Electron exchange involving a sulfur-stabilized ruthenium radical cation

Article abstract of DOI:10.1021/ic700580u

Half-sandwich Ru(II) amine, thiol, and thiolate complexes were prepared and characterized by X-ray crystallography. The thiol and amine complexes react slowly with acetonitrile to give free thiol or amine and the acetonitrile complex. With the thiol complex, the reaction is dissociative. The thiolate complex has been oxidized to its Ru(III) radical cation and the solution EPR spectrum of that radical cation recorded. Cobaltocene reduces the thiol complex to the thiolate complex. The ¹H and ³¹P NMR signals of the thiolate complex in acetonitrile become very broad whenever the thiolate and thiol complexes are present simultaneously. The line broadening is primarily due to electron exchange between the thiolate complex and its radical cation; the latter is generated by an unfavorable redox equilibrium between the thiol and thiolate complexes. Pyramidal inversion of sulfur in the thiol complex is fast at room temperature but slow at lower temperatures; major and minor conformers of the thiol complex were observed by ³¹P NMR at -98°C in CD ₂Cl₂.

Full text of DOI:10.1021/ic700580u

Inorg. Chem. 2007, 46, 5805−5812
Electron Exchange Involving a Sulfur-Stabilized Ruthenium Radical
Cation
Anthony P. Shaw, Bradford L. Ryland, Jack R. Norton,* Daniela Buccella, and Alberto Moscatelli
Department of Chemistry, Columbia UniVersity, New York, New York 10027
Received March 27, 2007
Half-sandwich Ru(II) amine, thiol, and thiolate complexes were prepared and characterized by X-ray crystallography.
The thiol and amine complexes react slowly with acetonitrile to give free thiol or amine and the acetonitrile complex.
With the thiol complex, the reaction is dissociative. The thiolate complex has been oxidized to its Ru(III) radical
cation and the solution EPR spectrum of that radical cation recorded. Cobaltocene reduces the thiol complex to the
1
thiolate complex. The H and 31P NMR signals of the thiolate complex in acetonitrile become very broad whenever
the thiolate and thiol complexes are present simultaneously. The line broadening is primarily due to electron exchange
between the thiolate complex and its radical cation; the latter is generated by an unfavorable redox equilibrium
between the thiol and thiolate complexes. Pyramidal inversion of sulfur in the thiol complex is fast at room temperature
but slow at lower temperatures; major and minor conformers of the thiol complex were observed by 31P NMR at
−98 °C in CD2Cl2.
Introduction
Chiral nitrogen and sulfur ligands are potential resolving
agents for chiral-at-metal transition-metal complexes, and
ligand lability is important in catalytic ionic hydrogenations.1
We set out to assess the differences in structure and reactivity
of Ru(II) complexes with amine, thiol, and thiolate ligands
(Figure 1) and to study their substitution kinetics and redox
chemistry.
Transition-metal thiol complexes are often unstable be-
cause coordinated thiols can be acidic, and some of their
complexes are easily oxidized.2 Treichel found the iron thiol
complex [CpFe(CO)2SHPh]BF4 to be a strong acid,3 while
Puerta found that electron rich half-sandwich Ru(II) thiol
complexes are easily oxidized.4
Figure 1.
due to electron exchange between the thiolate complex and
its radical cation.
Experimental Section
We have observed line broadening in the NMR spectra
of mixtures of the thiolate and thiol complexes. By conduct-
ing low-temperature NMR experiments and studying the
redox chemistry of the thiol and thiolate complexes 1 and
3, we have determined that the line broadening is largely
General Procedures. All air-sensitive compounds were prepared
and handled under a N2/Ar atmosphere using standard Schlenk and
inert-atmosphere box techniques. Hexanes were deoxygenated and
dried by two successive activated alumina columns under argon.
Benzene, toluene, ether, and THF were distilled from Na and
benzophenone under a nitrogen atmosphere. HPLC grade acetoni-
trile (J. T. Baker) was sparged with N2 and dried over 4 Å molecular
sieves. Water was deionized with a Barnstead NANOpure water
system. CD2Cl2 was dried over CaH2, degassed by three freeze-
pump-thaw cycles, and then purified by vacuum transfer at room
temperature. CDCl3 was degassed by three freeze-pump-thaw
cycles and dried over 4 Å molecular sieves. UV-visible spectra
were obtained on a HP 8453 spectrophotometer with a 1 cm quartz
cell.
* To whom correspondence should be addressed. E-mail: jrn11@
columbia.edu.
(1) (a) Guan, H.; Iimura, M.; Magee, M. P.; Norton, J. R.; Zhu, G. J. Am.
Chem. Soc. 2005, 127, 7805-7814. (b) Voges, M. H.; Bullock, R.
M. J. Chem. Soc., Dalton Trans. 2002, 759-770.
(2) Treichel, P. M.; Crane, R. A.; Haller, K. N. J. Organomet. Chem.
1991, 401, 173-180.
(3) Treichel, P. M.; Rosenhein, L. D. Inorg. Chem. 1981, 20, 942-944.
(4) Coto, A.; de los R´ıos, I.; Tenorio, M. J.; Puerta, M. C.; Valerga, P. J.
Chem. Soc., Dalton Trans. 1999, 4309-4314.
10.1021/ic700580u CCC: $37.00
Published on Web 06/15/2007
© 2007 American Chemical Society
Inorganic Chemistry, Vol. 46, No. 14, 2007 5805

Shaw et al.
Cyclic voltammetry was performed with a BAS CV-50W
potentiostat. The supporting electrolyte for all solutions was 0.10
M [Bu4N]PF6 in acetonitrile. The cell consisted of a platinum (1.6
mm diameter) or glassy carbon disk (5.0 mm diameter) working
electrode, a silver wire reference electrode (0.01 M AgNO3 + 0.10
M [Bu4N]PF6), and a platinum wire auxiliary electrode (0.10 M
[Bu4N]PF6). Fc/Fc+ (0.001 M) was used as an external reference
and was found to be +87 mV with respect to our reference
electrode. All samples were prepared under a N2/Ar atmosphere
and sparged with Ar before analysis. Analyte concentrations were
0.001 M, and all scans were recorded at 100 mV/s.
(300 MHz, CDCl3): δ 2.33-2.47 (m, Ph2PCH2CH2PPh2, 2H),
2.55-2.70 (m, Ph2PCH2CH2PPh2, 2H), 4.54 (s, Cp, 5H), 7.10-
7.19 (m, Ar, 4H), 7.24-7.43 (m, Ar, 12H), 7.82-7.91 (m, Ar, 4H).
31P {1H} NMR (121.5 MHz, CDCl3): δ 79.90 (s).
[CpRu(dppe)(R-methylbenzylthiol)][BPh4] (1). Sodium tet-
raphenylborate (445 mg, 1.30 mmol) and racemic R-methyl-
benzylthiol10 (180 µL, 1.30 mmol) were refluxed with CpRu(dppe)-
Cl (600 mg, 1.0 mmol) in 50 mL of methanol for 90 min. The
precipitate was filtered, washed with water and methanol, and dried
under vacuum to give the yellow-green product (910 mg, 0.89
1
mmol) in 89% yield. H NMR (400 MHz, CD2Cl2): δ 1.06 (d,
Variable Temperature NMR Studies. Probe temperatures
below room temperature were calibrated with a Wilmad chemical
shift thermometer (99.97% methanol + 0.03% HCl).5 Temperatures
above room temperature were calibrated with an ethylene glycol
chemical shift thermometer.6
CH3, JH-H ) 6.8 Hz, 3H), 1.85-1.92 (m, SH, 1H), 2.30-2.64 (m,
Ph2PCH2CH2PPh2, 4H), 2.71-2.81 (m, CH, 1H), 4.59 (s, Cp, 5H),
6.82-6.90 (t, Ar, 6H), 6.97-7.04 (t, Ar, 8H), 7.15-7.22 (t, Ar,
2H), 7.26-7.36 (m, Ar, 12H), 7.37-7.67 (m, Ar, 17H). 31P {1H}
NMR (161.9 MHz, CD2Cl2): AB pattern, δ 78.28 (d, JP-P ) 25.8
Hz), 79.29 (d, JP-P ) 25.8 Hz). Anal. Calcd for C63H59BP2SRu:
C, 74.04; H, 5.82; S, 3.14. Found: C, 74.06; H, 5.84; S, 3.04.
[CpRu(dppe)(R-methylbenzylamine)][BPh4] (2). Sodium tet-
raphenylborate (161 mg, 0.47 mmol) and racemic R-methyl-
benzylamine (61 µL, 0.47 mmol) were refluxed with CpRu(dppe)-
Cl (235 mg, 0.39 mmol) in 20 mL of methanol for 60 min. The
precipitate was filtered, washed with water and methanol, and dried
under vacuum to give the yellow product (270 mg, 0.27 mmol) in
Ligand Substitution Kinetics in CD3CN. The ruthenium thiol
or amine complex (0.005 mmol) and the internal standard anisole
(1.0 µL) were added to 0.75 mL of CD3CN in an NMR tube under
an inert atmosphere. The tube was sealed and inserted into a
preheated NMR probe calibrated at 341 K. The disappearance of
the amine complex was monitored via the height of the Cp singlet
(δ 4.51). The disappearance of the thiol complex was monitored
via the height of the coordinated thiol methyl doublet (δ 1.06).
Ligand Substitution Kinetics in CD3NO2. The ruthenium thiol
complex (0.012 mmol) and 4 equiv of CH3CN were dissolved in
1.0 mL of CD3NO2. The tube was sealed and inserted into a
preheated NMR probe calibrated at 341 K. The reaction was
monitored via the ratio of the height of the free thiol doublet (δ
1.66) to that of the tetraphenylborate resonance (δ 6.84). The
reaction was repeated with 16 equiv of CH3CN.
RuCl2(PPh3)3 was prepared in high yield on a large scale by a
procedure derived from published methods.7 RuCl3‚3H2O (15.0 g,
57.0 mmol) was refluxed with PPh3 (90.0 g, 343.5 mmol) in 3.25
L of deoxygenated anhydrous methanol for 2 h and 30 min. The
liquid was decanted, and the remaining black solid loaded on a
Schlenk frit, washed with ether (3 × 100 mL), and dried under
vacuum overnight to give the product (52.0 g, 54.2 mmol) in 95%
yield. 31P {1H} NMR (161.9 MHz, CDCl3): δ -3.96 (s). 13C{1H}
NMR (75 MHz, CDCl3): δ 127.47 (m), 128.68 (m), 129.37 (s),
132.17 (m), 133.92 (m), 135.36 (m).
69% yield. 1H NMR (400 MHz, CD3CN): δ 0.61 (d, CH3, JH-H
)
6.4 Hz, 3H), 1.48 (br, NH2, 1H), 1.62 (br, NH2, 1H), 2.21-2.65
(m, Ph2PCH2CH2PPh2, 3H), 2.72-2.90 (m, Ph2PCH2CH2PPh2, 1H),
3.20-3.30 (m, CH, 1H), 4.51 (s, Cp, 5H), 6.59-6.64 (m, Ar, 2H),
6.81-6.87 (m, Ar, 4H), 6.96-7.02 (m, Ar, 8H), 7.22-7.30 (m,
Ar, 13H), 7.32-7.39 (m, Ar, 2H), 7.40-7.45 (m, Ar, 3H), 7.45-
7.49 (m, Ar, 3H), 7.51-7.58 (m, Ar, 3H), 7.58-7.66 (m, Ar, 3H),
7.71-7.78 (m, Ar, 2H), 7.84-7.90 (m, Ar, 2H). 31P {1H} NMR
(161.9 MHz, CD3CN): AB pattern, δ 81.96 (d, JP-P ) 25.4 Hz),
82.88 (d, JP-P ) 25.4 Hz). Anal. Calcd for C63H60NBP2Ru: C,
75.29; H, 6.02; N, 1.39. Found: C, 75.34; H, 5.92; N, 1.49.
CpRu(dppe)(R-methylbenzylthiolate) (3). The thiol complex
1 (440 mg, 0.43 mmol) was stirred in 10 mL of CH3CN, and NEt3
(0.88 mL, 6.45 mmol) was added in one portion, giving a yellow
solution which was placed in a refrigerator at -24 °C overnight.
The solvent was carefully decanted from the orange crystals that
formed. The crystals were washed with CH3CN (2 × 0.5 mL) and
dried under vacuum to give the analytically pure thiolate complex
(285 mg, 0.41 mmol) in 94% yield. 1H NMR (300 MHz, CD3CN):
δ 0.78 (d, CH3, JH-H ) 6.6 Hz, 3H), 2.05-2.23 (m, Ph2PCH2CH2-
PPh2, 2H), 2.28-2.50 (m, Ph2PCH2CH2PPh2, 1H), 2.62-2.85 (m,
Ph2PCH2CH2PPh2, 1H), 3.10 (q, br, CH, 1H), 4.43 (s, Cp, 5H),
7.00-7.34 (m, Ar, 15H), 7.39-7.51 (br, Ar, 6H), 7.78-7.91 (m,
Ar, 4H). 31P {1H} NMR (121.5 MHz, CD3CN): AB pattern, δ 81.53
(d, JP-P ) 26.8 Hz), 82.83 (d, JP-P ) 26.8 Hz). The NMR spectra
are often broad but sharpened by the addition of a small amount of
Cp2Co. FAB+ MS: m/z 702.13 [M + 1]+. Anal. Calcd for C39H38P2-
SRu: C, 66.74; H, 5.46; S, 4.57. Found: C, 66.59; H, 5.62; S,
4.40.
CpRu(dppe)Cl8 was prepared by boiling a solution of RuCl2-
(PPh3)3 (6.0 g, 6.27 mmol) and sodium cyclopentadienide9 (0.59
g, 6.69 mmol) in 100 mL of dry THF for 3 h. An orange precipitate
3
formed after evaporation of /4 of the solvent and addition of 100
mL of hexanes, which was washed with hexanes and dried under
vacuum. The resulting solid was boiled with 1,2-bis(diphenylphos-
phino)ethane (2.50 g, 6.28 mmol) in 85 mL of toluene for 15 h.
The reaction mixture was filtered, and the filtrate was loaded onto
the top of a silica column (8 cm × 2 cm diameter). Excess
phosphines were eluted with benzene, and the product was eluted
with a 1:1 benzene-ether mixture. Evaporation of the solvent gave
1
the yellow product (2.16 g, 3.60 mmol) in 57% yield. H NMR
[CpRu(dppe)(R-methylbenzylthiolate)][PF6] (5-PF6) was pre-
pared as a 0.05 M solution in CH2Cl2 by treating 3 with an
equimolar quantity of [NO]PF6. Gas evolution appeared to be
complete within 1 h, but residual gas was removed by four freeze-
pump-thaw cycles. The X-band EPR of a 5 × 10-4 M sample
was performed on a Bruker EMX EPR spectrometer with a TE102
rectangular cavity: microwave frequency 9.731 GHz; microwave
(5) Van Geet, A. L. Anal. Chem. 1970, 42, 679-680.
(6) Raiford, D. S.; Fisk, C. L.; Becker, E. D. Anal. Chem. 1979, 51, 2050-
2051.
(7) (a) Ortiz-Frade, L. A.; Ruiz-Ram´ırez, L.; Gonza´lez, I.; Mar´ın-Becerra,
A.; Alcarazo, M.; Alvarado-Rodriguez, J. G.; Moreno-Esparza, R.
Inorg. Chem. 2003, 42, 1825-1834. (b) Hallman, P. S.; Stephenson,
T. A.; Wilkinson, G. Inorg. Synth. 1970, 12, 237-240.
(8) (a) Ashby, G. S.; Bruce, M. I.; Tomkins, I. B.; Wallis, R. C. Aust. J.
Chem. 1979, 32, 1003-1016. (b) Bruce, M. I.; Windsor, N. J. Aust.
J. Chem. 1977, 30, 1601-1604.
(9) Panda, T. K.; Gamer, M. T.; Roesky, P. W. Organometallics 2003,
22, 877-878.
(10) Stratmann, O.; Kaiser, B.; Fro¨hlich, R.; Meyer, O.; Hoppe, D. Chem.
Eur. J. 2001, 7, 423-435.
5806 Inorganic Chemistry, Vol. 46, No. 14, 2007

Electron Exchange InWolWing a S-Stabilized Ru Radical Cation
Table 1. Summary of Crystallographic Data
1
2
3
empirical formula
formula weight
crystal shape
crystal size (mm)
temp (K)
crystal system
space group
a (Å)
C63H59BP2RuS
1021.98
yellow plate
0.30 × 0.10 × 0.03
125(2)
monoclinic
P21/c
11.4497(6)
23.7658(13)
18.9148(10)
90
C63H60BNP2Ru
1004.94
yellow needle
0.30 × 0.15 × 0.15
125(2)
monoclinic
P21/c
11.5351(11)
23.674(2)
18.6041(17)
90
C39H38P2RuS
701.76
orange block
0.20 × 0.20 × 0.20
243(2)
monoclinic
P21/c
12.1405(5)
20.7461(8)
14.1678(5)
90
b (Å)
c (Å)
R (deg)
â (deg)
101.4350(10)
90
5044.8(5)
4
100.6360(10)
90
4993.1(8)
4
109.8180(10)
90
3357.1(2)
4
γ (deg)
V (Å3)
Z
d
calc (g/cm3)
1.346
1.337
1.388
λ (Mo KR) (Å)
0.71073
0.457
81097
0.71073
0.420
88343
0.71073
0.651
23868
µ (mm-1
)
data collected
unique data
15473
15473/2/637
1.008
18388
18388/3/648
1.024
7888
7888/0/388
1.014
data/restraints/params
GOF on F2
R1, wR2 [I >2σ(I)]
R1, wR2 (all data)
0.0511, 0.1040
0.0983, 0.1235
0.0337, 0.0796
0.0507, 0.0878
0.0342, 0.0747
0.0590, 0.0828
power 63.616 mW; modulation amplitude 4.00 G; center field 3350
G. Simulation was performed with Bruker WINEPR SimFonia
version 1.0 software.
We thus prepared compound 1 in high yield by treating
CpRu(dppe)Cl with rac-R-methylbenzylthiol in the presence
of NaBPh4 in refluxing methanol (eq 1). Treichel has
prepared cationic ruthenium tert-butyl thiol complexes under
similar conditions.2 The racemic amine complex 2 was
obtained analogously (eq 2).
[CpRu(dppe)CD3CN][BPh4]. The thiol complex 1 (0.01 mmol)
was dissolved in 0.75 mL of CD3CN and heated to 75 °C for 2 h
and 30 min. The 1H NMR spectrum showed free thiol and complete
conversion to the CD3CN complex. 1H NMR (300 MHz, CD3CN):
δ 2.49-2.64 (m, Ph2PCH2CH2PPh2, 4H), 4.72 (s, Cp, 5H), 6.80-
6.88 (m, Ar, 4H), 6.94-7.03 (m, Ar, 8H), 7.20-7.57 (m, Ar, 24H),
7.70-7.82 (m, Ar, 4H). 31P {1H} NMR (121.5 MHz, CD3CN): δ
79.33 (s). The addition of 1 equiv of Cp2Co left the spectra
unchanged.
X-ray Structure Determinations. Crystals of 1 and 2 were
grown by slow diffusion of a layer of methanol into concentrated
CH2Cl2 solutions. Crystals of 3 were grown by slow evaporation
of a concentrated CD3CN solution. The hydrogen atom on the sulfur
of 1 was not found. The thiol ligand of 1 was disordered, and the
minor component was refined isotropically. The amine ligand of 2
was also disordered. The minor component of the disordered amine
group was refined isotropically, and the protons on nitrogen were
placed in idealized positions; the protons on the nitrogen of the
major component were located and refined.
For 1 and 2, data were collected on a Bruker Apex II
diffractometer; for 3, data were collected on a Bruker P4 diffrac-
tometer. The structures were solved using direct methods and
standard difference map techniques and refined by full-matrix least-
squares procedures using SHELXTL version 5.10 software. Hy-
drogen atoms on carbon were included in calculated positions.
Compounds 1 and 2 are yellow-green and yellow, respec-
tively. They may be handled in air when dry and are
moderately soluble in CH3CN and CH2Cl2. Complexes 1 and
2 both display an AB pattern in their 31P NMR spectra; the
two phosphorus atoms are diastereotopic because of the chiral
thiol or amine ligand. X-ray analysis of crystals of 1 and 2
(Table 1, Figures 2 and 3) showed that, in both structures,
the phenyl group of the thiol or amine ligand is oriented
toward the Cp ring and away from the phosphine phenyl
groups. The Ru-S bond length of 2.3412(13) Å in 1 is com-
parable to that of 2.369(2) Å in [CpRu(PPh3)2(C6H5CH2CH2-
SH)][BF4]13 and 2.377(2) Å in [CpRu(PPh3)2(n-PrSH)][BF4]14
Ligand Substitution Kinetics. The thiol compex 1 and
the amine complex 2 form [CpRu(dppe)CD3CN][BPh4] over
the course of several days at room temperature in CD3CN.
Results and Discussion
Thiol and Amine Complexes. Halide dissociation from
the complexes CpRu(PR3)2Cl is known to occur readily in
polar solvents (methanol, acetonitrile, and dimethylsulfoxide)
and is promoted by the presence of a suitable halide acceptor
+
(Na+ or NH4 ).11,12 Cationic complexes [CpRu(PR3)2L]+
form in the presence of a coordinating ligand or solvent.8a
(13) Park, H.; Minick, D.; Draganjac, M.; Cordes, A. W.; Hallford, R. L.;
Eggleton, G. Inorg. Chim. Acta 1993, 204, 195-198.
(14) Amarasekera, J.; Rauchfuss, T. B. Inorg. Chem. 1989, 28, 3875-
3883.
(11) Treichel, P. M.; Komar, D. A.; Vincenti, P. J. Inorg. Chim. Acta 1984,
88, 151-152.
(12) Treichel, P. M.; Vincenti, P. J. Inorg. Chem. 1985, 24, 228-230.
Inorganic Chemistry, Vol. 46, No. 14, 2007 5807

Shaw et al.
Figure 4. Molecular structure of 3 (20% probability level). Hydrogen
atoms omitted for clarity. Selected distances (Å) and angles (deg): Ru-S
2.4019(7), Ru-P1 2.2613(6), Ru-P2 2.2483(6), Ru‚‚‚C8 3.50, Ru-S-
C8 110.80(9), P1-Ru-P2 82.65(2), P1-Ru-S 82.87(2), P2-Ru-S 87.86-
(3).
Figure 2. Molecular structure of 1 (20% probability level). Counterion
and hydrogen atoms omitted for clarity. The hydrogen atom on sulfur was
not found. Selected distances (Å) and angles (deg): Ru-S 2.3412(13), Ru-
P1 2.2911(7), Ru-P2 2.2842(7), Ru‚‚‚C8 3.60, Ru-S-C8 118.79(14), P1-
Ru-P2 83.79(3), P1-Ru-S 88.20(4), P2-Ru-S 86.22(4).
several by treating CpRu(PPh3)2Cl with thiolate anions in
THF.15 The orange thiolate complex 3 is obtained by treating
1 with excess NEt3 in acetonitrile (eq 4). As with complexes
1 and 2, the 31P NMR spectrum of 3 displays an AB pattern.
Crystals of 3 are readily formed by cooling or slowly
evaporating concentrated acetonitrile solutions. Slow evapo-
ration of a CD3CN solution gave a crystal suitable for X-ray
analysis (Table 1, Figure 4). Like the thiol and amine
complexes, the phenyl group of the thiolate is oriented away
from the phosphine phenyl groups. The Ru-S bond length
of 2.4019(7) Å in 3 is similar to the 2.420(4) Å found in
CpRu(dippe)SPh.4
Figure 3. Molecular structure of 2 (20% probability level). Counterion
and hydrogen atoms except those on nitrogen omitted for clarity. Selected
distances (Å) and angles (deg): Ru-N 2.211(3), Ru-P1 2.2842(4), Ru-
P2 2.2879(4), Ru‚‚‚C8 3.34, Ru-N-C8 128.5(3), P1-Ru-P2 83.445(14),
P1-Ru-N 85.7(2), P2-Ru-N 91.84(10).
Compound 3 is oxidized by [NO]PF6 to the Ru(III) radical
cation 5 (eq 5). Such half-sandwich 17-electron radical
cations of iron16,17 and ruthenium18,19 are well-known. The
EPR spectrum of 5-PF6 in CH2Cl2 (Figure 5) shows hyperfine
coupling to two 31P and 99/101Ru, although the spectrum is
broad. Broad solution EPR spectra have been observed by
D´ıaz for sulfur-stabilized iron radical cations, albeit without
hyperfine splitting.17a
1
Monitoring these reactions by H NMR spectroscopy gave
first-order rate constants (at 68 °C) of 5.2(2) × 10-4 s-1 for
1 and 5.5(1) × 10-4 s-1 for 2srate constants approximately
40 times greater than that determined by Treichel and
Vincenti for the solvolysis of CpRu(dppe)Cl in CD3CN under
similar conditions.12 In the polar but less coordinating solvent
CD3NO2, the rate proved to be independent of [CH3CN], so
the reaction of the thiol complex with CH3CN is a dissocia-
tive first-order process (eq 3). At 68 °C k1 is 2.6(1) × 10-4
s-1.
(15) Shaver, A.; Plouffe, P.-Y.; Bird, P.; Livingstone, E. Inorg. Chem. 1990,
29, 1826-1832.
(16) Iron radical cations: (a) Tilset, M.; Fjeldahl, I.; Hamon, J.-R.; Hamon,
P.; Toupet, L.; Saillard, J.-Y.; Costuas, K.; Haynes, A. J. Am. Chem.
Soc. 2001, 123, 9984-10000. (b) Hamon, P.; Toupet, L.; Hamon, J.-
R.; Lapinte, C. Organometallics 1992, 11, 1429-1431. (c) Treichel,
P. M.; Komar, D. A. J. Organomet. Chem. 1981, 206, 77-88.
(17) Sulfur-stabilized iron radical cations: (a) D´ıaz, C. Bol. Soc. Chil. Qu´ım.
1999, 44, 315-320. (b) D´ıaz, C.; Leal, C.; Yutronic, N. J. Organomet.
Chem. 1996, 516, 59-64. (c) Treichel, P. M.; Rosenhein, L. D.;
Schmidt, M. S. Inorg. Chem. 1983, 22, 3960-3965. (d) Treichel, P.
M.; Molzahn, D. C.; Wagner, K. P. J. Organomet. Chem. 1979, 174,
191-197.
(18) Ruthenium radical cations: (a) Guan, H.; Saddoughi, S. A.; Shaw, A.
P.; Norton, J. R. Organometallics 2005, 24, 6358-6364. (b) Hembre,
R. T.; McQueen, J. S. Angew. Chem., Int. Ed., Engl. 1997, 36, 65-
67. (c) Hembre, R. T.; McQueen, J. S.; Day, V. W. J. Am. Chem.
Soc. 1996, 118, 798-803. (d) Smith, K.-T.; Rømming, C.; Tilset, M.
J. Am. Chem. Soc. 1993, 115, 8681-8689.
(19) Sulfur-stabilized ruthenium radical cations: See refs 2 and 4, also:
Treichel, P. M.; Schmidt, M. S.; Crane, R. A. Inorg. Chem. 1991, 30,
379-381.
Deprotonation of the Thiol Complex 1 and Oxidation
of the Resulting Thiolate Complex 3. Thiolate complexes
CpRu(PPh3)2SR (R ) alkyl) are known; Shaver prepared
5808 Inorganic Chemistry, Vol. 46, No. 14, 2007

Electron Exchange InWolWing a S-Stabilized Ru Radical Cation
Figure 5. EPR spectrum (9.731 GHz, g ) 2.075) of 5-PF6 in CH2Cl2 at
room temperature (top). Simulation (bottom) used aRu ) 16.0 G; aP ) 21.4
G; aS ) 15.0 G; line broadening 13.70 G.
Figure 7. Cyclic voltammograms (platinum disk working electrode) of 3
(A) and 1 (B) in CH3CN at 100 mV/s. Potentials (mV) vs Fc/Fc+ in CH3-
CN.
Table 2. Redox Potentialsa
compound
potential
[-1429]
-544
(-1334, ∆73)
(-1912, ∆90)
process
1
3
reduction
oxidation
oxidation
oxidation
Cp2Co
Cp*2Co
a All potentials (mV) in CH3CN at 100 mV/s vs Fc/Fc+. A platinum
disk working electrode was used. Brackets indicate an irreversible peak.
Parentheses enclose E1/2 and Epa - Epc for quasireversible processes. Plain
text indicates E1/2 for a reversible process where Epa - Epc ) 59 mV and
Ia/Ic ) 1.
(Table 2), Cp2Co should reduce the radical cation 5, and
that prediction has been confirmed experimentally.
Figure 6. Overlaid UV-visible spectra of 3 (2 × 10-4 M) and 5-PF6 (2 ×
In contrast, the thiol complex 1 is irreVersibly reduced
(Figure 7). Subsequent scans show peaks for the 3/5 couple,
implying that 3 and presumably H2 are the products of the
irreversible reduction of 1. A 19-electron complex 4 is drawn
in eq 6 as an intermediate in a stepwise mechanism, although
there is no evidence precluding a ring slip to η3 or the
dissociation of one arm of the chelating diphosphine.20
10-4 M) in CH2Cl2.
Concentrated deep purple CH2Cl2 solutions of 5-PF6 stored
under an inert atmosphere in ambient light gradually
decompose (turning brown) over the course of several days
at room temperature.
(20) Several half-sandwich cyclopentadienyl complexes of group 8 metals
have been reported that are ostensibly 19-electron, but in no case has
their structure been determined by X-ray crystallography; it is possible
that they are really 17-electron complexes. (a) Hamon, P.; Hamon,
J.-R.; Lapinte, C. J. Chem. Soc., Chem. Commun. 1992, 1602-1603.
(b) Ruiz, J.; Lacoste, M.; Astruc, D. J. Am. Chem. Soc. 1990, 112,
5471-5483. (c) Aase, T.; Tilset, M.; Parker, V. D. J. Am. Chem. Soc.
1990, 112, 4974-4975. (d) Ruiz, J.; Guerchais, V.; Astruc, D. J. Chem.
Soc., Chem. Commun. 1989, 812-813. (e) Lapinte, C.; Catheline, D.;
Astruc, D. Organometallics 1984, 3, 817-819.
Dilute solutions of 3 and 5-PF6 are yellow and purple,
respectively, and may be distinguished by their UV-visible
spectra (Figure 6).
Cyclic Voltammetry Studies of 3 and 1. In CH3CN the
thiolate complex 3 is reversibly oxidized by one electron,
presumably to the same radical cation 5 (the cyclic volta-
mmogram is shown in Figure 7). From the 3/5 potential
Inorganic Chemistry, Vol. 46, No. 14, 2007 5809

Shaw et al.
Scheme 1
Scheme 2
Attempted Deprotonation and Reduction of the Amine
Complex 2. We have not been able to deprotonate the amine
complex 2 in acetonitrile, even with strong bases such as
DBN (1,5-diazabicyclo[4.3.0]non-5-ene) or TMG (1,1,3,3-
tetramethylguanidine). This is not surprising given that amide
complexes comparable to the one that would result are
extremely basic.25 Complex 2 is not reduced by Cp2Co, but
is by Cp*2Co, giving unidentified products and free amine.
NMR Line Broadening of the Thiolate Complex 3. We
attempted (unsuccessfully) to estimate the pKa of the thiol
complex 1 in CD3CN by treating it with various amines and
Organic thiols are known to undergo S-H bond cleavage
upon reduction at transition-metal electrodes.21 At a platinum
electrode, it is likely that 1 proceeds directly to 3 as shown
in eq 7.
We reduced 1 at a glassy carbon electrode (as a reviewer
suggested) in an attempt to observe a return oxidation peak
(4 f 1). Except for a shift in the irreversible cathodic peak
to -2110 mV (vs Fc/Fc+), the voltammogram was the same
as that obtained with a platinum electrode, and the thiolate
complex 3 was observed on subsequent scans as in
Figure 7.
1
monitoring the resulting 1/3 equilibria by H NMR. We
expected both NEt3 and DBN to deprotonate 1 (eq 9),
given that NEt3 was used for the synthesis of 3 in high yield
from 1.
D´ıaz has observed complexity in the electrochemistry of
[CpFe(dppe)SR][PF6] arising from the involvement of dis-
ulfides.22 We have not observed any analogous complications
in our ruthenium system; the 3/5 couple is straightforwardly
reversible.
However, when 1 was treated with a single equiv of NEt3
or DBN in an NMR tube, the 1H NMR peaks of the resulting
3 were broad, while the peaks of the resulting Et3NH+ or
DBNH+ were sharp. In a 31P NMR spectrum, 3 appeared
only as a broad featureless bump, although cooling the tube
to -51 °C resulted in the appearance of the two doublets
(JP-P ) 26.4 Hz, δ 82.9 and 80.1) of 3.26
Chemical Reduction of the Thiol Complex 1. Although
the irreversible cathodic peak for 1 (Table 2) occurs at a
more negative potential than the Cp2Co/Cp2Co+ couple, Cp2-
Co cleanly reduces 1 to 3 (eq 8). Only 3 and Cp2Co+ are
1
We suspected this line broadening arose from traces of 1.
The addition of 6 equiv of DBN to 1 (ensuring that
essentially no 1 remained) resulted in a sharp spectrum of
3. The addition of a small amount of Cp2Co to an equimolar
mixture of 1 and NEt3 eliminated broadening of the peaks
observed by the time the H NMR spectrum of a 1/Cp2Co
solution can be recorded (5-10 min).
1
of 3 in the H NMR and 31P NMR spectra, causing us to
suspect electron exchange between 3 and its radical cation
5. (We have experience with the effects of Ru(II)/Ru(III)
An analogous reaction of Cp2Co is that with strong acids,
giving Cp2Co+ and hydrogen. Koelle has shown that Cp2-
CoH+ is involved.23,24 The most plausible mechanism
consistent with his pulse radiolysis experiments is shown in
Scheme 1.
The reduction in eq 8 may proceed by electron transfer
from Cp2Co to 1, followed by disproportionation of 4
(Scheme 2). It is also possible that Cp2CoH+ is generated
by H• transfer from 4 to Cp2Co+ after or concurrent with
the initial reduction and then evolves hydrogen via the last
two equations in Scheme 1.
electron exchange on NMR spectra.18a
)
To generate a mixture of 3 and 5 uncomplicated by the
presence of 1, we added a small (substoichiometric) amount
of [NO]PF6 to a saturated solution of 3 in CD3CN; its Cp
1H NMR singlet and thiolate doublet broadened into the
baseline. The addition of Cp2Co made the spectra of 3 sharp
again, presumably by reducing the 5. Thus the broadening
aboVe is caused by fast electron exchange between 3 and 5.
Electron Transfer from the Thiolate Complex 3 to the
Thiol Complex 1. Treating 1 with half an equivalent of NEt3
gave a mixture of 1 and 3 that we used for a variable
temperature NMR study (eq 10).
(21) Bukhtiarov, A. V.; Mikheev, V. V.; Lebedev, A. V.; Tomilov, A. P.;
Kuz’min, O. V. Zh. Obshch Khim. 1988, 58, 684-692. English
translation: J. Gen. Chem. USSR. 1988, 58, 605-612.
(22) D´ıaz, C.; Araya, E.; Santa Ana, M. A. Polyhedron 1998, 17, 2225-
2230.
At room temperature, the 31P NMR signal of the thiolate
complex appeared as a broad featureless lump in the baseline,
(23) Koelle, U.; Infelta, P. P.; Gra¨tzel, M. Inorg. Chem. 1988, 27, 879-
883.
(24) The third equation in Scheme 1 reflects the ease with which a hydride
ligand can serve as the kinetic site for protonation of a hydride
complex: Papish, E. T.; Magee, M. P.; Norton, J. R. Protonation of
Transition Metal Hydrides to Give Dihydrogen Complexes: Mecha-
nistic Implications and Catalytic Applications. In Recent AdVances in
Hydride Chemistry; Poli, R., Ed.; Elsevier: New York, 2001; pp 39-
74.
(25) (a) Joslin, F. L.; Johnson, M. P.; Mague, J. T.; Roundhill, D. M.
Organometallics 1991, 10, 2781-2794. (b) Kaplan, A. W.; Ritter, J.
C. M.; Bergman, R. G. J. Am. Chem. Soc. 1998, 120, 6828-6829.
(26) At room temperature in CD3CN, 3 appears as an AB pattern centered
about δ 82.2 (∆ν ) 1.3 ppm, JP-P ) 26.8 Hz). The temperature
dependence of the 31P NMR spectrum was confirmed by obtaining
low-temperature spectra of a genuine sample.
5810 Inorganic Chemistry, Vol. 46, No. 14, 2007

Electron Exchange InWolWing a S-Stabilized Ru Radical Cation
while the AB pattern of the thiol complex was broadened
but still clearly visible. As the temperature was lowered, the
peaks of the thiolate complex sharpened (Figure 8). This
indicated the operation of a process that selectively broadened
the peaks of the thiolate complex.
We believe that 4 and 5 are generated by the following
redox reaction (eq 11).
We also believe that eq 11 is an equilibrium, albeit an
unfavorable one, that is established more rapidly than the
disproportionation of 4, as the extent of line broadening of
3 does not change appreciably over several days.27 Our
electrochemical studies imply that this equilibrium in eq 11
lies substantially to the left. Even though the amount of
radical cation 5 produced must be exceedingly small, the
resulting electron exchange (eq 12) is extremely fast,28 and
very little 5 is required to substantially broaden the NMR
peaks of 3.
Figure 8. Variable temperature 31P NMR spectra (121.5 MHz) of the
mixture in eq 10. The small singlet at δ 79.33 is due to a small amount of
[CpRu(dppe)CD3CN][BPh4].
Our NMR experiments are consistent with eq 12 as the
primary reason for the line broadening of the NMR signals
of 3 and with slower exchange processes29 broadening 1 to
a much lesser degree.
The electron exchange in eq 12 is expected to be rapid.
Low-spin octahedral Ru(II) and Ru(III) differ by a nonbond-
ing t2g electron, so the electron transfer is accompanied by
minimal changes in metal-ligand distances. The electron
transfer in eq 11 is expected to be much slower because it
requires movement of an antibonding eg electron and thus
significant changes in metal-ligand distances.
Can We Stop Electron Exchange in 1/3 Systems? The
addition of excess DBN to 1 generates 3 quantitatively, with
1
sharp H and 31P NMR spectra. The absence of 1 prevents
the generation of 5 by eq 11 and thus suppresses line
broadening by the electron exchange in eq 12.
Are Any Other Electron-Transfer Reactions Involved
in 1/3 Systems? A CD3CN solution of 1 and 3 slowly
produces [CpRu(dppe)CD3CN][BPh4] and free thiol (from
the reaction of 1 with the solvent), so we have considered
the possibility that the acetonitrile complex and the free thiol
are involved in redox chemistry. The acetonitrile complex
is not reduced by Cp2Co, which means it must be a poorer
oxidizing agent than the thiol complex 1. The free thiol does
not reduce 1 or 5, as evidenced by the unchanged NMR line
broadening of 1/3 mixtures over time as free thiol appears.
Both [CpRu(dppe)CD3CN][BPh4] and free thiol retain sharp
NMR peaks in mixtures of 1 and 3. Thus neither [CpRu-
(dppe)CD3CN][BPh4] nor free thiol are participating in redox
chemistry in 1/3 systems.
The addition of Cp2Co to a mixture of 1 and 3 produces
a similar result, the quantitative generation of 3 with sharp
1H and 31P NMR spectra. The Cp2Co converts the 1 into 3,
preventing the generation of 5 by eq 11; furthermore, the
Cp2Co reduces any 5 that is present. Again, line broadening
by the electron exchange in eq 12 is suppressed. The relevant
processes are shown in Scheme 3.
Inversion of Coordinated Sulfur. When the temperature
of a solution of 1 and 3 in acetonitrile was lowered, the
downfield 31P resonance of the thiol complex 1 broadened
(Figure 8). Because of the freezing point of acetonitrile, we
examined the thiol complex 1 alone in CD2Cl2. As the
temperature was lowered (Figure 9), the downfield 31P NMR
resonance first broadened and then split into two signals.
The upfield phosphorus resonance broadened at lower
temperatures and then split into two signals. At -98 °C, the
31P NMR spectra reveal major and minor conformers, surely
(27) Mixtures of 1 and 3 in CD3CN are unchanged over time except for
the slow reaction of 1 with solvent. Over the course of four days, the
thiolate complex resonances remain broad with no discernible changes,
but sharp peaks for free thiol and [CpRu(dppe)CD3CN][BPh4] appear
in the 1H and 31P NMR spectra. We conclude that 4 (eq 11) must not
disproportionate rapidly when it is present in such low concentrations.
(28) This system is in the “large hyperfine” or “slow exchange” limit of
the de Boer-MacLean equation for NMR line broadening due to e-
exchange. See ref 18a and de Boer, E.; MacLean, C. J. Chem. Phys.
1966, 44, 1334-1342.
(29) This could be due to the electron transfer equilibrium in eq 11 (most
likely), slow electron exchange between 1 and 4, or proton exchange
between 1 and 3.
Inorganic Chemistry, Vol. 46, No. 14, 2007 5811

Shaw et al.
Inversion of sulfur in transition-metal complexes typically
occurs rapidly at room temperature. Low inversion barriers
in electron-deficient complexes are due to stabilization of
the planar transition state by empty metal d orbitals.30 In
18-electron complexes the low inversion barriers must be
due to destabilization of the pyramidal ground state.31
Conclusions
The NMR spectra in acetonitrile of the thiolate complex
3 become broad whenever 3 and its conjugate acid, the
cationic thiol complex 1, are present simultaneously. Oxida-
tion of 3 by 1 generates a small amount of the radical cation
5, which then undergoes rapid electron exchange with 3.
Acknowledgment. This research was supported by NSF
grant CHE-0451385. We thank Prof. N. Turro for use of
the EPR facility supported by the National Science Founda-
tion (CHE-04-15516). We thank Prof. G. Parkin for as-
sistance with the X-ray structure determinations and the
National Science Foundation (CHE-0619638) for the acqui-
sition of an X-ray diffractometer.
Figure 9. Variable temperature 31P NMR spectra (121.5 MHz) of 1 in
CH2Cl2.
Scheme 3
Supporting Information Available: Complete details of the
crystallographic study. This material is available free of charge via
the Internet at http://pubs.acs.org.
IC700580U
(30) Abel, E. W.; Bhargava, S. K.; Orrell, K. G. Prog. Inorg. Chem. 1984,
32, 1-118.
the result of slowing pyramidal inversion of the sulfur of
the thiol ligand.
(31) Rogers, J. R.; Wagner, T. P. S.; Marynick, D. S. Inorg. Chem. 1994,
33, 3104-3110.
5812 Inorganic Chemistry, Vol. 46, No. 14, 2007