Polymerizations initiated by diradicals from cycloaromatization reactions

Article abstract of DOI:10.1021/ma050750z

Four cycloaromatization substrates each produce diradicals that lead to the initiation of polymerization of vinyl monomers. All the initiators produce significant amounts of polymer, especially with methacrylate monomers. Intramolecular termination of short diradical chains produces oligomeric byproducts and limits the amount of high polymer that is formed. The polymer yield can be increased through the addition of a chain transfer agent by presumably converting unproductive diradicals into pairs of monoradicals. The enediynes that contain terminal acetylenes are less effective initiators because they retard radical polymerization. Bergman cyclization substrates are better initiators than Myers cyclization substrates.

Full text of DOI:10.1021/ma050750z

7266
Macromolecules 2005, 38, 7266-7273
Polymerizations Initiated by Diradicals from Cycloaromatization
Reactions
Joseph D. Rule and Jeffrey S. Moore*
The Beckman Institute for Advanced Science and Technology and Department of Chemistry,
University of Illinois, Urbana, Illinois 61801
Received April 11, 2005; Revised Manuscript Received June 21, 2005
ABSTRACT: Four cycloaromatization substrates each produce diradicals that lead to the initiation of
polymerization of vinyl monomers. All the initiators produce significant amounts of polymer, especially
with methacrylate monomers. Intramolecular termination of short diradical chains produces oligomeric
byproducts and limits the amount of high polymer that is formed. The polymer yield can be increased
through the addition of a chain transfer agent by presumably converting unproductive diradicals into
pairs of monoradicals. The enediynes that contain terminal acetylenes are less effective initiators because
they retard radical polymerization. Bergman cyclization substrates are better initiators than Myers
cyclization substrates.
Introduction
However, the behavior of these polymerizations was
unusual.24 The polymer yields depended heavily on the
type of monomer, with methacrylates producing signifi-
cantly more polymer than styrene, methyl acrylate, or
other monomers. In addition, the polymer had a bimodal
molecular weight distribution. Mass spectrometry showed
that the shorter-chain fraction was consistent with
intramolecular termination of diradical intermediates
while mechanistic studies showed that the high molec-
ular weight fraction was formed from monoradical chain
growth. The mechanism that is most consistent with
these observations involves two competing reaction
routes for the diradicals formed from the initiator.
In the first route, the diradicals can add to monomer
units until the chain ends are sufficiently mobile to
interact with each other and intramolecularly termi-
nate. This route is the source of the low molecular
weight fraction. The kinetics of this intramolecular
termination have long been predicted to be extremely
fast relative to polymer growth from both radicals.5,6
Soon after the first reports of radical polymerization,
Flory promoted the idea of growing two polymer chains
from a single diradical molecule.1 At that time, the
thermal initiation of styrene was presumed to occur
through a diradical intermediate, and this hypothesis
led to significant study and debate regarding diradical-
initiated polymerizations.1-6 Eventually, it was gener-
ally accepted that thermal initiation of styrene results
primarily from monoradicals,7 but the role of diradicals
in styrene’s autoinitiation is still under research.8 Other
diradical-initiated polymerizations have also been ac-
tively studied over the past half century. However, the
diradicals previously studied including those from cyclic
azo compounds,9-11 tetraoxanes,12 and electronically
destabilized cyclopropanes11,13,14 have all failed to pro-
duce significant amounts of polymer.
A more recently developed source of diradicals is the
Bergman cyclization of enediynes to produce dide-
hydroarenes (eq 1).
In the second route, one of the two radicals inter-
molecularly abstracts a hydrogen atom, and two mono-
radical molecules are formed. These monoradicals can
then diffuse apart and produce high molecular weight
polymer as in typical radical polymerizations.27 This
mechanism predicts that additives which favor hydro-
gen atom abstraction should produce more monoradical
molecules and, therefore, more high polymer. This
prediction was confirmed in polymerizations initiated
by 1 in which the addition of butanethiol as a chain
transfer agent produced dramatically higher yields of
polymer than analogous polymerizations without the
thiol.24 Monomers that tend to favor chain transfer also
produce good polymer yields.
Here we report a systematic comparison of polymer-
izations initiated by 1 and three other enediyne mol-
ecules with significant structural differences and two
distinct cycloaromatization reactions. Through these
studies, we demonstrate that the behavior we originally
reported for 1 is general for several related diradical
sources and for various monomers. In addition, we
investigate some of the structural features of these
initiators that favor polymerization.
While this reaction has received significant attention
in synthetic and biochemical applications,15,16 its use in
materials has been limited to homopolymerizations of
enediynes to produce polyarenes or similar polymers.17-23
Until our recent communication,24 polymer initiation of
vinyl monomers through Bergman cyclization appears
only in the patent literature.25 In contrast to the
negative precedents in diradical-initiated polymeriza-
tions, our work with 3,4-benzocyclodec-3-ene-1,5-diyne
(1)26 demonstrated that diradicals formed from the
Bergman cyclization of this enediyne could initiate
significant radical polymerization.24
10.1021/ma050750z CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/23/2005

Macromolecules, Vol. 38, No. 17, 2005
Polymerization of Vinyl Monomers 7267
Scheme 1
Scheme 2
The enediynes chosen for these experiments were
constrained to cyclize at moderate temperatures (<160
°C) but also to be stable at room temperature. Enediynes
with internal acetylene groups tend to cyclize only at
very high temperatures, so they are not sufficiently
reactive.28 However, enediynes with terminal acetylene
groups cyclize in the appropriate temperature range.
1,2-Diethynylbenzene (2), which cyclizes with a half-
life of about 1 h at 152 °C,28 was chosen to represent
the class of benzannulated enediynes having terminal
acetylenes. (Z)-2,3-Diethylhexa-l,5-diyn-3-ene (3) was
chosen to represent the class of olefinic enediynes
having terminal acetylenes.29 While the kinetics of
cyclization for this compound are not reported, a close
structural analogue cyclizes with a half-life of about 2
h at 132 °C.29
reaction was the expected naphthalene derivative 5
(Scheme 1).
At this point, all four of the enediynes in this study
had been confirmed to undergo cyclizations that produce
diradicals, but to meaningfully compare their behavior
as initiators, their differences in reactivity had to be
considered. Therefore, in situ 1H NMR trials were used
to measure the cyclization kinetics at various temper-
atures and to determine the temperatures at which each
substrate cyclizes with a half-life of 16-20 h (Figure
1). The desired rates of Bergman cyclization for 1, 2,
and 3 were achieved at 100, 120, and 104 °C, respec-
tively. With the addition of 5 equiv of triethylamine,
enediyne 4 was found to rearrange with the desired rate
and undergo Myers cyclization at only 32 °C. By
performing the comparative experiments at these speci-
fied temperatures, the rates of radical production will
be approximately constant for all the tested enediynes.
Thus, any differences in behavior observed between the
initiators cannot be attributed to differences in cycliza-
tion rate.
To probe the effects of alternative diradical geometry,
a substrate for Myers cyclization was chosen. Myers
cyclization is the rearrangement of an eneyneallene to
form an aromatic ring with one radical on the ring and
another radical in a benzyl position (eq 2).
The ability of the enediynes to initiate polymeriza-
tions was investigated by heating them in neat mono-
mer at their specified temperatures (Table 1). Polymer-
izations of styrene and methyl acrylate were not
performed with initiator 2 because the high temperature
required (120 °C) produces excessive amounts of ther-
mally initiated polymer in control reactions. The poly-
mer yields achieved in this system varied greatly with
the monomers that were employed, and specifically, the
methacrylates gave the best conversions (vide infra).
The lower activation energy of these cyclizations relative
to Bergman cyclizations could lead to initiation of
polymerization at much lower temperatures.16 However,
the higher reactivity of eneyneallenes also leads to
instability that makes them less convenient to work
with. To avoid this problem, a stable enediyne (4) with
a propargylic sulfone group was used as an eneyneallene
precursor. This class of compounds is known to re-
arrange to eneyneallenes under basic conditions.30 Thus,
we surmised that with the addition of mild base (e.g.,
amine) the eneyneallene would form, which would then
rapidly undergo Myers cyclization to produce a diradical
intermediate (Scheme 1).
Results and Discussion
Enediyne 4 was synthesized through the route estab-
lished by Grissom and co-workers for similar compounds
(Scheme 2).30,31 DSC traces of 4 with and without added
triethylamine confirmed that it is stable to more than
200 °C as the enediyne, but with the addition of the
amine it reacts rapidly at temperatures above 60 °C.
To confirm that this reaction was the desired cyclo-
aromatization, enediyne 4 was mixed with 5 equiv of
triethylamine and an excess of 1,4-cyclohexadiene (as
a hydrogen atom donor). The primary product of this
Figure 1. Kinetics of disappearance of the enediynes as
1
measured by H NMR in d8-toluene with 1,4-cyclohexadiene
(1 M). In the case of 4, 5 equiv of triethylamine was added.

7268 Rule and Moore
Macromolecules, Vol. 38, No. 17, 2005
Table 1. Polymerizations Initiated by Enediynesa
monomer
initiator
[initiator] (mM)
T (°C)
polymer yield (%)
Mn
PDI
butyl methacrylate
1
22
23
25
22
0
27
29
31
23
33
36
0
21
24
28
0
24
24
22
0
30
26
26
32
0
26
34
34
25
0
100
120
104
32
100
100
120
104
32
93
16
3
714 000
933 000
338 000
805 000
1 129 000
482 000
760 000
519 000
499 000
328 000
284 000
1 589 000
626 000
228 000
607 000
3 441 000
265 000
233 000
101 000
312 000
1.38
2.47
1.59
2.10
1.21
2.52
2.82
1.73
2.29
2.18
2.78
2.07
1.89
1.58
4.48
2.22
1.84
1.49
7.61
2.35
2
3
4
5
none
2
87
5
2
4
36
45
2
methyl methacrylate
1
2
3
4
4
57
82
4
none
100
100
104
32
100
100
104
32
100
100
120
104
32
methyl acrylate
styrene
1
29
1
2
3
4
none
18
12
5
1
3
4
1
none
5
acrylonitrile
1
3
2
1
3
1
4
1
none
100
100
120
104
32
b
vinylidene chloride
1
1
2
1
3
1
4
1
none
100
b
a 2.5 h in neat monomer. b Trace.
this molecular weight profile is consistent with intra-
molecular termination of diradical intermediates (Table
2).
To better understand this intramolecular termination
process, individual compounds were isolated from simi-
lar oligomeric fractions and characterized. We previ-
ously reported structures 6 and 7 that were obtained
from vinylidene chloride and 1.24 A similar experiment
with acrylonitrile produced compound 8 (Figure 3) as
well as at least eight other trimeric isomers that were
not separable. All of these structures are consistent with
diradical intermediates that intramolecularly terminate.
This intramolecular termination limits the production
of high molecular weight polymer. However, if the
diradical intermediates undergo chain transfer to an-
other molecule before they intramolecularly terminate,
then two monoradical molecules are formed which each
have the potential to produce high polymer. Therefore,
the addition of chain transfer agents to these reactions
should facilitate monoradical formation and lead to
higher yields of polymer. This hypothesis was tested by
adding butanethiol to MMA polymerizations initiated
by each enediyne. In all cases, the addition of butane-
thiol increases the polymer yields significantly (Figure
4). In a control case without enediynes, butanethiol
increases the polymerization rate only slightly. While
the small increases seen with the control are similar to
those seen with 4, the data for 4 were taken at a much
lower temperature than the control, and a comparison
between these two series should not be made.
Figure 2. GPC traces from MMA polymerizations initiated
by each enediyne.
Also, compared to the other enediynes, initiator 1
produces significantly more polymer with all tested
monomers. These trials, particularly in the case of the
methacrylates, demonstrate generally that the cyclo-
aromatizations can be used to produce high polymer.
The previously observed intramolecular cyclization
behavior was then investigated for each substrate.24 As
seen in Figure 2, each poly(methyl methacrylate)
(PMMA) sample produced from the enediynes shows a
bimodal distribution. The low molecular weight fractions
are consistent with the theoretical kinetic analyses that
suggest that intramolecular termination is the most
likely result for polymer chains having radicals at both
ends.5,6 Because the chains are not predicted to grow
extensively before intramolecular termination, observ-
able amounts of oligomeric material are expected from
this termination.
To confirm that this enhancement is general for
monomers other then MMA, enediyne 1 was used to
initiate a similar series of polymerizations with different
monomers. The results (Figure 5) show that higher
concentrations of butanethiol lead to faster conversion
of monomer in all cases. The chain transfer agent’s
dramatic effect on the rate of polymerization in this
To confirm this interpretation, the oligomeric frac-
tions in each case were isolated. Field desorption mass
spectrometry revealed that these products are chains
of initiator molecules and monomer units with no
additional chain end groups (i.e., hydrogen atoms), and

Macromolecules, Vol. 38, No. 17, 2005
Table 2. FDMS Data for MMA Oligomers Formed through Intramolecular Termination
enediyne 1 enediyne 2 enediyne 3 enediyne 4
Polymerization of Vinyl Monomers 7269
monomer
units
FWa
(g/mol)
relative
abundance (%)
FWa
(g/mol)
relative
abundance (%)
FWa
(g/mol)
relative
abundance (%)
FWa
(g/mol)
relative
abundance (%)
1
280
380
480
580
360
460
560
<2
10
100
10
<2
<2
<2
226
326
426
526
252
352
452
<2
2
100
38
<2
<2
3
232
332
432
532
264
364
464
6
7
100
14
4
15
11
422
522
622
722
644
744
844
14
100
2
<2
<2
<2
<2
2
3
4
1b
2b
3b
a Molecular weights of one initiator molecule and the shown number of monomer units. b These peaks correspond to two enediyne
molecules. It is assumed that one of the endiynes acts as monomer, and it is counted as one of the listed number of monomer units.
dependence in these reactions (Table 1). The key step
required for polymer growth is the conversion of diradi-
cals into pairs of monoradicals. Therefore, monomers
that most favor chain transfer are expected to be those
that produce the most polymer. Unfortunately, the
transfer constants to monomer (CM) cannot be used to
make good predictions in this system because they do
not account for the highly reactive radicals that are
initially formed on the aromatic rings. If it is reasonably
assumed that the didehydroarenes behave like phenyl
radicals,15 then hydrogen atom abstraction is more
likely to occur with the aromatic radicals than with the
monomer-derived radicals.32 Unfortunately, the relative
abilities of aromatic diradicals to abstract hydrogen
atoms from different monomers is not known. However,
because of the formation of a stabilized allylic radical,
it seems reasonable that hydrogen atom abstraction
from methacrylates would be more favorable than
abstraction from monomers without an R-methyl group.
This claim is supported by the report that tert-butoxy
radicals can abstract hydrogen atoms from R-methyl
groups of methacrylates with the ratio of abstraction
to radical addition being as high as 0.63.33 Although the
radicals resulting from cycloaromatizations are signifi-
cantly different from tert-butoxy radicals, similar ab-
straction of the relatively labile R-methyl hydrogen
atoms is not unreasonable. The facility of this hydrogen
atom abstraction from the R-methyl group is the most
likely cause of the high polymer yields achieved with
the methacrylates.34
While the enediynes in this study share the general
behaviors that were just discussed, they also show
significant differences. The most obvious of these dif-
ferences is the efficiency with which they initiate
polymerization. As seen in Table 1, enediyne 1 leads to
much higher polymer yields than the other enediynes.
Apparently, 2 and 3 are poor initiators relative to 1
because of their tendency to retard polymerizations.
Phenylacetylene retards MMA and styrene polymeriza-
tions as it copolymerizes with them.35,36 Because of the
structural similarity between phenylacetylene and ene-
diynes having terminal acetylenes (i.e., 2 and 3), it
seemed likely that they would behave similarly. To
measure possible retardation, MMA polymerizations in
the presence of enediynes and AIBN were monitored at
60 °C using in situ 1H NMR (Figure 6). At this
temperature, none of the enediynes cyclize appreciably,
so they are not expected to contribute to initiation.
Therefore, the polymer is initiated only by the radicals
produced from AIBN, and the effects of the enediynes
are limited to side reactions. By comparing the rates in
the presence of the enediynes with the control case with
only AIBN, the retardative effects could be observed
(Table 3).
Figure 3. ORTEP drawing of oligomeric byproduct 8 resulting
from the reaction of enediyne 1 with acrylonitrile.
Figure 4. Effect of butanethiol on polymerizations with
enediynes (6 mM) in neat MMA at 100 °C for 1 h. In the case
of 4, 5 equiv of triethylamine was added relative to the
enediyne.
Figure 5. Effect of butanethiol on polymerizations with 1 (6
mM) in neat monomer at 100 °C for 1 h.
system supports the theory that the high polymer
results from unproductive diradicals being converted to
pairs of monoradicals via chain transfer.
The general mechanism of these diradical-initiated
polymerizations may explain the dramatic monomer

7270 Rule and Moore
Macromolecules, Vol. 38, No. 17, 2005
Scheme 3
Scheme 4
Figure 6. Copolymerization of MMA and enediynes initiated
1
by AIBN (190 mM) at 60 °C monitored by in situ H NMR.
Table 3. MMA Polymerization Rates in the Presence of
Enediynesa
enediyne
relative Rp
none
1.00
0.88
0.38
0.03
1.00
1.08
1
2
3
4
radical concentration is high, but many of the radicals
do not initiate polymerization. Thus, the relatively
rapidly forming diradicals from 4 may be undergoing
significantly more intramolecular termination than
those formed from 1.
1,2-dipent-1-ynylbenzene
a Performed at 60 °C in d6-benzene. [Enediyne] ) 75 mM,
[MMA] ) 720 mM, [AIBN] ) 190 mM.
This hypothesis is supported by the mass spectrom-
etry data on the isolated oligomeric MMA (Table 2). The
Bergman cyclization substrates (1, 2, and 3) produce
oligomers mostly composed of an initiator molecule with
three added monomer units. In contrast, the Myers
precursor 4 primarily produced initiator molecules with
only one or two added monomer units. The difference
in behavior may result from the different geometries of
the respective diradicals. This argument is best il-
lustrated through the oligomeric products 6 and 7 which
were both likely formed through a common intermediate
(9) having a benzylic radical (Scheme 3). Once this
benzylic radical is formed, intramolecular termination
appears to be very favorable. Because the eneyneallene
derived from 4 forms a benzylic radical (10) as it
cyclizes, it requires only one monomer addition step to
reach the intermediate 11 (Scheme 4). Diradical 11 is
closely analogous to diradical 9, so it would similarly
be expected to strongly favor intramolecular termina-
tion. This facilitated route to intramolecular termination
limits the opportunities for 10 to undergo intermolecular
hydrogen atom abstraction, so 4 results in fewer pro-
ductive pairs of monoradicals and less polymer.
Clearly, the enediynes with terminal acetylenes (2
and 3) significantly retard the polymerizations, while
1, 4, and 1,2-dipent-1-ynylbenzene show no variation
outside of experimental error. This suggests that the
terminal alkynes are susceptible to attack by MMA-
derived radicals while the internal alkynes are not. If
this is the case, then 2 and 3 will be consumed during
the polymerizations while the other enediynes will not
be affected. By monitoring the concentrations of the
enediyne during the experiments above, this hypothesis
was confirmed. Only the enediynes with terminal
acetylenes are consumed significantly. Additional evi-
dence for the behavior of 3 as a comonomer is seen in
the molecular weights of the MMA oligomers that were
formed through intramolecular termination (Table 2).
In the case of 3, some of these oligomers have molecular
weights consistent with a diradical that added to
another initiator molecule as well as monomer before
terminating. These kinetics measurements lead to the
overall conclusion that 2 and 3 (and, in all likelihood,
other enediynes bearing terminal acetylenes) are poor
initiators of radical polymerization because of their
tendencies to copolymerize and act as retarders.
The difference in polymer production between 1 and
4 is not explained by the copolymerization and retarda-
tion that complicate the use of 2 and 3, so another
significant effect must be involved. To confirm that this
effect is not due to the large difference in the temper-
atures at which 1 and 4 were tested, enediyne 4 was
used to initiate polymerizations at 57 and 82 °C (Table
1). Thus, a comparison between the two initiators can
be made with less dependence on temperature. Of
course, by increasing the temperatures above 32 °C, the
cyclization of 4 is expected to be much faster than that
of 1 at 100 °C, so the rates of radical formation are no
longer equivalent. The higher rate of radical production
with 4 is reflected in the lower molecular weights of its
products. However, the yields achieved with 4 were still
considerably lower than those with 1. The lower mo-
lecular weight achieved with 4 despite its slower
polymerization is consistent with a system where the
Conclusions
While the Bergman and Myers cyclization products
can initiate radical polymerization, these reactions are
complicated by the diradical nature of the initiators. In
particular, the diradicals tend to terminate intra-
molecularly to produce oligomeric byproducts. By favor-
ing hydrogen atom abstraction events that convert the
diradicals into pairs of monoradicals, this intramolecu-
lar termination can be avoided and high polymer can
be produced.
In comparing various enediynes, it was found that
those with terminal acetylenes tend to retard radical
polymerizations. Also, the diradicals produced through
Myers cyclization have a geometry that seems to favor
intramolecular termination more than those produced
by Bergman cyclization. Therefore, Bergman cyclization

Macromolecules, Vol. 38, No. 17, 2005
Polymerization of Vinyl Monomers 7271
sidual initiator was removed through silica gel chromatogra-
phy. The resulting oligomeric mixtures were analyzed by
FDMS.
of enediynes having internal alkynes appears to be the
most favorable cycloaromatization reaction for polymer
initiation.
Representative Oligomerization of MMA Initiated by
1. Enediyne 1 (0.040 g, 0.22 mmol) was weighed into an oven-
dried Schlenk tube and placed under a nitrogen atmosphere.
Chlorobenzene (18 mL) and methyl methacrylate (1.88 g, 18.8
mmol) were added via syringe. The solution was degassed with
three consecutive freeze-pump-thaw cycles and placed in a
100 °C bath for 5 h. Nearly all the solvent and residual
monomer were then removed under vacuum. The resulting
polymer/oligomer mixture was analyzed by GPC. The product
mixture was then precipitated into MeOH from a THF
solution, and the PMMA was removed by filtration and dried
Experimental Section
Methyl methacrylate, butyl methacrylate, methyl acrylate,
styrene, acrylonitrile, methacrylonitrile, and vinylidene chlo-
ride (all from Aldrich) were dried over CaH2 and distilled prior
to use. Triethylamine, methylene chloride, benzene, and
chlorobenzene (all from Fisher) were similarly dried over CaH2
and distilled. AIBN was recrystallized from MeOH. Pd(PPh3)2Cl2
(Aldrich, 98%), CuI (Aldrich, 98%), 1-pentyne (Aldrich, 99%),
methanesulfonyl chloride (Aldrich, 99.7%), thiophenol (Aldrich,
99%), m-chloroperoxybenzoic acid (Aldrich, ∼70%), 1,4-cyclo-
hexadiene (Aldrich, 97%), and anthracene (97%) were used as
received. Anhydrous toluene (Aldrich, 99.8%) and anhydrous
THF (Aldrich, 99.9%) were used as received. The other solvents
were used as received from Fisher without purification or
rigorous exclusion of moisture.
Gel permeation chromatography (GPC) measurements were
performed in THF at 25 °C with a Waters 515 HPLC pump, a
Viscotek TDA model 300 triple detector, and a series of three
Viscogel 7.8 × 300 mm columns (2 × GMHXL16141 and 1 ×
G3000HXL16136). Molecular weight data were determined
using Viscotek’s TriSEC software. The light scattering, mass,
and viscosity constants were determined from a single 96 kDa
narrow polystyrene standard and checked against other known
polystyrene standards for accuracy. The column exclusion limit
was 1.0 × 107 Da, and the flow rate was 1.0 mL/min. 1H NMR
spectra were taken on a Varian Unity 500 instrument. Mass
spectrometry was performed on a Micromass 70-VSE instru-
ment in FD mode.
General Polymerization Procedure. The enediyne (0.022
mmol) was weighed into an oven-dried Schlenk tube and
placed under a nitrogen atmosphere. Monomer (1 mL) was
then injected into the tube, and the solution was degassed
through three consecutive freeze-pump-thaw cycles. To ac-
count for any monomer lost during the degassing cycles, the
tube was weighed and the amount of monomer remaining was
calculated by comparison with the weight of the empty tube.
The solution was placed in an oil bath at the appropriate
temperature for 2.5 h. In some cases, monomer was heated
above its boiling point in these sealed systems, and extreme
caution must be used due to the pressure that develops. The
resultant poly(methyl methacrylate), poly(butyl methacrylate),
and polystyrene samples were dissolved in THF and precipi-
tated into MeOH. Poly(methyl acrylate) samples were dis-
solved in THF and precipitated into petroleum ether. Poly-
acrylonitrile and poly(vinylidene chloride) were isolated by
removing residual monomer under vacuum.
Representative Polymerization of MMA Initiated by
1. Enediyne 1 (3.9 mg, 0.022 mmol) was weighed into an oven-
dried Schlenk tube and placed under a nitrogen atmosphere.
Methyl methacrylate (1.0 mL) was then injected into the tube,
and the solution was degassed through three consecutive
freeze-pump-thaw cycles. To account for any MMA lost
during the degassing cycles, the tube was weighed and the
amount of monomer remaining (0.764 g, 7.6 mmol) was
calculated by comparison with the weight of the empty tube.
The solution was placed in an oil bath at 100 °C for 2.5 h. The
resultant PMMA was dissolved in THF and precipitated into
MeOH. The precipitate was collected by filtration and dried
under vacuum to give a white polymer (0.668 g, 87%). GPC
(g/mol): Mn ) 482 000, Mw ) 1 215 000.
to give a white polymer (0.49 g, 26%). GPC (g/mol): Mn
)
46 000; Mn ) 78 000. The filtrate was concentrated under
vacuum to yield crude oligomer (0.050 g). Residual initiator
(0.011 mg, 0.06 mmol, 28%) was removed by column chroma-
tography using a 1:4 mixture of methylene chloride and
petroleum ether, respectively. The remaining oligomer was
forced to elute from the column by using a 1:7:32 mixture of
methanol, methylene chloride, and petroleum ether, respec-
tively. A mixture of products was obtained. FDMS (m/z): 380
(10%), 480 (100%), 580 (10%).
Oligoacrylonitrile Formation (8). Enediyne 1 (98 mg,
0.55 mmol) was weighed into an oven-dried 100 mL Schlenk
tube and placed under an Ar atmosphere. Acrylonitrile (10 mL,
0.152 mol) was added via syringe. The solution was degassed
with three consecutive freeze-pump-thaw cycles and placed
in a 100 °C bath for 20 h. DMF was added to disperse the
polymer, which was then precipitated into MeOH and filtered
to give 1.49 g of polymer (19%). The filtrate was evaporated
to leave 189 mg of material that was separated by column
chromatography (1:4:5 EtOAc/petroleum ether/ methylene
chloride). 69 mg of trimeric products (FDMS m/z ) 339) was
isolated, but only one product (8) was isolated with sufficient
purity for characterization. 12 mg, 7% (relative to 1). 1H NMR
(500 MHz, CD2Cl2, δ): 7.54 (1 H, dd, J ) 8, 1.1 Hz), 7.45 (1 H,
ddd, J ) 7.5, 7.5, 1.4 Hz), 7.40 (1 H, ddd, J ) 7.5, 7.5, 1.3 Hz),
7.22 (1 H, dd, J ) 7.7, 1.3 Hz), 3.49 (1 H, dddd, J ) 11.6, 11.6,
7.1, 7.1 Hz), 3.19 (1 H, dd, J ) 15.4, 11.6 Hz), 3.11 (1 H, d, J
) 6.7 Hz), 2.75 (1 H, dd, J ) 15.4, 6,6 Hz), 2.48 (1 H, dd, J )
13.4, 7.6 Hz), 2.46 (1 H, m), 2.26 (1 H, dd, J ) 6.7, 6.7 Hz),
2.24 (1 H, m), 2.18 (1 H, ddd, J ) 13.5, 11.5, 6.9 Hz), 2.09 (1
H, dd, J ) 9.5, 7.2 Hz), 1.93 (1 H, m), 1.85 (1 H, dd, J ) 9.5,
6.2 Hz), 1.81 (1 H, m), 1.71 (2 H, m), 1.69 (1 H, m), 1.60 (1 H,
m). 13C NMR (125 MHz, CD2Cl2, δ): 143.98, 138.22, 133.92,
132.49, 127.85, 127.55, 127.39, 125.45, 124.32, 123.65, 121.17,
119.27, 119.00, 41.81, 33.94, 33.80, 33.71, 29.01, 26.65, 26.29,
22.73, 22.04, 14.89.
Preparation of 3-(2-Pent-1-ynylphenyl)prop-2-yn-1-ol.
In a 50 mL Schlenk tube, Pd(PPh3)2Cl2 (98 mg, 0.14 mmol)
and CuI (89 mg, 0.47 mmol) were dissolved in toluene (40 mL).
3-(2-Iodophenyl)-2-propyn-1-ol31 (1.20 g, 4.65 mmol) and 1-pen-
tyne (5.00 g, 73.4 mmol) were then added via syringe. The
solution was degassed under vacuum, and triethylamine (2
mL, 14 mmol) was added. After stirring for 18 h, ethyl ether
was added and the mixture was filtered. The solvent was
removed and column chromatography with a 1:9 mixture of
ethyl acetate and petroleum ether resulted in 600 mg (65%)
of 3-(2-pent-1-ynylphenyl)prop-2-yn-1-ol as a dark orange oil.
TLC Rf 0.45 (1:3 ethyl acetate/petroleum ether). FTIR (neat):
3347 (br), 3061, 2963, 2933, 2871, 2231, 1481, 1443, 1025, 758
cm-1 1H NMR (500 MHz, CDCl3, δ): 7.40 (m, 2H, ar), 7.22
.
(m, 2H, ar), 4.53 (s, 2H, -CH2-OH), 2.45 (t, J ) 6.9 Hz, 2H,
-CH2-CH2-), 1.76 (br s, 1H, -OH), 1.65 (m, 2H, -CH2-CH3),
1.08 (t, J ) 7.3 Hz, 3H, -CH3). 13C NMR (500 MHz, CDCl3,
δ): 132.2 (ar), 132.1 (ar), 128.4 (ar), 127.4 (ar), 126.8 (q, ar),
125.0 (q, ar), 94.9 (≡C-CH2-), 90.8 (≡C-CH2-), 84.9 (≡C-
C-), 79.6 (≡C-C-), 52.0 (-CH2-OH), 22.3 (-CH2-), 21.8
(-CH2-), 13.7 (-CH3). HRMS-EI (m/z): M+ calcd for C14H14O,
198.1045; found, 198.1047.
General Oligomerization Procedure. The appropriate
enediyne (0.22 mmol) was weighed into an oven-dried Schlenk
tube and placed under a nitrogen atmosphere. Chlorobenzene
(18 mL) and methyl methacrylate (1.88 g, 18.8 mmol) were
added via syringe. The solution was degassed with three
consecutive freeze-pump-thaw cycles and placed in a 100 °C
bath for 5 h. Nearly all the solvent and residual monomer were
removed under vacuum. The product mixture was then dis-
solved in THF and precipitated into MeOH to remove the
polymer. The filtrate was concentrated under vacuum. Re-
Preparation of Methanesulfonic Acid 3-(2-Pent-1-
ynylphenyl)prop-2-ynyl Ester. Methanesulfonyl chloride
(410 mg, 3.6 mmol) was added dropwise to a solution of 3-(2-

7272 Rule and Moore
Macromolecules, Vol. 38, No. 17, 2005
pent-1-ynylphenyl)prop-2-yn-1-ol (550 mg, 2.77 mmol) and
triethylamine (390 mg, 3.9 mmol) in methylene chloride (14
mL) at 0 °C. After stirring for 1 h at 0 °C, filtration through a
silica plug with a 1:1 mixture of ethyl acetate and petroleum
ether gave 730 mg (95%) of the desired mesylate as a dark
orange oil. FTIR (neat): 2964, 2936, 2873, 2231, 1482, 1444,
1365, 1177, 993, 974, 936, 803, 761 cm-1. 1H NMR (500 MHz,
CDCl3, δ): 7.42 (m, 2H, ar), 7.29 (m, 1H, ar), 7.23 (m, 1H, ar),
5.12 (s, 2H, -CH2-O-), 3.19 (s, 3H, -S-CH3), 2.44 (t, J )
7.0 Hz, 2H, -CH2-CH2-), 1.65 (m, 2H, -CH2-CH3), 1.08 (t,
J ) 7.3 Hz, 3H, -CH3). 13C NMR (500 MHz, CDCl3, δ): 132.5
(ar), 132.3 (ar), 129.4 (ar), 127.6 (ar), 127.4 (q, ar), 123.6 (q,
ar), 95.6 (≡C-CH2-), 88.7 (≡C-CH2-), 84.1 (≡C-C-), 79.3
(≡C-C-), 58.8 (-CH2-O-), 39.5 (-S-CH3), 22.3 (-CH2-),
21.7 (-CH2-), 13.7 (-CH3). HRMS-EI (m/z): M+ calcd for
C14H14O, 276.0820; found, 276.0827.
CH2-), 1.58 (m, 2H, -CH2-CH3), 0.94 (t, J ) 7.4 Hz, 3H,
-CH3). 13C NMR (500 MHz, CDCl3, δ): 139.6 (q, ar), 138.3 (q,
ar), 134.0 (ar), 133.7 (q, ar), 132.1 (ar), 131.7 (q, ar), 129.2 (2C,
ar), 129.0 (2C, ar), 127.9 (ar), 127.8 (ar), 127.3 (ar), 126.9 (ar),
125.9 (ar), 125.0 (q, ar), 59.9 (-CH2-S-), 34.8 (-CH2-), 23.8
(-CH2-), 14.2 (-CH3). HRMS-EI (m/z): M+ calcd for C14H14O,
324.1184; found, 324.1180.
Cyclization Kinetics Measurements. In a nitrogen-filled
glovebox, the appropriate enediyne (0.079 mmol) and an-
thracene (5 mg, 0.028 mmol) were dissolved in d8-toluene (0.71
mL). The tube was capped with a septum and removed from
the glovebox. 1,4-Cylohexadiene (0.075 mL, 0.79 mmol) was
added via syringe. After an initial spectrum was taken, the
tube was placed in a heated oil bath. The tube was briefly
removed from the bath periodically to take NMR spectra. After
approximately two half-lives, the rate of cyclization was
calculated with first-order kinetics based on the amount of
time that the sample had been in the oil bath. The concentra-
tion of enediyne was calculated based on the integrated signal
of the anthracene internal standard.
AIBN-Initiated Polymerizations of MMA. A stock solu-
tion was prepared with AIBN (0.20 M) and anthracene (0.093
M) in d6-benzene. NMR samples were prepared with this stock
solution (0.60 mL), MMA (0.050 mL, 0.047 mmol), and the
appropriate enediyne (0.049 mmol). A control sample with no
enediyne was prepared with the same stock solution. The
samples were capped with septa and removed from the
glovebox. The solutions were heated to 60 °C in the NMR
probe, and a spectrum was taken every minute. The concen-
trations of MMA and enediyne were calculated based on the
integrated signals of the anthracene internal standard.
Preparation of 1-Pent-1-ynyl-2-(3-phenylsulfanylprop-
1-ynyl)benzene. A solution of NaOH in water (0.22 mL, 15
M, 3.3 mmol) was added to thiophenol (360 mg, 3.3 mmol) in
THF (50 mL). A solution of methanesulfonic acid 3-(2-pent-1-
ynylphenyl)prop-2-ynyl ester (650 mg, 2.35 mmol) in THF (5
mL) was then added dropwise to the thiophenol solution at
room temperature, and the reaction was stirred for 1 h.
Column chromatography with petroleum ether gave 522 mg
(77%) of 1-pent-1-ynyl-2-(3-phenylsulfanylprop-1-ynyl)benzene
as a yellow oil. TLC Rf 0.40 (1:19 ethyl acetate/petroleum
ether). FTIR (neat): 3059, 2962, 2932, 2871, 2229, 1584, 1481,
1
1440, 1227, 758, 739, 690 cm-1. H NMR (500 MHz, CDCl3,
δ): 7.52-7.54 (m, 2H, ar), 7.38 (m, 1H, ar), 7.31-7.34 (m, 3H,
ar), 7.17-7.26 (m, 3H, ar), 3.89 (s, 2H, -CH2-S-), 2.39 (t, J
) 7.0 Hz, 2H, -CH2-CH2-), 1.62 (m, 2H, -CH2-CH3), 1.05
(t, J ) 7.4 Hz, 3H, -CH3). 13C NMR (500 MHz, CDCl3, δ):
135.8 (q, ar), 132.2 (ar), 132.0 (ar), 130.4 (2C, ar), 129.1 (2C,
ar), 128.1 (ar), 127.3 (ar), 127.0 (ar), 126.9 (q, ar), 125.5 (q,
ar), 94.8 (≡C-CH2-), 88.9 (≡C-CH2-), 82.8 (≡C-C-), 79.6
(≡C-C-), 24.1 (-CH2-S-), 22.4 (-CH2-), 21.8 (-CH2-),
13.7 (-CH3). HRMS-EI (m/z): M+ calcd for C14H14O, 290.1129;
found, 290.1129.
Acknowledgment. We are grateful to the Fannie
and John Hertz Foundation and the Air Force Office of
Scientific Research for financial support.
Supporting Information Available: DSC data on the
cyclization of 4, UV absorbance data for PMMA initiated by
1, kinetics of enediyne consumption in the presence of AIBN,
and crystallographic data for 8. This material is available free
of charge via the Internet at http://pubs.acs.org.
Preparation of 1-(3-Benzenesulfonyl-prop-1-ynyl)-2-
pent-1-ynylbenzene (4). A solution of 1-pent-1-ynyl-2-(3-
phenylsulfanylprop-1-ynyl)benzene (652 mg, 3.78 mmol) in
methylene chloride (3 mL) was added dropwise at 0 °C to a
solution of m-chloroperoxybenzoic acid in methylene chloride
(40 mL), and stirring was continued for 1 h at 0 °C. Column
chromatography with a 1:4 mixture of ethyl acetate and
petroleum ether followed by a second column with methylene
chloride gave 247 mg (45%) of 1-(3-benzenesulfonylprop-1-
ynyl)-2-pent-1-ynylbenzene as well as 209 mg of the corre-
sponding sulfoxide. The sulfoxide was further oxidized under
analogous conditions and purified with column chromatogra-
phy using methylene chloride to give an additional 136 mg of
the sulfone for a total yield of 69%. TLC Rf 0.60 (methylene
chloride). FTIR (neat): 3062, 2963, 2934, 2905, 2872, 2230,
References and Notes
(1) Flory, P. J. J. Am. Chem. Soc. 1937, 59, 241-253.
(2) Mayo, F. R. J. Am. Chem. Soc. 1943, 65, 2324-2329.
(3) Mayo, F. R.; Gregg, R. A.; Matheson, M. S. J. Am. Chem. Soc.
1951, 73, 1691-1700.
(4) Johnson, D. H.; Tobolsky, A. V. J. Am. Chem. Soc. 1952, 74,
938-943.
(5) Haward, R. N. Trans. Faraday Soc. 1950, 46, 204-210.
(6) Zimm, B. H.; Bragg, J. K. J. Polym. Sci. 1952, 9, 476-478.
(7) Mayo, F. R. J. Am. Chem. Soc. 1953, 75, 6133-6141.
(8) Khuong, K. S.; Jones, W. H.; Pryor, W. A.; Houk, K. N. J.
Am. Chem. Soc. 2005, 127, 1265-1277.
(9) Overberger, C. G.; Lapkin, M. J. Am. Chem. Soc. 1955, 77,
4651-4657.
(10) Cohen, S. G.; Hsiao, S.; Sakland, E.; Wang, C. J. Am. Chem.
Soc. 1957, 79, 4400-4405.
(11) Toplikar, E. G.; Herman, M. S.; Padias, A. B.; Hall, H. K.;
Priddy, D. B. Polym. Bull. (Berlin) 1997, 39, 37-43.
(12) Borsig, E.; Vadnalova, O.; Kolar, P.; Lazar, M. Chem. Zvesti.
1976, 30, 328-335.
(13) Li, T.; Willis, T. J.; Padias, A. B.; Hall, H. K. Macromolecules
1991, 24, 2485-2487.
(14) Li, T.; Padias, A. B.; Hall, H. K. Polym. Bull. (Berlin) 1991,
25, 537-541.
(15) Wang, K. K. Chem. Rev. 1996, 96, 207-222.
(16) Grissom, J. W.; Gunawardena, G. U.; Klingberg, D.; Huang,
D. H. Tetrahedron 1996, 52, 6453-6518.
(17) John, J. A.; Tour, J. M. J. Am. Chem. Soc. 1994, 116, 5011-
5012.
(18) John, J. A.; Tour, J. M. Tetrahedron 1997, 53, 15515-15534.
(19) Smith, D. W.; Babb, D. A.; Snelgrove, R. V.; Townsend, P.
H.; Martin, S. J. J. Am. Chem. Soc. 1998, 120, 9078-9079.
(20) Chen, X. H.; Tolbert, L. M.; Hess, D. W.; Henderson, C.
Macromolecules 2001, 34, 4104-4108.
1481, 1447, 1326, 1167, 1139, 1086, 873, 761, 744, 688 cm-1
.
1H NMR (500 MHz, CDCl3, δ): 8.07 (m, 2H, ar), 7.68 (m, 1H,
ar), 7.58 (m, 2H, ar), 7.38 (m, 1H, ar), 7.29 (m, 1H, ar), 7.25
(m, 1H, ar), 7.19 (m, 1H, ar), 4.22 (s, 2H, -CH2-S-), 2.38 (t,
J ) 7.0 Hz, 2H, -CH2-CH2-), 1.58 (m, 2H, -CH2-CH3), 1.01
(t, J ) 7.3 Hz, 3H, -CH3). 13C NMR (500 MHz, CDCl3, δ):
138.1 (q, ar), 134.4 (ar), 132.4 (ar), 132.2 (ar), 129.3 (4C, ar),
128.9 (ar), 127.4 (ar), 127.2 (q, ar), 124.1 (q, ar), 95.3 (≡C-
CH2-), 86.8 (≡C-CH2-), 80.1 (≡C-C-), 79.2 (≡C-C-), 49.9
(-CH2-S-), 22.2 (-CH2-), 21.7 (-CH2-), 13.7 (-CH3).
HRMS-EI (m/z): M+ calcd for C14H14O, 322.1028; found,
322.1026.
Myers Cyclization of 4 to 2-Benzenesulfonylmethyl-
3-propylnaphthalene (5). A solution of 4 (20 mg, 0.062
mmol), 1,4-cyclohexadiene (0.50 mL, 423 mg, 5.3 mmol), and
triethylamine (31 mg, 0.31 mmol) in benzene (2 mL) was
stirred for 15 h at 37 °C followed by 23 h at 67 °C. Column
chromatography with a 1:9 mixture of ethyl acetate and
petroleum ether gave 12 mg (60%) of 5. TLC Rf 0.21 (1:9 ethyl
acetate/petroleum ether). 1H NMR (500 MHz, CDCl3, δ): 7.74
(d, J ) 8.1, 1H, ar), 7.59-7.65, (m, 5H, ar), 7.40-7.49 (m, 5H,
ar), 4.55 (s, 2H, -CH2-S-), 2.55 (t, J ) 7.8 Hz, 2H, -CH2-

Macromolecules, Vol. 38, No. 17, 2005
Polymerization of Vinyl Monomers 7273
(21) Perera, K. P. U.; Shah, H. V.; Foulger, S. H.; Smith, D. W.
Thermochim. Acta 2002, 388, 371-375.
(29) Lockhart, T. P.; Commita, P. B.; Bergman, R. G. J. Am. Chem.
Soc. 1981, 103, 4082-4090.
(22) Prasanna, K.; Perera, U.; Krawiec, M.; Smith, D. W. Tetra-
hedron 2002, 58, 10197-10203.
(30) Grissom, J. W.; Klingberg, D. Tetrahedron Lett. 1995, 36,
6607-6610.
(23) Johnson, J. J.; Bringley, D. A.; Wilson, E. E.; Lewis, K. D.;
Beck, L. W.; Matzger, A. J. J. Am. Chem. Soc. 2003, 125,
14708-14709.
(31) Grissom, J. W.; Klingberg, D.; Huang, D. H.; Slattery, B. J.
J. Org. Chem. 1997, 62, 603-626.
(32) Pryor, W. A.; Fiske, T. R. Trans. Faraday Soc. 1969, 65,
1865-1872.
(24) Rule, J. D.; Wilson, S. R.; Moore, J. S. J. Am. Chem. Soc.
2003, 125, 12992-12993.
(33) Sato, T.; Otsu, T. Makromol. Chem. 1977, 178, 1941-1950.
(34) The previous patent literature suggests that the electron
density of the monomer (i.e. the Alfrey-Price e value) plays
a highly important role in determining reactivity with the
radicals resulting from cycloaromatization. The methacry-
lates have similar e values (0.28 and 0.40) to those of methyl
acrylate (0.64) and vinylidene chloride (0.34), yet Table 1
shows that the methacrylates display much higher reactivity.
Therefore, we believe that effects of monomer electron density
are playing a small role relative to propensity for hydrogen
atom abstraction.
(25) Drumright, R. E.; Terbrueggen, R. H.; Priddy, D. B.; Koster,
R. A. U.S. Patent 5,618,900, 1997.
(26) Semmelhack, M. F.; Neu, T.; Foubelo, F. J. Org. Chem. 1994,
59, 5038-5047.
(27) This mechanism predicts that the aromatic system formed
through Bergman cyclization will be at the end of one of the
two monoradical chains resulting from chain transfer. If this
is the case, the polymeric product should contain residual
amounts of initiator byproducts (i.e., aromatic systems). This
prediction is supported in MMA polymerizations initiated by
enediyne 1. The resulting high molecular weight polymer
shows UV absorbance in the range of 280-310 nm which is
consistent with the presence of such aromatic systems. These
data can be found in the Supporting Information.
(28) Grissom, J. W.; Calkins, T. L.; McMillen, H. A.; Jiang, Y. H.
J. Org. Chem. 1994, 59, 5833-5835.
(35) Higashiura, K.; Oiwa, M. J. Polym. Sci., Part A 1968, 6,
1857-1869.
(36) Simonescu, C. I.; Natansohn, A.; Percec, V. J. Macromol. Sci.,
Pure Appl. Chem. 1981, A14, 643-657.
MA050750Z