Deactivation of a ruthenium(II) N-heterocyclic carbene p-cymene complex during transfer hydrogenation catalysis

Article abstract of DOI:10.1007/s11243-017-0189-x

A ruthenium (II) N-heterocyclic carbene (NHC) complex was synthesized to investigate ligand dissociation as a possible deactivation pathway for the catalytic cycle of a transfer hydrogenation reaction. Diiodo(1,3-dimethylbenzimidazole-2-ylidene)(p-cymene)ruthenium(II) was synthesized for use as the catalytic species and characterized using physico-chemical, spectroscopic methods, and single crystal X-ray diffraction. The transfer of hydrogen from isopropanol to acetophenone was followed using ¹H NMR. We observed 94% conversion of the substrate to the alcohol product after 1?h. We also found that the p-cymene complex decomposed during the catalytic reaction to the extent of 80% deactivation after 1?h, based on ¹H NMR spectrometry. From Gaussian calculations, an ultraviolet–visible spectrum that is in excellent agreement with the actual spectrum was computed, giving insight into the nature of the absorptions observed experimentally.

Full text of DOI:10.1007/s11243-017-0189-x

Transit Met Chem
DOI 10.1007/s11243-017-0189-x
Deactivation of a ruthenium(II) N‑heterocyclic carbene p‑cymene
complex during transfer hydrogenation catalysis
John R. Miecznikowski1 · Nicholas A. Bernier1,4 · Christopher A. Van Akin1,5
Sheila C. Bonitatibus1,6 · Maura E. Morgan1 · Rami M. Kharbouch1 ·
Brandon Q. Mercado2 · Matthew A. Lynn3
·
Received: 1 August 2017 / Accepted: 16 October 2017
© Springer International Publishing AG 2017
Abstract A ruthenium (II) N-heterocyclic carbene (NHC)
complex was synthesized to investigate ligand dissociation
as a possible deactivation pathway for the catalytic cycle of
a transfer hydrogenation reaction. Diiodo(1,3-dimethylben-
zimidazole-2-ylidene)(p-cymene)ruthenium(II) was synthe-
sized for use as the catalytic species and characterized using
physico-chemical, spectroscopic methods, and single crystal
X-ray difraction. The transfer of hydrogen from isopropanol
to acetophenone was followed using 1H NMR. We observed
94% conversion of the substrate to the alcohol product after
1 h. We also found that the p-cymene complex decomposed
during the catalytic reaction to the extent of 80% deactiva-
tion after 1 h, based on 1H NMR spectrometry. From Gauss-
ian calculations, an ultraviolet–visible spectrum that is in
excellent agreement with the actual spectrum was computed,
giving insight into the nature of the absorptions observed
experimentally.
Introduction
The reduction of organic compounds such as aldehydes and
ketones can be performed in a variety of ways, including
through the use of a stoichiometric reagent, a catalyst, elec-
trolysis, and via hydrogen atom transfer via a radical mecha-
nism [1]. Catalytic transfer hydrogenation reactions are an
alternative to direct hydrogenation of substrates whereby
the need for molecular hydrogen is avoided [2]. Transfer
hydrogenation is a useful complement to catalytic reduction
by molecular hydrogen since it frequently shows a diferent
pattern of functional group selectivity [3]. In transfer hydro-
genation reactions, the ionic H+/H− transfer from a donor to
Electronic supplementary material The online version of
this article (https://doi.org/10.1007/s11243-017-0189-x) contains
supplementary material, which is available to authorized users.
1
* John R. Miecznikowski
jmiecznikowski@fairfeld.edu
Department of Chemistry and Biochemistry, Fairfeld
University, 1073 North Benson Road, Fairfeld, CT 06824,
USA
Nicholas A. Bernier
nbernier@ucla.edu
2
Department of Chemistry, Yale University, 225 Prospect
Street, New Haven, CT 06520-8107, USA
Christopher A. Van Akin
3
christopher.vanakin@student.fairfeld.edu
Department of Science and Mathematics, National Technical
Institute for the Deaf, Rochester Institute of Technology, 52
Lomb Memorial Drive, Rochester, NY 14623, USA
Sheila C. Bonitatibus
sheila.bonitatibus@student.fairfeld.edu
4
Present Address: Department of Chemistry
Maura E. Morgan
and Biochemistry, University of California, Los Angeles, 607
Charles E. Young Drive East, Los Angeles, CA 90095, USA
maura.morgan@student.fairfeld.edu
Rami M. Kharbouch
5
Present Address: University of Connecticut School
rkharbouch@student.fairfeld.edu
of Medicine, 263 Farmington Avenue, Farmington,
CT 06030, USA
Brandon Q. Mercado
brandon.mercado@yale.edu
6
Present Address: Department of Chemistry, Boston
University, 590 Commonwealth Avenue, Boston, MA 02215,
USA
Matthew A. Lynn
malntm@rit.edu
Vol.:(0123456789)
1 3

Transit Met Chem
a substrate forms the basis for many useful C–C, C–O, and
C–N bond-forming reactions [4–7].
temperature = 120 °C. The ligand precursor, 1,3-dimeth-
ylbenzimidazole iodide was synthesized by a previously
reported procedure [24]. UV–visible spectra were collected
on a Cary 100 UV–visible spectrometer. IR spectra were
collected using a Thermo Nicolet AVATAR 380 FT-IR with
a SMART SPECULATR reference adaptor.
N-heterocyclic carbenes are widely used ligands for metal
complexes that have the potential to serve as homogenous
catalysts. These compounds are ideal for use with larger sub-
strates that may not be able to interact well with surfaces in
heterogenous catalysis due to steric reasons [8]. N-heterocy-
clic carbenes ligands are sometimes considered phosphine
alternatives, primarily because their metal complexes cata-
lyze typical organic reactions similar to those observed for
complexes possessing trialkylphosphine ligands. In addition,
N-heterocyclic carbine complexes are not as susceptible to
metal–ligand cleavage [9–11]. Nolan and co-workers frst
reported the frst N-heterocyclic carbene (NHC) complex
for transfer hydrogenation [12].
Synthesis
Synthesis of diiodo(1,3‑dimethylbenzimidazole‑2‑
ylidene)(p‑cymene)ruthenium(II)
A reaction vessel was charged with 1,3 dimethyl benzimi-
dazole iodide (0.28 g, 1.0 mmol), silver(I) oxide (0.23 g,
1.0 mmol), and 30 mL of dichloromethane. The reaction
was heated at refux in the absence of light for 20 h. The
reaction solution was fltered through Celite to remove the
formed solids, and the solvent was removed under reduced
pressure. Yield: 0.16 g.
Common transition metal precatalysts for transfer hydro-
genation contain Ir(I), Ir(III), Rh(I), Rh(III), Ru(II), and
Ni(0) metal centers [12–21]. Ruthenium NHC complexes
that contain arene ligands have been reported for transfer
hydrogenation catalysis [15, 22].
The loss of a p-cymene ligand that is coordinated to
a transition metal center is our proposed deactivation
pathway for this catalyst. Crabtree and co-workers have
reported the loss of a Cp* ligand under commonly modi-
fed transfer hydrogenation conditions [23]. Understand-
ing the nature of catalyst decomposition is both interest-
ing and important given that the ability to reuse a catalyst
is especially important for precious metal complexes. We
want to investigate the disruption of the catalytic cycle
via ligand dissociation or decomposition during a transfer
hydrogenation reaction. A highlight of the present work is
the quantifcation of the loss of a p-cymene ligand during
transfer hydrogenation catalysis when the precatalyst is a
ruthenium(II) N-heterocyclic carbene complex that con-
tains a p-cymene ligand.
Next, potassium iodide (0.66 g, 4.0 mmol) was added to
the reaction vessel containing the silver benzimidazole com-
pound (0.14 g, 0.19 mmol) and [Ru(p-cymene)Cl2]2 (Strem
Chemicals) (0.093 g, 0.15 mmol) with 20 mL acetonitrile.
The reaction was heated at refux for 20 h, and the solvent
was removed under reduced pressure. The product was dis-
solved in tetrahydrofuran, and the solution was fltered to
remove undissolved solids. The tetrahydrofuran solvent was
then removed under reduced pressure. Yield: 0.045 g (46%).
Crystals for X-ray difraction were grown by the slow vapor
difusion of pentane into a solution of the ruthenium com-
plex in tetrahydrofuran. Anal. Calc. for C19H24I2N2Ru·2H2O
(635.29): C, 34.0; H, 4.2; N, 4.2. Found: C, 34.3; H, 3.9;
N, 3.9.
1H NMR (DMSO-d6, 300 MHz) δ 7.75(m, 2H, ben-
zimidazole, CH), 7.47 (m, 2H, benzimidazole, CH),
6.38 (d, (3J = 6.3 Hz), 1H, p-cymene CH), 6.25 (m, 2H,
p-cymene CH), 6.14 (d, (3J = 6.0 Hz), 1H, p-cymene
CH), 4.20 (s, 3H, imidazole CH3), 4.09 (s, 3H, imidazole
CH3), 3.00 (m, 1H, iPr-H), 2.19 (s, 3H, CH3 p-cymene),
1.25 (d, (3J = 6.9 Hz), 3H, p-cymene iPr CH3), 1.05 (d,
(3J = 6.6 Hz), 3H, p-cymene iPr CH3). 13C {1H} NMR
(DMSO-d6, 75 MHz), δ 179.50 (Cipso), 123.79 (benzimida-
zole CH), 123.40 (benzimidazole CH), 118.30 (C), 111.34
(benzimidazole CH), 111.07 (benzimidazole CH), 105.34
(C), 92.92 (p-cymene CH), 92.83 (p-cymene CH), 88.59
(p-cymene CH), 88.51 (p-cymene CH), 40.74 (imidazole
CH3), 35.94 (imidazole CH3), 31.18 (iPr C-H), 23.56 (iPr
CH3), 20.36 (iPr CH3), 18.40 (p-cymene CH3). UV–visible
data: λ (nm), (ε (M−1cm−1): 359 (6760); 297 (19600); 294
(19300); 237 (23700); 234 (23400); 218 (22400). IR data:
ν (cm−1, intensity, assignment); 2962.45 (m); 2360.32 (m,
CO2); 1458.68 (w); 1365.35 (w); 1342.55 (w); 1259.64
Experimental
General procedures
Reagents and solvents were commercially available and were
purchased from Acros Organics, Fisher Scientifc, and Strem
Chemicals. All chemicals were used as received. NMR
spectra were recorded at 25 °C on a Bruker spectrometer
at 300 MHz (1H NMR) and 75 MHz (13C NMR) and refer-
enced to DMSO ( in ppm, J in Hz). Elemental analyses were
performed by Robertson Microlit Laboratories (Ledgewood,
NJ, USA). Residual solvent molecules were identifed by
1H NMR. Electrospray mass spectra were collected using a
direct fow injection (injection volume = 5 μL) on an Agilent
QTOF instrument in the positive ion mode. The optimized
conditions were capillary = 3000 kV, cone = 10 V, source
1 3

Transit Met Chem
Table 1 Crystal and structure refnement data for C19H24I2N2Ru [1]
(m); 1077.29 (s); 1015.13 (s); 861.36 (w); 793.91 (vs);
747.55 (m); 666.84 (w); 627.13 (w).
Formula
C19H24I2N2Ru
635.29
93(2)
0.71073
Monoclinic
P2(1)/c
15.6125(11)
7.8796(6)
16.7467(12)
97.872(9)
2040.8(3)
4
FW (g/mol)
Temperature (K)
Wavelength (Å)
Crystal system
Space group
a (Å)
Catalytic activity of the ruthenium (II) N‑heterocyclic
carbene complex
In a typical reaction, an Erlenmeyer fask was charged with
KOH (0.16 g, 3.0 mmol) and 30 mL isopropanol. Next, a
reaction vessel was charged with acetophenone (0.22 g,
1.8 mmol) and 1.0 mL of the prepared KOH solution
(0.098 mmol) and 9 mL of isopropanol. In a separate vial,
the prepared ruthenium precatalyst (0.012 g, 0.018 mmol)
was dissolved in a minimal amount of tetrahydrofuran.
The precatalyst solution was added to the reaction vessel
and heated at refux. Approximately 1 mL aliquots were
taken from the solution after a certain period of time. Each
sample was placed in approximately 20 mL hexanes, and
the hexanes mixture was fltered through silica. The hex-
anes were removed under reduced pressure, and 1H NMR
spectra of the products were acquired in CDCl3.
b (Å)
c (Å)
β (o)
Volume (Å)3
Z
r (calc) (g/cm3)
2.068
3.797
Abs (mm−1
)
F(000)
1208
Crystal size (mm3)
Theta range (o)
Ref/Uniq
0.250 × 0.190 × 0.080
1.317–25.092
3599/3599
0.081
R(int)
Abs correction
Max./Min.
Ref method
Data/restr/par
GOF on F2
Semi-empirical from equivalents
0.751/0.453
SHELXL-2014/7
3599/0/223
1.179
Stability of the ruthenium (II) N‑heterocyclic carbene
complex using 1,3,5‑trimethoxybenzene
R1 indices (I > 2 s)
wR2
0.0377
0.1014
1.253 and − 0.839
An Erlenmeyer fask was charged with KOH (0.010 g,
0.18 mmol) and 20 mL isopropanol. Next, a 25-mL reac-
tion vessel was charged with acetophenone (0.023 g,
0.20 mmol) and 1,3,5-trimethoxybenzene (0.0055 g,
0.033 mmol). Next, 0.20 mL of the prepared KOH
(0.0018 mmol) solution was added, along with an addi-
tional 1.8 mL isopropanol and the ruthenium precatalyst
(0.022 g, 0.034 mmol). The reaction was put on refux
for a period of time. The solvent was then removed under
Peak/hole (e/Å−3
)
Crystallographic analyses
Low-temperature difraction data (ω-scans) were collected
on a Rigaku R-AXIS RAPID difractometer coupled to an
R-AXIS RAPID imaging plate detector with Mo Kα radia-
tion (λ = 0.71073 Å) for the structure of 1. The difraction
images were processed and scaled using the Rigaku Crystal-
Clear software [25]. The structure was solved with SHELXT
and was refned against F2 on all data by full-matrix least
squares with SHELXL [26]. All non-hydrogen atoms were
refned anisotropically. Hydrogen atoms were included in
the model at geometrically calculated positions and refned
using a riding model. The isotropic displacement parameters
of all hydrogen atoms were fxed to 1.2 times the U value
of the atoms to which they are linked (1.5 times for methyl
groups). The full numbering scheme of 1 can be found in
the Supporting Information. Full details of the X-ray struc-
ture determination are in the CIF included as Supporting
Information. CCDC number 1530788 contains the supple-
mentary crystallographic data for this paper. These data can
be obtained free of charge from The Cambridge Crystallo-
graphic Data Center via www.ccdc.cam.ac.uk/data_request/
1
reduced pressure, and a H NMR of the product was
acquired in DMSO-d6.
Stability of the ruthenium (II) N‑heterocyclic carbene
complex using 1,3,5‑trimethylbenzene
An Erlenmeyer fask was charged with KOH (0.010 g,
0.18 mmol) and 20 mL isopropanol. Next, a 25-mL reac-
tion vessel was charged acetophenone (0.022 g, 0.19 mmol)
and 1,3,5-trimethylbenzene (0.0081 g, 0.071 mmol). Next,
0.20 mL of the prepared KOH (0.0018 mmol) solution
was added, along with an additional 1.8 mL isopropanol
and the ruthenium precatalyst (0.022 g, 0.034 mmol). The
reaction was put on refux for a period of time. The solvent
was then removed under reduced pressure, and a 1H NMR
of the product was acquired in DMSO-d6.
1 3

Transit Met Chem
cif. All other pertinent crystallographic details such as h, k,
l ranges, 2θ ranges, and R-factors can be found in Table 1.
Maroon-red crystals single crystals suitable for X-ray
diffraction were grown by allowing pentane vapor to
slowly difuse into a tetrahydrofuran solution containing
the ruthenium(II) complex. The solid-state structure of 1
is shown in Fig. 1. The complex features an η6 p-cymene
ligand and a ruthenium(II) center. The Ru(II) center is also
coordinated to a 1,3-dimethyl-benzimidazole ligand and two
iodide ligands.
Refnement details
These data were refned as a 2-component twin. (1 0 0 0 − 1
0 0 0 − 1) was used with an HKLF5 formatted refection
list. This twin law afected approximately 16.3(2)% of the
refections.
Various bond lengths and angles for complex 1 are
given in Table 2. The R-factor of 1 was 0.0377. The ruthe-
nium–carbon (N-heterocyclic carbene) bond distance of
2.072(7) (Å) is in excellent agreement with previously
published examples in the literature for related complexes
[30–33]. In addition, the Ru–I bond distances of 2.7520(8)
and 2.2456(8) (Å) compare well with a previously published
example by Hahn and co-workers [33].
Gaussian calculations
Gaussian 09 [27] was used to perform a geometry optimi-
zation and time-dependent density functional theory (TD-
DFT) calculation using the B3LYP hybrid functional on the
subject molecule. The 6-31G(d) basis set was employed for
H, C, and N, and the LANL2DZ basis set, including efec-
tive core potentials, was used for Ru and I. Symmetry was
not imposed on the system. A total of 100 electronic transi-
tions in the electronic spectrum were computed with the
TD-DFT method.
Complex 1 was characterized using 1H, 13C, and HSQC
NMR spectroscopy. The key resonances of the product are
the p-cymene aromatic proton resonances which appear as
doublets at 6.38 (1H), 6.25 (2H), and 6.14 (1H). There is
1
evidence for water in the H NMR spectrum, which was
acquired in DMSO-d6. Also, the presence of water is consist-
ent with the elemental analysis results of the bulk sample.
Any attempts to remove the residual water from the sample
were unsuccessful.
Results and discussion
ESI–MS data for 1 were collected with a cone voltage
of 10 V. The predominant feature in the mass spectrum of
1 corresponds to the [Ru(NHC)(p-cymene)I]+ parent ion,
indicating that one of the iodide anions dissociated during
the analysis. For all of these spectra, the isotopic pattern in
the mass spectrometry data fts the assigned structure.
Complex 1 was also characterized with UV–visible
spectroscopy. The UV–visible spectrum for 1, which was
acquired in acetonitrile, is shown in Fig. 2 (solid line). Fig-
ure 2 also contains the calculated UV–visible spectrum,
which has been red-shifted by 29 nm for presentation, for 1
(dashed line).
Synthesis and characterization
The synthesis of 1 was accomplished according to the steps
shown in Scheme 1. In the preparation of 1, 1,3-dimethyl
benzimidazole iodide was generated following a published
route [24] and was subsequently reacted with silver(I) oxide
in dichloromethane [28, 29]. After fltering the product
through Celite, the solvent was evaporated under reduced
pressure. The resulting Ag(I) benzimidazole product was
then reacted with potassium iodide and [Ru(p-cymene)Cl2]2
in acetonitrile to aford 1. As such, we fnd that complex 1
is soluble in acetonitrile, methanol, dimethyl sulfoxide, and
tetrahydrofuran. The reactions (Scheme 1) were performed
in air and the ruthenium metalation proceeded with a yield
of 46%. Complex 1 is air stable but may decompose in the
presence of light over a period of three months.
In the UV–visible spectrum, we observe charge-transfer
bands in the ultraviolet region. Consistent with previous
reports [34–38] of Ru(II) piano-stool compounds, the
molar absorptivities of none of the features in our experi-
mental spectrum are consistent with the assignment of d–d
bands. In the literature, absorptions in the 200–350 nm
Scheme 1 Synthesis of 1
1 3

Transit Met Chem
Fig. 1 Solid-state structure
of 1. All non-hydrogen atoms
shown are depicted with 50%
thermal contours
Table 2 Selected bond lengths
and angles (esd) for 1
the absorptions were in the ca. 380–450 nm range although
Gichumbi and co-workers [35] attributed the absorption
feature in this range to MLCT transitions. In their report,
Małecki and colleagues [36] included TD-DFT calcula-
tions that indicated that d–d transitions occurred at wave-
lengths greater than 400 nm and had small oscillator
strengths.
(1)
2.7520(8)
I(1)–Ru(1) (Å)
I(2)–Ru(1) (Å)
Ru(1)–C(11) (Å)
Ru(1)–C(3) (Å)
Ru(1)–C(5) (Å)
Ru(1)–C(6) (Å)
Ru(1)–C(4) (Å)
Ru(1)–C(2) (Å)
Ru(1)–C(1) (Å)
2.7456(8)
2.072(7)
2.180(7)
2.184(8)
2.188(8)
2.207(8)
2.255(8)
2.278(8)
Given its combination of ligands with their varied coor-
dination modes and electron donor and acceptor attributes,
assigning the electronic spectrum of the low-symmetry com-
plex 1 can rightly be expected to be complicated. In strong
agreement with the experimentally determined spectrum,
our calculations show three absorption regions between
200 nm and 400 nm. The maxima of the calculated spectrum
match rather well with those observed experimentally (205
vs. 234/237 nm, 264 vs. 294/297 nm, and 332 vs. 359 nm)
with the peaks of the calculated spectrum occurring ca.
30 nm lower than in the experimental spectrum. The TD-
DFT-computed spectrum also fnds nine absorptions located
between 390 nm and 620 nm although all of these are quite
C(11)–Ru(1)–I(2) 91.50(19)
C(11)–Ru(1)–I(1) 93.4(2)
I(2)–Ru(1)–I(1)
84.72(2)
range have variously been attributed to metal-to-ligand
(MLCT), ligand-to-metal (LMCT), and ligand-to-ligand
(LLCT, both inter- and intraligand) charge transfers. In
cases where d–d transitions have been assigned [34, 36],
1 3

Transit Met Chem
Fig. 2 Experimental and
calculated UV–visible spectra
of 1. The calculated UV–visible
spectrum was obtained using
TD-DFT Gaussian 09 calcula-
tions and has been red-shifted
by 29 nm in this composite
image
Table 3 Percent orbital contributions for the metal and ligand moie-
ties in the frontier molecular orbitals of 1
(Ru, the two I atoms, the carbene ligand, and p-cymene)
of 1. All of the carbene and p-cymene orbital character is
π-type, and all of that from the I atoms is from their valence
p orbitals. As for the Ru atom, from a basic understanding of
the molecular orbital theory of a d6 metal complex it might
be supposed for 1 that the three highest occupied and the two
lowest unoccupied MOs are those that should be termed the
“d orbitals” and that the assignment of transitions involving
the “metal” orbitals involves any of these fve MOs. Review
of Table 3 shows that although the two lowest unoccupied
MOs do have more orbital contribution from Ru than from
elsewhere, only one of the ten highest occupied MOs con-
tains majority Ru character. In fact, orbital character from
the iodine atoms makes up more than half of each of the
four highest MOs; the HOMO is calculated to contain 80%
I character with only 8% from Ru. Considering the Ru and
I orbital contributions together, eight of the highest energy
occupied molecular orbitals have majority Ru-I character;
the more stable ones of these orbitals contain Ru–I-bonding
character, while the ones closer in energy to the HOMO are
Ru–I antibonding. The signifcant orbital contributions of
both Ru and I in many of the frontier MOs of 1 should not
be a surprise given the similar 4d5s5p valence atomic struc-
ture of these two elements, but it does mean that describ-
ing specifc absorptions as resulting from metal-based d–d
transitions or arising from the metal in a MLCT band cannot
be precisely done for this system. Instead, the frontier “d
orbitals” should really be considered to be Ru–I admixtures.
Given this caveat for 1, we nevertheless provide some
indication of the kinds of electronic transitions that give
rise to the several features in the absorbance spectrum.
In the 200–240 nm range, LLCT excitations from orbit-
als (HOMO-6 and HOMO-10) with significant carbene
Molecular orbital Ru
2 I
Dimethylbenzimi- p-cymene
dazole-2-ylidene
HOMO-10
HOMO-9
HOMO-8
HOMO-7
HOMO-6
HOMO-5
HOMO-4
HOMO-3
HOMO-2
HOMO-1
HOMO
40.7%
7.4% 36.8%
15.1%
18.1
15.6
25.5
9.1
11.4
19.7
11.2
14.5
15.4
9.7
28.1
27.6
7.9
73.7
68.5
0.7
52.5
38.1
41.4
1.9
26.8
41.3
28.5
25.8
21.3
8.1
45.2
34.3
19.4
17.7
22.8
0.4
25.7
43.0
28.4
0.6
13.6
31.9
57.0
58.2
60.7
79.9
26.0
6.5
3.7
3.3
4.7
88.5
48.2
7.0
3.3
1.5
2.7
2.3
LUMO
0.6
LUMO + 1
LUMO + 2
LUMO + 3
LUMO + 4
LUMO + 5
31.7
65.1
4.6
7.6
4.1
1.8
0.1
6.9
98.7
weak. The absorptions between 390 nm and 450 nm give rise
to a weak tail in the spectrum.
Assigning the nature of the electronic transitions in the
absorption spectrum necessitates review of the computed
excited states and the molecular orbitals (MOs) that are
involved. Full tabulations of all of the excitation energies,
oscillator strengths, and orbital compositions are provided in
the supplementary information. Shown in Table 3 are sum-
mations of the orbital contributions of the four fragments
1 3

Transit Met Chem
character into orbitals with carbene and p-cymene character
are calculated to occur. Near 260 nm, the computed excita-
tions involve electrons going from either carbene (HOMO-
5)- or Ru/I (HOMO-4 through HOMO-1)-based orbitals to
carbene-based orbitals (LUMO + 2 and LUMO + 5), indi-
cating a mixture of ligand-to-ligand and metal-to-ligand
transitions. Near 330 nm, excitations are from Ru/I-based
MOs (HOMO-2 and HOMO-3) with majority I character
into various low-lying unoccupied MOs (LUMO through
LUMO + 4), some of which have heavy Ru character
while others are primarily on the p-cymene and carbene
ligands for a mixture of what might ordinarily be termed
as metal-to-metal and metal-to-ligand transitions. Lastly, in
the 390–440 nm range the excitations are from the highest
occupied MOs (HOMO-4 to HOMO) to the lowest unoccu-
pied orbitals (LUMO and LUMO + 1), thereby most closely
resembling an actual d-to-d charge transfer band. We see lit-
tle evidence in the experimental spectrum for these “d-to-d”
excitations save for a broad tail from approximately 400 to
600 nm.
the 1H NMR of the catalyst. However, the active catalyst was
not observed using 1H NMR.
The ruthenium(II) p-cymene complex, 1, can withstand
air and moisture. It was not necessary to run reactions using
precatalyst 1 under a nitrogen atmosphere nor were dry sol-
vents required for the reactivity experiments. Using aceto-
phenone as a substrate, signifcant product conversion was
observed. Conversion after three and 40 min was 28 and
92%, respectively. Conversion after fve minutes and ten
minutes was 50 and 70%, respectively. Conversion after 15
and 30 min was 76 and 87%, respectively. The conversion
was essentially complete after 1 h.
Complex 1 could not be completely recycled for the trans-
fer hydrogenation of acetophenone. An aliquot of the reac-
tion mixture was removed after 1 h of transfer hydrogenation
at which 94% conversion had been achieved. After 1 h of
reaction time, an additional 1.8 mmol (0.22 g) of acetophe-
none was added. Allowing the reaction to continue for a
second hour, we observed 86% conversion of acetophenone
to 1-phenylethanol.
Complex deactivation
Transfer hydrogenation reactivity
Since we have an understanding of the activity complex
1 has for the transfer hydrogenation of acetophenone, we
wanted to test the stability of 1 during the catalytic reac-
tion. Based upon electrospray mass spectrometry data, we
are assuming that the precatalyst has dissociated an iodide
ligand to form the active catalyst. As noted before, attempts
to obtain an 1H NMR of the catalyst after the reaction were
unsuccessful. We and others have hypothesized that the loss
of the p-cymene ligand during catalysis will cause the cata-
lyst to deactivate, thereby hindering its reactivity [23]. We
therefore wanted to determine how much p-cymene dissoci-
ated from the starting precatalyst during transfer hydrogena-
tion. To accomplish this, we conducted the transfer hydro-
genation reaction in the presence of catalyst, base, and an
internal standard, 1,3,5-trimethoxybenzene. The results are
summarized in Table 4. The relative integration of the meth-
oxy protons was set to 9H at δ 3.6 ppm.
Having established a synthetic protocol for this organome-
tallic complex and gained an understanding of its structural
and electronic characteristics, we turned our attention to
probing its catalytic activity. Complex 1 was screened for
transfer hydrogenation activity for the reduction of aceto-
phenone. The solvent for the reaction was 2-propanol, and
the added base was KOH. The substrate/precatalyst/base
1
ratio was 100:1:5. For the reduction of acetophenone, H
NMR spectrometry was used to follow the disappearance
of the methyl protons (singlet, δ 2.6 ppm) and the appear-
ance of the methyl protons in the alcohol product (doublet,
δ 1.6 ppm). We observed 94% conversion after 1 h. This
conversion corresponds to a turnover number of 94. Prod-
1
uct formation was detected by H NMR spectroscopy, by
comparison with authentic material. Control reactions with
substrate plus base (no precatalyst) and substrate plus pre-
catalyst (no base) were conducted for 1 h. In both cases,
less than 5% conversion was observed. Complex 1 exhibits
transfer hydrogenation activity toward acetophenone which
is comparable to previously published ruthenium(II) N-het-
erocyclic carbene complexes [15, 39–41]. Complex 1 was
also screened for the catalytic reduction of an electron-poor
substrate, 4-nitroacetophenone and of an electron-rich sub-
strate, 4-methoxyacetophenone following the same catalytic
conditions, which were substrate/precatalyst/base ratio of
100:1:5. Complex 1 showed less than fve percent conver-
sion of 4-nitroacetophenone and 4-methoxyacetophenone.
Another objective was to obtain a 1H NMR of the reaction
mixture after 1 h of reaction time with the goal of obtaining
In addition, we conducted control reactions. We heated
the precatalyst in DMSO-d6 for 1 h at 85 °C. After this
heating time, we acquired a 1H NMR at room temperature
Table 4 Deactivation of 1
Entry Time (min) % Decom-
after various elapsed times in
position
the transfer hydrogenation of
of 1
acetophenone
1
2
3
4
15
30
45
60
47
59
65
80
1 3

Transit Met Chem
(supporting information) and we did not notice any changes
in the resonances of the complex compared to the initial 1H
NMR spectrum. In addition, we heated complex 1 (0.010
mmoles) in the presence of KOH (0.029 mmoles) in iso-
propanol for 1 h at refux. We measured the 1H NMR of the
product in DMSO-d6, and we noticed that the spectrum was
diferent compared to the 1H NMR spectrum of complex 1
(supporting information). We are assuming that the isoprox-
ide ion is reacting with complex 1.
for catalytic activity for the transfer hydrogenation of aceto-
phenone. We achieved 94% conversion after 1 h of reaction
time. To further understand the properties of 1, we tested
its stability during the catalytic reaction. We observed after
1 h of reaction time that 80% of the p-cymene ligand had
decomposed. The loss of a p-cymene ligand is an important
decomposition pathway that has to be considered in transi-
tion metal catalysis. A computational study of the UV–Vis
spectrum of 1 shows considerable mixing of Ru and I orbital
character into the frontier molecular orbitals, giving rise to
the usual types of “d–d” and MLCT excitations albeit with
some of the “d” and “metal” orbitals more properly viewed
as Ru/I hybrids, some of which have more I character than
Ru.
We saw after 1 h of reaction time that 80% of the
p-cymene ligand had dissociated (supporting information).
This was determined with 1H NMR spectroscopy and fol-
lowing the change of the p-cymene resonance that integrates
to one proton at δ 6.38 ppm. This resonance integrated to
0.20 protons after 1 h of reaction time. One could also follow
the p-cymene resonance that integrates to two protons at δ
6.25 ppm. This resonance integrated to 0.43 protons after
1 h of reaction time. This key experiment confrmed that the
p-cymene ligand dissociated during catalysis. The loss of an
arene ligand may be prevalent in homogenous catalysis in
general. Hydrogenation of the p-cymene ligand during the
reaction was not observed since we did not observe 1H NMR
signals for the hydrogenated product. We did notice unbound
Acknowledgements JRM would like to thank Fairfeld University for
awarding him a generous sabbatical leave when part of this research
was accomplished. JRM thanks Professor Robert Crabtree for hosting
him during his sabbatical leave and for helpful suggestions. This work
was supported by generous funding from the Fairfeld University Sum-
mer Research Kuck Fund (NAB, SCB, and RMK). The authors (NAB,
CAVA, SCB, MEM, and RMK) acknowledge support from Fairfeld
University Hardiman and Lawrence Scholarships for undergraduate
research expenses. MAL is grateful for receipt of an NTID Faculty
Evaluation and Development (FEAD) grant. We all thank the reviewers
for helpful suggestions as we revised the manuscript.
1
p-cymene at δ 5.86 ppm present in the H NMR that cor-
responded to 1.16 protons after 1 h of reaction time. This
resonance should integrate to four protons if the p-cymene
were not coordinated to the metal center.
References
We also used mesitylene, 1,3,5-trimethylbenzene, instead
of 1,3,5-trimethoxybenzene as an internal standard for the
deactivation studies. We saw after 60 min of reaction time
that 61% of the p-cymene ligand had dissociated (supporting
information). This was determined with 1H NMR spectros-
copy and following the change of the p-cymene resonance
that integrates to one proton at δ 6.38 ppm. This resonance
integrated to 0.39 protons after 1 h of reaction time. One
could also follow the p-cymene resonance that integrates to
two protons at δ 6.25 ppm. This resonance integrated to 0.85
protons after 1 h of reaction time. We did notice unbound
1. Gladiali S, Mestroni G (1998) Transition metals for organic syn-
thesis, Beller M, Bolm C, Eds. Wiley-VCH: Weinheim, Germany,
Vol 2, 97–119
2. Zassinovich G, Mestroni G, Gladiali S (1992) Asymmetric hydro-
gen transfer reactions promoted by homogenous transition metal
complexes. Chem Rev 92:1051–1069. https://doi.org/10.1021/
cr00013a015
3. Johnstone RAW, Wilby AH, Entwistle ID (1985) Heterogenous
catalytic transfer hydrogenation and its relation to other meth-
ods for reduction of organic compounds. Chem Rev 85:129–170.
https://doi.org/10.1021/cr00066a003
4. Bullock RM (2004) Catalytic ionic hydrogenations. Chem Eur J
10:2366–2374. https://doi.org/10.1002/chem.200305639
5. Guillena G, Ramón DJ, Yus M (2007) Alcohols as electrophiles
in C–C bond-forming reactions: the hydrogen autotransfer pro-
cess. Angew Chem Int Ed 46:2358–2364. https://doi.org/10.1002/
anie.200603794
1
p-cymene at δ 5.86 ppm present in the H NMR that cor-
responded to 1.54 protons after 1 h of reaction time.
6. Nixon TD, Whittlesey MK, Williams JMJ (2009) Transition metal
catalyzed reactions of alcohols using borrowing hydrogen meth-
odology. Dalton Trans. https://doi.org/10.1039/B813383B
7. Dobereiner GE, Crabtree RH (2009) Dehydrogenation as a sub-
strate-activating strategy in homogenous transition-metal cataly-
sis. Chem Rev 110:681–703. https://doi.org/10.1021/cr900202j
8. Tanoue K, Yamashita M (2015) Synthesis of pincer iridium
Conclusions
We have reported the successful two-step synthesis of an
N-heterocyclic carbene ruthenium(II) complex as well as its
use as a catalyst in the transfer hydrogenation of acetophe-
none. The synthesis of the catalyst frst requires the prepara-
tion of a silver(I) carbene complex followed by transmeta-
lation with [Ru(p-cymene)Cl2]2 to form complex 1, which
we have characterized with NMR spectroscopy, UV–visible
spectroscopy, elemental analyses, IR spectroscopy and sin-
gle crystal X-ray crystallography. Complex 1 was screened
i
complexes bearing a boron atom and Pr-substituted phospho-
rous atoms: applications to catalytic transfer dehydrogenation of
alkanes. Organometallics 34:4011–4017. https://doi.org/10.1021/
acs.organomet.5b00376
9. Arduengo AJ, Dias HVR, Harlow RI, Kline M (1992) Electronic
stabilization of nucleophilic carbenes. J Am Chem Soc 114:5530.
https://doi.org/10.1021/ja00040a007
1 3

Transit Met Chem
10. Lappert MF (1988) The coordination chemistry of electron rich alk-
enes (enetetramines). J. Organometallic Chem 358:185. https://doi.
org/10.1016/0022-328X(88)87079-7
Vreven T, Throssell K, Montgomery Jr. JA, Peralta JE, Ogliaro F,
Bearpark M, Heyd JJ, Brothers E, Kudin KN, Staroverov VN, Keith
T, Kobayashi R, Normand J, Raghavachari K, Rendell A, Burant
JC, Iyengar SS, Tomasi J, Cossi M, Millam JM, Klene M, Adamo
C, Cammi R, Ochterski JW, Martin RL, Morokuma K, Farkas O,
Foresman JB, Fox DJ (2013) Gaussian 09, Revision D.01, Gaussian,
Inc.: Wallingford CT
11. Bourissou D, Guerret O, Gabbaï FP, Bertrand G (2000) Stable car-
benes. Chem Rev 100:39–92. https://doi.org/10.1021/cr940472u
12. Hillier AC, Lee HM, Stevens ED, Nolan SP (2001) Cationic iridium
complexes bearing imidazol-2-ylidene ligands as transfer hydrogen-
ation catalysts. Organometallics 20:4246. https://doi.org/10.1021/
om0103456
28. Wang HMJ, Lin IJB (1998) Facile synthesis of silver(I)-carbene
complexes. Useful carbene transfer agents. Organometallics 17:972.
https://doi.org/10.1021/om9709704
13. Wang D, Astruc D (2015) The golden age of transfer hydrogena-
tion. Chem Rev 115:6621–6686. https://doi.org/10.1021/acs.
chemrev.5b00203
29. Ghattas W, Müller-Bunz H, Albrecht M (2010) [Ru(bpy)3]2+ ana-
logues containing an N-heterocyclic carbene ligand. Organometal-
lics 29:6782–6789. https://doi.org/10.1021/om100925j
14. Noyori R, Hashiguchi S (1997) Asymmetric transfer hydrogenation
catalyzed by chiral ruthenium complexes. Acc Chem Res 30:97–
102. https://doi.org/10.1021/ar9502341
30. Sabater S, Mata JA, Peris E (2012) Heterobimetallic iridium-ruthe-
nium assemblies through an ambidentate triazole-diylidene ligand:
electrochemical properties and catalytic behavior in a cascade reac-
tion. Organometallics 31:6450. https://doi.org/10.1021/om300675p
31. Zhang Y, Chen C, Ghosh SC, Li Y, Hong SH (2010) Well defned
N-heterocyclic carbine based ruthenium catalysts for direct amide
synthesis from alcohols and amines. Organometallics 29:1374–
1378. https://doi.org/10.1021/om901020h
15. Yaşar S, Karaca EO, Șahin C, Özdemir I, Șahin O, Büyükgüngör
O (2015) Novel ruthenium(II)-N-heterocyclic carbene complexes:
synthesis, characterization, and catalytic application. J Organometal-
lic Chem. https://doi.org/10.1016/j.organchem.2015.04.012
16. Hintermair U, Campos J, Brewster TP, Pratt LM, Schley ND, Crab-
tree RH (2014) Hydrogen-transfer catalysis with Cp*IrIII complexes:
the infuence of the ancillary ligands. ACS Catalysis 4:99–108.
https://doi.org/10.1021/cs400834
32. Kumar S, Narayanan A, Rao MN, Shaikh MM, Ghosh P (2011)
Ruthenium complexes of chelating amido-functionalized N-hetero-
cyclic carbene ligands: synthesis, structure, and DFT studies. J Chem
Sci 123(6):791–798. https://doi.org/10.10007/s12039-011-0173-5
33. Kauhfhold O, Flores-Figueroa A, Pape T, Hahn FE (2009) Template
synthesis of ruthenium complexes with saturated and benzannulated
NH, NH-stabilized N-heterocyclic carbene ligands. Organometallics
28:896–901. https://doi.org/10.1021/om800964n
17. Albrecht M, Miecznikowski JR, Samuel A, Faller JW, Crabtree RH
(2002) Chelated iridium(III) bis-carbene complexes as air-stable
catalysts for transfer hydrogenation. Organometallics 21:3596–3604.
https://doi.org/10.1021/om020338x
18. Pàmies O, Bäckvall J-E (2001) Studies on the mechanism of
metal-catalyzed hydrogen transfer from alcohols to ketones.
Chem Eur J 7:5052–5058. https://doi.org/10.1002/1521-
3765(20011203)7:23<5052:AID-CHEM5052>3.0.CO;2-Z
19. Samec JSM, Bäckvall J-E (2002) Ruthenium-catalyzed trans-
fer hydrogenation of imines by propan-2-ol in benzene.
Chem Eur J 8:2955–2961. https://doi.org/10.1002/1521-
3765(20020703)8:13<2955:AID-CHEM2955>3.0.CO;2-Q
20. Kuhl S, Schneider R, Fort Y (2003) Transfer hydrogenation of
imines catalyzed by a nickel(0)/NHC complex. Organometallics
22:4184–4186. https://doi.org/10.1021/om034046n
34. Grau J, Noe V, Ciudad C, Prieto MJ, Font-Bardia M, Calvet T,
Moreno V (2012) New π-arene ruthenium(II) piano-stool com-
plexes with nitrogen ligands. J Inorg Biochem 109:72–81. https://
doi.org/10.1016/j.jinorgbio.2012.01.003
35. Gichumbi JM, Friedrich HB, Omondi B (2016) Synthesis and
characterization of piano-stool ruthenium complexes with N, N′-
pyridine imine bidentate ligands and their application in styrene
oxidation. J Organomet Chem 808:87–96. https://doi.org/10.1016/j.
jorganchem.2016.02.015
21. Farrell K, Müller-Bunz H, Albrecht M (2015) Synthesis, isomeri-
zation and catalytic transfer hydrogenation activity of rhodium(III)
complexes containing both chelating dicarbenes and diphosphine
ligands. Organometallics 34:5723–5733. https://doi.org/10.1021/
acs.organomet.5b00809
36. Małecki JG, Jaworska M, Kruszynski R (2006) Synthesis, molecular,
crystal and electronic structure of [(C6H6)RuCl2(picoline)]. Polyhe-
dron 25:2519–2524. https://doi.org/10.1016/j.poly.2006.02.016
37. Bratsos I, Urankar D, Zangrando E, Genova-Kalou P, Košmrlj J,
Alessio E, Turel I (2011) 1-(2-Picolyl)-substituted 1,2,3-triazole as
novel chelating ligand for the preparation of ruthenium complexes
with potential anticancer activity. Dalton Trans 40:5188–5199.
https://doi.org/10.1039/c0dt01807d
22. Fernandez FE, Puerta MC, Valerga P (2012) Ruthenium(II) picolyl-
NHC complexes: synthesis, characterization, and catalytic activity
in amine N-alkylation and transfer hydrogenation reactions. Orga-
nometallics 31:6868–6879. https://doi.org/10.1021/om300692a
23. Campos J, Hintermair U, Brewster TP, Takase MK, Crabtree RH
(2014) Catalyst activation by loss of cyclopentadienyl ligands in
hydrogen transfer catalysis with Cp*IrIII complexes. ACS Catalysis
4:973–985. https://doi.org/10.1021/cs401138f
38. Shirin Z, Pramanik A, Ghosh P, Mukherjee R (1996) Stable
cyclohexadienyl complexes of ruthenium in a piano stool geometry
containing a tridentate nitrogen donor ligand. First structural char-
acterization of the (η5-cyanocyclohexadienyl)ruthenium(II) complex
and spectroelectrochemical correlation. Inorg Chem 35:3431–3433.
https://doi.org/10.1021/ic950750n
24. Grimmett MR (1997) Imidazole and benzimidazole synthesis, Aca-
demic Press, 202
39. Horn S, Gandolf C, Albrecht M (2011) Transfer hydrogenation of
ketones and activated olefns using chelating NHC ruthenium com-
plexes. Eur J Inorg Chem 18:2863–2868. https://doi.org/10.1002/
ejic.201100143
25. CrystalClear and CrystalStructure; Rigaku/MSC: The Woodlands,
TX, 2005
26. Sheldrick GM (2008) A short history of SHELX. Acta Cryst.
A64:112–122. https://doi.org/10.1107/S0108767307043930
27. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Scalmani G, Barone V, Petersson GA, Nakatsuji
H, Li X, Caricato M, Marenich A, Bloino J, Janesko BG, Gomperts
R, Mennucci B, Hratchian HP, Ortiz JV, Izmaylov AF, Sonnenberg
JL, Williams-Young D, Ding F, Lipparini F, Egidi F, Goings J, Peng
B, Petrone A, Henderson T, Ranasinghe D, Zakrzewski VG, Gao J,
Rega N, Zheng G, Liang W, Hada M, Ehara M, Toyota K, Fukuda
R, Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O, Nakai H,
40. Nini D, Hor TSA (2011) Syntheses and structures of ruthenium(II)
N,S-heterocyclic carbene diphosphine complexes and their catalytic
activity towards transfer hydrogenation. Chem Asian J. https://doi.
org/10.1002/asia.201000930
41. Yigit M, Yigit B, Ozdemir I, Cetinkaya E, Cetinkaya B (2006)
Active ruthenium-(N-heterocyclic carbene) complexes for hydro-
genation of ketones. Appl Organomet Chem 20(5):322–327. https://
doi.org/10.1002/aoc.1054
1 3