Kinetic isotope effects for electron-transfer pathways in the oxidative C-H activation of hydrocarbons

Article abstract of DOI:10.1021/ja974205s

Fast hydrogen atom transfers from various methylbenzenes (ArH) to photoactivated quinones Q* show primary kinetic isotope effects k(H)/k(D) of 2.4-5.6. The quantitative effects of added inert salt on the kinetics and on the yields of the intermediate cation radical ArH(+*) demonstrate that hydrogen transfer proceeds via a two-step sequence involving an initial electron transfer to form the ion-radical pair [ArH(+*), Q(-*)] which subsequently undergoes proton transfer according to the electron-transfer mechanism in Scheme 1.

Full text of DOI:10.1021/ja974205s

2826
J. Am. Chem. Soc. 1998, 120, 2826-2830
Kinetic Isotope Effects for Electron-Transfer Pathways
in the Oxidative C-H Activation of Hydrocarbons
T. M. Bockman, S. M. Hubig, and J. K. Kochi*
Contribution from the Chemistry Department, UniVersity of Houston, Houston Texas 77204-5641
ReceiVed December 12, 1997
Abstract: Fast hydrogen atom transfers from various methylbenzenes (ArH) to photoactivated quinones Q*
show primary kinetic isotope effects k(H)/k(D) of 2.4-5.6. The quantitative effects of added inert salt on the
kinetics and on the yields of the intermediate cation radical ArH+• demonstrate that hydrogen transfer proceeds
via a two-step sequence involving an initial electron transfer to form the ion-radical pair [ArH+•, Q-•] which
subsequently undergoes proton transfer according to the electron-transfer mechanism in Scheme 1.
Introduction
Chart 1
Facile carbon-hydrogen bond activation is critical to the
synthetically important oxidative functionalizations of toluene
and related hydrocarbons (ArH) to produce topical benzalde-
hydes, benzoic acids, and various benzyl derivatives with
different (1 - ꢀ) organic and inorganic oxidants.1,2 However,
there is a recurring mechanistic ambiguity as to whether the
initial C-H scission occurs (a) in a single step involving direct
hydrogen-atom transfer3 or (b) via prior electron transfer
followed by proton transfer.4,5 Deuterium labeling8 has been
used to address this important mechanistic question, and the
observation of H(D) isotope effects on the kinetics of oxidation
has been taken as conclusive evidence for the one-step formula-
tion.9 However, we now demonstrate that such a simple
conclusion is incorrect and that (primary) deuterium isotope
(1) (a) Wiberg, K. B.; Evans, R. J. Tetrahedron 1960, 8, 313. (b) Hartford,
W. H.; Darrin, M. Chem. ReV. 1958, 58, 1. (c) Onopchenko, A.; Schulz, J.
G. D.; Seekircher, R. J. Org. Chem. 1972, 37, 1414. (d) Heiba, E. I.; Dessau,
R. M.; Koehl Jr., W. J. J. Am. Chem. Soc. 1968, 90, 1082. (e) For a review
on oxidations by metal complexes, see: Mijs, W. J.; de Jonge, C. R. H. I.
Organic Syntheses by Oxidation with Metal Compounds; Plenum: New
York, 1986.
(2) (a) Becker, H.-D. In The Chemistry of the Quinone Compounds, Part
1; Patai, S., Ed.; Wiley: New York, 1974; p 335. (b) Dauben, H. J., Jr.;
Gedecki, F. A.; Harmon, K. M.; Pearson, D. L. J. Am. Chem. Soc. 1957,
79, 4557. (c) Longone, D. T.; Smith, G. L. Tetrahedron Lett. 1962, 205.
(d) Lee, I. S. H.; Jeong, E. H.; Kreevoy, M. M. J. Am. Chem. Soc. 1997,
119, 2722.
(3) (a) Wagner, P. J.; Truman, R. J.; Puchalski, A. E.; Wake, R. J. Am.
Chem. Soc. 1986, 108, 7727. (b) Wagner, P. J.; Thomas, M. J.; Puchalski,
A. E. J. Am. Chem. Soc. 1986, 108, 7739.
(4) (a) Traylor, T. G.; Nakano, T.; Miksztal, A. R.; Dunlap, B. E. J. Am.
Chem. Soc. 1987, 109, 3625. (b) Jones, G., II; Mouli, N. J. Phys. Chem.
1988, 92, 7174. (c) Wagner, P. J. Top. Curr. Chem. 1976, 66, 1.
(5) The initial electron-transfer step in case (b) is generally reversible,
so the latter case represents a classic example of a preequilibrium.6
Unambiguous proof of preequilibria is a challenging experimental task,
which is not readily accomplished by conventional kinetic measurements.
Instead, additional approaches such as temperature studies, interception of
precursors, and direct observation of the complete reaction sequence by
time-resolved spectroscopic techniques must be brought to bear.7
(6) Bunnett, J. F. In InVestigation of Rates and Mechanisms of Reactions,
Part 1; Bernasconi, C. F., Ed.; Wiley: New York, 1986; p 268 f.
(7) (a) Baggott, J. E. In Photoinduced Electron Transfer; Fox, M. A.,
Chanon, M., Eds.; Elsevier: New York; 1988, Part B, p 385. (b) Strehlow,
H.; Knoche, W. Fundamentals of Chemical Relaxation; Verlag Chemie:
New York, 1977.
effects of appreciable magnitude are perfectly consistent with
an electron-transfer mechanism for oxidative C-H bond activa-
tion.
Since quinones are effective dehydrogenation reagents as well
as electron acceptors,2a,10 we will focus our studies on the use
of laser-activated quinones Q* with the enhanced reduction
potentials (E* ) in Chart 1. Time-resolved absorption spec-
red
troscopy then allows the direct observation of all the short-lived
intermediates involved in the diffusional interaction of Q* with
the various methylbenzene donors (ArH with oxidation poten-
tials E0ox) in Chart 1.
Results and Discussion
I. Spontaneous Activation of Quinone and Facile Hydro-
gen Transfer from Methylbenzenes. The photoactivation of
quinone is spontaneously effected in dichloromethane by a 25-
(9) (a) Gardner, K. A.; Mayer, J. M. Science 1995, 269, 1849. (b)
Gardner, K. A.; Kuenert, L. A.; Mayer, J. M. Inorg. Chem. 1997, 36, 2069.
(c) Ho¨fler, C.; Ru¨chardt, C. Liebigs Ann. 1996, 183.
(10) Rathore, R.; Hubig, S. M.; Kochi, J. K. J. Am. Chem. Soc. 1997,
119, 11468.
(8) (a) Carlson, B. W.; Miller, L. L. J. Am. Chem. Soc. 1985, 107, 479.
(b) Pajak, J.; Brower, K. R. J. Org. Chem. 1985, 50, 2210. (c) Dinktu¨rk,
S.; Ridd, J. H. J. Chem. Soc., Perkin Trans. 2 1982, 1961.
S0002-7863(97)04205-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/12/1998

KIE for Electron-Transfer ActiVation of Hydrocarbons
J. Am. Chem. Soc., Vol. 120, No. 12, 1998 2827
Figure 1. Spectral changes attendant upon hydrogen transfer from
0.005 M durene to Q* generated by the 355-nm excitation of 0.002 M
Q(CX) in dichloromethane at 23 °C. Inset shows the digital fit (smooth
line) of the Q* decay and Q(H)• growth to first-order kinetics with
kobs ) 3.6 and 4.3(( 0.5) × 106 s-1 as followed at 500 and 335 nm,
respectively.
Figure 2. Pseudo-first-order plots of kobs with durene (b) and durene-
d14 (O) in dichloromethane, as described in Figure 1.
Table 1. Kinetic Isotope Effect for Hydrogen Transfer from
Methylbenzenes to Q* a
Q/ArHb k2(H) [M-1 s-1
]
k2(D) [M-1 s-1 k2(H)/k2(D) (kH/kD)PT
]
ps laser pulse at 355 nm to yield the characteristic spectrum of
Q* with unit efficiency.10,11 The subsequent first-order decay
of Q* in the presence of ArH is accompanied by the simulta-
neous appearance of the reduced Q(H)•12 and the benzylic
radical (Ar•)13 in equimolar amounts. As such, hydrogen
transfer according to eq 1 is conveniently monitored by the
temporal evolution of the diagnostic absorption spectra of both
Q* and Q(H)• previously reported.14
CX/DUR
CX′/DUR
CX/HMB
CA/TOL
CA/DUR
CA/BIP
2.4 × 108
3.7 × 108
d
7.9 × 107
1.3 × 108
d
3.0 (3.3)
2.8 (c)
d (3.1)
2.6 (1.6)
1.7 (2.4)
0.99
2.4
5.6
4.4
e
1.8 × 107
1.2 × 1010
4.2 × 108
6.8 × 106
7.2 × 109
4.3 × 108
2.8
f
a Measured as bimolecular rate constants k2 for Q* and ArH upon
the 355-nm excitation (15 mJ per pulse) of 0.005 M Q and various
concentrations of ArH in dichloromethane as described in the text.
b Quinone and methylbenzene identified in Chart 1. c Not measured.
d Measured only under diffusion-free conditions (as described in the
Experimental Section). e No free ions observed. f No hydrogen transfer
observed from biphenyl (BIP), as described in ref 22.
k2
9
Q* + ArH
8 Q(H)• + Ar•
(1)
Hydrogen transfer from durene as a typical ArH (in excess)
is illustrated in Figure 1 by the monotonic disappearance of
Q*, and it is accompanied by the concomitant production of
the reduced quinone radical Q(H)•.14 The rate of hydrogen
transfer is conveniently monitored by the temporal evolution
of the diagnostic spectra of Q* and Q(H)•.14 In Figure 1, the
spectral band of Q* with λmax ) 500 nm decreases smoothly
as the sharp band of the reduced quinone Q(H)• at λmax ) 435
nm increases. The assignment of the 435-nm band to Q(H)• is
confirmed by the simultaneous appearance of its UV band at
360 nm.10 The isosbestic points at 370, 400, and 440 nm
indicate that the transformation occurs uniformly,15 and the
decay of Q* and the rise of Q(H)• both follow first-order kinetics
with the same rate constant (kobs) as shown by the digital fits in
the inset to Figure 1 to confirm a simple kinetic formulation.
The linear variation in kobs as a function of ArH concentration
is shown in Figure 2, and the rate constant for the bimolecular
reaction in eq 1 is simply given by the slope of this pseudo-
first-order plot. [See the Experimental Section for details of
this analysis.] The values of the bimolecular rate constants (k2)
for various combinations of methylarenes and quinones are
compiled in Table 1.
II. Kinetic Isotope Effect for Hydrogen Transfer. The
effect of isotopic substitution is readily evaluated by comparing
the bimolecular rate constant k2(D) for perdeuterated ArD with
the rate constant k2(H) for its protiated parent. For example,
Figure 2 shows the distinctly slower rates which govern the
abstraction of deuterium from durene-d14, and the isotope effect
on k2 is readily evaluated from the slope of the kinetics plot.
The isotope effects (k2H/k2D) for various methylbenzenes are
listed in Table 1 (fourth column). It is noteworthy that the
kinetic isotope effects require the presence of a benzylic
hydrogen, since there is no difference in the rates with which
Q* is reduced by biphenyl or biphenyl-d10 (entry 6).22
III. Interception of the Transient Intermediates by Added
Salt. Let us now address the question as to whether the
observed kinetic isotope effects in Table 1 rule out an electron-
transfer mechanism. To probe the possibility of such an
alternative pathway, we employ inert salt to intercept the ion-
(11) Hubig, S. M.; Bockman, T. M.; Kochi, J. K. J. Am. Chem. Soc.
1997, 119, 2926.
(12) (a) Kobashi, H.; Funabashi, M.-A.; Kondo, T.; Morita, T.; Okada,
T.; Mataga, N. Bull. Chem. Soc. Jpn. 1984, 57, 3557. (b) Kobashi, H.;
Okada, T.; Mataga, N. Bull. Chem. Soc. Jpn. 1986, 59, 1975. (c) Jones, G.,
II; Haney, W. A.; Phan, X. T. J. Am. Chem. Soc. 1988, 110, 1922.
(13) (a) Christensen, H. C.; Sehested, K.; Hart, E. J. J. Phys. Chem. 1973,
77, 983. (b) The absorption of the trimethylbenzyl radical Ar• at λmax
)
310 nm is obscured by the overlapping absorptions of the quinone radical
and other long-lived products of the hydrogen transfer.
(14) (a) For the quinone transients, the diagnostic absorption maxima
are as follows: CA* (510 nm, ꢀ ) 7600 M-1 cm-1), CA(H)• (435 nm, ꢀ
) 7700 M-1 cm-1), CA- • (450 nm, ꢀ ) 9600 M-1 cm-1), CX* (500 nm,
ꢀ ) 5300 M-1 cm-1), CX(H)• (340 nm, ꢀ ) 5300 M-1 cm-1), and CX- •
(430 nm, ꢀ ) 6800 M-1 cm-1) as described in ref 10. (b) Note that the
decay rate of Q* in the absence of ArH is much slower (kobs < 105 s-1
than the rates of hydrogen transfer in eq 1. See ref 10.
)
(15) Moore, J. W.; Pearson, R. G. Kinetics and Mechanism, 3rd ed.;
Wiley: New York, 1981; p 49.

2828 J. Am. Chem. Soc., Vol. 120, No. 12, 1998
Bockman et al.
Figure 3. Salt effects with TBA+PF6 on (A) the ion yields of ArH+• and (B) the first-order rate constant for the reaction of Q*(CX) with HMB
(b) and HMB-d18 (O) in dichloromethane by the procedure described in Figure 1.
-
Table 2. Effect of Added Salt on the Quantum Yield of ArH+•
and the Pseudo-First-Order Rate Constant
radical pair intermediate prior to proton transfer in the following
way. Upon the addition of the neutral electrolyte, tetra-n-
[salt] (M)c
butylammonium hexafluorophosphate (TBA+ PF6-), the tran-
sient spectrum of Q(H)• in Figure 1 is replaced by that of the
anion radical Q-•. Most importantly, the characteristic spectral
band of the arene cation radical, ArH+•,16 which is not present
in the absence of TBA+ PF6-, now becomes a dominant feature.
For example, in the presence of TBA+ PF6-, a dichloromethane
solution of hexamethylbenzene and quinone produces high
yields of the quinone anion radical (with λmax ) 330 and 430
nm) and the methylbenzene cation radical (with λmax ) 495
nm) in equimolar amounts. In other words, added salt drasti-
cally alters the original spectral transformation from Q* to the
radical Q(H)• in eq 1 to one that generates the anion Q-•, i.e.
a
Φion
0.02 0.05 0.10 0.20
0.20 0.37 0.55 0.72 3.6 4.2
kobsb [107M-1 s-1
]
Q/ArHd
0 0.05 0.10 0.20
CX/HMB
4.3
3.7
4.1
4.0
CX/HMB-d18 0.52 0.76 0.84 0.90 1.7 3.5
CA/DUR 0.72 0.78 0.88 0.96
CA/DUR-d14 0.84 1.00 1.00 1.00
CX/DUR
0.10 0.17 0.26 0.39 1.7 2.3
2.8
1.2
3.3
1.6
CX/DUR-d14 0.17 0.22 0.39 0.57 0.6 0.9
CA/HMB
0.57 0.78 0.87 1.00
CA/HMB-d18 0.85 0.99 1.00 1.00
a Yield of cation radicals ArH+• as determined by transient acti-
nometry, see text. b Observed (pseudo-first-order) rate constant ((0.05)
for the decay of excited quinone (Q*) upon the 355-nm excitation (15
mJ per pulse) of 0.005M Q and 0.008 M hexamethylbenzene (HMB)
or 0.12 M durene (DUR) in dichloromethane at 23 °C. c Tetrabutyl-
ammonium hexafluorophosphate (see also ref 18 and the Experimental
Section). Φion ) 0 in the absence of salt with all Q/ArH combinations.
d Quinone and methylbenzene identified in Chart 1.
+
-
TBA PF6
Q* + ArH
8 Q-• + ArH+•
(2)
The quantitative effects of added salt on the yields of the ions
Q-• and ArH+• in eq 2 are evaluated in Table 2 as the quantum
yields Φion by using benzophenone as the transient actinom-
eter10,17 (see the Experimental Section). Figure 3A demonstrates
the yield of ArH+• to increase monotonically with increasing
salt concentration18 in accord with the progressive replacement
of Q(H)• with Q-• in the transient spectra.
In contrast to the strong effect of added salt on the product
yields, the effect on their rates of formation is insignificant.19
Figure 3B shows the rate constant for hydrogen transfer from
hexamethylbenzene to Q* (filled circles) to be essentially
unaffected by the addition of TBA+ PF6-. Similarly, there is
only a minor effect of added salt on the kinetics (kobs) of the
hydrogen abstraction from durene in Table 2.
(16) For the methylbenzene transients, the diagnostic absorption maxima
are as follows: HMB+• (495 nm; ꢀ ) 2500 M-1 cm-1), DUR+• (480 nm;
ꢀ ) 2400 M-1 cm-1), and TOL (420 nm). See (a) Bockman, T. M.;
Karpinski, Z. J.; Sankararaman, S.; Kochi, J. K. J. Am. Chem. Soc. 1992,
114, 1970. (b) Sehested, K.; Holcman, J.; Hart, E. J. J. Phys. Chem. 1977,
81, 1363.
(17) Hurley, J. K.; Sinai, N.; Linschitz, H. Photochem. Photobiol. 1983,
38, 9.
-
(18) Aggregation of TBA+ PF6 is not significant in this concentration
range. See: Sigvartsen, T.; Gestblom, B.; Noreland, E.; Songstad, J. Acta
Chem. Scand. 1989, 43, 103 (compare the results for TBA+PF6-, TBA+ClO4-
,
-
and PPN+PF6 in the Experimental Section).
(19) (a) Gordon, J. E. The Organic Chemistry of Electrolyte Solutions;
Wiley: New York, 1975; p 99 f. See also: (b) Masnovi, J. M.; Kochi, J.
K. J. Am. Chem. Soc. 1985, 107, 7880. (c) Yabe, T.; Kochi, J. K. J. Am.
Chem. Soc. 1992, 114, 4491.
(20) Note that any subsequent proton transfer for the separated ion
radicals occurs by second-order kinetics and is therefore negligibly slow.
(21) (a) Schlesener, C. J.; Amatore, C.; Kochi, J. K. J. Am. Chem. Soc.
1984, 106, 7472. (b) Masnovi, J. M.; Sankararaman, S.; Kochi, J. K. J.
Am. Chem. Soc. 1989, 111, 2263.
IV. Electron-Transfer Pathway for Hydrogen Transfer.
The foregoing effect of an additive in eq 2 (to completely
transform the reaction products while leaving the overall rate
essentially unaffected) is incompatible with a one-step mech-
anism for direct hydrogen transfer (eq 1), since it demands the
existence of another intermediate.6 Furthermore, the particular
effect of added salt implies that this intermediate must be an
ionic species. From the known behavior of quinones in their
excited states to readily oxidize electron-rich organic donors to
their cation radicals,10 we identify the ionic intermediate as the
ion pair [Q-•, ArH+•] which is known to be readily intercepted
(22) Kinetic isotope effects on simple electron transfers, without a follow-
D
up proton-transfer step, are small (kETH/kET e 1.3). See: (a) Pal, H.;
Nagasawa, Y.; Tominaga, K.; Yoshihara, K. J. Phys. Chem. 1996, 100,
11964. (b) Doolen, R.; Simon, J. D. J. Am. Chem. Soc. 1994, 116, 1155
(compare the results with biphenyl (entry 6 in Table 1)).

KIE for Electron-Transfer ActiVation of Hydrocarbons
J. Am. Chem. Soc., Vol. 120, No. 12, 1998 2829
by added salt.19
kS
\
[ArH+•, Q-•] + TBA+, PF6- y
z TBA+, Q-• + ArH+•, PF6
-
(3)
We therefore conclude that the critical step to form Q(H)• and
Ar• as the products of hydrogen transfer is to be identified as
proton transfer (kPT) from ArH+• to Q-• within the contact ion
pair in eq 4.20
kET
kPT
Q* + ArH 8 [Q-•, ArH+•] 8 Q(H)• + Ar•
(4)
The pathway in eq 4 might appear to be incompatible with
the observed isotope effects in Table 1, since the initial electron-
transfer step is not expected to show an appreciable isotope
effect. To resolve this paradox, let us consider the cumulative
effect of added salt on the kinetic isotope effect. In the absence
of any salt, Q* reacts appreciably slower with the perdeuterated
methylarenes, and the difference is reflected in the kinetic
isotope effect (3.1 for HMB) which is more or less that for
deprotonation of various ArH+• by pyridine bases.21 Most
Figure 4. Linear correlation of the competition between proton transfer
(kPT) and ion separation (kS) as derived from the salt effect on ArH+•
yields in Figure 3A according to the mechanism in Scheme 1.
Scheme 1
strikingly, the progressive addition of TBA+PF6 causes a
-
kET
kPT
Q(H)• + Ar•
Q* + ArH
Q– •, ArH+•
corresponding diminution in the deuterium isotope effects
(compare the open circles relative to the filled circles in Figure
3B), such that at the (highest) salt concentration of 0.2 M the
isotope effect completely vanishes in Figure 3B! Equally
striking is the concomitant increase in the ion yields of ArH+•
with increasing salt concentration in Figure 3A to establish the
salt-induced competition between ion-pair separation (kS) and
proton transfer (kPT), i.e.
k–ET
ks[salt]
Q– + ArH+•
•
the theoretical intercept of Φion ) 1.0 solidifies this prediction,
and the slope (kPT/kS) leads to the isotope effect for proton
transfer (kH/kD)PT listed in the last column in Table 1.24
Summary and Conclusions
kPT
ArH+• + Q-• 7kS
9
[ArH+•, Q-•] 8 Ar• + Q(H)• (5)
We have unambiguously demonstrated that substantial kinetic
isotope effects are observed for the hydrogen transfer in eq 1,
even when it proceeds via a two-step sequence involving
reversible electron transfer followed by proton transfer according
to Scheme 1. This mechanism is applicable to the oxidative
C-H activation of most hydrocarbons which involve endergonic
electron transfer (∆G g 0), so that reverse electron transfer
(k-ET) is competitive with proton transfer (kPT). Although
analogous electron-transfer formulations are applicable to a wide
variety of other organic reactions,25 their experimental verifica-
tion (as elucidated herein) will be difficult, since the rapidity
of the follow-up steps generally precludes the direct observation
of the ion-radical pair. Indeed, time-resolved spectroscopy,
Such a formulation emphasizes the effect of added salt on the
kinetic isotope effect to parallel its effect on the ion yield, which
is consistently greater for the deuterated analogue (ArD). The
difference, however, diminishes with increasing salt concentra-
tions, as the yield of cation radical approaches unity for both
isotopomers.
Salt effects in eq 5 predict the ion-pair separation in eq 2 to
be complete at sufficiently high salt concentrations, and this
will lead to the loss of the kinetic isotope effect. Thus the
difference in reactivity between ArH and ArD in Table 1 must
stem entirely from the proton-transfer step (kPT) in eq 4.22 Since
the overall decay rate of Q* is affected by isotopic substitution,
the kinetics of this follow-up step must feed back onto the initial
electron-transfer step. For this condition to hold, electron
transfer must be reversible. As such, the inclusion of the salt-
induced separation and reverse electron transfer leads to the
modification of the ET mechanism in eq 4 to that shown in
Scheme 1.
(24) If the ion-radical pair in Scheme 1 is assumed to be present in a
dynamic steady-state concentration, the bimolecular rate constant, k2, is given
by k2 ) kET(kPT + kS[salt])/(k
+ kPT + kS[salt]). Two limiting cases
-ET
arise from this equation. For case (i), kPT . k
+ kS[salt], k2 is simply
-ET
given by kET. No kinetic isotope effect is expected in this case. For case
(ii), k -ET . kPT + kS[salt], k2 is given by (kET/k-ET)(kPT + kS[salt]). In this
case, the kinetic isotope effect (in the absence of added salt) will be identical
to that of the intrinsic proton transfer, i.e., (kH/kD)PT in Table 1. Thus the
measured kinetic isotope effect on k2 will fall between these extremes. (b)
Digital simulation of the complete (complex) kinetics in Scheme 1, using
the method of Weigert,24c leads to the same conclusion. (c) Weigert, F. J.
Comput. Chem. 1987, 11, 273.
In this scheme, the isotope effect on the proton transfer (kPT)
is “diluted” by the competing ion separation (kS) as well as the
reversible electron transfer (k-ET). Thus, the isotope effect
decreases as the free-ion yields increase, and such an inverse
relationship is verified in Figure 3. For the competition in
Scheme 1, the ion yield is readily formulated as Φion ) kS-
[salt]/(kPT + kS[salt]),23 which leads to the linearized correlation
(25) The mechanistic ambiguity addressed in this study applies to other
hydrogen-atom transfer reactions, as well as a wide variety of organic
processes for which an initial electron-transfer process may be formulated.
For H-atom transfers, see: (a) Patz, M.: Fukuzumi, S. J. Phys. Org. Chem.
1997, 10, 129. For other reactions, see: (b) Olah, G. A.; Malhotra, R.;
Narang, S. C. Nitration: Methods and Mechanism; VCH: New York, 1989;
p 166 f (nitration). (c) Baciocchi, E.; Galli, C. J. Phys. Org. Chem. 1995,
8, 563 (halogenation). (d) Tolbert, L. M.; Sun, X. J.; Ashby, E. C. J. Am.
Chem. Soc. 1995, 117, 2681 (nucleophilic substitution). (e) Freilich, S. C.;
Peters, K. S. J. Am. Chem. Soc. 1985, 107, 3819 (cycloaddition). (f) Lopez,
L.; Troisi, L. Tetrahedron Lett. 1989, 30, 3097 (rearrangement of epoxides).
For organometallic examples, see: (g) Astruc, D. Electron Transfer and
Radical Processes in Transition-Metal Chemistry; VCH: New York, 1995.
Φion ) 1 + (kPT/kS) [salt]-1, and this is confirmed in Figure
-1
-1
4 by the linear plot of the reciprocal ion yields (Φion
)
versus the reciprocal salt concentration [salt]-1. Most notably,
(23) The bimolecular formulation of the salt effect in Scheme 1 is likely
to be an oversimplification, since it assumes that added salt reacts as a
simple molecular species and does not take into account the behavior of
the separated ions.

2830 J. Am. Chem. Soc., Vol. 120, No. 12, 1998
Bockman et al.
these plots yielded the bimolecular rate constants (k2) in Table 1. (B)
The kinetics plots of the observed first-order rate constants versus the
ArH concentrations did not remain linear at high [ArH] but approached
a plateau value for concentrations greater than 3.0 × 10-1 M. This
asymptotic behavior of the kinetics plots was diagnostic of a preequi-
librium intermediate, which was previously identified as the encounter
complex [Q*, ArH] of the photoactivated quinone and the polymeth-
ylbenzene.10 Thus the limiting values of kobs at high [ArH] represented
the intrinsic decay (electron-transfer) rate constant of the encounter
complex. Since the encounter complex decayed after its diffusive
formation from Q* and ArH, the limiting decay rate constants
represented electron transfer in the absence of diffusion. The ratios of
these rate constants for the protiated and deuterated polymethylbenzenes
yielded the kinetic isotope effect under these diffusion-free conditions,
and they are listed in Table 1 in parentheses.
combined with the efficacy of added salt, has provided us with
a unique opportunity to intercept and quantify this elusive
intermediate. More typically, the transient ion pair is unobserv-
able, and we are left with a reaction that appears to proceed via
a single (concerted) step.
Experimental Section
Materials. Chloranil, hexamethylbenzene, and durene in Chart 1
were obtained from Aldrich and purified by recrystallization from
ethanol.26 Toluene was distilled from sodium and benzophenone in
an atmosphere of argon. Toluene-d8 (Aldrich), durene-d14 (KOR
Isotopes), hexamethylbenzene-d18 (MSD Isotopes), biphenyl-d10 (Al-
drich), and tetra-n-butylammonium hexafluorophosphate and perchlorate
(Aldrich) were used as received. Dichloromethane (reagent grade) was
stirred over concentrated H2SO4, washed with aqueous NaHCO3, and
distilled serially from P2O5 and from CaH2 under an argon atmosphere.
The synthesis of 2,5-dichloroxyloquinone (CX) was described previ-
ously.10 2,6-Dichloroxyloquinone (CX′) was prepared as follows: 2,6-
dimethyl-1,4-dimethoxybenzene (Aldrich) was chlorinated overnight
with chlorine gas in carbon tetrachloride at room temperature, and 2,6-
dichloro-3,5-dimethyl-1,4-dimethoxybenzene was obtained in 90%
yield. Demethylation with BBr3 in dichloromethane yielded the
corresponding hydroquinone, which was subsequently oxidized with
N2O4 to 2,6-dichloroxyloquinone (CX′) in 86% yield. The quinone
was purified by recrystallization from ethanol.
Determination of the Bimolecular Rate Constants for Hydrogen
Transfer. The laser-flash experiments were carried out using the third
harmonic (355 nm) of a Q-switched Nd:YAG laser (10-ns fwhm) and
a kinetic spectrometer described in detail elsewhere.16a General
Procedure. Upon laser excitation of the quinone (5.0 × 10-3 M) in
dichloromethane, the decay of the triplet quinone was observed at 500
nm in the presence of varying concentrations (10-3-10-1 M) of
polymethylbenzene. The transient decay was fitted to first-order
kinetics, and the observed rate constants (kobs) were plotted against the
arene concentration. (A) The plot was linear for arene concentrations
in the range of 10-3 to approximately 5 × 10-2 M, and the slope of
Determination of the Quantum Yields for Ion-Radical Forma-
tion. The yields of the ion radicals in dichloromethane solutions
containing varying amounts of added salt (TBA+ PF6-) were determined
by using benzophenone in benzene as the transient actinometer.10,17
Thus, samples of the actinometer solution and the quinone/arene
solution in dichloromethane containing 0.02, 0.05, 0.10, and 0.20 M
salt with matching absorbances at 355 nm were excited with the 10-ns
pulse of the Nd:YAG laser. The yields of the ion radicals were
determined by quantitative comparison of the maximum absorbance
of triplet benzophenone (ꢀ530 ) 7220 M-1 cm-1 17 with the maximum
)
absorbance of the cation radical (e.g., HMB+• and DUR+•, at 495 and
480 nm, respectively, both with ꢀ ) 2500 M-1 cm-1).16 The observed
-
salt effect for TBA+ PF6 was the same as that for two other salts,
(TBA+ ClO4-) and [PPN+ PF6-], and values of Φion for the three salts
in the concentration range (i) 0.05, (ii) 0.10, and (iii) 0.20 M are given
serially as follows: (i) 0.52, (0.52), [0.56]; (ii) 0.70 (0.70), [0.75]; (iii)
0.84, (0.85), [0.89] for CX* and HMB in dichloromethane at 23 °C.
Acknowledgment. We thank R. Rathore for providing the
dichloroxyloquinones in Chart 1, and the National Science
Foundation and the R. A. Welch Foundation for financial
support.
(26) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals, 3rd ed.; Pergamon: Oxford, 1988.
JA974205S