A Rhenium-Functionalized Metal–Organic Framework as a Single-Site Catalyst for Photochemical Reduction of Carbon Dioxide

Article abstract of DOI:10.1002/ejic.201600064

A metal–organic framework (MOF) based on (bpy)Re(CO)₃Cl-containing elongated dicarboxylate ligands and Zr₆(μ₃-O)₄(μ₃-OH)₄clusters was synthesized and used as an effective single-site catalyst to photochemically reduce carbon dioxide to carbon monoxide and formate and to provide mechanistic insights into the photocatalytic CO₂reduction process.

Full text of DOI:10.1002/ejic.201600064

DOI: 10.1002/ejic.201600064
Communication
Carbon Dioxide Reduction
A Rhenium-Functionalized Metal–Organic Framework as a
Single-Site Catalyst for Photochemical Reduction of Carbon
Dioxide
Ruiyun Huang,[a][‡] Yu Peng,[b,c][‡] Cheng Wang,*[a] Zhan Shi,[c] and Wenbin Lin*[a,b]
Abstract: A metal–organic framework (MOF) based on fective single-site catalyst to photochemically reduce carbon di-
(bpy)Re(CO)3Cl-containing elongated dicarboxylate ligands and oxide to carbon monoxide and formate and to provide mecha-
Zr6(μ3-O)4(μ3-OH)4 clusters was synthesized and used as an ef- nistic insights into the photocatalytic CO2 reduction process.
Introduction
photocatalytically active ligands. We envisioned that elongating
the Re-based ligands by adding two phenyl spacers to them
would open up the interior channels of the MOF material and
meet the steric requirements of the (bpy)Re(CO)3Cl moieties.
The stability of MOF materials is the limiting factor for many
of their applications. A number of Zr-based UiO MOFs have
been synthesized by using tunable dicarboxylate bridging li-
gands and Zr6(μ3-O)4(μ3-OH)4 cuboctahedral SBUs and are
among the most stable MOF materials due to their strong Zr–
O bonds.[9] Herein we report a highly stable and porous UiO-
As a new class of crystalline hybrid materials with high specific
surface area, tunable pore size, and pore walls that can be func-
tionalized, metal–organic frameworks (MOFs) have attracted
great attention for potential applications in gas storage,[1] non-
linear optics,[2] compound separation,[3] heterogeneous cataly-
sis,[4] drug delivery,[5] biomedical sensing,[6] and other areas.
MOFs can be functionalized through ligand doping, post-syn-
thetic modification, or using predesigned functional bridging
linkers.[7] Using the doping strategy, we successfully incorpo-
rated [Re(CO)3(5,5′-dcbpy)Cl] (at 4 % wt doping level; 5,5′-dcbpy
is 5,5′-dicarboxy-2,2′-bipyridine) into the UiO-67 MOF built from
Zr6(μ3-O)4(μ3-OH)4 secondary building units (SBUs) and 4,4′-bi-
phenyldicarboxylate (BPDC) ligands.[8] The resulting doped UiO-
67 MOF was found to be a good heterogeneous photocatalyst
for CO2 reduction to CO using visible light. However, the open
channels in this MOF are very small, which not only limits the
diffusion rates of reactants and products in the reaction, but
also sets an upper limit on the doping concentration of the
sterically demanding active (bpy)Re(CO)3Cl reaction centers.
Such drawbacks precluded a more detailed mechanistic study
of our first MOF-based molecular photocatalytic system and
prompted our interest in building a MOF structure with 100 %
type MOF with
a framework formula of {Zr6(O)4(OH)4-
[Re(CO)3Cl(bpydb)]6} (1) based on the elongated linear
(bpy)Re(CO)3Cl-containing dicarboxylate ligand (H2L) and
Zr6(μ3-O)4(μ3-OH)4 SBUs [Figure 1a, bpydb is 4,4′-(2,2′-bipyr-
idine-5,5′-diyl)dibenzoate].
[a] Collaborative Innovation Center of Chemistry for Energy Materials,
State Key Laboratory of Physical Chemistry of Solid Surfaces,
Department of Chemistry,
College of Chemistry and Chemical Engineering, Xiamen University,
Xiamen 361005, P. R. China
E-mail: wangchengxmu@xmu.edu.cn
http://121.192.177.81:90/wblin/INDEX.html
[b] Department of Chemistry, University of Chicago,
929 E. 57th Street, Chicago, IL 60637, USA
E-mail: wenbinlin@uchicago.edu
http://linlab.uchicago.edu/index.html
[c] State Key Laboratory of Inorganic Synthesis and Preparative Chemistry,
College of Chemistry, Jilin University,
Changchun 130012, China
Figure 1. Crystal structure of 1 (a) showing an octahedral cage and (b) show-
ing a tetrahedral cage with SBUs depicted as polygons. (c) Structure of 1 as
viewed along the [100] direction of the unit cell. (d) PXRD pattern of 1 (black)
and that simulated from the CIF file of 1 (red).
[‡] These authors have contributed equally.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/ejic.201600064.
Eur. J. Inorg. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Results and Discussion
Synthesis and Characterization
The Re-derived ligand H2L was synthesized by refluxing
[Re(CO)5Cl] with 4,4′-(2,2′-bipyridine-5,5′-diyl)dibenzoic acid
(H2bpydb) in methanol in a nitrogen atmosphere for 7 days.
Yellow crystals of 1 were obtained by heating H2L with ZrCl4
and TFA in DMF at 100 °C for 3 days (Scheme 1). Single-crystal
X-ray structure determination indicated that 1 features a three-
dimensional fcu topology as other UiO MOFs (Figure 1a–c).[10]
The phase purity of 1 was confirmed by the powder X-ray dif-
fraction (PXRD) patterns of 1 (Figure 1d). Inductively coupled
plasma mass spectrometry (ICP-MS) of the digested MOF sam-
ple gave a ratio of Zr/Re = 1.17, consistent with that in the
crystal structure within experimental error. Thermogravimetric
analysis (TGA) of 1 shows a weight loss of about 58 % from
20 °C to about 250 °C, corresponding to the removal of guest
solvent molecules in the MOF channel. The TGA trace then con-
tinues in a plateau up to 450 °C before the decomposition of
the framework (Figure S3).
Figure 2. (a) Plots of CO and H2 turnover numbers vs. the concentration of
water for 1; (b) CO and H2 turnover number vs. time for 1; (c) FTIR spectra of
H2L (red) and 1 before (blue) and after (black) catalysis; (d) Photo of the color
change for 1 before (left) and after (right) light irradiation (the yellow crystals
turned green).
actions between two Re molecules[14] in the catalytic cycle. The
presence of multiple catalytic pathways in one system makes it
difficult to attribute the catalytic selectivity and activity to any
single reaction pathway. The present CO2 reduction catalyzed
by 1 can only occur by the unimolecular mechanism as a result
of site-isolation of the catalyst immobilized in the MOF frame-
work. This study provides us a unique opportunity to examine
the activity and selectivity of single-site Re catalysts without
interference from bimolecular mechanisms.
The photocatalytic turnover numbers (TONs) of CO and H2
in the reactions catalyzed by 1 are listed in Table 1. The product
selectivity of CO vs. H2 of the reaction depends on the reaction
medium. CO2 reduction to CO was preferred over H2 production
when acetonitrile (MeCN) was used as the solvent, while proton
reduction to H2 was favored in tetrahydrofuran (THF) (see
Table 1 entry 1 vs. entry 2). As shown in Table 1, the CO TON
Scheme 1. Synthesis of MOF-1.
Photochemical Reduction of CO2
The series of Re(CO)3(bpy)X (X = halide) complexes are well of 1 in 6 hours (6.44) is much higher than that of the homoge-
known for their ability to selectively photocatalyze the reduc- neous counterpart (1.12). This enhanced TON is presumably due
tion of CO2 to CO.[11] The photocatalytic activity of 1 towards to the stabilization effect of active-site isolation of the immobi-
CO2 reduction was carefully evaluated in CO2-saturated organic lized single-site Re catalysts in the framework. The homogene-
solvents with triethylamine (TEA) as a sacrificial reducing agent ous control can not only undergo both the unimolecular and
and a 410 nm LED as the light source. The yellow crystals bimolecular catalysis but is also more prone to deactivation/
turned green after irradiation with LED light for 5 min, indicat- decomposition by bimolecular pathways. This stabilization ef-
ing rapid activation of the precatalyst to initiate catalytic turn- fect in thise framework is consistent with our previous observa-
over (Figure 2d). The amount of CO and H2 generated in the tion of the shorter Re-ligand-doped UiO-67 MOF.[8]
reaction was quantified by analyzing the headspace gas by GC
with CH4 as an internal standard. The amount of formate gener-
ated in the reaction was quantified by 1H NMR spectroscopy by
using a deuterated solvent in the reaction.
Control experiments in the absence of CO2 or LED light led
to no CO detection (Table 1, entries 5 and 6), confirming that
the CO detected in the catalytic runs did not come from the
chemical or photochemical decomposition of the Re(CO)3Cl
Previous mechanistic studies on the [Re(CO)3(bpy)Cl]-cata- centers. As expected, in the absence of sacrificial reducing
lyzed reduction of CO2 suggested either a unimolecular path- agent, only noncatalytic amounts of CO were generated
way involving a [ReI(bpy)(CO)3(COOH)] intermediate[12] or bimo- (Table 1, entry 4). Interestingly, we found that addition of water
lecular pathways involving
a
CO2-bridged Re dimer, to acetonitrile can significantly enhance the activity of 1. The
[(CO)3(bpy)ReI](CO2)[ReI(bpy)(CO)3],[13] or outer-sphere redox re- optimized solvent system for the maximum TON for CO is an
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Table 1. Investigation of MOF-1 for the photochemical reduction of CO2.
process. This prompted us to examine the decomposition proc-
ess of the (bpy)Re(CO)3Cl catalysts by precluding bimolecular
pathways and isolating the solid catalysts after reactions. We
have previously established MOFs as powerful platforms to
study the details of the modification and deactivation of single-
site molecular catalysts for water oxidation.[16] After photocatal-
ysis, 1 was separated from the supernatant by centrifugation
and examined by PXRD. The resultant PXRD pattern indicates
only partial retention of the ordered framework structure, sug-
gesting significant distortion of the framework during the catal-
ysis (Figure S8), presumably as a result of the deactivation/de-
composition of the Re centers in the framework structure.
Entry
Photocatalyst
H2-TON[a]
TON for CO
1
2
3
4
5
6
MOF-1/MeCN
MOF-1/THF
Re(CO)3Cl(bpyde)
MOF-1/no TEA
MOF-1/no CO2
MOF-1/no light
0.40
4.15
0.18
0.003
0.73
0
6.44
0.32
1.12
0.19
0
0
[a] The TON was calculated from the results of GC analysis after irradiation
for 6 h.
We have further established the stability of the framework
structure during the catalysis by a combination of ICP-MS and
PXRD analyses. The framework is very stable when not too
much water is added to the reaction mixture. The active MOF
catalyst decomposes under the reaction conditions as a result
of decomposition of the Re complex instead of the framework
structure. When too much water is added to the reaction mix-
ture, however, the framework itself begins to decompose be-
cause of the presence of hydroxide from the reaction between
water and triethylamine.
acetonitrile solution with a water concentration of 0.83
M (Fig-
ure 2a). At a higher water concentration, the crystals appeared
to partially dissolve, leading to less stable homogeneous cata-
lysts with significantly lower TONs (Figure 2a).
Luminescence spectra of 1 were also taken to shed light on
the details of the photochemical process. An emission peak at
around 643 nm was observed, upon excitation at 465 nm, due
3
to a MLCT state. Addition of water to the MOF suspension in
anhydrous MeCN did not lead to a substantial change in inten-
sity in the luminescence spectra (Figure S5). On the other hand,
adding triethylamine to the suspension effectively quenched
the emission intensity resulting from a redox reaction between
triethylamine and the 3MLCT state of the Re ligand, which is
the first step of Re-catalyzed CO2 reduction. There was also a
blueshift in the emission spectra (Figure S5). This blueshift ac-
companied a color change in the MOF catalyst to green, similar
to that observed in the catalytic runs. Following the color
change, the species became very sensitive to oxygen and
quickly returned to its original yellow color upon contact with
air. Given this sensitivity to oxygen, we believe that this species
is the reduced ReI(bpydc–·)(CO)3(solvent) with a radical anion
on the bipyridine ring. Interestingly, the addition of triethyl-
amine with water quenches the emission intensity far more pro-
foundly than adding triethylamine alone to the anhydrous
acetonitrile suspension of the MOF. This is consistent with the
observed acceleration effect of water on the CO2 reduction ac-
tivity and is possibly related to proton-coupled electron transfer
(PCET) in the presence of water.[15]
ICP-MS analysis of the supernatant indicated that 6 % Zr and
27 % Re leached into the solution. This result was further con-
firmed by the detection of 91 % Zr and 69 % Re by ICP-MS in
the recovered MOF solids. At the same time, both the reaction
1
supernatant and the recovered solids were studied by H NMR
spectroscopy. Formate was detected in the supernatant with
CD3CN as solvent, characterized by a singlet peak at δ =
8.45 ppm, which was further confirmed by spiking the sample
with sodium formate. The amount of formate generated was
quantified with mesitylene as an internal standard, leading to a
–
formate TON of 0.25 and a CO/HCO2 molar ratio of 26.
The recovered MOF solid was dissolved both in 10 % NaOD/
D2O solution and D3PO4/[D6]DMSO solution. 1H NMR spectra in
these solvents showed the generation of multiple species with
complicated spectra. The different solvents used for NMR spec-
troscopic analysis gave information on different species from
digestion in line with the different solubilities of these species.
1
A singlet peak at 9.3–9.2 ppm in the H NMR spectrum of the
original Re ligand is associated with the proton adjacent to the
nitrogen atom on the pyridine ring. The chemical shift of this
proton is relatively sensitive to the coordination of pyridine.
Without coordination to Re, this chemical shift moves to higher
field below 9.0 ppm. In the 1H NMR spectrum in D3PO4/
[D6]DMSO, peaks at 9.3–9.2 ppm were still observed, suggesting
the preservation of Re coordination in some of the molecules
(Figure S9). This peak is split into at least three peaks, indicating
generation of different Re species. On the other hand, in the 1H
NMR spectrum in NaOD/D2O, peaks above 9.0 ppm have very
small integration areas. Instead, peaks in the δ = 6.0–4.5 ppm
range indicate partial hydrogenation of aromatic rings in some
of the molecules (Figure S10).
The time-dependent CO2 reduction experiment was per-
formed by analyzing the CO production at different points in
time by GC. The TONs for CO quickly reached 5.7 in 1 hour and
steadily increased to 6.4 in 6 hours, indicating a fast turnover
at the initial stage of the reaction followed by catalyst deactiva-
tion. The photocatalytic reduction is highly selective toward
CO2 to generate CO and H2 with a ratio of approximately 15.
The time-dependent CO2 reduction of the homogeneous sys-
tem shows a very similar trend to that of the MOF (Figure S6).
React-IR was used to monitor the photocatalytic CO2 reduc-
tion in situ. An increased absorption signal for CO together with
–
the decreased absorption for CO2 and HCO3 confirms that the
CO comes from the photoreduction of CO2 and related species
(Figure S7).
FTIR spectra showed changes of the vibrational frequencies
The immobilization of the (bpy)Re(CO)3Cl centers in the UiO of the carbonyl groups on the Re cores before and after cataly-
MOF enhanced the TON for CO by almost six times, but the Re sis. There are three CO stretching peaks at 2019, 1910, and
catalyst still decomposed following a fast catalytic activation 1881 cm–1 for the pristine MOF, corresponding to the A′(1), A′
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
(2), and A′′ vibrational modes, respectively (Figure S11).[17] The
A′(2) mode showed up as a shoulder peak. After catalysis, the
peak of the A′(2) mode became more prominent, and the peak
associated with the A′′ mode slightly shifted to 1886 cm–1 (Fig-
ure 2c). The presence of only three CO stretching peaks after
catalysis indicates that the remaining Re centers are still coordi-
nated with three carbonyl groups and very likely still retain a
fac-coordination geometry. The chloride might be replaced by
other coordinating ligands.
Combining the information from the 1H NMR and FTIR spec-
troscopic studies, we envisioned a possible decomposition
pathway of the Re catalyst: partial hydrogenation of the pyr-
idine ring followed by Re decomplexation from the ligand. Such
a proposal can explain a number of our experimental observa-
tions.
Re(CO)3Cl(bpydb) (H2L): Re(CO)5Cl (127.8 mg, 0.354 mmol) and
bpydb (140 mg, 0.354 mmol) were added to methanol (75 mL) and
degassed for 30 min in a nitrogen atmosphere. The mixture was
heated to reflux for 7 d, and the solid product was collected by
filtration. Methanol was removed under vacuum to afford the pure
1
product. Yield: 210 mg (86 %). H NMR (400 MHz, CDCl3): δ = 9.26
(s, 2 H), 9.01 (d, 2 H), 8.77 (d, 2 H),8.17 (d, 4 H), 8.11 (d, 4 H) ppm. The
single-crystal X-ray structure of the compound was also determined
(Figure S2 and Table S2)[18]
MOF-1: ZrCl4 (1 mg, 4.29 μmol) and Re(CO)3Cl(bpydb) (3 mg,
4.29 μmol) were added to a mixture of DMF (300 μL) and TFA (6 μL),
sealed in a 2 mL glass vial, and heated at 100 °C for 72 h. After
cooling to room temperature, yellow crystals were obtained by cen-
trifugation. They were washed with DMF and methanol. Yield:
2.9 mg (46.6 %).
CCDC 1449520 (for MOF-1) and 1449521 (for H2L) contain the sup-
plementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre
One remaining question about this single-site CO2 reduction
catalyst is the origin of the second electron. Two equivalents of
electrons are required to reduce CO2 to CO, but the redox reac-
tion of the 3MLCT state of the Re catalyst is a one-electron
process. In the homogeneous system, the second electron can
come from another Re molecule through inner-sphere or outer-
sphere electron transfer. In the MOF structure with isolated Re
ligands, however, electron transfer from adjacent Re centers is
expected to be slow (although it is still a possible pathway).
Alternatively, the decomposition product of triethylamine (TEA)
can provide the second electron. After being photooxidized by
Acknowledgments
We thank the National Natural Science Foundation of the P.R.
China (21471126), the National Thousand Talents Program of
the P.R. China, the 985 Program of Chemistry and Chemical En-
gineering Disciplines of Xiamen University, and the US National
Science Foundation (NSF) (DMR-1308229) for funding support.
Y. P. acknowledges support from the China Scholarship Council.
We thank Mr. Alexander E. Hess for experimental help.
3
the MLCT state of the Re catalyst, the generated TEA·+ radical
cation can quickly lose a proton to give an enamine radical with
very high reducing power. This species can give the second
electron to the Re–CO2 adduct intermediates to complete the
cycle. The resulting enamine is then hydrolyzed to give acet-
aldehyde and diethylamine, both of which were detected by 1H
NMR spectroscopy in the supernatant of the reaction mixture
in CD3CN.
Keywords: Metal–organic frameworks · Heterogeneous
catalysis · Homogeneous catalysis · Photochemistry · Carbon
dioxide reduction
[1] a) J. Sculley, D. Yuan, H.-C. Zhou, Energy Environ. Sci. 2011, 4, 2721; b)
L. J. Murray, M. Dinca, J. R. Long, Chem. Soc. Rev. 2009, 38, 1294; c) K.
Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm,
T. H. Bae, J. R. Long, Chem. Rev. 2012, 112, 724.
[2] C. Wang, T. Zhang, W. Lin, Chem. Rev. 2012, 112, 1084–1104.
[3] a) E. D. Bloch, W. L. Queen, R. Krishna, J. M. Zadrozny, C. M. Brown, J. R.
Long, Science 2012, 335, 1606; b) Y. Peng, Y. Li, Y. Ban, H. Jin, W. Jiao, X.
Liu, W. Yang, Science 2014, 346, 1356.
[4] a) K. Manna, T. Zhang, F. X. Greene, W. Lin, J. Am. Chem. Soc. 2015, 137,
2665; b) M. Zhao, S. Ou, C.-D. Wu, Acc. Chem. Res. 2014, 47, 1199; c) J.
Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen, J. T. Hupp, Chem.
Soc. Rev. 2009, 38, 1450; d) K. Mo, Y. Yang, Y. Cui, J. Am. Chem. Soc. 2014,
136, 1746.
[5] a) C. He, D. Liu, W. Lin, Chem. Rev. 2015, 115, 11079; b) K. Lu, C. He, W.
Lin, J. Am. Chem. Soc. 2015, 137, 7600.
[6] a) C. He, K. Lu, W. Lin, J. Am. Chem. Soc. 2014, 136, 12253; b) D. Liu, K.
Lu, C. Poon, W. Lin, Inorg. Chem. 2014, 53, 1916.
[7] a) S. M. Cohen, Chem. Rev. 2012, 112, 970; b) F. Song, C. Wang, J. M.
Falkowski, L. Ma, W. Lin, J. Am. Chem. Soc. 2010, 132, 15390; c) D. Sun,
Y. Gao, J. Fu, X. Zeng, Z. Chen, Z. Li, Chem. Commun. 2015, 51, 2645.
[8] C. Wang, Z. Xie, K. E. deKrafft, W. Lin, J. Am. Chem. Soc. 2011, 133, 13445.
[9] a) D. Feng, Z.-Y. Gu, J.-R. Li, H.-L. Jiang, Z. Wei, H.-C. Zhou, Angew. Chem.
Int. Ed. 2012, 51, 10307; Angew. Chem. 2012, 124, 10453; b) J. H. Cavka,
S. Jakobsen, U. Olsbye, N. Guillou, C. Lamberti, S. Bordiga, K. P. Lillerud,
J. Am. Chem. Soc. 2008, 130, 13850.
Conclusions
We have synthesized a new porous Re-functionalized UiO MOF
of fcu topology that serves as a highly effective single-site pho-
tocatalyst for the reduction of CO2 to CO and formate. The Re-
MOF showed higher TONs for CO than the corresponding ho-
mogeneous catalyst, but still underwent decomposition due to
partial hydrogenation of the bpy ligand and subsequent Re de-
complexation. The incorporation of the Re catalyst into a stable
MOF framework affords isolated catalytic sites, which enables
detailed studies of the catalytic selectivity and mechanism of
the unimolecular catalytic pathway, without interference from
parallel bimolecular catalysis in the homogeneous system. This
work highlights exciting opportunities in using MOFs as a plat-
form to study the mechanisms and deactivation/decomposition
pathways of molecular catalysts.
[10] CCDC 1449520; space group F23; a = 38.7410(2) Å; Z = 8.
[11] a) J. Hawecker, J.-M. Lehn, R. Ziessel, J. Chem. Soc., Chem. Commun. 1983,
536; b) A. J. Morris, G. J. Meyer, E. Fujita, Acc. Chem. Res. 2009, 42, 1983.
[12] D. H. Gibson, X. Yin, H. He, M. S. Mashuta, Organometallics 2003, 22, 337.
Experimental Section
4,4′-(2,2′-Bipyridine-5,5′-diyl)dibenzoic Acid (bpydb): Ligand
bpydb was synthesized as previously reported.[16]
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[13]
[14]
[17] D. M. Dattelbaum, K. M. Omberg, J. R. Schoonover, R. L. Martin, T. J.
Meyer, Inorg. Chem. 2002, 41, 6071.
[18] CCDC 1449521; space group Cc; a = 27.622(2) Å; b = 9.5198(7) Å; c =
12.8992 Å; ꢀ = 106.806(5)°, Z = 4.
a) D. H. Gibson, X. Yin, J. Am. Chem. Soc. 1998, 120, 11200; b) Y. Hayashi,
S. Kita, B. S. Brunschwig, E. Fujita, J. Am. Chem. Soc. 2003, 125, 11976.
H. Takeda, K. Koike, H. Inoue, O. Ishitani, J. Am. Chem. Soc. 2008, 130,
2023.
[15] O. S. Wenger, Coord. Chem. Rev. 2015, 282–283, 150.
[16] C. Wang, J.-L. Wang, W. Lin, J. Am. Chem. Soc. 2012, 134, 19895.
Received: January 24, 2016
Published Online: ■
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Carbon Dioxide Reduction
A
(bpy)Re(CO)3Cl-containing metal–
organic framework (MOF) acts as an ef-
fective single-site catalyst to photo-
chemically reduce carbon dioxide to
carbon monoxide and formate. Mecha-
nistic insights into the monomolecular
pathway of the photocatalytic CO2 re-
duction have been obtained without
interference from bimolecular path-
ways, thanks to site-isolation of the
catalytic centers in MOFs.
R. Huang, Y. Peng, C. Wang,*
Z. Shi, W. Lin* ...................................... 1–6
A
Rhenium-Functionalized Metal–
Organic Framework as a Single-Site
Catalyst for Photochemical Reduc-
tion of Carbon Dioxide
DOI: 10.1002/ejic.201600064
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim