Synthesis, Crystal Structures, and Laser Flash Photolysis of tert-Butyl Aroylperbenzoates

Article abstract of DOI:10.1021/jo030168d

tert-Butyl aroylperbenzoates (1-7) were synthesized. Single-crystal structures for 2 and 5 show that the perester and benzophenone carbonyl groups are almost coplanar in each. Laser flash photolysis (LFP, λ_ex = 355 nm) of 1-5 in CCl₄

Full text of DOI:10.1021/jo030168d

Syn th esis, Cr ysta l Str u ctu r es, a n d La ser F la sh P h otolysis of
ter t-Bu tyl Ar oylp er ben zoa tes1
Bipin K. Shah and Douglas C. Neckers*
Center for Photochemical Sciences, Bowling Green State University, Bowling Green, Ohio 43403
neckers@photo.bgsu.edu
Received May 19, 2003
tert-Butyl aroylperbenzoates (1-7) were synthesized. Single-crystal structures for 2 and 5 show
that the perester and benzophenone carbonyl groups are almost coplanar in each. Laser flash
photolysis (LFP, λex ) 355 nm) of 1-5 in CCl4 produces the corresponding aroylphenyl radicals
(9-13). The lifetimes of the para aroyl-substituted phenyl radicals (9-12) are similar (∼0.4 µs),
but each is shorter lived than the meta aroyl-substituted phenyl radical (13). LFP of 2, 6, and 7
also produces different (tert-butyldioxycarbonylbenzoyl)benzyl radicals (8, 14, and 15, λmax ≈ 320
nm). The lifetimes of each in CCl4 have been found to be ∼17-18 µs. The effect of substituents on
the quantum yield of decomposition of 1-7 and the lifetimes of 9-13 is discussed.
In tr od u ction
assigned to the aroylphenyl radicals, and bimolecular
rates of quenching of the 4-benzoylphenyl and 3-(4′-
methylbenzoyl)phenyl radicals by double-bond-containing
monomers have been found to be in the range of 107-
Although thousands of tons of vinyl polymers are made
annually by diaroyl peroxide and peroxyester-catalyzed
free-radical polymerization, peroxides are rarely used as
photoinitiators, even when their bond homolysis reactions
can be made efficient by sensitization. Photochemical
processes involving aromatic ketones and either the
Norish Type I cleavage or bimolecular electron transfer
processes that produce radicals are generally used for
photoinitiated monomer to polymer conversions.2 tert-
Butyl aroylperbenzoates (BP) are good candidates as
initiators for photochemical vinyl polymerization because
of the known direct dissociation of their weak -O-O-
bond. Their photochemical decomposition and activities
as photoinitiators have been investigated.3,4 They have
been shown to be efficient photochemical sources of
aroylbenzoyloxyl and aroylphenyl radicals.3,4
108 M-1 s-1 6
.
The photochemical behavior of BPs containing one or
two dissociable groups, e.g., tert-butyl 4-(4′-methylben-
zoyl)perbenzoate (1) and tert-butyl 4-(4′-chloromethyl-
benzoyl)perbenzoate (2), has also been compared re-
cently.7 2, which has both C-Cl and perester groups
conjugated with the benzophenone chromophore, pro-
duces not only the 4-(4′-chloromethylbenzoyl)benzoyloxyl
and 4-(4′-chloromethylbenzoyl)phenyl radicals but also
the benzyl-type radical [4-(4′-tert-butyldioxycarbonylben-
zoyl)benzyl radical], the latter forming due to direct
cleavage of the C-Cl bond.7
tert-Butyl 4-(3′-methylbenzoyl)perbenzoate (3), tert-
butyl 4-(4′-bromobenzoyl)perbenzoate (4), tert-butyl 3-(4′-
methylbenzoyl)perbenzoate (5), tert-butyl 3-(4′-bromo-
methylbenzoyl)perbenzoate (6), and tert-butyl 3-(3′-bro-
momethylbenzoyl)perbenzoate (7) (Figure 1) were re-
cently synthesized and studied. Herein we report quan-
tum yields of dissociation of these BPs and the lifetimes
of the corresponding aroylphenyl radicals obtained by
LFP. We also present spectroscopic evidence for the
formation of benzyl-type radicals and consider their
lifetimes.
The kinetics of formation and decay of the singlet and
triplet states of BPs have been studied by laser flash
photolysis (LFP).5,6 Lifetimes of their triplet states are
less than ∼1 ns.5,6 The 550 nm absorption obtained from
the 355 nm LFP (pulse width ∼7 ns) of BPs has been
* To whom correspondence should be addressed. Telephone: (419)
372-2034. Fax: (419) 372-0366.
(1) Contribution No. 502 from the Center for Photochemical Sci-
ences.
(2) (a) Roffey, C. G. Photopolymerization of Surface Coatings; J ohn
Wiley & Sons: New York, 1982; p 67. (b) Lai, Y.; Quinn, E. T. In
Photopolymerization Fundamentals and Applications; Scranton, A. B.,
Bowman, C. N., Peiffer, R. W., Eds.; ACS Symposium Series 673;
American Chemical Society: Washington, DC, 1997; p 35.
(3) Thijs, L.; Gupta, S. N.; Neckers, D. C. J . Org. Chem. 1979, 44,
4123.
(4) (a) Gupta, S. N.; Thijs, L.; Neckers, D. C. J . Polym. Sci., Polym.
Chem. Ed. 1981, 19, 855. (b) Gupta, I.; Gupta, S. N.; Neckers, D. C. J .
Polym. Sci., Polym. Chem. Ed. 1982, 20, 147. (c) Allen, N. S.; Hardy,
S. J .; J acobine, A. F.; Glaser, D. M.; Yang, B.; Wolf, D.; Catalina, F.;
Navaratnam, S.; Parsons, B. J . J . Appl. Polym. Sci. 1991, 42, 1169.
(5) Morlino, E. A.; Bohorquez, M. D.; Neckers, D. C.; Rodgers, M.
A. J . J . Am. Chem. Soc. 1991, 113, 3599.
Resu lts a n d Discu ssion
Qu a n tu m Yield s of Dissocia tion . BPs 1-7 show the
characteristic nfπ* transition of the benzophenone chro-
mophore. Each has a maximum at ∼350 nm in both
benzene and CCl4. When irradiated at 350 nm, each
produces similar products that are derived from the
corresponding aroylbenzoyloxyl and aroylphenyl radicals.
(6) Shah, B. K.; Neckers, D. C. J . Am. Chem. Soc., Submitted.
(7) Shah, B. K.; Neckers, D. C. J . Org. Chem. 2002, 67, 6117.
10.1021/jo030168d CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/04/2003
8368
J . Org. Chem. 2003, 68, 8368-8372

tert-Butyl Aroylperbenzoates
F IGURE 2. ORTEP drawings (50% probability ellipsoids) of
2 and 5.
F IGURE 1. tert-Butyl aroylperbenzoates.
a
TABLE 1. Qu a n tu m Yield s of Bon d Dissocia tion (Od iss
)
Decomposition of the CH2-Br bond occurs preferen-
tially to that of the perester bond. Consequently, because
of the heavy-atom-induced cage radical pair recombina-
tion of the bromine atom with the benzyl-type radical,
one observes a reduced φdiss of the perester bond in these
molecules. The heavy atom may enhance the depopula-
tion of the triplet state by inducing nonradiative decay
of the triplet state,8 and this may lower φdiss as observed
in the case of 4-(bromomethyl)benzophenone.7 However,
this is ruled out in the case of the BPs. When bromine is
present in a nondissociable form, the effect was not
observed, as is reflected by the higher φdiss of the perester
bond of 4.
φdiss
BP
-O-O-
C-X
1b
2b
3
4
5
0.82
0.63
0.83
0.87
0.92
0.33
0.43
0.26
6
7
0.49
0.39
Intensity (I0) ) 9.2 × 1016 quanta L-1 s-1; solvent ) C6D6;
a
concentration ) 7.25-8.42 × 10-3 M; error ) within 10%. Ref 7.
b
Thus, decomposition of 1-7 follows the general mecha-
nism of homolytic dissociation of the perester bond due
to the rapid intramolecular triplet energy dissipation.7
In the case of BPs containing the CH2-X group (2, 6,
and 7), dissociation of the C-X bond is also observed as
indicated by formation of dehalogenated products.
Cr ysta l Str u ctu r es. Photochemical dissociation of
peroxides in the solid state has been the subject of various
studies.9 Although esters generated from tert-alkyl hy-
droperoxides are comparatively stable,10 crystal structure
data on peroxides and peresters are rather limited.11 This
is undoubtedly due to the decay that often occurs when
their crystals are exposed to the X-ray radiation.12 BPs
1-7 form high-quality crystals and are stable at room
temperature. X-ray crystal structures of 2 and 5 have
been determined in order to assess the structural effect
imparted by the perester moiety when it is present at
the para or meta position with respect to the benzophe-
none carbonyl group.
The quantum yields of dissociation (φdiss) of the perester
and C-X bonds as measured in C6D6 are presented in
Table 1. The φdiss value of the perester bond of 5 was the
highest among that of 1-7. However, there are no major
differences in the φdiss values for the perester cleavage of
1-5, indicating that the φdiss of the perester bond is not
significantly affected by the perester moiety. It makes
little difference whether the perester is present at the
para or meta position with respect to the benzophenone
carbonyl group. Similarly, the position of the methyl
group or even replacing the methyl group with a bromine
atom has little or no influence on the φdiss of the perester
bond.
The single-crystal structures of 2 and 5 with a labeling
scheme of atoms are shown in Figure 2. They have been
found to crystallize in different space groups. Crystals
of 2 are monoclinic, space group C2, while those of 5 are
monoclinic, space group P2(1). The perester carbonyl
group [C(14)-O(2)] assumes anti conformation with
The higher φdiss value observed for either the C-X bond
or the perester bond is the limiting quantum yield of
decomposition for 2, 6, and 7.7 The higher φdiss values of
6 and 7 are 0.49 and 0.43, respectively, which are much
lower than that of other BPs. This indicates that their
decomposition is a slow process. The φdiss values of the
perester bond of 6 (0.33) and 7 (0.43) are also substan-
tially lower than those of the other BPs. This is due to
the presence of the benzylic bromide function, which has
been found to decrease the φdiss of the perester bond.7 A
benzylic bromide function at either the para or meta
position has a similar effect.
(8) (a) Inoue, H.; Sakurai, T.; Hoshi, T. Bull. Chem. Soc. J pn. 1991,
64, 3340. (b) Darmanyan, A. P. Chem. Phys. Lett. 1983, 102, 329. (c)
Cowen, D. O.; Koziar, J . C. J . Am. Chem. Soc. 1974, 96, 1229.
(9) (a) Feng, X. W.; McBride, J . M. J . Am. Chem. Soc. 1990, 112,
6151. (b) McBride, J . M.; Vary, M. W. Tetrahedron 1982, 38, 765. (c)
Karch, N. J .; Koh, E. T.; Whitsel, B. L.; McBride, J . M. J . Am. Chem.
Soc. 1975, 97, 6729.
(10) (a) Bouillon, G.; Lick, C.; Schank, K. In The Chemistry of
Functional Groups, Peroxides; Patai, S., Ed.; J ohn Wiley & Sons: New
York, 1983; p 279. (b) Exner, O. In The Chemistry of Functional Groups,
Peroxides; Patai, S., Ed.; J ohn Wiley & Sons: New York, 1983; p 85.
(11) Antolini, L.; Benassi, R.; Ghelli, S.; Folli, U.; Sbardellati, S.;
Taddei, F. J . Chem. Soc., Perkin Trans. 2 1992, 1907.
(12) Golic, L.; Leban, I. Acta Crystallogr., Sect. C 1984, 40, 447.
J . Org. Chem, Vol. 68, No. 22, 2003 8369

Shah and Neckers
SCHEME 1. F or m a tion of Ar oylp h en yl a n d Ben zyl-Typ e Ra d ica ls
respect to the benzophenone carbonyl group [C(7)-O(1)]
in 2, while the carbonyl groups assume syn conformation
in 5.
The geometry of the perester group is similar in both
compounds with perester bond [O(3)-O(4)] distances of
1.473(4) Å in 2 and 1.465(3) Å in 5. The dihedral angles
between the two mean planes created by C(1)-C(7) and
C(7)-C(13) have been found to be 49.1 and 130.2° in 2
and 5, respectively. No short van der Waals contacts
between molecules were observed. The shortest distance
between the molecules of 2 is found to be 3.99 Å. The
same for 5 is 6.14 Å, which is the distance of one of its
unit cell dimensions.
The φdiss values of the perester bond are 0.63 for 2 and
0.92 for 5 in benzene. The crystal structures show that
the carbonyl and perester groups, as well as the aromatic
ring between them [C(7)-O(1), C(8)-C(13), and C(14)-
O(3)-O(4)], are almost planar in both 2 and 5. The torsion
angle between the two carbonyl groups has been found
to be 22.2 and 12.9° in 2 and 5, respectively. Although
scission of the perester bond in solution occurs at the
triplet excited-state manifold, planarity of the two car-
bonyl groups and the perester bond also explains why
the intramolecular triplet energy dispersion is so facile
in these molecules, leading to the higher φdiss of the
perester bond.13
F IGURE 3. Transient absorption spectra obtained from 355
nm laser flash photolysis of 1 (1.75 × 10-3 M) in CCl4, recorded
(b) 20 ns and (O) 2 µs after the laser pulse. Inset: kinetic trace
monitored at 550 nm (from 1.18 × 10-3 M 1 in CCl4).
La ser F la sh P h otolysis. The 355 nm LFP (pulse
width ∼7 ns, 19 mJ /pulse) of 1-4 in CCl4 resulted in two
absorption bands having maxima at 320 and 550 nm
(Figures 3 and 4). The 550 nm band can be assigned to
the corresponding aroylphenyl radicals (9-12, Scheme
1) since the 550 nm absorption bands obtained from the
355 nm LFP of tert-butyl 4-(benzoyl)perbenzoate and 5
have been assigned to the 4-benzoylphenyl and 3-(4′-
methylbenzoyl)phenyl (13) radicals, respectively. Details
on this assignment have been previously provided.6
The lifetimes of 9-13 recorded in CCl4 are included in
Table 2, and the rates of quenching (kq) of two para
radicals (9 and 12) by various quenchers are presented
F IGURE 4. Transient absorption spectra obtained from 355
nm laser flash photolysis of 4 (1.79 × 10-3 M) in CCl4, recorded
(b) 20 ns and (O) 2 µs after the laser pulse. Inset: kinetic trace
monitored at 550 nm (from 1.29 × 10-3 M 4 in CCl4).
in Table 3. The meta-centered radical (13) is longer lived
than the para-centered radicals (9-12). However, there
are no significant differences in the lifetimes of 9-12 in
similar concentration ranges of the precursor BPs. The
4-benzoylphenyl radical also has a similar lifetime (0.41
µs) in CCl4.6 This may be because the substituents are
too remote to impart an observable effect on such reactive
radicals. The kq values of 9 and 12 by 1,4-cyclohexadiene
and by the precursor BPs have been observed to be in
(13) There may be considerable differences in the geometries of these
molecules when they are in their solid state and in solution, especially
because of the flexible nature of the appended perester moiety.
However, the high quantum yield of O-O bond dissociation suggests
that molecules assume appropriate geometries in their triplet states
such that the activation energy in causing scission of the O-O bond
is minimal. A flat geometry in their triplet states might be imagined
in which case triplet energy dissipation may be very facile. Density
functional theory (DFT) calculations on some of these molecules are
in the process of being carried to obtain information about the structure
of their triplet states.
8370 J . Org. Chem., Vol. 68, No. 22, 2003

tert-Butyl Aroylperbenzoates
TABLE 2. Lifetim es (τ) of Ar oylp h en yl Ra d ica ls
(Solven t ) CCl4)
precursor BP
radical
(concentration, M)
τ, µs
9
10
11
12
13a
1.18 × 10-3
1.77 × 10-3
1.27 × 10-3
1.29 × 10-3
1.41 × 10-3
0.44 ((0.01)
0.43 ((0.01)
0.34 ((0.01)
0.39 ((0.01)
0.62 ((0.02)
a
Ref 6.
TABLE 3. Ra te Con sta n ts (kq) of Qu en ch in g of th e
4-(4′-Meth ylben zoyl)p h en yl (9) a n d
4-(4′-Br om oben zoyl)p h en yl (12) Ra d ica lsa
kq × 109, M-1 s-1
quencher
oxygen
1,4-cyclohexadiene
precursor BP
9
12
F IGURE 5. Traces at 320 nm obtained from the 355 nm laser
flash photolysis of 1 and 2; initial absorbance of both BPs at
355 nm ) 0.33.
1.34 ((0.03)
1.13 ((0.04)
1.59 ((0.10)
0.88 ((0.04)
1.14 ((0.06)
0.87 ((0.03)
TABLE 4. Lifetim es (τ) of Ben zyl-Typ e Ra d ica ls
(Solven t ) CCl4)
a
Solvent ) CCl4; concentration of precursor BPs was between
8.44 × 10-4 and 2.65 × 10-3 M.
precursor BP
radical
(concentration, M)
τ, µs
the range of 0.87-1.59 × 109 M-1 s-1. Similar values of
kq have been obtained for the 4-benzoylphenyl radical and
5.6 The rate constant of quenching of a photochemically
generated phenyl radical14 by the precursor benzoyl
peroxide has been found to be 2.1 × 107 M-1 s-1, and this
is lower than our values. H-abstraction by the aroyl-
pehnyl radicals from the precursor BPs seems to be the
likely reason for this.
In addition to a fast-decaying component, a slow-
decaying component was also observed at 320 nm from
LFP of all BPs except 4. In the case of 1 and 2, the slow-
decaying components showed identical lifetimes (τ ≈ 17
µs), indicating that they belong to an identical transient
that can be obtained from both 1 and 2. This ∼320 nm
absorption has been assigned to the corresponding ben-
zyl-type radicals [4-(4′-tert-butyldioxycarbonylbenzoyl)-
benzyl) 8, 3-(4′-tert-butyldioxycarbonylbenzoyl)benzyl 14,
and 3-(3′-tert-butyldioxycarbonylbenzoyl)benzyl 15] on
the basis of the following facts: (a) the benzyl radical has
a known absorption at 320 nm,15 (b) the half-life time of
a similar benzyl radical has been reported to be 10 µs,16
and (c) no slow-decaying component was observed at 320
nm from LFP of tert-butyl (4-benzoyl)perbenzoate and 4,
which do no have methyl or CH2-X groups.
8
14
15
1.77 × 10-3
1.51 × 10-3
2.29 × 10-3
16.96 ((1.07)
17.05 ((1.18)
18.13 ((2.25)
higher for 2 than for 1 (Figure 5). This is as expected,
since 8 forms directly due to cleavage of the C-Cl bond
in the case of 2, while in the case of 1, it forms when
different radicals abstract an H-atom from the methyl
group. Similar observations were made in the case of 5
and 6 where 14 forms directly due to scission of the C-Br
bond from 6.The lifetimes of 8, 14, and 15 have also been
found to be similar. The same resonance factor is avail-
able to stabilize these radicals. Each was quenched by
oxygen, as was the fast-decaying component at the 320
nm. The rate of quenching of these radicals could not be
obtained because of the interference of the fast-decaying
component. We also observed no measurable signal at
550 nm during LFP of either 6 or 7 where formation of
14 or 15, respectively, is competitive and predominates.
Investigation of the photopolymerization initiation activi-
ties of some of these BPs is underway.
Con clu sion s
In the photodecompostion of 1-5, the φdiss of the
perester bond was found to be the highest for 5. This
compound has the perester moiety positioned meta with
respect to the benzophenone carbonyl group. However,
neither the position of the methyl group nor the replace-
ment of the methyl group with a bromine atom showed
any significant effect on the φdiss of the perester bond as
reflected by similar values of the φdiss of 1-4 in benzene.
Single-crystal structures show that the carbonyl and
perester groups assume anti conformation in 2, while syn
confirmation is observed in 5. The carbonyl and perester
groups, as well as the aromatic ring separating them, are
almost planar in both 2 and 5. The presence of a benzylic
bromide function either at the para or meta position
significantly lowers the φdiss of the perester bond. This is
supported by LFP studies. The 355 nm LFP of 1-5
produced the corresponding aroylphenyl radicals (9-13,
The assignment is further supported in that in opti-
cally matched experiments, the optical density of the
slow-decaying component at 320 nm was found to be
(14) Scaiano, J . C.; Stewart, L. C. J . Am. Chem. Soc. 1983, 105, 3609.
(15) Meisel, D.; Das, P. K.; Hug, G. L.; Bhattacharyya, K.; Fes-
senden, R. W. J . Am. Chem. Soc. 1986, 108, 4706.
(16) Font-Sanchis, E.; Miranda, M. A.; Pe´rez-Prieto, J .; Scaiano, J .
C. J . Org. Chem. 2002, 67, 6131.
J . Org. Chem, Vol. 68, No. 22, 2003 8371

Shah and Neckers
for data integration, and corrections for absorption and decay
were carried out using the SADABS program.22 The X-ray
structures were determined by direct methods,23 and refine-
ment was done by full matrix least squares22 on F2 using all
3549 unique data for 2 and 3558 unique data for 5. The
refinement included anisotropic thermal parameters for non-
hydrogen atoms with isotropic thermal parameters for all
hydrogen atoms. In the case of 2, anomalously large thermal
parameters were observed as a result of two alternative
orientations for the C-atoms of the tert-butyl group. Similar
statistical disorder has been previously observed for the crystal
structure of 2,4,6-tri-tert-butylperbenzoate.11 Although these
data were collected at room temperature, such statistical
disorder was not observed in the case of 5. The final refinement
converged to wR2 ) 0.1592 (for F2, all data) and R1 ) 0.0642
[F, 2558 reflections with I > 2σ(I)] in the case of 2 and wR2 )
0.1142 (for F2, all data) and R1 ) 0.0516 [F, 2582 reflections
with I > 2σ(I)] in the case of 5.
La ser F la sh P h otolysis. Nanosecond flash photolysis
studies were performed using a kinetic spectrometric detection
system previously described.24 The excitation pulse (355 nm,
19 mJ /pulse) was the third harmonic of a Q-switched Nd:YAG
laser. The excitation pulse width was ∼7 ns. Transients
produced were followed temporally and spectrally by a com-
puter-controlled kinetic spectrophotometer. The sample solu-
tions showing an absorbance in the range of 0.15-0.35 at the
excitation wavelength in 1 cm2 quartz cuvettes were used and
degassed continuously with argon during experiments. Fresh
samples were used for obtaining each kinetic trace. Twice-
distilled CCl4 was used as the solvent for LFP. The kq values
were calculated by monitoring the lifetimes at 550 nm and
plotting 1/lifetime against the concentration of quenchers.
λmax 550 nm). The lifetimes of para radicals (9-12)
measured in 1.18-1.77 × 10-3 M solution of 1-4 in CCl4
have been found to be ∼0.4 µs, which is lower than the
lifetime of the meta radical (13 ∼0.62 µs, 1.41 × 10-3
M
5 in CCl4). The 550 nm absorption of the aroylphenyl
radicals was not observed from LFP of 6 and 7 in which
formation of benzyl-type radical (14 and 15) predomi-
nates. The lifetimes of the benzyl-type radicals (8, 14,
and 15, λmax ≈ 320 nm) in CCl4 have been found to be
∼17-18 µs.
Exp er im en ta l Section
Reagents and solvents were obtained from commercial
suppliers and used as received unless otherwise noted. Sol-
vents used for syntheses, e.g., benzene, toluene, and cyclohex-
ane were dried over sodium/benzophenone under argon before
use. Column chromatography and thin-layer chromatography
(TLC) were performed using standard-grade silica gel (32-63
µm, 60 Å) and silica gel plates (200 µm). Melting points are
uncorrected. Nuclear magnetic resonance (NMR) spectra were
recorded on 200 and 400 MHz spectrometers. CDCl3 was the
solvent for NMR unless otherwise noted, and chemical shifts
are reported in parts per million (ppm) for 1H NMR on the δ
scale relative to TMS at 0.0 ppm. GC/MS data were collected
using spectrometers having 30 m × 0.25 mm × 0.25 µm
1
columns. The φdiss values were determined by H NMR using
C6D6 as the solvent and hexamethylsiloxane as the internal
standard. The detailed procedure has been reported previ-
ously.7 Product analyses were done using GC/MS as described
previously.7
Syn th esis. The synthesis of 1 and 2 has been reported
earlier.7 3-7 have been synthesized following the general
method of synthesis of peresters from their corresponding
aroylbenzoic acids (16-19).17-20 A detail of synthesis and
purification methods of BPs 3-7 and their spectroscopic data,
as well as synthesis of 16-19, has been provided in Supporting
Information.
X-r a y Cr ysta llogr a p h y. Data collection was performed at
room temperature with Mo KR using a Bruker AXS SMART
platform diffractometer.21 Intensity data were collected using
three different φ settings and 0.3° increment ω scans, 2θ <
56.58° for 2 and 2θ < 56.56° for 5, which corresponds to more
than a hemisphere of data. The SAINT21 program was used
Ack n ow led gm en t. B.K.S. would like to thank Mc-
Master Endowment for a fellowship. Help from Dr. K.
Kirschbaum and Mr. B. Fneich during the X-ray mea-
surement is also greatly acknowledged. We thank the
National Science Foundation Division of Material Re-
search (DMR 9803006) for financial support of this
work.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails of synthesis and purification and spectral data of BPs
3-7, synthetic procedures of 16-19, 1H and 13C NMR of 3-7,
UV-vis spectra of 1-7, X-ray crystal structure data of 2 and
5, and various kinetic traces and fittings for lifetimes and rate
constants. This material is available free of charge via the
Internet at http://pubs.acs.org.
(17) Vogel, A. I. Practical Organic Chemistry; Longman: New York,
1978; p 973.
J O030168D
(18) Haddach, M.; McCarthy, J . R. Tetrahedron Lett. 1999, 40, 3109.
(19) Parham, W. E.; Sayed, Y. A. J . Org. Chem. 1974, 39, 2053.
(20) Murakami, Y.; Hara, H.; Okada, T.; Hashizume, H.; Kii, M.;
Ishihara, Y.; Ishikawa, M.; Shimamura, M.; Mihara, S.; Kato, G.;
Hanasaki, K.; Hagishita, S.; Fujimoto, M. J . Med. Chem. 1999, 42,
2621.
(22) Sheldrick, G. M. SADABS, Program for the Empirical Absorp-
tion Correction of Area Detector Data; University of Go¨ttingen: Go¨t-
tingen, Germany, 1996.
(23) Sheldrick, G. M. SHELXTL (ver. 6.12); Bruker AXS Inc.:
Madison, WI, 2000.
(21) Bruker SMART (Ver. 5.05) and SAINT-Plus (Ver. 7.08); Bruker
AXS Inc.: Madison, WI, 1999.
(24) Ford, W. E.; Rodgers, M. A. J . J . Phys. Chem. 1994, 98, 3822.
8372 J . Org. Chem., Vol. 68, No. 22, 2003