Photoreversible multiple additions of hydrogen to a highly unsaturated platinum-rhenium cluster complex

Article abstract of DOI:10.1021/ja067028g

The compound Pt₃Re₂(CO)₆(PBU ^t₃)₃, 1, was obtained from the reaction of Re₂(CO)₁₀ with Pt(PBu^t₃)₂ in octane solvent at reflux. Compound 1 consists of a trigonal bipyramidal cluster of five metal atoms with three platinum atoms in the trigonal plane and the two rhenium atoms in the apical positions. The metal cluster is formally unsaturated by 10 electrons. Compound 1 sequentially adds 3 equiv of hydrogen at room temperature/1 atm to form the series of compounds Pt₃Re ₂(CO)₆(PBu^t₃)₃(μ-H) ₂, 2, Pt₃Re₂(CO)₆(PBu ^t₃)₃(μ-H)₄, 3, and Pt ₃Re₂(CO)₆(PBu^t₃) ₃(μ-H)₆, 4. A small but significant kinetic isotope effect was observed, k_H/k_D = 1.3. The rate of addition of hydrogen is unaffected by the presence of a 20-fold excess of free PBu ^t₃ in solutions of 1. Compounds 2-4 each consist of a trigonal bipyramidal cluster of three platinum and two rhenium atoms similar to that of 1. The hydrido ligands in 2-4 bridge the platinum-rhenium bonds and are arranged to give structures having overall C_2v, symmetry for 2 and 3 and approximate D_3h symmetry for 4. Some of the hydrido ligands were expelled from 4 in the form of hydrogen upon exposure of solutions to UV-vis irradiation to yield compound 3 and then 2 in reasonable yields, but the elimination of all hydrido ligands to yield 1 was achieved only under the most forcing UV irradiation and then only with a major loss of the complex due to decomposition. The electronic structures of 1-4 were investigated by DFT calculations. Additional DFT calculations have suggested some mechanisms for the activation of hydrogen at multicenter metal sites without ligand eliminations prior to the hydrogen additions.

Full text of DOI:10.1021/ja067028g

Published on Web 01/05/2007
Photoreversible Multiple Additions of Hydrogen to a Highly
Unsaturated Platinum-Rhenium Cluster Complex
Richard D. Adams,*,† Burjor Captain,† Chad Beddie,‡ and Michael B. Hall*,‡
Contribution from the Department of Chemistry and Biochemistry, UniVersity of South Carolina,
Columbia, South Carolina 29208, and Department of Chemistry, Texas A&M UniVersity,
College Station, Texas 77843-3255
Received September 30, 2006; E-mail: adams@mail.chem.sc.edu; hall@science.tamu.edu
Abstract: The compound Pt3Re2(CO)6(PBut3)3, 1, was obtained from the reaction of Re2(CO)10 with Pt-
(PBut3)2 in octane solvent at reflux. Compound 1 consists of a trigonal bipyramidal cluster of five metal
atoms with three platinum atoms in the trigonal plane and the two rhenium atoms in the apical positions.
The metal cluster is formally unsaturated by 10 electrons. Compound 1 sequentially adds 3 equiv of hydrogen
at room temperature/1 atm to form the series of compounds Pt3Re2(CO)6(PBut3)3(µ-H)2, 2, Pt3Re2(CO)6-
(PBut3)3(µ-H)4, 3, and Pt3Re2(CO)6(PBut3)3(µ-H)6, 4. A small but significant kinetic isotope effect was
observed, kH/kD ) 1.3. The rate of addition of hydrogen is unaffected by the presence of a 20-fold excess
of free PBut3 in solutions of 1. Compounds 2-4 each consist of a trigonal bipyramidal cluster of three
platinum and two rhenium atoms similar to that of 1. The hydrido ligands in 2-4 bridge the platinum-
rhenium bonds and are arranged to give structures having overall C2v symmetry for 2 and 3 and approximate
D3h symmetry for 4. Some of the hydrido ligands were expelled from 4 in the form of hydrogen upon exposure
of solutions to UV-vis irradiation to yield compound 3 and then 2 in reasonable yields, but the elimination
of all hydrido ligands to yield 1 was achieved only under the most forcing UV irradiation and then only with
a major loss of the complex due to decomposition. The electronic structures of 1-4 were investigated by
DFT calculations. Additional DFT calculations have suggested some mechanisms for the activation of
hydrogen at multicenter metal sites without ligand eliminations prior to the hydrogen additions.
Introduction
certain metal complexes will react with hydrogen to condense
and form polynuclear metal cluster complexes containing
There is currently much interest in the development of a
hydrogen-based system for fuels and energy storage for the
future.1 There are active searches for molecules and materials
that can reversibly absorb large quantities of hydrogen.2,3 Studies
have shown that in the presence of bulky phosphine ligands
unusually large numbers of hydrido ligands.4,5 Recent studies
have shown that even some cyclopentadienyl complexes of the
lanthanides can form polynuclear metal complexes containing
large numbers of hydrido ligands.6 Reversibility is a property
that must be central to any application of new materials for
hydrogen storage. There are only a few examples of thermally
reversible hydrogen additions to metal complexes that are not
accompanied by the addition and elimination of other ligands.4a,b,7,8
One of the most interesting families of these compounds are
† University of South Carolina.
‡ Texas A&M University.
(1) (a) DOE Hydrogen Storage Targets. http://www.eere.energy.gov/hydroge-
nandfuelcells/mypp/pdfs/storage.pdf. (b) A Multiyear Plan for the Hydrogen
R&D Program: Rationale, Structure, and Technology Road Maps; U.S.
Department of Energy: Washington, DC, 1999; http://www.eere.energy-
.gov/ hydrogenandfuelcells/pdfs/bk28424.pdf, p 32. (c) National Hydrogen
Energy Road Map; U.S. Department of Energy: Washington, DC, 2002;
http://www.sc.doe.gov/bes/reports/files/NHE_rpt.pdf, p 17. (d) Ogden,
J. M. Phys. Today 2002, 55, 69-75. (e) Ogden, J. M. Annu. ReV. Energy
EnViron. 1999, 24, 227-279. (f) Ogden, J. Sci. Am. 2006, 295, 94-101.
(2) (a) Schlapbach. L; Zu¨ttel, A. Nature 2001, 414, 353-358. (b) Zecchina,
A.; Borgida, S.; Vitillo, J. G.; Ricchiardi, G.; Lamberti, C.; Spoto, G.;
Bjørgen, M.; Lillerud, K. P. J. Am. Chem. Soc. 2005, 127, 6361-6366. (c)
Nijkamp, M. G.; Raaymakers, J. E. M. J.; van Dillen, A. J.; de Jong, K. P.
Appl. Phys. A 2001, 72, 619-623. (d) Weitkamp, J.; Fritz, M.; Ernst, S.
Int. J. Hydrogen Energy 1995, 20, 967-970.
(4) (a) Brayshaw, S. K.; Ingleson, M. J.; Green, J. C.; McIndoe, J. S.; Raithby,
P. R.; Kociok-Ko¨hn, G.; Weller, A. S. J. Am. Chem. Soc. 2006, 128, 6247-
6263. (b) Brayshaw, S. K.; Ingleson, M. J.; Green, J. C.; Raithby, P. R.;
Kociok-Ko¨hn, G.; McIndoe, J. S.; Weller, A. S. Angew. Chem., Int. Ed.
2005, 44, 6875-6878. (c) Ingleson, M. J.; Mahon, M. F.; Raithby, P. R.;
Weller, A. S. J. Am. Chem. Soc. 2004, 126, 4784-4785.
(5) (a) Frost, P. W.; Howard, J. A. K.; Spencer, J. L.; Turner, D. G. J. Chem.
Soc., Chem. Commun. 1981, 1104-1106. (b) Gregson, D.; Howard, J. A.
K.; Murray, M.; Spencer, J. L. J. Chem. Soc., Chem. Commun. 1981, 716-
717. (c) Chodosh, D. F.; Crabtree, R. H.; Felkin, H.; Morehouse, S.; Morris,
G. E. Inorg. Chem. 1982, 21, 1307-1311. (d) Smidt, S. P.; Pfaltz, A.;
Martinez-Viviente, E.; Pregosin, P. S.; Albinati, A. Organometallics 2003,
22, 1000-1009. (e) Lemmen, T. H.; Folting, K.; Huffman, J. C.; Caulton,
K. G. J. Am. Chem. Soc. 1985, 107, 7774-7775. (f) Goeden, G. V.; Caulton,
K. G. J. Am. Chem. Soc. 1981, 103, 7354-7355.
(6) (a) Tardif, O.; Nishiura, M.; Hou, Z. M. Organometallics 2003, 22, 1171-
1173. (b) Cui, D. M.; Tardif, O.; Hou, Z. M. J. Am. Chem. Soc. 2004, 126,
1312-1313. (c) Luo, Y.; Baldamus, J.; Tardif, O.; Hou, Z. M. Organo-
metallics 2005, 24, 4362-4366.
(7) Kubas, G. J. Metal Dihydrogen and σ-Bond Complexes; Kluwer/Academic/
Plenum Publishers: New York, 2001; Chapter 3.
(3) (a) Rosi, N. L.; Eckert, J.; Eddaoudi, M.; Vodak, D. T.; O’Keerre, M.;
Yaghi, O. M. Science 2003, 300, 1127-1129. (b) Zhao, X.; Xiao, B.;
Fletcher, A. J.; Thomas, K. M.; Bradshaw, D.; Rosseinsky, M. J. Science
2004, 306, 1012-1015. (c) Rowsell, J. L. C.; Eckert, J.; Yaghi, O. M.
J. Am. Chem. Soc. 2005, 127, 14904-14099. (d) Rowsell, J. L. C.; Spencer,
E. C.; Eckert, J.; Howard, J. A. K.; Yaghi, O. M. Science 2005, 309, 1350-
1354. (e) Rowsell, J. L. C.; Yaghi, O. M. Angew. Chem., Int. Ed. 2005,
44, 4670-4679. (f) Dinca, M.; Long, J. R. J. Am. Chem. Soc. 2005, 127,
9376-9377. (g) Kaye, S. S.; Long, J. R. J. Am. Chem. Soc. 2005, 127,
6506-6507. (h) Sun, D.; Ma, S.; Ke, Y.; Collins, D. J.; Zhou, H.-C.
J. Am. Chem. Soc. 2006, 128, 3896-3897.
9
986
J. AM. CHEM. SOC. 2007, 129, 986-1000
10.1021/ja067028g CCC: $37.00 © 2007 American Chemical Society

Multiple Additions of Hydrogen to a Pt−Re Cluster
A R T I C L E S
unsaturated intermediate.8e Bulky ligands, such as PBu3 , can
t
Scheme 1
stabilize metal complexes in electronically unsaturated states;
see above. In these cases, it is possible that hydrogen addition
can proceed without a ligand elimination step.7,8c
In recent studies we have shown that it is possible to prepare
electronically unsaturated bimetallic cluster complexes by
introducing the 12-electron Pt(PBut3) or Pd(PBut3) group into
metal carbonyl clusters through use of the reagents Pd(PBut3)2
and Pt(PBut3)2.10-13 In many cases, the M(PBut3), M ) Pd or
Pt, groups in these complexes are highly mobile and readily
move about the surface of the cluster complex itself.11 To some
of these compounds we have been able to introduce simple small
8c,9,14
molecules such as H2
and alkynes.15 Here, we present a
new example of these unsaturated complexes, the highly
t
unsaturated compound Pt3Re2(CO)6(PBu3 )3, 1, which sequen-
Scheme 2a
tially adds up to 3 equiv of hydrogen at room temperature/1
atm of H2 in a process that can be reversed partially by
application of UV-vis irradiation. A preliminary report of a
portion of this work has been published.16
Experimental Section
General Data. Reagent grade solvents were dried by the standard
procedures and were freshly distilled prior to use. Infrared spectra were
recorded on a Thermo Nicolet Avatar 360 FT-IR spectrophotometer.
1H NMR and 31P{1H} NMR spectra were recorded on a Varian Mercury
400 spectrometer operating at 400.1 and 161.9 MHz, respectively. The
1H NMR spectrum of 2 was recorded on a Varian Inova 500
spectrometer operating at 500.2 MHz. 31P{1H} NMR spectra were
externally referenced against 85% ortho-H3PO4. NMR simulations were
performed using the SpinWorks program.17 Electrospray mass spec-
trometric measurements were obtained on a MicroMass Q-Tof spec-
trometer. Pt(PBut3)2 and Re2(CO)10 were purchased from Strem and
were used without further purification. Product separations were
performed by thin layer chromatography (TLC) in air on Analtech 0.25
and 0.5 mm silica gel 60 Å F254 glass plates.
Preparation of Pt3Re2(CO)6(PBut3)3, 1. Under an atmosphere of
nitrogen, Re2(CO)10 (29.0 mg, 0.044 mmol) and Pt(PBut3)2 (160.0 mg,
0.267 mmol) were dissolved in 15 mL of octane. The reaction mixture
was heated to reflux for 3 h. The solvent was removed in vacuo, and
the product was separated by TLC by using a 3:1 hexane/methylene
chloride solvent mixture to yield 13.9 mg (18%) of brown 1. A
considerable amount of Re2(CO)10, 8.8 mg (30%), was recovered along
with one minor dirhenium product, Re2(CO)9(PBut3), 4.0 mg (11%), a
a Key: platinum (blue), rhenium (green), phosphorus (gold), oxygen (red),
carbon (pink), hydrogen (gray).
the hexarhodium complexes [Rh6(PR3)6H12][BArF ] (R ) iPr,
4 2
Cy; ArF ) [B{C6H3(CF3)2}4]) that were recently reported by
Weller and co-workers.4 These complexes add an additional 2
equiv of hydrogen, reversibly, to yield the hexadecahydride
complexes [Rh6(PR3)6H16][BArF ] (R ) iPr, Cy; ArF
)
4 2
[B{C6H3(CF3)2}4]); see Scheme 1.4a,b Another example is the
t
unsaturated dirhenium diplatinum complex Pt2Re2(CO)7(PBu3 )2-
(µ-H)2, which adds 1 equiv of H2 at room temperature to yield
t
the tetrahydrido complex Pt2Re2(CO)7(PBu3 )2(µ-H)4.8c Interest-
ingly, this hydrogen addition can be reversed by both thermal
and photochemical means; see Scheme 2. Photochemical
stimulation is an effective method for releasing hydrogen from
mononuclear metal complexes,9 but there are very few examples
where hydrogen can be eliminated photolytically from a
polynuclear metal complex.8c
Electronic unsaturation is an important condition for the
addition of small molecules such as H2 to polynuclear metal
complexes.4,7,8a-d In most cases, hydrogen addition reactions
follow a ligand elimination step that creates an electronically
phosphine derivative of Re2(CO)10. Spectral data for 1: IR νCO (cm-1
,
in hexane) 1985 (s), 1892 (s); 1H NMR (400 MHz, toluene-d8, 25 °C,
3
TMS) δ )1.63 ppm (d, J(P,H) ) 13 Hz, 81 H, CH3); 31P{1H} NMR
(162 MHz, toluene-d8, 25 °C, 85% ortho-H3PO4) δ ) 109.23 ppm (m,
3
3P) [2J(Pt,Pa) ) 332 Hz, J(Pa,Pb) ) 44 Hz] [1J(Pt,Pb) ) 4162 Hz,
3J(Pa,Pb) ) 44 Hz] [1J(Pt,Pb) ) 4531 Hz] [2J(Pt,Pa) ) 1253 Hz]; ES+/
MS m/z calcd for M + H, [Pt3Re2P3O6C42H81 + H]+, 1733, found 1733.
The isotope pattern is consistent with the presence of three platinum
and two rhenium atoms. Anal. Calcd for 1: 29.07, C; 4.71, H. Found:
(8) (a) Farrugia, L. J.; Green, M.; Hankey, D. R.; Orpen, A. G.; Stone,
F. G. A. J. Chem. Soc., Chem. Commun. 1983, 310-312. (b) Adams,
R. D.; Wang, S. Organometallics 1986, 5, 1272-1274. (c) Adams, R. D.;
Captain, B.; Smith, M. D. Angew. Chem., Int. Ed. 2006, 45, 1109-1112.
(d) Aubart, M. A.; Chandler, B. D.; Gould, R. A. T.; Krogstad, D. A.;
Schoondergang, M. F. J.; Pignolet, L. H. Inorg. Chem. 1994, 33, 3724-
3734. (e) Safarowic, F. J.; Bierdeman, D. J.; Keister, J. B. J. Am. Chem.
Soc. 1996, 118, 11805-11812. (f) Arif, A. M.; Bright, T. A.; Jones, R. A.;
Nunn, C. M. J. Am. Chem. Soc. 1988, 110, 6894-6895.
(9) (a) Perutz, R. N. Pure Appl. Chem. 1998, 70, 2211-2220. (b) Geoffroy,
G. L.; Wrighton, M. S. Organometallic Photochemistry; Academic Press:
New York, 1979; Chapter 7. (c) Daniel, C. J. Phys. Chem. 1991, 95, 2394-
2398. (d) Geoffroy, G. L. Prog. Inorg. Chem. 1980, 27, 123-151. (d) Ma,
Y.; Bergman, R. G. Organometallics 1994, 13, 2548-2550. (e) Janowicz,
A. H.; Bergman, R. G. J. Am. Chem. Soc. 1983, 105, 3929-3939.
(10) (a) Adams, R. D.; Captain, B.; Fu, W.; Hall, M. B.; Manson, J.; Smith,
M. D.; Webster, C. E. J. Am. Chem. Soc. 2004, 126, 5253-5267. (b)
Adams, R. D.; Captain, B.; Fu, W.; Smith, M. D. J. Am. Chem. Soc. 2002,
124, 5628-5629. (c) Adams, R. D.; Captain, B.; Zhu, L. Inorg. Chem.
2006, 45, 430-436.
(11) (a) Adams, R. D.; Captain, B.; Fu, W.; Pellechia, P. J.; Smith, M. D. Angew.
Chem., Int. Ed. 2002, 41, 1951-1953. (b) Adams, R. D.; Captain, B.; Fu,
W.; Pellechia, P. J.; Smith, M. D. Inorg. Chem. 2003, 42, 2094-2101. (c)
Adams, R. D.; Captain, B.; Fu, W.; Pellechia, P. J.; Zhu, L. Inorg. Chem.
2004, 43, 7243-7249.
(12) Adams, R. D.; Captain, B.; Hall, M. B.; Smith, J. L., Jr.; Webster, C. E.
J. Am. Chem. Soc. 2005, 127, 1007-1014.
(13) Adams, R. D.; Captain, B.; Herber, R. H.; Johansson, M.; Nowik, I.; Smith,
J. L., Jr.; Smith, M. D. Inorg. Chem. 2005, 44, 6346-6358.
(14) Adams, R. D.; Captain, B.; Zhu, L. Inorg. Chem. 2005, 44, 6623-6631.
(15) Adams, R. D.; Captain, B.; Zhu, L. Organometallics 2005, 24, 2419-
2423.
(16) Adams, R. D.; Captain, B. Angew. Chem., Int. Ed. 2005, 44, 2531-2533.
(17) Marat, K. SpinWorks version 2.4. Department of Chemistry, University of
Manitoba, Winnipeg, Manitoba, Canada R3T 2N2; ftp://davinci.chem.u-
manitoba.ca/pub/marat/SpinWorks/.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 987

A R T I C L E S
Adams et al.
4. Spectral data for 3: IR νCO (cm-1, in hexane) 2000 (vs), 1910 (s),
1901 (s); 1H NMR (400 MHz, toluene-d8, 25 °C, TMS) δ ) 1.63 ppm
3
3
(d, J(P,H) ) 13 Hz, 27 H, CH3), 1.47 ppm (d, J(P,H) ) 13 Hz, 54
H, CH3), -7.42 ppm (m, J(Pt,H) ) 760 Hz, 4H, hydride); 31P{1H}
1
NMR (162 MHz, toluene-d8, 25 °C, 85% ortho-H3PO4) δ ) 110.65
ppm (m, 2P), 108.41 ppm (m, 1p) [3J(PA,PB) ) 55.0 Hz]; ES+/MS m/z
calcd for M + H, [Pt3Re2P3O6C42H85 + H]+, 1737, found 1737. The
isotope pattern is consistent with the presence of three platinum and
two rhenium atoms.
Determination of the Kinetic Isotope Effect (KIE) for the
Addition of H2 to 1. At room temperature, deuterium gas was bubbled
(flow rate ∼27 mL/min) through the solution of 6.0 mg of 1 dissolved
in 25 mL of hexane. During this time IR spectra were recorded
periodically. Under identical conditions, hydrogen was bubbled through
a solution of 1 (6.0 mg) dissolved in 25 mL of hexane, and the IR
spectra were recorded periodically. At certain times the IR spectra of
the two samples appeared to be virtually the same. The KIE was
determined to be the ratio of the reaction times for D2/H2 for the two
reactions when the spectra appeared to be the same. For example, the
IR spectrum for the D2 addition reaction after 2 h was approximately
the same as the IR spectrum after 1.5 h for the H2 addition reaction;
thus, the D2/H2 isotope effect kH/kD was determined to be 1.3(1).
Figure 1. IR spectra in the CO absorption region showing the progress of
the reaction of 1 with hydrogen to form 2-4 as a function of time. Reaction
times t in hours are shown in the insets on the left.
Addition of H2 to 1 in the Presence of Free PBut3. Compound 1
(11.0 mg, 0.006 mmol) was dissolved in 40 mL of hexane. This solution
was divided into two equal portions, and to one of these solutions was
added PBut3 (16 µL, 0.064 mmol). With stirring, hydrogen gas was
then bubbled through both solutions at equal rates for 3 h. A series of
IR spectra for the two samples were recorded during the 3 h period.
Comparison of the IR spectra recorded at the same time intervals
indicated no significant differences in the rate of formation of
compounds 2-4 in the presence or absence of PBut3.
29.98, C; 4.89, H. Spectral and crystallographic data for Re2(CO)9-
(PBut3): IR νCO (cm-1, in hexane) 2109 (m), 2004 (vs), 1970 (m), 1920
1
(s); H NMR (400 MHz, toluene-d8, 25 °C, TMS) δ ) 1.19 ppm (d,
3J(P,H) ) 12 Hz, 27 H, CH3); 31P{1H} NMR (162 MHz, toluene-d8,
25 °C, 85% ortho-H3PO4) δ ) 88.47 ppm (s, 1P); monoclinic, space
group P21/n, unit cell a ) 11.7770(11) Å, b ) 14.9566(15) Å, c )
45.768(5) Å, R ) 90.00°, â ) 92.098(2)°, γ ) 90.00°, V ) 8056.3-
(14) Å3, and Z ) 12, temperature 294(2) K, radiation Mo KR (λ )
0.71073 Å), number of reflections (>2σ) 10887, number of parameters
917, R ) 0.0331, Rw ) 0.0745.
UV-Vis Irradiation of 4. Compound 4 (10.1 mg, 0.0058 mmol)
was dissolved in 40 mL of toluene in a 100 mL three-neck Pyrex flask
equipped with a reflux condenser and a gas inlet. A slow stream of
nitrogen (flow rate ∼27 mL/min) was allowed to flow through the flask
while the solution was irradiated for 7.5 h using a medium-pressure
mercury UV lamp (1000 W). The progress of this reaction can be
monitored by IR. The toluene solvent was then removed in vacuo, and
the residue was dissolved in ∼40 mL of hexane. The hexane solution
was then filtered through a glass frit into a 100 mL Schlenk flask. The
hexane solution was concentrated to ∼5 mL and then transferred to a
10 mL Schlenk tube. This solution was further concentrated to 1-2
mL and then placed in the freezer (-25 °C) overnight to yield dark
crystals of 2 (2.7 mg, 27%) (note: compound 2 is unstable on silica
gel and decomposes in air): IR νCO (cm-1, in hexane) 1992 (vs), 1905
Addition of Hydrogen to 1. Compound 1 (11.6 mg, 0.0067 mmol)
was dissolved in 25 mL of hexane. With stirring, hydrogen was then
allowed to purge through the solution at 25 °C for 4 h. The solvent
was removed in vacuo, and the product was separated by TLC using
a 3:1 hexane/methylene chloride solvent mixture to yield 10.5 mg (90%)
of orange Pt3Re2(CO)6(PBut3)3(µ-H)6, 4. Spectral data for 4: IR νCO
1
(cm-1, in hexane) 2009 (s), 1914 (vs); H NMR (400 MHz, toluene-
3
d8, 25 °C, TMS) δ ) 1.50 ppm (d, J(P,H) ) 13 Hz, 81 H, CH3),
-4.34 ppm (quartet, 1J(Pt,H) ) 723 Hz, 2J(P,H) ) 4 Hz, 6H, hydride);
31P{1H} NMR (162 MHz, toluene-d8, 25 °C, 85% ortho-H3PO4) 111.42
ppm (m, 3P) [2J(Pt,Pa) ) 222 Hz, 3J(Pa,Pb) ) 56 Hz] [1J(Pt,Pb) ) 2986
3
3
Hz, J(Pa,Pb) ) 56 Hz] [1J(Pt,Pb) ) 3208 Hz, J(Pa,Pb) ) 56 Hz], see
the Discussion for an explanation; ES+/MS m/z calcd for M + H, [Pt3-
Re2P3O6C42H87 + H]+, 1739, found 1739. The isotope pattern is
consistent with the presence of three platinum and two rhenium atoms.
Anal. Cacld for Pt3Re2(CO)6(PBut3)3(µ-H)6·C6H6: 31.73, C; 5.16, H.
Found: 31.94, C; 5.21, H. If one follows the reaction of 1 with hydrogen
at 25 °C/1 atm by IR spectroscopy, CO absorptions due to the formation
of a series of three products, Pt3Re2(CO)6(PBut3)3(µ-H)2, 2, Pt3Re2(CO)6-
(PBut3)3(µ-H)4, 3, and Pt3Re2(CO)6(PBut3)3(µ-H)6, 4, are observed,
formed by the addition of 1, 2, and 3 equiv of H2 to 1. Plots of a series
of these spectra as a function of time are shown in Figure 1. Absorptions
due to compounds 2 and 3 are observed at 1992 (vs), 1905 (s), and
1893 (s) cm-1 and 2000 (vs), 1910 (s), and 1901 (s) cm-1, respectively.
Compound 2 is sensitive to air and silica gel. It was obtained in a pure
form only by irradiating solutions of 4; see below. Compound 3 can
be isolated in a low yield by stopping the reaction before completion;
see the next section.
1
(s), 1893 (s); H NMR (400 MHz, toluene-d8, 25 °C, TMS) δ ) 1.56
ppm (d, 3J(P,H) ) 13 Hz, 27 H, CH3), 1.52 ppm (d, 3J(P,H) ) 13 Hz,
1
2
54 H, CH3), -5.91 ppm (dt, J(Pt,H) ) 821 Hz, J(P,H) ) 7.6 Hz,
3J(P,H) ) 1.4 Hz, 2H, hydride); 31P{1H} NMR (162 MHz, toluene-d8,
3
25 °C, 85% ortho-H3PO4) δ ) 111.1 ppm (d, J(P,P) ) 53 Hz, 2P),
105.6 ppm (t, J(P,P) ) 53 Hz, 1P); ES+/MS m/z calcd for M + H,
3
[Pt3Re2P3O6C42H83 + H]+, 1735, found 1735. The isotope pattern is
consistent with the presence of three platinum and two rhenium atoms.
Photolysis of Compound 4 in Quartz Glassware. Compound 4
(10.5 mg, 0.0060 mmol) was dissolved in 30 mL of hexane in a 50
mL three-neck quartz flask equipped with a reflux condenser and a
gas inlet. A slow stream of nitrogen (flow rate ∼27 mL/min) was
allowed to flow through the flask while the solution was irradiated for
4 h by using a high-pressure mercury UV lamp (American Ultraviolet
Co.) at the 250 wpi setting. The IR spectrum of the reaction mixture
showed compound 2 to be the major product; however, a small
absorption for compound 1 was also observed. The solvent was removed
in vacuo, and the product was separated by TLC using a 3:1 hexane/
methylene chloride solvent mixture to yield a trace amount, ∼0.5 mg
(∼ 0.5%), of compound 1. The major compound 2 decomposed during
Preparation of Pt3Re2(CO)6(PBut3)3(µ-H)4, 3. Compound 1 (16.3
mg, 0.0094 mmol) was dissolved in 25 mL of hexane. With stirring,
hydrogen was then allowed to purge through the solution at 25 °C for
1.5 h. The solvent was removed in vacuo, and the products were
separated by TLC using a 4:1 hexane/methylene chloride solvent
mixture to yield 2.1 mg (13%) of brown 3 and 2.4 mg (15%) of orange
9
988 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Multiple Additions of Hydrogen to a Pt
−
Re Cluster
A R T I C L E S
Table 1. Crystallographic Data for Compounds 1 and 2
the separation, and bands corresponding to compounds 3 and 4 were
also present in trace amounts.
1
2
Detection of H2 by 1H NMR. A 10.5 mg amount of 2 was dissolved
in 3 mL of toluene-d8 in a 10 mm NMR tube. The NMR tube was
evacuated and filled with nitrogen four times and then evacuated one
last time. The evacuated NMR tube was then irradiated using a medium-
pressure mercury UV lamp (1000 W) for 5 h. 1H NMR showed a singlet
at 4.51 ppm, indicating the presence of H2.
empirical formula
fw
cryst syst
lattice params
a (Å)
Pt3Re2P3O6C42H81‚C6H14 Pt3Re2P3O6C42H81‚C6H14
1818.82
1818.82
monoclinic
monoclinic
13.9117(13)
24.058(2)
18.0919(17)
92.210(2)
6050.7(10)
P21/n (No. 14)
4
13.9272(13)
23.991(2)
18.0905(16)
91.807(2)
6041.5(9)
P21/n (No. 14)
4
b (Å)
c (Å)
Crystallographic Analyses. Dark single crystals of 1 and 3 suitable
for X-ray diffraction analyses were obtained by slow evaporation of
solvent from a hexane/methylene chloride solvent mixture at -25 °C.
Dark single crystals of 2 were obtained by slow evaporation of hexane
at -25 °C. Red single crystals of 4 were obtained by slow evaporation
from a benzene/octane solution at 8 °C. Each data crystal was glued
onto the end of a thin glass fiber. X-ray intensity data were measured
by using a Bruker SMART APEX CCD-based diffractometer using
Mo KR radiation (λ ) 0.71073 Å). The raw data frames were integrated
with the SAINT+ program by using a narrow-frame integration
algorithm.18 Correction for Lorentz and polarization effects was also
applied with SAINT+. An empirical absorption correction based on
the multiple measurement of equivalent reflections was applied using
the program SADABS. All structures were solved by a combination
of direct methods and difference Fourier syntheses and refined by a
full-matrix least-squares method on F2, using the SHELXTL software
package.19 All non-hydrogen atoms were refined with anisotropic
displacement parameters. Hydrogen atoms were placed in geometrically
idealized positions and included as standard riding atoms during the
least-squares refinements. Crystal data, data collection parameters, and
results of the analyses are listed in Tables 1 and 2.
â (deg)
V (Å3)
space group
Z value
F
calcd (g/cm3)
1.997
2.000
µ(Mo KR) (mm-1
temp (K)
)
11.016
11.033
296
296
2θmax (deg)
50.06
50.06
no. of obsd reflns
(I > 2σ(I))
7040
5660
no. of params
559
529
GOFa
1.004
1.004
max shift in cycle
residuals:a R, Rw
abs correction
max/min
largest peak in final 1.567
diff map (e-/Å3)
0.002
0.001
0.0501, 0.1010
multiscan
1.000/0.588
0.0515, 0.1068
multiscan
1.000/0.431
2.327
2
a R ) ∑hkl(||Fo| - |Fc||)/∑hkl|Fo|. Rw ) [∑hklw(|Fo| - |Fc|)2/∑hklwFo ]1/2
,
w ) 1/σ2(Fo). GOF ) [∑hklw(|Fo| - |Fc|)2/(ndata - nvari)]1/2
.
Table 2. Crystallographic Data for Compounds 3 and 4
Compounds 1, 2, and 4 crystallized in the monoclinic crystal system.
For all three compounds, the systematic absences in the intensity data
identified the unique space group P21/n. For compound 1 one molecule
of hexane from the crystallization solvent cocrystallized with the
complex. The solvent was included in the analysis and was satisfactorily
refined with anisotropic thermal parameters. For compound 2 one
molecule of hexane from the crystallization solvent cocrystallized with
the complex. The solvent was included in the analysis and was
satisfactorily refined with isotropic thermal parameters. For compound
4 one molecule of benzene and half a molecule of octane from the
crystallization solvent cocrystallized with the complex. The solvent was
included in the analysis and was satisfactorily refined with anisotropic
thermal parameters. The six hydrido ligands in 4 were located and
refined successfully with isotropic thermal parameters.
3
4
empirical formula
Pt3Re2P3O6C42H81‚C6H14
Pt3Re2P3O6C42H87‚
C6H6‚1/2C8H18
1873.92
fw
1818.82
cryst syst
lattice params
a (Å)
b (Å)
c (Å)
orthorhombic
monoclinic
24.044(2)
18.2218(16)
13.8953(12)
90
6087.9(9)
Pnma (No. 61)
4
14.7099(6)
23.4704(9)
17.7176(7)
91.031(1)
6116.0(4)
P21/n (No. 14)
4
â (deg)
V (Å3)
space group
Z value
F
calcd (g/cm3)
1.984
10.949
296
2.035
10.902
100
µ(Mo KR) (mm-1
temp (K)
)
Compound 3 crystallized in the orthorhombic crystal system. The
systematic absences in the intensity data were consistent with either
space group Pnma or space group Pna21. The former space group was
chosen initially for structure solution tests. This space group was
confirmed by the successful solution and refinement of the structure.
One molecule of hexane from the crystallization solvent was cocrys-
2θmax (deg)
56.72
5638
56.60
13055
no. of obsd reflns
(I > 2σ(I))
no. of params
308
1.044
0.002
0.0392, 0.0856
multiscan
1.000/0.286
1.512
641
1.060
0.001
0.0342, 0.0757
multiscan
1.000/0.634
1.955
GOFa
max shift in cycle
residuals:a R, Rw
abs correction
max/min
largest peak in final
diff map (e-/Å3)
(18) SAINT+ Version 6.2a. Bruker Analytical X-ray System, Inc., Madison,
WI, 2001.
(19) Sheldrick; G. M. SHELXTL Version 6.1. Bruker Analytical X-ray Systems,
Inc., Madison, WI, 1997.
(20) All calculations were conducted using the Gaussian03 suite of programs:
Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo,
J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.;
Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.;
Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels,
A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson,
B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03,
Revision B.4. Gaussian, Inc., Pittsburgh, PA, 2003.
2
a R ) ∑hkl(||Fo| - |Fc||)/∑hkl|Fo|. Rw ) [∑hklw(|Fo| - |Fc|)2/∑hklwFo ]1/2
,
w ) 1/σ2(Fo). GOF ) [∑hklw(|Fo| - |Fc|)2/(ndata - nvari)]1/2
.
tallized with the complex in the asymmetric unit. The solvent was
included in the analysis and was satisfactorily refined with anisotropic
thermal parameters.
Details of the Computational Treatments. All calculations were
performed using the Gaussian 03 suite of programs.20 Optimized gas-
phase geometries for the QM model (see below) were obtained using
the B3PW91 DFT method.21,22 The basis set (BS1) used for geometry
(21) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 989

A R T I C L E S
Adams et al.
Figure 2. QM and QM/MM models used in the theoretical investigations. Text in blue indicates the regions that are modeled with molecular mechanics.
optimizations and energy calculations was implemented as follows: for
rhenium and platinum, the valence double-ú LANL2DZ23 basis set was
supplemented with sets of 6p functions for transition metals developed
by Couty and Hall,24 for phosphorus, the LANL2DZdp basis set was
used,24-26 for carbon and oxygen, the 6-31G(d′)27 basis set was used,
for the hydrogen atoms attached to phosphorus, the 6-31G basis set
was used,27 and for dihydrogen and for the hydrogen atoms that bonded
to the platinum and rhenium, the 6-31G(d′,p′) basis set was used.27 All
structures were calculated in singlet spin states using the restricted
B3PW91 method. Calculating the harmonic vibrational frequencies and
noting the number of imaginary frequencies confirmed the nature of
all intermediates (Nimag ) 0) and transition-state structures (Nimag ) 1).
Two computational models were used for the theoretical calcula-
tions: The first model, which will be referred to as the QM model,
used a chemical model in which PH3 ligands were used instead of the
PBut3 ligands and the geometry optimizations and energy calculations
were conducted at the B3PW91/BS1 level of theory. The second model,
which will be referred to as the QM/MM model, used the experimental
chemical model; however, geometry optimizations and energy calcula-
tions were conducted using a two-level ONIOM procedure,28 in which
B3PW91/BS1 was used as the higher level of theory and the UFF
molecular mechanics force field29 was used as the lower level of theory.
For the ONIOM procedure, Re2Pt3(CO)6(PH3)3 was used as the structure
of the high level (QM region), while Re2Pt3(CO)6(PBut3)3 was used as
the structure of the low level (MM region), as shown in Figure 2. Thus,
the ONIOM-extrapolated electronic energy of the complex is given as
follows:
ONIOM-extrapolated energy of [Re2Pt3(CO)6(PBut3)3] )
B3PW91/BS1[Re2Pt3(CO)6(PH3)3] + UFF[Re2Pt3(CO)6
(PBut3)3] - UFF[Re2Pt3(CO)6(PH3)3]
For QM/MM calculations that included dihydrogen or hydride ligands,
the additional hydrogen atoms were also included in the QM region.
An example of a Gaussian 03 input file for a QM/MM geometry
optimization of Re2Pt3(CO)6(PBut3)3(µ-H)6 is provided in the Supporting
Information. It should be noted that the choice of Re2Pt3(CO)6(PH3)3
as the region for the high level of theory for the QM/MM model and
as the chemical model for the QM model facilitates a comparison of
the results obtained by the two approaches in terms of how the steric
profile of the But groups influences the potential energy surface.
Results and Discussion
(22) (a) Burke, K.; Perdew, J. P.; Wang, Y. In Electronic Density Functional
Theory: Recent Progress and New Directions; Dobson, J. F., Vignale, G.,
Das, M. P., Eds.; Plenum: New York, 1998; pp 81-111. (b) Perdew, J. P.
In Electronic Structure of Solids ’91; Ziesche, P., Eschrig, H., Eds.;
Akademie Verlag: Berlin, 1991; p 11. (c) Perdew, J. P.; Chevary, J. A.;
Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.; Fiolhais, C.
Phys. ReV. B 1992, 46, 6671-6687. (d) Perdew, J. P.; Chevary, J. A.;
Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.; Fiolhais, C.
Phys. ReV. B 1993, 48, 4978. (e) Perdew, J. P.; Burke, K.; Wang, Y. Phys.
ReV. B 1996, 54, 16533-16539.
The new compound Pt3Re2(CO)6(PBut3)3, 1, was obtained in
18% yield as the only heterometallic product from the reaction
of Re2(CO)10 with Pt(PBut3)2 in octane solvent at reflux
(125 °C) for 3 h. Significant amounts of Re2(CO)10 were
recovered from the reaction after this period, but longer reaction
periods do not increase the yield of 1. The only other product
formed in this reaction was a monophosphine derivative of Re2-
(CO)10, Re2(CO)9(PBut3), obtained in 11% yield.30 The Re2-
(CO)9(PBut3) product was characterized crystallographically. Its
structure is similar to that of Re2(CO)10 with the exception of
a PBut3 ligand that occupies a coordination site trans to the
Re-Re bond on one of the two rhenium atoms; see the
Supporting Information for further details, Figure S1.
(23) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270-283. (b) Wadt,
W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284-298. (c) Hay, P. J.; Wadt,
W. R. J. Chem. Phys. 1985, 82, 299-310.
(24) Couty, M.; Hall, M. B. J. Comput. Chem. 1996, 17, 1359-1370.
(25) Basis sets were obtained from the Extensible Computational Chemistry
Environment Basis Set Database, Version 02/25/04, as developed and
distributed by the Molecular Science Computing Facility, Environmental
and Molecular Sciences Laboratory, which is part of the Pacific Northwest
Laboratory, P.O. Box 999, Richland, WA 99352, and funded by the U.S.
Department of Energy. The Pacific Northwest Laboratory is a multiprogram
laboratory operated by Battelle Memorial Institute for the U.S. Department
of Energy under Contract DE-AC06-76.RLO 1830. Contact Karen Schu-
chardt for further information.
(26) Check, C. E.; Faust, T. O.; Bailey, J. M.; Wright, B. J.; Gilbert, T. M.;
Sunderlin, L. S. J. Phys. Chem. A 2001, 105, 8111-8116.
(28) (a) Maseras, F.; Morokuma, K. J. Comput. Chem. 1995, 16, 1170-1179.
(b) Humbel, S.; Sieber, S.; Morokuma, K. J. Chem. Phys. 1996, 105, 1959-
1967. (c) Matsubara, T.; Sieber, S.; Morokuma, K. Int. J. Quantum Chem.
1996, 60, 1101-1109. (d) Svensson, M.; Humbel, S.; Froese, R. D. J.;
Matsubara, T.; Sieber, S.; Morokuma, K. J. Phys. Chem. 1996, 100, 19357-
19363. (e) Svensson, M.; Humbel, S.; Morokuma, K. J. Chem. Phys. 1996,
105, 3654-3661. (f) Dapprich, S.; Koma´romi, I.; Byun, K. S.; Morokuma,
K.; Frisch, M. J. J. Mol. Struct.: THEOCHEM. 1999, 461-462, 1-21.
(g) Vreven, T.; Morokuma, K. J. Comput. Chem. 2000, 21, 1419-1432.
(h) Vreven, T.; Morokuma, K.; Farkas, O¨ .; Schlegel, H. B.; Frisch, M. J.
J. Comput. Chem. 2003, 24, 760-769.
(27) (a) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54,
724-728. (b) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys.
1972, 56, 2257-2261. (c) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta
1973, 28, 213-222. (d) Petersson, G. A.; Al-Laham, M. A. J. Chem. Phys.
1991, 94, 6081-6090. (e) Petersson, G. A.; Bennett, A.; Tensfeldt, T. G.;
Al-Laham, M. A.; Shirley, W. A.; Mantzaris, J. J. Chem. Phys. 1988, 89,
2193-2218. (f) Foresman, J. B.; Frisch, Æ. Exploring Chemistry with
Electronic Structure Methods, 2nd ed.; Gaussian, Inc.: Pittsburgh, PA;
1996; p 110. The 6-31G(d′,p′) basis set has the d polarization functions
for C, N, O, and F taken from the 6-311G(d) basis set, instead of those
which use the exponent with the original arbitrarily assigned value of 0.8
used in the 6-31G(d) basis set.
(29) Rappe´, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A.; Skiff,
W. M. J. Am. Chem. Soc. 1992, 114, 10024-10035.
(30) See the Supporting Information for additional data on Re2(CO)9(PBut3).
9
990 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Multiple Additions of Hydrogen to a Pt−Re Cluster
A R T I C L E S
Figure 4. An enlarged version of the 31P{1H} NMR spectrum for compound
1, showing the 195Pt satellites. The center resonance A is truncated at this
level of expansion.
Figure 3. ORTEP diagram of the molecular structure of 1 showing 30%
thermal ellipsoid probability. The methyl groups on the PBut3 ligands are
omitted for clarity.
Table 3. Comparison of the Metal-Metal Bond Distances (Å) for
Compounds 1-4
1
2
3
4
Re(1)-Pt(1)
Re(1)-Pt(2)
Re(1)-Pt(3)
Re(2)-Pt(1)
Re(2)-Pt(2)
Re(2)-Pt(3)
av Re-Pt
Pt(1)-Pt(2)
Pt(1)-Pt(3)
Pt(2)-Pt(3)
av Pt-Pt
2.6275(7)
2.6728(7)
2.6431(7)
2.6690(7)
2.6158(7)
2.6617(7)
2.6483(7)
2.7202(7)
2.7104(7)
2.7369(7)
2.7225(7)
2.6915(8)
2.7477(9)
2.7343(8)
2.7507(9)
2.6975(9)
2.7162(9)
2.7230(9)
2.6771(8)
2.7521(8)
2.7497(8)
2.7263(8)
2.8132(4)
2.8183(4)
2.7887(4)
2.8132(4)a
2.8183(4)a
2.7887(4)a
2.8067(4)
2.7165(5)
2.7366(5)
2.7406(5)
2.7312(5)
2.9202(3)
2.8989(3)
2.9004(3)
2.9170(3)
2.9134(3)
2.9050(3)
2.9092(3)
2.7026(3)
2.7095(3)
2.7259(3)
2.7127(3)
Figure 5. Illustration of the four magnetic isomers present in compound
1. Pt and Pt* represent NMR-inactive Pt isotopes, spin 0, and the NMR-
active isotope 195Pt, spin 1/2, respectively.
) 2.7104(7) Å, Pt(2)-Pt(3) ) 2.7369(7) Å, av ) 2.7225(7) Å,
are longer than those in 5, 2.6114(7), 2.6078(9), and 2.5930(9)
Å. The IR spectrum of 1 shows only two CO stretching
absorptions: 1985 (s) and 1892 (s) cm-1. On the basis of D3h
symmetry, the higher energy absorption can be assigned to an
A2′′ mode and the lower energy absorption to an E′ mode. The
1H NMR spectrum exhibits only one resonance for the methyl
a The bond distances are generated by symmetry equivalence in the
molecule.
Compound 1 was isolated in a pure form by TLC with some
loss due to decomposition. The compound was characterized
by a single-crystal X-ray diffraction analysis, and an ORTEP
diagram of its molecular structure is shown in Figure 3.
Compound 1 contains a trigonal bipyramidal cluster of five
metal atoms, three platinum and two rhenium. The three
platinum atoms occupy the trigonal plane. The two rhenium
atoms occupy the apical positions. Each platinum atom contains
one PBut3 ligand, and each rhenium atom contains three linear
terminal carbonyl ligands. Ignoring the tert-butyl groups, the
molecule has approximate D3h symmetry. According to its
formula, compound 1 contains only 62 valence electrons and
is electronically unsaturated by 10 valence electrons according
to conventional electron counting theories.31,32 Selected bond
distances for 1 are given in Table 3. The six Re-Pt bonds are
unusually short. These bond distances range from 2.6158(7) to
2.6728(7) Å, av ) 2.6483(7) Å. The Re-Pt bond distance in
the linear compound PtRe2(CO)12 is 2.8309(5) Å.33 The Pt-Re
bond distances are similar to those in the electronically
unsaturated tetrahedral Pt3Re complex [Pt3Re(CO)3(µ-dppm)3]+,
5, which are 2.6839(7), 2.6488(8), and 2.6850(8) Å.34 The
Pt-Pt distances in 1, Pt(1)-Pt(2) ) 2.7202(7) Å, Pt(1)-Pt(3)
3
groups, δ ) 1.63 ppm (d, J(P,H) ) 13 Hz, 81 H, CH3). The
31P{1H} NMR spectrum also shows only one resonance,
δ )109.23 ppm, with satellites due to 195Pt couplings; see Figure
4. The various 195Pt isotopomers of 1, a-d, are shown in Figure
5. Isomer a with no 195Pt NMR active isotopes present (spin 0)
gives rise to a single resonance (A in Figure 4) at 109.23 ppm.
Isomer b with only one 195Pt NMR nucleus (spin 1/2) gives
rise to two doublets centered at 108.2 and 110.3 ppm
3
[2J(Pt,Pa) ) 332 Hz, J(Pa,Pb) ) 44 Hz] (B in Figure 4) and
two triplets at 96.4 and 122.1 ppm [1J(Pt,Pb) ) 4162 Hz,
3J(Pa,Pb) ) 44 Hz] (C in Figure 4). Isomer c gives rise to the
resonances at 95.2 and 123.2 ppm [1J(Pt,Pb) ) 4531 Hz] (D in
Figure 4) and at 105.4 and 113.1 ppm [2J(Pt,Pa) ) 1253 Hz]
(E in Figure 4). The satellites for isomer d are not seen due to
its very low relative abundance. The 195Pt satellite pattern for 1
is similar to that observed for the compound [Pt(PBut3)(CO)]3.35
Because of its high degree of electronic unsaturation, the reaction
of 1 with hydrogen was investigated.
Compound 1 reacts with hydrogen at 25 °C/ 1 atm to yield
a series of three products: Pt3Re2(CO)6(PBut3)3(µ-H)2, 2, Pt3-
(31) A trigonal bipyramidal cluster of five metal atoms in which each of the
metal atoms formally has an 18-electron configuration will have a total of
72 valence electrons.32
(32) Mingos, D. M. P. Introduction to Cluster Chemistry; Prentice Hall:
Engelwood Cliffs, NJ, 1990; Chapter. 2.
(33) Urbancic, M. A.; Wilson, S. R.; Shapley, J. R. Inorg. Chem. 1984, 23,
2954-2958.
(34) (a) Xiao, J.; Kristof, E.; Vittal, J. J.; Puddephatt, R. J. J. Organomet. Chem.
1995, 490, 1-6. (b) Xiao, J.; Puddephatt, R. J.; Manojlovic-Muir, L.; Muir,
K. W.; Torabi, A. A. J. Am. Chem. Soc. 1994, 116, 1129-1130.
(35) Goel, R. G.; Ogini, W. O.; Srivastava, R. C. J. Organomet. Chem. 1981,
214, 405-417.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 991

A R T I C L E S
Adams et al.
Figure 7. 31P{1H} NMR spectrum for compound 4.
Figure 6. ORTEP diagram of the molecular structure of Pt3Re2(CO)6-
(PBut3)3(µ-H)6, 4, showing 50% thermal ellipsoid probability. The methyl
groups on the PBut3 ligands have been omitted for clarity.
Re2(CO)6(PBut3)3(µ-H)4, 3, and Pt3Re2(CO)6(PBut3)3(µ-H)6, 4,
formed by the addition of 1, 2, and 3 equiv of H2 to 1. A series
of IR spectra which show the progress of the reaction as a
function of time as monitored in the CO stretching region are
presented in a stacked form in Figure 1.
After a period of 4 h, the air-stable compound 4 was isolated
by TLC in 90% yield. Compound 4 was characterized by a
1
combination of IR, H NMR, 31P NMR, mass spectra and a
single-crystal X-ray diffraction analysis. An ORTEP diagram
of the molecular structure of 4 is shown in Figure 6. Compound
4 is structurally similar to 1 except that it contains six additional
hydrido ligands. All six hydrido ligands were located and refined
in the structural analysis. There is one hydrido ligand bridging
each of the six Pt-Re bonds of the Pt3Re2 trigonal bipyramidal
cluster. As expected, the bridging hydrido ligands produce an
increase in the length of the Pt-Re bonds relative to those in
1, Pt(1)-Re(1) ) 2.9202(3) Å, Pt(2)-Re(1) ) 2.8989(3) Å,
Pt(3)-Re(1) ) 2.9004(3) Å, Pt(1)-Re(2) ) 2.9170(3) Å, Pt-
(2)-Re(2) ) 2.9134(3) Å, Pt(3)-Re(2) ) 2.9050(3) Å, av )
2.9092(3) Å, an average increase of 0.26 Å.36 The Pt-Pt
distances in 4 are similar to those in 1: Pt(1)-Pt(2) ) 2.7026-
(3) Å, Pt(1)-Pt(3) ) 2.7095(3) Å, Pt(2)-Pt(3) ) 2.7259(3)
Å, av ) 2.7127(3) Å. Because of their close structural similarity,
the IR spectrum of 4 in the CO region is very similar to that of
1, except that the two absorptions have been shifted to higher
Figure 8. ORTEP diagram of the molecular structure of 2 showing 30%
thermal ellipsoid probability. The methyl groups on the PBut3 ligands have
been omitted for clarity. The hydrido ligands were not located in the
structural analysis.
Compound 3 can be isolated in a low yield by stopping the
reaction before the conversion to 4 is complete and then
separating by TLC. However, compound 2 is unstable to TLC
and cannot be obtained in a significant amount by this method.
A better route to 2 is via the photoelimination of hydrogen from
4, see below, which provides 2 as the principal product.
Compounds 2 and 3 were both characterized by a combination
of IR, 1H and 31P NMR, mass spectra and single-crystal X-ray
diffraction analyses. The arrangements of the non-hydrogen
atoms in 2 and 3 are virtually the same as those of 1 and 4; see
Figures 8 and 9, except that the Pt-Re bonds become progres-
sively and significantly longer as the number of hydrido ligands
is increased. The hydrido ligands were not located in the
structural analyses of 2 and 3, presumably due to disorder, but
they are believed to bridge the Pt-Re bonds because these bonds
undergo progressive lengthening relative to those of 1.36 In
particular, the average Pt-Re bond length in 2 is 2.7230(9) Å,
and the average Pt-Re bond length in 3 is 2.8067(4) Å; see
Table 3. The Pt-Pt bond distances in 2, 2.7263(8) Å, and 3,
2.7312(5) Å, are only slightly longer than those in 1 and 4.
On the basis of their IR and NMR spectra and the energies
of structures calculated using DFT calculations, it is proposed
that compounds 2 and 3 both have structures with C2V symmetry
as shown in Figure 10. Due to their lower symmetry, the IR
spectra of 2 and 3 both show three CO stretching absorptions
at 1992 (vs), 1905 (s), and 1893 (s) cm-1 for 2 and 2000 (vs),
1910 (s), and 1901 (s) cm-1 for 3. The presence of two lower
1
energy: 2009 (s) and 1914 (vs) cm-1. The H NMR spectrum
3
of 4 is similar to that of 1, δ ) 1.50 (d, J(P,H) ) 13 Hz, 81
H, CH3), except for the presence of a new high-field multiplet
at δ ) -4.34 with coupling to the phosphorus and platinum
atoms (1J(Pt,H) ) 723 Hz, 2J(P,H) ) 4 Hz, 6H) that is attributed
to the six equivalent hydrido ligands. The multiplicity is a result
of complex second-order coupling effects that are due to the
magnetic nonequivalence of the hydrido ligands. The 31P{1H}
NMR spectrum of 4 is very similar to that of 1. Isomer a (see
Figure 5) gives rise to a single resonance at 111.42 ppm (A in
Figure 7). Isomer b gives rise to two doublets centered at 110.7
3
and 112.1 ppm [2J(Pt,Pa) ) 222 Hz, J(Pa,Pb) ) 56 Hz] (B in
Figure 7) and two triplets at 102.2 and 120.6 ppm [1J(Pt,Pb) )
2986 Hz, 3J(Pa,Pb) ) 56 Hz] (C in Figure 7), and isomer c gives
rise to two doublets at 101.5 and 121.3 ppm [1J(Pt,Pb) ) 3208
Hz, 3J(Pa,Pb) ) 56 Hz] (D in Figure 7). The expected lines for
Pa as well as those for isomer d are not seen due to their very
low relative intensity.
(36) Teller, R. G.; Bau, R. Struct. Bonding 1981, 44, 1-82.
9
992 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Multiple Additions of Hydrogen to a Pt−Re Cluster
A R T I C L E S
Figure 11. (a) 1H NMR spectrum for compound 2. (b) Simulated 1H NMR
spectrum for the second-order spin system A2MX2 for compound 2.
(c) Simulated 1H NMR spectrum for the second-order spin system
AA′MXX′ for compound 2.
Figure 9. ORTEP diagram of the molecular structure of 3 showing 30%
thermal ellipsoid probability. The methyl groups on the PBut3 ligands have
been omitted for clarity. The hydrido ligands were not located in the
structural analysis.
Figure 12. (a) Observed 31P{1H} NMR spectrum for compound 3.
(b) Simulated 31P{1H} NMR spectrum for the AB2 second-order spin system
for compound 3.
Figure 10. Structures for compounds 2 and 3 showing the proposed
locations for the hydrido ligands based on spectral analyses and DFT
calculations.
energy absorptions in both compounds can be attributed to a
splitting of the degenerate lower energy E-mode that exists in
the higher symmetry structure of 1. The CO absorptions for
1-4 shift progressively to higher energy as the number of
hydrido ligands is increased.
1
The H NMR spectrum of 2 shows two doublets in a 1:2
ratio for the two inequivalent PBut3 groups, δ ) 1.56 ppm (d,
3
3J(P,H) ) 13 Hz, 27 H, CH3), 1.52 ppm (d, J(P,H) ) 13 Hz,
54 H, CH3), and appropriately the 31P{1H} NMR spectrum
shows a doublet and a triplet in a 2:1 ratio representing the
Figure 13. Space-filling model of the structure of 1 showing an opening
between the tert-butyl groups of the P(But)3 ligands above the Pt-Pt bond.
Pt atoms are shown in blue, Re in green, O in red, C in gray, and H in
white.
3
A2X first-order spin system, δ ) 111.1 ppm (d, J(P,P) ) 53
3
Hz), 105.6 ppm (t, J(P,P) ) 53 Hz). The resonance of the
hydrido ligands in 2 shows coupling to both Pt and P at δ )
inequivalent PBut3 ligands, δ ) 1.56 ppm (d, 3J(P,H) ) 13 Hz,
27 H, CH3), 1.52 ppm (d, 3J(P,H) ) 13 Hz, 54 H, CH3);
however, the 31P{1H} NMR spectrum does not show the
expected doublet and triplet pattern in a 2:1 ratio expected for
an AX2 spin system as found in 2. The 31P{1H} NMR spectrum
of 3, shown in Figure 12a, is instead, a classical AB2 second-
order spin system, ignoring the couplings to 195Pt, due to the
small chemical shift difference between the two phosphorus
resonances.37The simulated spectrum shown in Figure 12b was
obtained by using the chemical shifts for PA, 110.65 ppm, and
PB, 108.41 ppm. The 31P-31P coupling constant used for the
1
2
3
-5.91 (dt, J(Pt,H) ) 821 Hz, J(P,H) ) 7.6 Hz, J(P,H) )
1.4 Hz, 2H); see Figure 11a. Spin simulation for the hydride
resonance was carried out by using the values reported above.
The simulated spectrum shown in Figure 11b represents the A2-
MX2 spin system. For a comparison the pattern representing
an AA′MXX′ spin system is shown in Figure 11c. For the two
hydrido ligands bonded to the same platinum atom, the
simulated spectrum should be consistent with the A2MX2 pattern
in Figure 11b. However, if the two hydride ligands were bonded
to different platinum atoms, the expected spin system would
be the AA′MXX′ pattern shown in Figure 11c. It is clear that
the observed spectrum is most consisitent with a structure having
the two hydrido ligands bonded to the same platinum atom.
3
1
simulation was J(PA,PB) ) 55.0 Hz. The H NMR spectrum
of 3 shows only a single hydride resonance with coupling to
both platinum and phosphorus, -7.42 ppm (m, 1J(Pt,H) ) 760
Hz, 4 H). This multiplet is a very complex second-order pattern
due to the magnetic nonequivalence of the hydride ligands and
was not simulated. The phosphorus decoupled 1H NMR
spectrum of 3 shows only a sharp singlet, except for the
1
As in compound 2, the H NMR spectrum of 3 shows two
doublets in a 1:2 ratio for the methyl groups of the two
(37) Pavia, D. L.; Lampman, G. M.; Kriz, G. S. Introduction to Spectroscopy,
2nd ed.; Harcourt Brace: Orlando, FL, 1996; Chapter 5.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 993

A R T I C L E S
Adams et al.
Figure 14. Side and top views of the A′ HOMO, the degenerate pairs of the E′′ LUMO, the E′ LUMO + 1, and the A′′ LUMO + 2 generated from a
single-point B3PW91/BS1 calculation of Re2Pt3(CO)6(PH3)3 with approximate D3h symmetry of the non-hydrogen atoms (isovalue 0.04).
couplings to 195Pt for the hydride resonance, indicating that the
multiplet pattern of the central resonance is due to complex 31P-
1H couplings.
The addition of D2 to 1 is significantly slower than the
addition of H2. A KIE of approximately 1.3(1) for the formation
of 2-d2 from 1 was determined by following the addition by IR
spectroscopy. It appeared that similar isotopes effects were
operative for the subsequent conversions of 2-d2 to 3-d4 and
3-d4 to 4-d6 by reactions with D2 in the overall conversions,
but these reactions were not measured separately.
The rate of addition of H2 to 1 was also followed in a hexane
solution containing a 20-fold excess of free PBut3. The reaction
was followed by IR spectroscopy of the CO absorptions of the
four compounds. There was no discernible difference in the rates
of formation of 2-4 in the presence or absence of the free PBut3.
This is interpreted as being consistent with an addition mech-
anism that does not involve the dissociation of a PBut3 ligand
Figure 15. Schematic representation of the rotation of the platinum d
orbitals that is observed in the B3PW91/BS1 HOMO of Re2Pt3(CO)6(PH3),
prior to the addition of the hydrogen molecule. Examination of
which decreases the Pt-Pt triangle edge antibonding character and increases
a space-filling model of 1 shows that there is probably enough
space between the tert-butyl groups in the region above the Pt-
Pt bonds in the plane of the Pt3 triangle for a H2 molecule to
penetrate the ligand sphere and approach the metal atoms; see
Figure 13.
Electronic Structure of 1. To understand the nature of the
electronic unsaturation in 1, we analyzed the molecular orbitals
generated from B3PW91/BS1 calculations. To simplify the MO
analysis, the QM model, Re2Pt3(CO)6(PH3), was used as the
model for 1, and furthermore, the input structure for the single-
point calculation was set up by basing the coordinates for the
Re, Pt, P, C, and O atoms on the crystal coordinates and then
adding the H atoms to generate the PH3 fragments such that
the Pt-Pt-Pt bonding character in the center of the Pt3 triangle. The two
Re(CO)3 units and hydrogen atoms of the PH3 ligands are not shown.
the molecule had precise C3h symmetry, while the non-hydrogen
atoms had approximate D3h symmetry. Similar to previously
reported Fenske-Hall calculations,16 the DFT molecular orbital
analysis revealed the existence of five low-lying unoccupied
molecular orbitals (LUMOs) in the form of a degenerate pair
at -2.5 eV having E′′ symmetry, the LUMO, a second
degenerate pair, LUMO + 1, at -2.1 eV having E′ symmetry,
and one nondegenerate orbital at -1.6 eV which has A2′′
symmetry; see Figure 14. These five orbitals could accommodate
the anticipated 10 “missing” valence bonding electrons. The
9
994 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Multiple Additions of Hydrogen to a Pt−Re Cluster
A R T I C L E S
Scheme 3. Proposed General Pathway for the Addition of 3 equiv of Dihydrogen to 1
highest occupied molecular orbital (HOMO) of A1′ symmetry
lies at -5.8 eV. This orbital is dominated by Pt-Pt interactions
in the Pt3 plane, and interestingly, the Pt d orbitals in the Pt3
plane are “twisted” to generate a bonding region in the center
of the Pt3 triangle and to reduce the antibonding component of
the orbital along the edges of the Pt-Pt triangle (Figures 14
and 15).
Calculated Structures of the 0H, 2H, 4H, and 6H Com-
plexes with the QM and QM/MM Models. Three equivalents
of dihydrogen can be added to the 1 cluster to generate 4. IR
analysis indicates that the addition of each molecule of
dihydrogen occurs sequentially, and furthermore, the addition
of 20 equiv of PBut3 to the solution does not slow the reaction,
which suggests that PBut3 dissociation does not occur during
the addition of dihydrogen. Thus, a plausible pathway for the
addition of 3 equiv of dihydrogen to 1 involves three successive
steps in which each equivalent of dihydrogen may coordinate
to the cluster and is activated and converted into two hydride
ligands (Scheme 3).
geometries that resulted from using the QM and QM/MM
models are similar to the X-ray structures and are presented in
Figures 16 and 17.
Because the positions of the hydrido ligands were not located
in the X-ray analysis of 2H and 4H, the structures and relative
energies of a number of isomers that contained two and four
hydrido ligands were calculated using the QM and QM/MM
models to locate the preferential binding positions of the hydrido
ligands.38 The lowest energy isomers of the 2H complex for
both the QM and QM/MM models contained the two hydrido
ligands in bridging positions along the two Re-Pt bonds from
the same platinum atom. All other isomers of 2H that were
located that did not contain the two hydride ligands in bridging
The positions of the hydrido ligands were located in the X-ray
analysis of 4 (6H), and IR and NMR evidence suggests C2V
symmetry for the hydrido ligands in the dihydride 2 (2H) and
the tetrahydride 3 (4H) complexes, but the positions of the
hydrido ligands were not located in the X-ray analysis of 2 and
3 complexes, and there is no experimental information regarding
the structures of potential dihydrogen intermediates (0H-H2, 2H-
H2, 4H-H2). Therefore, computational methods were used to
investigate possible structures of the intermediates in this
pathway, with the ultimate goal of providing an enhanced
understanding of the process of hydrogen activation and storage
on 1.
Initially, the geometries and relative energies of the starting
complex 1 (0H) and the overall product of the addition of 3
equiv of hydrogen 4 (6H) were optimized using a QM model,
which employed the B3PW91/BS1 level of theory and a
truncated model in which PH3 was used in place of PBut3, and
a QM/MM model which employed the ONIOM(B3PW91/BS1:
UFF) level of theory where the But groups were modeled with
the UFF molecular mechanics level of theory. The initial
geometries that were used as starting points for the geometry
optimizations for the 0H and 6H complexes were based on the
experimental X-ray structures, and the subsequent optimized
Figure 16. QM-model-optimized structures and relative energies of the
0H, 2H, 4H, and 6H complexes. The relative energies were compared to
the energy of the lowest energy 0H structure + 3H2, 0.0 kcal/mol. Key:
platinum (green), rhenium (beige), phosphorus (purple), oxygen (red), carbon
(gray), hydrogen (white).
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 995

A R T I C L E S
Adams et al.
Figure 18. Correlation energy diagram for the 0H, 2H, 4H, and 6H
complexes.
Correlation Energy Diagram. With the information regard-
ing the preferred locations of the hydride ligands in the 2H and
4H complexes in hand, we sought to identify the frontier orbitals
that are affected upon each addition of dihydrogen to the cluster
of 1. For this molecular orbital analysis, the smaller QM model
was used, and to simplify further the analysis of the molecular
orbitals, the 6H complex was constrained to C3h symmetry and
the 2H and 4H complexes were constrained to Cs symmetry.
The 2H, 4H, and 6H complexes were then optimized with the
symmetry constraints, and the resulting B3PW91/BS1 orbitals
were examined (Figure 18). The orbitals for the 0H complex
used in the analysis were obtained as described previously (vide
supra).
Upon addition of the first equivalent of dihydrogen, the E′′
and E′ orbitals are split due to the reduction of symmetry. The
two added electrons are added to one of the orbitals, A′′, derived
from the E′′ set. The electrons from the addition of the second
equivalent of dihydrogen to form 4H are added to the second
A′′ orbital from the original E′′ set and complete the filling of
that set. Subsequently, the third equivalent of dihydrogen
provides two electrons which fill another A′′ orbital from 4H,
so that 6H contains filled E′′ and A′′ orbitals that were originally
unoccupied in 0H (Figure 18). Notably, the 0H E′′ and A′′
orbitals that become filled upon the addition of the 3 equiv of
H2 have bonding character along Re-Pt edges, while the 0H
E′ orbitals, which remain unoccupied throughout this series of
complexes, do not have bonding character along the Re-Pt
bonds (Figure 14).
Computational Investigation of H2 Addition and Binding.
Another aspect of the reactivity of these RePt cluster complexes
toward dihydrogen that we sought to explain by using the QM
and QM/MM computational models was the nature of the
addition and coordination of H2 to 0H, 2H, and 4H to generate
the 0H-H2, 2H-H2, and 4H-H2 complexes, respectively, that
might represent intermediates in the hydrogen activation reac-
tions.
With the QM model, optimized structures were located for
H2 binding to both Pt and Re atoms in the 0H and 2H
complexes, producing 0H-H2 and 2H-H2 structures, while for
the 4H complex, an optimized structure, 4H-H2, was only
located for H2 binding to Re. In terms of free energy, H2 binding
to the 0H, 2H, and 4H complexes to generate the 0H-H2, 2H-
H2, and 4H-H2 complexes is endoergic in all cases (Figures 19
and 20 and Figures S2-S4 in the Supporting Information). In
Figure 17. QM/MM-model-optimized structures and relative energies of
the 0H, 2H, 4H, and 6H complexes. The relative energies were compared
to the energy of the lowest energy 0H structure + 3H2, 0.0 kcal/mol. Key:
platinum (green), rhenium (beige), phosphorus (purple), oxygen (red), carbon
(gray), hydrogen (white).
positions along the two Re-Pt bonds from the same platinum
atom were higher in free energy by more than 12 and 16 kcal/
mol for the QM and QM/MM models, respectively. Similarly,
the lowest energy isomers of the 4H complex for both the QM
and QM/MM models contained all four of the hydrido ligands
in bridging positions along Re-Pt bonds, with two hydrides
attached to each of two Pt atoms (Figures 16 and 17). All other
isomers of 4H that were located that did not contain all four of
the hydride ligands in bridging positions along Re-Pt bonds,
with two hydrides attached to each of two Pt atoms, were higher
in free energy by more than 10 and 22 kcal/mol for the QM
and QM/MM models, respectively. The QM and QM/MM
structures of the lowest free energy 2H and 4H isomers are
consistent with the NMR experiments which indicate that the
2H and 4H complexes display approximate C2V symmetry in
solution.
The lowest relative energy structures of the 0H, 2H, 4H, and
6H complexes that were optimized using the QM model and
the QM/MM model are presented in Figures 16 and 17,
respectively. For the QM model (Figure 16), the addition of
each equivalent of H2 to the bimetallic Re2Pt3 cluster is
exergonic, such that compared to that of 0H#1 + 3H2, the
relative gas-phase free energies of 2H-sPt#1 + 2H2, 4H-sPt#1
+ H2, and 6H-sPt#1 are -15.4, -26.0, and -35.1 kcal/mol,
respectively. For the QM/MM model (Figure 17), the addition
of each equivalent is also exergonic, such that compared to that
of 0H#1 + 3H2, the relative gas-phase free energies of 2H-
sPt#1 + 2H2, 4H-sPt#1 + H2, and 6H-sPt#1 are -0.6, -6.0,
and -15.4 kcal/mol, respectively. The calculated relative free
energies for both the QM and QM/MM models are consistent
with the experimental observations that compounds 2 (2H), 3
(4H), and 4 (6H) are readily formed upon the reaction of 1
(0H) with excess H2.
(38) Details of the search for alternative isomers of the 2H and 4H complexes
are provided in the Supporting Information.
9
996 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Multiple Additions of Hydrogen to a Pt−Re Cluster
A R T I C L E S
Figure 20. QM-optimized dihydrogen complexes, transition states for H2
activation, and resulting local minima for the 0H-H2-Pt (a), 2H-H2-Pt-
trans (b), and 2H-H2-Pt-cis (c) series of complexes. Relative energies of
the 0H-H2-Pt series (a) are compared to the energy of the lowest energy
0H complex + H2, 0.0 kcal/mol. Relative energies of the 2H-H2-Pt-trans
(b) and 2H-H2-Pt-cis (c) series are compared to the energy of the lowest
energy 2H complex + H2, 0.0 kcal/mol. Key: platinum (green), rhenium
(beige), phosphorus (purple), oxygen (red), carbon (gray), hydrogen (white).
Figure 19. QM-optimized dihydrogen complexes, transition states for H2
activation, and resulting local minima for the 0H-H2-Re (a), 2H-H2-Re
(b), and 4H-H2-Re (c) series of complexes. Relative energies of the 0H-
H2-Re series (a) are compared to the energy of the lowest energy 0H
complex + H2, 0.0 kcal/mol. Relative energies of the 2H-H2-Re series (b)
are compared to the energy of the lowest energy 2H complex + H2, 0.0
kcal/mol. Relative energies of the 4H-H2-Re series (c) are compared to the
energy of the lowest energy 4H complex + H2, 0.0 kcal/mol. Key: platinum
(green), rhenium (beige), phosphorus (purple), oxygen (red), carbon (gray),
hydrogen (white).
In the process of binding H2 to the 0H and 2H complexes to
generate the corresponding 0H-H2 and 2H-H2 complexes, with
both the QM and QM/MM models, the trigonal bipyramidal
structure of the core Re2Pt3 unit generally remains intact.
However, with the QM model for H2 binding to the Re atom of
the 4H complex that generates 4H-H2-Re#1 (Figure 19c), the
trigonal bipyramidal structure of the Re2Pt3 is significantly
distorted such that the complex has opened to facilitate H2
binding. In the QM/MM model of the 4H-H2-Re#1 complex
(Figure 21c), there is less distortion of the Re2Pt3 from the
trigonal bipyramidal structure, which is likely because the steric
bulk of the tert-butyl group blocks the opening of the top of
the Re2Pt3 to the extent observed with the smaller QM model.
Computational Investigation of H2 Activation. In addition
to locating the structures of the H2 adducts that were described
above, transition states were located for dihydrogen activation
(cleavage) from the XH-H2-Re complexes for both the QM
model (center column, Figure 19) and QM/MM model (center
column, Figure 21). With the QM model, the free energies for
cleavage are 2.1, 5.2, and 3.3 kcal/mol higher than the energies
of the preceding dihydrogen complexes for the 0H, 2H, and
4H systems, respectively. Although the free energies of the
products of the cleavage (last column, Figure 19) are below
the TS free energies, these products are still higher in free energy
than the reactant cluster + H2 (the energy reference).
these complexes H2 binds to Pt more strongly than to Re
(compare the first columns of Figures 19 and 20). In spite of
the result that H2 binding to the Pt atoms generates more stable
complexes, repeated attempts to locate a 4H-H2 complex in
which H2 was bound to Pt failed.
With the QM/MM model, optimized structures were located
for 0H-H2, 2H-H2, and 4H-H2 complexes in which H2 is bound
to a Re atom (see Figure 21 and Figures S5-S7 in the
Supporting Information). Including the steric bulk of the But
groups further disfavors H2 binding by 19, 14, and 7 kcal/mol
for 0H, 2H, and 4H, respectively (compare the first columns
of Figures 19 and 21). Despite repeated attempts, freely
optimized structures in which H2 was bound to the Pt were not
located for the QM/MM model. Relaxed potential energy scans
with loosened convergence criteria indicated that the electronic
energy increased as the H2-Pt distance was decreased. This
suggests that 0H-H2, 2H-H2, and 4H-H2 complexes in which
the H2 is closely bound to the Pt either are not local minima or
are very shallow local minima on the QM/MM potential energy
surface.
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 997

A R T I C L E S
Adams et al.
Figure 21. QM/MM-optimized dihydrogen complexes, transition states for H2 activation, and resulting local minima in the 0H-H2-Re (a), 2H-H2-Re (b),
and 4H-H2-Re (c) series of complexes. Relative energies of the 0H-H2-Re series (a) are compared to the energy of the lowest energy 0H complex + H2,
0.0 kcal/mol. Relative energies of the 2H-H2-Re series (b) are compared to the energy of the lowest energy 2H complex + H2, 0.0 kcal/mol. Relative
energies of the 4H-H2-Re series (c) are compared to the energy of the lowest energy 4H complex + H2, 0.0 kcal/mol. Key: platinum (green), rhenium
(beige), phosphorus (purple), oxygen (red), carbon (gray), hydrogen (white).
In spite of the larger steric bulk in the QM/MM model, the
hydrogen activation barriers are 2.2, 4.3, and 2.1 kcal/mol,
respectively, like those in the smaller model. Now, however,
the product of H2 cleavage (last column, Figure 21) shows
products that are much more stable than the XH-H2-Re and
even lower in free energy than the separated reactants.
yielded minima in which both resultant hydride ligands were
bridging Re-Pt bonds. Thus, it appears that including the
PBut3 ligands through the use of the QM/MM model pre-
cludes the formation of some shallow local minima that are
observed on the potential energy surface of the smaller QM
model. These results suggest that, in the experimental system,
once the dihydrogen enters the cluster, it is cleaved and the H
atoms readily migrate to the RePt edges, the lowest energy
positions.
For the small QM model, low-energy transition states for
hydrogen activation at Pt were also located (Figure 20), and
the resulting products all contain one hydride ligand that bridges
a Pt-Pt bond and another hydride ligand that is in a terminal
position on a Pt atom.
The analogous H2-Pt complexes and transition states for
dihydrogen activation on Pt could not be found for the larger
QM/MM model. However, QM/MM structures analogous to the
QM model local minima with H terminal and bridging Pt were
located and optimized (Figure 22). Although it is probable that
any QM/MM hydrogen activation transition states in which
A comparison of the hydrogen activation steps from the XH-
Re complexes + H2 that were obtained with the QM and QM/
MM models reveals that the higher barriers in the more sterically
crowded (experimental) system arise primarily because the
binding of dihydrogen is much more endoergic in the QM/MM
model than in the QM model. For both the QM and QM/MM
models, in all cases the transition states are between 2 and 6
kcal/mol higher in free energy than the corresponding dihydro-
gen complexes, which indicates that hydrogen activation should
occur readily after dihydrogen reaches Re centers in the 0H,
2H, and 4H complexes. In another notable difference, the
hydrogen activation steps calculated using the QM model for
the 0H-H2-Re, 2H-H2-Re, and 4H-H2-Re complexes generate
local minima as products in which both resulting hydride ligands
are not bridging Re-Pt bonds, whereas for the QM/MM model
of the corresponding complexes, analogous local minima were
not located. Instead, optimization of input structures that are
slightly on the product side of the QM/MM transition states
(39) Efremenko, I. J. Mol. Catal. A 2001, 173, 19-59.
(40) (a) Moc, J.; Musaev, D. G.; Morokuma, K. J. Phys. Chem. A 2000, 104,
11606-11614. (b) Efremenko, I.; German, E. D.; Sheintuch, M. J. Phys.
Chem. A 2000, 104, 8089-8096. (c) Wang, Y.; Cao, Z.; Zhang, Q. Chem.
Phys. Lett. 2003, 376, 96-102.
9
998 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007

Multiple Additions of Hydrogen to a Pt−Re Cluster
A R T I C L E S
Figure 24. IR spectra in the CO absorption region showing the pro-
gressive transformation of 4 to 3 and then 2 as a function of time under the
influence of UV-vis irradiation. Reaction times t in hours are shown on
the left.
Figure 22. QM/MM-optimized structures of 2H, 4H, and 6H complexes
that contain a terminal hydride ligand attached to a Pt atom and another
hydride ligand that bridges a Pt-Pt bond. The relative energy of 2H-bPtPt-
tPt#1 is compared to the energy of the lowest energy 0H complex + H2,
0.0 kcal/mol. Relative energies of 4H-bPtPt-tPt-cis#1 and 4H-bPtPt-tPt-
trans#1 are compared to the energy of the lowest energy 2H complex +
H2, 0.0 kcal/mol. The relative energy of 6H-bPtPt-tPt#1 is compared to
the energy of the lowest energy 4H complex + H2, 0.0 kcal/mol. Key:
platinum (green), rhenium (beige), phosphorus (purple), oxygen (red), carbon
(gray), hydrogen (white).
for the nature of hydrogen activation on metal surfaces39 and
have been invoked in multicenter activations of hydrogen on
small ligand-free clusters.40 Here, our results suggest that the
Re2Pt3(CO)6(PBut3)3 cluster system is able to add dihydrogen
without ligand dissociation or without the presence of an obvious
“vacant site” by utilizing these multiple metal atom hydrogen
activation pathways. Recent computational studies of the
complex [Rh6(PR3)6H16][BArF ] have indicated the existence
4 2
of bridging H2 coordination for some of the dihydrogen ligands
in this complex.4a
Elimination of Hydrogen from 4. Efforts were made to
induce the elimination of the hydrido ligands of 4 in the form
of hydrogen. There was no evidence for elimination of the
hydrido ligands even when 4 was heated to 175 °C. However,
under the influence of UV-vis irradiation at room temperature,
a solution of 4 in a Pyrex reaction flask was slowly converted
back to 3 and then 2 over a period of hours. Stacked plots of
these spectra as a function of time are shown in Figure 24. There
was very little formation of compound 1 under these conditions.
Spectroscopically, the conversion to 2 appears to be very high
after 7 h, but because of the instability of this compound, it
was possible to isolate it in pure form in only 27% yield.
However, this is the best method to obtain 2 in a pure form.
When irradiation of 4 was performed in a quartz reaction vessel,
it was possible to observe the formation of small amounts of 1,
but the yield was very low, and most of the compounds
decomposed under these conditions. We do not know why the
photoelimination of hydrogen from 2 is so ineffective. We can
only speculate that photoelimination of hydrogen from 2 is more
difficult than the photoelimination of hydrogen from 3 and 4
and the radiation that passes through the Pyrex glassware is
not sufficiently energetic to produce the hydrogen elimination
from 2. When a sample of 4 was irradiated similarly in a 10
mm NMR tube in toluene-d8 solvent, it was possible to observe
Figure 23. Schematic representation of the two-center (1) and three-center
(2) activation of dihydrogen observed in the transition states TS 0H-H2-
Pt#1 and TS 0H-H2-Re#1.
hydrogen activation occurs in the Pt3 plane would be higher in
free energy than the metastable hydride product complexes
shown in Figure 22, it is possible that there are transition states
which lead to these complexes and that these transition states
might not be much higher in free energy than the products, as
in the QM model. Thus, it is possible that hydrogen addition in
these Re2Pt3 clusters could be occurring through multiple
hydrogen activation pathways.
It is worth noting that, in the hydrogen activation transition
states that were located by using the QM and QM/MM models,
the cleavage of the H-H bond often involves more than one
metal atom. For example, two Pt atoms participate in the
heterolytic dihydrogen cleavage in the QM model transition state
TS 0H-H2-Pt#1 (Figure 20), and two Pt atoms and one Re atom
participate in the heterolytic dihydrogen cleavage in the QM
and QM/MM model transition states TS 0H-H2-Re#1 (Figures
19 and 21, respectively). Figure 23 shows a schematic repre-
sentation of hydrogen cleavage mediated through multiple metal
atoms. Activation pathways such as these may have implications
1
the formation of hydrogen in the H NMR spectrum from its
characteristic resonance at δ ) 4.51 ppm.
Acknowledgment. This research was supported by the Office
of Basic Energy Sciences of the U.S. Department of Energy
9
J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007 999

A R T I C L E S
Adams et al.
under Grant No. DE-FG02-00ER14980. We thank Strem for
donation of a sample of Pt(PBut3)2 and Johnson Matthey for a
donation of K2PtCl4. The work at Texas A&M University was
supported by the NSF (Grants CHE 0518074 and DMS
0216275), The Welch Foundation (Grant A-0648), and Texas
A&M University. We thank Dr. Perry J. Pellechia for assistance
with the NMR spectra and Dr. Mark D. Smith for collecting
low-temperature X-ray diffraction data.
Supporting Information Available: Experimental data and
CIF files for each of the structural analyses, additional optimized
structures of dihydrogen adducts and other isomers of the 2H
and 4H complexes, and an example of a Gaussian 03 input file
used for the QM/MM calculations. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA067028G
9
1000 J. AM. CHEM. SOC. VOL. 129, NO. 4, 2007