Addition of polysubstituted alkenes, aromatic alkynes, and dienes to a metal-stabilized thiyl radical via carbon-sulfur bond formation: Electrochemical, chemical, and computational investigations

Article abstract of DOI:10.1016/j.ica.2013.08.003

The addition of polysubstituted alkenes, alkynes, and dienes to the metal-stabilized thiyl radical complex [Ru-1]⁺ was evaluated by electrochemical, chemical, and computational methods. The [Ru-1]⁺ radical complex was generated by two successive one-electron oxidations of the metal thiolate precursor PPN[Ru(DPPBT)₃], PPN[Ru-1]^-, (PPN = bis(triphenylphosphine)iminium; DPPBT = diphenylphosphinebenzenethiolato). Rate constants were experimentally determined by cyclic voltammetric methods over multiple scan rates using [Ru-1]^- solutions containing unsaturated substrates. Rate constants for polysubstituted alkenes range from 1.1 × 10¹ to 8.8 × 10² M^-1 s ^-1, which are at 3-5 orders of magnitude slower than their monosubstituted counterparts. Rate constants are slower for substituted alkynes varying from 2.3 × 10² to 1.2 × 10⁴ M ^-1 s^-1, which are ～100 times slower than the corresponding alkenes. Selected complexes were spectroscopically characterized following synthesis by chemical oxidation methods. Addition of asymmetrically substituted alkenes and alkynes yield [Ru-1·substrate]⁺ products as a mixture of isomers, which were assigned based on ³¹P NMR and density functional theory calculations (B3LYP/LANL2DZ + 6-31 g). The solid state structure of [Ru-1·cyclohexene]PF₆ was determined by single crystal X-ray diffraction.

Full text of DOI:10.1016/j.ica.2013.08.003

Inorganica Chimica Acta 408 (2013) 1–8
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Addition of polysubstituted alkenes, aromatic alkynes, and dienes to a
metal-stabilized thiyl radical via carbon–sulfur bond formation:
Electrochemical, chemical, and computational investigations
⇑
Kagna Ouch Sampson, Davinder Kumar, Mark S. Mashuta, Craig A. Grapperhaus
Department of Chemistry, University of Louisville, Louisville, KY 40292, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 June 2013
Received in revised form 1 August 2013
Accepted 2 August 2013
Available online 19 August 2013
The addition of polysubstituted alkenes, alkynes, and dienes to the metal-stabilized thiyl radical complex
[Ru-1]+ was evaluated by electrochemical, chemical, and computational methods. The [Ru-1]+ radical
complex was generated by two successive one-electron oxidations of the metal thiolate precursor
PPN[Ru(DPPBT)3], PPN[Ru-1]À, (PPN = bis(triphenylphosphine)iminium; DPPBT = diphenylphosphine-
benzenethiolato). Rate constants were experimentally determined by cyclic voltammetric methods over
multiple scan rates using [Ru-1]À solutions containing unsaturated substrates. Rate constants for poly-
substituted alkenes range from 1.1 Â 101 to 8.8 Â 102 MÀ1 sÀ1, which are at 3–5 orders of magnitude
slower than their monosubstituted counterparts. Rate constants are slower for substituted alkynes vary-
ing from 2.3 Â 102 to 1.2 Â 104 MÀ1 sÀ1, which are ꢀ100 times slower than the corresponding alkenes.
Selected complexes were spectroscopically characterized following synthesis by chemical oxidation
methods. Addition of asymmetrically substituted alkenes and alkynes yield [Ru-1Ásubstrate]+ products
as a mixture of isomers, which were assigned based on 31P NMR and density functional theory calcula-
tions (B3LYP/LANL2DZ + 6–31 g). The solid state structure of [Ru-1Ácyclohexene]PF6 was determined
by single crystal X-ray diffraction.
Keywords:
Metal-stabilized thiyl radicals
C–S bond formation
Electrochemistry
Alkenes
Alkynes
Dienes
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
[Ru-1]+ and the substrate with concomitant oxidation of the C–C
multiple bond. Similar reactions have been noted by Webster,
Metal-stabilized thiyl radicals are a special class of oxidized me-
tal-thiolates in which an unpaired electron resides in a molecular
orbital with significant contributions from both the metal and sul-
fur [1]. Previously we reported the formation of a Ru-stabilized thi-
yl radical complex, [Ru-1]+, upon oxidation of the metal–thiolate
precursor [Ru-1]À by subsequent one electron steps [2]. Density
functional theory calculations on [Ru-1]+ suggest a singlet diradi-
cal ground state with spin-densities of 0.84 on Ru and 0.99 over
the three sulfur donors consistent with a metal-stabilized thiyl
radical description [3]. The sulfur-centered spin-density imparts li-
gand-centered reactivity similar to that of organic thiyl radicals,
which are known to react with unsaturated hydrocarbons with
applications including cis/trans isomerization, sulfide synthesis,
and polymerization [4–7].
Goh and co-workers upon chemical oxidation of a ruthenium thio-
late [10]. Electrochemical studies employing a variety of mono-
functionalized and cyclic alkenes revealed enhanced reaction
rates with electron-donating substituents and decreased rates with
electron-withdrawing groups [1]. For cyclic alkenes, increased
reaction rates correlate with increased ring strain. All alkene addi-
tions to [Ru-1]+ in this and prior studies are irreversible, although
the related rhenium derivative [Re-1]+ adds alkenes in a less rapid,
but reversible process [11].
In the current manuscript, we extend the scope of substrate
reactivity for [Ru-1]+. In prior studies, we evaluated the effect of
substituent groups on monofunctionalized and cyclic alkenes and
the formation of spectroscopically distinct isomers for asymmetri-
cally substituted substrates (Scheme 1) [1,9]. In the preferred iso-
mer, the R group sits nearest the S trans to S, while the minor
isomer finds R nearest the S trans to P. Herein, we investigate a ser-
ies of polysubstituted alkenes (including cis- and trans-isomers),
aromatic and disubstituted alkynes, and dienes. We also report a
general synthetic method employing chemical oxidants to
generate [Ru-1Ásubstrate]+ products, which were characterized
spectroscopically. The structure of one derivative, [Ru-1Ácyclohex-
ene]PF6, was determined by single crystal X-ray diffraction.
Previously we reported the addition of ethylene, [3,8] mono-
functionalized and cyclic alkenes, [1,8] and 1-octyne [9] to the me-
tal-stabilized thiyl radical [Ru-1]+. The resulting [Ru-1Ásubstrate]+
product complexes are formed via C–S bond formation between
⇑
Corresponding author. Tel.: +1 5028525932; fax: +1 5028528149.
E-mail address: grapperhaus@louisville.edu (C.A. Grapperhaus).
0020-1693/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ica.2013.08.003

2
K.O. Sampson et al. / Inorganica Chimica Acta 408 (2013) 1–8
cyclic voltammograms as described in Supplementary information
using the DigiSim simulation program [15].
In a typical experiment, a 2.0 mM solution of [PPN][Ru-1] in
acetonitrile was added to the three electrode electrochemical cell.
An initial voltammogram was collected at a scan rate of 200 mV/s
using a pre-equilibrium time of 15 s, an initial and final potential of
À600 mV versus the pseudo reference electrode, and a switching
potential of +1200 mV. Within this window two events assigned
as [Ru-1]À/0 and [Ru-1]+/0, respectively, were observed. These
peaks were used to scale the voltammogram versus ferrocenium/
ferrocene (Fc+/Fc) using previously reported potentials of À830
and À20 mV for the two redox events, respectively (Fig. S1). Next,
an excess of substrate was added to give a 0.20 M solution of the
substrate. The solution was mixed by purging with nitrogen gas
for 3 min. For reactive substrates, collection of the cyclic voltam-
mogram reveals three redox events. Two events are attributed to
[Ru-1]À/0 and [Ru-1]+/0, although the latter is shifted to lower
potentials due to the rapid chemical reaction of [Ru-1]+. A new re-
dox event assigned to a RuIII/RuII dithioether complex was ob-
served between +260 and +320 mV (versus ferrocenium/
ferrocene). Data were collected at several scan rates for evaluation
of the chemical addition rate constant. Full simulation details in
text and tabular (Table S1) form are provided in the Supplementary
data along with representative experimental and simulated cyclic
voltammograms (Fig. 1 and Figs. S2–S8).
Scheme 1. Addition of unsaturated hydrocarbons to [Ru-1]+.
2. Experimental
2.1. Materials
2.4. X-ray crystallographic methods
A cut yellow plate 0.21 Â 0.17 Â 0.04 mm3 crystal of [Ru-
1Ácyclohexene]PF6 was mounted on a CryoLoop for collection of
X-ray data on an Agilent Technologies Gemini CCD diffractometer.
The CrysAlisPRO (CCD) [16] software package (v 1.171.36.28) was
Chemical reagents were obtained from commercial sources and
used without further purification unless otherwise stated. Re-
agent-grade solvents were dried utilizing the standard techniques
and freshly distilled under nitrogen prior to use. Ferrocenium
hexafluorophosphate (FcPF6) was purchased from Aldrich, recrys-
tallized, [12] and stored in an argon filled dry box. Complexes
[PPN][Ru-1] [13] and [HNEt3][Ru-1] [14] were prepared as previ-
ously reported. Deuterated acetonitrile was purchased from Cam-
bridge Isotope Laboratories, Inc. and used as received. All
reactions were performed using standard Schlenk techniques un-
der an argon atmosphere or in an argon filled dry box unless other-
wise noted.
used to acquire a total of 674 thirty-second frame
sures of data at 100 K to a 2h max = 55.24° using monochromated
Mo K radiation (0.71073 Å) from a sealed tube. Frame data were
x-scan expo-
a
processed using CrysAlisPRO (RED) [16] (v 1.171.36.28) to
determine final unit cell parameters: a = 12.4717(6) Å, b =
12.7535(6) Å, c = 18.2603(9) Å,
= 89.193(4)°, V = 1866.7(3) Å3, Dcalc = 1.538 Mg/m3, Z = 2 to pro-
duce raw hkl data that were then corrected for absorption (trans-
mission minimum/maximum 0.903/1.00;
= 0.664 mmÀ1
a = 89.638(4)°, b = 74.032(4)°,
c
=
l
)
2.2. Physical methods
using SCALE
3
ABSPACK [17]. The structure was solved by Patterson
ꢀ
methods in the space group P1 using SHELXS-90 [18] and refined
by least squares methods on F2 using SHELXL-97 [18] incorporated
into the SHELXTL [19] (v 6.14) suite of programs. Phenyl, methylene
and methine hydrogen atoms were placed in their geometrically
generated positions and refined as a riding model that were in-
cluded as fixed contributions with U(H) = 1.2 Â Ueq (attached C
IR spectra were measured with a Thermo Nicolet Avatar 360
spectrometer at 4 cmÀ1 resolution. Electronic absorption spectra
were obtained with an Agilent 8453 diode array spectrometer uti-
lizing a custom designed cell with 0.5 cm path length quartz sam-
ple compartment. NMR spectra were recorded on
a Varian
400 MHz spectrometer and referenced to TMS (1H NMR) or 85%
phosphoric acid (31P NMR). Elemental analyses were performed
by Midwest Microlab (Indianapolis, IN). Electrospray ionization
mass spectra (ESI-MS) were collected by the Mass Spectrometry
Application and Collaboration Facility in the Chemistry Depart-
ment at Texas A&M University.
2.3. Electrochemical measurements
Cyclic voltammograms were recorded using a PAR 273 poten-
tiostat with a three electrode cell (glassy carbon working electrode,
platinum wire counter electrode and Ag/AgCl pseudo reference
electrode). Samples were measured in acetonitrile solution with
tetra-n-butylammonium hexafluorophosphate (TBAHFP) (0.1 M)
as supporting electrolyte. All potentials were scaled to a ferroce-
nium/ferrocene standard (Fc+/Fc) as an internal reference. Experi-
mental rate constants were determined by digital simulation of
Fig. 1. Experimental (À) and simulated (---) cyclic voltammograms of [Ru-1]À in
presence of cis-stilbene at a scan rate 400 mV/s.

K.O. Sampson et al. / Inorganica Chimica Acta 408 (2013) 1–8
3
atom). For 14961 unique reflections (R(int) 0.040) the final aniso-
tropic full matrix least-squares refinement on F2 for 682 variables
converged at R1 = 0.055 and wR2 = 0.109 with a Goodness-of-fit of
1.08. Crystallographic details are summarized in Table 1.
cannula. After stirring for 3 h, the solvent was removed by rotary
evaporation to yield a yellow green residue. Crude product was
washed with an excess of hot water (ꢀ300 mL) and diethyl ether
(25 mL). Yield: 85 mg (76%). Single crystals suitable for X-ray dif-
fraction studies were grown through slow evaporation into 1:1
CH2Cl2:hexane. E1/2 (RuIII/RuII) = 269 mV. +ESI-MS for [C60H52P3S3-
Ru]+: theoretical m/z (z = 1) 1063.1487; observed 1063.1683. Ele-
ment Anal. Calc. for C60H52P4S3F6Ru: C, 59.64; H, 4.35%. Found: C,
2.5. Computational methodology
All computational processes were performed using the GAUSSIAN
09 suite of programs [20]. Density functional theory (DFT) calcula-
tions employed the B3LYP functional which utilizes three parame-
ters Becke Functional (B3) with the Lee–Yang–Parr (LYP)
generalized gradient correlation functional. A mixed basis set, 6-
31 G for carbon, hydrogen, sulfur and phosphorus and LANL2DZ
for ruthenium, was used. Furthermore, all optimized geometries
are genuine minima with no imaginary vibrational frequencies.
The intrinsic coordinates for all calculations were derived from
[Ru-1Ávinyl] [21] with replacement of the vinyl substituent with
the appropriate alkene derived moiety. To determine the barrier
for conversion between the corresponding k and d isomers, the
dihedral angle was frozen between the atoms 105 (carbon), 2 (sul-
fur), 3 (sulfur), and 104 (carbon) during the geometry optimization.
59.9; H, 4.75%. Electronic absorption: kmax (e
, MÀ1 cmÀ1): 367 nm
(6993), 440 nm (shoulder). FT-IR (KBr pellet, cmÀ1) 3052, 2925,
2852, 1434, 1086, 845, 559, 518. 31P{1H} NMR (162 MHz, CD3CN):
d 36.5 (J = 29, 313 Hz, Pax1) 39.5 (J = 29, 313 Hz, Pax2), 54.8 (J = 29,
29 Hz, Peq).
2.6.2. [Ru-1Ácis stilbene][PF6] (2)
To a 40 mL solution of [HNEt3][Ru-1] (100 mg, 0.0935 mmol) in
acetonitrile with cis-stilbene (1.67 mL, 9.35 mmol) was slowly
added to a 30 mL of FcPF6 (0.0619 g, 0.187 mmol) in acetonitrile
via cannula transfer. The mixture was cooled in an ice bath and al-
lowed to stir overnight. The color of solution transformed from yel-
low to yellow-green. The solvent was removed through use of a
rotary evaporator, followed by washing with an excess of hot water
(300 mL) and diethyl ether (100 mL). Yield: 69 mg (57%). E1/2 (RuIII/
RuII) = 361 mV. +ESI-MS for [C68H54P3S3Ru]+: theoretical m/z (z = 1)
1161.1644; observed 1161.1980. Element Anal. Calc. for C68H54P4-
S3F6Ru: C, 62.52; H, 4.17%. Found: C, 63.17; H, 5.63%. Electronic
2.6. Chemical syntheses
2.6.1. [Ru-1Ácyclohexene][PF6] (1)
To a suspension of 100 mg (0.0935 mmol) of [HNEt3][Ru-1] in
40 mL of dry acetonitrile was added 0.947 mL of cyclohexene
(9.35 mmol) via syringe. The solution was cooled in an ice bath,
absorption: kmax (e
, MÀ1 cmÀ1): 366 nm (10093). FT-IR (KBr pellet,
cmÀ1) 3052, 2917, 2848, 1434, 1091, 837, 694, 522. 31P{1H} NMR
and
a
blue solution of ferrocenium hexafluorophosphate
(162 MHz, CD3CN): major isomer (70%) d 35.6 (J = 29, 315 Hz,
(0.0619 mg, 0.187 mmol) in 30 mL of acetonitrile was added via
Pax1) 42.9 (J = 29, 315 Hz, Pax2), 52.5 (J = 29, 29 Hz, Peq); minor iso-
mer (30%) d 35.2 (J = 30, 317 Hz, Pax1) 43.7 (J = 30, 317 Hz, Pax2),
54.0 (J = 30, 30 Hz, Peq).
Table 1
Crystal data and structure refinement for [Ru-1Ácyclohexene]PF6.
2.6.3. [Ru-1Átrans stilbene][PF6] (3)
Empirical formula
Formula weight
T (K)
C61H54Cl2F6P4RuS3
1293.07
100.0
The complex is prepared by the method mentioned above ex-
cept that trans-stilbene (1.69 g, 0.187 mmol) was added instead
of cis-stilbene. Yield: 85 mg (70%). E1/2 (RuIII/RuII) = 337 mV. +ESI-
MS for [C68H54P3S3Ru]+: theoretical m/z (z = 1) 1161.34; observed
1161.17. Element Anal. Calc. for C68H54P4S3F6Ru: C, 62.52; H,
k (Å)
0.71073
Crystal system
Space group
Unit cell dimensions
a (Å)
b (Å)
c (Å)
triclinic
ꢀ
P1
12.4717(6)
12.7535(6)
18.2603(9)
89.638(4)
4.17%. Found: C, 58.63; H, 4.30%. Electronic absorption: kmax (e,
MÀ1 cmÀ1): 370 nm (6811), 441 nm (shoulder). FT-IR (KBr pellet,
cmÀ1) 3040, 2921, 2851, 1484, 1091, 833, 698, 523. 31P{1H} NMR
(162 MHz, CD3CN): major isomer (69%) d 35.6 (J = 30, 303 Hz,
a
(°)
b (°)
74.032(4)
P
ax1) 38.5 (J = 30, 303 Hz, Pax2), 57.9 (J = 30, 30 Hz, Peq); minor iso-
c
(°)
89.193(4)
2792.1(2)
2
1.538
0.664
1320
0.21 Â 0.17 Â 0.04
yellow plate
3.23–30.04
À16 6 h 6 17
À16 6 k 6 17
À24 6 l 6 25
47997
V (Å3)
mer (31%) d 35.2 (J = 30, 318 Hz, Pax1) 43.7 (J = 30, 318 Hz, Pax2),
54.1 (J = 30, 30 Hz, Peq).
Z
Dcalc (Mg/m3)
Absorption coefficient (mmÀ1
F (000)
)
2.6.4. [Ru-1Áphenylacetylene][PF6] (4)
Crystal size (mm3)
Crystal color, habit
To a 100 mg sample of [HNEt3][Ru-1] (0.0935 mmol) in 40 mL
acetonitrile in a Schlenk flask was transferred phenylacetylene
(1.38 mL, 9.35 mmol) via syringe. After addition, the flask was
cooled in an ice bath and a blue solution of FcPF6 (0.0619 mg,
0.0817 mmol) in 30 mL acetonitrile was added. The reaction solu-
tion was stirred overnight after which the solvent was removed
by rotary evaporation to produce a yellow green residue. The res-
idue was washed three times with hot water (ꢀ100 mL) and
diethyl ether (25 mL). The crude product was recrystallized via a
mixture of THF/hexane yielding a yellow solid (49 mg, 42%). E1/2
(RuIII/RuII) = 361 mV. +ESI-MS for [C62H48P3S3Ru]+: theoretical m/z
(z = 1) 1083.1209; observed: 1083.1174. Element Anal. Calc. for
h Range for data collection (°)
Index ranges
Reflections collected
Independent reflections
Completeness to h = 27.62°
Absorption correction
Maximum and minimum transmission
Refinement method
14961 [R(int) = 0.0407]
99.8%
Semi-empirical from equivalents
1.000 and 0.903
Full-matrix least-squares on F2
14961/0/682
1.093
R1 = 0.0549, wR2 = 0.1090
R1 = 0.0992, wR2 = 0.1265
1.799 and À1.603
Data/restraints/parameters
Goodness-of-fit (GOF) on F2
Final R indices [I > 2
R indices (all data)a,b
Largest difference peak and hole (e ÅÀ3
r
(I)]a,b
C
62H48P4S3F6Ru: C, 60.62; H, 3.95. Found: C, 57.32; H, 4.28%. Elec-
tronic absorption: kmax (e
, MÀ1 cmÀ1): 305 nm (6919), 437 nm
)
(shoulder). FT-IR (KBr pellet, cmÀ1) 3052, 2917, 2848, 1434,
1091, 837, 694, 523. 31P{1H} NMR (162 MHz, CD3CN): major isomer
(57%) d 36.5 (J = 30, 312 Hz, Pax1) 42.7 (J = 30, 312 Hz, Pax2), 56.7
a
R1
wR2 = {
GOF = S = {
=
R
||Fo| À |Fc||/
R|Fo|.
[w(Fo À Fc2)2]/
R
[w(Fo2)2]}1/2
;
where
w = q/r
2(Fo2) + (qp)2 + bp.
b
2
R
2
R
[w(Fo À Fc2)2]/(n À p)1/2
.

4
K.O. Sampson et al. / Inorganica Chimica Acta 408 (2013) 1–8
tions and decreases the current magnitude of the return cathodic
peak. In addition, a new redox event assigned as the formal
RuIII/II redox couple for the product complex is observed between
+260 and +320 mV (E4) depending on the substrate as shown in
Fig. 1. The potential is shifted by +1090 to +1150 mV relative to
E1 consistent with the modification of two thiolate donors to thio-
ethers [8,13,22,23].
For each substrate, a series of cyclic voltammograms were re-
corded at multiple scan rates from 100 to 1000 mV/s. The series
of voltammograms were then simulated simultaneously using
the DIGISIM software package [15] to extract kinetic and equilibrium
parameters according to an EECE process (Scheme 3) for each sub-
strate. Initial guesses and constraints of simulation parameters
were set using previously described methods [1,9,11] (see Support-
ing Information). Experimentally determined rate constants (kf) for
the chemical step (C) are summarized in Table 2. For substrates
that react too slowly to be observed directly by cyclic voltammetry
a maximum rate constant was established with reactivity con-
firmed by synthetic methods.
Scheme 2. Selected alkene, alkyne, and diene substrates.
Analysis of the rate constant data (Table 2) reveals a large steric
contribution for polysubstituted alkenes as compared to the mono-
functionalized alkenes. For monofunctionalized alkenes, the fastest
rate constants were observed for ‘‘electron-rich’’ styrenes and vinyl
ethers, which displayed rate constants 10–1000 times fasters than
n-hexene and the electron-deficient acrylonitrile [24,25]. However,
the additional steric bulk of a second phenyl substituent to styrene
dramatically decreases the rate constant from 4.6 Â 107 MÀ1 sÀ1 to
8.8 Â 102 MÀ1 sÀ1 for cis-stilbene and to <1.1 Â 101 MÀ1 sÀ1 for
trans-stilbene. Significant steric effects are also observed for the
smaller substituents such as methyl. The trisubstituted 2-methyl-
2-butene adds with a rate constant of 1.4 Â 102 MÀ1 sÀ1, whereas
the tetrasubstituted 2,3-dimethyl-2-butene reaction occurs to
slowly to be directly detectable by cyclic voltammetry (kf < 3.2
 101 MÀ1 sÀ1).
(J = 30, 30 Hz, Peq); minor isomer (43%) d 36.0 (J = 30, 312 Hz, Pax1
43.0 (J = 30, 312 Hz, Pax2), 57.2 (J = 30, 30 Hz, Peq).
)
2.6.5. [Ru-1Áphenylpropyne][PF6] (5)
The complex is synthesized the same method as described
above except 1-phenyl-1-propyne (0.112 mL, 9.35 mmol) was
added instead of phenylacetylene. Yield: 88 mg (76%). E1/2 (RuIII/
RuII) = 368 mV. +ESI-MS for [C63H50P3S3Ru]+: theoretical: m/z
(z = 1) 1097.1315; observed 1097.1331. Element Anal. Calc. for C63-
H50P4S3F6Ru: C, 60.86; H, 4.06. Found: C, 58.5; H, 4.33%. Electronic
absorption: kmax (e
, MÀ1 cmÀ1): 374 nm (5302), 446 nm (shoulder).
FT-IR (KBr pellet, cmÀ1) 3052, 2917, 2848, 1434, 1086, 841, 694,
523. 31P{1H} NMR (162 MHz, CD3CN): major isomer (55%) d 37.1
(J = 25, 315 Hz, Pax1) 43.3 (J = 25, 315 Hz, Pax2), 55.8 (J = 25, 25 Hz,
A comparison of rate constants for a series of alkynes suggests
both steric and electronic factors are also significant for these
substrates. The monofunctionalized phenylacetylene adding 10
Peq); minor isomer (45%) d 35.7 (J = 25, 316 Hz, Pax1) 45.0 (J = 25,
316 Hz, Pax2), 55.5 (J = 25, 25 Hz, Peq).
3. Results and discussion
3.1. Electrochemical investigations
Table 2
Second-order rate constants (kf) for the addition of various substrates to [Ru-1]+
determined by electrochemical methods (see text).
Using previously described electrochemical methods, [1] a ser-
ies of unsaturated hydrocarbons (Scheme 2) including polysubsti-
tuted alkenes, alkynes, and dienes were added to the metal
stabilized thiyl radical [Ru-1]+. The reaction proceeds via an EECE
mechanism (Scheme 3) beginning with the formation of [Ru-1]+
Substrate
kf (MÀ1 sÀ1
)
Ref.
Monofunctionalized alkenes
Styrene
4.6 Â 107
7.2 Â 107
3.6 Â 107
2.0 Â 107
2.6 Â 106
7.0 Â 105
2.7 Â 104
[1]
[1]
[1]
[1]
[1]
[1]
[1]
m-Methyl styrene
p-Methyl styrene
n-Propyl vinyl ether
tert-Butyl vinyl ether
1-Hexene
via
subsequent
one-electron
oxidations
of
[Ru-1]À
(E1 = À830 mV) and Ru-1 (E2 = À22 mV). In the absence of unsatu-
rated substrates these are the only two events observed in the cyc-
lic voltammogram [2]. In the presence of unsaturated
hydrocarbons, a chemical reaction (C) occurs with [Ru-1]+ to gen-
erate the product complex [Ru-1Ásubstrate]+. The rapid chemical
step shifts the potential of the anodic current minimum for the for-
mal RuIV/III couple (E2) due to disruption of the Nernstian condi-
Acrylonitrile
Cyclic alkenes
Norbornene
Cyclopentene
Cyclohexene
4.0 Â 107
6.0 Â 106
2.9 Â 103
[1]
[1]
[1]
Polysubstituted alkenes
cis-Stilbene
trans-Stilbene
2-Methyl-2-butene
2,3-Dimethyl-2-butene
8.8 Â 102
61.1 Â 101
1.4 Â 102
63.2 Â 101
this work
this work
this work
this work
Alkynes
Phenyl acetylene
1-Octyne
1.2 Â 104
8.7 Â 103
2.3 Â 102
this work
this work
this work
1-Phenyl-1-propyne
Dienes
1,3-Butadiene
2,3-Dimethyl-1,3-butadiene
2.4 Â 105
1.1 Â 103
this work
this work
Scheme 3. EECE mechanism.

K.O. Sampson et al. / Inorganica Chimica Acta 408 (2013) 1–8
5
Table 3
Characterization table for new [Ru-1Ásubstrate]+ products.
Complex
Yld.
(%)
+ESI-MS
(m/z)
E1/2 RuIII/II
(mV)
Isomer ratio
31P NMR major
d ppm (J Hz)
31P NMR minor
d ppm (J Hz)
54.8 (29, 29)
39.5 (313, 29)
36.5 (313, 29)
52.5 (29, 29)
42.9 (315, 29)
35.6 (315, 29)
57.9 (30, 30)
38.5 (303, 30)
35.6 (303, 30)
56.7 (30, 30)
42.7 (312, 30)
36.5 (312, 30)
55.8 (25, 25)
43.3 (315, 25)
37.1 (315, 25)
[Ru-1Ácyclohexene]+
[Ru-1Ácis-stilbene]+
76
57
70
42
76
1063.17
1161.20
1161.17
1083.12
1097.13
269
361
337
361
368
100:0
70:30
69:31
57:43
55:45
54.0 (30, 30)
43.7 (317, 30)
35.2 (317, 30)
54.1 (30, 30)
43.7 (318, 30)
35.2 (318, 30)
57.2 (30, 30)
43.0 (312, 30)
36.0 (312, 30)
55.5 (25, 25)
45.0 (316, 25)
35.7 (316, 25)
[Ru-1Átrans-stilbene]+
[Ru-1Áphenylacetylene]+
[Ru-1Áphenylpropyne]+
times faster than 1-octyne consistent with expected electronic ef-
fects. However, addition of a second substituent in 1-phenyl-1-
propyne retards the rate of addition by a factor of 50.
For conjugated dienes, rate constants also decrease substan-
tially with increased steric constraints. The rate constant for 1,3-
butadiene, 2.4 Â 105 MÀ1 sÀ1, is close to that of 1-hexene; 7.0
 105 MÀ1 sÀ1. The addition of methyl substituents in 2,3-di-
methyl-1,3-butadiene lowers the rate constant by more than 200
fold to 1.1 Â 103 MÀ1 sÀ1. However, the rate constant still is ꢀ8
times greater than the more trisubstituted alkene 2-methyl-2-bu-
tene. While the similarities in rate constants between the 1,3-but-
adienes and their mono-alkene counterpart may suggest a 2 + 2
addition, products resulting from [2 + 4] addition cannot be ex-
cluded on the basis of the rate data. Previously, Fekl reported
[2 + 4] addition of conjugated dienes to [Pt(tfd)2] (tfd = S2C2(CF3)2)
upon oxidation [26]. Attempts to isolate products in the current
study have yielded
investigation.
a complex mixture that remains under
3.2. Chemical syntheses
Representative complexes were prepared by chemical oxidation
of [HNEt3][Ru-1] with two equivalents of ferrocenium hexafluoro-
phosphate in the presence of the appropriate unsaturated
hydrocarbon (Scheme 2). The reactions proceed through a pair of
one-electron oxidation steps of Ru(II) to a formally Ru(IV) complex,
which then reacts with the unsaturated hydrocarbon to generate
the Ru(II) dithioether complex [Ru-1Ásubstrate]PF6. Reaction
yields varied from 42% to 76% for the five substrates examined
(Table 3). For each complex, electrochemical investigations reveal
a single redox event with a RuIII/RuII half-potential (E1/2) consistent
with the electrochemical studies above (Table 3). All complexes
were further identified by mass spectrometry and 31P NMR as
described below. A single product, [Ru-1Ácyclohexene]PF6, was
selected for X-ray diffraction studies.
Fig. 2. (top) 31P NMR spectrum of [Ru-1Ácyclohexene]PF6 in CD3CN (marked with
black j). (bottom) 31P NMR spectrum of [Ru-1Ácis-stilbene]PF6 in CD3CN showing
both the major (marked with black j) and minor (marked with red d) isomers. (For
interpretation of the references to colour in this figure legend, the reader is referred
to the web version of this article.)
bene]+, [Ru-1Áphenylacetylene]+, and [Ru-1Á1-phenyl-1-propy-
ne]+ also display m/z parent peaks within 0.26–29 ppm of the
theoretical M+ parent ion peak with isotopic patterns consistent
with a single Ru atom per complex.
3.3. Characterization
3.3.1. Mass spectrometry
3.3.2. 31P NMR investigations
The chemically synthesized [Ru-1Ásubstrate]PF6 complexes
were characterized by +ESI-MS. (Table 3 and Figs. S9–S13). The ob-
served parent ion peak for [Ru-1Ácyclohexene]+, m/z = 1063.1683
amu, is within 18 ppm of the theoretical peak, m/z = 1063.1487
amu, for the M+ ion. Further, the isotopic envelope is consistent
with theoretical predictions confirming the presence of Ru. No
other ruthenium containing products are observed in the mass
spectrum. The +ESI-MS of [Ru-1Ácis-stilbene]+, [Ru-1Átrans-stil-
The 31P NMR of the chemically synthesized [Ru-1Ásubstrate]PF6
complexes shows second-order spectra. Chemical shifts and
coupling constants consistent with the proposed structures are
summarized in Table 3. The 31P NMR for the addition product
[Ru-1Ácylcohexene]+ (Fig. 2, top) displays a dominant second-or-
der spectrum with a chemical shift of 54.8 (J = 29, 29 Hz) for the
equatorial P and chemical shifts of 36.5 (313, 29 Hz) and 39.5
(313, 29 Hz) ppm for the axial P donors (Fig. 2, top). These values

6
K.O. Sampson et al. / Inorganica Chimica Acta 408 (2013) 1–8
are similar to those for [Ru-1ÁC2H4]+ indicating a single product
with a similar phosphorus environment [1,8].
Table 4
Selected bond distances (Å) and bond angles (°) for [Ru-1Ácyclohexene]PF6.
The 31P NMR for the product of the cis-substituted alkene, [Ru-
1Ácis-stilbene]+ (Fig. 2, bottom), reveals two overlapping second-
order spectra. The mass spectrum and the cyclic voltammogram
of [Ru-1Ácis-stilbene]+ indicate a single product suggesting the
two products observed in the 31P NMR spectra are diastereomers.
The major product (70%) corresponds to a second-order spectrum
with chemical shifts of 35.6 (315, 29 Hz, Pax1), 42.9 (315, 29 Hz,
Bond
Distance (Å)
Bond
Angle (°)
Ru–S1
Ru–S2
Ru–S3
Ru–P1
Ru–P2
Ru–P3
S2–C55
S3–C60
C55–C60
2.3866(9)
2.3722(9)
2.3363(9)
2.3351(9)
2.4172(9)
2.3273(9)
1.865(4)
1.886(4)
1.515(5)
S1–Ru–S2
S1–Ru–S3
S2–Ru–S3
P1–Ru–P2
P1–Ru–P3
P2–Ru–P3
Ru–S2–C55
Ru–S3–C60
S2–C55–C60
92.77(3)
175.70(3)
85.07(3)
95.54(3)
94.72(3)
168.71(3)
103.00(12)
107.57(12)
109.2(3)
P
ax2) and 52.5 (29, 29 Hz, Peq) ppm. The minor product (30%) dis-
plays similar second-order spectrum with resonances at
57.9 ppm, 38.5 ppm, 35.6 (30, 303 Hz, Pax1), 38.5 (30, 303 Hz,
ax2) ppm, and 57.9 (30, 30 Hz, Peq) ppm. The identity of the two
a
P
hexyl unit (C55–C60) are arranged in a chair conformation with
the sulfur substituents S2 and S3 located in the axial position of
C55 and equatorial position of C60, respectively.
isomers is discussed in the Computational Studies section below.
The 31P NMR of [Ru-1Átrans-stilbene]+ and those of the alkyne
addition products Ru-1Áphenylacetylene]+ and [Ru-1Á1-phenyl-1-
propyne]+ each display a pair of second-order spectra similar to
[Ru-1Ácis-stilbene]+ (Figs. S14–S16). The relative intensities of
the diastereomers (Table 3) of the cis- and trans-stilbene products
are similar (70:30), whereas the alkyne addition products show
greater parity (55–57:43–45). For all spectra, the chemical shift
of the equatorial P donor falls between 52.5 and 57.9 ppm with ax-
ial-equatorial coupling constants of 25–30 Hz, while the axial P do-
nors are found in pairs between 35.2–37.1 and 38.5–45.0 ppm with
axial-axial coupling constants ranging from 303 to 315 Hz. For the
alkyne addition products, the 31P NMR spectra are similar to those
previously reported for [Ru-1Áoctyne]+ [9]. In that study, the iso-
mer with the substituted carbon positioned nearest the S trans to
S was identified as the major product.
The Ru–P and Ru–S bond distances in [Ru-1Ácyclohexene]+ vary
from 2.3273(9) to 2.4172(9) Å and 2.3363(9) to 2.3866(9) Å,
respectively. Selected bond distances and angles are listed in
Table 4. The metal–ligand bond distances are similar to the previ-
ously reported complexes [Ru-1ÁC2H4]+ and [Ru-1Áp-methylsty-
rene]+ [1,8]. As in those complexes, the addition of the alkene
substrate results in formation of C–S single bonds as denoted by
the S2–C55 and S3–C60 bond distances of 1.865(4) and
1.886(4) Å, respectively. This corresponds with a C55–C60 bond
distance of 1.515(5) Å, which signifies the oxidation of the alkene
double bond in the product complex. Overall, the proposed
covalent addition of the cyclohexene substrate across the reactive
cis-sulfur sites is consistent with the X-ray structure.
3.3.4. Computational studies
3.3.3. X-ray structure of [Ru-1Ácyclohexene][PF6]
The tris-chelate precursor [Ru-1]À exists in solution as an enan-
Single crystals of [Ru-1Ácyclohexene]PF6 in the triclinic space
ꢀ
tiomeric mixture (K/D) of the meridional stereoisomer. Addition of
group P1 were obtained by slow evaporation of dichlorometh-
ane:hexane mixtures of the complex. An ORTEP [27] representation
of the metal containing cation is provided in Fig. 3 with full crystal
data and structure refinement details listed in Table 1. The [Ru-
1Ácyclohexene]+ complex contains a six coordinated Ru(II) ion li-
gated by three PS chelates in a pseudo-octahedral environment
with the three sulfur donors arranged in a meridional fashion as
in related complexes [1,8,9,11,14,28]. The cyclohexene substrate
has been added across the cis-sulfur donors resulting in the forma-
tion of a neutral, tetradentate dithioether chelate comprised of P2–
S2–S3–P3 arranged as the
a isomer. The six-carbons of the cyclo-
Fig. 4. Eight possible isomers of [Ru-1Ácis-stilbene]PF6 and related complexes
organized according to chelate ring configuration (columns) and carbon stereo-
chemistry (rows).
Fig. 3. ORTEP representation of the cationic metal complex of [Ru-
1Ácyclohexene]PF6.

K.O. Sampson et al. / Inorganica Chimica Acta 408 (2013) 1–8
7
with electron-withdrawing groups [1]. In all instances, rate con-
stants for polysubstituted alkenes are significantly lower than the
mono-functionalized counterparts, regardless of electron-donor
ability. The rate constants for alkynes are found to be approxi-
mately 100 times lower than those of corresponding alkenes. These
results are consistent with a kinetic preference for ethylene addi-
tion over all other alkenes allowing the selective binding of ethyl-
ene to [Ru-1]+ even in the presence of other unsaturated
substrates.
The addition of substituted alkene substrates lead to a mixture
of isomers. For monofunctionalized alkenes the unsubstituted car-
bon preferentially adds to the sulfur with higher spin density con-
sistent with Tedder’s Rules [29,30]. Similar product distributions
are observed for monofunctionalized alkynes. For disubstituted al-
kenes, both the cis- and trans-alkenes can lead to as many as four
distinct isomers. The [Ru-1Ácis-stilbene]+ product is observed a
mixture of d/k-SR and d/k-RS, whereas only the k-RR and d-SS iso-
mers are observed for [Ru-1Átrans-stilbene]+ due to steric con-
straints. The observation of distinct reaction products for cis- and
trans-stilbene indicates the substrate does not rearrange during
the addition process. This is consistent with a concerted, [2 + 2]
mechanism or a step-wise process in which the second C–S bond
is formed faster than the rate of C–C bond rotation in the interme-
diate. Further efforts to refine the reaction mechanism are
underway.
Fig. 5. Calculated relative energies for the ring inversion of the RS isomer of [Ru-
1Ácis-stilbene]+ as a function of the C105–-S2–S3–C104 torsion angle.
vicinal substituted alkenes imparts three stereogenic centers. Two
are carbon centered (R/S), while the third depends on the chelate-
ring configuration (k, d). Representations of the eight possible iso-
mers resulting from these three centers are shown in Fig. 4. Four of
these isomers (k-RS, d-RS, k-SR, and d-SR) are possible for cis substi-
tuted alkenes and four (k-RR, d-RR, k-SS, and d-SS) for trans-substi-
tuted alkenes.
To determine the relative energies of the isomers in Fig. 4, den-
sity functional theory calculations using the B3LYP functional with
a mixed basis set (6-31 G for C, H, S, P and LANL2DZ for Ru) were
performed for cis- and trans-stilbene as described in the Section 2.
For cis-stilbene, the relative energies of the four isomers are:
d-SR = 0.0 kcal/mol; k-SR = +1.9 kcal/mol; d-RS = +3.3 kcal/mol;
k-RS = +4.2 kcal/mol. The two lower energy structures, d-SR and
k-SR, are related by inversion of the chelate ring and are separated
by only 1.9 kcal/mol. Likewise, the two higher energy structures,
d-RS and k-RS, are chelate ring invertomers separated by a small
energy gap, 0.9 kcal/mol. Inversion of the chelate ring (d ? k) for
the RS pair was modeled by systematic variation of the C–S–S–C
torsion angle revealing a small energy barrier of ꢀ2 kcal/mol
(Fig. 5), which is easily surmounted in solution on the NMR time-
scale. A similar barrier is found for the SR configuration (Fig. S17).
While interconversion of the chelate ring is viable on the NMR
timescale, inversion of the carbon stereogenic centers SR ? RS
would require cleavage of the C–S bonds, which is not observed
for [Ru-1Ásubstrate]+ complexes. The two sets of 31P resonances
observed for [Ru-1Ácis-stilbene]+ are therefore assigned to d/k-SR
and d/k-RS.
Acknowledgments
Acknowledgment is made to the Donors of the American Chem-
ical Society Petroleum Research Fund (51566-ND3) for support of
this work. The authors are thankful to the Cardinal Research
Cluster at the University of Louisville for providing computational
facilities. M.S.M. thanks the Department of Energy (DEFG02-
08CH11538) and the Kentucky Research Challenge Trust Fund for
upgrade of our X-ray facilities.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2013.08.003.
References
[1] K. Ouch, M.S. Mashuta, C.A. Grapperhaus, Inorg. Chem. (2011) 9904.
[2] C.A. Grapperhaus, S. Poturovic, Inorg. Chem. 43 (2004) 3292.
[3] C.A. Grapperhaus, P.M. Kozlowski, D. Kumar, H.N. Frye, K.B. Venna, S. Poturovic,
Angew. Chem., Int. Ed. 46 (2007) 4085.
[4] C. Walling, W. Helmreich, J. Am. Chem. Soc. 81 (1958) 1144.
[5] C. Chatgilialoglu, C. Ferreri, M. Ballestri, Q.G. Mulazzani, L. Landi, J. Am. Chem.
Soc. 122 (2000) 4593.
[6] Y. Ichinose, K. Oshima, K. Utimoto, Chem. Lett. (1988) 669.
[7] J. Lalevee, X. Allonas, J.P. Fouassier, J. Org. Chem. 71 (2006) 9723.
[8] C.A. Grapperhaus, K.B. Venna, M.S. Mashuta, Inorg. Chem. 46 (2007) 8044.
[9] K. Ouch, M.S. Mashuta, C.A. Grapperhaus, Eur. J. Inorg. Chem. 2012 (2012) 475.
[10] R.Y.C. Shin, M.E. Teo, W.K. Leong, J.J. Vittal, J.H.K. Yip, L.Y. Goh, R.D. Webster,
Organometallics 24 (2005) 1483.
[11] C.A. Grapperhaus, K. Ouch, M.S. Mashuta, J. Am. Chem. Soc. 131 (2009) 64.
[12] N.G. Connelly, W.E. Geiger, Chem. Rev. 96 (1996) 877.
[13] C.A. Grapperhaus, S. Poturovic, M.S. Mashuta, Inorg. Chem. 41 (2002) 4309.
[14] J.R. Dilworth, Y.F. Zheng, S.F. Lu, Q.J. Wu, Transition Met. Chem. 17 (1992) 364.
[15] M. Rudolph, S.W. Feldberg, DigiSim (3.03b), Bioanalytical Systems Inc., West
Lafayette, IN, 2004.
Calculations of the [Ru-1Átrans-stilbene]+ isomers predict
nearly degenerate structures for k-RR and d-SS with an energy sep-
aration of 0.3 kcal/mol. Although the energy gap between these
two structures is small, there no low energy pathway can be found
to interconvert these products. The chelate ring invertomers of
these two structures, d-RR and k-SS, failed to converge. This is
attributed to disfavored positioning of both sterically bulky phenyl
substituents in pseudo-axial positions (Fig. 4). The two sets of 31P
signals for [Ru-1Átrans-stilbene]+ are therefore assigned to k-RR
and d-SS.
[16] Agilent. CrysAlisPRO (Version 1.171.36.28), Agilent Technologies, Yarnton,
Oxfordshire, England (2013).
4. Conclusions
[17] Agilent. SCALE3 ABSPACK included in CrysAlisPRO RED (V 1.171.36.28),
(2013).
[18] G.M. Sheldrick, Acta Crystallogr., Sect. A 46 (1990) 467.
[19] SHELXTL v6.14, Program Library for Structure Solution and Molecular Graphics
Bruker Advanced X-ray Solutions Inc., Madison, WI, 2000.
[20] GAUSSIAN 09, Gaussian Inc., Wallingford, CT, 2009.
[21] R. Chauhan, M.S. Mashuta, C.A. Grapperhaus, Inorg. Chem. 51 (2012) 7913.
[22] P.J. Farmer, J.H. Reibenspies, P.A. Lindahl, M.Y. Darensbourg, J. Am. Chem. Soc.
115 (1993) 4665.
In the current manuscript, we report significant steric factors on
the rate of addition of unsaturated substrates, including polysub-
stituted alkenes, alkynes, and dienes, to the metal-stabilized thiyl
radical [Ru-1]+. The results complement earlier studies on elec-
tronic effects with monofunctionalized alkenes that reveal en-
hanced rates with electron-donating groups and decreased rates

8
K.O. Sampson et al. / Inorganica Chimica Acta 408 (2013) 1–8
[23] C.A. Grapperhaus, C.S. Mullins, M.S. Mashuta, Inorg. Chim. Acta 358 (2005)
623.
[24] H. Sakurai, S. Hayashi, A. Hosomi, Bull. Chem. Soc. Jpn. 44 (1971) 1945.
[25] M.J. Jones, G. Moad, E. Rizzardo, D.H. Solomon, J. Org. Chem. 54 (1989) 1607.
[26] M.J. Kerr, D.J. Harrison, A.J. Lough, U. Fekl, Inorg. Chem. 48 (2009) 9043.
[27] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
[28] P. Pérez-Lourido, J. Romero, J.A. Garcıa-Vázquez, J. Castro, A. Sousa, L. Cooper,
´
J.R. Dilworth, R.L. Richards, Y. Zheng, J.A. Zubieta, Inorg. Chim. Acta 356 (2003)
193.
[29] J.M. Tedder, J.C. Walton, Tetrahedron 36 (1980) 701.
[30] J.M. Tedder, Angew. Chem., Int. Ed. 21 (1982) 401.