Barrierless electron transfer bond fragmentation reactions

Article abstract of DOI:10.1021/ja030438+

The ultrafast N-O bond fragmentation in a series of N-methoxypyridyl radicals, formed by one-electron reduction of the corresponding N-methoxypyridiniums, has been investigated as potentially barrierless electron-transfer-initiated chemical reactions. A model for the reaction involving the electronic and geometric factors that control the shape of the potential energy surface for the reaction is described. On the basis of this model, molecular structural features appropriate for ultrafast reactivity are proposed. Femtosecond kinetic measurements on these reactions are consistent with a kinetic definition of an essentially barrierless reaction, i.e., that the lifetime of the radical is a few vibrational periods of the fragmenting bond, for the p-methoxy-N-methoxypyridyl radical.

Full text of DOI:10.1021/ja030438+

Published on Web 10/06/2004
Barrierless Electron Transfer Bond Fragmentation Reactions
Edward D. Lorance, Wolfgang H. Kramer, and Ian R. Gould*
Contribution from the Department of Chemistry and Biochemistry, Arizona State UniVersity,
Tempe, Arizona 85287-1604
Received July 17, 2003; E-mail: igould@asu.edu
Abstract: The ultrafast N-O bond fragmentation in a series of N-methoxypyridyl radicals, formed by one-
electron reduction of the corresponding N-methoxypyridiniums, has been investigated as potentially
barrierless electron-transfer-initiated chemical reactions. A model for the reaction involving the electronic
and geometric factors that control the shape of the potential energy surface for the reaction is described.
On the basis of this model, molecular structural features appropriate for ultrafast reactivity are proposed.
Femtosecond kinetic measurements on these reactions are consistent with a kinetic definition of an
essentially barrierless reaction, i.e., that the lifetime of the radical is a few vibrational periods of the
fragmenting bond, for the p-methoxy-N-methoxypyridyl radical.
Scheme 1
Introduction
Bond fragmentation upon one-electron oxidation or reduction
is of both mechanistic1 and technological interest.2 In particular,
reactions of radical anions formed upon one-electron reduction
have been studied as paradigm bond-breaking processes.3,4 Also
of interest in photochemical processes has been the development
of very fast fragmentation reactions,5 the importance of which
is illustrated in Scheme 1. In this example, exothermic electron
transfer (ket) occurs from an excited donor (D) to a positively
charged electron acceptor X-Y+ to form the donor radical
cation, D•+, and a radical X-Y•. Fragmentation of this radical
to form a new radical X• and a neutral molecule Y (kfr) occurs
in competition with return electron transfer (k-et). Clearly, the
larger kfr, the less energy-wasting return electron transfer occurs.
Ideally, the fragmentation reaction, kfr, should be so fast that it
occurs in concert with electron attachment and with no barrier.
An obvious question is how can such concerted and barrierless
reactions be designed?
The dissociative electron-transfer reactions of alkyl and aryl
halide and other radical anions provide the most useful guide.4,6
The usual thermodynamic cycle description of redox-activated
bond cleavage predicts that, for a series of otherwise equal
molecules, fragmentation of a radical anion should become more
energetically favorable with increasingly negative reduction
potential of the parent molecule and thus presumably be faster.7
(1) See, for example: (a) Maslak, P. Top. Curr. Chem. 1993, 168, 1. (b)
Saveant, J.-M. Acc. Chem. Res. 1993, 26, 455. (c) Albini, A.; Fasani, E.;
d’Alessandro, N. Coord. Chem. ReV. 1993, 125, 269. (d) Saveant, J.-M.
AdV. Electron-Transfer Chem. 1994, 4, 53. (e) Gaillard, E. R.; Whitten, D.
G. Acc. Chem. Res. 1996, 29, 292.
(2) (a) Crivello, J. V. AdV. Polym. Sci. 1984, 62, 1. (b) Reiser, A. PhotoreactiVe
Polymers, the Science and Technology of Resists; Wiley: New York, 1989.
(c) Saeva, F. D Top. Curr. Chem. 1990, 156, 59. (d) DeVoe, R. J.; Olofson,
P. M.; Sahyun, M. R. V. AdV. Photochem. 1992, 17, 313. (e) Saeva, F. D.
AdV. Electron-Transfer Chem. 1994, 4, 1. (f) Paczkowski, J.; Neckers, D.
C. In Electron Transfer in Chemistry; Balzani, V., Ed.; Wiley-VCH: New
York, 2001; Vol. 5, p 516. (g) Gould, I. R.; Shukla, D.; Giesen, D.; Farid,
S. HelV. Chim. Acta 2001, 84, 2796.
(3) (a) Parker, V. D. Acta Chem. Scand. B 1981, B35, 595. (b) Hammerich,
O.; Parker, V. D. Acta Chem. Scand. B 1981, B37, 851. (c) Koppang, M.
D.; Woolsey, N. F.; Bartak, D. E. J. Am. Chem. Soc. 1985, 107, 4692. (d)
Maslak, P.; Guthrie, R. D. J. Am. Chem. Soc. 1986, 108, 2628. (e) Maslak,
P.; Narvaez, J. N. Angew. Chem. 1990, 4, 302. (f) Ruhl, J. C.; Evans, D.
H.; Hapiot, P.; Neta, P. J. Am. Chem. Soc. 1991, 113, 5188. (g) Wu, F.;
Guarr, T. F.; Guthrie, R. D. J. Phys. Org. Chem. 1992, 5, 7. (h) Saveant,
J. M. J. Phys. Chem. 1994, 98, 3716. (i) Guthrie, R. D.; Patwardhan, M.;
Chateauneuf, J. E. J. Phys. Org. Chem. 1994, 7, 147. (j) Mathivanan, N.;
Johnston, L. J.; Wayner, D. D. M. J. Phys. Chem. 1995, 99, 8190. (k)
Maslak, P.; Theroff, J. J. Am. Chem. Soc. 1996, 118, 7235. (l) Andrieux,
C. P.; Saveant, J.-M.; Tallec, A.; Tardivel, R.; Tardy, C. J. Am. Chem.
Soc. 1996, 118, 9788. (m) Andrieux, C. P.; Combellas, C.; Kanoufi, F.;
Saveant, J.-M.; Thiebault, A. J. Am. Chem. Soc. 1997, 119, 9527. (n)
Benassi, R.; Bertarini, C.; Taddei, F. J. Chem. Soc., Perkin Trans. 2 1997,
2263. (o) Borsari, M.; Fontanesi, C.; Gavioli, G. Curr. Top. Electrochem.
1997, 5 167. (p) Zheng, Z.-R.; Evans, D. H.; Chan-Shing, E. S.; Lessard,
J. J. Am. Chem. Soc. 1999, 121, 9429. (q) Eberson, L. Acta Chem. Scand.
1999, 53, 751. (r) Beregovaya, I. V.; Shchegoleva, L. N. Chem. Phys. Lett.
2001, 348, 501. (s) Andrieux, C. P.; Farriol, M.; Gallardo, I.; Marquet, J.
J. Chem. Soc., Perkin Trans. 2 2002, 985. (t) Costentin, C.; Robert, M.;
Saveant, J.-M. J. Am. Chem. Soc. 2003, 125, 105.
(4) See, for example: (a) Andrieux, C. P.; Le Gorande, A.; Saveant, J.-M. J.
Am. Chem. Soc. 1992, 114, 6892. (b) Grimshaw, J.; Langan, J. R.; Salmon,
G. A. J. Chem. Soc., Faraday Trans 1994, 90, 75. (c) Andrieux, C. P.;
Robert, M.; Saeva, F. D.; Saveant, J.-M. J. Am. Chem. Soc. 1994, 116,
7864. (d) Saveant, J.-M. AdV. Electron-Transfer Chem. 1994, 4, 53. (e)
Adcock, W.; Andrieux, C. P.; Clark, C. I.; Neudeck, A.; Saveant, J.-M.;
Tardy, C. J. Am. Chem. Soc. 1995, 117, 8285. (f) Andrieux, C. P.; Saveant,
J.-M.; Tallec, A.; Tardivel, R.; Tardy, C. J. Am. Chem. Soc. 1997, 119,
2420. (g) Andrieux, C. P.; Saveant, J.-M.; Tardy, C. J. Am. Chem. Soc.
1997, 119, 11546. (h) Zhong, D.; Zewail, A. H. Proc. Natl. Acad Sci. U.S.A.
1999, 96, 2602. (i) Costentin, C.; Hapiot, P.; Medebielle, M.; Saveant, J.-
M. J. Am. Chem. Soc. 2000, 122, 5623. (j) Pause, L.; Robert, M.; Saveant,
J.-M. J. Am. Chem. Soc. 2000, 122, 9829. (k) Soriano, A.; Silla, E.; Tunon,
I. J. Chem. Phys. 2002, 116, 6102. (l) Antonello, S.; Crisma, M.; Formaggio,
F.; Moretto, A.; Taddei, F.; Toniolo, C.; Maran, F. J. Am. Chem. Soc. 2002,
124, 11503. (m) Cardinale, A.; Isse, A. A.; Gennaro, A.; Robert, M.;
Saveant, J.-M. J. Am. Chem. Soc. 2002, 124, 13533.
9
10.1021/ja030438+ CCC: $27.50 © 2004 American Chemical Society
J. AM. CHEM. SOC. 2004, 126, 14071-14078
14071

A R T I C L E S
Scheme 2
Lorance et al.
energy diagrams for electron transfer from two different donors,
D1 and D2, to a fragmentable acceptor, X-Y+, as a function of
the X-Y separation distance, r(X-Y). The dashed curves
represent the energies of the donor/acceptor pairs before electron
transfer and have minima at the equilibrium bond distance of
the X-Y+, re(X-Y+). D1 is a stronger electron donor than D2,
and thus the dashed curve for this pair is higher in energy. The
energy curve for fragmentation of the X-Y• formed by electron
transfer from the donors shows exothermic fragmentation over
a small barrier. Electron transfer from D1 to X-Y+ to form
X-Y• at its equilibrium geometry, re(X-Y•), is exothermic and
occurs at the crossing point of the upper dashed and solid curves
as indicated. Fragmentation of the intermediate X-Y• subse-
quently occurs over the small barrier, again as indicated.
Electron transfer from D2 to X-Y+ to form X-Y• at its
equilibrium geometry is endothermic. Electron transfer to X-Y+
occurs by stretching the X-Y bond until the lower dashed curve
and the solid curve cross as indicated. At this point the solid
curve is repulsive with respect to the X-Y bond. The overall
reductive cleavage path for D2 is obviously concerted because
it involves only one transition state, but it is also clearly not
barrierless. Indeed, reduction by D2 may be characterized by a
sufficiently large barrier (the crossing point of the lower dashed
and the solid curves) to be essentially useless in a photochemical
process.
Thus, for cleavage to proceed via the smallest energy barrier,
electron transfer using a reductant such as D1 is necessary. This
would also allow the complete fragmentation energy surface
of the reactive species to be explored, since X-Y• is formed at
a geometry in which the energy curve is still in the attractive
region. In this way, the factors controlling the barrier to
fragmentation on the solid energy curve could be studied. The
use of a photochemical reductant D1 would also allow increased
time resolution compared to electrochemical methods. Indeed,
Eberson has previously pointed to the need for experiments with
femtosecond time resolution to address the problem of barri-
erless reactions.10 Such experiments form the basis of the present
work.
This prediction has been confirmed by experiment, especially
in the benzyl halide series.6 Although some time-resolved
measurements have been reported,8 these reactions have mainly
been studied using electrochemical methods.4 The electrochemi-
cal time scale, however, is slow compared to that of the fastest
possible fragmentation reactions, and this has led to some
questions about the nature of very fast dissociative processes,
particularly when results of electrochemical and photochemical
processes are compared.9-12
The principle issues seem to be the definition of a concerted
dissociative reaction and whether this implies that the fragmen-
tation process occurs without any barrier. A kinetic definition
of a truly barrierless reaction might be that the reduced (or
oxidized) species has a lifetime of only one or two vibrational
periods of the fragmenting bond. An alternative thermodynamic
definition would be that this species has no potential energy
minimum with respect to the fragmentation coordinate. In a
series of detailed investigations of such processes, however,
Saveant has clearly shown that the thermodynamics of the initial
electron-transfer process has to be considered when asking
questions such as these.13 Consider Scheme 2, which shows
(5) (a) Maslak, P.; Kula, J.; Chateauneuf, J. E. J. Am. Chem. Soc. 1991, 113,
2304. (b) Zou, C.; Miers, J. B.; Ballew, R. M.; Dlott, D. D.; Schuster, G.
B. J. Am. Chem. Soc. 1991, 113, 7823. (c) Perrier, S.; Sankararaman, S.;
Kochi, J. K. J. Chem. Soc., Perkin Trans. 2 1993, 825. (d) Lucia, L. A.;
Wang, Y.; Nafisi, K.; Netzel, T. L.; Schanze, K. S. J. Phys. Chem. 1995,
99, 11801. (e) Maslak, P.; Narvaez, J. N.; Vallombroso, T. M., Jr. J. Am.
Chem. Soc. 1995, 117, 12373. (f) Hassoon, S.; Sarker, A.; Polykarpov, A.
Y.; Rodgers, M. A. J.; Neckers, D. C. J. Phys. Chem. 1996, 100, 12386.
(g) Bockman, T. M.; Hubig, S. M.; Kochi, J. K. J. Org. Chem. 1997, 62,
2110. (h) Bockman, T. M.; Hubig, S. M.; Kochi, J. K. J. Am. Chem. Soc.
1998, 120, 6542. (i) Rasmusson, M.; Akesson, E.; Eberson, L.; Sundstrom,
V. J. Phys. Chem. 2001, 105, 2027. (j) Nath, S.; Singh, A. K.; Palit, D. K.;
Sapre, A. V.; Mittal, J. P. J. Phys. Chem. A 2001, 105, 7151.
(6) For a summary, see ref 4d.
We recently reported on rate constants for N-O bond
fragmentation in a series of N-methoxyheterocycle radicals
formed by one-electron reduction of the corresponding N-
methoxy compounds (corresponding to X-Y+), as illustrated
in eq 1 for the case of the parent pyridinium, 1.14,15 One-electron
(7) (a) Popielarz, R.; Arnold, D. R. J. Am. Chem. Soc. 1990, 112, 3068. (b)
Maslak, P.; Vallombroso, T. M.; Chapman, W. H., Jr.; Narvaez, J. N.
Angew. Chem., Int. Ed. Engl. 1994, 33, 73. (c) Wayner, D. D. M.; Parker,
V. D. Acc. Chem. Res. 1993, 26, 287.
(8) (a) Neta, P.; Behar, D. J. Am. Chem. Soc. 1980, 102, 4798. (b) Neta, P.;
Behar, D. J. Am. Chem. Soc. 1981, 103, 103. (c) Behar, D.; Neta, P. J.
Am. Chem. Soc. 1981, 103, 2280. (d) Norris, R. K.; Barker, S. D.; Neta, P.
J. Am. Chem. Soc. 1984, 106, 3140. (e) Kimura, N.; Takamuku, S. Bull.
Chem. Soc. Jpn. 1986, 59, 3653. (f) Kimura, N.; Takamuku, J. Am. Chem.
Soc. 1995, 117, 8023.
(9) (a) Shaik, S.; Danovich, D.; Sastry, G. N.; Ayala, P. Y.; Schlegel, H. B. J.
Am. Chem. Soc. 1997, 119, 9237. (b) Robert, M.; Saveant, J.-M. J. Am.
Chem. Soc. 2000, 122, 514.
(10) Eberson, L. Acta Chem. Scand. 1999, 53, 751.
(11) Wang, X.; Saeva, F. D.; Kampmeier, J. A. J. Am. Chem. Soc. 1999, 121,
4364.
(12) Pause, L.; Robert, M.; Saveant, J.-M. ChemPhysChem 2000, 1, 199.
(13) These ideas are discussed in many papers from the Saveant group, including
refs 4c,d,g,j,m and (a) Andrieux, C. P.; Merz, A.; Saveant J.-M. J. Am.
Chem. Soc. 1985, 107, 6097. (b) Constentin, C.; Hapiot, P.; Medebielle,
M.; Saveant, J.-M. J. Am. Chem. Soc. 1999, 121, 4451, (c) Pause, L.; Robert,
M.; Saveant, J.-M. J. Am. Chem. Soc. 1999, 121, 7158. (d) Constentin, C.;
Robert, M.; Saveant, J.-M. J. Phys. Chem. 2000, 104, 7492.
reduction of 1 forms the pyridyl radical 1R (corresponding to
X-Y•), which undergoes fragmentation to form the neutral
pyridine and a methoxy radical. A wide range of rate constants
was determined for these reactions, depending upon the
particular N-methoxyheterocycle,14 and we wondered if it would
be possible to rationally increase the rate constant for fragmen-
tation toward the barrierless limit. Here we describe further
(14) Lorance, E. D.; Kramer, W. H.; Gould, I. R. J. Am. Chem. Soc. 2002, 124,
15225.
(15) See also similar work by Kochi et al.: (a) Lee, K. Y.; Kochi, J, K. J.
Chem. Soc., Perkin Trans. 2 1992, 1011. (b) Bockman, T. M.; Lee, K. Y.;
Kochi, J, K. J. Chem. Soc., Perkin Trans. 2 1992, 1581.
9
14072 J. AM. CHEM. SOC. VOL. 126, NO. 43, 2004

Barrierless Electron Transfer Bond Fragmentation
A R T I C L E S
Figure 1. Correlation diagrams for cleavage of (A) unsubstituted N-methoxy pyridyl radical and (B) a corresponding radical substituted with a strong
electron-donating group, D. The dots represent electrons not involved in pairwise bonding. Three reactant (r1-r3) and product configurations (p1 and a
doubly degenerate p2) are important; see text for their description. The dotted lines represent diabatic correlations in the planar radical, and the curves
represent adiabatic potential energy surfaces formed upon configuration mixing in the nonplanar radicals in N-methoxyheterocycle radicals.
Scheme 3
studies on the N-O bond fragmentation reaction. A more
detailed theoretical analysis of the N-O bond fragmentation
process than given previously is discussed,14 and femtosecond
kinetic methods for the study of fragmentation reactions of this
type are described.
be barrierless. Consider the situation in Figure 1B. An electron-
donating group affects mainly the energy of the π* orbital,
raising the energies of those configurations in which the π*
orbital is occupied, i.e., r1, r2, and p2. As indicated, this has the
effect of increasing the reaction exothermicity and decreasing
but not eliminating the energy of the diabatic crossing. The
barrier only disappears in the unlikely situation that the energy
of p2 becomes lower than r1.
Theoretical Model for Bond Fragmentation
A qualitative state energy diagram of N-O bond cleavage
for reduction of the parent N-methoxypyridinium compound is
shown in Figure 1A. Addition of an electron results in the
formation of the pyridyl radical (r1, Figure 1A). The electron is
accepted into the lowest energy orbital, in this case a π* orbital
of the pyridyl system. However, increasing the N-O bond
distance diabatically in a planar radical with this electronic
configuration simply raises the energy of the electrons associated
with the N-O σ bond and leads to the methoxy radical and an
n-π* excited state of pyridine (p2), rather than the ground-state
pyridine and methoxy radical products (p1). As discussed
previously,14 certain electronic configurations of the radical
involving population of the N-O σ* orbital decrease in energy
with increasing N-O bond distance. One obvious excited-state
contributor has one electron in the π* and two unpaired
electrons, one in each of the σ and σ* orbitals (r2). Increasing
the N-O bond distance in this planar radical produces the same
pyridine n-π* excited state at infinite N-O separation as the
ground state. Another, in which the N-O σ orbital has one
electron and the σ* orbital is doubly occupied (r3), is also
repulsive with respect to the N-O coordinate and when planar
can form the ground-state products.
Mixing of either the reactant excited configuration r3 with
the reactant ground-state r1, or mixing the excited configurations
r2 and r3 with each other, is symmetry forbidden when the radical
is planar. Mixing as a function of N-O bond bending, however,
generates adiabatic electronic states (solid lines in Figure 1).16
This feature is common to many bond-breaking reactions in
which the ring symmetry plane must be broken to allow the
mixing of the otherwise orthogonal σ and π systems.17-19 Of
particular interest in this regard is the recent work by Hynes et
al., who described fragmentation of the radical anion of
p-cyanophenyl chloride in terms of stretching and out-of-plane
bending of the C-Cl bond.18
Bending the N-O bond out of the plane of the ring is thus
an important part of the reaction, Scheme 3. Indeed, we
previously showed that all of the N-methoxyheterocycle radicals
are bent at the local radical minimum, with increased bending
(16) (a) Salem, L. Electrons in Chemical Reactions; Wiley-Interscience: New
York, 1982. (b) Michl, J.; Bonacic-Koutecky, V. Electronic Aspects of
Organic Photochemistry; Wiley-Interscience: New York, 1990. (c) Shaik,
S. S.; Schlegel, H. B.; Wolfe, S. Theoretical Aspects of Physical Organic
Chemistry; Wiley: New York, 1992. (d) Pross, A. Theoretical and Physical
Principles of Organic Chemistry; Wiley: New York, 1995.
(17) (a) Pearson, R. G. Symmetry Rules for Chemical Reactions; Wiley: New
York, 1976; Chapter 5. (b) Reference 16b, Chapter 3.
Thus, a barrier is predicted for fragmentation of the initially
formed π* pyridyl radical, corresponding to the energy required
to attain the crossing point of the dotted lines originating with
the r1 and r3 configurations in the correlation diagram. According
to these diabatic correlations, however, the reaction will never
(18) (a) Laage, D.; Burghardt, I.; Sommerfeld, T.; Hynes, J. T. ChemPhysChem
2003, 4, 61. (b) Laage, D.; Burghardt, I.; Sommerfeld, T.; Hynes, J. T. J.
Phys. Chem. A 2003, 107, 11271. (c) Burghardt, I.; Laage, D.; Hynes, James
T. J. Phys. Chem. A 2003, 107, 11292.
(19) Note that in a nonplanar conformation and in the absence of mixing, the r2
state now correlates with the p1.
9
J. AM. CHEM. SOC. VOL. 126, NO. 43, 2004 14073

A R T I C L E S
Scheme 4
Lorance et al.
Finally, the orbital and state considerations clearly point to
electron-donating groups in the p-position to decrease the
electronic barrier to reaction, and so 7 and 8 were included in
the present work to explore these effects. The N-ethyl com-
pounds 9 and 10 were also required to model return electron
transfer in the geminate pairs.
Kinetic Measurements
The goal is to identify bond fragmentation reactions that may
occur at the fastest possible rates, and indeed, the fragmentation
rate constants for the radicals studied here are all greater than
1010 s-1, and some are greater that 1012 s-1. Direct measurement
of the rate constants requires the radicals to be formed faster
than they react. Previously, excitation of charge-transfer (CT)
complexes has been used to measure the rate constants for
fast fragmentation reactions of this kind on the picosecond
and, in a few cases, subpicosecond time scale.5,14,15 Several
experimental issues related to these kinds of measurements,
particularly the effect of solvent relaxation, have not been
discussed previously, and so the method is described in some
detail here.
being associated with increased reaction rate constant.14 An
important additional consequence for the correlations of Figure
1B is that when mixing is extensive and the reaction is
sufficiently exothermic, the splitting of the adiabats can
potentially eliminate the barrier to reaction. This indicates that
some combination of large reaction exothermicity and extensive
configuration mixing by bending could be used to design a truly
barrierless reaction.
This description of the reaction provides a more detailed
understanding in terms of the important orbital energies than a
simple thermodynamic cycle approach7 and allows more subtle
substituent effects to be predicted. In addition to lowering the
energy crossing point of the diabats, an electron-donating
substituent in the p-position of the pyridyl should result in
increased σ* character as a consequence of state mixing, a lower
activation energy, earlier transition state, and faster rate. An
electron-withdrawing substituent on the pyridyl ring will lower
the energy of the r1, r2, and p2, decreasing the σ* character,
resulting in less bending and a decreased rate. A delocalizing
substituent such as an aromatic group will have the effect of
lowering the energy of the π* orbital, decreasing σ* character,
and reducing the reaction rate compared to the simple pyridyl
system.
The N-methoxypyridinium compounds of Scheme 4 are
reasonably strong electron acceptors. Addition of a suitable one-
electron donor, D, to a solution of the N-alkoxypyridinium
results in the formation of a CT complex (see Supporting
Information for an example). Excitation into the CT absorption
band of this complex using a pulsed laser forms a donor radical
cation/pyridyl radical pair within the time period of the
excitation pulse. Subsequent reactions are those of the (D•+
X-Y•) geminate pair, Scheme 1.14
The important processes usually considered for geminate pairs
of this type are return electron transfer to reform the ground-
state CT complex, k-et, and fragmentation of the radical,
kfr.14,15,20 Another process of geminate pairs, separation to form
freely diffusing radicals in solution, is too slow for the particular
systems studied here to appreciably compete with the other
two.20
The electronic absorptions in the visible region of the pyridyl
radicals studied here are too weak to be observed in the presence
of strong absorptions of the donor radical cation, D•+.14 The
kinetics of the pyridyl radicals are thus determined by monitor-
ing D•+. According to Scheme 1, the D•+ signal decays
exponentially to a permanent value (on the ultrafast time scale)
as a result of the removal of D•+ by the return electron-transfer
reaction, k-et. The permanent signal is a consequence of those
pyridyl radicals that fragment within the geminate pair, thus
avoiding return electron transfer. The observed rate constant
for decay of D•+ is given by eq 2a. Here, A0 is the absorbance
of D•+ at time zero, and Ainf is the absorbance of the D•+ that
does not decay on the ultrafast time scale. A quantum yield,
Structures and Experimental Approach
Consistent with this orbital and state picture of the reaction,
the previous work on the substituted pyridyl radicals clearly
indicated a decrease in rate constant for fragmentation with
increasing delocalization and, more importantly, the use of
electron-withdrawing groups.14 Illustrative examples are the
p-phenypyridinium and p-cyanopyridinium compounds, 2 and
3, Scheme 4. The p-withdrawing group in 3R limits the reaction
rate as a result of the withdrawing effect. Withdrawing substit-
uents in the m-position are expected to have a smaller negative
electronic effect on the reaction barrier, and so the m-substituted
4 was chosen for this study in which faster fragmentation rates
were sought. The p-phenyl group limits the reaction rate as a
result of the delocalization effect, and so the m- and o-substituted
compounds 5 and 6 were selected for study. A phenyl group in
the o-position in 6 is expected to be less delocalizing than a
p-phenyl substituent as a result of steric interaction with the
N-methoxy group, which should cause the ring to be twisted
out of conjugation with the pyridyl ring.
Φ
ions, for formation of long-lived D•+ can be defined as in eq
2b. Values for the elementary rate constants k-et and kfr are
(20) (a) Gould, I. R.; Young, R. H.; Farid, S. J. Phys. Chem. 1991, 95, 2068.
(b) Gould, I. R.; Young, R. H.; Farid, S. In Photochemical Processes in
Organized Molecular Systems; Honda, K., Ed.; Elsevier: New York, 1991.
(c) Gould, I. R.; Mueller, J. L.; Farid, S. Z. Phys. Chem. (Munich) 1991,
170, 143. (d) Gould, I. R.; Noukakis, D.; Gomez-Jahn, L.; Young, R. H.;
Goodman, J. L.; Farid, S. Chem. Phys. 1993, 176, 439. (e) Arnold, B. R.;
Noukakis, D.; Farid, S.; Goodman, J. L.; Gould, I. R. J. Am. Chem. Soc.
1995, 117, 4399. (f) Gould, I. R.; Farid, S. Acc. Chem. Res. 1996, 29, 522.
9
14074 J. AM. CHEM. SOC. VOL. 126, NO. 43, 2004

Barrierless Electron Transfer Bond Fragmentation
A R T I C L E S
related to the observed rate constant for removal of D•+, kobs
,
eq 2c, and are obtained as indicated in eqs 2d and 2e.
A(t) ) (A0 - Ainf) e-k t + Ainf
Φions ) Ainf/A0
(2a)
obs
(2b)
(2c)
kobs ) k-et + kfr
kfr ) kobsΦions
(2d)
(2e)
k-et ) kobs - kfr
Excitation of the CT complexes was performed using a
femtosecond pump-probe laser system.21 According to eqs 2,
the rate constant for fragmentation can only be determined with
any accuracy if it competes appreciably with return electron
transfer. If kfr is much smaller than k-et, then the yield of long-
lived D•+ is small, and kfr is a small fraction of the observed
rate constant for geminate pair decay, kobs. If kfr is much larger
than k-et, then only a small decay as a function of time is
observed, and kobs cannot be determined accurately.5h,14,15
Because the fragmentation rate constants measured here were
so large, very fast return electron transfer was necessary to
compete appropriately. Return electron transfer occurs in the
Marcus inverted region for radical ion pairs such as these,20,22
which means that the fastest reactions occur when the exother-
micity of the return electron-transfer reaction is as small as
possible. This requires the use of an electron donor D with as
low an oxidation potential as possible. We have used p-
trianisylamine (TAA) as the electron donor. The oxidation
Figure 2. Absorption spectra as a function of time for excitation of the
charge-transfer complex formed between p-trianisylamine (TAA) and
p-phenyl-N-methoxypyridyl tetrafluoroborate (2) in acetonitrile at room
temperature, showing both decay and temporal evolution of the spectra from
an absorbance maximum at ca. 740 nm to a maximum at ca. 715 nm.
potential of TAA is 0.52 V vs SCE.22 Amines with lower
oxidation potentials than this are subject to atmospheric oxida-
tion and are inconvenient to use. The TAA•+ has an absorption
maximum at 715 nm in acetonitrile, with an extinction coef-
ficient at this wavelength of 45,000.22 It thus represents an ideal
choice for an electron donor.
Formation of CT complexes was readily observed between
all of the pyridinium salts of Scheme 4 and TAA as the donor.
Femtosecond excitation into the CT absorption bands at 410
nm resulted in transient absorptions in the region expected for
TAA•+. Typical data are illustrated in Figures 2 and 3 for the
p-phenylpyridyl radical, 2R. After the pulse, absorption decay
due to return electron transfer was observed to a constant level,
as expected, except that the observed time-dependent absorption
decays were different at different wavelengths. This was due
to a change in the shape of the absorption band with time. The
effect is illustrated in Figure 2. At early times an absorption
maximum around 740 nm was observed, which had shifted to
ca. 715 nm by the end of the decay time period. The latter
wavelength is very close to that observed previously for TAA•+
in acetonitrile.22 The spectral shifts are assigned to time-
Figure 3. (O) Normalized absorbance of p-trianisylamine radical cation
(TAA•+), integrated over the wavelength range 659-797 nm (see text)
versus time for pulsed laser excitation at 410 nm of the charge-transfer
complex formed between TAA and p-phenyl-N-methoxypyridinium tet-
rafluoroborate (2) in acetonitrile at room temperature. The solid curve
represents a fit to the data as described in the text and Experimental Section,
using values for kobs and Φions of 1.8 × 1012 s-1 and 0.15, respectively, and
a Gaussian instrument response function of 600 fs width. The values of ket
and kfr obtained from these data (eqs 2) are 1.5 × 1012 and 2.7 × 1011 s-1
,
respectively.
dependent solvent relaxation around the geminate pair.23 In small
nitrile solvents these should occur roughly on a 1-2 ps time
scale, as observed.23,24
The time dependence of the concentration of the TAA•+ was
determined by first correcting for the wavelength chirp of the
(23) (a) Johnson, A. E.; Levinger, N. E.; Kliner, D. A. V.; Tominaga, K.; Barbara,
P. F. Pure Appl. Chem. 1992, 64, 1219. (b) Zeglinski, D. M.; Waldeck, D.
H. J. Phys. Chem. 1988, 92, 692. (c) Maroncelli, M. J. Mol. Liq. 1993, 57,
1. (d) Fleming, G. R.; Cho, M. Annu. ReV. Phys. Chem. 1996, 47, 109. (e)
Chang, Y. J.; Simon, J. D. Springer Ser. Chem. Phys. 1996, 62, 253. (f)
Cramer, C. J.; Truhlar, D. G. Chem. ReV. 1999, 99, 2161.
(24) The change in solvation state during the time period of the processes being
monitored should result in a time-dependent return electron-transfer rate
constant and thus nonexponential kinetics. However, the change in the
electron-transfer rate constant must be small, since we are able to fit the
data well using the exponential eq 2a (see, for example, Figures 3 and 4).
We assume that we are, in fact, measuring an averaged k-et that has a
small distribution in values.
(21) Sakomura, M.; Lin, S.; Moore, T. A.; Moore, A. L.; Gust, D.; Fujihira, M.
J. Phys. Chem. A 2002, 106, 2218.
(22) Gould, I. R.; Ege, D.; Moser, J. E.; Farid, S. J. Am. Chem. Soc. 1990, 112,
4290.
9
J. AM. CHEM. SOC. VOL. 126, NO. 43, 2004 14075

A R T I C L E S
Lorance et al.
Table 1. Kinetic Parameters for Geminate Pairs of N-Methoxypyridyl Radicals and Radical Cations of p-Trianisylamine in Acetonitrile at
Room Temperature
a No decay of geminate pair observed; see text for discussion of these values. b Previously published data.14 c Not measured with TAA.14
white light probe beam21,25 and then integrating the absorbance
band at the different time delays. A plot of this wavelength-
corrected integrated absorbance versus time is shown in Figure
3, together with a fit to the data according to eq 2a, after
convolution with a Gaussian instrument response function.
Corresponding experiments were performed for the radicals 1R
and 4R-8R, with the experimental data for 5R and 6R
illustrated in Figure 4. Importantly, excitation of a CT complex
of TAA with the p-methoxy and m-methoxy compounds 7 and
8 resulted in no obserVable return electron transfer, Figure 4
and Table 1. The kinetic parameters obtained by analyzing the
time-resolved kinetic data according to eqs 2 for 1-8 are
summarized in Table 1.
Mechanistic Implications
Figure 4. Normalized absorbance of p-trianisylamine radical cation,
Our goal is the development of ultrafast fragmentation
integrated over the wavelength range 659-797 nm (see text), versus time
reactions that compete efficiently with return electron transfer.
The rate constants for return electron transfer in the present cases
are all very large, ca. 1012 s-1 (Table 1). The more efficiently
the fragmentation reaction competes with return electron
transfer, the higher the yield of long-lived TAA•+, e.g., Figures
3 and 4. Note that the raw data shown in the figures are
somewhat deceptive, since convolution effects make the ap-
parent yields of long-lived TAA•+ absorption higher than the
actual yields (Table 1). For all of the reactions studied here,
for pulsed laser excitation at 410 nm of the charge-transfer complexes
formed between p-trianisylamine (TAA) and p-methoxy-N-methoxypyri-
dinium tetrafluoroborate (7, b), m-phenyl-N-methoxypyridinium tetrafluo-
roborate (5, (), and o-phenyl-N-methoxypyridinium tetrafluoroborate (6,
O) in acetonitrile at room temperature. The solid curves represents a fit to
the data for 5 and 6 to eq 2a, as described in the text and Experimental
Section, using values for the parameters given in Table 1. No fit is given
for 7 because no decay is observed in this case.
however, fragmentation was sufficiently rapid that efficient
formation of separated radicals occurs even with ultrafast return
electron transfer.
(25) Yamaguchi, S.; Hamaguchi, H.-O. Appl. Spectrosc. 1995, 49, 1513.
9
14076 J. AM. CHEM. SOC. VOL. 126, NO. 43, 2004

Barrierless Electron Transfer Bond Fragmentation
A R T I C L E S
The unsubstituted pyridyl radical 1R turns out to be one of
the fastest fragmenting radicals. As discussed above, delocal-
ization of the spin in 2R decreases the reaction rate. The effect
of moving the phenyl from the para to the meta position (5R)
is to increase the rate constant for fragmentation compared to
2R by a factor of ca. 3. The phenyl group in the ortho position
(6R) results in even faster fragmentation. In this case, steric
interactions with the N-methoxy group cause the phenyl group
to be sufficiently twisted that essentially no delocalization
occurs.
definition, a barrierless fragmentation reaction has no real
vibration frequency along the bond-breaking coordinate. A
maximum value can be obtained from the N-O bond stretching
frequency of the parent pyridinium, for which a value of 390
cm-1 can be estimated,26 corresponding to a time constant of
ca. 100 fs. The N-O bond must be weaker in the radical,
corresponding to a vibrational time constant greater than 100
fs. Thus, the lifetime of the radicals 7R and 8R are determined
to be a maximum of four vibrational periods and probably less.
Although it is not possible to definitively say whether the
reactions are barrierless or not, they are certainly very close,
presumably as a consequence of the exothermicity and state
splitting effects discussed above.
The p-cyanopyridinium is the most useful radical precursor
in a practical sense because the strong withdrawing group allows
this compound to be reduced by the widest range of excited-
state sensitizers. The para withdrawing group, however, results
in the slowest reaction included here (3R, Table 1). We
wondered if the strong withdrawing group could be retained in
the structure, without the negative effect on the fragmentation
rate constant. As with the m-phenyl radical above, the m-cyano
radical 4R reacts almost as fast as the unsubstituted parent
radical.
Summary
N-O bond fragmentation in N-methoxypyridyl radicals can
occur on the subpicosecond time scale. The reaction is slowed
by electron-withdrawing and electron-delocalizing groups. The
way in which these substituents influence the fragmentation rate
can be manipulated with steric effects and electronic effects and
can be understood in terms of the important orbital and state
correlations. Electron-donating groups increase the reaction rate
close to the time scale for vibration of the fragmenting bond,
and fragmentation of the p-methoxy-substituted radical may be
barrierless.
The interesting observations are for the two methoxy-
substituted radicals, 7R and 8R. The complete lack of return
electron transfer observed for these radicals implies either
extremely fast fragmentation, slow return electron transfer, or
some combination of the two. In either case, fragmentation of
the N-O bond occurs with no detectable energy-wasting return
electron transfer, i.e., essentially 100% efficiency, but this
observation alone is insufficient to determine whether the
fragmentation reaction is barrierless. Furthermore, in the absence
of observed return electron transfer, it is not possible to
determine an absolute value for the rate constant for fragmenta-
tion. However, some reasonable limits can be set.
Experimental Section
Materials. Acetonitrile was spectrograde (Omnisolv) and used as
received. The N-methoxypyridinium compounds 2 and 3 were available
from previous studies.14 General synthetic procedures for preparation
of the N-methoxy and N-ethyl pyridiniums have been given previously.14
All of the pyridinium compounds were prepared and used as the
tetrafluoroborate salts. p-Trianisylamine was a gift from D. Weiss
(Eastman Kodak Company) and was used as received.
Two model compounds were studied, i.e., the corresponding
N-ethyl-substituted compounds 9 and 10, to obtain further
information on the return electron-transfer process. Excitation
of CT complexes of these acceptors with TAA results in
immediate formation of a TAA•+/pyridyl radical geminate pair
as before. In this case, however, no bond fragmentation occurs,
and only return electron transfer is observed.14 The return
electron-transfer rate constants obtained for the radicals 9R and
10R with TAA•+ are 1.2 × 1011 and 5.2 × 1011 s-1, respectively.
Simulations of the observed time-resolved absorptions for 7R
and 8R were performed using the return electron-transfer rate
constants for 9R and 10R, respectively, and various fragmenta-
tion rate constants. The best fits to the data are obtained using
values for kfr equal to or greater than 2.5 × 1012 s-1 (details
are given in Supporting Material). Furthermore, we previously
found that return electron transfer in geminate pairs of the
N-methoxy radicals was actually faster than in the corresponding
N-ethyl radicals, presumably because the N-methoxy-substituted
pyridiniums are easier to reduce than the N-ethylpyridiniums.14
The return electron-transfer rate constants for 9R and 10R above
thus represent minimum values for corresponding return electron
transfer in the 7R and 8R pairs. If return electron transfer in
the 7R and 8R pairs is faster than in the 9R and 10R pairs,
then the fragmentation rate constants are even higher than 2.5
N-Methoxypyridinium Tetrafluoroborate (1). Following the gen-
eral procedure,14 trimethyloxonium tetrafluoroborate (3.48 g, 23.6
mmol) and pyridine N-oxide hydrate (2.00 g, 21.0 mmol) were stirred
in 80 mL of dichloromethane for 4 h; 2.80 g (14.2 mmol, 68%) 1 was
1
obtained as a colorless oil. H NMR (300 MHz, CD3CN): δ (ppm)
4.42 (s, 3 H), 8.14 (t, 2 H, J ) 7.2 Hz), 8.57 (t, 1 H, J ) 7.5 Hz), 9.01
(dd, 2 H, J ) 1.2, 7.5 Hz). 13C NMR (75 MHz, CD3CN): δ (ppm)
70.2, 130.0, 141.0, 145.7. MS m/z: 110.0 (M+).
N-Methoxy-3-cyanopyridinium Tetrafluoroborate (4). Following
the general procedure, trimethyloxonium tetrafluoroborate (0.72 g, 4.87
mmol) and 4-cyanopyridine N-oxide (0.56 g, 4.63 mmol) were stirred
in 30 mL of dichloromethane for 4 h. Recrystallization from methanol
afforded 0.81 g (3.65 mmol, 79%) of 4. 1H NMR (300 MHz, CD3CN):
δ (ppm) 4.53 (s, 3 H), 8.35 (dd, 1 H, J ) 6.9, 8.1 Hz), 8.94 (dd, 1 H,
J ) 0.9, 6.6 Hz), 9.47 (d, 1 H, J ) 5.9 Hz), 9.91 (s, 1 H). 13C NMR
(75 MHz, CD3CN): δ (ppm) 69.2, 114.5, 116.6, 129.4, 143.7, 144.4,
147.4. MS m/z: 135.1 (M+).
N-Methoxy-3-phenylpyridinium Tetrafluoroborate (5). Following
the general procedure,14 trimethyloxonium tetrafluoroborate (2.19 g,
14.8 mmol) and 3-phenylpyridine N-oxide (2.20 g, 12.8 mmol) were
stirred in 40 mL of dichloromethane for 4 h. Recrystallization from
methanol/dichloromethane afforded 3.38 g (12.4 mmol, 97%) of 5. 1H
NMR (300 MHz, CD3OH): δ (ppm) 3.14 (s, 3 H), 6.18 (d, 2 H, J )
1.8 Hz), 6.20 (d, 1 H, J ) 2.1 Hz), 6.41-6.45 (m, 2 H), 6.80 (dd, 1 H,
J ) 6.6, 7.8 Hz), 7.40 (d, 1 H, J ) 8.4 Hz), 7.72 (dt, 1 H, J ) 0.9, 6.6
Hz), 8.08 (t, 1 H, J ) 2.4 Hz). 13C NMR (75 MHz, CD3OH): δ (ppm)
× 1012 s-1
.
Thus, we estimate the lifetime for the radicals 7R and 8R to
be probably less than 400 fs. The vibration frequency of the
N-O bond in the reacting radicals is not known, and by
(26) This corresponds to the frequency of the normal mode that most ap-
proximates the N-O stretch from a B3PW91/6-31+G* computation on 7.
Details of these computations are given in: Lorance, E. D.; Hendrickson,
K.; Gould, I. R. J. Org. Chem. Submitted for publication.
9
J. AM. CHEM. SOC. VOL. 126, NO. 43, 2004 14077

A R T I C L E S
Lorance et al.
70.4, 118.2, 128.3, 130.0, 130.4, 131.4, 133.4, 139.3, 143.3, 143.7.
MS m/z: 186.1 (M+).
to give an optical density close to 1.0 at the laser excitation wavelength.
Typically the amine concentration was about 0.06 M, and the
methoxypyridinium compounds were about 0.3 M. The laser repetition
rate was 10 kHz, and to avoid buildup of too high a concentration of
the TAA radical cation (which would excessively attenuate the probe
beam), ca. 0.05 M triethylamine was added to reduce this species back
to the neutral TAA between pulses. Experiments performed in the
presence and absence of triethylamine showed no difference in the
kinetics of the geminate pairs (except that the solutions had to be
replaced more frequently in the absence of the TEA). The presence of
the TEA also did not affect the absorption of the CT complexes formed
between the N-methoxypyridinium compounds and the TAA. The time-
resolved absorbance spectral data was corrected for chirp in the probe
beam using the method described previously.21 The corrected absor-
bance spectra were integrated over the range 659-797 nm. The resulting
data were fitted to eq 2a after convolution of the equation with a
Gaussian instrument response function. The width of the Gaussian was
a fitting factor and was typically ca. 600 fs. In some cases a small
(less than 10%) additional exponential component was required to obtain
a good fit to the data, with an ill-defined time constant from 10 to 100
ps. Although not accurately determined, this long time constant is clearly
not related to the dynamics of the geminate amine radical cation/pyridyl
radical pairs, which are much shorter lived, and is assigned to decay
of a small amount of amine radical cation formed by two-photon
ionization of uncomplexed amine, which presumably decays by reaction
with solvated electrons. Experiments conducted with only trianisylamine
and with trianisylamine and triethylamine gave similar spectra and decay
times.
N-Methoxy-2-phenylpyridinium Tetrafluoroborate (6). Following
the general procedure,14 trimethyloxonium tetrafluoroborate (2.19 g,
14.8 mmol) and 2-phenylpyridine N-oxide (2.20 g, 12.8 mmol) were
stirred in 40 mL of dichloromethane for 4 h. Recrystallization from
1
methanol/dichloromethane afforded 3.3 g (12.1 mmol, 94%) of 6. H
NMR (300 MHz, CD3CN): δ (ppm) 4.01 (s, 3 H), 7.65-7.74 (m, 3
H), 7.80-7.84 (m, 2 H), 8.08-8.15 (m, 2 H), 8.55-8.60 (m, 1 H),
9.03 (d, J ) 5.1 Hz, 1 H). 13C NMR (75 MHz, CD3CN): δ (ppm)
69.7, 129.1, 129.4, 130.2, 131.0, 132.3, 133.3, 142.6, 146.5, 153.4.
MS m/z: 186.1 (M+).
N,4-Dimethoxypyridinium Tetrafluoroborate (7). Following the
general procedure,14 trimethyloxonium tetrafluoroborate (1.10 g, 7.43
mmol) and 4-methoxypyridine N-oxide hydrate (0.95 g, 6.64 mmol)
were stirred in 30 mL of dichloromethane for 4 h; 0.81 g (4.24 mmol,
1
64%) of 7 was obtained as a colorless oil. H NMR (300 MHz, CD3-
CN): δ (ppm) 4.07 (s, 3 H), 4.27 (s, 3 H), 7.45 (d, 2 H, J ) 7.8 Hz),
8.74 (d, 2 H, J ) 7.8 Hz). 13C NMR (75 MHz, CD3CN): δ (ppm)
62.4, 73.5, 121.8, 146.1, 174.9. MS m/z: 140.4 (M+).
N,3-Dimethoxypyridinium Tetrafluoroborate (8). Following the
general procedure,14 trimethyloxonium tetrafluoroborate (1.10 g, 7.43
mmol) and 4-methoxypyridine N-oxide hydrate (0.95 g, 6.64 mmol)
were stirred in 30 mL of dichloromethane for 4 h; 0.81 g (4.24 mmol,
1
64%) of 8 was obtained as a colorless oil. H NMR (300 MHz, CD3-
CN): δ (ppm) 3.42 (s, 3 H), 3.82 (s, 3 H), 7.44-7.49 (m, 1 H), 7.56-
7.60 (m, 1 H), 8.36 (m, 1 H), 8.62 (m, 1 H). 13C NMR (75 MHz,
CD3CN): δ (ppm) 58.0, 70.0, 129.0 (2 C), 130.8, 133.3, 159.5. MS
m/z: 140.2 (M+).
N-Ethyl-4-methoxypyridinium Tetrafluoroborate (9). Following
the general procedure,14 triethyloxonium hexafluorophosphate (4.8 g,
18.8 mmol) and 3-methoxypyridine (1.43 g, 1.3 mL, 13.0 mmol) were
stirred in 30 mL of dichloromethane for 4 h; 3.38 g (11.9 mmol, 91%)
of 9 was obtained as a colorless oil. 1H NMR (300 MHz, CD3CN): δ
(ppm) 1.57 (t, 3 H, J ) 7.8 Hz) 4.13 (s, 3 H), 4.47 (q, 2 H, J ) 7.8
Hz), 7.48 (d, 2 H, J ) 7.8 Hz), 8.65 (d, 2 H, J ) 7.8 Hz). 13C NMR
(75 MHz, CD3CN): δ (ppm) 16.3, 56.4, 58.5, 113.4, 146.6, 172.7.
MS m/z: 138.2 (M+).
N-Ethyl-3-methoxypyridinium Tetrafluoroborate (10). Following
the general procedure,14 triethyloxonium hexafluorophosphate (2.60 g,
11.5 mmol) and 3-methoxypyridine (1.43 g, 1.3 mL, 13.0 mmol) were
stirred in 30 mL of dichloromethane for 4 h; 2.58 g (9.1 mmol, 79%)
of 10 was obtained as a colorless solid. 1H NMR (300 MHz, CD3CN):
δ (ppm) 1.58 (t, 3 H, J ) 7.8 Hz) 3.99 (s, 3 H), 4.52 (q, 2 H, J ) 7.8
Hz), 7.88-7.92 (m, 1 H), 7.98-8.03 (m, 1 H), 8.28-8.32 (m, 1 H),
8.36-8.38 (m, 1 H). 13C NMR (75 MHz, CD3CN): δ (ppm) 16.6, 58.3,
56.4, 129.7, 131.4, 133.0, 137.5, 159.9. MS m/z: 138.2 (M+).
Ultrafast Transient Absorption Experiments. The laser apparatus
has been described previously.21 Experiments were performed in 1-cm3
cuvettes with continuous stirring. The concentrations of the p-
trianisylamine and the N-methoxypyridinium compounds were adjusted
Errors in the various rate constants were estimated from repeated
measurements and are different for the different radicals. The errors
on the rate constants for fragmentation are somewhat higher for those
radicals with yields of ions that were particularly high or low. A
maximum error of ca. 25% is estimated for the rate constants in each
case.
Acknowledgment. A research grant from The Research
Corporation is gratefully acknowledged. The authors thank J.
P. Dinnocenzo (University of Rochester) for several helpful
discussions and J. T. Hynes (University of Colorado, Boulder)
for preprints of his work on related fragmentation reactions.
Instrumental support from the ASU Center for the Study of Early
Events in Photosynthesis is acknowledged.
Supporting Information Available: Absorption spectrum of
a typical charge-transfer complex and simulations of the time-
resolved absorption data for the m-methoxy-N-methoxypyridyl
(8R) and p-methoxy-N-methoxypyridyl (7R) radicals. This
material is available free of charge via the Internet at
http://pubs.acs.org.
JA030438+
9
14078 J. AM. CHEM. SOC. VOL. 126, NO. 43, 2004