Nickel-thiolate complex catalyst assembled in one step in water for solar H2 production

Article abstract of DOI:10.1021/ja208555h

We report the use of a simple complex assembled from Ni(II) salt and 2-mecaptoethanol in one step in water as the efficient catalyst in a molecular hydrogen system which can be sensitized by a low-cost xanthene dye, Erythrosin B. An excellent quantum efficiency of 24.5% is attained at 460 nm. This simple system is expected to contribute toward the development of economical and environmentally benign solar hydrogen production systems.

Full text of DOI:10.1021/ja208555h

Communication
pubs.acs.org/JACS
Nickel−Thiolate Complex Catalyst Assembled in One Step in Water
for Solar H Production
Wei Zhang, Jindui Hong, Jianwei Zheng, Zhiyan Huang, Jianrong (Steve) Zhou, and Rong Xu*
2
†
†
‡
§
§
,†
†
School of Chemical & Biomedical Engineering, Nanyang Technological University, 62, Nanyang Drive, Singapore 637459, Singapore
‡
Institute of High Performance Computing, Agency for Science, Technology and Research, 1 Fusionopolis Way, #16-16 Connexis,
1
§
38632, Singapore
School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore 637371, Singapore
*
S Supporting Information
one step in water as the efficient catalyst in a molecular
hydrogen system which can be sensitized by low-cost xanthene
dyes. This simple system is expected to contribute toward the
development of economical and environmentally benign solar
hydrogen production systems.
ABSTRACT: We report the use of a simple complex
assembled from Ni(II) salt and 2-mecaptoethanol in one
step in water as the efficient catalyst in a molecular
hydrogen system which can be sensitized by a low-cost
xanthene dye, Erythrosin B. An excellent quantum
efficiency of 24.5% is attained at 460 nm. This simple
system is expected to contribute toward the development
of economical and environmentally benign solar hydrogen
production systems.
The molecular hydrogen system consisting of all earth-
abundant elements was assembled by mixing nickel(II) acetate
(
3 mM) and 2-mercaptoethanol (ME) (30 mM) to form the
catalyst complex in the aqueous solution containing triethanol-
amine (TEOA, 15 vol%) as the sacrificial reagent, followed by
adding Erythrosin B (EB, 2.25 mM) as the PS (Scheme 1). The
hotocatalytic hydrogen production from water represents
an important process in sustainable solar energy
Scheme 1. A Simple Process to Assemble the EB−Ni−ME
Molecular System from Earth-Abundant Elements
P
conversion for the future. This process can be realized by
using the molecular hydrogen system which can provide a
homogeneous reaction environment toward optimal availability
1
of active catalytic sites for solar hydrogen production. Such a
system typically comprises a metal complex photosensitizer
2
+
(
PS), e.g., Ru(bpy)3 , for solar light harvesting and a metal-
based catalyst, such as metallic colloidal Pt or Pt-complexes for
receiving the electrons from the excited PS directly or via an
electron relay. Despite the promising findings so far, the
frequent use of precious metals and/or complicated structures
associated with the photosensitizers or catalysts limits their
potential wide applications.
resultant system is denoted as EB−Ni−ME. Toxic solvents like
acetonitrile or toluene commonly used in other molecular
hydrogen systems to better dissolve the metal complexes are
not required in our system. In the aqueous solution, Ni ions
and the ME ligands upon mixing and dissolving readily form
the dark brown colored complex. The majority of the Ni−ME
complex appears to be homogeneous in the solution although a
very small portion of the precipitates can be separated by high
speed centrifugation.
Back in 1975, Balzani et al. conceptually showed that
transition metal complexes are in principle suitable catalysts for
2
+
2
the photodissociation of water. A few promising examples of
using transition metal complex based catalysts have been
3
a,b
reported so far, including metalloporphyrin complexes,
III
3c,d
[
Co (dmgH) pyCl],
and the bioinspired nickel phosphine
Nonetheless, these molecular complexes
2
3
e,f
type complexes.
have relatively complicated structures and may not be easily
synthesized. It is highly desirable to use a simple process toward
the preparation of low-cost molecular systems for photo-
catalytic hydrogen production potentially at a large scale.
In our recent study, inorganic NiS nanoparticles were found
capable of replacing the Pt cocatalyst in the well-known Pt/CdS
semiconductor photocatalysts while exhibiting an excellent
Figure 1A shows the time course of hydrogen evolution over
the EB−Ni−ME system. The experimental parameters
2+
including the concentrations of EB, Ni ions, and TEOA;
Ni/S ratio; and the pH have been optimized. It is found that
the EB−Ni−ME system can efficiently generate 9.5 mmol of
hydrogen gas in the first 12 h of irradiation of visible light (λ >
20 nm) corresponding to a high average production rate of
92 μmol/h. A moderate quantum efficiency of 12% is obtained
4
7
4
quantum efficiency of 51% under visible light. This
phenomenon inspired us to employ the low-cost transition
metal−sulfur complexes as the catalysts in the molecular
hydrogen systems. Herein, we report the use of a simple
complex assembled from a Ni(II) salt and 2-mecaptoethanol in
Received: September 10, 2011
Published: December 1, 2011
©
2011 American Chemical Society
20680
dx.doi.org/10.1021/ja208555h | J. Am. Chem.Soc. 2011, 133, 20680−20683

Journal of the American Chemical Society
Communication
concentration of the Ni−ME complex in TOEA solution was
2
+
estimated to be around 90% of that in DI water when [Ni ]
was in the range of 0.3−1.0 mM (Table S2). Positive ion ESI-
MS analysis was further carried out to investigate the structure
of the Ni−ME complex formed in the solutions. It was found
that the Ni−ME complex formed in DI water has a formula of
[
Ni(SCH CH OH) ] (Figure S4A and Table S3). The MS
2 2 2 6
spectrum of nickel acetate in TEOA solution shows a dominant
peak at m/z = 823 (Figure S4B) which should correspond to
the colorless tetramer formed between Ni(II) and TEOA.
Finally, the spectrum of Ni−ME in TEOA solution suggests the
presence of both Ni−ME (m/z = 1298) and tetramer Ni−
TEOA (m/z = 823) complexes (Figure S4C). In addition, a
peak at m/z = 1448 was observed, which can be assigned to
Figure 1. (A) Time course of photocatalytic hydrogen evolution over
EB−Ni−ME system (light source: 300 W Xe lamp, λ > 420 nm, 100
mL solution, pH = 8.5). (B) UV−vis absorption spectrum of EB and
the quantum efficiencies of hydrogen evolution under photons with
different wavelengths over the EB−Ni−ME system.
[
Ni(SCH CH OH) ] associated with one TEOA (m/z = 1448
2 2 2 6
−
1298 = 150, MW of TEOA = 149). Although the peak of the
[
Ni(TEOA)] complex (m/z = 823) is rather strong compared
4
at 420 nm with a band-pass filter. Our system exhibits relatively
good stability in a single run compared to other molecular
to those of [Ni(SCH CH OH) ] complexes, such results are
2 2 2 6
not reliable for quantitative analysis due to the need to produce
charged ions of all complexes and fragmentation of molecular
ions under the ESI conditions. As mentioned above based on
the UV data, it could be estimated that about 10% of Ni(II)
forms a complex with TEOA and 90% of Ni(II) remains in the
1g,3c,d
systems reported,
as the major slowdown happens only
after 16 h which is due to degradation of EB (to be discussed
shortly). The total amount of H produced from this system is
2
about 12.3 mmol in 24 h. Even under photons of longer than
5
00 nm wavelength, our system can still efficiently generate 8.2
complex with ME as [Ni(SCH CH OH) ] . Based on the
2 2 2 6
5
mmol of hydrogen in 24 h (Figure S1). The measured quantum
efficiencies under photons at different wavelengths are shown
Figure 1B. The highest quantum efficiency of 24.5% is obtained
at 460 nm, which is shorter than the wavelength of the highest
absorption of EB at 524 nm in the visible light range. At 550
nm, the quantum efficiency is still as high as 11.0%. To the best
of our knowledge, these values are among the highest reported
for the molecular hydrogen systems in the visible light range.
To confirm that hydrogen is generated from water instead of
above results and literatures, it is proposed that the Ni−ME
complex formed in TEOA solution adopts a cyclic structure
(Figure S5A), although we could not exclude the existence of a
linear structure in equilibrium (Figure S5B). TEOA can also
coordinate with a small fraction of Ni(II) but does not affect
the coordination in the majority of [Ni(SCH CH OH) ] .
2 2 2 6
EB has four iodide substitutions on its xanthene ring. Similar
to other heavy halogen-substituted xanthene dyes, the C−I
bond in EB could be cleaved quickly through reductive
quenching leading to its photodegradation and deterioration of
other components, the photoreaction in a D O solution while
2
6
keeping other conditions identical was conducted based on
activity. However, in our EB−Ni−ME system, EB is relatively
4
similar procedures reported in our previous study. The same
stable. As shown in Figure S6A, the intensity of the absorption
does not change significantly even at 8 h of irradiation. The
peak position blue shifts from 524 to 502 nm, which is
associated with the partial loss of iodide of EB. At 24 h, the
absorption peak has become much weaker and the peak
position further shifts to 488 nm corresponding to mainly the
absorption of Fluorescein after total dehalogenation of EB. The
ESI-MS analysis of the reaction mixture provided the consistent
results about the progressive loss of iodide groups from EB
during the light irradiation (Table S4). The absorption ability
of the Ni−ME complex in the range of 460 to 600 nm overlaps
with that of EB (Figure S6B) indicating that Ni−ME can
quench the excited EB efficiently.
amount of D gas was obtained in 24 h, indicating that the only
2
hydrogen source in the produced hydrogen gas is water.
Several reports can be found from the literature dealing with
the structures of the complexes between Ni(II) and primary
5
thiolates. De Brabander et al. propsed a linear “core + link”
polymeric complex between Ni(II) and ME in dilute aqueous
5a
solutions based on a titration method. On the other hand,
Gould and Harding precipitated the crystal from an alkaline
solution of concentrated Ni(II) and ME. Their X-ray diffraction
data indicated a cyclic hexameric complex with six Ni(II) ions
forming a planar ring and the Ni(II) centers bridged by twelve
5
b
ME thiolates. A few other groups also proposed the
hexagonal structure between Ni(II) and ME or other
Our Ni−ME complex has been shown to be an efficient
electrocatalyst for the reduction of protons to molecular
hydrogen based on the cyclic voltammetric study. As shown in
5
c−e
thiolate.
In order to determine the structure of the Ni−
ME complex formed under our conditions and to investigate
the influence of TEOA, UV−vis absorption and electrospray
ionization-mass spectrometry (ESI-MS) analyses were carried
out. Figure S2 shows the UV−vis absorption spectra of the Ni−
ME complex in DI water (pH = 8.5, adjusted by NaOH) and an
aqueous solution containing TEOA (pH = 8.5). In both cases,
four characteristic charge transfer bands can be observed at
nearly the same wavelength of 258, 330, 408, and 525 nm,
indicating that TEOA does not affect coordination between
Ni(II) and ME thiolate. Nevertheless, the absorption intensities
of Ni−ME in TEOA solution were slightly lower than those in
DI water. We further diluted both solutions and compared their
absorption intensities as shown in Figure S3 and Table S1. The
3e,f
Figure 2A, similar to other Ni-complexes,
the Ni−ME
complex displays an irreversible reduction wave at around
−0.69 V (vs SCE) which should be associated with the
reduction of Ni(II) to the reactive Ni(I) center. After adding
acetic acid (HAc) to the solution, the current was greatly
enhanced and the electrocatalytic proton reduction potential
was shifted slightly more negative to −0.73 V. Similar
observations were reported by other groups for transition
3a,e,f
metal complexes of nickel and iron.
The oxidation and
reduction potentials of EB obtained by the cyclic voltammetric
studies are +0.88 and −1.05 V, respectively. The reduction
potential of TEOA was measured to be −0.55 V. By comparing
2
0681
dx.doi.org/10.1021/ja208555h | J. Am. Chem.Soc. 2011, 133, 20680−20683

Journal of the American Chemical Society
Communication
conservation rule (Wigner Rule), the PS in its long-lived
excited triplet state can sensitize the catalyst toward forming its
own long-lived excited triplet state and, thus, enhance the
photocatalytic activity. The optimum performance achieved
with the use of EB as the PS suggests the balance of these two
aspects in EB. Due to dehalogenation of EB during reaction
(
verified through UV−vis and ESI-MS analysis as discussed
above), the loss of iodide groups from EB eventually resulted in
a lower H production activity as the irradiation time was
2
increased. Compared with EB, Fluorescein (FL) is a relatively
more photostable dye, although the activity of the FL−Ni−ME
system was much lower (Figure 3A). Hence, we conducted a
long-term photoreaction using the FL−Ni−ME system to
investigate the stability of the Ni−ME catalyst. As shown in
Figure S8, after 4.5 days of continuous reaction, no significant
Figure 2. (A) Cyclic voltammogram of Ni−ME complex formed
between nickel(II) acetate (30 mM) and ME (60 mM) in water; a
glassy carbon working electrode; scan rate at 100 mV/s. (B) Proposed
reaction scheme of the photocatalytic hydrogen production over the
EB−Ni−ME molecular system.
decrease in H production activity was observed, indicating that
2
these redox potentials, it is suggested that an oxidative
quenching pathway of the photosensitized EB by TEOA is
supported. Based on the above results, the reaction scheme of
the EB−Ni−ME molecular system for photocatalytic hydrogen
the Ni−ME complex catalyst is stable.
Figure 3B shows the hydrogen evolution over the molecular
2
+
2+
systems containing EB as the PS, Ni , or Co complexed with
different sulfur-containing small ligands as the catalysts. Overall,
production in TEOA sacrificial solution is proposed in Figure
2
+
2+
B. EB2 (a divalent anion when dissolved in water) absorbs
−
complexes of Ni exhibit better activities than those of Co .
Typically, Ni−ME is about three times higher in activity than
Co−ME. Among the nickel complexes, ME is the better
capping agent than mercaptoacetic acid (MA) and thiourea
2
3
2−
visible light efficiently to form the excited triplet state, *EB .
3
2−
The excited electron in *EB is transferred to the Ni−ME
−
complex, leaving the oxidized EB . Following the photo-
(
TU). ME is widely used in the synthesis of water-soluble ZnS
sensitization and electron injection to Ni−ME, a reactive
reduced Ni(I) center is formed which plays a critical role in
electron transfer to a proton for hydrogen production. Further,
the presence of the basic sulfur groups surrounding the Ni
center could facilitate the formation of Ni-hydride for
subsequent electrochemical desorption leading to hydrogen
9
nanoparticles for its excellent capping ability, while an excess
amount of MA can easily precipitate Ni in aqueous solution.
Within all the combinations investigated, Ni capped with ME
homogeneously in solution is shown to be an efficient catalyst
for generating hydrogen.
2
+
2
+
evolution. The oxidized EB− is reduced back to EB by
4
,7
2−
To better understand the reaction mechanism and the
activity difference among the Ni complexes, we performed ab
initio calculations by using the hybrid B3LYP functional
together with an Lanl2dz basis set, in conjunction with a
TEOA as the sacrificial electron donor.
Besides EB, other xanthene dyes have also been investigated
and their molecular structures are shown in Figure S7. Figure
1
0a,b
polarizable continuum model of solvation (PCM),1
as
0c
implemented in the Gaussian 09 program package.
The
structures of Ni−ME (both cyclic and linear) and Ni−MA were
first optimized in water. Two interesting features can be seen
from the simulation. First, the highest occupied molecular
orbitals (HOMOs) and the lowest unoccupied molecular
orbitals (LUMOs) are located at Ni atoms and the surrounding
S atoms in the Ni−ME complex, as shown in Figure 4A. Similar
Figure 3. Time course of hydrogen evolution using (A) different
xanthene dyes as the PSs and Ni−ME as the catalyst and (B) EB as the
PS, and Ni2 and Co complexed with different ligands, ME, MA
+
2+
(
3
mercaptoacetic acid) and TU (thiourea) as the catalysts. Light source:
00 W Xe lamp, λ > 420 nm, 100 mL solution, pH = 8.5.
3
A indicates that, among these dyes with different halogen
substitutions to the xanthene ring, EB outperforms the others
in activity. The halogen-free Fluorescein exhibits a much lower
hydrogen evolution rate compared to other xanthene dyes
examined. Different halogenation substitutions affect the
ground and excited states of xanthene dyes resulting in
harvesting photons with different wavelengths and generating
electrons of different potentials. On the other hand, the heavy
iodine and bromine atoms can facilitate the formation of long-
lived triplet states of xanthene dyes from their excited singlet
Figure 4. HOMO and LUMO orbitals in (A) cyclic Ni−ME complex
and (B) Ni−MA clusters.
results were obtained for the linear Ni−ME cluster as shown in
Figure S9. Such results confirm that the Ni and S atoms in the
Ni−ME complex are important. However, the HOMOs and
LUMOs in the Ni−MA complex are located not only at Ni and
8
states through intersystem crossing. Based on the spin
2
0682
dx.doi.org/10.1021/ja208555h | J. Am. Chem.Soc. 2011, 133, 20680−20683

Journal of the American Chemical Society
Communication
S atoms but also at O atoms (Figure 4B), indicating that O
atoms in the Ni−MA complex may also play a role.
U.S.A. 2009, 106, 10460−10465. (b) Zhang, P.; Wang, M.; Li, C. X.;
Li, X. Q.; Dong, J. F.; Sun, L. C. Chem. Commun. 2010, 46, 8806−
8
808. (c) Lazarides, T.; McCormick, T.; Du, P. W.; Luo, G. G.;
Second, Mulliken charge analysis shows that Ni atoms in the
Ni−ME complex are negatively charged (Table S5), suggesting
that the Ni center in Ni−ME has an electron-rich environment
favorable for reduction of protons. For comparison, Ni atoms in
the Ni−MA complex are positively charged. Furthermore, we
simulated EB−Ni−ME and EB−Ni−MA complexes in water
respectively. The binding energy of EB with Ni−ME is 0.35 eV
stronger than that of EB with Ni−MA, suggesting that the
photogenerated electron in EB is more easily transferred to
Ni−ME than to Ni−MA. Moreover, as shown in Table S5, the
S atoms in EB−Ni−ME are negatively charged while those in
EB−Ni−MA are positively charged. Half of the Ni atoms in
EB−Ni−ME bear negative charges while all Ni atoms in EB−
Ni−MA are positively charged, showing that Ni−ME is
Lindley, B.; Eisenberg, R. J. Am. Chem. Soc. 2009, 131, 9192−9194.
d) McCormick, T. M.; Calitree, B. D.; Orchard, A.; Kraut, N. D.;
Bright, F. V.; Detty, M. R.; Eisenberg, R. J. Am. Chem. Soc. 2010, 132,
5480−15483. (e) Le Goff, A.; Artero, V.; Jousselme, B.; Tran, P. D.;
Guillet, N.; Metaye, R.; Fihri, A.; Palacin, S.; Fontecave, M. Science
(
1
2
009, 326, 1384−1387. (f) McLaughlin, M. P.; McCormick, T. M.;
Eisenberg, R.; Holland, P. L. Chem. Commun. 2011, 47, 7989−7991.
(4) Zhang, W.; Wang, Y. B.; Wang, Z.; Zhong, Z. Y.; Xu, R. Chem.
Commun. 2010, 46, 7631−7633.
(
5) (a) De Brabander, H. F.; Van Poucke, L. C.; Eeckhaut, Z. Inorg.
Chim. Acta 1972, 6, 459−462. (b) Gould, R. O.; Harding, M. M. J.
Chem. Soc. A 1970, 6, 875−881. (c) Jian, F. F.; Jiao, K.; Li, Y.; Zhao, P.
S.; Lu, L. D. Angew. Chem., Int. Ed. 2003, 115, 5900−5902. (d) Wark,
T. A.; Stephan, D. W. Organometallics 1989, 8, 2863−2843. (e) Xiao,
H. L.; Jian, F. F.; Zhang, K. J. Bull. Korean Chem. Soc. 2009, 30, 846−
848.
superior to Ni−MA as a reduction center for H production.
2
These simulation results support our experimental data.
In conclusion, the simple EB−Ni−ME molecular system
assembled from earth-abundant elements in water in one step
shows outstanding photocatalytic efficiencies for hydrogen
evolution under visible light. It is believed that such findings
present a promising opportunity toward the development of
low-cost and environmentally benign solar hydrogen produc-
tion systems to meet the increasing future energy demand.
(
6) (a) Shimidzu, T.; Iyoda, T.; Koide, Y. J. Am. Chem. Soc. 1985,
107, 35−41. (b) Lambert, C. R.; Kochevar, I. E. Photochem. Photobiol.
1997, 66, 15−25.
(7) Assuncao, N. A.; Giz, M. J.; Tremiliosi, G.; Gonzalez, E. R. J.
Electrochem. Soc. 1997, 144, 2794−2800.
(8) Shimidzu, T.; Iyoda, T.; Koide, Y. J. Am. Chem. Soc. 1985, 107,
3
(
5−41.
9) Nanda, J.; Sapra, S.; Sarma, D. D.; Chandrasekharan, N.; Hodes,
G. Chem. Mater. 2000, 12, 1018−1024.
10) (a) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55,
(
ASSOCIATED CONTENT
Supporting Information
■
117−129. (b) Pascualahuir, J. L.; Silla, E.; Tunon, I. J. Comput. Chem.
*
S
1994, 15, 1127−1138. (c) Frisch, M. J.; et al. Gaussian 09, revision A.1;
2
009.
Experimental methods, Figures S1−S9, and Table S1−S5.
Complete ref 10c. This material is available free of charge via
the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
rxu@ntu.edu.sg
■
ACKNOWLEDGMENTS
■
This work was supported by AcRF grants: ARC25/08 from
Ministry of Education, Singapore. The authors thank X. Wang
for the contribution in cyclic voltammetry studies.
REFERENCES
■
(
1) (a) Du, P. W.; Schneider, J.; Jarosz, P.; Eisenberg, R. J. Am. Chem.
Soc. 2006, 128, 7726−7727. (b) Elvington, M.; Brown, J.; Arachchige,
S. M.; Brewer, K. J. J. Am. Chem. Soc. 2007, 129, 10644−10645.
(
c) Gartner, F.; Sundararaju, B.; Surkus, A. E.; Boddien, A.; Loges, B.;
Junge, H.; Dixneuf, P. H.; Beller, M. Angew. Chem., Int. Ed. 2009, 48,
962−9965. (d) Goldsmith, J. I.; Hudson, W. R.; Lowry, M. S.;
9
Anderson, T. H.; Bernhard, S. J. Am. Chem. Soc. 2005, 127, 7502−
7510. (e) Ozawa, H.; Haga, M. A.; Sakai, K. J. Am. Chem. Soc. 2006,
128, 4926−4927. (f) Rau, S.; Schafer, B.; Gleich, D.; Anders, E.;
Rudolph, M.; Friedrich, M.; Gorls, H.; Henry, W.; Vos, J. G. Angew.
Chem., Int. Ed. 2006, 45, 6215−6218. (g) Tinker, L. L.; McDaniel, N.
D.; Curtin, P. N.; Smith, C. K.; Ireland, M. J.; Bernhard, S. Chem.
Eur. J. 2007, 13, 8726−8732. (h) Zhang, X. J.; Jin, Z. L.; Li, Y. X.; Li, S.
B.; Lu, G. X. J. Phys. Chem. C 2009, 113, 2630−2635. (i) Du, P.;
Schneider, J.; Fan, L.; Zhao, W.; Patel, U.; Castellano, F. N.; Eisenberg,
R. J. Am. Chem. Soc. 2008, 130, 5056−5058. (j) Sakai, K.; Ozawa, H.
Coord. Chem. Rev. 2007, 251, 2753−2766. (k) Tinker, L. L.; McDaniel,
N. D.; Bernhard, S. J. Mater. Chem. 2009, 19, 3328−3337.
(
2) Balzani, V.; Moggi, L.; Manfrin, M. F.; Bolletta, F.; Gleria, M.
Science 1975, 189, 852−856.
3) (a) Kluwer, A. M.; Kapre, R.; Hartl, F.; Lutz, M.; Spek, A. L.;
Brouwer, A. M.; van Leeuwen, P.; Reek, J. N. H. Proc. Natl. Acad. Sci.
(
2
0683
dx.doi.org/10.1021/ja208555h | J. Am. Chem.Soc. 2011, 133, 20680−20683