Organocatalytic photoreduction of Zn(II) to zinc metal

Article abstract of DOI:10.1039/c3cc47633b

The photoreduction of Zn(II) to Zn metal catalyzed by 8-hydroxyquinoline derivatives is reported. Structure-activity relationships, ideal reaction conditions, and the reaction kinetics are described and a mechanism involving an in situ formed quinolate complex is portrayed. the Partner Organisations 2014.

Full text of DOI:10.1039/c3cc47633b

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Organocatalytic photoreduction of Zn(II) to zinc
metal†
Cite this: Chem. Commun., 2014,
50, 5196
Anthony C. Brooks, Katherine Basore and Stefan Bernhard*
Received 4th October 2013,
Accepted 19th December 2013
DOI: 10.1039/c3cc47633b
www.rsc.org/chemcomm
The photoreduction of Zn(II) to Zn metal catalyzed by 8-hydroxy- with triethylamine performing as the sacrificial reductant.5 This work
quinoline derivatives is reported. Structure–activity relationships, ideal describes the substitution of this expensive and toxic noble metal
reaction conditions, and the reaction kinetics are described and a complex with an organocatalyst. It is found that 8-hydroxyquinoline
mechanism involving an in situ formed quinolate complex is portrayed. (Hq) derivatives form coordinated Zn species that are responsible for
light absorption and catalytic activity. This work uses the vast body of
The photon-driven reduction of metal ions is an unusual but highly
work on Zn-based quinolate complexes6 as a guideline for synthetic
useful photosynthetic reaction with potential applications in solar
modification to the ligand framework.
fuels, organic synthesis and metal coatings. For instance, coupling
Similarly to the electroluminescence tuning of quinolate LEDs,
such a process with a complementary OHÀ oxidation would reverse
structure–activity relationships were observed between the 8-hydroxy-
the discharge chemistry observed in highly efficient metal air
quinoline ligands and the photocatalytic reduction of Zn(II). Zinc
batteries1 and convert solar radiation to stored electrochemical
metal was initially observed to be produced when vials containing
potential. On the other hand, the produced metal particles with
ZnCl2, triethylamine (TEA, electron donor) and catalytic quantities of
pristine surfaces are highly reactive and can also function efficiently
8-hydroxyquinoline (Hq) were exposed to 465 nm LED light under
in organic reductive coupling reactions of aryls and olefins or in
inert atmosphere in a homemade photoreactor and zinc metal
the preparation of arylmetal species.2 Additionally, metal photo-
was quantified by established procedure.5 The omission of ZnCl2,
lithographic deposition techniques would allow for the fabrica-
TEA, Hq, or light prevented the production of zinc metal. Given the
tion of low work function electrodes for organic electronics and
mechanistic evidence observed in this work, a basic reaction scheme
photovoltaic applications.
involving the reductive quenching of a Zn(II) monoquinolate by TEA
These potential uses make the photoreduction of metals an
appealing research topic. Earlier studies predominately targeted cations
and subsequent catalytic reduction of solid ZnCl2 to afford Zn0 via a
chloride stabilized Zn(I) intermediate is proposed in Fig. 1.
with high reduction potentials or used UV radiation for the photo-
The substitution of different 8-hydroxyquinoline ligands varying
generation of base metals. For instance, Ag+ (Ered(Ag+/Ag) = +0.80 V vs.
at the 5, 7, and 2 positions was found to significantly affect the yield
SCE) was photocatalytically reduced to silver metal with [Ru(bpy)3]2+ as
of the desired photoproduct. Fig. 2 illustrates the basic structure and
the photosensitizer.3 Cations of less noble metals such as Pb(II) or Cd(II)
variable positions of the quinoline ligands used in this study while
(e.g. Ered(Cd2+/Cd) = À0.40 V vs. SCE) were photoreduced in systems
that required high-intensity UV radiation, an approach that is incom-
patible with a solar fuel scheme.4 To our knowledge, no molecular
photoreactions have been described in the literature that use visible
light with an electrochemically highly reactive metal such as Zn, Al, or
even Li. The lack of air-stability of lithium and the complex three-
electron redox process required for the photoreduction of aluminum
renders Zn(II) the target of choice for exploring this type of reaction.
Along these lines, the Bernhard group recently documented the visible-
light driven reduction of Zn(II) to zinc metal by an Ir(III) photocatalyst
Department of Chemistry, Carnegie Mellon University, 4400 Fifth Avenue,
Pittsburgh, PA, 15213, USA. E-mail: bern@cmu.edu; Tel: +1 412 468 8172
† Electronic supplementary information (ESI) available: 1H NMR, CV, rate
kinetics, Zn produced vs. [H2O]. See DOI: 10.1039/c3cc47633b
Fig. 1 General schematic of photocatalyzed Zn(II) reduction.
5196 | Chem. Commun., 2014, 50, 5196--5199
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
Fig. 2 Structure of 8-hydroxyquinoline derivatives.
Fig. 4 Dependence of the yield of photoreduced zinc metal on the zinc
salt starting material. Reactions conditions included 3 mmol 5,7-dCl-Hq
ligand, 2.95 mmol Zn(II) salt, 1 ml TEA, and 9 ml MeCN per vial.
conditions. Stoichiometric quantities of H2O are required in the
subsequent oxidative degradation of the ammonium radical cation
generated by sacrificial TEA oxidation.9 It is also important to note that
heptanes and toluene were the most effective solvents when 10%
MeCN was added. These nonpolar solvents are non-coordinating and
will thus not compete with MeCN for any open coordination sites on
Zn(II) or any intermediate. The success of nonpolar solvents despite
Fig. 3 Yield of photoreduced zinc metal upon the addition of different
8-hydroxyquinoline ligands after 68 h. Each reaction vial contained 3 mmol the limited solubility of ZnCl2 is also consistent with a heterogeneous
ligand, 3 mmol ZnCl2, 1 ml TEA, and 9 ml MeCN.
reaction, an observation that is corroborated on later in this article
with the kinetic evidence.
The reaction mechanism was further investigated in an attempt
to identify the active catalytic quinolate species. In situ catalyst
Fig. 3 documents their activity. Most noticeable is the increased yield
of zinc metal with halide-substituted quinoline ligands.
ligation is likely due to the widely known chelation of Zn(II) by
The electron withdrawing halides are located on the phenol
8-hydroxyquinolines.10 Fig. 6 plots the relative quinolate concentra-
moiety of the quinoline ligand where the HOMO resides.6d The
tions derived from peak integration areas of 1H NMR spectra versus
lower energy HOMOs that result cause enhanced luminescence
an increasing ligand to zinc(II) ratio. A single compound, likely a
quenching of halogenated quinolates by TEA and, as a conse-
mono-substituted [Zn(5,7-dCl-q)Ln] (where L = H2O, OHÀ, ClÀ or
quence, higher yields of Zn0. Among these ligands, 5,7-dCl-Hq
MeCN) complex is the dominant quinolate species when ZnCl2 is in
exhibits the highest catalytic turnover numbers (TONs), and it
excess of 5,7-dCl-Hq, such as under typical photoreduction condi-
was therefore employed for the rest of the studies.
tions. Cyclic voltammetry excludes the formation of Zn(5,7-dCl-q)2
Beyond the structure of the quinoline ligand, other photoreaction
when Zn(II) is in excess of ligand due to the lack of a second
conditions were manipulated in order to understand the involved
mechanism. The use of zinc halide salts as a starting reagent was
found to afford significantly more zinc metal than when salts with
acetate, sulphate, or tetrafluoroborate anions were employed (Fig. 4).
Zinc halide salts are likely more successful due to their ability to
stabilize a short lived Zn(I) intermediate, as postulated earlier.5
Indeed, the preferential use of ZnCl2 in the synthesis of aryl-
diimine and Cp* stabilized Zn(I)–Zn(I) complexes suggest ClÀ may
help mediate Zn(II) reduction.7,8
The originally used acetonitrile (MeCN) solvent environment
proved to be optimal; reactions performed in alcohols, ethers, esters,
and hydrocarbons exhibit far inferior metrics compared to acetonitrile.
In fact, addition of a small percentage of acetonitrile dramatically
improves the Zn yield in ethyl acetate, toluene, and heptanes (Fig. 5).
Acetonitrile has been shown to positively shift the reduction potential
of Zn(II); the resulting increased DE between the LUMOs of Zn(II) and
the photocatalyst makes MeCN the most effective solvent.5 Water was
shown to inhibit Zn(II) photoreduction at concentrations above
300 mM but it was also found that the use of anhydrous ZnCl2 failed
to produce any zinc metal under otherwise identical reaction
Fig. 5 Effect of solvent on Zn(II) photoreduction. There was in each vial
3 mM 5,7-dCl-Hq, 420 mg ZnCl2, 1 ml TEA, and 9 ml solvent.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5196--5199 | 5197

View Article Online
ChemComm
Communication
Fig. 7 Dependence of initial rate of Zn(II) on ZnCl2 per reaction vial. Each
vial contained 3 mmol ligand, 1 ml TEA, 9 ml MeCN and ZnCl2.
Fig. 6 Relative quinolate concentrations derived from 1H NMR spectra as
300 mM 5,7-dCl-Hq is titrated with ZnCl2 in 9 : 1 MeCN : TEA.
yield is likely due to the limited solubility of ZnCl2, which blocks light
absorption. A first order dependence of the reaction rate on ligand
concentration was also found. The rate of photoreduction initially is
proportional to concentration, but levels off at higher loadings due to
the saturation of light absorption of the photoactive catalyst species.
A R2 value of 0.99 is observed for the logarithmic fit derived from
Beer–Lambert’s law, which indicates that light absorption is satu-
rated at higher catalyst concentrations (ESI†).
To summarize, this work presents the photoreduction of Zn(II)
to zinc metal by a zinc quinolate catalyst that is formed in situ.
The highest yields of zinc metal were obtained using ZnCl2 and
5,7-dichloro-8-hydroxyquinoline in acetonitrile. The active catalyst
was identified as a mono-substituted zinc quinolate by 1H NMR and
luminescence quenching. Future work will focus on the replace-
ment of TEA for nonsacrificial Zn(II) photoreduction. The authors
would like to acknowledge support from the National Science
Foundation through CHE-1055547.
irreversible reduction peak generally observed for Znq2.11 Indeed,
1H NMR of dissolved solids in a typical photoreaction reveal only one
quinolate species, with another-likely the degraded or active catalyst-
forming after light exposure (ESI†). The two other quinolates in Fig. 6
correspond to 5,7-dCl-q and Zn(5,7-dCl-q)2, the latter of which was
synthesized by established procedure for reference purposes.12 The
formation of [Zn(5,7-dCl-q)Ln], and Zn(5,7-dCl-q)2 were reversible;
Keq values of formation were calculated be 1026 MÀ1 and 257 MÀ1
respectively. The parent 5,7-dCl-Hq is fully deprotonated in the TEA
environment and does not absorb 465 nm light, rendering both its
presence and participation in the photoreduction process unlikely.
Fluorescence quenching was employed to confirm that a mono-
substituted zinc quinolate is the active, zinc(II) reducing photo-
catalyst. The sacrificial electron donor TEA is oxidized at +0.934 V
vs. SHE13 and is thus incapable of reducing neither a ground state
zinc quinolate nor unligated Zn(II) directly. As a consequence,
reductive luminescence quenching is necessary. While quenching
by TEA was not detected in 5,7-dCl-q nor Zn(5,7-dCl-q)2, it was
observed in [Zn(5,7-dCl-q)Ln], indicating that it is the active catalyst.
The quenching effect only follows a linear Stern–Vollmer relation-
ship at TEA concentrations below 0.25% v/v and no additional
quenching is observed beyond 1% v/v TEA (ESI†).
Catalyst durability and kinetics were also investigated to gain a
greater understanding of the Zn(II) photoreduction mechanism. It
was found that at 300 mM 5,7-dCl-Hq, Zn(II) reduction largely ceases
after 200 h and fits a first order decay model with R2 = 0.96. After
10 days an average of 242 mmol zinc was produced, equivalent to
81 TONs (ESI†). This constitutes roughly 10% of the Zn present in
the vial and the increasing coating of the vessel with light-blocking
Zn metal could be a contributing factor to this decrease in reactivity.
Rate studies were also performed, and over the first 8 h a
dependence of the zinc production rate on the concentrations of
both ZnCl2 and 5,7-dCl-Hq were measured. Similar to our earlier
work with an Ir(III) catalyst, a minimum concentration of ZnCl2 is
required for the reaction to proceed, represented as the x intercept in
Fig. 7. Interestingly, the x intercept roughly corresponds with the
solubility of ZnCl2 in the reaction mixture, which strongly suggests a
heterogeneous reaction pathway. It can be observed that the initial
reaction rates increase with the addition of more ZnCl2 until it
plateaus and then decreases at very high loadings. This decrease in
Notes and references
1 Metal-air batteries overview: (a) K. Harting, U. Kunz and T. Turek,
Z. Phys. Chem., 2012, 226, 151; (b) J.-S. Lee, S. T. Kim, R. Cao,
N.-S. Choi, M. Liu, K. T. Lee and J. Cho, Adv. Energy Mater., 2011,
1, 34; (c) Y. Li, M. Gong, Y. Liang, J. Feng, J.-E. Kim, H. Wang,
G. Hong, B. Zhang and H. Dai, Nat. Commun., 2013, 4, 1805.
2 (a) D. A. Everson, B. A. Jones and D. J. Weix, J. Am. Chem. Soc., 2012,
134, 6146; (b) A. Krasovskiy, C. Duplais and B. H. Lipshutz, J. Am. Chem. Soc.,
2009, 131, 15592; (c) S. Wang, Q. Qian and H. Gong, Org. Lett., 2012, 14, 3352.
3 D. G. Whitten, Acc. Chem. Res., 1980, 13, 83.
ˇ
ˇ ´
´
4 (a) I. A. Ruvarac-Bugarcic, Z. V. Saponjic, S. Zec, T. Rajh and
´
J. M. Nedeljkovic, Chem. Phys. Lett., 2005, 407, 110; (b) A. L. Stroyuk,
V. V. Shvalagin and S. Y. Kuchmii, J. Photochem. Photobiol., A, 2005,
173, 185; (c) H. Kyung, J. Lee and W. Choi, Environ. Sci. Technol., 2005,
´ ´
39, 2376; (d) L. X. Chen, T. Rajh, O. Micic, Z. Wang, D. M. Tiede and
M. Thurnauer, Nucl. Instrum. Methods Phys. Res., Sect. B, 1997, 133, 8.
5 A. C. Brooks, K. Basore and S. Bernhard, Inorg. Chem., 2013, 52, 5794.
´
´
¨
6 (a) N. Donze, P. Pechy, M. Gratzel, M. Schaer and L. Zuppiroli, Chem.
Phys. Lett., 1999, 315, 405; (b) L. S. Sapochak, F. E. Benincasa,
R. S. Schofield, J. L. Baker, K. K. C. Riccio, D. Fogarty,
H. Kohlmann, K. F. Ferris and P. E. Burrows, J. Am. Chem. Soc.,
2002, 124, 6119; (c) Y. Hamada, T. Sano, M. Fujita, T. Fujii, Y. Nishio
and K. Shibata, Jpn. J. Appl. Phys., 1993, 32, L514; (d) R. Pohl and
P. Anzenbacher Jr., Org. Lett., 2003, 5, 2769.
7 (a) J. Yu, X.-J. Yang, Y. Liu, Z. Pu, Q.-S. Li, Y. Xie, H. F. Schaefer and B. Wu,
Organometallics, 2008, 27, 5800; (b) A. Grirrane, I. Resa, A. Rodriguez,
E. Carmona, E. Alvarez, E. Gutierrez-Puebla, A. Monge, A. Galindo, D. del
´
Rıo and R. A. Andersen, J. Am. Chem. Soc., 2007, 129, 693.
8 For a review on Zn–Zn complexes: T. Li, S. Schulz and P. W. Roesky,
Chem. Soc. Rev., 2012, 41, 3759.
5198 | Chem. Commun., 2014, 50, 5196--5199
This journal is ©The Royal Society of Chemistry 2014

View Article Online
Communication
ChemComm
9 E. D. Cline, S. E. Adamson and S. Bernhard, Inorg. Chem., 2008, 12 (a) H. Jang, L.-M. Do, Y. Kim, T. Zyung and Y. Do, Synth. Met., 2001,
47, 10378.
121, 1667; (b) A. Sharma, D. Singh, J. K. Makrandi, M. N. Kamalasanan,
R. Shrivastva and I. Singh, Mater. Lett., 2007, 61, 4614.
10 C. J. Fahrni and T. V. O’Halloran, J. Am. Chem. Soc., 1999, 121, 11448.
11 L. M. A. Monzon, F. Burke and J. M. D. Coey, J. Phys. Chem. C, 2011, 13 Y. L. Chow, W. C. Danen, S. F. Nelsen and D. H. Rosenblatt, Chem.
115, 9182.
Rev., 1978, 78, 243.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 5196--5199 | 5199