Addition of palladium and platinum tri-tert-butylphosphine groups to Re-Sn and Re-Ge Bonds

Article abstract of DOI:10.1021/ic051077x

The reaction of Re₂(CO)₈[μ-η²-C(H)= C(H)Buⁿ](μ-H) with Ph₃SnH at 68°C yielded the new compound Re₂(CO)₈-(μ-SnPh₂)₂ (10) which contains two SnPh₂ ligands bridging two Re(CO)₄ groups, joined by an unusually long Re-Re bond. Fenske-Hall molecular orbital calculations indicate that the bonding in the Re₂Sn₂ cluster is dominated by strong Re-Sn interactions and that the Re-Re interactions are weak. The ¹¹⁹Sn Moessbauer spectrum of 10 exhibits a doublet with an isomer shift (IS) of 1.674(12) mm s^-1 and a quadrupole splitting (QS) of 2.080(12) mm s^-1 at 90 K,characteristic of Sn(IV) in a SnA₂B₂ environment. The IS is temperature dependent, -1.99(14) ×10^-4 mm s^-1 K^-1; the QS is temperature independent. The temperature-dependent properties are consistent with the known Gol'danskii-Kariagin effect. The germanium compound Re₂(CO)₈(μ-GePh₂) ₂ (11) was obtained from the reaction of Re₂(CO) ₈[μ-η²-C(H)=C(H)Buⁿ](μ-H) with Ph₃GeH. Compound 11 has a structure similar to that of 10. The reaction of 10 with Pd(PBu^t₃)₂ at 25°C yielded the bis-Pd(PBu^t₃) adduct, Re₂(CO) ₈(μ-SnPh₂)₂[Pd(PBu^t ₃)]₂ (12); it has two Pd(PBu^t₃) groups bridging two of the four Re-Sn bonds in 10. Fenske-Hall molecular orbital calculations show that the Pd(PBu^t₃) groups form three-center two-electron bonds with the neighboring rhenium and tin atoms. The mono- and bis-Pt(PBu^t₃) adducts, Re₂(CO) ₈(μ-SnPh₂)₂[Pt(PBu^t₃)] (13) and Re₂(CO)₈(μ-SnPh₂) ₂[Pt(PBu^t₃)]₂ (14), were formed when 10 was treated with Pt(PBu^t₃)₂. A mono adduct of 11, Re₂(CO)₈(μ-GePh₂) ₂[Pt(PBu^t₃)] (15), was obtained similarly from the reaction of 11 with Pt(PBu^t₃)₂.

Full text of DOI:10.1021/ic051077x

Inorg. Chem. 2005, 44, 6346−6358
Addition of Palladium and Platinum Tri-tert-Butylphosphine Groups to
Re Sn and Re Ge Bonds
−
−
Richard D. Adams,*,† Burjor Captain,† Rolfe H. Herber,*,‡ Mikael Johansson,† Israel Nowik,‡
Jack L. Smith, Jr.,† and Mark D. Smith†
Contribution from Department of Chemistry and Biochemistry, UniVersity of South Carolina,
Columbia, SC 29208, and Racah Institute of Physics, The Hebrew UniVersity of Jerusalem,
91904 Jerusalem, Israel
Received June 29, 2005
The reaction of Re2(CO)8[
-SnPh2)2 (10) which contains two SnPh2 ligands bridging two Re(CO)4 groups, joined by an unusually long
µ-η d µ-H) with Ph3SnH at 68 °C yielded the new compound Re2(CO)8-
2-C(H) C(H)Bun](
(
µ
Re
−
Re bond. Fenske
−Hall molecular orbital calculations indicate that the bonding in the Re2Sn2 cluster is dominated
by strong Re
−Sn interactions and that the Re
−
Re interactions are weak. The 119Sn Mo¨ssbauer spectrum of 10
exhibits a doublet with an isomer shift (IS) of 1.674(12) mm s- and a quadrupole splitting (QS) of 2.080(12) mm
1
1
4
s- at 90 K,characteristic of Sn(IV) in a SnA2B2 environment. The IS is temperature dependent,
−1.99(14) ×
10-
mm s- K-1; the QS is temperature independent. The temperature-dependent properties are consistent with the
1
known Gol’danskii
reaction of Re2(CO)8[
The reaction of 10 with Pd(PBut3)2 at 25
it has two Pd(PBut3) groups bridging two of the four Re
show that the Pd(PBut3) groups form three-center two-electron bonds with the neighboring rhenium and tin atoms.
−
Kariagin effect. The germanium compound Re2(CO)8(
2-C(H) C(H)Bun](
-H) with Ph3GeH. Compound 11 has a structure similar to that of 10.
C yielded the bis-Pd(PBut3) adduct, Re2(CO)8( -SnPh2)2[Pd(PBut3)]2 (12);
Sn bonds in 10. Fenske Hall molecular orbital calculations
µ-GePh2)2 (11) was obtained from the
µ
-η
d
µ
°
µ
−
−
The mono- and bis-Pt(PBut3) adducts, Re2(CO)8(
µ
-SnPh2)2[Pt(PBut3)] (13) and Re2(CO)8(
µ
-SnPh2)2[Pt(PBut3)]2 (14),
were formed when 10 was treated with Pt(PBut3)2. A mono adduct of 11, Re2(CO)8(
obtained similarly from the reaction of 11 with Pt(PBut3)2.
µ
-GePh2)2[Pt(PBut3)] (15), was
Introduction
reforming.3,4 Recent studies have shown that polynuclear
metal complexes can be excellent precursors for new
polymetallic nanoparticle catalysts.5
We have shown that bimetallic cluster complexes contain-
ing palladium and platinum can be easily prepared from the
Platinum-rhenium bimetallic catalysts have attracted great
interest because of their superior properties for the important
process known as petroleum reforming.1 Tin is also widely
used to modify the activity of noble metal catalysts,2 and
platinum-tin catalysts have been shown to exhibit superior
properties for a variety of processes including petroleum
(3) (a) Cortright, R. D.; Levin, P. E.; Dumesic, J. A. Ind. Eng. Chem.
Res. 1998, 37, 1717. (b) Cortright, R. D.; Dumesic, J. A. J. Catal.
1994, 148, 771. (c) Cortright, R. D.; Dumesic, J. A. J. Catal. 1995,
157, 576. (d) Cortright, R. D.; Dumesic, J. A. Appl. Catal. A 1995,
129, 101. (e) Vilella, I. M. J.; de Miguel, S. R.; de Lecea, C. S.-M.;
Linares-Solano, A.; Scelza, O. A. Appl. Catal. A 2005, 281, 247. (f)
Rodriquez, D.; Sanchez, J.; Arteaga, G. J. Mol. Catal. A 2005, 228,
309. (g) Llorca, J.; Homs, N.; Leon, J.; Sales, J.; Fierro, J. L. G.;
Ramierez de la Piscina, P. Appl. Catal. A 1999, 189, 77.
(4) (a) Burch, R. J. Catal. 1981, 71, 348. (b) Burch, R.; Garla, L. C. J.
Catal. 1981, 71, 360. (c) Srinivasan, R.; Davis, B. H. Platinum Met.
ReV. 1992, 36, 151. (d) Fujikawa, T.; Ribeiro, F. H.; Somorjai, G. A.
J. Catal. 1998, 178, 58.
(5) (a) Thomas, J. M.; Johnson, B. F. G.; Raja, R.; Sankar, G.; Midgley,
P. A. Acc. Chem. Res. 2003, 36, 20. (b) Thomas, J. M.; Raja, R.;
Johnson, B. F. G.; Hermans, S.; Jones, M. D.; Khimyak, T. Ind. Eng.
Chem. Res. 2003, 42, 1563. (c) Johnson B. F. G. Top. Catal. 2003,
24, 147. (d) Moreno-Manas, M.; Pleixats, R. Acc. Chem Res. 2003,
36, 638.
* To whom correspondence should be addressed. E-mail: Adams@
mail.chem.sc.edu (R.D.A.); herber@VMS.HUJI.AC.IL (R.H.H.).
† University of South Carolina.
‡ The Hebrew University of Jerusalem.
(1) (a) Xiao, J.; Puddephatt, R. J. Coord. Chem. ReV. 1995, 143, 457. (b)
Augustine, S. M.; Sachtler, W. M. H. J. Catal. 1989, 116, 184. (c)
Nacheff, M. S.; Kraus, L. S.; Ichikawa, M.; Hoffman, B. M.; Butt, J.;
Sachtler, W. M. H. J. Catal. 1987, 106, 263.
(2) (a) Huber, G. W.; Shabaker, J. W.; Dumesic, J. A. Science 2003, 300,
2075. (b) Holt, M. S.; Wilson, W. L.; Nelson, J. H. Chem. ReV. 1989,
89, 11. (b) Coupe´, J. N.; Jorda˜o, E.; Fraga, M. A.; Mendes, M. J.
Appl. Catal. A 2000, 199, 45. (c) Hermans, S.; Raja, R.; Thomas, J.
M.; Johnson, B. F. G.; Sankar, G.; Gleeson, D. Angew. Chem., Int.
Ed. 2001, 40, 1211. (e) Hermans, S.; Johnson, B. F. G. Chem.
Commun. 2000, 1955.
6346 Inorganic Chemistry, Vol. 44, No. 18, 2005
10.1021/ic051077x CCC: $30.25
© 2005 American Chemical Society
Published on Web 08/12/2005

Addition of Pd- and Pt-PBut3 Groups to Re-Sn and Re-Ge Bonds
reactions of the palladium and platinum bis-tri-tert-bu-
tylphosphine compounds, M(PBut3)2, M ) Pd or Pt with
preformed metal cluster complexes.6 For example, the
reaction of Pd(PBut3)2 with the ruthenium carbonyl cluster
complexes Ru3(CO)12 and Ru6(CO)17(µ6-C) yields the adducts
Ru3(CO)12[Pd(PBut3)]3 (1) and Ru6(CO)17(µ6-C)[Pd(PBut3)]2
(2), respectively.6a
cleavage of one or more phenyl groups from the tin atom to
form bridging SnPh2 ligands, bridging SnPh ligands, or both.
The cleaved phenyl groups combine with the hydrogen atoms
and are eliminated as benzene. For example, the complex
Ru5(CO)8(C6H6)(µ-SnPh2)4(µ5-C) (6), formed from the reac-
tion of Ph3SnH with Ru5(CO)12(C6H6)(µ5-C), contains four
SnPh2 groups that bridge the Ru-Ru bonds along the base
of an Ru5 square pyramid.7a,b The addition of Ph3SnH to
rhodium and iridium carbonyl cluster complexes yields
products with an even higher tin content. For example, the
complex Rh3(CO)3(SnPh3)3(µ-SnPh2)3(µ3-SnPh)2 (7), formed
from the reaction of Rh4(CO)12 and Ph3SnH, contains eight
tin ligands: three terminal SnPh3 groups, three edge-bridging
SnPh2 groups, and two triply bridging SnPh groups.7c It has
also been shown that Ph3GeH can engage in similar reactions
with metal carbonyl clusters to yield products such as
Ru5(CO)11(µ-GePh2)4(µ5-C) (8) and Ir4H4(CO)4(µ-GePh2)4-
(µ4-GePh)2 (9), which contain bridging GePh2 ligands.8
Compound 9 also contains two quadruply bridging GePh
ligands.
Pd(PBut3) and Pt(PBut3) groups can exhibit facile dynami-
cal activity. For example, the Pt(PBut3) adduct of Ru5(CO)15-
(µ5-C), Ru5(CO)15(µ5-C)[PtPBut3)] (3), exists as both open
and closed isomers, 3a and 3b, of the metal clusters in
solution, and they are in rapid equilibrium on the NMR time
scale at room temperature (eq 1).6b
The bimetallic compounds Ir6(CO)10[Pt(PBut3)]4 (4)6c and
Pt3Re2(CO)6(PBut3)3 (5)6d were synthesized from the reactions
of Pt(PBut3)2 with Ir4(CO)12 and Re2(CO)10, respectively.
Both are electronically unsaturated and were found to activate
hydrogen under mild conditions.
Herein, we demonstrate for the first time that rhenium can
be chemically combined both with platinum or palladium
and tin to make the first ternary ReSnM complexes, M )
Pd or Pt, which could potentially be used as precursors to
stoichiometrically precise ReSnM catalysts. A preliminary
report on this work has been published.9
We have also prepared a number of new bimetallic
complexes containing tin ligands by the reactions of Ph3-
SnH with transition metal carbonyl cluster complexes.7 This
process begins with the oxidative addition of the Sn-H bond
to the transition metal clusters and is followed by the
Experimental Section
General Data. All reactions were performed under a nitrogen
atmosphere. 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 spectro-
photometer. 1H NMR spectra were recorded on a Varian Mercury
spectrometer operating at 400.1 MHz. 31P{1H} NMR spectra were
(6) (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. (b) Adams,
R. D.; Captain, B.; Fu, W.; Pellechia, P. J.; Smith, M. D. Inorg. Chem.
2003, 42, 2094. (c) Adams, R. D.; Captain, B.; Hall, M. B.; Smith, J.
L., Jr.; Webster, C. E. J. Am. Chem. Soc. 2005, 127, 1007. (d) Adams,
R. D.; Captain, B. Angew. Chem., Int. Ed. 2005, 44, 2531. (e) Adams,
R. D.; Captain, B.; Pellechia, P. J.; Smith, J. L., Jr. Inorg. Chem. 2004,
43, 2695. (f) Adams, R. D.; Captain, B.; Fu, W.; Smith, J. L., Jr.;
Smith, M. D. Organometallics 2004, 23, 589. (g) Adams, R. D.;
Captain, B.; Fu, W.; Hall, M. B.; Smith, M. D.; Webster, C. E. Inorg.
Chem. 2004, 43, 3921. (h) Adams, R. D.; Captain, B.; Smith, M. D.
J. Cluster Sci. 2004, 15, 139. (i) Adams, R. D.; Captain, B.; Fu, W.;
Smith, M. D. J. Organomet. Chem. 2003, 682, 113. (j) Adams, R. D.;
Captain, B.; Pellechia, P. J.; Zhu, L. Inorg. Chem. 2004, 43, 7243.
(7) (a) Adams, R. D.; Captain, B.; Fu, W.; Smith, M. D. Inorg. Chem.
2002, 41, 2302. (b) Adams, R. D.; Captain, B.; Fu, W.; Smith, M. D.
Inorg. Chem. 2002, 41, 5593. (c) Adams, R. D.; Captain, B.; Smith,
J. L., Jr.; Hall, M. B.; Beddie, C. L.; Webster, C. E. Inorg. Chem.
2004, 43, 7576.
(8) (a) Adams, R. D.; Captain, B.; Fu, W. Inorg. Chem. 2003, 42, 1328.
(b) Adams, R. D.; Captain, B.; Smith, J. L., Jr. Inorg. Chem. 2005,
44, 1413.
(9) Adams, R. D.; Captain, B.; Johansson, M.; Smith, J. L., Jr. J. Am.
Chem. Soc. 2005, 127, 489.
Inorganic Chemistry, Vol. 44, No. 18, 2005 6347

Adams et al.
recorded on a Varian Mercury 400 spectrometer operating at 162.0
MHz and were externally referenced against 85% ortho-H3PO4.
Elemental analyses were performed by Desert Analytics (Tucson,
AZ). Product separations were performed by TLC in air on Analtech
0.5 mm silica gel 60 Å F254 glass plates. Re2(CO)10, Pd(PBut3)2,
and Pt(PBut3)2 were obtained from Strem Chemicals, Inc. Ph3SnH
and Ph3GeH were purchased from Aldrich and were used without
further purification. Re2(CO)8[µ-η2-C(H)dC(H)Bun](µ-H) was pre-
pared according to a previously reported procedure.10
Synthesis of Re2(CO)8(µ-SnPh2)2 (10). Ph3SnH (200 mg, 0.57
mmol) was dissolved in 5 mL of hexane, and the solution was then
added to a solution of Re2(CO)8[µ-η2-C(H)dC(H)Bun](µ-H) (75
mg, 0.11 mmol) in 30 mL of hexane. The reaction mixture was
heated to reflux for 2 h. After the mixture was cooled, the solvent
was removed in vacuo. The product was separated by TLC using
a 3:1 hexane/CH2Cl2 solvent mixture to yield 64.8 mg (52% yield)
of yellow Re2(CO)8(µ-SnPh2)2 (10). Spectral data for 10. IR (cm-1
in hexane): νCO 2060 (m), 2008 (s), 1981 (m), 1972 (m). 1H NMR
(CD2Cl2): δ 7.69 (m, 8 H, Ph), 7.43 (m, 12 H, Ph). Anal. Calcd:
33.65, C; 1.76, H. Found: 34.02, C; 2.06, H.
Synthesis of Re2(CO)8(µ-GePh2)2 (11). Ph3GeH (170 mg, 0.57
mmol) was added to a solution of Re2(CO)8[µ-η2-C(H)dC(H)Bun]-
(µ-H) (75 mg, 0.11 mmol) in 30 mL of heptane. The reaction
mixture was heated to reflux for 2 h. After the mixture was cooled,
the solvent was removed in vacuo, and the product was separated
by TLC using a 3:1 hexane/CH2Cl2 solvent mixture to yield 37.1
mg (32% yield) of pale yellow Re2(CO)8(µ-GePh2)2 (11). Spectral
data for 11. IR (cm-1 in hexane): νCO 2068 (m), 2011 (s), 1987
(m), 1982 (m). 1H NMR (CD2Cl2): δ 7.63 (m, 8 H, Ph), 7.40 (m,
12 H, Ph). Anal. Calcd: 36.60, C; 1.92, H. Found: 36.37, C; 2.17,
H.
Synthesis of Re2(CO)8(µ-SnPh2)2[Pd(PBut3)]2 (12). Pd(PBut3)2
(56 mg, 0.11 mmol) was added to a solution of 10 (25 mg, 0.022
mmol) in 25 mL of CH2Cl2. The reaction mixture was stirred at
room temperature for 30 min. The solvent was then removed in
vacuo, and the product was separated by TLC using a 2:1 hexane/
CH2Cl2 solvent mixture to yield 24.9 mg (67%) of orange Re2-
(CO)8(µ-SnPh2)2[Pd(PBut3)]2 (12). Spectral data for 12. IR (cm-1
in hexane): νCO 2074 (w), 2047 (w), 2035 (w), 1998 (s), 1994
(m), 1980 (s), 1962 (m), 1955 (vs), 1821 (w). 1H NMR (CD2Cl2):
δ 7.79 (m, 8 H, Ph), 7.30 (m, 12 H, Ph), 1.24 (d, 54 H, CH3, 3JP-H
) 13 Hz). 31P{1H} NMR (CD2Cl2): δ 85.4 (s, 2 P). Anal Calcd:
38.22, C; 4.24, H. Found: 37.85, C; 3.93, H.
Synthesis of Re2(CO)8(µ-GePh2)2[Pt(PBut3)] (15). Pt(PBut3)2
(75 mg, 0.125 mmol) was added to a solution of 11 (25 mg, 0.024
mmol) in 25 mL of octane. The reaction mixture was heated to
reflux for 12 h. After the mixture was cooled, the solvent was
removed in vacuo. The product was separated by TLC by using a
2:1 hexane/CH2Cl2 solvent mixture to yield 6.2 mg (18% yield) of
orange Re2(CO)8(µ-GePh2)2[Pt(PBut3)] (15). Spectral data for 15.
IR (cm-1 in hexane): νCO 2081 (w), 2051 (w), 2000 (s), 1984 (m),
1970 (m), 1786 (w). 1H NMR (CD2Cl2): δ 7.77 (m, 8 H, Ph), 7.30
(m, 12 H, Ph), 1.26 (d, 27 H, CH3, 3JP-H ) 13 Hz). 31P{1H} NMR
(CD2Cl2): δ 114.7 (s, 1 P, 1JPt-P ) 5868 Hz). Anal. Calcd: 36.51,
C; 3.27, H. Found: 37.20, C; 3.34, H.
Crystallographic Analyses. Yellow crystals of 10 (triclinic form)
and 11 suitable for X-ray diffraction analyses were grown by slow
evaporation of solutions in hexane/CH2Cl2 solvent mixtures at 5
°C. Yellow crystals of 10 (monoclinic form), orange crystals of
12, 13, and 15, and red crystals of 14 were grown from solutions
in benzene/octane solvent mixtures at 5 °C. Each data crystal was
glued onto the end of a thin glass fiber. X-ray intensity data were
measured with a Bruker SMART APEX CCD-based diffractometer
using Mo KR radiation (λ ) 0.71073 Å). The raw data frames were
integrated with the SAINT+ program11 using a narrow-frame
integration algorithm. Corrections for Lorentz and polarization
effects were also applied with SAINT+. An empirical absorption
correction based on the multiple measurement of equivalent
reflections was applied in each analysis by using the program
SADABS. All structures were solved by a combination of direct
methods and difference Fourier syntheses and refined by the full-
matrix least-squares method on F2 using the SHELXTL software
package.12 All non-hydrogen atoms in the main residue were refined
with anisotropic displacement parameters. All 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 the results of the analyses are listed in
Tables 1-3.
Crystals of 10, grown from CH2Cl2/hexane, crystallized in the
triclinic crystal system. The space group P1h was assumed and
confirmed by the successful solution and refinement of the structure.
Crystals of 10 grown from benzene/octane crystallized in a
monoclinic modification. The systematic absences in the data were
consistent with the space groups C2/c and Cc. The former was
chosen and confirmed by the successful solution and refinement
of the structure.
Synthesis of Re2(CO)8(µ-SnPh2)2[Pt(PBut3)] (13) and Re2-
(CO)8(µ-SnPh2)2[Pt(PBut3)]2 (14). Pt(PBut3)2 (40 mg, 0.067 mmol)
was added to a solution of 10 (25 mg, 0.022 mmol) in 25 mL of
hexane. The reaction mixture was heated to reflux for 6 h. The
solvent was removed in vacuo, and the products were separated
by TLC using a 2:1 hexane/CH2Cl2 solvent mixture to yield, in
order of elution, 12.8 mg (38% yield) of orange Re2(CO)8(µ-
SnPh2)2[Pt(PBut3)] (13) and 1.0 mg (2% yield) of red Re2(CO)8-
(µ-SnPh2)2[Pt(PBut3)]2 (14). Spectral data for 13. IR (cm-1 in
hexane): νCO 2075 (w), 2046 (w), 1997 (s), 1980 (m), 1962 (m),
1789 (w). 1H NMR (CD2Cl2): δ 7.76 (m, 8 H, Ph), 7.38 (m, 12 H,
Ph), 1.32 (d, 27 H, CH3, 3JP-H ) 13 Hz). 31P{1H} NMR (CD2Cl2):
Compounds 11 and 14 crystallized in the monoclinic crystal
system. For 11, the systematic absences in the data were consistent
with the space groups C2, Cm, and C2/m. The space group C2/m
was chosen initially and confirmed by the successful solution and
refinement of the structure. For 14, the space group P21/c was
identified uniquely on the basis of the systematic absences in the
intensity data. The crystal of 14 contains two molecules of benzene
from the crystallization solvent that cocrystallized with the complex.
Both benzene molecules were successfully located and refined with
anisotropic displacement parameters. Compounds 12 and 15
crystallized in the triclinic crystal system. In each case, the space
group P1h was assumed and confirmed by the successful solution
and refinement of the structure.
1
δ 118.6 (s, 1 P, JPt-P ) 5932 Hz). Anal. Calcd: 34.32, C; 3.08,
H. Found: 34.67, C; 2.91, H. Spectral data for 14. IR (cm-1 in
CH2Cl2): νCO 2047 (vw), 2021 (w), 1974 (s), 1942 (m), 1776 (w).
1H NMR (CD2Cl2): δ 7.79 (m, 8 H, Ph), 7.28 (m, 12 H, Ph), 1.28
(d, 54 H, CH3, 3JP-H ) 13 Hz). 31P{1H} NMR (CD2Cl2): δ 139.5
The crystal for compound 13 was identified as a nonmerohedral
twin composed of two domains related by a 180° rotation around
the reciprocal [0 -1 0] axis. The GEMINI13 program was used to
(11) SAINT+, version 6.2a; Bruker Analytical X-ray Systems, Inc.:
Madison, WI, 2001.
1
(s, 2 P, JPt-P ) 6048 Hz). MS: e/z 1977 (M + K), 1938 (M+).
(12) Sheldrick, G. M. SHELXTL, version 6.1; Bruker Analytical X-ray
Systems, Inc.: Madison, WI, 1997.
(10) Nubel, P. O.; Brown, T. L. J. Am. Chem. Soc. 1984, 106, 644.
6348 Inorganic Chemistry, Vol. 44, No. 18, 2005

Addition of Pd- and Pt-PBut3 Groups to Re-Sn and Re-Ge Bonds
Table 1. Crystallographic Data for Compound 10
empirical formula
fw
Re2Sn2O8C32H20
1142.26
Re2Sn2O8C32H20
1142.26
Re2Sn2O8C32H20
1142.26
cryst syst
lattice parameters
a (Å)
b (Å)
c (Å)
triclinic
monoclinic
monoclinic
9.5962(5)
12.6495(7)
14.5038(8)
104.3420(10)
90.5590(10)
105.3800(10)
1639.25(15)
P1h
12.4615(8)
10.7519(7)
23.8216(15)
90
100.8370(10)
90
3134.8(3)
C2/c
4
2.42
9.322
100(2)
56.60
3741
12.7904(6)
10.8809(5)
24.0014(12)
90
102.2830(10)
90
3263.8(3)
C2/c
4
2.325
8.954
296(2)
56.56
R (deg)
â (deg)
γ (deg)
V (Å3)
space group
Z
2
F
calc (g cm-3
)
2.314
8.914
296(2)
56.56
6488
397
1.019
µ(Mo KR) (mm-1
temp (K)
)
2Θmax (deg)
no. of observations
no. of params
GOFa
3597
199
1.068
199
1.214
max shift in cycle
residuals:b R1, wR2
abs correction
max/min
0.002
0.001
0.001
0.0398, 0.0956
SADABS
1.000/0.485
2.484
0.0192, 0.0498
SADABS
0.155/0.121
0.572
0.0211, 0.0536
SADABS
0.639/0.143
0.906
largest peak (e Å-3
)
a
GOF ) [∑hklw(|Fo| - |Fc|)2/(ndata - nvari)]1/2
.
b R1 ) ∑hkl(|Fo| - |Fc||)/∑hkl|Fo|; wR2 ) [∑hklw(|Fo| - |Fc|)2/∑hklwF2 ]1/2; w ) 1/σ2(Fo).
o
Table 2. Crystallographic Data for Compounds 11, 12, and 14
11
12
14
empirical formula
fw
Re2Ge2O8C32H20
1050.06
Re2Sn2Pd2P2O8C56H74
1759.67
Re2Pt2Sn2P2O8C56H74‚2C6H6
2093.27
cryst syst
lattice parameters
a (Å)
b (Å)
c (Å)
monoclinic
triclinic
monoclinic
14.0667(11)
10.6056(8)
12.0970(10)
90
117.3840(10)
90
1602.5(2)
C2/m
2
10.8119(7)
12.4429(8)
12.9946(8)
80.1170(10)
68.9880(10)
71.9270(10)
1547.85(17)
P1h
14.0204(7)
14.6505(7)
17.9884(9)
90
108.7180(10)
90
3499.5(3)
P21/c
2
R (deg)
â (deg)
γ (deg)
V (Å3)
space group
Z
1
F
calc (g cm-3
)
2.176
9.435
296(2)
1.888
5.357
296(2)
1.987
8.226
296(2)
µ(Mo KR) (mm-1
temp (K)
)
2Θmax (deg)
50.04
1410
124
1.152
50.06
4901
334
1.042
50.06
5434
388
1.048
no. of observations
no. of paramrs
GOFa
max shift in cycle
residualsb: R1, wR2
abs correction
max/min
0.001
0.000
0.001
0.0254, 0.0593
SADABS
0.686/0.490
1.204
0.0365, 0.0887
SADABS
0.852/0.610
1.809
0.0229, 0.0571
SADABS
0.518/0.153
1.226
largest peak (e Å-3
)
a GOF ) [∑hklw(|Fo| - |Fc|)2/(ndata - nvari)]1/2
.
b R1 ) ∑hkl(||Fo| - |Fc||)/∑hkl|Fo|; wR2 ) [∑hklw(|Fo| - |Fc|)2/∑hklwF2 ]1/2; w ) 1/σ2(Fo).
o
determine the orientation matrices for the twin domains. The
intensity data were processed with SAINT version 7.00 for
multicomponent crystals.11 In real space, the reflection indices for
the two domains are related by the twin law (by rows) [1 0 0/0 -1
0/-0.649 0 -1]. The data were corrected for absorption effects
with the TWINABS subroutine in SAINT. An initial structural
model was obtained by direct methods using the reflections from
the major twin component only. This model was subsequently
refined against the twinned data with excellent results (R1 ) 0.038),
using SHELXTL.12 All data from both domains were used in the
refinement. On the basis of the batch scale factor for the two
domains, the data crystal is composed of 76.7% of the major domain
and 23.3% of the minor domain. The compound crystallizes in the
space group P21/n, and there is one complete complex in the
asymmetric unit. The final difference map extrema are +0.99/-
1.12 e- Å-3, located near the Re(1) atom.
Molecular Orbital Calculations. All molecular orbital calcula-
tions reported herein were performed by using the Fenske-Hall
method.14 The calculations were performed utilizing a graphical
user interface developed15 to build inputs and view outputs from
stand-alone Fenkse-Hall (version 5.2) and MOPLOT216 binary
executables. Contracted double-ú basis sets were used for the Re
5d, Pd 4d, Sn 5p, Ge 4p, P 3p, and C and O 2p atomic orbitals.
(13) GEMINI, version 1.02; Bruker Analytical X-ray Systems, Inc.:
Madison, WI, 1999.
(14) Hall, M. B.; Fenske, R. F. Inorg. Chem. 1972, 11, 768.
Inorganic Chemistry, Vol. 44, No. 18, 2005 6349

Adams et al.
Table 3. Crystallographic Data for Compounds 13 and 15a
13
15
empirical formula
fw
Re2PtSn2PO8C44H47
1539.66
Re2PtGe2PO8C44H47
1447.46
cryst syst
lattice parameters
a (Å)
b (Å)
c (Å)
monoclinic
triclinic
11.5715(5)
23.1038(10)
17.8097(7)
90
102.1550(10)
90
4654.6(3)
P21/n
4
11.7436(4)
14.5720(5)
14.9178(6)
105.0460(10)
109.3420(10)
94.8710(10)
2284.91(14)
P1h
R (deg)
â (deg)
γ (deg)
V (Å3)
Figure 1. ORTEP diagram of the molecular structure of Re2(CO)8(µ-
SnPh2)2, 10 (from the triclinic crystalline form), showing 40% thermal
ellipsoid probability.
space group
Z
2
F
calc (g cm-3
)
2.197
9.318
100(2)
2.104
9.713
296(2)
µ(Mo Ka) (mm-1
temp (K)
)
effected using an R-Fe absorber at room temperature. All IS values
are reported with respect to the centroid of a BaSnO3 absorber
spectrum acquired at 296 K. The minimal line widths observed in
the ME spectra are on the order of 0.82 mm s-1, and the difference
between this value and the theoretical minimum width (0.642 mm
s-1) is ascribed to radiation effects in the two year old 119mSn source,
as well as to minor extraneous line broadening because of
uncontrolled environmental effects.
2Θmax (deg)
50.06
9569
524
1.016
56.61
9535
532
1.027
no. of observations
no. of params
GOFa
max shift in cycle
residualsb: R1, wR2
abs correction
max/min
0.001
0.002
0.0381, 0.0745
SADABS
1.000/0.569
0.992
0.0321, 0.0694
SADABS
0.379/0.248
1.658
largest peak (e Å-3
)
a GOF ) [∑hklw(|Fo| - |Fc|)2/(ndata - nvari)]1/2
.
b R1 ) ∑hkl(||Fo| - |Fc||)/
Results and Discussion
∑
hkl|Fo|; wR2 ) [∑hklw(|Fo| - |Fc|)2/∑hklwF2 ]1/2; w ) 1/σ2(Fo).
o
Very few examples of rhenium-tin complexes have been
reported to date.18 The new dirhenium-ditin complex Re2-
(CO)8(µ-SnPh2)2 (10) was obtained in a 52% yield from the
reaction of Re2(CO)8[µ-η2-C(H)dC(H)Bun](µ-H) with an
excess of Ph3SnH at 68 °C. The analogous germanium
complex Re2(CO)8(µ-GePh2)2 (11) was obtained similarly in
a 32% yield from the reaction of Re2(CO)8[µ-η2-C(H)dC(H)-
Bun](µ-H) with Ph3GeH at 97 °C (eq 2).
Hydrogen was used in place of both phenyl and tert-butyl groups
in these calculations. The Fenske-Hall scheme is a nonempirical
approximate method that is capable of calculating molecular orbitals
for very large transition metal systems and has built-in fragment
analysis routines that allow one to assemble transition metal cluster
structures from the ligand-containing fragments.
Mo1ssbauer Measurements. A sample of 10, consisting of
yellow crystals as received, was transferred to a plastic sample
holder with minimum exposure to the laboratory environment,
rapidly cooled to 78 K, loaded into the precooled cryostat, and
subjected to Mo¨ssbauer spectroscopic (ME) examination as a
function of temperature. While such ME spectra yield reliable values
for the two pertinent hyperfine interactions, the isomer shift (IS)
and the quadrupole splitting (QS), they are not appropriate for the
determination of the lattice dynamical properties of the Sn atom
because of the large “texture effect” (i.e., the preferential crystal
orientation with respect to the optical axis of the ME measurements).
The initial data so acquired will be referred to as 10a. After an
initial set of measurements, the large crystals were warmed to room
temperature, thoroughly ground to a fine powder, and mixed with
boron nitride before being replaced in the sample holder to
overcome this limitation. This procedure reduces the texture effect
to a minimal value and permits an accurate determination of the
vibrational properties of the metal atom (Sn) as a function of
temperature, as will be discussed below. The data acquired with
this sample will be referred to as 10b. The remainder of the ME
experimental procedure was as described earlier17 with the addition
of the use of a real-time digital output monitoring program, which
permitted temperature control to better than (0.1 K over the time
interval of the ME measurements. Spectrometer calibration was
Both compounds were characterized by a combination of
IR, 1H NMR, elemental, and single-crystal X-ray diffraction
analyses. Compounds 10 and 11 are isostructural in the solid
state, but they are not isomorphous. In fact, compound 10
was obtained in two crystalline modifications: a monoclinic
(C2/c) and a triclinic (P1h) form (both different from that of
11). Both forms were investigated by single-crystal X-ray
diffraction analyses. Diffraction data for the triclinic form
were collected at 296 K, and data for the monoclinic form
were collected at both 296 and 100 K. An ORTEP diagram
of the molecular structure of 10 (triclinic form) is shown in
Figure 1. The molecular structure of 11 is not significantly
different. Selected intramolecular distances and angles are
listed in Table 4. Both compounds contain two Re(CO)4
groups that are linked by two bridging XPh2 ligands, X )
(15) Manson, J.; Webster, C. E.; Hall, M. B. JIMP, development version
0.1.v117 (built for Windows PC and Redhat Linux); Department of
Chemistry, Texas A&M University: College Station, TX; http://
www.chem.tamu.edu/jimp/, accessed July 2004.
(16) MOPLOT2 for orbital and density plots from linear combinations of
Slater or Gaussian type orbitals, version 2.0; Dennis L. Lichtenberger,
Department of Chemistry, University of Arizona: Tucson, AZ, 1993.
(17) Nowik, I.; Herber, R. H. Inorg. Chim. Acta 2000, 310, 191.
(18) (a) Preut, H.; Haupt, H.-J.; Florke, U. Acta Crystallogr., Sect. C 1984,
40, 600. (b) Haupt, H.-J.; Balsaa, P.; Schwab, B.; Florke, U.; Preut,
H. Z. Anorg. Allg. Chem. 1984, 22, 513. (c) Narayanan, B. A.; Kochi,
J. K. Inorg. Chim. Acta 1986, 122, 85. (d) Westerberg, D. E.;
Sutherland, B. E.; Huffman, J. C.; Caulton, K. G. J. Am. Chem. Soc.
1988, 110, 1642. (e) Huie, B. T.; Kirtley, S. W.; Knobler, C. B.; Kaesz,
H. D. J. Organomet. Chem. 1981, 213, 45.
6350 Inorganic Chemistry, Vol. 44, No. 18, 2005

Addition of Pd- and Pt-PBut3 Groups to Re-Sn and Re-Ge Bonds
Table 4. Selected Intermolecular Distances (Å) and Angles (deg) for Compounds 10 and 11a
compound 10
compound 11
C2/m at 296 K
P1h at 296 K
Re(1)-Re(1)*
C2/c at 296 K
Re(1)-Re(1)*
Re(1)-Sn(1)
Re(1)-Sn(1)*
C-O (ave)
C2/c at 100 K
Re(1)-Re(1)*
Re(1)-Sn(1)
Re(1)-Sn(1)*
C-O (ave)
3.1971(4)
2.7429(4)
2.7675(5)
1.14(1)
3.1685(3)
2.7567(3)
2.7643(3)
1.14(1)
3.1542(3)
2.7491(3)
2.7682(3)
1.14(1)
Re(1)-Re(1)*
3.0510(6)
2.5923(6)
2.5923(6)
1.13(1)
Re(1)-Sn(1)
Re(1)-Sn(1)*
C-O (ave)
Re(1)-Ge(1)
Re(1)-Ge(1)*
C-O (ave)
Re(1)-Sn(1)-Re(1)*
Re(1)*-Re(1)-Sn(1)
70.926(11) Re(1)-Sn(1)-Re(1)*
54.895(10) Re(1)*-Re(1)-Sn(1)
70.045(7) Re(1)-Sn(1)-Re(1)*
55.088(6) Re(1)*-Re(1)-Sn(1)
69.735(7) Re(1)-Ge(1)-Re(1)*
55.416(6) Re(1)*-Re(1)-Ge(1)
72.10(2)
53.951(11)
Re(1)*-Re(1)-Sn(1)* 54.179(9) Re(1)*-Re(1)-Sn(1)* 54.866(6) Re(1)*-Re(1)-Sn(1)* 54.849(5) Re(1)*-Re(1)-Ge(1)* 53.951(11)
Sn(1)-Re(1)-Sn(1)*
Re-C-O (ave)
109.074(11) Sn(1)-Re(1)-Sn(1)*
177.2(12) Re-C-O (ave)
109.955(7) Sn(1)-Re(1)-Sn(1)*
177.2(7) Re-C-O (ave)
110.265(7) Ge(1)-Re(1)-Ge(1)* 107.90(20)
177.3(6)
Re-C-O (ave)
177.1(7)
a Estimated standard deviations in the least significant figure are given in parentheses.
Table 5. Mo¨ssbauer Parameters and Derived Values for 10
parameter
T
units
multiplier
IS (90)
QS (90)
-dIS/dT
-d ln A/dT
ΘM
1.674 ( 0.012
2.080 ( 0.012
1.99 ( 0.15
16.8 ( 1.2
47 ( 2
mm s-1
mm s-1
88-250 K mm s-1 K-1
10-4
10-3
88-250 K K-1
K
Da
Meff
209 ( 14
k2 x2
1.89 ( 0.03
2.58 ( 0.05
3.57 ( 0.13
4.29 ( 0.28
88.5 K
150 K
210 K
250 K
above. The hyperfine parameters and their temperature
dependencies are summarized in Table 5. The IS value of
1.674(12) mm s-1 at 90 K is consistent with the Sn atom
being in the +IV oxidation state, and the QS value of 2.080-
(12) mm s-1 is characteristic of the geometry of the SnA2B2
nearest-neighbor environment. A C2 axis of symmetry passes
through the tin atoms and bisects the Re-Re bond. These
hyperfine parameters may be compared to those of Ph2SnCl2
of similar geometry for which IS is 1.385(27) mm s-1 and
QS is 2.841(24) mm s-1 at 80 K.20 The temperature
dependence of the IS value is summarized graphically in
Figure 3 in which the open and full points reflect the data
for the crystalline (10a) and powder samples (10b), respec-
tively. This temperature dependence over the accessible
temperature range (89 < T < 150 K) is reasonably well fit
by a linear regression (slope ) -1.99(14) × 10-4 mm s-1
K-1, cc ) 0.94 for 16 data points) that leads to an effective
mass21 (Meff) of 209(14) Da. The difference between this
value and the “bare” Sn atom mass reflects the covalency
of the tin-ligand bonding interactions. As will be discussed
below, Meff is related to the dynamics of the Sn atom in 10b.
In contrast, the QS parameter is not a temperature-
dependent quantity over the accessible temperature range and
all data lie within an error limit of (0.032 mm s-1 of the
mean. This temperature independence is readily understood
on the basis of the X-ray data at 100 and 296 K, which show
that the Re-Sn-Re and C-Sn-C bond angles change by
only 1.19 and 2.90 degrees, respectively, over this temper-
ature interval. Similarly, the Re-Sn and Sn-C bond
Figure 2. 119Sn Mo¨ssbauer spectrum of 10 at 88.5 K. The isomer shift is
with reference to a room-temperature BaSnO3 absorber spectrum.
Sn and Ge for 10 and 11, respectively. According to electron
counting procedures, they should both contain a Re-Re
single bond so that each rhenium atom can attain the usual
18-electron configuration. The Re-Re distance in 10,
however, is quite long, 3.1685(3) Å, in the monoclinic from
at 296 K and is even longer, 3.1971(4) Å, in the triclinic
form at 296 K, but it is still short enough to allow for some
direct Re-Re interactions and the pairing of the unpaired
electron on each Re(CO)4 group, see the molecular orbital
calculations below for details. The fact that the Re-Re
distance is different by nearly 0.03 Å in the two crystalline
modifications at room temperature is consistent with the
notion that this bond is not a strong one. The only differences
between the molecules in the two crystalline forms of 10
are the weak intermolecular packing forces and the rotational
orientations of the phenyl rings on the tin ligands. The Re-
Re bond in 11 (3.0510(6) Å) is significantly shorter than
that in 10 and only slightly longer than that found for the
unbridged Re-Re single bond in Re2(CO)10 (3.042(1) Å).19
Mo1ssbauer Spectrum of 10. A typical 119Sn ME spectrum
of the finely ground crystalline sample of 10b at 88.5 K is
shown in Figure 2. As expected, the spectrum consists of a
simple doublet with well-separated components. The area
ratio, R ) A(+)/A(-) (where the + and - correspond to
the spectrum centroid), of this spectrum is 1.002(4), indicat-
ing the high efficacy of the grinding procedure discussed
(20) (a) Herber, R. H. J. Inorg. Nucl. Chem. 1973, 35, 67. (b) Debye, N.
W. G.; Linzer, M. J. Chem. Phys. 1974, 61, 4770. (c) Williams, D.
E.; Kocher, C. W. J. Chem. Phys. 1970, 52, 1480. (d) Herber, R. H.;
Leahy, M. F. J. Chem. Phys. 1977, 67, 2718.
(19) Churchill, M. R.; Amoh, K. N.; Wasserman, H. J. Inorg. Chem. 1981,
20, 1609.
(21) Herber, R. H. In Chemical Mo¨ssbauer Spectroscopy; Herber, R. H.,
Ed.; Plenum Press: New York, 1984; Chapter VII.
Inorganic Chemistry, Vol. 44, No. 18, 2005 6351

Adams et al.
Figure 3. Temperature dependence of the isomer shift of 10. The open
data points refer to the crystalline sample; the closed data points are for
the finely ground sample.
Figure 5. k2 x2 parameters of 10. The open data points refer to the values
parallel to the C2V axis, while the closed data points pertain to the average
perpendicular values. The line corresponds to the k2 x2
for Meff ) 209
ave
Da and ΘM ) 78 K. The two star symbols are the values calculated from
the X-ray crystallographic data at 100 and 296 K.
yields a value of 5.0(3) for this parameter. The satisfactory
agreement between these values, keeping in mind the
limitation in the temperature range accessible to the ME
measurements of such samples, validates the dynamical
analysis reported in the present study. In this context, it is
worth noting that the IS and QS parameters of 10a and 10b
are identical (within experimental errors). The area ratio, R,
which is unity in the low-temperature limit, becomes a temp-
erature-dependent quantity at higher temperatures. This temp-
erature-dependent phenomenon, known as the Gol’danskii-
Kariagin effect (GKE),22 results from the anisotropy of the
metal atom motion and has been observed in numerous 57Fe
and 119Sn ME spectra. From the magnitudes of R and f at a
given temperature, it is possible to calculate this vibrational
anisotropy. The pertinent data for 10b are summarized in
Figure 5. The open data points pertain to k2 x2 parallel to
the 2-fold symmetry axis passing through the metal atom
and bisecting the Re-Re bond, while the filled data points
reflect the data for the perpendicular values. Moreover, the
Figure 4. Temperature dependence of the logarithm of the area under the
resonance curve (normalized to the 90 K value) for 10b.
distances change by only 0.0062 and 0.0050 Å, respectively,
and these changes are within the error limits of the ME
experiments.
The temperature dependence of the area under the
resonance curve, which for a thin randomly oriented absorber
is identical to the temperature dependence of the recoil-free
fraction, f, is summarized graphically in Figure 4 for 10b.
The ln A data show some curvature, especially at low
temperatures. It should be noted that, for a thin absorber, f
) exp(-k2 x2 ) where k is the wave vector of the 119mSn
gamma ray and has a numerical value of 1.210 × 109 cm
and x2 is the mean-square amplitude-of-vibration of the
Mo¨ssbauer active atom. From the high-temperature limiting
slope of d ln(f/dT) and that of the temperature dependence
of the IS, it is possible to calculate21 the effective Mo¨ssbauer
lattice temperature, ΘM, and for 10b, this is found to be to
be 47(4) K. This unusually low value for ΘM is reflected in
the limited temperature range over which the ME data could
be acquired. In this context, it is appropriate to note that
k2 x2 is also calculable from the Ui,j values obtained in a
single-crystal X-ray diffraction experiment. For 10b, these
data have been determined at 100 and 296 K (see Supporting
Information) and are 1.56 and 5.91, respectively. The low-
temperature ME data give k2 x2 ) 1.7(1) at 100 K.
Extrapolation of the high-temperature ME data to 296 K
two parameters are related to k2 x2
by the relationship
ave
k2 x2
) 1/3k2 x2
+ 2/3k2 x2 , assuming approximate
ave
para
perp
axial symmetry. Since the ME data at 88.5 K show that R )
1.002, the tin atom motion at this temperature must be nearly
isotropic, consistent with the 100 K X-ray Ui,j values which
are 11(1), 13(1), and 8(1) × 10-19 cm2 respectively, with
the off-diagonal elements effectively equal to zero. The k2 x2
data derived from the ME experiments (which are limited
to T < ∼260 K because of the small values of f at the higher
temperatures) can be fitted separately to a second-order
polynomial function, and this procedure yields a value of
4.52 for k2 x2
and a value of 5.58 for k2 x2
at 296 K,
perp
para
indicative of a significant vibrational anisotropy of the tin
at this temperature. The differences in the root-mean-square
(22) (a) Gol’danskii, V. I.; Makarov, E. F. In Chemical Applications of
Mo¨ssbauer Spectroscopy; Gol’danskii, V. I., Herber, R. H., Eds.;
Academic Press: New York, 1968; pp 102-107 and references therein.
(b) Herber, R. H.; Nowik, I. Solid State Sci. 2002, 4, 691. (c) Herber,
R. H.; Nowik, I. Hyperfine Interact. 2001, 136-137, 699. (d) Herber,
R. H.; Chandra, S. J. Chem. Phys. 1971, 54, 1847.
6352 Inorganic Chemistry, Vol. 44, No. 18, 2005

Addition of Pd- and Pt-PBut3 Groups to Re-Sn and Re-Ge Bonds
Figure 6. Coordinate system used for the Fenske-Hall molecular orbital
calculations.
Figure 8. Energy level diagram for the Re2(CO)8 fragment.
b3g, and 1b2u symmetry (which are Re-Re antibonding
orbitals). These six orbitals are all filled. The two sets of
two higher-lying orbitals are transformed into four nonde-
generate orbitals of b3u, 2ag, b1g, and 2b2u symmetry. The
HOMO (b3u) and LUMO (2ag) are Re-Re bonding orbitals.
The b1g and 2b2u are unfilled antibonding orbitals.
The key orbitals on the SnH2 fragment are the HOMO
“sp2”-like hybrid orbital and the LUMO py orbital that lies
perpendicular to the SnH2 plane. The orbitals for the
combined [SnH2]2 fragment (under D2h symmetry) are
derived from the symmetric and antisymmetric combinations
of the sp2-like hybrid orbital (ag and b3u) and the py orbital
(b2u and b1g), see right-hand side of Figure 9.
Figure 7. Molecular orbital energy level diagram for the Re(CO)4
fragment.
amplitude-of-vibration of the Sn atom are ∼ 0, 0.057, 0.103,
and 0.149 Å at 88.9, 150, 200, and 250 K, respectively.
Molecular Orbital Calculations on Re2Sn2 and Re2Ge2.
We have previously shown that the Rh-Rh and Ir-Ir bonds
are unusually long when bridged by a SnPh2 group.7c To
examine the effect of the two edge-bridging SnPh2 and GePh2
groups on the rhenium-rhenium single bond in 10 and 11,
we performed Fenske-Hall molecular orbital calculations.
For these calculations, SnH2 and GeH2 groups were used in
place of the complete SnPh2 and GePh2 ligands in 10 and
11, respectively. A schematic of the coordinate system used
for these calculations is given in Figure 6. The Re2(CO)8-
(SnH2)2 cluster was constructed from two Re(CO)4 fragments
and two SnH2 fragments. This model possesses D2h sym-
metry, and the discussion of its bonding interactions will be
based on this symmetry point group. The d7 Re(CO)4
fragment, which has C2V local symmetry, contains three low-
lying filled orbitals derived from the “t2g” set of Oh symmetry
and two higher-lying orbitals derived from the “eg” set which
contains one electron, see Figure 7. There is considerable p
The orbitals of the Re2(CO)8 fragment combine with
orbitals of the [SnH2]2 fragment of appropriate symmetry
and energy to generate the bonding MOs for the Re2(CO)8-
(SnH2)2 cluster (Figure 9). Favorable interactions of orbitals
from the [SnH2]2 unit lowers the energy of the unoccupied
2ag and b1g orbitals on the Re2(CO)8 fragment, and these
orbitals are then filled by four electrons that are donated by
the SnH2 groups. As was the case with the Rh3(CO)6(SnH3)3-
(µ-SnH2)3 model cluster reported in a previous report,7c the
metal-metal bonding of 10 is dominated by strong Re-Sn
interactions. This can be seen in the contour diagrams for
the b1g (HOMO), 2ag, 1b2u, 1ag, and b3u orbitals shown in
Figures 10-12. Significant Re-Re interactions are found
only in the 2ag orbital (the HOMO - 2, Figure 11), which
explains why the Re-Re distance is unusually long. The
b3g, au, b2g, and b1u orbitals on the Re2(CO)8 fragment have
no interactions with the orbitals on the [SnH2]2 fragment.
As noted above, the Re-Re bond distance in 11 is
considerably shorter than that in 10. For comparison, we have
also calculated the energies of the Fenske-Hall molecular
orbitals of Re2(CO)8(µ-GeH2)2 using the positional param-
eters for 11 for all nonhydrogen atoms. A correlation diagram
that compares the energies of the MOs for the Re2(CO)8(µ-
SnH2)2 and Re2(CO)8(µ-GeH2)2 model clusters is shown in
Figure 13. For comparison, we will focus on the b1g (HOMO)
and the 2ag (HOMO - 2). The b3g (HOMO - 1) has the
2
2
character in the singly occupied HOMO dx -y orbital and s
2
and p character in the LUMO dz orbital. These metal orbitals
interact with the carbonyl 2π* (LUMO) orbital.
A linear combination of the two Re(CO)4 fragments in a
dinuclear Re2(CO)8 unit having D2h symmetry is shown in
Figure 8. The two sets of three low-lying orbitals are
transformed into nondegenerate orbitals of b1u, 1ag, and b2g
symmetry (which are Re-Re bonding orbitals) and of au,
Inorganic Chemistry, Vol. 44, No. 18, 2005 6353

Adams et al.
Figure 11. Contour diagram for the 2ag molecular orbital. This orbital
contains the only significant Re-Re bonding in the cluster.
[Pd(PBut3)]2 (12) in 67% yield. Compound 12 was charac-
1
terized by a combination of IR, H and 31P{1H} NMR,
elemental, and single-crystal X-ray diffraction analyses. An
ORTEP diagram of the molecular structure of 12 is given in
Figure 14. Selected intramolecular distances and angles are
listed in Table 6. In the solid state, the compound is
crystallographically centrosymmetrical. This compound can
be viewed as a bis-Pd(PBut3) adduct of 10 because no ligands
were lost from 10 during the course of the reaction. The
structure contains two Pd(PBut3) groups that bridge the
oppositely positioned Re-Sn bonds of the starting cluster
10. One carbonyl ligand from each of the original Re(CO)4
groups in 10 has moved into a semi-bridging position on
each of the Re-Pd bonds. The Re(1)-Pd(1) bond distance
(2.8580(5) Å) is similar in length to one of the two Re-Pd
bonds found in the compound [(η5-C4H4BPh)(CO)3RePd]2
(2.866(1) Å)23 (the other Re-Pd bond in this compound is
much shorter, 2.666(1) Å) but longer than those observed
for the compounds [NEt4]2[Re7(CO)21{Pd(η3-PhC3H4)}(C)]
(2.872(5) Å, average)24 and [RePd(CO)3{Ph2P(CH2)2PPh2}-
(η5-7-CB10H11)] (2.7858(4) Å).25 The Pd-Sn bond distance
(2.7185(7) Å) is significantly longer than those in the
literature (e.g., 2.6082(3) Å as observed for the compound
Pd(PBut2CH2CH2PBut2)(µ-H)SnMe3).26 The addition of the
Pd(PBut3) groups does not significantly alter the length of
the palladium bridged Re-Sn bonds (from 2.7675(5) Å in
10 to 2.7674(5) Å in 12), but curiously, the two Re-Sn
bonds not directly associated with the palladium atom, Re-
(1)-Sn(1)*, have increased in length substantially (from
2.7429(4) Å in 10 to 2.8215(5) Å in 12). The Re-Re bond
distance (3.262(1) Å) is even longer than that found in 10.
Molecular Orbital Calculations on Re2Sn2Pd2. The most
interesting feature of the bonding of Pd(PBut3) and Pt(PBut3)
groups in the adducts, such as 1, 2, and 12, is that the
palladium atom is electron deficient and is formally a zero-
electron donor.6a,c,g In 12, the palladium-metal bonds that
Figure 9. Energy level diagram for the Re2(CO)8(µ-SnH2)2 model cluster.
Figure 10. b1g HOMO of the Re2(CO)8(SnH2)2 model cluster showing
strong Re-Sn bonding.
same energy in both model compounds. Both the b1g and
2ag orbitals are significantly lower in energy in the Re2(CO)8-
(µ-GeH2)2 model, but most interestingly, we see that the 2ag
orbital was lowered by 0.7 eV on going from the Sn
compound to the Ge compound, while the b1g orbital was
lowered by only 0.6 eV. Because the 2ag orbital is predomi-
nantly Re-Re bonding, this would indicate that the Re-Re
bond in the Ge form is stronger than that in the Sn form.
This is consistent with the observation of a shorter Re-Re
bond distance in 11 than in 10.
Reaction of 10 with Pd(PBut3)2. The importance of the
Re-Sn bonding in 10 is further demonstrated by its reaction
with Pd(PBut3)2. The reaction of 10 with an excess of Pd-
(PBut3)2 at 25 °C yielded the complex Re2(CO)8(µ-SnPh2)2-
(23) Braunstein, P.; Englert, U.; Herberich, G. E.; Neuschu¨tz Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1010.
(24) Henly, T. J.; Wilson, S. R.; Shapley, J. R. Inorg. Chem. 1988, 27,
2551.
(25) Blandford, I.; Jeffrey, J. C.; Jelliss, P. A.; Stone, F. G. A. Organo-
metallics 1998, 17, 1402.
(26) Trebbe, R.; Schager, F.; Goddard, R.; Po¨rschke, K.-R. Organometallics
2000, 19, 521.
6354 Inorganic Chemistry, Vol. 44, No. 18, 2005

Addition of Pd- and Pt-PBut3 Groups to Re-Sn and Re-Ge Bonds
Figure 12. Contour diagrams for the 1b2u, 1ag, and b3u orbitals, respectively, showing Re-Sn bonding.
Table 6. Selected Intermolecular Distances (Å) and Angles (deg) for
Compounds 12 and 14a
12
14
Re(1)-Re(1)*
Re(1)-Sn(1)
Re(1)-Sn(1)*
Pd(1)-Re(1)
Pd(1)-Sn(1)
Pd(1)-P(1)
3.262(1)
Re(1)-Re(1)*
Re(1)-Sn(1)
Re(1)-Sn(1)*
Pt(1)-Re(1)
Pt(1)-Sn(1)
Pt(1)-P(1)
3.2008(3)
2.7730(4)
2.8078(3)
2.7972(2)
2.8166(4)
2.3084(12)
1.15(1)
2.7674(5)
2.8215(5)
2.8580(5)
2.7185(7)
2.4093(17)
1.14(2)
C-O (ave)
C-O (ave)
Re(1)-Sn(1)-Pd(1) 133.46(2)
Re(1)-Sn(1)-Pt(1)
126.406(12)
Re(1)-Sn(1)-Re(1)* 71.409(13) Re(1)-Sn(1)-Re(1)* 69.993(9)
Pd(1)-Re(1)-Sn(1)* 164.370(18) Pt(1)-Re(1)-Sn(1)* 159.520(11)
Sn(1)-Re(1)-Sn(1)* 108.591(13) Sn(1)-Re(1)-Sn(1)* 110.007(9)
Re-C-O (ave)
175.8(12)
Re-C-O (ave)
176.6(7)
a Estimated standard deviations in the least significant figure are given
in parentheses.
Pd(PH3) groups were used instead of the actual P(But3)3
groups found in 12. These groups were added to the MO
model developed above for 10. The key orbitals of the Pd-
(PH3) fragment consist of the empty spz hybrid orbital and
2
the filled dz orbital. An energy level diagram for the resulting
Re2(CO)8(µ-SnH2)2[Pd(PH3)]2 cluster (using approximate C2h
symmetry) is shown in Figure 15. Viewed in combination,
the two Pd(PH3) groups form a pair of filled ag and bu orbitals
Figure 13. Correlation diagrams for the molecular orbitals of 10 and 11.
2
from the dz orbitals and a similar pair of empty ag and bu
orbitals from the sp hybrid orbitals. These ag orbitals interact
with the 3ag HOMO (in the C2h symmetry) of the Re2(CO)8-
(µ-SnH2)2 fragment to form the 3ag orbital (HOMO) and the
2ag orbital (HOMO - 2). A contour diagram of the 3ag
orbital (Figure 16) shows that the Re-Sn overlap decreases
for the bonds not directly associated with the palladium, and
this is consistent with the observed lengthening of these
bonds in 12. The 2ag orbital (Figure 17), which contains the
major direct Re-Re interactions, also contains considerable
interactions with the two Pd atoms consisting mainly of the
1ag orbital on the [Pd(PH3)]2 fragment. The only other
molecular orbital of interest is the 2bu orbital (Figure 18),
which could be viewed as two three-center interactions
between the Pd(PH3) fragments and the associated Re-Sn
bonds. This type of interaction is similar to the one observed
previously for the Ir4(CO)12[Pt(PH3)]2 model cluster.6c Bond-
ing interactions between the palladium and tin atoms in the
frontier orbitals are small, which could explain why the Pd-
Sn bonds are so long.
Figure 14. ORTEP diagram of the molecular structure of Re2(CO)8(µ-
SnPh2)[M(PBut3)]2 (M ) Pd for 12 and Pt for 14) showing 30% thermal
ellipsoid probability.
are formed utilize electrons provided by the Re2Sn2 cluster,
and certain of these bonds are thus delocalized and intrinsi-
cally electron deficient. To understand how the bridging
palladium phosphine groups bond to the Re2Sn2 cluster 12,
we performed Fenske-Hall molecular orbital calculations.
Inorganic Chemistry, Vol. 44, No. 18, 2005 6355

Adams et al.
Figure 17. 2ag molecular orbital showing Re-Re and Re-Pd bonding.
Figure 15. Energy level diagram for the Re2(CO)8(µ-SnH2)2[Pd(PH3)]2
model cluster.
Figure 18. Anti-symmetric 2bu molecular orbital showing the three-center
bonding between the Pd(PH3) groups and the Re-Sn bond.
1
a combination of IR, H and 31P{1H}-NMR, and single-
crystal X-ray diffraction analyses. An ORTEP diagram of
the molecular structure of 13 is given in Figure 19. Selected
intramolecular bond distances and angles are listed in Table
7. The structure of 13 consists of a single Pt(PBut3) group
that occupies a bridging position on the Re(1)-Sn(1) bond
of the starting cluster 10. A single carbonyl ligand bridges
the Re-Pt bond. The Re(1)-Pt(1) bond distance is 2.7932-
(5) Å is similar in length to the nonhydride bridged Re-Pt
bond distances found in Re2Pt(CO)8(PPh3)2(µ-H)2 (2.788(1)
Å),27 [Re3Pt(CO)14(µ-H)2]- (2.7778(14) Å and 2.849(2) Å),28
Re2Pt(CO)8(COD)(µ-H)2 (COD ) 1,5-cyclooctadiene) (2.895-
(1) and 2.741(1) Å),29 and Re3Pt(CO)14(µ-H)3 (2.776(1) Å).29
Figure 16. The antibonding 3ag HOMO of the Re2(CO)8(µ-SnH2)2[Pd-
(PH3)]2 cluster.
Addition of Pt(PBut3) Groups to 10 and 11. Compound
10 also reacted with Pt(PBut3)2 at 68 °C to yield both the
mono- and bis-Pt(PBut3) adducts Re2(CO)8(µ-SnPh2)2[Pt-
(PBut3)] (13) and Re2(CO)8(µ-SnPh2)2[Pt(PBut3)]2 (14) in 38
and 2% yields, respectively. Unfortunately, only trace
amounts of 14 were obtained and efforts to increase the yield
were unsuccessful. Both compounds were characterized by
(27) Beringhelli, T.; Ceriotti, A.; D’Alfonso, G.; Della Pergola, R.
Organometallics 1990, 9, 1053.
(28) Bergamo, M.; Beringhelli, T.; D’Alfonso, G.; Ciani, G.; Moret, M.;
Sironi, A. Organometallics 1996, 15, 1637.
(29) Antognazza, P.; Beringhelli, T.; D’Alfonso, G.; Minoja, A. Organo-
metallics 1992, 11, 1777.
6356 Inorganic Chemistry, Vol. 44, No. 18, 2005

Addition of Pd- and Pt-PBut3 Groups to Re-Sn and Re-Ge Bonds
Scheme 1
stable to be isolated and characterized. However, a mono-
Pt(PBut3) adduct of 11, Re2(CO)8(µ-GePh2)2[Pt(PBut3)] (15),
was obtained in an 18% yield from its reaction with
Pt(PBut3)2 at 127 °C. Under these conditions, there was no
evidence for the formation of a bis-Pt(PBut3) adduct of 11,
and reactions at higher temperatures resulted only in un-
characterizable decomposition products. Compound 15 was
characterized by a combination of IR, 1H and 31P{1H}-NMR,
elemental, and single-crystal X-ray diffraction analyses. This
compound is isostructural with its tin homolog, 13, see Figure
19. Selected intramolecular bond distances and angles are
listed in Table 7. The platinum atom bridges the Re(1)-
Ge(1) bond. The Re(1)-Pt(1) bond distance (2.8428(3) Å)
is even longer than that in 13. The Pt(1)-Ge(1) distance of
2.6324(5) Å is significantly longer than that observed for
typical Pt-Ge single bonds e.g., Pt-Ge ) 2.4495(7) Å in
cis-PtMe(GePh3)(PMe2Ph)2,30 2.437(1) Å in PtEt(GePh3)(PMe2-
Ph)2,31 and 2.4400(4) Å in PtH(GePh3)(PPh3)2,32 indicating
that it is very weak. The Re-Re bond distance in 15 (3.0927-
(3) Å) is significantly longer than in 11 (3.0510(6) Å). As
with 13, the distances between the rhenium atoms and the
platinum-bridged Ge(1) atom have increased significantly
in length to 2.6439(5) Å and 2.6661(5) Å, while the distances
between the rhenium atoms and Ge(2) are slightly shorter
(2.5765(5) Å and 2.5773(5) Å) than those observed in 11
(Re-Ge ) 2.5923(6) Å). As with 13, the increase in all bond
distances associated with the platinum atom can be attributed
to the delocalization of the electron density in those bonds
into the new bonds formed by the associations with the
platinum atom. The addition of Pd(PBut3) and Pt(PBut3)
groups to 10 and 11 are summarized in Scheme 1. We were
not able to isolate and characterized any mono-Pd(PBut3)
adducts of 10 or 11, but in the formation of 12, a mono
adduct must undoubtedly be traversed.
Figure 19. ORTEP diagram of the molecular structures of Re2(CO)8(µ-
XPh2)[Pt(PBut3)] (X ) Sn for 13 and Ge for 15) showing 30% thermal
ellipsoid probability.
Table 7. Selected Intermolecular Distances (Å) and Angles (deg) for
Compounds 13 and 15a
13
15
Re(1)-Re(2)
Re(1)-Sn(1)
Re(1)-Sn(2)
Re(2)-Sn(1)
Re(2)-Sn(2)
Pt(1)-Re(1)
Pt(1)-Sn(1)
Pt(1)-P(1)
3.2034(6)
2.8113(8)
2.7343(8)
2.8143(8)
2.7364(9)
2.7932(5)
2.7505(8)
2.321(2)
1.14(3)
Re(1)-Re(2)
Re(1)-Ge(1)
Re(1)-Ge(2)
Re(2)-Ge(1)
Re(2)-Ge(2)
Pt(1)-Re(1)
Pt(1)-Ge(1)
Pt(1)-P(1)
3.0927(3)
2.6439(5)
2.5773(5)
2.6661(5)
2.5765(5)
2.8428(3)
2.6324(5)
2.3337(13)
1.15(2)
C-O (ave)
C-O (ave)
Re(1)-Re(2)-Sn(1)
Re(1)-Re(2)-Sn(2)
55.246(18) Re(1)-Re(2)-Ge(1)
54.126(18) Re(1)-Re(2)-Ge(2)
54.045(12)
53.137(12)
Pt(1)-Re(1)-Re(2) 110.453(17) Pt(1)-Re(1)-Re(2) 110.643(8)
Pt(1)-Sn(1)-Re(2) 124.78(3)
Pt(1)-Re(1)-Sn(2) 159.22(2)
Re-C-O (ave)
Pt(1)-Ge(1)-Re(2) 134.33(2)
Pt(1)-Re(1)-Ge(2) 162.714(15)
177(1)
Re-C-O (ave)
177(1)
a Estimated standard deviations in the least significant figure are given
in parentheses.
The Pt(1)-Sn(1) bond distance is 2.7505(8) Å. As in 10,
the Re-Re bond is very long, 3.2034(6) Å. The distances
between both rhenium atoms and the platinum-bridged tin
atom, Sn(1), are considerably longer, 2.8143(8) Å and
2.8113(8) Å, than the distances between the rhenium atoms
and Sn(2), 2.7364(9) Å and 2.7343(8) Å. The latter are
slightly shorter than those observed in 10. The lengthening
of the Re-Sn bonds to Sn(1) may be the result of electron
delocalization caused by the presence of the bridging Pt-
(PBut3) group.
Conclusions
Compounds 10 and 11 further expand the chemistry
involving the multiple addition of tin and germanium ligands
to metal carbonyl cluster complexes. Compounds 12-15
extend the families of complexes with electron deficient
metal-metal bonds formed by the addition of M(PBut3) (M
) Pd, Pt) groups to metal cluster compounds. These
compounds represent the first examples of adduct formations
between these M(PBut3) groups and transition metal-main
group metal bonds. Fenske-Hall molecular orbital calcula-
tions have been performed to help to understand the unusual
Compound 14 is isostructural with 12, see Figure 14.
Selected intramolecular bond distances and angles are listed
in Table 6. The lengthening effect observed for the unbridged
Re-Sn bonds in 12 is also present in 14 (Re(1)-Sn(1) )
2.8078(3)), however the effect is not as large. The platinum-
bridged Re-Sn bonds (2.7730(4) Å) and the Pt-Re bonds
(2.7972(2) Å) are all similar in length to those observed in
12. Unlike the addition of Pd(PBut3) to 10, the addition of
the Pt(PBut3) groups does not cause a significant increase in
the Re-Re bond distance (3.2034(6) Å for 13 and 3.2008-
(3) Å for 14).
(30) Ozawa, F.; Hikida, T.; Hasebe, K.; Mori, T. Organometallics 1998,
17, 1018.
(31) Hasebe, K.; Kamite, J.; Mori, T.; Katayama, H.; Ozawa, F. Organo-
metallics 2000, 19, 2022.
Compound 11 also reacted with Pd(PBut3)2 at room
temperature, but the products obtained were not sufficiently
(32) Habereder, T.; No¨th, H. Appl. Organomet. Chem. 2003, 17, 525.
Inorganic Chemistry, Vol. 44, No. 18, 2005 6357

Adams et al.
metal-metal bonding in these complexes. The metal-metal
bonding in 10 is dominated by the Re-Sn interactions. The
Re-Re interaction is weak, and the Re-Re distance is
unusually long. The Pd-Sn, Pt-Sn, and Pt-Ge bonds are
also weak. Interestingly, in 12 the unbridged Re-Sn bonds
were noticeably weakened when the Pd(PBut3) group was
added to the neighboring Re-Sn bond.
Pt(PBut3)2. The authors thank Drs. M. B. Hall and C. E.
Webster for instruction in the use of the Fenske-Hall
molecular orbital program package and Prof. W. Glaberson
and Dr. Brettschneider for the use of their DASWIN program
(available at www.Phys.HUJI.AC.IL/∼glabersn). M.J. thanks
the Swedish Research Council for support.
Supporting Information Available: CIF files are available for
each of the structural analyses. This material is available free of
charge via the Internet at http://pubs.acs.org.
Acknowledgment. This research was supported by the
Office of Basic Energy Sciences of the U.S. Department of
Energy under Grant No. DE-FG02-00ER14980 and the USC
Nanocenter. We thank Strem for a donation of a sample of
IC051077X
6358 Inorganic Chemistry, Vol. 44, No. 18, 2005