Photochemistry and spectroscopy of "stable organic radicals": Steric and electronic effects in intermolecular photoinduced electron transfer

Article abstract of DOI:10.1021/jo0017988

The intermolecular reactivities of amino-substituted perchlorotriphenylmethyl radicals 1-3 were studied, with particular emphasis on electron transfer (ET) reactions. The natural fluorescence lifetimes and the rates of the electron-transfer quenching were studied with several electron donors and acceptors. Fluorescence quenching studies demonstrate the importance of the redox potentials of the ET pair on the observed steric and electronic properties.

Full text of DOI:10.1021/jo0017988

3886
J . Org. Chem. 2001, 66, 3886-3892
P h otoch em istr y a n d Sp ectr oscop y of “Sta ble Or ga n ic Ra d ica ls”:
Ster ic a n d Electr on ic Effects in In ter m olecu la r P h otoin d u ced
Electr on Tr a n sfer
Marina Canepa, Marye Anne Fox,* and J ames K. Whitesell*
Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695-8204
mafox@ncsu.edu
Received December 27, 2000
The intermolecular reactivities of amino-substituted perchlorotriphenylmethyl radicals 1-3 were
studied, with particular emphasis on electron transfer (ET) reactions. The natural fluorescence
lifetimes and the rates of the electron-transfer quenching were studied with several electron donors
and acceptors. Fluorescence quenching studies demonstrate the importance of the redox potentials
of the ET pair on the observed steric and electronic properties.
In tr od u ction
ever, chemistry characteristics of solvent-separated radi-
cals ions are observed.
The “inert carbon free radicals,” such as perchloro-
triphenylmethyl (PTM) radical, are passive toward bond-
formation because of steric shielding of the central carbon
atom by the chlorinated phenyl rings.1 They do not
dimerize under any circumstances, they are insensitive
to oxygen, and they do not react with typical radical
reagents or with such highly reactive polar as concen-
trated mineral acids, dihalogens, etc. However, these
“inert” radicals are photosensitive and are active in
electron-transfer, where steric shielding is ineffective.1
As such, derivatives of PTM radicals are useful reac-
tion partners in the study of steric and electronic effects
on reversible electron-transfer chemistry. In bimolecular
electron transfer (ET) between electron donors and
acceptors in solution, the closest approach of the reac-
tants in the transition-state defines the donor/acceptor
separation and orbital overlap limits on the intrinsic rate
of electron exchange. Steric hindrance has been reported2
to be very important to a comprehensive understanding
of the factors governing charge-transfer (CT). Significant
steric effects were observed on the kinetics of CT to
ketones and quinones from aromatic donors.3 Electron
transfer from hindered and unhindered arenes to p-
chloranil also showed steric-dependent kinetics and
established a clear distinction between inner and outer-
sphere et in the electronic coupling of a donor and an
acceptor in the transition state complex preceding the
observed electron transfer.3
Several chemical reactions are known to be sufficiently
fast to compete with charge recombination. For example,
some nucleophilic reactions can compete with single
electron-transfer (SET) process. Both nucleophilic and
SET processes involve a charge shift, and the principal
factors that favor SET include: 1) a strong electrostatic
attraction between the nucleophile and the electrophile;
2) minimal steric hindrance to close approach in the
transition state; 3) substantial delocalization of the odd
electron in the radical ion pairs; and 4) the bond strength
of the putative chemical bond that could be formed
through radical combination.
Syn th esis. Radicals 1-3 were prepared by ammonia-
tion and reduction of the carbenium salt 4, which in turn
was obtained by oxidation of the parent radical PTM with
SbCl5 in SO2Cl2.5 As shown in Scheme 1, the synthesis
of the amino radical 1 was performed following the
method of Ballester et al.5 Treatment of cation 4 in CH2-
Cl2 with NH3 gave a benzimine which reacts with SnCl2
in ether to give the 4-aminotetradecachlorotriphenylmethyl
radical 1. Radical 1 reacts with acetyl chloride or pivaloy
lchloride to produce 4-[(acetyl)amino]tetradecachloro-
triphenylmethyl radical 2, or 4-[(pivaloyl)amido]tetra-
decachlorotriphenylmethyl radical 3, Scheme 2. The
latter species 3 was also prepared following the procedure
reported by J ulia et al.6
Sp ectr a l Ch a r a cter iza tion of 1-3. EP R. The EPR
spectra of radicals 15 and 3 in degassed THF exhibit a
single resonance with two pairs of 13C satellites. The
spectrum of radical 2 shows two lines and computer
simulation allowed calculation of the H coupling constant
(Table 1, Figure 1). Coupling with N was not observed
because the introduction of the acetyl group sterically
inhibits spin delocalization, placing the proton further
from the nodal plane of the lone-electron π orbital while
increasing hyperconjugation. The electron-withdrawing
character of the carbonyl group diminishes the avail-
ability of the nitrogen electron pair, thus indirectly
Photoinduced electron-transfer reactions are often
performed in polar solvents, such as acetonitrile, in which
separation of the initial geminate radical-ion pair can
occur.4 The contact radical-ion pairs are short-lived, and
if physical separation does not take place, secondary
chemical reactions would have to be very rapid in order
to compete with back electron transfer and other rapid
first-order deactivation pathways. Once separated, how-
(1) (a) Ballester, M.; Riera, J .; Castaner, J .; Badia, C.; Monso, J . M.
J . Am. Chem. Soc. 1971, 93, 2215. (b) Ballester, M., Adv. Phys. Org.
Chem. 1989, 25, 354-399.
(2) Pischel, U.; Zhang, X.; Hellrung, B.; Haselbach, E.; Muller, P.;
Nau, W. M. J . Am. Chem. Soc. 2000, 122, 2027-2034.
(3) Hubig, S. M.; Rathore, R.; Kochi, J . K. J . Am. Chem. Soc. 1999,
121, 617-626.
(5) Ballester, M.; Riera, J .; Castaner, J .; Rodriguez, A.; Rovira, C.;
Veciana, J . J . Org. Chem. 1982, 47, 4498-4505.
(6) Teruel, L.; Viadel, L.; Carilla, J .; Fajari, L.; Brillas, E.; Sane, J .;
Rius, J .; J ulia, L. J . Org. Chem. 1996, 61, 6063-6066.
(4) Gould, I. R.; Farid, S. J . Phys. Chem. 1993, 97, 13067-13072.
10.1021/jo0017988 CCC: $20.00 © 2001 American Chemical Society
Published on Web 04/27/2001

Intermolecular Photoinduced Electron Transfer
J . Org. Chem., Vol. 66, No. 11, 2001 3887
Sch em e 1. Syn th esis of 4-Am in otetr a d eca ch lor otr ip h en ylm eth y Ra d ica l
Sch em e 2. Syn th esis of 4-[(a cetyl)Am in o]tetr a d eca ch lor otr ip h en ylm eth yl a n d
4-[(p iva loyl)Am id o]tetr a d eca ch lor otr ip h en ylm eth yl Ra d ica ls
Ta ble 1. EP R Sp ectr a l Da ta of Ra d ica ls 1-3
than the acetyl group in radical 2. Such a twisting
reduces the proton splitting constants because of dimin-
ished delocalization of the unpaired electron, Figure 2.
In contrast, the spectrum of radical 1 shows substantial
coupling with N, since the nitrogen electron pair is
delocalized by resonance with the aromatic ring, Figure
3.
splittings, G
13C
radicals
lines
Na
Ha
Ra
aromatica
1
2
3
1
2
1
1.00
25.0
29.6
29.9
10.3, 12.9
10.6, 13.3
10.3, 12.6
1.65
Electr on ic Absor p tion Sp ectr a . Absorption spectra
of these radicals are characterized by two degenerate one-
electron transitions: from the SOMO to the LUMO and
from the HOMO to the SOMO. The spectra of the amino
radical 1 in a nonpolar solvent shows a marked batho-
chromic shift with solvent polarity, indicating strong
solvation of the excited state relative to the ground state.5
The spectra of radicals 2 and 3 in cyclohexane or in
DMSO show instead a hypsochromic shift with incrased
solvent polarity, partly because of inhibited resonance
from the amino group to the aromatic system caused by
a
Derived by spectral simulation using a Lorentzian line shape
and a line width of 1.3 G.
diminishing still further the observed spin coupling with
the nitrogen.
In radical 3 the bulky pivaloyl group introduces steric
twisting7,8 of the phenyl substituents even more strongly
(7) Schreiner, K.; Berndt, A.; Baer, F. Mol. Phys. 1973, 26, 929-
939.
(8) Hudson, A.; Kent, H. J .; J ackson, R. A.; Treweek, R. F. J . Chem.
Soc., Faraday Trans. 2 1974, 70, 892-894.

3888 J . Org. Chem., Vol. 66, No. 11, 2001
Canepa et al.
F igu r e 1. EPR spectrum of radical 2 (s) and its computer
simulation (- - -).
F igu r e 3. EPR spectrum of radical 1 (s) and its computer
simulation (- - -).
F igu r e 4. UV-vis spectra of 2 in DMSO (s) and in cyclo-
hexane (- - -).
F igu r e 2. EPR spectrum of radical 3 (s) and its computer
simulation (- - -).
3, only a single reversible reduction wave was observed,
Figures 7 and 8. The introduction of the amido group
shifts the half-redox potential (E1/2) by 260 mV and 335
mV negative, respectively, for 2 and 3. Thus, in the
ground state, both 2 and 3 are expected to be good
electron acceptors but poor donors. The insertion of the
additional methylene makes 3 even a better electron
acceptor than 2. The ease of reduction of 3 might be
attributed partly to the weaker hyperconjugation by the
pivaloyl group than with a simple methyl group. A methyl
group produces stronger electron release by hyperconju-
gation9 and the formation of the anion derived from 2 is
less favorable than from 3. There is already evidence that
hypeconjugation plays an important role in the chemical
reactivity of free radicals9 and in 2 and 3 this effect
seems to be considerable. These shifts are particularly
evident with the parent perchlorotriphenylmethyl radical
(PTM).10
the steric blockage of the preferred conformation by the
acetyl group by the two ortho-chlorines and partly
because of electron-withdrawal by the carbonyl groups,
Figure 4. The steric bulk near the carbonyl group in
radical 3 suppresses this solvent effect, Figure 5. The
spectra of radicals 2 and 3 in cyclohexane are unshifted
from the maximum observed for 1.
Electr och em istr y. The redox potentials of radicals
1-3 were determined by cyclic voltammetry (CV) of 1
mM radical solutions in degassed acetonitrile at room
temperature, Table 2. Tetrabutylammonium tetrafluo-
roborate (0.01 M) was used as supporting electrolyte.
The cyclic voltammogram of the amino radical 1
exhibits a reversible single-electron reduction wave at
-510 mV and one reversible single-electron oxidation
wave at + 500 mV, Figure 6. Therefore, this radical can,
in principle, act as an electron donor or as an electron
acceptor, and its absorption spectrum would indicate
enhanced redox reactivity in the excited state. In 2 and
(9) Symons, M. C. R. Tetrahedron 1962, 18, 333.
(10) Fox, M. A.; Gaillard, E.; Chen, C. J . Am. Chem. Soc. 1987, 109,
7088-7094.

Intermolecular Photoinduced Electron Transfer
J . Org. Chem., Vol. 66, No. 11, 2001 3889
F igu r e 7. Cyclic voltammogram of radical 2 (10-3 M in
acetonitrile containing 0.01 M tetrabutylammonium tetrafluo-
roborate). Potentials are reported against a Ag/AgNO3 refer-
ence and were measured in the dark.
F igu r e 5. UV-vis spectra of 3 in DMSO (s) and in cyclo-
hexane (- - -).
Ta ble 2. Red ox P oten tia lsa (vs Ag/Ag+) for Ra d ica ls 1-3
a n d P TM
radicals
E1/2 (mV)
1
2
3
500; -510
-250
-175
P TM
-430
a
Measured as 1 mM radical solutions containing tetrabutyl-
ammonium tetrafluoroborate (0.01 M) in degassed acetonitrile; Pt
cathode, 100 mV/sec.
F igu r e 8. Cyclic voltammogram of radical 3 (10-3 M in
acetonitrile containing 0.01 M tetrabutylammonium tetrafluo-
roborate). Potentials are reported against a Ag/AgNO3 refer-
ence and were measured in the dark.
Qu en ch in g Exp er im en ts. Fluorescence quenching of
radicals 1-3 by a redox partner follows a Stern-Volmer
relationship11
I0/I ) 1 + KSV ) 1 + kτ0[Q]
where I0 and I are the relative fluorescence intensities
in absence and presence of the quencher Q, KSV is the
Stern-Volmer quenching rate constant, τ0 is the fluo-
rescence lifetime of the fluorophore in absence of the
quencher, and k is the bimolecular rate constant of
quenching. Although electron transfer is assumed to be
the dominant nonradiative decay pathway, solvent polar-
ity is believed to alter the quenching mechanism.12-14 In
polar solvents, quenching is governed by solvation of the
F igu r e 6. Cyclic voltammogram of radical 1 (10-3 M in
acetonitrile containing 0.01M tetrabutylammonium tetrafluo-
roborate). Potentials are reported against a Ag/AgNO3 refer-
ence and were measured in the dark.
Tim e-Resolved Mea su r em en ts. The fluorescence
lifetimes of radicals 1-3 are reported in Table 3. The
amino-substituted radicals 1 and 3 exhibit very similar
lifetimes in the presence of O2, but under the same
conditions the lifetime of the acetylaminoradical 2 is
significantly shorter. This is surprising in that the excited
doublet state10 of the PTM radical was not quenched by
oxygen or by electron acceptors such as duroquinone (E1/2
) -0.235 mV), even though its ground-state reduction
potential (-430 mV) was less negative than required.
(11) Braun, A. M.; Maurette, M. T.; Olivero, E. Photochem. Technol.
1991, 41.
(12) Mataga, N.; Nakashima, N. Spectrosc. Lett. 1975, 8, 275.
(13) Knibbe, H.; Rolling, K.; Schafer, F. P.; Weller, A. J . Chem. Phys.
1967, 47, 1184.
(14) Froehlich, P.; Wehry, E. L. Modern Fluorescence Spectroscopy
2; 1976; Chapter 8.

3890 J . Org. Chem., Vol. 66, No. 11, 2001
Canepa et al.
Ta ble 3. F lu or escen ce Lifetim esb (λexc ) 337 n m ) in
Cycloh exa n e a t th e λm a x Obser ved in th e Em ission
Sp ectr u m of Ea ch Com p ou n d (608 < λem < 634 n m )
radicals
λem (nm)
τ (ns)
1
2
3
634
613
608
605
8.9
4.6
8.2
7.0a
P TM
Reference 10. b Typical errors in observed fluorescence lifetime
a
are 10%.
Ta ble 4. Kin etic Con sta n tsa for F lu or escen ce Qu en ch in g
of 1-3 Ca lcu la ted fr om th e Ster n -Volm er P lots
k*10 [M-1 sec-1
]
13
radical
solvent
quencher
1
2
3
1
2
3
2
3
cyclohexane
cyclohexane
cyclohexane
acetonitrile
acetonitrile
acetonitrile
acetonitrile
acetonitrile
Ph3N
Ph3N
Ph3N
Ph3N
Ph3N
Ph3N
p-chloranil
p-chloranil
2.69
1.97
1.88
2.95
0.64
0.84
3.97
0.81
F igu r e 9. Stern-Volmer plot of fluorescence quenching of
radical 1 by triphenylamine (ν) and by p-chloranil (τ) in
acetonitrile.
a
Typical errors in calculated rate constants are 10%.
resulting ions and only a minor steric effect could be
observed. In nonpolar solvents, exciplex formation is
likely to be dominant in the observed quenching. Steric
hindrance should therefore be very significant, since the
transition-state geometry limits the intrinsic rate of the
electron exchange.
The samples of radical 1 in cyclohexane were freshly
prepared before each measurement because they showed
appreciable decomposition under irradiation, with fluo-
rescence intensity rapidly decreasing. In contrast, solu-
tions of 1 in acetonitrile were very stable. Both an
oxidative and reductive quencher were used in these
experiments. Triphenylamine and p-chloranil are known
to be excellent electron-transfer reagents (E0 ) 0.70 V
ox
vs SCE2,15-17 and E0red ) 0.02 V vs SCE, respectively). In
addition, p-chloranil is reported3 to have a long-lived (µs)
excited (triplet) state that exhibits a high reduction
potential (E* ) 2.15 V vs SCE).
red
F igu r e 10. Stern-Volmer plot of fluorescence quenching of
radical 2 by p-chloranil (0) and by triphenylamine (O) in
acetonitrile.
To calculate the rate constant for quenching in accord
with the Stern-Volmer relationship, we collected the
fluorescence intensities of each radical in solution in the
absence of the quencher (I0) and monitored fluorescence
intensities (I) after each addition of the quencher. The
reaction of radical 1 in acetonitrile with triphenylamine
allowed us to calculate the quenching rate constant
(Table 4) within the range of applicability of the
Stern-Volmer relationship, Figure 9. The experiments
were performed both in aerated and in deoxygenated
solvent. Consistent results were obtained under either
condition.
of intensity, and we could calculate the derived rate
constant, Table 4.
Fluorescence of solutions of 2 and 3 both in cyclohexane
and acetonitrile were monitored by adding the same
quenchers as with 1, and the corresponding Stern-
Volmer plots were obtained, Figures 10 and 11. (The raw
data, establishing the range over which the Stern-
Volmer law remains linear, are available in the Support-
ing Information.)
Analogous quenching of radical 1 in degassed aceto-
nitrile by chloranil showed anomalous behavior. Each
addition of the quencher caused the fluorescence intensity
(I) to significantly increase, Figure 9. Neither emission
from the solvent nor from the quencher produce this
effect when examined separately.
Discu ssion
The above data can be interpreted by the general
Scheme 3 illustrating electronic and solvent effects on
the quenching rate.
The rate constant for fluorescence quenching (k) is
therefore a function of k1, k-1, k2, and k3.
In cyclohexane (whether deoxygenated and not), quench-
ing of 1 by triphenylamine gave the expected decrease
k2 + k3
k ) k1
(
)
(15) Dalton, J . C.; Snyder, J . J . J . Am. Chem. Soc. 1975, 97, 5192-
5197.
(16) Lishan, D. G.; Hammond, G. S.; Yee, W. A. J . Phys. Chem. 1981,
85, 3435-3440.
(17) Turro, N. J .; Engel, R. J . Am. Chem. Soc. 1969, 91, 7113-7121.
k-1 + k2 + k3
In cyclohexane (a nonpolar solvent), k3 is negligible but
it is very significant in polar solvents as acetonitrile.

Intermolecular Photoinduced Electron Transfer
J . Org. Chem., Vol. 66, No. 11, 2001 3891
methyl group which produces the most powerful electron
releasing effect in this series.
Further evidence for the importance of the substitutent
group in fluorescence quenching is found by studying
radical 3. The pivaloyl group in the radical 3 does not
release electrons effect as powerfully as does the methyl
group in radical 2. The rate constant for SET from 3 in
acetonitrile to p-chloranil (k ) 0.81*1013 [M-1 sec-1]) is
much lower than for 2 (k ) 3.97*1013 [M-1 sec-1]) where
the carbocation was strongly stabilized by hyperconju-
gation by the methyl group. Therefore, in this solvent,
fluorescence quenching by triphenylamine as electron
donor is favored in radical 3 (k ) 0.84*1013 [M-1 sec-1])
over radical 2 (k ) 0.64*1013 [M-1 sec-1]).
Devia tion s fr om th e Ster n -Volm er La w . At low
concentrations, the quenching of the fluorescence by a
quencher molecule in solution follows the classic Stern-
Volmer relationship. Since the fluorescence lifetime is
independent of the quencher concentration, Stern-
Volmer plots will be linear as long as the rate constant
k is independent of quencher. It has been known for
many years that associative quenching reactions can lead
to curved Stern-Volmer plots.18,19 Both positive and
negative deviations have been observed, and several
plausible explanations have been proposed.20-26 Negative
curvature involves a decrease in KSV and is associated
with a change in the absorption and fluorescence spec-
trum of the fluorophore.20 Deviations from the Stern-
Volmer equation in such reactions have been explained
by the existence of multiple fluorescing states27 or by
compound formation.20 In our experiments we observed
appreciable Stern-Volmer deviations upon adding large
quantities of quencher. Reactions with radicals 2 and 3
exhibited positive deviations (Figure 10), which has been
explained in the literature as either “static” quenching19
at high quencher concentration, or as deviations from free
diffusion in solution.28,29 The latter effect is based on the
assumption that the quenching reaction is very rapid
and, hence, diffusion contolled.30-32 On the other hand,
radical 1 possesses an amino group not protected as an
amide as in radicals 2 and 3. If photoexcited, radical 1
can undergo reactions with a range of quenchers.33 Thus,
the plots show negative deviations in the experiments
with triphenylamine both in cyclohexane and in aceto-
nitrile. In particular in acetonitrile, the classical polar
solvent, the reaction between the photoactivated radical
1 and p-chloranil is so fast that it was not possible for us
to observe the net quenching, Figure 9.
F igu r e 11. Stern-Volmer plot of fluorescence quenching of
radical 3 by triphenylamine in cyclohexane (T) and in aceto-
nitrile (∆).
Sch em e 3a
a
R•* is the radical in the excited state. Q is the quencher, (a)
is an encounter complex, (b) is an exciplex, and (c) is an ion pair.
Oxidative quenching by an electron acceptor is shown on the left;
reductive quenching by an electron donor is shown on the right.
Therefore, in a less polar solvent, k2 determines the
quenching rate. This process involves geometric and
solvent relaxation of the encounter complex to an equi-
librium exciplex and an increase in steric hindrance plays
an important role on the observed quenching rate.
In cyclohexane, the rate constants for 2 (k ) 1.97*1013
[M-1 sec-1]) are lower than those for 1 (k ) 2.69*1013 [M-1
sec-1]) under the same conditions. This behavior has been
attributed to the increased steric bulk of the substituent
rather than to an electronic effect, since a specific
geometry is required between the radical and the quench-
er in order to generate an emissive exciplex. In accord
with our hypothesis, we found the lowest rate constant
for radical 3 (k ) 1.88*1013 [M-1 sec-1]).
(18) J ette, E. R.; West, W. Proc. R. Soc. London Ser. A 1928, 121,
299-312.
In acetonitrile, steric effects are less relevant. Here,
the quenching process is governed by electron transfer
which requires only a loose encounter complex. Therefore,
it is possible to isolate electronic effects on the quenching.
The fluorescence quenching with triphenylamine as
electron donor is favored in radical 1 (k ) 2.95*1013 [M-1
sec-1]) over radical 2 (k ) 0.64*1013 [M-1 sec-1]), but a
higher value for the rate constant is found for 2 with
p-chloranil (k ) 3.97*1013 [M-1 sec-1]). In this polar
solvent, electron-withdrawal by the substituent plays an
important role in the resonance stabilization of the
ground state of 2. Thus, the rate constant for SET to 2
in acetonitrile from Ph3N is lower than with the amino
radical 1. Electron-transfer between radical 2 and p-
chloranil in acetonitrile is faster, since the formation of
the carbocation is favored by hyperconiugation by the
(19) Frank, I. M.; Wawilow, S. I. Z. Phys. 1931, 69, 100-110.
(20) Boaz, H.; Rollefson, G. K. J . Am. Chem. Soc. 1950, 72, 3435.
(21) Keizer, J . J . Am. Chem. Soc. 1983, 105, 1494.
(22) Zang, H.; Durocher, G. J . Lumin. 1995, 63, 75.
(23) Shafirovich, V. Y.; Courtney, S. H.; Ya, N.; Geacintov, N. E. J .
Am. Chem. Soc. 1995, 117, 4920.
(24) Eftnik, M. R.; J ia, Y.; Hu, D.; Ghiron, C. A. J . Phys. Chem. A
1995, 99, 5713.
(25) Wilkinson, F.; Abdel-Shafi, A. A. J . Phys. Chem. A 1997, 101,
5509.
(26) Nath, S.; Pal, H.; Palt, D. K.; Sapre, A. V.; Mittal, J . P. J . Phys.
Chem. A 1998, 102, 5822.
(27) Rollefson, G. K.; Boaz, H. J . Phys. Colloid. Chem. 1948, 52,
518-527.
(28) Keizer, J . J . Phys. Colloid. Chem. 1981, 85, 940-941.
(29) Keizer, J . J . Phys. Chem. 1982, 86, 5052-5067.
(30) Noyes, R. M. J . Am. Chem. Soc. 1957, 79, 551-555.
(31) Noyes, R. M. J . Phys. Chem. 1961, 65, 763-765.
(32) Noyes, R. M. Prog. React. Kinet. 1961, 1, 131-160.
(33) Guha, D.; Mitra, S.; Das, R.; Mukherjee, S. Ind. J . Chem. 1999,
38, 760-767.

3892 J . Org. Chem., Vol. 66, No. 11, 2001
Canepa et al.
Am m on olysis of Ca r ben iu m Sa lt 4. Dry NH3(g) was
passed slowly (4 min) through a suspension of salt 4 (0.100 g)
in CH2Cl2 anhydrous (10 mL). The reaction was performed
under extremely dry conditions: the glassware was dried
under vacuum and then washed with argon many times; a CaO
trap was used to dehydrate the NH3(g) after the CaO was
preheated at 800 °C overnight. The resulting mixture was
filtered and the filtrate was evaporated to dryness giving NH-
tetradecachlorofuchsonimine which was used in the next
reaction without further purification.
Con clu sion s
The reactivity of a new series of amino-substituted
perchlorotriphenylmethyl radicals with electron acceptors
and donors has been described. In general, fluorescence
quenching follows a normal Stern-Volmer relationship,
and clear evidence for steric and electronic effects on the
rate of electron transfer were obtained. Significant
mechanistic diversion observed in polar and nonpolar
solvents allowed us to distinguish between steric and
electronic effects in these single electron-transfer reac-
tions.
4-Am in otetr a d eca ch lor otr ip h en ylm eth yl Ra d ica l (1).
Anhydrous SnCl2 (0.040 g) was added to a solution of crude
fuchsonimine (0.080 g) in ethyl ether anhydrous (18 mL) and
the mixture was stirred in the dark under Ar overnight. The
green solution was then filtered, washed with aqueous HCl
and with water, dried and evaporated. The residue was
purified by column flash-chromatography (silica gel hexane-
CCl4) giving amino radical: 0.050 g (62%), dark green needles.
The structure was confirmed by using IR spectroscopy (ger-
manium crystal): 3500, 3395, 1600, 1530, 1440, 1375, 1360,
1340, 1330, 1325, 1295, 1260, 820, 775, 760, 735, 710, 650
Exp er im en ta l Section
Ap p a r a ti. NMR spectra were recorded with a Varian
Gemini 200 MHz instrument. IR spectra were recorded with
a
Bio-Rad FT-500 with a UMH-600 microscope using a
germanium crystal ATR objective. Absorption spectra were
recorded on a Shimadzu PC-3101PC instrument and ESR
spectra were recorded on an IBM-Bruker E200SRC spectrom-
eter.
cm-1
.
4-[(acetyl)Am in o]tetr adecach lor otr iph en ylm eth yl Rad-
ica l (2). A solution of amino radical 2 (0.050 g) in acetyl
chloride (5 mL) was left (3 days) under argon in the dark. On
elimination of the solvent a red residue corresponding to the
radical 4 was obtained: 90%.
UV-vis (C6H12) 385, 565 nm; (Me2SO) 388, 514 nm. IR
(germanium crystal): 2950, 1800, 1675, 1530, 1360, 1330,
1260, 1225, 815, 730, 710, 650 cm-1. Anal. Calcd. for C21H4-
Cl14NO: C, 32.2; H, 0.5; Cl, 63.4; N, 1.8; O, 2.0. Found: C,
35.1; H, 1.8; Cl, 52.8; N, 1.7; O, 3.4.
4-[(ter t-Bu tyla cetyl)a m in o]tetr a d eca ch lor otr ip h en yl-
m eth yl Ra d ica l (3). A solution of tert-butylacetyl chloride
(0.027 g, 0.198 mmol) in CHCl3 (1 mL) was added dropwise to
a cold stirred (0 °C) solution of radical 2 (0.100 g, 0.135 mmol)
and triethylamine (0.024 g, 0.235 mmol) in CHCl3 (1.5 mL)
under argon. The resulting mixture was stirred further at 0
°C for 1 h and then at room temperature for 22 h. Evaporation
of the solvent gave a residue that was placed in CHCl3, washed
with diluted aqueous HCl and water, and dried with anhy-
drous Na2SO4. The solution was filtered and evaporated to give
a residue which was purified by flash-chromatography (silica
gel hexane-CCl4) to afford radical 5 (0.010 g, 10%).
IR (germanium crystal): 2950, 2850, 1725, 1660, 1520, 1460,
1390, 1330, 1260, 1220, 1150, 1120, 1100, 820, 740, 720, 700,
660 cm-1. UV-vis (C6H12): 385, 580 nm; (Me2SO) 385, 580
nm. Anal. Calcd. for C25H12Cl14NO: C, 35.8; H, 1.4; Cl, 59.2;
N, 1.7; O, 1.9. Found: C, 40.4; H, 3.5; Cl, 44.6; N, 1.15; O, 2.0.
Fluorescence spectra of solutions were measured on a PTI
QuantaMaster Model C-60/2000 spectrofluorometer. Lifetime
measurement were made on a PTI StrobeMaster fluorescence
spectrometer with a photon counting stroboscopic detector.
Cyclic voltammetry (CV) was performed with a BAS-100
electrochemical analyzer on a platinum electrode in a standard
three electrode cell. Measurements were reported against a
Ag/AgNO3 reference electrode and were done in the dark.
Solutions were approximately 1 mM radical in acetonitrile
containing 0.1 M tetra-n-butylammonium tetrafluoroborate.
Ma ter ia ls. Solven ts. SO2Cl2 was freshly distilled before
each reaction and was allowed to react under Ar. Chloroform
(anhydrous 99.9% Aldrich) was used without further purifica-
tion. Ethyl ether and THF were distilled under Ar from
metallic Na, with benzophenone being used as the indicator.
Methylene chloride was distilled over P2O5 under Ar. Cyclo-
hexane (HPLC grade, 99.9% Aldrich) was used without further
purification, as was DMSO spectroscopic grade (99.9%, Ald-
rich) and anhydrous carbon tetrachloride (99%, Aldrich).
Acetonitrile HPLC grade (Fisher) for the CV measurements
and quenching experiments was purified by refluxing over
P2O5, followed by distillation from P2O5 first and then from
CaH2. The purified solvent was stored under Ar over molecular
sieves activated in an oven.
Rea gen ts. The handling of radicals in solution was per-
formed in the dark.
RH-quasi-Perchlorotriphenylmethane (PTM-H) and perchlo-
rotriphenylmethyl radical (PTM) were prepared by the method
of Ballester et al.1
Sa m p le for th e Qu en ch in g Exp er im en ts. Triphenyl-
amine was purified by recrystallization from petroleum ether.
Chloranil was recrystallized from toluene.
Ack n ow led gm en t. This work was supported by the
National Science Foundation. The authors are grateful
to Prof. D. A. Shultz and Mr. Rosario M. Fico J r. for
EPR measurements and to Dr. K. W. Kittredge for
assistance with lifetime measurements. Elemental analy-
ses were obtained at Atlantic Microlab.
The radicals were purified by column flash-chromatography
under Ar and stored under vacuum in the dark. The solutions
were prepared for each radical both in cyclohexane and in
acetonitrile with a concentration on the order of 10-6 M.
P er ch lor otr iph en ylca r ben iu m Hexa ch lor oa n tim on a te
(4). SbCl5 (2.8 mL) was added slowly at room temperature to
a solution of PTM radical (0.61 g) in SO2Cl2 (78.5 mL) and the
resulting mixture was left undisturbed under Ar in the dark
(96 h). The small dark green cubes which formed were
separated, washed with anhydrous CCl4, and dried under
vacuum to give the salt. Precipitation from the mother liquors
by the addition of CCl4 afforded more pure salt (0.75 g, 86.5%).
Su p p or tin g In for m a tion Ava ila ble: Tables 1-3, quench-
ing experiments of radical 1 in acetonitrile and cyclohexane;
Tables 4-6, quenching experiments of radical 2; Tables 7-9,
quenching experiments of radical 3; Figures 12 and 13, Stern-
Volmer plots of fluorescence quenching for radical 1 and 2 in
cyclohexane with Ph3N; Figure 14, Stern-Volmer plot for
radical 3 in acetonitrile with quencher p-chloranil. This
material is available free of charge via the Internet at
http://pubs.acs.org.
IR (polypropylene) 340 cm-1
.
J O0017988