Systematic investigation of relationship between strength of NH?S hydrogen bond and reactivity of molybdoenzyme models

Article abstract of DOI:10.1021/ic302149p

A series of monooxomolybdenum(IV) and dioxomolybdenum(VI) complexes containing two intramolecular NH?S hydrogen bonds were synthesized and characterized by ¹H NMR, UV-visible, IR, and Raman spectroscopies, and electrochemical measurements. Elimination of the steric effects enabled the accurate and quantitative evaluation of NH?S hydrogen bonds. Strong correlations among the strength of the hydrogen bond, the strength of the Mo^VI-O bond, and the redox potential were clearly shown. The hydrogen bonds stabilize the intermediate in the reaction between the monooxomolybdenum(IV) and Me₃NO, resulting in acceleration of the reaction or retardation of trans-cis rearrangement. The proposed intermediate and the reaction mechanism, discussed on the basis of theoretical calculations, provided a unified explanation of the reaction.

Full text of DOI:10.1021/ic302149p

Article
pubs.acs.org/IC
Systematic Investigation of Relationship between Strength of NH···S
Hydrogen Bond and Reactivity of Molybdoenzyme Models
Taka-aki Okamura,* Yasuhito Ushijima, Yui Omi, and Kiyotaka Onitsuka
Department of Macromolecular Science, Graduate School of Science, Osaka University, Toyonaka, Osaka 560-0043, Japan
S
* Supporting Information
ABSTRACT: A series of monooxomolybdenum(IV) and
dioxomolybdenum(VI) complexes containing two intramolec-
ular NH···S hydrogen bonds were synthesized and charac-
1
terized by H NMR, UV−visible, IR, and Raman spectros-
copies, and electrochemical measurements. Elimination of the
steric effects enabled the accurate and quantitative evaluation of
NH···S hydrogen bonds. Strong correlations among the
strength of the hydrogen bond, the strength of the MoVIO
bond, and the redox potential were clearly shown. The
hydrogen bonds stabilize the intermediate in the reaction between the monooxomolybdenum(IV) and Me3NO, resulting in
acceleration of the reaction or retardation of trans−cis rearrangement. The proposed intermediate and the reaction mechanism,
discussed on the basis of theoretical calculations, provided a unified explanation of the reaction.
on the substituent group R in (NEt4)2[MoIVO{S2-3,6-
(RCONH)2C6H2}2], occurring in the order Ph3C ≫ CH3 >
CF3 > (bdt) > t-Bu. This suggested a multiplicity of factors
INTRODUCTION
■
The dimethyl sulfoxide reductase (DMSOR) family of
molybdenum enzymes, which contains two molybdopterins,
determining the rate constant. Generally, the reduction of
catalyzes oxygen atom transfer reactions between water and
substrates using MoIV and MoVI oxidation states, with some
Me3NO by a monooxomolybdenum(IV) complex proceeds via
a two-step reaction.23 First, Me3NO approaches the molybde-
num center from the position trans to the oxo ligand, forming a
Me3NO adduct as an intermediate. Next, trans−cis rearrange-
ment occurs to simultaneously afford a cis-dioxo complex and
Me3N. When the concentration of Me3NO is sufficiently high
or the substituent groups are bulky enough (i.e., R = t-Bu) to
sterically hinder rotation, the trans−cis rearrangement is the
rate-determining step, and saturation of the reaction rate is
observed. In the case R = Ph3C, the situation was very different.
The adequately bulky substituent groups prevented rotation
and distorted the monooxomolybdenum(IV) complex, which
allowed the direct cis attack of Me3NO.23,24 Unfortunately, the
resulting dioxomolybdenum(VI) complexes proved to be too
unstable for isolation, and therefore, the contribution of the
NH···S in the oxidized state could not be described.
Recently, we successfully isolated a new dioxomolybdenum-
(VI) complex, which contained two intramolecular NH···S
hydrogen bonds, as a single isomer. Through this complex, we
revealed the selective and effective stabilization of MoVIO by
hydrogen bonds.25 In that report, a carbamoyl group
(RNHCO) was used to introduce an amide moiety because
of synthetic facility; however, that led to the formation of a
flexible NH···S hydrogen bond and resulted in a small
contribution to the reduction of Me3NO in polar solvent.
exceptions.1,2 DMSOR and trimethylamine-N-oxide reductase
(TMAOR) reduce Me3NO to Me3N, which results in the
formation of a MoVIO bond.1,3 Many model compounds
have demonstrated oxo-transfer reactions from Me3NO to
desoxo- or monooxomolybdenum(IV) giving monooxo- or
dioxomolybdenum(VI) species, respectively.4−9 Previously, we
reported a dioxomolybdenum(VI) benzene-1,2-dithiolate (bdt)
complex that oxidized benzoin to benzil with the accompanying
production of water, and proceeded in a catalytic cycle similar
to a biomimetic enzymatic reaction.10−12
NH···S hydrogen bonds are widely distributed in the active
sites of enzymes and metalloproteins, as described previ-
ously.13,14 Our systematic investigations revealed the function
of the hydrogen bonds using various model compounds of
rubredoxins,13,15 ferredoxins,16 cytochrome P450s,17,18 molyb-
doenzymes,19,20 and copper proteins.21 The hydrogen bonds
shift the redox potential to more positive values and stabilize
M−S bonds.16,18,22 We found that electron-withdrawing groups
enhanced the hydrogen bond, resulting in a greater positive
shift of the redox potential, where the change was
approximately proportional to the number of hydrogen
bonds.13,16,18
The introduction of four intramolecular NH···S hydrogen
bonds into a monooxomolybdenum(IV) complex,
(NEt4)2[MoIVO(bdt)2], accelerated the reduction of Me3NO
to Me3N with the accompanying production of the
corresponding dioxomolybdenum(VI) complex; however, the
results were complicated. The rate constants were dependent
Received: October 2, 2012
Published: December 21, 2012
© 2012 American Chemical Society
381
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
× 2), and sat. NaCl aq (30 mL × 3). The solution was dried over
Na2SO4 and concentrated to dryness under the reduced pressure. The
crude product was purified by column chromatography on silica gel
with n-hexane/ethyl acetate (49/1) as eluent. The first fraction was
collected and concentrated to dryness under reduced pressure to give
orange oil. The oil was crystallized from methanol to give colorless
crystals. Yield: 2.90 g (41%). 1H NMR (CDCl3): δ 8.60 (br, 1H, NH),
8.06 (d, J = 8.17 Hz, 1H, Ar−H), 7.30 (t J = 8.23 Hz, 1H, Ar−H), 7.00
(dd, J = 8.05, 1.04 Hz, 1H, Ar−H), 3.49 (sep, J = 6.65 Hz, 1H, CH),
3.36 (sep, J = 6.71 Hz, 1H, CH), 1.38 (d, J = 6.65, 6H, CH3), 1.25 (d, J
= 6.71, 6H, CH3). ESI-MS ([M] = N3CONHC6H3(SC3H7)2): m/z
290.1 ([M-HN3+Na+]+, calcd. 290.1), 268.1 ([M-HN3+H+]+, calcd.
268.1). FT-IR (KBr): ν(NH) 3315, νas(NNN) 2142, ν(CO)
1722 cm−1. Anal. Calcd for C13H18N4OS2: C, 50.30; H, 5.84; N, 18.05.
Found: C, 50.34; H, 5.73; N, 18.04.
In this paper, a series of new ligands with acylamino groups
(RCONH; R = CH3, t-Bu, CF3, CPh3, C(C6H4-4-t-Bu)3) were
designed and synthesized (Figure 1b). The location of the
Figure 1. Schematic drawing of (a) molybdopterin and (b) the model
ligand having an intramolecuar NH···S hydrogen bond.
2,3-Bis(isopropylthio)aniline (2). This compound was synthe-
sized by a similar procedure to the literature.30 Compound 1 (602 mg,
1.94 mmol) was dissolved in a mixture of 2.1 M NaOH aq (6.0 mL)
and dioxane (6.0 mL). The solution was stirred at room temperature
for 45 min. Hydrochloric acid (6 M, 6.0 mL) was added, and the
solution was stirred for 30 min. Aqueous NaOH solution (2.1 M, 15.0
mL) was added, and the solution was stirred for 35 min. The mixture
was extracted with diethyl ether. The organic layer was washed with
water (20 mL × 2) and sat. NaCl aq (15 mL × 3), and then dried over
Na2SO4. The solution was concentrated to dryness under reduced
pressure to give clear oil. The oil was crystallized from diethyl ether to
RCONH group at the 3-position of the 1,2-benzenedithiolato
ligand provides a rigid and stable intraligand NH···S hydrogen
bond. Moreover, the five-membered hydrogen bond is
reminiscent of the intraligand interaction in the molybdopterin
cofactor (Figure 1a). Alteration of the R group can adequately
adjust the strength of the hydrogen bond electrostatically by
inductive effects. These newly designed ligands enabled
quantitative evaluation of the strength of the hydrogen bonds,
and they excluded steric hindrance in the molybdenum
complexes, resulting in a systematic investigation of the
relationship between the strength of the hydrogen bond and
the reactivity of the molybdoenzyme models.
1
afford colorless crystals. Yield: 425 mg (91%). H NMR (CDCl3): δ
7.05 (t, J = 7.93 Hz, 1H, Ar−H), 6.62 (dd, J = 7.93, 0.43 Hz,1H, Ar−
H), 6.52 (dd, J = 7.93, 1.22 Hz,1H, Ar−H), 4.51 (br, 2H, Ar-NH2),
3.43 (sep, J = 6.77 Hz, 1H, CH), 3.41 (sep, J = 6.77 Hz, 1H, CH), 1.38
(d, J = 6.77 Hz,6H, CH3), 1.26 (d, J = 6.65 Hz,6H, CH3). Anal. Calcd
for C12H19NS2: C, 59.70; H, 7.93; N, 5.80. Found: C, 59.70; H, 7.94;
N, 5.79.
EXPERIMENTAL SECTION
■
All procedures were performed under Ar atmosphere by Schlenk
technique except extraction.
N-{2,3-Bis(isopropylthio)phenyl}acetamide (3a). This com-
pound was synthesized by a similar method to the literature.31 The
amine 2 (412 mg, 1.71 mmol) was dissolved in acetic anhydride (10
mL) at room temperature. After stirring at 90 °C for 105 min,
unreacted acetic anhydride was hydrolyzed by the addition of water (6
mL) and heating at 60 °C for 30 min. The solution was concentrated
to dryness under reduced pressure. The residue was dissolved in
diethyl ether (50 mL). The solution was washed with 2% HCl aq (10
mL × 3), 4% NaHCO3 aq (10 mL × 3), water (10 mL × 5), and sat.
NaCl aq (10 mL × 2), and then dried over Na2SO4. The solution was
concentrated to dryness under reduced pressure to give a colorless oily
residue. It was crystallized from diethyl ether to give colorless crystals.
Yield: 510 mg (quant.). 1H NMR (CDCl3): δ 8.91 (br, 1H, NH), 8.25
(d, J = 8.17 Hz, 1H, Ar−H), 7.28 (t, J = 8.17 Hz, 1H, Ar−H), 6.99 (d,
J = 8.17 Hz, 1H, Ar−H), 3.49 (sep, J = 6.85 Hz, 1H, CH), 3.37 (sep, J
= 6.85 Hz, 1H, CH), 2.21 (s, 1H, CH3), 1.38 (d, J = 6.85, 6H, CH3),
1.26 (d, J = 6.85, 6H, CH3). FT-IR (CH2Cl2): ν(NH) 3338 cm−1.
Anal. Calcd for C14H21NOS2: C, 59.32; H, 7.47; N, 4.94. Found: C,
59.32; H, 7.49; N, 4.92.
Materials. Tris(4-tert-butylphenyl)methane,26 2,3-bis-
(isopropylthio)benzoic acid,27 (NEt4)[MoVO(SPh)4],28
(NEt4)2[MoIVO(S-4-ClC6H4)4],29 and Ph3CCOCl13 were prepared
by the reported methods.
Synthesis of Tris(4-tert-butylphenyl)acetic Acid. Lithiation
was carried out in a similar manner as described in the literature.26
Tris(4-tert-butylphenyl)methane (16.5 g, 40.0 mmol) was dissolved in
a mixture of dry tetrahydrofuran (THF, 200 mL) and hexamethyl-
phosphoramide (HMPA, 66 mL) and cooled at −70 °C. To this
solution was added n-butyllithium in n-hexane (1.67 M, 26 mL, 43
mmol). The color of the solution turned dark blood red. The solution
was stirred at −40 °C for 2 h. Thereafter, dry CO2 gas was bubbled
through the reaction mixture at −30 °C for 30 min, and the solution
was warmed up to room temperature to give a yellow solution. The
reaction mixture was treated carefully with a piece of ice and conc. HCl
aq cooling in an ice bath. The resulting white powder was filtered,
washed with THF and n-hexane, dried over P2O5 in vacuo. Yield: 18.2
1
g (quant.). Mp: > 300 °C. H NMR (DMSO-d6): δ 13.0 (br, 1H,
N-{2,3-Bis(isopropylthio)phenyl}pivalamide (3b). To a THF
solution (10 mL) of 2 (142 mg, 0.59 mmol) were added dropwise
NEt3 (0.2 mL, 1.48 mmol) and t-BuCOCl (0.18 mL, 1.48 mmol) in an
ice bath. The solution was stirred overnight at room temperature and
concentrated to dryness under reduced pressure to give an orange
powder. The residue was dissolved in ethyl acetate (40 mL). The
solution was washed with 2% HCl aq (10 mL × 3), 4% NaHCO3 aq
(10 mL × 3), water (10 mL × 3), and sat. NaCl aq (10 mL × 3), and
then dried over Na2SO4. The solution was concentrated to dryness
under reduced pressure to give orange needles. Recrystallization from
COOH), 7.31 (d, J = 8.5 Hz, 6H, Ar−H), 7.06 (d, J = 8.5 Hz, 6H, Ar−
H), 1.27 (s, 27H, t-Bu). FT-IR (KBr): ν(CO) 1699 cm−1. Anal.
Calcd for C32H40O2: C, 84.16; H, 8.83. Found: C, 81.57; H, 9.06.
N-{2,3-Bis(isopropylthio)phenyl}carbamoyl Azide (1). To a
solution of 2,3-bis(isopropylthio)benzoic acid (6.19 g, 22.9 mmol) in
CH2Cl2 (75 mL) was added SOCl2 (60 mL, 830 mmol) and DMF
(1.0 mL), and the mixture was refluxed for 3 h. The solution was
concentrated to dryness under reduced pressure. The resulting reddish
orange residue was dissolved in acetone (100 mL). Cold aqueous
solution (25 mL) of NaN3 (8.99 g, 138 mmol) was added to the
solution quickly cooling in an ice bath, and the mixture was stirred at
room temperature for 1 h. During the reaction, evolution of nitrogen
gas was observed. Cold water (10.0 mL) was added to this solution,
and the mixture was stirred for 1.5 h. The solution was stirred at 50 °C
for another 30 min and concentrated to dryness under reduced
pressure. The residue was dissolved in ethyl acetate. The solution was
washed successively with 4% NaHCO3 aq (40 mL × 2), water (30 mL
1
diethyl ether gave colorless blocks. Yield: 192 mg (quant.). H NMR
(CDCl3): δ 9.35 (br, 1H, NH), 8.34 (dd, J = 8.00, 1.20 Hz, 1H, Ar−
H), 7.28 (t, J = 8.00 Hz, 1H, Ar−H), 6.98 (dd, J = 8.00, 1.20 Hz, 1H,
Ar−H), 3.49 (sep, J = 6.65 Hz, 1H, CH), 3.39 (sep, J = 6.65 Hz, 1H,
CH), 1.38 (d, J = 6.70 Hz, 6H, CH3), 1.34 (s, 9H, t-Bu), 1.26 (d, J =
6.70 Hz, 6H, CH3). ESI-MS ([M] = t-BuCONHC6H3(SC3H7)2): m/z
326.2 (M + H+)+, 348.2 (M + Na+)+. FT-IR (CH2Cl2): ν(NH) 3343
382
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
cm−1. Anal. Calcd for C17H27NS2: C, 62.72; H, 8.36; N, 4.30. Found:
C, 62.63; H, 8.21; N, 4.33.
room temperature (r.t.) for 4 days. The precipitated yellow powder
was collected by centrifugation and washed with water. The product
was dried over P2O5 under reduced pressure. Yield: 103.6 mg (85%).
1H NMR (CDCl3): δ 8.74, 8.71 (br, 1H, NH (cis, trans)), 8.62, 8.58
(d, J = 8.48, Hz, 1H, Ar−H (cis, trans)), 7.55, 7.50 (dd, J = 7.56, 1.46
Hz, 1H, Ar−H (cis, trans)), 7.37, 7.36 (t, J = 8.05 Hz, 1H, Ar−H (cis,
trans)), 2.32, 2.29 (s, 3H, CH3 (cis, trans)). ESI-MS ([M] = (1,2-S2-3-
CH3CONHC6H3)2): m/z 417.2 (M + Na+)+.
N-{2,3-Bis(isopropylthio)phenyl}trifluoroacetamide (3c).
This compound was synthesized by a similar method to the
literature.13 To a solution of 2 (435 mg, 1.8 mmol) in THF (15
mL) was added trifluoroacetic anhydride (1.25 mL) at 0 °C. The
solution was stirred overnight at 25 °C. Water (8 mL) was added, and
the resulting solution was concentrated to dryness under reduced
pressure. The residue was dissolved in a mixture of diethyl ether and
water. The organic layer was separated and washed with 2% HCl aq
(10 mL × 2), 4% NaHCO3 aq (20 mL × 2), water (10 mL × 2), and
sat. NaCl aq (10 mL × 2). The solution was dried over Na2SO4 and
concentrated to dryness under reduced pressure to give brown oil.
(1,2-S2-3-t-BuCONHC6H3)2 (4b). A solution of 3b (183 mg, 0.56
mmol) in THF (2.0 mL) was added dropwise to a solution of sodium
(54 mg, 2.30 mmol) in liquid NH3 (∼50 mL). After stirring for 20
min, the reaction was quenched with NH4Cl (69 mg, 1.30 mmol). The
solvent was removed, and the residue was dissolved in water (30 mL).
The solution was washed with diethyl ether (20 mL) and stirred in the
air at r.t. for 2 days. The precipitation was collected by centrifugation
and washed with water and sat. NaCl aq. The product was dried over
P2O5 under reduced pressure to give a pale yellow powder. Yield: 63
1
Yield: 556 mg (92%). H NMR (CDCl3): δ 9.93 (br, 1H, NH), 8.22
(dd, J = 8.23, 1.10 Hz, 1H, Ar−H), 7.35 (t, J = 8.23 Hz, 1H, Ar−H),
7.12 (dd, J = 8.23, 1.10 Hz, 1H, Ar−H), 3.51 (sep, J = 6.71 Hz, 1H,
CH), 3.39 (sep, J = 6.65 Hz, 1H, CH), 1.38 (d, J = 6.65, 6H, CH3),
1
1. 26 (d,
J = 6. 71, 6H, CH3 ) . E SI -M S ( [M] =
mg (47%). H NMR (CDCl3): δ 9.20, 9.14 (br, 1H, NH (cis, trans)),
CF3CONHC6H3(SC3H7)2): m/z 338.0 (M + H+)+, 360.1 (M +
Na+)+. FT-IR (CH2Cl2): ν(NH) 3286 cm−1. Anal. Calcd
C14H18F3NOS2: C, 49.83; H, 5.38; N, 4.15. Found: C, 50.17; H,
5.36; N, 4.11.
8.65, 8.59 (dd, J = 8.40, 1.40 Hz, 1H, Ar−H (cis, trans)), 7.54, 7.48
(dd, J = 8.40, 1.40 Hz, 1H, Ar−H (cis, trans)), 7.38, 7.36 (t, J = 8.40
Hz, 1H, Ar−H (cis, trans)), 1.39 (s, 9H, t-Bu). ESI-MS ([M] = (1,2-S2-
+
3-t-BuCONHC6H3)2): m/z 479.2 (M + H+)+, 496.1 (M + NH4 )+,
N-{2,3-Bis(isopropylthio)phenyl}-2,2,2-triphenylacetamide
(3d). A solution of Ph3CCOCl (102 mg, 0.33 mmol) in THF was
added dropwise to a solution of 2 (53 mg, 0.22 mmol) and NEt3 (0.05
mL, 0.4 mmol) in THF (2 mL) at 0 °C. The solution was stirred
overnight at 70 °C and concentrated to dryness under reduced
pressure. The residue was dissolved in a mixture of diethyl ether and
water. The organic layer was separated and washed with 2% HCl aq
(20 mL × 2), 4% NaHCO3 aq (20 mL × 2), water (20 mL × 2), and
sat. NaCl aq (20 mL × 2). The solution was dried over Na2SO4 and
concentrated to dryness under reduced pressure to give a colorless
powder. The powder was recrystallized from ethyl acetate to give
colorless needles. Yield: 59 mg (51%). 1H NMR (CDCl3): δ 9.29 (br,
1H, NH), 8.48 (dd, J = 8.29, 1.16 Hz, 1H, Ar−H), 7.31 (m, 15H, Ar−
H(CPh3)), 7.18 (m, 1H, Ar−H), 6.95 (m, 1H, Ar−H), 3.47 (sep, J =
6.65 Hz, 1H, CH), 3.11 (sep, J = 6.71 Hz, 1H, CH), 1.35 (d, J = 6.65
Hz, 6H, CH3), 0.85 (d, J = 6.71 Hz, 6H, CH3). FT-IR (CH2Cl2):
ν(NH) 3293 cm−1. Anal. Calcd. C32H33NOS2: C, 75.10; H, 6.50; N,
2.74. Found: C, 75.00; H, 6.39; N, 2.82.
N-{2,3-Bis(isopropylthio)phenyl}-2,2,2-tris(4-tert-
butylphenyl)acetamide (3e). Tris(4-tert-butylphenyl)acetic acid
(386 mg, 0.85 mmol) was dissolved in SOCl2 (3 mL, 40 mmol),
and the solution was stirred at 80 °C for 1.5 h. The solution was
concentrated to dryness under reduced pressure to give a white
powder, which was dissolved in CHCl3 (5 mL). The solution was
added dropwise to a solution of 2 (204 mg, 0.85 mmol) and NEt3 (0.2
mL, 1 mmol) in CHCl3 (3 mL) at 0 °C. The solution was stirred
overnight at 70 °C and concentrated to dryness under reduced
pressure. The residue was dissolved in a mixture of n-hexane/diethyl
ether (1:1) and water. The organic layer was washed successively with
2% HCl aq, 4% NaHCO3 aq, water, and sat. NaCl aq. The solution
was dried over Na2SO4 and concentrated to dryness under reduced
pressure to give a powder. The product was recrystallized from diethyl
ether/MeOH to give colorless needles. Yield: 250 mg (48%). 1H
NMR (CDCl3): δ 9.32 (br, 1H, NH), 8.50 (dd, J = 8.17, 1.10 Hz, 1H,
Ar−H), 7.31, 7.21 (m, 12H, Ar−H (C(4-t-BuC6H4)3)), 7.27 (t, J =
8.17 Hz, 1H, Ar−H), 6.94 (dd, J = 8.17, 1.10 Hz, 1H, Ar−H), 3.45
(sep, J = 6.65 Hz, 1H, CH), 3.11 (sep, J = 6.65 Hz, 1H, CH), 1.35 (d, J
= 6.65 Hz, 6H, CH3), 1.31 (s, 27H, t-Bu), 0.87 (d, J = 6.65 Hz, 6H,
CH3). FT-IR (CH2Cl2): ν(NH) 3294 cm−1. Anal. Calcd.
C44H57NOS2: C, 77.71; H, 8.45; N, 2.06. Found: C, 76.13; H, 8.28;
N, 1.17.
501.3 (M + Na+)+.
(1,2-S2-3-NH2C6H3)2. The compound 4a (205.5 mg, 0.52 mmol)
was dissolved in conc. HCl aq (∼5 mL), and the solution was stirred at
50 °C for 2 days. The solution was basified by NaOH aq in an ice bath
to give yellow powder. The precipitate was collected by centrifugation
and washed with water. The yellow powder was dried over P2O5 under
reduced pressure. Yield: 227.5 mg (quant). 1H NMR (CDCl3): δ 7.15
(dd, J = 7.44, 1.46 Hz, 1H, Ar−H), 7.08 (t, J = 8.05, 7.44 Hz, 1H, Ar−
H), 6.77 (dd, J = 8.05, 1.46 Hz, 1H, Ar−H), 4.72 (br, 2H, NH2). ESI-
MS([M] = (1,2-S2-3-NH2C6H3)2) m/z 309.4 (M − H+)−.
(1,2-S2-3-CF3CONHC6H3)2 (4c). Trifluoroacetic anhydride (1 mL)
was added to a solution of (1,2-S2-3-NH2C6H3)2 (102.7 mg, 0.33
mmol) in THF (25 mL) cooling in an ice bath. The reaction mixture
was stirred at 30 °C overnight. The unreacted (CF3CO)2O was
hydrolyzed by the addition of water. The solution was concentrated to
dryness under reduced pressure to give an off-white powder. The
powder was dissolved in diethyl ether and washed with 2% HCl aq, 4%
NaHCO3 aq, water, and sat. NaCl aq. The organic layer was dried over
Na2SO4. The solution was concentrated to dryness under reduced
pressure to give an off-white powder, which was dried over P2O5 under
reduced pressure. Yield: 44 mg (27%). 1H NMR (CDCl3): δ 9.58, 9.48
(br, 1H, NH (cis, trans)), 8.59, 8.56 (dd, J = 8.36, 1.34 Hz, 1H, Ar−H
(cis, trans)), 7.70, 7.68 (dd, J = 8.36, 1.34 Hz, 1H, Ar−H (cis, trans)),
7.49, 7.48 (t, J = 8.36 Hz, 1H, Ar−H (cis, trans)). ESI-MS ([M] = (1,2-
S2-3-CF3CONHC6H3)2): m/z 501.4 (M − H+)−.
(1,2-S2-3-Ph3CCONHC6H3)2 (4d). A solution of Ph3CCOCl (370
mg, 1.21 mmol) in THF (5 mL) was added dropwise to a solution of
(1,2-S2-3-NH2C6H3)2 (185 mg, 0.60 mmol) and NEt3 (0.2 mL, 1.4
mmol) in THF (5 mL) in an ice bath, and the reaction mixture was
stirred at 70 °C for 4 days. A yellow precipitate was deposited
gradually, which was collected with filtration and washed with THF to
give a yellow powder. The powder was dried over P2O5 under reduced
1
pressure. Yield: 210 mg (41%). H NMR (CDCl3): δ 9.13 (br, 1H,
NH), 8.72 (dd, J = 8.11, 1.71 Hz, 1H, Ar−H), 7.35 (m, 17H, Ar−H +
Ar−H(CPh3)). ESI-MS ([M] = (1,2-S2-3- Ph3CCONHC6H3)2): m/z
849.4 (M − H+)−.
{1,2-S2-3-(4-t-Bu-C6H4)3CCONHC6H3}2 (4e). A solution of (4-t-
Bu-C6H4)3CCOCl (570 mg, 1.20 mmol) in CH2Cl2 (5 mL) was added
dropwise to a solution of (1,2-S2-3-NH2C6H3)2 (187 mg, 0.60 mmol)
and NEt3 (0.2 mL, 1.4 mmol) in CH2Cl2 (5 mL) cooling in an ice
bath, and the reaction mixture was stirred at 50 °C for 4 days. A yellow
precipitate was deposited gradually, which was collected by filtration,
and washed with ethyl acetate to give a yellow powder. The powder
was dried over P2O5 under reduced pressure. Yield: 140 mg (25%). 1H
NMR (CDCl3): δ 9.29 (br, 1H, NH), 8.78 (d, J = 5.67 Hz, 1H, Ar−
H), 7.36−7.24 (m, 14H, Ar−H + Ar−H (C(4-t-BuC6H4)3), 1.33 (s,
(1,2-S2-3-CH3CONHC6H3)2 (4a). A solution of 3a (177 mg, 0.62
mmol) in THF (1.5 mL) was added dropwise to a solution of sodium
(152 mg, 6.61 mmol) in liquid NH3 (about 50 mL). After stirring for
30 min, the reaction was quenched with NH4Cl (180 mg, 3.37 mmol).
Upon quenching, the color turned from blue to colorless. The solvent
was removed, and the residue was redissolved in water (60 mL). The
solution was washed with diethyl ether (30 mL) and stirred in air at
2 7 H , t - B u ) . E S I - M S ( [ M ]
= { 1 , 2 - S 2 - 3 - ( 4 - t - B u -
C6H4)3CCONHC6H3}2) m/z = 1209.6 (M + Na)+.
383
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
(NEt4)2[MoIVO(1,2-S2-3-CH3CONHC6H3)2] (5a). A mixture of
(NEt4)2[MoIVO(S-4-ClC6H4)4] (36 mg, 0.038 mmol) and 4a (15
mg, 0.038 mmol) in MeCN (3 mL) was stirred at room temperature
overnight. The insoluble materials were filtered out. The filtrate was
concentrated to a volume of 1 mL, and diethyl ether was added to the
solution to give an orange powder. The powder was collected, and
washed with 1,2-dimethoxyethane (DME), then dissolved in
acetonitrile. The solution was filtered, and the filtrate was concentrated
under reduced pressure to a minimum volume. Upon cooling in the
refrigerator, an orange powder was deposited. This powder was
separated and dried in vacuo. Yield: 10 mg (33%). 1H NMR
(CD3CN): δ 8.93 (br, 1H, NH), 7.72 (d, J = 7.7 Hz, 1H, Ar−H), 7.23
(d, J = 7.7 Hz, 1H, Ar−H), 6.71 (t, J = 7.7 Hz, 1H, Ar−H), 2.97 (q, J =
7.2 Hz, 8H, NEt4), 2.13 (s, 3H, CH3), 1.03 (tt, J = 7.2, 1.8 Hz, 12H,
NEt4). ESI-MS (M2− = [MoIVO(1,2-S2-3-CH3CONHC6H3)2]2−): m/z
509.3 (M2− + H+)−, 638.0 (M2− + (NEt4)+)−. Absorption spectrum
(DMF): λmax (ε, M−1 cm−1) 331 (8400), 382 (sh) (850), 448 (370)
nm. Anal. Calcd for C32H54MoN4O3S4: C, 50.11; H, 7.10; N, 7.30.
Found: C, 48.48; H, 6.68; N, 6.89.
The disagreement of elemental analysis for molybdenum complexes
is probably caused by their air sensitivity and low crystallinity.
Unfortunately, trials for crystallization of the complexes failed. The
powder was oxidized easily and increase in weight was observed during
weighing. Formal addition of oxygen to the chemical formula
impr oved the results. The calculated values for
C32H54MoN4O3S4·(O2)0.8: C, 48.49; H, 6.87; N, 7.07, which agree
with the found ones.
C32H48F6MoN4O3S4: C, 43.93; H, 5.53; N, 6.40. Found: C, 43.46;
H, 5.34; N, 6.36.
(NEt4)2[MoIVO(1,2-S2-3-Ph3CCONHC6H3)2] (5d). A mixture of
(NEt4)2[MoIVO(S-4-ClC6H4)4] (93 mg, 0.10 mmol) and 4d (83 mg,
0.10 mmol) in DME (5 mL) was stirred at room temperature
overnight. Grayish orange powder was deposited from the reaction
mixture, which was collected and washed successively with DME,
acetonitrile, toluene, and acetonitrile. The orange powder was dried in
1
vacuo. Yield: 27.3 mg (23%). H NMR (CD3CN): δ 9.32 (br, 1H,
NH), 7.90, 7.87 (dd, J = 7.7, 1.2 Hz, 1H, Ar−H (cis, trans)), 7.50−7.25
(m, 16H, Ar−H + CPh3(cis, trans)), 6.78, 6.77 (t, J = 7.7 Hz,, 1H, Ar−
H (cis, trans)), 2.97 (q, J = 7.3 Hz, 8H, CH2(NEt4)), 1.06 (tt, J = 7.3,
1.8 Hz, 12H, CH3(NEt4)). ESI-MS (M2− = [MoIVO(1,2-S2-3-
CPh3CONHC6H3)2]2−) m/z 964.4 (M2−+ H+)−, 1094.3 (M2−
+
(NEt4)+)−. Absorption spectrum (DMF): λmax (ε, M−1 cm−1) 342
(8300), 402 (700), 448 (sh) (450) nm. Anal. Calcd for
C68H78MoN4O3S4: C, 66.75; H, 6.43; N, 4.58. Found: C, 64.80; H,
6.09; N, 4.66.
(NEt4)2[MoIVO{1,2-S2-3-(4-t-BuC6H4)3CCONHC6H3}2] (5e). A
mixture of (NEt4)2[MoIVO(S-4-ClC6H4)4] (211 mg, 0.22 mmol)
and 4e (267 mg, 0.25 mmol) in DME (8 mL) was stirred at room
temperature for 2 h. The grayish red powder was precipitated from the
reaction mixture, which was collected and washed with DME and
propionitrile. The obtained orangish white powder was dried in vacuo.
1
Yield: 91 mg (26%). H NMR (CD3CN): δ 9.43 (br, 1H, NH), 7.90
(dd, J = 7.6, 1.1 Hz, 1H, Ar−H), 7.39 (dd, J = 39.1, 21.7 Hz,12H, Ar−
H((4-t-BuC6H4)3C)), 7.26 (dd, J = 7.6, 1.1 Hz, 1H, Ar−H), 6.76 (t, J
= 7.6 Hz, 1H, Ar−H), 3.10 (q, J = 7.3 Hz, 8H, CH2(NEt4)), 1.35 (s,
27H, t-Bu), 1.16 (tt, J = 7.3, 1.9 Hz, 12H, CH3(NEt4)). ESI-MS (M2−
= [MoIVO(1,2-S2-3-(4-t-BuC6H4)3CCONHC6H3)2]2−) m/z 1300.9
(M2− + H+)−, 1430.8 (M2− + (NEt4)+)−. Absorption spectrum
(DMF): λmax (ε, M−1 cm−1) 289 (33100), 343 (8100), 400 (sh) (780),
446 (sh) (440) nm.
(NEt4)2[MoIVO(1,2-S2-3-t-BuCONHC6H3)2] (5b). This compound
was synthesized by a similar method to the literature.24 To a solution
of (NEt4)[MoVO(SPh)4] (100 mg, 0.15 mmol) in acetonitrile (3 mL)
were added a solution of 4b (70 mg, 0.15 mmol) and
tetraethylammonium borohydride (22 mg, 0.15 mmol) in acetonitrile
(3 mL) and water (0.6 mL). The solution was stirred overnight, and
then the solution was concentrated to dryness under reduced pressure.
The residue was washed with DME, and dissolved in acetonitrile. After
removal of insoluble materials with filtration, the filtrate was
concentrated under reduced pressure and then refrigerated to give
orange powder. This powder was dried in vacuo. Yield: 52 mg (49%).
1H NMR (CD3CN): δ 9.40, 9.38 (br, 1H, NH (cis, trans)), 7.89, 7.88
(dd, J = 7.6, 1.2 Hz, 1H, Ar−H (cis, trans)), 7.32, 7.31 (dd, J = 7.6, 1.2
Hz, 1H, Ar−H (cis, trans)), 6.80, 6.79 (t, J = 7.6 Hz, 1H, Ar−H (cis,
trans)), 2.99 (q, J = 7.2 Hz, 8H, CH2 (NEt4)), 1.40, 1.39 (s, 9H, t-Bu
(cis, trans)), 1.08 (tt, J = 7.2, 1.7 Hz, 12H, CH3 (NEt4)). ESI-MS (M2−
= [MoIVO(1,2-S2-3-t-BuCONHC6H3)2]2−) m/z 296.3 (M2−), 592.5
(M2− + H+)−, 722.3 (M2−+ (NEt4)+)−. Absorption spectrum (DMF):
λmax (ε, M−1 cm−1) 331 (8800), 382 (sh) (930), 448 (440) nm. Anal.
Calcd for C38H66MoN4O3S4: C, 53.62; H, 7.82; N, 6.58. Found: C,
49.78; H, 7.56; N, 5.94.
(NEt4)2[MoVIO2(1,2-S2-3-CH3CONHC6H3)2] (6a). To a DMF
solution (2 mL) of 5a (2.9 mg, 0.0038 mmol) was added a DMF
solution (0.2 mL) containing 0.57 mg (0.0076 mmol) of trimethyl-
amine N-oxide at room temperature. The pale yellow solution turned
immediately to red. The solution was concentrated to dryness under
reduced pressure. The residue was washed with DME and dried under
reduced pressure. The product was used for the spectral measurements
without further purification. 1H NMR (CD3CN): δ 8.43 (br, 1H, NH),
7.70 (d, J = 7.8 Hz, 1H, Ar−H), 6.87 (dd, J = 7.8, 1.3 Hz, 1H, Ar−H),
6.60 (t, J = 7.8 Hz, 1H, Ar−H), 3.12 (q, J = 7.3 Hz, 8H, NEt4), 2.03 (s,
3H, CH3), 1.18 (tt, J = 7.3, 1.9 Hz, 12H, NEt4). ESI-MS (M2−
[MoVIO2(1,2-S2-3-CH3CONHC6H3)2]2−): m/z 654.1 (M2−
=
+
(NEt4)+) −. Absorption spectrum (DMF): λmax (ε, M−1 cm−1) 303
(sh) (14000), 338 (10500), 378 (sh) (5700), 524 (2300) nm.
The following dioxomolybdenum(VI) complexes were synthesized
by a similar method described for 6a.
The following method also gave the identical product, which was
confirmed by spectral measurements. A mixture of (NEt4)2[MoIVO(S-
4-ClC6H4)4] (39 mg, 0.041 mmol) and 4b (20 mg, 0.041 mmol) in
DME (3 mL) was stirred at room temperature overnight. The
precipitated orange powder was collected and washed with DME. The
crude product was purified by reprecipitation from acetonitrile/diethyl
ether to give pure 5b. Yield: 11 mg (32%).
(NEt4)2[MoVIO2(1,2-S2-3-t-BuCONHC6H3)2] (6b). 1H NMR
(CD3CN): δ 8.82 (br, 1H, NH), 7.81 (dd, J = 7.8, 1.3 Hz, 1H, Ar−
H), 6.86 (dd, J = 7.8, 1.3 Hz, 1H, Ar−H), 6.61 (t, J = 7.8 Hz, 1H, Ar−
H), 3.11 (q, J = 7.3 Hz, 8H, NEt4), 1.23 (s, 9H, t-Bu), 1.17 (tt, J = 7.3,
1.9 Hz, 12H, NEt4). ESI-MS (M2− = [MoVIO2(1,2-S2-3-t-
BuCONHC6H3)2]2−): m/z 738.3 (M2− + (NEt4)+) −. Absorption
spectrum (DMF): λmax (ε, M−1 cm−1) 303 (sh) (14600), 338 (11700),
377 (sh) (6400), 522 (230) nm.
(NEt4)2[MoIVO(1,2-S2-3-CF3CONHC6H3)2] (5c). A mixture of
(NEt4)2[MoIVO(S-4-ClC6H4)4] (126 mg, 0.13 mmol) and 4c (67
mg, 0.13 mmol) in DME (5 mL) was stirred at room temperature for
3 h. Orange powder was precipitated from the reaction mixture, which
was collected and washed with DME. The powder was purified by a
similar procedure described for 5a. The resulting orange powder was
dried in vacuo. Yield: 29 mg (27%). 1H NMR (CD3CN): δ 10.02 (br,
1H, NH (cis, trans)), 7.75 (dd, J = 7.7, 1.2 Hz, 1H, Ar−H (cis, trans)),
7.47, 7.46 (dd, J = 7.7, 1.2 Hz, 1H, Ar−H (cis, trans)), 6.88 (t, J = 7.7
Hz, 1H, Ar−H (cis, trans)), 3.07 (q, J = 7.3 Hz, 8H, CH2(NEt4)), 1.14
(tt, J = 7.3, 1.9 Hz, 12H, CH3(NEt4)). ESI-MS (M2− = [MoIVO(1,2-
(NEt4)2[MoVIO2(1,2-S2-3-CF3CONHC6H3)2] (6c). 1H NMR
(CD3CN): δ 9.54 (br, 1H, NH), 7.68 (dd, J = 7.8, 1.3 Hz, 1H, Ar−
H), 6.69 (dd, J = 7.8, 1.3 Hz, 1H, Ar−H), 6.61 (t, J = 7.8 Hz, 1H, Ar−
H), 3.14 (q, J = 7.3 Hz, 8H, NEt4), 1.19 (tt, J = 7.3, 1.9 Hz, 12H,
NEt4). Absorption spectrum (DMF): λmax (ε, M−1 cm−1) 284 (sh)
(20700), 349 (6200), 510 (860) nm.
(NEt4)2[MoVIO2(1,2-S2-3-Ph3CCONHC6H3)2] (6d). 1H NMR
(CD3CN): δ 8.76 (br, 1H, NH), 7.80 (dd, J = 7.8, 1.3 Hz, 1H, Ar−
H), 7.31−7.16 (m, 15H, CPh3), 6.86 (dd, J = 7.8, 1.3 Hz, 1H, Ar−H),
6.62 (t, J = 7.8 Hz, 1H, Ar−H), 2.96 (q, J = 7.3 Hz, 8H, CH2(NEt4)),
S2-3-CF3CONHC6H3)2]2−) m/z 616.3 (M2− + H+)−, 746.0 (M2−
+
(NEt4)+)−. Absorption spectrum (DMF): λmax (ε, M−1 cm−1) 286
1.06 (tt, J = 7.3, 1.9 Hz, 12H, CH3(NEt4)). ESI-MS (M2−
=
(26600), 342 (5800), 435 (670) nm. Anal. Calcd for
[MoVIO2(1,2-S2-3-Ph3CCONHC6H3)2]2−) m/z 981.3 (M2− + H+)−.
384
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
Absorption spectrum (DMF): λmax (ε, M−1 cm−1) 283 (sh) (22800),
357 (9400), 410 (sh) (4600), 525 (2100) nm.
ONMe3 on the software ChemBio3D Ultra (ver. 11.0), and the energy
was minimized by MM2 calculation to give initial models for DFT
calculations. The geometry was roughly optimized using a 3-21G basis
set and then reoptimized by the higher level basis sets described above.
(NEt4)2[MoVIO2{1,2-S2-3-(4-t-BuC6H4)3CCONHC6H3}2] (6e). 1H
NMR (CD3CN): δ 8.82 (br, 1H, NH), 7.76 (dd, J = 7.8, 1.3 Hz, 1H,
Ar−H), 7.26 (dd, J = 47.2, 29.5 Hz,12H, Ar−H((4-t-BuC6H4)3C)),
6.79 (dd, J = 7.8, 1.3 Hz, 1H, Ar−H), 6.53 (t, J = 7.8 Hz, 1H, Ar−H),
3.11 (q, J = 7.3 Hz, 8H, CH2(NEt4)), 1.26 (s, 27H, t-Bu), 1.16 (tt, J =
7.3, 1.9 Hz, 12H, CH3(NEt4)). ESI-MS (M2− = [MoVIO2(1,2-S2-3-(4-
t-BuC6H4)3CCONHC6H3)2]2−) m/z 1576.4 (M2− + 2(NEt4)+ + H+)+,
1706.5 (M2− + 3(NEt4)+)+. Absorption spectrum (DMF): λmax (ε, M−1
cm−1) 356 (8900), 410 (sh) (3400), 528 (1500) nm.
RESULTS
■
Synthesis of the Ligands. A series of 3-acylaminobenze-
nedithiolato ligands was synthesized in a stepwise fashion, as
shown in Scheme 1. The well-known Curtius rearrangement,
1
Physical Measurements. H NMR spectra were obtained on a
Scheme 1. Synthetic Route of the Disulfides and the
a
JEOL GSX-400 spectrometer and a JEOL ECA-500 in CDCl3 or
CD3CN at 30 °C. GOESY spectra were recorded on a VARIAN VNS-
600. ESI-MS spectrometric analyses were performed on a Finnigan
MAT LCQ-MS instrument using acetonitrile solution. IR spectro-
scopic measurements in solution were done on Jasco FT/IR-4000
spectrometer. Raman spectra were measured at 298 K on Jasco NR-
1800 with Liq. N2 cooled CCD detector. Exciting radiation was
provided by Ar+ ion laser (514.5 nm). The measurement of cyclic
voltammograms (CVs) in DMF solution were carried out on a BAS
100B/W instrument with a three-electrode system: glassy carbon
working electrode, a Pt-wire auxiliary electrode, and saturated calomel
electrode (SCE). The scan rate was 100 mV/s. Concentration of the
sample was 2.5 mM, containing 0.1 M of n-Bu4NClO4 as a supporting
electrolyte. Potential was determined at room temperature vs SCE as a
reference. The result was cross-referenced by using the ferrocene/
ferrocenium couple as a calibrant. The redox potential of Fe(Cp)2 was
observed at +0.48 V vs SCE under the same condition. UV−visible
absorption spectra were recorded using a SHIMADZU UV-3100PC
spectrometer.
Molybdenum Complexes
Structural Determination. Each single crystal of 1, 3a, 3b, and 3d
was selected carefully and mounted on MicroMount 200 μm with
Nujol, which was frozen immediately in a stream of cold nitrogen at
200 K. Data collection was made on a Rigaku RAPID II Imaging Plate
area detector with Mo−Kα radiation (0.71075 Å) using MicroMax-
007HF microfocus rotating anode X-ray generator and VariMax-Mo
optics. The structure was solved by SIR-9232 and expanded Fourier
technique using SHELXL-97.33 All non-hydrogen atoms were refined
anisotropically. All hydrogen atoms except amide NH of 1, 3b, and 3d
were located at the calculated positions using the riding model.
Kinetic Analysis. Reaction systems containing the
monooxomolybdenum(IV) complex and Me3NO were monitored
spectrophotometrically in the region 250−700 nm. A typical
measurement was carried out using a 1-mm UV cell containing a
solution (0.3 mL) of monooxomolybdenum(IV) complex (1 mM in
DMF; 0.3 mL) at 27 °C. After thermal equilibrium, a Me3NO solution
(60 mM, 180 mM, 240 mM, 300 mM; 10 μL) was injected through a
silicone rubber cap, and the cell contents were quickly mixed by
shaking. All calculations for the data analysis were performed at 525
nm for 5a, 524 nm for 5b, 525 nm for 5c, 521.5 nm for 5d, and 527.5
nm for 5e.
Density Functional Theory (DFT) Calculations. Geometry
optimizations, frequency calculations, and natural bond orbital (NBO)
analysis were performed using Becke’s three parameter hybrid
fuctionals (B3LYP) in the Gaussian 0334 program package. The
basis set used for Mo was MINI-135 supplemented by the 5p AO with
the same exponent as that for the 5s AO.36 For the other atoms (H, C,
N, O, F, S), a 6-31G** basis set was employed. The coordinates of the
c r y s t a l s t r u c t u r e s , ( N E t 4 ) 2 [ M o I V O { 1 , 2 - S 2 - 3 , 6 -
(CH3CONH)2C6H2}2],24 and (NEt4)2[WVIO{1,2-S2-3,6-
(CH3CONH)2C6H2}2]37 were used for the initial structures with
some modifications. The intermediates of the reaction between the
molybdenum complexes with Me3NO were calculated using the
following procedure. At first, the optimized structures of [MoO-
(bdt)2]2− (bdt = 1,2-benzenedithiolato), trans-[MoIVO(1,2-S2-3-t-
BuCONHC6H3)2]2− (trans-5b), cis-[MoIVO(1,2-S2-3-t-Bu-
CONHC6 H3 )2 ]2 − (cis-5b), [MoI V O{1,2-S2 -3,6-(t-Bu-
CONH)2C6H2}2]2− (7b), and [MoIVO(1,2-S2-3-CF3CONHC6H3)2]2−
(5c) were obtained. The obtained structures were connected with
a
(i) SOCl2/CH2Cl2, then NaN3 aq/acetone. (ii) dil. NaOH aq/
dioxane, HCl aq, then NaOH aq. (iii) RCOCl or (RCO)2O. (iv) Na/
liq. NH3, then air oxidation. (v) conc. HCl aq, then NaOH aq. (vi)
(NEt4)2[MoO(S-4-ClC6H4)4] in DME (or CH3CN), or NEt4BH4,
then (NEt4)[MoO(SPh)4] in CH3CN/H2O. (vi) Me3NO/DMF.
which effectively converts a carboxylic acid to an amine, was
used to introduce an amino group to the benzene ring.38−40 In
the general procedure, a carboxylic acid is converted to an acid
chloride, which is allowed to react with NaN3 to give a
carbazide (RCON3).38 The carbazide is rearranged to the
amine with the evolution of nitrogen gas by heating under
acidic conditions. In our case, the carbazide could not be
isolated and was converted to the carbamoyl azide in the
presence of excess NaN3 without heating. Carbamoyl azides are
known, and their properties have been reviewed.41 Some
derivatives have been structurally characterized by X-ray
analysis.42,43 The isolated N-{2,3-bis(isopropylthio)phenyl}-
carbamoyl azide (1) was fully characterized by H NMR, IR
spectroscopy, elemental analysis, and X-ray analysis.
The molecular structure of 1 is shown in Figure 2, and the
crystallographic parameters are listed in Supporting Informa-
tion, Table S1. The geometrical parameters were typical, and no
significant differences could be found from the reported
structures.42,43 It was noteworthy that the NH group was
1
385
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
S2−C4 angles (99.70−103.35°) were smaller than the C11−
S1−C1 angles (104.91−105.22°). The longer S2−C distances
and smaller C−S2−C angles indicate that S2 has higher p-
character than S1.
1H NMR spectra of 3a−3e in CDCl3 are shown in
Supporting Information, Figure S2. The downfield-shifted
NH signals again suggested the presence of NH···S hydrogen
bonds. The NH chemical shifts were observed in the order 3c >
3b > 3e > 3d > 3a. For 3a, the NH and 4-H signals were
broadened in comparison with the other compounds. Because
the 4-H proton is located in the neighborhood of the carbonyl
group, the proton is deshielded by the π-electrons of the amide
group depending on the torsion angle around the C13−N1
bond in Supporting Information, Figure S1. The broadened
signal of 3a was probably caused by the thermal rotational
motion of the amide plane. The presence of an NH···S bond
was established by IR spectroscopy under similar conditions to
those shown in Supporting Information, Table S2. The strength
of the NH···S hydrogen bond was evaluated by the reported
method using the values of the corresponding disulfide, (S-2-
RCONHC6H4)2, in CH2Cl2.13,17,24 The difference indicates the
relative strength of the hydrogen bond. The order of strength
was 3c > 3b > 3e ≈ 3d > 3a, which was consistent with the 1H
NMR results. The strength of the NH···S hydrogen bond was
dependent on the electron density of the sulfur atom, which
was influenced by the neighboring atom or functional groups.
In metal complexes, the nature of the metal ions or the other
binding ligands, oxidation state, coordination number, and
geometry also affect the strength.14 In this case, the i-Pr group
acted as an electron-donating group to sufficiently increase the
electron density on sulfur to form a NH···S hydrogen bond.
The removal of the protecting isopropyl groups of 3a and 3b
following air oxidation gave dibenzo[c,g][1,2,5,6]tetrathiocin
derivatives, 4a and 4b, with two disulfide linkages. Attempted
deprotection of 3c−3e by the same method was unsuccessful,
but another procedure gave positive results. Acylation of the
amine that was prepared by the hydrolysis of 4a gave 4c−4e.
1H NMR spectra of 4a−4c revealed a mixture of trans and cis
isomers, whereas 4d and 4e exhibited one set of peaks. This
indicated that the bulky substituent prevented the formation of
the cis isomer to avoid steric congestion. Because either isomer
gives the same dithiolato ligand by reduction, separation of
isomers was unnecessary in this step.
Figure 2. Molecular structure of N-[2,3-bis(isopropylthio)phenyl]-
carbamoyl azide (1).
directed toward the sulfur atom with a H···S distance of 2.53(2)
Å, suggesting the presence of a weak NH···S hydrogen bond.
1
The NH group exhibited a H NMR signal at 8.60 ppm and
ν(NH) at 3315 cm−1 in the IR spectrum. The relatively
1
downfield-shifted H signal and lower wavenumber ν(NH)
characterize the acidic NH group forming the NH···S hydrogen
bond.
The Curtius rearrangement generally proceeds at high
temperatures or requires protic or Lewis acids as catalysts.44
In our case, the rearrangement occurred under mild conditions,
although the reaction mixture was heated to complete the
reaction for practicality. The synthesis of carbamoyl azides from
acid chlorides and aqueous NaN3 at room temperature has
been reported.45 Another paper reported the isolation of
carbazide derivatives under mild conditions; however, the IR
spectra exhibited a peak at 3294 cm−1 which suggests the
formation of a carbamoyl azide, which was not mentioned in
the paper.46 These reported compounds contained O or S
atoms close to the N atom adjacent to the carbonyl C of the
carbazide as a common structure. We present here a proposed
mechanism (shown in Supporting Information, Scheme S1).
Electron-rich sulfur probably stabilizes the electron-deficient
nitrene intermediate concomitant with the evolution of
nitrogen gas. The polarized intermediate is readily converted
to an isocyanate by 1,2-rearrangemement, which then easily
reacts with excess azide anion to give product 1.
Synthesis of the Molybdenum Complexes.
Monooxomolybdenum(IV) complexes were synthesized by a
method similar to that previously reported.20,24,25 The
precursors of the ligands, 4a−4e, were reacted with
(NEt4)2[MoIVO(S-4-ClC6H4)4]29 in 1,2-dimethoxyethane
(DME) or acetonitrile at ambient temperature. The reported
method using (NEt4)[MoVO(SPh)4] and NEt4BH4 effectively
afforded the monooxomolybdenum(IV) complexes, but this
reaction required excess water to avoid the formation of
tris(dithiolene) derivatives.24 Both methods gave the products,
which were isolated as crystalline powders, in comparable
yields. The reaction produced some byproducts caused by
incomplete ligand-exchange reaction or the formation of
dinuclear and unknown complexes. The byproducts were
soluble in DME, therefore, removed by exhaustive washing,
resulting in the low yields of the products. Unfortunately, many
crystallization trials failed for our complexes. As described in
the next section, 5a−5d contained both trans and cis isomers,
but 5e, with extremely bulky substituents, was found to be a
single isomer, most likely the trans isomer to avoid steric
The carbamoyl azide 1 was smoothly hydrolyzed by dilute
alkaline solution following the evolution of CO2 gas under
acidic conditions to give the amine 2 in high yield. Acylation of
2 by the corresponding acid chloride or acid anhydride gave
3a−3e in satisfactory yields. These compounds were
1
characterized by H NMR, IR spectroscopy, and elemental
analysis. For 3a, 3b, and 3d, molecular structures were
determined by X-ray analysis. The molecular structures are
shown in Supporting Information, Figure S1, and the
crystallographic parameters are listed in Supporting Informa-
tion, Table S1. These structures were essentially the same
except that the amide plane of 3a was twisted (by
approximately −50°) from the benzene ring, whereas the
amide planes of 3b and 3d were nearly coplanar with the
benzene ring. The amide NH group of 3a was directed to the
neighboring carbonyl group to form an intermolecular
NH···OC hydrogen bond. The coplanarity of 3b and 3d
suggested the presence of intramolecular NH···S hydrogen
bonds. The S2−C12 distances (1.769−1.779 Å) were longer
than the S1−C11 distances (1.764−1.770 Å), and the C12−
386
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
hindrance. Both isomers reacted with Me3NO to give a single
isomer of 6a−6e, as illustrated in Scheme 1.
same properties within the resolution of our measurements.
The validity of this estimation was confirmed by the theoretical
calculations described in the discussion.
Solution Structures of the Monooxomolybdenum(IV)
1
Complexes. The aromatic regions of the H NMR spectra of
IR and Resonance Raman Spectra. To evaluate the
strength of the NH···S hydrogen bond, IR spectra in the solid
state were obtained and listed in Table 1, with the values of the
corresponding disulfides (S-2-RCONHC6H4)2 in CH2Cl2. As
described in previous papers, a large negative shift value,
Δν(NH), indicates the presence of a strong NH···S hydrogen
bond.13,17,24 The estimated order of the strength of the
hydrogen bonds was 5c > 5b > 5a > 5e ≈ 5d. The obviously
stronger NH···S of 5c reasonably explains the downfield-shifted
5a−5e in CD3CN are shown in Figure 3. Peak assignments and
1
NH signal in the H NMR spectrum. The Δν(NH) values
(−50 to −89) of 5a−5d were larger than those (−36 to −52)
of the corresponding complexes (NEt4)2[MoO{S2-3,6-
(RCONH)2C6H2}2] (R = CH3, 7a; t-Bu, 7b; CF3, 7c; CPh3,
7d), with four NH···S hydrogen bonds, by about 1.2−1.8
times.24 This fact suggests that the electron of the non-
hydrogen-bonded thiolate was delocalized over the dithiolene
moiety, increasing the electron density of the other thiolate and
resulting in a stronger hydrogen bond. Such unsymmetrical
electronic structures of these 3-acylaminobenzenedithiolates are
of fundamental significance in the present study. The push−
pull-like adjustment of these ligands should contribute to the
stabilization of the unsymmetrical structure of the molybdenum
complexes, as described below.
1
Figure 3. H NMR spectra of (a) 5a, (b) 5b, (c) 5c, (d) 5d, and (e)
The NH···S hydrogen bond formed more effectively in the
dioxomolybdenum(VI) complexes 6a−6e. The order of
hydrogen bond strength, 6c > 6b > 6a > 6d > 6e, was similar
to 5a−5e. Because dioxomolybdenum(VI) complexes with four
NH···S hydrogen bonds, (NEt4)2[MoVIO2{S2-3,6-
(RCONH)2C6H2}2],24 were too unstable to be isolated,
tungsten analogues were used as references. Substitution of
molybdenum by tungsten resulted in comparable or weaker
NH···S hydrogen bonds in monooxometal(IV) complexes,
(NEt4)2[MIVO{S2-3,6-(RCONH)2C6H2}2] (M = Mo, W). The
differences between tungsten and molybdenum, ΔW−Moν(NH),
were 0 for R = CH3 and 11 for R = t-Bu. In each
dioxotungsten(VI) analogue, (NEt4)2[WVIO2{S2-3,6-
(RCONH)2C6H2}2], two distinct ν(NH)s were observed and
assigned to those at positions trans and cis to the terminal oxo
ligand. The shifts from the disulfides, Δν(NH) values, were
−12 and −66 for R = CH3, and −24 and −65 for R = t-Bu. The
larger negative values were ascribed to the stronger NH···S
hydrogen bond at the trans position.37 The Δν(NH) values for
6a and 6b were more negative than the tungsten analogues by 8
5e in CD3CN at 30 °C. The asterisk denotes DMF as a contaminant.
proposed structures of the two isomers are included. Two sets
of peaks, assignable to the trans and cis isomers, were found for
5a−5d. The chemical shift differences between the two isomers
were smaller than those of the previously reported
(NE4)2[MoIVO(1,2-S-3-t-BuNHCOC6H3)2].25 The enlarged
NH peak in 5a indicates a 2:1 mixture of the isomers. The
major peak is presumably the sterically favorable trans isomer.
The signals of 5a were slightly broader than those of the others,
which suggests the presence of the thermal motion of the amide
plane as found in 3a. The most downfield-shifted NH peak of
5c suggests the strongest NH···S hydrogen bond among these
1
complexes, although solvation effects should be considered. H
NMR spectra revealed the presence of trans and cis isomers;
however, the other spectra (IR, Raman) or redox potentials did
not show distinct peaks for the two isomers. These results
indicate that the difference between the two isomers is very
small or indistinguishable; therefore, two isomers exhibit the
Table 1. IR Bands of ν(NH) (cm−1) in the Molybdenum Complexes in the Solid State Compared with the Corresponding
Disulfide, (S-2-RCONHC6H4)2 in a CH2Cl2 Solution
a
b
R
Mo complexes
(S-2-RCONHC6H4)2
Δν(NH)
MoIVO
5a
5b
5c
5d
5e
6a
6b
6c
6d
6e
CH3
t-Bu
CF3
3321
3331
3269
3291
3289
3308
3302
3229
3272
3285
3382
3397
3358
3341
3342
3382
3397
3358
3341
3342
−61
−66
−89
−50
−53
−74
−95
−129
−69
−57
CPh3
C(C6H4-4-t-Bu)3
MoVIO2
CH3
t-Bu
CF3
CPh3
C(C6H4-4-t-Bu)3
a
b
Nujol. In CH2Cl2 (10 mM).
387
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
and 30 cm−1, respectively. The results clearly demonstrated that
the unsymmetrical substitution effectively localized the negative
charge on the hydrogen-bonded thiolate anion.
Resonance Raman and IR spectra of 5a−5e in the solid state
were measured, and ν(MoO)s are summarized in Table 2.
Table 2. Raman and IR Bands of Molybdenum Complexes in
the Solid State
a
ν(MoO) /cm−1
MoIVO
Raman
IR
5a
5b
5c
5d
5e
910 (+8)
915 (+13)
914 (+12)
920 (+18)
911 (+9)
902
910 (+5)
914 (+9)
913 (+8)
917 (+12)
910 (+5)
905
Figure 4. Cyclic voltammograms of (NEt4)2[MoIVO(1,2-S2-3-
CH3CONHC6H3)2] (5a) in DMF. Two scans were obtained in the
different ranges of the potential.
a
b
Table 3. Redox Potentials in DMF
8
c
b
Raman
complex
E1/2
ΔE1/2
MoVIO2
νs(MoO)/cm−1
νas(MoO)/cm−1
5a
5b
5c
5d
5e
−0.21
−0.19
−0.14
−0.24
−0.28
−0.38
+0.17
+0.19
+0.24
+0.14
+0.10
6a
6b
6c
6d
6e
870 (+12)
873 (+15)
878 (+20)
868 (+10)
870 (+12)
858
840 (+11)
842 (+13)
844 (+15)
835 (+6)
839 (+10)
829
c
8
d
a
[MoIVO] = 2.5 mM. [n-Bu4NClO4] = 0.1 M. Potentials are reported
9
b
a
Differences from the value of (NEt4)2[MoIVO(bdt)2] (8) are shown
vs SCE at Pt working electrode. The scan rate was 100 mV/s. The
c
shifts from the value of (NEt4)2[MoIVO(bdt)2] (8). Ref 47.
b
c
in the parentheses. Refs 23, 47. Differences from the value of
d
(NEt4)2[MoVIO2(bdt)2] (9) are shown in the parentheses. Ref 12.
relationship between the positive shift of the redox potential
and the number of NH···S hydrogen bonds, has been found for
related compounds.13,16,18
To evaluate the contribution of the NH···S hydrogen bonds,
the differences from the value of (NEt4)2[MoIVO(bdt)2]
(8)23,47 are also shown, parenthetically. All complexes showed
higher ν(MoO) than that of 8, where the order was 5d > 5b
≈ 5c > 5e ≈ 5a. Each complex exhibited a weaker ν(MoO)
in comparison with the corresponding (NEt4)2[MIVO{S2-3,6-
(RCONH)2C6H2}2] (7a−7d) with four NH···S hydrogen
bonds, where ν(MoO)s were found at 914−928 cm−1.24
The individual hydrogen bonds of 5a−5d were stronger than
those of the corresponding 7a−7d; however, the overall
contributions did not surpass the combined power of the four
hydrogen bonds in 7a−7d. Dioxomolybdenum(VI) complexes
6a−6e exhibited characteristic bands, that is, symmetric and
asymmetric ν(MoO)s, ascribed to the cis-dioxo config-
uration. These values are listed in Table 2 together with the
differences from those of (NEt4)2[MoVIO2(bdt)2] (9).12 The
higher wavenumber shifts were 10−20 cm−1 and occurred in
the order 6c > 6b > 6a ≈ 6e ≥ 6d, which roughly agrees with
the strength of the hydrogen bond.
Electrochemical Properties. Cyclic voltammograms of 5a
in DMF are shown in Figure 4, and the redox potentials of 5a−
5e and 8 are listed in Table 3. A quasi-reversible Mo(IV)/
Mo(V) redox couple and irreversible oxidation to Mo(VI)
species were observed as well-known redox behaviors for
analogous complexes.19,23,24 The redox potentials were found in
the order 5c > 5b > 5a > 5d > 5e, which was consistent with
the strength of the NH···S hydrogen bonds. The positive shifts
of 5a−5d were smaller than the corresponding (NEt4)2[MoO-
{S2-3,6-(RCONH)2C6H2}2] (7a−7d), where the alphabetic
notation follows 5a−5d.24 In the case of R = CF3, ΔE1/2 values
were +0.24 V for 6c and +0.46 V for 7c. When the number of
the hydrogen bonds was halved from four to two, the ΔE1/2
value decreased by half. Such additivity, that is, a linear
Absorption Spectra. UV−visible spectra of 5a−5e and
6a−6e in DMF are shown in Figure 5. The spectra of 5a and
5b were essentially identical and were identical to those of
(NEt4)2[MoIVO(bdt)2] (8), except for a small shift and change
in the absorbance.23 The absorption band at ∼330 nm was
ascribed to the ligand-to-metal charge-transfer (LMCT) band.23
The peak maxima at 331 nm of 5a and 5b were equal to that of
(NEt4)2[MoIVO(tdt)2] (tdt = 3,4-toluenedithiolato) rather than
8 (328 nm). This small red shift indicated that the acylamino
group acts as a weak electron-donating group. The shape of
each spectrum from 5a−5d resembled the corresponding
reported spectra for 7a−7d.24 The shape predominantly
depended on the nature of the substituent, not on the number.
The LMCT bands overlapped with the ligand absorptions;
thus, the distinct maximum for 5a was found as a shoulder peak
in 7a. The broad weak peaks at ∼450 nm were assignable to d−
d bands. The similarity indicates the absence of significant
geometrical changes in these complexes.24
Molybdenum(VI) complexes have d0 configuration; thus, the
lowest-energy absorptions at about 510−530 nm were assigned
to LMCT bands. Previously, the excitation profiles revealed
that the bands included both thiolate-to-Mo and oxo-to-Mo
LMCT bands.48 The higher-energy absorptions, that is, 300 nm
< λ < 450 nm, were also more intense than in the case of 5a−
5d, and thus, they should be ascribed to LMCT bands.
Consequently, the absorption bands of 6a−6e were character-
istic of dioxomolybdenum(VI) bis(dithiolate) complexes and
proved the formation of the products.
Reduction of Trimethylamine N-oxide. Stoichiometric
reactions between 5a−5e and trimethylamine N-oxide in DMF
were monitored by absorption spectroscopy. Observed rate
388
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
which are summarized in Table 4 along with the data for related
complexes.24 The constants for 5d and 5e are almost equal
Table 4. Observed Initial Reaction Rate Constants (kobs) in
the Reaction between Monooxomolybdenum(IV)
Complexes and Me3NO in DMF at 27 °C
kobs (10−4s−1
)
[Me3NO] = 2 [Me3NO] = 6 [Me3NO] = 8 [Me3NO] = 10
complex
mM
mM
mM
mM
5a
5b
5c
5d
5e
57
9
248 48
261 69
317 79
420 48
377 72
543 24
77 14
8
3
9
2
25
28
49
8
3
1
90 26
97 25
158 23
212 43
144 35
a
b
7a
∼140
130
1
a
7b
3.3 0.1
35
3.3 0.1
a
7c
1
a
7d
260 10
5.4 0.1
1700 400
a
8
57
4
a
b
Ref 24. The value was read from the graph in the literature.
within the experimental errors. The reaction rate constants kobs
were in the order 5b > 5a > 5d ≈ 5e > 5c. This trend was
similar to the strength of the NH···S hydrogen bonds, except
for 5c. The results were very different from 7a−7d. These
complicated trends were related to bulkiness or steric hindrance
and the intermediates of the reaction. A comprehensive analysis
and explanation are described, with theoretical calculations, in
another section. The dependence of kobs on the concentration
of the substrate Me3NO indicates that these reactions are
essentially second-order, except for 5c (Supporting Informa-
tion, Figure S3). Second-order kinetics analysis gave reasonable
results (Supporting Information, Figure S4). The second-order
rate constants, k2, were 4.1 0.7 M−1 s−1 for 5a, 6.7 0.3 M−1
s−1 for 5b, 2.1 0.4 M−1 s−1 for 5d, and 2.2 0.4 M−1 s−1 for
5e. The rate constant k2 for 5a was larger than that of 7a (2.3
M−1 s−1) with four NH···S hydrogen bonds.37 Disparity in the
reactivities between trans and cis isomers was hardly detectable.
When a trans/cis mixture of 5c was allowed to react with
Figure 5. UV−visible spectra of (a) 5a−5e and (b) 6a−6e in DMF.
Each substituent R is shown in the parentheses.
constants kobs were obtained by pseudo-first-order kinetic
analyses. The spectral change for the reaction of 5a with 2
equivalents Me3NO to give 6a is shown in Figure 6. The
absorption maximum at 525 nm due to the LMCT band of 6a
increased gradually. At the initial stage (up to 30% conversion),
a logarithmic plot of the decrease in 5a against the reaction
time gave a linear correlation with the slope, −kobs, as shown in
the inset in Figure 6. In a similar manner, the rate constants
were obtained for various Me3NO concentrations (2−10 mM),
1
Me3NO while being monitored by H NMR spectroscopy, the
trans/cis ratio was retained during the reaction.
Structures of the Dioxomolybdenum(VI) Complexes.
1H NMR spectra of 6a−6e in CD3CN at 30 °C are shown in
Figure 7. Each complex exhibited one set of peaks, suggesting
the presence of a single isomer with equivalent ligands.
According to the interpretation in the previous paper, it was
reasonable that the NH···S hydrogen-bonded thiolates were
located at the position trans to the oxo ligand, as illustrated in
Figure 7.25 This isomer is designated ″A″ in this paper. Another
isomer with equivalent ligands is termed ″B″, where the
hydrogen bond is cis to the oxo ligand, as shown in Supporting
Information, Figure S7. Both isomers have a 2-fold axis through
the bisected line of the O−Mo−O angle. A comparison of the
1H NMR spectra for 5b and 6b is shown in Supporting
Information, Figure S5. Two sets of peaks for the coordinated
ligands in 5b changed to one set of peaks in 6b. The NH, 6-H,
and t-Bu signals of 5b were shifted upfield by 0.57, 0.47, and
0.17 ppm, respectively. IR spectra revealed that the NH···S
hydrogen bond of 6b was stronger than that of 5b, as described
above (Table 1). If the other contributions are ignored, the NH
signal should be shifted downfield from 5b to 6b. Thus, the
upfield shift must be caused by spatial shielding effects of the
Figure 6. UV−visible spectral change of 5a to 6a on the stoichiometric
reaction between 5a and Me3NO in DMF at 27 °C ([MoIVO] = 1
mM, [Me3NO] = 2 mM). Pseudo-first order kinetics of 5a to 6a is
shown in the inset, where the fitting line is y = ln(1 − [MoVIO2]/
[MoIVO]0) = −kobs·t and kobs = 57 × 10−4 s−1.
389
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
shorter than Mo−S1. The Mo−S1 bond of 5b was apparently
shorter than that of 8; however, the bond order demonstrated
that the bond was weakened by the NH···S hydrogen bond,
while the Mo−S2 bond was strengthened. These values
suggested that the NH···S hydrogen bond slightly weakens
Mo−S1 by direct contact to decrease the electron density on
sulfur and strengthens the neighboring Mo−S2 bond by
electronic interaction via the benzene ring. This effect was
observed in previous studies on (NEt4)2[MoIVO(S2-3-t-
BuNHCOC6H3)2] and was discussed using X-ray structures
and DFT calculations.25 The unsymmetrically localized
negative charge on the hydrogen-bonded sulfur increases the
positive charge on the Mo atom, which is consistent with the
positive shift of the redox potentials of 5b. The lower energy
level of the HOMO in 5b than that in 8 also agreed with the
observed one-electron oxidation potentials of the molybdenum-
(IV) complexes. The shorter or stronger MoO bond means
significant donation from the oxo group to the Mo ion, that is, a
decrease in the electron density of the oxo ligand. The natural
charge of O, therefore, appears slightly less negative. The
difference in the calculated total energy between the trans and
cis isomers was less than 0.1 kcal mol−1. This minor difference
agrees with the experimental observation of a 1:1 mixture of
trans/cis isomers. The difference of the redox potentials
between trans- and cis-5b was estimated from the energy level
of HOMO listed in Supporting Information, Table S3. The
calculated value is ∼3 mV, which is negligibly small and
experimentally undetectable. The estimated differences of the
vibrational energy between trans- and cis-5b were ∼1.7 cm−1 for
ν(NH) and ∼0.3 cm−1 for ν(MoO) from the frequency
calculations. These values are too small to detect each signal
separately in broad IR and Raman spectra.
The dioxomolybdenum(VI) complexes 6a−6e were present
in solution as the unique isomer A, where the hydrogen-bonded
thiolate was trans to the terminal oxo ligand, as estimated from
the spectroscopic analysis. To confirm this conclusion, the
optimized structures of isomers A and B for 6b were calculated.
The structures of 6b-A and 6b-B, in addition to 9, are shown in
Supporting Information, Figure S7. The geometric and
electrostatic parameters are summarized in Supporting
Information, Table S4. For these complexes, the sulfur atom
at the position trans to oxo ligand is labeled S1, connected to
C1, and the other is S2. The total energy of 6b-A was calculated
to be less than that of 6b-B by approximately 5.38 kcal mol−1.
These results demonstrated clearly that 6b-A is a unique
isomer, although the difference in the calculated energies was
not very large. This difference was comparable to the previously
reported value.25 Unlike trans-5b and cis-5b, the two isomers
showed significant differences with each other. The NH···S
hydrogen bond distance of 6b-A was shorter than that of 6b-B.
Both the bond index and bond order of 6b-A clearly indicated
an increase in covalency in comparison with that of the
corresponding monooxomolybdenum(IV) complex, 5b. These
results were in good agreement with the values of ν(NH) in the
IR spectra. The MoO bond of 6b-A was significantly shorter
than those of 6b-B and 9, which was consistent with the Raman
spectral observations. In 6b-A, Mo−S1 bond was obviously
longer than Mo−S2 by 0.2256 Å. The long Mo−S1 bond
localizes the negative charge of the thiolate anion on the sulfur
atom (calculated natural charge is −0.25738), which is
favorable for the formation of the NH···S hydrogen bond.
Conversely, the hydrogen bond stabilizes the localized electron
on sulfur and enables the thiolate to maintain the
1
Figure 7. H NMR spectra of (a) 6a, (b) 6b, (c) 6c, (d) 6d, and (e)
6e in CD3CN at 30 °C. An asterisk (*) indicates DMF as a
contaminant.
delocalized π-electrons on the ligand. A similar situation was
discussed previously.25 Interligand interactions in 6b are
illustrated in Supporting Information, Figure S6. The t-Bu
group was located over the benzene ring of the other ligand in
the same molecule. The spatial proximity resulted in the
observation of NOEs. The GOESY spectrum indicated that the
t-Bu group was close to the 5- and 6-H protons by the favorable
interligand orientation, while intraligand contacts were
observed as intense signals (Supporting Information, Figure
S6). These results demonstrated clearly that isomer A
represented the unique configuration of 6a−6e. The NH···S
hydrogen bond must be situated at the position trans to the
terminal oxo ligand.
DISCUSSION
■
Thermodynamic Stability of the Isomers for the
MoIVO and MoVIO2 Complexes. H NMR spectra revealed
1
that each isolated 5a−5d comprised a mixture of trans and cis
isomers. The differences in the thermodynamic stabilities
between the trans and cis isomers were estimated by theoretical
calculations. Geometry optimizations were performed by DFT
calculations, and the optimized structures of trans and cis
isomers of 5b are shown in Supporting Information, Figure S7.
Selected geometric and electrostatic parameters are summar-
ized in Supporting Information, Table S3 with the parameters
for 8 as a reference. As two ligands are identical, only one set of
parameters is shown for each complex. The hydrogen-bonded
sulfur was designated S1, linked to C1, and the other was S2−
C2. In the table, the Wiberg bond index and atom−atom
overlap-weighted natural atomic orbital (NAO) bond order are
shown to estimate the covalency of the bond. The bond index
is approximately equal to the conventional bond order. The
positive values of atom−atom overlap-weighted NAO bond
order indicates a bonding character or attractive interaction
between the two atoms. The length and strength of the NH···S
hydrogen bonds of trans- and cis-5b were approximately the
same. For either trans- or cis-5b, the Mo−S and MoO bonds
were shorter than those of 8. In each ligand, Mo−S2 was
390
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
extraordinarily long Mo−S1 bond. On the other hand, the short
Mo−S2 indicates significant electron-donation from S2 to Mo,
resulting in the lower negative natural charge on S2
(−0.12336). In 6b-B, the difference between Mo−S1 and
Mo−S2 was smaller than those in 6b-A and 9. The originally
localized negative charge on S1 was distributed between S1 and
S2 as an averaged charge. The configuration of B has the
opposite effect on the stabilization of 6b.
complexes. Electron-rich thiolate is an intrinsically strong donor
that interferes with donation by the oxo ligand to the
molybdenum center. When the thiolate is hydrogen-bonded
by an amide NH, its donating ability is decreased, ultimately as
in the case of thioketone. This effect results in a strong MoO
bond.
In this paper, the contributions of the NH···S hydrogen bond
were evaluated using substituent effects. The strength of the
NH···S hydrogen bond was in the order 6c > 6b > 6a > 6d >
6e, estimated by Δν(NH) in the IR spectra. The strength of the
MoO bond was in the order 6c > 6b > 6a > 6e ≈ 6d,
according to Δνs(MoO) in the Raman spectra. The
difference in νs(MoO) between 6d and 6e was about 2
cm−1. The peaks were broad and noisy, and the resolution was
1−2 cm−1; thus, they were approximately the same or within
experimental error. A plot of Δνs(MoO) values versus
Δν(NH) for 6a−6e is shown in Figure 8b, where the plot for 9
is placed at the origin, assuming νs(MoO) = Δν(NH) = 0. A
strong correlation (r2 = 0.96) was clearly demonstrated. The
plot of 6e deviated the most from the fitting line. As mentioned
above, this deviation includes experimental error, and thus, a
strict evaluation is difficult; however, another factor may
possibly apply. The two extremely bulky substituents of 6e are
close to each other in the unique isomer A. Steric interactions
probably result in a slight deformation from the original
geometry, which must perturb the trans influence.
Relation between the Strength of NH···S Hydrogen
Bonds and the Redox Potential. NH···S hydrogen bonds
are known to shift redox potentials in a positive manner. The
relationship between the strength of the hydrogen bond and
the positive shift has been analyzed qualitatively.13,16,18,19,24
A
quantitative analysis is presented here. Using the data in Tables
1 and 3, ΔE1/2 values were plotted against −Δν(NH), where
the values for 8 were assumed as Δν(NH) = ΔE1/2 = 0 (i.e., the
origin) (Figure 8a). The plot clearly indicates a strong
Reaction Mechanism for the Reduction of Me3NO. For
the reduction of Me3NO by monooxomolybdenum(IV)
complexes, two reaction mechanisms have been proposed,
depending on the substituents.23,24 In the traditional
mechanism, Me3NO first attacks Mo at the position trans to
the oxo ligand to give an intermediate Me3NO adduct. The
ONMe3 ligand moves to the cis position through a trans−cis
rearrangement, accompanied by dissociation of the O−N bond
to afford a cis-dioxomolybdenum(VI) complex. If the
substituent is bulky or the intermediate is stabilized, the
trans−cis rearrangement is the rate-determining step at high
concentrations of Me3NO. The reaction essentially obeys
second-order kinetics; however, the apparent reaction rate
reaches a maximum at high concentrations of Me3NO when the
substituent is bulky, such as Ph3Si23 or t-BuCONH.24 This
saturation of the reaction rate against the concentration of
substrate is reminiscent of Michaelis−Menten kinetics, which
are usually observed in enzymatic reactions. The reaction of 7a,
with relatively small CH3CONH groups, also showed such
saturation when [Me3NO] > 6 mM.24 In the case of 7d, with
four extraordinarily bulky Ph3CCONH groups, the trans−cis
rearrangement hardly occurred. Instead, the direct cis attack of
Me3NO proceeded. This is the alternative mechanism reported
in the previous paper.24 The bulky substituent groups induced a
deformation of the MoOS4 core and provided a vacant site at
the cis position for the attack of Me3NO, resulting in the
remarkably fast reaction..
Figure 8. Correlation (a) between Δν(NH) and ΔE1/2 in the
monooxomolybdenum(IV) complexes and (b) between Δν(NH) and
Δνs(MoO) in the dioxomolybdenum(VI) complexes. The original
data are shown in Tables 1−3.
correlation, with r2 = 0.94. The plot of 5e deviated slightly
from the fitting line. Complex 5e has extremely bulky and
hydrophobic (4-t-BuC6H4)3C groups. Generally, the negative
shift of the redox potential in a solvent with a lower dielectric
constant is known both theoretically and experimentally.49−51
Our results suggest that the bulky, hydrophobic substituent
partially masks the molybdenum center and creates a
hydrophobic environment with a low dielectric constant.
Relation between the Strength of NH···S Hydrogen
Bonds and MoVIO Bonds. The contribution of the NH···S
hydrogen bond to the MoO bond at the trans position is
essentially the same as that described in the previous paper.25
This is explained by the so-called “trans influence”. O and S
atoms, situated trans to one another, donate competitively to
the Mo center in such dioxomolybdenum(VI) bis(dithiolate)
In the cases of 5d and 5e, very fast reactions (as in 7d) were
not observed (Table 4). Reduction of the number of the
substituent groups from four to two probably enabled the
trans−cis rearrangement. Interestingly, 5c displayed saturation
of kobs, although rearrangement seemed to be sterically possible.
These reactions were considered to proceed by the traditional
mechanism, which included the attack of Me3NO at the
position trans to the oxo ligand followed by trans−cis
rearrangement. To estimate the reaction mechanism, optimized
structures of the putative intermediates were obtained by DFT
391
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
calculations. The structures of the Me3NO-adducts of trans-5b,
cis-5b, and 7b are shown in Supporting Information, Figure S8.
The geometrical parameters for these compounds, as well as
the related adducts of 8 and 5c, are listed in Supporting
I n f o r m a t i o n , T a b l e S 5 .
A
c o m p a r i s o n o f
monooxomolybdenum(IV) complexes (Supporting Informa-
tion, Table S3) with the corresponding Me3NO-adducts
(Supporting Information, Table S5) showed that the MoO
bond was shortened and the Mo−S bonds were elongated by
the approach of Me3NO, which resulted in a slightly distorted
octahedral geometry of the Me3NO-adducts. The molybdenum
ion was deviated from the S4 plane to form the square
pyramidal geometry of the MoOS4 core. The ONMe3 ligand
weakly coordinated to the molybdenum center with a Mo−O
bond longer than 2.4 Å. The OMo−O bonds were
approximately linear (about 174 or 175°). The Mo−O−N
angles were bent at an angle of about 136−138°. Generally, the
M−O−N (M: metal) bonds found in transition metal
complexes52,53 and organometallic aluminum complexes54 are
bent with an angle of about 120−130°. However, a weak Ba−O
contact found in [(Me3NO)2Ba2Cu4(OCMe2CF3)8] resulted in
an approximately linear Ba−O−N angle (∼175°).55 The
optimized structures demonstrated that the Mo−ONMe3
interaction was a weak, but stable, coordinate bond. The O
Mo−O bond was slightly bent toward the S2 atom, and the
Mo−O−N angle was bent to the opposite side, that is, toward
S4. The OMo−O−N connection was arranged in a zigzag
pattern. The Me3N moiety was located in the vicinity of S4 to
avoid steric congestion with the bulky substituent. Except for
cis-5b-ONMe3, the shortest Mo−S bond was Mo−S2 and the
longest was Mo−S4. The angle of OMo−S2 was the widest,
and that of OMo−S4 was the narrowest. The relationship
between the Mo−S distance and the OMo−S angle has been
discussed for the monooxomolybdenum(V) tetrakis-
(arenethiolate) complex with distorted geometry between
square-pyramidal and trigonal-bipyramidal.19 A similar ten-
dency was found in the optimized structure of trans-5b
(Supporting Information, Table S3). As mentioned above, the
hydrogen bond stabilizes the long Mo−S bond. For trans-5b-
ONMe3 and 5c-ONMe3, the longest Mo−S4 bond did not have
an adjacent hydrogen bond, which probably caused an unstable
Mo−S bond between the electron-rich S and electron-deficient
Mo. The electron-deficient molybdenum was stabilized by the
strongest Mo−S2 bond at the trans position. In the case of cis-
5b-ONMe3, the hydrogen-bonded Mo−S2 bond was trans to
the Mo−S4 bond. The relatively weak donor S2 could not
effectively stabilize the Mo−S4 bond via the molybdenum ion
as described for trans-5b-ONMe3. Instead of the interligand
interaction, the distortion of the geometry was reduced, and the
localized electron density was delocalized via the intraligand
path. The distortion of the intermediates and the open space
between the substituent groups of trans- and cis-5b-ONMe3
promote the preferable motion toward trans−cis rearrangement.
A proposed mechanism for the reduction of Me3NO by 5a−5e,
including the trans−cis rearrangement, is given in Figure 9.
Because there was no significant difference between the total
Figure 9. Proposed mechanism for the reduction of Me3NO by 5a−5e
via trans−cis rearrangement. The intermediates are stabilized by the
NH···S hydrogen bond, especially for 5c (R = CF3).
moiety rises at the back and across the open space.
Simultaneously, the right ligand rotates clockwise around the
Mo−S bond at the front to move the hydrogen-bonded thiolate
to the bottom. The trans−cis rearrangement involves an
intramolecular two electron transfer from MoIV to ONMe3,
accompanying the formation of a MoVIO bond and the
release of NMe3. The widest OMo−S angle and the shortest
Mo−S bond on the far side of the left ligand is retained during
the rearrangement and becomes the widest OMo−S and the
shorter Mo−S of the corresponding dioxomolybdenum(VI)
complex. For the cis isomer, the mechanism is similar to that for
the trans isomer, except that the final rotation is counter-
clockwise. In this case, the three steps probably proceed in a
concerted fashion.
In the intermediates, the NH···S hydrogen bonds obviously
stabilized both MoO and Mo−O(NMe3) bonds. In
particular, the bonds of 7b-ONMe3 and 5c-ONMe3 were
effectively shortened (Supporting Information, Table S5).
These data indicate that both the number and the strength of
the NH···S hydrogen bonds contributed to the stabilization of
the intermediates. The space-filling model of 7b-ONMe3
revealed that steric congestion prevented or retarded the
rotation of the ligand that is required for rearrangement
(Supporting Information, Figure S8c). These results reasonably
explain the small observed rate constant (kobs = 3.3 0.1 s−1;
[Me3NO] ≥ 2 mM) for 7b. A significantly strong hydrogen
bond sufficiently stabilized the intermediate. In the case of 5c,
the value of kobs was saturated at about 9 s−1 for [Me3NO] ≥ 2
mM. The rearrangement of 5c-ONMe3 was retarded by the
strong hydrogen bond in spite of apparently sufficient space
available for the rearrangement.
1
energy of trans- and cis-5b-ONMe3, as well as the H NMR
observation of the retention of the trans/cis ratio during the
reaction, the two pathways are comparable. The arrows indicate
the motion of the moiety and the rearrangement occurs
successively, in the order of the encircled numbers. In the trans
isomer, the weakest Mo−S bond rotationally moves to the left
side to open the space at the opposite side. Next, the Me3NO
The alternative cis-attack mechanism was proposed for the
reaction of Me3NO and 7d (with four Ph3CCONH groups) in
the previous paper.24 The mechanism is reconsidered here
392
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
using DFT calculations. To limit computational costs, a
complex without any substituents, [MoO(bdt)2]2− (8), was
used in the calculations. When the constraint of Mo−O(NMe3)
distance was not used, the optimization converged to the
starting model or the product, [MoO2(bdt)2]2− (9) and Me3N,
depending on the starting model. The optimized structures are
shown in Supporting Information, Figure S9, with the fixed
distance of Mo···O. From a distance of ∼4 Å and diagonally
from above, molecular Me3NO was approached step-by-step to
the molybdenum center. In the range of 4−2.1 Å, the final
structure was the Me3NO-adduct. In accordance with the
approach of Me3NO, the MoO2S4 core was distorted to
octahedral geometry and the Mo−S distance trans to the oxo
ligand was elongated. The positively charged Me3N moiety
tended to be drawn by the basic terminal oxo ligand. At Mo···O
= 2.1 Å, the N−O−Mo angle was about 128°, which agrees
with the reported values for transition metal complexes.52,53
Within 1.8 Å of Mo−O contact, the structure converged to the
dioxomolybdenum(VI) complex and Me3N. In the range of
1.9−2.0 Å, the construct could not converge to a suitable
structure. The total energy of the optimized structure increased
with the approach of Me3NO and dropped immediately upon
cleavage of the O−N bond. These results suggest that the cis
attack does not pass through a metastable intermediate such as
a Me3NO-adduct. The approach of Me3NO and the cleavage of
the N−O bond probably occur simultaneously, with con-
comitant formation of the dioxomolybdenum(VI) complex.
S2) and IR (Table S2) spectra of 3a−3e, dependence of kobs on
[Me3NO] (Figure S3), second order kinetics for the reaction
1
from 5a to 6a (Figure S4), H NMR spectra of 5b and 6b
(Figure S5), GOESY spectrum of 6b (Figure S6), optimized
structures for 5b, 6b (Figure S7) and geometrical parameters
(Tables S3, S4), optimized structures of intermediates (Figures
S8, S9) and their geometrical parameters (Table S5). This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: tokamura@chem.sci.osaka-u.ac.jp.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by JSPS KAKENHI Grant Number
23655049.
REFERENCES
■
(1) Hill, R. Chem. Rev. 1996, 96, 2757−2816.
(2) Pushie, M. J.; George, G. N. Coord. Chem. Rev. 2011, 255, 1055−
1084.
(3) Messerschmidt, A., Huber, R., Poulos, T., Wieghardt, K., Eds.;
Handbook of Metalloproteins; John Wiley & Sons, Ltd: Chichester,
U.K., 2001; Vol. 2.
(4) Enemark, J. H.; Cooney, J. J. A.; Wang, J.-J.; Holm, R. H. Chem.
Rev. 2004, 104, 1175−1200.
CONCLUSIONS
■
The monooxomolybdenum(IV) and dioxomolybdenum(VI) 3-
acylaminobenzenedithiolate complexes having two intramolec-
ular NH···S hydrogen bonds clearly demonstrated the
contributions of NH···S hydrogen bonds to the properties
and the reactivities of the complexes. The reactivities of the
trans and cis isomers of the monooxomolybdenum(IV)
complexes were essentially identical. Each dioxomolybdenum-
(VI) complex was found as a unique isomer where the
hydrogen-bonded thiolate was trans to the terminal oxo ligand.
Reduction of the number of the substituent groups from four to
two successfully excluded the influence of steric factors and
resulted in a good correlation between the strength of the
hydrogen bond and the strength of MoO or the redox
potential. The contribution from the substituents followed the
order CF3 > t-Bu > CH3 > CPh3 > C(C6H4-4-t-Bu)3 for
Mo(VI) and CF3 > t-Bu > CH3 > C(C6H4-4-t-Bu)3 ≈ CPh3 for
Mo(IV). Rate constants followed approximately the same
order, except for CF3: t-Bu > CH3 > CPh3 ≈ C(C6H4-4-t-Bu)3
≫ CF3. These data indicate that the strong NH···S hydrogen
bond in 5c (R = CF3) effectively stabilized the intermediate and
retarded the trans−cis rearrangement in the proposed
mechanism. DFT calculations supported this consideration.
Consequently, the present data clearly demonstrate the
contribution of the NH···S hydrogen bonds to the properties
and the reactivities of the molybdoenzyme models, and suggest
that the hydrogen bond is one of the important factors in the
fine and delicate tuning of the reactivity of metalloenzymes in
biological systems.
(5) Holm, R. H.; Solomon, E. I.; Majumdar, A.; Tenderholt, A.
Coord. Chem. Rev. 2011, 255, 993−1015.
(6) Ng, V. W. L.; Taylor, M. K.; Young, C. G. Inorg. Chem. 2012, 51,
3202−3211.
(7) Smith, P. D.; Millar, A. J.; Young, C. G.; Ghosh, M.; Basu, P. J.
Am. Chem. Soc. 2000, 122, 9298−9299.
(8) Stiefel, E. I., Ed.; Dithiolene Chemistry Synthesis, Properties, and
Applications; John Wiley & Sons, Inc.: Hoboken, NJ, 2004; Vol. 52.
(9) Sugimoto, H.; Tsukube, H. Chem. Soc. Rev. 2008, 37, 2609−2619.
(10) Yoshinaga, N.; Ueyama, N.; Okamura, T.; Nakamura, A. Chem.
Lett. 1990, 19, 1655−1656.
(11) Ueyama, N.; Yoshinaga, N.; Okamura, T.; Zaima, H.; Nakamura,
A. J. Mol. Catal. 1991, 64, 247−256.
(12) Ueyama, N.; Oku, H.; Kondo, M.; Okamura, T.; Yoshinaga, N.;
Nakamura, A. Inorg. Chem. 1996, 35, 643−650.
(13) Okamura, T.; Takamizawa, S.; Ueyama, N.; Nakamura, A. Inorg.
Chem. 1998, 37, 18−28.
(14) Ueyama, N.; Okamura, T. In Organometallic Conjugation;
Nakamura, A., Ueyama, N., Yamaguchi, K., Eds.; Kodansha: Tokyo,
Japan, 2002; pp 265−306.
(15) Ueyama, N.; Okamura, T.; Nakamura, A. J. Chem.Soc., Chem.
Commun. 1992, 1019−1020.
(16) Ueyama, N.; Yamada, Y.; Okamura, T.; Kimura, S.; Nakamura,
A. Inorg. Chem. 1996, 35, 6473−6484.
(17) Ueyama, N.; Nishikawa, N.; Yamada, Y.; Okamura, T.;
Nakamura, A. J. Am. Chem. Soc. 1996, 118, 12826−12827.
(18) Ueyama, N.; Nishikawa, N.; Yamada, Y.; Okamura, T.; Oka, S.;
Sakurai, H.; Nakamura, A. Inorg. Chem. 1998, 37, 2415−2421.
(19) Ueyama, N.; Okamura, T.; Nakamura, A. J. Am. Chem. Soc.
1992, 114, 8129−8137.
(20) Okamura, T.; Taniuchi, K.; Lee, K.; Yamamoto, H.; Ueyama, N.;
Nakamura, A. Inorg. Chem. 2006, 45, 9374−9380.
(21) Okamura, T.; Ueyama, N.; Nakamura, A.; Ainscough, E. W.;
Brodie, A. M.; Waters, J. M. J. Chem.Soc., Chem. Commun. 1993, 1658−
1659.
(22) Baba, K.; Okamura, T.; Yamamoto, H.; Yamamoto, T.; Ueyama,
N. Inorg. Chem. 2008, 47, 2837−2848.
ASSOCIATED CONTENT
* Supporting Information
X-ray crystallographic data for 1, 3a, 3b, and 3d in CIF format,
crystallographic data (Table S1), molecular structures (Figure
S1), proposed mechanism for 1 (Scheme S1), 1H NMR (Figure
■
S
393
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394

Inorganic Chemistry
Article
(23) Oku, H.; Ueyama, N.; Kondo, M.; Nakamura, A. Inorg. Chem.
1994, 33, 209−216.
(24) Baba, K.; Okamura, T.; Suzuki, C.; Yamamoto, H.; Yamamoto,
T.; Ohama, M.; Ueyama, N. Inorg. Chem. 2006, 45, 884−901.
(25) Okamura, T.; Tatsumi, M.; Omi, Y.; Yamamoto, H.; Onitsuka,
K. Inorg. Chem. 2012, 51, 11688−11697.
(50) Ueyama, N.; Terakawa, T.; Nakata, M.; Nakamura, A. J. Am.
Chem. Soc. 1983, 1983, 7098−7102.
(51) Ohno, R.; Ueyama, N.; Nakamura, A. Inorg. Chim. Acta 1990,
169, 253−255.
(52) Nubel, P. O.; Wilson, S. R.; Brown, T. L. Organometallics 1983,
2, 515−525.
(53) Jin, S.; Nieuwenhuyzen, M.; Wilkins, C. J. J. Chem. Soc., Dalton
Trans. 1992, 2071−2078.
(26) Gibson, H. W.; Lee, S.-H.; Engen, P. T.; Lecavalier, P.; Sze, J.;
Shen, Y. X.; Bheda, M. J. Org. Chem. 1993, 58, 3748−3756.
(27) Hahn., F. E.; Seidel., W. W. Angew. Chem., Int. Ed. Engl. 1995,
34, 2700−2703.
(54) Kai, Y.; Yasuoka, N.; Kasai, N.; Kakudo, M. Bull. Chem. Soc. Jpn.
1972, 45, 3388−3396.
(55) Borup, B.; Streib, W. E.; Caulton, K. G. Inorg. Chem. 1997, 36,
5058−5063.
(28) Boyd, I. W.; Dabce, I. G.; Murray, K. S.; Wedd, A. G. Aust. J.
Chem. 1978, 31, 279−284.
(29) Kondo, M.; Ueyama, N.; Fukuyama, K.; Nakamura, A. Bull.
Chem. Soc. Jpn. 1993, 66, 1391−1396.
(30) Marinescu, L.; Thinggaard, J.; Thomsen, I. B.; Bols, M. J. Org.
Chem. 2003, 68, 9453−9455.
(31) Ueyama, N.; Okamura, T.; Yamada, Y.; Nakamura, A. J. Org.
Chem. 1995, 60, 4893−4899.
(32) Altomare, A.; Burla, M. C.; Camalli, M.; Cascarano, M.;
Giacovazzo, C.; Guagliardi, A.; Polidori, G. J. Appl. Crystallogr. 1994,
27, 435.
(33) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112−122.
(34) 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.; 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. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian03, Revision
E.01; Gaussian, Inc.: Pittsburgh, PA, 2004.
(35) Huzinaga, S., Ed.; Gaussian Basis Sets for Molecular Calculations;
Elsevier: Amsterdam, The Netherlands, 1984.
(36) Yamanaka, S.; Kawamura, T.; Noro, T.; Yamaguchi, K. J. Mol.
Struct. 1994, 310, 185−196.
(37) Baba, K.; Okamura, T.; Yamamoto, H.; Yamamoto, T.; Ohama,
M.; Ueyama, N. Inorg. Chem. 2006, 45, 8365−8371.
(38) Hall, D. W.; Richards, J. H. J. Org. Chem. 1963, 28, 1549−1554.
(39) Newkome, G. R.; Gross, J.; Patri, K. J. Org. Chem. 1997, 62,
3013−3014.
(40) Okamura, T.; Iwamura, T.; Nozaki, K.; Ohno, T.; Yamamoto,
H.; Ueyama, N. Mol. Cryst. Liq. Cryst. 2002, 379, 431−436.
(41) Lieber, E.; R., L. M., Jr.; Rao, C. N. R. Chem. Rev. 1965, 65,
377−384.
(42) Galeazzi, R.; Geremia, S.; Mobbili, G.; Orena, M. Tetrahedron:
Asymmetry 1996, 7, 3573−3584.
(43) Krow, G. R.; Huang, Q.; Lin, G.; Centafont, R. A.; Thomas, A.
M.; Gandla, D.; DeBrosse, C.; Carroll, P. J. J. Org. Chem. 2006, 71,
2090−2098.
(44) Fahr, E.; Neumann, L. Angew. Chem., Int. Ed. Engl. 1965, 4, 595.
(45) Yamaguchi, S.; Uchiuzoh, Y.; Sanada, K. J. Heterocycl. Chem.
1995, 32, 419−423.
(46) Liao, Y.; Venhuis, B. J.; Rodenhuis, N.; Timmerman, W.;
Wikstrom, H. J. Med. Chem. 1999, 42, 2235−2244.
̈
(47) Boyde, S.; Ellis, S. R.; Garner, C. D.; Clegg, W. J. Chem. Soc.,
Chem. Commun. 1986, 1541−1543.
(48) Oku, H.; Ueyama, N.; Nakamura, A. Inorg. Chem. 1995, 34,
3667−3676.
(49) Kassner, R. J.; Yang, W. J. Am. Chem. Soc. 1977, 99, 4351−4355.
394
dx.doi.org/10.1021/ic302149p | Inorg. Chem. 2013, 52, 381−394