Activation energies for the singlet excited state processes of substituted benzenes: Para, meta, and ortho isomers of methylbenzonitrile and methylanisole in acetonitrile

Article abstract of DOI:10.1021/ja048780p

The rate constants of decay of the excited singlet states of the methylbenzonitriles (1-3) and the methylanisoles (4-6) have been determined by the measurement of fluorescence lifetimes over a broad range of temperatures (-45 to +65 °C) in acetonitrile. By fitting this data to a nonlinear expression that includes the Arrhenius equation, rate constants for the activated process (reaction) and the unactivated ones (fluorescence and intersystem crossing) can be reliably obtained. Available literature data for benzene, toluene, and ortho-xylene were also analyzed. The results indicate that the excited singlet state of substituted benzenes is quite reactive and forms a prefulvene biradical intermediate efficiently (quantum yield = 0.69 for benzene itself) by an activated route. In contrast, the efficiency of isolable product formation is quite low because the dominant process for this intermediate is returned to starting material. These observations explain why Ermolaev's rule does not apply to benzene derivatives. Copyright

Full text of DOI:10.1021/ja048780p

Published on Web 06/29/2004
Activation Energies for the Singlet Excited State Processes of Substituted
Benzenes: para, meta, and ortho Isomers of Methylbenzonitrile and
Methylanisole in Acetonitrile
Carlos M. Gonza´lez and James A. Pincock*
Department of Chemistry, Dalhousie UniVersity, Halifax, NoVa Scotia, Canada B3H 4J3
Received March 3, 2004; E-mail: james.pincock@dal.ca
Scheme 1. Reaction Pathways of S1 and RIs Exemplified by the
Prefulvene Biradicals, A, B, and C
The photochemistry and photophysics of organic substrates upon
excitation in fluid solution is controlled, almost without exception,
by the reactivity of the first excited singlet state (S1). As shown in
Scheme 1, the competitive processes and their associated quantum
yields are: kf (Φf), kisc (Φisc), kic (Φic), and kr (Φr). The reaction
process, kr, often results in the formation of a reactive intermediate,
RI, that may return to S0 (kicr) or proceed to observable products
(kp).
Ermolaev’s rule1 for S1 of aromatic compounds, which have been
fundamental in the development of organic photochemistry, states
that (Φf + Φisc) ) 1. This rule applies reasonably well to polycyclic
aromatics: naphthalene (Φf ) 0.19, Φisc ) 0.75),2 anthracene (0.30,
0.71),2 and phenanthrene (0.13, 0.85).2 However, it fails completely
for benzene and many of its derivatives: benzene (0.06, 0.25),2
toluene (0.14, 0.51),2 and ortho-xylene (0.16, 0.58).2 The reason
for this “quantum yield deficiency” is well-established; some
combination of kic and kr contributes to the decay of S1. In fact,
many experimental results for substituted benzenes demonstrate that
RI is a prefulvene biradical formed by meta bonding (A, Scheme
1).3 These results are supported by calculations for benzene itself,
indicating that a barrier exists between S1 and a conical intersection
that leads to A.4 Therefore, eq 1, where Φd is the quantum yield
for the sum of (kr + kic) ) kd, must be used. Nevertheless,
estimations of Φisc for substituted benzenes using Ermolaev’s rule
have recently been reported.5
of substituted benzenes.7 An example is shown in eq 3 for the
methylbenzonitriles. As shown by deuterium labeling and as the
equation indicates, the nitrile-substituted carbon is the active one.
Therefore, the critical RI from the para isomer is the prefulvene
biradical B (Scheme 1), which rearranges to C, providing a pathway
to the meta isomer. In fact, products derived from C have been
obtained from irradiations of both 1 and 2 in 2,2,2-trifluoroethanol.
Moreover, this phototransposition reaction has been shown to occur
from S1. Therefore, these reactions are ideal candidates for a study
of temperature effects. We now report an analysis for literature
results for benzene, toluene, and ortho-xylene8 and for new results
for the methylbenzonitriles (1, 2, 3) and the methylanisoles (4, 5,
6), chosen because they do not undergo the phototransposition
reaction of eq 3. Are the anisole isomers unreactive relative to the
benzonitriles because a larger kinetic barrier separates S1 from the
corresponding substituted prefulvene or do larger barriers exist on
the surface that allow the prefulvenes to rearrange (B to C for the
methylbenzonitriles, Scheme 1)? This question can be answered
by variable temperature studies of fluorescence of the two sets of
substrates which probe only the first possibility and not the second.
Φf + Φisc + Φd ) 1
(1)
The effect of temperature on the rate constants of excited state
processes has received far less attention than it merits. However,
recently Lewis and co-workers have shown that quantum yield and
fluorescence measurements over a broad range of temperatures
provide a powerful tool for understanding the details of excited-
state processes for substituted styrenes.6 The basic idea is that the
total rate constant of decay (kt) of S1 is given by eq 2, with kd )
kr + kic and using the reasonable assumption that the photophysical
processes kf and kisc have no temperature dependence. Therefore,
the Arrhenius parameters A and Ea refer to kd, equal to kr if kr .
kic. This last assumption is reasonable if the S1 to S0 energy gap is
large, leading to unfavorable Franck-Condon overlap and small
values of kic. The reaction process kr can itself result in internal
conversion if the geometric changes that lead to RI do not prevent
return to S0 (kicr). The values of kt can be conveniently obtained
using fluorescence decay as a way of monitoring [S1] as a function
of time.
Rate constants of decay (kt ) 1/τs) for compounds 1-6 were
obtained from the lifetimes (τs) of the exponential decays of
fluorescence (nanosecond single photon counting) over the tem-
perature range of -45 to +65 °C in acetonitrile. A plot of the data
according to eq 2 for 1 is shown in Figure 1, along with a nonlinear
least-squares fit. Plots for 2 and 3 are shown in the Supporting
Information (Figure S1 and S2). Similar plots were made from
literature data for benzene, toluene, and ortho- xylene in hydro-
carbon solvents.8,9 The Arrhenius parameters are given in Table 1,
kt ) kf + kisc + kic + kr ) kf + kisc + kd
) (kf + kisc) + A exp(-Ea/RT)
(2)
Recently, we have published a series of articles on the pho-
totransposition reactions (ortho, meta, para) in acetonitrile of a range
9
8870
J. AM. CHEM. SOC. 2004, 126, 8870-8871
10.1021/ja048780p CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Arrhenius Parameters, Rate Constants (25 °C), and Quantum Yields for Reactions of S1
compd
A/1011 s-1
E /kJ mol-1 ) m3
(k + k )/107 s-1 ) m1
kd/107 s-1
Φd
k/107 s-1
k /107 s-1
Φf
k /107 s-1
Φisc
Φprod
a
f
isc
t
f
isc
benzene
toluene
o-xylene
1
2
3
5.0 ( 3.9
4.3 ( 2.4
23.8 ( 2.3
27.0 ( 1.5
25.4 ( 2.6
26.7 ( 0.8
25.4 ( 1.1
19.5 ( 2.1
0.35 ( 0.51
1.6 ( 0.1
1.5 ( 0.1
4.6 ( 0.1
8.0 ( 0.4
9.4 ( 0.1
3.2
1.0
0.70
2.7
3.3
0.8
0.69
0.34
0.26
0.29
0.28
0.08
3.6
2.9
3.1
9.3
12
0.22
0.41
0.50
1.5
2.6
2.2
0.06
0.14
0.16
0.16
0.22
0.22
0.89
1.5
1.8
0.25
0.52
0.58
(0.55)
(0.40)
(0.65)
0.02
2.1 ( 1.8
0.01
0.025
0.003
13.4 ( 0.4
10.0 ( 0.4
0.21 ( 0.15
10
fluorescence data for the anisoles were quite different from the
toluonitriles as the kt values decreased only slightly with temper-
ature; the averages over the entire temperature range were 14 ( 2,
13 ( 2, and 15 ( 2 × 10-7 s-1, for 4, 5, and 6, respectively. These
observations provide strong circumstantial evidence that if Ea is
large enough to prevent kr from contributing to kt, then kt = kf +
kisc. Therefore, the quantum yield deficiency for substituted benzenes
may only be significant for those that decay efficiently through kr
to RI, i.e., kr . kic and Φd = Φr.
Finally, the quantum yields of product formation Φprod are quite
small for substituted benzenes. Values are given in Table 1 for
benzene (benzvalene formation),13 ortho-xylene (formation of meta-
xylene),14 1 (formation of 2 and 3),7 and 2 (formation of 1 and 3).7
These values are all significantly lower than Φd. Therefore, the
formation of RI, in agreement with calculations,4 is a major pathway
for internal conversion by kicr (Scheme 1). In conclusion, the effect
of temperature on fluorescence is a valuable tool for probing the
reactivity of S1 of substituted benzenes.
Figure 1. Arrhenius plot for the rate constant of decay (kt) of S1 for 1
versus T.
along with values for (kf + kisc), the intercept at 0 K of the plots
and kd, calculated at 25 °C.
For the alkylbenzene derivatives, both Φf and Φisc values have
been determined previously,10 and therefore kt (25 °C) can be
divided into each of its component processes, kf, kisc, and kd, using
eq 1 and kx ) Φxkt. These values are given in Table 1. The values
of (kf + kisc) obtained from these quantum yield measurements (1.1,
1.9, and 2.3 × 10-7 s-1 for benzene, toluene, and ortho-xylene,
respectively) agree quite well with those from the Arrhenius plots
(0.35 ( 0.51, 1.6 ( 0.1, and 1.5 ( 0.1 × 10-7 s-1, respectively)11
especially considering the long extrapolation necessary for the latter.
The kd values can also be calculated at 25 °C from the Arrhenius
parameters and then summed into (kf + kisc) to give kt. These kt
values (3.6, 2.6, and 2.2 × 107 s-1 for benzene, toluene, and ortho-
xylene, respectively) also agree remarkably well with those obtained
from independent fluorescence lifetime measurements (3.6, 2.9, and
3.1 × 107 s-1). This comparison of using either nonlinear fits to
the Arrhenius equation or directly measured quantum yields for
determining rate constants for the reactions of S1 gives us
considerable confidence that the Arrhenius plot method is reliable.
An examination of the rate constants in Table 1 reveals that all
substrates have kt values considerably higher than (kf + kisc) values.
As expected, Ermolaev’s rule does not apply but eq 1 does. In fact,
Φd values, obtained from eq 1 for the alkyl benzenes and from
kd/kt values for compounds 1-3 in Table 1, indicate that, in all
cases, kd ) (kic + kr) is a significant contributor to the decay of S1.
The agreement between [kt/(kf + kisc + kd)] ) 9.3/7.3, 12/11.3,
and 10/10.2 for 1, 2, and 3, respectively, obtained from fluorescence
decay for kt and from the Arrhenius plots for (kf + kisc + kd) is
again excellent.12
Acknowledgment. We thank NSERC of Canada for financial
support and Sepracor Canada Ltd., Windsor, N.S., for the generous
donation of chemicals.
Supporting Information Available: Arrhenius plots for compounds
2 and 3. This material is available free of charge via the Internet at
http://pubs.acs.org.
References
(1) Ermolaev, V. L.; Sveshnikova, E. B. Acta Phys. Pol. 1968, 34, 771.
(2) Handbook of Photochemistry, 2nd ed.; Murov, S. L., Carmichael, I., Hug,
G. L., Eds.; Marcel Dekker: New York, 1993; Table 1, pp 5-53.
(3) Gilbert, A. In CRC Handbook of Organic Photochemistry and Photo-
biology; Horspool, W. M., Song, P.-S., Eds.; CRC Press: New York,
1995; Chapter 18.
(4) Palmer, I. J.; Ragazos, I. N.; Bernardi, F.; Olivucci, M.; Robb, M. A. J.
Am. Chem. Soc. 1993, 115, 673-686.
(5) Nijegorodov, N.; Mabbs, R.; Winkoun, D. P. Spectrochim. Acta, Part A
2003, 59, 595-606.
(6) (a) Lewis, F. D.; Zuo, X. J. Am. Chem. Soc. 2003, 125, 2046-2047. (b)
J. Am. Chem. Soc. 2003, 125, 8806-8813. (c) Lewis, F. D.; Zuo, X.;
Kalgutkar, R. S.; Wagner-Brennan, J. M.; Miranda, M. A.; Front-Sanchis,
E.; Perez-Prieto, J. J. Am. Chem. Soc. 2001, 123, 11883-11889.
(7) (a) MacLeod, P. J.; Pincock, A. L.; Pincock, J. A.; Thompson, K. A. J.
Am. Chem. Soc. 1998, 120, 6443-6450. (b) Foster J.; Pincock, A. L.;
Pincock, J. A.; Rifai, S.; Thompson, K. A. Can. J. Chem. 2000, 78, 1019-
1029.
(8) Cundall, R. B.; Pereire, L. C. J. Chem. Soc., Faraday Trans. 1972, 68,
1152-1163.
(9) The kt and T values were estimated from graphical presentations of the
data in ref 8.
(10) (a) Carroll, F. A.; Quina, F. H. J. Am. Chem. Soc. 1976, 98, 1-6. (b)
Quina, F. H.; Carroll, F. A. J. Am. Chem. Soc. 1976, 98, 6-9.
(11) The larger error in the value from the Arrhenius plot for benzene is a
consequence of the larger scatter in the low temperature points possibly
because cyclohexane as solvent would be frozen; methylcyclohexane (mp
) -126 °C) was used for the toluene and ortho-xylene measurements.
(12) The low value of A for 3 is unusual and perhaps in error as a consequence
of the small change in kt over the temperature range studied (Figure S2).
(13) Wilzbach, K. E.; Harkness, A. L.; Kaplan, L. J. Am. Chem. Soc. 1968,
90, 1116-1118.
The question then arises as to the partitioning of kd between kic
and kr. Literature data for anisole (Φf ) 0.24, Φisc ) 0.64)2 and
fluorobenzene (Φf ) 0.11, Φisc ) 0.80)2 indicate that kic is only a
minor contributor to the decay of S1 for these two chromophores.
Moreover, neither the methylanisoles (4-6) nor the methylfluo-
robenzenes undergo phototransposition reactions.7 As well, the
(14) Anderson, D. J. Phys. Chem. 1970, 74, 1686-1690.
JA048780P
9
J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004 8871