Development of a versatile methodology for the synthesis of poly(2,5-benzophenone) containing coil-rod-coil triblock copolymers

Article abstract of DOI:10.1021/ma0352983

A short, simple, versatile methodology was developed for the synthesis of poly(2,5-benzophenone) containing coil-rod-coil triblock copolymers. After a model study, poly(2,5-benzophenone) macroinitiators were synthesized via end-capping poly(2,5-dichlorobe

Full text of DOI:10.1021/ma0352983

3642
Macromolecules 2004, 37, 3642-3650
Development of a Versatile Methodology for the Synthesis of
Poly(2,5-benzophenone) Containing Coil-Rod-Coil Triblock Copolymers
Er ik C. Ha gber g,† Br a n d on Good r id ge,† Oza n Ugu r lu ,‡ Scott Ch u m bley,‡ a n d
Va ler ie V. Sh ea r es*,†,§
Department of Chemistry and Department of Material Science and Engineering, Iowa State University,
Ames, Iowa 50011
Received September 3, 2003; Revised Manuscript Received February 23, 2004
ABSTRACT: A short, simple, versatile methodology was developed for the synthesis of poly(2,5-
benzophenone) containing coil-rod-coil triblock copolymers. After a model study, poly(2,5-benzophenone)
macroinitiators were synthesized via end-capping poly(2,5-dichlorobenzophenone) synthesized by Ni(0)-
catalyzed polymerization and functionalization of the chain ends utilizing phase transfer chlorination.
Varying molecular weights of polystyrene-b-poly(2,5-benzophenone)-b-polystyrene were synthesized using
the macroinitiators in the atom transfer radical polymerization of styrene. The materials were
characterized by gel permeation chromatography, nuclear magnetic resonance, differential scanning
calorimetry, and transmission electron microscopy. Applications were explored through dynamic
mechanical analysis and the construction of a distributed junction photovoltaic device.
In tr od u ction
However, the importance of polydispersity in systems
containing a polydisperse rod block has never fully been
explored. Because of the incompatibility of the rod block
in the random coil phase, phase separation is expected,
but the degree of control of the resulting morphology is
not known.1
Rod-coil and coil-rod-coil block copolymers are a
unique and interesting class of new materials. The
combination of the flexible random coil segments with
rigid-rod blocks gives rise to many potential applications
in optoelectronic devices, nanopatterning, and mechan-
ical reinforcing agents. The utility of these materials
arises from the block copolymer architecture. The inher-
ent incompatibility resulting from entropic factors be-
tween the rigid-rod-like block and the flexible random
coil block leads to phase separation at much lower
molecular weights than coil-coil systems.1 The flexible
coillike blocks also increase the solubility and process-
ability of the copolymer over that of the rodlike homo-
polymer. Rod-coil block copolymers containing materi-
als such as oligophenylenevinylene (OPV),2,3 oligo-
phenyleneethylene (OPE),4 and poly-p-phenylene (PPP)5
have been synthesized. Additionally, coil-rod-coil tri-
block copolymer architectures containing aromatic het-
erocyclic rigid rod materials such as poly(benzobis-
thiazole) have been reported.6-9
Ordered, phase-separated structures with dimensions
as small as 10 nm have been constructed. The ability
to pattern materials such as PPV, PPE, and PPP on
nanometer scales leads to potential utility in the
electronics industry. For example, coil-rod-coil block
copolymers have exhibited narrowed emission spectra
over that of the rod homopolymer due to isolated
domains of the optically active material.8,10 Further-
more, the high modulus of the rigid backbones has been
exploited through the use of these materials to construct
nanocomposites.11
The primary challenge in working with rod-coil block
copolymers lies in the synthesis of rods with controlled
polydispersity and controlled functionality at the chain
end. The difficulties arise from the long stepwise
syntheses necessary to build functionalized rods of
useful molecular weight. Poly(ethylene oxide)-b-poly-
(phenylenevinylene)-b-poly(ethylene oxide) is a good
example.2 This material, containing monodisperse rod
and coil blocks, exhibits interesting self-assembly in
solution but requires seven separate synthetic steps to
produce triblock copolymers containing PPV eightmers.
When not a requirement for the desired application, the
removal of the narrow polydispersity constraint upon
the rod block expands the possible synthetic pathways.
By the addition of molecules monofunctional with
respect to the polymerization conditions, chain end
functionalized rodlike polymers can be synthesized in
a single step. Polymerization methods such as this
eliminate the necessity for long, laborious, stepwise
synthetic methods. The resulting materials can be
converted to the desired block copolymers by grafting
or by using them as macroinitiators. The rod material
of interest in our research is poly(2,5-benzophenone).
The benefits of this polymer include good thermal
properties, excellent mechanical properties, and con-
ductivity when doped.12 In addition, many functional
polymeric materials may be designed via functionalized
monomers, functionalization of the polymer, or the
addition of end-capping agents to the polymeriza-
tion.13,14 As such, this would be a valuable material to
incorporate into coil-rod-coil copolymers if a feasible
synthetic strategy could be designed.
Narrow overall polydispersity is generally considered
a prerequisite for the formation of ordered structures
from most block copolymers consisting of two coil blocks.
† Department of Chemistry.
‡ Department of Material Science and Engineering.
§ Current address: University of North Carolina at Chapel Hill,
Department of Chemistry, 15-26 Venable Hall, CB#3290, Chapel
Hill, NC 27599-3290.
Herein, we describe the details of a short, simple,
versatile synthetic methodology to produce coil-rod-
coil triblock copolymers. As an example material, we
have synthesized polystyrene-b-poly(2,5-benzophenone)-
* Corresponding author: e-mail ashby@email.unc.edu.
10.1021/ma0352983 CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/20/2004

Macromolecules, Vol. 37, No. 10, 2004
Poly(2,5-benzophenone) Containing Triblock Copolymers 3643
4′-Isop r op ylb en zop h en on e-Ca p p ed P oly(2,5-b en zo-
p h en on e). Under an inert atmosphere, NiCl2 (0.1 equiv), Zn
(3.1 equiv), PPh3 (0.4 equiv), and bipyridine (0.1 equiv) were
weighed and added to a flask equipped with an overhead
stirrer. DMAc (10 equiv) was added, and the mixture was
stirred and heated at 80 °C until a deep red color was observed.
Varying ratios of 2,5-dichlorobenzophenone and end-capping
agent, 4-chloro-4′-isopropylbenzophenone, were added (1 equiv
total). Reactions were stirred and heated until no change in
color or viscosity was observed. The polymer was precipitated
into a 4:1 methanol:hydrochloric acid solution to remove excess
zinc. The polymers were then washed with sodium bicarbonate
solution and water and dried under vacuum. 1H NMR: δ )
1.25 (broad singlet, chain end CH3), 3.0 (broad singlet, chain
end CH), 6.7-8.2 (broad multiplet, aryl backbone CHs). 13C
NMR: δ ) 23.7 (CH3), 34.3 (CH), 126.3 (CH), 126.8 (CH), 127.3
(CH), 128.1 (CH), 128.7 (CH), 130.0 (CH), 130.1 (CH), 130.3
(CH), 132.1 (CH), 132.9 (CH), 135.3 (q-C), 136.8 (q-C), 137.1
(q-C), 138.6, (q-C), 139.1 (q-C), 141.2 (q-C), 143.9 (q-C), 154.0
(q-C), 195.8 (CO), 198.2 (CO). Anal. Calcd for C162H110O12: C,
86.52; H, 4.94. Found: C, 86.07; H, 5.32.
b-polystyrene containing a polydisperse rigid-rod block
and monodisperse coil blocks. Rigid poly(2,5-benzo-
phenone) blocks of varying molecular weight were
synthesized by end-capping the polymer formed by the
Ni(0)-catalyzed polymerization of 2,5-dichlorobenzo-
phenone with a monochloro end-capping agent. The end-
capping agent was chosen after a model study to
determine a simple and quantitative route to chlorinate
the chain ends postpolymerization. The rigid-rod ma-
terials were used then as macroinitiators for the atom
transfer radical polymerization (ATRP) of styrene to
construct triblock copolymers. The phase separation
behavior of the resulting triblock materials was exam-
ined by transmission electron microscopy (TEM). In
addition, potential applications of these materials were
explored. The mechanical properties were analyzed by
dynamic mechanical analysis (DMA), and a distributed
junction photovoltaic device was constructed after sul-
fonation of the polystyrene block.
Ch a in E n d Ch lor in a t ion of 4′-Isop r op ylb en zo-
p h en on e-Ca p p ed P oly(2,5-ben zop h en on e). The procedure
was followed as with the chlorination of 2,5-dichloro-4′-
isopropylbenzophenone, except that after 24 h the reaction
mixture was precipitated into acidic methanol, filtered, and
washed with water. The yellow powder was then reprecipitated
from chloroform into acidic methanol. The resulting material
was a pale yellow powder. 1H NMR: δ ) 2.05 (chain end CH3),
6.9-8.1 (broad multiplet, aryl backbone). 13C NMR: δ ) 34.1
(CH3), 67.5 (q-C), 125.6 (CH), 126.9 (CH), 127.5 (CH), 128.1
(CH), 130.0 (CH), 130.1 (CH), 130.3 (CH), 130.7 (CH), 131.1
(CH), 133.0 (q-C), 136.9 (q-C), 137.0 (q-C), 138.1 (q-C), 138.8
(q-C), 139.6 (q-C), 141.2 (q-C), 143.5 (q-C), 150.4 (q-C), 196.7
(CO), 198.6 (CO). Anal. Calcd for C162H108O12Cl2: C, 81.00; H,
4.50. Found: C, 81.86; H, 4.65.
P o ly s t y r e n e -b -p o ly (2,5-b e n zo p h e n o n e )-b -p o ly s t y -
r en e. Chain end chlorinated 4′-isopropylbenzophenone-capped
poly(2,5-benzophenone) (1 equiv in chain ends), CuCl (1 equiv),
and Cu (0.1 equiv) were added to a Schlenk flask under an
inert atmosphere. The flask was sealed with a septum. Styrene
was added via syringe along with anisole (1 mL of solvent/1
mL of monomer) and pentamethyldiethyltriamine (PMDETA)
(2 equiv). The reaction was heated and stirred at 110 °C for
24 h. After the time period, THF was added, and the solution
was precipitated into stirring acidic methanol, filtered, and
reprecipitated from chloroform into acidic methanol. The
resulting material was a pale yellow powder. 1H NMR: δ )
1.48 (polystyrene (PS) backbone CH2), 2.26 (PS backbone CH),
6.63 (broad, PS phenyl), 7.13 (broad, PS phenyl), 7.43 (broad,
polybenzophenone (PB)), 7.74 (broad, PB). 13C NMR: δ ) 42.02
(PS backbone), 44.37 (PS backbone), 125.65 (PS phenyl), 126.74
(PB), 127.71 (PS phenyl), 128.00 (PS phenyl), 129.03 (PB),
129.85 (PB), 130.32 (PB), 130.71 (PB), 132.84 (PB), 137.29
(PB), 139.25 (PB), 145.39 (PS phenyl), 145.71 (PB), 146.11
(PB), 197.35 (PB).
Exp er im en ta l Section
Ma ter ia ls. All materials were purchased from Aldrich and
used without further purification unless otherwise noted. N,N-
Dimethylacetamide and anisole were dried over CaH2 (Fisher)
and distilled under reduced pressure. Styrene was passed over
basic alumina and distilled under reduced pressure. Tri-
phenylphosphine was recrystallized from cyclohexane. 2,2′-
Bipyridine and 4,4′-di(5-nonyl)-2,2′-bipyridine (dNbipy) were
recrystallized from ethanol. The commercial bleach, Chlorox,
was titrated with concentrated hydrochloric acid to pH 8.5
immediately prior to use.
2,5-Dich lor ob en zop h en on e. The synthesis of 2,5-di-
chlorobenzophenone has been described previously.15
4-Ch lor o-4′-a lk ylben zop h en on e. Aluminum chloride (1
equiv) was added to a stirring solution of 4-chlorobenzoyl
chloride (1.1 equiv) and toluene (or cumene) (1 equiv) in
nitromethane at 0 °C. The solution was allowed to warm to
room temperature and stir for 24 h. The solution was then
precipitated on ice and allowed to stir overnight. The off-white
powder was filtered, dissolved in ethanol, treated with acti-
vated carbon, and filtered. The resulting white crystals were
recrystallized twice more from ethanol. 4-Chloro-4′-methyl-
benzophenone (96% yield). 1H NMR: δ ) 2.45 (s, 3H), 7.29 (d,
2H, J ) 8.4 Hz), 7.45 (d, 2H, J ) 8.4 Hz), 7.69 (d, 2H, J ) 8.4
Hz), 7.74 (d, 2H, J ) 8.4 Hz). 13C NMR: δ ) 21.9 (CH3), 128.8
(CH), 129.3 (CH), 130.4 (CH), 131.6 (CH), 134.8 (q-C), 136.5
(q-C), 138.8 (q-C), 143.8 (q-C), 195.5 (CO). Anal. Calcd for
C
14H11OCl: C, 72.89; H, 4.82. Found: C, 72.76; H, 4.94.
4-Chloro-4′-isopropylbenzophenone (74% yield). 1H NMR: δ )
1.29 (d, 6H, J ) 6.9 Hz), 2.99 (septet, 1H, J ) 6.9 Hz), 7.34 (d,
2H, J ) 8.4), 7.45 (d, 2H, J ) 8.1 Hz), 7.74 (m, 4H, J ) 8.1,
8.4 Hz). 13C NMR: δ ) 23.7 (CH3), 34.3 (CH), 126.5 (CH), 128.5
(CH), 130.3 (CH), 131.4 (CH), 134.9 (q-C), 136.2 (q-C), 138.6
(q-C), 154.3 (q-C), 195.2 (CO). Anal. Calcd for C16H15OCl: C,
74.26; H, 5.85. Found: C, 74.31; H, 5.82.
Su lfon a ted P olystyr en e-b-p oly(2,5-ben zop h en on e)-b-
p olystyr en e. Fuming sulfuric acid (5 mL) was added dropwise
to stirring polystyrene-b-poly(2,5-benzophenone)-b-polystyrene
(0.5 g) dissolved in methylene chloride (5 mL). The reaction
was allowed to stir for approximately 5 min until the lower
methylene chloride layer was clear and colorless. The layers
were separated, and the aqueous layer was neutralized with
3 M NaOH. The sodium salt of the sulfonated copolymer was
precipitated from water into stirring methanol. 1H NMR: δ )
1.5 (broad multiplet), 6.7 (broad singlet), 7.6 (broad singlet).
13C NMR: δ ) 20.1 (sulfonated PS backbone CH2), 23.4
(sulfonated PS backbone CH), 133.3 (sulfonated PS phenyl
CH), 135.3 (sulfonated PS phenyl CH), 150.3 (sulfonated PS
phenyl q-C), 158.3.
4-Ch lor o-4′-(ch lor om eth yl)ben zoph en on e an d 4-Ch lor o-
4′-(ch lor od im et h ylm et h yl)b en zop h en on e. 4-Chloro-4′-
methylbenzophenone (11 mmol) or 4-chloro-4′-isopropyl-
benzophenone (10 mmol) was dissolved in 3 mL of methylene
chloride. Bleach (Chlorox, pH adjusted to 8.5, 2 equiv in
hypochlorite) and benzyltriethylammonium chloride (4 equiv)
were added, and the mixture was stirred at 1000 rpm for 24
h. Five mL of methylene chloride was added, and the aqueous
layer was removed. The methylene chloride solution was
washed twice with water. The methylene chloride was then
removed, yielding white crystals. 4-Chloro-4′-(chlorodimethyl-
methyl)benzophenone (100% yield). 1H NMR: δ ) 2.08 (s, 6H),
7.46 (d, 2H, J ) 8.7 Hz), 7.70 (d, 2H, J ) 8.7 Hz), 7.76 (m, 4H,
J ) 8.7, 8.7 Hz). 13C NMR: δ ) 34.4 (CH3), 69.0 (q-C), 125.9
(CH), 128.8 (CH), 130.2 (CH), 131.7 (CH), 135.9 (q-C), 136.5
(q-C), 139.2 (q-C), 151.0 (q-C), 195.1 (CO). Anal. Calcd for
Ch a r a cter iza tion . Molecular weights, relative to narrow
polystyrene standards, were measured using a Waters gel
permeation chromatography (GPC) system consisting of a
Waters 510 pump, a Waters 717 autosampler, a Wyatt Optilab
DSP refractometer, and a Wyatt Dawn EOS light scattering
C
16H14OCl2: C, 65.54; H, 4.82. Found: C, 64.85; H, 5.08.

3644 Hagberg et al.
Macromolecules, Vol. 37, No. 10, 2004
Sch em e 1. Syn th esis of Ben zyl Ch lor id e-Ter m in a ted
P oly(2,5-ben zop h en on e)
Ni(0) coupling reaction. These polybenzophenones will
be referred to as MIbenzyl (macroinitiator with benzyl
chloride end-caps). End-capping agent mole fractions of
0.2, 0.1, and 0.05 were employed. The molecular weights
obtained were 3.1 × 103 g/mol with a polydispersity
index (PDI) of 1.9, 3.5 × 103 g/mol with a PDI of 2.1,
and 3.9 × 103 g/mol with a PDI of 2.4. While the
molecular weight differences are not large, numerous
experiments at specific mole fractions of end-capping
agent gave consistent results.
Although macroinitiators with quantitatively chlori-
nated chain ends could not be synthesized in this
manner, these materials were utilized to explore the
reaction conditions and the feasibility of the method
(Scheme 2). Because of the limited solubility of poly-
(2,5-benzophenone) in common solvents, polymeriza-
tions were run in two suitable solvents, N,N-dimethyl-
acetamide (DMAc) and anisole. The results of these
polymerizations are summarized in Table 1. Although
the MIbenzyl series lacked quantitatively functionalized
chain ends, triblocks of varying molecular weights were
synthesized. With DMAc as solvent, reaction times of
48 h were necessary to achieve high yields. When
anisole was used as a solvent, 96% yield was achieved
in 14 h. This behavior is most likely due to solvent-
induced changes in the catalyst structure.16
detector. The measurements were taken at 40 °C with THF
as the mobile phase on four columns (Polymer labs PLgel 100,
500, 1 × 104, and 1 × 105 Å). Glass transitions were determined
at the inflection point of the endotherm with a Perkin-Elmer
Pyris 1 differential scanning calorimetry system (DSC) with
a heating rate of 10 °C/min. Transmission electron microscopy
(TEM) was performed on a 1200EX J EOL STEM with a 120
kV accelerating potential and a Philips CM30 STEM with a
100 kV accelerating potential. Dynamic mechanical analysis
(DMA) measurements were conducted on a Perkin-Elmer DMA
7e in a three-point bending mode with a 100 mN static force
and a 110 mN dynamic force. Photovoltaics were constructed
by casting a thin polymer film upon an indium tin oxide-coated
2 in. × 2 in. glass plate by controlled solvent evaporation over
a period of 2 days. The polymer-coated plates were then
immersed for 1 h in a sodium dispersion in mineral oil. The
plates were then rinsed with dry hexanes and dried, and then
a second ITO-coated glass plate was placed over the polymer
film. Rudimentary evaluation of the devices was accomplished
by measuring the voltage and current developed across the
two ITO electrodes in the dark and when exposed to full
sunlight.
The triblock materials from both solvents had bimodal
distributions due to the macroinitiator lacking benzyl
halide chain ends. The incorporation of the poly(2,5-
benzophenone) into the triblock copolymer was con-
firmed by extracting the samples with ethyl acetate.
Poly(2,5-benzophenone) is not soluble in ethyl acetate,
while the diblock and triblock copolymers are soluble.
This mixture was added to ethyl acetate and centri-
fuged, and the clear solution was removed. The soluble
material was precipitated into methanol and isolated.
1H NMR of this material confirmed the incorporation
of the poly(2,5-benzophenone) into the triblock copoly-
mer with broad signals at 7.35 and 7.7 ppm. Further-
more, GPC of the material recovered from solution was
monomodal with a number-average molecular weight
of 100.1 × 103 g/mol and PDI of 1.4. The GPC trace of
the polymer prior to fractionation was bimodal with an
Mn of 115.7 × 103 g/mol and a PDI of 1.6.
Resu lts a n d Discu ssion
The goal of this work was to develop a simple,
versatile methodology for the synthesis of poly(2,5-
benzophenone) containing coil-rod-coil triblock copoly-
mers. The utilization of end-capping agents yielded
chain-end-functionalized rod materials in a single step,
avoiding the necessity of time-intensive multistep meth-
ods. These rodlike materials can be used as macro-
initiators or grafted with other chain-end-functionalized
polymers. Previously, our group explored the end-
capping of poly(2,5-benzophenone)s by addition of mol-
ecules that are monofunctional with respect to the Ni(0)
coupling reaction.10 The monofunctional end-caps can
be chosen so that they contain a functionality, not
reactive in Ni(0) coupling but reactive under other
conditions. In this work it was chosen to end-cap the
polymers with molecules containing functionalities that
can act as initiating sites in atom transfer radical
polymerization (ATRP). Functionalities that act as
ATRP initiators are known to be reactive under a
variety of Ni-catalyzed coupling conditions. Despite this
likely incompatibility, it was desired to utilize the
simplest possible synthetic route. Therefore, p-chloro-
benzyl chloride was explored as a potential end-capping
agent (Scheme 1). Addition of this compound yields poly-
(2,5-benzophenone)s with limited chain-end function-
alization due to reduction of the benzyl halide in the
To synthesize macroinitiator with quantitatively func-
tionalized chain ends, the reduction of the benzyl
chloride in the Ni(0) coupling reaction had to be
overcome. To accomplish this, an end-capping system
with good compatibility to the Ni(0) coupling reaction
was designed. Aryl chloride-containing species with a
benzylic carbon, which could be chlorinated postpoly-
merization, were investigated. The end-capping agents
chosen were 4-chloro-4′-alkylbenzophenones. These ben-
zophenone derivatives are structurally similar to the
monomer, giving them reactivities close to that of the
monomer. End-capping was attempted with p-chloro-
toluene but resulted in materials with broad multimodal
molecular weight distributions. This result in the pres-
ence of p-chlorotoluene was most likely due to differ-
ences in reactivity, which will be discussed later.
A simple, selective chlorination route is also necessary
to functionalize the chain ends postpolymerization.
Phase-transfer chlorination is a simple high yield route
to chlorination of benzylic carbons. It has previously
been shown that bleach and benzyltriethylammonium
chloride can effectively chlorinate benzylic carbons in
both small molecules and macromolecules.17,18 When

Macromolecules, Vol. 37, No. 10, 2004
Poly(2,5-benzophenone) Containing Triblock Copolymers 3645
Sch em e 2. ATRP of Styr en e Usin g P oly(2,5-ben zop h en on e) Ma cr oin itia tor
Ta ble 1. P olym er iza tion Resu lts a n d Con d ition s for th e
ATRP of Styr en e Usin g th e Ben zyl Ch lor id e
En d -F u n ction a lized Ma cr oin itia tor a
Mn × 10-3
%
sample M/Ib time (h) solvent
(g/mol)c
PDIc yieldd
TB1
TB2
TB3
60
80
80
48
48
14
DMAc
DMAc
anisole
32.5
115.7
85.9
1.5
1.6
1.6
74
92
96
a
Polymerization conditions: 3.1 × 103 g/mol macroinitiator, 2
b
equiv of dNbipy, 110 °C. M/I ) mol monomer/mol initiator
(theoretical number of chain ends). c Data obtained by GPC.
d
Determined by mass recovered.
Ta ble 2. Con d ition s a n d Resu lts of th e P h a se Tr a n sfer
Ch lor in a tion of 4′-Alk ylben zop h en on e Der iva tives
F igu r e 1. 1H NMR of 4-chloro-4′-isopropylbenzophenone
before and after chlorination.
time
(h)
bleach
equiv
BnEt3NCl
equiv
temp
(°C)
%
substrate
Me
yielda
significant negative F+ value, and thus the presence of
electron-withdrawing groups, such as a carbonyl in the
para position, slows the chlorination reaction.17 To
increase the rate of chlorination, electron density must
be added to the benzylic carbon. It was found that
4-chloro-4′-isopropylbenzophenone could be quantita-
tively chlorinated under the best reaction conditions for
the methyl derivative (Scheme 3). The addition of the
methyl substituents on the benzylic carbon increases
the electron density at the reaction site enough to
overcome the electron-withdrawing effect of the car-
8
4
8
24
24
4b
24
24
2
8
8
8
8
8
2
2
2
2
2
2
2
2
4
4
RT
RT
RT
RT
40
80
RT
RT
41
37
41
47
39
5
76
100
i-Pr
Determined by 1H NMR. Reaction mixture turned black.
a
b
Sch em e 3. P h a se Tr a n sfer Ch lor in a tion of
Alk ylben zop h en on e Mod el Com p ou n d s
1
bonyl. The reaction can be easily followed by H NMR
before and after chlorination (Figure 1). Prior to chlo-
rination, the methyls of the isopropyl group have a shift
of δ 1.3 ppm. After chlorination, the peak was shifted
downfield to δ 2.0 ppm due to the chlorine atom in the
benzylic position.
4-Chloro-4′-isopropylbenzophenone was utilized as an
end-capping agent in the Ni(0)-catalyzed polymerization
of 2,5-dichlorobenzophenone (Scheme 4). These materi-
als will be referred to as MIiso (macroinitiator with
isopropylbenzophenone end-caps). Varying the feed ratio
from 0.2 to 0.025 mole fraction end-capping agent
yielded polymers with molecular weights ranging from
1.7 × 103 to 3.3 × 103 g/mol (Table 3). Once again,
although the variation in molecular weights is not large,
repeated experiments consistently gave these molecular
weights. The incorporation of the end-capping agent was
4-chloro-4′-methylbenzophenone was reacted using the
conditions described in the literature (2 equiv of bleach
and 2 equiv of benzyltriethylammonium chloride), a
yield of only 41% was obtained (Table 2). Increasing the
amount of bleach and varying the time from 4 to 24 h
at room temperature only increased the yield to 47%.
Higher temperatures were also ineffective, with 40 °C
giving a yield of 39% in 24 h. When the temperature
was increased to 80 °C, a yield of only 5% was observed
at 4 h. However, when 4 equiv of benzyltriethylammo-
nium chloride was utilized with 2 equiv of bleach, the
yield increased to 76%. Unfortunately, further increase
in the amount of ammonium salt did not lead to
increased yields with the methyl derivative. It is known
that the phase transfer chlorination reaction exhibits a
1
observed by H NMR (Figure 2). With increasing mole
fraction of end-capping agent in the feed, an increase
in the signal arising from the methyl of the isopropyl
group (δ 1.3 ppm) was observed. Attempts with lower
amounts of end-capping agent in the feed led to broad

3646 Hagberg et al.
Macromolecules, Vol. 37, No. 10, 2004
Sch em e 4. Syn th esis of 4′-Isop r op ylben zop h en on e-Ca p p ed P oly(2,5-ben zop h en on e)
Ta ble 3. Molecu la r Weigh t Da ta for
Isop r op ylben zop h en on e-Ca p p ed
P oly(2,5-ben zop h en on e)s
end-cap feeda
(mole fraction)
Mn × 10-3
sample
(g/mol)b
PDIb
MI1
MI2
MI3
MI4
0.2
0.1
0.05
0.025
1.7
2.4
2.9
3.3
1.7
1.8
2.1
2.1
a
End-cap feed ) mol end-cap/(mol end-cap + mol monomer).
b
Data determined by GPC.
F igu r e 3. 1H NMR of 4′-isopropylbenzophenone chain-end-
functionalized poly(2,5-benzophenone) before and after chlo-
rination.
Sch em e 5. P h a se Tr a n sfer Ch lor in a tion of
4′-Isop r op ylben zop h en on e-Ca p p ed
P oly(2,5-ben zop h en on e)
F igu r e 2. 1H NMR 4′-isopropylbenzophenone-functionalized
benzophenones resulting from varying feed ratios of monomer
and end-capping agent.
multimodal molecular weight distributions and poor
molecular weight control. The loss of control with end-
cap amounts less than 2.5% is presumably due to side
reactions such as the reduction of aryl chlorides and
reduction of ketones.13,19 It is also interesting to note
the difference in molecular weights of the MIbenzyl
series compared to the MIiso series (Table 3). With the
same feed ratio of monomer and end-capping agent, the
materials end-capped with chlorobenzyl chloride are
higher molecular weight. At 0.2 mole fraction end-
capping agent, a molecular weight of 3.1 × 103 g/mol
was achieved in the MIbenzyl series. At the same mole
fraction of end-capping agent in the MIiso series, the
resulting material had a molecular weight of 1.7 × 103
g/mol. Similar results were observed for the other end-
capping agent feed ratios. Percec et al. have observed a
diminished reactivity to Ni(0)-catalyzed coupling of
monomers with electron-donating substituents ortho to
the leaving group.20 Monomers with electron-withdraw-
ing groups ortho to the reactive site show an increased
reactivity. The same effect would be expected for sub-
stituents in the para position. Indeed, this is observed
in the reactivities of the end-capping agents. Because
of a diminished reactivity, chlorobenzyl chloride reacts
with monomer at a slower rate than monomer reacts
with itself, leading to higher molecular weights. The
benzophenone end-capping agent is structurally similar
to the monomer and therefore reacts with monomer at
or near the same rate that monomer reacts with
monomer. The electron-withdrawing capability of the
end-capping agents also provides insight into the dif-
ficulty in the use of p-chlorotoluene as an end-capping
agent in Ni(0)-catalyzed coupling. Of the three end-
capping agents considered here, p-chlorotoluene is the
most deactivated to Ni(0) coupling. The differences
between the reactivities of the benzophenone monomer
and the p-chlorotoluene end-capping agent lead to poor
control of the chain end functionality.
Utilizing the conditions determined in the model
chlorination study, it was possible to chlorinate the
chain ends of poly(2,5-benzophenone) terminated with
4′-isopropylbenzophenone functionalities (Scheme 5).
The chlorination reaction proceeded in quantitative
yield and was followed by 1H NMR. As shown in Figure
3, prior to chlorination, the methyls of the isopropyl are
observed as a broad doublet at δ 1.3 ppm. After
chlorination, the signal is shifted to δ 2.0 ppm due to
the presence of the chlorine atom at the benzylic
position.

Macromolecules, Vol. 37, No. 10, 2004
Poly(2,5-benzophenone) Containing Triblock Copolymers 3647
Ta ble 5. Gla ss Tr a n sition Tem p er a tu r es of
Ma cr oin itia tor s a n d Tr iblock Cop olym er s
Sch em e 6. ATRP of Styr en e Usin g
P oly(2,5-ben zop h en on e) Ma cr oin itia tor
MI Mn × 10-3 MI TB Mn × 10-3
Tg PS Tg PB
b
(g/mol)a
1.7
Tg
(g/mol)c
PDIc blockb blockb
139
7.9
12.9
8.0
42.2
23.8
34.6
34.6
36.0
1.4
1.4
1.6
1.5
1.2
1.7
1.5
1.2
103
106
103
101
102
110
105
100
2.4
2.8
3.3
149
159
158
155
154
158
152
151
147
a
b
Data determined by GPC for the macroinitiator. Glass
transition determined by DSC; MI ) macroinitiator, PS
)
polystyrene, PB ) polybenzophenone. c Data determined by GPC
for the triblock copolymer.
sion, the poly(2,5-benzophenone) film was stained to a
greater extent than the polystyrene film. Polystyrene
and polybenzophenone were also confirmed to be im-
miscible. The two materials were blended and analyzed
by TEM. Before and after annealing at 180 °C for 48 h,
two phases were observed.
Ta ble 4. Va r ia tion of Tr iblock Cop olym er Weigh t by Use
of th e Mon om er to In itia tor Ra tioa
MI Mn × 10-3
TB Mn × 10-3
Representative TEM micrographs are shown in Fig-
ures 4-6 for triblocks containing varying lengths of rod
block and coil block. Rod blocks of molecular weight 1.7
× 103 g/mol led to the formation of a random distribu-
tion of spheroids of relatively uniform size (∼50 nm) at
7.3 vol % (Figure 4). As the rod block volume was
increased to 11.7% and then 19.4%, larger more size
disperse spheroids were formed. The series containing
rod blocks of 2.4 × 103 g/mol exhibited a similar trend
with the average size of the polybenzophenone domains
increasing with increasing volume fraction. Analysis of
the triblock materials containing 2.7 × 103 g/mol rod
blocks showed polybenzophenone spheroids at 10% rod
block, but upon increasing the volume fraction to 15.7%,
a highly disordered morphology was formed (Figure 5).
Unlike the previous examples, rod blocks of 3.3 × 103
g/mol led to the formation of more disordered phases at
lower rod volume fractions with 9.2 vol % forming size
disperse spheroids (Figure 6).
Although not highly ordered, clear phase separation
was observed in all examples. Overall, volume fractions
of less than 10% led to the formation of polybenzo-
phenone spheroids of relatively uniform size. Larger
volume fractions formed spheroids, but the size of the
phases varied greatly. Upon increasing the rod size and
or the volume fraction, more disordered, but phase-
separated, structures were observed (Figures 5b and
6b). As shown in the 1.7 × 103 g/mol series, the potential
for formation of more ordered structures at higher rod
volume fractions still exists (Figure 4b,c). Increasing the
volume fraction from 11.7% rod to 19.4% rod led to a
decrease in the variation of the spheroid size.
sample
(g/mol)b
PDIb M/Ic
(g/mol)d
PDId
TB1
TB2
TB3
TB4e
1.7
1.7
1.7
2.4
1.69
1.69
1.69 157
1.82 314
39
78
7.9
12.9
19.0
42.0
1.4
1.4
1.9
1.5
a
Polymerization conditions ) isopropyl functionalized macro-
b
initiator, PMDETA, anisole, 110 °C, 24 h. Data determined by
GPC for the polybenzophenone macroinitiator. c M/I ) moles
d
monomer/moles chain ends of macroinitiator. Data determined
by GPC for triblock copolymer. e 48 h reaction.
These macroinitiators were used in the ATRP of
styrene (Scheme 6). It was found that under optimized
conditions clean and efficient initiation of styrene
yielded triblock copolymers. DMAc again proved to be
an ineffective solvent, while the reaction in anisole with
PMDETA as the ligand gave good yields of the triblock
copolymer. Variation of the monomer-to-initiator ratio
was utilized to synthesize polystyrene blocks of varying
molecular weight (Table 4). As the monomer-to-initiator
ratio was increased from 39 to 157, polymers of increas-
ing molecular weight were synthesized. For large mono-
mer-to-initiator ratios (approximately >150), it was
necessary to increase the reaction time in order to reach
high conversions. It also was found that yields less than
75% were desirable due to increased viscosity at higher
conversions that led to broad multimodal molecular
weight distributions. Increasing the amount of solvent
reduced the rate early in the reaction, preventing high
yields from being obtained.
Differential scanning calorimetry of these materials
showed two glass transition temperatures for triblock
copolymers with rod blocks greater than 1.7 × 103 g/mol
(Table 5). Furthermore, the transitions observed were
characteristic of the transitions of the two homopoly-
mers, indicating that there is little blending of the two
phases. For copolymers with rod blocks of 1.7 × 103
g/mol, the only glass transition observed was for poly-
styrene.
The overall polydispersity of these materials varied
from 1.2 to 1.9. This variation seemed to have little
effect upon the resulting morphologies. In fact, the
copolymer in Figure 5b has an overall polydispersity of
1.3, while the material in Figure 4a has an overall
polydispersity of 1.9. The material with the larger
polydispersity exhibited a higher degree of order. The
polydispersity of the rod block also seemed to have little
effect upon the phase separation. The rod block poly-
dispersity in Figure 4a was 1.7 and in Figure 6a was
2.1. Both materials exhibited similar morphologies of
relatively uniform spheroids.
To elucidate the morphologies formed, TEM studies
were undertaken. Thin films were cast from toluene
upon a water surface and picked up onto carbon-coated
grids. The films were then annealed at 180 °C for 48 h
and stained with RuO4 vapor. To identify the compo-
nents upon staining, thin films of the two materials
were immersed in a 5% solution of RuO4. After immer-
øN values were determined from solubility param-
eters according to Sperling,21 and the data were plotted

3648 Hagberg et al.
Macromolecules, Vol. 37, No. 10, 2004
F igu r e 4. TEM micrographs of polystyrene-b-poly(2,5-benzophenone)-b-polystyrene with 1.7 × 103 g/mol rod blocks: (a) 7.3 vol
% rod block; (b) 11.7 vol % rod block; (c) 19.4 vol % rod block.
F igu r e 5. TEM micrographs of polystyrene-b-poly(2,5-benzo-
phenone)-b-polystyrene with 2.7 × 103 g/mol rod blocks: (a)
10.0 vol % rod block; (b) 15.7 vol % rod block.
F igu r e 7. Partial phase separation diagram for polystyrene-
b-poly(2,5-benzophenone)-p-polystyrene: Filled points: uni-
form spheroids, 50 nm and smaller. Unfilled points: large size
disperse spheroids. [, 3.3 × 103 g/mol rod block; 2, 2.9 × 103
g/mol rod block; b, 2.7 × 103 g/mol rod block; 9, 2.4 × 103 g/mol
rod block; ×, 1.7 × 103 g/mol rod block. Dashed line represents
the spinodal for coil-coil block copolymer systems.
F igu r e 6. TEM micrographs of polystyrene-b-poly(2,5-benzo-
phenone)-b-polystyrene with 3.3 × 103 g/mol rod block: (a) 8.3
vol % rod block; (b) 9.2 vol % rod block.
to give a partial phase separation diagram (Figure 7).
All block copolymers presented here exhibit phase
separation and lay outside the spinodal known for coil-
coil systems. This decreased miscibility correlates well
with what is predicted by theory, with the rod block
driving the system toward phase separation by limiting
the number of chain conformations the coil block can
adopt when the two are blended. It is also interesting
to note that all materials presented here exhibit phase
separation at relatively small rod volume fractions. This
fact is contrary to what has been predicted by theoretical
models, which have shown the spinodal of the rod-coil
systems shifting toward higher rod volume fractions
compared to coil-coil systems.22,23 It is, however, neces-
sary to present a complete phase separation diagram
before definitive conclusions can be drawn.
Simple synthetic routes to a wide variety of these
materials create the potential for their use in mechan-
ical and electronic applications. As an example of this
class of materials utility to enhance mechanical proper-
ties, dynamic mechanical analysis was performed on the
triblock copolymers. When compared to similar molec-
ular weight polystyrene, the triblock copolymer ( Mn
) 55.7 × 103 g/mol, PDI 1.9, 4.4 vol % rod block)
exhibited a large increase in storage modulus (∼50%)
as well as a 15 °C increase in the tan delta maximum
(Figure 8). Furthermore, the maximum value of the tan
F igu r e 8. DMA results of polystyrene-b-poly(2,5-benzo-
phenone)-b-polystyrene and polystyrene.
delta was decreased over that of neat polystyrene,
indicating increased elastic properties. This level of
improvement in the mechanical properties is not typi-
cally achieved in polymer composites at loading levels
less than 10%, although nanocomposite materials have
shown similar properties at low loading levels.24 While
these polystyrene-based materials are only examples
and are unlikely to be used in such mechanical roles,
triblocks of materials such as Nylon or polyesters have
this potential. Utilizing the end-capping methodology
presented here, the rod blocks could be functionalized
to initiate the polymerization of monomers such as
lactones or incorporated into the triblock copolymer via
grafting with other chain-end-functionalized materials.
The phase-separated morphology exhibited here has,
in addition to use in mechanical applications, the
potential to be utilized in the electronics industry to
exploit the conductive backbone of the polybenzo-
phenone. One such application is construction of dis-

Macromolecules, Vol. 37, No. 10, 2004
Poly(2,5-benzophenone) Containing Triblock Copolymers 3649
ITO-coated glass slide was placed over the polymer film.
When the device was exposed to full sunlight, a stable
voltage of 250 mV and current density of 0.39 A/cm2
were observed. Presently, extensive characterization of
the current-voltage properties of the devices is being
conducted.
Con clu sion s
A method for the synthesis of coil-rod-coil triblock
copolymers was demonstrated by the synthesis of poly-
styrene-b-poly(2,5-benzophenone)-b-polystyrene. Poly-
(2,5-benzophenone) macroinitiators were synthesized
utilizing 4-chloro-4′-isopropylbenzophenone as an end-
capping agent followed by phase transfer chlorination
to quantitatively chlorinate the chain ends. Differing
molecular weights of triblock copolymer were synthe-
sized in anisole with PMDETA as the ligand by varying
the monomer-to-initiator ratio. All triblock materials
synthesized exhibited phase separation with rod volume
fractions smaller than 10% producing a morphology
consisting of relatively uniform sized poly(2,5-benzo-
phenone) spheroids. Furthermore, all materials phase
separated at volume fractions much smaller than more
traditional coil-coil systems. Potential applications of
this class of materials were also demonstrated. DMA
analysis showed the triblock copolymers exhibited im-
proved mechanical performance over that of neat poly-
styrene. It was also shown that through a simple
modification an architecture could be designed to exploit
the conjugated rod block in photovoltaic applications.
Photovoltaic devices built utilizing the sulfonated tri-
block copolymer exhibited a voltage of 250 mV and a
current density of 0.39 A/cm2 in sunlight. Currently,
efforts are underway to synthesize high molecular
weight poly(2,5-benzophenone) macroinitiators. In ad-
dition, differing end-capping agents and methods of
chain end functionalization are being explored.
F igu r e 9. TEM micrographs of polystyrene-b-poly(2,5-benzo-
phenone)-b-polystyrene before (a) and after sulfonation (b).
Sch em e 7. Syn th esis of Su lfon a ted
P olystyr en e-b-p oly(2,5-ben zop h en on e)-b-p olystyr en e
tributed junction photovoltaics. The key feature to these
materials is a two-phase system with dimensions on the
scale of tens to hundreds of nanometers. One phase
should exhibit good electron mobility and the other good
hole mobility. Coil-rod-coil block copolymer systems
are excellent candidates for such applications due to the
conjugated rod block and the phase separation at the
submicrometer length scale. As an example synthesis
of these materials for such an application, polystyrene-
b-poly(2,5-benzophenone)-b-polystyrene was reacted with
fuming sulfuric acid (Scheme 7). The resulting material
was soluble in all polar organic solvents, such as THF,
methanol, and water. Upon extraction of the polymer
solution with NaOH, the resulting sodium salt was only
soluble in water. The triblock copolymer functionalized
with sodium sulfonate groups exhibited one Tg at 255
°C. No transition due to poly(2,5-benzophenone) was
observed. Sulfonation of the poly(2,5-benzophenone)
block is unlikely under the conditions used as no
sulfonation of poly(2,5-benzophenone) homopolymer was
observed under the same reaction conditions.
Ack n ow led gm en t. This research was supported by
the U.S. Department of Energy, Office of Basic Energy
Sciences, Division of Chemical Sciences, under Contract
W-7405-Eng-82.
Refer en ces a n d Notes
(1) Klok, H.; Lecommandoux, S. Adv. Mater. 2001, 13, 1217.
(2) Wang, H.; Wang, H. A.; Urban, V. S.; Littrell, K. C.;
Thiyagarajan, P.; Yu, L. J . Am. Chem. Soc. 2000, 122,
6855.
(3) Tew, G. N.; Pralle, M. U.; Stupp, S. I. J . Am. Chem. Soc. 1999,
121, 9852.
(4) Huang, W. Y.; Matsuoka, S.; Kwei, T. K.; Okamoto, Y.; Hu,
X.; Rafailovich, M. H.; Sokolov, J . S. Macromolecules 2001,
34, 7809.
(5) Tsolakis, P. K.; Koulouri, E. G.; Kallitis, J . K. Macromolecules
TEM studies of the resulting sulfonated materials
showed a two-phase morphology, different from that of
the nonsulfonated triblock copolymer (Figure 9). While
the polystyrene-b-poly(2,5-benzophenone)-b-polystyrene
copolymer exhibited a spheroidal morphology, the sul-
fonated derivative exhibited a much more disordered
morphology. The dimensions of the two phases are near
200 nm.
1999, 32, 9054.
(6) Tsai, T. T.; Arnold, F. E.; Hwang, W. F. J . Polym. Sci., Part
A: Polym. Chem. 1989, 27, 2839.
(7) Cooper, K. L.; Arnold, F. E. Polym. Prepr. (Am. Chem. Soc.,
Div. Polym. Chem.) 1992, 33 (1), 1006.
(8) Chen, L. X.; J enekhe, S. A. Polym. Mater. Sci. Eng. 1995,
73, 304.
(9) Baek, J .-B.; J uhl, S. B.; Lyons, C. B.; Farmer, B. L.; Tan, L.-
S. Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.) 2002,
43 (2), 1130.
The sulfonated block copolymer was dissolved in
water and cast upon indium tin oxide (ITO)-coated 2
in. × 2 in. glass plates. The films (approximately 250
nm in thickness) were slowly dried over a period of 2
days and then submerged in a bath of sodium dispersed
in mineral oil for 1 h. In an inert atmosphere, the glass
plates were washed with dry hexanes, and then a second
(10) Marsitzky, D.; Klapper, M.; Mullen, K. Macromolecules 1999,
32, 8685.
(11) Pae, Y.; Harris, F. W. J . Polym. Sci., Part A: Polym. Chem.
2000, 38, 4247.
(12) Marrocco, M. Gagne, R. R. U.S. Patent 5,227,457, 1993.
(13) Bloom, P. D.; Sheares, V. V. Macromolecules 2001, 34, 1627.
(14) Bloom, P. D.; Sheares, V. V. J . Polym. Sci., Part A: Polym.
Chem. 2001, 39, 3505.

3650 Hagberg et al.
Macromolecules, Vol. 37, No. 10, 2004
(15) Phillips, R. W.; Sheares, V. V.; Samulski, E. T.; DeSimone,
J . M. Macromolecules 1994, 27, 2354.
(16) Matyjaszewski, K.; Xia, J . Chem. Rev. 2001, 101, 2921.
(17) Fonouni, H. E.; Krishan, S.; Kuhn, D. G.; Hamilton, G. A. J .
Am. Chem. Soc. 1983, 105, 7672.
(18) Mohanraj, S.; Ford, W. T. Macromolecules 1986, 19, 2470.
(19) Colon, I.; Kelsey, D. R. J . Org. Chem. 1986, 51, 2627.
(20) Grob, M. C.; Feiring, A. E.; Auman, B. C.; Percec, V.; Zhao,
M.; Hill, D. H. Macromolecules 1996, 29, 7284.
(21) Sperling, L. H. Introduction to Physical Polymer Science, 3rd
ed.; J ohn Wiley and Sons: New York, 2001; pp 63-75.
(22) Friedel, P.; J ohn, A.; Popspiech, D.; J ehnichen, Netz, R. R.
Macrmol. Theory Simul. 2002, 11, 785.
(23) Li, W.; Gersappe, D. Macromolecules 2001, 34, 6783.
(24) Zweifel, H., Ed. Plastics Additives Handbook, 5th ed.; Hanser/
Gardner: Cincinnati, OH, 2001; pp 935-939.
MA0352983