Semiconductive Amine-Functionalized Co(II)-MOF for Visible-Light-Driven Hydrogen Evolution and CO2 Reduction

Article abstract of DOI:10.1021/acs.inorgchem.8b01265

A Co-MOF, [Co₃(HL)₂·4DMF·4H₂O] was simply synthesized through a one-pot solvothermal method. With the semiconductor nature, its band gap was determined to be 2.95 eV by the Kubelka-Munk method. It is the first trinuclear Co-MOF employed for photocatalytic hydrogen evolution and CO₂ reduction with cobalt-oxygen clusters as catalytic nodes. Hydrogen evolution experiments indicated the activity was related to the photosensitizer, TEOA, solvents, and size of catalyst. After optimization, the best activity of H₂ production was 1102 μmol/(g h) when catalyst was ground and then soaked in photosensitizer solution before photoreaction. To display the integrated design of Co-MOF, we used no additional photosensitizer and cocatalyst in the CO₂ reduction system. When -NH₂ was used for light absorption and a Co-O cluster was used as catalyst, Co-MOF exhibited an activity of 456.0 μmol/(g h). The photocatalytic mechanisms for hydrogen evolution and CO₂ reduction were also proposed.

Full text of DOI:10.1021/acs.inorgchem.8b01265

Article
Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX
pubs.acs.org/IC
Semiconductive Amine-Functionalized Co(II)-MOF for Visible-Light-
Driven Hydrogen Evolution and CO Reduction
2
†
†
†
†
†
†
Wei-Ming Liao, Jian-Hua Zhang, Zheng Wang, Yu-Lin Lu, Shao-Yun Yin, Hai-Ping Wang,
†
,†
†,‡
Ya-Nan Fan, Mei Pan,* and Cheng-Yong Su
†MOE Laboratory of Bioinorganic and Synthetic Chemistry, Lehn Institute of Functional Materials, School of Chemistry, Sun
Yat-Sen University, Guangzhou 510275, China
‡
State Key Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences,
Shanghai 200032, China
*
S Supporting Information
ABSTRACT: A Co-MOF, [Co (HL) ·4DMF·4H O] was simply
3
2
2
synthesized through a one-pot solvothermal method. With the
semiconductor nature, its band gap was determined to be 2.95 eV
by the Kubelka−Munk method. It is the first trinuclear Co-MOF
employed for photocatalytic hydrogen evolution and CO
2
reduction with cobalt−oxygen clusters as catalytic nodes. Hydrogen
evolution experiments indicated the activity was related to the
photosensitizer, TEOA, solvents, and size of catalyst. After
optimization, the best activity of H production was 1102 μmol/
2
(
g h) when catalyst was ground and then soaked in photosensitizer
solution before photoreaction. To display the integrated design of
Co-MOF, we used no additional photosensitizer and cocatalyst in
the CO2 reduction system. When −NH2 was used for light
absorption and a Co−O cluster was used as catalyst, Co-MOF exhibited an activity of 456.0 μmol/(g h). The photocatalytic
mechanisms for hydrogen evolution and CO reduction were also proposed.
2
7
INTRODUCTION
group reported a Cu I -based cluster MOF, Cu-I-bpy (bpy=
4
(
2
2
■
,4′-bipyridine), showing highly efficient activity (7.09 mmol/
g h)) for hydrogen evolution without cocatalyst and
photosensitizer.
The research on photocatalytic CO reduction has been
concerned because of its useful chemical conversion of CO2
into chemicals and fuels, otherwise, slowing down the global
warming effect. MOFs have been employed for photocatalytic
CO reduction for its accessible voids and stability, and the
related reports were summarized. Among them, only several
clusters-based MOF catalyst were explored. Li group
synthesized an amine-functionalized NH -MIL-125(Ti)
which could produce HCOO from CO photoreduction in
MeCN/TEOA (5/1) solution under visible-light irradiation.
Later, they also investigated the performance of NH -Uio-
6(Zr), NH -MIL-101(Fe), NH -MIL-53(Fe), and NH -MIL-
8B(Fe) catalysts for CO reduction, in which Fe-based MOFs
showed a dual catalytic pathways based on direct excitation of
The concerns on global warming and the energy crisis have
prompted researchers to develop sustainable and clean energy
sources. Visible-light-driven catalytic studies on hydrogen or
oxygen evolution, CO reduction, and organic synthesis have
2
2
1
been explored in many kinds of materials, such as metal oxide
2
and metal−organic frameworks (MOFs). In general, MOFs
often act as supporter, photosensitizer (PS) unit, catalyst unit,
or both photosensitizer and catalysis units in photocatalytic
2
3
2a,8
reaction. Among these, some cluster-based MOFs as node
9
catalysts for hydrogen evolution have been explored for their
4
efficient activity and stability in catalysis. In 2009, Mori
2
5
−
group reported the first metal−oxygen cluster MOF, [Ru (p-
2
2
BDC) ] (p-BDC = 1,4-benzenedicarboxylate), for photo-
2
n
catalytic water reduction to produce hydrogen molecules in the
2
presence of Ru(bpy)32 (bpy = 2,2′-bipyridine), MV (N,N′-
dimethyl-4,4′-bipyridinium), and EDTA-2Na. The above
Ru(II)-MOF catalyst was based on noble metal, whereas
several noble-metal-free cluster MOFs were also investigated.
To mimic the [NiFe]-hydrogenase, the Du group synthesized a
+
2+
6
8
2 2 2
2
9c
Fe−O clusters and exciting −NH functionality, respectively.
These examples emphasize the potential for CO photo-
reduction by MOFs including a −NH functional PS unit and
metal-cluster catalytic unit.
2
2
2
2
D [Ni S]-based MOF, [Ni (PymS) ] (PymSH = pyrimidine-
2 2 4 n
6
2
-thiol), exhibiting high stability in aqueous conditions. They
also prepared a 2D double-layered MOF, [Ni(Hmna)(mna)-
H O) ]·(H O) (H mna = 2-mercaptonicotinic acid) for H
2
(
2
10
2
11
2
production with a TON of 685.4 with fluorescein as
photosensitizer after 48-h photoreduction. Recently, Du’s
Received: May 9, 2018
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.8b01265
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. Crystal structure of Co-MOF (LIFM-45): (a) {Co O } cluster center; (b) coordination environment of ligand; and (c) 3D packing
3
14
mode along a axis. Color indication: Co, green; O, pink; N, blue; C, gray. All hydrogen atoms were omitted for clarity.
Noble-metal free Co(II) complexes were applied to
flow with a heating rate of 5 °C/min. PXRD data were recorded on a
Rigaku SmartLab X-ray diffractometer with Cu−Kα radiation (λ =
.54056 Å) at room temperature. N adsorption isotherms (77 K) of
photocatalytic hydrogen evolution and CO reduction for
2
10
1
their high photoactivity in homogeneous solution. But few
works concerning Co(II)-based MOF heterogeneous catalyst
2
Co-MOF were carried out on a Quantachrome Autosorb-iQ2-MP
instrument. SC-XRD data were collected on a Rigaku Oxford
SuperNova X-RAY diffractometer system equipped with a Cu sealed
tube (λ = 1.54178 Å) at 50 kV and 0.80 mA at 293 K. With the twin
nature of obtained crystal, TWIN and BASF command was used to
refine the crystal data. As solvents (DMF and water) were highly
disordered in the framework, SQUEEZE process was employed by
has been developed for CO reduction, and PS was included in
2
1
1
these systems. Herein, we reported the first trinuclear Co-
MOF, formulated as [Co (HL) ·4DMF·4H O] (H L= 2′-
3
2
2
4
amino-[1,1′:4′,1″-terphenyl]-3,3″,5,5″- tetracarboxylic acid)
for both photocatalytic hydrogen evolution and CO
2
12
reduction. Various experiment conditions were employed to
achieve the optimized results, and it reaches 1102 μmol/(g h)
PLATON. The electron count of solvent was then determined to be
200 in each unit cell, which indicated of four DMF and four water
molecules. Scanning electron microscope (SEM) measurements were
performed on Hitachi SU8010 scanning electron microscope.
in DMF/H O (4/2) solution with TEOA as sacrificial agent
2
for H2 production. In order to perform the integrated
functionality of frameworks, CO2 reduction experiments
without additional PS and cocatalyst were performed and it
obtained a considerable selectively catalytic activity of 456.0
Synthesis of [Co (HL) ·4DMF·4H O] (Co-MOF, LIFM-45). The
3
2
2
ligand 2′-amino-[1,1′:4′,1″-terphenyl]-3,3″,5,5″- tetracarboxylic acid
1
3
(
H L) was prepared according to the previous references and
4
1
characterized by H NMR (400 MHz, DMSO-d , 25 °C, δ): 13.34 (s,
−
6
μmol/(g h) for HCOO . The photocatalytic mechanisms
4
2
H), 8.45 (d, 2H), 8.41 (s, 2H), 8.24 (s, 2H), 7.24 (s, 1H), 7.20 (d,
H), 7.06 (d, 2H), 5.25 (s, 2H). Then, 0.0105 g (0.025 mmol) of H4L
based on the redox of trinuclear cobalt cluster were also
explored.
and 0.0146 g (0.05 mmol) of Co(NO ) ·6H O was added into a 20
3
2
2
mL glass bottle equipped with cap. A mixture solution of 3 mL of
EXPERIMENTAL SECTION
Chemicals and Materials. Ru(bpy) Cl was purchased from
HWRK Chem. 3,5-bis(methoxycarbonyl) phenylboronic acid pinacol
ester was purchased from Shanghai UCHEM Inc. Cesium carbonate,
Tetrakis (triphenylphosphine) palladium and Co(NO ) ·6H O were
DMF/H O (v/v, 1/1) was then injected. The bottle was sealed and
■
2
sonicated for 2 min followed by a 24 h heating at 358 K. After the
solvothermal reaction, the bottle was cooled naturally. The bulk
3
2
crystals were separated and washed with 10 mL of DMF/H O (v/v,
2
1
/1) solvents for three times and dried under vacuum at room
3
2
2
−
1
temperature. (Yield: 52%, based on H L) FT-IR (KBr, cm ): ν =
purchased from Alfa Aesar. Triethanolamine (TEOA) was purchased
from Sigma-Aldrich. All chemicals and solvents purchased from the
commercial sources were of AR grade and used without further
purification.
4
3392 (m), 3139 (m), 1708 (m), 1656 (s), 1621 (s), 1575 (s), 1450
(m), 1400 (s), 1374 (s), 1249 (w), 1103 (w), 779(m), 746(m), 725
(m) (Figure S1). Found: C, 48.51%; H, 4.18%; N, 5.87%. Calcd: C,
Characterization and Methods. Fourier transform infrared (FT-
IR) spectra were performed on a Nicolet/Nexus-670 spectrometer in
the wavenumber range between 4000 and 400 cm− using the KBr
disk. Elemental analysis (EA) measurement was measured on a Vario
48.81%; H, 4.39%; N, 6.10% indicating formula for C44
4(C NO)·4H
shown in Table S1 (CCDC:1834360).
Evaluation of Photocatalytic Hydrogen Evolution and CO
H
24
N
2
O
16
·
H
7
O. Crystal data of Co-MOF was collected and is
3
2
1
2
1
EL CHNS elemental analyzer. The H NMR spectra were recorded
Reduction. For H
Ru(bpy) Cl as photosensitizer (PS), 4 mg of Co-MOF as catalyst, 1
mL of TEOA as sacrificial agent, and mixed solvents were added into
a 55 mL glass bottle. The bottle was then sealed and degassed with N
2
evolution in a typical experiment, 4 mg of
on a Bruker Avance 400 NMR spectrometer, applying TMS as the
internal standard and using MestReNova software to analyze the
spectra. The UV−vis absorption spectra were measured on a
SHIMADZU UV-3600. Photoluminescence spectra were measured
on EDINBURGH FLS980 fluorescence spectrophotometer. TG curve
was carried out on a STA 449 F3 Jupiter instrument under a nitrogen
3
2
2
gas for 20 min. The visible-light-driven hydrogen evolution was
2
performed on a white LED lamp (λ = 420−780 nm, 100 mW/cm ).
For CO reduction, 4 mg of Co-MOF as photocatalyst, 5 mL of
2
B
DOI: 10.1021/acs.inorgchem.8b01265
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 2. Photoelectronic properties of Co-MOF: (a) UV−vis absorption spectra of H L and Co-MOF in solid state. (b) Cyclic voltammogram of
4
Co-MOF. Measured conditions: Ag/AgCl was used as the reference electrode, glassy carbon was used as the working electrode, and a platinum
1
/2
plate was used as counter electrode. Solvent, DMF; electrolyte, Bu NPF (0.1 M); scan rate, 0.1 V/s. (c) (αhν) versus hv curve for Co-MOF.
4
6
The band gap is calculated by the Kubelka−Munk method. (d) Mott−Schottky plots for Co-MOF. The frequency is in the range of 500−1500 Hz.
2
+
MeCN or N,N-dimethylacetamide (DMAc) as solvent, and sacrificial
agent were added into a 55 mL glass bottle. The bottle was then
sealed and bubbled with CO for 20 min. The visible-light-driven CO
asymmetric unit of Co-MOF consists of one and a half Co
3−
ions, one HL ligand, two uncoordinated DMF molecules and
two water molecules. In the coordination environment of metal
center, there are two kinds of Co ions (Figure 1a). Co1 is
2
2
reduction experiment was performed under a white LED lamp (λ =
2+
2
4
00−780 nm, 100 mW/cm ) or 300 W Xe lamp (λ > 400 nm). The
hexa-coordinated by oxygen atoms coming from six different
photocatalytic gas products were detected on a gas chromatograph
3
−
HL ligands, showing octahedral geometry. While Co2 is
penta-coordinated by five oxygen atoms, among which three
are from monodentate coordination of carboxylate groups and
the other two are from chelated coordination of carboxylate
group, respectively. Each HL ligand acts as a μ -bridge to
(
(
GC9790, Fuli Analytical Instrument Co., Ltd.) equipped by TCD
H2 gas) and FID (CO gas). The liquid product HCOO was
−
determined by using an ion chromatograph (882 Compact IC pro,
Metrosep) equipped with Metrosep A supp 5 250/4.0 column
according to reported works.
9c,14
3−
6
Electrochemical Measurement. Electrochemical measurement
was finished on a CHI 660e electrochemical workstation. Cyclic
voltammetry (CV) measurement for Co-MOF were performed with a
three-electrode cell, in which glassy carbon as working electrode, Pt as
auxiliary electrode, and Ag/AgCl as reference electrode; 0.1 M
tetrabutylammonium hexafluorophosphate (Bu NPF ) was used as
2+
link six Co ions, in which two carboxylate groups adopt μ -
2
1
1
2
η :η , the other one adopts μ -η and the fourth one adopts μ -
1
1
1
η -bridging mode, respectively (Figure 1b). In general, in the
Co-MOF crystal structure, {Co O } coordinated cluster-
centers are formed. Through the linkage between Co ions
3
14
2
+
4
6
3−
electrolyte in dry and degassed DMF solvent during the experiment.
For ease of measurement, Co-MOF crystals were first ground for 2
min in a mortar. Co-MOF powder was achieved, and 3 mg was
dispersed in 0.5 mL of DMF solvent and 10 μL of Nafion. Ten of
microliters dispersion was injected onto the glassy carbon electrode
and dried under infrared lamp for 30 min. Mott-schottky plots for Co-
MOF in different frequencies (500, 1000, and 1500 Hz, respectively)
were measured by Impedance-Potential technique. The amplitude was
set to be 10 mV.
and HL ligands based on these clusters, 3D-network porous
1
2
structure is constructed (Figure 1c). Calculated by PLATON
program, the total void volume is determined to be 1803 Å,
approximately 34% of the total crystal volume.
Elemental analysis (EA), thermogravimetric (TG) analysis,
powder X-ray diffraction (PXRD) and Fourier transform
infrared (FT-IR) spectrum were performed to characterize Co-
MOF. Knowing from TG curve, it shows that the first 5.2%
weight loss of Co-MOF happens before 100 °C, this should be
ascribed to the loss of water molecules (Figure S2a). And the
second weight loss (21.2%) starting from 150 °C should be
attributed to loss of DMF molecules. The third weight loss
happens above 430 °C resulted from the collapsion of
framework and decomposition of organic ligand. This result
is consistent with those of EA and crystal data. PXRD pattern
showed that the diffraction peaks of as synthesized Co-MOF
matches well with the simulated one in spite of difference in
peak intensity (Figure S2b). Scanning electron microscope
(SEM) measurement indicates the Co-MOF crystals are like
RESULTS AND DISCUSSION
Crystal Structure and Characterization. Co-MOF
■
(
LIFM-45), formulated as [Co (HL) ·4DMF·4H O], was
3
2
2
synthesized by the reaction of cobalt nitrate hexahydrate and
H L (2′-amino-[1,1′:4′,1″-terphenyl]-3,3″,5,5″- tetracarboxylic
4
acid) ligand under solvothermal condition at 358 K for 24 h.
Appropriate crystal was selected for single-crystal X-ray
diffraction (SCXRD), and crystal structure was achieved
from refinement (Table S1 and Figure 1). Co-MOF crystal
crystallizes in monoclinic system, space group C2/c. The
C
DOI: 10.1021/acs.inorgchem.8b01265
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
thick blades in morphology with several hundred micrometers
in length (Figure S2c). To explore the porous property of Co-
experiments were performed to optimize the photocatalytic
activity. And some results achieved in different solvents and
ratios were shown and compared (Figure 3). No hydrogen was
MOF, N adsorption−desorption isotherms were measured at
2
7
7 K (Figure S3a). According to the data of adsorption
isotherm of N2 for Co-MOF, it indicates of not simple
microporous adsorption. Pore size distribution was calculated
by DFT method (Figure S3c), which indicates that there are
both micropores (around 0.9 nm) and mesopores (between 3
and 5 nm) in Co-MOF samples. It shows that the adsorption
amount has not reached the platform in the N2 77 K
adsorption experiment. We also simulate N sorption under 77
2
15
K by Materials Studio software (Figure S4). The simulated
isotherm shows the trend of multiple steps, which indicates the
existence of different sizes of pores. In this N 77 K sorption
2
simulation, the adsorption amount reaches its maximum value
at p/p = 0.48, which is different from the experiment result.
0
However, the maximum adsorption uptake amount is basically
coincident between the calculated and experimental results.
And in general, both the experiment and simulation indicate of
microporous and mesoporous feature in Co-MOF, which
suggests the potential photocatalysis in Co-MOF. Brunauer−
Figure 3. Hydrogen evolution results for Co-MOF in different kinds
and ratios of solvents after different times. The volume ratio of DMAc
(
N,N-dimethylacetamide), isopropanol or DMSO to H O was 5/1.
2
2
Emmett−Teller (BET) area was determined to be 343 m /g.
The volume ratio of DMF or MeCN to H O was 4/2. The times
2
After immersed in mixed DMF/H O (4/2,v/v) solution for 12
2
below zero means the experiment was performed without light
h, the PXRD pattern is almost the same to that of as
synthesized, which indicates the crystallinity of Co-MOF is
irradiation.
remained (Figure S5). N adsorption and desorption isotherms
2
found when these experiments performed without light
irradiation. The hydrogen amount produced from these
experiments shows a similar changing trend which increases
quickly in the first hours but gradually slows down and
eventually reaches maximum. Totally in 12 h, the experiment
at 77 K were also measured. The highly similar adsorption
behavior comparing to that of as-synthesized indicates the
framework and porosity are remained (Figure S3b).
Optical and Eletrochemical Properties. The solid-state
UV−vis absorption spectra of H L ligand and Co-MOF were
4
performed in DMF/H O (v/v, 4/2) produced the most
2
measured and compared (Figure 2). To H L, the board
4
amount of hydrogen (10.0 μmol), while the one in DMSO/
absorption below 450 nm should be ascribed to the K and B
bands of aromatic groups. After coordination of cobalt and
H O (v/v, 5/1) produced the least amount (1.37 μmol). The
2
decreasing reaction ratio late period of experiment should be
ascribed to the decomposition of PS because it could be
activated by addition of supplementary PS (Figure S6).
Therefore, the first period experiment in 3 h might be rational
to compare the catalytic activity in different solvents (Table
S2).
H L ligand in Co-MOF, these bands exhibit a little red-shift.
4
2
+
Meanwhile, significant d−d transition absorption of Co
appears at around 560 nm. To investigate the potential
semiconductive property of Co-MOF, band gap energy (Eg)
was determined to be 2.95 eV by Tauc plot through Kubelka−
Munk method (Figure 2c). This low E value is quite
We found H O content in solvents would affect the activity
g
2
comparable to some other metal−oxygen cluster-based MOF
on producing hydrogen, for example in DMF/H O, when the
2
1
6
materials. Mott-schottky method was usually applied to
ratio varied from 5.9/0.1 (mL/mL) to 4/2, the activity
enhanced from 165 to 488 μmol/(g h) gradually, then reduced
to 127 μmol/(g h) at ratio of 3/3 (Table S2, entries S7−S10).
17
estimate the flat band position. Herein, the flat band position
of Co-MOF could be determined to be −1.54 V vs. Ag/AgCl
(
−1.34 V vs. NHE, Figure 2d). As it suggested that the bottom
Similar phenomenon happened in MeCN/H O and Isopro-
2
of conduction band (CB) in n-type semiconductors was 0.10 V
panol/H O solvents, and they achieved the best activity (68
2
18
more negative than flat band position, the CB of Co-MOF is
estimated to be −1.44 V vs. NHE. Therefore, valence band
and 394 μmol/(g h), respectively) at solvent ratio of 4/2 and
5/1, respectively (Table S2, entries S5 and S19). The amount
of solution also affects the photocatalytic activity. When we
doubled the reaction solution, activity was reduced to 45
μmol/(g h), and enhanced activity (366 μmol/(g h)) was
found when we halved the solution (Table S2, entries S11 and
S12). Furthermore, increasing the catalyst amount or
decreasing the PS amount would lead to a reduced catalytic
activity, while decreasing the catalyst could enhance the
activity (Table S2, entries S13 to S15). The influence of all
these factors on catalytic activity should be ascribed to the
effect on light absorption and electron transition during the
photocatalytic process.
(
VB) is calculated to be +1.51 V.
Visible-Light-Driven H Production. Visible-light-driven
2
hydrogen evolution experiments were performed in mixed
solvents at room temperature with Co-MOF as catalyst,
Ru(bpy) Cl as PS, and TEOA as sacrificial agent. When 4 mg
3
2
of PS, 4 mg of Co-MOF, and 1 mL of TEOA were added into
mL of MeCN/H O (5/1, v/v) solvents, hydrogen product
6
2
was detected under light irradiation and the activity was 190
μmol/(g h) in the first 3 h (Table S2, entry S1). Without PS or
light in the experiment, no any hydrogen product was
observed, which emphasized the photocatalytic nature (Table
S2, entries S2 and S3). And it should be noted that little
hydrogen was found without catalyst, which was ascribed to PS
To achieve better hydrogen amount, we also investigated the
influence of catalyst pretreatment on catalytic activity. For
example, if Co-MOF catalyst was ground for 3 min, the thick
blade solid transformed to granular morphology and the size
(
Table S2, entry S4). As solvents, PS, catalyst and sacrificial
agent were all essential to the hydrogen evolution, a series of
D
DOI: 10.1021/acs.inorgchem.8b01265
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
decreased from several hundred micrometers in length to
several micrometers in diameter (Figure S2c, d). After
grinding, PXRD pattern of the particles was consistent with
the original Co-MOF, which indicated the grinding operation
did not destroy the MOF structure (Figure S2b). The obtained
powder was taken for photocatalysis and achieved activity of
Table 2. Results of Visible-Light-Driven Photocatalytic CO2
Reduction for Co-MOF in 6 h
a
μmol/(g h)
entry
sacrificial agent
solvent
CO
H2
HCOO−
b
1
0.1 g of BIH
MeCN
MeCN
MeCN
MeCN
DMAc
DMAc
0
0
1.9
1.1
1.4
0
0
0
0
5.9
7.5
0
0
0
b
5
95 μmol/(g h) in 3 h with DMF/H O (5/1, mL/mL) as
2
1 mL of TEOA
0.1 g of BNAH
0.1 g of BNAH
0.1 g of BNAH
0.1 g of BNAH
2
b
solvents, and this value was much better than that without
grinding before reaction (Table 1, entries 1 and 2). Nearly
3
78.2
264.4
456.0
0
c
4
c
5
d
Table 1. Hydrogen Evolution Results with Different
6
a
b
Catalyst Pretreatment (ground for 3 min or soaked with the
Conditions: catalyst (Co-MOF, 4 mg), solvent (5 mL). Irradiated
c d
a
PS solution for 12 h) before Experiments
by LED lamp. Irradiated by 300 W Xe lamp. Without light
irradiation.
entry
pretreatment
grinding
soaking
grinding+soaking
grinding+soaking
DMF/H O (mL/mL) activity (μmol/(g h)
2
1
2
3
4
5
5/1
5/1
5/1
5/1
4/2
203
595
488
951
1102
should be noted that the kind of sacrificial agent played an
important role in catalytic system. When BIH (1,3-dimethyl-2-
phenyl-2,3-dihydro-1H-benzo[d]imidazole) or TEOA was
used as sacrificial agent, no gas or liquid product was observed.
Further, when the light source changed to 300 W Xe lamp, the
a
Conditions: 4 mg of PS, 4 mg of Co-MOF, and 1 mL of TEOA.
−
amount of HCOO increased up to 264.4 μmol/(g h),
whereas CO production decreased to 1.1 μmol/(g h) (Table 2,
entry 4). Comparatively, some hydrogen gas was also found.
The solvent affected the catalytic activity as well, because the
products exhibited a large enhancement when DMAc was used−
as solvent (Table 2, entry 5). The catalytic activity for HCOO
twice the advantage should be ascribed to the enhanced
amount of accessible cobalt center on the solid surface. The
exposed metal sites promote the direct contact of the catalytic
components, resulting in more efficient electron transfer in the
photocatalytic process and thus enhanced catalytic activity.
With potential porous nature, soaking catalyst in PS and
TEOA solution before light irradiation might be another
method to provide a platform for efficient contact among
catalytic components. We thus performed this soaking
experiment and it exhibited large increment in catalytic activity
−
was 456.0 μmol/(g h) and the selectivity of HCOO /CO and
−
HCOO /H was high up to 326 and 61, respectively. The
2
value of 456.0 μmol/(g h) was comparable to some similar
9
metal−oxygen cluster based MOF catalysts.
−
In order to testify the origin of carbon source for HCOO
product, we carried out C isotope labeling experiments. The
experimental condition is similar to Entry 4 in Table 2, except
1
3
(
488 μmol/(g h)) with comparison to that without soaking
1
2
13
(Table 1, entry 3). Further, combined grinding and soaking
that CH CN is changed to CD CN and CO to CO . After
3 3 2 2
1
3
strategy leads to a nearly superimposed influence on catalytic
activity (951 μmol/(g h)). Through adjusting the appropriate
solvent ratio, Co-MOF achieved its highest catalytic activity of
the photocatalytic reaction, we tested the C NMR of the
1
3
−
solution, and found that the signal of H COO was observed
at 169.8 ppm, and the signals at 168.2 and 159.4 ppm should
be attributed to CO3 and H CO (Figure S7).
1
3
2−
13
−
1
102 μmol/(g h) DMF/H O (4 mL/2 mL) solvents during 3
2
3
h photocatalytic hydrogen evolution experiment (Table 1,
entry 5). SEM image of Co-MOF powder after photocatalytic
reaction indicated of maintained granular particle morphology
We further investigated the photocatalytic mechanism of
Co-MOF for H evolution and CO , photophysical properties
2
2
were measured. To the best of our knowledge, when
2
+
(
Figure S2e). And the unchanged PXRD pattern after catalysis
[Ru(bpy)3] was used as photosensitizer, its excited state,
2
+
indicates of good crystallinity in comparison to original Co-
MOF (Figure S2b). To evaluate the photostability of Co-
MOF, we have performed cycle experiments. The second and
third cycle performances of Co-MOF were 1213 and 1017
μmol/(g h), respectively, which indicates good photostability
of Co-MOF catalyst.
[Ru(bpy) ] *, might be either reductively or oxidatively
3
quenched by an electron donor or acceptor resulting in a
1
9
decreasing intensity in fluorescence. In our hydrogen
evolution, TEOA used as electron donor agent, but it could
2
+
not reductively quench [Ru(bpy) ] * excited state, because
3
no reduction in fluorescence was observed when TEOA was
To manifest the design of photocatalyst including both PS
and catalyst unit, we investigated the photocatalytic CO2
reduction of Co-MOF without additional PS (Table 2). All
added into Ru(bpy) Cl solution (Figure S8). Therefore, the
photocatalytic hydrogen evolution might be ascribed to
oxidatively quenching mechanism.
3
2
of the experiment mixture was purged with CO for 20 min
To confirm this assumption, we investigated the thermody-
namics feasibility of the photoinduced electron transition
(PET) that happened from Ru(bpy) Cl to Co-MOF. First, a
2
before photoreaction to remove the air. The products were
determined by GC chromatography (CO, H ) and ion
2
3
2
−
chromatography (HCOO ). When 4 mg of Co-MOF
cyclic voltammetry experiment of Co-MOF was carried out in
drying and degassed DMF solution including 0.1 M Bu NPF .
photocatalyst, 5 mL of MeCN solvent, and 0.1 g of BNAH
4
6
(
1-benzyl-1,4-dihydronicotinamide) were taken for the experi-
Co-MOF powder was dropped on surface of the working
electrode and dried under infrared lamp before measurement.
The result showed an irreversible and a reversible reduction
peaks at −0.83 and −0.96 V vs Ag/AgCl, respectively (Figure
2b). These two peaks should be ascribed to the single electron
−
ment, it achieved 1.9 μmol of HCOO and 0.05 μmol of CO
in 6 h irradiation with a LED lamp, whereas no hydrogen gas
was found. It exhibited selective catalysis nature, and the
selectivity of HCOO /CO was up to 38. Correspondingly, the
catalytic activity of Co-MOF for producing CO and HCOO
−
−
II
II
II
I
II
II
transfer from [Co Co Co ] to [Co Co Co ] and from
20
I
II
II
I
II
I
was 1.9 and 78.2 μmol/(g h), respectively (Table 2, entry 3). It
[Co Co Co ] to [Co Co Co ], respectively. Then free
E
DOI: 10.1021/acs.inorgchem.8b01265
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
0
energy change (ΔG ) could be estimated by the Rehm−Weller
in PS solution before photoreaction. As integrated photo-
catalyst of Co-MOF, it includes PS and catalyst units and thus
2
1
equation
provides a moderate activity for CO reduction. Proposed
°
2
2
ΔG = E − Ered − E − e /εd
(1)
ox
T
mechanisms indicate the proton and CO reduction results
2
II
II
II
2
3
+
from the redox cycle of trinuclear [Co Co Co ] cluster.
Though the activity is to be enhanced, we believe that the
present study has addressed the importance of noble-metal-free
cluster MOFs for further artificial photosynthesis.
Herein, Eox should be the oxidation potential of Ru(bpy)
+1.53 V vs NHE), Ered was the reduction potential of Co-
19
(
3
MOF (−0.63 V vs NHE), and E was the MLCT excited state
T
estimated to be 2.20 V from tangent to the high-energy side of
emission spectrum of Ru(bpy)32 (Figure S5). The term e /εd
+
2
ASSOCIATED CONTENT
Supporting Information
could be negligible, because the dielectric constant (ε) of the
polar solvents (such as, DMF and H O) was quite large.
Therefore, the ΔG value was calculated to be −0.04 V< 0,
which indicated that PET progress from Ru(bpy)3 to Co-
MOF was thermodynamically favorable. Then the photo-
catalytic mechanism for hydrogen evolution could be described
as following (Figure 4a). [Ru(bpy)3]2 is excited under light
■
*
S
2
0
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.8b01265.
2
+
IR, TG, PXRD, SEM, and other figures for testifying the
photocatalytic process (PDF)
+
Accession Codes
CCDC 1834360 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cam-
bridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
E-mail: panm@mail.sysu.edu.cn.
■
*
ORCID
Mei Pan: 0000-0002-8979-7305
Cheng-Yong Su: 0000-0003-3604-7858
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Figure 4. Proposed photocatalytic mechanisms of (a) hydrogen
■
evolution and (b) CO reduction for Co-MOF under visible-light
This work was supported by NSFC (21771197,
21720102007), STP Project of Guangzhou (201504010031),
FRF for the Central Universities, and NSF of Guangdong
Province (S2013030013474).
2
irradiation.
2
+
irradiation and reaches its excited state [(Ru(bpy) ] *. The
3
excited state is then quenched by Co-MOF catalyst, in which
2
+
3+
[
[
[
Ru(bpy) ] * is oxidized to [Ru(bpy) ]
and the
REFERENCES
3
3
■
II
II
II
I
II
II
Co Co Co ] center is reduced to [Co Co Co ]. Thereafter,
(
1) (a) Qi, J.; Zhao, K.; Li, G.; Gao, Y.; Zhao, H.; Yu, R.; Tang, Z.
3
+
2+
Ru(bpy)3] is reduced by TEOA to reproduce [Ru(bpy) ] ,
3
Multi-shelled CeO hollow microspheres as superior photocatalysts
2
I
II
II
and the reductive state [Co Co Co ] was for proton
reduction.
for water oxidation. Nanoscale 2014, 6, 4072−4077. (b) Zhao, K.;
Zhao, S.; Qi, J.; Yin, H.; Gao, C.; Khattak, A. M.; Wu, Y.; Iqbal, A.;
Wu, L.; Gao, Y.; Yu, R.; Tang, Z. Cu O clusters grown on TiO
nanoplates as efficient photocatalysts for hydrogen generation. Inorg.
Chem. Front. 2016, 3, 488−493. (c) Gao, C.; Meng, Q.; Zhao, K.; Yin,
H.; Wang, D.; Guo, J.; Zhao, S.; Chang, L.; He, M.; Li, Q.; Zhao, H.;
Huang, X.; Gao, Y.; Tang, Z. Co
controllable facets enabling highly efficient visible-light photocatalytic
When no additional PS is used in photocatalytic CO
2
2
2
reduction process, the organic linker HL3 ligand in Co-
−
MOF acts an antenna for absorption of visible light (Figure
b). Under irradiation, HL3− ligand achieves its excited state
4
II
II
II
O hexagonal platelets with
3 4
and transfers electron to Co−O cluster. [Co Co Co ] center
I
II
II
obtains the electron and is reduced to [Co Co Co ] for
reduction of CO . Adv. Mater. 2016, 28, 6485−6490.
−
2
reducing CO to CO or HCOO in the presence of BNAH
2
(2) (a) Meyer, K.; Ranocchiari, M.; van Bokhoven, J. A. Metal
organic frameworks for photo-catalytic water splitting. Energy Environ.
Sci. 2015, 8, 1923−1937. (b) Wang, W.; Xu, X.; Zhou, W.; Shao, Z.
Recent progress in metal-organic frameworks for applications in
electrocatalytic and photocatalytic water splitting. Adv. Sci. 2017, 4,
sacrificial agent. BNAH acts as electron donor to ensure the
favorable electron transition during the photoreaction.
CONCLUSION
The first trinuclear Co-MOF is synthesized and performed for
photocatalytic hydrogen evolution and CO reduction. This
■
1
600371. (c) He, H.; Perman, J. A.; Zhu, G.; Ma, S. Metal-organic
frameworks for CO chemical transformations. Small 2016, 12, 6309−
2
2
6
324. (d) Chen, Y.; Wang, D.; Deng, X.; Li, Z. Metal-organic
Co-MOF exhibits its semiconductive behavior and the band
gap is determined to be 2.95 eV. Various experiment
conditions were optimized and it achieved the best activity
frameworks (MOFs) for photocatalytic CO reduction. Catal. Sci.
2
Technol. 2017, 7, 4893−4904. (e) Zhang, Y.; Guo, J.; Shi, Y.; Zhu, Y.;
Hou, K.; Zheng, Y.; Tang, Z. Tunable chiral metal organic frameworks
toward visible light-driven asymmetric catalysis. Sci. Adv. 2017, 3,
(
1102 μmol/(g h)) when catalyst was ground and then soaked
F
DOI: 10.1021/acs.inorgchem.8b01265
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
e1701162. (f) Liao, W.-M.; Zhang, J.-H.; Wang, Z.; Yin, S.-Y.; Pan,
M.; Wang, H.-P.; Su, C.-Y. Post-synthetic exchange (PSE) of UiO-67
frameworks by Ru/Rh half-sandwich units for visible-light-driven H2
efficient visible-light driven CO reduction to CO in CH CN/H O
solution. Angew. Chem., Int. Ed. 2017, 56, 738−743.
(11) (a) Wang, S.; Yao, W.; Lin, J.; Ding, Z.; Wang, X. Cobalt
2
3
2
evolution and CO reduction. J. Mater. Chem. A 2018, 6, 11337−
imidazolate metal-organic frameworks photosplit CO under mild
2
2
1
(
1345.
reaction conditions. Angew. Chem., Int. Ed. 2014, 53, 1034−1038.
(b) Wang, Y.; Huang, N.-Y.; Shen, J.-Q.; Liao, P.-Q.; Chen, X.-M.;
Zhang, J.-P. Hydroxide ligands cooperate with catalytic centers in
3) (a) Wang, C.; deKrafft, K. E.; Lin, W. Pt nanoparticles@
photoactive metal-organic frameworks: efficient hydrogen evolution
via synergistic photoexcitation and electron injection. J. Am. Chem.
Soc. 2012, 134, 7211−7214. (b) Wen, M.; Mori, K.; Kamegawa, T.;
Yamashita, H. Amine-functionalized MIL-101(Cr) with imbedded
platinum nanoparticles as a durable photocatalyst for hydrogen
production from water. Chem. Commun. 2014, 50, 11645−11648.
metal-organic frameworks for efficient photocatalytic CO
reduction.
2
J. Am. Chem. Soc. 2018, 140, 38−41.
(12) Spek, A. L. Single-crystal structure validation with the program.
J. Appl. Crystallogr. 2003, 36, 7−13.
(13) (a) Cai, J.; Wang, H.; Wang, H.; Duan, X.; Wang, Z.; Cui, Y.;
Yang, Y.; Chen, B.; Qian, G. An amino-decorated NbO-type metal-
(
c) Pullen, S.; Ott, S. Photochemical hydrogen production with metal-
organic frameworks. Top. Catal. 2016, 59, 1712−1721. (d) Hou, C.-
organic framework for high C H storage and selective CO capture.
2
2
2
C.; Li, T.-T.; Cao, S.; Chen, Y.; Fu, W.-F. Incorporation of a
RSC Adv. 2015, 5, 77417−77422. (b) Liao, W.-M.; Zhang, J.-H.; Yin,
S.-Y.; Lin, H.; Zhang, X.; Wang, J.; Wang, H.-P.; Wu, K.; Wang, Z.;
Fan, Y.-N.; Pan, M.; Su, C.-Y. Tailoring exciton and excimer emission
in an exfoliated ultrathin 2D metal-organic framework. Nat. Commun.
[
Ru(dcbpy)(bpy)2]2+ photosensitizer and a Pt(dcbpy)Cl2 catalyst
into metal-organic frameworks for photocatalytic hydrogen evolution
from aqueous solution. J. Mater. Chem. A 2015, 3, 10386−10394.
2
(
018, 9, 2401.
14) Zhang, S.; Li, L.; Zhao, S.; Sun, Z.; Luo, J. Construction of
interpenetrated ruthenium metal-organic frameworks as stable
photocatalysts for CO reduction. Inorg. Chem. 2015, 54, 8375−8379.
(
4) (a) Rimoldi, M.; Howarth, A. J.; DeStefano, M. R.; Lin, L.;
Goswami, S.; Li, P.; Hupp, J. T.; Farha, O. K. Catalytic zirconium/
hafnium-based metal-organic frameworks. ACS Catal. 2017, 7, 997−
1
014. (b) Chinapang, P.; Okamura, M.; Itoh, T.; Kondo, M.;
2
(
15) Materials Studio Release Notes, Release 6.1.0; Accelrys Software,
Inc.: San Diego, 2012.
16) (a) Silva, C. G.; Corma, A.; García, H. Metal-organic
frameworks as semiconductors. J. Mater. Chem. 2010, 20, 3141.
b) Hendon, C. H.; Tiana, D.; Fontecave, M.; Sanchez, C.; D’Arras,
Masaoka, S. Development of a framework catalyst for photocatalytic
hydrogen evolution. Chem. Commun. 2018, 54, 1174−1177.
(
(
5) Kataoka, Y.; Sato, K.; Miyazaki, Y.; Masuda, K.; Tanaka, H.;
Naito, S.; Mori, W. Photocatalytic hydrogen production from water
(
using porous material [Ru (p-BDC) ] . Energy Environ. Sci. 2009, 2,
2
2 n
L.; Sassoye, C.; Rozes, L.; Mellot-Draznieks, C.; Walsh, A.
Engineering the optical response of the titanium-MIL-125 metal-
organic framework through ligand functionalization. J. Am. Chem. Soc.
3
(
97.
6) Feng, Y.; Chen, C.; Liu, Z.; Fei, B.; Lin, P.; Li, Q.; Sun, S.; Du, S.
Application of a Ni mercaptopyrimidine MOF as highly efficient
catalyst for sunlight-driven hydrogen generation. J. Mater. Chem. A
2
(
013, 135, 10942−10945.
17) Ramírez-Ortega, D.; Melen
Gonzalez, I.; Arroyo, R. Semiconducting properties of ZnO/TiO
́
dez, A. M.; Acevedo-Pena, P.;
2
(
015, 3, 7163−7169.
́
7) Shi, D.; Zheng, R.; Sun, M. J.; Cao, X.; Sun, C. X.; Cui, C. J.; Liu,
2
composites by electrochemical measurements and their relationship
C. S.; Zhao, J.; Du, M. Semiconductive copper(I)-organic frameworks
for efficient light-driven hydrogen generation without additional
photosensitizers and cocatalysts. Angew. Chem., Int. Ed. 2017, 56,
with photocatalytic activity. Electrochim. Acta 2014, 140, 541−549.
(18) Chen, X.; Shen, S.; Guo, L.; Mao, S. S. Semiconductor-based
photocatalytic hydrogen generation. Chem. Rev. 2010, 110, 6503−
570.
19) Narayanam, J. M.; Stephenson, C. R. Visible light photoredox
catalysis: applications in organic synthesis. Chem. Soc. Rev. 2011, 40,
02−113.
20) Mandal, S.; Shikano, S.; Yamada, Y.; Lee, Y. M.; Nam, W.;
1
4637−14641.
8) (a) Zhang, T.; Lin, W. Metal-organic frameworks for artificial
photosynthesis and photocatalysis. Chem. Soc. Rev. 2014, 43, 5982−
993. (b) Shen, L.; Liang, R.; Wu, L. Strategies for engineering metal-
organic frameworks as efficient photocatalysts. Chin. J. Catal. 2015,
6, 2071−2088. (c) Wang, S.; Wang, X. Small 2015, 11, 3097−3112.
d) Liao, W.-M.; Zhang, J.-H.; Hou, Y.-J.; Wang, H.-P.; Pan, M.
6
(
(
5
1
(
3
(
Llobet, A.; Fukuzumi, S. Protonation equilibrium and Hydrogen
Production by a dinuclear cobalt-hydride complex reduced by
cobaltocene with trifluoroacetic acid. J. Am. Chem. Soc. 2013, 135,
Visible-light-driven CO2 photo-catalytic reduction of Ru(II) and
Ir(III) coordination complexes. Inorg. Chem. Commun. 2016, 73, 80−
1
(
5294−15297.
21) Hirahara, M.; Masaoka, S.; Sakai, K. Syntheses, character-
ization, and photochemical properties of amidate-bridged Pt(bpy)
8
(
9.
9) (a) Fu, Y.; Sun, D.; Chen, Y.; Huang, R.; Ding, Z.; Fu, X.; Li, Z.
An amine-functionalized titanium metal-organic framework photo-
2
+
dimers tethered to Ru(bpy)3 derivatives. Dalton Trans. 2011, 40,
967−3978.
catalyst with visible-light-induced activity for CO reduction. Angew.
2
3
Chem., Int. Ed. 2012, 51, 3364−3367. (b) Sun, D.; Fu, Y.; Liu, W.; Ye,
L.; Wang, D.; Yang, L.; Fu, X.; Li, Z. Studies on photocatalytic CO
2
reduction over NH -Uio-66(Zr) and its derivatives: towards a better
2
understanding of photocatalysis on metal-organic frameworks. Chem. -
Eur. J. 2013, 19, 14279−14285. (c) Wang, D.; Huang, R.; Liu, W.;
Sun, D.; Li, Z. Fe-based MOFs for photocatalytic CO reduction: role
2
of coordination unsaturated sites and dual excitation pathways. ACS
Catal. 2014, 4, 4254−4260.
(
10) (a) Guo, Z.; Cheng, S.; Cometto, C.; Anxolabehere-Mallart, E.;
Ng, S. M.; Ko, C. C.; Liu, G.; Chen, L.; Robert, M.; Lau, T. C. Highly
efficient and selective photocatalytic CO Reduction by iron and
2
cobalt quaterpyridine complexes. J. Am. Chem. Soc. 2016, 138, 9413−
9
416. (b) Artero, V.; Chavarot-Kerlidou, M.; Fontecave, M. Splitting
water with cobalt. Angew. Chem., Int. Ed. 2011, 50, 7238−7266. (c) Di
Giovanni, C.; Gimbert-Surinach, C.; Nippe, M.; Benet-Buchholz, J.;
Long, J. R.; Sala, X.; Llobet, A. Dinuclear cobalt complexes with a
decadentate ligand scaffold: hydrogen evolution and oxygen reduction
catalysis. Chem. - Eur. J. 2016, 22, 361−369. (d) Ouyang, T.; Huang,
H. H.; Wang, J. W.; Zhong, D. C.; Lu, T. B. A dinuclear cobalt
cryptate as a homogeneous photocatalyst for highly selective and
G
DOI: 10.1021/acs.inorgchem.8b01265
Inorg. Chem. XXXX, XXX, XXX−XXX