Active Role of a Germylene Ligand in Promoting Reactions of Platinum Complexes with Oxygen and Sulfur Dioxide

Article abstract of DOI:10.1021/ic9808044

The reaction of (Et₃P)₂PtGe[N(SiMe₃)₂]₂ with dioxygen yields (Et₃P)₂Pt(μ-η²-O ₂)Ge[N(SiMe₃)₂]₂ (1). Exposure of 1 to light resulted in a rearrangement to (Et₃P)₂PtO₂Ge[N(SiMe₃) ₂]₂ (2a), the first example of a bidentate, dianionic germanate ligand. The isomerization was judged to occur via an intramolecular O-O bond scission and rotation of the Pt-Ge bond. No free germylene was detected, and the reaction was found to be zero order. An analogue of 2a was prepared by direct reaction of (Ph₃P)₂PtO₂ with Ge[N(SiMe₃)₂]₂ yielding (Ph₃P)₂PtO₂-Ge[N(SiMe₃) ₂]₂ (2b). Addition of SO₂ to 1 results in the formation of the bridging sulfate (Et₃P)₂Pt(μ-η²-SO ₄)-Ge[N(SiMe₃)₂]₂ (3). An infrared spectroscopy study of the sulfate reaction was performed using oxygen-18. The results indicate that direct insertion of SO₂ into the O-O bond does not occur. Formaldehyde was also observed to insert into the Pt-O bond of 1 giving (Et₃P)₂Pt(μ-η²-OCH ₂OO)Ge[N(SiMe₃)₂]₂ (5).

Full text of DOI:10.1021/ic9808044

Inorg. Chem. 1998, 37, 6461-6469
6461
Active Role of a Germylene Ligand in Promoting Reactions of Platinum Complexes with
Oxygen and Sulfur Dioxide
K. E. Litz, M. M. Banaszak Holl,* J. W. Kampf, and G. B. Carpenter†
Department of Chemistry, The University of Michigan, Ann Arbor, Michigan 48109-1055
ReceiVed July 13, 1998
The reaction of (Et3P)2PtGe[N(SiMe3)2]2 with dioxygen yields (Et3P)2Pt(µ-η2-O2)Ge[N(SiMe3)2]2 (1). Exposure
of 1 to light resulted in a rearrangement to (Et3P)2PtO2Ge[N(SiMe3)2]2 (2a), the first example of a bidentate,
dianionic germanate ligand. The isomerization was judged to occur via an intramolecular O-O bond scission
and rotation of the Pt-Ge bond. No free germylene was detected, and the reaction was found to be zero order.
An analogue of 2a was prepared by direct reaction of (Ph3P)2PtO2 with Ge[N(SiMe3)2]2 yielding (Ph3P)2PtO2-
Ge[N(SiMe3)2]2 (2b). Addition of SO2 to 1 results in the formation of the bridging sulfate (Et3P)2Pt(µ-η2-SO4)-
Ge[N(SiMe3)2]2 (3). An infrared spectroscopy study of the sulfate reaction was performed using oxygen-18. The
results indicate that direct insertion of SO2 into the O-O bond does not occur. Formaldehyde was also observed
to insert into the Pt-O bond of 1 giving (Et3P)2Pt(µ-η2-OCH2OO)Ge[N(SiMe3)2]2 (5).
Introduction
Chart 1. Common Coordination Modes of Dioxygen
Oxidation of low-valent organometallic complexes and transi-
tion metal catalysts by molecular oxygen is a field of both
academic and industrial importance for several reasons. Metal
peroxo complexes are believed to be important intermediates
in the Pt-catalyzed synthesis of sulfuric acid and are thought to
precede the partial oxidation of saturated and unsaturated
hydrocarbons by transition metal catalysts.1 Metal peroxo
complexes also play a role in the uptake and utilization of
dioxygen by biological and biomimetic transition-metal com-
plexes.2 Oxygen is also known to poison catalysts and degrade
organotransition metal compounds.3 The many ways that oxygen
can interact with metal complexes are underscored by the wide
variety of coordination modes displayed in Chart 1.
were reported approximately 30 years ago.6 The first bimetallic
type X peroxo complex containing two transition metals, [Ir2I2-
(CO)2(µ-η2-O2)(dppm)2]2, was published by Cowie et al. in
1990.7 This compound was unique in having a peroxo moiety
bridging a metal-metal bond. All previous bridging peroxo
groups had been of types III-VII. Cowie’s bridging peroxide
also forms a bound sulfate in the presence of SO2 (eq 2). Main
Many dioxygen complexes have been observed to react with
SO2 yielding metal sulfate adducts.4,5 The first studies with
group 10 metals, which included synthesis of (Ph3P)2PtO2 and
subsequent reaction with SO2 to form the bound sulfate (eq 1),
(2)
(1)
† Department of Chemistry, Brown University, Providence, RI 02912.
(1) Thorough reviews of this subject can be found in the following: (a)
Sheldon, R. A.; Kochi, J. K. Metal-Catalyzed Oxidation of Organic
Compounds; Academic Press: New York, 1981. (b) Pines, H. The
Chemistry of Catalytic Hydrocarbon ConVersion; Academic Press:
New York, 1981; pp 213-230. (c) Shilov, A. E. Catalysis by Metal
Complexes: ActiVation of Saturated Hydrocarbons by Transition Metal
Complexes; D. Reidel Publishing: Boston, MA, 1984; pp 62-141.
(d) Jorgensen, K. A. Chem. ReV. 1989, 89, 431. (e) Simandi, L. I.
Catalysis By Metal Complexes: Catalytic ActiVation of Dioxygen by
Metal Complexes; Kluwer Academic Publishers: Boston, MA, 1992.
(2) Selected reviews can be found in the following: (a) Martell, A. E.;
Sawyer, D. T. Oxygen Complexes and Oxygen ActiVation by Transition
Metals; Plenum: New York, 1988. (b) Tyeklar, Z.; Karlin, K. D. Acc.
Chem. Res. 1989, 22, 241. (c) Sawyer, D. T., Ed. Oxygen Chemistry;
Oxford University Press: New York, 1991. (d) Lippard, S. J.; Berg,
J. M. Principles of Bioinorganic Chemistry; University Science
Books: Mill Valley, CA, 1994.
group type X peroxo complexes are known for digermenes and
disilenes; however, their reactivity with SO2 has not been
reported.8,9 As part of a general study of germylenes as “active”
ligands for transition metals, we explored the reactivity of
(Et3P)2PtGe[N(SiMe3)2]2 with dioxygen. The germylene ligand
(6) (a) Takahasi, S.; Sonogashira, K.; Hagihara, N. J. Chem. Soc. Jpn.
1966, 87, 610. (b) Wilke, G.; Shott, H.; Heimbach, P. Angew. Chem.,
Int. Ed. Engl. 1967, 6, 92. (c) Cook, C. D.; Jauhal, G. S. Inorg. Nucl.
Chem. Lett. 1967, 3, 31.
(7) (a) Xiao, J.; Santarsiero, B. D.; Vaartstra, B. A.; Cowie, M. J. Am.
Chem. Soc. 1993, 115, 3212. (b) Vaartstra, B. A.; Xiao, J.; Cowie,
M. J. Am. Chem. Soc. 1990, 112, 9425.
(8) Masamune, S.; Batcheller, S. A.; Park, J.; Davis, W. M. J. Am. Chem.
Soc. 1989, 111, 1888.
(9) (a) Michalczyk, M. J.; West, R.; Michl, J. J. Chem. Soc., Chem.
Commun. 1984, 1525. (b) Fink, M. J.; De Young, D. J.; West, R.;
Michl, J. J. Am. Chem. Soc. 1983, 105, 1070. (c) Fink, M. J.; Haller,
K. J.; West, R.; Michl, J. J. Am. Chem. Soc. 1984, 106, 822.
(3) Hegedus, L. L.; McCabe, R. W. Catalyst Poisoning; Marcel Dekker
Inc.: New York, 1984.
(4) Valentine, J. S. Chem. ReV. 1973, 73, 235.
(5) Kubas, G. J. Acc. Chem. Res. 1994, 27, 183.
10.1021/ic9808044 CCC: $15.00 © 1998 American Chemical Society
Published on Web 11/19/1998

6462 Inorganic Chemistry, Vol. 37, No. 25, 1998
Litz et al.
(Et3P)2PtO2Ge[N(SiMe3)2]2 (2a). In a Pyrex bomb reactor, a 10 mL
hexane solution containing 500 mg (0.61 mmol) (Et3P)2PtGe-
[N(SiMe3)2]2 was stirred, blanketed with 1 atm O2, and exposed to a
100 W Sylvania Par38 mercury lamp. After 10 min, the initially golden
orange solution became olive colored. Hexane was removed in Vacuo
leaving a tan oil which crystallized upon standing for 24 h. Colorless
crystals were obtained by washing with 1 mL of acetonitrile (307 mg,
59% yield): 1H NMR (C6D6) δ 0.70 (s, 36 H, Si(CH3)3), 0.88 (m, 18
H, CH2CH3), 1.46 (m, 12 H, CH2CH3); 13C NMR (C6D6) δ 15.8 (m,
CH2CH3), 8.2 (m, CH2CH3), 6.6 (s, Si(CH3)3); 31P{1H} NMR (C6D6) δ
was found to cooperate in an intimate fashion with the central
platinum center in the binding of O2. Given the recent
precedence for reversible germanium-oxygen bond forma-
tion,10,11 we anticipated that this system might allow for
reversible binding of O2 and/or the reaction products of the metal
peroxo complexes.
In this paper, we report the synthesis and characterization of
the first mixed transition-metal/main-group peroxo complex,
(Et3P)2Pt(µ-η2-O2)Ge[N(SiMe3)2]2 (1). Complex 1 photochemi-
cally rearranges to (Et3P)2PtO2Ge[N(SiMe3)2]2 (2a), a complex
containing a unique bidentate germanate ligand. In addition, 1
undergoes reaction with SO2 and H2CdO leading to the bridging
adducts (Et3P)2Pt(µ-η2-SO4)Ge[N(SiMe3)2]2 (3) and (Et3P)2Pt-
(µ-η2-OCH2OO)Ge[N(SiMe3)2]2 (5), respectively. The mecha-
nism of sulfate formation was probed using 18O2 labeling
experiments and infrared spectroscopy. Reduction of 3 with Na/
Hg amalgam regenerated (Et3P)2PtGe[N(SiMe3)2]2. The activa-
tion of dioxygen, its conversion to sulfate, and facile removal
demonstrate a complete stepwise cycle for the conversion of
1
1.8 (s, w/195Pt satellites, JPt-P ) 3284 Hz). Anal. Calcd for C24H66-
GeN2O2P2PtSi4: C, 33.65; H, 7.76; N, 3.27. Found: C, 33.73; H, 7.71;
N, 3.23.
(Ph3P)2PtO2Ge[N(SiMe3)2]2 (2b). In a round-bottom flask protected
from light by aluminum foil, a 5 mL ethereal solution of 1.8 g (Ph3P)2-
PtO2 (2.39 mmol) and 0.943 g Ge[N(SiMe3)2]2 was allowed to stir until
all solids dissolved and the solution became dark orange-brown with
formation of a white precipitate. After 1 h, all volatiles were removed
in Vacuo. The solids were washed with 2 × 5 mL pentane leaving
1.70 g (62% yield) off-white microcrystals: 1H NMR (C6D6) δ 0.64
(s, 36 H, Si(CH3)3), 6.91 (m, 24 H, PPh3), 7.69 (m, 6 H, PPh3); 13C-
{1H} NMR (C6D6) δ 6.15 (s, Si(CH3)3), 127.8 (d, phenyl), 128.1 (s,
phenyl), 130.2 (s, phenyl), 135.1 (q, phenyl); 31P{1H} NMR (C6D6) δ
SO2 to [SO4]2-
.
1
6.87 (s, w/195Pt satellites, JPt-P ) 3461 Hz). Anal. Calcd for C48H66-
Experimental Section
GeN2O2P2PtSi4: C, 50.35; H, 5.81; N, 2.45. Found: C, 50.47; H, 6.12;
N, 2.43. IR (KBr pellet) (cm-1): 3058 vs, 2952 s, 2896 s, 1483 m,
1434 s, 1349 m, 1279 s, 1244 vs, 1188 s, 1131 s, 1096 vs, 1033 w,
998 w, 913 vs, 864 vs, 843 vs, 745s, 688 vs, 527 vs. MS (CI, CH4):
1146 amu, isotope distribution consistent with that calculated for C48H67-
GeN2O2P2PtSi4.
All manipulations were performed using air-free techniques and dry,
deoxygenated solvents unless otherwise stated. (Et3P)2PtGe[N(SiMe3)2]2,11a
(Ph3P)2PtO2,12 (Et3P)2PtCO3,13 and Ge[N(SiMe3)2]2 were prepared
14
according to published procedures. Oxygen (99.6%) and benzil were
utilized as received from Aldrich Chemical. Anhydrous sulfur dioxide
was used as received from Matheson Gas Products. Benzene-d6,
benzene, diethyl ether, THF, and pentane solvents were all degassed
and distilled over sodium benzophenone ketyl. Acetonitrile was
degassed, dried by refluxing over P2O5, and distilled onto activated 4
Å molecular sieves. The 95% 18O2 was obtained from Icon Services.
All other reagents were used as received from Aldrich Chemical. IR
spectra were obtained on a Nicolet 5DXB FT-IR or a Bio-Rad FTS40
(Et3P)2Pt(µ-η2-SO4)Ge[N(SiMe3)2]2 (3). Complex 1 was generated
in situ by allowing (Et3P)2PtGe[N(SiMe3)2]2 (0.200 g 0.242 mmol) to
react with 1 atm dioxygen for 25 min while protected from light. When
the orange color faded, all volatiles were removed in Vacuo, the solids
were dissolved in toluene, and 1 atm of SO2 was allowed to blanket
the stirred solution for 2 h. The white solid that precipitated from
solution was recovered by filtration and washed with hexane (0.120 g,
54% yield): 1H NMR (CDCl3) δ 0.35 (s, 36 H, Si(CH3)3), 1.16 (m, 18
H, CH2CH3), 1.98 (m, 6 H, CH2CH3), 2.10 (m, 6 H, CH2CH3); 13C-
{1H} NMR (CDCl3) δ 6.96 (s, SiMe3), 8.45 (s, CH2CH3), 9.54 (s,
CH2CH3), 14.60 (d, CH2CH3), 21.06 (d, CH2CH3); 31P{1H} NMR (C6D6)
δ 21.7 (d, w/195Pt satellites, 1JPt-P ) 1950 Hz, 2JP-P ) 16.3 Hz), -7.6
1
FT-IR as NaCl/Nujol mulls or as pressed KBr or CsI pellets. H, 13C,
and 31P NMR spectra were obtained in the listed deuterated solvents
on a Bruker AM-360 spectrometer (360.1, 90.6, and 145.8 MHz) and
referenced to residual protons, solvent carbons, and external 85% H3-
PO4 in D2O, respectively. Microanalyses were performed by University
of Michigan Analytical Services.
1
2
(d w/195Pt satellites, JPt-P ) 4020 Hz, JP-P ) 17 Hz); IR (Nujol/
NaCl) (cm-1) 1283 s, 1153 s, 892 s. Anal. Calcd for C24H66GeN2O2P2-
PtSi4: C, 31.30; H, 7.22; N, 3.04. Found: C, 30.51; H, 6.96; N, 2.90.
(Et3P)2Pt(µ-η2-OC(O)O)Ge[N(SiMe3)2]2 (4). A 5 mL THF solution
containing 200 mg of (Et3P)2PtCO3 and 160 mg of Ge[N(SiMe3)2]2
was stirred for 2 h. The solvent was removed in vacuo. A white powder
was recovered by filtration from pentane (211 mg, 59% yield): 1H
NMR (CDCl3) δ 0.35 (s, 36H, Si(CH3)3), 1.12 (m, 18H, CH2CH3), 1.94
(m, 6H, CH2CH3), 2.02 (m, 6H, CH2CH3); 13C{1H} NMR (CDCl3) δ
162.7 (d, trans-3JP-C ) 2.41 Hz, CO3), 20.5 (m, CH2CH3), 14.6 (m,
CH2CH3), 8.9 (m, CH2CH3), 7.9 (m, CH2CH3), 6.8 (s, Si(CH3)3); 31P-
(Et3P)2Pt(µ-η2-O2)Ge[N(SiMe3)2]2 (1). A hexane solution containing
400 mg (0.48 mmol) of (Et3P)2PtGe[N(SiMe3)2]2 was stirred, blanketed
with 1 atm of O2, and protected from light. The initially golden yellow
solution developed a light green hue after 20 min. An aliquot analyzed
by 1H NMR revealed quantitative conversion to 1. The solvent volume
was reduced to one-third. The product was isolated by cold filtration
on a fritted glass filter disk as a bright yellow-green solid (272 mg,
66% yield). 1 was observed to be moisture and light sensitive: 1H NMR
(C6D6) δ 0.69 (s, 36 H, Si(CH3)3), 0.82 (m, 18 H, CH2CH3), 1.34 (m,
6 H, CH2CH3), 1.55 (m, 6 H, CH2CH3); 13C{1H} NMR (C6D6) δ 21.2
(m, CH2CH3), 14.8 (m, CH2CH3), 9.5 (m, CH2CH3), 8.3 (m, CH2CH3),
7.4 (s, Si(CH3)3); 31P{1H} NMR (C6D6) δ 16.2 (d, w/195Pt satellites,
1
{1H} NMR (CDCl3) δ 21.01 (d, w/195Pt satellites, JPt-P ) 1934 Hz,
1
2
2JP-P ) 17 Hz), -6.6 (d w/195Pt satellites, JPt-P ) 3772 Hz, JP-P 17
Hz); IR (KBr) (cm-1) 1652/1623 ν(CdO). Anal. Calcd for C25H66-
GeN2O3P2PtSi4: C, 33.94; H, 7.52; N, 3.17. Found: C, 33.93; H, 7.21;
N, 2.96.
2
1
1JPt-P ) 1839 Hz, JP-P ) 10.6 Hz), -5.3 (d w/195Pt satellites, JPt-P
) 3125 Hz, 2JP-P 10.3 Hz). Anal. Calcd for C24H66GeN2O2P2PtSi4: C,
33.65; H, 7.76; N, 3.27. Found: C, 33.70; H, 7.43; N, 3.15.
(Et3P)2Pt(µ-η2-OCH2OO)Ge[N(SiMe3)2]2 (5). THF was condensed
into a 10 mL round-bottom two-neck flask charged with 0.200 g (0.24
mmol) (Et3P)2PtGe[N(SiMe3)2]2, and the solution was allowed to stir
under 1 atm O2 for 1 h while protected from light. The volatiles were
removed in vacuo to remove excess dioxygen. The contents were
redissolved in THF, and 0.075 g of paraformaldehyde was added. The
reaction was allowed to stir for 3 h. Upon completion, the solvent was
removed in Vacuo, and the contents of the round-bottom flask were
filtered with benzene to separate from excess paraformaldehyde. The
product was recrystallized from cold hexane as an off-white powdery
solid (0.060 g, 29% yield): 1H NMR (C6D6) δ 0.72 (s, 36 H, Si(CH3)3),
0.77 (m, 18 H, PCH2CH3), 1.42 (m, 6 H, PCH2CH3), 1.71 (m, 6 H,
(10) (a) Koe, J. R.; Tobita, H.; Suzuki, T.; Ogino, H. Organometallics 1992,
11, 150. (b) Koe, J. R.; Tobita, H.; Ogino, H. Organometallics 1992,
11, 2479. (c) Ogino, H.; Tobita, H. AdV. Organomet. Chem. 1998,
42, 223.
(11) (a) Litz, K. E.; Henderson, K.; Gourley, R. W.; Banaszak Holl, M.
M. Organometallics 1995, 14, 5008. (b) Litz, K. E.; Bender, J. E.;
Kampf, J. W.; Banaszak Holl, M. M. Unpublished results.
(12) Nyman, C. J.; Wymore, C. E.; Wilkinson, G. J. Chem. Soc. (A) 1968,
561.
(13) Hayward, P. J.; Blake, D. M.; Wilkinson, G.; Nyman, C. J. J. Am.
Chem. Soc. 1970, 92, 5873.
(14) Gynane, M. J. S.; Harris, D. H.; Lappert, M. F.; Power, P. P.; Riviere,
P.; Riviere-Baudet, M. J. Chem. Soc., Dalton Trans. 1977, 2004.
4
3
PCH2CH3), 5.65 (d with 195Pt satellites: JP-H ) 10.0 Hz, JPt-H
)

Reactions of Pt Complexes with O2 and SO2
Inorganic Chemistry, Vol. 37, No. 25, 1998 6463
40.0 Hz); 13C{1H} NMR (C6D6) δ 7.86 (s, Si(CH3)3), 8.49 (m,
PCH2CH3), 8.78 (m, PCH2CH3) 14.4 (m, PCH2CH3), 19.5 (m, PCH2-
CH3), 99.3 (br s, Pt-OCH2); 31P{1H} NMR (C6D6) δ 19.50 (d with
range 5-45°. The largest peak in the final difference map was 5.750
e Å-3 and was associated with Pt.
C24H66N2O2Si4P2GePt (2a). A colorless crystal having dimensions
1.0 × 0.6 × 0.5 mm was mounted in a flame-sealed glass capillary.
X-ray data collection was performed using a Siemens P4 single-crystal
diffractometer. Three standard reflections were measured after every
97 reflections; a 16.5% decrease in standard intensities was observed
and corrected. Data reduction included profile fitting and a semiem-
pirical absorption correction based on ψ-scans.17 The structure was
determined by Patterson methods. Atoms heavier than carbon were
refined with anisotropic displacement parameters. All expected hydro-
gen atoms were introduced in ideal positions. Final refinement on F2
was carried out using SHELXL 93.
C26H70N2O4.5SSi4P2GePt (3‚0.5THF). A colorless thin, needlelike
crystal, having the dimensions 0.06 × 0.24 × 0.46 mm, was mounted
in paratone N hydrocarbon oil and placed in a cold nitrogen stream
(158(2) K). X-ray data collection was performed on a Siemens SMART
diffractometer with a CCD area detector equipped with a normal focus
Mo-target X-ray tube (λ ) 0.710 73 Å) operated at 2000 W power (50
kV, 40 mA). The frames were integrated with Siemen’s SAINT software
package with a narrow frame algorithm. All non-hydrogen atoms were
allowed to refine anisotropically with hydrogen atoms placed in
idealized positions. An absorption correction was applied using
SADABS.18
2
1
195Pt satellites: JP-P ) 14.7 Hz, JPt-P ) 1833 Hz), -6.74 (d with
2
1
195Pt satellites: JP-P ) 14.7 Hz, JPt-P ) 3494 Hz); IR (Nujol/NaCl)
[cm-1, (intensity)] 1251 (vs), 1159 (m), 1110 (m), 1033 (m), 941 (m),
899 (s), 871 (m), 765 (m), 667 (w). Anal. Calcd: C, 33.86; H, 7.73;
N, 3.16. Found: C, 33.86; H, 7.55; N, 3.04.
Kinetics of Isomerization of 1 to 2. An evacuated, flame-sealed 5
mm Pyrex NMR tube containing a 0.25 mL of C6D6 solution of 0.050
g (0.058 mmol) of 1 was irradiated at a distance of 17 cm from a 100
W Sylvania Par38 mercury lamp at a temperature of 30 °C. Kinetic
measurements were obtained at 0, 300, 480, 720, and 960 s intervals
by integration of the 31P NMR signal. The final value was monitored
at 81% completion. Isomerization was zero order with an observed
rate constant of 0.000854 s-1 with a least-squares fit to the linear
equation y ) 0.000854x + 0.0580 (r ) 0.998).
Photolysis of 1 with a Germylene Trap. As above, a benzene-d6
solution containing 0.050 g of 1 (0.058 mmol) and 10 equiv of benzil
(0.120 g, 0.580 mmol) was photolyzed 17 cm from a 100 W Sylvania
Par 38 mercury lamp at a temperature of 30 °C. Quantitative conversion
to 2 was observed. No evidence of trapped germylene product was
detected.
Reduction of 3. A 5 mg (0.005 mmol) amount of 3 was reduced in
an excess of sodium/mercury amalgam in THF. After being stirred 1
h, the initially colorless solution became olive-drab in color. THF was
removed in Vacuo. The remaining solids were extracted from the
amalgam with C6D6 and filtered in the drybox through a Pasteur pipet
containing packed glass wool. 1H NMR demonstrated quantitative
conversion to (Et3P)2PtGe[N(SiMe3)2]2. 1H NMR (C6D6): δ 0.52 (s, 36
H, SiMe3), 1.00 (m, 18 H, PCH2CH3), and 1.45 (m, 12 H, PCH2CH3);
consistent with the previously published data for (Et3P)2PtGe-
[N(SiMe3)2]2.
C27H66Cl6GeN2O3P2PtSi4 (4‚2CHCl3). X-ray-quality crystals were
grown from concentrated chloroform solutions of 4 by slow evaporation
of solvent. A small, colorless crystal (dimensions 0.38 × 0.30 × 0.16
mm) was selected, mounted on a glass fiber in paratone N hydrocarbon
oil, and placed in a cold stream of nitrogen (153(2) K) on the
diffractometer. X-ray data collection was performed on a Siemens
SMART diffractometer with a CCD area detector equipped with a
normal focus Mo-target X-ray tube (λ ) 0.710 73 Å) operated at 2000
W power (50 kV, 40 mA). The frames were integrated with Siemen’s
SAINT software package with a narrow frame algorithm All non-
hydrogen atoms were allowed to refine anisotropically, with hydrogen
atoms placed in idealized positions. An absorption correction was
applied using SADABS. The largest peak in the final difference map
was 5.249 e Å-3 and was associated with Pt.
Oxygen-18 Study of the Reaction between 1 and SO2. Oxygen-
18 labeled 1 (18O2-1) was prepared according to the procedure given
above using 95% 18O2 (ICON Services). Oxygen-18 labeled 3 (18O2-3)
was prepared by addition of unlabeled SO2 to 18O2-1 in the manner
described for the synthesis of 3. Infrared spectra for 1, 18O2-1, 3, and
18O2-3 were recorded from 4000 to 400 cm-1 as KBr pellets and from
600 to 200 cm-1 as CsI pellets.
Results
A yellow-orange solution of (Et3P)2PtGe[N(SiMe3)2]2 was
exposed to dioxygen resulting in the rapid formation of a
colorless solution. Protection of the reaction from ambient light
allowed formation of a new product containing cis inequivalent
PEt3 groups as indicated by 1H, 13C, and 31P NMR spectroscopy.
The spectroscopic signature of an unsymmetrical, square-planar
Pt(II) complex supported assigning this new product as the
peroxo complex, (Et3P)2Pt(µ-η2-O2)Ge[N(SiMe3)2]2 (1). IR
spectra of peroxo bimetallic complexes typically show an O-O
stretch between 800 and 950 cm-1.4 However, the assignment
of the O-O stretch in 1 was made difficult due to the overlap
of strong Ge-N and Si-N stretches in that region. Synthesis
of 18O2-labeled 1 allowed the observation of a new band at 796
cm-1 which has been tentatively assigned as the 18O-18O
stretch. To confirm our structural assignment and to examine
the impact of the Pt-Ge moiety on O-O activation, an X-ray
structure determination of 1 was performed.
X-ray-quality crystals were grown from an ether solution
maintained at -10 °C and protected from light (Table 1). The
presence of a bridging peroxo moiety spanning a Pt-Ge bond
was confirmed (Figure 1). The four-membered ring displays
noticeable distortion as indicated by the O-Pt-Ge-O dihedral
angle of 13.4°. This puckering of the ring is related to the
rotation of the GeN2 fragment away from a perpendicular
orientation to the PtP2 fragment by 32°. The variation in bond
length required in the four-membered ring (Pt1-O1 ) 2.050(8)
X-ray Crystal Structure Determinations. Table 1 provides a
general summary of crystallographic parameters. For structures 1 and
2a, extensive checks were made to confirm the choice of space group
and check for higher symmetry. The metric symmetry was established
using the LePage algorithm.15 A search for additional missing symmetry
was performed utilizing subroutines available as part of the PLATON
suite of programs (PLATON, Program for the Automated Analysis of
Molecular Geometry, A. L. Spek, Utrecht University, The Netherlands,
1997).16 Neither structure exhibited problems during the refinement
(correlations between least-squares parameters, unusually distorted
geometries) typical of missing symmetry. Both structures were therefore
determined to contain 2 crystallographically independent molecules per
asymmetric unit.
C24H66N2O2Si4P2GePt (1). A colorless crystal of 1 having dimen-
sions 0.42 × 0.32 × 0.26 mm was mounted on a glass fiber with
paratone N hydrocarbon oil and placed in a cold stream of nitrogen at
148(2) K. X-ray data collection was performed on a Siemens R3/v
automated diffractometer with graphite-monochromatized Mo KR
radiation (λ ) 0.710 73 Å) and solved by direct methods with the
programs SHELXTL PLUS and SHELXL-93. A 15% decrease in
standard intensities was observed and corrected. The data were corrected
for absorption using a semiempirial method based on ψ-scans.17 All
non-hydrogen atoms were allowed to refine anisotropically, with
hydrogen atoms placed in idealized positions. Additional details: scan
method ω, scan rate 10°/min, background-to-scan ratio 0.5, ω scan
(15) LePage, Y. J. J. Appl. Crystallogr. 1982, 15, 225.
(16) Spek, A. L. Acta Crystallogr. 1990, 46A, C34.
(17) XEMP, V. 3.43; Siemens Analytical X-ray Instruments, Inc.: Madison,
WI, 1988.
(18) Blessing, R. H. Acta Crystallogr. 1995, A51, 33.

Table 1. X-ray Data
compd
1
2a
3
4
empirical formula
fw
C24H66GeN2O2P2PtSi4
856.77
C24H66GeN2O2P2PtSi4
856.77
C24H66GeN2O4P2PtSSi4‚0.5C4H8O
956.88
C25H64GeN2O3P2PtSi4‚2CHCl3
1121.50
temp
148(2) K
298(2) K
158(2) K
158(2) K
wavelength
cryst system
space group
unit cell dimens
0.710 73 Å
triclinic
P1h (No. 2)
a ) 11.689(2) Å, R ) 108.74(1)°
b ) 18.753 (2) Å,â ) 97.56(1)°
c ) 19.160(2) Å, γ ) 95.44(1)°
3900.5(9) Å3, 4
4.580 mm-1
0.710 73 Å
triclinic
P1h (No. 2)
0.710 73 Å
monoclinic
P21/c (No. 14)
a ) 11.6826(1) Å
b ) 22.6072(2) Å, â ) 94.5360(10)°
c ) 17.19130(10) Å
4526.19(5) Å3, 4
4.003 mm-1
0.710 73 Å
monoclinic
P21/c (No. 14)
a ) 12.346(3) Å, R ) 68.97(2)°
b ) 17.752(3) Å, â ) 81.76(2)°
c ) 20.521(5) Å, γ ) 74.07(2)°
4032(2) Å3, 4
a ) 23.904(6) Å
b ) 11.046(3) Å, â ) 107.272(13)°
c ) 19.613(4) Å
4946(2) Å3, 4
V, Z
abs coeff
4.431 mm-1
3.947 mm-1
F(000)
1744
1744
1952
2256
cryst size (mm)
θ range for data collcn
limiting indices
0.42 × 0.32 × 0.26 mm
1.0 × 0.6 × 0.5 mm
0.06 × 0.24 × 0.46 mm
0.38 × 0.30 × 0.16 mm
2.50-22.50°
1.87-25°
2.16-29.42°
2.57-26°
-12 e h e 12, -18 e k e 18,
-1 e l e 20
10 637
-1 e h e 14, -19 e k e 19,
-24 e l e 24
16 039
-15 e h e 15, -30 e k e 29,
-23 e l e 22
46 853
-29 e h e 28, -1 e k e 13,
-1 e l e 24
12 299
reflcns collcd
indepdt reflcns
abs corr
9823
ψ-scans
13904 [R(int) ) 0.0574]
ψ-scans
11512 [R(int) ) 0.0339]
SADABS
9704 [R(int) ) 0.0761]
SADABS
max and min transm
refinement method
data/restraints/param
goodness-of-fit on F2
final R indices [I > 2σ(I)
R indices (all data)a
extinction coeff
1.341and 0.636
full-matrix least-squares on F2
9814/0/688
0.292 and 0.106
full-matrix least-squares on F2
13902/0/409
0.915 and 0.614
full-matrix least-squares on F2
11492/0/420
0.937 and 0.338
full-matrix least-squares on F2
9700/0/435
1.085
0.887
1.092
0.980
R1 ) 0.0550, wR2 ) 0.1421
R1 ) 0.0750, wR2 ) 0.1488
no corr applied
5.750 and -2.520 e Å-3
Siemens P4U,
R1 ) 0.0633, wR2 ) 0.1498
R1 ) 0.1263, wR2 ) 0.1715
no corr applied
R1 ) 0.0264, wR2 ) 0.0728
R1 ) 0.0315, wR2 ) 0.0755
0.000 22(5)
R1 ) 0.0675, wR2 ) 0.1693
R1 ) 0.0876, wR2 ) 0.1755
no corr applied
5.249 and -3.430 e Å-3
Siemens P4U,
largest diff peak and hole
diffractometer
1.054 and -1.710 e Å-3
Siemens P4
1.563 and -0.998 e Å-3
Siemens SMART-CCD /LT-2
equipped with LT-2
equipped with LT-2
a R1 ) ∑(|Fo| - |Fc|)/∑|Fo|; wR2 ) [∑w(Fo - Fc2)/∑w(Fo )2]1/2
.
2
2

Reactions of Pt Complexes with O2 and SO2
Inorganic Chemistry, Vol. 37, No. 25, 1998 6465
Figure 2. ORTEP of 2a (50% probability). Selected bond lengths (Å)
and angles (deg) are reported for one of the two independent molecules
in the unit cell. No chemically significant deviations were noted: Pt1-
O1 ) 2.070(6); Pt1-O2 ) 2.032(8); Pt1-P1 ) 2.226(3); Pt1-P2 )
2.224(3); Ge1-O1 ) 1.738(8); Ge1-O2 ) 1.784(7); Pt1-Ge1 )
2.770(2); Ge1-N1 ) 1.841(10); Ge1-N2 ) 1.861(9); O2-Pt1-O1
) 78.9(3); O2-Pt1-P2 ) 89.4(2); O1-Pt1-P2 ) 168.2(2) O2-Pt1-
P1 ) 169.4(2); O1-Pt1-P1 ) 90.5(2); P2-Pt1-P1 ) 101.29(14);
Pt1-O1-Ge1 ) 92.9(3); Pt1-O2-Ge1 ) 92.8(3); O1-Ge1-O2 )
95.4(3); N1-Ge1-N2 ) 111.3(4); N1-Ge1-O1 ) 111.3(4); N2-
Ge1-O1 ) 112.7(4); N2-Ge1-O2 ) 112.4(4); N1-Ge1-O2 )
112.9(4).
Figure 1. ORTEP of 1 (50% probability). Selected bond lengths (Å)
and angles (deg) are reported for one of the two independent molecules
in the unit cell. No chemically significant deviations were noted: Pt1-
P1 ) 2.318(3); Pt1-P2 ) 2.219(3); Pt1-O1 ) 2.050(8); O1-O2 )
1.503(11); O2-Ge1 ) 1.847(8); Ge1-N1 ) 1.847(8); Ge1-N2 )
1.888(8); Pt1-Ge1 ) 2.4286(13); P1-Pt1-P2 ) 96.17(11); P2-Pt1-
Ge1 ) 107.79(9); P1-Pt1-Ge1 ) 156.00(8); P1-Pt1-O1 ) 88.5(2);
O1-Pt1-Ge1 ) 67.8(2); Pt1-O1-O2 ) 107.0(5); O1-O2-Ge1 )
96.9(5); O2-Ge1-Pt1 ) 83.3(2); N1-Ge1-N2 ) 110.2(4); N1-
Ge1-Pt1 ) 119.9(3); N2-Ge1-Pt1 ) 126.8(3); O2-Ge1-N1 )
102.4(4); O2-Ge1-N2 ) 103.1(4).
standard deviation of the mean O-O bond length (1.45(7) Å)
for peroxides bridging between two metal centers. The Pt-Ge
distance of 2.4286(13) Å falls exactly in the range previously
observed for a number of Pt-germyl complexes.11a,19
Scheme 1. Complete Reaction Cycle Showing Formation
of Bridging Peroxide, Oxidation of SO2, and Reduction of
Sulfate To Yield Starting Complex with Photochemical
Isomerization of the Peroxide to a Germanate Moiety Also
Shown
When the reaction of (Et3P)2PtGe[N(SiMe3)2]2 with O2 was
performed in the presence of ambient light or when crystals of
1 were stored in ambient light, a slow conversion to a complex
having symmetrically equivalent triethylphosphine ligands oc-
curred. This conversion was not observed to occur when 1 was
protected from light, even upon heating benzene solutions to
80 °C. However, upon exposure of the solutions to UV-light,
rapid and quantitative conversion to (Et3P)2PtO2Ge[N(SiMe3)2]2
(2a) resulted (Scheme 1). Structural assignment of 2a was based
upon spectroscopic data and elemental analysis which indicated
the same molecular composition as 1. A germyl moiety and
cis-ligated equivalent triethylphosphines bound to a square-
1
planar platinum(II) center were indicated by H, 13C, and 31P
NMR analysis. By analogy to the photochemistry observed for
peroxo complexes of disilenes and digermenes,8,9 it was
hypothesized that a similar rearrangement could have taken place
to form a Pt(II) germanate complex. This assignment was
confirmed by X-ray crystallographic analysis (Figure 2, Table
1). The four-membered ring in 2a displays considerably less
distortion than that found in 1 as evidenced by the nearly planar
0.53(3)° O-Pt-O-Ge dihedral angle. The coordination sphere
at germanium remains four-coordinate but changes from the
esoteric trigonal-pyramidal geometry to a more common
tetrahedral coordination polyhedra formed by two oxygen atoms
and two nitrogen atoms. Coordination about the Pt center is
roughly square-planar. The largest deviation is found in the
O(1)-Pt-O(2) angle of 78.9(3)°. Like similarly bonded silicon
complexes, the metals on opposite corners of the square are
held in rather close proximity to each other. The Pt-Ge distance
of 2.770(2) Å is 0.34 Å longer than that observed for the
platinum-germyl bond in complex 1. By way of comparison,
the Ge-Ge bond length the cyclodigermoxane synthesized by
Masamune et al. was found to be 2.617(1) Å, about 0.16 Å
Å; Pt1-Ge1 ) 2.4286(13) Å; O1-O2 ) 1.503(11) Å; Ge1-
O1 ) 1.847(8) Å) probably drives this distortion as steric
crowding of the triethylphosphine and (trimethylsilyl)amide
groups does not appear to be a limiting factor on the basis of a
comparison with the structures of the three-coordinate metal
germylene complexes and other metallocycles.11a,19 As has been
typical for the four-membered metallacyles synthesized in this
system to date, the coordination geometry about germanium can
best be described as a distorted trigonal pyramidal with the base
formed by Pt and two N atoms and the cap formed by oxygen.
The O-O bond distance is somewhat shorter than that observed
by Cowie et al. in the iridium system (1.58(2) Å) and that
observed by Bhaduri et al. in [Ph3P)2Pt(µ-O2)(µ-OH)]2ClO4
(1.547(21) Å).20 The distance observed in 1 is within one
(19) Litz, K. E.; Kampf, J. W.; Banaszak Holl, M. M. J. Am. Chem. Soc.
1998, 120, 7484.
(20) Bhaduri, S.; Casella, L.; Ugo, R.; Raithby, P. R.; Zuccaro, C.;
Hursthouse, M. B. J. Chem. Soc., Dalton Trans. 1979, 1624.

6466 Inorganic Chemistry, Vol. 37, No. 25, 1998
Litz et al.
longer than a typical Ge-Ge bond.8 Both of these germanium
containing rings contrast sharply with the observation of West
et al. that four-membered ring cyclosiloxanes have closer than
expected Si-Si distances.21
The reaction of (Ph3P)2PtO2 and Ge[N(SiMe3)2]2 was exam-
ined to determine if a peroxo complex or germanate complex
would form. The reaction was protected from light to prevent
degradation of a peroxo complex. Germanate complex 2b was
1
assigned as the sole product of this reaction (eq 3). H, 13C,
(3)
Figure 3. ORTEP of 3 (50% probability). Bond lengths (Å) and angles
(deg): Pt1-Ge1 ) 2.4663(3); Pt1-P1 ) 2.2271(8); Pt1-P2 )
2.3374(8); Pt1-O2 ) 2.096(2); Ge1-O1 ) 1.911(2); Ge1-N1 )
1.875(3); Ge1-N2 ) 1.869(3); S1-O1 ) 1.538(2); S1-O2 )
1.517(2); S1-O3 ) 1.443(3); S1-O4 ) 1.442(3); P1-Pt1-P2 )
98.76(3); P1-Pt1-O2 ) 175.84(7); P2-Pt1-O2 ) 80.90(6); Ge1-
Pt1-P1 ) 103.46(2); Ge1-Pt1-P2 ) 157.65(2); N1-Ge1-N2 )
112.01(13); N2-Ge1-O1 ) 98.80(11); N1-Ge1-O1 ) 95.77(12);
N2-Ge1-Pt1 ) 119.65(9); N1-Ge1-Pt1 ) 125.36(9); O1-Ge1-
Pt1 ) 93.02(7); O2-Pt1-Ge1 ) 77.12(6); O2-S1-O1 ) 105.02(12);
Pt1-O2-S1 ) 122.56(13); S1-O1-Ge1 ) 118.4(13); O3-S1-O4
) 114.8(2); O2-S1-O4 ) 108.58(14); O2-S1-O3 ) 110.5(2); O1-
S1-O4 ) 109.24(14); O1-S1-O3 ) 108.32(14).
and 31P NMR spectra revealed the complex to possess cis-
equivalent triphenylphosphine ligation to a Pt(II) center. Direct
formation of 2b suggested that isomerization from 1 to 2a may
proceed via ejection of germylene and subsequent recombination
of (Et3P)2PtO2 and Ge[N(SiMe3)2]2 to form 2a. Another
alternative mechanism involving the photodissociation of the
metallacycle into O2 and (Et3P)2PtGe[N(SiMe3)2]2, followed by
photolytic dissociation of germylene, formation of (Et3P)2PtO2,
and subsequent reaction to form 2a, was also considered on
the basis of the analogy to results previously observed for
(Et3P)2Pt(µ-η2-H2CO)Ge[N(SiMe3)2]2.11b To test for the presence
of free germylene during the isomerization of 1 to 2a, chemical
trapping experiments were performed. Photolysis of 1 with 10
equiv of benzil, an efficient germylene trap,11b,22 gave no
evidence of free germylene during the isomerization of 1 to
2a. The rate of isomerization was measured for a 0.058 mmol
solution of 1 in C6D6 exposed to a 100 W SylvaniaPar 38
mercury lamp. A rate constant of 0.00085 s-1 was obtained,
and the reaction was found to be zero order with respect to
metal concentration. Taken together, the kinetic data and the
fact that free germylene is not generated during the isomerization
suggest an intramolecular conversion from 1 to 2a.
The chemical behavior of the unique peroxo bridging found
in 1 was tested with a variety of reagents. Unlike many type II
peroxo complexes of Ni, Pd, and Pt, 1 did not react with CO2,
ketones, alkenes, alkynes, or alkyl- and arylphosphines.4,23 The
lack of reactivity cannot be explained solely as a steric effect
because reagents such as CO2 were observed to react with the
more sterically bulky nitrosobenzene adduct.19 Instead, these
observations imply an electronically based selectivity that differs
between type II Pt peroxo complexes and the type X peroxo
complex reported here. Two reagents were observed to react
with 1, SO2 and H2CdO, and the results from these reactions
are presented below.
basis of spectroscopic analysis, the structure was assigned as
the bridging sulfate complex, (Et3P)2Pt(µ-η2-SO4)Ge[N(SiMe3)2]2
(3) (Scheme 1). An X-ray crystallographic analysis was
performed to confirm this assignment (Figure 3, Table 1).
The five-membered ring formed by Pt-O-S-O-Ge atoms
represents the most prominent structural feature of 3. The
dihedral angle formed by the O-Pt-Ge-O portion of this ring
is 35.08(9)°. The Pt-O distance of 2.096(2) Å is longer than
observed for complexes 1 or 2a (2.050(8); 2.070(6), 2.032(8)
Å). The Ge-O bond length of 1.911(2) Å is also longer than
those observed for 1 or 2a (1.847(8); 1.738(8), 1.784(7) Å) and
longer than that seen in the previously characterized complexes
(Et3P)2Pt(µ-η2-CO2)Ge[N(SiMe3)2]2 (1.883(5) Å) and (Et3P)2-
Pt(µ-η2-PhNO)Ge[N(SiMe3)2]2 (1.820(4) Å).11a,19 The sulfate
shows significant deviation in the S-O bond lengths (1.538(2),
1.517(2), 1.443(3), 1.442(3) Å) as compared to “free” sulfate
ions which typically show distances in the 1.46-1.48 Å range.
Similar changes in bond length have been observed previously
for sulfate bridging across two platinum centers.24 The coordi-
nation sphere about germanium resembles a distorted trigonal
pyramid, and the coordination about platinum is roughly square-
planar.
Reaction of excess SO2 with a benzene solution of 1 resulted
in the rapid formation of an insoluble solid. Characterization
of the white precipitate by 1H, 13C, and 31P NMR spectroscopy
revealed quantitative conversion of 1 to a new complex
possessing cis-inequivalent triethylphosphines and a germyl
moiety coordinated to platinum in a square planar fashion. IR
spectroscopy revealed the presence of strong SdO stretches
indicative of a sulfate group at 1279 and 1159 cm-1. On the
Given the general difficulty in designing catalytic conversions
of SO2 to [SO4]2-,5 we were interested in the feasibility of
removing the sulfate fragment from 3 to regenerate (Et3P)2PtGe-
[N(SiMe3)2]2. Unlike Cowie’s peroxo complex, 3 is well suited
for simple reduction reactions as there are no halide ligands to
cause potential side reactions. Reduction of 3 in THF solution
with Na/Hg amalgam quantitatively removed the sulfate moiety
and quantitatively regenerated (Et3P)2PtGe[N(SiMe3)2]2 as
demonstrated by 1H NMR spectroscopy (Scheme 1). To better
understand the interaction of strongly coordinating anions such
as sulfate with the metal-germylene complexes, we synthesized
a carbonate complex. Addition of 1 equiv of Ge[N(SiMe3)2]2
(21) Fink, M. J.; Haller, K. J.; West, R.; Michl, J. J. Am. Chem. Soc. 1984,
106, 822.
(22) (a) Tokitoh, N.; Manmura, K.; Okazaki, R. Organometallics 1994,
13, 167. (b) Weidenbruch, M.; Stu¨rmann, M.; Kilian, H.; Pohl, S.;
Saak, W. Chem. Ber. Recueil 1997, 130, 735.
(23) (a) Nyman, C. J.; Wilkinson, G. J. Chem. Soc., Chem. Commun. 1967,
407. (b) Hayward, P. J.; Blake, D. M.; Nyman, C. J.; Wilkinson, G.
J. Chem. Soc., Chem. Commun. 1969, 987. (c) Otsuka, S.; Nakamura,
A.; Tatsuno, Y.; Miki, M. J. Am. Chem. Soc. 1972, 94, 3761.
(24) (a) Cotton, F. A.; Falvello, L. R.; Han, S. Inorg. Chem. 1982, 21,
2889. (b) Zhilyaev, A. N.; Shikhaleeva, E. V.; Katsen, S. B.;
Baranovsky, I. B. Zh. Neorg. Khim. 1987, 32, 118.

Reactions of Pt Complexes with O2 and SO2
Inorganic Chemistry, Vol. 37, No. 25, 1998 6467
Figure 4. ORTEP of 4 (50% probability). Selected bond lengths (Å)
and angles (deg): Pt1-Ge1 ) 2.4174(10); Pt1-P1 ) 2.355(2); Pt1-
P2 ) 2.229(2); Pt1-O1 ) 2.091(6); Ge1-O2 ) 1.864(6); C25-O1
) 1.319(10); C25-O2 ) 1.337(10); C25-O3 ) 1.249(10); Ge1-N1
) 1.868(8); Ge1-N2 ) 1.889(8); P1-Pt1-P2 ) 104.52(9); N1-Ge1-
N2 ) 113.7(3); P1-Pt1-O1 ) 79.8(2); P2-Pt1-O1 ) 174.6(2); P2-
Pt1-Ge1 ) 96.07(7); P1-Pt1-Ge1 ) 158.96; Pt1-O1-C25 )
121.4(5); O1-C25-O2 ) 120.5(7); O1-C25-O3 ) 120.3(8); O3-
C25-O2 ) 119.2(8); C25-O2-Ge1 ) 116.5; O2-Ge1-Pt1 )
95.0(2); O2-Ge1-N1 ) 98.6(3); O2-Ge1-N2 ) 95.6(3).
Figure 5. Infrared spectra of 1 (A), 18O2-1 (B), 3 (C), and 18O2-3 (D)
(KBr pellets) from 1350 to 550 cm-1
.
graphically characterized Ge-O bonds lengths in molecular
species have been observed between 1.73 and 1.83 Å. The Pt-O
bonds of 3 and 4 are both longer than the mean length observed
for square-planar platinum complexes (2.05(4) Å). Attempts at
reducing the carbonate complex to generate (Et3P)2PtGe-
[N(SiMe3)2]2 failed. Reduction with Na/Hg amalgam did not
result in any reaction, and reduction with cesium attacked the
bis(trimethylsilyl)amide groups on the germylene. The reaction
of 1 with CO did not provide an alternate route to 4.
to a solution of (Et3P)2PtCO3 in THF resulted in insertion of
the germylene into the Pt-O bond and formation of (Et3P)2-
Pt(µ-η2-OC(O)O)Ge[N(SiMe3)2]2 (4) (eq 4). Characterization
The formation of sulfate complex 3 was studied by infrared
spectroscopy by comparison of isotopically labeled and unla-
beled compounds. 18O2-1 was prepared by addition of 18O2 (95%
18O) to (Et3P)2PtGe[N(SiMe3)2]2. Addition of unlabeled SO2 to
18O2-1 resulted in the partially labeled sulfate complex, 18O2-3.
The symmetric SdO stretch in 3 appears at 1159 cm-1 (Figure
5C, region 1). The asymmetric SdO stretch is obscured by the
strong Si(CH3)3 deformations at 1251 cm-1. The partially
labeled sulfate complex possesses two symmetric SdO stretches
of roughly equal intensity at 1152 and 1131 cm-1 (calculated
shift ) 1114 cm-1) indicating label incorporation at a single
external SdO site (Figure 5D, region 1). In addition, the Ge-O
stretch at 1096 cm-1 in 3 shifts to 1040 cm-1 in 18O2-3 indicative
of no exchange at the Ge-O bond (Figure 5, region 2). The
infrared spectra of 1 and 18O2-1 between 1300 and 550 cm-1
were identical with the exception of an as yet unassigned
absorbance at 934 cm-1 in 18O2-1 and the 18O-18O stretch
tentatively assigned at 796 cm-1 (Figure 5A,B). In the infrared
region between 600 and 200 cm-1, Pt-O stretches are reported
(4)
by 1H, 13C, and 31P NMR spectroscopy indicated a square-planar
complex with cis-inequivalent triethylphosphine ligands. A
single-crystal X-ray diffraction study confirmed the overall
geometry and provided important metrical parameters for
comparison to sulfate complex 3 (Figure 4). The five-membered
ring contains a planar CO3 fragment and possesses a 19.98°
OPtGeO dihedral angle. The Pt-O and Ge-O bond lengths
are 2.091(6) and 1.864(6) Å, respectively, and the carbonate
C-O distances are 1.319(10), 1.337(10), and 1.249(10) Å. Note
that the C25-O1 and C25-O2 bond lengths are significantly
longer and C25-O3 is significantly shorter than those previous
observed for similar bridging carbonates.25 This suggests that
the carbonate in complex 4 lies the farthest along the continuum
toward formal single bonds and a double bond for the carbon-
oxygen interaction. In concert with the Ge-O bond length, these
structural parameters suggest that the carbonate is more strongly
bound than the sulfate. By way of contrast, the 1.911(2) Å
Ge-O bond length of the sulfate is one of the longest observed,
exceeded only by a trichloromethyl acetate bridged dimer of
tetraphenyldigermane (2.073(3) and 2.314(3) Å)26 and a tri-
fluoromethanesulfonate-stabilized diphenylgermylene ligand in
a rhenium complex (2.034(4) Å).27 To date, 90% of crystallo-
to occur near 400 cm-1 28
Unlabeled 1 possessed a broad
.
absorption in this region centered at 417 cm-1 (Figure 6A).
Comparison to 18O2-1 reveals a band shifted as predicted by
harmonic oscillator approximation at 402 cm-1 (Figure 6B).
Therefore we assigned the peak at 417 cm-1 as a Pt-O stretch.
The other absorptions observed likely result from ring modes,
also known to occur in this region. The IR spectrum of 3, shown
in Figure 6C, was noted to have a broad absorption with a
shoulder at 419 and 411 cm-1, assigned as Pt-O modes. The
partially labeled sulfate complex, 18O2-3, also contained a Pt-O
stretch at 416 cm-1. Since the peak at 416 cm-1 was unshifted,
this suggests that the Pt-18O bond is cleaved and an unlabeled
O-atom from SO2 is incorporated.
(25) (a) Reinking, M. K.; Ni, J.; Fanwick, P. E.; Kubiak, C. P. J. Am. Chem.
Soc. 1989, 111, 6459. (b) Cotton, F. A.; Labella, L.; Shang, M. Y.
Inorg. Chim. Acta 1992, 197, 149.
(26) Simon, D.; Haberle, K.; Drager, M. J. Organomet. Chem. 1984, 267,
133.
Insertion of formaldehyde into the Pt-O bond of 1 was also
pursued. A THF solution of 1 was generated in situ followed
(27) Lee, K. F.; Arif, A. M.; Gladysz, J. A. Organometallics 1991, 10,
751.
(28) Socrates, G. Infrared Characteristic Group Frequencies; John Wiley
and Sons: New York, 1992.

6468 Inorganic Chemistry, Vol. 37, No. 25, 1998
Litz et al.
Figure 6. Infrared spectra of 1 (A), 18O2-1 (B), 3 (C), and 18O2-3 (D) (CsI pellets) from 600 to 200 cm-1
.
by addition of paraformaldehyde. A new doublet with Pt
satellites was observed in the 1H NMR at 5.65 ppm, consistent
with the insertion of a formaldehyde unit into the Pt-O bond.
The magnitude of the coupling constants (4JP-H ) 10.0 Hz,
3JPt-H ) 40.0 Hz) indicated that the formaldehyde oxygen was
bound to platinum. The 13C NMR spectrum showed no
discernible Pt coupling for the formaldehyde-derived methylene
at 99.3 ppm. The remainder of the spectral data, including 31P
NMR spectra and IR spectra, support assignment of the complex
as (Et3P)2Pt(µ-η2-OCH2OO)Ge[N(SiMe3)2]2 (5) (eq 5). This
peroxide complexes. This was firmly established by addition
of (Ph3P)2PtO2 to Ge[N(SiMe3)2]2, producing the germanate
analogue 2b. Direct synthesis of 2b raises some interesting
issues about complexes 1 and 2a. The reaction of O2 with
(Et3P)2PtGe[N(SiMe3)2]2 must occur with the intact Pt-Ge
bond. If predissociation of germylene occurred followed by
formation of the type II platinum peroxo, complex 2a would
be generated instead of complex 1. Predissociation of germylene
can also be excluded because none was trapped by excess benzil
present during the isomerization from 1 to 2a. Eliminating the
presence of free germylene in the process is important because
we have recently observed the UV-photolysis of (Et3P)2Pt(µ-
η2-H2CO)Ge[N(SiMe3)2]2.11b Upon irradiation, the bridging
formaldehyde moiety undergoes a photocycloreversion reaction
at the four-membered ring generating (Et3P)2PtGe[N(SiMe3)2]2
and formaldehyde. Continued photolysis cleaves the Pt-Ge
bond of (Et3P)2PtGe[N(SiMe3)2]2 to give (Et3P)2Pt and free
germylene. The (Et3P)2Pt fragment causes the decarbonylation
reaction of the formaldehyde. The free germylene formed by
this photochemical reaction is readily trapped by benzil.
Therefore, mechanisms involving cycloreversion of 1, followed
by loss of germylene, and formation of a type II peroxo are
inconsistent. The formation of 2a from 1 likely involves
cleavage of the O-O bond by light, rotation, and subsequent
formation of the additional Pt-O and Ge-O bonds.
(5)
orientation of insertion is consistent with the previously observed
result for the reaction of formaldehyde with (Et3P)2Pt(µ-η2-
N(Ph)O)Ge[N(SiMe3)2]2 which gave (Et3P)2Pt(µ-η2-OCH2N-
(Ph)O)Ge[N(SiMe3)2]2. In this case, the orientation of formal-
dehyde insertion was confirmed using X-ray crystallography.19
Discussion
Reaction of O2 with (Et3P)2PtGe[N(SiMe3)2]2 forms peroxide
product 1 if the reagents are protected from light. The reactivity
observed for 1 indicates this structural hybrid between the main
group peroxodigermanes and the peroxoiridium dimer prepared
by Cowie et al. (eq 2) can undergo transformations typical to
both classes of compounds. 1 photoisomerizes to 2a in a fashion
analogous to the main group bridging peroxides yet also
undergoes Pt-O bond insertion chemistry typically observed
for transition metal peroxides. This behavior contrasts with that
of Cowie’s peroxo complex which was observed to decompose
upon photoirradiation but underwent insertion reactions with a
wider variety of substrates.7
Given the similarity of 1 to (Ph3P)2PtO2, comparison of the
closely related binding modes and reactivity is informative. In
particular, does the change in dioxygen coordination alter the
mechanism by which the sulfate reaction proceeds? Mechanistic
studies of the reaction of SO2 with (Ph3P)2PtO2 argue for a
peroxysulfite or peroxysulfuryl intermediate which rearranges
to yield the bidentate sulfate product (Scheme 2).29 Insertion
into the Pt-O bond, instead of direct attack at the O-O bond,
is believed to occur because the isotopic label ends up in both
the terminal and metal-bound positions of the sulfate. Mehandru
(29) (a) Horn, R. W.; Weissberger, E.; Collman, J. P. Inorg. Chem. 1970,
9, 2367. (b) Valentine, J.; Valentine, D.; Collman, J. P. Inorg. Chem.
1971, 10, 219.
Direct synthesis of bidentate, dianionic germanates, such as
found in 2a, is possible via reaction of germylenes with type II

Reactions of Pt Complexes with O2 and SO2
Inorganic Chemistry, Vol. 37, No. 25, 1998 6469
Scheme 2. Mechanism of Sulfate Formation in Type II and
Type X Peroxides with Spectator Ligands Not Shown
Figure 7. Proposed intermediate for SO2 insertion into 1.
direct insertion into the O-O bond of the peroxide as the
mechanism for sulfate formation. Unfortunately, neither ex-
perimental study has been able to distinguish between a
peroxysulfite or peroxysulfuryl intermediate. Although we have
been unable to detect any intermediates in the conversion of 1
to 3, we have isolated and structurally characterized an
isoelectronic insertion product for the reaction of SO2 with
(Et3P)2Pt(µ-η2-PhNO)Ge[N(SiMe3)2]2.19 This analogy suggests
that the five-membered ring structure depicted in Figure 7 may
be the intermediate for SO2 insertion into bimetallic peroxo 1.
We were able to demonstrate removal of the sulfate moiety
of 3 by reduction with Na/Hg amalgam. The initial metal starting
material, (Et3P)2PtGe[N(SiMe3)2]2, is regenerated in quantitative
yield. As Kubas has noted, the removal of sulfate and regenera-
tion of starting materials has been problematic in designing
catalytic conversions of SO2 to [SO4]2-.5 The sulfate analogue
prepared by addition of SO2 to Cowie’s peroxo-Ir dimer is
unsuitable for such a reduction strategy due to the presence of
iodide ligands. Although 1 is not an attractive candidate for the
catalytic abatement of SO2, it serves to illustrate the efficacy
of strategies that employ reversible Ge-O bond formation and
Pt-Ge centers as catalytic sites.
and Anderson studied the mechanism of this reaction theoreti-
cally and concluded the five-membered peroxysulfite intermedi-
ate was more likely than the four-membered peroxysulfuryl
intermediate.30 They based this conclusion on the high-energy
transition state they predicted for conversion of the peroxysul-
furyl intermediate to the sulfate. Although the peroxysulfite
intermediate was calculated to be less stable, the lower-energy
transition state for conversion to sulfate was deemed the more
important factor.
The labeling study presented in this paper for the reaction of
1 and SO2 to form sulfate complex 3 demonstrates this type of
bridging peroxide follows the same general mechanism previ-
ously proposed by Collman et al. wherein SO2 inserts into the
Pt-O bond and rearranges to the sulfate complex.29 The oxygen
in the Pt-O-S linkage in sulfate complex 3 comes from SO2,
as no shift in the Pt-O region was observed when comparing
3 to 18O2-3. The IR spectrum of 8O2-3 contains a feature at 1159
cm-1, consistent with a Sd16O stretch, and a feature at 1131
cm-1, consistent with a Sd18O stretch. The new sulfate SdO
stretch at 1131 cm-1 at roughly equal intensity to that at 1159
cm-1 results from essentially complete transfer of the Pt bound
18O in the initial peroxide complex to an external SdO in 3.
These results indicate type X peroxides undergo the sulfate
reaction in a fashion similar to type II peroxides. Both this work
and the previous work by Collman et al. definitively excludes
Acknowledgment. Alfa-Aesar is thanked for a loan of K2-
PtCl4. M.M.B.H. thanks the University of Michigan for support
of this work and D. Coucouvanis for helpful discussions.
Supporting Information Available: Tables of data collection and
refinement parameters, atomic coordinates and thermal parameters
including hydrogens, complete bond distances and angles, and aniso-
tropic displacement parameters (35 pages). X-ray crystallographic data
files for 1, 2a, 3, and 4, in CIF format, are available on the Internet
only. Ordering and access information is given on any current masthead
page.
(30) Mehandru, S. P.; Anderson, A. B. Inorg. Chem. 1985, 24, 2570.
IC9808044