Formation and reactivity of a persistent radical in a dinuclear molybdenum complex

Article abstract of DOI:10.1021/ja078115r

The reactivity of the S-H bond in Cp*Mo(μ-S)₂(μ-SMe) (μ-SH)MoCp* (S₄MeH) has been explored by determination of kinetics of hydrogen atom abstraction to form the radical Cp*Mo(μ-S) ₃(μ-SMe)MoCp* (S₄Me?), as well as reaction of hydrogen with the radical-dimer equilibrium to reform the S-H complex. From the temperature dependent rate data for the abstraction of hydrogen atom by benzyl radical, ΔH^? and ΔS^? were determined to be 1.54 ± 0.25 kcal/mol and -25.5 ± 0.8 cal/mol K, respectively, giving k_abs = 1.3 × 10⁶ M^-1 s^-1 at 25°C. In steady state abstraction kinetic experiments, the exclusive radical termination product of the Mo₂S₄ core was found to be the benzyl cross-termination product, Cp*Mo(μ-S) ₂(μ-SMe)(μ-SBz)MoCp* (S₄MeBz), consistent with the Fischer-Ingold persistent radical effect. S₄Me? was found to reversibly dimerize by formation of a weak bridging disulfide bond to form the tetranuclear complex (Cp*Mo(μ-S)₂(μ-SMe)MoCp*) ₂(μ-S₂) ((S₄Me)₂). The radical-dimer equilibrium constant has been determined to be 5.7 × 10 ⁴ ± 2.1 × 10⁴ M^-1 from EPR data. The rate constant for dissociation of the dimer was found to be 1.1 × 10³ s^-1 at 25°C, based on variable temperature ¹H NMR data. The rate constant for dimerization of the radical has been estimated to be 6.5 × 10⁷ M^-1 s^-1 in toluene at room temperature, based on the dimer dissociation rate constant and the equilibrium constant for dimerization. Structures are presented for (S ₄Me)₂, S₄MeBz, and the cationic Cp*Mo(μ-S₂)(μ-S)(μ-SMe)MoCp*(OTf) (S ₄Me⁺), a precursor of the radical and the alkylated derivatives. Evidence for a radical addition/elimination pathway at an Mo ₂S₄ core is presented.

Full text of DOI:10.1021/ja078115r

Published on Web 06/20/2008
Formation and Reactivity of a Persistent Radical in a
Dinuclear Molybdenum Complex
Aaron M. Appel,† Suh-Jane Lee,† James A. Franz,*,† Daniel L. DuBois,†
M. Rakowski DuBois,† Jerome C. Birnbaum,† and Brendan Twamley‡
Pacific Northwest National Laboratory, P.O. Box 999, Richland, Washington 99352, and
Department of Chemistry, UniVersity of Idaho, Moscow, Idaho 83844-2343
Received October 23, 2007; Revised Manuscript Received April 16, 2008; E-mail: james.franz@pnl.gov
Abstract: The reactivity of the S-H bond in Cp*Mo(µ-S)2(µ-SMe)(µ-SH)MoCp* (S4MeH) has been explored
by determination of kinetics of hydrogen atom abstraction to form the radical Cp*Mo(µ-S)3(µ-SMe)MoCp*
(S4Me•), as well as reaction of hydrogen with the radical-dimer equilibrium to reform the S-H complex.
From the temperature dependent rate data for the abstraction of hydrogen atom by benzyl radical, ∆H‡
and ∆S‡ were determined to be 1.54 ( 0.25 kcal/mol and -25.5 ( 0.8 cal/mol K, respectively, giving kabs
) 1.3 × 106 M-1 s-1 at 25 °C. In steady state abstraction kinetic experiments, the exclusive radical
termination product of the Mo2S4 core was found to be the benzyl cross-termination product, Cp*Mo(µ-
S)2(µ-SMe)(µ-SBz)MoCp* (S4MeBz), consistent with the Fischer-Ingold persistent radical effect. S4Me•
was found to reversibly dimerize by formation of a weak bridging disulfide bond to form the tetranuclear
complex (Cp*Mo(µ-S)2(µ-SMe)MoCp*)2(µ-S2) ((S4Me)2). The radical-dimer equilibrium constant has been
determined to be 5.7 × 104 ( 2.1 × 104 M-1 from EPR data. The rate constant for dissociation of the
dimer was found to be 1.1 × 103 s-1 at 25 °C, based on variable temperature 1H NMR data. The rate
constant for dimerization of the radical has been estimated to be 6.5 × 107 M-1 s-1 in toluene at room
temperature, based on the dimer dissociation rate constant and the equilibrium constant for dimerization.
Structures are presented for (S4Me)2, S4MeBz, and the cationic Cp*Mo(µ-S2)(µ-S)(µ-SMe)MoCp*(OTf)
(S4Me+), a precursor of the radical and the alkylated derivatives. Evidence for a radical addition/elimination
pathway at an Mo2S4 core is presented.
Introduction
demonstrated in our report of (CpMoS)2(S2CH2) as an electro-
catalyst for hydrogen production at low overpotentials.16 Study-
ing the formation and cleavage of S-H bonds in Mo2S4-based
complexes is central to understanding the wide range of
observed reactivities.
Homogenous molybdenum sulfide complexes have been
widely studied due to their importance in understanding
fundamental reaction channels available to molybdenum sulfide
based hydrodesulfurization (HDS) catalysts1–3 and the important
role that molybdenum sulfide complexes play in enzymatic
reactions.3–5 Cp2Mo2S4 complexes have been extensively studied
because of their ability to activate hydrogen and undergo bond-
forming reactions with a variety of hydrocarbons.6–15 In addition
to the many examples of hydrogen utilization reactions, these
complexes can also be used for the production of hydrogen, as
Investigations of hydrogen atom transfer kinetics and ther-
modynamics are also important for hydrogenation,17–20 chain
transfer polymerization,21–25 and initiation of other radical
(12) Brunner, H.; Meier, W.; Wachter, J.; Weber, P.; Ziegler, M. L.;
Enemark, J. H.; Young, C. G. J. Organomet. Chem. 1986, 309, 313–
318.
† Pacific Northwest National Laboratory.
(13) Kubas, G. J.; Ryan, R. R.; Kubatmartin, K. A. J. Am. Chem. Soc.
1989, 111, 7823–7832.
‡ University of Idaho.
(1) Grange, P.; Vanhaeren, X. Catal. Today 1997, 36, 375–391.
(2) Bianchini, C.; Meli, A. Acc. Chem. Res. 1998, 31, 109–116.
(3) Donahue, J. P. Chem. ReV. 2006, 106, 4747–4783.
(4) Hille, R. Chem. ReV. 1996, 96, 2757–2816.
(14) Birnbaum, J.; Godziela, G.; Maciejewski, M.; Tonker, T. L.; Halti-
wanger, R. C.; Rakowski Dubois, M. Organometallics 1990, 9, 394–
401.
(15) Newell, R.; Ohman, C.; Rakowski DuBois, M. Organometallics 2005,
24, 4406–4415.
(5) Brondino, C. D.; Rivas, M. G.; Romao, M. J.; Moura, J. J. G.; Moura,
I. Acc. Chem. Res. 2006, 39, 788–796.
(16) Appel, A. M.; DuBois, D. L.; Rakowski DuBois, M. J. Am. Chem.
Soc. 2005, 127, 12717–12726.
(6) Rakowski Dubois, M.; Haltiwanger, R. C.; Miller, D. J.; Glatzmaier,
G. J. Am. Chem. Soc. 1979, 101, 5245–5252.
(17) Ru¨chardt, C.; Gerst, M.; Ebenhoch, J. Angew. Chem., Int. Ed. Engl.
1997, 36, 1406–1430.
(7) Rakowski Dubois, M.; Vanderveer, M. C.; Dubois, D. L.; Haltiwanger,
R. C.; Miller, W. K. J. Am. Chem. Soc. 1980, 102, 7456–7461.
(8) Rakowski Dubois, M.; Dubois, D. L.; Vanderveer, M. C.; Haltiwanger,
R. C. Inorg. Chem. 1981, 20, 3064–3071.
(18) Sweany, R. L.; Halpern, J. J. Am. Chem. Soc. 1977, 99, 8335–8337.
(19) Halpern, J. Pure Appl. Chem. 1986, 58, 575–584.
(20) Eisenberg, D. C.; Norton, J. R. Isr. J. Chem. 1991, 31, 55–66.
(21) Gridnev, A. A.; Ittel, S. D. Chem. ReV. 2001, 101, 3611–3660.
(22) Tang, L.; Norton, J. R. Macromolecules 2006, 39, 8236–8240.
(23) Tang, L.; Norton, J. R. Macromolecules 2006, 39, 8229–8235.
(24) Choi, J.; Tang, L.; Norton, J. R. J. Am. Chem. Soc. 2007, 129, 234–
240.
(9) Brunner, H.; Meier, W.; Wachter, J.; Guggolz, E.; Zahn, T.; Ziegler,
M. L. Organometallics 1982, 1, 1107–1113.
(10) Curtis, M. D.; Williams, P. D. Inorg. Chem. 1983, 22, 2661–2662.
(11) Brunner, H.; Kauermann, H.; Meier, W.; Wachter, J. J. Organomet.
Chem. 1984, 263, 183–192.
9
8940 J. AM. CHEM. SOC. 2008, 130, 8940–8951
10.1021/ja078115r CCC: $40.75
2008 American Chemical Society

Persistent Radical in a Mo2S4 Complex
A R T I C L E S
reactions.26–28 The formation of persistent radicals,29–36 either
due to slow or readily reversible dimerization, can result in living
polymerization or catalytic hydrogenation by reverse-radical-
disproportionation (eq 1), depending upon the strength of the
Chart 1. Structural Representations and Abbreviations of
Cp*2Mo2S4 Complexes in This Work
•
MH + R2CdCR′2 f M • + R2CsCHR′2
(1)
bond to hydrogen. Equation 2 illustrates the putative mechanism
for stepwise hydrogenolytic cleavage of carbon-carbon bonds
facilitated by reactive M-H or S-H bonds of heterogeneous
molybdenum sulfide catalysts (represented by Mo2S4H2). This
is suggested to be a general mechanism for hydrogenation (eq
3) and hydrogenolysis during hydropyrolysis of carbonaceous
materials such as biomass and coal.17
Mo2S4H2
ArsR
8 ArH + • R Mo S H•8 ArH + HR
(2)
2
4
Mo2S4H2
8 R2CsCHR′2 Mo S H•8 R2CHsCHR′2 (3)
•
2
4
R2CdCR′2
H2
Mo2S4
98 Mo2S4H2
(4)
We have reported studies of the hydrogen atom abstraction
rate by benzyl radical from Cp2Mo2S4H2 and the relevance of
these kinetic studies to reverse-radical-disproportionation based
hydrogenation (eq 3) for MonS2n HDS catalysts and C-C bond
hydrogenolysis (eq 2).37 However, the radical resulting from
hydrogen atom abstraction from Cp2Mo2S4H2 can donate a
second hydrogen atom or undergo disproportionation to form
Cp2Mo2S4 and Cp2Mo2S4H2, complicating the kinetics and
preventing unequivocal observation of S-S bond formation in
self-reaction of Cp2Mo2S4H•. To simplify the kinetic studies,
one of the µ-SH hydrogen atoms was replaced with a methyl
group in order to eliminate disproportionation by hydrogen
transfer. Next, self-combination of the radical was impeded by
use of the more sterically hindered pentamethylcyclopenta-
dienide (Cp*) rather than the unsubstituted Cp ligand.38 These
two approaches were expected to yield a stable or persistent
radical. Both approaches have been used in this work to make
S4MeH (see Chart 1) and related complexes to investigate the
homolysis and formation of S-H and S-S bonds in Mo2S4
complexes.
Results and Discussion
Synthesis and Structure of S4Me+. S4H2 was prepared by the
literature methods7,9 and used for the synthesis of S4 by a new
route through acid induced H2 elimination (Scheme 1, reactions
a and b). In this approach, reacting S4H2 with an excess of
trifluoroacetic acid resulted in formation of S4H+, which was
then deprotonated using triethylamine to give S4. The spectro-
scopic data of this product matched the previously reported
data.9 However, rather than producing a mixture of S4, anti-S4,
and syn-S4, this method selectively produced S4, which was then
used as a starting material for the synthesis of mono and
dialkylated S4H2 analogs.
S4Me+ was prepared by reaction of methyl triflate with S4
(Scheme 1, reaction e) and isolated as the triflate salt rather
than the previously reported iodide11 in order to simplify
electrochemical studies. The crystal structure of S4Me+ shows
a bridging disulfide with an average sulfur-sulfur distance of
2.06 Å for the two cations observed per asymmetric unit. The
disulfide bond is consistent with previously reported structures
for S4 (2.10 Å)9 and a similar monoalkylated cationic complex
(2.13 Å).39 The S-CH3 average distance is 1.81 Å, and the
Mo-Mo average distance is 2.63 Å. Figure 1 shows one of the
two S4Me+ cations per asymmetric unit in the structure, with
selected bond lengths and angles listed in Table 1. Crystal-
lographic data and additional figures with the complete contents
of the asymmetric unit are in the Supporting Information.
Synthesis and Reactivity of S4MeH. S4MeH was prepared
from S4 by reaction with methyl lithium followed by protonation
of the formed anion (Scheme 1, reactions c and d). The 1H NMR
spectrum of this complex was very similar to those of S4H2
(25) Tsarevsky, N. V.; Matyjaszewski, K. Chem. ReV. 2007, 107, 2270–
2299.
(26) Yet, L. Tetrahedron 1999, 55, 9349.
(27) Jasperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. ReV. 1991, 91, 1237–
1286.
(28) Smith, D. M.; Pulling, M. E.; Norton, J. R. J. Am. Chem. Soc. 2007,
129, 770–771.
(29) Daikh, B. E.; Finke, R. G. J. Am. Chem. Soc. 1992, 114, 2938–2943.
(30) Bravo, A.; Bjorsvik, H. R.; Fontana, F.; Liguori, L.; Minisci, F. J.
Org. Chem. 1997, 62, 3849–3857.
(31) Fischer, H. Macromolecules 1997, 30, 5666–5672.
(32) Kothe, T.; Marque, S.; Martschke, R.; Popov, M.; Fischer, H. J. Chem.
Soc., Perkin Trans. 2 1998, 1553–1559.
(33) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1976, 9, 13–19.
(34) Fischer, H. J. Am. Chem. Soc. 1986, 108, 3925–3927.
(35) Fischer, H. Chem. ReV. 2001, 101, 3581–3610.
(36) Focsaneanu, K. S.; Scaiano, J. C. HelV. Chim. Acta 2006, 89, 2473–
2482.
(37) Franz, J. A.; Birnbaum, J. C.; Kolwaite, D. S.; Linehan, J. C.;
Camaioni, D. M.; Dupuis, M. J. Am. Chem. Soc. 2004, 126, 6680–
6691.
(38) Watkins, W. C.; Jaeger, T.; Kidd, C. E.; Fortier, S.; Baird, M. C.;
Kiss, G.; Roper, G. C.; Hoff, C. D. J. Am. Chem. Soc. 1992, 114,
907–914.
(39) Lopez, L. L.; Gabay, J.; Haltiwanger, R. C.; Green, K.; Allshouse, J.;
Casewit, C.; Rakowski Dubois, M. Organometallics 1993, 12, 4764–
4770.
9
J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008 8941

A R T I C L E S
Appel et al.
Scheme 1. Synthetic Routes to S4, S4MeH, S4Me+, and S4MeBz
by Reactions a through f
Figure 1. Structure of S4Me+ with 30% thermal displacement. Only one
of the two unique S4Me+ cations is shown, and hydrogen atoms, counterions,
and solvent molecules are omitted for clarity.
Table 1. Selected Bond Distances and Angles for One of the Two
Unique S4Me+ Cations per Asymmetric Unit
distances, Å
angles, deg
Mo(1)-Mo(2)
2.6311(7)
2.050(2)
3.145
2.910
3.136
1.820(6)
2.4255(16)
2.4310(15)
2.4627(15)
2.2984(17)
2.4300(16)
2.4452(15)
2.4512(15)
2.2953(16)
2.325
C(21)-S(3)-Mo(1)
C(21)-S(3)-Mo(2)
Mo(1)-S(1)-Mo(2)
Mo(1)-S(2)-Mo(2)
Mo(2)-S(3)-Mo(1)
Mo(2)-S(4)-Mo(1)
S(1)-Mo(1)-S(2)
S(1)-Mo(2)-S(2)
S(1)-S(2)-Mo(1)
S(1)-S(2)-Mo(2)
S(2)-Mo(1)-Mo(2)
S(2)-Mo(2)-Mo(1)
S(2)-S(1)-Mo(1)
S(2)-S(1)-Mo(2)
111.9(2)
112.2(2)
65.62(4)
65.31(4)
64.75(4)
69.89(5)
49.95(6)
49.74(6)
64.89(6)
64.74(6)
57.61(4)
57.09(4)
65.17(6)
65.52(6)
S(1)-S(2)
S(2)· · ·S(3)
S(3)· · ·S(4)
S(4)· · ·S(1)
C(21)-S(3)
Mo(1)-S(1)
Mo(1)-S(2)
Mo(1)-S(3)
Mo(1)-S(4)
Mo(2)-S(1)
Mo(2)-S(2)
Mo(2)-S(3)
Mo(2)-S(4)
Mo-C (average)
Chart 2. Structural Representations for the Isomers of the Sulfur
Core in Cp2Mo2S4R2 Complexes, As Viewed Down the Mo-Mo
Axis
and S4Me2, including the presence of two different observable
isomers (see Chart 2).7,12 The two isomers are observed in an
approximately 2:1 ratio and are assigned as isomers A and B,
as they have been computationally shown to be the lowest in
energy for Cp2Mo2S4H2.37 The spectrum contained the Cp*
resonance at 2.20 ppm (in CD3CN), the methanethiolate
resonances at 1.07 and 0.99 ppm (isomers B and A, respec-
tively), and the hydrosulfide resonances at -2.33 and -2.51
ppm (isomers B and A, respectively).
S4MeH was used for kinetic studies of the S-H hydrogen
atom abstraction rate of benzyl radical through photolysis of
dibenzyl ketone (DBK) in the presence of S4MeH (Scheme 2,
reaction a), as previously described for Cp2Mo2S4H2.37 Figure
2 shows the Eyring plot of the resulting data, and Table 2
contains the rate parameters along with those for Cp2Mo2S4H2
for comparison.
1H NMR the products resulting from cross-termination of S4Me•
radical with benzyl radical and self-termination of S4Me• to
form a dimer. In these experiments, the primary radical
termination product observed for the resulting S4Me• was the
cross-termination product, S4MeBz, suggesting slow or revers-
ible dimerization, consistent with the persistent radical effect.29–36
The cross-termination reaction is shown in Scheme 2, reaction
c, and Scheme 3, eq 9. The depletion of S4MeH and the growth
of S4MeBz as a function of photolysis time can be observed
from the methanethiolate, benzylthiolate (methylene), and
After determining the rate of hydrogen atom abstraction by
benzyl radical, additional photolysis experiments were run at a
higher concentration of S4MeH in order to directly observe by
9
8942 J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008

Persistent Radical in a Mo2S4 Complex
A R T I C L E S
Table 2. Rate Parameters for the Reaction of Benzyl Radical with µ-SH in S4MeH and Previously Reported Kinetic Results for
Comparisona
hydrogen donor
k (M-1 s-1) 298 K
T range (K)
mean temp (Tm, K)
∆S‡ (eu) (Tm, K)
log Ac (M-1 s-1
)
∆H‡ (kcal/mol)
Eac (kcal/mol)
S4MeH
1.3 × 106
2.6 × 106
1.3 × 105
3.1 × 104
295-353
295-372
298-376
295-374
325
331
337
341
-25.5 ( 0.8
-19.2 ( 1.8
-23.2 ( 0.8
-24.7 ( 1.6
7.68 ( 0.17
9.07 ( 0.38
8.21 ( 0.17
7.89 ( 0.35
1.54 ( 0.25
2.96 ( 0.58
3.57 ( 0.26
3.96 ( 0.54
2.13 ( 0.25
3.62 ( 0.58
4.24 ( 0.26
4.64 ( 0.54
b
Cp2Mo2S4H2
2-naphthalenethiolb
1-octanethiolb
a All measurements were made in benzene, errors are 2σ. b Franz, J. A.; Birnbaum, J. C.; Kolwaite, D. S.; Linehan, J. C.; Camaioni, D. M.; Dupuis,
M. J. Am. Chem. Soc. 2004, 126, 6680-6691. c A ) Arrhenius pre-exponential factor, Ea ) Arrhenius expression, k ) Ae-E /RT.
a
Scheme 2. Formation and Reactions of S4Me•
Scheme 3. Primary Reactions in Steady-State Kinetics of Reaction
of Benzyl Radical and S4MeH
but exhibit the additional methylene and aromatic resonances
from the benzyl group and the loss of the hydrosulfide
resonances. The aromatic resonances were observed at 7.08,
7.03, and 6.94 ppm, the methylene resonances at 2.82 and 2.78
ppm (assigned as isomers B and A, respectively), Cp* reso-
nances were observed at 2.16 and 2.14 ppm (isomers A and
B), and the methanethiolate resonances at 1.14 and 1.00 ppm
(isomers B and A). To confirm this formulation, crystals of the
synthetically prepared S4MeBz were grown by slow evaporation
from a benzene solution, and the structure is shown in Figure
4 with selected bond lengths and angles given in Table 3.
hydrosulfide NMR signals (Figure 3). Two isomers are observed
for each complex, giving two methyl and hydrosulfide reso-
nances for S4MeH and two methylene and methyl resonances
for S4MeBz (see Chart 2, above).
In order to verify the primary termination product, S4MeBz
was synthesized by the reaction of S4Me+ with benzylmagne-
1
sium chloride (Scheme 1, Reaction f) and compared by H
NMR, giving matching synthetic and photolytic products with
the exception of the A:B isomeric ratios, which were ap-
proximately 10:1 for the synthetic product and 1:2 for the
photolytic product. However, after several days in solution the
isomeric ratio for A:B reached 3:2, which is the same ratio
reported for Cp2Mo2S4H2.7 The Cp* and methanethiolate
resonances of S4MeBz were very similar to those of S4MeH,
Figure 3. 1H NMR spectra of the photolysis of DBK and S4MeH, showing
the depletion of S4MeH and formation of S4MeBz from the methylene (left),
methyl (center), and hydrosulfide (right) regions as a function of photolysis
time. Two isomers are observed for each complex (see Chart 2). The sample
was photolyzed in C6D6 at room temperature up to 150 s total photolysis
time.
Figure 2. Temperature-dependent rate data for abstraction of hydrogen
atom by benzyl radical from S4MeH in benzene.
9
J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008 8943

A R T I C L E S
Appel et al.
Scheme 4. Transient Intermediate, Suggested To be cis-S4MeBz,
Is Observed Both in the Reaction (a) of Benzyl Radical with
S4MeH and in the Photoreaction (g) of trans-S4MeBz with Benzyl
Radicala
Figure 4. Structure of S4MeBz with 30% thermal displacement. Only the
primary position for the bridging sulfide and thiolate ligands is shown for
one of the two unique S4MeBz moieties in the asymmetric unit. Hydrogen
atoms, counterions, and solvent molecules are omitted for clarity.
Table 3. Selected Bond Distances and Angles for One of the Two
Unique S4MeBz Moieties per Asymmetric Unit
distances, Å
angles, deg
Mo(1)-Mo(2)
2.5758(5)
1.835(15)
1.860(4)
Mo(1)-S(1A)-Mo(2)
Mo(1)-S(3A)-Mo(2)
Mo(2)-S(2A)-Mo(1)
65.75(3)
66.74(3)
62.44(2)
62.47(2)
S(2A)-C(21A)
S(4A)-C(22A)
Mo(1)-S(1A)
Mo(1)-S(2A)
Mo(1)-S(3A)
Mo(1)-S(4A)
Mo(2)-S(1A)
Mo(2)-S(2A)
Mo(2)-S(3A)
Mo(2)-S(4A)
2.3719(10) Mo(2)-S(4A)-Mo(1)
2.4891(10) C(21A)-S(2A)-Mo(1) 113.0(6)
2.3376(10) C(21A)-S(2A)-Mo(2) 113.6(2)
2.4877(10) C(22A)-S(4A)-Mo(1) 116.22(17)
2.3745(10) C(22A)-S(4A)-Mo(2) 112.00(13)
2.4818(10) C(23)-C(22A)-S(4A)
2.3466(10)
2.4806(10)
111.9(3)
a In both cases, cis-S4MeBz undergoes slow thermal isomerization (d)
to trans-S4MeBz. Pathway (d) may occur via addition/scission pathways
(e,f).
Mo-C (average) 2.343
The asymmetric unit contains two S4MeBz and two benzene
molecules, and the bridging sulfide and alkylthiolate ligands in
each S4MeBz unit are disordered into two positions with
occupancies of 89% and 11%. The average distances are 2.58
Å for Mo-Mo, 1.84 Å for S-CH3, and 1.85 Å for S-CH2Ph.
The average Mo-S distances are 2.36 Å, 2.49 Å, and 2.48 Å
for the sulfides, methanethiolates, and benzylthiolates, respec-
tively. The crystal structure contains only isomer A (see Chart
2), which is expected to correspond to the primary isomer
observed in the NMR spectra for the synthetic product. Although
the second isomer observed in the NMR spectrum is not
observed in the crystal structure, it is predicted to be isomer B,
based on the previously calculated isomers for Cp2Mo2S4H2.37
In addition to isomers A and B, a third product was observed
in the photolysis experiments, and this product may be one of
the cis isomers of S4MeBz (Scheme 4, Reaction c). While the
cis isomers are usually not observed for the S4R2 complexes,
benzyl addition by a radical pathway may result in the formation
of different isomers than are observed by other synthetic routes.
The observed isomer is expected to be isomer C or D in Chart
2, as the relative energies of these isomers for Cp2Mo2S4H2 have
been previously calculated to be similar, whereas isomer E is
expected to be significantly higher in energy.37
A and B suggests that the product is very likely one of the less
stable cis cross-termination product isomers. Further evidence
for the intermediacy of cis-S4MeBz was developed by photolysis
of a solution of the two isomers of trans-S4MeBz in benzene
(Scheme 4, Reaction g). Without added DBK, no photoreaction
was observed. However, with added DBK, the identical
intermediate, proposed to be one of the possible isomers of cis-
S4MeBz, was formed and was observed to convert in a dark
reaction to the two isomers of trans-S4MeBz. Thus, addition
of benzyl radical to trans-S4MeBz and scission to give both
isomers (Scheme 4, pathways e,f) appears facile, possibly by a
slow chain reaction involving addition/scission of benzyl radical
to the Mo2S4 core.
Reduction of S4Me+. By cyclic voltammetry, two reversible
waves are observed for the reduction of S4Me+ in acetonitrile,
with half-wave potentials of -0.93 and -1.30 V vs ferrocene
and peak separations of 59 and 62 mV at 0.5 V/s. These data
suggest that a stable product could be isolated from a one-
electron reduction of the cation at a potential between the two
waves, with the overall reaction shown in Scheme 2, reaction
b. S4Me+ was reduced in acetonitrile both by controlled potential
electrolysis, passing 3.9 C (4.3 C expected for one electron
reduction), and by chemical reduction using 1 equiv. of
bis(benzene) chromium(0)40 to give a purple precipitate with
Investigation of this third product, which exhibited a Cp*
methyl resonance at 1.90 ppm, revealed that it slowly converts
to trans-S4MeBz isomers A and B (approximately 75% conver-
sion was observed in one week). The slow conversion to isomers
(40) Connelly, N. G.; Geiger, W. E. Chem. ReV. 1996, 96, 877–910.
9
8944 J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008

Persistent Radical in a Mo2S4 Complex
A R T I C L E S
Figure 6. EPR spectrum of 5 mM (S4Me)2 in toluene at 23 °C with g )
2.003.
Chart 3. Structural Representations for the Possible Isomers of the
Sulfur Core in the S4Me• Radical, As Viewed Down the Mo-Mo
Axis
Figure 5. Structure of (S4Me)2 with 30% thermal displacement, where
the two S4Me• units are symmetry related (2-fold axis). Only one of the
disordered sulfide and methanethiolate positions is shown, and the hydrogen
atoms and solvent molecules are omitted for clarity.
Table 4. Selected Bond Distances and Angles for (S4Me)2
distances, Å
angles, deg
Mo(1)-Mo(2)
2.5940(5)
2.059(3)
1.782(8)
2.4760(16)
2.3537(18)
2.454(2)
2.2966(17)
2.5326(17)
2.3994(17)
2.464(2)
C(21A)-S(3)-Mo(1)
C(21A-S(3)-Mo(2)
Mo(1)-S(1)-Mo(2)
Mo(1)-S(2)-Mo(2)
Mo(1)-S(3)-Mo(2)
Mo(1)-S(4)-Mo(2)
S(1A)-S(1)-Mo(1)
S(1A)-S(1)-Mo(2)
116.4(4)
115.3(4)
62.37(4)
66.15(5)
63.67(5)
67.71(5)
141.03(6)
106.82(10)
S(1)-S(1A)
S(3)-C(21A)
Mo(1)-S(1)
Mo(1)-S(2)
Mo(1)-S(3)
Mo(1)-S(4)
Mo(2)-S(1)
Mo(2)-S(2)
Mo(2)-S(3)
Mo(2)-S(4)
Mo-C (average)
significantly tilted away from the neighboring S4Me•, forming
a dihedral angle of 12.7° between the C(1)-C(10) and
C(11)-C(20) planes. This tilting of the rings illustrates the steric
crowding occurring between the Cp* rings on the two S4Me•
units, and has further implications for the low temperature NMR
data, as discussed below. Additional crystallographic figures and
data are given in the Supporting Information.
2.3592(18)
2.353
Magnetic Studies of (S4Me)2. A solution of (S4Me)2 in toluene
was used to collect the EPR spectrum shown in Figure 6. While
the dimer itself is expected to be diamagnetic, dissociation into
S4Me• would give a paramagnetic complex, giving an EPR
spectrum similar to those reported for (CpMo)2(µ-S)(µ-
SCH3)S2CH2 and [MeCpMo(µ-S)(µ-SCH3)]2K, including the
matching spectroscopic data from either method. Additionally,
cyclic voltammetry in THF resulted in equivalent data from
S4Me+ and the products of the reductions. Recrystallization of
this solid by vapor diffusion of acetonitrile into a toluene
solution of the product gave crystals suitable for analysis by
X-ray diffraction. The crystal structure revealed a dimer of two
S4Me• units, making a tetranuclear complex with a bridging
disulfide bond between the Mo2S4 cores. The structure of the
dimer is shown in Figure 5, with selected bond lengths and
angles in Table 4.
The two S4Me• units in the dimer are symmetry related (2-
fold axis) and have a Mo(1)-Mo(2) distance of 2.59 Å. The
methanethiolate on each monomer is disordered into three
positions, giving an average S-CH3 bond length of 1.79 Å.
Isomers A and B (Chart 2) can be assigned using the relative
orientations of the bridging disulfide bond and the S-C bond
of the methanethiolate in one S4Me• unit, where two of the
disordered methanethiolate positions correspond to isomer A
(75% occupancy), and the remaining position corresponds to
isomer B (25% occupancy). The disulfide (S(1)) and two
bridging sulfide ligands (S(2) and S(4)) in each monomer are
disordered into two different sets of positions. The average bond
length for the disulfide bridging the two S4Me• units is 2.14 Å,
which is consistent with previously reported bridging disulfide
bond lengths in analogous dimers,14 as well as other internal
disulfide bonds (see S4Me+, above). While steric strain for dimer
formation is not evident from the length of the bridging disulfide
bond, the Cp* rings at each end of the Mo2S4 units are
5
hyperfine coupling to I ) /2 molybdenum nuclei.41,42 In the
case of the S4Me• radical, the EPR spectrum is expected to be
more complex than those previously reported. The spectrum
may be a composite spectrum of as many as three possible
isomeric forms of the radical, as illustrated in Chart 3. Isomers
A and B are analogous to the isomers discussed above for S4R2
complexes, except with a disulfide bond as the point of reference
rather than a bond to a second substituent. A and B are expected
to be close in energy, reminiscent of the two predominant
isomers of Cp2Mo2S4H2.7,37 Isomer C, however, is a distinct
structure, lacking the disulfide bond. Computational studies
reveal isomer C to be 6 kcal/mol less stable than A, suggesting
that isomer C does not contribute significantly to the EPR
spectrum, as it will be inadequately populated to be observable.43
On the basis of these computational results, the observed EPR
spectrum is expected to result from a combination of isomers
A and B. The Supporting Information contains further details
of the EPR spectrum and simulations of this complex spectrum
using two isomers. Further evidence that the complex EPR
(41) Casewit, C. J.; Haltiwanger, R. C.; Noordik, J.; Rakowski Dubois,
M. Organometallics 1985, 4, 119–129.
(42) Casewit, C. J.; Rakowski Dubois, M. Inorg. Chem. 1986, 25, 74–80.
(43) UB3LYP electronic structure calculations reveal C to be 6 kcal/mole
less stable than A. Results to be published separately.
9
J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008 8945

A R T I C L E S
Appel et al.
spectrum resulted from S4Me• and not an unrelated impurity is
shown in the collection of EPR spectra at multiple concentrations
and the reaction of the sample with hydrogen, both discussed
below.
The solid state dimer structure and the solution EPR spectra
suggest the interconversion of (S4Me)2 and S4Me• (Scheme 2,
reaction d). In order to determine the extent of dimer formation
in solution (eq 12), the Evans method was used to measure the
magnetic susceptibility by 1H NMR.44,45 In this method, a sealed
capillary tube containing a pure solvent is placed in an NMR
tube containing a solution of the analyte in the same solvent.
The difference in chemical shift between the pure solvent and
the solvent containing analyte is proportional to the magnetic
susceptibility of the analyte.
Figure 7. Experimental EPR spectra for 16 mM (blue), 4.1 mM (red),
and 1.1 mM (green) solutions of (S4Me)2 after correcting for the concentra-
tion dependence of the radical-dimer equilibrium. The spectra were collected
in toluene at 23 °C.
[(S4Me)2]
[S4Me]2
Kdim
)
(12)
However, a 0.018 M solution of (S4Me)2 in toluene resulted
in a single peak for the toluene methyl group, indicating a low
level of paramagnetism. The lower limit for the equilibrium
constant for dimer formation (Kdim) can be estimated from the
NMR spectrum given that the difference in chemical shift
between the pure solvent peak and the solvent peak for the
analyte solution is less than the observed line width, 3 Hz.
Assuming that the radical S4Me• would have one unpaired
electron with a spin-only µeff of 1.73, as would be expected
given the g value of 2.003 observed in the EPR spectrum, the
maximum concentration of radical in this solution is 8.1 × 10-4
products. The formation of S4Me2 and anti-S4 suggests that a
primary decomposition pathway for S4Me• is disproportionation
by methyl transfer, much like the reported transfer for
(MeCpMo)2(S)(SCH3)(S2CH2) at reflux in THF to give
(MeCpMoS2)2(S2CH2) and (MeCpMo(SCH3)2)2(S2CH2).47 If
(S4Me)2 disproportionates by methyl transfer, then S4Me2 and
S4 would be the expected products; however, it would be
predicted that S4 would undergo facile isomerization to anti-S4
given that S4 can be thermally9 and photochemically48 converted
to syn-S4 and anti-S4, and that computationally, the anti isomer
has been determined to be the most stable isomer of Cp2Mo2S4.37
Below 10 °C, the broad Cp* resonance in the 1H NMR
spectrum stopped shifting and began to resolve. Similarly,
methanethiolate peaks were not discernible at room temperature
but were observed below -1 °C, showing two primary isomers
in a 1.7:1 ratio, similar in both chemical shift and relative ratio
to those observed for S4MeH. At -1 °C, the Cp* resonance
resolved into two broad peaks. Upon further cooling to -33
°C, each of these two peaks resolved into four singlets, with
one of the singlets having twice the area of the others (relative
areas of 3:3:3:6). From the crystal structure of (S4Me)2, it can
be seen that the bridging disulfide is asymmetric relative to the
two Cp* rings on each unit, as illustrated in Figure 8, resulting
in two different chemical environments for the Cp* rings,
consistent with the NMR data at -1 °C. Furthermore, both the
observed close proximities of the Cp* methyl groups in the
crystal structure and the inequivalency of the methyl groups on
each Cp* that is observed in the NMR spectra at -33 °C suggest
that the Cp* methyl groups are sufficiently crowded to hinder
ring rotation and thereby rapid site exchange of the methyl
groups while in the (S4Me)2 form.
M, giving Kdim > 2.7 × 104 M-1
.
In order to better quantify the radical-dimer equilibrium, the
EPR spectrum of solutions of (S4Me)2 were collected at 16,
4.1, and 1.1 mM concentrations, and spectra of TEMPO at 8.4
× 10-5 M were collected under identical conditions. The spectra
of both TEMPO and (S4Me)2 were then double integrated to
determine the concentration of S4Me•, resulting in an average
equilibrium constant of 5.7 × 104 ( 2.1 × 104 M-1 from 9
experiments. This equilibrium constant corresponds to a dimer
bond dissociation free energy of 6.5 kcal/mol at 25 °C. The
determined equilibrium constant is in line with the limit provided
by the magnetic susceptibility experiments and is just outside
of the range observable in that experiment. While the line shape
for the EPR spectra does not change with concentration, the
total intensity does. Figure 7 shows the experimental EPR
spectra after correcting for the expected change in radical
concentration based upon the change in initial dimer concentra-
tion, as described by Eq 12. Specifically, decreasing the total
(S4Me)2 concentration by a factor of 4 results in a factor of 2
decrease in S4Me• concentration.
Variable Temperature NMR Spectroscopy of (S4Me)2. At
temperatures of +10 °C or higher, a broad singlet was observed
The use of low temperature NMR data has been previously
reported for the determination of the dissociation rate constants
for [Cp*Cr(CO)3]2 and [CpCr(CO)3]2.46 For these species, the
rate constants for dissociation were reported to be proportional
to the line width for the average Cp or Cp* resonances over
the temperature range in which the resonances were not
paramagnetically shifted or broadened. In the case of (S4Me)2,
the low temperature NMR spectra reveal the expected inequiva-
lence of the Cp* methyl resonances. In order to determine the
rate constants for exchange from the 1H NMR spectra, the
1
for the Cp* resonance in the H NMR of (S4Me)2 in toluene-
d8. This peak shifted downfield with increasing temperature or
decreasing concentration, a consequence of the increase in the
radical to dimer ratio resulting from the greater extent of
dissociation at higher temperatures or in more dilute solutions.46
At temperatures above 50 °C, substantial decomposition oc-
curred within 15 min, resulting in the formation of equal
amounts of S4Me2 and anti-S4, as well as additional unidentified
(44) Evans, D. F. J. Chem. Soc. 1959, 2003–2005.
(45) Sur, S. K. J. Magn. Reson. 1989, 82, 169–173.
(46) Woska, D. C.; Ni, Y. P.; Wayland, B. B. Inorg. Chem. 1999, 38, 4135–
4138.
(47) Casewit, C. J.; Rakowski Dubois, M. J. Am. Chem. Soc. 1986, 108,
5482–5489.
(48) Bruce, A. E.; Tyler, D. R. Inorg. Chem. 1984, 23, 3433–3434.
9
8946 J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008

Persistent Radical in a Mo2S4 Complex
A R T I C L E S
exchange and the methanethiolate exchange and results in a
linear plot, suggesting that both exchange processes occur by
the same mechanism, dimer dissociation and redimerization.
This is further supported by the fact that rapid interconversion
of the methanethiolate isomers by inversion at the sulfur centers
is not observed for Cp*2Mo2S4R2 complexes, as is evident in
the reported chromatographic separation for the isomers of
S4Me2 and the lack of exchange for the isomeric hydrosulfido
resonances of Cp2Mo2S4H2 up to 90 °C.7,12 The normally slow
inversion of sulfur centers suggests that a different mechanism
is the source of the methanethiolate exchange in this system.
Dimer dissociation and the following redimerization would
result in exchange of both the methanethiolate resonances and
the Cp* methyl groups, and this process is already known to
occur as a result of the observed equilibrium from the magnetic
susceptibility and EPR data, and the matching cyclic voltam-
metry data for S4Me+ and (S4Me)2 in THF. If S4Me• and
(S4Me)2 did not interconvert rapidly at room temperature, then
the cyclic voltammograms for S4Me+ and (S4Me)2 would not
be expected to match. Given the equilibrium constant of 5.7 ×
104 M-1 for Kdim from the EPR data, the dimer is the
predominant form at millimolar concentrations at room tem-
perature, indicating that the dimerization rate constant is larger
than the dissociation rate constant. Consequently, dissociation
appears to be the rate-limiting step in the exchange process,
consistent with the observed concentration independent low
temperature NMR data. Using the kinetic parameters from the
Eyring plot for the exchange process, given that these parameters
appear to be for the dissociation of (S4Me)2, the rate constant
for dissociation can be calculated to be 1.1 × 103 s-1 at 25 °C.
Using the rate constant for the dissociation of the dimer and
the equilibrium constant for dimerization (1.1 × 103 s-1 and
5.7 × 104 M-1 at 25 °C), the rate constant for dimerization can
be calculated to be 6.3 × 107 M-1 s-1 at 25 °C in toluene. In
the photolysis experiments, the exclusive formation of cross-
termination products and the lack of formation of the self-
termination product (S4Me)2 is consistent with the Fischer-Ingold
persistent radical effect,29–36 and in this case, may be the result
of both reversible dimerization and appreciably reduced self-
termination rates.
Figure 8. Ball and stick and space-filling models of (S4Me)2 using
crystallographic data, illustrating Cp* inequivalence due to the bridging
disulfide (top) and the steric constraint around each Cp* (bottom), both
consistent with the variable temperature NMR data. Hydrogen atoms have
been omitted for clarity.
Reaction of (S4Me)2 with Hydrogen. Previously reported
disulfide bridged dimers have been reduced in a hydrogen
atmosphere to give the monomeric S-H complexes.14 In the
analogous reaction, a solution of (S4Me)2 in toluene-d8 was
purged with hydrogen to test for conversion to S4MeH (Scheme
2, reaction e). The reaction occurred readily at room temperature,
with nearly 60% completion after 10 min total reaction time
and quantitative conversion of (S4Me)2 to S4MeH in 30 min.
In a similar experiment, an EPR spectrum of a solution of
(S4Me)2 in toluene was collected, and then the sample was
purged with hydrogen and the EPR spectrum again collected,
resulting in the EPR signal being reduced to the baseline. Both
of these experiments show that the (S4Me)2/S4Me• mixture can
be converted to the diamagnetic S4MeH by reaction with
hydrogen, and that both the NMR and EPR signals are consistent
with the assignments to (S4Me)2 and S4Me•.
The reaction of (S4Me)2 with hydrogen offers a clean, simple
route for the synthesis of S4MeH from S4Me+ by chemical
reduction and hydrogenation and does not require the chro-
matographic separation of S4MeH from S4Me2 and S4H2 that
is needed after methylation and protonation of S4. The conver-
sion of (S4Me)2 to S4MeH under 1 atm of hydrogen at room
temperature also demonstrates that the effective S-H homolytic
Figure 9. Eyring plot of site exchange rate from (S4Me)2 low temperature
NMR data, giving ∆H‡ of 17.1 ( 1.5 kcal/mol and ∆S‡ of 12.6 ( 6.0
cal/mol K.
methanethiolate and Cp* methyl resonances were simulated
from +10 to -33 °C using SpinWorks49 (details of the
simulations are included in the Supporting Information). An
Eyring plot of these data is shown in Figure 9 and results in
∆H‡ of 17.1 ( 1.5 kcal/mol and ∆S‡ of 12.6 ( 6.0 cal/mol K.
The Eyring plot contains data for both the Cp* methyl group
(49) Marat, K. SpinWorks, v. 2.5.5; 2005.
9
J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008 8947

A R T I C L E S
Appel et al.
capillary column and a flame-ionization detector. Elemental analyses
were performed by Desert Analytics in Tucson, Arizona.
bond dissociation free energy for S4MeH to the dimer/radical
mixture is greater than 52 kcal/mol, as the reaction goes to
completion (the sum of two S-H bond dissociation free energies
is greater than the homolytic bond dissociation free energy of
H2, which is approximately 104 kcal/mol).50 Given that gas
phase homolytic bond dissociation enthalpies are expected to
be 5 kcal/mol higher than solution bond dissociation free
energies,50 the S-H bond enthalpy is estimated to be greater
than 57 kcal/mol. Complete thermochemical studies of the
homolytic and heterolytic S-H bond dissociation free energies
of S4H2 and S4MeH are currently in progress. These studies
will provide additional insight into the reactivities of the S-H
bonds in Mo2S4 clusters.
Electrochemical data were collected using a CH Instruments
model 660C computer-aided three-electrode potentiostat in either
acetonitrile or THF with 0.3 M tetraethylammonium tetrafluoro-
borate or tetrabutylammonium tetrafluoroborate, respectively. For
cyclic voltammetry, the working electrode was a glassy carbon disk,
the counter electrode was a glassy carbon rod, and a silver chloride
coated silver wire was used as a pseudoreference electrode and
was separated from the main compartment by a Vycor disk (1/8
in. diameter) obtained from Bioanalytical Systems, Inc. Ferrocene
or decamethylferrocene was used as an internal reference with all
potentials reported versus the ferrocene/ferrocenium couple. In bulk
electrolysis experiments, the working electrode was a 1.2-cm
diameter, 2.5-cm tall cylindrical piece of 30 pore per inch reticulated
vitreous carbon purchased from ERG. The counter electrode was a
3 mm graphite rod purchased from Alfa-Aesar, and a silver chloride
coated silver wire was used as a pseudoreference electrode and
was separated from the main compartment by a 7-mm Vycor disk
purchased from Princeton Applied Research (Cat. No. G0070).
Materials. Reagents were purchased commercially and used
without further purification unless otherwise specified. All reactions,
syntheses, and manipulations of Mo2S4 complexes were carried out
under nitrogen using standard Schlenk techniques or in a glovebox.
All of the preparative chromatographic separations were run using
silica gel. Solvents were dried by activated alumina column in an
Innovative Technology, Inc., PureSolv system or by distillation from
calcium hydride (acetonitrile) or sodium and benzophenone (THF).
NMR solvents were distilled from calcium hydride or dried over
activated sieves, degassed, and stored in a glovebox. Benzene for
photolysis experiments was triply fractionally distilled to reduce
trace toluene to ∼10-7 M. Dibenzyl ketone (DBK) was recrystal-
lized from methanol. S4H2 was synthesized from [Cp*Mo(CO)2]2
by the literature methods7,9 and purified by column chromatography
using dichloromethane. A new synthetic route to S4 is described
below, and the spectroscopic data matches the previously reported
data.9 Additionally, the triflate salt of S4Me+ shows very similar
spectral data to the previously reported iodide analog, [Cp*Mo(µ-
SMe)(µ-S)(µ-S2)MoCp*](I).11
Conclusions
A variety of reactions have been demonstrated at the sulfur
centers in Mo2S4-based homogeneous analogs of HDS catalysts,
including hydrogen atom transfer, hydrogen addition, alkyl
group addition and transfer, and disulfide formation and
cleavage. All of these reactions are key steps in HDS catalysis
and are important for wide range of chemistry. For Mo2S4
complexes, the formation and cleavage of disulfide bonds help
to control the reactivity and form stable complexes. The
observed disulfide bridged tetranuclear complex, (S4Me)2, shows
a very weak disulfide bond dissociation free energy of 6.5 kcal/
mol. This weak disulfide bond formation results in a radical-
dimer equilibrium and gives rise to selective reactivity with
carbon centered radicals, typical of persistent free radicals. While
increasing the steric bulk of the complex and substituting a
methyl group for one of the hydrogen atoms did not completely
prevent dimerization, disproportionation is substantially slower,
although not eliminated. Understanding the kinetics and ther-
modynamics for the formation and cleavage of bonds to
hydrogen are key to understanding and developing many
catalytic and chemical processes, and the reactivities observed
in this study are particularly relevant to kinetics of S-H bond
cleavage in MonS2n clusters. Building on the chemistry and
observed reactivities in this work, experimental and computa-
tional studies of the thermodynamics of S-H and S-C bonds
in Cp*2Mo2S4 complexes are in progress and will be the subject
of future publications.
Synthesis of Cp*Mo(µ-S2)(µ-S)2MoCp*, S4. S4H2 (0.75 g, 1.3
mmol) was dissolved in 100 mL of dichloromethane, and tri-
fluoroacetic acid (0.40 mL, 5.4 mmol) was added, resulting in a
color change from red to violet. After stirring overnight, triethyl-
amine (0.75 mL, 5.4 mmol) was added, changing the solution from
violet to blue. The solvent was removed, and the solid was then
purified by column chromatography using 9:1 toluene and diethyl
1
ether. Yield: 0.45 g, 60%. H NMR (CD3CN, 500 MHz): δ 2.13
Experimental Procedures
(s, Cp*).
Synthesis of [Cp*Mo(µ-SMe)(µ-S)(µ-S2)MoCp*](OTf), S4Me+.
Methyl trifluoromethane sulfonate (0.043 mL, 0.38 mmol) was
added to a solution of S4 (0.22 g, 0.38 mmol) in 20 mL toluene.
The solution changed from blue to red-brown, was stirred for 10
min, and the solvent was evaporated. The solid was purified by
column chromatography with 1:4 acetonitrile and toluene to remove
a yellow-brown fraction, followed by elution with acetonitrile to
give a purple fraction. Yield: 0.215 g, 75%. Anal. Calcd for
C22H33S5O3F3Mo2: C, 35.01; H, 4.41. Found: C, 35.35; H, 4.31.
1H NMR (CD3CN, 500 MHz): δ 2.30 (s, 30.0 H, Cp*), 1.36 (s,
2.9 H, SCH3). MS (m/z, ESI positive ion): 605.95 (S4Me+). CV,
E1/2, V (in acetonitrile): -0.93 V (∆Ep ) 59 mV), -1.30 V (∆Ep
) 62 mV) at 0.5 V/s. CV, E1/2, V (in THF): -0.89 V (∆Ep ) 109
mV), -1.50 V (∆Ep ) 110 mV) at 0.5 V/s. Crystals suitable for
X-ray diffraction were grown by slow diffusion of pentane into a
toluene solution of S4Me+ at room temperature under nitrogen.
Instrumentation. Photolysis experiments were run using a 1-kW
Hanovia high pressure xenon arc lamp, and samples were prepared
in degassed benzene or benzene-d6 in Pyrex tubes. 1H NMR spectra
were acquired using a Varian Inova 500 or VXR-300 spectrometer,
and the chemical shifts were referenced to TMS using the residual
solvent peak as a secondary reference (1.94 ppm for acetonitrile-
d3, 2.09 ppm for toluene-d8, and 7.15 ppm for benzene-d6). EPR
spectra were recorded using a Bruker EMX 9 GHz spectrometer
with g values measured relative to solid DPPH (g ) 2.0036).
Magnetic susceptibilities were determined by the solution NMR
method developed by Evans.44,45 Electrospray ionization (ESI) mass
spectra were collected using a Thermo Finnigan TSQ 7000 or an
in-house-developed 11.5 T FTICR MS.51 GC analyses were carried
out using a Hewlett-Packard 5890 Series II GC equipped with a
splitless injector and a 30 m × 0.25 mm DB-5 high performance
Synthesis of Cp*Mo(µ-S)2(µ-SMe)(µ-SH)MoCp*, S4MeH.
Methyllithium (0.20 mL of 1.6 M in diethyl ether, 0.32 mmol) was
added to a solution of S4 (0.14 g, 0.24 mmol) in 30 mL benzene,
changing the blue solution to purple. The solution was stirred for
30 min and then acetic acid (18 µL, 0.31 mmol) was added, resulting
(50) Wayner, D. D. M.; Parker, V. D. Acc. Chem. Res. 1993, 26, 287–
294.
(51) Harkewicz, R.; Belov, M. E.; Anderson, G. A.; Pasa-Tolic, L.;
Masselon, C. D.; Prior, D. C.; Udseth, H. R.; Smith, R. D. J. Am.
Soc. Mass Spectrom. 2002, 13, 144–154.
9
8948 J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008

Persistent Radical in a Mo2S4 Complex
A R T I C L E S
in a red solution that was stirred for 1 h, and then the solvent was
removed under vacuum. The solid was purified by column
chromatography using benzene, yielding a mixture of methylated
and protonated Mo2S4 complexes. This mixture was separated by
preparative TLC using 1:4 dichloromethane and benzene, giving
an approximately 1:5:5 ratio of S4H2, S4MeH, and S4Me2. Yield:
0.065 g, 44%. Anal. Calcd for C21H34S4Mo2: C, 41.58; H, 5.65; S,
2.31, 2.30, 2.18, 2.16, 2.15, and 2.14 ppm integrating to a total of
30.0 H (with the peaks at 2.32 and 2.14 having twice the area of
the other six). Methane thiolate peaks were observed at 1.00 ppm
(singlet, 1.8 H), and 0.75 ppm (two peaks, separated by 0.006 ppm,
1.1 H total). At +10 °C or higher, the methane thiolate peaks
coalesced and were no longer observable, and the Cp* methyl peaks
coalesced into one broad peak that shifted downfield with increasing
temperature or decreasing concentration. Above 50 °C, significant
decomposition occurred within 15 min, resulting in the formation
of equal amounts of S4Me2 and anti-S4, as well as additional
unidentified products.
Reaction of (S4Me)2 with Hydrogen. A solution of (S4Me)2 in
toluene-d8 was purged with H2 (1 atm) for 5 min and then 1H NMR
spectra were acquired, showing 60% conversion to S4MeH after 5
min and complete conversion by 30 min. No other products were
observed. Similarly, a solution of (S4Me)2 in toluene was run by
EPR, purged with H2 (1 atm) for 10 min and run again, showing
loss of the EPR signal.
Rate Constants for the Reaction of Benzyl Radical with
S4MeH. Hydrogen atom abstraction kinetics were determined by
the previously reported method using S4MeH (1.3 × 10-4 M) as
the hydrogen atom donor.37 Benzyl radicals were produced by two
second photolysis of DBK (0.02 M), and the rate of hydrogen atom
abstraction was determined by competition relative to the rate of
formation of bibenzyl. Rate constants were calculated using density-
corrected concentrations of toluene, bibenzyl, initial S4MeH
concentration (DH) using eq 11 of Scheme 3.
1
21.1. Found: C, 41.65; H, 5.59; S, 20.4. H NMR (CD3CN, 500
MHz): δ 2.20 (s, 30.0 H, Cp*), 1.07 (s, 1.0 H, SCH3, isomer B),
0.99 (s, 1.8 H, SCH3, isomer A) -2.33 (s, 0.3 H, SH, isomer B),
-2.51 (s, 0.6 H, SH, isomer A). MS (m/z, ESI positive ion): 605.95
(S4Me+).
Synthesis of Cp*Mo(µ-S)2(µ-SMe)(µ-SBz)MoCp*, S4MeBz.
Benzylmagnesium chloride (0.040 mL of 2.0 M in THF, 0.080
mmol) was added to a solution of S4Me+ (0.058 g, 0.077 mmol)
in 20 mL of THF. The resulting red-orange solution was stirred
for 30 min, filtered through a plug of glass wool, and the solvent
was removed under vacuum to give a red solid. Yield: 0.042 g,
77%. Anal. Calcd for C28H40S4Mo2: C, 48.27; H, 5.79; S, 18.4.
1
Found: C, 48.74; H, 5.51; S, 18.6. H NMR (C6D6, 500 MHz):
7.08 (t, 2.2 H, ArH), 7.03 (t, 2.0 H, ArH), 6.94 (d, 0.9 H, ArH),
2.82 (s, 0.1 H, SCH2Ph, isomer B), 2.78 (s, 1.9 H, SCH2Ph, isomer
A), 2.16 (s, 27.6 H, Cp*, isomer A), 2.14 (s, 2.4 H, Cp*, iso-
mer B), 1.14 (s, 0.2 H, SCH3, isomer B), 1.00 (s, 2.8 H, SCH3,
isomer A). Crystals suitable for X-ray diffraction were grown from
a benzene solution of S4MeBz by slow evaporation at room
temperature under nitrogen.
Photolysis of DBK and S4MeH for Product Analysis. A
Synthesis of (Cp*Mo(µ-S)2(µ-SMe)MoCp*)2(µ-S2), (S4Me)2.
Method 1: Chemical Reduction. S4Me+ (0.215 g, 0.285 mmol)
and bis(benzene)chromium (0) (0.058 g, 0.279 mmol) were stirred
in 15 mL of acetonitrile for 5 h. The resulting purple precipitate
was collected by filtration, dried under vacuum, and used without
purification. Yield: 0.155 g, 92%. Anal. Calcd for C21H33S4Mo2:
C, 41.65; H, 5.49; S, 21.2. Found: C, 40.56; H, 5.40; S, 20.8.
Subsequent analyses of the same sample with time resulted in
progressively lower values, suggesting the sample was highly
reactive, as per discussions with Desert Analytics. MS (m/z, ESI
positive ion): 605.95 (S4Me+). CV, E1/2, V (in THF): -0.90 V
(∆Ep ) 96 mV), -1.49 V (∆Ep ) 102 mV) at 0.5 V/s. Crystals
suitable for X-ray diffraction were grown by slow diffusion of
acetonitrile into a toluene solution of (S4Me)2 at room temperature
under nitrogen.
solution of S4MeH (0.0015 M) and DBK (0.014 M) was photolyzed
for 0 to 150 s and analyzed by H NMR to obtain the spectra in
Figure 3. The products observed were toluene, bibenzyl, phenyl
1
acetaldehyde, S4MeBz, and an additional product with Cp*
1
resonance at 1.90 ppm. Monitoring this mixture by H NMR for
one week showed ∼75% conversion of the product at 1.90 ppm to
S4MeBz.
Photolysis of S4MeBz and DBK in Benzene-d6 To Detect
Photo-Conversion. A solution of S4MeBz and DBK was photo-
lyzed for 300 s and analyzed by 1H NMR. Approximately 20% of
the S4MeBz was converted to the product with a Cp* resonance
1.90 ppm, matching that described the photolysis of DBK and
S4MeH.
Photolysis of S4MeBz in Benzene-d6 To Detect Photo-
Decomposition. A solution of S4MeBz without DBK was photolyzed
for 450 s and analyzed by 1H NMR. The ratio of isomers A and B
shifted from ∼2:1 to ∼2:3 during the photolysis. No other changes
were observed.
Method 2: Bulk Electrolysis. In a typical experiment, S4Me+
(0.037 g, 0.049 mmol) was dissolved in 0.3 M tetraethylammonium
tetrafluoroborate in acetonitrile and reduced by constant potential
electrolysis at -1.06 V vs FeCp2 (100 mV negative of the peak
for the reduction wave), passing 3.9 C (4.3 C was expected for a
one electron reduction). The resulting precipitate was collected by
filtration and washed with minimal acetonitrile to remove electro-
lyte. Spectroscopic data matched the product of method 1.
Magnetic Susceptibility of (S4Me)2. A solution of (S4Me)2
(0.018 M in toluene at 26 ( 1 °C) was used to determine the
Photolysis of S4MeH in Benzene-d6 To Detect Photo-
Decomposition. A solution of S4MeH in was photolyzed for 300 s
1
and analyzed by H NMR. The ratio of isomers A and B shifted
from ∼2:1 to 1:1 during the photolysis. No other changes were
observed.
Photolysis of (S4Me)2 in Benzene-d6 To Detect Photo-
Decomposition. A solution of (S4Me)2 was photolyzed for 250 s
1
magnetic susceptibility by the Evans method using H NMR.44,45
1
The solution was transferred into a 5-mm NMR tube with a sealed
capillary of pure toluene. Only one methyl peak could be observed
for toluene (3 Hz line width), whereas two would be expected for
a paramagnetic sample at this concentration.
and analyzed by H NMR. No changes were observed.
Single Crystal X-ray Diffraction. Crystals of S4Me+, S4MeBz,
and (S4Me)2 were removed from the flask, a suitable crystal was
selected, attached to a glass fiber and data were collected at 90(2)
K using a Bruker/Siemens SMART APEX instrument (Mo KR
radiation, λ ) 0.71073 Å) equipped with a Cryocool NeverIce low
temperature device. Data were measured using omega scans 0.3°
per frame for 30, 30, and 5 s respectively, and a full sphere of data
was collected for all. A total of 2400 frames were collected with a
final resolution of 0.83 Å. Details of the data collections and
refinements are given in Table 5. Further details are provided in
the Supporting Information.
EPR of (S4Me)2. EPR spectra were collected for (S4Me)2 in
toluene at three concentrations (0.016, 0.0041, and 0.0011 M at 23
( 3 °C): g ) 2.003. TEMPO (8.4 × 10-5 M) was run in the same
EPR tube in a separate experiment, and the data for TEMPO and
(S4Me)2/S4Me· were double integrated and used to determine the
concentration of S4Me· in each sample. The data from nine
experiments were combined to give Kdim ) 5.9 × 104 ( 2.3 × 104
M-1, where the error is two times the standard deviation.
1
Variable Temperature NMR of (S4Me)2. H NMR spectra of
S4Me+. Cell parameters were retrieved using SMART soft-
ware.52 The data were rotationally twinned and were deconvoluted
using CELL_NOW,53 giving a two component twin relationship:
2° rotation about the reciprocal axis -0.959, -0.127, 1.000, with
(S4Me)2 were collected in toluene-d8 at 0.006 or 0.001 M total
concentration, assuming all (S4Me)2. At or below -33 °C, the
spectrum was resolved into eight Cp* methyl singlets at 2.33, 2.32,
9
J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008 8949

A R T I C L E S
Appel et al.
Table 5. Crystal Data and Refinement Parameters for S4Me+, S4MeBz, and (S4Me)2
complex
S4Me+
S4MeBz
(S4Me)2
formula
fw (amu)
T (K)
C47.50 H70 F6 Mo4 O6 S10
1555.40
90(2)
C34 H46 Mo2 S4
774.83
90(2)
C42 H66 Mo4 S8
1211.19
90(2)
λ (Mo KR) (Å)
crystal system
space group
unit cell dimensions
a (Å)
0.71073
monoclinic
C2/c
0.71073
triclinic
P1
0.71073
tetragonal
P 41 21 2
j
33.703(5)
20.664(3)
23.485(4)
90
132.887(2)
90
11984(3)
8
1.724
1.229
10.5202(4)
12.5398(5)
26.2385(10)
87.631(2)
84.018(2)
78.866(1)
3377.0(2)
4
1.524
1.012
11.6316(4)
11.6316(4)
35.549(2)
90
90
90
4809.6(4)
4
1.673
1.396
2456
b (Å)
c (Å)
R (deg)
ꢀ (deg)
γ (deg)
volume (Å3)
Z
F, calcd (Mg/m3)
µ, (mm-1
F(000)
)
6280
1592
crystal size (mm3)
crystal color and habit
θ range
0.54 × 0.08 × 0.04
red needle
1.65 to 25.25°
-40 e h′29
0 e k e 24
0 e l e 28
136982
0.28 × 0.14 × 0.06
red fragment
1.81 to 25.25°
-12 e h e 12
-15 e k e 15
-31 e l e 31
48347
0.33 × 0.24 × 0.11
red plate
1.84 to 25.25°
-13 e h e 13
-13 e k e 13
-42 e l e 42
68317
index ranges
reflections collected
independent reflections
completeness to theta ) 25.25°
max. and min. transmission
data/restraints/parameters
goodness-of-fit on F2
final Ra indices
26319 [R(int) ) 0.0000]
12243 [R(int) ) 0.0326]
4348 [R(int) ) 0.0361]
99.9%
100.0%
100.0%
0.9525 and 0.5565
26319/7/675
0.966
R1 ) 0.0651,
wR2 ) 0.1555
2.950 and -1.899
0.9417 and 0.7647
12243/7/764
1.048
R1 ) 0.0362,
wR2 ) 0.0774
0.758 and -0.532
0.8616 and 0.6559
4348/1/286
1.059
R1 ) 0.0249,
wR2 ) 0.0604
0.764 and -0.453
[I > 2σ(I)]
largest diff. peak and hole (e-/Å3)
2
2 2
a R1 ) Σ|Fo| - |Fc|/Σ|Fo|; wR2 ) {Σ[ w(Fo - Fc ) ]/Σ [w(Fo2)2]}1/2
.
a refined twinning ratio of 0.2769(5). The matrix used to relate the
second orientation to the first domain is: 1.004, 0.041, 0.009,
-0.029, 0.999, -0.028, -0.005, -0.001, 0.995. Each cell com-
ponent was refined using SAINTPlus54 on all observed reflections.
Data reduction and correction for Lp and decay were performed
using the SAINTPlus software. Absorption corrections were applied
using TWINABS.55 The structure was solved by direct methods
and refined by least-squares method on F2 using the SHELXTL
program package.56 The structure was solved in the space group
C2/c (#15) by analysis of systematic absences. One triflate group
is disordered and was modeled in two positions with occupancies
of 57/43% using restraints. Only the sulfur atom was refined
anisotropically. The toluene solvent molecule was modeled using
rigid coordinates, held isotropic with 50% occupancy. The sulfur
bridges in the Mo3/Mo4 unit were disordered and modeled in two
positions with occupancies of 87/13% (S5, S6, S7) and refined
anisotropically with restraints. All other non-hydrogen atoms were
refined anisotropically. No decomposition was observed during data
collection.
refined by least-squares method on F2 using the SHELXTL program
package.56 The structure was solved in the space group P1 (#2) by
j
analysis of systematic absences. All non-hydrogen atoms were
refined anisotropically. The central sulfur system is disordered with
the two positions for the S, Me, and methylene groups as well as
one benzyl group (89:11% occupancy). Soft restraints were applied
to the minor component to stabilize geometry. No decomposition
was observed during data collection.
(S4Me)2. Cell parameters were retrieved using SMART soft-
ware52 and refined using SAINTPlus54 on all observed reflections.
Data reduction and correction for Lp and decay were performed
using the SAINTPlus software. Absorption corrections were applied
using SADABS.57 The structure was solved by direct methods and
refined by least-squares method on F2 using the SHELXTL program
package.56 The structure was solved in the space group P4(1)2(1)2
(#92) by analysis of systematic absences. The FLACK parameter
was refined using TWIN/BASF resulting in a value of 0.00(5). The
central sulfur atoms S1-S4 were disordered 50% in the first moiety
and 50% for S1b, S2b, and S4b. S3 was further disordered into
two positions, S3b and S3c, at 25% occupancy each. The disordered
methyl groups were site refined to give C21a, 50%, and C21b, C21c,
25%. There are several close inter X. . .Y contacts due to this
disorder which implies disorder of the Cp* ring. As the SMe
disorder is 25%, further disorder modeling of the Cp* methyl groups
was not pursued. Disordered carbon atoms as well as S3, S3b, and
S3c were held isotropic. Soft restraints were applied to S-C
distances. All other non-hydrogen atoms were refined anisotropi-
cally. No decomposition was observed during data collection.
S4MeBz. Cell parameters were retrieved using SMART soft-
ware52 and refined using SAINTPlus54 on all observed reflections.
Data reduction and correction for Lp and decay were performed
using the SAINTPlus software. Absorption corrections were applied
using SADABS.57 The structure was solved by direct methods and
(52) SMART: v. 5.632, Bruker AXS: Madison, WI, 2005.
(53) Sheldrick, G. M. CELL_NOW: 2002.
(54) SAINTPlus: v. 7.23a, Data Reduction and Correction Program; Bruker
AXS: Madison, WI, 2004.
(55) Sheldrick, G. M. TWINABS: v.1.05, an empirical absorption correction
program, Bruker AXS, Inc.: Madison, WI, 2002.
(56) Sheldrick, G. M. SHELXTL: v. 6.14, Structure Determination Software
Suite, Bruker AXS, Inc.: Madison, WI, 2004.
Acknowledgment. This work was supported by the U.S.
Department of Energy’s (DOE) Office of Basic Energy Sciences,
Chemical Sciences Program. The Pacific Northwest National
Laboratory is operated by Battelle for DOE. The Bruker (Siemens)
(57) SADABS: v. 2004/1, an empirical absorption correction program,
Bruker AXS, Inc.: Madison, WI, 2004.
9
8950 J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008

Persistent Radical in a Mo2S4 Complex
A R T I C L E S
SMART APEX diffraction facility was established at the University
of Idaho with the assistance of the NSF-EPSCoR program and the
M. J. Murdock Charitable Trust, Vancouver, WA, USA. The authors
thank Blandina Valenzuela and Rui Zhang for analysis of samples
by ESI-MS.
crystallographic figures, as well as X-ray data collection,
refinement results, atomic coordinates, anisotropic displacement
parameters, and complete bond distances and angles in CIF
format for S4Me+, S4MeBz, and (S4Me)2. This material is
available free of charge via the Internet at http://pubs.acs.
org.
Supporting Information Available: Tables of kinetic data,
details of the dynamic NMR simulations, and additional
JA078115R
9
J. AM. CHEM. SOC. VOL. 130, NO. 28, 2008 8951