrahydrofuran
(THF) from sodium benzophenone) and stored in dried,
˚
N2-filled flasks over 4 A molecular sieves. Nitrogen
was purged through these solvents (including dimethyl for-
mamide (DMF)) before use. Solvent was transferred to the
reaction vessel via stainless cannulaunder positivepressureof
N2. The reagents sodium hydroxide, ferrous dichloride
(22) McBride, D. W.; Stafford, S. L.; Stone, F. G. A. Inorg. Chem. 1962, 1,
386–388.

9590 Inorganic Chemistry, Vol. 48, No. 19, 2009
Tsai et al.
Reaction of Complex 3-PPN and [Na][C3H3N2]. Complex
3-PPN (0.4073 g, 0.5 mmol) and [Na][C3H3N2] (0.0450 g,
0.5 mmol) were loaded in a 50 mL Schlenk flask and dissolved
in THF. The mixture solution was stirred for 5 min at 0 ꢀC. The
reaction was monitored by FTIR. The IR spectrum (1776 m,
1712 s cm-1 (νNO) (THF)) suggests the formation of the known
[PPN][(NO)2Fe(C3H3N2)2].12 The mixture solution was then
filtered through Celite to remove the insoluble solid. Hexane
was added to precipitate the known red-brown solid [PPN]-
[(NO)2Fe(C3H3N2)2]. IR νNO: 1776 m, 1712 s cm-1 (THF).
Absorption spectrum (THF) [nm, λmax(M-1 cm-1, ε)]: 326,
716.12
C42H35N4O4P2SFe: C, 62.25; H, 4.32; N, 6.92. Found: C,
62.49; H, 4.39; N, 6.49.
X-Ray Absorption Measurements. All X-ray absorption
experiments were carried out at the National Synchrotron
Radiation Research Center (NSRRC), Hsinchu, Taiwan. Fe
K-edge was recorded at room temperature. For Fe K-edge
measurements, the experiments were performed in transmission
mode at the BL-17C X-ray Wiggler beamline with a double-
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 8.006 KeV using a gas-ionization detector. A reference
Fe foil is always used simultaneously for the calibration of
photon energy. The first inflection point at 7112.0 eV of the
Fe foil spectrum is used for energy calibration. The ion cham-
bers used to measure the incident (I0) and transmitted (I) photon
intensities were filled with a mixture of N2 and He gases and a
mixture of N2 and Ar gases, respectively.
Preparation of [PPN][(NO)2Fe(OPh)(-SC4H3S)] (4). A THF
solution of [Na][-SC4H3S] (0.0695 g, 0.5 mmol) was added
dropwise into a stirred solution of complex 2 (0.4210 g, 0.5 mmol)
in THF at 0 ꢀC. The mixture solution was stirred for 5 min at
0 ꢀC. The IR νNO stretching frequency (1740 s, 1681 s cm-1
(THF)) was assigned to the formation of [PPN][(NO)2-
Fe(OPh)(-SC4H3S)] (4). The resulting mixture was filtered
through Celite to remove the insoluble solid, and then hexane
was added into the filtrate to precipitate the red-brown solid
[PPN][(NO)2Fe(OPh)(-SC4H3S)] (4) (yield 0.3019 g, 70%).
Diffusion of hexane into the THF solution of complex 4 under
nitrogen at -15 ꢀC led to red-brown crystals suitable for single-
crystal X-ray diffraction. IR νNO: 1740 s, 1681 s cm-1 (THF).
Absorption spectrum (THF) [nm, λmax(M-1 cm-1, ε)]: 486
(5955), 572 (2146), 806 (233)). Anal. Calcd for C46H38FeN3-
O3P2S2: C, 63.99; H, 4.41; N, 4.87. Found: C, 63.67; H, 4.83; N,
5.15.
Crystallography. Crystallographic data of complexes 1, 2, 3,
4, 5, and 6 were summarized in Supporting Information Tables
S4-S12. The crystals chosen for X-ray diffraction studies
measured 0.30 ꢀ 0.25 ꢀ 0.25 mm for complex 1, 0.30 ꢀ 0.25 ꢀ
0.05 mm for complex 2, 0.30 ꢀ 0.10 ꢀ 0.10 mm for complex 3,
0.25 ꢀ 0.20 ꢀ 0.10 mm for complex 4, 0.30 ꢀ 0.15 ꢀ 0.05 mm for
complex 5, and 0.40 ꢀ 0.30 ꢀ 0.20 mm for complex 6, respec-
tively. Each crystal was mounted on a glass fiber and quickly
coated in epoxy resin. Unit-cell parameters were obtained by
least-squares refinement. Diffraction measurements for com-
plexes 1, 2, 3, 4, 5, and 6 were carried out on a SMART CCD
(Nonius Kappa CCD) diffractometer with graphite-monochro-
Reaction of Complex 2 and [Na][SEt]. A THF solution (5 mL)
of [Na][SEt] (0.0420 g, 0.5 mmol) was injected drop by drop into
a THF solution of complex 2 (0.4205 g, 0.5 mmol) under
nitrogen at 0 ꢀC. The IR νNO stretching frequencies shifting
from 1739 m and 1674 s cm-1 (complex 2) to 1808 w, 1774 s, and
1749 s cm-1 (THF) implicated the formation of the known
[(NO)2Fe(μ-SEt)]2.17 The mixture solution was filtered through
Celite, and the filtrate was concentrated under a vacuum.
Hexane was then added to the filtrate to precipitate the known
red-yellow [(NO)2Fe(μ-SEt)]2 (yield 0.0885 g, 50%).
˚
mated Mo KR radiation (λ=0.71073 A) and between 1.06 and
28.34ꢀ for complex 1, between 1.36 and 28.31ꢀ for complex 2,
between 1.54 and 28.38ꢀ for complex 3, between 1.36 and 28.35ꢀ
for complex 4, between 1.68 and 28.28ꢀ for complex 5, and
between 1.46 and 25.03ꢀ for complex 6. Least-squares refine-
ment of the positional and anisotropic thermal parameters of all
non-hydrogen atoms and fixed hydrogen atoms was based on F2.
A SADABS absorption correction was made.23 The SHELXTL
structure refinement program was employed.24
EPR Measurements. EPR measurements were performed at
the X-band using a Bruker EMX spectrometer equipped with a
Bruker TE102 cavity. The microwave frequency was measured
with a Hewlett-Packard 5246 L electronic counter. X-band EPR
spectra of complex 1 at 77 K in THF were obtained with a
microwave power of 19.678 mW (1.978 mW for complex 2,0.171mW
for complex 3, 20.117 mW for complex 4, 3.809 mW for complex
5, and 19.824 mW for complex 6), a frequency of 9.631 GHz for
complex 1 (9.625 GHz for complex 2, 9.432 GHz for complex
3, 9.456 GHz for complex 4, 9.471 GHz for complex 5, and
9.605 GHz for complex 6), and a modulation amplitude of 0.80 G
at 100 kHz.
Reaction of Complex 2 and [Na][p-OPhF]. Complex 2 (0.4205
g, 0.5 mmol) and sodium p-flourophenoxide ([Na][p-OPhF])
(0.1340 g, 1.0 mmol) were dissolved in THF (10 mL) and stirred
for 5 min under nitrogen at ambient temperature. The reaction
was monitored with FTIR. The IR νNO stretching frequencies
shifting from 1739 m and 1674 s cm-1 to 1751 m and 1685 s cm-1
implicated the formation of [PPN][(NO)2Fe(p-OPhF)2] (5). The
mixture solution was filtered through Celite, and the filtrate was
concentrated under a vacuum. Diethyl ether-hexane (10:15 mL)
was then added to precipitate the red-brown solid [PPN][(NO)2-
Fe(p-OPhF)2] (5) (yield 0.2982 g, 68%). Diffusion of diethyl
ether-hexane into the THF solution of complex 5 under nitro-
gen at -15 ꢀC led to red-brown crystals suitable for X-ray
diffraction. IR νNO: 1751 m, 1685 s cm-1 (THF). Absorption
spectrum (THF) [nm, λmax(M-1 cm-1, ε)]: 439 (14347), 509
(2550). Anal. Calcd for C48H38FeF2N3O4P2: C, 65.71; H, 4.33;
N, 4.79. Found: C, 65.97; H, 4.12; N, 5.15.
Magnetic Measurements. The magnetic data were recorded
on a SQUID magnetometer (MPMS5 Quantum Design
Company) under a 0.5 T external magnetic field in the temperat-
ure range of 2-300 K for complex 2. The magnetic susceptibility
data were corrected with temperature-independent para-
magnetism (TIP; 2 ꢀ 10-4 cm3 mol-1) and ligands’ diamagnet-
ism with the tabulated Pascal’s constants.25
Preparation of [PPN][(NO)2Fe(SPh)(ONO)] (6). The addit-
ion of a THF solution (5 mL) of [Na][SPh] (0.0132 g, 0.1 mmol)
into a THF solution (5 mL) of complex 1 (0.0747 g, 0.1 mmol)
drop by drop under N2 atmosphere at 0 ꢀC. The resulting
mixture was stirred for 15 min and then filtered through Celite
to remove the insoluble solid, presumably, [Na][NO2]. The
addition of hexane (15 mL) to the filtrate led to precipitation
of dark red solid [PPN][(NO)2Fe(SPh)(ONO)] (6). Diffusion of
hexane into the THF solution of complex 6 under nitrogen at
-15 ꢀC led to red crystals suitable for X-ray diffraction. IR
Computation Method. All calculations were carried out with
the ORCA electronic structure package, version 2.6.63.26 The
coordinates used for geometry optimization were based on
the experimental structures taken from the X-ray diffraction
::
::
(23) Sheldrick, G. M. SADABS; University of Gottingen: Gottingen,
Germany, 1996.
(24) Sheldrick, G. M. SHELXTL; Siemens Analytical X-ray Instruments
Inc.: Madison, WI, 1994.
(25) Kahn, O. Molecular Magnetism; VCH: New York, 1993.
(26) Neese, F. ORCA, version 2.6.35; University of Bonn: Bonn, Germany,
2008.
ν
NO: 1752 m, 1697 s cm-1 (THF); 1739, 1686 (νNO), 1418, 1070
(νΟ-N-O) cm-1 (KBr). Absorption spectrum (THF) [λmax, nm
(ε, M-1 cm-1)]: 462 (1900), 759 (365). Anal. Calcd for

Article
Inorganic Chemistry, Vol. 48, No. 19, 2009 9591
experiments.7e The geometry optimizations were done in re-
dundant internal coordinates. The BP86 functional together
with the aug-cc-pVTZP basis set on the Fe, S, N, and O atoms
and SV(P) basis set on the C and H atoms were used in the
geometry optimization,27 and the Coulomb term was approxi-
mated with the resolution-of-the-identity (RI) approximation to
gain computation efficiency.28
screening model (COSMO) as implemented in ORCA.29 Isosur-
face plots of the MOs were generated using the Molekel program
with an isovalue surface at 0.04 au.30 The natural population
analysis of the B3LYP-level wave function obtained from
ORCA calculations was done with the GENNBO package by
Weinhold and co-workers.31
The SCF calculations were tightly converged (10-7 Eh in
energy, 10-6 in the density change, and 10-6 in maximum
element of the DIIS error vector). The geometry search for all
complexes was carried out in redundant internal coordinates
without imposing symmetry constraints. In all cases, the geo-
metries were considered converged after the energy change was
less than 1 ꢀ 10-6 Eh; the gradient norm and maximum gradient
element were smaller than 1 ꢀ 10-4 Eh/Bohr and 3 ꢀ 10-5 Eh/
Bohr, and the root-mean square and maximum displacements of
all atoms were smaller than 6 ꢀ 10-4 Bohr and 1 ꢀ 10-3 Bohr,
respectively. All single-point energy and vibrational frequencies
were calculated at the B3LYP level with the same basis sets used
in the geometry optimization, and the solvent effect of tetra-
hydrofuran (ε = 7.25) was treated with the conductor-like
Acknowledgment. We gratefully acknowledge financial sup-
port from the National Science Council of Taiwan and facilities
from the National Center for High-Performance Computing.
The authors thank Miss Pei-Lin Chen and Mr. Ting-Shen Kuo
for the single-crystal X-ray structural determinations and
Dr. Tsu-Chien Weng for valuable discussions.
Supporting Information Available: X-ray crystallographic
files in CIF format for the structure determinations of
[PPN][(NO)2Fe(ONO)2], [PPN][(NO)2Fe(OPh)2], [Na-18-crown-
6-ether][(NO)2Fe(OPh)(C3H3N2)], [PPN][(NO)2Fe(OPh)(-SC4H4-
S), [PPN][(NO)2Fe(p-OPhF)2], and [PPN][(NO)2Fe(SPh)(ONO)].
Coordinates for optimized geometries. This material is available free
of charge via the Internet at http://pubs.acs.org.
::
(27) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100. (b) Perdew, J. P.
(29) (a) Andzelm, J.; Kolmel, C.; Klamt, A. J. Chem. Phys. 1995, 103,
::
Phys. Rev. B 1986, 33, 8822–8824. (c) Schafer, A.; Horn, H.; Ahlrichs, R.
9312–9320. (b) Sinnecker, S.; Rajendran, A.; Klamt, A.; Diedenhofen, M.; Neese,
F. J. Phys. Chem. A 2006, 110, 2235–2245.
(30) Molekel, Advanced Interactive 3D-Graphics for Molecular Sciences.
€
J. Chem. Phys. 1992, 97, 2571–2577. (d) Eichkorn, K.; Treutler, O.; Ohm, H.;
::
Haser, M.; Ahlrichs, R. Chem. Phys. Lett. 1995, 240, 283–289. (e) Dunning,
http:www.cscs.ch/molkel/ (accessed September 2009).
T. H. Jr. J. Chem. Phys. 1989 90, 1007–1023.
(31) Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J. E.;
Bohmann, J. A.; Morales, C. M.; Weinhold, F. NBO 5.0; Theoretical
Chemistry Institute, University of Wisconsin: Madison, WI, 2001. http://www.
chem.wisc.edu/∼nbo5 (accessed September 2009).
(28) (a) Baerends, E. J.; Ellis, D. E.; Ros, P. Chem. Phys. 1973, 2, 41–51.
(b) Dunlap, B. I.; Connolly, J. W. D.; Sabin, J. R. J. Chem. Phys. 1979, 71, 3396–
::
3402. (c) Vahtras, O.; Almlof, J. E.; Feyereisen, M. W. Chem. Phys. Lett. 1993,
213, 514–518.