Exploring the mechanism of aqueous C-H activation by Pt(II) through model chemistry: Evidence for the intermediacy of alkylhydridoplatinum(IV) and alkane σ-adducts

J. Am. Chem. Soc. 1996, 118, 5961-5976

5961

Exploring the Mechanism of Aqueous C-H Activation by

Pt(II) through Model Chemistry: Evidence for the Intermediacy

of Alkylhydridoplatinum(IV) and Alkane σ-Adducts

Shannon S. Stahl, Jay A. Labinger,* and John E. Bercaw*

Contribution from the Arnold and Mabel Beckman Laboratories of Chemical Synthesis,

California Institute of Technology, Pasadena, California 91125

ReceiVed January 12, 1996

Abstract: The protonolysis mechanisms of several alkylplatinum(II) complexes [(tmeda)PtMeCl (2) (tmeda )

N,N,N′,N′-tetramethylethylenediamine), (tmeda)Pt(CH₂Ph)Cl (5), (tmeda)PtMe₂(11), and trans-(PEt₃)₂Pt(CH₃)Cl (15)]

in CD₂Cl₂and CD₃OD have been investigated. These reactions model the microscopic reverse of C-H activation

by aqueous Pt(II). Each of the four systems (2 in CD₃OD, 5 in CD₂Cl₂, 11 in CD₃OD, and 15 in CD₃OD) exhibits

different behavior in the protonolysis reaction as observed by low-temperature ¹H NMR spectroscopy. Protonolysis

of 2 in methanol-d₄proceeds with no observable intermediates. Reversible reaction between 5 and HCl in CD₂Cl₂

at -78 °C produces (tmeda)Pt(CH₂Ph)(H)Cl₂(6), which undergoes reductive elimination of toluene at higher

temperatures. Treatment of 11 with HCl in methanol at -78 °C generates (tmeda)PtMe₂(H)Cl (12), which incorporates

deuterium from solvent (CD₃OD) into the methyl groups prior to reductive elimination of methane. Finally, 15

reacts with H⁺in methanol to liberate methane with no intermediates observed. However, hydrogen/deuterium

exchange takes place between the solvent (CD₃OD) and Pt-Me prior to methane loss. Each of these reactions was

evaluated further to determine the kinetics of the reaction, activation parameters, and isotope effects. Based on the

results, a common mechanistic sequence is proposed to operate in all the reactions: (1) chloride- or solvent-mediated

protonation of Pt(II) to generate an alkylhydridoplatinum(IV) intermediate, (2) dissociation of solvent or chloride to

generate a cationic, five-coordinate platinum(IV) species, (3) reductive C-H bond formation producing a platinum-

(II) alkane σ-complex, and (4) loss of alkane either through an associative or dissociative substitution pathway. The

characteristics of each system differ due to changes in the relative stabilities of the intermediates and/or transition

states upon varying the solvent or alkylplatinum species.

Identifying practical routes to alkane functionalization con-

tinues to be actively pursued.^1-3Although many organometallic

systems have been identified which selectively actiVate C-H

bonds, most rely on the generation of a high-energy, reactive

intermediate which decomposes rapidly if exposed to oxidants

necessary for catalytic turnover. Recently, several promising,

rather robust systems have been developed based on late

transition metals (e.g., Pd, Pt, and Hg).^4-7These metal

complexes undergo electrophilic activation and subsequent

oxidation of alkanes. The first such homogeneous alkane

oxidation system was discovered by Shilov and co-workers >20

years ago;⁸it utilized a mixture of Pt(II) and Pt(IV) salts in

aqueous solution to convert methane to mixtures of methanol

and methyl chloride. Although various features have thus far

prevented its practical utility, this system remains one of few

examples of catalytic (in Pt(II)) routes to alkane functionaliza-

tion. Recently, we^9-12,21and others^13-20have been exploring

some of the mechanistic features of this system in order to better

understand this remarkable reactivity.

In agreement with Shilov’s original proposal,¹³we believe

the catalytic mechanism consists of three primary transforma-

tions (Scheme 1): (1) electrophilic activation of the alkane by

Pt(II) to generate an alkylplatinum(II) intermediate, (2) two-

electron oxidation of alkylplatinum(II) to generate an alkyl-

platinum(IV) species, and (3) reductive elimination of R-X (X

) Cl or OH for example) to liberate the functionalized alkane

and Pt(II) catalyst.

In recent years, we have elucidated several details related

to the final two steps of the mechanism.^9-12Unfortunately,

very little is known about the nature of C-H activation, the

step which appears to govern the overall rate as well as the

(9) Labinger, J. A.; Herring, A. M.; Lyon, D. K.; Luinstra, G. A.; Bercaw,

J. E.; Horvath, I. T.; Eller, K. Organometallics 1993, 12, 895.

(10) Luinstra, G. A.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.

1993, 115, 3004.

(11) Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J.

E. Organometallics 1994, 13, 755.

(12) Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J.

E. J. Organomet. Chem. 1995, 504, 75.

Abstract published in AdVance ACS Abstracts, June 1, 1996.

(1) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.; Peterson, T. H.

Acc. Chem. Res. 1995, 28, 154.

(2) SelectiVe Hydrocarbon ActiVation; Davies, J. A., Watson, P. L.,

Liebman, J. F., and Greenberg, A., Eds.; VCH: New York, 1990.

(3) ActiVation and Functionalization of Alkanes; Hill, C. L., Ed.; John

Wiley & Sons: New York, 1989.

(4) Gretz, E.; Oliver, T. F.; Sen, A. J. Am. Chem. Soc. 1987, 109, 8109.

(5) Kao, L.-C.; Hutson, A. C.; Sen, A. J. Am. Chem. Soc. 1991, 113,

700.

(13) Kushch, L. A.; Lavrushko, V. V.; Misharin, Y. S.; Moravsky, A.

P.; Shilov, A. E. NouV. J. Chim. 1983, 7, 729.

(14) Horvath, I. T.; Cook, R. A.; Millar, J. M.; Kiss, G. Organometallics

1993, 12, 8.

(15) Sen, A.; Lin, M.; Kao, L.-C.; Hutson, A. C. J. Am. Chem. Soc.

1992, 114, 6385.

(16) Sen, A.; Benvenuto, M. A.; Lin, M.; Hutson, A. C.; Basickes, N. J.

Am. Chem. Soc. 1994, 116, 998.

(17) Hutson, A. C.; Lin, M.; Basickes, N.; Sen, A. J. Organomet. Chem.

1995, 504, 69.

(6) Labinger, J. A.; Herring, A. M.; Bercaw, J. E. J. Am. Chem. Soc.

1990, 112, 5628.

(7) Periana, R. A.; Taube, D. J.; Evitt, E. R.; Loffler, D. G.; Wentrcek,

P. R.; Voss, G.; Masuda, T. Science 1993, 259, 340.

(8) Goldshleger, N. F.; Eskova, V. V.; Shilov, A. E.; Shteinman, A. A.

Zh. Fiz. Khim. 1972, 46, 1353.

(18) Zamashchikov, V. V.; Litvinenko, S. L.; Uzhik, O. N.; Galat, V. F.

Zh. Obs. Khim. 1986, 56, 2417.

(19) Zamashchikov, V. V.; Popov, V. G.; Litvinenko, S. L. IzV. Akad.

Nauk SSSR, Ser. Khim. 1993, 42, 389.

(20) Zamashchikov, V. V.; Popov, V. G.; Rudakov, E. S.; Mitchenko,

S. A. Dokl. Akad. Nauk SSSR 1993, 333, 34.

5962 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

Scheme 1

Related alkylplatinum(II) model complexes containing sta-

bilizing donor ligands (e.g., amines or phosphines) proved more

amenable to study. Protonolysis studies of such complexes have

been conducted by other groups in the past; however, several

different mechanisms have been proposed.^27-31Furthermore,

when we began our work, no intermediates had been identified

in such reactions, and thus mechanistic proposals were based

on less direct evidence such as rate laws, isotope effects, or

competition studies. Recently, however, three groups have

independently identified the presence of alkylhydridoplatinum-

(IV) intermediates in the protonolysis of alkylplatinum(II)

species in organic solvents.^21,32-34We report herein studies of

the protonolysis reaction intended to examine the viability of

alkane adducts (σ-complexes) and/or alkylhydridoplatinum(IV)

species as intermediates in alkane activation by Pt(II) in aqueous

solution.

Results

selectivity of the catalytic cycle. Recently we have been

attempting to delineate some of the mechanistic features of this

reaction.

Two possible pathways have been proposed for C-H

activation leading to an alkylplatinum(II) intermediate:^9,22-24

(1) oxidative addition of the C-H bond at Pt(II) yielding an

alkylhydridoplatinum(IV) species which is subsequently depro-

tonated (eq 1) or (2) deprotonation of an intermediate Pt(II)-

alkane σ-complex (eq 2). Neither intermediate has been

observed directly.

Protonolysis of (tmeda)Pt(CH₃)Cl (2) (tmeda ) N,N,N′,N′-

Tetramethylethylenediamine) in Methanol. The protonolysis

of (tmeda)Pt(CH₃)Cl (2) with HCl in methanol proceeds cleanly

to generate (tmeda)PtCl₂and methane (eq 3).

Monitoring the reaction at low temperature (-50 to -25 °C)

in CD₃OD by ¹H NMR spectroscopy reveals pseudo-first-

order kinetics for the conversion of 2 to (tmeda)PtCl₂(3);

no intermediates are obserVed.³⁵The rate is first order in

both hydrogen ion and chloride concentration (Figures 1 and

2, respectively) at constant ionic strength (µ ) 1.0 M ) [HOTf]

+ [LiClO₄] + [LiCl]) over the concentration ranges [HOTf]

) 0 to 0.38 M and [Cl^-] ) 0.13 to 0.66 M. Significantly,

the dependence on chloride exhibits a non-zero intercept

(Figure 2). Activation parameters were determined from an

Eyring plot: ∆H ) 19.4 ( 1.5 kcal‚mol^-1and ∆S ) 14 ( 5

eu.³⁶

In the course of our investigations, we have encountered

difficulties exploring direct reactions of aqueous Pt(II) salts

with alkanes. Even in the absence of Pt(IV), oxidation pro-

ducts are generated with immediate precipitation of platinum

metal, accompanied by extensive H/D exchange between

alkanes and deuterated solvent. Unlike previous reports,^22,25

we have not observed H/D exchange prior to formation of

Pt(0).

Protonolysis of 2 with LOTf (L ) H, D) in a 19.3:1 mixture

of CD₃OD:CD₃OH at 0 °C generates a mixture of methane

isotopomers, CH₄:CH₃D ) 1:2.13, leading to a kinetic deuterium

(26) Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Mol. Catal.

In press.

(27) Belluco, U.; Giustiniani, M.; Graziani, M. J. Am. Chem. Soc. 1967,

89, 6494.

(28) Alibrandi, G.; Minniti, D.; Romeo, R.; Uguagliati, P.; Calligaro,

L.; Belluco, U.; Crociani, B. Inorg. Chim. Acta 1985, 100, 107.

(29) Alibrandi, G.; Minniti, D.; Romeo, R.; Uguagliati, P.; Calligaro,

L.; Belluco, U. Inorg. Chim. Acta 1986, 112, L15.

(30) Jawad, J. K.; Puddephatt, R. J.; Stalteri, M. A. Inorg. Chem. 1982,

21, 332.

(31) Siegbahn, P. E. M.; Crabtree, R. H. J. Am. Chem. Soc. 1996, 118,

4442.

(32) De Felice, V.; De Renzi, A.; Panunzi, A.; Tesauro, D. J. Organomet.

Chem. 1995, 488, C13.

(33) Hill, G. S.; Rendina, L. M.; Puddephatt, R. J. Organometallics 1995,

14, 4966.

(34) Arylhydridoplatinum(IV) species have been identified previously:

(a) Arnold, D. P.; Bennett, M. A. Inorg. Chem. 1984, 23, 2110. (b) Wehman-

Ooyevaar, I. C. M.; Grove, D. M.; de Vaal, P.; Dedieu, A.; van Koten, G.

Inorg. Chem. 1992, 31, 5484.

We subsequently began investigating the microscopic reverse

of the C-H activation step, namely protonolysis of alkylplati-

num(II) species. The catalytic intermediate, [Cl₃Pt^IICH₃]^2-(1),

is presumed to be generated upon 2e^-reduction of [Pt^IVMeCl₅]^2-,

however 1 immediately liberates methane when the reaction

is carried out in a protic solvent (i.e., CH₃OH and H₂O).

Furthermore, in aprotic media we found that 1 is unstable

with respect to disproportionation into [PtCl₄]^2-and [Cl₂Pt^II-

(CH₃)₂]^{2- 12,26}

.

(21) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1995,

117, 9371.

(22) Shilov, A. E. ActiVation of Saturated Hydrocarbons by Transition

Metal Complexes; Reidel: Dordrecht, 1984.

(23) Shilov, A. E. In ActiVation and Functionalization of Alkanes; Hill,

C. L., Ed.; John Wiley & Sons: New York, 1989; p 1.

(24) Crabtree, R. H. Chem. ReV. 1995, 95, 987.

(25) Goldshleger, N. F.; Tyabin, M. B.; Shilov, A. E.; Shteinman, A. A.

Zh. Fiz. Khim. 1969, 43, 2174.

(35) Monitoring the reaction by UV-visible spectroscopy was attempted,

but the product, (tmeda)PtCl₂, is insoluble and precludes the acquisition of

reproducible data.

(36) Reaction conditions (Eyring plot for protonolysis of 2 in CD₃OD):

[2] ) 11.9 mM; [Cl^-] ) 0 M; [H⁺] ) 0.18 M; µ ) 1.18 M; temperature

range -25 to -49 °C.

Mechanism of Aqueous C-H ActiVation by Pt(II)

J. Am. Chem. Soc., Vol. 118, No. 25, 1996 5963

Scheme 2

elimination proceeds at -30 °C (Scheme 2). Consequently, this

species was used to evaluate further mechanistic details.

Oxidative addition of HCl to 5 is rapid and reversible: addition

of DCl (as a diethyl ether-d₁₀solution) to 5 in CD₂Cl₂at -78

°C produces 6-d₁. Subsequent addition of a large excess of

HCl results in the immediate appearance of the hydride

resonance for 6 with no other spectroscopic changes.

Figure 1. Plot of the [H⁺] dependence of the rate of protonolysis of

2 in CD₃OD. [H⁺] ) 0-32 equiv; [2] ) 11.9 mM; [Cl^-] ) 0 M; µ )

1.0 M; temperature ) -35 °C.

Reductive elimination of toluene follows a first-order depen-

dence on [6]. The temperature dependence of the reductive

elimination rates allowed determination of activation param-

eters: ∆H ) 14.0 ( 2.5 kcal‚mol^-1, ∆S ) -18.5 ( 7.0 eu³⁸

(Figure 3). The rate of toluene formation from 6 is substantially

increased by addition of Bro¨nsted or Lewis acids (HCl, HOTf,

SnCl₄), and in the case of triflic acid, the rate of reductive

elimination of toluene is first order with respect to [HOTf] (0-

263 mM; 0-20 equiv) (Figure 4). Moreover, direct protonation

of 5 with HOTf (i.e., rather than addition of HOTf to a solution

of 6) at -80 °C proceeds immediately to toluene without

evidence of (tmeda)Pt(CH₂Ph)(H)(Cl)(OTf). A series of experi-

ments were carried out in which the [HOTf] was held constant

(5 equiv relative to [6]) and the HCl concentration was varied

between 0.11 and 0.62 M (9.4-51 equiv). As is apparent from

Figure 5, toluene elimination is inhibited by added HCl. The

dependence of the rate on chloride alone was also examined;

however, surprisingly, the addition of tetrabutylammonium

chloride to the reaction mixture results in immediate deproto-

nation of 6 and formation of 5 and tetrabutylammonium

bichloride (eq 4).

Figure 2. Plot of k_obsversus [Cl^-] for the protonolysis of 2 in CD₃-

OD. [Cl^-] ) 0-55.4 equiv; [2] ) 11.9 mM; [H⁺] ) 0.12 M; µ ) 1.0

M; temperature ) -35 °C.

isotope effect of 9.1 ((0.5). Comparing the pseudo-first-order

rate constant for protonolysis of 2 in CD₃OH with HOTf (k_obs

) 1.8 × 10^-3s^-1) versus CD₃OD with DOTf (k_obs) 0.77 ×

10^-3s^-1) at -40 °C leads to a different isotope effect: k_H/k_D

) 2.3 ((0.5).

Protonolysis of (tmeda)Pt(R)Cl (R ) CH₃, CH₂Ph) in

Methylene Chloride. Protonolysis of 2 also proceeds rapidly

in dichloromethane at room temperature, releasing methane and

also generating 3. In contrast to reaction with HCl in methanol,

addition of excess HCl (dissolved in diethyl ether-d₁₀) at -78

°C to a CD₂Cl₂solution of 2 first results in oxidative addition

to generate (tmeda)Pt(CH₃)(H)Cl₂(4). Upon warming this

solution to -60 °C, the reductive elimination of methane ensues

along with formation of (tmeda)PtCl₂(Scheme 2).

Although 4 is not sufficiently stable to isolate at room

temperature, it has been characterized by ¹H NMR spectroscopy

at low temperature (Table 1). The most distinctive characteristic

is the platinum(IV) hydride resonance which appears far upfield

The presence of [NBu₄][Cl-H-Cl] was verified by comparing

a solution FTIR spectrum of the reaction mixture to a solid-

state infrared spectrum reported for [NMe₄][Cl-H-Cl].³⁹

Integration of the the Pt^IV-H peak relative to one of the

benzyl protons for 6 obtained by addition of a mixture of HCl

and DCl to 5 reveals an inVerse equilibrium deuterium isotope

(δ -23.6 ppm) with corresponding platinum satellites (¹J

¹⁹⁵Pt-H

1

) 1290 Hz, ¹⁹⁵Pt 33.8% natural abundance, S ) /₂). These

data are consistent with other Pt(IV) hydrides containing

nitrogen and halide ligands.^32,33,37The methyl resonance of 4

2

195

appears as a singlet (δ 1.6 ppm) with J

) 66.5 Hz.

Pt-H

The benzyl analog of 2, (tmeda)Pt(CH₂Ph)Cl (5), reacts

similarly with HCl, except the benzylhydridoplatinum(IV)

complex (6) (Table 1) is more stable than 4 toluene reductive

(38) Reaction conditions (Eyring plot for reductive elimination of toluene

from 6 in CD₂Cl₂): [6] ) 13 mM; [HCl] ) 65 mM; temperature range

-24 to -43 °C.

(39) Two strong, very broad absorptions were observed at ∼1550 and

1000 cm^-1. Solid-state IR spectrum of [NMe₄][Cl-H-Cl]: Waddington,

T. C. J. Chem. Soc. 1958, 1708.

(37) Wehman-Ooyevaar, I. C. M.; Grove, D. M.; de Vaal, P.; Dedieu,

A.; van Koten, G. Inorg. Chem. 1992, 31, 5484.

5964 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

1

Table 1. Selected H NMR Spectroscopic Data for Compounds Discussed in Text

chemical shift/ppm

(multiplicity); coupling/Hz

0.434 (s); ²J_Pt-H) 78.9

2.84 (s); ³J_Pt-H) 51

2.70 (s); ³J_Pt-H) <10

2.77 (m)

compound

(tmeda)Pt(CH₃)Cl (2)

solvent

temp (°C)

peak assignment

CD₂Cl₂

25

Pt-CH₃

Pt-N-CH₃

Pt-N-CH₂-

2.54 (m)

(tmeda)Pt(CH₃)(H)Cl₂(4)

CD₂Cl₂/Et₂O-d₁₀(10:1)

-70

Pt-CH₃

Pt-H

Pt-N-CH₃

1.57 (s); ²J_Pt-H) 66.5

-23.6 (s); ¹J_Pt-H) 1292

3.02; ³J_Pt-H∼ 57

2.89; ³J_Pt-H) <10

2.86; ³J_Pt-H∼ 33

(tmeda)Pt(CH₂Ph)Cl (5)

CD₂Cl₂(300 MHz)

25

Pt-CH₂Ph

Pt-N-CH₃

2.96 (s); ²J_Pt-H) 109.8

2.76 (s); ³J_Pt-H) 83

2.70 (s); ³J_Pt-H) <10

2.47 (m) [second multiplet obscured

by peaks at 2.76 and 2.70]

4.8 (d); ²J_Pt-H) 107, ²J_Ha-Hb) 10

3.2 (d); ²J_Pt-H) 92

-23.9 (s); ¹J_Pt-H) 1230

3.0; ³J_Pt-H∼ 32

Pt-N-CH₂-

(tmeda)Pt(CH₂Ph)(H)Cl₂(6)

CD₂Cl₂

-60

Pt-CH_aH_bPh

Pt-H

Pt-N-CH₃

2.9; ³J_Pt-H) <10

2.8; ³J_Pt-H) <10

2.7; ³J_Pt-H∼ 56

(tmeda)Pt(CH₂Ph)₂(7)

CD₂Cl₂

25

Pt-CH₂Ph

Pt-N-CH₃

Pt-N-CH₂-

Pt-CH_aH_bPh

Pt-H

2.71 (s), ²J_Pt-H) 122

2.56 (s); ³J_Pt-H) 20

2.54 (s); ³J_Pt-H) <10

3.44 (d); ²J_Pt-H) 95, ²J_Ha-Hb) 10

3.31 (d); ²J_Pt-H) 117

-23.1 (s); ¹J_Pt-H) 1630

2.76 (s), ³J_Pt-H) <15

2.51 (s), ³J_Pt-H) 27

2.63 (br s)

(tmeda)Pt(CH₂Ph)₂(H)Cl (8)

CD₂Cl₂/Et₂O-d₁₀(10:1)

-20

Pt-N-CH₃

Pt-N-CH₂-

2.91 (br s)

(4,4′-Me₂-2,2′-bpy₂)Pt(CH₃)₂(9)

CD₂Cl₂

25

Pt-CH₃

Pt-(py-CH₃)

Pt-CH₃

Pt-H

Pt-(py-CH₃)

Pt-CH₃

0.90 (s); ²J_Pt-H) 85

2.36 (s)

(4,4′-Me₂-2,2′-bpy₂)Pt(CH₃)₂(H)Cl (10)

CD₂Cl₂/Et₂O-d₁₀(10:1)

-65

1.25 (s); ²J_Pt-H) 66

-22.1 (s); ¹J_Pt-H) 1610

2.44 (s)

(tmeda)Pt(CH₃)₂(H)Cl (12)

CD₃OD

-50

0.87 (s, 3 H); ²J_Pt-H) 66

-23.1(s); ¹J_Pt-H) 1726

2.78 (s); ³J_Pt-H) 24

2.72 (s); ³J_Pt-H) 10

2.91 (br s)

Pt-H

Pt-N-CH₃

Pt-N-CH₂-

Pt-CH₃

Pt-H

[(tmeda)Pt(CH₃)₂(H)(CD₃OH)](OTf) (13)

CD₃OD

-55

0.84 (s); ²J_Pt-H) 66

-23.1 (s); ¹J_Pt-H) 1721

2.79 (s); ³J_Pt-H) 27

2.70 (s); ³J_Pt-H) <10

2.91 (br s)

Pt-N-CH₃

Pt-N-CH₂-

Pt-CH₃

Pt-H

trans-(PEt₃)₂Pt(CH₃)Cl (15)

(PEt₃)₂Pt(CH₃)(H)Cl₂(16)

CDCl₃

25

-65

0.325 (t); ³J_P-H) 6; ²J_Pt-H) 84

0.935 (t); ³J_P-H) 5; ²J_Pt-H) 65

-18.8 (t); ²J_P-H) 6; ¹J_Pt-H) 1223

CD₂Cl₂/Et₂O-d₁₀(10:1)

effect, (K_H/K_D) ) 0.51 ( 0.05 (-28 °C), for equilibrium 5.

to 2 or 5. We gained further insight by investigating the reaction

of HCl with dialkylplatinum(II) species, since the products

exhibit greater symmetry. Protonation of (tmeda)Pt(CH₂Ph)₂

(7) produces (tmeda)Pt(CH₂Ph)₂(H)Cl (8) which exhibits two

inequivalent resonances for its tmeda methyl protons in the ¹H

NMR spectrum at -45 °C (Table 1). Furthermore, addition of

HCl to (4,4′-dimethyl-2,2′-bipyridyl)Pt(CH₃)₂(9) leads to a

dialkylhydridoplatinum(IV) complex (10) with only a single

resonance in the ¹H NMR spectrum corresponding to the

methyls on the bipyridyl ligand (see Table 1). These results

indicate trans addition of HCl to 7 and 9, and thus, by extension,

to 2 and 5 as well (see Discussion section).

Comparison of the resulting concentrations of PhCH₃versus

PhCH₂D generated upon decomposition of this mixture of

isotopomers indicates a kinetic deuterium isotope effect for

reductive elimination k_H/k_D) 1.55 ( 0.10 at both -28 and 0

°C.

Protonolysis of (tmeda)Pt(CH₂Ph)₂, (4,4′-Dimethyl-2,2′-

bipyridyl)Pt(CH₃)₂, and (tmeda)Pt(CH₃)₂in Methylene Chlo-

ride. Since the Pt(IV) complexes described above were not

characterized by X-ray crystallography, we were unable to

unambiguously establish the stereochemistry of HCl addition

During this study, we discovered that dialkylhydridoplatinum-

(IV) complexes are more stable than the corresponding monoal-

kylhydridoplatinum(IV) species with respect to alkane loss.

Whereas (tmeda)Pt(CH₂Ph)(H)Cl₂(6) eliminates toluene at

approximately -30 °C, a comparable rate for reductive elimina-

tion from (tmeda)Pt(CH₂Ph)₂(H)Cl (8) does not occur until -5

Mechanism of Aqueous C-H ActiVation by Pt(II)

J. Am. Chem. Soc., Vol. 118, No. 25, 1996 5965

Figure 3. Eyring plot for the reductive elimination of toluene from 6

in CD₂Cl₂. Calculated activation parameters: ∆H ) -14.0 ( 2.5 kcal/

mol, ∆S ) -18.5 ( 7.0 eu. [6] ) 13 mM; [HCl] ) 65 mM;

temperature ) -24 to -43 °C.

Figure 6. Sequence of ¹H NMR spectra demonstrating the appearance

of shoulders upfield of the original Pt(IV)-CH₃resonance from 12 in

CD₃OD along with an overall decrease in the peak integration. The

upfield shoulders are assigned as the sequentially deuterated methyl

groups: a ) Pt-CH₃, b ) Pt-CH₂D, and c ) Pt-CHD₂.

Scheme 3

Figure 4. [HOTf] dependence of the rate of reductive elimination of

toluene from 5 in CD₂Cl₂. [HOTf] ) 0-20 equiv (relative to [6]); [6]

) 13 mM; [HCl] ) 65 mM; temperature ) -61 °C.

even in the absence of acid at room temperature,⁴⁰the following

procedure was adopted: 11 was dissolved in a minimal amount

of CD₂Cl₂and then cooled to -78 °C before adding methanol-

d₃(CD₃OH:CD₂Cl₂) 13:1). Subsequent addition of excess

HCl to this solution at -78 °C leads to formation of the

oxidative addition product, (tmeda)Pt(CH₃)₂(H)Cl (12) (Table

1). Addition of HOTf also leads to a stable alkylhydridoplati-

num(IV) complex, presumably the methanol coordinated prod-

uct, [(tmeda)Pt(CH₃)₂(H)(CD₃OH)](OTf) (13) (Table 1). Sig-

nificantly, monitoring the ¹H NMR spectrum of 12-d₁in CD₃OD

as it is warmed to -40° C revealed deuterium incorporation

into the methyl positions (Scheme 3, Figure 6). When warmed

to room temperature to allow methane loss prior to complete

deuterium exchange, the full range of methane isotopomers is

observed (Figure 7).

H/D exchange rates were obtained by monitoring the decrease

in total integration of H NMR signals corresponding to Pt-

CL₃(L ) H, D) for a solution of 12 in CD₃OD. The activation

parameters for H/D exchange were derived from an Eyring

plot: ∆H ) 15.8 ((1.8) kcal/mol and ∆S ) -4.7 ((7.5)

eu.⁴¹At constant ionic strength, added chloride inhibits H/D

1

Figure 5. Plot of k_obsversus [HCl] revealing that HCl inhibits the

reductive elimination of toluene from 6 in CD₂Cl₂when [HOTf] is

present. [HCl] ) 9.4-51 equiv (relative to [6]); [6] ) 12 mM; [HOTf]

) 61 mM; temperature ) -62 °C.

°C. The methyl derivatives (tmeda)Pt(CH₃)(H)Cl₂(4) and

(tmeda)Pt(CH₃)₂(H)Cl (12) show a more pronounced effect

methane loss from 4 occurs rapidly at -60 °C, but not until

+10 °C from 12.

Protonolysis of (tmeda)Pt(CH₃)₂(11) in CD₃OD. Upon

observing the higher stability of dialkylhydridoplatinum(IV)

species, we investigated the protonolysis of (tmeda)Pt(CH₃)₂

(11) in methanol. Since 11 appears to decompose in methanol

(40) No methane is generated in this side reaction. The reactivity is likely

related to that previously observed for similar compounds dissolved in protic

solvents (alcohols and water): (a) Monaghan, P. K.; Puddephatt, R. J.

Organometallics 1984, 3, 444. (b) Holtcamp, M. W.; Labinger, J. A.;

Bercaw, J. E. Unpublished results.

(41) Reaction conditions (Eyring plot for deuterium incorporation into

12 in CD₃OD:CD₂Cl₂(13:1)): [12] ) 21 mM; [H⁺] ) 0.21 M; [Cl^-] )

0.20 M; µ ) 1.21 M; temperature range -31 to -55 °C.

5966 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

Figure 9. Plot of k_obsversus 1/[Cl^-] for methane elimination from 12

in CD₃OH:CD₂Cl₂(13:1). [Cl^-] ) 0-0.80 M; [12] ) 21 mM; [H⁺] )

0.21 M; µ ) 1.21 M; temperature ) -28 °C.

Proton incorporation into the platinum-methyl positions of

(tmeda)Pt(CD₃)₂(H)Cl (12-d₆) in CH₃OH and CD₄loss from

12-d₇in CH₃OD were monitored by ²D NMR. Kinetic isotope

effects were calculated by comparing these rates with those for

deuterium incorporation and CH₄loss from 12: k_H/k_D) 1.9

((0.2) at -47.6 °C for H/D incorporation into Pt-CL₃; k_H/k_D

) 0.29 ((0.05) at -26.8 °C for methane loss.

Figure 7. Full range of methane isotopomers that arise upon warming

a reaction mixture containing 12 in CD₃OD prior to complete deuterium

incorporation into the platinum-methyl groups.

Protonolysis of trans-(PEt₃)₂Pt(CH₃)Cl (15) in Methylene

Chloride and Methanol. Addition of HCl (as a Et₂O-d₁₀

solution) to a CD₂Cl₂solution of trans-(PEt₃)₂Pt(CH₃)Cl (15)

at -78 °C generates (PEt₃)₂Pt(CH₃)(H)Cl₂(16) (eq 6).

The ¹H NMR spectrum for 16 clearly reveals the hydride

resonance at δ -18.8 ppm with coupling to both platinum and

phosphorus (Table 1). Consistent with the formation of a

platinum(IV) complex, the resonances associated with the

methyl group shift downfield and the two-bond platinum-

hydrogen coupling constant decreases relative to 15. As the

solution is warmed to approximately -45 °C loss of methane

is observed along with formation of (PEt₃)₂PtCl₂(17) (eq 7).

Figure 8. Plot of k_obsversus 1/[Cl^-] for deuterium incorporation into

12 in CD₃OD:CD₂Cl₂(13:1). [Cl^-] ) 0.10-0.60 M; [12] ) 21 mM;

[H⁺] ) 0.21 M; µ ) 1.21 M; temperature ) -43 °C.

exchange, and the rate was found to depend on the inverse of

[Cl^-] (see Figure 8).

By monitoring the ¹H NMR resonance of methylplatinum(IV)

in CD₃OH (i.e., a solvent in which no H/D exchange occurs),

12 was found to lose methane cleanly at approximately -25

°C exhibiting first-order kinetics with activation parameters:

∆H ) 19.1 ((0.6) kcal/mol and ∆S ) 4.1 ((2.2) eu.⁴²The

rate of methane elimination from 12 is also inversely propor-

tional to the chloride concentration (Figure 9).

As expected, (tmeda)Pt(CH₃)Cl (2) is not observed during

the conversion of 12 to 3; 2 decomposes rapidly under these

conditions (see above). However, a different methylplatinum-

(II) intermediate (14) does appear as the reaction proceeds. The

amount of 14 observed in solution decreases as the initial

chloride concentration is increased; furthermore, loss of methane

from 14 does not take place until approximately +15 °C well

above the temperature for methane loss from 12 or 2. 14 was

identified as [(tmeda)Pt(CH₃)(methanol)]⁺by independent

Protonolysis of 15 in CD₃OH revealed yet another reaction

pattern. Addition of DOTf to 15 in CD₃OD at -78 °C (µ ) 1

M) produces no reaction, i.e., neither an alkylhydridoplatinum-

(IV) intermediate nor methane is generated. But as the solution

is warmed to -25 °C, deuterium incorporation into the Pt-

CH₃sites takes place with no Pt(IV) intermediate observed

(Scheme 4); only 15 and its isotopomers are detected in solution

1

by H NMR spectroscopy prior to methane loss. The kinetics

were monitored by integrating the total Pt-CH_3-nD_n¹H NMR

signal over time (Figure 10).

1

synthesis (from treating 2 with AgOTf) and comparison of H

NMR parameters and protonolysis behavior. Fortunately,

formation of 14 does not interfere with the kinetics for

conversion of 12 to methane and 3.

In the presence of DOTf, with no excess chloride added to

the reaction mixture, full deuterium incorporation takes place

with no competition from methane loss. As the chloride

concentration is increased, methane loss begins to compete with

H/D exchange. Rate data for the loss of methane from 15 were

obtained by monitoring the protonolysis reaction in CD₃OH (i.e.,

(42) Reaction conditions (Eyring plot for methane elimination from 12

in CD₃OH:CD₂Cl₂(13:1)): [12] ) 21 mM; [H⁺] ) 0.21 M; [Cl^-] ) 0.20

M; µ ) 1.21 M; temperature range -15 to -35 °C.

Mechanism of Aqueous C-H ActiVation by Pt(II)

J. Am. Chem. Soc., Vol. 118, No. 25, 1996 5967

Figure 10. Sequence of ¹H NMR spectra demonstrating the appearance

of new resonances upfield of the original Pt^II-CH₃resonance from 15

along with an overall decrease in the peak integration. The upfield

resonances are assigned as the sequentially deuterated methyl groups:

a ) Pt-CH₃, b ) Pt-CH₂D, and c ) Pt-CHD₂.

Figure 11. Plot of k_obsversus [Cl^-] for deuterium incorporation into

15 in CD₃OD (∆) and for the protonolysis of 15 in CD₃OH ( ).

Conditions for H/D exchange: [Cl^-] ) 0.10-0.50 M; [15] ) 15 mM;

[H⁺] ) 0.15 M; µ ) 1.15 M; temperature ) -23 °C. Conditions for

protonolysis: [Cl^-] ) 0.067-0.78 M; [15] ) 15 mM; [H⁺] ) 0.15

M; µ ) 1.15 M; temperature ) -10 °C.

Scheme 4

mixture of [Pt^IVMeCl₅]^2-and CrCl₂in CD₃OD:CD₃OH (4:1)

from -78 to 0 °C, the reagents dissolve and methane evolution

is observed. The methane generated is composed of a mixture

of CH₄, CH₃D, and CH₂D₂(∼2.6:2.2:1).⁴⁵

Discussion

Relevance of Model Complexes. Based on the principle of

microscopic reversibility, a mechanistic study of the protonolysis

of alkylplatinum(II) could provide insight into C-H activation

by Pt(II). Although we had originally hoped to conduct our

mechanistic studies on the catalytic intermediate [Cl₃Pt^IICH₃]^2-

(1), we encountered a series of complications while studying

this species (see Introduction).^12,26We therefore chose to

investigate the protonolysis of (tmeda)Pt(CH₃)Cl (2). The

relevance of this model is supported by the ability of tmeda-

ligated platinum species to undergo reactions directly analogous

to those proposed for catalytic alkane oxidation (steps I, II, and

III in Scheme 5).¹²Further support derives from the observation

of isotopic exchange prior to liberation of methane when 1 is

generated by reduction of 19 in CD₃OD, behavior closely

resembling that of several of the model systems. Nevertheless,

whenever one investigates model chemistry, there always exists

a certain degree of ambiguity with respect to its relevance to

the “real” system.

Protonolysis of (tmeda)Pt(CH₃)Cl (2) in CD₃OD. Although

the protonolysis of alkylplatinum(II) species has been investi-

gated previously by a variety of other groups, no consensus

has been reached on the detailed mechanism of the reaction.

Only indirect evidence such as kinetic analysis, isotope effects,

and competition experiments was available in these studies until

recently. Protonolysis of 2 in CD₃OD is no exception.

This reaction’s first-order dependence on chloride and proton

concentration (Figures 1 and 2) combined with the non-zero

intercept in Figure 2 supports a two-term rate law (eq 8), namely

in the absence of H/D exchange) at -5 °C in the presence of

0.14 M chloride.⁴³

The rate of H/D exchange exhibits a first-order dependence

on [Cl^-] (Figure 11). Loss of methane from 15 in CD₃OH is

also dependent on chloride concentration (approximately first

order) (Figure 11). In contrast to the chloride dependence for

H/D exchange, methane loss exhibits no measurable rate when

[Cl^-] ) 0 (Figure 11).

Activation parameters were determined for both reactions:

for H/D exchange, ∆H ) 14.5 ((1.5) kcal/mol and ∆S )

-14.8 ((4.5) eu,⁴⁴and for loss of methane, ∆H ) 16.8 ((0.8)

kcal/mol and ∆S ) -12.5 ((2.6) eu.

Preparation of trans-(PEt₃)₂Pt(CD₃)Cl (15-d₃) allowed the

determination of kinetic isotope effects for both H/D exchange

and loss of methane. Both isotope effects obtained for this

system are inVerse: k_H/k_D) 0.80 ((0.05) at -26.8 °C for H/D

exchange, and k_H/k_D) 0.11 ((0.02) at -12.0 °C for methane

loss.

Reduction of [Pt^IVMeCl₅]^2-in Methanol-d_3,4and Water-

d_0,2. Reduction of [Pt^IVMeCl₅]^2-(19) with 2 equiv of CrCl₂in

a 10:1 mixture of D₂O:H₂O at 0 °C generates CH₄and CH₃D

(3.2:3.6), from which a kinetic deuterium isotope effect is

calculated: k_H/k_D) 8.8 ((0.5). Upon warming a heterogeneous

(43) In the absence of added chloride ([Cl^-] < 0.14 M), side reactions

complicate the kinetics. Upon methane loss from 15, the product (PEt₃)₂-

Pt(Cl)(OTf) can abstract chloride from 15 to generate a new phosphine-

ligated methylplatinum(II) species, (PEt₃)₂Pt(CH₃)(OTf) (18), which is much

less reactive toward protonolysis. Small amounts of 18 can be identified in

these reaction mixtures by ¹H NMR spectroscopy, and the spectroscopic

data are consistent with independently prepared samples of 18. These results

will be discussed in further detail in a future paper: Holtcamp, M. W.;

Labinger, J. A.; Bercaw, J. E. Unpublished results.

rate ) k₂[2][H⁺] + k₃[2][H⁺][Cl^-]

(8)

one term independent of [Cl^-] and the other with a first-order

(45) Along with formation of methane, another methylplatinum species

is produced in the reaction mixture as observed by ¹H NMR spectroscopy.

This product appears consistent with [PtMe₂Cl₄]^2-, formed by nucleophilic

attack by Pt(II) (from [PtMeCl₃]^2-, 1) at carbon in [PtMeCl₅]^2-displacing

[PtCl₄]^2-(see ref 26). This reaction competes with the protonolysis of 1.

(44) Reaction conditions (Eyring plot for deuterium incorporation into

15 in CD₃OD): [15] ) 15 mM; [H⁺] ) 0.15 M; [Cl^-] ) 0.14 M; µ ) 1.15

M; temperature range -13 to -40 °C.

5968 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

Scheme 5

desolvation in the entropy of activation. The isotope effects

are also ambiguous. Isotope effects obtained from a competition

experiment in protic solvents are well-known to be strongly

affected by fractionation factors.^49,50This consideration ac-

counts for the discrepancy between the isotope effects obtained

under competitive (i.e., comparing the ratio of methane isoto-

pomers generated in a mixture of CD₃OH and CD₃OD k_H/k_D

) 9.1) versus non-competitive conditions (i.e., comparing the

ratio of protonolysis rate constants measured for the two separate

solvents k_H/k_D) 2.3). Due in part to this complication, these

values do not distinguish between the possible mechanisms.

Protonolysis of Alkylplatinum(II) Species in CD₂Cl₂.

Simply changing the reaction solvent from CD₃OD to CD₂Cl₂

led to an important result: the identification of an alkylhydri-

doplatinum(IV) complex when tmeda was incorporated as a

stabilizing ligand. Such intermediates had not been observed

in alkylplatinum(II) protonolysis studies when we began our

work. Since then, two other groups have identified similar

species by hydrolyzing trialkylsilyl halides (to generate HX in

situ; X ) Cl, Br, I) in solutions of dialkylplatinum(II) complexes

at low temperature. These reports also involved nitrogen-donor

ligands (bipyridyl or phenanthroline ligands) to stabilize the

platinum(IV) species. HCl addition to trans-(PEt₃)₂Pt(CH₃)Cl

(15) (i.e., a complex containing soft ligands) in CD₂Cl₂(eq 6)

also yields the platinum(IV) product at low temperature. These

observations appear to refute earlier proposals, based on an

extensive study of phosphine-ligated complexes (including 15),

that protonolysis of alkylplatinum(II) always proceeds via a

concerted mechanism (specifically, transition state C).²⁸

We next considered the detailed mechanism of alkane

reductive elimination. Nearly all of our mechanistic work

focused on loss of toluene from (tmeda)Pt(CH₂Ph)(H)Cl₂(6).

It was straightforward to demonstrate that oxidative addition

of HCl to alkylplatinum(II) was rapid and reversible on the time

scale of hydrocarbon elimination. The negative entropy of

activation (∆S ) -18.5 ( 7.0 eu) obtained for this reaction

suggests a mechanism other than simple reductive elimination

from 6-coordinate Pt(IV). Reductive elimination reactions often

exhibit near-zero or positive entropies of activation due to the

formation of two products from one reactant. Formation of an

ionic intermediate can lead to a negative entropy of activation

due to significant ordering of the solvent molecules;⁵¹in our

system, chloride dissociation could lead to such an intermediate.

Our first attempt to verify such a mechanism by checking for

chloride inhibition was thwarted, however, because excess

chloride deprotonates 6 to regenerate 5 (eq 4). We did obtain

support for this mechanism by adding HOTf or SnCl₄to a

solution of 6; addition of either reagent greatly accelerates the

rate of toluene elimination.

dependence on [Cl^-]. Various rate laws have been obtained in

related studies, and that shown in eq 8 is not without precedent.

This rate law can be interpreted in several different ways,

however. Initially, this rate law was invoked to support the

intermediacy of an alkylhydridoplatinum(IV) species (intermedi-

ate A) which subsequently reductively eliminates alkane.²⁷

According to this mechanism, the chloride-dependent term in

the rate law reflects stabilization of the Pt(IV) intermediate by

chloride coordination. The chloride-independent term, then,

implies a parallel, solvent-stabilized version of intermediate A

(which is kinetically invisible).

It has been recognized, however, that this rate law is also

consistent with chloride-assisted, direct electrophilic attack at

the Pt-C bond (transition states B or C).²⁸The four-centered

“σ-bond metathesis”-style mechanism (B) has been commonly

invoked in electrophilic attack of alkylmercury(II) and other

species which do not have an accessible higher oxidation

state,^46-48and a recent theoretical study has analyzed this

pathway for C-H activation by platinum(II).³¹Transition state

C was proposed by Alibrandi et al. following an extensive

kinetic study of alkylplatinum(II) species.²⁸

A possible mechanism for the reductive elimination of toluene

from 6 in the presence of HOTf is shown in Scheme 6. This

mechanism suggests that chloride must dissociate from 6 to

generate a five-coordinate cationic intermediate 20 prior to

reductive elimination. Triflic acid facilitates loss of chloride;

Lewis acids such as SnCl₄function similarly. Prerequisite

ligand dissociation has been identified in previous studies of

alkyl hydride reductive elimination,^52-54although this mecha-

nistic feature is by no means universal.^55-58

With respect to the activation of alkanes to generate the

alkylplatinum(II) intermediate, structure A reflects a mechanism

involving alkane oxidative addition/alkylhydridoplatinum(IV)

deprotonation (eq 1), whereas B and C are transition states along

reaction profiles for non-redox processes as in the alternative

(eq 2).

(49) Bell, R. P. The Proton in Chemistry; Cornell University Press:

Ithaca, NY, 1973.

(50) Melander, L.; Saunders, W. H. Reaction Rates of Isotopic Molecules;

Krieger: Malabar, FL, 1987; p 202.

(51) A very large negative entropy of activation (-39 ( 4 eu) has been

obtained for ethane elimination from dimethylpalladium(IV) in which iodide

dissociation has been shown to precede alkane loss: Byers, P. K.; Canty,

A. J.; Crespo, M.; Puddephatt, R. J.; Scott, J. D. Organometallics 1988, 7,

1363.

The activation parameters are of little value for mechanistic

interpretation because of the significant role of ion solvation/

(46) Abraham, M. H. Mechanisms of Electrophilic Substitution at

Saturated Carbon; Amsterdam, 1973.

(47) Kochi, J. K. Organometallic Mechanisms and Catalysis; Academic

Press: New York, 1978; pp 292-340.

(48) Nugent, W. A.; Kochi, J. K. J. Am. Chem. Soc. 1976, 98, 5979.

(52) Milstein, D. J. Am. Chem. Soc. 1982, 104, 5227.

Mechanism of Aqueous C-H ActiVation by Pt(II)

J. Am. Chem. Soc., Vol. 118, No. 25, 1996 5969

Scheme 6

Scheme 7

Figure 12. 1/k_obsvs [HCl] for reductive elimination of toluene from

6 in the presence of varying [HCl] and constant [HOTf]. For reaction

conditions see Figure 5.

formation from 5 + LCl (L ) H, D)), other isotope effects for

Schemes 6 and 7 can be calculated. If Scheme 6 is the correct

mechanism, a kinetic deuterium isotope effect for the reductive

elimination of toluene from 6 is k_H/k_D) 3.1 ((0.6) (KIE_overall

Table 2. Predicted Rate Expressions for Mechanisms Proposed in

Schemes 6 and 7

) EIE_HClox.addn.× KIE_k). This value is similar to those observed

3

in alkane reductive elimination of platinum(II) alkyl hy-

drides.^55,56,59On the other hand, if Scheme 7 is the operating

mechanism, the isotope effect for k₅is simply the observed

overall isotope effect, k_H/k_D) 1.55. This value appears rather

low for a single-step mechanism involving direct proton

transfer.⁶⁰Therefore, based on the kinetic argument described

above and these isotope effect calculations, we favor the

mechanism proposed in Scheme 6.

Scheme 6

Scheme 7

pre-equilibrium k₃K₂[HOTf][6]/[HCl] k₅K₄[HOTf][6]/[HCl]

first step

steady state

k₂k₃[HOTf][6]/

{k₃+ k_-2[HCl]}

k₄k₅[HOTf][6]/

{k_-4[HCl] + k₅[HOTf]}

An alternative mechanism that might also be considered is

shown in Scheme 7. Here 6 is merely generated in an

unproductive side reaction, while alkane formation requires

reductive elimination of HCl from 6 followed by direct

electrophilic attack by H⁺at the platinum alkyl bond of 5.

Indeed, both schemes look very similar kinetically, and the

derived rate expressions have been outlined in Table 2. Both

pre-equilibrium and steady-state situations have been considered.

Because full equilibration between 6 and 5 was demonstrated

by isotopic labeling (Scheme 2), the “steady-state” option for

Scheme 7 is eliminated. The other three mechanistic possibili-

ties were distinguished kinetically by evaluating the effect of

[HCl] on the reaction rate with [HOTf] held constant. As

demonstrated in Figure 5, the reaction is inhibited by HCl.

Figure 12 shows the same data, presented as 1/k_obsvs [HCl].

The non-zero intercept in this plot supports the mechanism in

Scheme 6 with steady-state kinetic behavior. Both pre-

equilibrium rate laws predict that a plot of 1/k_obsvs 1/[HCl]

would go through the origin. According to this mechanism,

the negative entropy of activation would likely arise from ion

solvation effects.

Stereochemistry of HCl Addition. The spectroscopic data

obtained for 4 and 6 (Table 1) are unable to distinguish between

trans versus cis addition of HCl to 2 and 5, respectively. In

theory, four possible stereoisomers could arise from this reaction

(structures D-G). Trans addition of HCl would lead to

structure D, whereas cis addition could produce any of the

structures D-G. Of course, any of the structures are also

accessible via isomerization during or after the addition of HCl.

Compound 6 contains four spectroscopically distinct methyl

groups on the tmeda ligand (Table 1), consistent with structures

D and F.

Evaluating HCl addition to related dialkylplatinum(II) species

provided further insight. There are still four possible products

however (structures H-K). In the absence of isomerization,

Isotope effects determined for this system (using calibrated

mixtures of HCl and DCl) also support the mechanism in

Scheme 6. Based on the isotope effects that were determined

in this system (K_H/K_D) 0.51 for pre-equilibrium oxidative

addition of LCl to 5, and k_H/k_D) 1.55 for overall toluene

(53) Basato, M.; Morandini, F.; Longato, B.; Bresadola, S. Inorg. Chem.

1984, 23, 649.

(54) Basato, M.; Morandini, F.; Bresadola, S. Inorg. Chem. 1984, 23,

3972.

(55) Abis, L.; Sen, A.; Halpern, J. J. Am. Chem. Soc. 1978, 100, 2915.

(56) Michelin, R. A.; Faglia, S.; Uguagliati, P. Inorg. Chem. 1983, 22,

1831.

(57) Buchanan, J. M.; Stryker, J. M.; Bergman, R. G. J. Am. Chem. Soc.

1986, 108, 1537.

structure H results from trans addition of HCl, I and J from

(59) Hackett, M.; Ibers, J. A.; Whitesides, G. M. J. Am. Chem. Soc. 1988,

110, 1436.

(60) Moore, J. W.; Pearson, R. G. Kinetics and Mechanism, 3rd ed.;

John Wiley & Sons: New York, 1981; p 367.

(58) Jones, W. D.; Feher, F. J. Acc. Chem. Res. 1989, 22, 91.

5970 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

Figure 13. Qualitative reaction coordinate diagram for the reaction of HCl with 2 and 11. While the barrier to reductive elimination from 12

(∆G₁₂) is greater than from 4 (∆G₄), the overall barrier to the reaction of HCl with 11 is lower than with 2 (∆∆G ).

cis addition. The remaining stereoisomer, K, can only arise

from isomerization following HCl addition. The ¹H NMR

spectrum of (tmeda)Pt(CH₂Ph)₂(H)Cl (7) exhibits two inequiva-

lent resonances for its tmeda methyl protons (Table 1). This

spectrum is consistent with either structure H or K. Protonation

of (4,4′-dimethyl-2,2′-bipyridyl)Pt(CH₃)₂(9) leads to a dialkyl-

hydridoplatinum(IV) complex (10) with only a single resonance

(tmeda)Pt(CH₃)(H)Cl₂(4) and (tmeda)Pt(CH₃)₂(H)Cl (12), prior

to reductive elimination. The relative stabilities of 4 and 12

with respect to reductive elimination indicate that ∆G₁₂

>

∆G₄. However as long as the transition state for reductive

elimination from 12 is lower in energy than the transition state

for elimination from 4, the reaction will selectively generate

(tmeda)Pt(CH₃)Cl until either HCl or (tmeda)Pt(CH₃)₂is

consumed.

1

in the H NMR spectrum corresponding to the methyls on the

bipyridyl ligand (Table 1). This result is only consistent with

the trans HCl addition product H, a conclusion also reached

by Puddephatt and co-workers in an identical reaction between

HCl and (4,4′-di-tert-butyl-2,2′-bipyridyl)Pt(CH₃)₂.³³By con-

trast, Panunzi and co-workers have identified structure J as the

product arising from HX (X ) Cl, Br) addition to (dmphen)-

Pt(CH₃)₂(dmphen ) 2,9-dimethyl-1,10-phenanthroline) in

acetone/diethyl ether solution.³²As noted by both groups, this

product geometry is likely due to the steric constraints of the

dmphen ligand used in this reation. Nevertheless, by analogy

to the products of HCl addition to dialkylplatinum(II) species,

7 and 9, we believe the addition proceeds with trans stereo-

chemistry. The spectroscopic data for (PEt₃)₂Pt(CH₃)(H)Cl₂

(16) are also consistent with trans HCl addition to 15, although

this has not been verified by experiments with related dialkyl

complexes.

Stability of Dialkylhydridoplatinum(IV) versus Monoalkyl-

hydridoplatinum(IV): A Digressionary Anomaly. We have

previously reported that preparation of (tmeda)Pt(CH₃)Cl (2)

is readily accomplished in high yield by addition of 1 equiv of

HCl to (tmeda)Pt(CH₃)₂(11) in CH₂Cl₂(Scheme 5). Based on

the decreased stability of (tmeda)Pt(CH₃)(H)Cl₂(4) relative to

(tmeda)Pt(CH₃)₂(H)Cl (12), we might have expected that once

2 begins forming in solution, it would be consumed more rapidly

than 11, leaving a 1:1 mixture of 11 and (tmeda)PtCl₂. Because

mixing equal amounts of 11 and (tmeda)PtCl₂in CD₂Cl₂

revealed only a minor amount of comproportionation after 24

h,⁵³this is not a kinetically competent mechanism to account

for the clean formation of 2 under preparative conditions.

This apparently anomalous result is perhaps best explained

by the energy diagram in Figure 13. We demonstrated above

that protonation of the alkylplatinum(II) complexes is rapid and

reversible. In other words, a mixture containing (tmeda)Pt-

(CH₃)Cl, (tmeda)Pt(CH₃)₂, and HCl in CH₂Cl₂will rapidly

equilibrate between both alkylhydridoplatinum(IV) species,

Protonolysis of (tmeda)Pt(CH₃)₂(11) in CD₃OD. Because

the protonolysis of (tmeda)Pt(CH₃)Cl (2) in methanol reveals

no intermediate prior to loss of methane, whereas an alkylhy-

dridoplatinum(IV) intermediate is observed in CD₂Cl₂, we must

consider whether changing the solvent changes the protonolysis

mechanism. However, observation of the dialkylhydridoplati-

num(IV) species (tmeda)Pt(CH₃)₂(H)Cl (12) upon treating

(tmeda)Pt(CH₃)₂(11) with HCl in methanol verifies the viability

of Pt(IV) intermediates in polar/protic solvent. Not only is a

platinum(IV) intermediate observed, but deuterium from CD₃-

OD is incorporated into the platinum-bound methyl groups prior

to reductive elimination of methane, implicating the potential

involvement of an alkane σ-complex.

We propose the mechanism shown in Scheme 8 for proto-

nolysis of 11 in methanol. The fundamental steps involve (1)

pre-equilibrium oxidative addition of HCl to generate 12, (2)

dissociation of chloride to produce a cationic, five-coordinate

intermediate (21), (3) reductive C-H bond formation leading

to the Pt(II) methane σ-complex (22) (the nature of this

interaction will be discussed below), and finally, (4) unimo-

lecular dissociation of alkane resulting in the cationic methyl-

platinum(II) species (14). Trapping of 14 by chloride to form

2 results in the immediate liberation of the second equivalent

of methane. (2 has already been shown to be unstable under

these reaction conditions.) Small to significant concentrations

1

of 14 can be observed by H NMR, depending on the initial

chloride concentration.

According to this proposal, methane dissociation is the rate

determining step; all preceding steps can fully equilibrate prior

to loss of methane. In order for H/D exchange to take place,

the σ-complex (22) must rearrange to coordinate a different

C-H(D) bond prior to reversion to 21.

Recently, significant evidence has been provided for alkane

σ-complexes mediating the solution-phase reductive elimination

and oxidative addition of alkanes at transition metal centers.^61-73

Mechanism of Aqueous C-H ActiVation by Pt(II)

J. Am. Chem. Soc., Vol. 118, No. 25, 1996 5971

Scheme 8

temperature matrix studies.^90,91Furthermore, analogous η²-

dihydrogen and silane complexes along with 3-center-2-electron

agostic C-H-M interactions⁹²are now well-known and many

have been structurally characterized.

Deuterium scrambling between hydride and alkyl positions

prior to alkane elimination has been observed previously in other

transition metal complexes.^64,65,67-73In some cases the ex-

change proceeds rapidly enough to allow full equilibration of a

single deuterium atom between the two different sites prior to

methane loss. Equilibrium isotope effects determined in such

cases reveal a preference for deuterium to reside in the alkyl

site, as expected.^65,67,69,71In our case, such equilibration leads

to complete isotopic exchange because the platinum hydride

also exchanges with the large excess of deuterium in the solvent.

Complete proton incorporation into 12-d₆is also observed when

the reaction is carried out in CD₃OH. That H/D exchange takes

place intramolecularly was verified by monitoring deuterium

incorporation into 12 under approximately 15 atm of CH₄. No

evidence for incorporation of CH₄was obtained; loss of CL₄in

Scheme 8 is irreversible.

Although such isotopic exchange has been invoked to support

the intermediacy of σ-complexes in the reductive elimination

of alkanes, it is often difficult to definitively exclude alternative

mechanistic possibilities. Such is the case here, too. The

mechanism for H/D exchange need not lie on the reaction

coordinate for reductive elimination. For example, it might

proceed by formation of a carbene intermediate generated via

R elimination or via deprotonation by an external base (e.g.,

solvent) (eqs 9 and 10, respectively). Reversal of either reaction

could account for H/D exchange: in the former case, exchange

of either platinum hydride with solvent deuterons followed by

carbene insertion, or in the later case, deuteration of the platinum

carbene by solvent deuterons.

For example, Bergman, Moore, and co-workers have obtained

transient IR spectroscopic data directly supporting such an

intermediate in the oxidative addition of alkanes at a Cp*Rh-

(CO) fragment.^61-63The viability of alkane adducts has

received additional support from a variety of other sources:

theory^74-78and gas phase,^79-84condensed phase,^85-89and low-

(61) Schultz, R. H.; Bengali, A. A.; Tauber, M. J.; Weiller, B. H.;

Wasserman, E. P.; Kyle, K. R.; Moore, C. B.; Bergman, R. G. J. Am. Chem.

Soc. 1994, 116, 7369.

(62) Bengali, A. A.; Schultz, R. H.; Moore, C. B.; Bergman, R. G. J.

Am. Chem. Soc. 1994, 116, 9585.

(63) Bengali, A. A.; Arndtsen, B. A.; Burger, P. M.; Schultz, R. H.;

Weiller, B. H.; Kyle, K. R.; Moore, C. B.; Bergman, R. G. Pure Appl.

Chem. 1995, 67, 281.

(64) Buchanan, J. M.; Stryker, J. M.; Bergman, R. G. J. Am. Chem. Soc.

1986, 108, 1537.

(65) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc. 1986, 108, 7332.

(66) Jones, W. D.; Feher, F. J. J. Am. Chem. Soc. 1986, 108, 4814.

(67) Gould, G. L.; Heinekey, D. M. J. Am. Chem. Soc. 1989, 111, 5502.

(68) Bullock, R. M.; Headford, C. E. L.; Kegley, S. E.; Norton, J. R. J.

Am. Chem. Soc. 1985, 107, 727.

(69) Bullock, R. M.; Headford, C. E. L.; Hennessy, K. M.; Kegley, S.

E.; Norton, J. R. J. Am. Chem. Soc. 1989, 111, 3897.

(70) Parkin, G.; Bercaw, J. E. Organometallics 1989, 8, 1172.

(71) Wang, C.; Ziller, J. W.; Flood, T. C. J. Am. Chem. Soc. 1995, 117,

1647.

(72) Mobley, T. A.; Schade, C.; Bergman, R. G. J. Am. Chem. Soc. 1995,

117, 7822.

The magnitude of the normal isotope effect that we obtained

for H/D exchange (k_H/k_D) 1.9 ((0.2)) appears unable to

accommodate either carbene mechanism which should lead to

inVerse isotope effects. (The predicted inverse nature of these

isotope effects arises solely from our definition of k_Hand k_D.

Both eqs 9 and 10 for H/D exchange involving a carbene

intermediate should lead to faster rates for deuterium incorpora-

tion into -CH₃(k_D) versus proton incorporation into -CD₃

(k_H).)

The inverse dependence of the rates of H/D exchange and

methane elimination on chloride concentration (Figures 8

and 9) implies that chloride must dissociate from 12 prior to

further reaction. A similar observation (involving iodide disso-

(73) Green, M. L. H. Personal communication.

(74) Cundari, T. R. J. Am. Chem. Soc. 1992, 114, 10557.

(75) Cundari, T. R. J. Am. Chem. Soc. 1994, 116, 340.

(76) Low, J. J.; Goddard, W. A. Organometallics 1986, 5, 609.

(77) Song, J.; Hall, M. B. Organometallics 1993, 12, 3118.

(78) Ziegler, T.; Tschinke, V.; Fan, L.; Becke, A. D. J. Am. Chem. Soc.

1989, 111, 9177.

(79) Brown, C. E.; Ishikawa, Y.; Hackett, P. A.; Rayner, D. M. J. Am.

Chem. Soc. 1990, 112, 2530.

(80) Carroll, J. J.; Weisshaar, J. C. J. Am. Chem. Soc. 1993, 115, 800.

(81) Hop, C. E. C. A.; McMahon, T. B. J. Am. Chem. Soc. 1991, 113,

355.

(82) Ishikawa, Y.; Brown, C. E.; Hackett, P. A.; Rayner, D. M. Chem.

Phys. Lett. 1988, 150, 506.

(83) Schroder, D.; Fiedler, A.; Hrusak, J.; Schwarz, H. J. Am. Chem.

Soc. 1992, 114, 1215.

(84) Wasserman, E. P.; Moore, C. B.; Bergman, R. G. Science 1992,

255, 315.

(87) Klassen, J. K.; Selke, M.; Sorensen, A. A. J. Am. Chem. Soc. 1990,

112, 1267.

(88) Simon, J. D.; Xie, X. J. Phys. Chem. 1989, 93, 291.

(89) Yang, C. K.; Peters, K. S.; Vaida, V. Chem. Phys. Lett. 1986, 125,

566.

(85) Dobson, G. R.; Hodges, P. M.; Healy, M. A.; Poliakoff, M.; Turner,

J. J.; Firth, S.; Asali, K. J. J. Am. Chem. Soc. 1987, 109, 4218.

(86) Dobson, G. R.; Asali, K. J.; Cate, C. D.; Cate, C. W. Inorg. Chem.

1991, 30, 4471.

(90) Billups, W. E.; Chang, S.-C.; Hauge, R. H.; Margrave, J. L. J. Am.

Chem. Soc. 1993, 115, 2039.

(91) Perutz, R. N.; Turner, J. J. J. Am. Chem. Soc. 1975, 97, 4791.

(92) Crabtree, R. H. Angew. Chem., Int. Ed. Engl. 1993, 32, 789.

5972 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

Table 3. Deuterium Kinetic Isotope Effects Measured for the

Reductive Elimination of Alkanes from Various Alkylhydrido

Transition Metal Complexes

Scheme 9

temp isotope

(°C) effect

entry

compd

ref

Normal Isotope Effects

1

2

3

4

5

cis-(PPh₃)₂Pt(Me)(H)

-25

-28

40

3.3

3.1

2.2

1.5

1.3

55

this work

56

59

52

(tmeda)Pt(CH₂Ph)(H)Cl₂

cis-(PPh₃)₂Pt(CH₂CF₃)(H)

(P-P)Pt(CH₂CMe₃)(H)^a

mer-(PMe₃)₃Rh(CH₂COCH₃)(H)Cl

ciation) has been made by Canty and Puddephatt et al. and

Goldberg et al. when investigating the reductive elimination of

ethane from trimethyl palladium(IV) and -platinum(IV) species,

respectively.^51,93-96Canty and Puddephatt suggest that upon

iodide dissociation, an agostic interaction might develop between

one of the methyl C-H bonds and the coordinatively unsaturated

palladium center. This interaction could stabilize the redirection

of a carbon sp³orbital away from the metal center and toward

the second methyl group, ultimately leading to C-C bond

formation. In our case, an agostic interaction could stabilize

the C-H bond formation. The intermediate alkane σ-complex

then might involve an η³(H-C-H) interaction with the metal

center (25_σ) (Scheme 9). Such an alkane binding mode was

evaluated in theoretical work by Low and Goddard as an

intermediate in the reductive elimination of methane from Pt-

(II).⁷⁶With such an intermediate, H/D exchange does not

require that the bound alkane rearrange to coordinate a different

C-H(D) bond as the oxidative C-H bond cleavage (25_σ

24) may select either coordinated bond. It should perhaps be

noted that 25_σcan be considered a resonance structure of a

carbene complex (25_c). Unfortunately, we have no direct data

that allow us to evaluate the nature of this intermediate further.

69

31

Inverse Isotope Effects

[Cp₂Re(CH₃)(H)]⁺Cl^-

6

7

8

9

75

0.8

0.74 71

0.7

0.51 66

0.5 65

0.29 this work

67

(Cn)Rh(PMe₃)(CH₃)(H)

Cp*Ir(PMe₃)(Cy)(H)

Cp₂W(CH₃)(H)

130

73

100

51

-30

-27

64

68, 69

70

10 Cp*₂W(CH₃)(H)

11 Cp*Rh(PMe₃)(3,5-C₆H₃Me₂)(H)

12 Cp*Rh(PMe₃)(Et)(H)

13 (tmeda)Pt(CH₃)₂(H)Cl

^aP-P ) bis(dicyclohexylphosphino)ethane

only do the values vary widely (some of this variation is of

course due to differences in temperature), but they do not appear

to correlate directly with the relative exothermicity of the

reaction. The lack of correlation between these values suggests

that reaction exothermicity alone cannot explain the discrepancy

in isotope effects. Interestingly, for all reported cases of inverse

isotope effects (entries 6-13), H/D exchange is observed

between the alkyl (aryl) and hydride positions, constituting

independent evidence for the presence of an intermediate

preceding alkane loss. Thus, while H/D exchange does not

necessarily lie on the reaction coordinate for reductive elimina-

tion, the data in Table 3 provide substantial support for such a

link. It is still possible that σ-complexes mediate the reductive

elimination of reactions which reveal normal isotope effects,

but in these cases, the rate-determining step would involve

formation of the σ-complex rather than dissociation of alkane

from the metal center. That is, the transition state would lie

much earlier along the reaction coordinate for C-H bond

formation.

We therefore believe that H/D exchange in (tmeda)Pt(CH₃)₂-

(H)Cl (12) supports the presence of a methane σ-complex that

lies on the reaction coordinate for reductive elimination.

Without an intermediate in this reaction, the inverse isotope

effect would be very difficult to explain. The isotope effect

for methane loss, then, is a composite value arising from an

equilibrium isotope effect (reversible σ-complex formation) and

a kinetic isotope effect (dissociation of alkane). Lack of direct

observation of the σ-complex prevents us from deconvoluting

this value.

The isotope effect for methane loss from (tmeda)Pt(CH₃)₂-

(H)Cl (12) is substantially inverse (k_H/k_D) 0.29 ((0.05)).

Inverse isotope effects have been reported by a number of groups

who investigated the reductive elimination of alkanes from

transition-metal centers; however, this is the smallest value

obtained so far.^{64,65,67,69-71}As has been discussed thoroughly

in several of these references, inverse isotope effects often

suggest the presence of an unobserved equilibrium preceding

the rate-determining step. A single-step mechanism may also

produce an inverse isotope effect;^97,98however, one would

expect such a value to arise only from a highly endothermic

reaction in which the transition state is very “product-like,”

namely, one in which a considerable amount of C-H bond

formation has already taken place.

Several instances of normal isotope effects (including the one

discussed above: (tmeda)Pt(CH₂Ph)(H)Cl₂(5) in CD₂Cl₂, k_H/

k_D) 3.1) for C-H reductive elimination have also been

reported.^52,55,56,59The data in Table 3 compare all the isotope

effects that we have found for C-H reductive elimination; the

table is arranged in the order of decreasing isotope effect. Not

Protonolysis of trans-(PEt₃)₂Pt(CH₃)Cl (15) in CD₃OD.

We initially chose to evaluate the protonolysis of trans-(PEt₃)₂-

Pt(CH₃)Cl (15) because of the inverse isotope effect reported

by Romeo et al.⁹⁹Furthermore, 15 has been the subject of

several protonolysis studies in the past,^27,28,99and we wanted

to compare its reactivity directly with our tmeda-ligated platinum

complexes. Previous investigations utilized UV-visible spec-

troscopy to monitor the reaction, and clean isosbestic points

suggested no intermediates were present in the reaction.

However, upon investigating the protonolysis of 15 in CD₃OD

(93) Byers, P. K.; Canty, A. J.; Crespo, M.; Puddephatt, R. J.; Scott, J.

D. Organometallics 1988, 7, 1363.

(94) Canty, A. J. Acc. Chem. Res. 1992, 25, 83.

(95) Goldberg, K. I.; Yan, J. Y.; Winter, E. L. J. Am. Chem. Soc. 1994,

116, 1573.

(96) Goldberg, K. I.; Yan, J. Y.; Breitung, E. M. J. Am. Chem. Soc.

1995, 117, 6889.

1

by H NMR, we observed deuterium incorporation into the

methyl sites prior to methane loss (Scheme 4). Since there is

no buildup of the intermediate responsible for this exchange, it

would be invisible in the electronic spectrum. In contrast to

(97) Bigeleisen, J. Pure Appl. Chem. 1964, 8, 217.

(98) Melander, L. Acta Chem. Scand. 1971, 25, 3821.

(99) Romeo, R.; Minniti, D.; Lanza, S.; Uguagliati, P.; Belluco, U. Inorg.

Chem. 1978, 17, 2813.

Mechanism of Aqueous C-H ActiVation by Pt(II)

J. Am. Chem. Soc., Vol. 118, No. 25, 1996 5973

Scheme 10

Scheme 11

the protonolysis of (tmeda)Pt(CH₃)₂(11), no platinum(IV)

intermediate is observed; 15 and its isotopomers are the only

species observed prior to elimination. We propose that deute-

rium incorporation results from the formation of an alkane

σ-complex prior to methane dissociation.

Protonolysis of 15 might proceed via direct protonation at

the platinum-carbon bond (see structures B and C above), or

alternatively, it may proceed through a mechanism analogous

to Scheme 8. In the latter case, the platinum(IV) intermediate

would not be observed if it is less stable than the starting

alkylplatinum(II) complex 15. The former conclusion was

favored by Alibrandi et al. in their extensive study of a series

of alkylplatinum(II) species.²⁸They propose a mechanism

involving direct protonolysis of 15 through transition state C

involving both chloride-dependent (k_Cl) and chloride-indepen-

dent (k_H) pathways (Scheme 10). Nevertheless, we believe that

a mechanism similar to Scheme 8, involving a phosphine-ligated

alkylhydridoplatinum(IV) intermediate (16), is at least as

consistent with the data, and also contributes to a more unified

picture for the protonolysis of alkylplatinum(II) species (see

below). Indeed, such a mechanism was proposed by Belluco

and co-workers in their first study of the protonolysis of 15.²⁷

The proposed platinum(IV) intermediate, (PEt₃)₂Pt(CH₃)(H)Cl₂

(16), can in fact be observed in CD₂Cl₂. Although 16 appears

to be less stable than (tmeda)Pt(CH₃)₂(H)Cl (12) (eliminating

methane at -45 versus +10 °C, respectively, in CD₂Cl₂), its

formation suggests that it might be an energetically viable (albeit

unobserved) intermediate in methanol.

Unfortunately, the kinetic isotope effects are of little utility

in identifying the correct mechanism. Because the reagents and

the transition state leading to H/D exchange are virtually

identical for the two mechanisms (reductive elimination forming

a σ-complex versus direct electrophilic attack at the Pt-C bond),

the isotope effect for H/D exchange (k_H/k_D) 0.80) is unable to

distinguish between them. The isotope effect for methane loss

is even more inverse (k_H/k_D) 0.11); this result merely suggests

that formation of a strong C-H bond is nearly complete in this

transition state.

to generate an alkylhydridoplatinum(IV) intermediate (W), (2)

dissociation of solvent or chloride to generate a cationic, five-

coordinate platinum(IV) species (X), (3) reductive C-H bond

formation producing an alkane σ-complex (Y), and (4) loss of

alkane either through an associative or dissociative substitution

pathway. Based on this proposal, the origin of the differences

between the four systems lies in the fact that changing the

solvent or the initial Pt(II) species alters the relative energy of

the reagents, intermediates, and transition states, thereby chang-

ing the rate-determining step and/or dictating which intermedi-

ates may be observed.

This proposal also accounts for the differences in the chloride

dependency of the reactions. Both the protonolysis of 2 in

methanol and H/D exchange in 15 show positive dependence

of the reaction rates on [Cl^-] (see Figures 2 and 11); however,

the reactions still proceed with measurable rate in the absence

of chloride. These results reflect the fact that formation of W,

X, and Y may be mediated by either solvent or chloride, with

chloride being a more effective mediator for these reactions than

solvent.

On the other hand, loss of methane from intermediate Y in

the protonolysis of 15 exhibits no measurable rate in the absence

of chloride (Figure 11). This result suggests that alkane loss

occurs via an associatiVe chloride substitution pathway.

The reactions starting from an alkylhydridoplatinum(IV)

intermediate (6 in CD₂Cl₂and 12 in methanol) exhibit an inVerse

dependence of the reaction rates on [Cl^-]. This result is most

easily explained by pre-equilibrium chloride dissociation from

W_Clpreceding the rate-determining step (see above). Because

both H/D exchange and methane loss from 12 exhibit this

inverse chloride dependence, we believe this reflects a dissocia-

tiVe substitution pathway, since an associative pathway, as

observed for 15, should cancel the chloride dependence.

Shown in Figures 14a-d are qualitative reaction coordinate

Analysis: Common Mechanism? In this work we have

evaluated the reactivity of four different systems: [1] (tmeda)-

Pt(CH₃)Cl (2) in methanol, [2] (tmeda)PtRCl (R ) CH₃, CH₂-

Ph (5)) in dichloromethane, [3] (tmeda)Pt(CH₃)₂(11) in

methanol, and [4] trans-(PEt₃)₂Pt(CH₃)Cl (15) in methanol.

Despite the significant differences among the systems, we

believe all of the data support a common mechanism (Scheme

11): (1) chloride- or solvent-mediated protonation of Pt(II) (V)

5974 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

into the methyl groups and the inverse isotope effect for methane

loss (see above). These results are shown schematically in

Figure 14c.

The protonolysis of (PEt₃)₂Pt(CH₃)Cl (15) also exhibits some

unique features. Again we believe the data are consistent with

the mechanism proposed in Scheme 11 with the differences

arising from the relative energy of the reagents, intermediates,

and transition states (Figure 14d). As with the protonolysis of

11, the rate-determining step is loss of alkane from the

σ-complex (Y). In this case, however, no Pt(IV) intermediate

(W) is observed; 15 is the species observed under the reaction

conditions. This feature likely arises from the substitution of

relatively “hard” amine donors (i.e., tmeda) with “soft” phos-

phine ligands. Soft ligands are commonly used to stabilize

lower oxidation states. The relative energies of intermediates

W, X, and Y remain unknown.

Despite the data that we have obtained for the four different

systems, several detailed features of this mechanism remain

uncertain. For example, although chloride loss precedes alkane

reductive elimination, our kinetic data cannot distinguish

between elimination from solvated six-coordinate and five-

coordinate cationic intermediates. Furthermore, many aspects

of the HCl oxidative addition mechanism remain unanswered.

Although the reaction appears to proceed with trans stereo-

chemistry (suggesting a heterolytic mechanism), it is not known

whether the mechanism involves initial proton attack at Pt(II)

followed by chloride association or the reverse order of addition.

In the latter case, chloride or solvent are in fact necessary in

order to generate the reactive five-coordinate intermediate (X).

In the former case, however, X is rapidly trapped in a non-

productive equilibrium by chloride or solvent. It must then be

regenerated in a subsequent step prior to alkane elimination.

Another area of uncertainty concerns the nature of the alkane

σ-complex, because of our inability to observe this intermediate.

Figure 14. Qualitative reaction coordinate diagrams (based on Scheme

11) for the protonolysis reactions of (a) (tmeda)PtMeCl (2) in methanol,

(b) (tmeda)PtMeCl (2) in dichloromethane, (c) (tmeda)PtMe₂(11) in

methanol, and (d) trans-(PEt₃)₂PtMeCl (15) in methanol. (See discussion

in text.)

diagrams for each of the four systems analyzed. The diagrams,

based on Scheme 11, account for the different observations made

in each case. For the protonolysis of 2 in methanol, we obtain

no direct evidence for intermediates in the reaction. Neverthe-

less, the data are still consistent with the mechanism proposed

in Scheme 11 (Figure 14a). Although many specific features

of the diagram remain speculative, the important details are the

following: (1) the reagents are significantly more stable than

any other species preceding the rate-determining step (i.e., only

2 is observed prior to product formation) and (2) the rate-

determining step lies before dissociation of alkane from Y.

Based on the [Cl^-] dependence of the reaction, we assign

formation of W as the rate-determining step; such an energy

profile reveals why no intermediates can be observed. It is also

consistent with the observation that 14 is less susceptible to

protonolysis than 2, as generation of W from a cationic V should

be more difficult.

Conclusion: Implications for Alkane Activation by Pt(II)

We began this study hoping to gain insights into the

mechanism of aqueous C-H activation by Pt(II), the micro-

scopic reverse of the reactions discussed here. Based on our

results, we believe we can now speculate further on some of

the detailed aspects of this reaction. The mechanism for

protonolysis of alkylplatinum(II) complexes involves a signifi-

cant number of ligand association, dissociation, and substitution

steps, and furthermore, the reaction proceeds through both ionic

and neutral intermediates (Scheme 11). These features suggest

an important role for the solvent (water) in Shilov’s alkane

oxidation system: it must be able to accommodate neutral and/

or ionic intermediates in order to facilitate the various mecha-

nistic steps. Significant stabilization or destabilization of any

of the reaction intermediates relative to the others could

dramatically decrease the rate of C-H activation. Interestingly,

nearly all of the “electrophilic” alkane oxidation systems utilize

highly polar solvents (H₂O, H₂SO₄, CF₃COOH).^4-7This may

be rationalized at least in part by comparing Figures 14a and

14b reading right-to-left destabilization of ionic species in

nonpolar solvent (Figure 14b) translates to a higher barrier for

C-H activation.

Upon carrying out the same reaction in dichloromethane

(rather than methanol), significant differences are observed. In

this case alkylhydridoplatinum(IV) (W) is the most stable

species under the reaction conditions. Although W is in rapid

equilibrium with V prior to loss of alkane, it is the only species

observed aside from the products (Z); Figure 14b reflects this

change. The differences between Figures 14a and 14b can be

accounted for by the change in solvent polarity, i.e., methanol

is better able to stabilize the charged reagents and intermediates.

Consequently, upon shifting to dichloromethane, the charged

intermediate (X) and the preceding transition state become much

less stable, thus making its formation the rate-determining step.

Protonolysis of (tmeda)Pt(CH₃)₂(11) in methanol differs

significantly from that of (tmeda)Pt(CH₃)Cl (2). Although the

only difference is replacement of chloride with a more electron-

donating methyl group in the platinum coordination sphere, this

change is apparently sufficient to stabilize Pt(IV) (W) relative

The nature of the platinum ligands also appears crucial; they

must be sufficiently labile to allow the various ligand exchange

reactions to proceed at rapid rates. The initial interaction

between the alkane and Pt(II) involves a ligand-substitution

reaction (alkane for chloride or solvent). Although we cannot

address whether the substitution takes place via an associative

or dissociative mechanism, this step is likely the origin of

chloride inhibition of alkane oxidation. In other words, chloride

1

to Pt(II) (V), because W becomes the species observed by H

NMR at low temperature. The other significant feature of this

reaction is that alkane dissociation from the σ-complex (Y) is

now the rate-determining step. Although Y is never seen

directly, its existence is supported by deuterium incorporation

Mechanism of Aqueous C-H ActiVation by Pt(II)

J. Am. Chem. Soc., Vol. 118, No. 25, 1996 5975

intervals. Spectra were then processed and integrated using WIN-

NMR® software; the Pt-Me peak was integrated relative to the tBuOH

resonance. Data analysis was carried out using KaleidaGraph®.

Kinetic Isotope Effects for the Protonolysis of (tmeda)PtMeCl

(2) in CD₃OD. See previous section for reaction details. The

competitive isotope effect was measured by comparing the integration

of CH₃D to CH₄generated from the reaction carried out in a mixture

of CD₃OD and CD₃OH. The non-competitive isotope effect was

determined by comparing the rates of the reaction carried out in CD₃-

OH and CD₃OD.

competes favorably with alkane for a coordination site on

Pt(II); that alkane can compete at all still seems quite remark-

able. We cannot exclude the possibility that chloride plays a

more intimate role in these processes as well, perhaps along

the lines of structure B above as suggested by theoretical

studies.³¹

Our results implicate the presence of both alkane σ-complexes

and alkylhydridoplatinum(IV) as intermediates in the C-H

activation reaction.¹⁰⁰Contrary to our previous suggestion,⁹it

appears that the alkylplatinum(II) species is generated by

deprotonation of alkylhydridoplatinum(IV) (eq 1), not the

σ-complex (eq 2). According to this proposal, the σ-adduct is

an intermediate preceding the oxidative addition of the C-H

bond. Such a mechanism is presumably facilitated by the

accessibility of the Pt(IV) oxidation state. The other electro-

philic alkane oxidation systems (such as those involving

palladium or mercury) likely operate through a different pathway

considering the significant instability of their M(IV) oxidation

states.

Preparation of Alkylhydridoplatinum(IV) species (4, 6, 8, 10, 12,

16) in CD₂Cl₂. HCl(g) was dissolved in Et₂O-d₁₀(using a gas bulb

containing a known amount of HCl) and the solution was then titrated

to determine the HCl concentration. (An aliquot of the solution was

added to a few milliters of water containing phenolphthalein, and then

titrated against a 0.100 N solution of NaOH.) Approximately 10 equiv

of the HCl solution were added via syringe to a 5 mm NMR tube

containing the appropriate alkylplatinum(II) reagent dissolved in CD₂-

Cl₂at -78 °C. After mixing the contents while keeping the tube as

cold as possible, the tube was placed in the precooled NMR probe.

1

Relevant H NMR data are recorded in Table 1.

¹H NMR Kinetics for the Reductive Elimination of Toluene from

6 in CD₂Cl₂. HCl/Et₂O-d₁₀(60 µL, 0.76 M) was added to a solution

of 5 (4 mg, 0.0091 mmol) in CD₂Cl₂at -78 °C. Pentachloroethane

(4 µL) was also added as an integration standard. Sequential ¹H NMR

spectra for the reductive elimination of toluene were obtained using

the preprogrammed subroutine discussed above for the protonolysis of

2 in CD₃OD. The furthest downfield aromatic protons of 6 (δ 7.42

ppm , br s, 2 H) were integrated to obtain the kinetics data. A solution

of HOTf was also prepared in Et₂O-d₁₀in order to monitor the [HOTf]

dependence of the reaction; however, this solution had to be kept cold

and used shortly after preparation to avoid its rapid decomposition at

room temperature. A typical kinetics run involved adding HCl/Et₂O-

d₁₀(25 µL, 1.88 M) to 5 (4 mg, 0.0091 mmol) dissolved in CD₂Cl₂

(0.5 mL) at -78 °C. After thoroughly mixing the solution at low

temperature, HOTf/Et₂O-d₁₀(43 µL, 1.07 M) was added. The solution

was again thoroughly mixed prior to placement in the precooled probe.

To check the effect of a nonprotic Lewis acid, a solution of SnCl₄in

CD₂Cl₂was prepared and was similarly added to a pre-formed solution

of 6.

Clearly these systems are highly complex, involving a number

of intermediates. To date we have significant, but incomplete,

understanding of how their relative stabilities and reactivities

depend upon structure and reaction conditions. Studies continue

toward our long-range goal: to exploit detailed understanding

of the mechanism and energetics of electrophilic alkane oxida-

tion for the rational design of improved catalysts.

Experimental Section

General Considerations. Protonolysis reactions involving 5 and

11 were carried out under an inert atmosphere using standard Schlenk

techniques or in a glovebox, but for other reactions, this was

2

unnecessary. ¹H and H NMR spectra were obtained using General

Electric QE300 and Bruker AM500 spectrometers. Low-temperature

kinetics were obtained on the AM500 spectrometer using an automated

sequence to record sequential spectra. NMR kinetics were all carried

out in NMR tubes equipped with a screw cap and silicone/PTFE septum

available from Wilmad Glass Co. Infrared spectra were recorded on a

Perkin-Elmer 1600 series FTIR spectrometer. Solvents were dried prior

to use: Et₂O-d₁₀over Na/benzophenone, CD₂Cl₂over CaH₂followed

by 4 Å molecular sieves, and CD₃OD (and other isotopomers) over 4

Å molecular sieves. The following compounds were prepared according

to literature procedures or analogous methods: (tmeda)PtMe₂(11),

(tmeda)PtMeCl (2), (tmeda)Pt(CH₂Ph)Cl (5), and trans-(PEt₃)₂Pt(CH₃)-

Cl (15), (4,4′-dimethyl-2,2′-bipyridyl)Pt(CH₃)₂(9).^101,102

Determination of Isotope Effects Involving 6 in CD₂Cl₂. Solutions

containing DCl and mixed HCl/DCl solutions were prepared in a

manner analogous to that described for HCl above. To obtain the

equilibrium and overall kinetic isotope effects, the mixed HCl/DCl

solution (130 µL, 2.09 M) was added to a solution of 5 (4 mg, 0.0091

mmol) in rigorously dried CD₂Cl₂(0.5 mL), and the ratio of CH₄and

CH₃D was obtained by integrating the corresponding peaks in the

1

resulting H NMR spectrum. The syringe used to dispense the HCl/

¹H NMR Kinetics for the Protonolysis of (tmeda)PtMeCl (2) in

CD₃OD. 2 (3 mg, 0.0083 mmol) and various amounts of LiCl and

LiClO₄were added to a series of NMR tubes according to the amounts

determined for the particular experiment. For [Cl^-]-dependence

experiments, the [LiCl] was varied between 0 and 0.66 M. For [H⁺]-

dependence experiments, [HOTf] was varied between 0 and 0.38 M.

LiClO₄was always added such that the ionic strength (µ ) [HOTf] +

[LiCl] + [LiClO₄]) equaled 1 M. This was followed by addition of

the solvent (CD₃OD) containing a known amount of tBuOH as an

integration standard. In order to monitor the kinetics at low temperature,

the individual tubes were cooled to -78 °C prior to the addition of

HOTf. After HOTf addition (total volume of reaction mixture ) 700

µL.), the contents of the tube were mixed thoroughly (while keeping

the tube as cold as possible), and the tube was then placed in the pre-

cooled NMR probe. After adjusting the necessary shims, a micropro-

gram was initiated to automatically acquire spectra at regular time

DCl solution was rinsed several times with approximately 100 µL of

the solution to avoid complications arising from H/D exchange with

the protic sites on the glass syringe. The ratio of HCl to DCl was

calibrated by adding the solution (using the same pre-rinsed syringe)

to an excess of CH₃Li‚LiBr (relative to [H⁺] + [D⁺]) dissolved in THF-

d₈. Integration of the CH₄and CH₃D gave the HCl/DCl ratio, and this

ratio was used to determine the equilibrium and kinetic isotope effects.

Isotope effect and calibration experiments were typically done in

triplicate.

¹H NMR Kinetics for H/D Exchange and Methane Elimination

in 12 in CD₃OD. In an inert atmosphere glovebox, 11 (5 mg, 0.015

mmol) was loaded into a 5 mm NMR tube and then removed from the

glovebox. CD₂Cl₂(50 µL) was added to dissolve 11, and the solution

was then cooled to -78 °C. LiCl and LiClO₄(of varying ratios

depending on the experiment) were then dissolved in methanol (total

solvent volume for reaction ) 0.7 mL), and the solution was added to

the cooled NMR tube. For [Cl^-]-dependence experiments, the [LiCl]

was varied between 0.10 and 0.60 M for H/D exchange and between

0 and 0.80 for methane elimination. The ionic strength was held

constant ([HOTf] + [LiCl] + [LiClO₄] ) 1.21 M) for all the reactions.

After thoroughly mixing the solution at low temperature, a triflic acid

solution (in methanol) was added to the NMR tube. The solution was

again mixed thoroughly before placing it in the precooled NMR probe

to monitor the kinetics. The kinetics were monitored for both the H/D

(100) Independent work by Zamashchikov et al.²⁰also supports the

intermediacy of alkylhydridoplatinum(IV) in C-H activation; however, their

argument is based on isotope effects and relies on the assumption that

σ-complex formation would have no isotope effect. In light of strong

evidence to the contrary (see refs 61-63), their conclusions appear open

to question.

(101) Clark, H. C.; Manzer, L. E. J. Organomet. Chem. 1973, 59, 411.

(102) Appleton, T. G.; Hall, J. R.; Williams, M. A. J. Organomet. Chem.

1986, 303, 139.

5976 J. Am. Chem. Soc., Vol. 118, No. 25, 1996

Stahl et al.

exchange reaction and methane loss by integrating the Pt(IV)-Me

resonance from 12 relative to an internal standard (pentachloroethane).

The H/D exchange reactions were carried out in CD₃OD, while reactions

involving methane loss were done in CD₃OH. Again preprogrammed

subroutines were used to record the spectra; the program for monitoring

methane loss included a solvent presaturation sequence to delete the

-OH resonance.

OD. Reactions involving 12-d₆were monitored by ²D NMR, and again

spectra were accumulated automatically. Kinetic data were obtained

by monitoring the loss in the integration intensity of the Pt-CD₃

resonance relative to an internal standard (toluene-d₈). When monitor-

ing elimination of CD₄in CH₃OD, the solvent resonance was suppressed

using the “1-3-3-1” pulse sequence developed by Hore.^103,104

Isotope Effect for Protonolysis of “[PtMeCl₃]^2-” in Water and

Methanol. The dry reagents [NMe₄]₂[PtMeCl₅] (100 mg, 0.19 mmol)

and CrCl₂(46 mg, 0.37 mmol) were combined and thoroughly mixed

in a small flask equipped with a stirbar. The flask was placed on the

high-vacuum line, evacuated, and cooled to liquid N₂temperature. A

10:1 mixture of degassed D₂O:H₂O was then condensed into the flask

and allowed to warm. Upon thawing, the solution turned dark green

(reaction with Cr(II)) and gas evolution was immediately apparent along

with formation of a black precipitate. The gas was collected and

quantified using a Toepler pump, and the CH₄/CH₃D mixture was

characterized and their ratio determined using ¹H NMR spectroscopy.

Alternatively, the dry reagents were loaded into an NMR tube equipped

with a J. Young valve (Wilmad Glass Co.) under inert atmosphere.

Methanol was then vacuum transferred into an NMR tube at -78 °C.

The heterogeneous mixture was mixed while warming to 0 °C.

Methane evolution was again immediately apparent along with forma-

tion of black precipitate. The methane generated was analyzed by

vacuum tranfer of all volatiles into a separate NMR tube.

¹H NMR Kinetics for H/D Exchange and Methane Loss from

15 in CD₃OD. In an NMR tube, 15 (5 mg, 0.010 mmol) was combined

with the appropriate amount of LiCl and LiClO₄(ratio varied depending

on the experiment). The ionic strength was held constant ([HOTf] +

[LiCl] + [LiClO₄] ) 1.15 M) for all experiments. In order to determine

the [Cl^-] dependence of the reactions, [LiCl] was varied between 0.10

and 0.50 M for H/D exchange and between 0.067 and 0.78 M for

methane elimination. The reagents were dissolved in methanol (CD₃-

OD for monitoring H/D exchange and CD₃OH for methane loss) and

then cooled to -78 °C. After addition of a methanol solution of triflic

acid (DOTf/CD₃OD for H/D exchange and HOTf/CD₃OH for methane

loss), the reaction was mixed thoroughly and then placed in the NMR

probe. Kinetics data were obtained by monitoring the loss in the

integration intensity of the Pt-Me resonance relative to an internal

standard (pentachloroethane). Again preprogrammed subroutines

were used to record the spectra; the program for monitoring methane

loss included a solvent presaturation sequence to delete the -OH

resonance.

Acknowledgment. This work was supported by the Office

of Naval Research and the Army Research Office. S.S.S. thanks

the National Science Foundation for a predoctoral fellowship.

Kinetic Isotope Effects for H/D Exchange and Methane Elimina-

tion from 12 and 15 in Methanol. Kinetic isotope effects for H/D

exchange and methane elimination from 12 in CD₃OD were obtained

by comparing these rates with those of analogous reactions of 12-d₆in

CH₃OH and CH₃OD, respectively. Standardized solutions of HOTf

in CH₃OH and DOTf in CH₃OD were prepared for these reactions.

Reaction conditions were identical to the reactions carried out in CD₃-

JA960110Z

(103) Hore, P. J. J. Magn. Reson. 1983, 54, 539.

(104) Hore, P. J. J. Magn. Reson. 1983, 55, 283.

Article Doi

Exploring the mechanism of aqueous C-H activation by Pt(II) through model chemistry: Evidence for the intermediacy of alkylhydridoplatinum(IV) and alkane σ-adducts

DOI: 10.1021/ja960110z

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja960110z

Full text of DOI:10.1021/ja960110z

Products guided by the article

R&D Labs maybe for 171359-86-7

Relevant to this article

Hot Product