Bridging Efficiency within Multinuclear Homogeneous Catalysts in the Photocatalytic Reduction of Carbon Dioxide

Article abstract of DOI:10.1002/cctc.201500674

A trinuclear complex consisting of one [Ru(dmb)₃]²⁺ (dmb=4,4′-dimethyl-2,2′-bipyridine) (Ru) and two [Re(dmb)(CO)₃Cl] (Re) building blocks, [Re(CO)₃Cl(dmb-dmb)Ru(dmb)(dmb-dmb)Re(CO)₃Cl](PF₆)₂ (Re-Ru-Re), is presented. Photophysical properties of Re-Ru-Re and the individual components with different or no covalent linkages are thoroughly investigated and compared. To elucidate the role of the single covalent bonds, photocatalytic reduction of CO₂ is performed with the trinuclear complex and a series of model systems featuring systematic absence of linkages between the metal centers. Photoluminescence spectra and quantum yields reveal efficient energy transfer from the excited state of Re to Ru if these fragments are covalently linked. Moreover, intramolecular electron transfer from the one-electron reduced species of Ru to Re occurs if there is covalent bonding, leading to a higher photostability and thus the highest turnover number in photocatalytic CO₂ reduction of 199 for the trinuclear complex Re-Ru-Re within the systems under investigation. Optimized experimental conditions reveal the highest turnover number (315) reported to date for Re^I/Ru^II-based homogeneous catalysts in photocatalytic CO₂ reduction.

Full text of DOI:10.1002/cctc.201500674

DOI: 10.1002/cctc.201500674
Full Papers
Bridging Efficiency within Multinuclear Homogeneous
Catalysts in the Photocatalytic Reduction of Carbon
Dioxide
[
a]
Simon Meister, Richard O. Reithmeier, Alexander Ogrodnik, and Bernhard Rieger*
2
+
A trinuclear complex consisting of one [Ru(dmb)3] (dmb=
,4’-dimethyl-2,2’-bipyridine) (Ru) and two [Re(dmb)(CO) Cl]
energy transfer from the excited state of Re to Ru if these frag-
ments are covalently linked. Moreover, intramolecular electron
4
3
(
Re) building blocks, [Re(CO) Cl(dmbÀdmb)Ru(dmb)(dmbÀ transfer from the one-electron reduced species of Ru to Re
3
dmb)Re(CO) Cl](PF ) (ReÀRuÀRe), is presented. Photophysical
occurs if there is covalent bonding, leading to a higher photo-
stability and thus the highest turnover number in photocata-
3
6 2
properties of ReÀRuÀRe and the individual components with
different or no covalent linkages are thoroughly investigated
and compared. To elucidate the role of the single covalent
lytic CO reduction of 199 for the trinuclear complex ReÀRuÀ
2
Re within the systems under investigation. Optimized experi-
bonds, photocatalytic reduction of CO is performed with the
mental conditions reveal the highest turnover number (315) re-
2
I
II
trinuclear complex and a series of model systems featuring sys-
tematic absence of linkages between the metal centers. Photo-
luminescence spectra and quantum yields reveal efficient
ported to date for Re /Ru -based homogeneous catalysts in
photocatalytic CO reduction.
2
Introduction
Photocatalysis has emerged as an elegant methodology for en-
the Ru moiety is typically excited and subsequently reduced
by BNAH, followed by electron transfer from the reduced Ru
[1]
ergetically disfavored reactions. It has become a viable ap-
proach for the production of solar fuels, paving the way for
complex to the Re catalyst, where CO reduction takes place.
2
[2]
[3]
the valorization of H O or CO2. The latter has proven useful
Several attempts have been made to further increase catalytic
performance by covalently linking oxidation and reduction
2
as an alternative and sustainable C feedstock, especially if the
1
[
9,10]
[11]
reduction of CO allows for the exploitation of established syn-
sites
and by connecting two reduction sites.
In both
2
[
4]
thetic pathways. The selective reduction of CO to CO can be
cases, the length of the linker was revealed to govern the turn-
over number (TON) and turnover frequency (TOF). It was found
that an ethyl tether between two reduction catalysts promoted
a binuclear mechanism, whereas the same linker between a re-
duction and an oxidation site facilitated electron transfer from
2
I
performed with Re catalysts, as reported by Hawecker, Lehn,
[5]
and Ziessel in the early 1980s. These complexes of the gener-
al formula [Re(NN)(CO) X] (NN=a,a’-dipyridyl or phenanthro-
3
À
À
À
line; X=Cl , Br , or SCN ) are able to act as photosensitizer,
oxidation site, and reduction site all at once. After excitation,
a sacrificial amine [typically triethanolamine (TEOA) or triethyla-
BNAH to CO , increasing both TON and TOF. Interestingly, con-
2
jugated linkers were found to be unsuitable in CO reduction,
2
[6]
mine] is oxidized and an electron transferred to CO2. Recently,
our group was able to show that TEOA radicals and excessive
irradiation contribute to catalyst deactivation, accounting for
although a more efficient electron transfer would be expected.
In fact, the electronic properties of the complexes were
changed in such a way that catalytic performance deteriora-
[7]
[10c]
low turnovers and poor catalyst stability. One approach to
enhance catalytic performance is the separation of photosensi-
tizer, oxidation site, and reduction site. This may be achieved
ted.
Consequently, it can be concluded that the ideal photo-
catalyst for CO reduction should consist of at least one photo-
2
sensitizer unit and two reduction sites, which are all in spatial
proximity and connected by short alkyl chains to support fast
electron transfer and a binuclear mechanism. Our focus was to
II
by using Ru complexes as photosensitizer and oxidation site,
for example, leaving the task of selective CO reduction to the
2
I
[8]
Re catalyst. Major improvements are observed in terms of
determine which covalent bond was crucial for efficient CO re-
2
I
II
turnovers and stability if mixed Re /Ru catalytic systems with
-benzyl-1,4-dihydronicotinamide (BNAH) as sacrificial amine
duction. However, the wide range of reaction conditions and
catalytic setups rendered comparison of different studies diffi-
cult. Herein, we present four similar trinuclear catalytic systems,
1
[9]
and CO as electron acceptor are used. In this type of system,
2
each of them consisting of two [Re(dmb)(CO) Cl] (dmb=4,4’-
3
[
a] S. Meister, R. O. Reithmeier, Dr. A. Ogrodnik, Prof. Dr. B. Rieger
Wacker Lehrstuhl für Makromolekulare Chemie
TU München - Technische Chemie
dimethyl-2,2’-bipyridine) reduction catalysts and one
2+
[
Ru(dmb)3] photosensitizer and oxidation site, differing only
in the way these centers are covalently linked (Figure 1). Pho-
tophysical properties of the trinuclear systems and their single
components were investigated in detail and photocatalytic re-
Lichtenbergstr. 4, 85748 Garching (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201500674.
ChemCatChem 2015, 7, 3562 – 3569
3562
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
I
II
2
Figure 1. Model systems used for CO reduction incorporating two Re centers and one Ru center with different bridging modes.
duction of CO was performed to elucidate the relevance of
(CO) Cl(dmbÀdmb)] for both complexes, after which the pho-
2
3
each single covalent connection of the two metal centers.
tosensitizer was assembled. For the synthesis of ReÀRuÀRe,
two equivalents of [Re(CO) Cl(dmbÀdmb)] were brought to re-
3
action with one equivalent of [Ru(cod)Cl ] (cod=1,5-cyclooc-
2
n
Results and Discussion
tadiene) in a microwave reactor with moderate irradiation
50 W), whereas for the synthesis of ReÀRu, dmb was used in-
(
Synthesis and characterization
stead of [Re(CO) Cl(dmbÀdmb)]. The exclusion of H O during
3
2
[7]
The complexes [Re(dmb)(CO) Cl] (Re) and [Ru(dmb) ](PF )
6 2
this step prevented substitution of the chloro ligands at the Ru
3
3
[12]
[12]
(
Ru) were prepared according to procedures previously re-
center. In the following step, MeOH was added to facilitate
À
ported in literature. [Re(CO) Cl(dmbÀdmb)Re(CO) Cl] (ReÀRe)
Cl dissociation and coordination of either dmb for the synthe-
3
3
was prepared following the procedure reported for the bromi-
sis of ReÀRuÀRe or [Re(CO) Cl(dmbÀdmb)] for the synthesis of
3
[
11]
À
um derivative. A three-step synthesis for [Re(CO) Cl(dmbÀ ReÀRu. Salt metathesis yielded the PF salts of the desired
3
6
dmb)Ru(dmb) ](PF ) (ReÀRu) and the new trinuclear complex
complexes. All complexes were characterized fully by using
2
6 2
1
[
(
Re(CO) Cl(dmbÀdmb)Ru(dmb)(dmbÀdmb)Re(CO) Cl](PF )
H NMR, ESI-MS, and UV/Vis spectroscopies and elemental anal-
3
3
6 2
ReÀRuÀRe) was developed, which is shown for the latter in
ysis. For the complexes containing carbonyl groups, IR spectra
were recorded additionally (see the Experimental Section).
Scheme 1. The first step comprised the synthesis of [Re-
Scheme 1. Synthesis of the trinuclear complex ReÀRuÀRe.
ChemCatChem 2015, 7, 3562 – 3569
www.chemcatchem.org
3563
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Photophysical properties
The metal centers in the bi- and trinuclear complexes ReÀRe,
ReÀRu, and ReÀRuÀRe are connected by non-conjugated
bridging ligands, which should hinder electronic communica-
tion between the metal centers in the electronic ground
[
10c,13]
state.
This is confirmed in UV/Vis spectra of the complex
mixtures Re+Re+Ru, ReÀRe+Ru, ReÀRu+Re, and ReÀRuÀ
Re, which strongly resemble one another (Figure 2). The spec-
tra are of additive nature, that is, the absorbance at a given
wavelength is identical to the sum of the absorbance of Ru
and two times the absorbance of Re, indicating that electronic
communication in the ground state does not occur. Absorption
maxima and corresponding extinction coefficients are given in
Table 1. Comparison with the spectra of isolated Ru or Re re-
veals that the broad absorption in the visible region at l=
Figure 2. UV/Vis spectra of the four trinuclear model systems. Spectra of
Re+Re and Ru are depicted for comparison.
4
20–500 nm stems from the photosensitizer Ru, whereas the
Re moieties absorb only in the
UV region up to l=420 nm. The
lowest energy transitions with
absorption maxima at l=
Table 1. Photophysical data for the four trinuclear model systems and their single components including ab-
sorption (labs) and emission (lem) maxima, Stern–Volmer constants (KSV), lifetimes of the excited states (t),
4
60 nm (Ru) and l=363 nm (Ru
quenching rate constants (k ), and photoluminescence quantum yields (Fem) for Ru (if not stated otherwise)
Q
and Re) can be assigned to
for excitation at l=365 and 520 nm.
metal-to-ligand charge transfer
[
a]
[a]
[b]
[b]
[b]
em
[b]
[b,c]
[d]
[b,e]
[b,f]
System
[Re]
mm]
[Ru]
l
abs
e
l
K
SV
À1
t
k
Q
F
em
F
em
(
MLCT; dp!p*), whereas the
3
À1
À1
À1
À1
[
[mm]
[nm] [10 m cm
]
[nm] [m
]
[ns]
[m ns
]
strong absorption in the UV
range at lꢀ290 nm is associat-
ed with intraligand processes
Re+Re+Ru 50
25
460
62
289
460
63
89
460
363
89
460
63
90
462
61
89
461
60
289
462
27
89
16.2
16.2
125.4
17.3
16.6
132.6
17.2
16.1
128.1
16.5
16.5
119.9
17.1
11.9
107.7
16.5
12.0
107.9
16.6
12.6
93.4
3.9
633
633
634
634
634
633
633
609
7923 879
8043 881
8183 892
8083 899
7704 890
8120 876
9104 888
9.01
0.065
0.068
0.103
0.125
0.108
0.073
0.083
0.007
0.085
0.083
0.090
0.077
0.089
0.082
0.083
0.000
3
[
14]
ReÀRe+Ru
ReÀRu+Re
ReÀRuÀRe
ReÀRu
50
50
50
25
25
0
25
25
25
25
25
25
0
9.13
9.17
8.99
8.66
9.27
10.25
–
(
p!p*).
The normalized emission
3
2
spectra of the four trinuclear
mixtures appear almost identi-
cal, with an emission maximum
at l=633–634 nm (Table 1 and
Figure 3). The spectra are domi-
nated by a strong emission from
Ru, which appears at a maximum
of l=633 nm with a relatively
high quantum yield (0.083). The
contribution of Re with a maxi-
mum at l=609 nm is rather
weak due to its low quantum
yield (0.007) but still in a detect-
able range. In Figure 3, the non-
normalized spectra of isolated
Re (50 mm) and Ru (25 mm) are
displayed, showing their original
relative intensities. Assuming
the absence of electronic com-
munication between the com-
2
3
2
3
2
Re+Ru
Ru
3
[
g]
[g]
[i]
3
2
[h]
[i]
Re
25
364
291
–
27
14.8
[
a] Concentration of metal centers; [b] Measured in Ar-saturated DMF solution (e=extinction coefficient); [c] Ex-
citation with a short laser pulse (<15 ns) at l=355 nm; [d] Calculated from KSV and t; [e] lex =365 nm; [f] lex
520 nm; [g] From Ref. [16c], converted for DMF as described in Ref. [17]; [h] From Ref. [6a] (in MeCN); [i] Quan-
tum efficiencies for photoluminescence of Re.
=
plexes in the trinuclear mixture without covalent linkages, the
calculated sum of the spectra of Ru and Re+Re should reflect
the emission spectrum of Re+Re+Ru. However, the emission
intensity of Re+Re+Ru is lower than the calculated sum, but
through normalization it is possible to superimpose the mea-
sured and calculated spectra (Figure S2). The loss of intensity
correlates with a lower quantum yield for Re+Re+Ru com-
pared with Ru. This can be attributed to a filter effect: at the
excitation wavelength (365 nm), a 50 mm solution of Re (i.e.,
Re+Re) and a 25 mm solution of Ru absorb approximately
equally, causing the absorbance of Re+Re+Ru to be twice as
high as that of the single components. Considering the loga-
rithmic correlation between absorbance and absorption, the
25 mm trinuclear mixture (50 mm Re+25 mm Ru) absorbs less
ChemCatChem 2015, 7, 3562 – 3569
www.chemcatchem.org
3564
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
imposed with that of pure Ru, indicating that no emission
from Re in the trinuclear complex occurs (Figure S3). However,
with decreasing emissions from Re, quantum yields increase,
which is only feasible if energy transfer from excited Re to Ru
occurs. A similar phenomenon has been observed before in
[
15]
a related complex. Consequently, covalent linkage of Re and
3
I
Ru enhances production of the Ru MLCT state because Re
moieties act as light antennae if irradiation is performed with
UV or blue light. To further substantiate the proposed energy
transfer mechanism, the determination of quantum yields was
repeated by using green light (l=520 nm) for the excitation
of the complexes. Therein, quantum yields should no longer
be affected by energy transfer from Re to Ru because Re
cannot be excited. As expected, quantum yields are identical
for all mixtures within the range of error (Table 1). Deviations
might be a result of imprecise integration of emission signals
due to overlapping of excitation (l=520 nm) and emission
(
l=633 nm) wavelengths (Figure S5).
Photoluminescence is quenched efficiently with BNAH.
Stern–Volmer plots reveal very similar Stern–Volmer constants
KSV) for all four systems under investigation (Figure 4). The life-
(
Figure 3. Normalized photoluminescence spectra of the four trinuclear
model systems (lex =365 nm, concentration=25 mm for every listed metal).
Original (non-normalized) spectra for Re+Re and Ru are shown for compari-
son. Re+Re+Ru is normalized to the sum of Re+Re and Ru. ReÀRe+Ru,
ReÀRu+Re, and ReÀRuÀRe are normalized to Re+Ru+Re and only differ
from the sum of Re+Re and Ru in the region of l=500–580 nm.
Figure 4. Stern–Volmer plots of deoxygenated DMF solutions of the four tri-
nuclear model systems. lex =365 nm, quencher (Q)=BNAH.
than double (1.7-fold) the amount of photons compared with
the 25 mm solution of Ru or the 50 mm solution of Re. Under
the experimental conditions used in this study, only 83% of
the respective complexes in the trinuclear mixture are excited
compared with the homonuclear solutions (see the Supporting
Information for detailed calculations), and thus emission inten-
sity is somewhat lower.
times (t) of the excited states range between 880 and 900 ns,
3
[14b,16]
which corresponds to the MLCT lifetime of Ru.
The decay
of photoluminescence is shown exemplarily for ReÀRuÀRe in
Figure 5. Apparently, covalent linkage of the complexes does
not influence quenching efficiency [with respect to the
quenching rate constant (k )], which is reflected in similar K
For a better comparison, the emission spectra of the trinu-
clear mixtures are normalized to the calculated sum of Re+Re
and Ru spectra in Figure 3. The actual quantum yields are
given in Table 1 and non-normalized plots are shown in Fig-
ure S4. As mentioned above, superimposition is possible for
the mixture without any covalent linkages (Figure S2). The
other spectra show slight differences in the range of l=500–
Q
SV
and t values (Table 1). Stern–Volmer plots of all systems listed
in Table 1 are presented in the Supporting Information (Fig-
ure S8).
Catalytic performance
5
80 nm, suggesting decreasing emissions from the triplet
The trinuclear mixtures presented in Figure 1 were tested in
3
MLCT ( MLCT) of Re in the order ReÀRe+Ru>ReÀRu+Re>
ReÀRuÀRe. The emission spectrum of ReÀRuÀRe can be super-
photocatalytic CO reduction to evaluate the impact of cova-
2
lent linkages between each of the single complexes. In mixed
ChemCatChem 2015, 7, 3562 – 3569
www.chemcatchem.org
3565
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
À1
ed by a moderate TOF (1.3 min ) and TON (40) (Table 2). Con-
I
necting the Re centers (ReÀRe+Ru) causes an approximately
À1
threefold increase in both TOF (3.8 min ) and TON (129). CO
2
reduction at the Re catalyst is known to be rate-limiting in
Table 2. TOFs and TONs of the four trinuclear model systems (0.03 mm)
2
in CO -saturated DMF solution with BNAH (0.1m) and TEOA (1.7m).
II
Values were calculated per Ru center (for each mixture). Irradiation was
performed at l=520 nm.
À1
System
TOF [min
]
TON
Re+Re+Ru
ReÀRe+Ru
ReÀRu+Re
ReÀRuÀRe
1.3
3.8
4.3
4.2
n.d.
40
129
174
199
315
[a]
ReÀRuÀRe
Figure 5. Decay of photoluminescence after excitation of complex
ReÀRuÀRe with a short laser pulse (<15 ns) at l=355 nm.
[
a] 0.05 mm, irradiation intensity reduced to 2.5% compared to all other
experiments..
I
II
Re/Ru systems with BNAH as sacrificial electron donor and ir-
radiation in the visible range above l=500 nm, Ru is reduced
typically after excitation, producing a one-electron reduced
[
9]
such systems. Owing to the limited stability of the photosen-
sitizer, it is crucial how fast CO is converted at the Re moiety.
2
(
OER) species. Subsequently, the OER species of Re is built
Apparently, the binuclear CO2 reduction mechanism, which
I
[11]
through inter- or intramolecular electron transfer and CO re-
emerges if Re centers are forced into spatial proximity, con-
2
[9]
duction takes place at the Re catalyst. Usually, TEOA is added
in addition to BNAH to deprotonate the oxidized BNAH and
tributes significantly to acceleration of the CO reduction pro-
2
I
II
cess, even in mixed Re /Ru systems. Thus, in the timeframe
during which active photosensitizer is present (i.e., before de-
generation), more CO can be produced in the presence of the
faster reduction catalyst ReÀRe. However, a larger effect on
catalyst performance is observed if photosensitizer (Ru) and re-
duction catalyst (Re) are covalently linked. The activity (TOF)
[9,15a]
thus prevent electron back-transfer.
For the photocatalytic
experiments, the respective model system (Figure 1) was dis-
solved in DMF (0.03 mm), BNAH (0.1m) and TEOA (1.7m) were
added, and the solution was saturated with CO and irradiated
2
under ambient conditions with an LED light source emitting at
l=520 nm. The systems under investigation showed moderate
to good activity during the first hour of irradiation (Figure 6).
Deterioration during the second hour of irradiation was due to
degeneration of the photosensitizer Ru, which was confirmed
by experimental results from the periodic addition of Ru (Fig-
ure S6).
À1
reaches a rate of 4.2–4.3 min for both ReÀRu+Re and ReÀ
RuÀRe. Interestingly, the rate increase by enabling intramolec-
ular electron transfer from Ru to Re, which is possible through
[
10a]
the ethyl bridge,
seems to be independent of the number
of Re sites connected to the same Ru center. Similar TOFs for
both ReÀRu+Re and ReÀRuÀRe suggest that the Re moieties
As expected, the system with the mononuclear mixture
in ReÀRuÀRe are not close enough to promote binuclear CO
2
(
Re+Re+Ru) shows the lowest activity and stability, illustrat-
reduction. Nevertheless, the contribution from intramolecular
electron transfer to the overall activity is slightly larger than
the contribution from the binuclear mechanism, reflected in
À1
a higher TOF for ReÀRu+Re (4.3 min ) than for ReÀRe+Ru
À1
(
3.8 min ). Notably, domination of the binuclear mechanism
can be excluded for ReÀRu+Re because there is no spatial
proximity between the Re centers. Regarding catalyst long-
term stability, the TON can be increased by connecting more
Re sites to the photosensitizer, resulting in up to 200 turnovers
for the trinuclear catalyst under these conditions. Covalent
linkage of reduction catalysts to the photosensitizer enhances
electron transfer from Ru to Re, thus decreasing the lifetime of
[
9]
I
the reduced state at Ru (OER). As reported before for Re cat-
alysts and proposed for other photocatalysts, degeneration of
[
7]
the catalyst by excitation of the OER species is possible. This
is considered to be the main deactivation route for the degen-
eration of the photosensitizer. Therefore, shortening the life-
time of the OER photosensitizer accounts for higher catalyst
stability and increasing TONs. Consequently, the new catalyst
Figure 6. Time vs. conversion plots of the four trinuclear model systems
0.03 mm). Turnovers were calculated based on Ru content in each mixture.
Irradiation was performed at l=520 nm.
(
ChemCatChem 2015, 7, 3562 – 3569
www.chemcatchem.org
3566
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
ReÀRuÀRe is found to be the catalyst with the best per-
formance within the systems under investigation.
tennae. This is an important property, considering that many
light sources (including the sun) emit in the visible and UV re-
gions. Efficient multinuclear photocatalysts should be able to
use any part of the catalyst for the absorption process and re-
actions should be limited to a defined catalytic site to achieve
high selectivity. Thus, energy and electron transfer are two im-
portant parameters, whose optimization is crucial for catalyst
performance. The presented trinuclear catalyst combines elec-
tron and energy transfer between the respective catalytic sites
perfectly, with potential for efficient performance over a broad
irradiation spectrum. However, triethanolamine would have to
be substituted by a base that does not quench the excited
states of rhenium and ruthenium. Studies with irradiation at
various wavelengths using a different base are the subject of
further investigations in our laboratories.
To further increase the TON of the most active catalytic
system, experimental conditions were varied. Assuming that
I
CO reduction at the Re moiety is still rate-limiting in the trinu-
2
clear catalyst ReÀRuÀRe and taking into account recently pub-
lished results, it should be possible to further increase the TON
[7]
if a lower irradiation intensity is applied. Thereby, the OER
photosensitizer will be less prone to degeneration through
excess irradiation, with excitation being still sufficient to gener-
ate the OER species. In the following experiment, only 2.5% of
the original irradiation intensity was used. For comparability
with other research groups’ results, catalyst concentration was
adjusted to 0.05 mm. After 16 h of irradiation, the TON reached
2
44, exceeding those of comparable catalysts comprising cova-
I
II
[9,10,15a,18]
lently linked Re /Ru complexes.
Further irradiation re-
vealed ongoing activity for more than 24 h, resulting in a final
I
II
TON of 315, the highest TON reported to date for Re/Ru -
based catalysts (Figure S7).
Experimental Section
Instrumentation and measurements
Conclusions
All manipulations were performed by using standard Schlenk tech-
A
new
trinuclear
rhenium(I)/ruthenium(II)
complex
1
13
niques if not stated otherwise. H NMR and C NMR spectra were
measured on a Bruker AVIII-300 spectrometer at 298 K and refer-
[
(
Re(CO) Cl(dmbÀdmb)Ru(dmb)(dmbÀdmb)Re(CO) Cl](PF )
3
3
6 2
[19]
ReÀRuÀRe; dmb=4,4’-dimethyl-2,2’-bipyridine) has been syn-
enced to residual solvent signals. IR spectroscopy was performed
on a Bruker Vertex-70 FTIR spectrometer at RT in bulk material. ESI-
MS spectra were recorded on a Varian LC-MS 500 spectrometer.
UV/Vis spectra were measured on a Varian Cary 50 spectrophotom-
eter in a UV quartz cuvette (10 or 1 mm). Emission spectra were
measured on an Avantes Avaspec-2048 spectrometer at 258C by
using a Prizmatix Mic-LED-365 for excitation at l=365 nm or
a Luxeon K2 Emitter LED for excitation at l=520 nm. Photolumi-
nescence quantum yields were determined relative to the standard
[Ru(dmb) ](PF ) (Fem =0.083); for a detailed description, see the
thesized and its photophysical properties investigated thor-
oughly. Its photocatalytic activity in carbon dioxide reduction
was studied and compared to similar systems, in which differ-
ent modes of covalent linkage of the ruthenium and rhenium
centers were present. Thereby, the effect of every covalent link
between two or more complexes was revealed. Owing to slow
carbon dioxide reduction at the rhenium catalyst coinciding
with a fast degeneration of the ruthenium-based photosensi-
tizer, it was found that connecting two reduction catalysts en-
hanced photocatalytic performance in mixed rhenium(I)/ruthe-
nium(II) systems. This was attributed to a binuclear reduction
mechanism, which promoted faster carbon dioxide reduction.
However, covalent linkage of reduction catalyst and photosen-
sitizer was found to be slightly more relevant to yield higher
turnover frequencies and numbers (TONs) because intramolec-
ular electron transfer was enabled. This accounted for a re-
duced lifetime of the one-electron reduced photosensitizer,
rendering the system less prone to deactivation as a result of
photobleaching by excitation of the reduced photosensitizer.
Consequently, the new trinuclear complex ReÀRuÀRe was
found to be the catalyst with the highest TON (199) within the
four systems under investigation. Optimized experimental con-
ditions afforded the highest TON (315) reported to date for
rhenium(I)/ruthenium(II) mixed catalytic systems.
3
6 2
[16c,17]
Supporting Information.
Lifetimes of the excited states were
determined by fitting the exponential decay of the photolumines-
cence after a short laser pulse (<15 ns) with a wavelength of l=
3
55 nm (Nd:YAG). Stern–Volmer plots were measured from deoxy-
genated DMF solutions of the model systems and their compo-
nents (see Table 1). Elemental analysis was performed in the micro-
analytical laboratory at the Technische Universität München. GC
analysis was performed by using a Varian 490 gas chromatograph
equipped with a 1 m COX column and a GC thermal conductivity
detector. He (5.0) was used as the carrier gas. Irradiation experi-
ments were performed in a 160 mL Schlenk tube (internal diame-
ter=3 cm) with an LED light source emitting at l=(520Æ50) nm
(
for details, see the Supporting Information). Reaction vessels were
wrapped in black foil prior to sample preparation and unwrapped
just before irradiation experiments. DMF solutions containing the
catalytic mixture (0.03 mm, i.e., 0.03 mm Ru and 0.06 mm Re moi-
II
I
eties), BNAH (0.1m), and TEOA (1.7m) were saturated with CO by
2
bubbling for at least 15 min, sealed with a septum, and the pres-
sure was adjusted to 140 kPa. Irradiation experiments were per-
formed in a dark room. For GC analysis, irradiation was suspended
to draw 100 mL samples from the headspace above the solution
and inject them directly into the micro gas chromatograph. Irradia-
tion was continued after each GC sampling. TONs were determined
from the point at which catalyst activity declined fully, if not stated
otherwise, and were defined as TON=n /ncat. , in which cat. repre-
sents the investigated catalytic system or mixture. TOFs were cal-
culated from the linear slope of initial catalytic activity.
Moreover, an interesting phenomenon was observed in the
study of photoluminescence spectra and quantum yields: exci-
tation of the trinuclear complex ReÀRuÀRe with UV light re-
sulted in very efficient triplet–triplet energy transfer from the
triplet metal-to-ligand charge transfer of Re to Ru. Thus, be-
sides promoting an intramolecular electron transfer from the
one-electron reduced species of Ru to Re, the covalent con-
nection of rhenium moieties enabled them to act as light an-
CO
ChemCatChem 2015, 7, 3562 – 3569
www.chemcatchem.org
3567
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Materials
cooling to RT, the solvent was evaporated under reduced pressure
and the residual solid dissolved in H O (20 mL). A solution of
2
Dry solvents were obtained from an MBraun solvent purification
system (MP-SPS-800) or dried by using standard methods. NMR sol-
vents were purchased from Sigma–Aldrich and used without fur-
NH PF (750 mg, 4.6 mmol, 25.5 equiv.) in H O (20 mL) was added
4
6
2
and the precipitate collected by filtration over a fine frit. After
washing with H O (2”25 mL) and diethyl ether (3”25 mL), the
2
ther purification. The gases CO (4.5), Ar (4.8), and He (5.0) were
2
solid was dried under vacuum to yield a red powder (144 mg,
purchased from Westfalen. Dmb, pentacarbonylchlororhenium(I),
1
0
.15 mmol, 86% yield). H NMR (300 MHz, CD CN, 298 K): d=8.32
3
[
Ru(cod)Cl ] , 1,2-dibromoethane, NH PF , DMF, and TEOA were ob-
2 n 4 6
(
1
s, 6H), 7.51 (d, J=5.8 Hz, 6H), 7.20 (d, J=5.8 Hz, 6H), 2.51 ppm (s,
tained from Sigma–Aldrich or ABCR and used without further pu-
3
À1
À1
8H); UV/Vis (DMF): lmax (e 10 m cm )=290 (186.0), 462 nm
rification. BNAH was obtained from TCI. [Re(dmb)(CO) Cl] (Re) was
2+
3
(
32.5); ESI-MS (CH CN): m/z (%): 327.3 [MÀ2PF ] ; elemental analy-
[7]
3
6
prepared as described recently.
À1
sis calcd (%) for C H F N P Ru·1H O (961.72 gmol ): C 44.96, H
36
36 12
6
2
2
3
.98, N 8.74; found: C 44.80, H 3.96, N 8.79.
[
Re(CO) Cl(dmbÀdmb)]: The product was prepared by following
3
[13]
Syntheses
a slightly modified method from Wallendael et al. A solution of
[
Re(CO) Cl] (300 mg, 0.83 mmol, 1.0 equiv.) in MeOH (abs., 300 mL)
5
1
,2-bis[4-(4’-methyl-2,2’-bipyridyl)]ethane (dmbÀdmb): The product
was added dropwise to a refluxing solution of dmbÀdmb (1.06 g,
2.89 mmol, 3.4 equiv.) in toluene (abs., 850 mL) over 4 h. The solu-
tion was refluxed for another 4 h, cooled to RT, and the solvent
subsequently removed under reduced pressure. The residue was
extracted with acetonitrile (3”35 mL) and the solvent removed
under reduced pressure. The resulting solid was dissolved in
chloroform (25 mL) and the solution filtered and transferred to
a flask containing hexane (300 mL). The yellow precipitate was col-
lected by filtration and washed with diethyl ether (6”5 mL). After
was prepared by following a modified method from Elliott, Freitag,
and Blaney.
1
anes, 7.0 mL, 11.2 mmol, 1.0 equiv.) was added at À208C. After
cooling to À788C, a solution of dmb (2.05 g, 11.1 mmol, 1.0 equiv.)
in THF (40 mL) was added over 25 min. The solution was stirred for
[20]
To a solution of diisopropylamine (1.53 mL,
0.9 mmol, 1.0 equiv.) in THF (25 mL), n-butyllithium (1.6m in hex-
1
h at À788C and for another 30 min at À108C. After cooling to
À788C again, 1,2-dibromoethane (2.0 mL, 23.2 mmol, 2.1 equiv.)
was added. The suspension was warmed to RT and H O (50 mL)
2
drying under reduced pressure,
a yellow powder (353 mg,
was added. By addition of NaHCO , the pH was adjusted to 8 and
3
1
0
2
0
.53 mmol, 63% yield) was yielded. H NMR (300 MHz, CDCl3,
the mixture was extracted with diethyl ether (4”50 mL) and di-
chloromethane (3”50 mL). The organic phases were combined
and the solvent evaporated under reduced pressure to yield
a white powder. Recrystallization from ethyl acetate afforded color-
98 K): d=8.88 (dd, J=15.0, 5.6 Hz, 2H), 8.57 (ddd, J=20.6, 5.0,
.8 Hz, 2H), 8.30 (d, J=19.8 Hz, 2H), 7.96 (d, J=13.5 Hz, 2H), 7.33
ddd, J=7.2, 5.5, 1.3 Hz, 2H), 7.17 (ddd, J=5.0, 1.7, 0.8 Hz, 1H),
(
1
7.14 (dd, J=5.0, 1.8 Hz, 1H), 3.22–3.03 (m, 4H), 2.52 (s, 3H),
2
less crystals of dmbÀdmb (1.42 g, 3.87 mmol, 77% yield). H NMR
À1
.45 ppm (s, 3H); IR (CO): n˜ =2015, 1877 cm ; ESI-MS (CH CN): m/z
3
(
300 MHz, CDCl , 298 K): d=8.56 (t, J=5.3 Hz, 4H), 8.31 (s, 2H),
3
+
+
+
(
%): 636.8 [MÀCl] , 672.8 [M+H] , 694.8 [M+Na] ; elemental anal-
8
2
1
1
3
.23 (s, 2H), 7.13 (ddd, J=9.4, 5.4, 1.6 Hz, 4H), 3.09 (s, 4H),
À1
1
3
ysis calcd (%) for C27
H
22ClN
O
4
3
Re (672.15 gmol ): C 48.25, H 3.30,
.45 ppm (s, 6H); C NMR (75 MHz, CDCl , 298 K): d=156.52,
3
N 8.34; found: C 48.33, H 3.47, N 8.14.
55.91, 150.91, 149.32, 149.06, 148.29, 124.87, 123.94, 122.13,
+
21.23, 36.27, 21.31 ppm; ESI-MS (toluene): m/z (%): 367.0 [M+H] ,
[Re(CO) Cl(dmbÀdmb)Re(CO) Cl] (ReÀRe): [Re(CO) Cl] (335.8 mg,
3
3
5
+
89.0 [M+Na] .
0.93 mmol, 2.0 equiv.) and dmbÀdmb (170.1 mg, 0.46 mmol,
1
.0 equiv.) were suspended in toluene (abs., 10 mL) and refluxed
[
Ru(dmb) Cl ]: The product was prepared by following a modified
2 2
[12]
for 5 h. After cooling to RT, the yellow solid formed was filtered off
and washed with toluene (1”10 mL) and diethyl ether (2”20 mL).
Drying under reduced pressure afforded a yellow powder (447 mg,
0
8
2
4
method from Rau et al.
1
[Ru(cod)Cl2]n (152.0 mg, 0.54 mmol,
.0 equiv.), dmb (199.3 mg, 1.08 mmol, 2.0 equiv.), and LiCl
(
(
(
192.1 mg, 4.53 mmol, 8.4 equiv.) were suspended in dry DMF
30 mL) and heated in a closed vessel setup in a microwave reactor
1258C, 50 W, dynamic mode, 2 h). After cooling to 508C, the sol-
1
.46 mmol, 98% yield). H NMR (300 MHz, [D ]acetone, 298 K): d=
6
.98 (d, J=5.8 Hz, 2H), 8.91 (d, J=5.6 Hz, 2H), 8.63 (s, 2H), 8.46 (s,
H), 7.72 (dd, J=5.8, 1.7 Hz, 2H), 7.60 (d, J=5.8 Hz, 2H), 3.44 (s,
vent was removed under reduced pressure. The residual solid was
suspended in H O (500 mL), filtered over a fine frit, and washed
with a mixture of H O and acetonitrile (9:1, 100 mL). The crude
product was eluted from the frit with chloroform (500 mL) and the
solvent removed under reduced pressure. Recrystallization from
À1
H), 2.59 ppm (s, 6H); IR (CO): n˜ =2022, 1920, 1884 cm ; UV/Vis
2
3
À1
À1
(
(
DMF): lmax (e 10 m cm )=292 (67.5), 367 nm (16.4); ESI-MS
2
+
CH CN): m/z (%): 943.4 [MÀCl] ; elemental analysis calcd (%) for
3
À1
C H Cl N O Re (977.84 gmol ): C 36.85, H 2.27, N 5.73; found: C
30
22
2
4
6
2
3
6.57, H 2.30, N 5.53.
acetonitrile (5 mL) afforded a purple to black powder (112 mg,
2
1
0.7 mmol, 38% yield). H NMR (300 MHz, CD CN, 298 K): d=9.83
[Re(CO) Cl(dmbÀdmb)Ru(dmb) ](PF )
(ReÀRu):
[Ru(dmb) Cl ]
2 2
3
3
2
6 2
(
d, J=5.8 Hz, 2H), 8.21 (s, 2H), 8.06 (s, 2H), 7.48 (d, J=5.6 Hz, 2H),
(56.3 mg, 0.10 mmol, 1.0 equiv.) and [Re(CO) Cl(dmbÀdmb)]
3
7
2
5
5
.40 (d, J=6.0 Hz, 2H), 6.82 (d, J=5.1 Hz, 2H), 2.62 (s, 6H),
(70.0 mg, 0.10 mmol, 1.0 equiv.) were dissolved in MeOH (abs.,
3
À1
À1
.40 ppm (s, 6H); UV/Vis (DMF): lmax (e 10 m cm )=381 (4.4),
15 mL) and heated to 608C for 40 h. After cooling to 08C, aqueous
NH PF6 (0.15m, 18.5 mL, 2.8 mmol, 28 equiv.) was added and an
orange to red precipitate formed, which was collected by centrifu-
+
+
67 nm (4.1); ESI-MS (CH CN): m/z (%): 505.2 [MÀCl] , 540.1 [M] ,
3
4
+
63.1 [M+Na] .
gation. The precipitate was resuspended in a mixture of H O and
EtOH (25:1, 25 mL), centrifuged, and collected again. This proce-
dure was repeated 3 times. Then, the residue was dissolved in ace-
2
[
Ru(dmb) ](PF ) (Ru): The product was prepared by following
3 6 2
2
+
a method similar to that of the synthesis of [Ru(RÀbpy) (LÀL)]
2
[12]
derivatives described by Rau et al.
[Ru(cod)Cl ] (50.0 mg,
2 n
tonitrile (2 mL), precipitated in H O (40 mL), and collected by cen-
2
0
.18 mmol, 1.0 equiv.) and dmb (98.64 mg, 0.54 mmol, 3.0 equiv.)
trifugation. This procedure was repeated 3 times. After this, the
precipitate was dissolved in acetonitrile (2 mL) and precipitated in
diethyl ether (40 mL). Decantation and subsequent drying under
reduced pressure afforded an orange to red powder (54.8 mg,
were suspended in dry DMF (30 mL) and heated in a closed vessel
setup in a microwave reactor (1258C, 50 W, dynamic mode, 2 h).
The reaction mixture was transferred to a flask, H O (30 mL) added,
and the solution stirred in an oil bath at 1008C overnight. After
2
ChemCatChem 2015, 7, 3562 – 3569
www.chemcatchem.org
3568
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
1
0
8
2
(
(
[
.04 mmol, 38% yield). H NMR (300 MHz, CD CN, 298 K): d=8.90–
3
Acknowledgements
.74 (m, 2H), 8.40–8.19 (m, 8H), 7.60–7.04 (m, 14H), 3.23 (s, 4H),
.61–2.41 ppm (m, 18H); IR (CO): n˜ =2017, 1884 cm ; UV/Vis
DMF): lmax (e 10 m cm )=289 (224.1), 461 nm (34.4); ESI-MS
À1
S.M. acknowledges the Beilstein Institut for a PhD scholarship
and sincerely thanks Teresa Meister for extensive proofreading
and fruitful discussions.
3
À1
À1
2
+
CH CN/H O): m/z (%): 418.3 [MÀ2PF [Re(CO) Cl]]
,
571.3
3
2
2
6
3
+
+
MÀ2PF ] , 1287.3 [MÀPF ] ; elemental analysis calcd (%) for
6
6
À1
C H ClF N O P ReRu·2H O (1467.65 gmol ): C 41.74, H 3.43, N
51
46
12
8
3
2
2
7
.63; found: C 41.62, H 3.30, N 7.54.
Keywords: bridging
ligands
·
covalent
linkages
·
[
[
Re(CO) Cl(dmbÀdmb)RuCl (dmbÀdmb)Re(CO) Cl] (ReÀRuCl ÀRe):
photocatalysis · rhenium · ruthenium
3
2
3
2
Ru(cod)Cl ] (41.7 mg, 0.15 mmol, 1.0 equiv.) and [Re(CO) Cl(dmbÀ
2
n
3
dmb)] (200.0 mg, 0.30 mmol, 2.0 equiv.) were suspended in dry
DMF (30 mL) and heated in a closed vessel setup in a microwave
reactor (1258C, 50 W, dynamic mode, 2 h). After cooling to 508C,
the solvent was removed under reduced pressure. The residue was
washed with diethyl ether (3”30 mL) and dried under reduced
pressure to yield a purple powder (223 mg, 0.15 mmol, 99% yield).
IR (CO): n˜ =2016, 1877 cm ; UV/Vis (CH CN): l (e 10 m cm )=
3
1
[1] a) D. M. Schultz, T. P. Yoon, Science 2014, 343, 6174; b) R. Reithmeier, C.
Bruckmeier, B. Rieger, Catalysts 2012, 2, 544–571.
[2] K. S. Joya, Y. F. Joya, K. Ocakoglu, R. van de Krol, Angew. Chem. Int. Ed.
2013, 52, 10426–10437; Angew. Chem. 2013, 125, 10618–10630.
[
[
3] A. J. Morris, G. J. Meyer, E. Fujita, Acc. Chem. Res. 2009, 42, 1983–1994.
4] a) M. Cokoja, C. Bruckmeier, B. Rieger, W. A. Herrmann, F. E. Kühn,
Angew. Chem. 2011, 123, 8662–8690; b) M. Aresta, A. Dibenedetto, A.
Angelini, Chem. Rev. 2014, 114, 1709–1742; c) M. Aresta, A. Dibenedet-
to, Dalton Trans. 2007, 2975–2992.
À1
3
À1
À1
3
max
+
71 (17.0), 567 nm (7.9); ESI-MS (CH CN): m/z (%): 1480.6 [MÀCl] ,
3
+
517.3 [M+H] ; elemental analysis calcd (%) for C H Cl N O Re Ru
1516.27 gmol ): C 42.77, H 2.92, N 7.39; found: C 42.56, H 3.14, N
.18.
[5] J. Hawecker, J.-M. Lehn, R. Ziessel, J. Chem. Soc. Chem. Commun. 1983,
5
4
44
4
8
6
2
À1
(
7
536–538.
[
6] a) K. Kalyanasundaram, J. Chem. Soc. Faraday Trans. 2 1986, 82, 2401–
415; b) J. Hawecker, J.-M. Lehn, R. Ziessel, Helv. Chim. Acta 1986, 69,
1990–2012.
[7] S. Meister, R. O. Reithmeier, M. Tschurl, U. Heiz, B. Rieger, ChemCatChem
015, 7, 690–697.
2
[
(
Re(CO) Cl(dmbÀdmb)Ru(dmb)(dmbÀdmb)Re(CO) Cl](PF )
3
3
6 2
ReÀRuÀRe): ReÀRuCl ÀRe (160 mg, 0.11 mmol, 1.0 equiv.) and
2
2
dmb (194 mg, 1.06 mmol, 10.0 equiv.) were dissolved in a mixture
of MeOH (abs., 60 mL) and DMF (abs., 60 mL). The solution was
heated to 658C under exclusion of light and air for 3 d. After evap-
oration of the solvent under reduced pressure, the resulting
orange solid was extracted with acetonitrile (1”5 mL, 2”2 mL).
The solution was passed through a syringe filter and dropped into
a flask containing toluene (200 mL). Cooling the solution to 08C
for 1 h afforded an orange precipitate, which was collected by fil-
tration over a fine frit, washed with toluene (3”5 mL), and rinsed
from the frit with acetonitrile. The solvent was evaporated under
reduced pressure and the orange solid (153 mg, 0.09 mmol) dis-
[
[
8] H. Hukkanen, T. T. Pakkanen, Inorg. Chim. Acta 1986, 114, L43–L45.
9] B. Gholamkhass, H. Mametsuka, K. Koike, T. Tanabe, M. Furue, O. Ishitani,
Inorg. Chem. 2005, 44, 2326–2336.
[
10] a) K. Koike, S. Naito, S. Sato, Y. Tamaki, O. Ishitani, J. Photochem. Photo-
biol. A 2009, 207, 109–114; b) S. Sato, K. Koike, H. Inoue, O. Ishitani,
Photochem. Photobiol. Sci. 2007, 6, 454–461; c) Z.-Y. Bian, S.-M. Chi, L. Li,
W. Fu, Dalton Trans. 2010, 39, 7884–7887.
[11] C. Bruckmeier, M. W. Lehenmeier, R. Reithmeier, B. Rieger, J. Herranz, C.
Kavakli, Dalton Trans. 2012, 41, 5026–5037.
[12] S. Rau, B. Schäfer, A. Grüßing, S. Schebesta, K. Lamm, J. Vieth, H. Gçrls,
D. Walther, M. Rudolph, U. W. Grummt, E. Birkner, Inorg. Chim. Acta
2004, 357, 4496–4503.
solved in EtOH (abs., 90 mL). A solution of NH PF6 (300 mg,
1
tions were filtered and subsequently mixed together. After cooling
to 08C for 1 h, the precipitate was collected by filtration over
a fine frit and eluted from the frit with acetonitrile (100 mL). H O
(
plex, and subsequently acetonitrile was added until a clear orange
solution remained. The solvent was evaporated under reduced
pressure until a cloudy orange precipitate was forming and the
rest of the solvent lost most of its color. The precipitate was col-
lected by centrifugation, decantation, and subsequent dilution in
acetonitrile to recover the complex from the centrifugation tube.
Evaporation of the solvent under reduced pressure and drying
4
[
[
13] S. Van Wallendael, R. J. Shaver, D. P. Rillema, B. J. Yoblinski, M. Stathis,
T. F. Guarr, Inorg. Chem. 1990, 29, 1761–1767.
.84 mmol, 17.4 equiv.) in EtOH (50 mL) was prepared. Both solu-
14] a) E. M. Kober, T. J. Meyer, Inorg. Chem. 1982, 21, 3967–3977; b) B.
Durham, J. V. Caspar, J. K. Nagle, T. J. Meyer, J. Am. Chem. Soc. 1982,
104, 4803–4810; c) S. van Wallendael, D. P. Rillema, Coord. Chem. Rev.
1991, 111, 297–318; d) W. Kaim, S. Kohlmann, Inorg. Chem. 1990, 29,
2909–2914; e) G. D. Hager, G. A. Crosby, J. Am. Chem. Soc. 1975, 97,
2
120 mL) was added, which caused slight precipitation of the com-
7
031–7037.
15] a) Z.-Y. Bian, K. Sumi, M. Furue, S. Sato, K. Koike, O. Ishitani, Dalton Trans.
009, 983–993; b) V. Balzani, Tetrahedron 1992, 48, 10443–10514.
16] a) K. Kalyanasundaram, Coord. Chem. Rev. 1982, 46, 159–244; b) A. Juris,
V. Balzani, F. Barigelletto, S. Campagna, P. Belser, A. v. Zelewesky, Coord.
Chem. Rev. 1988, 84, 85–277; c) S. Ji, W. Wu, W. Wu, H. Guo, J. Zhao,
Angew. Chem. Int. Ed. 2011, 50, 1626–1629; Angew. Chem. 2011, 123,
[
[
2
under vacuum afforded an orange powder (110 mg, 0.06 mmol,
1
664–1667.
1
5
4
2
4% yield). H NMR (300 MHz, CD CN, 298 K): d=8.95–8.71 (m,
[17] C. Würth, M. Grabolle, J. Pauli, M. Spieles, U. Resch-Genger, Nat. Protoc.
2013, 8, 1535–1550.
[18] Z.-Y. Bian, K. Sumi, M. Furue, S. Sato, K. Koike, O. Ishitani, Inorg. Chem.
2008, 47, 10801–10803.
[19] G. R. Fulmer, A. J. M. Miller, N. H. Sherden, H. E. Gottlieb, A. Nudelman,
B. M. Stoltz, J. E. Bercaw, K. I. Goldberg, Organometallics 2010, 29, 2176–
3
H), 8.53–8.15 (m, 10H), 7.70–6.99 (m, 16H), 3.39–3.03 (m, 8H),
À1
.69–2.38 ppm (m, 18H); IR (CO): n˜ =2017, 1880 cm ; UV/Vis
3
À1
À1
(
DMF): lmax (e 10 m cm )=290 (232.2), 461 nm (32.5); ESI-MS
2
+
(
(
CH CN/H O): m/z (%): 815.4 [MÀ2PF ] ; elemental analysis calcd
3
2
6
À1
%) for C H Cl F N O P Re Ru·2H O (1955.56 gmol ): C 40.54, H
66 56 2 12 10 6 2 2 2
2
179.
20] C. M. Elliott, R. A. Freitag, D. D. Blaney, J. Am. Chem. Soc. 1985, 107,
647–4655.
3
.09, N 7.16; found: C 40.45, H 2.94, N 7.28.
[
4
Received: June 12, 2015
Published online on September 11, 2015
ChemCatChem 2015, 7, 3562 – 3569
www.chemcatchem.org
3569
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim