Synergistic catalysis for light-driven proton reduction using a polyoxometalate-based Cu-Ni heterometallic-organic framework

Article abstract of DOI:10.1039/c8cc09285k

Synergistic effects of bimetallic Ni and Cu supported on metal-organic polymer composites based on Wells-Dawson P₂W₁₈O₆₂^6- clusters as photosensitizer units were identified, and we report a novel approach for addressing these issues for dehydrogenation and hydrogen production reactions.

Full text of DOI:10.1039/c8cc09285k

ChemComm
View Article Online
View Journal
COMMUNICATION
Synergistic catalysis for light-driven proton
reduction using a polyoxometalate-based Cu–Ni
Cite this:DOI: 10.1039/c8cc09285k
heterometallic–organic framework†
Received 22nd November 2018,
Accepted 11th January 2019
a
a
ab
Wenlong Sun,a Cheng He,
Tao Liu
and Chunying Duan
*
DOI: 10.1039/c8cc09285k
rsc.li/chemcomm
Synergistic effects of bimetallic Ni and Cu supported on metal–organic as a promising approach to stabilize and heterogenize these redox
6ꢀ
polymer composites based on Wells–Dawson P2W18O62 clusters as active catalysts for photocatalytic hydrogen evolution; in most
photosensitizer units were identified, and we report a novel approach for reported examples, noble metal (Ru/Ir)-coordinated pyridine
addressing these issues for dehydrogenation and hydrogen production organic derivatives are required as photosensitizers.15–17
reactions.
Considering the above results, we successfully synthesized
the first example of a polyoxometalate-based polymer containing
The development of novel and highly efficient materials for bimetallic Ni–Cu ions as the active nodes for oxidative dehydro-
dehydrogenation and hydrogen (H2) production from organic genation and reductive hydrogen production. The polymer design
6ꢀ
transformations is fundamentally important in the development of included the incorporation of the Wells–Dawson P2W18O62
oxidative and reductive coupling reactions as well as in potential cluster as a photosensitizer unit into the pores of a well-
applications such as hydrogen-storage materials and the conversion established redox-active mixed-metal copper and nickel organic
of solar energy into hydrogen fuel.1–10 Transition metal (Ni and Cu) polymer. We proposed a reaction mechanism for this transforma-
composite materials, which are earth-abundant and relatively tion as shown in Scheme 1. The photocatalytic dehydrogenation
inexpensive materials, have been used for the reduction of and hydrogen evolution that occurs because of the unique photo-
6ꢀ
protons for hydrogen evolution. Bimetallic Ni–Cu materials have redox properties of the excited state of the P2W18O62 anion
been reported to exhibit higher catalytic activity in thermal H2 enabled the heteropolyacid to catalyse the direct oxidation of the
evolution than can be achieved with materials composed of Ni or substrates via ultraviolet irradiation, and the heteropolyacid formed
Cu alone.11–13 These bimetallic Ni–Cu supported materials have also heteropolyblue. The resulting heteropolyblue could be excited by
been shown to be effective photocatalysts for reductive hydrogen UV irradiation to reach a new excited state and further reduce the
evolution in the presence of an additional photosensitizer.11,13 redox-active Cu/Ni sites. The water-coordinated Cu/Ni centres with
The identification of new cocatalyst-free and self-sensitized a suitable redox potential reduce protons for achieving hydrogen
polymers for dehydrogenation and hydrogen production without evolution, completing the oxidative dehydrogenation and reductive
any other additives is of great importance.
hydrogen production over the entire catalytic cycle.14
A variety of materials have been exploited for these processes.1–13
Single-crystal X-ray diffraction analysis revealed that Cu–Ni–
Among such species, metal–organic polymers have emerged as POMs crystallize in the monoclinic space group P6/m. In our
an interesting class of porous crystalline materials that can easily multifunctionalised coordinated polymers, the [P2W18O62]6ꢀ
possess photosensitizer and metal activating sites that make
them catalytically active.14–17 Polyoxometalates are well known
photosensitive materials that have the potential to facilitate multi-
electron transfer reactions for light-driven hydrogen evolution.14
Imbedding POMs into the pores of frameworks has re-emerged
a Chemical School of Zhang Dayu State Key Laboratory of Fine Chemicals,
Dalian University of Technology, Dalian 116024, China. E-mail: cyduan@dlut.edu.cn
b Collaborative Innovation Center of Chemical Science and Engineering, Tianjin 300071,
China
† Electronic supplementary information (ESI) available: Experimental details, crystal
structure and data for catalytic reactions. CCDC 1875777. For ESI and crystallographic
Scheme 1 Diagram illustrating the process of the photocatalytic reaction.
data in CIF or other electronic format see DOI: 10.1039/c8cc09285k
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun.

View Article Online
Communication
ChemComm
centres is efficient and direct, which allows the new polymers to
photocatalytically split water into hydrogen; the PLATON program
was used to calculate the void volume at approximately 210 Å.3
Confocal laser scanning microscopy exhibited strong green fluores-
cence (lex = 520 nm) attributable to the emission of the fluorescein
dye (Fig. S6, ESI†), confirming that the uptake of the dye molecules
by the crystals of the MOF was unsuccessful. The dye molecules
cannot deeply penetrate the internal crystal framework. Further-
more, the results demonstrated that the catalytic process should
occur on the surface of the material.
We could obtain Cu–POM and Ni–POM single-crystal structures,
but scaling up the syntheses of Cu–POM or Ni–POM was difficult,
which is not conducive to industrialization or catalytic experiments.
Cu–Ni–POM not only possesses Cu–POM and Ni–POM structural
motifs but can also be produced on a large scale, which is beneficial
for industrialization and catalytic experiments. ICP-MS analysis of
the Cu–Ni–POM showed that it contained equal portions of copper
and nickel (1 : 1). The solid-state UV-vis absorption spectra of
Zn–POM14 exhibited a polyoxometalate-based broad absorption
6ꢀ
Fig. 1 (a) Diagram of the coordination environment of the P2W18O62
cluster, and Cu2+ and Ni2+ ions. (b) View of a single molecule cage,
showing the constituents and fragments, i.e., P2W18O626ꢀ, OX, CuII, NiII
6ꢀ
band at 200–400 nm, which was assigned to P2W18O62 (Fig. S7,
and PBPY. (c) View of a two-dimensional layer Kagome structure. (d) View
´
of the front face of a three-dimensional structure showing embedded
P2W18O626ꢀ units within the pores along the b axis direction. (e) Side view
ESI†). The solid-state UV-vis absorption spectrum of Cu–Ni–POM
exhibited a polyoxometalate-based broad absorption band at
200–400 nm and an additional visible metal (Cu)-to-ligand-to-
metal (Ni) charge-transfer band at 500–1000 nm (Fig. S11, ESI†).
The spectrum of Cu–Ni–POM showed an emission band at approxi-
mately 409 nm (Fig. S9, ESI†). In this case, the free energy change
of a one-dimensional pore structure (nanotube) in (d) formed by the
6ꢀ
ligands with embedded [P2W18O62
]
units in the interior.
ions act as photosensitive and photooxidation centres,14,18 and (E0–0) between the ground state and the vibrationally related excited
the Cu–Ni–POM was designed with Cu(II)/Ni(II) as the metal state was approximately 3.10 eV (Fig. S12, ESI†), and this high
nodes and 4,40-(1,4-phenylene)bis-pyridine (PBPY) and oxalate excitation energy created an active excited state for the photo-
(OX) linkers to balance the charge of the [P2W18O62]6ꢀ anions oxidation of substrates or electron donors. Electrochemical
(Fig. 1a). The Cu/Ni ions were coordinated in an octahedral measurements of Cu–Ni–POM showed six redox peaks, with four
geometry, where Cu(1)/Ni(1) is connected to four oxygen atoms of them attributable to [P2W18O62]6ꢀ; the first two redox couples
from two OX molecules and two nitrogen donors from two (I and II) correspond to one-electron redox processes that can induce
PBPY ligands, and Cu(2)/Ni(2) is coordinated to four oxygen oxidative reactions, and the last two redox couples (III and IV)
atoms from one water molecule and two OX molecules and two correspond to two-electron redox processes (Fig. 2a).14,15 Peaks
nitrogen donors from two PBPY ligands (Fig. 1a). The unique V and VI correspond to the Cu2+ and Ni2+ redox peaks (Fig. 2b),
coordination geometry of the Cu(2)/Ni(2) centres with open respectively. Indeed, the addition of Et3NH+ to the Cu–Ni–POM
axial sites to allow H2O coordination is the source of the redox in aqueous media leads to new waves near the redox peaks at
activity and proton-activation sites for the electrochemical approximately ꢀ1.5 (VI), ꢀ0.95 (V) and ꢀ0.7 (IV, V) that increase
hydrogen evolution from water.14
in intensity with increasing acid concentration (Fig. 2c), indicating
These two types of Cu/Ni ions are alternately connected via that the Cu–Ni–POM acts as an electrocatalyst for proton
corner-sharing oxygen atoms from OX anions to form a two- reduction.19,20
´
dimensional sheet with a Kagome structure (Fig. 1c). The two-
The ultraviolet-light-driven catalytic activity of Cu–Ni–POM
´
dimensional Kagome structure within the polymers consists of for hydrogen evolution is given in Table S2 (ESI†). The optimal
triangular and hexagonal structures (Fig. S2, ESI†). The adjacent conditions were found to be a H2O/Me2CO (1 : 2 by volume)
sheets are connected by rigid PBPY ligands via Cu–N/Ni–N bonds solution containing Cu–Ni–POM and MeOH (25% v/v). When
to form three-dimensional cationic pores along the axes (Fig. S3 Zn–POM or K6P2W18O62 was used as the photocatalyst, little
and S4, ESI†). Along the b axis (Fig. 1e), the 1D nanotubes have a hydrogen evolution was detected under the same experimental
diameter of 12.3 Å and can serve as a cage for a single molecule. conditions (Fig. 3b). A mixture solution of K6P2W18O62 and
The tubes are made of the constituents and fragments, i.e., Ni(NO3)2 (TON approximately 210) or K6P2W18O62 and Cu(NO3)2
[P2W18O62]6ꢀ, OX, CuII, NiII and PBPY (Fig. 1b), and each of the (TON approximately 110) hydrogen gas were generated exhibited a
isolated voids in the 1D nanotube is occupied by [P2W18O62]6ꢀ higher turnover number than were generated when using Zn–POM
anions (Fig. 1d and e) arranged like train cars. The redox-active (TON approximately 18) or K6P2W18O62 (TON approximately 18) as
metal centres and photosensitizing [P2W18O62]6ꢀ centres are suffi- the photocatalyst after irradiation for 65 h (Fig. 3b). These results
ciently close in proximity that the potential photoinduced electron confirm that copper and nickel are the catalytic sites for the
transfer from the photosensitizer to the redox-active Cu(2)/Ni(2) reduction of protons to hydrogen. A mixture of K6P2W18O62 and
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2019

View Article Online
ChemComm
Communication
hydrogen gas after irradiation for 65 h (turnover frequency
ca. 833 mmol gꢀ1 hꢀ1), and the calculated turnover number of the
entire reaction was approximately 550 per mole of Cu–Ni–POM
catalyst (Fig. 3a and b). We compare with the relative studies,15,17,21
these similar heterogeneous (MOFs) photocatalysis hydrogen
evolution reaction relatively low efficiency need the presence of
noble metal catalyst, and the similar homogeneous (POMs)
photocatalysis reaction need the presence of an additional noble
metal photosensitizer. Cu–Ni–POM represents a single catalyst
that exhibits a higher activity without any other additives.
The XPS spectrum indicates that after the photocatalytic
reaction, the WVI centres in the Cu–Ni–POM anions are converted
into WV, forming an excited-state POM species14 (Fig. 2d). Photo-
catalysis by Cu–Ni–POM compounds was first characterized by
colour changes of the reaction mixtures from colourless to blue
suspensions, which is an indication of the reduction of W(VI). The
Fig. 2 (a and b) Solid-state cyclic voltammograms of Cu–Ni–POM at a
scan rate of 50 mV sꢀ1. (c) Cyclic voltammograms of Cu–Ni–POM
containing 0.1 mM TBAPF6 (black line) in the presence of different con-
centrations of HNEt3Cl [5 mmol (red line) and 10 mmol (blue line) at a scan colour remained if the system was exposed to light. However,
rate of 50 mV sꢀ1]. (d) X-ray photoelectron spectroscopy spectra of W 4f in
heteropoly before irradiation (black line) and after irradiation (red line).
re-oxidation of W(V) to W(VI) occurred within hours of exposure to
air or oxygen with concomitant decolouration of the blue suspen-
sion (Fig. S15, ESI†).14 These experimental results reveal that the
excited state POM first receives electrons from the MeOH to form
Ni(NO3)2/Cu(NO3)2 at the same concentration produced only a very
the heteropolyblue species. The latter absorbs photons to reach its
low yield of hydrogen compared to the aforementioned mixed
excited state and undergoes photoinduced electron transfer to
copper–nickel-based Cu–Ni–POM system (TON = 550) after irradia-
the copper or nickel centres. Coordination of a water molecule to
tion for 65 h (Fig. 3b). This result confirms that hydrogen produc-
the copper or nickel centres and the formation of low-valence
tion is more rapid in a mixed-metal system than in a single metal
copper or nickel species further reduce protons to generate
system. The almost linear increase in the hydrogen evolution with
hydrogen, whereas the excited-state heteropolyblue-POM species
increasing concentration of Cu–Ni–POM in the time depen-
are converted back to ground state POM species to complete the
dence curves of hydrogen evolution suggested that the photo-
overall catalytic cycle.
catalytic systems were quite stable (Fig. 3a). A catalyst loading
of approximately 1 mg produced approximately 1.2 mL of
Heteropolyblue has two absorption bands; the visible
absorption band is centred at 650 nm, and the broad ultraviolet
absorption band is at 200–350 nm (Fig. S8, ESI†). In this work,
we use the heteropolyblue ultraviolet absorption broadband at
200–350 nm with 200–400 nm UV photons to excite the POM to
obtain electrons to generate heteropolyblue and ultraviolet
photons to excite heteropolyblue (UV absorption band in
200–350 nm) photoinduced electron transfer to reduce the
mix-metal Cu and Ni ions to produce hydrogen. In addition,
the photocatalytic dehydrogenation of tetrahydroisoquinoline
and hydrogen production with Cu–Ni–POM were compared
with those of Co–POM (previous work).14 A solution containing
Cu–Ni–POM (1 mg, 0.1 mmol) and tetrahydroisoquinoline
(3 mmol) in H2O/Me2CO (1 : 3) generated 3,4-dihydroisoquinoline
(1) (turnover number ca. 1300) and hydrogen gas (turnover number
ca. 1250) using 200–400 nm irradiation for 68 hours (Fig. 3d).
A catalyst loading of approximately 1 mg produced 2.8 mL of
hydrogen gas after irradiation for 68 h (Fig. 3c). The Cu–Ni–POM
Fig. 3 (a) Volume of evolved hydrogen as a function of time using Cu–Ni–
POM (0 mmol, 0.05 mmol or 0.1 mmol based on the molecular cage unit) in a
H2O/Me2CO (1 : 2 by volume) solution containing MeOH (25% v/v) after light
(this work) exhibited a higher activity than Co–POM in the catalytic
dehydrogenation of tetrahydroisoquinoline and hydrogen produc-
tion (Fig. 3d). In the Cu–Ni–POM, the Cu, Ni and POM act as
irradiation for 65 h (200–400 nm). (b) Calculated turnover numbers of different
catalysts in photoreductive hydrogen production in a H2O/Me2CO (1 : 2 by electrocatalysts for proton reduction (Fig. 2c), because the catalytic
volume) solution containing MeOH (25% v/v) after light irradiation for 65 h
dehydrogenation of tetrahydroisoquinoline and hydrogen produc-
tion occur more rapidly from the heterometal Cu–Ni ions and POM
concerted catalysis centres than with Co–POM alone (previous
(200–400 nm). (c) Volume of evolved hydrogen as a function of time using
Cu–Ni–POM (0.1 mmol). (d) Calculated turnover numbers of Cu–Ni–POM (this
work) and Co–POM (previous work) in the catalytic dehydrogenation of
work). These experiments confirm that the Cu–Ni–POM framework,
which allows a concerted mechanism, exhibited higher activity.
tetrahydroisoquinoline to produce 3,4-dihydroisoquinoline (1) and hydrogen
gas in H2O/Me2CO (1 : 3 by volume) as the solvent. V = 5.5 mL.
This journal is ©The Royal Society of Chemistry 2019
Chem. Commun.

View Article Online
Communication
ChemComm
Table 1 Reaction of aryl iodides with nitrogen-containing heterocyclesa
In summary, we have demonstrated a new cocatalyst-free
and self-sensitized catalytic system for hydrogen production by
exploiting the catalytic activity of bimetallic Cu–Ni supported
on metal–organic polymer composites based on the polyoxo-
6ꢀ
metalate P2W18O62 as a photosensitizer unit. In addition,
Cu–Ni–POM bimetallic species are efficient and stable photo-
catalysts for dehydrogenation and hydrogen evolution under
irradiation with UV light. To the best of our knowledge, this is
the first work employing a POM material with integrated
bimetallic Cu–Ni species to produce H2 fuel and a valuable
organic compound. We believe that this strategy provides
promising and practical solutions for converting solar energy
into hydrogen fuel and for conducting organic oxidation reac-
tions to generate valuable compounds.
This work was supported by the National Natural Science
Foundation of China (No: U1608224 and 21531001).
a
Reaction conditions: 1a–1l (0.5 mmol), 2a–2l (1 mmol), catalyst
(0.4 mmol% based on the molecule cage unit in Fig. 1b), NaOH
(1.0 mmol), DMSO (1 mL), 120 1C, 12 h. Isolated yield.
Conflicts of interest
The C–N bond forming experiments confirm that the con-
certed reaction with the Cu–Ni–POM catalysis was rapid. The
experimental results are listed in Table S1 (ESI†). Either
Cu(NO3)2 (10%) or MOF-199 (2.5%) was used as the catalyst
under the same reaction conditions, and they provided obviously
lower yields than those achieved with Cu–Ni–POM (0.4%). The
results demonstrated that the concerted reactions with the
copper metal (catalyst) and nickel metal/POM (cocatalyst) were
more efficient. Under the optimal conditions, the scope of the
Cu–Ni–POM catalyst system was explored using various aryl
iodides and N-containing heterocycles. As shown in Table 1, in
general, the N-containing heterocycles with high numbers of
nitrogen atoms (1H-1,2,4-triazole) gave the coupling products in
lower yields (3k, 3l). We proposed that the Cu–Ni–POM active
site coordinated to the nitrogen atoms in the nitrogen-dense
substrates; therefore, substrates containing fewer nitrogen
atoms exhibited higher activities than those with high nitrogen
contents (Table 1). The coupling reaction proceeded smoothly
for most of the aryl iodides having electron-rich, electron-neutral
or electron-deficient groups (but electron-rich and electron-
neutral groups were favoured) and afforded the corresponding
products in good to excellent yields (3a, 3b, 3e, 3f, 3g, 3j and 3k).
There are no conflicts to declare.
Notes and references
1 G. E. Dobereiner, A. Nova, N. D. Schley, N. Hazari, S. J. Miller,
O. Eisenstein and R. H. Crabtree, J. Am. Chem. Soc., 2011, 133, 7547.
2 S. Pullen, H. H. Fei, A. Orthaber, S. M. Cohen and S. Ott, J. Am. Chem.
Soc., 2013, 135, 16997.
3 Y. Qin and M. Oestreich, Angew. Chem., Int. Ed., 2017, 56, 7716.
4 X. Chen, S. Shen, L. Guo and S. S. Mao, Chem. Rev., 2010, 110, 6503.
5 K. H. He, F. F. Tan, C. Z. Zhou, G. J. Zhou, X. L. Yang and Y. Li,
Angew. Chem., Int. Ed., 2017, 56, 3080.
6 T. Zhang and W. B. Lin, Chem. Soc. Rev., 2014, 43, 5982.
7 J. Zhang, Y. Chen and X. Wang, Energy Environ. Sci., 2015, 8, 3092.
8 D. Z. Zee, T. Chantarojsiri, J. R. Long and C. J. Chang, Acc. Chem.
Res., 2015, 48, 2027.
9 F. C. Leng, H. Liu, M. L. Ding, Q. P. Lin and H. L. Jiang, ACS Catal.,
2018, 8, 4583.
10 D. Y. Shi, R. Zheng, M. J. Sun, X. R. Cao, C. X. Sun, C. J. Cui, C. S. Liu,
J. W. Zhao and M. Du, Angew. Chem., Int. Ed., 2017, 56, 14637.
11 Y. Yamada, S. Shikano, T. Akita and S. Fukuzumi, Catal. Sci.
Technol., 2015, 5, 979.
12 Y. W. Zhang, W. Y. Huang, S. E. Habas, J. N. Kuhn, M. E. Grass,
Y. Yamada, P. D. Yang and G. A. Somorjai, J. Phys. Chem. C, 2008,
112, 12092.
13 Y. J. Hao, S. Z. Kang, X. Liu, X. Q. Li, L. X. Qin and J. Mu, ACS
Sustainable Chem. Eng., 2017, 5, 1165.
In general, the electron-deficient substrates gave the coupling 14 W. L. Sun, B. W. An, B. Qi, T. Liu, M. Jin and C. Y. Duan, ACS Appl.
Mater. Interfaces, 2018, 10, 13462.
15 Z. M. Zhang, T. Zhang, C. Wang, Z. K. Lin, L. S. Long and W. B. Lin,
products in lower yields (3c, 3d, 3h, 3i and 3l), and the electron-
rich indole substrate exhibited a higher activity than the
J. Am. Chem. Soc., 2015, 137, 3197.
electron-deficient indole substrate (Table 1). To better verify 16 X. J. Kong, Z. K. Lin, Z. M. Zhang, T. Zhang and W. B. Lin, Angew.
Chem., Int. Ed., 2016, 55, 6411.
17 J. Tian, Z. Y. Xu, D. W. Zhang, H. Wang, S. H. Xie, D. W. Xu,
the structure of the catalyst surface, scanning electron micro-
scopy (SEM) measurements were performed. The SEM images of
Y. H. Ren, H. Wang, Y. Liu and Z. T. Li, Nat. Commun., 2016,
the initial Cu–Ni–POM displayed particle sizes of approximately
0.16–0.19 mm, which showed a catalytic yield of 79% of 3a
after 12 hours. Fine crystalline material with particle sizes of
7, 11580.
18 S. S. Wang and G. Y. Yang, Chem. Rev., 2015, 115, 4893.
19 X. Jing, Y. Yang, C. He, Z. D. Chang, J. N. H. Reek and C. Y. Duan,
Angew. Chem., Int. Ed., 2017, 56, 11759.
0.5–2.5 mm, which was prepared by grinding Cu–Ni–POM crystals, 20 X. Y. Dong, M. Zhang, R. B. Pei, Q. Wang, D. H. Wei, S. Q. Zang,
Y. T. Fan and T. C. W. Mak, Angew. Chem., Int. Ed., 2016, 55, 2073.
displayed a catalytic yield of 97% of 3a after 12 hours (Fig. S16,
21 H. J. Lv, W. W. Guo, K. F. Wu, Z. Y. Chen, J. Bacsa, D. G. Musaev,
ESI†). These results demonstrated that the surface area is a crucial
Y. V. Geletii, S. M. Lauinger, T. Q. Lian and C. L. Hill, J. Am. Chem.
factor in achieving a high yield.
Soc., 2014, 136, 14015.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2019