BF3-activated oxidation of alkanes by MnO4 -

Article abstract of DOI:10.1021/ja0552951

The oxidation of alkanes and arylalkanes by KMnO₄ in CH ₃CN is greatly accelerated by the presence of just a few equivalents of BF₃, the reaction occurring readily at room temperature. Carbonyl compounds are the predominant products in the oxidation of secondary C-H bonds. Spectrophotometric and kinetics studies show that BF₃ forms an adduct with KMnO₄ in CH₃CN, [BF₃·MnO ₄]^-, which is the active species responsible for the oxidation of C-H bonds. The rate constant for the oxidation of toluene by [BF₃·MnO₄]^- is over 7 orders of magnitude faster than by MnO₄^- alone. The kinetic isotope effects for the oxidation of cyclohexane, toluene, and ethylbenzene at 25.0 °C are as follows: k_C6H12/k_C6D12 = 5.3 ± 0.6, k_C7H8/k_C7D8 = 6.8 ± 0.5, k_C8H10/k _C8D10 = 7.1 ± 0.5. The rate-limiting step for all of these reactions is most likely hydrogen-atom transfer from the substrate to an oxo group of the adduct. A good linear correlation between log(rate constant) and C-H bond energies of the hydrocarbons is found. The accelerating effect of BF₃ on the oxidation of methane by MnO₄ has been studied computationally by the Density Functional Theory (DFT) method. A significant decrease in the reaction barrier results from BF₃ coordination to MnO₄^-. The BF₃ coordination increases the ability of the Mn metal center to achieve a d¹ Mn(VI) electron configuration in the transition state. Calculations also indicate that the species [2BF₃·MnO₄]^- is more reactive than [BF₃·MnO₄]^-.

Full text of DOI:10.1021/ja0552951

Published on Web 02/11/2006
-
BF3-Activated Oxidation of Alkanes by MnO4
William W. Y. Lam,† Shek-Man Yiu,† Joyce M. N. Lee,† Sammi K. Y. Yau,†
Hoi-Ki Kwong,† Tai-Chu Lau,*,† Dan Liu,‡ and Zhenyang Lin*,‡
Contribution from the Department of Biology and Chemistry, City UniVersity of Hong Kong,
Tat Chee AVenue, Kowloon Tong, Hong Kong, China, and Department of Chemistry, The Hong
Kong UniVersity of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, China
Received August 13, 2005; E-mail: bhtclau@cityu.edu.hk; chzlin@ust.hk
Abstract: The oxidation of alkanes and arylalkanes by KMnO4 in CH3CN is greatly accelerated by the
presence of just a few equivalents of BF3, the reaction occurring readily at room temperature. Carbonyl
compounds are the predominant products in the oxidation of secondary C-H bonds. Spectrophotometric
and kinetics studies show that BF3 forms an adduct with KMnO4 in CH3CN, [BF3‚MnO4]-, which is the
active species responsible for the oxidation of C-H bonds. The rate constant for the oxidation of toluene
by [BF3‚MnO4]- is over 7 orders of magnitude faster than by MnO4 alone. The kinetic isotope effects for
-
the oxidation of cyclohexane, toluene, and ethylbenzene at 25.0 °C are as follows: kC H /kC D ) 5.3 (
6
12
6 12
0.6, kC H /kC D ) 6.8 ( 0.5, kC H /kC D10 ) 7.1 ( 0.5. The rate-limiting step for all of these reactions is most
7
8
7
8
8
10
8
likely hydrogen-atom transfer from the substrate to an oxo group of the adduct. A good linear correlation
between log(rate constant) and C-H bond energies of the hydrocarbons is found. The accelerating effect
of BF3 on the oxidation of methane by MnO4- has been studied computationally by the Density Functional
-
Theory (DFT) method. A significant decrease in the reaction barrier results from BF3 coordination to MnO4
.
The BF3 coordination increases the ability of the Mn metal center to achieve a d1 Mn(VI) electron
configuration in the transition state. Calculations also indicate that the species [2BF3‚MnO4]- is more reactive
than [BF3‚MnO4]-.
Introduction
acids on metal-oxo species are well-known; however the use
of Lewis acids to enhance the reactivity of these species is much
Enzymes such as cytochrome P-450 and methane monooxy-
genases are able to oxidize unactivated C-H bonds under mild
conditions.1-3 The search for highly reactive metal-oxo species
in chemical systems that can mimic the reactivity of the iron-
oxo intermediates in these enzymes remains a challenge to
chemists.4-8 Metal-oxo species such as chromate and perman-
ganate have long been used as oxidants for a variety of organic
functional groups.9 However, their use in the oxidation of
alkanes is limited; refluxing conditions are often required to
oxidize unactivated C-H bonds. Our approach to generate
highly oxidizing reagents is to make use of Lewis acids to
activate metal-oxo species. The activating effects of Brønsted
less studied. We have recently reported that the oxidation of
alkanes by anionic oxo species of ruthenium,10 iron,11 manga-
nese,12 chromium,12 and osmium13 is greatly enhanced by Lewis
acids. The Lewis acid assisted oxidation of various other organic
substrates by permanganate has also been subsequently reported
by Lee and co-workers.14,15 Mayer and co-workers have recently
studied the kinetics and mechanisms of the oxidation of various
arylalkanes by nBu4NMnO4 in organic solvents.16-18 We report
here a kinetics and mechanistic study of the BF3-activated
oxidation of alkanes and arylalkanes by KMnO4. Our results
-
show that BF3 forms an adduct with MnO4 in CH3CN that
oxidizes hydrocarbons at rates over 7 orders of magnitude faster
than that by MnO4- alone. Initial results of this work have been
communicated.12 The accelerating effect of BF3 on the oxidation
of methane by MnO4- has been studied computationally by the
Density Functional Theory (DFT) method.
† City University of Hong Kong.
‡ The Hong Kong University of Science and Technology.
(1) Ortiz de Montellano, P. R., Ed. Cytochrome P450. Structure, Mechanism
and Biochemistry; Plenum Press: New York, 1995.
(2) McLain, J. L.; Lee, J.; Groves, J. T. In Biomimetic Oxidations Catalyzed
by Transition Metal Complexes; Meunier, B., Ed.; Imperial College Press:
London, 2000; pp 91-170.
(3) Baik, M.-H.; Newcomb, M.; Friesner, R. A.; Lippard, S. J. Chem. ReV.
2003, 103, 2385-2420.
(10) Lau, T. C.; Mak, C. K. J. Chem. Soc., Chem. Commun. 1993, 766-767.
(11) Ho, C. M.; Lau, T. C. New J. Chem. 2000, 24, 587-590.
(12) Lau, T. C.; Wu, Z. B.; Bai, Z. L.; Mak, C. K. J. Chem. Soc., Dalton Trans.
1995, 695-696.
(13) Yiu, S. M.; Wu, Z. B.; Mak, C. K.; Lau, T. C. J. Am. Chem. Soc. 2004,
126, 14921-14929.
(4) Hu, Z.; Gorun, S. M. In Biomimetic Oxidations Catalyzed by Transition
Metal Complexes; Meunier, B., Ed.; Imperial College Press: London, 2000;
pp 269-308.
(5) Robert, A.; Meunier, B. In Biomimetic Oxidations Catalyzed by Transition
Metal Complexes; Meunier, B., Ed.; Imperial College Press: London, 2000;
pp 543-562.
(14) Xie, N.; Binstead, R. A.; Block, E.; Chandler, W. D.; Lee, D. G.; Meyer,
T. J.; Thiruvazhi, M. J. Org. Chem. 2000, 65, 1008-1015.
(15) Lai, S.; Lee, D. G. Tetrahedron 2002, 58, 9879-9887.
(16) Gardner, K. A.; Mayer, J. M. Science 1995, 269, 1849-1851.
(17) Gardner, K. A.; Kuehnert, L. L.; Mayer, J. M. Inorg. Chem. 1997, 36,
2069-2078.
(6) Shilov, A. E.; Shul’pin, G. B. Chem. ReV. 1997, 97, 2879-2932.
(7) Crabtree, R. H. J. Chem. Soc., Dalton Trans. 2001, 2437-2450.
(8) Fokin, A. A.; Schreiner, P. R. Chem. ReV. 2002, 102, 1551-1594.
(9) Hudlicky, M. Oxidations in Organic Chemistry; American Chemical
Society: Washington, DC, 1990.
(18) Mayer, J. M. Acc. Chem. Res. 1998, 31, 441-450.
9
10.1021/ja0552951 CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 2851-2858
2851

A R T I C L E S
Lam et al.
Experimental Section
Materials. CAUTION: Care should be taken in handling BF3/
MnO4- in organic solVents, since the system is Very reactiVe. Although
we haVe not encountered any problems so far, the amount of KMnO4/
nBu4NMnO4 used each time should be less than 50 mg. Potassium
permanganate (Ajax Chemicals, AR) was used as received. nBu4NMnO4
was prepared according to a literature method.14 A Boron trifluoride-
acetonitrile complex solution (BF3‚CH3CN) (Fluka, 15-18%) was
stored at -20 °C and was used without further purification. The
concentration was determined to be 15.2% by hydrolysis to H3BO3
and HF followed by titration with NaOH. p-Xylene (Aldrich, 99+%
anhydrous), 4-methylanisole (Aldrich, 99%), 4-chlorotoluene (Aldrich,
98% GC), d8-toluene (Aldrich, 99.5 at. %D), d10-ethylbenzene (Acroˆs,
98+ at. %D), and d12-cyclohexane (Aldrich, 99.5 at. %D) were used
as received. Cyclohexane (Lab-scan, AR), cycloheptane (Aldrich, 98%),
cyclooctane (Aldrich, 99%), cyclopentane (Lab-scan, AR), toluene
(Riedel-deHae¨n, 99.7%), ethylbenzene (Riedel-deHae¨n, 99%), and
cumene (Merck) were washed with cold concentrated H2SO4 followed
by distilled water and then 5% aqueous NaHCO3. They were then dried
with MgSO4 and distilled from CaH2 under argon. They were passed
through a column of alumina before use. Triphenylmethane (Aldrich,
99%) was recrystallized from ethanol. Diphenylmethane (Aldrich, 99%)
was sublimed before use. Acetonitrile (Lab-scan, AR) was stirred
overnight with KMnO4 and then distilled; it was distilled again over
CaH2 under argon.19
Oxidation of Hydrocarbons. A solution of BF3‚CH3CN in CH3-
CN was added with vigorous stirring to a solution of KMnO4 (0.02
mmol) in CH3CN containing cyclohexane. After 5 min 100 µL of H2O
were added, and the resulting mixture was analyzed by GC and GC-
MS using chlorobenzene as the internal standard. A Hewlett-Packard
6890 gas chromatograph with an FFAP capillary column was used.
GC-MS measurements were carried out on an HP 6890 gas chro-
matograph interfaced to an HP 5973 mass selective detector.
Manganese Oxidation State Determination. The oxidation state
of the manganese product was determined by an iodometric method.20
The manganese oxidation state was found to be 3.94 ( 0.16 in the
oxidation of cyclohexane.
Figure 1. Plot of % yield vs [BF3]/[MnO4-] in the oxidation of cyclohexane
by BF3/MnO4-. (O) Reaction carried out in air. (b) Reaction carried out
under argon. [MnO4-] ) 4.3 × 10-3 M, [cyclohexane] ) 1.0 M. The yields
are calculated by assuming MnO4- acts as a three-electron oxidant: Yield
) 100% × (2/3[cyclohexanol]f + 4/3[cyclohexanone]f)/[MnO4-]i (f ) final,
i ) initial).
CN was added with vigorous stirring to a solution of KMnO4
(4.3 × 10-3 M) containing cyclohexane (1.0 M) in CH3CN at
23 °C. The purple color of solution was discharged within
seconds. The oxidation state of the manganese after the reaction
was determined to be 3.94 ( 0.16; this together with the
appearance of a characteristic broad absorbance in the visible
spectrum (vide infra) is consistent with the formation of colloidal
MnO2. Analysis of the organic products by GC and GC-MS
after ca. 5 min indicated the presence of 8% cyclohexanol and
73% cyclohexanone.21 The yields were calculated based on the
system acting as a three-electron oxidant. No increase in yields
were found by reducing the MnO2 to Mn2+ using acidic iodide;
apparently strong adsorption of products onto MnO2 did not
occur. In the absence of BF3, KMnO4 is stable in cyclohexane/
CH3CN for at least several hours at room temperature. The
yields increased with the [BF3]/[MnO4-] ratio, the maximum
yield occurring when 20 equiv of BF3 were used (Figure 1, Table
S1). The yields also increased significantly when the reaction
was carried out in air instead of under argon (Figure 1, Table
S2), indicating the intermediacy of cyclohexyl radicals. A few
experiments were done using nBu4NMnO4, and the yields were
found to be similar to those using KMnO4.
Kinetics. The kinetics of the reaction were studied by using a
Hewlett-Packard 8452A UV-vis spectrophotometer or an Applied
Photophysics DX-17MV stopped-flow spectrophotometer. The con-
centrations of the hydrocarbons were at least in 10-fold excess of that
of MnO4-. Reactions were initiated by mixing BF3‚CH3CN in CH3CN
with a freshly prepared solution of KMnO4 and hydrocarbon in CH3-
CN under argon. The reaction progress was monitored by observing
absorbance changes at 526 nm (λmax of MnO4-). Pseudo-first-order rate
constants, kobs, were obtained by nonlinear least-squares fits of At vs
time t according to the equation At ) Af + (A0 - Af) exp(-kobst), where
A0 and Af are the initial and final absorbances, respectively.
Kinetic Isotope Effects (KIEs). The KIEs for cyclohexane oxidation
were studied by using an equimolar mixture of cyclohexane and d12-
cyclohexane as substrate. The products were analyzed by GC and GC-
MS. The peaks due to the possible products cyclohexanol, d12-
cyclohexanol, cyclohexanone, and d10-cyclohexanone are well resolved
in the GC spectrum under our experimental conditions. The KIE was
calculated by the following equation: kC H /kC D ) (moles of c-C6H11-
OH + moles of c-C6H10O)/(moles of c-C6D11OD + moles of c-C6D10O).
KIEs for cyclohexane, toluene, and ethylbenzene were also deter-
mined by comparing the rate constants for the oxidation of the protio
vs deutero substrates using UV/vis spectrophotometric methods.
The kinetic isotope effects (KIE), determined from the
competitive oxidation of cyclohexane and d12-cyclohexane, were
found to be 4.7 ( 0.5.
Spectrophotometric Changes. When BF3‚CH3CN in CH3-
CN (8.0 × 10-4 M) was mixed with KMnO4 (1.0 × 10-4 M)
in CH3CN at 25.0 °C (in a two-compartment cuvette), there
was an initial rapid decrease in absorbance in the region 500-
580 nm, followed by a much slower and larger decay (Figure
S1). The initial decay, which was too fast to be followed by
stopped-flow spectrophotometry, corresponded to an apparent
decomposition of ca. 15% MnO4-; however quenching of the
reaction after about 5 s with excess acidic iodide followed by
6
12
6 12
Results and Discussion
Oxidation of Cyclohexane by BF3/MnO4-. In a typical
experiment, a solution of BF3‚CH3CN (8.6 × 10-2 M) in CH3-
(19) Armarego, W. L. F.; Perrin, D. D. Purification of Laboratory Chemicals,
4th ed.; Reed Educational and Professional Publishing Ltd.: Oxford, 1996.
(20) Lee, D. G.; Perez-Benito, J. F. J. Org. Chem. 1988, 53, 5725-5728.
(21) In the original communication BF3‚MeCO2H was used.12 A lower yield of
42% cyclohexanone was obtained under different reaction conditions. We
have also vigorously purified the solvent and substrates in the present study.
BF3‚CH3CN was used in the present study (instead of BF3‚MeCO2H) to
avoid complications from acetic acid.
9
2852 J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006

-
BF3-Activated Oxidation of Alkanes by MnO4
A R T I C L E S
Figure 3. Spectrophotometric changes at 20-s intervals during the BF3 (8
-
Figure 2. Plot of ꢀapp vs (ꢀapp - ꢀMnO )/[BF3] (slope ) -(1.42 ( 0.06) ×
4
-
× 10-4 M) activated oxidation of cyclohexane (0.1 M) by MnO4 (1 ×
10-1, y-intercept ) (1.43 ( 0.03), r ) -0.995).
10-4 M) in CH3CN at 298.0 K.
measuring the concentration of I3- by UV-vis spectrophotom-
etry indicated that over 95% of MnVII was still present. This
decay also increased with the amount of BF3. These results,
together with other evidence presented below, suggest that the
first rapid decay is due to the formation of a Lewis acid-base
adduct as shown in eq 1, while the second step corresponds to
decomposition of the adduct.
[BF3‚MnO4]-. The second step, however, is much faster than
that in the absence of cyclohexane, and the rate also increases
with cyclohexane concentration. The second step is assigned
to the oxidation of cyclohexane by [BF3‚MnO4]-. The optical
spectra in the range 400-640 nm show the disappearance of
-
MnO4 and the appearance of a broad absorbance due to
colloidal MnO2.14,17
K
BF3‚CH3CN + MnO4- y
\
z [BF3‚MnO4]- + CH3CN (1)
Kinetics. The kinetics of the second step in the BF3-activated
-
oxidation of cyclohexane by MnO4 were monitored by
observing the absorbance changes at 526 nm (λmax of MnO4-).
In the presence of at least 10-fold excess of BF3 and cyclohex-
ane, clean pseudo-first-order kinetics were observed for over
three half-lives (Figure 4), despite the formation of MnO2
colloids. First-order plots are usually slightly curved in oxida-
tions by permanganate.14,17 The pseudo-first-order rate constant,
The equilibrium constant, K, for the adduct formation between
MnO4- and BF3 was determined by using a spectrophotometric
method. A series of solutions were prepared by mixing 4 ×
10-4 M KMnO4 in CH3CN with various amounts of BF3‚CH3-
CN (4 × 10-4-3 × 10-3 M), and the absorbance at 526 nm
(A526) of each solution in a 1-cm cell was taken immediately
after mixing (within 5 s). The data were treated using eqs 2
and 3.22
k
obs, is independent of the concentration of MnO4- (5 × 10-5
-
2 × 10-4 M) but depends linearly on the concentration of the
cyclohexane (1 × 10-3-2 × 10-1 M) (Figure 4). The relatively
ꢀapp - ꢀ
-
small y-intercept of the second-order plot (1.02 × 10-3 s-1
)
MnO4
ꢀapp ) ꢀadduct
-
(2)
(3)
indicates that decomposition of [BF3‚MnO4]- is insignificant
compared with the oxidation of cyclohexane. Attempts were
also made to independently determine the rate constant for the
decomposition of the adduct in the absence of cyclohexane. In
this case the decay of the adduct appeared to be rather
complicated and did not follow simple first-order kinetics, and
a rate constant of approximately 2 × 10-3 s-1 (298.0 K, [BF3]
) 2 × 10-3 M) was obtained.
K[BF3]
-
ꢀapp ) A526/[MnO4 ]i
-
ꢀapp, ꢀadduct, ꢀ
, and [MnO4-]i are the apparent molar
MnO4
absorptivity, molar absorptivity of the adduct, molar absorptivity
of MnO4- (2530 M-1 cm-1 at 526 nm), and initial concentration
of MnO4-, respectively. A plot of ꢀapp vs (ꢀapp - ꢀ
)/[BF3]
-
MnO4
gives a straight line as shown in Figure 2, and from the slope
The effects of BF3 on k2 were studied by varying the
concentration of BF3 from 2 × 10-4 to 3 × 10-3 M, and
saturation kinetics were observed (Figure 5). The kinetic results
are consistent with an initial equilibrium formation of an adduct
as shown in eq 1, followed by rate-limiting oxidation of
cyclohexane by the adduct, as shown in eq 4.
K was determined to be 704 ( 31 M-1 at 25.0 °C. From the
intercept ꢀadduct (526 nm) was found to be 1430 ( 31 M-1 cm-1
Attempts to detect the adduct 2BF3‚MnO4 were problematic;
at higher concentrations of BF3 (>0.1 M) the decomposition
of the adduct became very rapid.
.
-
Similar spectrophotometric changes occurred when BF3 was
added to KMnO4 in the presence of cyclohexane, i.e., there was
a very rapid decrease in absorbance, followed by a much slower
decay (Figure 3). The change in absorbance for the first step
increases with BF3 concentration but is independent of cyclo-
hexane concentration, consistent with the formation of the adduct
[BF3‚MnO4]- + c-C6H12
9
k8 products
(4)
k2 is related to K and k by eq 5.
kK[BF3]
k2 )
(5)
(22) Ramette, R. W. Chemical Equilibrium and Analysis. Addison-Wesley
Publishing Company, Ltd.: Massachusetts, 1981; pp 426-439.
1 + K[BF3]
9
J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006 2853

A R T I C L E S
Lam et al.
Figure 4. (a) Kinetic trace for the BF3 (2 × 10-3 M) activated oxidation of cyclohexane (0.1 M) by MnO4- (5 × 10-5 M) in CH3CN at 298.0 K. The inset
shows the corresponding plot of ln(At - Af) vs t (ca. three half-lives, slope ) kobs). (b) Second-order plot of kobs vs [c-C6H12]. Slope ) k2 ) (6.95 ( 0.38)
× 10-1 M-1 s-1, y-intercept ) (1.03 ( 0.02) × 10-3 s-1, r ) 0.9971.
Figure 5. Plot of k2 vs [BF3] for the BF3 activated oxidation of cyclohexane (0.10 M) by MnO4- (5 × 10-5 M) in CH3CN at 298.0 K. The inset shows the
corresponding plot of 1/k2 vs 1/[BF3] (slope ) (1.23 ( 0.01) × 10-3, y-intercept ) (7.33 ( 0.21) × 10-1, r ) 0.9996).
The plot of 1/k2 versus 1/K is linear (Figure 5); k and K are
found to be (1.21 ( 0.03) M-1 s-1 and (752 ( 44) M-1
respectively at 25.0 °C by using a nonlinear least-squares fit of
the data to eq 5. The value of K is in good agreement with the
value 704 ( 31 M-1 obtained by direct spectrophotometric
determination. The overall rate law is shown in eq 6.
kcal mol-1 and 21 ( 2 cal K-1 mol-1 respectively. The
activation parameters ∆Hq and ∆Sq for the k step, obtained from
the plot of ln k/T versus 1/T, are 10.7 ( 1.3 kcal mol-1 and
-22 ( 2 cal K-1 mol-1, respectively.
The kinetics for the oxidation of c-C6D12 were also investi-
gated; kC H /kC D was found to be 5.3 ( 0.6 at 25.0 °C, in
6
12
6 12
good agreement with the value 4.7 ( 0.5 obtained by a
competition method (vide supra).
kK[BF3]
-
V )
[MnO4 ][c-C6H12]
(6)
Oxidation of Other Alkanes and Arylalkanes. The oxida-
tion of other alkanes (c-C5H12, c-C7H14, and c-C8H16) has also
been investigated. In all cases the corresponding ketone is the
major product (Table 1).
1 + K[BF3]
The maximum [BF3] used in kinetic experiments was around
3 × 10-3 M. When higher concentrations were used the
reactions became much faster and irreproducible. Experiments
done at 0.1 M BF3 indicated that k2 is over 10 times larger than
k, suggesting the existence of a more reactive species, which is
probably [2BF3‚MnO4]-.
The kinetics were studied over a 30 °C temperature range.
From the plot of ln K versus 1/T, ∆H° and ∆S° for the
equilibrium adduct formation (eq 1) are found to be 2.3 ( 0.5
The BF3/KMnO4 oxidation of various arylalkanes, including
toluene, substituted toluenes, ethylbenzene, cumene, diphenyl-
methane, and triphenylmethane, has also been studied. No new
absorption in the UV-vis region arises when BF3‚CH3CN is
mixed with these aromatic compounds (in the absence of
MnO4-) in CH3CN, suggesting that there is negligible interaction
between BF3 and these substrates.
9
2854 J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006

-
BF3-Activated Oxidation of Alkanes by MnO4
A R T I C L E S
Table 1. Product Yields for the Oxidation of Hydrocarbons by
KMnO4/BF3
Table 2. Rates and Activation Parameters for the Oxidation of
Toluene by [BF3‚MnO4]- and nBu4N[MnO4]
a
a
-b
substrate
productb (% yield)
[BF3
‚
MnO4]-
MnO4
cyclohexane
cyclooctane
cyclohexanol (8), cyclohexanone (73)
cyclooctanol (4), cyclooctanone (76)
cyclopentanol (11), cyclopentanone (39)
benzaldehyde (20)
acetophenone (45)
2-phenyl-2-propanol (26)
benzophenone (46)
rate constant at 298 K/M-1 s-1
∆Hq/kcal mol-1
9.2
4.2 × 10-7
21.0 ( 1.0
-(16 ( 3)
24.6 ( 0.9
12.0 ( 0.6
-(14 ( 3)
12.6 ( 0.4
cyclopentane
toluene
ethylbenzene
iso-propylbenzene
diphenylmethane
triphenylmethane
∆Sq/cal mol-1 K-1
Ea/kcal mol-1
a Reaction in CH3CN. b Reaction in neat toluene. Data from ref 17.
triphenylmethanol (20)
Table 3. Representative Rate Data for the Oxidation of
- a
Hydrocarbons by BF3‚MnO4
a KMnO4, 4.3 × 10-3 M; BF3, 8.6 × 10-2 M; substrate, 1.0 M. T ) 23
°C. Time ) 5 min. All reactions were carried out in CH3CN under argon.
O2 was strictly excluded from the reaction by four freeze-pump-thaw
cycles. b The yields were calculated based on the system acting as a three-
electron oxidant.
1
1
substrate
k (BF3/KMnO4)/M-1 s-
BDE/kcal mol-
c-C5H10
c-C6H12
c-C7H14
c-C8H16
6.90 × 10-1
96.4b
95.4b
94.0b
92.6b
88.5c
84.7c
84.4d
82c
1.21
4.55
6.38
9.22
PhCH3
The product from the oxidation of toluene consists of 20%
benzaldehyde, no change in the yield was observed and no
benzoic acid was detected even after adding acidic iodide to
reduce MnO2 to Mn2+. The detection limit of benzoic acid by
GC under our conditions is around 5%. No products arising
from benzene ring oxidation could be detected. Also no products
such as methyl substituted diphenylmethanes or methylben-
zophenone, which arise from the benzyl cation, could be found
(detection limit < 1%), suggesting that [BF3‚MnO4]- does not
oxidize toluene via a hydride abstraction mechanism.23 The
“unassisted” oxidation of toluene by nBu4NMnO4 produces
predominantly benzoic acid;17 such a difference in product
PhCH2CH3
PhCH(CH3)2
Ph2CH2
2.60 × 101
1.32 × 101
1.36 × 102
1.88 × 102
2.55 × 101
4.55 × 101
5.85
Ph3CH
81c
p-H3CO-PhCH3
p-H3C-PhCH3
p-Cl-PhCH3
a T ) 298 K, in CH3CN. b Reference 26. c References 17 and 27.
d Reference 28.
The kinetics for the oxidation of the other substrates were
carried out at [MnO4-] ) 5 × 10-5 M, [substrate] ) 1 × 10-3
-
2 × 10-1 M, and [BF3] ) 2.0 × 10-3 M. The k values were
calculated from k2 by using eq 5 with K ) 752 M-1. Results
are shown in Table 3. The KIE for the oxidation of ethylbenzene
is kC H /kC D ) 7.1 ( 0.5 (at 25.0 °C).
distribution is probably due to a much faster rate of oxidation
-
by [BF3‚MnO4]- than MnO4
.
The oxidation of ethylbenzene, isopropylbenzene, diphenyl-
methane, and triphenylmethane gave acetophenone (45%),
2-phenyl-2-propanol (26%), benzophenone (46%), and triphen-
ylmethanol (20%), respectively. No other products could be
detected.
8
10
8 10
In the oxidation of various para-substituted toluenes, electron-
donating substituents are found to cause an increase in rates
while electron-withdrawing substituents cause a decrease in
rates. Although the correlations are poor, the rates are monotonic
with σp or σ+ values, with F ) -(1.38 ( 0.22) and F+ ) -(0.69
( 0.25) (Figure S3). This is in contrast to oxidation by Bu4-
NMnO4 alone, where both electron-donating and electron-
withdrawing substituents are found to cause an increase in rates.
The yields for the oxidation of the alkanes and arylalkanes
by BF3/MnO4- range from 20 to 80%. Lower yields are obtained
for the oxidation of the arylalkanes, despite their lower C-H
bond energies. Presumably further oxidation of the alcohol and
ketone products occurs for these substrates, leading to unidenti-
fied ring degradation species. A similar result is also found for
n
Mechanism. Our results suggest that BF3 forms an adduct
with KMnO4 in CH3CN, which is the active species responsible
for the oxidation of the hydrocarbons. This adduct is too reactive
to isolate or characterize; however a number of much less-
oxidizing metal-oxo species have been known to form stable
adducts with boranes such as B(C6F5)3.24,25 The X-ray crystal
structures of these adducts indicate that the boron is bonded to
an oxo ligand, i.e., MdOsB(C6F5)3. In general the BsO
distance is consistent with a single bond, while there is
substantial lengthening of the MdO bond. It is reasonable to
assume that BF3 also forms an adduct with MnO4- via bonding
to an oxo ligand. Electron-withdrawing by the Lewis acidic BF3
n
the unassisted oxidation of arylalkanes by Bu4NMnO4.17
Kinetic studies for the oxidation of these substrates by
KMnO4/BF3 have also been carried out. For the oxidation of
toluene, the rate law is the same as that of cyclohexane; k and
K are found to be (9.22 ( 0.25) M-1 s-1 and (753 ( 42) M-1
respectively at 25.0 °C. The value of K is the same as that for
cyclohexane, as expected if it represents the equilibrium constant
for adduct formation, which should be independent of substrate.
Notably, k, which represents the rate constant for the oxidation
of toluene by [BF3‚MnO4]-, is over 7 orders of magnitude larger
n
than that by Bu4NMnO4 (4.2 × 10-7 M-1 s-1 at 25.0 °C).17
would enhance the oxidizing power of MnVII
.
∆Sq values for the two pathways are similar, the much faster
rate for the BF3-activated pathway arises from a much lower
∆Hq (Table 2). On the other hand, the KIE for the BF3-activated
pathway, kC H /kC D ) 6.8 ( 0.5 (at 25.0 °C), is comparable to
The rate-determining step (rds) in the oxidation of these
substrates most likely involves H-atom abstraction by the adduct
7
8
7 8
the value 6 ( 1 (at 45 °C) for the unactivated pathway.17
(24) Wolff, F.; Choukroun, R.; Lorber, C.; Donnadieu, B. Eur. J. Inorg. Chem.
2003, 628-632.
(25) (a) Galsworthy, J. R.; Green, M. L. H.; Mu¨ller, M.; Prout, K. J. Chem.
Soc., Dalton Trans. 1997, 1309-1313. (b) Galsworthy, J. R.; Green, J. C.;
Green, M. L. H.; Mu¨ller, M. J. Chem. Soc., Dalton Trans. 1998, 15-19.
(c) Doerrer, L. H.; Galsworthy, J. R.; Green, M. L. H.; Leech, M. A. J.
Chem. Soc., Dalton Trans. 1998, 2483-2487.
(23) (a) Lockwood, M. A.; Wang, K.; Mayer, J. M. J. Am. Chem. Soc. 1999,
121, 11894-11895. (b) Larsen, A. S.; Wang, K.; Lockwood, M. A.; Rice,
G. L.; Won, T.-J.; Lovell, S.; Sad´ılek, M.; Turecˇek, F.; Mayer, J. M. J.
Am. Chem. Soc. 2002, 124, 10112-10123.
9
J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006 2855

A R T I C L E S
Lam et al.
Table 4. Reaction Barriers (∆Eq in kcal/mol) and Reaction
Energies (∆E in kcal/mol) for the Three Reactions Calculated
∆
Eq
∆E
reaction I
reaction II
reaction III
32.1
26.2
20.5
-34.2
-38.2
-56.9
t
t
abstraction by OH•, BuO•, and BuOO•.17 If we assume that
the rates of abstraction by BF3/KMnO4 also follow this
correlation, then the OsH bond strength in [F3BsOdMnVI(O)2-
(OsH)]- can be estimated to be around 95 kcal/mol from the
plot of log k vs the strength of the OsH bond formed by the
oxidants.17 This value is 15 kcal/mol larger than the OsH bond
strength in HMnO4 (80 kcal mol-1).
Computational Study. The accelerating effect of BF3 on the
oxidation of alkanes by MnO4- has been studied computation-
ally. A previous theoretical study34 on the oxidation of CH4 by
MnO4- showed that the crucial step in the oxidation of CH4 by
MnO4- is the transformation of MnO4- + CH4 f [O2Mn(OH)-
(OCH3)]-; the barrier for the transformation was calculated to
be 32.3 kcal/mol. A hydrogen abstraction and oxygen rebound
mechanism was postulated. The previous theoretical study also
showed that the species involved in the reaction are well
described by closed shell wave functions.34 To examine the BF3
accelerating effect, the following three model reactions were
investigated with the aid of closed-shell self-consistent field
(SCF) calculations.
Figure 6. Plot of log k′ vs BDE of the substrates for the BF3 activated
oxidation of hydrocarbons by MnO4- in CH3CN at 298.0 K (slope ) -(1.94
( 0.11) × 10-1, y-intercept ) (1.77 ( 0.09) × 101, r ) -0.9900).
to produce a carbon radical (eq 7), similar to oxidation by
-
“unactivated” MnO4
.
[F3B-OMnVII(O)3]- + RH f [F3B-OMnVI(O)2(OH)]- +
R• slow (7)
[F3B-OMnVI(O)2(OH)]- + R• f [F3B-OMnV(O)2]- +
ROH (8)
Reaction I: MnO4- + CH4 f [O2Mn(OH)(OCH3)]-
OH• rebound to R• will form ROH (eq 8), which is then
rapidly oxidized to the corresponding aldehyde or ketone by
[F3B-OMnV(O)2]- and [F3B-OMnVII(O)3]-. The observation
that O2 affects the product yields is consistent with the formation
of radicals in the rds.29
Reaction II: MnO4(BF3)- + CH4 f
[O2Mn(OH)(OCH3)(BF3)]-
Reaction III: MnO4(BF3)2- + CH4 f
Correlation between Rate Constants and CsH Bond
Strengths (BDE). Following the work of Mayer,17 log k′ is
plotted against CsH BDE of the hydrocarbons. k′ is k corrected
for both the number of reactive hydrogen atoms in each substrate
and the stoichiometry of the reaction.30 A good linear correlation
is found (Figure 6), which is consistent with a hydrogen atom
transfer mechanism for the BF3-activated oxidation of hydro-
carbons by MnO4-. Similar correlations have been observed in
the oxidation of hydrocarbons by CrO2Cl2,31 nBu4NMnO4,17 and
trans-[RuVI(N2O2)(O)2]2+ 32 and in the oxidation of phenols by
[O2Mn(OH)(OCH3) (BF3)2]-
Geometry optimizations at the B3LYP level density functional
theory,35,36 with the 6-311+G** basis sets for the nonmetal
atoms and the LanL2DZ basis sets for Mn,37,38 have been
performed to calculate the energies and structures of all species
involved in the three reactions. The results of the calculations
are listed in Table 4, the calculated transition states are shown
in Figure 7, and the structural details of all the calculated species
are given in the Supporting Information. The barrier for reaction
I is calculated to be 32.1 kcal/mol, consistent with the result
from the previous theoretical study,34 while, in reactions II and
III, the barriers are 26.2 and 20.5 kcal/mol, respectively. The
lower barriers for reactions II and III explain the experimental
observation that the oxidation of alkanes by MnO4- is acceler-
ated by BF3. The calculated barriers for methane oxidation are
higher than the experimental values (Table 2) for toluene
oxidation, as expected due to the higher C-H bond energy of
methane (104.8 kcal/mol). It should also be noted the barriers
for methane oxidation are gas-phase calculations, which should
be somewhat different from the solution values.
trans-[RuVI(N2O2)(O)2]2+ 33
. The rates of hydrogen atom transfer
n
from arylalkanes to Bu4NMnO4 also correlate with rates of
(26) Luo, Y. R. Handbook of Bond Dissociation Energies in Organic Com-
pounds; CRC Press: Boca Raton, FL, 2003.
(27) (a) Parker, V. D. J. Am. Chem. Soc. 1992, 114, 7458-7462. (b) Bordwell,
F. G.; Cheng, J. P.; Ji, G. Z.; Satish, A. V.; Zhang, X. J. Am. Chem. Soc.
1991, 113, 9790-9795.
(28) CRC Handbook of Chemistry and Physics, 82nd ed.; Lide, D. R., Ed. CRC
Press: Boca Raton, FL.
(29) The fact that the reaction rate is not affected may be due to more efficient
trapping of R• by O2 than by Mn(VII) under our experimental conditions
([Mn(VII)] ) 1 × 10-5-5 × 10-5 M in kinetics studies).
-
(30) According to the proposed mechanism, the rate of disappearance of MnO4
in the reaction with cyclohexane should be 4/3 times the rate of hydrogen
4
-
atom transfer, because
/ mol of MnO4 are required to make 1 mol of
cyclohexanone. Also ther3e are 12 reactive H atoms in cyclohexane; hence
k′ ) k × /12 × /4. For Ph3CH, however, the observed rate is only 2/3 the
rate of hydrogen atom transfer, since the product is Ph3COH; in this case
k′ ) k × 3/2. We thank one of the reviewers for pointing this out.
(31) Cook, G. K.; Mayer, J. M. J. Am. Chem. Soc. 1995, 117, 7139-7156.
(32) Lam, W. W. Y.; Yiu, S. M.; Yiu, D. T. Y.; Lau, T. C.; Yip, W. P.; Che,
C. M. Inorg. Chem. 2003, 42, 8011-8018.
(34) Strassner, T.; Houk, K. N. J. Am. Chem. Soc. 2000, 122, 7821-7822.
(35) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5152. (b) Miehlich, B.;
Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200-206. (c)
Lee, C.; Yang, W.; Parr, G. Phys. ReV. B 1988, 37, 785-789.
(36) Frisch, M. J.; et al. Gaussian 03, revision B05; Gaussian, Inc.: Pittsburgh,
PA, 2003.
1
3
(33) Yiu, D. T. Y.; Lee, M. F. W.; Lam, W. W. Y.; Lau, T. C. Inorg. Chem.
2003, 42, 1225-1232.
(37) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213-222.
(38) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299-310.
9
2856 J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006

-
BF3-Activated Oxidation of Alkanes by MnO4
A R T I C L E S
Figure 8. Spatial plots of the highest occupied molecular orbitals (HOMO),
together with their orbital energies, for the three transition states of reactions
I, II, and III.
extensively delocalized on both the metal center and the CH3
group. In view of the significant contribution from the CH3
group to each HOMO orbital in each transition state, we can
envision that the H-CH3 bond breaking is related to a hydrogen
abstraction process, giving a radical CH3 and a formal d1 Mn-
(VI) center in the transition state. A similar hydrogen abstraction
resulting in a radical pair species has been emphasized in a
recent theoretical study of the methane activation on molybde-
num oxides.40
The decrease in the reaction barriers resulted from the BF3
coordination is driven by the increasing ability of the metal
center to achieve a d1 Mn(VI) electron configuration in the
transition state. Figure 8 shows that significant Mn-O anti-
bonding character can be found in each of the HOMOs. From
the orbital energies shown in Figure 8, we can see that
coordination of BF3 significantly stabilizes the HOMOs and in
turn facilitates the hydrogen abstraction process that passes the
hydrogen’s electron to the metal center. In other words, we can
say that the BF3 coordination to the lone pairs on the oxo groups
reduces the antibonding character in the d orbital which
Figure 7. Optimized structures of transition states together with the reaction
barriers. TSI, TSII, and TSIII are the transition states of reactions I, II,
and III, respectively.
The calculated transition state structures (Figure 7) for the
three reactions show no significant changes in the structural
parameters related to the bond breaking and formation. The
coordination of BF3 makes the transition states earlier with
respect to reactants. It should be noted that we also studied
transition structures having other BF3-coordination sites. In other
words, transition structures in which BF3 is coordinated to
oxygens other than those shown in Figure 7 were calculated.
The results of these calculations show that coordination to other
sites has smaller effects in decreasing the reaction barrier. On
the basis of the transition state structure for reaction I (TSI),
we also calculated the single-point energy of its triplet state
and found that the triplet state is less stable than the singlet
state by 22.1 kcal/mol, further supporting that employing the
closed-shell calculations is appropriate.
accommodates the metal d electron in the transition state.
-
Formally there are no d electrons in the metal center of MnO4
,
and therefore the stabilizing effect to the reactant (MnO4-) is
much less significant when compared with the transition state.
From the reaction energies shown in Table 4, we can see
that binding the first BF3 makes the reaction more favorable
by 4.0 kcal/mol, slightly smaller than the stabilization energy
(5.9 kcal/mol) gained by the transition state. However, binding
the second BF3 makes the reaction more favorable by 18.7 kcal/
Figure 8 shows the plots39 of the HOMO orbitals of the three
transition states for the three reactions. Each HOMO orbital is
(40) (a) Fu, G.; Xu, X.; Lu, X.; Wan, H. J. Am. Chem. Soc. 2005, 127, 3989-
3996. (b) Fu, G.; Xu, X.; Lu, X.; Wan, H. J. Phys. Chem. B 2005, 109,
6416-6421.
(39) Schaftenaar, G. Molden V3.5; CAOS/CAMM Center Nijimegen: Toer-
nooiveld, Nijimegen, Netherkands, 1999.
9
J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006 2857

A R T I C L E S
Lam et al.
mol, significantly greater than the stabilization energy (5.7 kcal/
mol) gained by the transition state. The steric repulsion between
the Me and BF3 groups in the product structure may significantly
reduce the stabilization gained by binding the first BF3. The
steric effect is less important in the transition state structures.
Finally, we wish to point out that based on the calculation
the lowest activation energy is achieved by the 2:1 adduct
[2BF3‚MnO4]-. Experimentally there is some evidence for this
adduct; at [BF3] > 0.1 M, the oxidation of cyclohexane becomes
much faster than that expected for the 1:1 adduct [BF3‚MnO4]-.
Unfortunately the kinetics are irreproducible, thus preventing
detailed studies.
effects have been attributed to a significant decrease in the
reaction barriers resulting from the BF3 coordination. The BF3
coordination increases the ability of the Mn metal center to
achieve a d1 MnVI electron configuration in the transition state.
The lowest activation energy is achieved by the 2:1 adduct
[2BF3‚MnO4]-.
Acknowledgment. The work described in this paper was
supported by the Research Grants Council of Hong Kong (CityU
1097/01P) and the City University of Hong Kong (7001363).
We thank the reviewers for valuable comments.
Supporting Information Available: Complete ref 35, table
of rate data, UV/vis spectra, product yields, and the Cartesian
coordinates of all the DFT calculated structures. This material
is available free of charge via the Internet at http://pubs.acs.org.
Conclusion
BF3 forms an adduct with MnO4- that oxidizes hydrocarbons
at rates over 7 orders of magnitude faster than that by MnO4
alone. Experimental and computational studies support a
hydrogen atom abstraction mechanism. The BF3 accelerating
-
JA0552951
9
2858 J. AM. CHEM. SOC. VOL. 128, NO. 9, 2006