Chlorine photoelimination from a diplatinum core: Circumventing the back reaction

Full text of DOI:10.1021/ja807222p

Published on Web 12/18/2008
Chlorine Photoelimination from a Diplatinum Core: Circumventing the Back
Reaction
Timothy R. Cook, Yogesh Surendranath, and Daniel G. Nocera*
Department of Chemistry, 6-355, Massachusetts Institute of Technology, 77 Massachusetts AVenue,
Cambridge, Massachusetts 02139-4307
Received September 11, 2008; E-mail: nocera@mit.edu
Solar energy may be stored by using it to rearrange the bonds
of H2O and HX (X ) Cl, Br) to H2/O2 and H2/X2, respectively.1,2
From a mechanistic standpoint, the oxidative half-reaction of HX
is more attractive than H2O because it only involves the coupling
of a two-electron oxidation to H2 generation. The water oxidation
half-reaction requires four electrons, and necessarily involves the
intimate coupling of electron to proton transfer.3,4 Unlike the 1e-
chemistry of redox symmetric bimetallic complexes,2,5 we have
Figure 1. Thermal ellipsoid plots of 1 (left) 2 (center) and 3 (right) drawn
at the 50% probability level; 1 and 3 crystallize with half a molecule per
asymmetric unit. (O)-CH2CF3 and (N)-CH3 groups of the phosphazanes
are omitted for clarity.
developed a method for hydrogen production from HX solutions
utilizing a bimetallic photocatalyst that operates in two discrete two-
electron steps for hydrogen evolution and halogen elimination.6,7
A hydrido-halide Rh2II,II(H)2(X)2 species is generated from the
0,0
Br) and µ-X2 complexes (X ) I).20 We find that the d8 · · ·d8 cores
addition of two HX molecules to a parent Rh2 complex. H2
of these complexes can disproportionate to deliver d7-d9 mixed-
photoelimination is facile to furnish a two-electron mixed valence
complex, Rh20,II(X)2; photoelimination of halogen, however, from
valence complexes in the presence of phosphazane ligands. Suitable
Rh20,II(X)2 is not. Hence the overall efficiency for the H2 photocycle
is limited by the efficiency of halogen photoelimination. More
generally, overcoming the strong metal-halide bond is an important
overall determinant of H2 production in most HX photocycles. If
no pathway for bond activation is accessible, productive chemistry
ceases upon hydrogen evolution.8 Even if halogen photoelimination
is achieved, halogen must be trapped since the back-reaction of
the primary photoelimination products is rapid and favorable.9 For
this reason, the overall photoefficiency typically scales with the
efficiency of the chemical trap. Traps are problematic in that the
trap-X bond provides the thermodynamic driving force,10,11 thus
obviating energy storage implications. In addition, the detailed
process by which photoelimination proceeds is obscured as both
X· and X2 can react with most traps. Improved efficiencies for H2
production therefore require increased quantum yields for M-X
bond activation and new strategies to prevent the back reaction of
primary photoproducts.
Pt0 and PtII sources (e.g., (tBudba)3Pt2 and Pt(cod)Cl2, tBudba ) 4′-
tert-butyl dibenzylideneacetone, cod ) 1,5-cyclooctadiene) react
with the phosphazane tfepma (tfepma ) ((CF3CH2O)2P)2NCH3) to
I,I
yield Pt2I,I(tfepma)2Cl2 (1) in analogy to the synthesis of Pt2
phosphine complexes.21 The solid state structure of 1, shown in
Figure 1, reveals a short Pt-Pt contact of 2.6187(6) Å, similar to
that of other M2(P-P)2X2 bimetallics and consistent with a
metal-metal bond.17,22
A
31P{1H} resonance at 113.57 ppm
disappears upon the addition of 1 equiv of PhI ·Cl2 to a solution of
1 and two resonances appear at 100.52 and 60.62 ppm, correspond-
ing to phosphorus atoms coordinated to the PtI and PtIII, respectively,
of Pt2I,III(tfepma)2Cl4 (2). X-ray diffraction studies confirm the two-
electron mixed-valent nature of 2 (Figure 1). The Pt(1)-Pt(2)
distance of 2.6187(7) Å establishes that a metal-metal bond is
maintained upon oxidation. The coordination of Cl(3) and Cl(4) to
Pt(2), (dPt(2)sCl(3) ) 2.323(2) Å and dPt(2)sCl(4) ) 2.338(2) Å)
completes the pseudo-octahedral geometry about the PtIII center.
The structure of 2 is unique among the oxidation products of
M2(P-P)2X2 bimetallics, as it represents the first example of a group
10 two-electron mixed-valent compound. Oxidation of 2 with 1
equiv of PhI ·Cl2, or treatment of 1 with 2 equiv of PhI ·Cl2, affords
a red solution of Pt2III,III(tfepma)2Cl6 (3). The 31P{1H} NMR
spectrum of 3 shows a single resonance at 52.22 ppm. As with 1
and 2, X-ray diffraction analysis of the complex indicates the
presence of a direct PtsPt interaction (dPtsPt ) 2.7037(3) Å) for
two metals possessing an octahedral coordination environment.
The diplatinum-based suite of d7-d7, d9-d7, and d9-d9 com-
plexes defined by 3, 2, and 1, respectively, complement the
previously described dirhodium suite of complexes of the same
d-electron count.9 As shown in Supporting Information, Figure S7,
the electronic profiles of 1-3 are dominated by excited states of
dσ* parentage as observed for the spectroscopy of d7-d7, d9-d7,
and d9-d9 dirhodium complexes.23 As the dσ* absorption profile
possesses both M-M and M-X antibonding character, electronic
excitation into the empty dσ* orbital is expected to promote halogen
Our most recent investigations of HX photocycles have focused
on improving the efficiency of the halogen elimination step of the
photocycle by incorporating oxidizing metals, such as Au and Pt
into bimetallic cores.12,13 Whereas a PtAu bimetallic trihalogen
complex shows markedly improved photoefficiencies for photo-
elimination,13 it nonetheless requires a halogen trap. We now report
a system that addresses both challenges of a high photoefficiency
III,III
for X2 elimination in the absence of a trap. Photolysis of a Pt2
complex with UV or visible light (355-510 nm) eliminates halogen
to deliver a two-electron mixed-valent Pt2I,III with a high quantum
yield in solution. Photolysis of the complex in the solid state yields
Cl2 with no need for a halogen trap.
Though bimetallic complexes of group 10 metals bridged by
diphosphine ligands typically promote insertion reactions,14-18 they
can drive oxidative chemistry.19 With regard to the latter, we were
intrigued by the oxidative chemistry of Pd2(P-P)2X2 (P-P )
dimethylphosphinomethane, diethylphosphinomethane) by halogens
to yield valence symmetric face-to-face Pd2II.II(P-P)2X4 (X ) Cl,
9
28 J. AM. CHEM. SOC. 2009, 131, 28–29
10.1021/ja807222p CCC: $40.75
2009 American Chemical Society

C O M M U N I C A T I O N S
Figure 2. Electronic absorption spectra of 3 (solid), 2 (dashed), and the
solid state photolysis product (dotted) in CH2Cl2. λirr > 350 nm.
Figure 3. Mass spectrometry analysis of the gas evolved from solid state
photolysis of 3. Traces correspond to 35Cl (black), 35Cl35Cl (red), 37Cl
(purple), 35Cl37Cl (blue), 37Cl37Cl (green) mass fragments. Inset shows the
fractional amount of 35Cl (red) and 37Cl (black) present.
elimination as we have observed for the M-X bond photoactivation
of Rh2 and PtAu13 systems.
9
of a chemical trap. Since Cl2 addition proceeds facilely under the
conditions of photolysis, we have shown authentic energy storage
reactivity in the absence of a halogen trap. Further work is underway
to unify the H2 and X2 photochemistry and to elucidate mechanistic
details of MsX bond activation for the compounds reported here.
Irradiation of 43 µM solutions of 3 in benzene with either 405 or
510 nm light in the presence of 2,3-dimethyl-1,3-butadiene (DMBD)
leads to rapid conversion to 2 as monitored by both UV-vis
spectroscopy (Figure S8) and 31P{1H} NMR (Figure S13). The
quantum yield for Pt-Cl bond activation is high. Figure S9 shows
the measured photoreaction quantum yields (Φp) as related to the
DMBD concentration. The observed quantum yield of 38% for halogen
photoelimination far exceeds all previously reported values.6,13 Since
both 405 and 510 nm light give the same efficiency for halogen
elimination, a common excited-state must be reached by the 385 and
520 nm transition bands. Photolysis of 3 in benzene in the absence of
DMBD results in reductive fragmentation of the bimetallic core to
yield Pt(tfepma)Cl2. Conversely, photolysis of 3 in THF in the absence
of DMBD results in its quantitative conversion to 2. Based on
dirhodium photochemistry,6,7 the photoeliminated halogen is efficiently
trapped by R-H abstraction from THF.
Acknowledgment. This research was supported by the NSF
(Grant CHE-0750239). Grants from the NSF also provided instru-
ment support to the DCIF at MIT (CHE-9808061, DBI-9729592).
Note Added after ASAP Publication. Typographical errors in the
compound descriptions presented in the online abstract (ASAP
12/18/08) have been corrected. The revised abstract was reposted on
January 7, 2009.
Supporting Information Available: Tables of bond lengths and
angles, full experimental details and crystallographic information files
(CIF). This material is available free of charge via the Internet at http://
pubs.acs.org.
The resulting suite of compounds preserves a metal-metal bond
I,III
across a four-electron transformation with a Pt2 species mixed-
References
valence species as the linchpin. The presence of metal-metal
bonding across the series provides an accessible excited-state with
σ* character from which photochemistry can originate, in much
the same way as has been previously reported for Rh2 dimers.9,23
A halogen trap is circumvented if the photoreaction is performed
in the solid state. Irradiation of solid powder samples of 3 in vacuo
and in ambient conditions results in elimination of Cl2 to produce
(1) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci.U.S.A. 2006, 103, 15729.
(2) Esswein, A. J.; Nocera, D. G. Chem. ReV. 2007, 107, 4022.
(3) Betley, T A.; Surendranath, Y.; Childress, M. V.; Alliger, G. E.; Fu, R.;
Cummins, C. C.; Nocera, D. G. Philos. Trans. R. Soc. London, B 2008,
363, 1293.
(4) Kanan, M. W.; Nocera, D. G. Science 2008, 321, 1072.
(5) Gray, H. B.; Maverick, A. W. Science 1981, 214, 1201.
(6) Heyduk, A. F.; Nocera, D. G. Science 2001, 293, 1639.
(7) Esswein, A. J.; Veige, A. S.; Nocera, D. G. J. Am. Chem. Soc. 2005, 127,
16641.
(8) Mann, K. R.; Lewis, N. S.; Miskowski, V. M.; Erwin, D. K.; Hammond,
G. S.; Gray, H. B. J. Am. Chem. Soc. 1977, 99, 5525.
(9) Heyduk, A. F.; Macintosh, A. M.; Nocera, D. G. J. Am. Chem. Soc. 1999,
121, 5023.
(10) Bellachioma, G.; Cardaci, G.; Macchioni, A.; Venturi, C.; Zuccaccia, C.
J. Organomet. Chem. 2006, 691, 3881.
(11) Fortune, J.; Manning, A. R. Inorg. Chem. 1980, 19, 2590.
(12) Esswein, A. J.; Dempsey, J. L.; Nocera, D. G. Inorg. Chem. 2007, 46,
2362.
(13) Cook, T. R.; Esswein, A. J.; Nocera, D. G. J. Am. Chem. Soc. 2007, 129,
10094.
(14) Jain, V. K.; Jain, L. Coord. Chem. ReV. 2005, 249, 3075.
(15) Azam, K. A.; Ferguson, G.; Ling, S. S. M.; Parvez, M.; Puddephatt, R. J.;
Srokowski, D. Inorg. Chem. 1985, 24, 2799.
(16) Benner, L. S.; Balch, A. L. J. Am. Chem. Soc. 1978, 100, 6099.
(17) Kullberg, M. L.; Kubiak, C. P. Inorg. Chem. 1986, 25, 26.
(18) Wong, T. Y. H.; Barnabas, A. F.; Sallin, D.; James, B. R. Inorg. Chem.
1995, 34, 2278.
1
(based on UV-vis, 31P{1H} and H analysis, Figures 2, S12, and
S14) 2 in 70% synthetic yield along with its isomer, the cis,trans-
Pt2II,II(tfepma)2Cl4 compound (see Supporting Information for
structural and NMR data) in 15% yield. Generation of chlorine gas
in these solid state experiments was confirmed by a mass
spectrometric (MS) analysis of the gas in the headspace of a sealed,
evacuated high-pressure glass vessel in which 3 was irradiated. Mass
fragments corresponding to 35,35Cl2, 37,37Cl2, 35,37Cl2, 35Cl ·, and 37Cl·
are all observed (Figure 3); the Cl · results from Cl2 fragmentation
upon electron impact ionization. The experimental abundance of
35Cl and 37Cl of 0.756(1) and 0.244(1), respectively, is in good
agreement with the natural abundance of these isotopes in chlorine
gas (75.78(4) and 24.22(4), respectively24). Since the possibility
of a bimolecular reaction is removed in the solid state and no trap
is present to collect ejected Cl· radicals, an intramolecular
elimination of Cl2 from the excited-state of 3 is invoked.
(19) Hunt, C. T.; Balch, A. L. Inorg. Chem. 1981, 20, 2267.
(20) Pamplin, C. B.; Rettig, S. J.; Patrick, B. O.; James, B. R. Inorg. Chem.
2003, 42, 4117.
(21) Pringle, P. G.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1983, 889.
(22) Holloway, R. G.; Penfold, B. R.; Colton, R.; McCormick, M. J. J. Chem.
Soc., Chem. Commun. 1976, 485.
(23) Kadis, J.; Shin, Y.-g. S.; Dulebohn, J. I.; Ward, D. L.; Nocera, D. G. Inorg.
Chem. 1996, 35, 811.
(24) Lide, D. R. CRC Handbook of Chemistry and Physics, 85th ed.; CRC Press:
Boca Raton, FL, 2004; pp 1-15.
The development of HX splitting photocatalysis is expanded by
the results reported herein: (1) high quantum yields for halogen
photoelimination have been achieved and (2) the evolution of Cl2
upon solid-state photolysis of 3 represents the first example of the
thermodynamically unfavorable halogen elimination without the use
JA807222P
9
J. AM. CHEM. SOC. VOL. 131, NO. 1, 2009 29