Reactivity of platinum - Oxygen bonds: Kinetic and mechanistic studies of the carbonylation of platinum aryloxide complexes and the formation of (aryloxy)carbonyls

Article abstract of DOI:10.1021/ja952680p

The reactions of [Pt(triphos)(Cl)][Cl] (1) {triphos = bis[2-(diphenylphosphino)ethyl]phenylphosphine} with NaOC₆H₄-p-R, in the presence of NaPF₆, yields the aryloxy complexes [Pt(triphos)(OC₆H₄-p-R)] [PF₆] (R = OMe (2a), Me (2b), H (2c), F (2d), Cl (2e)). Upon reaction of 2a-e with carbon monoxide at pressures from 10 to 134 psi in acetonitrile the (aryloxy)carbonyl complexes [Pt(triphos)(C(O)OC₆H₄-p-R)][PF₆] (3a-e) were obtained. The molecular structure of [Pt(triphos)(C(O)OC₆H₄-p-Me)][PF₆] (3b) was determined by X-ray diffraction. Complex 3b crystallized in the monoclinic space group P2₁|n (no. 14) with a = 10.797(1) ?, b = 19.927(3) ?, c = 19.113(2) ?, β= 98.07(1)°, V = 4071 (2) Aβ³, and Z = 4. The structure was solved and refined to R = 0.035 and R_w = 0.040 for 3958 reflections with I > 3σ(I). The kinetics of the carbonylation of 2a-e to form 3a-e were studied by ³¹P{¹H} NMR. Rates of carbonylation exhibit a first order dependence on [CO], but are independent of the concentration of free aryloxide in solution. Rates of aryloxide ligand exchange were also found to be significantly faster than rates of carbonylation. The rates of carbonylation depend on the para-substituent of the aryloxy ligand and follow the order F (2d) > Me (2b) > OMe (2a). These observations are interpreted in terms of a carbonylation mechanism that proceeds via a migratory insertion pathway, rather than by nucleophilic attack at coordinated carbon monoxide by free or dissociated aryloxide.

Full text of DOI:10.1021/ja952680p

4846
J. Am. Chem. Soc. 1996, 118, 4846-4852
Reactivity of Platinum-Oxygen Bonds: Kinetic and
Mechanistic Studies of the Carbonylation of Platinum Aryloxide
Complexes and the Formation of (Aryloxy)carbonyls
David W. Dockter,‡ Phillip E. Fanwick,† and Clifford P. Kubiak*
Contribution from 1393 Brown Laboratory of Chemistry, Department of Chemistry,
Purdue UniVersity, West Lafayette, Indiana 47907
ReceiVed August 7, 1995. ReVised Manuscript ReceiVed NoVember 21, 1995X
Abstract: The reactions of [Pt(triphos)(Cl)][Cl] (1) {triphos ) bis[2-(diphenylphosphino)ethyl]phenylphosphine}
with NaOC6H4-p-R, in the presence of NaPF6, yields the aryloxy complexes [Pt(triphos)(OC6H4-p-R)][PF6] (R )
OMe (2a), Me (2b), H (2c), F (2d), Cl (2e)). Upon reaction of 2a-e with carbon monoxide at pressures from 10
to 134 psi in acetonitrile the (aryloxy)carbonyl complexes [Pt(triphos)(C(O)OC6H4-p-R)][PF6] (3a-e) were obtained.
The molecular structure of [Pt(triphos)(C(O)OC6H4-p-Me)][PF6] (3b) was determined by X-ray diffraction. Complex
3b crystallized in the monoclinic space group P21/n (no. 14) with a ) 10.797(1) Å , b ) 19.927(3) Å, c ) 19.113(2)
Å, â ) 98.07(1)°, V ) 4071(2) Å3, and Z ) 4. The structure was solved and refined to R ) 0.035 and Rw ) 0.040
for 3958 reflections with I > 3σ(I). The kinetics of the carbonylation of 2a-e to form 3a-e were studied by
31P{1H} NMR. Rates of carbonylation exhibit a first order dependence on [CO], but are independent of the
concentration of free aryloxide in solution. Rates of aryloxide ligand exchange were also found to be significantly
faster than rates of carbonylation. The rates of carbonylation depend on the para-substituent of the aryloxy ligand
and follow the order F (2d) > Me (2b) > OMe (2a). These observations are interpreted in terms of a carbonylation
mechanism that proceeds via a migratory insertion pathway, rather than by nucleophilic attack at coordinated carbon
monoxide by free or dissociated aryloxide.
Introduction
complexes has been slower to develop.9,11,16,21,23,26-36 Bercaw
and Bryndza et al. have shown that homolytic Pt-OR bond
We report synthetic and mechanistic studies of platinum
aryloxy and (aryloxy)carbonyl complexes. The chemistry of
late transition metal aryloxides was once considered to be limited
by the incompatibility of the “hard” basic ligands and “soft”
metal centers.1-3 In the wake of the earlier development of
metal alkyl chemistry, there now exists a growing body of
research on such complexes.4,5 While a significant number of
examples of late transition metal alkoxy complexes now exist
in the literature,6-25 the chemistry of late transition metal aryloxy
(15) Milstein, D.; Calabrese, J. C.; Williams, I. D. J. Am. Chem. Soc.
1986, 108, 6387-6389.
(16) Alsters, P. L.; Baesjou, P. J.; Janssen, M. D.; Kooijman, H.; Sicherer-
Roetman, A.; Spek, A. L.; Koten, G. v. Organometallics 1992, 11, 4124-
4135.
(17) Glueck, D. S.; Winslow, L. J. N.; Bergman, R. G. Organometallics
1991, 10, 1462-79.
(18) Andrews, M. A.; Gould, G. L. Organometallics 1991, 10, 387-9.
(19) Lunder, D. M.; Lobkovsky, E. B.; Streib, W. E.; Caulton, K. G. J.
Am. Chem. Soc. 1991, 113, 1937-1938.
(20) Osakada, K.; Kim, Y.-J.; Tanaka, M.; Ishiguro, S.-i.; Ohshiro, K.;
Yamamoto, A. Inorg. Chem. 1991, 30, 197-200.
(21) Mandal, S. K.; Ho, D. M.; Orchin, M. Organometallics 1993, 12,
1714-1719.
† To whom correspondence pertaining to crystallographic studies should
be addressed.
‡ Present address: Corporate Technology, Phillips Petroleum Co.,
Bartlesville, OK 74004.
(22) Smith, G. D.; Hanson, B. E.; Merola, J. S.; Waller, F. J. Organo-
metallics 1993, 12, 568-570.
X Abstract published in AdVance ACS Abstracts, May 1, 1996.
(1) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533.
(2) Pearson, R. G. J. Chem. Educ. 1968, 45, 643.
(3) Hartley, F. R. The Chemistry of Platinum and Palladium; Halsted
Press: New York, 1973; p 169.
(23) Simpson, R. D.; Bergman, R. G. Organometallics 1993, 12, 781-
96.
(24) Aldridge, T. K.; Stacy, E. M.; MacMillin, D. R. Inorg. Chem. 1994,
33, 722-727.
(25) To´th, I.; Elsevier, C. J. J. Chem. Soc., Chem. Commun. 1993,
529531.
(26) Woerpel, K. A.; Bergman, R. G. J. Am. Chem. Soc. 1993, 115, 7884.
(27) Osakada, K.; Ohshiro, K.; Yamamoto, A. Organometallics 1991,
10, 404-10.
(28) Churchill, M. R.; Fettinger, J. C.; Rees, W. M.; Atwood, J. D. J.
Organomet. Chem. 1986, 304, 361.
(4) Bryndza, H. E.; Domaille, P. J.; Tam, W.; Fong, L. K.; Paciello, R.
A.; Bercaw, J. E. Polyhedron 1988, 7, 1441-52.
(5) Bryndza, H. E.; Tam, W. Chem. ReV. 1988, 88, 1163-88.
(6) Sen, A.; Chen, J.-T.; Vetter, W. M.; Whittle, R. R. J. Am. Chem.
Soc. 1987, 109, 148-156.
(7) Kubota, M.; Sly, W. G.; Santarsiero, B. D.; Clifton, M. S.; Kuo, L.
Organometallics 1987, 6, 1257-1259.
(8) Bellon, P. L.; Manassero, M.; Porta, F.; Sansoni, M. J. Organomet.
Chem. 1974, 80, 139-145.
(29) Hayashi, Y.; Yamamoto, T.; Yamamoto, A.; Komiya, S.; Kushi,
Y. J. Am. Chem. Soc. 1986, 108, 385-391.
(9) Osakada, K.; Kim, Y.-J.; Yamamoto, A. J. Organomet. Chem. 1990,
382, 303.
(30) Cowan, R. L.; Trogler, W. C. J. Am. Chem. Soc. 1989, 111, 4750-
61.
(10) Lunder, D. M.; Lobkovsky, E. B.; Streib, W. E.; Caulton, K. G. J.
Am. Chem. Soc. 1991, 113, 1937-1938.
(31) Bugno, C. D.; Pasquali, M.; Leoni, P.; Sabation, P.; Braga, D. Inorg.
Chem. 1989, 28, 1390-1394.
(32) Ni, J.; Kubiak, C. P. In Homogeneous Transition Metal Catalyzed
Reactions; Moser, W. R., Slocum, D. W., Eds.; Advances in Chemistry
Series; American Chemical Society: Washington, D C, 1992; Vol. 230,
pp 515-528.
(33) Rees, W. M.; Churchill, M. R.; Fettinger, J. C.; Atwood, J. D.
Organometallics 1985, 4, 2179-2185.
(34) Seligson, A. L.; Cowan, R. L.; Trogler, W. C. Inorg. Chem. 1991,
30, 3371-81.
(11) Kim, Y. J.; Osakada, K.; Takenaka, A.; Yamamoto, A. J. Am. Chem.
Soc. 1990, 112, 1096-104.
(12) Bryndza, H. E.; Calabrese, J. C.; Wreford, S. S. Organometallics
1984, 3, 1603-4.
(13) Bryndza, H. E.; Calabrese, J. C.; Marsi, M.; Roe, D. C.; Tam, W.;
Bercaw, J. E. J. Am. Chem. Soc. 1986, 108, 4805-13.
(14) Bryndza, H. E.; Kretchmar, S. A.; Tulip, T. H. J. Chem. Soc., Chem.
Commun. 1985, 14, 977-8.
S0002-7863(95)02680-1 CCC: $12.00 © 1996 American Chemical Society

ReactiVity of Platinum-Oxygen Bonds
J. Am. Chem. Soc., Vol. 118, No. 20, 1996 4847
Table 1. 31P{1H} NMR Data for Complexes 2a-e,
[Pt(triphos)(OC6H4-p-R)][PF6]
energies in alkoxy complexes are comparable to the Pt-C (sp3)
bond energies of the alkyl complexes.4 Compared to the alkoxy
complexes, the chemistry of late transition metal aryloxy
complexes is expected to reflect a weaker M-O bond, inasmuch
as the homolytic and heterolytic bond strengths for aryloxide
complexes are expected to be lower.
There are two principal mechanisms by which late transition
metal aryloxy complexes can be carbonylated to their respective
(aryloxy)carbonyls. The first is classical migratory insertion,37
where the aryloxy ligand migrates to a carbon monoxide cis to
the aryloxy ligand, eq 1. Bryndza concluded that this mecha-
PB, trans-phosphinec PB, trans-phosphinec
R
δ(PB)
1J(Pt-PB) δ(PA) 1J(Pt-PA) Hammett
(compound) (ppm)a
(Hz)
(ppm)a
(Hz)
σb
Cl (2e)
F (2d)
H (2c)
Me (2b)
OMe (2a)
75.69
75.48
75.53
75.26
75.09
2870.4
2840.9
2835.4
2808.8
2795.4
40.89
40.80
40.32
40.31
40.40
2608.8
2633.8
2620.0
2641.2
2649.5
0.23
0.06
0.00
-0.17
-0.27
O
C
CO
LnM OAr
LnM OAr
LnM(CO)OAr
(1)
a Recorded at 80.98 MHz in CD3CN at 293 K; chemical shifts are
relative to external 85% H3PO4 at 0.00 ppm. b The Hammett σ-values47
are included as a reference to the electron-donating nature of the
nism is important in the carbonylation of the platinum alkoxy
complexes Pt(dppe)(OMe)(R) (R ) OMe, Me; dppe ) 1,2-bis-
(diphenylphosphino)ethane).38 This conclusion was based on
NMR evidence of a five-coordinate carbon monoxide adduct,
insertion rates that are faster than methoxide dissociation, and
crossover studies that show no incorporation of labeled meth-
oxide. A second type of mechanism that must be considered
in view of the lower M-OAr bond energies expected for late
transition metal aryloxy complexes is nucleophilic addition of
the aryloxide to coordinated carbon monoxide. This involves
substitution of a weakly bound aryloxy ligand by carbon
monoxide, followed by nucleophilic addition of the displaced
aryloxide to the carbonyl ligand, eq 2. Atwood et al. found
2
aryloxide. c The J(PA-PB) coupling constant for 2a-e is 3 Hz.
enced to external TMS via residual solvent peaks. 31P NMR spectra
were referenced to external 85% H3PO4 and were all proton decoupled.
Infrared spectra were obtained using a Perkin-Elmer 1710 Fourier
transform spectrometer with a 1700/PC data station.
All reactions and manipulations were carried out under a nitrogen
atmosphere using a nitrogen-filled glovebox (Vacuum Atmospheres)
or by standard Schlenk line techniques. The NaOC6H4R reagents were
prepared from the corresponding phenol and a 5-fold excess of NaH
in THF. Unreacted NaH was filtered off, the filtrate concentrated, and
hexane added to precipitate the white product. Pt(cod)Cl2 (cod ) 1,5-
cyclooctadiene) was prepared by a published literature procedure.39
Synthesis of [Pt(triphos)(Cl)][Cl] (1). This complex was prepared
by modification of the previously reported synthesis of Pt(dmpe)Cl2
(dmpe)1,2-bis(dimethylphosphino)ethane).40,41 A solution of Pt(cod)-
Cl2 (1.63 g, 4.33 mmol) in methylene chloride (30 mL) was added to
a solution of triphos (2.32 g, 4.33 mmol) in methylene chloride (30
mL). After 30 min, the solvent was removed in Vacuo to give a white
solid which was washed with diethyl ether and recrystallized from
methylene chloride/heptane. Compound 1 was isolated as a white solid
and dried under vacuum. Yield: 2.84 g, 81.4%. 1H NMR (CD2Cl2,
300 MHz, δ, ppm): 2.2 (m, 2H, P-CH2), 2.7 (m, 2H, P-CH2), 3.1 (m,
2H, P-CH2), 3.8 (m, 2H, P-CH2), 7.5-8.4 (m, 25H, triphos phenyl).
31P{1H} NMR (CD3CN, 80.98 MHz, δ, ppm): A2BX, (PA) 42.27,
1J(PA-Pt) ) 2483.6 Hz; (PB) 86.34, 1J(PB-Pt) ) 3027.8 Hz,
2J(PA-PB) ) 1.5 Hz.
CO
LnM-OAr y z LnM(CO)+ + -OAr f LnM(CO)OAr (2)
that nucleophilic addition of aryloxide to the coordinated carbon
monoxide ligands of [Ir(CO)2(PPh3)2]+ occurs during the
reaction of Ir(CO)(PPh3)2(OR) (R ) Me, Ph) with carbon
+
monoxide.33 Spectroscopic observation of the Ir(CO)3(PPh3)2
intermediate in quantities dependent on phenoxide basicity
supported this conclusion.
We describe the synthesis of a series of new platinum aryloxy
complexes and their reactions with carbon monoxide to give
(aryloxy)carbonyl complexes. We present evidence, based on
kinetic studies of the carbonylation reactions, concerning the
mechanism of formation of the (aryloxy)carbonyl complexes.
Preparation of Aryloxide Complexes. 31P{1H} NMR for com-
plexes 2a-e are presented in Table 1. The designation PA refers to
phosphorus nuclei cis to the aryloxide, and PB refers to the phosphorus
nucleus trans to the aryloxide.
Experimental Section
General Considerations. K2PtCl4 was purchased from Aldrich
Chemical Co. or obtained on loan from Johnson-Matthey. Triphos
was purchased from Strem Chemical Co. Carbon monoxide (research
grade) was purchased from Matheson. All phenols and sodium hydride
were purchased from Aldrich. These materials were used without
further purification. Sodium hexafluorophosphate was purchased from
Aldrich, dried under vacuum at 150 °C for 36 h, and stored in an inert
atmosphere glovebox.
Synthesis of [Pt(triphos)(OC6H4-p-OMe)][PF6] (2a). To an ac-
etonitrile (7 mL) solution of NaPF6 (0.1898 g, 1.1300 mmol) and [Pt-
(triphos)(Cl)][Cl] (0.2459 g, 0.3072 mmol) was added NaOC6H4-p-
Me (0.0523 g, 0.3233 mmol), giving a light yellow solution. After 10
min the solvent was removed in Vacuo. The remaining solid was
redissolved in methylene chloride, the solution filtered, and the filtrate
dried in Vacuo to give a yellow solid. Yield: 0.2645 g, 85.0%. 1H
NMR (CD3CN, 200 MHz, δ, ppm): 3.4 (s, 3H, OCH3), 2.1-3.3 (m,
8H, P-CH2), 6.2 (d, 2H, CH), 6.4 (d, 2H, CH), 7.2-8.2 (m, 25H, triphos
phenyl). Anal. Calcd for C41H40O2F6P4Pt: C, 49.96; H, 3.99.
Found: C, 49.02; H, 3.94.
Acetonitrile and methylene chloride were freshly distilled from CaH2
under a nitrogen atmosphere. Anhydrous n-heptane was purchased from
Aldrich. Deuterated acetonitrile (99.9%) was purchased from Cam-
bridge Isotope Labs. Deuterated acetonitrile and methylene chloride
were stored under nitrogen and used as received.
Synthesis of [Pt(triphos)(OC6H4-p-Me)][PF6] (2b). This complex
was prepared similarly to 2a from NaPF6 (0.1137 g, 0.6770 mmol),
[Pt(triphos)(Cl)][Cl] (0.1924 g, 0.2405 mmol), and NaOC6H4-p-Me
1H and 31P NMR spectra were recorded on a Varian XL-200 or
General Electric QE-300 spectrometer. 1H NMR spectra were refer-
(39) McDermott, J. X.; White, J. F.; Whitesides, G. M. J. Am. Chem.
Soc. 1976, 98, 6521-6528.
(40) Anderson, G. K.; Lumetta, G. J. Inorg. Chem. 1987, 26, 1518-
1524.
(41) Although first prepared in 1971 by King et. al. (Inorg. Chem. 1971,
10, 1841-50), this synthesis requires less time. The yields are similar to
the earlier preparation and give a product spectroscopically identical to that
reported for [Pt(triphos)(Cl)][Cl].
(35) Hartwig, J. F.; Andersen, R. A.; Bergman, R. G. Organometallics
1991, 10, 1875-87.
(36) Ladipo, F. T.; Kooti, M.; Merola, J. S. Inorg. Chem. 1993, 32, 1681-
1688.
(37) Flood, T. C.; Jensen, J. E.; Statler, J. A. J. Am. Chem. Soc 1981,
103, 4410-14.
(38) Bryndza, H. E. Organometallics 1985, 4, 1686-7.

4848 J. Am. Chem. Soc., Vol. 118, No. 20, 1996
Dockter et al.
Table 2. 31P{1H} NMR Data for Complexes 3a-c,
[Pt(triphos)(C(O)OC6H4-p-R)][PF6]
(gauge pressure) of carbon monoxide. Although the reaction proceeds
to completion, the product from this NMR scale reaction was not
isolated.
PB, trans-phosphinec PA, cis phosphinec
Synthesis of [Pt(triphos)(C(O)OC6H4-p-F)][PF6] (3d). This com-
plex was prepared similarly to 3a from a methylene chloride (1.5 mL)
solution containing [Pt(triphos)(OC6H6-p-F)][PF6] (2d) (0.0151 g,
0.0153 mmol). Yield: 0.0076 g, 48.0%. IR (KBr): ν(CO) ) 1673
R
δ(PB
1J(Pt-PB) δ(PA) 1J(Pt-PA) Hammett
(compound) (ppm)a
(Hz)
(ppm)a
(Hz)
σb
Cl (3e)
F (3d)
H (3c)
Me (3b)
OMe (3a)
90.86
90.88
90.79
90.71
90.55
1665.8
1658.5
1648.0
1642.4
1643.1
40.04
40.01
39.75
39.70
39.60
2607.6
2611.1
2622.3
2627.9
2626.4
0.23
0.06
0.00
-0.17
-0.27
cm-1
.
Reaction of [Pt(triphos)(OC6H4-p-Cl)][PF6] (2e) with Carbon
Monoxide. The complex [Pt(triphos)(C(O)OC6H6-p-Cl)][PF6] (3e) was
prepared in situ from a methylene chloride (1.5 mL) solution containing
[Pt(triphos)(OC6H6-p-Cl][PF6] (2e) (0.0450 g, 0.0466 mmol) pressurized
with carbon monoxide. This reaction was found by 31P{1H} NMR to
proceed only to 90% completion with 100 psi (gauge pressure) of carbon
monoxide, and the complex was not isolated.
a Recorded at 80.98 MHz in CD3CN at 293 K; chemical shifts are
relative to external 85% H3PO4 at 0.00 ppm. b The Hammett σ values47
are included as a reference to the electron-donating nature of the
2
aryloxide. c The J(PA-PB) coupling constant for 3a-e is 6 Hz.
Collection of X-ray Diffraction Data for [Pt(triphos)(C(O)OC6H4-
p-Me)][PF6] (3b). Crystals suitable for a single-crystal X-ray diffrac-
tion study were grown by placing a methylene chloride solution of
[Pt(triphos)(OC6H4-p-Me)][PF6] (2b) into a pressure valve NMR tube.
The solution was then carefully layered with dry heptane, and 100 psi
of carbon monoxide was introduced into the tube. A pale yellow crystal
of C42H40PtF6O2P4 having approximate dimensions of 0.35 × 0.28 ×
0.24 mm was mounted on a glass fiber in a random orientation. The
structure was solved using the Patterson method which revealed the
position of the Pt atom. The remaining atoms were located in
succeeding difference Fourier syntheses. Hydrogen atoms were located
and added to the structure factor calculations, but their positions were
not refined. The final cycle of refinement converged with R ) 0.035
and Rw ) 0.040. Plots of ∑w(|Fo| - |Fc|)2 versus |Fo|, reflection order
in data collection, sin θ/λ, and various classes of indices showed no
unusual trends. The highest peak in the final difference Fourier had a
height of 0.67 e/Å3. The crystallographic data are summarized in Table
3. An ORTEP drawing of 3b is presented in Figure 1. Selected bond
distances and angles are listed in Table 4.
Kinetic Studies. Rates of carbonylation of 2a-d were followed
by 31P{1H} NMR under carbon monoxide pressure at ambient temper-
atures.43 In a typical experiment, a pressure NMR tube (Wilmad) was
charged in an inert atmosphere box with an acetonitrile-d3 solution of
one of the platinum aryloxy complexes 2a-d. Experiments with excess
aryloxide ligand required a 1:1 acetonitrile-d3/THF solvent mixture.
An initial 31P{1H} NMR spectrum was taken before the tube was
evacuated and pressurized (10-134 psi gauge) with carbon monoxide.
31P{1H} NMR spectra were then taken at regular intervals. Signals
corresponding to the starting aryloxy complexes 2a-d and the
carbonylated (aryloxy)carbonyl product, 3a-d were integrated, and
these integrations were used in the kinetic data analysis. Spin-lattice
relaxation times (T1) for 2a were determined by inversion recovery for
PA (4.8 s) and PB (3.4 s). To obtain accurate integrations, a single 35°
pulse was used with a 6 s recycle rate and a 2 s acquisition time. The
pressure NMR tubes were individually calibrated so that sample
volumes could be easily determined from sample height. The total
volume of the tubes was also calibrated. The concentration of carbon
monoxide in acetonitrile was determined by Henry’s law.44
(0.0369 g, 0.2532 mmol) and was isolated as a yellow solid. Yield:
0.2100 g, 89.0%. 1H NMR (CD3CN, 300MHz, δ, ppm): 2.1 (s, 3H,
PhCH3), 2.2-3.6 (m, 8H, P-CH2), 6.5 (d, 2H, CH), 6.8 (d, 2H, CH),
7.2-8.2 (m, 25H, triphos phenyl). Anal. Calcd for C41H40OF6P4Pt:
C, 50.16; H, 4.11. Found: C, 50.04; H, 4.48.
Synthesis of [Pt(triphos)(OPh)][PF6] (2c). This complex was
prepared similarly to 2a from NaPF6 (0.1049 g, 0.6249 mmol), [Pt-
(triphos)(Cl)][Cl] (0.1667 g, 0.2082 mmol), and NaOPh (0.0248 g,
0.2193 mmol) and was isolated as a light yellow solid. Yield: 0.1730
g, 86.1%. 1H NMR (CD3CN, 300 MHz, δ, ppm): 2.1-3.4 (m, 8H,
P-CH2), 6.5 (m, 3H, CH), 7.0 (d, 2H, CH), 7.2-8.2 (m, 25H, triphos
phenyl). Anal. Calcd for C40H38OF6P4Pt: C, 49.65; H, 3.96. Found:
C, 49.69; H, 4.12.
Synthesis of [Pt(triphos)(OC6H4-p-F)][PF6] (2d). This complex
was prepared similarly to 2a from NaPF6 (0.4245 g, 2.5275 mmol),
[Pt(triphos)(Cl)][Cl] (0.4394, 0.5488 mmol), and NaOC6H4-p-F (0.1089
g, 0.8122 mmol) and was isolated as a light yellow solid. Yield: 0.5146
g, 95.1%. 1H NMR (CD3CN, 300 MHz, δ, ppm): 2.1-3.4 (m, 8H,
P-CH2), 6.3 (m, 2H, CH), 6.3 (m, 2H, CH), 7.2-8.1 (m, 25H, triphos
phenyl). Anal. Calcd for C40H37OF7P4Pt‚1/2CH2Cl2: C, 47.31; H, 3.73.
Found: C, 47.71; H, 3.71.42
Synthesis of [Pt(triphos)(OC6H4-p-Cl)][PF6] (2e). This complex
was prepared similarly to 2a from NaPF6 (0.24 g, 1.43 mmol), [Pt-
(triphos)(Cl)][Cl] (0.1601 g, 0.2000 mmol), and NaOC6H4-p-Cl (0.0602
g, 0.3999 mmol) and was isolated as a light yellow solid. Yield: 0.1740
g, 87.0%. 1H NMR (CD3CN, 300 MHz, δ, ppm): 2.1-3.5 (m, 8H,
P-CH2), 6.4 (d, 2H, CH), 6.5 (d, 2H, CH), 7.3-8.1 (m, 25H, triphos
phenyl). Anal. Calcd for C40H37OClF6P4Pt: C, 47.94; H, 3.72.
Found: C, 47.75; H, 3.73.
Preparation of (Aryloxy)carbonyl Complexes. 31P{1H} NMR for
complexes 3a-e in acetonitrile-d3 were recorded in situ using high-
pressure NMR tubes pressurized with carbon monoxide. Results are
presented in Table 2. The designation where PA refers to phosphorus
nuclei cis to the (aryloxy)carbonyl and PB refers to phosphorus nucleus
trans to the (aryloxy)carbonyl is retained.
Synthesis of [Pt(triphos)(C(O)OC6H4-p-OMe)][PF6] (3a). A pres-
sure valve NMR tube (Wilmad), containing a methylene chloride (1.5
mL) solution of [Pt(triphos)(OC6H4-p-OMe)][PF6] (2a) (0.0319 g,
0.0315 mmol), was carefully layered with heptane and pressurized with
100 psi (gauge pressure) of carbon monoxide. The tube was allowed
to stand for 3 days at room temperature until yellow crystals formed.
IR (KBr): ν(CO) ) 1655 cm-1. Anal. Calcd for C42H40O3F6P4Pt: C,
49.18; H, 3.93. Found: C, 49.17; H, 3.95.
Results and Discussion
Synthesis of Platinum Aryloxide Complexes. The synthesis
of the platinum aryloxy complexes [Pt(triphos)(OC6H4-p-R)]-
[PF6] 2a-e proceeds by a metathesis reaction involving [Pt-
(triphos)(Cl)][Cl] and NaOC6H4-p-R, eq 3. The anion exchange
is done concurrently by performing the reaction in the presence
of excess NaPF6. 31P{1H} NMR provides an excellent means
of characterizing platinum triphos complexes. The very small
Synthesis of [Pt(triphos)(C(O)OC6H4-p-Me)][PF6] (3b). This
complex was prepared similarly to 3a from a methylene chloride (1.5
mL) solution containing [Pt(triphos)(OC6H4-p-Me)][PF6] (2b)(0.1326
g, 0.1351 mmol). Yield: 0.0623 g, 45.7%. IR (KBr): ν(CO) ) 1677
cm-1. Anal. Calcd for C42H40O2F6P4Pt: C, 49.96; H, 3.99. Found:
C, 49.35; H, 4.01.
(43) Ambient room temperature in the NMR probe is ca. 22 °C. During
the reaction the NMR tube was not removed from the probe. Due to probe
design, the probe cooling air provides a thermostatic environment. Experi-
ments with the NMR chemical shift thermometer have shown that the probe
temperature would vary by less than (1 °C over the length of the
experiment. Sample heating during the actual NMR experiment, due to
decoupling, should be minimal as low power WALTZ decoupling was used.
(44) Fogg, P. G. T.; Gerrard, W. Solubility of Gases in Liquids, A Critical
EValuation of Gas/Liquid Systems in Theory and Practice; John Wiley and
Sons: New York, 1991.
Reaction of [Pt(triphos)(OPh)][PF6] (2c) with Carbon Monoxide.
The complex [Pt(triphos)(C(O)OPh)][PF6] (3c) was prepared in situ
from a methylene chloride (1.5 mL) solution containing [Pt(triphos)-
(OPh)][PF6] (2c) (0.0450 g, 0.0466 mmol) pressurized with 100 psi
1
(42) Solvent incorporation verified by H NMR.

ReactiVity of Platinum-Oxygen Bonds
J. Am. Chem. Soc., Vol. 118, No. 20, 1996 4849
Table 3. Structure Determination Summary for [Pt(triphos)C(O)OC6H4-p-Me][PF6] (3b)
formula
formula weight
crystal system
space group
a, Å
b, Å
c, Å
PtP4F6O2C42H40
1009.76
crystal dimensions, mm
temperature, K
0.35 × 0.28 × 0.24
293
monoclinic
P21/n (no. 14)
10.797(1)
19.927(3)
19.113(2)
98.07(1)
4071(2)
radiation (wavelength)
Mo KR (0.710 73 Å)
36.97
-12 to +12, -22 to 0, 0 to 21
6839
6626
3958
496
µ, cm-1
h, k, l range
no. of data collected
no. of unique data
no. of data with I > 3.0σ(I)
no. of variables
â, deg
V, Å3
Z
d
4
R
Rw
0.035
0.040
calc, g cm-3
1.647
influence in which the increased electron-withdrawing nature
of the para-substituent leads to increases in 1J(PB-Pt) coupling
constants. The increasing PB-Pt interaction occurs in response
to the presence of progressively weaker Pt-O(C6H4-p-R) bonds,
trans to PB, as the electron-withdrawing nature of the para
substituent is increased. This effect of the para-substituents
1
on the J(PB-Pt) coupling constants correlates with Hammett
σ-values.47 This correlation gives a best fit equation of 1J(PB-
Pt) ) (149.3 Hz)σ + 2834.61; R2 ) 0.997. The slope of this
correlation, 149.3 Hz, is not a conventional Hammett δ, which
refers to the susceptibility of a reaction’s rate to electronic
effects. However, correlation of NMR parameters with Ham-
mett σ provides a useful framework for comparison. Further-
more, correlations between Hammett σ and 1H NMR parameters
of substituted phenols48 and arylcarbinols49,50 are reported in
the literature.
Figure 1. ORTEP drawing of [Pt(triphos)C(O)OC6H4Me][PF6] (3b)
showing 50% probability ellipsoids and the atomic labeling. Hydrogen
atoms are not shown, and ligand phenyl carbons are shown as points
for clarity.
Carbonylation Reactions. When [Pt(triphos)(OC6H4-p-R)]-
[PF6] is reacted with as little as 10 psi of carbon monoxide, the
31P{1H} NMR shows a downfield shift of the phosphorus atom
trans to the aryloxide, PB, along with a decrease in the
1J(Pt-PB) coupling constant. These changes are characteristic
of carbon monoxide insertion.14,32,38 The replacement of the
sp3 oxygen with an sp2 carbon puts a ligand with more
σ-character, and larger trans influence, opposite PB. The
increased σ-donation from the (aryloxy)carbonyl ligand is
reflected in a smaller 1J(Pt-PB) coupling constant, for the
phosphorus trans to the (aryloxy)carbonyl. For example, when
carbon monoxide inserts into the Pt-O bond of [Pt(triphos)-
(OC6H4-p-Me)][PF6], the 1J(Pt-PB) coupling constant decreases
from 2808.8 to 1642.4 Hz, and the chemical shift of PB moves
downfield from 75.3 to 90.7 ppm. The 31P{1H} NMR data for
the carbonylated products 3a-e are summarized in Table 2.
For the phosphorus atoms cis to the (aryloxy)carbonyl, the small
Table 4. Selected Bond Distances (Å) and Torsion and Bond
Angles (deg) for [Pt(triphos)C(O)OC6H4-p-Me][PF6] (3b)
Bond Distances
Pt-P(1)
2.296(2) O(41)-C(40)
2.254(2) O(41)-C(41)
2.292(2) C(41)-C(42)
2.111(8) C(41)-C(46)
1.08(1)
1.28(1)
1.44(1)
1.38(2)
1.36(1)
Pt-P(2)
Pt-P(3)
Pt-C(40)
O(40)-C(40)
Bond and Torsion Angles
85.38(8) Pt-C(40)-O(40)
168.60(8) Pt-C(40)-O(41)
P(1)-Pt-P(2)
P(1)-Pt-P(3)
P(1)-Pt-C(40)
P(2)-Pt-P(3)
P(2)-Pt-C(40)
P(3)-Pt-C(40)
122.1(7)
110.2(6)
126.0(8)
124.(1)
96.5(2)
O(40)-C(40)-O(41)
85.03(8) O(41)-C(41)-C(42)
174.9(2) O(41)-C(41)-C(46)
92.5(2)
115.6(9)
P(1)-Pt-C(40)-O(40) 75.4(9)
C(40)-O(41)-C(41) 125.7(8) P(1)-Pt-C(40)-O(41) 118.6(6)
1
observed changes in chemical shift and J(Pt-PA) values can
be interpreted in terms of only minor overall structural changes
occurring upon reaction with carbon monoxide. When the
carbonylation is followed by 31P{1H} NMR, starting material
is seen to cleanly convert to product; no other products or
intermediates are seen, Figure 2. The IR spectra of compounds
3a,b,d show (aryloxy)carbonyl ν(CO) bands at 1677, 1655, and
1673 cm-1, respectively, consistent with published spectra of
other alkoxycarbonyls.14,33,38
Crystal and Molecular Structure of [Pt(triphos)(C(O)-
OC6H4-p-Me)][PF6] (3b). The crystal and molecular structure
of the [(p-methylaryl)oxy]carbonyl complex 3b was determined.
This is the first X-ray structure determination of an (aryloxy)-
carbonyl complex. Selected bond distances and angles are
summarized in Table 3. An ORTEP drawing of the molecular
cation [Pt(triphos)(C(O)OC6H4-p-Me)]+ is given in Figure 1.
MeCN, NaPF6
[Pt(triphos)(Cl)][Cl] + NaOC6H4-p-R
8
1
[Pt(triphos)(OC6H4-p-R)][PF6] + NaCl (3)
2a: R ) OMe 2d: R ) F
2b: R ) Me
2c: R ) H
2e: R ) Cl
2J(PA-PB) couplings are consistent with the expected square
planar structure of these complexes.45 Furthermore, the 1J(PB-
Pt) coupling constants are sensitive to the nature of the ligand
trans to PB.46 For example, as the electron-withdrawing nature
of the para-substituent on the arene ring of the aryloxy ligands
1
is increased from R ) OMe (2a) to R ) Cl (2e), J(PB-Pt)
systematically increases from 2795 to 2870 Hz; 31P{1H} NMR
data for 2a-e are summarized in Table 1. The trend evident
in the 31P{1H} NMR data is rationalized by a “push-pull” trans
(47) McDaniel, D. H.; Brown, H. C. J. Org. Chem. 1958, 23, 420.
(48) Ouellette, R. J. Can. J. Chem. 1965, 43, 707-709.
(49) Bobko, E.; Tolerico, C. S. J. Org. Chem. 1983, 48, 1368-1369.
(50) Ouellette, R. J.; Marks, D. L.; Miller, D. J. Am. Chem. Soc. 1967,
89, 913-917.
(45) Karplus, M. J. Chem. Phys. 1959, 30, 11.
(46) Hartley, F. R. The Chemistry of Platinum and Palladium; Halsted
Press: New York, 1973; p 303.

4850 J. Am. Chem. Soc., Vol. 118, No. 20, 1996
Dockter et al.
Figure 2. Overlaid 31P{1H} spectra showing the disappearance of
resonances corresponding to the aryloxide 2b and the appearance of
the (aryloxy)carbonyl 3b.
The central d8 platinum(II) atom has a distorted square planar
Figure 3. Dependence of the observed rate of [(p-methylaryl)oxy]-
carbonyl formation on carbon monoxide pressure for [Pt(triphos)(C(O)-
OC6H4-p-Me)][PF6] (2b).
geometry; the two trans angles are P(1)-Pt-P(3)
)
168.60(8)° and P(2)-Pt-C(40) ) 174.9(2)°. Several differ-
ences are apparent between 3b and the related, structurally
characterized alkoxycarbonyls: trans-Pt(C(O)OEt)2(PPh3)2,8 Pt-
(C(O)OMe)2(dppe),14 trans-Pt(C(O)OMe)(C(O)Ph)(PPh3)2,6 and
Pt(C(O)OEt)(CtCC(O)OMe)(PPh3)2.7 The most striking dif-
ference is the C(40)-O(40) bond distance of 1.08(1) Å, similar
in length to weakly back-bonded terminal carbonyls. It must
be noted that the thermal ellipsoid of the carbonyl oxygen is
distorted along the C(40)-O(40) bond axis, such that this bond
length is artificially shortened. For the complex Pt(dppe)(C(O)-
OMe)2, Bryndza reports carbonyl bond distances of 1.191 and
1.202 Å.14 Alkoxycarbonyl C-O bond distances in the range
of 1.191-1.228(7) Å are seen for all of the aforementioned
complexes. The C(40)-O(41) bond distance of 1.28(1) Å in
3b is also approximately 0.077(10) Å shorter than the average
distance of 1.357(9) Å seen for the related complexes. Interest-
ingly the longest and shortest bond distances, 1.372 and 1.350
Å, for this group are seen in Pt(C(O)OMe)2(dppe).14 The Pt-
C(40) bond distance is 2.111(9) Å, approximately 0.1 Å longer
than Pt-C distances reported for the related complexes. For
the complex Pt(dppe)(C(O)OMe)2 Bryndza reports the Pt-C
bonds to be 2.057 and 2.065 Å,14 while Sen reports the Pt-C
bond as 2.032(9)Å for Pt(PPh3)2(C(O)Ph)(C(O)OMe).6 The
origin of these differences is probably not due to the presence
of an aromatic ring, but rather it is more likely due to the cationic
nature of complex 3b. If the presence of an aromatic ring were
a primary factor, one would expect some deviation in the aryl
carbon-oxygen O(41)-C(41) bond length of 1.44(1) Å. The
O(41)-C(41) distance is comparable to the C-O bond length
of 1.401(10) Å, typical of aromatic esters.51 The C-O bonds
Figure 4. Change in the observed rate of [(p-methylaryl)oxy]carbonyl
formation vs carbon monoxide concentration for [Pt(triphos)(C(O)-
OC6H4-p-Me)][PF6] (2b).
first order in [CO], Figures 3 and 4. The reaction shows no
observable dependence on the presence of excess aryloxide in
solution. The presence of 10, 20, and 40 equiv of aryloxide
added to the solution does not affect the rate of the reaction.
These observations are consistent with associative kinetics and
suggest quite clearly that the carbonylations of 2a-d do not
proceed by a rate-determining dissociation of the aryloxide
ligand.
While these observations rule out a dissociative pathway, two
associative pathways can still be considered. The first pathway
involves classical migratory insertion: axial association of
carbon monoxide with the square planar complex followed by
insertion into the Pt-OAr bond, eq 4. The second pathway
8
in trans-Pt(C(O)OEt)2(PPh3)2 and Pt(C(O)OEt)(CtCC(O)-
7
OMe)(PPh3)2 at 1.459(11) Å and 1.436(16) Å are also
comparable. Apparently, the cationic nature of the platinum
center in 3b results in poorer orbital overlap and a longer Pt-C
bond. The result is more electron density on the carbonyl carbon
atom, C(40), and shorter C(40)-O(40) and C(40)-O(41) bonds.
A final structural point regarding 3b is that the (aryloxy)carbonyl
group adopts a geometry that is tilted 75.4(9)° with respect to
the PtP3C plane, while comparable alkoxycarbonyls adopt a
nearly perpendicular orientation.6,7,14,52
Kinetics of Carbonylation. The carbonylations of 2a-d
were followed by 31P{1H} NMR at ambient temperatures, Figure
2. The kinetics of carbonylation are first order in [2a-d] to
better than 4 half-lives. Varying the carbon monoxide pressures
from 50 to 134 psi also indicated that the observed rates were
involves an associative substitution of aryloxide for carbon
monoxide, eqs 5 and 6, followed by nucleophilic addition of
the displaced aryloxide to the coordinated carbon monoxide,
eq 7.
In order to distinguish between the two mechanisms that
predict associative kinetics with carbon monoxide, aryloxide
ligand exchange and crossover were examined. When 1 equiv
of NaOC6H4-p-R′ (R′ ) OMe, H) is added to a solution of [Pt-
(51) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen, A.
G.; Taylor, R. J. Chem. Soc., Perkin Trans. 2 1987, S1-S19.
(52) Zhong, Z.; Stang, P. J.; Arif, A. M. Organometallics 1990, 9, 1703-
1706.

ReactiVity of Platinum-Oxygen Bonds
J. Am. Chem. Soc., Vol. 118, No. 20, 1996 4851
Table 5. Pseudo-First-Order Rate Constants for the Carbonylation
of the Aryloxy Complexes 2a-d
kobs
Hammett
compound
(×10-4 M-1 s-1
)
σ47
[Pt(triphos)(OC6H4F)][PF6] (2d)
[Pt(triphos)(OC6H4H)][PF6] (2c)
[Pt(triphos)(OC6H4Me)][PF6] (2b)
[Pt(triphos)(OC6H4OMe)][PF6] (2a)
1.7 ( 0.2
2.3 ( 0.2
1.3 ( 0.2
1.0 ( 0.2
0.06
0.00
-0.17
-0.27
(triphos)(OC6H4-p-Me)][PF6] (2b), the 31P{1H} NMR shows two
products at equilibrium in solution. These two products can
be identified as [Pt(triphos)(OC6H4-p-Me)][PF6] (2b) and
[Pt(triphos)(OC6H4-p-R′)][PF6] (2a or 2c). The reaction is slow
enough to enable the observation of two distinct species on the
NMR time scale, and the reaction is complete within 20 min.
This brackets the rate constant for exchange between 102 and
10-1 s-1. These NMR studies also facilitate the measurement
of relative stability constants, Keq, for the different platinum
aryloxy complexes 2a-e, eq 8. Equilibrium constants of 0.75
[Pt-OC6H4-p-R)]+ + -OC6H4-p-R′ {Keq}
[Pt-OC6H4-p-R′)]+ + -OC6H4-p-R (8)
Figure 5. Dependence of the observed rate of (aryloxy)carbonyl
formation on the aryloxide substituent.
( 0.2 for the [Pt(triphos)(OC6H4-p-Me)]+(2b)/[Pt(triphos)-
(OC6H4-p-OMe)]+ (2a) system, 0.31 ( 0.2 for the [Pt(triphos)-
(OPh)]+ (2c)/[Pt(triphos)(OC6H4-p-Me)]+ (2b) system, and 0.23
( 0.2 for the [Pt(triphos)(OPh)]+ (2c)/[Pt(triphos)(OC6H4-p-
OMe)]+ (2a) system were obtained. These equilibrium con-
stants favor the platinum aryloxide complexes of the more
nucleophilic aryloxide ligands and should be considered within
the context of the essentially thermoneutral behavior reported
earlier for the exchange of platinum methoxides with methanol.4
When the equilibrated mixtures of platinum aryloxide complexes
are reacted with carbon monoxide at 100 psi, both insertion
products appear together at rates seen for the same aryloxide
complexes alone. Aryloxide exchange is at least a factor of
1000 times faster than carbon monoxide insertion. Since
aryloxide ligand exchange in these systems is so facile, aryloxide
displacement by carbon monoxide is also expected to be rapid,
eq 6. Thus, unless the presence of carbon monoxide profoundly
affects the rate of this exchange, it is difficult to seriously
consider aryloxide dissociation as the rate-determining step in
the carbonylation of 2a-e. Inasmuch as square planar ligand
substitutions are normally associative, we expect phenoxide
exchange to be associative. However, we see no indication that
phenoxide inhibits the carbon monoxide insertion reaction. This
suggests that the binding of carbon monoxide to the square
planar complex is stronger than phenoxide binding. This is to
be expected as phenoxide may only donate into a high-energy
p orbital, while CO may also interact with the lower energy
filled dπ orbitals. The crossover experiment showing that
equilibrated mixtures of aryloxide compounds are carbonylated
to give both insertion products at the same rates expected for
the isolated species is inconsistent with the CO-induced aryl-
oxide dissociative mechanism, eqs 5-7. In the crossover
experiment this mechanism would predict preferential dissocia-
tion of the better leaving group and nucleophilic addition by
the more nucleophilic aryloxide already in solution. The
expected result would thus be preferential formation of the
insertion product of the more nucleophilic aryloxide, contrary
to our experimental observations.
The rate of carbon monoxide insertion is also related to the
electronic effects introduced by the para-substituents of the
aryloxides, Table 5. The rate of carbonylation for the more
nucleophilic aryloxides is slower than that for the less nucleo-
philic aryloxides, Figure 5. These rates also correlate well with
Hammett σ values.53 Clearly, the opposite trend would be
observed if nucleophilic addition of the displaced aryloxide, eq
7, were the rate-determining step. The presence of excess
aryloxide would also affect the rate if nucleophilic addition to
coordinated carbon monoxide were rate-determining. This was
not observed. These results remove nucleophilic addition of
displaced aryloxide to coordinated carbon monoxide, eq 7, as a
reasonable rate-determining step. Overall, the facility of aryl-
oxide exchange and relative rates of carbonylation of equili-
brated mixtures of different aryloxide complexes together with
the decreased rates of carbonylation for platinum complexes
of more nucleophilic aryloxide ligands cannot be reconciled with
the carbon monoxide-induced aryloxide dissociation mechanism
summarized in eqs 5-7.
The relative stability constants of the aryloxide complexes
2a-c and the electronic influence of para-substituents on the
aryloxide ligands both indicate the importance of Pt-O bond
breaking in the carbonylation mechanism. When ∆G°, calcu-
lated from the relative stability constants, is compared to ∆Ea,
calculated from relative rates for carbonylation, a linear relation-
ship is found. For example, the ∆G° for the [Pt(triphos)-
(OPhMe)]+ (2b)/[Pt(triphos)(OPhOMe)]+ (2a) equilibrium is
calculated to be 0.71 kJ/mol, and ∆Ea is 0.48 kJ/mol. In the
[Pt(triphos)(OPhMe)]+ (2b)/[Pt(triphos)(OPh)]+ (2c) equilibri-
um, ∆G° is 2.87 kJ/mol and ∆Ea is 1.52 kJ/mol. Finally for
the [Pt(triphos)(OPh)]+ (2c)/[Pt(triphos)(OC6H4-p-OMe)]+ (2a)
equilibrium ∆G° is 3.59 kJ/mol and ∆Ea is 1.99 kJ/mol. This
linear free energy relationship suggests that differences in Pt-O
bond energy, determined from the relative stability constants,
(53) A plot of the rate of carbon monoxide insertion versus Hammet σ
gives a best fit line of kobs ) 5.16σ + 2.20.

4852 J. Am. Chem. Soc., Vol. 118, No. 20, 1996
Dockter et al.
prior to complete Pt-O bond cleavage.5 This effect would
stabilize the transition state and accelerate the rates for the more
nucleophilic aryloxides, a trend opposite that observed by us
experimentally. We believe that a transition state model like
that shown in Figure 6 is sensible for carbonylation of
aryloxides, but in our system the influence of Pt-O bond
breaking relative to C-O bond making dominates.
Figure 6. Interaction of the carbonyl π* orbital with the aryloxide
It is interesting to note that while we arrive at the same
mechanistic conclusion as Bryndza for the carbonylation of
platinum aryloxide complexes, the two systems have differing
exchange conditions. In the Pt(dppe)(OMe)2 system, methoxide
exchange is found to be slow with respect to insertion.38 In
the aryloxide systems 2a-e, aryloxide exchange is faster than
insertion. In both systems, the mechanism of carbonylation is
a classical migratory insertion of the alkyloxide or aryloxide
ligand to coordinated carbon monoxide.
oxygen lone pair.
contribute to the activation energy for carbonylation.54 This
supports the notion that Pt-O bond breaking is important in
the formation of the transition state. This is further supported
by the dependence of the observed rates of carbon monoxide
insertion on aryloxide ligand para-substitution. More nucleo-
philic aryloxide ligands are slower to carbonylate since a
stronger Pt-O bond must be spent. Together these data support
our assertion of a classical migratory insertion pathway, eq 4,
for the carbonylation of 2a-e. While little is known of the
structure of the transition state involved and the role of ion
solvation energies in the carbonylation of the cationic com-
plexes, our data suggest that differences in Pt-O ground state
bond energies are contributed to the differences in activation
energies. We note that migratory insertion of carbon monoxide
only occurs into the Pt-O bond of Pt(dppe)(Me)(OMe), and
not the available Pt-C bond.38 This preference was attributed
to the interaction of the oxygen lone pairs with the carbonyl
π* orbital, Figure 6. Thus, part of the C-O bond is formed
Acknowledgment is made to the Department of Energy
(Grant DE-FG22-93PC93208) and the National Science Foun-
dation (Grant CHE-9319173) for support of this research. We
are also grateful to Johnson Matthey for a generous loan of K2-
PtCl4 and to the NSF for support of the chemical X-ray
diffraction facility at Purdue.
Supporting Information Available: A plot of ∆Ea vs ∆G°
along with tables of positional and general temperature factors,
bond distances, bond angles, and torsional angles, and a
description of data collection and reduction for 3b (26 pages).
Ordering information is given on any current masthead page.
(54) A plot of ∆G° vs ∆Ea gives a best fit line of ∆Ea ) 0.55∆G°; R2
) 0.994.
JA952680P