C-H bond activation by cationic platinum(II) complexes: Ligand electronic and steric effects

Published on Web 01/15/2002

C-H Bond Activation by Cationic Platinum(II) Complexes:

Ligand Electronic and Steric Effects

H. Annita Zhong, 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 May 14, 2001

Abstract: A series of bis(aryl)diimine-ligated methyl complexes of Pt(II) with various substituted aryl groups

has been prepared. The cationic complexes [(ArNdCR-CRdNAr)PtMe(L)]⁺[BF₄]^-(Ar ) aryl; R ) H, CH₃;

L ) water, trifluoroethanol) react smoothly with benzene at approximately room temperature in trifluoro-

ethanol solvent to yield methane and the corresponding phenyl Pt(II) cations, via Pt(IV)-methyl-phenyl-

hydrido intermediates. The reaction products of methyl-substituted benzenes suggest an inherent reactivity

preference for aromatic over benzylic C-H bond activation, which can however be overridden by steric

effects. For the reaction of benzene with cationic Pt(II) complexes bearing 3,5-disubstituted aryl diimine

ligands, the rate-determining step is C-H bond activation, whereas for the more sterically crowded analogues

with 2,6-dimethyl-substituted aryl groups, benzene coordination becomes rate-determining. This switch is

manifested in distinctly different isotope scrambling and kinetic deuterium isotope effect patterns. The more

electron-rich the ligand is, as assayed by the CO stretching frequency of the corresponding carbonyl cationic

complex, the faster the rate of C-H bond activation. Although at first sight this trend appears to be at odds

with the common description of this class of reaction as electrophilic, the fact that the same trend is observed

for the two different series of complexes, which have different rate-determining steps, suggests that this

finding does not reflect the actual C-H bond activation process, but rather reflects only the relative ease

of benzene displacing a ligand to initiate the reaction; that is, the change in rates is mostly due to a ground-

state effect. The stability of the aquo complex ground state in equilibrium with the solvento complex increases

as the diimine ligand is made more electron-withdrawing. Several lines of evidence, including the mechanism

of degenerate acetonitrile exchange for the methyl-acetonitrile Pt(II) cations in alcohol solvents, suggest

that associative substitution pathways operate to get the hydrocarbon substrate into, and out of, the

coordination sphere; that is, the mechanism of benzene substitution proceeds by a solvent (TFE)-assisted

associative pathway.

Introduction

Pt(II) catalyzes oxidation of alkanes to alcohols by Pt(IV) at

120 °C.⁴This system is not yet practical because it requires an

Extensive research over the last 30 years aimed at the selective

functionalization of alkanes by transition metal complexes¹has

discovered many examples of inter- and intramolecular C-H

bond activation. Relatively few examples, however, lead to

actual alkane functionalization.²We have concentrated on an

example of electrophilic activation of alkanes by late transition

metal complexes,³the so-called Shilov system (eq 1), in which

expensive stoichiometric oxidant, the catalyst is unstable with

respect to Pt metal formation, and rates are too slow, but it does

exhibit patterns of regioselectivity (1° > 2° . 3°)^1das well as

chemoselectivity (C-H bonds of RCH₃are activated in prefer-

ence to C-H bonds of RCH₂OH)⁵that would be of considerable

practical interest if the above problems could be solved.

cat [PtCl₄]^2-

RCH₃+ [PtCl₆]^2-+ H₂O

8

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

Chem. Res. 1995, 28, 154. (b) SelectiVe Hydrocarbon ActiVation; Davies,

J. A., Watson, P. L., Liebman, J. F., Greenberg, A., Eds.; VCH: New York,

1990. (c) ActiVation and Functionalization of Alkanes; Hill, C. L., Ed.;

John Wiley & Sons: New York, 1989. (d) Shilov, A. E.; Shul’pin, G. B.

ActiVation and Catalytic Reactions of Saturated Hydrocarbons in the

Presence of Metal Complexes; Kluwer Academic Publishers: Dordrecht,

2000.

(2) For some examples, see: (a) Periana, R. A.; Taube, D. J.; Gamble, S.;

Taube, H.; Satoh, T.; Fujii, H. Science 1998, 280, 560-564. (b) Periana,

R. A.; Taube, D. J.; Evitt, E. R.; Loffler, D. G.; Wentrcek, P. R.; Voss, G.;

Masuda, T. Science 1993, 259, 340-343. (c) Waltz, K. M.; Hartwig, J. F.

J. Am. Chem. Soc. 2000, 122, 11358-11369. (d) Chen, H. Y.; Schlecht,

S.; Semple, T. C.; Hartwig, J. F. Science 2000, 287, 1995-1997. (e) Liu,

F.; Pak, E. B.; Singh, B.; Jensen, C. M.; Goldman, A. S. J. Am. Chem.

Soc. 1999, 121, 4086-4087. (f) Sen, A.; Benvenuto, M. A.; Lin, M.;

Hutson, A. C.; Basickes, N. J. Am. Chem. Soc. 1994, 116, 998-1003. (g)

Jia, C.; Kitamura, T.; Fujiwara, Y. Acc. Chem. Res. 2001, 34, 633-639.

H₂O,120 °C

+ [PtCl₆]^2-+ 2HCl (1)

RCH₂OH

(RCH₂Cl)

Studies by our group^5,6and others⁷implicate a catalytic cycle

consisting of three steps: (i) electrophilic activation of the alkane

to yield a Pt(II)-alkyl compound; (ii) oxidation to a Pt(IV)-alkyl

(3) Stahl, S.; Labinger, J. A.; Bercaw, J. E. Angew. Chem., Int. Ed. 1998, 37,

2181-2192.

(4) Gol’dshleger, N. F.; Es’kova, V. V.; Shilov, A. E.; Shteinman, A. A. Zh.

Fiz. Khim. 1972, 46, 1353-1354 (English translation 1972, 46, 785-786).

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

E.; Horvath, I. T.; Eller, K. Organometallics 1993, 12, 895-905.

9

1378 VOL. 124, NO. 7, 2002 J. AM. CHEM. SOC.

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

Scheme 1

Scheme 2

by [PtCl₆]^2-via electron transfer; and (iii) nucleophilic attack

by water or chloride ion at the alkyl to regenerate the Pt(II)

complex and release the functionalized alkane (Scheme 1).

Studies of protonolysis of R-Pt(II), the microscopic reverse of

C-H activation, were carried out on model systems to shed

some light on the nature of step (i) and led to the conclusion

that two intermediates are involved: a Pt(II)-C,H-η²-alkane

complex and a Pt(IV)-alkyl-hydrido complex.⁸These studies

further suggested that cationic complexes of the form [(N-N)-

PtR(L)]⁺, where N-N is a bidentate diamine or diimine and L

is a weakly bound solvent molecule or other ligand, should be

capable of activating C-H bonds according to eq 2.

metal center? One might intuitively expect that for these

formally electrophilic activations, a more electron-deficient

metal center would lead to a faster reaction. However, it is not

clear whether the term “electrophilic” has real mechanistic

significance or merely describes stoichiometry. Indeed, a recent

observation for a related reaction of a cationic Ir complex

suggests the opposite trend: the more electron-rich center

appears to react faster.¹²Changes in ligand properties might

even change the rate-determining step, with possible conse-

quences for selectivity. It is also unclear whether hydrocarbon

enters the coordination sphere via an associative or dissociative

mechanism.

To address some of these issues, we have initiated a series

of kinetic and mechanistic investigations on the C-H activation

of benzene by [(ArNdC(R)-C(R)NdAr)Pt(Me)(L)]⁺(BF₄)^-(R

) Me or H, L ) H₂O/TFE) in which both the steric and the

electronic properties of Ar, and hence the overall complex, are

varied over a significant range. We have also examined

substitution reactions in the same system (L ) MeCN). On the

basis of these and other recent findings, we can construct a

mechanistic description which appears consistent with all

observations, even though it may fall somewhat short of a

complete explanation.

Two different examples confirming this prediction, where

N-N is tetramethylethylenediamine and L is pentafluoropyri-

dine,⁹or N-N is Ar^fNdC(Me)-C(Me)dNAr^f(Ar^f) 3,5-

(CF₃)₂C₆H₃) and L is water and/or trifluoroethanol (TFE),¹⁰were

subsequently reported.

The complexes with N-N as an R-diimine ligand and L as

H₂O/TFE are particularly well suited for mechanistic investiga-

tion, since many reactions take place at a convenient rate at or

near room temperature, and since the steric and electronic

properties of the ligands can easily be varied. A detailed kinetics

and mechanistic study on the reaction of [(ArNdCMe-CMeNd

Ar)PtMe(L)]⁺(Ar ) 2,6-(CH₃)₂C₆H₃) with benzene supported

the involvement of the two intermediates cited above, as well

as a Pt(II)-(π-C,C-η²-benzene) complex, the formation of which

appears to be the rate-determining step.¹¹Yet several key

questions remain unanswered. In particular, how does the

reaction depend on the electronic and steric properties of the

Results

Synthesis of Diimine Ligands. The diimines ArNdCMe-

CMeNdAr (1, Ar ) 3-R³-4-R⁴-5-R⁵-C₆H₂; 2, Ar ) 2,6-(CH₃)₂-

4-R⁴-C₆H₂), formally derivatives of 1,4-diazabutadiene, are

prepared in moderate to good yields by condensing 2,3-

butanedione and the corresponding anilines in methanol with a

catalytic amount of formic acid or in toluene at 90 °C over 5 Å

molecular sieves¹³(Scheme 2). The latter procedure is particu-

larly effective for the more electron-withdrawing anilines, and

the use of 5 Å molecular sieves is crucial to obtain a reasonable

yield of the products.

(6) (a) Luinstra, G. A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J. E.

Organometallics 1994, 13, 755-756. (b) Luinstra, G. A.; Labinger, J. A.;

Bercaw, J. E. J. Am. Chem. Soc. 1993, 115, 3004-3005. (c) Luinstra, G.

A.; Wang, L.; Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Organomet.

Chem. 1995, 504, 75.

(7) (a) Kushch, L. A.; Lavrushko, V. V.; Misharin, Y. S.; Moravskii, A. P.;

Shilov, A. E. NouV. J. Chim. 1983, 7, 729-733. (b) Hutson, A. C.; Lin,

M.; Basickes, N.; Sen, A. J. Organomet. Chem. 1995, 504, 69-74. (c)

Horvath, I. T.; Cook, R. A.; Millar, J. M.; Kiss, G. Organometallics 1993,

12, 8-10.

(8) (a) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1996,

118, 5961-5976. (b) Stahl, S. S.; Labinger, J. A.; Bercaw, J. E. J. Am.

Chem. Soc. 1995, 117, 9371-9372.

(9) (a) Holtcamp, M. W.; Henling, L. M.; Day, M. W.; Labinger, J. A.; Bercaw,

J. E. Inorg. Chim. Acta 1998, 270, 467-478. (b) Holtcamp, M. W.;

Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc. 1997, 119, 848-849. (c)

Holtcamp, M. W.; Labinger, J. A.; Bercaw, J. E. Inorg. Chim. Acta 1997,

265, 117-125.

(10) Johansson, L.; Ryan, O. B.; Tilset, M. J. Am. Chem. Soc. 1999, 121, 1974-

1975.

For comparison we wanted to have some examples of the

analogous ligands without the backbone methyl substituents,

ArNdCH-CHNdAr. Although reactions between 3,5-disub-

stituted anilines and glyoxal¹⁴invariably yielded mixtures,

(12) (a)Tellers, D. M.; Skoog, S. J.; Bergman, R. G.; Gunnoe, T. B.; Harman,

W. D. Organometallics 2000, 19, 2428-2432. (b) Tellers, D. M.; Bergman,

R. G. J. Am. Chem. Soc. 2000, 122, 954-955. (c) Tellers, D. M.; Yung,

C. M.; Arndtsen, B. A.; Adamson, D. R.; Bergman, R. G. J. Am. Chem.

Soc. 2002, 124, 1400-1410.

(11) Johansson, L.; Tilset, M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem. Soc.

2000, 122, 10846-10855.

(13) Westheimer, F. H.; Taguchi, K. J. Org. Chem. 1971, 36, 1570-1572.

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1379

A R T I C L E S

Zhong et al.

Scheme 3

Scheme 5

Scheme 4

In many instances, addition of CO causes the solution to

change from orange or red to yellow almost instantaneously

(except for 9b, in which case the solution turned to wine red).

The Pt-CH₃resonances (in CD₂Cl₂) shift ∼0.2 ppm upfield as

compared to those of the corresponding dimethyl complexes,

while ²J_Pt-Hdecreases from ∼85 to ∼66 Hz. As a result of the

reduced symmetry, the backbone R groups are now inequivalent,

which is displayed in the NMR spectra, particularly in the ³J_Pt-H

values.

All the carbonyl cations appear to be indefinitely stable in

trifluoroethanol under 1 atm of CO, but 7d and 7e decompose

rapidly in the solid state or in methylene chloride in the absence

of excess CO. In TFE-d₃, the backbone methyls of 7e became

fully deuterated within an hour at room temperature. Isolated

9b appeared to contain a small amount of paramagnetic material

that significantly broadened the spectrum. The carbonyl IR

stretching frequencies of complexes 7-9 are reported in Table

1.

Generation of Platinum(II) Methyl-Aquo/Solvento Cat-

ions. Protonolysis of Pt(II) dimethyl complexes 4-6 by aqueous

HBF₄in TFE generates the corresponding methyl cations as an

equilibrium mixture of aquo and solvento adducts 10-12

(Scheme 6).^10,11

The stoichiometry need not be exactly 1:1, as the Pt-Me

groups of cationic complexes do not readily react with the slight

excess of acid. Subsequent addition of a slight excess of

acetonitrile (or carrying out the initial protonolysis in acetonitrile

solution) affords the acetonitrile adducts 13-15. The platinum

satellites for the Pt-CH₃signals for 10-12 are extremely broad,

in contrast to the relatively sharp satellites observed for the

corresponding carbonyl (7-9) and acetonitrile (13-15) adducts.

Protonolysis of 4-6 in TFE-d₃results in liberation of CH₄

as well as CH₃D, with concomitant formation of [Pt-CH₂D]

and [Pt-CH₃] cations (eq 3).

diimines 3 were prepared by reacting the corresponding imi-

nophosphoranes¹⁵with glyoxal trimer (Scheme 3).¹⁶

Preparation of Platinum Dimethyl Complexes. Diimine

ligands 1-3 react with bis(dimethyl(µ-dimethyl sulfide)plati-

num(II))¹⁷in toluene at room temperature to afford the corre-

sponding platinum dimethyl complexes 4-6 in high yields

(Scheme 4). The most useful feature for characterization is the

[Pt-CH₃] signal, with its accompanying ¹⁹⁵Pt satellites, in the

¹H NMR. The shift is somewhat ligand-dependent, but the two-

bond platinum coupling is virtually the same, ∼85 Hz, for all

these dimethyl complexes.

Preparation of Platinum(II) Methyl Carbonyl Cations.

Addition of 1 equiv of HBF₄(aq) to a solution of 4-6 in

trifluoroethanol generates a mixture of solvento and aquo

adducts (vide infra) of the platinum(II) methyl cation, which

upon exposure to 1 atm of carbon monoxide converts cleanly

to the corresponding platinum methyl carbonyl cation 7-9

(Scheme 5).

(14) Kliegman, J. M.; Barnes, R. K. J. Org. Chem. 1970, 35, 3140-3143. 2,6-

Dimethylaniline and its derivatives do react fairly cleanly with glyoxal (40%

aqueous) to yield the corresponding diimines.

(15) (a) Saito, T.; Ohkubo, T.; Kuboki, H.; Maeda, M.; Tsuda, K.; Karakasa,

T.; Satsumabayashi, S. J. Chem. Soc., Perkin Trans. 1 1998, 3065-3080.

(b) Nitta, M.; Soeda, H.; Iino, Y. Bull. Chem. Soc. Jpn. 1990, 63, 932-

934. (c) Hartmann, R. W.; Vom Orde, H. D.; Heindl, A.; Schoenenberger,

H. Arch. Pharm. (Weinheim, Ger.) 1988, 321, 497-501. (d) Soloshonok,

V. A.; Gerus, I. I.; Yagupol’skii, Y. L.; Kukhar, V. P. Zh. Org. Khim.

1987, 23, 2308-2313.

(16) Iminophosphoranes containing extremely electron-withdrawing substituents

(e.g, Ar ) 3,5-(CF₃)₂-C₆H₃) do not react with gloxyal trimer in refluxing

THF. The corresponding diimines can only be prepared by reacting the

iminophosphoranes with glyoxal monomer at room temperature in meth-

ylene chloride. For preparation of glyoxal monomer, see: Wang, Y.; Arif,

A. M.; Gladysz, J. A. Organometallics 1994, 13, 2164-2169.

(17) Hill, G. S.; Irwin, M. J.; Levy, C. J.; Rendina, L. M.; Puddephatt, R. J.

Inorg. Synth. 1998, 32, 149-153.

Within integration error limits, the ratio of CH₄to CH₃D

equals that of [Pt-CH₂D] to [Pt-CH₃]. No multiple deuteration

was observed for either the methane or the [Pt-methyl]. When

protonolysis of 4b was carried out in the presence of added

acetonitrile, the ratio of CH₃D to CH₄was found to increase

with the acetonitrile concentration (Figure 1).

9

1380 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

Table 1. Infrared Carbonyl Stretching Frequencies for [(N-N)Pt(CH₃)(CO)]⁺[BF₄]^-(7-9) in Methylene Chloride Solution

Table 2. Equilibrium Constants for [(N-N)Pt(R)(L)]⁺[BF₄]^-(10

and 11, R ) CH₃; 17 and 18, R ) C₆H₅; i, L ) TFE; ii, L ) H₂O)

in TFE/Water Mixtures at 20 °C

10i {^K^eq} 10ii

17i {^K^eq} 17ii

a (R³) R⁵) CMe₃, R⁴) OMe)

b (R³) R⁵) CMe₃, R⁴) H)

c (R³) R⁴) R₅) OMe)

3.9 × 10²

4.3 × 10²

7.5 × 10²

1.4 × 10³

2.8 × 10³

6.6 × 10²

7.4 × 10²

9.6 × 10²

1.8 × 10³

4.0 × 10³

d (R³) OMe, R⁵) CF₃, R⁴) H)

e (R³) R⁵) CF₃, R⁴) H)

11i {^K^eq} 11ii

18i {^K^eq} 18ii

a (R²) R⁶) Me, R⁴) Me)

b (R²) R⁶) Me, R⁴) H)

c (R²) R⁶) Me, R⁴) Br)

7.8 × 10²

9.5 × 10²

1.6 × 10³

1.5 × 10³

1.8 × 10³

3.1 × 10³

decomposition is accelerated by higher platinum concentrations

and retarded by added water; the mechanism for the formation

of the (µ-OH)₂dimer is not clear.

Figure 1. The ratio of CH₃D:CH₄generated in the protonolysis of 4b in

TFE-d₃as a function of acetonitrile concentration.

Scheme 6

Equilibria between Aquo and Solvento Complexes. Ad-

dition of 1 equiv of aqueous HBF₄(which contains ap-

proximately 5 mol of H₂O per mole of HBF₄) at 20 °C to a

suspension of 0.007 mmol of (4) in 0.7 mL of TFE-d₃results

in two Pt-CH₃signals in the ¹H NMR, in ratios ranging from

∼1.5:1 (10a) to >10:1 (10e). For 10a, addition of water further

increases the relative intensity of the major peak, which is

accordingly assigned to aquo complex 10aii. The equilibrium

between the solvento and aquo adducts (eq 5) greatly favors

the latter for all ligands examined here. The magnitude of the

equilibrium constant depends on the ligands (Table 2): the more

electron-withdrawing the diimine substituents (indicated by a

higher carbonyl stretching frequency for the analogous methyl-

carbonyl cation, vide supra), the larger the equilibrium constant.

Moreover, for the same diimine ligand, the equilibrium constant

for the platinum(II) phenyl cation (17 and 18) is greater than

Cations 10-12 decompose over the course of several weeks

at room temperature in TFE solution. For 10a/b and 12a, a single

(or major) species is formed, accompanied by methane libera-

tion. NMR (¹H and ¹⁹F) is consistent with a (µ-OH)₂dimeric

structure (16). This was confirmed for 16b, the decomposition

product of 10b, by a crystal structure determination.¹⁸The

analogous phenyl complexes 17 and 19 (vide infra) react

similarly, evolving benzene (eq 4). In contrast, the decomposi-

tion of 11 yielded a mixture of unidentifiable products. The

(18) Rostovtsev, V. V.; Labinger, J. A.; Bercaw, J. E., unpublished results.

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1381

A R T I C L E S

Zhong et al.

1

Figure 2. The tert-butyl region of the H NMR spectra, showing changes for the reaction between benzene and 10b ([D₂O] ) 2.72 M, [C₆H₆] ) 0.49 M,

[Pt] ) 0.01 M). The inset shows an exponential curve fit of the peak height at 1.25 ppm vs time.

for the methyl cation (10 and 11). The equilibrium constants

for 12 and 19 are too large to be determined accurately; they

are all greater than 3 × 10³.

Kinetics of Benzene C-H Bond Activation. ¹H NMR was

Reactions between Platinum(II) Methyl Cations and

used to monitor the disappearance of starting material and/or

the appearance of product, from which rates were determined.

Figure 2 shows the results of a typical experiment, for the

Benzene. Platinum(II) methyl cations 10-12 react cleanly with

benzene, or with partially or completely deuterated benzene

(vide infra), to form the corresponding platinum(II) phenyl

reaction between 10b ([D₂O] ) 2.72 M, where only aquo

complexes 17-19 respectively, with concurrent production of

adducts are detectable) and benzene in TFE-d₃. The values of

methane (eq 6).

Under reaction conditions where both the solvento and the

the calculated rate constants, k_obs, for the various complexes

under different sets of conditions can be found in the Supporting

aquo cations are observable by ¹H NMR, the two species

Information (Tables S1-S12). Additional kinetics experiments

disappear at the same rate, and the solvento and aquo adducts

of the phenyl products appear at the same rate. No other species

were observed up to 3 half-lives; the (µ-OH)₂dimers 16 can be

were carried out in several cases, particularly for 10a and 10b.

The reaction rates are not affected by ionic strength; for example,

the observed rate constant for the reaction between 10a and

benzene at [D₂O] ) 1.33 M, [C₆H₆] ) 0.25 M, and [Pt] ) 0.01

M is (1.8 ( 0.2) × 10^-4s^-1with no added NMe₄BF₄and (1.9

( 0.2) × 10^-4s^-1with [NMe₄BF₄] ) 0.12 M. Hence most of

the kinetic studies were run without controlling ionic strength.

However, it should be noted that the rates did show a slight

observed at later stages of the reaction. In all cases the rates of

(µ-OH)₂dimer formation (eq 4) from either the starting materials

10-12 or the products 17-19 are at least an order of magnitude

slower than those of benzene C-H bond activation, so the rate

constants reported below should not be affected by this

secondary reaction to any significant degree.

9

1382 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

Figure 3. The inverse of observed rate constants in the reactions of 10a

with benzene varies linearly with the water:benzene concentration ratio.

Figure 4. For the reactions of various Pt(II) methyl cations with benzene,

the inverse of observed rate constants is linear with respect to the water:

benzene concentration ratio. The slopes of the lines vary greatly from cation

to cation.

Figure 5. (a) A plot of k_obsvs [C₆D₆]/[D₂O] deviates from linearity at low

water concentrations ([Pt] ) 0.01 M, [C₆D₆] ) 0.25 M). (b) For reactions

of 10a and benzene at high water concentrations, k_obsis linear with respect

to [C₆D₆]/[D₂O] ([Pt] ) 0.01 M, [D₂O] ) 1.23-1.30 M).

decrease in the presence of weakly coordinating anions such as

triflate (Table S12).

As previously found for 11b,¹¹the rates of benzene C-H

bond activation are decreased by added water, and 1/k_obsis linear

with respect to [D₂O]/[C₆H₆] (Figure 3). The slope of the line

(and, thus, k_obs) depends significantly on the ligand (Figure 4).

The plot of k_obsversus [C₆H₆]/[D₂O] deviates from linearity at

very low water concentrations (Figure 5).

The temperature dependence was studied over the range of

0-55 °C for the reaction between 10b and C₆H₆and over the

range of 0-30 °C for the reaction between 10b and C₆D₆. The

water concentration was kept sufficiently high such that aquo

adducts account for >90% of the Pt(II) species. The overall

activation parameters were calculated from Eyring plots, such

Figure 6. Eyring plot for the reactions of 10b with C₆H₆and C₆D₆.

as the one shown in Figure 6, and give ∆H^q) 20 kcal mol^-1

∆S^q) 5 e.u. for C₆H₆activation and ∆H^q) 20.5 kcal mol^-1

,

consists mostly of CH₃D (∼60%) and CH₂D₂(∼30%). Even

less scrambling is observed for complexes 12. For example,

when 12a reacts with C₆D₆, CH₂D₂and higher isotopomers

∆S^q) 6 e.u. for C₆D₆activation. The entropy of activation

may be contrasted to that found for 11b, for which a ∆S^qof

-16 e.u. was measured.¹¹

1

account for <10% of the liberated methane. H NMR shows

that label is also incorporated into the methyl group of unreacted

starting material, up to ∼20% of [Pt-CH₂D]. For 11a and 11c

the results are very similar to those of 11b: reactions with C₆D₆

produce CH₂D₂and CHD₃as the major methane isotopomers,

although deuterium incorporation into the unreacted platinum-

methyl group is observed here as well.

Kinetic Deuterium Isotope Effects (KIEs). KIEs were

determined by three different methods: (1) in parallel reactions,

separately determining the rate constants for reactions of C₆H₆

and C₆D₆under the same conditions; (2) intermolecular

Deuterium Scrambling. Earlier studies on 11b¹¹found that

(a) <5% deuterium incorporation was observed in the methane

generated in the reactions with C₆H₆in TFE-d₃and (b) nearly

complete statistical scrambling of isotopes takes place in the

reaction with either C₆D₆or 1,3,5-C₆H₃D₃. For complexes 10,

again, there is no incorporation of label from solvent, but in

contrast there is only partial scrambling of protium and

deuterium among the methyl group and deuteriobenzenes. For

example, when 10b reacts with C₆D₆, the methane evolved

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1383

A R T I C L E S

Zhong et al.

Table 3. Kinetic Deuterium Isotope Effects for Benzene C-L (L ) H, D) Bond Activation Reactions

complex

T (°C)

method

k_H/k_D

complex

T (°C)

method

k_H/k_D

10a

10b

10c

10d

20

parallel

2.0

2.2

1.8

1.9

1.6

2.2

2.0

10e

11a

11b

11c

12a

12b

12c

20

35

20

parallel

2.2

1.1

5.0

3.6

5.9

1,3,5-C₆H₃D₃

1:1 C₆H₆:C₆D₆

parallel

1:1 C₆H₆:C₆D₆

parallel

1:1 C₆H₆:C₆D₆

Scheme 7

competition, determining (by ¹H NMR) the isotopic composition

of methane liberated in reaction with 1:1 C₆H₆:C₆D₆, and (3)

intramolecular competition, using 1,3,5-C₆H₃D₆. The results are

shown in Table 3. Note that the two competition methods give

values that are consistently lower than those measured by

parallel reactions (vide infra), although the differences are

comparable to the uncertainties.¹⁹

Reactions of 10b with Other Aromatic Hydrocarbons. The

regioselectivity for the C-H bond activation for alkyl-

substituted aromatic compounds appears to be affected by the

steric bulk of both the substrates and the ligand. Whereas 11b

reacts with p-xylene predominantly at the benzylic position (eq

7),²⁰the reaction between 10b and p-xylene results almost

exclusively in aryl C-H bond activation (eq 8).

benzylic activation product 21b to aromatic activation product

22b is observed, whereas at 55 °C, the selectivity drops to 3:1.

The product resulting from aromatic C-H bond activation

appears to be much less stable than that from benzylic C-H

activation: during the reaction between 10b and mesitylene,

the ¹H NMR resonances corresponding to 22b initially grow in

intensity and then disappear gradually, with formation of (µ-

OH)₂dimer 16b. Compared to benzene C-H bond activation,

the reaction between 10b and p-xylene is approximately 3 times

slower, while that with mesitylene is approximately 10 times

slower.

The kinetic deuterium isotope effects for reactions between

10b and p-xylene and mesitylene were measured (parallel

reactions method at 25 °C). These also appear to depend on

steric bulk of the substrates: k_H/k_D) 2.2 for benzene (C₆X₆),

1.5 for p-xylene (C₆X₃(CX₃)₂), and 1.2 for mesitylene (C₆X₃-

(CX₃)₃) (X ) H, D).

Electron-deficient or extremely bulky aromatic compounds

react sluggishly, or not at all, with 10b at room temperature.

For example, when 30 equiv of 1,4-di(tert-butyl)benzene was

added to a 0.01 M TFE-d₃solution of 10b ([D₂O] ) 0.05 M),

85% of the starting material remained after 24 h at room

temperature. The ¹H NMR of the reaction mixture showed

several broad, as yet unidentified new peaks. Similarly, 10b

showed little reactivity toward pentafluorotoluene or 1,4-

(CF₃)₂C₆H₄after several hours at room temperature.

Acetonitrile Exchange Reactions. To shed some light on

whether dissociative or associative substitution pathways operate

in these systems, we briefly investigated the isotopic exchange

between a Pt(II)-bound CH₃CN and free CD₃CN in several

alcoholic solvents (Scheme 8), using ¹H NMR spectroscopy to

follow the disappearance of bound acetonitrile and the appear-

ance of free CH₃CN.

In contrast, the reaction between mesitylene and 10b proceeds

mainly through benzylic C-H bond activation (Scheme 7).

In the latter case, the selectivity for benzylic activation

decreases at higher temperatures. At 25 °C, a ratio of 95:5 of

(19) The estimated uncertainties are ∼10-20% for the parallel reactions and

5-10% for the inter- and intramolecular competition reactions.

(20) Johansson, L.; Ryan, O. B.; Rømming, C.; Tilset, M. J. Am. Chem. Soc.

2001, 123, 6579-6590.

The rate constants for the exchange of 13b in CD₃OD, as a

function of CD₃CN concentration and temperature, are shown

9

1384 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

Figure 7. Plot of observed rate constants for exchange of free and bound

acetonitrile vs free [acetonitrile] for 13b at various temperatures.

Figure 9. A plot of observed rate constants vs free acetonitrile concentra-

tions for the isotopic exchange of bound (13b) and free acetonitrile in

different solvents at 40 °C.

Table 4. Second-Order Rate Constants for the Acetonitrile

Exchange Reactions for 13b in Various Alcohol Solvents at 40 °C

solvent

10⁴× k₁(s^-1

)

10⁴× k_s(M^-1s^-1

)

10⁴× k₂(M^-1s^-1

)

methanol

ethanol

2-propanol

TFE

8.4 ( 0.3

6.7 ( 0.4

4.0 ( 0.3

0.34 ( 0.01

0.39 ( 0.02

0.31 ( 0.01

3.6 ( 0.1

5.2 ( 0.2

8.6 ( 0.2

2.6 ( 0.1

<0.01

Table 5. Second-Order Rate Constants for the Acetonitrile

Exchange Reactions of Various Pt(II) Methyl Cations in Methanol

(30 °C) and TFE (40 °C)

10⁴× k₁

(s^-1

10⁴× k_s

10⁴× k₂

(M^-1s^-1

)

complex

solvent

T (°C)

)

(M^-1s^-1

)

13b

13d

14b

14c

13b

13d

14b

14c

methanol

TFE

30

40

2.7 ( 0.3

9.9 ( 0.4

2.6 ( 0.2

6.3 ( 0.3

0.11 ( 0.01

0.4 ( 0.02

0.11 ( 0.01

0.26 ( 0.02

1.5 ( 0.1

7.9 ( 0.3

1.8 ( 0.2

4.4 ( 0.3

<0.1

2.6 ( 0.1

Figure 8. Eyring plots for second-order acetonitrile exchange rate constants

for both the solvent-assisted pathway (k_s) and the direct-exchange pathway

(k₂) for 13b.

a

-

<0.01

2.8 ( 0.1

10.2 ( 0.1

Scheme 8

^aAcetonitrile exchange is too fast (k_ex> 50 × 10^-4s^-1) to measure

rates.

the small range of temperatures limit our confidence in these

values.

The exchange reactions in different solvents at 40 °C were

also investigated, with the results shown in Figure 9. Rate

constants were calculated in the same manner and are displayed

in Table 4. We also briefly examined the dependence of the

exchange rates on the ligands. Under similar conditions ([CD₃-

CN] ) 1.3 M, T ) 40 °C, solvent ) TFE-d₃), the exchange

between bound and free acetonitrile for 13d (Ar ) 3-OMe-5-

CF₃-C₆H₃) is ca. 6-7 times faster than that for 13b (Ar ) 3,5-

(CMe₃)₂-C₆H₃) and that for 14c (Ar ) 2,6-Me₂-4-Br-C₆H₂) is

ca. 3.5 times faster than that for 14b (Ar ) 2,6-Me₂-C₆H₃). In

contrast, the rates of exchange are approximately the same for

13b and 14b (Table 5).

In deuterated alcohols other than TFE-d₃, competing deu-

teration of the backbone methyl groups is observed. The rate

of deuteration not only depends on solvent and ligand, but is

also regioselective; the deuteration of the two nonequivalent

backbone methyl groups occurs at different rates in each case,

with the higher field NMR signal exhibiting more rapid

deuteration. For example, for 13b in CD₃OD at 30 °C, <5%

in Figure 7. The shapes of the curves indicate a rate law of the

form k_ex) k₁+ k₂[CD₃CN], where the first term represents a

dissociative or solvent-assisted pathway whose rate constant is

determined from the intercept of the extrapolated linear part of

the plot, and the second term represents a direct associative path

whose rate is obtained from the slope of the latter. Eyring plots

between 20 and 40 °C have been constructed for k_s() k₁/[CD₃-

OD]) and k₂(Figure 8); both lead to calculated negative

entropies of activation, although the precision of the data and

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1385

A R T I C L E S

Zhong et al.

deuteration occurred after 1.5 h; whereas for 13d, the higher

field backbone methyl resonance was completely deuterated

within 15 min, while <5% deuteration was observed for the

low field one. For 13b at 40 °C, <10% deuteration occurred

after 1 h in CD₃OD, but in (CD₃)₂CDOD, the high field

resonance was more than 80% deuterated after 15 min, and the

low field resonance was ∼40% deuterated after 1 h. Deuterium

exchange is even faster for carbonyl cations; in 7e, backbone

deuteration occurred even in TFE-d₃at room temperature,

whereas no observable deuteration occurred for 7a-d under

these conditions.

are ca. 80° to the N-Pt-N planes.¹⁸However, the bimetallic

structure of 16b with relatively short Pt-O bridges could cause

the phenyl rings to rotate further out of the plane to avoid steric

congestion. ¹H NMR signals of the complexes with 3,5-

disubstituted aryldiimines are also consistent with rapid rotation

around N-C^ipsobonds. For example, only one set of ligand and

[Pt-CH₃] NMR peaks is observed for 4d (which has asym-

metrically substituted aryl groups) down to -40 °C, implying

rotation about the C-N bond is fast on the NMR time scale.

We expect this is the case for all 3,5-disubstituted aryls.

The ¹⁹⁵Pt satellites are useful NMR features, not only in

facilitating assignments but also as qualitative probes of

electronic effect transmission. In particular, for the methyl

carbonyl cations the downfield signals for the backbone methyls

Discussion

Synthesis and Characterization of Platinum Complexes.

The procedures for ligand synthesis, formation of dimethyl-

platinum complexes, and protonolysis to monomethylplatinum

cations all appear to be quite general for a wide array of

substituted diaryldiimine ligands, thus making systematic ex-

amination of the effects of ligand electronic and steric properties

possible. The only exception is in the synthesis of 6c, where it

appeared necessary to periodically purge and remove free SMe₂,

which can compete with the diimine ligand 3c for the coordina-

tion to the Pt(II) center. It is particularly convenient that cationic

Pt(II)-methyl complexes are not readily protonolyzed,^9cwhich

permits their clean generation without requiring rigorous control

of stoichiometry in adding an equivalent of acid.

All complexes prepared are characterized straightforwardly

by ¹H NMR. The only important structural variable is the

orientation of the aryl rings with respect to the coordination

plane. Structural characterizations by Ruffo and co-workers on

cationic Pt-methyl-olefin complexes with Ar ) 2,6-diethylphe-

nyl revealed a near-orthogonal orientation between the aryl rings

and the Pt coordination plane.²¹NMR studies by the same

authors suggest there is hindered rotation around the N-C^ipso

bond as well.²²Similar conclusions have also been reached by

Eisenberg and co-workers, where they were readily able to

isolate the meso and rac isomers of a neutral (diimine)PtMe₂

complex with Ar ) 2-OMe-4,6-(t-Bu)₂-C₆H₂.²³In comparison,

few X-ray structures have been obtained for complexes contain-

ing diimine ligands with 3,5-disubstituted aryl substituents.

Recently the crystal structure of a related pyrrolyl-imine complex

(23) found the phenyl ring at 58° to the N-Pt-N plane.²⁴

4

(7 and 8) exhibit larger J_Pt-Hvalues than the upfield signals.

We tentatively assign the downfield resonances to the methyl

adjacent to the N trans to CO; the large trans influence of methyl

bonded to platinum would be expected to elongate the opposite

Pt-N bond and reduce the corresponding coupling constant.

Similarly, for 9, the upfield signals for the backbone H exhibit

³J_Pt-Hvalues (∼74 Hz) nearly twice as large as those for the

downfield ones (∼38 Hz) and are consequently assigned to the

H adjacent to the N trans to CO. Such effects have chemical as

well as spectroscopic consequences, as manifested by the

differential rates of deuteration of backbone methyls for

complexes 13 in deuterated alcohols. We will not consider these

effects in any detail here, except to note that they are consistent

with the substantial perturbations of C-H bond activation by

variation of diimine ligand electronic character, to be discussed

below.

CO Stretching Frequencies as a Measure of Electronic

Character. To examine electronic effects on C-H bond

activation, we sought an empirical measure of actual electronic

density at cationic Pt(II) as preferable to a semiempirical

prediction of expected ligand properties. As ν_COis a well-

established probe of electron density at a metal center, we

prepared the platinum methyl carbonyl cations with the various

diimine ligands. These bands are quite sharp for solution

samples, and the ν_COvalues are shown in Table 1 and plotted

26

against the Hammett substituent constants (σ ) σ_p+ σ_m)in

Figure 10. All three series, 7a-f, 8a-c, and 9a-c, give fairly

good linear correlations; however, the lines are well displaced

from one another. In other words, for the same aryl σ values,

the cations 8 show a higher CO stretching frequency, and thus

are comparatively electron-poor, relative to 7. For example, ν_CO

for 7f (Ar ) 3,5-Me₂C₆H₃) is 2105.7(3) cm^-1versus 2109.6(3)

cm^-1for 8b (Ar ) 2,6-Me₂C₆H₃), though one might expect

that they should be reasonably electronically similar (both having

dimethylaryl groups) and exhibit similar CO stretching frequen-

cies. Why do they not?

The latter finding is consistent with AM1 level calculations

on 4e, which predict that the phenyl rings are ca. 60° tilted out

of the N-Pt-N plane.²⁵On the other hand, the X-ray structure

of 16b (R³) R⁵) CMe₃, R⁴) H) showed the phenyl rings

Of course, as defined above the σ values for complexes 8 do

not take account of the ortho methyl groups, as ortho substituent

constants are considered unreliable owing to steric complica-

tions. Yet this cannot be the explanation, as σ for methyl is

negative (electron-releasing relative to hydrogen); hence includ-

ing a contribution for the ortho substituents would move the

correlation line for 8 to the left, further from that for 7. We

(21) (a) Ganis, P.; Orabona, I.; Ruffo, F.; Vitagliano, A. Organometallics 1998,

17, 2646-2650. (b) Fusto, M.; Giordano, F.; Orabona, I.; Ruffo, F.

Organometallics 1997, 16, 5981-5987.

(22) Zuccaccia, C.; Macchioni, A.; Orabona, I.; Ruffo, F. Organometallics 1999,

18, 4367-4372.

(23) Yang, K.; Lachicotte, R. J.; Eisenberg, R. Organometallics 1998, 17, 5102-

5113.

(26) The Chemist’s Companion: A Handbook of Practical Data, Techniques,

and References; Gordon, A. J., Ford, R. A., Eds.; John Wiley & Sons:

New York, 1972.

(24) Scollard, J. D.; Labinger, J. A.; Bercaw, J. E., manuscript in preparation.

(25) Brandow, C. G.; Zhong, H. A., unpublished results.

9

1386 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

Figure 11. Log(K_eq) of 10 and 11 vs the CO stretching frequencies of the

corresponding methyl/carbonyl cations 7 and 8.

Figure 10. Plot of CO stretching frequencies for platinum(II) methyl/

carbonyl cations 7, 8, and 9 vs the σ values of the aryl substituents.

Scheme 9

believe that the difference lies in steric factors; the 2,6-methyl

groups in 8 cause the aryl groups to be oriented perpendicular

to the coordination plane, minimizing crowding, whereas the

less bulky aryls of 7 are rotated more parallel to the plane, in

at least partial conjugation with the unsaturated [NdC(Me)-

C(Me)dN] group. This would be expected to result in more

effective transmission of the substituent effects from the aryl

group, via the diimine, to the metal center. AM1 semiempirical

calculations²⁵support this; free ligand 2b is predicted to have

the aryl ring locked perpendicular to the [N-Pt-N] plane, with

consequently little π-donation from the phenyl ring to the

nitrogen atoms, while in 1f the dihedral angle between the aryl

ring and the diimine backbone is predicted to be approximately

60°, and the calculated electron density at the nitrogen in 1f is

higher than in 2b.

Methyl substitution at backbone positions of the diimine

ligands is also electron-releasing, as can be seen by comparing

7b-d and 9a-c, for which the aryl groups are identical; the

CO stretching frequencies of the former (with backbone methyls)

are red-shifted by about 4-6 cm^-1

.

The basicity of alcohols decreases in the order Me₂CHOH

> EtOH > MeOH . TFE, consistent with the observation that

deuteration occurs fastest in deuterated 2-propanol and slowest

in TFE-d₃. Electron-withdrawing groups will increase the acidity

of R-methyl protons, so the more electron-withdrawing diimine

ligands should lead to a higher rate for deuterium incorporation

into the methyl backbone. The more rapid exchange for 13d

than for 13b is thus consistent with the higher ν_COfor 7d than

7b. The observation of exchange for 7e even in TFE-d₃at room

temperature reflects the greater electron-withdrawing power of

CO relative to acetonitrile.

The differential deuteration rates observed for the two

backbone methyl groups in the same molecule may be attributed

to the differing trans influence of methyl versus MeCN. Thus

a Pt-N bond trans to MeCN is expected to be shorter, resulting

in stronger donation to platinum, making the nitrogen more

electropositive and accelerating exchange at the neighboring

methyl group.

Ligand Electronic Effects on the Aquo/Solvento Equilib-

rium. All cations 10-12 generated by protonolysis of the

corresponding dimethyl complexes with aqueous acid in TFE

are formed as mixtures of solvento (i) and aquo (ii) adducts in

rapid equilibrium (eq 5). In all cases measured, the equilibrium

substantially favors the aquo adduct (Table 2). Furthermore,

correlation of K_eqwith ν_COof the corresponding carbonyl cation

shows that the preference for water over TFE increases as the

metal center is made more electron-poor (Figure 11). This trend

presumably reflects the greater electron-donating ability of

water, relative to TFE, toward cationic Pt(II) centers. Similarly,

the phenyl cations 17 and 18 are more selective toward water

binding than the corresponding methyl cations 10 and 11.

Methyl groups are extremely good σ donors to metal centers,

whereas phenyl groups are more electronegative (sp²- vs sp³-

hybridized carbon) and thus less electron-donating, rendering

the platinum center more electron-deficient.

Ligand Electronic Effects on Backbone Methyl Deutera-

tion. In some cases, deuterium is incorporated into the backbone

methyls of Pt(II) methyl acetonitrile adducts in (CD₃)₂CDOD,

CD₃CD₂OD, and CD₃OD. This probably occurs via Lewis acid-

catalyzed imine- enamine tautomerization (Scheme 9), whose

rate will depend on both the basicity of the solvent and the

acidity of the imine R-methyl proton.

Ligand Effects on Overall Reaction Rates of Arene C-H

Bond Activation. Pt(II) methyl cations 10-12 react cleanly

with benzene to give the corresponding phenyl cations 17-19,

accompanied by liberation of methane. The kinetics are

conveniently followed by ¹H NMR and exhibit clean first-order

behavior in [Pt]. At sufficiently high water concentrations, where

the Pt(II) aquo adducts account for >90% of the total platinum

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1387

A R T I C L E S

Zhong et al.

Figure 12. A plot of the logarithms of the observed rate constants in

reactions of methyl cations 10 with C₆D₆as a function of ν_COof the

corresponding methyl/carbonyl cations 7.

species, the dependence on concentrations of benzene and water

satisfy the apparent rate law: rate ) k_obs[benzene]/[water]. A

series of comparative experiments was performed to determine

and distinguish ligand electronic and steric effects on k_obs

.

The reaction rates of C₆D₆with cations 10 were measured at

20 °C under two sets of conditions. At the lower water

concentration ([D₂O] ) 0.05 M, [TFE-d₃] ) 14.1 M, and [C₆D₆]

) 0.25 M), both the aquo and the solvento complexes are

observed in the reaction mixtures, and at the higher water

concentration ([D₂O] ) 0.70 M, [TFE-d₃] ) 12.3 M, and [C₆D₆]

) 1.31 M), the aquo complexes are the only observable species.

The logarithms of the observed rate constants are plotted against

the IR CO stretching frequencies of the corresponding carbonyl

complexes (7) in Figure 12. Under either set of reaction

conditions, a good linear correlation is obtained, demonstrating

that the more electron-rich Pt(II) cations are more reactive

toward C₆D₆. The slope of the linear correlation does depend

on conditions, however. A similar correlation is found for cations

11 reacting with C₆D₆at 35 °C (Figure 13a) and for cations 12

at 20 °C (Figure 13b).

Figure 13. (a) A plot of the logarithms of the observed rate constants for

reactions of cations 11 with C₆D₆as a function of ν_COfor the corresponding

methyl/carbonyl cations 8. (b) A plot of the logarithms of the observed

rate constants for reactions of cations 12 with C₆D₆/C₆H₆as a function of

ν_COfor the corresponding methyl/carbonyl cations 9.

for the latter, so we will use Scheme 11 as the framework for

our discussion.

In this scheme, the aquo (Aii) and solvento (Ai) complexes

are in rapid equilibrium, with benzene displacing the more

weakly bound solvent ligand; direct attack of benzene on Aii

to displace water is assumed to be negligible, although a small

contribution from this route would probably not perturb the

kinetics enough to be detectable. π-Benzene complex B then

undergoes C-H bond cleavage (probably via a σ-complex C).

Limiting cases for the rate-determining step would be formation

of B (k₂> k_-1[TFE]) or C-H cleavage (k₂< k_-1[TFE]). In

either case, the deduced rate law (see Supporting Information

for derivation) predicts that a plot of 1/k_obsversus [H₂O]/[C₆H₆]

will be linear, which is observed in all cases (Figure 3). At high

water concentrations (K_eq[H₂O] . [TFE]), the water complex

Aii is the major species in solution, and a plot of k_obsversus

[C₆H₆]/[H₂O] is expected to be linear, as long as benzene and

water are not present in high enough concentrations to signifi-

cantly alter the solvent properties. This expectation is born out

by the experimental results. At low water concentrations (K_eq-

[H₂O] ≈ [TFE]), concentrations of Ai and Aii will both be

significant, so a plot of k_obsversus [C₆H₆]/[H₂O] would be

expected to deviate from linearity, as observed.

The steric effects of the diimine ligands have a profound effect

on the reactivity of the metal center. For example, 8a has a CO

stretching frequency of 2108 cm^-1, indicating that the corre-

sponding 11a is somewhat more electron-rich than 10d, whose

corresponding methyl-carbonyl cation 7d has an IR stretching

frequency of 2110 cm^-1. However, under the same reaction

conditions, 10d reacts an order of magnitude faster with C₆D₆

than does 11a (Figure 14a). In general, reactions of benzene

with complexes containing 2,6-dimethyl-substituted aryls (11)

are considerably slower than those with no such substitution

(10), indicating a substantial slowing effect of increasing steric

bulk. On the other hand, Figure 14a indicates that the reactions

of complexes 12 are considerably slower than those of 10 with

similar electron density, although 12 should be, if anything, less

crowded than 10. We do not currently have a fully satisfying

explanation for this apparent anomaly (vide infra).

Details of the Mechanism of Benzene C-H Bond Activa-

tion. The mechanism proposed previously to account for the

reaction of benzene with 11b is shown in Scheme 10.

For 11b, formation of π-benzene adduct B was inferred to

be the rate-determining step, primarily on the basis of the first-

order dependence on benzene concentration coupled with the

low KIE (∼1) and the virtually complete isotopic scrambling

observed (see below).¹¹It is of interest here to compare the

The observed rate law, particularly the inverse dependence

on water concentration, can be interpreted in terms of either a

dissociative or a solvent-assisted associative pathway, but several

considerations (to be discussed later) lead to a strong preference

9

1388 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

Scheme 10

Scheme 11

any complex and C₆H₆in TFE-d₃. This means that reversible

deprotonation of intermediate D (to give I) in Scheme 10 is

much slower than all other reactions, since otherwise deuterium

from solvent would exchange into the complex and thence into

the methyl group. Furthermore, the fact that no H/D exchange

occurs in the present system explains the observation that when

cations 10 are generated by protonolysis of 4 in TFE-d₃, only

[Pt-CH₂D] and [Pt-CH₃] complexes are formed, with corre-

sponding amounts of CH₄and CH₃D; no multiple deuteration

is detectable. As shown in Scheme 12, addition of D⁺(from

aqueous DBF₄) to the platinum dimethyl complexes gives a Pt-

(IV)-D species a, which undergoes C-D bond formation to

form b, which may either directly release CH₃D to afford c or

rearrange and cleave a C-H bond to yield a′.

Figure 14. (a) A plot of the logarithms of the observed rate constants in

reactions of Pt(II) cations of variable steric bulk with C₆D₆as a function of

ν

_CO. (b) A plot of the logarithms of the observed rate constants in reactions

of Pt(II) cations of variable steric bulk with C₆H₆as a function of ν_CO

.

Because there is no H/D exchange between the solvent and

Pt(IV) hydrido intermediates, only one deuterium atom is

involved for each molecule of methane released. This deuterium

atom is either incorporated into the liberated methane as CH₃D

or left behind as [Pt-CH₂D]. This stands in contrast to previous

protonolysis studies of Pt(II) alkyl complexes in CD₃OD, where

substantial deuterium incorporation into the Pt alkyl group was

observed.⁵The difference may reflect the reduced basicity of

TFE as compared to that of methanol.

If the rate of the scrambling process in Scheme 12 is fast

relative to the loss of methane, a statistical mixture of CH₃D

and CH₄(4:3) will be obtained (assuming negligible KIE for

C-H vs C-D activation and methane dissociation); if the

behavior of cations 10 (which lack 2,6-dimethyl substituents

on the aryl groups) with that previously found for 11b (as well

as the new findings for 11a and 11c). While the two series

appear quite similar in some regards, particularly with respect

to one major trend, the dependence of reactivity on electronic

character, other phenomena appear quite different. In the next

few sections, we will try to account for and reconcile both the

differences and the similarities within the basic framework of

Scheme 10.

Isotopic Exchange and Kinetic Deuterium Isotope Effects.

It should first be noted that no discernible amount of deuterium

is detected in the methane liberated from the reaction between

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1389

A R T I C L E S

Zhong et al.

Scheme 12

Figure 15. Reaction coordinate for reactions between 10, 11, 12 and

benzene.

complexes 10 and 11: the former have KIE values around 2,

while the latter are close to unity. For several examples of 10,

we also measured the KIEs by intermolecular (1:1 C₆H₆:C₆D₆)

and intramolecular (1,3,5-C₆H₃D₃) competition reactions; this

approach would be much more complicated (by intermolecular

competition) or impossible (by intramolecular competition) for

11 because of the near-statistical isotope scrambling. We find

that the KIEs measured from the competition reactions are

slightly, but consistently, smaller than those measured by parallel

reactions. This observation may be accounted for by the fact

that more than one molecule of benzene may be involved in

the reaction (vide supra) (Scheme 10). Thus, for example, in

an intermolecular competition reaction of 10 with C₆D₆or C₆H₆,

initial C₆D₆activation followed by complete reversion to starting

material accompanied by deuterium incorporation into [Pt-CH₃]

happens occasionally; the subsequent conversion to 17 will give

deuterated methane even if it involves C₆H₆. In the parallel

method, such a sequence (participation of more than one

benzene molecule) will have no consequences, since only rates

and not degrees of deuteration are measured. Thus the competi-

tive method overestimates the apparent relative frequency of

C₆D₆activation.

The differences between 10 and 11 in behavior for both

isotope exchange and KIE are explicable in terms of Figure

15, analogous to Scheme 10 but with the assumption of a change

in the rate-determining step. For 10, to account for a significant

KIE, the rate-determining step must involve the C-H (C-D)

bond: either the actual C-H cleavage step, C f D, or (less

likely) the coordination of the C-H bond, B f C. That would

also explain the relatively low level of isotopic exchange, as

the barrier to dissociation of benzene from B (and, by extension,

that of methane from E) must be lower than at least one barrier

within the manifold of reactions that effect such exchange. This

case is represented by the free energy profile for 10 f 17 in

Figure 15. The highest point on the energy surface is transition

state Z^q.

reverse is true, only CH₃D will be produced. As Figure 1 shows,

for 4b the ratio of CH₃D to CH₄increases with increasing

acetonitrile concentration, implying that elimination of methane

is associative. Similar observations for 4e and 5b have been

reported recently by Tilset and co-workers.²⁷

When cations 10 are treated with C₆D₆, the liberated methane

includes isotopomers with more than one deuterium atom. This

is consistent with the mechanism of Scheme 10, if reversible

interconversions among intermediates C-E and the accompa-

nying rearrangements to C′ and E′ occur at a rate at least

comparable to those of dissociation of methane from E and C.

However, in contrast to 11b, where isotope scrambling between

the methane and platinum-phenyl groups is essentially statisti-

cal,¹¹for 10 a much smaller degree of scrambling is found. The

ratio of approximately 2:1 CH₃D:CH₂D₂for 10b indicates that

E loses methane on the order of twice as fast as it reverts to C.

Up to 20% of the unreacted starting material becomes

deuterated, giving [Pt-CH₂D], during reactions between 10 and

C₆D₆, which implies that some of the time intermediate E, which

will have coordinated CH₃D after a single C-H activation/

cleavage sequence, reverts all the way back to starting materials.

The relatively low levels of deuterium incorporation into the

platinum methyl group are consistent with the previous finding,

on the basis of protonolysis of a mixed [Pt(Me)(Ph)] complex,

that the rate for benzene elimination is ∼4-5 times slower than

that of methane elimination, which translates to a ∆∆G^qof

∼0.8-0.9 kcal mol^-1at room temperature.¹¹This observation

also implies that some of the evolved methane will come from

complexes that have interacted with more than one molecule

of C₆D₆, which may have consequences for KIE measurements

(vide infra).

In contrast, for 11 the highest point on the surface must be

transition state Y^q, which governs the initial coordination of

benzene. Since this coordination involves a π bond of benzene,

no primary KIE at all would be anticipated. Furthermore, the

complete statistical isotope exchange implies that once the

manifold of intermediates C-E has been entered, loss of

benzene or methane is much slower than interconversions among

those intermediates. The differences in entropy of activation,

strongly negative for 11b, very small for 10b, also seem

Kinetic deuterium isotope effects (KIEs) were measured for

all complexes 10-12 by comparing the rates of disappearance

of starting material in the presence of C₆H₆versus those with

C₆D₆under otherwise identical conditions. The most striking

observation in Table 3 is the difference in behavior between

(27) Johansson, L.; Tilset, M. J. Am. Chem. Soc. 2001, 123, 739-740.

9

1390 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

Figure 16. Reaction coordinate for reactions between 11a and 11c and benzene, showing that the differences in rate within a series arise from ground-state

energy differences for the aquo/methyl cations (Aii). Similar profiles apply to 10 and 12, with the highest energy transition states being Z^q, not Y^q.

consistent with the former, but not the latter, involving rate-

determining associative (vide infra) substitution.

determining step; (2) the intermediates D thus formed are

formally Pt(IV), more electron-deficient than the Pt(II) cations;

and (3) the transition state Z^qleading to that intermediate should

have at least partial higher oxidation state character. Hence more

strongly electron-donating ligands can better stabilize the higher

oxidation state intermediates/transition states and accelerate the

reaction.

Why should this rate-determining step switch operate? A

steric explanation appears the most probable. Coordination of

benzene in η²-C,C mode places the benzene molecule right in

the coordination plane, subject to steric interactions with the

other ligands. For 11, with 2,6-dimethyl-substituted aryl groups,

these interactions can be expected to be more significant than

for 10, which has no substituents in the 2,6-positions. Hence in

11 there will be relative destabilization of both the intermediate

B itself and, presumably, the transition state (Y^q) leading thereto,

as compared to 10.

The KIEs for 12 are considerably larger than those for 10

(3.6-5.9 vs ∼2). One might argue that 12 is even less sterically

crowded than 10, by virtue of the missing backbone methyl

groups, which would lower transition state Y^qstill further

(Figure 15). This might suggest that 12 gives a better measure

of the inherent KIE for rate-determining C-H bond activation,

whereas for 10 the energies of Y^qand Z^qare sufficiently close

that C-H activation is not completely rate-determining, and the

measured KIE values are hybrids. However, given that the rates

of the reactions of 12 with benzene are slower than steric

considerations would lead us to expect, this conclusion must

remain tentative for the present.

Electronic Effects on Rates. Within each series 10-12 the

changes in substituents (in the 3,5-positions for 10 and 12 and

in the 4-positions for 11) are far enough removed from the metal

center that steric parameters should change little, and thus the

observed trend within each series should reflect a purely

electronic effect. For both series, as the Pt(II) centers become

more electron-rich (as measured by lower CO stretching

frequencies in the analogous carbonyl cations), benzene C-H

bond activation becomes faster. In principle, this trend could

be explained in a number of ways. For cations 10 and 12, for

example, one might argue that (1) C-H activation is the rate-

On the other hand, a closely parallel trend is observed also

for cations 11, for which C-H bond activation does not appear

to be rate-determining, but rather the replacement of TFE solvent

by benzene. It is conceivable that the replacement might

coincidentally follow the same order; that is, that transition state

Y^qis also stabilized by electron-donating ligands. While

acetonitrile self-exchange rates follow the opposite trend, faster

at more electron-deficient metal centers (vide infra), the

electronic demands for the coordination of benzene versus

acetonitrile as the displacing ligand might well be different.

However, it would be much more satisfying to find a single

explanation that accounts simultaneously for all the series.

Since the transition states for the series appear to be quite

different, the simplest explanation for common behavior is that

we are dealing primarily with a ground-state effect. That is, the

most important effect of changing the electronic properties of

the diimine ligand is upon the relative stability of the aquo

complexes [(ArNdCR-CRNdAr)Pt(Me)(H₂O)]⁺(R ) Me or

H, Aii in Scheme 10). If the energies of either of the transition

states Y^qor Z^qare less sensitive to changes in the diimine ligand

than that of Aii, then the rates of benzene activation by series

10/12 and 11 will exhibit the same trend even though they have

different rate-determining steps (shown for series 11 in Figure

16; similar profiles would be obtained for 10 and 12, with the

highest energy transition states being Z^q, not Y^q).

If we further postulate that the same is true for the energy of

solvento complex Ai, then the equilibrium constants between

Aii and Ai would be expected to vary with ligand in much the

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1391

A R T I C L E S

Zhong et al.

alcohols, probably due to differences in solvation energies of

the cations.²⁸

Table 5 shows that the more electron-poor complexes (13d

vs 13b, 14c vs 14b) exhibit higher exchange rate constants for

both the solvent-assisted (k_s) and direct (k₂) terms. This is the

opposite trend from that found for benzene activation and

probably indicates stabilization of a five-coordinate transition

state in the associative substitution. The most closely corre-

sponding species in Figure 16 would be the transition state for

substitution of water by TFE (X^q); note that the energy of that

species has no consequence whatsoeVer on the rate of benzene

activation, if the arguments represented by that energy diagram

are correct. Hence there is no inconsistency between the two

opposing trends. Comparison of acetonitrile exchange for 14b

and 13d (whose corresponding carbonyl cations have very

similar ν_COvalues) reveals a moderate steric effect as well,

around a factor of 4, not quite so large as the steric retardation

that appears to operate in benzene activation.

Reactions of Alkylaromatics. Our results combined with

those of Tilset²⁰suggest an inherent reactivity preference for

aromatic over benzylic C-H activation, which can however be

readily overridden by steric effects. Thus the relatively un-

crowded 10b reacts with p-xylene to give 20b, the product of

reaction at an aryl C-H, whereas more crowded 11b gives

primarily benzylic activation. With mesitylene, still more

sterically encumbered, 10b gives a mixture of products,

somewhat favoring benzylic activation (21b), and the aromatic

activation product 22b appears, from its accelerated decomposi-

tion to (µ-OH)₂dimer 16b, to be significantly destabilized. The

substrate-derived crowding in these two cases may also cause

a shift to rate-determining (or partially so) substrate coordination,

as the KIEs decrease significantly with steric bulk.

Figure 17. A log-log plot of the observed rate constants between C₆D₆

and methyl cations 10 and 11 vs the aquo/solvento equilibrium constants

for the corresponding methyl solvento a aquo cations 10i a 10ii and 11i

a 11ii.

same way as the rate constants. This is in fact the case, as shown

in Figure 17; the plots of log(k_obs) versus log(K_eq) give

reasonably straight lines, with slope close to unity, for the three

sets of data.

While this interpretation is consistent with the most prominent

features of the ligand-reactivity relationship, it is far from

conclusive, and some observations remain unexplained, par-

ticularly the low reactivity of 12. Figure 14 shows that both

12b and 11a are less reactive than 10d, though all have

comparable electron density. The difference between 11a and

10d can be explained on steric grounds as the former has 2,6-

dimethyl-substituted aryl groups, but 12b should be, if anything,

less crowded than 10d. Apparently the replacement of backbone

methyl groups with hydrogens has an additional, as yet

unexplained effect here, as well as on the KIEs (vide supra).

Acetonitrile Exchange. A rate expression of k_ex) k₁+

k₂[CD₃CN] has been obtained for acetonitrile isotopic exchange

reactions. The k₁term could represent either a unimolecular

dissociative pathway or a solvent-assisted associative pathway.

The highly negative entropy of activation for the k₁term seems

inconsistent with a dissociative pathway and indicative of a

solvent-assisted pathway, though this is not definitive; solvation

of the cations can play an important role. However, both terms

exhibit similar entropies of activation, suggesting that both

pathways operate by the same mechanism. That is, the isotopic

acetonitrile exchange consists of both a solvent-assisted as-

sociative pathway and a direct attack associative pathway.

Although the k₁values vary considerably as the solvent is

changed from methanol to ethanol to 2-propanol, when divided

by bulk solvent concentrations the resulting k_svalues are all

about the same (Table 3). Apparently the solvent term is not

very sensitive to the steric bulk of the solvent. On the other

hand, k₁for TFE is essentially zero, presumably reflecting the

very low basicity of that solvent. The second-order rate constants

for direct acetonitrile attack (k₂) do vary across the entire set of

The C-H activation chemistry of purely aliphatic hydrocar-

bons will be reported later.²⁹

Conclusions

The results reported here, combined with some earlier find-

ings, support three main conclusions: (1) For the reaction of

benzene with cationic Pt(II) complexes bearing 3,5-disubstituted

aryl diimine ligands (10), the rate-determining step is C-H bond

activation; whereas for the more sterically crowded analogues

with 2,6-dimethyl-substituted aryl groups (11), benzene coor-

dination becomes rate-determining. This switch is manifested

in distinctly different isotope scrambling and KIE patterns. (2)

The more electron-rich the ligand is, as assayed by the CO

stretching frequency of the corresponding carbonyl cationic

complex, the faster the rate of C-H bond activation. This at

first sight appears to be at odds with the common description

of this class of reaction as electrophilic. However, the fact that

the same trend is observed for the two different series of

complexes, which have different rate-determining steps, suggests

that this finding does not reflect the actual C-H bond activation

process, but rather reflects only the relative ease of benzene

displacing a ligand to initiate the reaction, which in turn appears

to be mostly a ground-state effect. It is hence not possible to

say much, if anything, about the “inherent” nature of C-H

activation on the basis of these results. (3) Several lines of

evidence suggest that associative substitution pathways operate

(28) Romeo, R.; Minniti, D.; Lanza, S. Inorg. Chem. 1980, 19, 3663-3668.

(29) Zhong, H. A.; Labinger, J. A.; Bercaw, J. E., unpublished results.

9

1392 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

MeOH. To this yellow solution, two drops of formic acid were added,

and a very pale yellow precipitate formed within 15 min. The mixture

was stirred at room temperature for 14 h, cooled, filtered, washed with

cold methanol (2 × 5 mL), and dried over an aspirator for 3 h. 1b was

here to get the hydrocarbon substrate into, and out of, the

coordination sphere. While associative substitution does pre-

dominate in the chemistry of square-planar Pt(II), there are

situations where dissociative mechanisms become preferred,³⁰

and one might think that the present cases, involving (presum-

ably) very weakly bonded arenes and alkanes, would fall into

that category. However, acetonitrile exchange seems to be

clearly associative. More to the point, the fact that addition of

acetonitrile, a better nucleophile than solvent TFE, suppresses

isotopic scrambling between benzene and methyl groups

strongly implies that the replacement of coordinated methane

is associative. Analogous behavior has recently been observed

for the displacement of coordinated arenes.²⁰From the principle

of microscopic reversibility, we infer that the displacement of

solvent (water, acetonitrile) by hydrocarbon also proceeds

associatively.

1

isolated as an extremely pale yellow powder (0.95 g, 87%). H NMR

(500 MHz, C₆D₆): δ ) 1.32 (s, 36H, C(CH₃)₃), 2.33 (s, 6H NdC-

4

CH₃), 6.90 (d, J_H-H) 1.8 Hz, 4H, Ar-H), 7.35 (t, J_H-H) 1.8 Hz,

2H, Ar-H). ¹³C {¹H} NMR (125 MHz, C₆D₆): δ ) 15.78 (NdC-

CH₃), 31.97 (C(CH₃)₃), 35.40 (C(CH₃)₃), 114.27 (o-Ar-C), 118.19 (p-

Ar-C),151.860, 152.24 (Ar-C), 168.57 (NdC-CH₃). ESMS, Calcd

for C₃₂H₄₈N₂H ([M + H]⁺): 461.3896. Found: 461.3906.

1,4-Bis(3,4,5-trimethoxyphenyl)-2,3-dimethyl-1,4-diaza-1,3-buta-

diene (^{OMe Ar}DAB^Me, 1c). 3,4,5-Trimethoxyaniline (0.9875 g, 5.39

3

mmol) and 2,3-butanedione (0.225 g, 2.61 mmol) were dissolved in

30 mL of MeOH. To this yellow solution was added several drops of

formic acid, and a bright yellow precipitate formed overnight. The

mixture was stirred at room temperature for 14 h, cooled, filtered,

washed with cold methanol (2 × 5 mL), and dried over an aspirator

for 3 h. 1c was isolated as a bright yellow powder (0.757 g, 70%). ¹H

NMR (500 MHz, C₆D₆): δ ) 2.32 (s, 6H, NdC-CH₃), 3.40 (s, 12H,

OCH₃), 3.90 (s, 6H, OCH₃), 6.19 (s, 4H, Ar-H). ¹³C {¹H} NMR (125

MHz, C₆D₆): δ ) 15.91 (NdC-CH₃), 56.15 (OCH₃), 61.06 (OCH₃),

97.46 (o-Ar-C), 136.26, 147.90, 155.06 (Aryl C’s), 169.04 (NdC-

CH₃). ESMS, Calcd for C₂₂H₂₈N₂O₆Na ([M + Na]⁺): 439.1845.

Found: 439.1848.

The implications of these findings with respect to the ultimate

goal, the development of a practical, selective alkane function-

alization catalyst, remain the subject of ongoing research in our

labs.

Experimental Section

General Considerations. All moisture-sensitive compounds were

manipulated using standard vacuum line, Schlenk or cannula techniques,

or in a drybox under a nitrogen atmosphere. Argon and dinitrogen gases

were purified by passage over columns of MnO on vermiculite and

activated molecular sieves. Trifluoroethanol was purchased from

Aldrich, purified and dried over a mixture of CaSO₄/NaHCO₃, then

either vacuum distilled or distilled under argon and stored over activated

molecular sieves under vacuum. Trifluoroethanol-d₃was purchased from

Aldrich, stored over activated molecular sieves and a small amount of

NaHCO₃under vacuum, and vacuum distilled into oven-dried J-Young

NMR tubes for kinetic studies. Benzene and benzene-d₆were vacuum

distilled from sodium benzophenone ketyl shortly before kinetic runs

and stored over activated molecular sieves. Toluene was vacuum

distilled from sodium benzophenone ketyl. Triethylamine was distilled

from CaH₂and stored under Ar. 3,5-di-tert-butyl-4-methoxyaniline and

the corresponding diimine ligand 1a were synthesized by Dr. Joseph

Sadighi. Bis(dimethyl(µ-dimethyl sulfide)platinum(II) was prepared

according to literature procedure.¹⁷All other solvents and reagents were

used as received without further purification.

NMR spectra were recorded on a GE QE300 (¹H, 300.1 MHz), a

Varian INOVA 500 (¹H, 499.852 MHz, ¹³C, 125.701 MHz), or a Varian

Mercury 300 (¹H, 299.8 MHz, ¹⁹F, 282.081 MHz, ¹³C, 75.4626 MHz)

spectrometer. IR spectra were recorded on a Perkin-Elmer 1600 series

FTIR spectrometer. Mass spectra were measured on a Micromass LCT

instrument with a Z-spray source at University of California Irvine,

using electrospray orthogonal acceleration time-of-flight technique.

Under the analytical conditions, protonolysis occurs at the platinum

center in the platinum dimethyl complexes 4-6, which were conse-

quently detected as the cationic species [(N-N)Pt(Me)(NCMe)]⁺

(NCMe was the solvent used in the analysis). Elemental analyses were

performed at Midwest MicroLab LLC. A number of samples gave

analytical results lower than the expected values (these were generally

very small samples that may have been contaminated with either a small

amount of silica (from the filtration frit on which they were isolated)

or TFE solvent), but all analyzed correctly in the electrospray mass

spectrum.

1,4-Bis(3-methoxy-5-(trifluoromethyl)phenyl)-2,3-dimethyl-1,4-

diaza-1,3-butadiene (^{OMeCF Ar}DAB^Me, 1d). 3-Methoxy-5-(trifluorom-

3

ethyl)aniline (1.27 g, 6.65 mmol) and 2,3-butanedione (0.286 g, 3.32

mmol) were dissolved in 30 mL of MeOH. To this yellow solution

was added several drops of formic acid, and the mixture was stirred at

room temperature for 2 days without forming any precipitation. The

volatiles were then removed on a rotavap, and 6.8 g of activated 5 Å

molecular sieves were added to a toluene solution of the above residues.

The mixture was heated at 80-90 °C for two nights. The molecular

sieves were then filtered away and washed with methylene chloride (4

× 10 mL). The volatiles were evaporated, and 20 mL of methanol was

added to the residue. After stirring at room temperature for an hour,

the insolubles were collected, washed with cold methanol (2 × 5 mL),

and dried over an aspirator for 3 h. 1d was isolated as an off-white

1

powder (0.75 g, 52%). H NMR (500 MHz, C₆D₆): δ ) 1.90 (s, 6H,

NdC-CH₃), 3.13 (s, 6H, OCH₃), 6.43 (m, 2H, Ar-H), 6.72 (m, 2H,

Ar-H), 6.90 (m, 2H, Ar-H). ¹³C {¹H} NMR (125 MHz, C₆D₆): δ )

15.52 (NdC-CH₃), 55.33 (OCH₃), 106.38 (³J_C-F) 3.8 Hz, Ar-C),

108.67 (³J_C-F) 3.67 Hz, Ar-C), 108.79 (⁴J_C-F) 0.98 Hz, Ar-C),

125.63 (¹J_C-F) 270.35 Hz, CF₃), 133.17 (²J_C-F) 32.3 Hz, Ar-C),-

153.56, 161.52 (Ar-C), 169.34 (NdC-CH₃). ¹⁹F NMR (282 MHz,

C₆D₆): δ ) -64.81. ESMS, Calcd for C₂₀H₁₈N₂O₂F₆H ([M + H]⁺):

433.1351. Found: 433.1357.

1,4-Bis(2,4,6-trimethylphenyl)-2,3-dimethyl-1,4-diaza-1,3-butadi-

ene (^{Me Ar}DAB^Me, 2a). 2,4,6-Trimethylaniline (13.521 g, 100 mmol)

3

and 2,3-butanedione (4.307 g, 50 mmol) were dissolved in 50 mL of

MeOH. To this yellow solution was added several drops of formic acid,

and a bright yellow precipitate formed overnight. The mixture was

stirred at room temperature for 24 h, filtered, washed with methanol

(2 × 5 mL), and dried over an aspirator for 4 h. 2a was isolated as a

1

bright yellow powder (14.78 g, 92%). H NMR (500 MHz, C₆D₆): δ

) 2.00 (s, 12H, o-CH₃) 2.07 (s, 6H, p-CH₃), 2.22 (s, 6H, NdC-CH₃),

6.85 (s, 4H, Ar-H). ¹³C {¹H} NMR (125 MHz, C₆D₆): δ ) 16.08 (Nd

C-CH₃), 18.24 (o-CH₃), 21.21 (p-CH₃), 124.91 (o-Ar-C), 129.44 (m-

Ar-C), 132.70 (p-Ar-C), 147.09 (ipso-Ar-C), 168.76 (NdC-CH₃).

ESMS, Calcd for C₂₂H₂₈N₂H ([M + H]⁺): 321.2331. Found: 321.2323.

1,4-Bis(2,6-dimethyl-4-bromo-phenyl)-2,3-dimethyl-1,4-diaza-1,3-

1,4-Bis(3,5-di-tert-butylphenyl)-2,3-dimethyl-1,4-diaza-1,3-buta-

diene (^{tBu Ar}DAB^Me, 1b). 3,5-Di-tert-butylaniline (0.9732 g, 4.74 mmol)

2

butadiene (^{Me BrAr}DAB^Me, 2c). To a mixture of 2,6-dimethyl-4-

2

and 2,3-butanedione (0.204 g, 2.36 mmol) were dissolved in 15 mL of

bromoaniline (5.103 g, 25.5 mmol) and 2,3-butanedione (1.078 g, 12.52

mmol) was added 100 mL of methanol. The aniline was not very soluble

in methanol. To this yellow solution was added several drops of formic

(30) Romeo, R. Comments Inorg. Chem. 1990, 11, 21-57.

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1393

A R T I C L E S

Zhong et al.

acid. The mixture was stirred at room temperature for 48 h, filtered,

and washed with methanol (5 × 5 mL). The filtrate was concentrated

to ∼30 mL, filtered, and washed. The insolubles were combined and

dried over an aspirator for 4 h. 2c was isolated as a pale yellow powder

(4.84 g, 86%). ¹H NMR (500 MHz, C₆D₆): δ ) 1.71 (s, 12H, o-CH₃),

1.82 (s, 6H, NdC-CH₃), 7.14 (s, 4H, Ar-H). ¹³C {¹H} NMR (125

MHz, C₆D₆): δ ) 16.02 (NdC-CH₃), 17.78 (o-CH₃),116.725 (Br-

Ar-C), 127.33 (o-Ar-C), 131.449 (m-Ar-C), 148.11 (ipso-Ar-C),

168.84 (NdC-CH₃). ESMS, Calcd for C₂₀H₂₂N₂Br₂H ([M + H]⁺):

449.0228 (⁷⁹Br), 451.0209 (⁸¹Br). Found: 449.0231, 451.0222.

3H, OCH₃), 6.69 (br s, 1H), 6.84 (br s, 1H), 6.9-7.2 (overlapping peaks,

9H, phosphine o,p-Ar-H), 7.22 (br s, 1H, Ar-H), 7.70 (ddd, 6H, J )

1.5 Hz, 7.5 Hz, 12 Hz, phosphine m-Ar-H). Iminophosphorane (1.16

g, 2.56 mmol), glyoxal trimer (90 mg, 0.428 mmol), and 2 g of activated

molecular sieves were added to a reaction flask containing dry THF.

The mixture was heated at ∼80 °C overnight. After 17 h, examination

of an aliquot of the mixture indicated little progress in reaction. Another

100 mg of glyoxal trimer was added to the mixture, the reaction flask

was closed off, and the reaction temperature was raised to ∼100 °C.

After 14 h, the reaction mixture was cooled, and another aliquot of the

1

1,4-Bis(3,5-di-tert-butylphenyl)-1,4-diaza-1,3-butadiene (^{tBu Ar}DAB^H,

2

mixture was taken. The conversion was ∼57%, but the H NMR also

indicated the existence of a third species (∼5% of the total product)

besides the starting iminophosphorane and the diimine. Since it was

not clear whether this species was merely a side-product or the

decomposition product of the diimine, the reaction mixture was worked

up immediately, and 200 mg of yellow solids was collected (38.5%,

contains ∼5% starting iminophosphoranes). ¹H NMR (300 MHz, C₆D₆):

δ ) 3.06 (s, 6H, OCH₃), 6.76 (br s, 2H, Ar-H), 6.98 (br s, 2H, Ar-

H), 7.09 (br s, 2H, Ar-H), 7.99 (s, 2H, NdC-H). ESMS, Calcd for

C₁₈H₁₄N₂O₂F₆H ([M + H]⁺): 405.1038. Found: 405.1051.

2

3a). Br₂(0.25 mL, 4.85 mmol) was added to a 30 mL CH₂Cl₂solution

of PPh₃(1.30 g, 4.96 mmol) at 0 °C under Ar. After the red color

disappeared, the solvent was removed by rotavap, leaving behind a

white powder. A toluene solution of 3,5-di-tert-butylaniline (1.00 g,

4.87 mmol) was added to the above white powder under Ar to form an

orange solution with white precipitate. Triethylamine (2 mL, 14.4 mmol)

was added to the mixture, and the solution turned light yellow with

more white precipitate appearing. The mixture was heated to 80-90

°C under argon, and the reaction was allowed to proceed overnight.

After 14 h, the reaction flask was cooled to room temperature, the

mixture was filtered to remove NEt₃HBr, and the insolubles were

washed with petroleum ether (3 × 20 mL). The filtrate was then

concentrated, and dry heptane was added to the oily residues. The

iminophosphorane (1.4 g, 62%) was isolated by recrystallization from

heptane. ¹H NMR (300 MHz, C₆D₆): δ ) 1.31 (s, 18H, C(CH₃)₃), 6.9-

7.1 (overlapping peaks, 12H, Ar-H), 7.80 (m, 6H, Ar-H). Dry THF

was added to the iminophosphorane (1.05 g, 2.26 mmol), glyoxal trimer

(77 mg, 0.37 mmol), and 1 g of 4 Å molecular sieves under Ar, and

the mixture was refluxed overnight. After cooling to room temperature,

THF was removed, and the residue was dissolved in 10 mL of CH₂-

Cl₂. The yellow solution was filtered through a pad of silica gel and

Celite to remove triphenylphosphine oxide. The silica gel was washed

with additional CH₂Cl₂. All CH₂Cl₂solution was combined, and the

solvent was removed. MeOH was added to the residue, and the yellow

insolubles were collected (350 mg, 72%). ¹H NMR (500 MHz, C₆D₆):

(

^{tBu OMeAr}DAB^Me)PtMe₂(4a). 1a (200 mg, 0.384 mmol) and bis-

(dimethyl(µ-dimethyl sulfide)platinum(II)) (109.9 mg, 0.191 mmol)

were added to a receiving flask equipped with a 180° valve. The flask

was cooled to -78 °C and evacuated, and 10 mL of toluene was vacuum

transferred onto the solids. The dry ice bath was then removed, and

the reaction mixture was allowed to warm gradually to room temper-

ature and was left stirring at room temperature for 3 nights (NB: The

reaction was probably done after 10 h). The solution turned quickly

from pale yellow to deep purple upon warming to room temperature.

At the end of the reaction, there was a considerable amount of purple

solid suspended in the purple solution. Toluene was removed in vacuo,

and 15 mL of petroleum ether was added to the purple solid residues.

The insolubles were collected, washed with petroleum ether (4 × 5

mL), and dried over an aspirator for several hours. 4a was collected as

1

a purple-red solid (265 mg, 92.5%). H NMR (500 MHz, CD₂Cl₂): δ

2

) 0.93 (s, 6H, J_Pt-H) 86.7 Hz, Pt-CH₃), 1.452 (s, 36H, C(CH₃)₃),

4

δ ) 1.27 (s, 37H, C(CH₃)₃), 7.35 (d, J_H-H) 1.5 Hz, 4H, Ar-H),

1.48 (s, 6H, NdC-CH₃), 3.73 (s, 6H, OCH₃), 6.865 (s, 4H, Ar-H).

¹³C {¹H} NMR (125 MHz, CD₂Cl₂): δ ) -13.21 (¹J_Pt-C) 789 Hz,

Pt-CH₃), 21.30 (NdC-CH₃), 32.37 (C(CH₃)₃), 36.45 (C(CH₃)₃), 64.91

(OCH₃), 120.68 (o-Ar-C), 142.96, 144.46, 157.87 (Ar-C), 171.08 (Nd

C-CH₃). The compound slowly decomposes in methylene chloride.

ESMS, Calcd for C₃₅H₅₅N₂O₂PtCH₃CN ([M - Me + NCMe]⁺):

770.4156 (¹⁹⁴Pt), 771.4180 (¹⁹⁵Pt), 772.4192 (¹⁹⁶Pt). Found: 770.4173,

771.4180, 772.4189. Anal. Calcd for C₃₆H₅₈N₂O₂Pt: C, 57.97; H, 7.84;

N, 3.76. Found: C, 52.2/53.53; H, 7.00/7.24; N, 3.33/3.34.

2

7.48 (t, ⁴J_H-H) 1.5 Hz, 2H, Ar-H), 8.67 (s, 2H, NdC-H). ¹³C {¹H}

NMR (125 MHz, C₆D₆): δ ) 31.84 (C(CH₃)₃), 35.38 (C(CH₃)₃), 116.38

(o-Ar-C), 122.188 (p-Ar-C), 151.60 (ipso-Ar-C), 152.622 (m-Ar-

C), 160.18 (NdC-H). ESMS, Calcd for C₃₀H₄₄N₂H ([M + H]⁺):

433.3583. Found: 433.3584.

1,4-Bis(3,4,5-trimethoxyphenyl)-1,4-diaza-1,3-butadienene(^{OMe Ar}DAB^H,

3

3b). The iminophosphorane was synthesized similarly starting with

3,4,5-trimethoxyaniline (2 g, 0.0109 mol), PPh₃Br₂(4.61 g, 0.0109 mol),

and NEt₃(4 mL, 0.0287 mol) in 10 mL of toluene. A 2:1 heptane:

toluene solution was added to the oily residue to afford 3.81 g of the

(

^{tBu Ar}DAB^Me)PtMe₂(4b). 4b was synthesized similarly from 1b (160

mg, 0.348 mmol) and bis(dimethyl(µ-dimethyl sulfide)platinum(II))

(100 mg, 0.174 mmol). 4b was collected as a purple-red solid (220

1

desired product (79%). H NMR (300 MHz, C₆D₆): δ ) 3.53 (s, 6H,

1

2

m-OCH₃), 3.61 (s, 3H, p-OCH₃), 5.92 (s, 2H, Ar-H), 7.45-7.60

(overlapping peaks, 9H, phosphine o- and p-Ar-H), 7.70-7.80 (m,

6H, Ar-H). 3b was prepared similarly to 3a with 3.8 g of iminophos-

phorane, 330 mg of glyoxyal trimer, and 1 g of 4 Å molecular sieves

in refluxing THF overnight. A substantial amount of iminophosphorane

was recovered, and 300 mg of product was isolated (20%). The ligands

mg, 92%). H NMR (500 MHz, CD₂Cl₂): δ ) 0.90 (s, 6H, J_Pt-H

)

85.3 Hz, Pt-CH₃), 1.36 (s, 36H, C(CH₃)₃), 1.48 (s, 6H, NdC-CH₃),

3

6.83 (d, J_H-H) 1.5 Hz, 4H, o-Ar-H), 7.32 (t, J_H-H) 1.5 Hz, 4H,

p-Ar-H). ¹³C {¹H} NMR (125 MHz, CD₂Cl₂): δ ) -13.13 (Pt-CH₃),

21.24(NdC-CH₃), 31.69 (C(CH₃)₃), 35.49 (C(CH₃)₃), 116.76 (o-Ar-

C), 120.18 (p-Ar-C), 147.68, 151.98 (Ar-C), 171.03 (NdC-CH₃).

The compound slowly decomposes in methylene chloride. ESMS, Calcd

for C₃₃H₅₁N₂PtCH₃CN ([M - Me + NCMe]⁺): 710.3945 (¹⁹⁴Pt),

711.3969 (¹⁹⁵Pt), 712.3980 (¹⁹⁶Pt). Found: 710.3945, 711.3951, 712.3969.

Anal. Calcd for C₃₄H₅₄N₂Pt: C, 59.54; H, 7.94; N, 4.08. Found: C,

58.29/58.95; H, 7.76/7.86; N, 4.07/4.01.

1

isolated contained 5% phosphine-containing complex. H NMR (300

MHz, CD₂Cl₂): δ ) 3.80 (s, 6H, p-OCH₃), 3.87 (s, 12H, m-OCH₃),

6.604 (s, 4H, Ar-H), 8.40 (s, 2H, NdC-H). ESMS, Calcd for

C₂₀H₂₄N₂O₆H ([M + H]⁺): 389.1713. Found: 389.1723.

1,4-Bis(3-methoxy-5-(trifluoromethyl)phenyl)-1,4-diaza-1,3-buta-

dienene (^{OMeCF Ar}DAB^H, 3c). The iminophosphorane was synthesized

(

^{OMe Ar}DAB^Me)PtMe₂(4c). 4c was synthesized similarly from 1c

3

similarly starting with 3-methoxy-5-trifluoromethyl-aniline (1.25 g,

0.00654 mol), PPh₃Br₂(2.76 g, 0.00654 mol), and NEt₃(2 mL, 0.0143

mol). After removing NEt₃HBr and solvent, 100 mL of 4:1 petroleum

ether:toluene mixture was added to the oily residue to yield 2.25 g

(70% after correcting for toluene content) of light tan solid. The product

(200 mg, 0.48 mmol) and bis(dimethyl(µ-dimethyl sulfide)platinum-

(II)) (138 mg, 0.24 mmol). 4c was collected as a purple-red solid (278

1

2

mg, 90%). H NMR (500 MHz, CD₂Cl₂): δ ) 1.05 (s, 6H, J_Pt-H

)

85.2 Hz, Pt-CH₃), 1.48 (s, 6H, NdC-CH₃), 3.82 (s, 6H, OCH₃), 3.84

(s, 12H, OCH₃), 6.24 (s, 4H, Ar-H). ¹³C {¹H} NMR (125 MHz, CD₂-

Cl₂): δ ) -13.29 (¹J_Pt-C) 798 Hz, Pt-CH₃), 21.40 (NdC-CH₃),

1

contains ∼8 wt % toluene. H NMR (300 MHz, C₆D₆): δ ) 3.18(s,

9

1394 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

(

^{OMe ArH}DAB)PtMe₂(6b). 6b was synthesized similarly from 3b (95

3

56.72 (OCH₃), 61.17 (OCH₃), 99.77 (o-Ar-C), 136.46, 144.12, 154.02

(Ar-C), 171.67 (NdC-CH₃). The compound slowly decomposes in

methylene chloride. ESMS, Calcd for C₂₃H₃₁N₂O₆PtCH₃CN ([M - Me

+ NCMe]⁺): 666.2075 (¹⁹⁴Pt), 667.2098 (¹⁹⁵Pt), 668.2107 (¹⁹⁶Pt).

Found: 666.2085, (¹⁹⁴Pt), 667.2090, 668.2101. Anal. Calcd for

C₂₄H₃₄N₂O₆Pt: C, 44.93; H, 5.34; N, 4.37. Found: C, 44.02; H, 5.22;

N, 4.16.

mg, 0.245 mmol) and bis(dimethyl(µ-dimethyl sulfide)platinum(II)) (70

mg, 0.122 mmol). The petroleum ether insolubles were collected,

washed with benzene/methylene chloride (to remove triphenylphosphine

oxide contained in 3b), diethyl ether (20 mL), and petroleum ether (4

× 20 mL), and dried over an aspirator for several hours. 6b was

collected as a dark green/tan solid (100 mg, 67%). ¹H NMR (500 MHz,

2

^{OMeCF Ar}DAB^Me)PtMe₂(4d). 4d was synthesized similarly from 1d

3

CD₂Cl₂): δ ) 1.88 (s, 6H, J_Pt-H) 86 Hz, Pt-CH₃), 3.96 (s, 6H,

(

p-OCH₃), 4.01 (s, 12H, m-OCH₃), 6.82 (s, 4H, Ar-H), 9.52 (s, 2H,

³J_Pt-H) 27 Hz). ¹³C {¹H} NMR (125 MHz, CD₂Cl₂): δ ) -11.41

(Pt-CH₃), 54.07 (m-OCH₃), 56.45 (p-OCH₃), 101.08(o-Ar-C), 128.51

(p-Ar-C), 145.80 (ipso-Ar-C), 153.62 (m-Ar-C), 161.21 (NdC-

H). ESMS, Calcd for C₂₁H₂₇N₂O₆PtCH₃CN ([M - Me + NCMe]⁺):

638.1761 (¹⁹⁴Pt), 639.1785 (¹⁹⁵Pt), 640.1793 (¹⁹⁶Pt). Found: 638.1774,

639.1793, 640.1801.

(100 mg, 0.23 mmol) and bis(dimethyl(µ-dimethyl sulfide)platinum-

(II)) (66.5 mg, 0.116 mmol). 4d was collected as a purple-red solid

(130 mg, 85.4%). ¹H NMR (500 MHz, CD₂Cl₂): δ ) 1.07 (s, 6H, ²J_Pt-H

) 87.2 Hz, Pt-CH₃), 1.41 (s, 6H, NdC-CH₃), 3.900 (s, 6H, OCH₃),

6.78 (m, 2H, Ar-H), 6.89 (m, 2H, Ar-H), 7.10 (m, 2H, Ar-H). ¹³C

{¹H} NMR (125 MHz, CD₂Cl₂): δ ) -12.91 (Pt-CH₃), 21.60 (Nd

C-CH₃), 56.51 (OCH₃), 109.58 (³J_C-F) 3 Hz, Ar-C), 111.36 (³J_C-F

) 3 Hz, Ar-C), 111.55, 124.30 (¹J_C-F) 275 Hz, CF₃), 132.64 (²J_C-F

) 32.7 Hz, CF₃-C), 149.74, 161.01 (Ar-C), 171.869 (NdC-CH₃).

¹⁹F NMR (C6D6): δ ) -63.70. The compound slowly decomposes

in methylene chloride. ESMS, Calcd for C₂₁H₂₁N₂O₂F₆PtCH₃CN ([M

- Me + NCMe]⁺): 682.1400 (¹⁹⁴Pt), 683.1423 (¹⁹⁵Pt), 684.1431

3

(

^{OMeCF Ar}DAB^H)PtMe₂(6c). 3c (80 mg, 0.198 mmol) and bis-

(dimethyl(µ-dimethyl sulfide)platinum(II)) (containing some (SMe₂)₂-

PtMe₂monomer, 61 mg, ∼0.1 mmol) were added to a reaction flask

equipped with a 180° valve. Methylene chloride (5 mL) was added to

the mixture in air. The solution turned quickly from pale yellow to

green. The volatiles were partially removed after one-half an hour at

-20 °C to remove SMe₂. The flask was then back-filled with Ar, and

2 mL more of methylene chloride was added to the reaction mixture.

The mixture was then stirred at room temperature for another one-half

an hour to an hour. The solvent was then removed at -20 °C. Methylene

chloride (0.5 mL) and petroleum ether (5 mL) were added to the dark

solid residue, and the insolubles were collected and washed with

petroleum ether (3 × 2 mL). 6c, the dark green/black solid, was dried

over an aspirator for 2 h (104 mg, 84%). NB: This reaction did not

work in toluene; the green color initially formed faded after 4 h at

room temperature, and an intractable mixture was left behind. ¹H NMR

(500 MHz, CD₂Cl₂): δ ) 1.91 (s, 6H, ²J_Pt-H) 87 Hz, Pt-CH₃), 3.94

(s, 6H, OCH₃), 7.21 (s, 2H, Ar-H), 7.24 (s, 2H, Ar-H), 7.26 (s, 2H,

Ar-H), 9.64 (s, 2H, ³J_Pt-H) 27 Hz). ¹³C {¹H} NMR (125 MHz, CD₂-

Cl₂): δ ) -11.00 (Pt-CH₃), 56.27 (OCH₃), 111.36 (³J_C-F) 3.8 Hz,

Ar-C), 112.11 (³J_C-F) 3.6 Hz, Ar-C), 112.42, 123.81 (¹J_C-F) 272

Hz, CF₃), 132.49 (²J_C-F) 33.5 Hz, CF₃-C), 151.58, 160.57 (Ar-C),

162.52 (NdC-H). ESMS, Calcd for C₁₉H₁₇N₂O₂F₆PtCH₃CN ([M -

Me + NCMe]⁺): 654.1086 (¹⁹⁴Pt), 655.1110 (¹⁹⁵Pt), 656.1118 (¹⁹⁶Pt).

(

¹⁹⁶Pt). Found: 682.1382, 683.1411, 684.1393. Anal. Calcd for

C₂₂H₂₄N₂O₂F₆Pt: C, 40.19; H, 3.68; N, 4.26. Found: C, 38.65/38.88;

H, 3.53/3.53; N, 3.95/4.04.

^{Me Ar}DAB^Me)PtMe₂(5a). 5a was synthesized similarly from 2a

3

(

(111.6 mg, 0.348 mmol) and bis(dimethyl(µ-dimethyl sulfide)platinum-

(II)) (100 mg, 0.174 mmol). 5a was collected as a purple-red solid

(160 mg, 84%). ¹H NMR (500 MHz, CD₂Cl₂): δ ) 0.79 (s, 6H, ²J_Pt-H

) 85.7 Hz, Pt-CH₃), 1.24 (s, 6H, NdC-CH₃), 2.11 (s, 12H, o-CH₃),

2.37 (s, 6H, p-CH₃), 7.01 (s, 4H, Ar-H). ¹³C {¹H} NMR (125 MHz,

CD₂Cl₂): δ ) -14.92 (¹J_Pt-C) 797 Hz, Pt-CH₃), 17.48 (o-CH₃),

20.18 (NdC-CH₃), 21.14 (p-CH₃), 128.84 (m-Ar-C), 128.87 (o-Ar-

C), 135.77 (p-Ar-C), 143.77 (ipso-Ar-C), 170.79 (NdC-CH₃). The

compound slowly decomposes in methylene chloride. ESMS, Calcd

for C₂₃H₃₁N₂PtCH₃CN ([M - Me + NCMe]⁺): 570.2380 (¹⁹⁴Pt),

571.2403 (¹⁹⁵Pt), 572.2412 (¹⁹⁶Pt). Found: 570.2391, 571.2369, 572.2413.

Anal. Calcd for C₂₄H₃₄N₂Pt: C, 52.83; H, 6.28; N, 5.13. Found: C,

40.75/50.99; H, 4.79/6.17; N, 3.89/4.88.

^{Me BrAr}DAB^Me)PtMe₂(5c). 5c was synthesized similarly from 2c

2

(

(78.4 mg, 0.174 mmol) and bis(dimethyl(µ-dimethyl sulfide)platinum-

(II)) (50 mg, 0.087 mmol). 5c was collected as a purple-red solid (100

mg, 85%). ¹H NMR (500 MHz, CD₂Cl₂): δ ) 0.89 (s, 6H, ²J_Pt-H) 86

Hz, Pt-CH₃), 1.20 (s, 6H, NdC-CH₃), 2.13 (s, 12H, o-CH₃), 7.37 (s,

4H, Ar-H). ¹³C {¹H} NMR (125 MHz, CD₂Cl₂): δ ) -14.46 (¹J_Pt-C

) 800 Hz, Pt-CH₃), 17.44 (o-CH₃), 20.45 (NdC-CH₃), 119.21 (Br-

Ar-C), 131.01 (m-Ar-C), 131.55 (o-Ar-C), 145.23 (ipso-Ar-C),

170.96 (NdC-CH₃). The compound slowly decomposes in methylene

chloride. ESMS, Calcd for C₂₁H₂₅N₂Br₂PtCH₃CN ([M - Me +

NCMe]⁺): 699.0300 (⁷⁹Br⁷⁹Br¹⁹⁵Pt), 700.0275 (⁷⁹Br⁸¹Br¹⁹⁴Pt), 702.0276

(⁸¹Br⁸¹Br¹⁹⁴Pt), 703.0278 (⁸¹Br⁸¹Br¹⁹⁵Pt), 704.0283 (⁸¹Br⁸¹Br¹⁹⁶Pt).

Found: 699.0306, 700.0275, 702.0278, 703.0297, 704.0309. Anal.

Calcd for C₂₂H₂₈N₂Br₂Pt: C, 39.13; H, 4.18; N, 4.15. Found: C, 38.24/

38.03; H, 4.07/4.01; N, 3.93/3.88.

Found: 654.1078, 655.1118, 656.1111.

Bu

[(^t^OMeArDAB^Me)PtMe(CO)]⁺[BF₄]^-(7a). To a suspension of 4a

2

(11.4 mg, 0.015 mmol) in ∼2 mL of trifluoroethanol (TFE) was added

an aqueous solution of HBF₄(2 µL, 0.015 mmol). After stirring at

room temperature for a few minutes, a homogeneous orange solution

was obtained. The reaction flask was degassed and backfilled with 1

atm of CO. The color of the solution changed to bright yellow almost

instantaneously. The mixture was stirred under 1 atm of CO for 24 h.

TFE was then removed at -20 to 0 °C in vacuo, and a small amount

of petroleum ether was added to the oily residue to effect solidification

(scratching the reaction flask helps). PE was then removed on the

vacuum line, and the resulting yellow solid (7 mg, 54%) was dried for

30 min. The product contained a small amount of TFE, which was

hard to remove. ¹H NMR (300 MHz, CD₂Cl₂): δ ) 0.72 (s, 3H, ²J_Pt-H

) 66 Hz, Pt-CH₃), 1.45, 1.46 (s, 18H each, C(CH₃)₃), 2.36, 2.44 (s,

3H each, NdC-CH₃), 3.73, 3.73 (s, 3H each, OCH₃), 6.95, 7.18 (s,

2H each, Ar-H). ¹³C {¹H} NMR (75 MHz, CD₂Cl₂): δ ) -10.45

(Pt-CH₃), 20.63, 22.32 (NdC-CH₃), 31.90, 31.96 (C(CH₃)₃), 36.48,

36.52 (C(CH₃)₃), 64.91, 64.99 (OCH₃), 119.90, 120.40, 137.49, 142.12,

145.82, 146.04, 159.17, 159.88 (Ar-C), 155.52, 175.98, 189.08. IR

(CH₂Cl₂): ν(CO) ) 2103.5 cm^-1. The compound decomposes in

methylene chloride. Anal. Calcd for C₃₆H₅₅N₂O₃PtBF₄: C, 51.13; H,

6.56. Found: C, 42.25; H, 5.27.

^{tBu Ar}DAB^H)PtMe₂(6a). 6a was synthesized similarly from 3a (150

2

(

mg, 0.347 mmol) and bis(dimethyl(µ-dimethyl sulfide)platinum(II))

(100 mg, 0.174 mmol). 6a was collected as a dark green/tan black solid

(187 mg, 82%). ¹H NMR (500 MHz, CD₂Cl₂): δ ) 1.39 (s, 36H,

2

3

C(CH₃)₃), 1.65 (s, 6H, J_Pt-H) 85.4 Hz, Pt-CH₃), 7.27 (d, J_H-H

)

1.5 Hz, 4H, o-Ar-H), 7.50 (t, ³J_H-H) 1.5 Hz, 4H, p-Ar-H), 9.41 (s,

2H, J_Pt-H) 24.1 Hz). ¹³C {¹H} NMR (125 MHz, CD₂Cl₂): δ )

3

-11.12 (Pt-CH₃), 31.69 (C(CH₃)₃), 35.53 (C(CH₃)₃), 118.12 (o-Ar-

C), 122.90 (p-Ar-C), 149.91 (ipso-Ar-C), 152.30 (m-Ar-C), 161.87

(NdC-H). ESMS, Calcd for C₃₁H₄₇N₂PtCH₃CN ([M - Me +

NCMe]⁺): 682.3632 (¹⁹⁴Pt), 683.3655 (¹⁹⁵Pt), 684.3666 (¹⁹⁶Pt).

Found: 682.3641, 683.3646, 684.3667. Anal. Calcd for C₃₂H₅₀N₂Pt:

C, 58.43; H, 7.66; N, 4.26. Found: C, 56.16/56.16; H, 7.85/7.37; N,

3.99/4.00.

[(^{tBu Ar}DAB^Me)PtMe(CO)]⁺[BF₄]^-(7b). 7b was synthesized simi-

2

larly from 4b (15.7 mg, 0.023 mmol) and an aqueous solution of HBF₄

(3 µL, 0.023 mmol) in 3 mL of TFE. The yellow product contained a

small amount of TFE, which was hard to remove. ¹H NMR (300 MHz,

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1395

A R T I C L E S

Zhong et al.

2

CD₂Cl₂): δ ) 0.69 (s, 3H, J_Pt-H) 69 Hz, Pt-CH₃), 1.36, 1.37 (s,

18H each, C(CH₃)₃), 2.35, 2.45 (s, 3H each, NdC-CH₃), 6.91 (d, ⁴J_H-H

(5 µL, 0.038 mmol) in 8 mL of TFE. 10 mg of brown solid was obtained

(43%). The product contained a small amount of TFE, which was hard

to remove. ¹H NMR (300 MHz, CD₂Cl₂): δ ) 0.72 (s, 3H, ²J_Pt-H) 68

Hz, Pt-CH₃), 2.30, 2.42 (s, 3H each, NdC-CH₃), 2.38 (s, 12H,

m-CH₃), 6.70 (br, 2H, m-Ar-H), 6.89 (br, 2H, o-Ar-H), 7.04 (br, 1H,

4

) 1.8 Hz, 2H, Ar-H), 7.13 (d, J_H-H) 1.8 Hz, 2H, Ar-H), 7.45,

7.46 (overlapping t, 2H total, Ar-H). ¹³C {¹H} NMR (75 MHz, CD₂-

Cl₂): δ ) -10.42 (Pt-CH₃), 20.57, 22.34 (NdC-CH₃), 31.32

(overlapping C(CH₃)₃), 35.48, 35.54 (C(CH₃)₃), 115.49, 116.32, 122.37,

123.0, 153.15, 153.40 (Ar-C), 189.16. Resonances for two Ar-C and

p-Ar-H), 7.06 (br, 1H, p-Ar-H). IR (CH₂Cl₂): ν(CO) ) 2105.7 cm^-1

.

[(^{Me Ar}DAB^Me)PtMe(CO)]⁺[BF₄]^-(8a). 8a was synthesized similarly

from 5a (16.7 mg, 0.031 mmol) and an aqueous solution of HBF₄(4

µL, 0.031 mmol) in 5 mL of TFE. 7 mg of yellow powder was obtained

(35%). The product contained a small amount of TFE, which was hard

to remove. ¹H NMR (300 MHz, CD₂Cl₂): δ ) 0.58 (s, 3H, ²J_Pt-H) 66

Hz, Pt-CH₃), 2.16, 2.31 (s, 6H each, o-Ar-CH₃), 2.21, 2.41 (s, 3H

each, NdC-CH₃), 2.36 (overlapping s, 6H total, p-Ar-CH₃), 7.06,

7.07 (s, 4H total, Ar-H). ¹³C {¹H} NMR (75 MHz, CD₂Cl₂): δ )

-10.40 (Pt-CH₃), 18.04, 18.07 (o-Ar-CH₃), 19.84, 21.22, 21.28, 21.45

(p-Ar-C and NdC-CH₃), 127.72, 129.61, 129.85, 129.96, 137.63,

138.77, 139.28, 143.23 (Ar-C), 179.25, 191.39. IR (CH₂Cl₂): ν(CO)

) 2108.3 cm^-1. The compound decomposes in methylene chloride,

giving methane and Pt black. Anal. Calcd for C₂₄H₃₁N₂OPtBF₄: C,

44.66; H, 4.84; N, 4.34. Found: C, 34.36/34.14; H, 3.74/3.67; N, 3.08/

3.12. The sample appeared to contain silica from the frit; although the

C:H:N are all low, they are in the correct ratio.

3

NdC-CH₃were not located. IR (CH₂Cl₂): ν(CO) ) 2104.6 cm^-1

.

The compound decomposes in methylene chloride. Anal. Calcd for

C₃₄H₅₁N₂PtBF₄: C, 51.98; H, 6.54; N, 3.57. Found: C, 49.04/48.84;

H, 6.18/6.21; N, 3.35/3.32.

[(^{OMe Ar}DAB^Me)PtMe(CO)]⁺[BF₄]^-(7c). 7c was synthesized simi-

3

larly from 4c (14.7 mg, 0.023 mmol) and an aqueous solution of HBF₄

(3 µL, 0.023 mmol) in 3 mL of TFE. 13 mg of orange powder was

obtained (80%). The product contained a small amount of TFE, which

1

was hard to remove. H NMR (300 MHz, CD₂Cl₂): δ ) 0.84 (s, 3H,

²J_Pt-H) 66 Hz, Pt-CH₃), 2.34, 2.47 (s, 3H each, NdC-CH₃), 3.82,

3.86, 3.88 (s, 6H each, OCH₃), 6.37, 6.54 (s, 2H each, Ar-H). ¹³C

{¹H} NMR (75 MHz, CD₂Cl₂): δ ) -10.35 (Pt-CH₃), 20.66

(overlapping NdC-CH₃), 56.56, 56.86, 61.10, 61.20 (OCH₃), 98.25,

98.64, 99.42, 99.75, 138.41, 143.02, 154.31, 154.42(Ar-C), 163.53,

177.29, 191.04. IR (CH₂Cl₂): ν(CO) ) 2105.8 cm^-1. The compound

slowly decomposes in methylene chloride. Anal. Calcd for C₂₄H₃₁N₂O₇-

PtBF₄: C, 38.88; H, 4.21; N, 3.78. Found: C, 36.64/36.70; H, 4.00/

3.99; N, 3.34/3.26.

[(^{Me Ar}DAB^Me)PtMe(CO)]⁺[BF₄]^-(8b). 8b was synthesized similarly

2

from 5b (19.8 mg, 0.038 mmol) and an aqueous solution of HBF₄(5

µL, 0.038 mmol) in 5 mL of TFE. 16 mg of yellow powder was

obtained (76%). The product contained a small amount of TFE, which

[(^{OMeCF Ar}DAB^Me)PtMe(CO)]⁺[BF₄]^-(7d). 7d was synthesized

3

similarly from 4d (15.1 mg, 0.023 mmol) and an aqueous solution of

HBF₄(3 µL, 0.023 mmol) in 3 mL of TFE. 10 mg of yellow/orange

powder was collected (62%). The product contained a small amount

of TFE, which was hard to remove. H NMR (300 MHz, CD₂Cl₂): δ

) 0.77 (s, 3H, J_Pt-H) 68 Hz, Pt-CH₃), 2.33, 2.47 (s, 3H each,

1

was hard to remove. H NMR (300 MHz, CD₂Cl₂): δ ) 0.57 (s, 3H,

²J_Pt-H) 66 Hz, Pt-CH₃), 2.22, 2.37 (s, 6H each, o-Ar-CH₃), 2.24,

2.44 (s, 3H each, NdC-CH₃), 7.20-7.40 (m, 6H total, Ar-H). ¹³C

{¹H} NMR (75 MHz, CD₂Cl₂): δ ) -10.55 (Pt-CH₃), 18.12, 18.17

(o-Ar-CH₃), 19.88, 21.49 (NdC-CH₃), 128.09, 128.79, 129.27,

129.30, 129.44, 130.00, 140.42, 145.78(Ar-C), 162.69, 179.25, 190.39.

IR (CH₂Cl₂): ν(CO) ) 2109.6 cm^-1. ESMS, Calcd for C₂₂H₂₇N₂OPt

([M]⁺): 529.1750 (¹⁹⁴Pt), 530.1774 (¹⁹⁵Pt), 531.1782 (¹⁹⁶Pt). Found:

529.1764, 530.1771, 531.1781. Anal. Calcd for C₂₂H₂₇N₂OPtBF₄: C,

42.80; H, 4.41; N, 4.54. Found: C, 41.17/40.94; H, 4.44/4.19; N, 4.18/

4.16. The compound decomposes in methylene chloride.

1

2

NdC-CH₃), 3.93, 3.92 (s, 3H each, OCH₃), 6.93, 7.09, 7.20, 7.22

1

(overlapping broad m, 6H total, Ar-H). H NMR (300 MHz, TFE-

d₃): δ ) 0.83 (s, 3H, ²J_Pt-H) 66 Hz, Pt-CH₃), 2.27, 2.42 (s, 3H each,

NdC-CH₃), 3.90, 3.91 (s, 6H total, OCH₃), 6.82, 6.91, 6.97, 7.08

(broad s, 1H each, Ar-H), 7.27, 7.29 (broad s, 2H total, Ar-H). ¹³C

{¹H} NMR (75 MHz, TFE-d₃): δ ) -10.50 (Pt-CH₃), 20.68, 22.56

3

(NdC-CH₃), 56.92, 56.95 (OCH₃), 110.68 (q, J_C-F≈ 3 Hz, o-Ar-

[(^{Me BrAr}DAB^Me)PtMe(CO)]⁺[BF₄]^-(8c). 8c was synthesized simi-

2

3

C), 111.18 (o-Ar-C), 112.01 (q, J_C-F≈ 3 Hz, o-Ar-C), 112.68 (o-

3

larly from 5c (25.8 mg, 0.038 mmol) and an aqueous solution of HBF₄

(5 µL, 0.038 mmol) in 5 mL of TFE. 16 mg of yellow powder was

obtained (60%). The product contained a small amount of TFE, which

Ar-C), 113.15 (overlapping q, J_C-F≈ 3 Hz, p-Ar-C), 135.90 (q,

²J_C-F≈ 23 Hz, C-CF₃), 136.35 (q, J_C-F≈ 23 Hz, C-CF₃), 145.15,

2

149.641, 163.11, 163.30 (Ar-C), 179.03, 191.28. Unable to find CF₃

resonances, which are probably buried under CF₃CD₂OD resonances.

IR (CH₂Cl₂): ν(CO) ) 2110.1 cm^-1. The compound decomposes

quickly in methylene chloride and slowly in the solid state.

1

was hard to remove. H NMR (300 MHz, CD₂Cl₂): δ ) 0.61 (s, 3H,

²J_Pt-H) 69 Hz, Pt-CH₃), 2.20, 2.35 (s, 6H each, Ar-CH₃), 2.25, 2.45

(s, 3H each, NdC-CH₃), 7.43, 7.44 (s, 2H each, Ar-H). ¹³C {¹H}

NMR (75 MHz, CD₂Cl₂): δ ) -10.22 (Pt-CH₃), 18.03, 18.07 (Ar-

CH₃), 20.12, 21.73 (overlapping NdC-CH₃), 122.09, 122.66, 130.59,

132.08, 132.13, 132.19, 132.27, 132.41(Ar-C), 179.91, 191.04. IR

(CH₂Cl₂): ν(CO) ) 2111.6 cm^-1. The compound slowly decomposes

in methylene chloride. ESMS, Calcd for C₂₂H₂₅N₂Br₂OPt ([M]⁺):

684.9958 (⁷⁹Br⁷⁹Br¹⁹⁴Pt), 685.9984 (⁷⁹Br⁷⁹Br¹⁹⁵Pt), 686.9958, 687.9968,

688.9960, 689.9961 (⁸¹Br⁸¹Br¹⁹⁵Pt), 690.9966 (⁷⁹Br⁷⁹Br¹⁹⁵Pt). Found:

684.9980, 685.9995, 686.9962, 687.9972, 688.9959, 690.0007, 690.9974.

Anal. Calcd for C₂₂H₂₅Br₂N₂OPtBF₄: C, 34.09; H, 3.25; N, 3.61.

Found: C, 33.30/33.29; H, 3.42/3.32; N, 3.44/3.40.

[(^{CF Ar}DAB^Me)PtMe(CO)]⁺[BF₄]^-(7e). 7e was synthesized similarly

3

from 4e (11.2 mg, 0.015 mmol) and an aqueous solution of HBF₄(2

µL, 0.015 mmol) in 2 mL of TFE. The product contained a significant

amount of TFE, which was hard to remove. ¹H NMR (300 MHz, TFE-

d₃): δ ) 0.77 (s, 3H, ²J_Pt-H) 66 Hz, Pt-CH₃), 2.31, 2.45 (s, 3H each,

NdC-CH₃, the peaks were completely deuterated within an hour at

4

room temperature), 7.65 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.80 (d,

⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 8.05 (t, ⁴J_H-H) 1.5 Hz, 1H, p-Ar-H),

8.08 (t, ⁴J_H-H) 1.5 Hz, 1H, p-Ar-H). ¹³C {¹H} NMR (75 MHz, TFE-

d₃): δ ) -10.09 (Pt-CH₃), 123.01 (q, ³J_C-F≈ 3 Hz), 124.86 (q, ³J_C-F

≈ 3 Hz), 127.6 (q, ¹J_C-F≈ 273 Hz), 135.84 (q, ²J_C-F≈ 21 Hz, C-CF₃),

[(^{tBu ArH}DAB^H)PtMe(CO)]⁺[BF₄]^-(9a). 9a was synthesized simi-

2

136.30 (q, J_C-F≈ 3 Hz, C-CF₃), 144.67, 148.99 (Ar-C), 180.43,

larly from 6a (10 mg, 0.015 mmol) and an aqueous solution of HBF₄

(2 µL, 0.015 mmol) in 3 mL of TFE. 5.2 mg of orange powder was

obtained (47%). The product contained a small amount of TFE, which

192.41. Unable to find one set of CF₃resonances and several Ar-C

resonances, which are probably buried under CF₃CD₂OD resonances;

unable to find resonances corresponding to diimine methyl backbone,

which were completely deuterated within an hour at room temperature

in TFE-d₃. This significantly reduces the intensity of the ¹³C signals.

IR (CH₂Cl₂): ν(CO) ) 2113.5 cm^-1. The compound decomposes

quickly in methylene chloride and in the solid state.

1

was hard to remove. H NMR (500 MHz, CD₂Cl₂): δ ) 1.14 (t, 3H,

²J_Pt-H) 69 Hz, Pt-CH₃), 1.37, 1.39 (s, 18H each, C(CH₃)₃), 7.12 (d,

4

⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.44 (d, J_H-H) 1.5 Hz, 2H, o-Ar-

4

H), 7.59 (t, J_H-H) 1.5 Hz, 1H, p-Ar-H), 7.64 (t, J_H-H) 1.5 Hz,

1H, p-Ar-H), 9.04 (t, ³J_Pt-H) 74 Hz, NdC-H), 9.29 (t, ³J_Pt-H) 38

Hz, NdC-H). ¹³C {¹H} NMR (125 MHz, CD₂Cl₂): δ ) -8.81 (Pt-

CH₃), 31.10, 31.14 (C(CH₃)₃), 35.44, 35.45 (C(CH₃)₃), 117.20, 117.33,

[(^{35Me Ar}DAB^Me)PtMe(CO)]⁺[BF₄]^-(7f). 7f was synthesized simi-

larly from 4f (19.8 mg, 0.038 mmol) and an aqueous solution of HBF₄

2

9

1396 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

124.80, 126.91, 153.20, 154.04, 163.40, 176.49 (Ar-C). Could not find

CO resonance and two ArC resonances. IR (CH₂Cl₂): ν(CO) ) 2108.8

cm^-1. Anal. Calcd for C₃₂H₄₇N₂OPtBF₄: C, 50.73; H, 6.25; N, 3.70.

Found: C, 47.67/47.62; H, 5.94/5.91; N, 3.45/3.36.

to overlapping solvent peaks, OCH₃), 6.82, 6.95, 7.08, 7.28 (overlapping

with 10dii).

10dii: δ ) 0.860 (s, 3H, ²J_Pt-H) Hz, Pt-CH₃), 1.825, 2.05 (s, 3H

each, NdC-CH₃), 3.93, 3.94 (s, 3H each, OCH₃), 6.82, 6.93, 6.97,

7.08 (broad s, 1H, Ar-H), 7.26 (broad s, 2H total, Ar-H, split into

two peaks in the presence of benzene).

[(^{OMe ArH}DAB^H)PtMe(CO)]⁺[BF₄]^-(9b). 9b was synthesized simi-

3

larly from 6b (13.8 mg, 0.023 mmol) and an aqueous solution of HBF₄

(3 µL, 0.023 mmol) in 3 mL of TFE. 5.2 mg of a deep reddish purple

powder was obtained (33%). The product contained a small amount of

2

11ai: δ ) 0.66 (s, 3H, J_Pt-H) 72 Hz, Pt-CH₃), 1.74, 1.89 (s, 3H

each, NdC-CH₃), 2.15, 2.29 (s, 6H each, o-Ar-CH₃), 2.34 (overlap-

ping s, 6H total, p-Ar-CH₃), 7.06, 7.11 (s, 4H total, Ar-H). (In the

1

TFE, which was hard to remove. H NMR (500 MHz, CD₂Cl₂): δ )

2

1.24 (t, 3H, J_Pt-H) 66 Hz, Pt-CH₃), 3.84, 3.88 (s, 3H, p-OCH₃),

presence of 100 µL of C₆D₆): δ ) 0.74 (s, 3H, J_Pt-H) 72 Hz, Pt-

3.89, 3.94 (s, 6H each, OCH₃), 6.56, 6.90 (s, 2H each, Ar-H), 8.98,

9.21 (s, 1H each, NdC-H). IR (CH₂Cl₂): ν(CO) ) 2110.3 cm^-1. The

compound slowly decomposes in methylene chloride. ESMS, Calcd

for C₂₂H₂₇N₂OPt ([M]⁺): 625.1445 (¹⁹⁴Pt), 626.1469 (¹⁹⁵Pt), 627.1477

CH₃), 1.58, 1.73 (s, 3H each, NdC-CH₃), 2.16, 2.27 (s, 6H each,

o-Ar-CH₃), 2.36, 2.38 (s, 3H each, p-Ar-CH₃), 7.05, 7.10 (s, 4H total,

Ar-H).

11aii: δ ) 0.64 (s, 3H, ²J_Pt-H) 72 Hz, Pt-CH₃), 0.63(s, Pt-CH₂D),

1.63, 1.81 (s, 3H each, NdC-CH₃), 2.15, 2.26 (s, 6H each, o-Ar-

CH₃), 2.34 (overlapping s, 6H total, p-Ar-CH₃), 7.06, 7.10 (s, 4H total,

Ar-H). (In the presence of 100 µL of C₆D₆): δ ) 0.73 (s, 3H, ²J_Pt-H

) 72 Hz, Pt-CH₃), 0.72 (s, Pt-CH₂D), 1.48, 1.66 (s, 3H each, Nd

C-CH₃), 2.16, 2.27 (s, 6H each, o-Ar-CH₃), 2.36, 2.38 (s, 3H each,

p-Ar-CH₃), 7.05, 7.10 (s, 4H total, Ar-H).

(

¹⁹⁶Pt). Found: 625.1447, 626.1460, 627.1473.

[(^{OMeCF ArH}DAE)PtMe(CO)]⁺[BF₄]^-(9c). 9c was synthesized simi-

3

larly from 6c (14.5 mg, 0.023 mmol) and an aqueous solution of HBF₄

(3 µL, 0.023 mmol) in 3 mL of TFE. 6.0 mg of product was obtained

(37%). The product contained a small amount of TFE, which was

1

2

hard to remove. H NMR (CD₂Cl₂): δ ) 1.17 (t, 3H, J_Pt-H) 68 Hz,

Pt-CH₃), 3.94, 3.97 (s, 3H each, OCH₃), 7.12, 7.17, 7.32, 7.35,

11bi: δ ) 0.66 (s, 3H, ²J_Pt-H) 72 Hz, Pt-CH₃), 1.77, 1.92 (s, 3H

each, NdC-CH₃), 2.21, 2.35 (s, 6H each, o-Ar-CH₃), 7.24-7.28 (m,

6H total, Ar-H).

3

7.40, 7.43 (broad s, 1H each, Ar-H), 9.10 (t, 1H, J_Pt-H) 71 Hz,

3

NdC-H), 9.32 (t, 1H, J_Pt-H) 39 Hz, NdC-H). IR (CH₂Cl₂): ν-

(CO) ) 2116.0 cm^-1. The compound quickly decomposes in methylene

2

11bii: δ ) 0.64 (s, 3H, J_Pt-H) 72 Hz, Pt-CH₃), 0.62 (s, Pt-

chloride.

CH₂D), 1.65, 1.83 (s, 3H each, NdC-CH₃), 2.21, 2.32 (s, 6H each,

o-Ar-CH₃), 7.24-7.28 (m, 6H total, Ar-H).

Synthesis and Characterization of Methyl Aquo/Solvento Cations

(10-12). The aquo/solvento adducts of 10-12 were prepared similarly

to procedures described by Tilset and co-workers.¹¹The isolated orange/

brown solids inevitably contained 5-15% decomposition products

(mostly the µ-OH dimer). The amount of the impurities can be

minimized by strict control of the temperature at which TFE is removed.

These impurities are inert under conditions where benzene was activated

and are not expected to affect the outcome of intermolecular/

intramolecular competition reactions. For most kinetic studies, cations

10-12 are generated in situ (vide infra), and the chemical shifts reported

below are for solutions in TFE-d₃(300 and 500 MHz) in the absence

of substrates unless otherwise stated. The Pt-CH₂D resonances are

typically 0.01-0.02 ppm upfield of those of the corresponding Pt-

2

11ci: δ ) 0.70 (s, 3H, J_Pt-H) 72 Hz, Pt-CH₃), 1.78, 1.91 (s, 3H

each, NdC-CH₃), 2.18, 2.30 (s, 6H each, o-Ar-CH₃), 7.46 (s, 4H

total, Ar-H).

2

11cii: δ ) 0.69 (s, 3H, J_Pt-H) 72 Hz, Pt-CH₃), 0.67 (s, Pt-

CH₂D), 1.65, 1.84 (s, 3H each, NdC-CH₃), 2.18, 2.27 (s, 6H each,

o-Ar-CH₃), 7.43 (s, 4H total, Ar-H). (In the presence of 100 µL of

2

C₆D₆): δ ) 0.73 (s, 3H, J_Pt-H) 72 Hz, Pt-CH₃), 1.46, 1.66 (s, 3H

each, NdC-CH₃), 2.13, 2.24 (s, 6H each, o-Ar-CH₃), 7.42 (s, 4H

total, Ar-H).

12ai: δ ) 1.25 (br s, 3H, Pt-CH₃), 1.38, 1.40 (s, 18H each,

C(CH₃)₃), 7.13 (d, ⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.34 (d, ⁴J_H-H) 1.5

Hz, 2H, o-Ar-H), 7.66 (t, ⁴J_H-H) 1.5 Hz, 1H, p-Ar-H), 7.77 (t, ⁴J_H-H

) 1.5 Hz, 1H, p-Ar-H), 8.69 (s, 1H, NdC-H), 8.71 (s, 1H, NdC-

H). (In the presence of 30 µL of C₆D₆): δ ) 1.29 (br s, 3H, Pt-CH₃),

1.42, 1.44 (s, 18H each, C(CH₃)₃), 7.16 (d, ⁴J_H-H) 1.5 Hz, 2H, o-Ar-

2

CH₃peaks. J_Pt-Hcan only be observed on the 300 MHz NMR

instrument. Addition of benzene can significantly change the chemical

shifts for the backbone methyl or H resonances.

2

10ai: δ ) 0.77 (s, 3H, J_Pt-H) 66 Hz, Pt-CH₃), 1.465, 1.48 (s,

4

H), 7.53 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.70 (t, J_H-H) 1.5 Hz,

18H each, C(CH₃)₃), 1.88, 2.03 (s, 3H each, NdC-CH₃), 3.75, 3.76

(s, 3H each, OCH₃), 6.92, 7.10 (s, 2H each, Ar-H).

4

1H, p-Ar-H), 7.83 (t, J_H-H) 1.5 Hz, 1H, p-Ar-H), 8.39 (s, 1H,

NdC-H), 8.43 (s, 1H, NdC-H).

10aii: δ ) 0.765 (s, 3H, ²J_Pt-H) 66 Hz, Pt-CH₃), 1.465, 1.49 (s,

18H each, C(CH₃)₃), 1.79, 2.01 (s, 3H each, NdC-CH₃), 3.75, 3.78

(s, 3H each, OCH₃), 6.91, 7.15 (s, 2H each, Ar-H).

12aii: δ ) 1.24 (br s, 3H, Pt-CH₃), 1.38, 1.41 (s, 18H each,

C(CH₃)₃), 7.13 (d, ⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.51 (d, ⁴J_H-H) 1.5

Hz, 2H, o-Ar-H), 7.65 (t, ⁴J_H-H) 1.5 Hz, 1H, p-Ar-H), 7.77 (t, ⁴J_H-H

) 1.5 Hz, 1H, p-Ar-H), 8.80 (br s, 1H, NdC-H), 8.85 (s, 1H, Nd

C-H). (In the presence of 30 µL of C₆D₆): δ ) 1.27 (br s, 3H, Pt-

CH₃), 1.42, 1.44 (s, 18H each, C(CH₃)₃), 7.12 (d, ⁴J_H-H) 1.5 Hz, 2H,

o-Ar-H), 7.49 (d, ⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.68 (t, ⁴J_H-H) 1.5

2

10bi: δ ) 0.733 (s, 3H, J_Pt-H) 69 Hz, Pt-CH₃), 1.35, 1.36 (s,

18H each, C(CH₃)₃), 1.85, 2.01 (s, 3H each, NdC-CH₃), 6.84 (d, ⁴J_H-H

) 1.5 Hz, 2H, o-Ar-H), 7.03 (d, ⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.53,

7.61 (t, 1H each, p-Ar-H).

2

4

10bii: δ ) 0.723 (s, 3H, J_Pt-H) 69 Hz, Pt-CH₃), 1.35, 1.37 (s,

Hz, 1H, p-Ar-H), 7.81 (t, J_H-H) 1.5 Hz, 1H, p-Ar-H), 8.53 (br s,

18H each, C(CH₃)₃), 1.77, 1.99 (s, 3H each, NdC-CH₃), 6.83 (d, ⁴J_H-H

) 1.5 Hz, 2H, o-Ar-H), 7.07 (d, ⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.52,

7.59 (t, 1H each, p-Ar-H).

1H, NdC-H), 8.55 (s, 1H, NdC-H).

12bii: (In the presence of 30 µL of C₆D₆): δ ) 1.40 (br s, 3H,

Pt-CH₃), 1.38 (s, Pt-CH₂D), 3.88, 3.94 (s, 3H each, p-OCH₃), 3.90,

3.96 (s, 6H each, o-OCH₃), 6.59, 6.97 (s, 2H each, Ar-H), 8.71, 8.80

(s, 1H each, NdC-H).

12cii: δ ) 1.34(br s, 3H, Pt-CH₃), 1.33 (s, Pt-CH₂D), 3.91, 3.94

(s, 3H each, OCH₃), 7.03, 7.15, 7.31, 7.35, 7.38, 7.48 (broad s, 1H

each, Ar-H), 8.96, 9.07 (s, 1H each, NdC-H). (In the presence of

30 µL of C₆D₆): δ ) 1.37 (br s, 3H, Pt-CH₃), 1.36 (s, Pt-CH₂D),

3.94, 3.97 (s, 3H each, OCH₃), 7.03, 7.15, 7.31, 7.35, 7.38, 7.48 (broad

s, 1H each, Ar-H), 8.75, 8.84 (s, 1H each, NdC-H).

2

10ci: δ ) 0.93 (s, 3H, J_Pt-H) Hz, Pt-CH₃), 1.97, 2.075 (s, 3H

each, NdC-CH₃), 3.90-3.92 (OCH₃, cannot be identified with

certainty because of overlapping with solvent peaks), 6.37, 6.50 (s,

2H each, Ar-H).

2

10cii: δ ) 0.91 (s, 3H, J_Pt-H) Hz, Pt-CH₃), 1.86, 2.08 (s, 3H

each, NdC-CH₃), 3.90-3.92 (OCH₃, cannot be identified with

certainty due to overlapping solvent peaks), 6.36, 6.52 (s, 2H each,

Ar-H).

2

10di: δ ) 0.875 (s, 3H, J_Pt-H) Hz, Pt-CH₃), 1.95, 2.08 (s, 3H

Syntheses of Methyl-Acetonitrile Cations (13-15). Acetonitrile

adducts 13-15 were synthesized according to procedures reported in

each, NdC-CH₃), 3.89-3.94 (cannot be identified with certainty due

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1397

A R T I C L E S

Zhong et al.

ref 11. Without added acetonitrile, 13-15 are in equilibrium with

solvento adducts in CD₃OD, CD₃CD₂OD, and (CD₃)₂CDOD. ¹H NMR

data follows.

a small amount of benzene (e.g., 15 µL) to TFE-d₃shifts the resonances

for the diimine backbone methyls or protons by as much as 0.3 ppm

and can significantly affect shimming.

2

13b (TFE-d₃): δ ) 0.682 (s, 3H, J_Pt-H) 72 Hz, Pt-CH₃), 1.36,

17ai: δ ) 1.34, 1.51 (s, 18H each, C(CH₃)₃), 1.94, 2.12 (s, 3H each,

NdC-CH₃), 3.57, 3.77 (s, 3H each, OCH₃), 7.19, 7.75 (s, 2H each,

Ar-H), resonances for Ph-H’s cannot be identified with certainty.

17aii: δ ) 1.33, 1.51 (s, 18H each, C(CH₃)₃), 1.88, 2.14 (s, 3H

each, NdC-CH₃), 3.59, 3.80 (s, 3H each, OCH₃), 6.71, 7.25 (s, 2H

each, Ar-H), 6.73 (m, 1H, Ph-H_p), 6.79 (t, 7.4 Hz, 2H, Ph-H_o), 6.87

(m, 2H, Ph-H_m).

1.39 (s, 18H each, C(CH₃)₃), 1.92 (s, 3H, NC-CH₃), 1.99, 2.02 (s, 3H

4

each, NdC-CH₃), 6.84 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 6.97 (d,

⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.56, 7.60 (t, 1H each, p-Ar-H).

2

13b (CD₃OD): δ ) 0.55 (s, 3H, J_Pt-H) 73 Hz, Pt-CH₃), 1.37,

1.41 (s, 18H each, C(CH₃)₃), 2.05, 2.08 (s, 3H each, NdC-CH₃), 2.14

4

(s, 3H, NC-CH₃), 6.91 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.05 (d,

⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.47, 7.52 (t, 1H each, p-Ar-H).

13b (CD₃CD₂OD): δ ) 0.56 (s, 3H, ²J_Pt-H) 75 Hz, Pt-CH₃), 1.37,

1.41 (s, 18H each, C(CH₃)₃), 2.09, 2.12 (s, 3H each, NdC-CH₃), 2.17

17bi: δ ) 1.23, 1.40 (s, 18H each, C(CH₃)₃), 1.90, 2.10 (s, 3H each,

NdC-CH₃), 6.61 (d, ⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.14 (d, ⁴J_H-H

)

1.5 Hz, 2H, o-Ar-H), 7.23, 7.68 (t, 1H each, p-Ar-H), resonances

for Ph-H’s cannot be identified with certainty.

4

(s, 3H, NC-CH₃), 6.93 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.06 (d,

⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.44, 7.49 (t, 1H each, p-Ar-H).

17bii: δ ) 1.22, 1.41 (s, 18H each, C(CH₃)₃), 1.85, 2.14 (s, 3H

2

13b ((CD₃)₂CDOD): δ ) 0.59 (s, 3H, J_Pt-H) 75 Hz, Pt-CH₃),

4

each, NdC-CH₃), 6.61 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.18 (d,

1.37, 1.41 (s, 18H each, C(CH₃)₃), 2.12, 2.15 (s, 3H each, NdC-CH₃),

⁴J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.26, 7.65 (t, 1H each, p-Ar-H), 6.68

(m, 1H, Ph-H_p), 6.73 (m, 2H, Ph-H_o), 6.82 (m, 2H, Ph-H_m).

4

2.19 (s, 3H, NC-CH₃), 6.94 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.08

4

(d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.42, 7.46 (t, 1H each, p-Ar-H).

17ci: δ ) 2.04, 2.19 (s, 3H each, NdC-CH₃), 3.68, 3.89 (s, 3H

each, p-OCH₃), 3.72, 3.922 (s, 6H each, o-OCH₃), 6.09, 6.55 (s, 2H

each, Ar-H), resonances for Ph-H’s cannot be identified with

certainty.

2

13d (TFE-d₃): δ ) 0.76 (t, 3H, J_Pt-H) 67 Hz, Pt-CH₃), 2.01,

2.09 (s, 3H each, NdC-CH₃), 2.07 (s, 3H, NC-CH₃), 3.90, 3.92 (s,

3H each, OCH₃), 6.78, 6.88, 6.89, 7.05, 7.24, 7.26 (broad s, 1H each,

Ar-H).

17cii: δ )1.97, 2.19 (s, 3H each, NdC-CH₃), 3.71, 3.90 (s, 3H

each, p-OCH₃), 3.70, 3.915 (s, 6H each, o-OCH₃), 6.08, 6.67 (s, 2H

each, Ar-H), 6.80 (m, 1H, Ph-H_p), 6.86 (tt, 8 Hz, 1.8 Hz, 2H, Ph-

H_o), 6.94 (m, 2H, Ph-H_m).

17di: δ ) 2.00, 2.16 (s, 3H each, NdC-CH₃), 3.70, 3.92 (s, 3H

each, OCH₃), 6.48, 6.67, 6.90, 7.00, 7.30, 7.37 (broad s, 1H, Ar-H),

resonances for Ph-H’s cannot be identified with certainty.

17dii: δ ) 1.90, 2.15 (s, 3H each, NdC-CH₃), 3.67, 3.92 (s, 3H

each, OCH₃), 6.46, 6.70, 6.91, 7.02, 7.28, 7.37 (broad s, 1H, Ar- H),

6.74 (m, 1H, Ph-H_p), 6.79 (t, 9 Hz, 2H, Ph-H_o), 6.85 (m, 2H, Ph-

H_m).

2

13d (CD₃OD): δ ) 0.61 (t, 3H, J_Pt-H) 67 Hz, Pt-CH₃), 2.09,

2.15 (s, 3H each, NdC-CH₃), 2.27 (s, 3H, NC-CH₃), 3.94, 3.97 (s,

3H each, OCH₃), 6.95, 7.00, 7.07, 7.15, 7.27, 7.31 (broad s, 1H each,

Ar-H).

2

14b (CD₃OD): δ ) 0.44 (s, 3H, J_Pt-H) 75 Hz, Pt-CH₃), 2.01,

2.05 (s, 3H each, NdC-CH₃), 2.06 (s, 3H, NC-CH₃), 2.23, 2.34 (s,

6H each, Ar-CH₃), 7.25-7.34 (m, 6H total, Ar-H).

2

14c (TFE-d₃): δ ) 0.62 (s, 3H, J_Pt-H) 75 Hz, Pt-CH₃), 1.88,

1.94 (s, 3H each, NdC-CH₃), 1.97 (s, 3H, NC-CH₃), 2.15, 2.26 (s,

6H each, Ar-CH₃), 7.42, 7.47 (s, 2H each, Ar-H).

2

14c (CD₃OD): δ ) 0.48 (s, 3H, J_Pt-H) 75 Hz, Pt-CH₃), 2.03,

18ai: δ ) 1.69, 1.87 (s, 3H each, NdC-CH₃), 2.13, 2.35 (s, 6H

each, o-Ar-CH₃), 2.15, 2.37 (s, 3H each, p-Ar-CH₃), 6.69, 7.14 (s,

4H total, Ar-H), 6.69-6.84 (m, 5H, Ph-H’s).

2.06 (s, 3H each, NdC-CH₃), 2.19 (s, 3H, NC-CH₃), 2.21, 2.33 (s,

6H each, Ar-CH₃), 7.48, 7.53 (s, 2H each, Ar-H).

NMR Data for (µ-OH)₂Dimer 16b. H NMR (TFE-d₃): 1.24 (s,

1

36H, C(CH₃)₃), 1.91 (s, 6H each, NdC-CH₃), 5.30 (br s, O-H), 7.03

(d, ⁴J_H-H) 1.5 Hz, 4H, o-Ar-H), 7.60 (t, ⁴J_H-H) 1.5 Hz, 4H, p-Ar-

H). ¹⁹F NMR (TFE-d₃): -152.0 (BF₄^-).

18aii: δ ) 1.61, 1.81 (s, 3H each, NdC-CH₃), 2.11, 2.35 (s, 6H

each, o-Ar-CH₃), 2.17, 2.39 (s, 3H each, p-Ar-CH₃), 6.72, 7.13 (s,

4H total, Ar-H), 6.69-6.84 (m, 5H, Ph-H’s).

Measurement of Kinetics for C-H Bond Activation of Aromatic

Substrates. Dry TFE-d₃was vacuum transferred into an oven-dried 5

mm thin-walled NMR tube with J-Young valve. Approximately 0.0076

mmol of (N-N)PtMe₂(4-6), 1 µL of aqueous HBF₄(48 wt %, 0.00765

mmol), and a predetermined amount of D₂O were then added to the

tube. The mixture was shaken to form a clear solution. ¹H NMR spectra

were then taken of the mixture to ensure clean conversion to aquo/

solvento adducts 10-12. A predetermined amount of substrate was

then added to the NMR tube, and after allowing the mixture to

equilibrate to the preset temperature in the probe, disappearance of the

starting material (and appearance of the products 17-19) was moni-

tored. Probe temperatures were calibrated with a methanol thermometer

and were maintained at (0.2 °C throughout data acquisition. The

observed rate constants are calculated by curve fitting to the expression

A_t) A_f+ (A₀- A_f) × exp(-k_obst), where A_tis the area under the peak

(or the peak height). The area under the peak is found by multiplying

the peak height by the full width at half maximum. The volume of the

reaction mixture is determined as V (mL) ) 0.01384H - 0.006754,

where H is the solvent height in millimeters. The water concentration

is calculated as follows: [H₂O] ) [(1 µL × 1.4 g mL^-1× 52% + y µL

× 1 g mL^-1)/18 g mol^-1- 0.00765 × n/(n + 1)]/V (mL), where 1.4

g mL^-1is the density of the aqueous HBF₄solution, 52% is the wt %

of water in this aqueous solution, y is the amount of extra water added,

1 g mL^-1is the density of water, and n is the ratio of aquo:solvento

adducts. The chemical shifts for the phenyl complexes 17-19 reported

below were measured in TFE in the presence of benzene. Addition of

18bi: δ ) 1.75, 1.96 (s, 3H each, NdC-CH₃), 2.20, 2.42 (s, 6H

each, o-Ar-CH₃), aryl peaks cannot be identified with certainty (many

overlapping peaks).

18bii: δ ) 1.67, 1.89 (s, 3H each, NdC-CH₃), 2.18, 2.40 (s, 6H

each, o-Ar-CH₃), 6.91, 7.30 (s, 2H each, o-Ar-H), 6.93, 7.29 (s, 1H

each, p-Ar-H), 6.72-7.00 (m, 5H, Ph-H’s).

18ci: δ ) 1.77, 1.95 (s, 3H each, NdC-CH₃), 2.14, 2.36 (s, 6H

each, o-Ar-CH₃), 7.03, 7.48 (s, 4H total, Ar-H), 6.69-6.84 (m, 5H,

Ph-H’s).

18cii: δ ) 1.67, 1.89 (s, 3H each, NdC-CH₃), 2.12, 2.34 (s, 6H

each, o-Ar-CH₃), 7.06, 7.46 (s, 2H each, Ar-H), 6.69-6.90 (m, 5H,

Ph-H’s).

19aii: δ ) 1.23, 1.46 (s, 18H each, C(CH₃)₃), 6.810 (d, ⁴J_H-H) 1.5

4

Hz, 2H, o-Ar-H), 7.52 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H), 7.45 (t,

⁴J_H-H) 1.5 Hz, 1H, p-Ar-H), 7.84 (t, ⁴J_H-H) 1.5 Hz, 1H, p-Ar-H),

8.46 (br s, 1H, NdC-H), 8.53 (s, 1H, NdC-H), 6.84 (m, 2H, Ph-

H_o). The other three Ph-H peaks are probably hidden by free benzene

peaks.

19bii: δ ) 3.78, 3.95 (s, 3H each, p-OCH₃), 3.65, 3.97 (s, 6H each,

o-OCH₃), 6.29, 6.99 (s, 2H each, Ar-H), 8.66, 8.79 (s, 1H each, Nd

C-H), 7.06-7.14 (m, Ph-H’s).

19cii: (In the presence of 30 µL of C₆H₆), δ ) 3.61, 3.94 (s, 3H

each, OCH₃), 6.62, 6.87, 7.08, 7.35, 7.44, 7.48 (broad s, 1H each, Ar-

H), 8.68, 6.82 (s, 1H each, NdC-H), cannot be identified with

certainty. (In the presence of 60 µL of C₆D₆): δ ) 3.60, 3.93 (s, 3H

9

1398 J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002

Bond Activation by Cationic Platinum(II) Complexes

A R T I C L E S

each, OCH₃), 6.62, 6.87, 7.08, 7.35, 7.44, 7.48 (broad s, 1H each, Ar-

H), 8.53, 8.66 (s, 1H each, NdC-H), cannot be identified with

certainty.

Acetonitrile Exchange Reactions. To an oven-dried NMR tube was

added 3.5 mg of the appropriate acetonitrile adduct 13-15 (2.5 mg in

(CD₃) ₂CDOD because of limited solubility). About 0.7 mL of the

deuterated solvent was then added, and a ¹H NMR spectrum was

recorded before addition of a predetermined amount of CD₃CN. Under

these conditions, the acetonitrile adducts were the only observable

species in solution. After allowing the mixture to equilibrate to a preset

temperature (∼5 min), the disappearance of coordinated CH₃CN and

the appearance of free CH₃CN were monitored, and the observed rate

of exchange was determined from the following expression: A_t) A_f

+ (A₀- A_f) × exp(-k_ext), where A_tis the area under the peak (or the

peak height). The volume of the solution is determined as described

above.

20bii: δ ) 1.16, 1.24 (s, 9H each, C(CH₃)₃), 1.39 (s, 18H, C(CH₃)₃),

1.74, 2.08 (s, 3H each, NdC-CH₃), 1.99, 2.51 (s, 3H each, p-xylene-

4

Me), 6.55 (t, J_H-H) 1.8 Hz, 1H, o-Ar-H), 6.64 (t, J_H-H) 1.8 Hz,

4

1H, o-Ar-H), 7.19 (d, 2H, o-Ar-H), 7.23, 7.64 (t, J_H-H) 1.8 Hz,

1H each, p-Ar-H), 6.43 (m, 1H, p-xylene-H), 6.55 (m, 2H, p-xylene-

H).

20bi: δ ) 1.20, 1.21 (s, 9H each, C(CH₃)₃), 1.39 (s, 18H, C(CH₃)₃),

1.86, 1.98 (s, 3H each, NdC-CH₃), 2.13, 2.38 (s, 3H each, p-xylene-

4

Me), 6.67 (m, 2H, o-Ar-H), 7.05 (d, J_H-H) 1.5 Hz, 2H, o-Ar-H),

4

7.23, 7.56 (t, J_H-H) 1.8 Hz, 1H each, p-Ar-H), 6.25 (m, 1H,

p-xylene-H), 6.60 (m, 2H, p-xylene-H).

Measurement of KIE via Inter- or Intramolecular Competition

Reactions. Dry TFE-d₃was added to an oven-dried NMR tube

containing preformed aquo/solvento adducts 10. A mixture of 1:1 C₆H₆:

C₆D₆or C₆D₃H₃was then added to the tube. After the reaction was

complete, the integration ratio of CH_nD_4-n:CH₄was then taken to give

the KIE.

Measurement of the Ratio of CH₃D to CH₄as a Function of

Added Acetonitrile in the Protonolysis of 4b. To a suspension of

0.0076 mol of 4b over dry TFE-d₃in oven-dried NMR tube was added

a predetermined amount of CD₃CN. To this mixture, 5 µL of a stock

solution of 1 µL of aqueous HBF₄(48 wt %) in 24 µL of TFE-d₃was

added. The tube was shaken, and after a clear orange solution was

formed, the ratio of CH₃D:CH₄was measured by integration. To account

for the percentage of protonolysis by H⁺(rather than D⁺), the H⁺

concentration was measured by integrating the [OH] resonance against

the aromatic protons of 4b. H⁺typically accounts for 1-3% of the

total H⁺/D⁺source. This percentage was subtracted from the percentage

of liberated CH₄. The adjusted ratio of CH₃D:CH₄was then plotted

against the acetonitrile concentrations to yield Figure 1.

20b: (MeCN adduct in CD₃NO₂), δ ) 1.16, 1.28 (s, 9H each,

C(CH₃)₃), 1.43 (s, 18H, C(CH₃)₃), 1.95 (s, 3H, NdC-CH₃), 2.02 (s,

3H, CH₃CN), 2.17, 2.27, 2.40 (s, 3H each, NdC-CH₃or p-xylene-

Me), 6.36 (m, 1H, p-xylene-H), 6.50 (m, 2H, p-xylene-H), 6.36 (t,

⁴J_H-H) 1.8 Hz, 1H, o-Ar-H), 6.77 (t, ⁴J_H-H) 1.5 Hz, 1H, o-Ar-H),

7.16 (d, 2H, o-Ar-H), 7.24 (t, ⁴J_H-H) 1.5 Hz, 1H, p-Ar-H), 7.63 (t,

⁴J_H-H) 1.8 Hz, 1H, p-Ar-H).

21b: (MeCN adduct in CD₃NO₂), δ ) 1.34, 1.38 (s, 18 H each,

3

C(CH₃)₃), 1.75 (t, J_H-H) 27 Hz, 3H, CH₃CN), 2.081, 2.135 (s, 3H,

2

NdC-CH₃), 2.15 (s, 6H, mesitylene-Me), 2.80 (t, J_H-H) 103 Hz,

2H, Pt-CH₂Ar), 6.51 (m, 2H, mesitylene- H), 6.59 (m, 1H, mesity-

lene-H), 6.93, 6.97 (d, ⁴J_H-H) 1.5 Hz, 2H each, o-Ar-H), 7.40, 7.48

4

(t, J_H-H) 1.5 Hz, 1H each, p-Ar-H).

21bii: (Aquo adduct in TFE-d₃), δ ) 1.38, 1.42 (s, 18H each,

C(CH₃)₃), 1.78, 1.92 (s, 3H, NdC-CH₃), 2.53 (s, 6H, mesitylene-

Me), 5.93 (br s, 2H, o-mesitylene-H, substantial H incorporation into

4

this position when mesitylene-d₁₂is used), 6.47, 7.06 (d, J_H-H) 1.5

Hz, 2H each, o-Ar-H), 7.56, 7.60 (t, ⁴J_H-H) 1.5 Hz, 1H each, p-Ar-

H), several peaks cannot be identified with certainty; they are probably

buried under free mesitylene peaks.

Acknowledgment. Support by the National Science Founda-

tion (Grant No. CHE-9807496), Akzo-Nobel, and BP are

gratefully acknowledged. We thank Dr. Joseph Sadighi for

preparing the diimine ligand 1a and for devising the synthetic

route for the preparation of diimine ligands 3.

22bii: Most peaks cannot be identified with certainty, except 1.23

(s, C(CH₃)₃) and 1.70, 2.07 (s, NdC-CH₃).

Measurement of Equilibrium Constants. Dry TFE-d₃was vacuum

transferred into an oven-dried J-Young tube. Approximately 0.0076

mmol of (N-N)PtMe₂(4-6) and 1 µL of aqueous HBF₄(48 wt %,

0.00765 mmol) were then added to the tube. After allowing the mixture

to equilibrate to 20 °C, the ratio of the aquo to solvento adducts, n,

was then determined by integration. The water concentration was

Supporting Information Available: Summary of arene C-H

bond activation kinetic data and derivation of the rate law for

Scheme 11 (PDF). This material is available free of charge via

the Internet at http://pubs.acs.org.

determined as follows: [(1 µL × 1.4 g mL^-1× (1-0.48)/18 g mol^-1

)

- 0.0076 mmol × n/(1 + n)]/V (mL). Volume was determined as

described above. The TFE-d₃concentration was taken as 14.07 M.

JA011189X

9

J. AM. CHEM. SOC. VOL. 124, NO. 7, 2002 1399

Article Doi

C-H bond activation by cationic platinum(II) complexes: Ligand electronic and steric effects

DOI: 10.1021/ja011189x

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja011189x

Full text of DOI:10.1021/ja011189x

Products guided by the article

R&D Labs maybe for 404373-68-8

Relevant to this article

Hot Product