Two-electron HCl to H2 photocycle promoted by Ni(II) polypyridyl halide complexes

Article abstract of DOI:10.1021/ja408787k

Photochemical HX splitting requires the management of two protons and the execution of multielectron photoreactions. Herein, we report a photoinduced two-electron reduction of a polypyridyl Ni(II) chloride complex that provides a route to H₂ evolution from HCl. The excited states of Ni complexes are too short to participate directly in HX activation, and hence, the excited state of a photoredox mediator is exploited for the activation of HX at the Ni(II) center. Nanosecond transient absorption (TA) spectroscopy has revealed that the excited state of the polypyridine results in a photoreduced radical that is capable of mediating HX activation by producing a Ni(I) center by halogen-atom abstraction. Disproportionation of the photogenerated Ni(I) intermediate affords Ni(II) and Ni(0) complexes. The Ni(0) center is capable of reacting with HX to produce H₂ and the polypyridyl Ni(II) dichloride, closing the photocycle for H₂ generation from HCl.

Full text of DOI:10.1021/ja408787k

Article
pubs.acs.org/JACS
Two-Electron HCl to H2 Photocycle Promoted by Ni(II) Polypyridyl
Halide Complexes
David C. Powers, Bryce L. Anderson, and Daniel G. Nocera*
Department of Chemistry and Chemical Biology, Harvard University, 12 Oxford Street, Cambridge, Massachusetts 02138, United
States
S
* Supporting Information
ABSTRACT: Photochemical HX splitting requires the management of
two protons and the execution of multielectron photoreactions. Herein,
we report a photoinduced two-electron reduction of a polypyridyl Ni(II)
chloride complex that provides a route to H2 evolution from HCl. The
excited states of Ni complexes are too short to participate directly in HX
activation, and hence, the excited state of a photoredox mediator is
exploited for the activation of HX at the Ni(II) center. Nanosecond
transient absorption (TA) spectroscopy has revealed that the excited state
of the polypyridine results in a photoreduced radical that is capable of mediating HX activation by producing a Ni(I) center by
halogen-atom abstraction. Disproportionation of the photogenerated Ni(I) intermediate affords Ni(II) and Ni(0) complexes.
The Ni(0) center is capable of reacting with HX to produce H2 and the polypyridyl Ni(II) dichloride, closing the photocycle for
H2 generation from HCl.
(L′ = 1,3-bis(2,6-diisopropylphenyl)imidazole-2-ylidene (eq
2)),12 Both Ni(I) complexes 2 and 4 are unreactive toward
HCl, precluding an HX splitting cycle from being achieved.
INTRODUCTION
■
The photodriven splitting of hydrohalic acids (HX) into H2 and
X2 is a promising approach to storing solar energy in the form
of a redox flow battery1 or chemical fuel.2 The conversion of
HX to H2 and X2 stores considerable energy. For the case of X
= Cl, HX splitting stores approximately the same amount of
energy per electron as the conversion of H2O to H2 and 1/2O2
but only requires the management of two electrons and protons
as opposed to the four that are required in water splitting.3
Despite the imperative of energy storage in its various forms,
the molecular details of HX-splitting chemistry has received
little attention. Extant HX-splitting photocatalysts are based on
precious metals, with a particular emphasis on rhodium and
iridium. Dirhodium phosphazane catalysts are especially
effective at promoting the photocatalytic evolution of H2
from HX4,5 owing to the ability of the diphosphazane ligand
to enforce two-electron-mixed valency3 and to accommodate
M(μ2-X)M−X bridging intermediates from which reductive
elimination may occur.6,7 Studies of these complexes as well as
those of Pt and Au reveal that turnover efficiencies are limited
by the strong M−X bond,8−10 which has been proposed to be
activated via ligand-to-metal charge transfer (LMCT) excited
states.6,8,9
The strength of the MX bond may be reduced by moving to
first-row transition metals, which possess the added benefit that
they are noncritical elements. Our previous efforts to achieve
either H2- or X2-forming photoreactions with Ni(II) complexes
establish that the requisite two-electron chemistry of HX
splitting is circumvented by one-electron photoreactions.
Photolysis of Ni(II) hydridohalide complex 1, affords Ni(I)
chloride 2 (L = 1,3-dimesitylimidazol-2-ylidene (eq 1)),11 and
photolysis of Ni2(II,II) complex 3 affords Ni2(I,I) complex 4
Figure 1 presents a strategy for an authentic two-electron H2
photocycle using HX as a substrate. In the case of nickel,
driving reductive elimination from a Ni(III) state is preferred
over a Ni(II) state based on the facility of two-electron
reductive elimination reactions from mononuclear Ni(III)
centers, as observed for Ni-catalyzed cross-coupling reac-
tions.13−18 In principle, a Ni(III)-like state can be accessed
from a metal-to-ligand charge transfer (MLCT) of a Ni(II)
halide complex. However, whereas long-lived MLCT excited
states are frequently encountered in the photochemistry of
second- and third-row transition metal complexes, first-row
transition metal complexes typically possess extremely short-
lived charge transfer states.19−21 The compression of the crystal
field splitting attendant upon moving from a second- to first-
row metal results in the placement of d−d ligand-field states to
Received: August 24, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
1
MHz for H acquisitions. NMR chemical shifts are reported in ppm
with the residual solvent resonance as internal standard. UV−vis
spectra were recorded at 293 K in quartz cuvettes on a Spectral
Instruments 400 series diode array and were blanked against the
appropriate solvent. Solution magnetic moments were determined
using the Evans method46 in 1:1 THF/CH3CN and measured using
19F NMR (hexafluorobenzene added); diamagnetic corrections were
estimated from Pascal constants.47 Perpendicular mode X-band EPR
spectra were recorded on Bruker ElexSys E500 EPR equipped with an
N2 dewer for measurements at 77 K. EPR spectra were referenced to
diphenylpicrylhydrazyl, DPPH (g = 2.0037). Spectra were simulated
using EasySpin 4.5.0 implemented with MATLAB to obtain g-values.
EPR spectra of [7]OTf were collected in both 2-methyltetrahydrofur-
an (2-MeTHF) and 9:1 EtOH/MeOH, and the observed features
showed no solvent dependence. Steady-state emission spectra were
recorded on an automated Photon Technology International (PTI)
QM 4 fluorimeter equipped with a 150 W Xe arc lamp and a
Hamamatsu R928 photomultiplier tube.
Figure 1. Catalytic cycle for HX splitting. Two-electron photo-
reduction of a metal dihalide generates an intermediate poised to
thermally reduce protons.
Steady-state photochemical reactions were performed using a 1000
W high-pressure Hg/Xe arc lamp (Oriel), and the beam was passed
through a water-jacketed filter holder containing the appropriate long-
pass filter, an iris, and a collimating lens. Nanosecond transient
absorption (TA) measurements were made at 10 Hz with the pump
light provided by the third harmonic (355 nm) of a Quanta-Ray
Nd:YAG laser (Spectra-Physics) or with 310 nm light generated from
the frequency doubled 620 nm signal of a Spectra-Physics Quanta-Ray
MOPO-700 with FDO-970 option pumped with the 355 nm light
from the aforementioned laser. The signal light passed through a Triax
320 spectrometer, where it was dispersed by a 300 groove/mm × 250
nm blazed grating and collected with either an intensified gated CCD
camera (ICCD, CCD 30-11, Andor Technology, 1024 × 256 pixels, 26
μm2) for TA spectra or a photomultiplier tube (PMT) for TA single-
wavelength kinetics. PMT outputs were collected and averaged with a
1 GHz oscilloscope (LeCroy 9384CM). A TTL pulse synchronized
with the Q-switch of the Infinity laser was delayed 99 ms before
triggering the shutter for the probe light. Electronic delays were
created with SRS DG535 delay generators (Stanford Research
Systems). These delay boxes, in combination with electronic shutters
(Uniblitz), were used to create the necessary pulse sequence. THF/
CH3CN (1:1) solutions of complexes 5 and bc were prepared in 50
mL Schlenk flasks in an N2-filled glovebox. Solutions were flowed
through a 3 mm diameter, 1 cm path length flow cell (Starna, type
585.2) using a peristaltic pump and positive argon pressure. Details of
picosecond TA spectroscopy can be found in the Supporting
Information.
lower energy than charge-separated states.22,23 The conse-
quence of this energetic reordering of excited states can be seen
by comparing the lifetimes of charge-separated states of Ru and
Fe polypyridyl complexes: Ru2+ polypyridyl complexes display
charge-separated state lifetimes of 600−850 ns,24−31 whereas
the related Fe2+ polypyridyl complexes display charge separated
lifetimes on the order of ∼100 fs.22,23
In order to overcome the inherently short excited state
lifetimes of nickel complexes, we targeted a design strategy in
which photochemistry is derived from a photoredox mediator.
Pyridine has received substantial attention as a potential small-
molecule redox mediator because of the observation that single-
electron transfer reactions can interconvert pyridinium ions and
pyridinyl radicals.32 Electrochemical and photoelectrochemical
reduction reactions of CO2 to methanol have been proposed to
proceed through pyridinyl radical intermediates, which engage
in proton-coupled electron transfer (PCET) reactions to
accomplish substrate reduction.33−35 The role of the electrode
surface and the precise pyridine oxidation state are topics of
current discussion.36−38 Using these results as a guidepost, we
turned our attention to bipyridine excited states, which are
known to engage in H-atom abstraction (HAA) reactions,39−42
to access a reducing pyridinyl radical for substrate reduction.
We now report a HX-to-H2 photocycle promoted by a
Ni(II)/Ni(0) couple. Reduction of a Ni(II) dihalide complex
by a polypyridyl photoredox mediator provides either Ni(I) or
Ni(0) complexes dependent on the particular reaction
conditions employed. The photogenerated Ni(0) species reacts
thermally with HCl to generate H2 and regenerate the Ni(II)
dihalide, demonstrating a closed photocycle for the evolution of
H2 from HCl mediated by an earth-abundant transition metal
complex.
Preparation of Ni(II) Complexes (LNiCl2). Complex 5. To a
suspension of NiCl2(dme) (181 mg, 8.23 × 10−4 mol, 1.00 equiv) in
THF was added bc (297 mg, 8.23 × 10−4 mol, 1.00 equiv) as a solid.
The mixture was stirred at 23 °C for 10 h, during which time a pink
precipitate was observed. The mixture was concentrated to dryness
and the residue was washed with THF and dried in vacuo to afford 403
1
mg of the title complex as a pink solid (96% yield). H NMR (500
MHz, THF-d8) δ (ppm): 80.05 (s, 2H), 28.12 (s, 2H), 18.5 (br s, 6H),
9.61 (d, J = 7.3 Hz, 4H), 8.81 (t, J = 7.3 Hz, 2H), 7.19 (dd, J = 7.3 Hz,
J = 7.3 Hz, 4H). μeff = 3.59 μB. Crystals suitable for single-crystal
diffraction analysis were obtained by slow evaporation of a
concentrated MeOH solution of the complex.
EXPERIMENTAL SECTION
■
1
Complex 9 was prepared analogously: 97% yield. H NMR (500
General Considerations. All reactions were carried out in an N2-
filled glovebox. Anhydrous solvents were obtained by filtration through
drying columns.43 NiCl2dme (dme = 1,2-dimethoxyethane) and
Ni(cod)2 (cod = 1,5-cyclooctadiene) were obtained from Strem
Chemicals, while 2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline
(bathocuproine, bc), 2,9-dimethyl-1,10-phenanthroline (dmphen),
6,6′-dimethyl-2,2′-bipyridine (dmbpy), and 2,2′-biquinoline (biq)
were obtained from Sigma Aldrich. All chemicals were used without
purification. NiCl2dmphen (8),44 NiCl2biq (10),44 and Ni(dmphen)2
(13)45 were prepared according to literature procedures.
MHz, THF-d8) δ (ppm): 77.3 (br s, 2H), 63.41 (s, 2H), 27.7 (br s,
6H), 23.07 (s, 2H). μeff = 3.57 μB. Crystals suitable for single-crystal
diffraction analysis were obtained by slow evaporation of a MeOH
solution of the complex.
Preparation of Ni(0) Complexes (L2Ni). Complex 6. To a
suspension of Ni(cod)2 (70.0 mg, 2.54 × 10−4 mol, 1.00 equiv) in
THF was added bc (93.8 mg, 5.09 × 10−4 mol, 2.00 equiv) as a solid.
The reaction mixture was warmed to 60 °C, at which temperature it
was stirred for 12 h. The reaction mixture turned from a pale yellow
suspension to an inky purple solution during the course of heating.
The reaction mixture was filtered through Celite. The Celite was
washed with THF, the filtrate was combined, and solvent was removed
Physical Methods. NMR spectra were recorded at the Harvard
University Department of Chemistry and Chemical Biology NMR
facility on a Varian Unity/Inova 500 spectrometer operating at 500
B
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 1. Crystal Data and Structure Refinement
5
6·THF
[7]OTf·3THF
[11]PF6
[12]ClO4·bc
14·0.5PhH
15·THF
formula
fw, g/mol
temp, K
cryst system
space group
color
C26H20Cl2N2Ni
490.06
C56H48N4NiO
851.70
C65H64F3N4NiO6S
1144.98
110(2)
C28H24F6N4NiP
620.19
C54H36ClN6NiO4
927.04
C39H27N4Ni
610.35
C22H21ClN2O
364.86
100(2)
monoclinic
P21/c
110(2)
monoclinic
P21/n
100(2)
100(2)
monoclinic
P2/n
100(2)
100(2)
monoclinic
P21/c
triclinic
triclinic
triclinic
P1
̅
P1
P1
̅
green
̅
pink
purple
blue
red-black
10.590(3)
10.945(3)
11.849(3)
76.804(4)
77.027(4)
77.385(4)
1282.6(6)
2
purple
blue-green
13.4506(6)
14.7269(6)
15.8071(7)
88.800(1)
67.020(1)
88.659(1)
2881.6(2)
6
a, Å
13.0496(9)
21.9451(15)
7.8238 (6)
90
18.007(3)
11.3520(17)
21.176(3)
90
17.4078(4)
11.0915(2)
29.2364(6)
90
13.6604(9)
11.1003(7)
13.7705(9)
90
8.6703(15)
13.872(2)
15.823(3)
77.753(3)
81.643(3)
75.191(3)
1789.7(5)
2
b, Å
c, Å
α, deg
β, deg
γ, deg
104.450(1)
90
96.196(3)
90
90.371(1)
90
98.787(1)
90
V, A3
2169.7(3)
4
4303.3(11)
4
5644.8(2)
6
2063.4(2)
4
Z
a
R1
0.039
0.057
0.052
0.085
0.033
0.039
0.071
b
wR2
0.098
0.170
0.127
0.277
0.084
0.084
0.187
c
GOF (F2)
0.87
1.07
1.05
1.05
1.05
1.02
1.05
a
b
c
2
R1 = ∑||Fo − |Fc||/∑|Fo|. wR2 = (∑(w(Fo2 − Fc2)2)/∑(w(Fo )2))1/2. GOF = (∑w(Fo2 − Fc2)2/(n − p))1/2 where n is the number of data and p
is the number of parameters refined.
Figure 2. Thermal ellipsoid plots of [11]PF6, [12]ClO4, 14, and 15. Except for compound 15, hydrogen atoms and solvent molecules have been
omitted for clarity. Ellipsoids are drawn at the 50% probability level. Thermal ellipsoid plots of 5, 6, and [7]OTf appear in Figure 3
in vacuo to afford 184 mg of 6 as an air-sensitive dark purple solid
(93% yield). 1H NMR (500 MHz, C6D6) δ (ppm): 7.95 (s, 4H), 7.85
(s, 4H), 7.65 (d, J = 7.3 Hz, 8H), 7.36 (t, J = 7.3 Hz, 4H), 7.18 (dd, J =
7.8 Hz, J = 7.8 Hz, 8H), 2.48 (s, 12H). Crystals suitable for single-
crystal diffraction analysis were obtained from a THF solution of the
complex layered with PhCH3.
2.2004; g3 = 2.1626. Crystals suitable for single-crystal diffraction
analysis were obtained from a MeOH solution of the complex.
Complex [12]ClO4: 74% yield. 1H NMR (500 MHz, CD3CN) δ
(ppm): 20.4 (br s, 4H), 19.7−19.4 (m, 12H), 10.7 (br s, 4H), 8.0 (s,
4H). Crystals suitable for single-crystal diffraction analysis were
obtained from a MeOH solution of the complex.
1
Complex 14 was prepared analogously: 81% yield. H NMR (500
Preparation of K·biq. To a solution of biq (250 mg, 9.72 × 10−4
mol, 1.00 equiv) in THF was added potassium (38.0 mg, 9.72 × 10−4
mol, 1.00 equiv) as a solid. The reaction mixture was stirred at 23 °C
for 4 h, during which time the reaction mixture assumed a dark red-
brown color. The reaction mixture was filtered through Celite and the
filtrate was concentrated in vacuo to afford 55.1 mg of the title
complex as an air-sensitive dark red-brown solid (94% yield).
Treatment of K·biq with HCl·dioxane in THF results in immediate
color change of the reaction solutions from red-brown to green.
Removal of solvent in vacuo affords 15 as a green solid (71%). Crystals
suitable for single-crystal diffraction analysis were obtained from a
THF solution of the compound.
MHz, benzene-d6) δ (ppm): 10.22 (d, J = 8.3 Hz, 4H), 9.19 (d, J = 8.8
Hz, 4H), 7.56 (d, J = 7.8 Hz, 4H), 7.29 (t, J = 7.3 Hz, 4H), 7.15
(overlap with solvent signal, 4H), 6.80 (d, J = 8.8 Hz, 4H). Crystals
suitable for single-crystal diffraction analysis were obtained from a
THF solution of the complex layered with PhCH3.
Preparation of Ni(I) Complexes (L2Ni+X−). Complex [7]OTf. To
a solution of Ni(bc)2 (6) (59.0 mg, 7.57 × 10−5 mol, 1.00 equiv) in
THF was added AgOTf (19.5 mg, 7.57 × 10−5 mol, 1.00 equiv) as a
suspension in THF. The reaction solution immediately turned from
dark purple to blue. The reaction mixture was stirred at 23 °C for 30
min before being filtered through Celite. The filtrate was concentrated
in vacuo and the residue was taken up in 1:1 THF/pentane. Complex
[7]OTf was isolated by filtration to afford 55.1 mg of an air-sensitive
X-ray Crystallographic Details. All structures were collected on a
Bruker three-circle platform goniometer equipped with an Apex II
CCD and an Oxford cryostream cooling device at 100 K. Radiation
was generated from a graphite fine focus sealed tube Mo Kα (0.710 73
Å) source. Crystals were mounted on a cryoloop or glass fiber pin
using Paratone N oil. Data were integrated using SAINT and scaled
with either a numerical or multiscan absorption correction using
SADABS. The structures were solved by direct methods using
SHELXS-97 and refined against F2 on all data by full matrix least-
squares with SHELXL-97. All non-hydrogen atoms were refined
1
dark blue solid (80% yield). H NMR (500 MHz, THF-d8) δ (ppm):
33.8 (br s, 4H), 12.51 (s, 4H), 8.00−7.96 (m, 12H), 7.07 (s, 8H), 0.8
(br s, 12H). EPR (9:1 EtOH/MeOH) g-value: g1 = 2.4524; g2
2.2039; g3 = 2.1655. μeff = 2.46 μB.
=
Complexes [11]PF6 and [12]ClO4 were prepared analogously.
Complex [11]PF6: 82% yield. 1H NMR (500 MHz, CD3CN) δ
(ppm): 33.0 (br s, 4H), 13.3 (s, 4H), 11.6 (br s, 4H), 0.4 (br s, 12H).
μeff = 2.53 μB. EPR (9:1 EtOH:MeOH) g-value: g1 = 2.4616; g2 =
C
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. Photolysis of Ni(II) complex 5 in THF afforded Ni(0) complex 6, while photolysis of 5 in CH3CN/1,4-cyclohexadiene afforded Ni(I)
complex [7]Cl. Thermal ellipsoid plots of 5, 6, and [7]OTf are shown. Ellipsoids are drawn at the 50% probability level.
anisotropically. Hydrogen atoms were placed at idealized positions and
refined using a riding model. Crystal data and refinement statistics are
summarized in Table 1, and thermal ellipsoid plots are collected in
Figures 2 and 3.
RESULTS AND DISCUSSION
■
Synthesis and Steady-State Photochemistry. Treat-
ment of NiCl2dme with bathocuproine (bc) afforded pink
Ni(II) dichloride complex 5, which is pseudo-tetrahedral in the
solid state and in solution. THF solutions of 5 exhibit a UV−vis
absorption spectrum that features d−d absorptions at 500, 852,
and 1005 nm (ε = 200, 70, and 100 M−1 cm−1, respectively).
Referenced to Td symmetry, the d−d transitions arise from 3T1
48,49
3
3
to A2 and T2 excitations;
the deviation of 5 from Td
symmetry gives rise to multiple transitions for each of these
absorption manifolds.50 The room-temperature magnetic
moment, measured by the Evans method, was determined to
be μeff = 3.6.51 Photolysis (λ > 305 nm) of 5 in THF in the
presence of both 2.0 equiv of 2,6-lutidine and 2.0 equiv of
exogenous bc caused the pale pink reaction solution to turn
dark purple (Figure 4a). The dark purple product was identified
1
as the Ni(0) complex 6 by comparison of the UV−vis and H
NMR spectra of the photochemical reaction mixture to a
spectrum obtained for an authentic sample of 6, which was
prepared by treatment of Ni(cod)2 with 2.0 equiv of bc. The
central C−C bond of the phenanthroline backbone (C2−C2′ in
bipyridine numbering) contracts from 1.441(4) Å in 5 to
1.411(5) Å in 6. For comparison, the corresponding bond in
free bc is 1.46 Å.52 Both the dark purple color and the C−C
bond contraction53 in 6 are consistent with substantial redox
noninnocence of the bc ligands in 6, suggesting 6 may be
described as a Ni(II) complex with two pendent ligand-based
radicals.
Figure 4. Spectral evolution during photolysis (λ > 305 nm) of (a)
Ni(II) complex 5 to Ni(0) complex 6 in THF and (b) Ni(II) complex
5 to Ni(I) complex [7]Cl in 1:1 CH3CN/1,4-cyclohexadiene.
independently prepared [7]OTf, which was obtained by
1
treatment of Ni(0) complex 6 with AgOTf. H NMR, EPR,
In contrast to the photolysis of 5 in THF, when Ni(II)
complex 5 is photolyzed in CH3CN in the presence of
exogenous bc and 1,4-cyclohexadiene added as an H-atom
donor (vide infra), a dark blue solution is obtained (Figure 4b).
The identity of the blue photoproduct was established to be
homoleptic Ni(I) complex [7]Cl by comparison of the
photochemical reaction mixture with the spectral data of
and electronic absorption spectroscopies were used to assign
the structure of [7]OTf. X-ray crystallographic analysis of a
single crystal established the structure of [7]OTf to be the
homoleptic Ni(I) structure drawn in Figure 3. The central C−
C bond of [7]OTf was measured to be 1.429(4) Å, which is
between the distances in Ni(0) complex 6 (C−C = 1.411(5) Å)
and Ni(II) complex 5 (C−C = 1.441(4) Å).
D
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
The disparate, solvent-dependent photoreactivity of 5, which
undergoes either one- or two-electron photoreduction depend-
ing on the reaction solvent, was probed by examining the
solvent-dependent behavior of the isolated Ni(I) complexes.
The results of these experiments are summarized in Figure 5.
(Figure S25). The effect of ligand structure on photoreduction
efficiency was probed by comparison of the quantum yields for
one-electron photoreduction of complexes 5, 8, and 9. Whereas
complexes 5, 8, and 9 are isostructural in solution, established
by both solution magnetic moment (5, μeff = 3.6; 8, μeff = 3.5; 9,
μeff = 3.6) and similarity of d−d transitions (Figure S18), their
photochemical quantum yields are 2.14% (5), 0.92% (8), and
0% (9). This trend follows the expected ease of reduction of
the ancillary polypyridyl ligand; while electrochemical reduction
of bipyridine derivatives is frequently irreversible, monoproto-
nated bipyridines display the anticipated substituent-dependent
redox potentials.54,55 The quantum yield for photoreduction of
biquinoline-supported Ni(II) complex 10 could not be
determined because of insolubility of 10 under photoreduction
conditions.
Thermally Promoted Proton Reduction Reactions.
With conditions in hand to achieve the two-electron photo-
reduction of Ni(II) chlorides, we examined the viability of
proton reduction by the reduced Ni complexes. Treatment of
Ni(0) complexes 6 and 13 with HCl in MeOH generates H2 in
60 and 85% yield, respectively, along with Ni(II) complex 5 and
8; the results of these experiments are summarized in Figure 7.
Figure 5. Solvent-dependent equilibrium favors Ni(I) complex 7+ in
CH3CN, while Ni(II) and Ni(0) complexes 5 and 6 are favored in
THF.
Treatment of Ni(0) complex 6 with AgOTf affords Ni(I)
complex [7]OTf, which is stable in both THF and CH3CN.
Treatment of a THF solution of [7]OTf with 1.0 equiv of n-
Bu4NCl prompted immediate disproportionation of 7+ to
Ni(II) complex 5 and Ni(0) complex 6, with concurrent
liberation of 0.5 equiv of bc (see Figure S9 in Supporting
1
Information for H NMR). In contrast, [7]Cl is stable with
respect to disproportionation in CH3CN; treatment of [7]OTf
with 1.0 equiv of n-BuN4Cl in CH3CN did not result in the
observation of either 5 or 6 (Figure S10). The observation of
solvent-dependent disproportionation of 7+ in the presence of
Cl− clarifies the observation of 6 from photolysis of 5 in THF
and the observation of 7+ from photolysis of 5 in CH3CN.
Solvent-dependent, chloride-induced disproportionation is also
observed in photochemical reactions; concentration of the blue
photoreaction mixture of Ni(I) complex [7]Cl produced by
photolysis in CH3CN and 1,4-cyclohexadiene and redissolution
in THF affords a purple solution of 5 and 6.
The effect of bis-imine ligand on the efficiency of
photoreduction was probed using NiCl2dmphen (8),
NiCl2dmbpy (9), and NiCl2biq (10) (Figure 6). Both
Figure 7. (a) Protonation of Ni(0) complex 6 affords Ni(II) chloride
5, as well as H2. (b) Protonation of Ni(0) complex 14 affords Ni(II)
chloride 12 together with pyridinyl radical 15 but no H2.
The H2 yields were determined by GC analysis and confirmed
by independent Toepler pump combustion analysis (Figure
S36). Formation of 5 and 8 following H2 evolving reactions of
6 and 13, respectively, establishes a closed H2-evolving
photocycle using an earth-abundant transition metal complex.
While 5 is thermally stable to HCl, photolysis of 5 in the
presence of protons leads to the loss of bc and <0.2 turnover of
H2 production.
The ligand requirements for a complex that can mediate both
oxidative and reductive half reactions of HX splitting are strict.
A closed photocycle cannot be achieved for 2,2′-biquinoline
complex 10. Whereas protonation of Ni(biq)2 (14) affords
Ni(II) complex 10, no H2 is observed. Instead of generating H2,
the reducing equivalents incipient in Ni(0) complex 14
promote the reduction of biq to afford pyridinium-stabilized
pyridinyl radical 15. Radical 15 could also be synthesized by
sequential treatment of biq with K0 and HCl and features a
d(C2−C2)′ of 1.435(6) Å, diagnostic of a monoreduced
biquinoline56 (compared to d(C2−C2)′ in 2,2′-biquinoline of
1.492(3) Å57). Neither thermolysis or photolysis of 15 liberates
H2. Similar ligand reduction chemistry has not been observed in
Figure 6. Quantum yield for photoreduction of Ni(II) complexes is
correlated with reduction potential of the bis-imine ligand. Insolubility
of 10 prevented determination of quantum yield.
NiCl2dmphen (8) and NiCl2biq (10) participate in photo-
reduction chemistry in analogy to complex 5 (Figures S22−
S24). In addition, isolated mononuclear Ni(I) complexes
Ni(dmphen)2PF6 ([11]PF6) and Ni(biq)2PF6 ([12]PF6)
exhibit chloride-induced disproportionation in THF to afford
Ni(dmphen)2 13 and 8 or Ni(biq)2 (14) and 10, respectively.
In contrast to complexes 5 and 8, regardless of the wavelength
of irradiation employed, NiCl2dmbpy (9) is not photoreduced
E
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
the protonation of 6 but may account for the nonquantitative
yield of H2 observed. Although photoreduction of Ni(II)
complexes supported by bipyridine ligands is promoted by
more reducing ligands, the extent of proton reduction is limited
by deleterious ligand reduction processes.
Time-Resolved Photochemical Experiments. In order
to probe our contention that photogenerated pyridinyl radicals
could mediate the requisite photoreduction chemistry, a
nanosecond transient absorption (TA) spectrum of a solution
of bc was recorded. The collected TA spectrum featured a
growth centered at 526 nm (Figure 8a). We assign the transient
vs [5] is linear, consistent with Stern−Volmer kinetics and with
a reaction between transient radical 16 and 5 (Figure 8b, inset).
The calculated bimolecular rate constant of 3.0 × 109 M−1 s−1 is
near the diffusion limit.60
The dynamics of the photochemistry of bc were probed with
ultrafast TA spectroscopy, which revealed two kinetic domains
in the evolution of 16. An initially formed signal, assigned as
1bc* (λmax = 565 nm) generated by n → π* excitation, evolves
without well-defined isosbestic points during the first 600 ps
following the laser pulse. On the basis of analogy to ultrafast
studies of 4,4′-bipyridine, we propose that this initial period
1
involves both HAA from bc* to afford 16 and intersystem
crossing to generate 3bc* (Figure S35a).39 Between 600 ps and
3 ns after the laser pulse, the observed TA spectrum evolves
with tightly anchored isosbestic points to generate 16 (Figure
S35b). Excited state dynamics are finished, and pyridinyl radical
16 is formed within 3 ns of the laser pulse. In contrast to the
relatively slow excited-state dynamics observed for bc, ultrafast
TA spectra obtained by flash laser photolysis of Ni(II) complex
5 confirm the expectation of fast excited-state dynamics for
first-row transition metal polypyridyls. A TA feature as 560 nm
is observed for complex 5 and decays to baseline within 15 ps
of the laser pulse (Figure 9).
Figure 9. Ultrafast dynamics of laser flash photolysis of Ni(II) complex
5.
Nanosecond TA spectra were also obtained for dmphen and
dmbpy. Flash laser photolysis of dmbpy revealed a feature
centered at 370 nm (Figure S31). Unlike the TA spectrum of
bc, the intensity of the observed feature in the TA spectrum of
dmbpy did not display significant solvent dependence between
THF and hexane. This observation, coupled with the similarity
of the observed TA signal to a reported TA spectrum of the
long-lived triplet excited state of 2,2′-bipyridine,61 suggests that
the observed TA spectrum arises from 3dmbpy*, not a pyridinyl
radical. The lifetime of 3dmbpy* was quenched in the presence
of Ni(II) complex 9, but no photoreduction was observed,
demonstrating that triplet−triplet annihilation does not lead to
reduction of Ni(II). The TA spectrum of dmphen displays a
feature centered at 440 nm (Figure S30), which correlates well
to a reported spectrum of phen*. Unlike the TA spectrum
obtained by flash laser photolysis of dmbpy, the intensity of the
feature at 440 nm does display solvent dependent intensity; the
intensity in hexane is ∼75% the intensity in THF. We do not
observe a feature attributable to the monoreduced ligand, but
given the observed difference in the intensity and the
observation that the triplet excited state of bipyridine is
incompetent to accomplish reduction of Ni(II), we propose
that the observed photoreduction of Ni(II) complex 8 is
Figure 8. (a) Transient absorption spectrum (red) obtained by laser
flash photolysis (355 nm pump) of a 1:1 THF/CH3CN solution of bc,
recorded at 40 ns delay: normalized absorption (blue) and emission
(black) spectra of bc. (b) Single wavelength (526 nm) kinetic decay of
transient intermediate. Inset: plot of τo/τ vs [5].
species as ligand radical 16, based on the TA spectra obtained
upon flash photolysis of bipyridines, which also are ascribed to
pyridinyl radicals generated by hydrogen-atom abstraction
(HAA) reactions from solvent by electronically excited
bipyridines.39−42 HAA abstraction from THF generates an
equivalent of furanyl radical, which can participate in a variety
of subsequent transformations.58,59 The intensity of the TA
signal at 526 nm is stronger in THF than in CH3CN, and no
TA signal is observed in hexane (Figure S27), which is
consistent with the relative H-atom donor capacities of these
solvents. Monitoring the decay of 16 at 526 nm provided a
lifetime of 21.66 0.43 μs (Figure 8b). The lifetime of 16 was
diminished in the presence of Ni(II) complex 5; a plot of τo/τ
62
3
F
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
mediated by pyridinyl radicals. Consistent with the observations
regarding solvent-dependent generation of pyridinyl radical
intermediates, steady-state photolysis of 5 in either CH3CN or
hexanes (poor H-atom donors) resulted in no photoreduction.
Because the reducing pyridinyl radicals are ultimately derived
from H-atom abstraction from solvent, the resultant photocycle
produces equivalent solvent oxidation byproducts.
Attempts to isolate the product of one-electron reduction of
phenanthroline derivativeseither bc or dmphendid not
afford isolable products. Treatment of biq with K0 provided
access to a biquinoline radical anion. The viability of Ni(II)
reduction mediated by ligand radicals was confirmed by
treatment of Ni(II) complex 10 with K·biq, which afforded
Ni(0) complex 14 (Figure 10).
Figure 11. H2-evolution cycle based on bipyridine photoredox
chemistry to generate H2 from HCl.
extremely short; ultrafast measurements revealed an excited
state that decayed to baseline in 15 ps. Nevertheless, laser flash
photolysis of bc revealed the presence of a long-lived
photogenerated intermediate, assigned to pyridinyl radical 16
based on solvent-dependent yields of 16 as well as on
comparison of the TA spectrum of bc to previously reported
TA spectra of similar bipyridines. Monitoring the lifetime of
pyridinyl radical 16 as a function of [5] revealed that the
lifetime of radical 16 is quenched by Ni(II) complex 5. We have
directly verified the viability of free ligand radicals to mediate
Ni-centered reduction by treatment of Ni(II) complex 10 with
K·biq, which results in the observation of Ni(0) complex 14. A
closed photocycle for H2 evolution from HCl is thus
established by protonation of photogenerated Ni(0) complex
6 with HCl, which generates 5 and H2. These results presented
herein thus demonstrate the feasibility of utilizing first-row
transition metal complexes to couple two-electron photo-
reduction reactions and H2-evolving proton reduction reac-
tions.
Figure 10. Treatment of Ni(II) complex 10 with stable ligand−radical
K·biq affords Ni(0) complex 14.
CONCLUDING REMARKS
■
Photocycles for H2 evolution from HCl require management of
multiproton coupled electron transfer reaction (multi-PCET)
events. HX-splitting photocatalysis has been achieved by taking
advantage of the propensity of second- and third-row transition
metals to participate in two-electron reactions. Similar HX-
splitting photocycles have not been developed with first-row
transition metal complexes. In addition to the facility of one-
electron processes at first-row transition metal complexes,
extremely short charge-separated state lifetimes are a challenge
that needs to be addressed in order to achieve controlled multi-
PCET photoreduction reactions. To address the problems of
one-electron redox transformations and short charge-separated
state lifetimes, we have targeted using photoredox mediators
such that the excited state lifetime of interest can be
independently tuned, as well as complexes in which one-
electron photoreduction is coupled to a disproportionation
reaction to enforce the requisite multielectron redox chemistry.
On the basis of the demonstrated ability of pyridine
derivatives to serve as one-electron shuttles in proton and
CO2 reduction cycles and the expectation that bipyridine
excited states could participate in HAA reactions to generate
the requisite reducing pyridinyl radicals, we targeted bipyridines
as photoredox mediators for the reduction of Ni(II) complexes.
Following this stratagem, we have defined the photocycle
shown in Figure 11 for photogeneration of H2 from HCl
mediated by earth-abundant Ni complexes. Ni(II) complex 5,
supported by a bathocuproine ligand, undergoes solvent-
dependent photoreduction to either Ni(I) complexes (7+) or
Ni(0) complexes (6). Selectivity between one- and two-
electron photoreduction was determined to arise from solvent-
dependent chloride-ion-induced disproportionation. The pro-
posed role of bipyridine as a photoredox mediator was probed
via a series of transient absorption (TA) experiments. The
excited state lifetime of Ni(II) complex 5 was found to be
ASSOCIATED CONTENT
* Supporting Information
■
S
Detailed experimental procedures, spectroscopic data for all
new compounds, and X-ray crystal data for complex [11]PF6;
crystallographic information in txt format. This material is
available free of charge via the Internet at http://pubs.acs.org.
X-ray data have been deposited in the Cambridge Crystal
Structure Database as entries 933283 (5), 933284 (6), 933285
(9), 933286 ([11]PF6), 933287 ([11]ClO4), 936146 ([12]-
ClO4), 936147 (14), and 936149 (15).
AUTHOR INFORMATION
Corresponding Author
dnocera@fas.harvard.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We gratefully acknowledge Matt B. Chambers, Thomas S.
Teets, and Andrew M. Ullman for helpful discussions, Shao-
Liang Zheng and Tamara M. Powers for X-ray crystallography
assistance, and the NSF for funding (Grant CHE-1332783).
G
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(37) Keith, J. A.; Carter, E. A. Chem. Sci. 2013, 4, 1490−1496.
D.C.P. is supported by a Ruth L. Kirchenstein National
Research Service award (Grant F32GM103211).
(38) Lim, C.-H.; Holder, A. M.; Musgrave, C. B. J. Am. Chem. Soc.
2013, 135, 142−154.
(39) Buntinx, G.; Naskrecki, R.; Poizat, O. J. Phys. Chem. 1996, 100,
19380−19388.
REFERENCES
■
(1) Yang, Z.; Zhang, J.; Kintner-Meyer, M. C. W.; Lu, X.; Choi, D.;
Lemmon, J. P.; Liu, J. Chem. Rev. 2011, 111, 3577−3613.
(2) Esswein, A. J.; Nocera, D. G. Chem. Rev. 2007, 107, 4022−4047.
(3) Nocera, D. G. Inorg. Chem. 2009, 48, 10001−10017.
(4) Heyduk, A. F.; Nocera, D. G. Science 2001, 293, 1639−1641.
(5) Elgrishi, N.; Teets, T. S.; Chambers, M. B.; Nocera, D. G. Chem.
Commun. 2012, 48, 9474−9476.
(40) Poizat, O.; Buntinx, G.; Valat, P.; Wintgens, V.; Bridoux, M. J.
Phys. Chem. 1993, 97, 5905−5910.
(41) Enomoto, K.; LaVerne, J. A. J. Phys. Chem. A 2008, 112, 12430−
12436.
̀
(42) Castella-Ventura, M.; Kassab, E.; Buntinx, G.; Poizat, O. Phys.
Chem. Chem. Phys. 2000, 2, 4682−4689.
(43) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518−1520.
(44) Butcher, R. J.; Sinn, E. Inorg. Chem. 1977, 16, 2334−2343.
(6) Powers, D. C.; Chambers, M. B.; Teets, T. S.; Elgrishi, N.;
Anderson, B. L.; Nocera, D. G. Chem. Sci. 2013, 4, 2880−2885.
(7) Gray, T. G.; Veige, A. S.; Nocera, D. G. J. Am. Chem. Soc. 2004,
126, 9760−9768.
(45) Fischer, R.; Langer, J.; Malassa, A.; Walther, D.; Gorls, H.;
̈
Vaughan, G. Chem. Commun. 2006, 2510−2512.
(8) Cook, T. R.; Surendranath, Y.; Nocera, D. G. J. Am. Chem. Soc.
2009, 131, 28−29.
(46) Evans, D. F. J. Chem. Soc. 1959, 2003−2005.
(47) Bain, G. A.; Berry, J. F. J. Chem. Educ. 2008, 85, 532−536.
(48) Benson, E. E.; Rheingold, A. L.; Kubiak, C. P. Inorg. Chem. 2010,
49, 1458−1464.
(9) Cook, T. R.; Esswein, A. J.; Nocera, D. G. J. Am. Chem. Soc. 2007,
129, 10094−10095.
(10) Teets, T. S.; Nocera, D. G. J. Am. Chem. Soc. 2009, 131, 7411−
(49) Donoghue, J. T.; Drago, R. S. Inorg. Chem. 1962, 1, 866−872.
(50) Drago, R. S. Phyiscal Methods for Chemists, 2nd ed.; Surfside
Scientific Publishers: Gainesville, FL, 1992; p 422.
(51) Cotton, F. A.; Wilkinson, G.; Murillo, C. A.; Bochmann, M.
Advanced Inorganic Chemistry, 6th ed.; Wiley: New York, 1999; p 840.
(52) Wang, J.; Ye, J.-W.; Wang, Y. Acta Crystallogr. E 2007, 63,
O2007−O2008.
7420.
(11) Lee, C. H.; Cook, T. R.; Nocera, D. G. Inorg. Chem. 2011, 50,
714−716.
(12) Lee, C. H.; Lutterman, D. A.; Nocera, D. G. Dalton Trans. 2013,
42, 2355−2357.
(13) Coronas, J. M.; Muller, G.; Rocamora, M. J. Organomet. Chem.
1986, 301, 227−236.
(53) Scarborough, C. C.; Sproules, S.; Weyhermuller, T.; DeBeer, S.;
̈
(14) Ceder, R. M.; Granell, J.; Muller, G.; Font-Bardía, M.; Solans, X.
Organometallics 1986, 15, 4618−4624.
Wieghardt, K. Inorg. Chem. 2011, 50, 12446−12462.
(54) Tokel-Takvoryan, N. E.; Hemingway, R. E.; Bard, A. J. J. Am.
Chem. Soc. 1973, 95, 6582−6589.
(15) Higgs, A. T.; Zinn, P. J.; Simmons, S. J.; Sanford, M. S.
Organometallics 2009, 28, 6142−6144.
(55) Krishnan, C. V.; Creutz, C.; Schwarz, H. A.; Sutin, N. J. Am.
Chem. Soc. 1983, 105, 5617−5623.
(16) Tsou, T. T.; Kochi, J. K. J. Am. Chem. Soc. 1979, 101, 7547−
7560.
(56) Piglosiewicz, I. M.; Beckhaus, R.; Wittstock, G.; Saak, W.; Haase,
D. Inorg. Chem. 2007, 46, 7610−7620.
(17) Matsunaga, P. T.; Hillhouse, G. L.; Rheingold, A. L. J. Am.
Chem. Soc. 1993, 115, 2075−2077.
(57) Folting, K.; Merritt, L. L., Jr. Acta Crystallogr. 1977, B33, 3540−
(18) Koo, K.; Hillhouse, G. L. Organometallics 1995, 14, 4421−4423.
(19) McCusker, J. K. Acc. Chem. Res. 2003, 36, 876−887.
(20) Juban, E. A.; Smeigh, A. L.; Monat, J. E.; McCusker, J. K. Coord.
Chem. Rev. 2006, 250, 1783−1791.
3542.
(58) Wallace, T. J.; Gritter, R. J. J. Org. Chem. 1962, 27, 3067−3071.
(59) Mignani, S.; Beaujean, M.; Janousek, Z.; Merenyi, R.; Viehe, H.
G. Tetrahedron 1981, 37, 111−115.
(60) Turro, N. J. Modern Molecular Photochemistry; University
Science Books: Sausalito, CA, 1991; p 314.
(61) Poizat, O.; Buntinx, G. J. Photochem. Photobiol., A 2007, 192,
172−178.
(62) Bandyopadhyay, B. N.; Harriman, A. J. Chem. Soc., Faraday
Trans. 1976, 663−674.
(21) Creutz, C.; Chou, M.; Netzel, T. L.; Okumura, M.; Sutin, N. J.
Am. Chem. Soc. 1980, 102, 1309−1319.
(22) Monat, J. E.; McCusker, J. K. J. Am. Chem. Soc. 2000, 122,
4092−4097.
(23) Ferrere, S.; Gregg, B. A. J. Am. Chem. Soc. 1998, 120, 843−844.
(24) Caspar, J. V.; Meyer, T. J. J. Am. Chem. Soc. 1983, 105, 5583−
5590.
(25) Balzani, V.; Bolletta, F.; Gandolfi, M. T.; Maestri, M. Top. Curr.
Chem. 1978, 75, 1−64.
(26) Kavarnos, G. J.; Turro, N. J. Chem. Rev. 1986, 86, 401−449.
(27) Treadway, J. A.; Loeb, B.; Lopez, R.; Anderson, P. A.; Keene, F.
R.; Meyer, T. J. Inorg. Chem. 1996, 35, 2242−2246.
(28) Damrauer, N. H.; Boussie, T. R.; Devenney, M.; McCusker, J. K.
J. Am. Chem. Soc. 1997, 119, 8253−8268.
(29) Damrauer, N. H.; Weldon, B. T.; McCusker, J. K. J. Phys. Chem.
A 1998, 102, 3382−3397.
(30) Damrauer, N. H.; McCusker, J. K. Inorg. Chem. 1999, 38, 4268−
4277.
(31) Strouse, G. F.; Schoonover, J. R.; Duesing, R.; Boyde, S.; Jones,
W. E., Jr.; Meyer, T. J. Inorg. Chem. 1995, 34, 473−487.
(32) Yasukouchi, K.; Taniguchi, I.; Yamaguchi, H.; Shiraishi, M. J.
Electroanal. Chem. 1979, 105, 403−408.
(33) Seshadri, G.; Lin, C.; Bocarsly, A. B. J. Electroanal. Chem. 1994,
372, 145−150.
(34) Barton, E. E.; Rampulla, D. M.; Bocarsly, A. B. J. Am. Chem. Soc.
2008, 130, 6342−6344.
(35) Cole, E. B.; Lakkaraju, P. S.; Rampulla, D. M.; Morris, A. J.;
Abelev, E.; Bocarsly, A. B. J. Am. Chem. Soc. 2010, 132, 11539−11551.
(36) Keith, J. A.; Carter, E. A. J. Am. Chem. Soc. 2012, 134, 7580−
7583.
H
dx.doi.org/10.1021/ja408787k | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX