Substituted poly(p-phenylene) thin films via Surface-Initiated Kumada-Type catalyst transfer polycondensation

Article abstract of DOI:10.1021/ma902710j

Conjugated polymer films of unsubstituted and alkoxy-substituted poly(p-phenylene) have been prepared via a Kumada-type catalyst-transfer polycondensation in a grafting-from configuration. A surface-bound external initiator was formed by reacting a Ni(O) complex with a thienyl bromide monolayer, and this surface-bound initiator was used to polymerize several Grignard monomers prepared from the corresponding 1,4-dihalobenzenes resulting in surface-bound, conjugated polymer chains with thickness up to 30 nm. A series of 1,4-diodo monomers with alkoxy side chains ranging from unsubstituted 1,4-diiodobenzene to 1,4-diiodo-2,5-dihexyloxybenzene were polymerized in order to explore the influence of steric bulk on the surface-initiated polymerization process. Within the series of molecules, it was observed that monomers with smaller, less bulky side chains were more easily polymerized from the surface, resulting in smooth, regular films for unsubstituted and methoxy-substituted polyphenylene. Islands of polymer growth were observed with the ethoxy-substituted material and only sparse, irregular growth for the hexyloxysubstituted polymer. The resulting films were characterized by AFM., infrared spectroscopy, ellipsometry, and UV-vis spectroscopy.

Full text of DOI:10.1021/ma902710j

Macromolecules 2010, 43, 2137–2144 2137
DOI: 10.1021/ma902710j
Substituted Poly(p-phenylene) Thin Films via Surface-Initiated
Kumada-Type Catalyst Transfer Polycondensation
Nicholas Marshall, S. Kyle Sontag, and Jason Locklin*
Department of Chemistry, Faculty of Engineering, and Center for Nanoscale Science and Engineering,
University of Georgia, Athens, Georgia 30602
Received December 8, 2009; Revised Manuscript Received January 21, 2010
ABSTRACT: Conjugated polymer films of unsubstituted and alkoxy-substituted poly(p-phenylene) have
been prepared via a Kumada-type catalyst-transfer polycondensation in a grafting-from configuration. A
surface-bound external initiator was formed by reacting a Ni(0) complex with a thienyl bromide monolayer,
and this surface-bound initiator was used to polymerize several Grignard monomers prepared from the
corresponding 1,4-dihalobenzenes resulting in surface-bound, conjugated polymer chains with thickness
up to 30 nm. A series of 1,4-diodo monomers with alkoxy side chains ranging from unsubstituted
1,4-diiodobenzene to 1,4-diiodo-2,5-dihexyloxybenzene were polymerized in order to explore the influence
of steric bulk on the surface-initiated polymerization process. Within the series of molecules, it was observed
that monomers with smaller, less bulky side chains were more easily polymerized from the surface, resulting in
smooth, regular films for unsubstituted and methoxy-substituted polyphenylene. Islands of polymer growth
were observed with the ethoxy-substituted material and only sparse, irregular growth for the hexyloxy-
substituted polymer. The resulting films were characterized by AFM, infrared spectroscopy, ellipsometry,
and UV-vis spectroscopy.
Introduction
nanoscale object exist. Such methods, however, have not been
forthcoming; few techniques have been developed for controll-
Conjugated polymers are an important subject of current
materials research due to their unique properties and applications
in solar cells, polymer light-emitting devices, sensors, and prin-
table circuitry.1-3 Many of the most important applications for
this class of material involve the use of a conjugated polymer in
thin film device architectures, for either their optical or electronic
properties. A distinct advantage of conjugated polymers is the
ability to tune their electronic properties through synthetic
design. However, synthetic design of conjugated polymers is
inherently constrained by the need to confer processability,
usually by adding bulky, aliphatic side chains to monomers.
Modifying a monomer in this way generally has the effect of
lowering the conductivity of the polymer, but use of monomers
without such modifications generally leads to intractable poly-
mers with poor solubility and low molecular weight. When
solution processing is limited, few techniques for formation of
conjugated polymer thin films exist.4,5 Among existing techni-
ques, oxidative electropolymerization of an electron-rich mono-
mer onto an electrode surface is the most common; however, this
method is only applicable to conducting substrates and suffers
from a lack of regiospecificity and morphological control. New
routes to conjugated polymer films are highly desirable and can
provide access to materials with electronic properties that are not
possible with current synthetic methodologies due to the inherent
insolubility that limits processability.
ably forming processable conjugated polymers and even fewer for
attaching them at a surface or to a molecular species.7-11 Among
the techniques that have been reported, the Kumada-type cata-
lyst-transfer chain growth polymerization of McCullough12,13
and Yokozawa14,15 shows considerable promise, having been
used to prepare end-functionalized polymers,16-19 polymers
grown through surface-initiated polymerization,20,21 and a vari-
ety of conjugated polymers and block copolymers in solution.22 A
related technique, Suzuki-type chain growth polycondensation as
first reported by Yokozawa et al.,23 has also been used for the
synthesis of several conjugated polymers and in the surface-
initiated polymerization of a 9,9-dialkylfluorene monomer.24
Kumada-type catalyst-transfer polycondensations depend on
a unique intramolecular reaction that occurs within the catalytic
cycle (Scheme 1). In a catalytic cycle similar to that in most metal-
mediated aryl-aryl coupling reactions, a Ni(0) species undergoes
oxidative addition to a carbon-halogen bond to form an
organometallic Ni(2þ) complex. This complex then undergoes
a transmetalation step with a source of nucleophilic carbon,
which results in an organometallic species containing two aryl
groups bound directly to the nickel center.25,26 This unstable
intermediate, following a cis-trans isomerization, can reduc-
tively eliminate a Ni(0) species with any attendant ligands,
forming an aryl-aryl C-C bond as a byproduct. A considerable
body of indirect evidence suggests that the eliminated biaryl
compound is initially coordinated to the Ni(0) center.27,28
Furthermore, the emergence of nanotechnology has led to an
interest in the synthesis of nanowires for interfacing species at the
nanoscale such as enzymes, molecules, and nanoparticles with
macroscale electrodes.6 Conjugated polymers are excellent candi-
dates for these applications as long as appropriate methods for
connecting the polymers to both the surface and the molecule or
In most applications of Kumada coupling, this initial coordi-
nation is unimportant, as the product ring quickly dissociates
from the nickel center or is displaced by more powerfully
coordinating ligands. However, it has been established that
coordination of the aromatic ring to the Ni center is the first
irreversible step in oxidative addition of the catalyst to the Ar-X
bond.29 Therefore, if the ring is functionalized with an aryl halide
*To whom correspondence should be addressed. E-mail: jlocklin@
chem.uga.edu.
r 2010 American Chemical Society
Published on Web 02/03/2010
pubs.acs.org/Macromolecules

2138 Macromolecules, Vol. 43, No. 5, 2010
Scheme 1. A Catalyst-Transfer Step in the Kumada Coupling Synthesis of a Polyphenylene
Marshall et al.
group of its own, the Ni(0) species will oxidatively add to that
Ar-X bond instead of being displaced, and the catalytic cycle will
continue. For the polymerization of an X-Ar-MgCl monomer
via the Kumada reaction, this intramolecular chain transfer
mechanism can lead to a chain-growth polymerization and is a
suitable strategy for surface-initiated polymerization if appro-
priate surface-bound initiators can be made.
In a recent report, we described a surface-initiated polymeri-
zation of poly(thiophene) and poly(p-phenylene) from a thienyl
bromide monolayer on gold via SI-KCTP.21 We were interested
in further developing this technique, using monomers with a
variety of side chains in order to tune the properties of the films
formed and exploring the use of different catalyst systems and
initiators to achieve greater control of the system by reducing
termination reactions. To this end, we attempted the polymeri-
zation of several 1,4-dihalo-2,5-dialkoxybenzene monomers
using a Kumada-type catalyst-transfer polycondensation from
a surface-bound nickel initiator. The preparation of this surface-
bound initiator follows a strategy similar to reports of externally
initiated polymerizations in solution by Luscombe et al. for
nickel-based initiators30 and Yokozawa et al. for palladium-
based initiators.23
Early attempts to prepare surface-grafted conjugated polymer
films from 1,4-dihexyloxybenzene monomers in our laboratory
only produced films with a nominal thickness of 1-2 nm after
polymerization. We attempted several different surface-bound
initiator metal/ligand combinations, which all gave similar
results. After multiple trials, we speculated that the poor perfor-
mance of this polymerization reaction from the surface might be
due to the steric bulk of the hexyl side chain, attached ortho to the
halomagnesium moiety on the monomer. To test this hypothesis,
we synthesized aryl monomers substituted with smaller alkoxy
side chains and found that they were readily polymerized from
the initiator-functionalized surface to yield conjugated polymer
films.
We believe that this technique enjoys considerable advantages
over electrochemical methods for the formation of surface-bound
conjugated polymers due to the more controlled nature of chain-
transfer coupling polycondensation as opposed to electrochemi-
cal oxidation, the ability to form undoped polymers, and the
applicability of the technique to the formation of polymer films
on nonconducting substrates.
1,4-Dibromo-2,5-dimethoxybenzene and 1,4-dibromo-2,5-
dihexyloxybenzene were synthesized from the corresponding
1,4-dialkoxybenzenes according to literature procedures.32
Synthetic Methodology. All syntheses were carried out under
an inert atmosphere of purified argon or nitrogen, using stan-
dard Schlenk techniques or a glovebox (Unilab BP with an
MB10 purification system, MBraun, Inc.). NMR spectra were
recorded using a Varian Mercury 300 NMR spectrometer
working at 300 MHz. Chemical shifts are reported relative to
an internal tetramethylsilane standard.
2-Allyl-5-bromothiophene. 2.0 g (0.93 mL, 8.3 mmol) of 2,5-
dibromothiophene was dissolved in 20 mL of THF and cooled in
an ice bath. 5.0 mL (8.3 mmol) of freshly titrated 1.65 M
isopropylmagnesium chloride solution in THF was added drop-
wise from a syringe, and the reaction was stirred 1 h. 0.9 mL
(10.4 mmol) of allyl bromide was added dropwise from a
syringe, and the reaction was stirred and allowed to come to
room temperature overnight. The THF was removed in vacuo,
and the resulting oil was extracted with 3 ꢀ 20 mL portions of
hexane. The organic phase was washed three times with de-
ionized water and once with brine, dried over anhydrous sodium
sulfate, and the solvent removed in vacuo. The resulting yellow
oil was purified by column chromatography (hexanes:silica) to
yield 0.592 g (35%) of a clear oil. 1H NMR (CDCl3, 300 MHz) δ
(ppm): 6.86 (d, 2H, J = 3.5 Hz), 6.56 (d, 2H, J = 3.7 Hz), 5.93
(m, 1H), 5.12 (m, 2H), 3.48 (d, 2H, J = 6.6 Hz).
3-(5-Bromothiophen-2-yl)propyltrichlorosilane. 0.59 g (8.3
mmol) of 2-allyl-5-bromothiophene was dissolved in 10 mL of
freshly distilled dichloromethane under nitrogen, and 10 equiv
(83 mmol, 3 mL) of trichlorosilane was added via syringe
through a rubber septum. A match-head-sized portion of chloro-
platinic acid hexahydrate was added, and the reaction was
stirred overnight. After reaction, the vessel was fitted with a
short-path distillation head, and the solvent and excess trichloro-
silane were distilled off via heating to 70 °C. The product was
subjected to vacuum at 0.5 Torr for 3 h, yielding a yellow oil which
was used without further purification. 1HNMR(CDCl3,300MHz)
δ (ppm): 6.87 (d, 2H, J = 3.6 Hz), 6.57 (d, 2H, J = 3.6 Hz), 2.86
(t, 2H, J = 7.4 Hz), 1.92 (m, 2H), 1.43 (m, 2H).
General Synthesis of 1,4-Diiodo-2,5-dialkoxybenzenes. 1,4-
Diiodo-2,5-dimethoxybenzene (I2DMB), 1,4-diiodo-2,5-di-
ethoxybenzene (I2DEB), and 1,4-diiodo-2,5-dihexyloxybenzene
(I2DHB) were synthesized according to the following proce-
dure: 6.63 g of hydroquinone (60.2 mmol) was dissolved in 50
mL of DMF, and 14 g of powdered KOH was added. The
reaction was stirred for 1 h, and 180 mmol of alkyl bromide was
added in one portion. The reaction was stirred overnight and
quenched by pouring over ice. The product was collected by
filtration, washed with 1 M NaOH solution and water, and
recrystallized from ethanol to give white flakes. (Commercially
available 1,4-dimethoxybenzene was used, so this step was not
performed in the synthesis of 1,4-diiodo-2,5-dimethoxybenzene.)
36 mmol of 1,4-dialkoxybenzene was dissolved in 36 mL of
carbon tetrachloride, and 180 mL of glacial acetic acid was
added. 9.05 g of iodine (36 mmol) was added, followed by 6 mL
of sulfuric acid and 18 mL of water. 4.04 g (18 mmol) of periodic
acid was added in one portion, and the reaction was heated to
gentle reflux overnight under nitrogen. The reaction was cooled
and diluted with 200 mL of dichloromethane, and the organic
layer was collected and washed 3 times with water, 1 time with
saturated sodium bicarbonate solution, 2 times with saturated
Experimental Section
Materials. 1,4-Diiodo-2,5-dialkoxybenzenes were synthe-
sized according to literature procedures as described below.31
Isopropylmagnesium chloride solution (2.0 M in THF), bis-
(cyclooctadiene)nickel(0), and periodic acid were obtained from
Sigma-Aldrich. 2,5-Dibromothiophene, allyl bromide, ethyl
bromide, hexyl bromide, 1,4-dimethoxybenzene, 1,4-diiodoben-
zene, hydroquinone, KOH, iodine, trichlorosilane, H2PtCl6,
sulfuric acid, acetic acid, carbon tetrachloride, tetrahydrofuran
(THF), toluene, dimethylformamide (DMF), dichloromethane,
and glass slides were obtained from VWR. Toluene and THF
were distilled from sodium ketyl and degassed by sparging
with Ar before use. Quartz substrates were obtained from
Technical Glass Products, Inc. (Painesville Township, OH).
Unless indicated otherwise, materials were used as received.

Article
Macromolecules, Vol. 43, No. 5, 2010 2139
sodium thiosulfate solution, and 1 time with brine. The organic
layer was dried over anhydrous sodium sulfate, and the solvent
was removed in vacuo. The resulting yellow solid was recrys-
tallized from chloroform/methanol. Yield: (I2DMB) 9.0 g, 60%
for one step; (I2DEB) 8.5 g, 34% overall; (I2DHB) 7.2 g, 23%
overall. 1H NMR (CDCl3, 300 MHz) δ (ppm): (I2DHB) 7.17 (s,
2H), 3.93 (t, 4H, J = 6.5 Hz), 1.80 (p, 4H, J = 6.5 Hz), 1.50 (p,
4H, J = 7.4 Hz), 1.35 (m, 8H), 0.91 (t, 6H, 7.0 Hz). (I2DEB) 7.19
(s, 2H), 4.01 (q, 4H, J = 7.1 Hz), 1.43 (t, 6H, 7.0 Hz). (I2DMB)
7.19 (s, 2H), 3.82 (s, 6H).
Preparation of Initiator Wafers. Silicon wafers (Silicon Quest)
and glass slides (VWR) were cut into 2 ꢀ 3 cm rectangles and
subjected to plasma cleaning using argon plasma, using the
“high” setting on a plasma cleaner (Harrick Plasma) for 3 min.
The substrates were then treated for 15 min in a UV-ozone
cleaner and transferred to a nitrogen-filled glovebox. 40 μL of
3-(5-bromothiophen-2-yl)propyltrichlorosilane was dissolved
in 10 mL of toluene, and the solution was poured over the
substrates, which were capped and left for 12 h. The substrates
were then removed from the solution, sonicated 1 min in
toluene, rinsed 3 times with toluene, sonicated 1 min in isopropyl
alcohol, rinsed with isopropyl alcohol, sonicated 1 min in
chloroform, and rinsed several times with chloroform. The
substrates were dried in a stream of nitrogen gas and transferred
back inside a glovebox.
Magnesiation of 1,4-Dihalo-2,5-dialkoxybenzene Monomers.
1 mmol of monomer was dissolved in 10 mL of freshly distilled
THF in a Schlenk flask under nitrogen, and 0.9 equiv of freshly
titrated isopropylmagnesium chloride solution was added drop-
wise from a syringe at RT. For aryl iodide-based monomers, the
reaction was stirred 1 h. For aryl bromides, the reaction was
stirred 24 h as per Yokozawa et al.32 After the given reaction
time, the Schlenk flask was sealed and transferred to a glovebox.
For polymerizations in the presence of LiCl salt, 42 mg (1 equiv)
of LiCl was added, and the monomer solution was stirred 10 min
before addition of external initiator-functionalized substrates.
The efficiency of magnesium-iodine exchange was verified by
quenching a portion of Grignard monomer with dimethylchloro-
silane. 1H NMR of the product confirmed complete conversion
of the 1,4-diiodo-2,5-dihexyloxybenzene monomer to 1-iodo-
4-dimethylsilyl-2,5-dihexyloxybenzene. Alternatively, a com-
mercial solution of isopropylmagnesium chloride lithium chlor-
ide complex (Aldrich, 1.3 M in THF) was also used for
magnesiation,32 without a noticeable difference in the resulting
films.
Formation of Surface-Bound External Initiator and Polymer-
ization. A toluene solution 20 mmol in Ni(COD)2 and 20 mmol
in 1,2-bis(diphenylphosphino)ethane (dppe) was made up by
dissolving 55 mg of Ni(COD)2 and 80 mg of dppe in 10 mL of
toluene inside a nitrogen-filled glovebox. Thienyl bromide-
functionalized slides were placed in a glass slide holder, and
the bright yellow Ni complex solution was poured over them.
The slides were allowed to react for 1 h, then the solution was
removed, and the slides were washed three times with fresh
toluene. The slides were immediately transferred to a solution of
Grignard monomer to begin polymerization. Polymerizations
were allowed to proceed overnight, after which time the sub-
strates were removed, washed with acetone, water, acetone, and
toluene, and then sonicated in chloroform to remove any
physisorbed polymer. After rinsing with chloroform, the films
were dried in a stream of nitrogen and characterized.
for all polymer layers. AFM imaging was performed in tapping
mode on a Multimode Nanoscope IIIa (Digital Instruments/
Veeco Metrology Group) using silicon AFM probes with a
nominal spring constant of 40 N/m and a resonant frequency
of 300 kHz. The scan rate used was 1 Hz. UV-vis spectra of
polymer substrates were taken using a Cary 50 UV-vis spectro-
photometer from Varian. Variable angle polarized light UV-vis
was performed using a custom-built variable angle sample
holder fitted with a rotatable polarizing filter. Infrared spectra
of films were acquired using a Thermo-Nicolet Model 6700
spectrometer equipped with a variable angle grazing angle
attenuated total reflection (GATR-ATR) accessory (vari-
GATR, Harrick Scientific) and processed using OMNIC 8.0
software (ThermoNicolet, Madison, WI).
Results and Discussion
Preliminary experiments on the surface-initiated polymeriza-
tion of Grignard monomers based on substituted 1,4-haloben-
zenes indicated that the thickness of the resulting films is strongly
influenced by the size of the side chain. In order to systematically
examine the effect of side chain bulk on surface-initiated Kumada
catalyst transfer polymerization, we used a series of p-diiodoben-
zene monomers with alkoxy side chains of various lengths as well
as the parent compound p-diiodobenzene. The substituted diiodo-
benzenes were prepared by standard electrophilic iodination
of the corresponding p-dialkoxybenzene. Aryl iodide-based
p-dialkoxy monomers were chosen for this project because of
our initial finding that the Grignard monomer based on unsubsti-
tuted p-diiodobenzene reacted readily to form poly(p-phenylene)
films with a related catalyst system.21 Magnesium-halogen
exchange of one iodide substituent to form the Grignard mono-
mer proceeds readily even in the presence of electron-donating
alkoxy substituents on the ring. Although Yokozawa et al. report
the successful chain-growth polymerization of the related aryl
bromide monomer 1,4-dibromo-2,5-dihexyloxybenzene,32 we did
not observe polymer growth on the surface with either this mono-
mer or its less hindered relative 1,4-dibromo-2,5-dimethoxyben-
zene. It is not entirely clear why the aryl bromide monomers fail
to form surface-bound polymer, but it may be due to the termi-
nating chain-transfer reaction that can occur in the presence
of unreacted isopropylmagnesium chloride. Another possibility is
that oxidative addition of the Ni(0) species to the stronger C-Br
bond is sufficiently slow that the catalyst reductively eliminates and
transfers off the chain rather than performing a catalyst-transfer
step.
Design and Synthesis of the Surface-Bound Initiator. The
aryl bromide monolayer used to prepare the external initia-
tor in this set of experiments was prepared using a simple
two-step synthesis outlined in Scheme 2, consisting of
magnesium-halogen exchange on the commercially avail-
able 2,5-dibromothiophene followed by alkylation of the
aryl Grignard with allyl bromide. The alkylation step has an
inherently low yield due to the sluggish activity of Grignard
reagents in nucleophilic substitution. Ordinarily, a Cu cata-
lyst is added to such reactions to catalyze alkylation,33 but in
this case this technique results in polymerization of the aryl
bromide Grignard intermediate. Hydrosilylation of the
alkene produced by the first step using H2PtCl6 as a catalyst
yields a trichlorosilane bearing an alkyl chain terminated
with the reactive thienyl bromide moiety. Formation of
monolayers of this product by immersing activated silicon
oxide surfaces in a toluene solution of the trichlorosilane
results in a surface-bound species from which an external
initiator for SI-KCTP can be prepared.
Thin Film Measurements. Null ellipsometry measurements of
thin films were performed on a Multiskop (Optrel GbR) with a
632.8 nm HeNe laser light source at 70°, and film thicknesses
were determined using integrated specific software. Film thick-
ness was obtained by fitting ellipsometric data using manufac-
turer-provided specialized software. At least three spots on each
wafer were measured, and the thickness was averaged. To obtain
thickness values of the samples, a simple box model was
employed and a refractive index of n = 1.5, k = 0 was used
The Ni(0) catalyst solution most likely consists of a Ni(0)
center complexed to a single dppe ligand and a single
remaining COD ligand,34 although an equilibrium certainly

2140 Macromolecules, Vol. 43, No. 5, 2010
Scheme 2. Synthesis of Thienyl Bromide Initiator and p-Alkoxybenzene Monomers (R = Me, Et, Hexyl)
Marshall et al.
Scheme 3. Synthesis of Surface-Bound Conjugated Poly(dialkoxy-p-phenylene)s
exists between the different complexed forms of Ni(0) due to
the rapid dissociation of ligands, including phosphines, from
zerovalent nickel.35 Luscombe et al.36 report that the Ni-
(COD)2 þ dppe catalyst system fails to react with chloro-
benzene to produce an external initiator. We attribute the
success of this catalyst system in our experiments to the
superior reactivity of 2-bromothiophenes over chloroben-
zene toward oxidative addition of the Ni(0) species.
Although oxidative addition of Ni(0) to thienyl halides has
not been thoroughly studied, it is known that Ni(PPh3)4 adds
rapidly at room temperature to 2,5-dibromothiophene37 and
that a Ni(0) catalyst generated from Ni(COD)2 and 4 equiv
of PPh3 adds quickly to 2-bromothiophene.21 Furthermore,
oxidative addition of Ni(0) to aryl halides has been reported
by Kochi et al.35 to follow a Hammett correlation such that
electron-withdrawing substituents accelerate the rate of re-
action. 2-Bromothiophenes are known to react rapidly in
reactions such as magnesium-halogen exchange38 which are
similarly assisted by electron-withdrawing groups, due in
part to the heterocyclic sulfur atom acting as a π-acceptor.
Finally, Luscombe et al.36 report that although the Ni-
(COD)2 þ dppe system does not oxidatively add to chloro-
benzene to produce an external initiator, it does react
with the Grignard monomer 2-bromo-5-chloromagnesio-3-
hexylthiophene to form polymer, implying that this catalyst
system is capable of oxidatively adding to 2-bromothio-
phenes even in the presence of a strong electron donor at
the 5-position.
Table 1. Polyphenylene Film Thicknesses As Determined by Ellip-
sometrya
film thickness
(nm)
rms roughness
(nm)
λmax
(nm)
absorbance
monolayer
PPh
PDMP
PDEP
3.6 ( 0.1
26.4 ( 0.1
17.0 ( 0.1
12 ( 1
0.9
7.9
3.1
6.0
2.4
371
307
347
347
0.4
0.08
0.04
0.01
PDHP
4.8 ( 0.1
a Each value is the average of three or more measurements. The rms
roughness was determined by analysis of corresponding AFM topo-
graphy images. λmax and absorbance values were obtained from UV-vis
measurements of polymers grown on quartz.
surface-bound catalyst yielded a 30 nm thick polymer film, as
measured by both ellipsometry and surface profiling with
AFM. The film displayed the characteristic blue fluorescence
of PPh. The new catalyst therefore appeared effective, and
we performed polymerizations with a variety of substituted
p-diiodobenzene monomers in order to investigate the effect
of monomer substituents on the surface-initiated polymeri-
zation reaction (Scheme 3).
The morphology, thickness, and surface coverage of the
polymer film produced by a SI-KCTP reaction depend
strongly on the size of the alkoxy substitutent on the aryl
monomer, as seen in ellipsometric thickness measurements
and AFM micrographs of the resulting films in each case.
The unsubstituted and methoxy-substituted monomers ap-
pear to add readily to the nickel-functionalized surface.
Table 1 shows thickness measurements of the polymer films
as determined by ellipsometry. We observed an average
value of 26 nm for PPh, 14 nm for PDMP, 8 nm for PDEP,
and 1 nm for PDHP, after subtracting the thickness of the
monolayer. The surface-initiated polymerization technique
applied to PPh in this work overcomes several inherent
limitations of existing deposition techniques, directly form-
ing thick PPh films on a nonconducting oxide surface.
Formation of Polymer Films. We first tested the surface-
bound catalyst system by using it to initiate SI-KCTP of
4-chloromagnesio-1-iodobenzene, a Grignard monomer
which we have previously shown to react with a surface-
bound catalyst prepared from Ni(COD)2 þ 4PPh3 to form
poly(p-phenylene) (PPh) films.21 The Grignard monomer is
formed easily at room temperature on addition of iPrMgCl
to p-diiodobenzene. Reaction of this monomer with the new

Article
Macromolecules, Vol. 43, No. 5, 2010 2141
Figure 1. UV-vis spectra of polymer thin films on quartz.
Figure 2. GATR-FTIR spectrum of PPh film, with arrows indicating
the location of bands corresponding to out-of-plane bending of the
main chain and end group.
Spectroscopic and AFM Analysis of Polymer Films.
UV-vis spectroscopy of the polymer films (Figure 1) grown
on quartz substrates correlates well with the observed thick-
ness of the films grown on silicon as measured by ellipso-
metry and AFM. The absorbance intensity of the lowest-
energy λmax follows the trend APPh . APDMP > APDEP
.
APDHP and is shown in Table 1 for films prepared under
identical conditions. The PPh film has an absorbance at 371
nm, which is a good match for previous literature reports of
the UV-vis spectrum for thin films of this polymer.39,40 The
appearance of fine structure (shoulder at 425 nm) indicates
crystalline structure in the film. The absorption onset at 440
nm (2.82 eV) is also slightly lower than that observed in other
cases.41-44 The positions of the observed absorptions for the
poly(alkoxy)phenylenes are consistent with the known elec-
tronic structure of the polymers, which are high-band-gap
semiconductors.45 PDEP and PDHP have similar lowest-
energy absorption maxima at 347 nm, which is consistent
with that reported for PDEP46 but lower than that reported
for PDHP (373 nm).45 The reason for this is probably the
very low degree of polymerization that occurs for PDHP
using this surface-initiated technique, as evidenced by the
low film thickness (2.4 nm) of PDHP. For films of PDMP, a
low-energy absorbance maximum of λ = 307 nm is observed.
Electropolymerized PDMP has previously been prepared by
Lamy et al.47 on a Pt electrode and characterized in situ
through spectroelectrochemical techniques. These electro-
polymerized films show a strong absorption ca. 435 nm,
which we do not observe with PDMP films grown by SI-
KCTP. The difference in film properties of PDMP may lie
with the ability to form PDMP in an entirely undoped form
using SI-KCTP, without recourse to the use of Lewis acids or
electrochemical oxidative polymerization.
Figure 3. FTIR spectra of alkoxy-substituted polymer films and mono-
layer. From top to bottom: PDMP, PDEP, PDHP, and initiator
monolayer.
out-of-plane bending peak to the intensity of the lower-
frequency out-of-plane bending modes for the monosubsti-
tuted end group.43 The peaks are indicated in Figure 2 by
arrows. The DP is then given by DP = 2R þ 2. R in this case
is ∼34, giving a DP of 68. Molecular modeling shows that the
˚
length of one repeat unit of PPh is ∼3.6 A, which for an
extended conformation of the polymer gives a chain length of
∼24 nm. This value is a reasonable match for the 26 nm
polymer film as measured by ellipsometry (Table 1).
Figures 2 and 3 show the infrared spectrum of unsubsti-
tuted and alkoxy-substituted PPh films grown on SiO2
substrates as measured by GATR-FTIR at a grazing angle
of 63°. Unsubstituted PPh (Figure 2) has a CdC asymmetric
ring mode at 1480 cm-1, an in-plane C-H bend at 1001
cm-1, and a very strong and distinctive C-H out-of-plane
Figure 3 shows the GATR-FTIR spectra of each alkoxy-
substituted phenylene used in this study. The spectra are not
baseline corrected, as the intensity of each absorbance
corresponds directly to the film thickness. The C-H stretch-
ing region is magnified 3ꢀ. The FTIR spectrum of the
monolayer is also shown in Figure 3 in order to ascertain
the absorbances due to the underlying silicon oxide layer.
Broad absorptions at 900 and 800 cm-1, as well as a broad
bend at 810 cm-1. An intense, broad absorption at 900 cm-1
,
and a broad absorption centered at 1100 with a shoulder at
1250 cm-1, are due to the underlying silicon oxide surface.
Analysis of the C-H out-of-plane bending region allows for
an estimate of the degree of polymerization of the PPh film,
based on the ratio R of the intensity of the C-H p-substituted
absorption centered at 1100 with a shoulder at 1250 cm-1
,
are due to the underlying silicon oxide surface. Visible in the
C-H region of PDMP is a strong peak at 2995 cm-1

2142 Macromolecules, Vol. 43, No. 5, 2010
Marshall et al.
corresponding to an aromatic C-H mode and distinguish-
able from the alkyl C-H stretching bands at 2955 and 2930
cm-1. Relatively low-frequency C-H vibrations at 2853 and
2831 cm-1 are also distinct from other alkyl vibrations and
are attributed to asymmetric C-H modes in the aryl methyl
ether. Similar peaks are visible in the C-H region in PDEP.
A shoulder is seen near 2995 cm-1 likely due to aromatic
C-H stretching, and the peak at 2970 cm-1 is due to a methyl
C-H stretching mode. Lower-frequency symmetric and
asymmetric methylene vibrations are found at 2927 and
2870 cm-1. Four CdC aromatic ring stretching bands are
visible at 1496, 1467, 1444, and 1401 cm-1 in PDMP and
1497, 1475, 1443, and 1397 cm-1 in PDEP. An intense
C-O-C stretch is visible at 1210 cm-1 in PDMP and at
1215 cm-1 in PDEP and is barely visible in PDHP as a
shoulder at 1215 cm-1. This mode is the only vibration with a
sufficiently intense signature to be distinguished from the
monolayer in the inefficiently grafted PDHP layer. Around
1060 cm-1 in both PDMP and PDEP are several sharp peaks
consistent with alkyl C-O stretching modes in aryl ethers. A
sharp peak at 805 cm-1 in both PDMP and a shoulder at the
same frequency in PDEP is attributed to the aromatic C-H
out-of-plane bend.
We have investigated the morphology and thickness of the
conjugated polymer films formed via SI-KCTP using AFM.
The PPh film has a regular but somewhat globular morpho-
logy, similar to that observed in PPh and polythiophene films
grown from metal substrates using thiol-based aryl bromide
initiators.21 This globular morphology is uniform over large
areas and appears typical of unsubstituted conjugated poly-
mers prepared via SI-KCTP. The rms roughness of the PPh
film is 7.9 nm, and the film does not show the ultrasmooth
morphology which is indicative of fully stretched polymer
chains with high initiation efficiency.
The methoxy-substituted monomer also reacts readily
with the surface-bound catalyst, producing a thick polymer
film with a remarkably smooth, regular morphology.
(Figures 5 and 6, Table 1). The rms roughness of the resulting
PDMP film is 3.1 nm. The ethoxy-substituted monomer
reacts with the surface-bound catalyst less effectively, pro-
ducing an unusual surface morphology visible via AFM
consisting of isolated islands of complete surface coverage
resembling that seen in the methoxy derivative (Figures 5 and
6). Between these areas are regions of lower coverage,
containing isolated polymer features 20-30 nm in height.
The hexyl-substituted monomer adds even less efficiently,
producing only sparsely distributed 20 nm islands of polymer
Figure 4. AFM topography image (1 ꢀ 1 μm) of a 30 nm PPh film.
Figure 5. AFM topography images (1 ꢀ 1 μm) of thin films of PDMP (left), PDEB (center), and PDHB (right).
Figure 6. Large area topography scan (5 ꢀ 5 μm) of PDMP film formed in the presence of LiCl salt (left) and without LiCl (right). The film made with
added LiCl is substantially smoother and more regular than the film without the additive.

Article
Macromolecules, Vol. 43, No. 5, 2010 2143
(Figure 5, right), which are nearly undetectable by FTIR and
absorb only weakly in the UV.
remain unreacted in solution. In the case of the aryl iodide
monomers used in this study, a slight excess of aryl diiodide
was added to ensure that no isopropylmagnesium chloride
would remain during the polymerization. The presence of
excess aryl halide has been shown to have little or no effect on
the catalyst-transfer step in Kumada coupling reactions.50 It
is also possible that slow oxidative addition of the nickel
catalyst to the stronger C-Br bond in the aryl bromide-
based monomer is leading to reductive elimination of the
nickel catalyst competing with intramolecular chain transfer.
Further exploration of this issue is underway in order to
determine optimal catalyst/monomer systems and polymer-
ization conditions.
Under all polymerization conditions with each monomer,
polymer was formed in solution and isolated by precipitation
with ethanol after 24 h, suggesting that the nickel catalyst
does not remain permanently associated with the surface-
bound polymer chain. This finding is consistent with the
results of Rawlins and Achord,51 who observe that the
synthesis of P3HT using the GRIM method with Ni-
(dppp)Cl2 as a catalyst does not proceed in a perfectly living
fashion, but rather reaches a maximum molecular weight
determined by monomer concentration. This result suggests
that it may be possible to tune film thickness by varying
monomer concentration. We will explore this possibility in
future work.
Effect of LiCl and Halide Type on SI-KCTP Preparation
of Polyphenylene Films. In the case of poly(p-alkoxy)-
phenylenes synthesized in solution using a Ni(2þ) catalyst,
Yokozawa et al. have found that the addition of lithium
chloride to the reaction mixture before magnesium-halogen
exchange is beneficial in the synthesis of PDHP from
1,4-dibromo-2,5-dihexyloxybenzene, producing a polymer
with higher molecular weight and lower polydispersity.32
Lanni and McNeil, on the other hand, have shown that under
their reaction conditions the polymerization of this monomer
is essentially unaffected by the presence of LiCl.48 The discre-
pancy between these two reports is attributed to a difference
in the initiators used in the two experiments; in the Yokozawa
report, a suspension of Ni(dppe)Cl2 was added directly to
the Grignard monomer, while Lanni and McNeil added
5-7 equiv of Grignard monomer to the catalyst solution before
beginning the main polymerization. It was suggested that the
difference is due to the effect of LiCl on the reaction of the
Grignard monomer with Ni(dppe)Cl2.48 Because of the differ-
ence in these results, we were interested in determining the
effect of LiCl on the surface-initiated polymerization of poly-
phenylenes.
We investigated the effect of LiCl on the formation of
PDMP films by adding LiCl at various points in the experi-
ment. The Grignard monomer solutions were formed three
ways: (1) by reacting the diiodobenzene species with com-
mercially available iPrMgCl-LiCl solution, (2) adding LiCl
to a solution of Grignard monomer prepared from the
corresponding diiodobenzene and iPrMgCl, and (3)
Grignard monomer prepared from the diiodobenzene and
iPrMgCl without any added LiCl salt. We have found no
significant effect of LiCl on the average thickness of the
resulting polymer films, obtaining a value of ca. 20 nm from
ellipsometry in all cases. However, the film produced from
monomer without added LiCl had a considerably rougher
surface morphology, with a rms roughness of 7.4 nm com-
pared to 3.1 nm for the film made from monomer with added
LiCl (Figure 6). This result is consistent with Yokozawa’s
report of improved polydispersity on adding LiCl in the
solution polymerization of the hexyl-substituted aryl bro-
mide monomer with Ni(dppe)Cl2.32 Lanni and McNeil
suggest that the improvement is due to enhanced reactivity
of the monomer with the catalyst, demonstrating that the
effect disappears when the polymerization is preinitiated
with a few equivalents of Grignard monomer.48 Our results
are reasonable in light of these findings. To initiate surface-
directed polymerization, the Grignard monomer must react
with the surface-bound catalyst. Surface-based initiation
appears to be a more sterically demanding process than
initiation by catalyst in solution based on the dramatic effect
of side chain bulk that we observe in this series of experi-
ments. It is well-known that LiCl has the effect of breaking
up aggregated Grignard species in solution, thus increasing
their reactivity;49 this effect could be responsible for the
smoother film formed in the presence of LiCl. The more
reactive Grignard monomer may add more readily to the
surface, resulting in more uniform initiation and lower
polydispersity of the surface-bound polymer chains.
Conclusions
We and others have encountered difficulties in forming films of
the versatile conjugated polymer poly(3-hexylthiophene) via SI-
KCTP of the Grignard monomer 2-bromo-5-chloromagnesio-
3-hexylthiophene from a monolayer.20,21,52 Kiriy et al. report the
successful surface-initiated Suzuki polymerization of 9,9-dialkyl-
fluorene,24 in which the solubilizing chains are on the 9-position
of the fluorene ring, distant from the boronic acid moiety.
Luscombe et al.53 have observed a steric effect in the solution
synthesis of P3HT; hindered 5-bromo-2-chloromagnesio-3-
hexylthiophene does not polymerize in the presence of Ni-
(dppp)Cl2 and aryl bromide-based external initiators, while the
isomer 2-bromo-5-chloromagnesio-3-hexylthiophene polymerizes
readily. In light of these results, it is possible that bulky substit-
uents adjacent to the metal or metalloid group in a surface-
initiated coupling reaction substantially hinder the reaction. This
putative effect of chain bulk would present a fundamental
problem in efforts to synthesize surface-bound polymer brushes
of conjugated polymers, since attachment of such chains to the
conjugated polymer backbone is a ubiquitous strategy for the
preparation of soluble derivatives of conjugated polymers.
Further experiments to test this hypothesis and prepare sur-
face-grafted conjugated polymers using a variety of polymeriza-
tion techniques and monomers are underway in our laboratory.
Relatively few techniques exist for the synthesis of surface-
bound conjugated polymers. The technique reported here, which
relies on simple and inexpensive catalysts and monomers, holds
considerable promise for the fabrication of new thin-film archi-
tectures and devices. Furthermore, the dramatic dependence
observed of the film morphology upon the nature of the solubiliz-
ing side chain suggests that SI-KCTP is a sterically demanding
reaction strongly affected by the substitution of the monomer.
Our results also raise the possibility of tuning the film properties
by varying the bulk or length of the side chain. Further investiga-
tion into the effect of the monomer side chain in this system is
currently under investigation in our laboratory.
We observed no polymer growth when aryl bromide-based
monomers were used regardless of whether LiCl was added,
possibly due to chain transfer induced by unreacted isopro-
pylmagnesium chloride. Yokozawa et al. report an 89%
yield for the polymerization of the aryl bromide Grignard
monomer based on 1,4-dibromo-2,5-dihexyloxybenzene,
suggesting that small amounts of magnesiating agent may
Acknowledgment. This work was supported by the Donors
of the American Chemical Society Petroleum Research Fund
(Grant 48917-DNI7).

2144 Macromolecules, Vol. 43, No. 5, 2010
Marshall et al.
References and Notes
(27) Beryozkina, T.; Senkovskyy, V.; Kaul, E.; Kiriy, A. Macromole-
cules 2008, 41, 7817–7823.
(28) Sheina, E. E.; Liu, J.; Iovu, M. C.; Laird, D. W.; McCullough,
R. D. Macromolecules 2004, 37, 3526–3528.
(29) Yoshikai, N.; Matsuda, H.; Nakamura, E. J. Am. Chem. Soc. 2008,
130, 15258–15259.
(30) Bronstein, H. A.; Luscombe, C. K. J. Am. Chem. Soc. 2009, 131,
12894–12895.
(31) Fang, Z.; Wang, S.; Zhao, L.; Xu, Z.; Ren, J.; Wang, X.; Yang, Q.
Mater. Chem. Phys. 2008, 107, 305–309.
(32) Miyakoshi, R.; Shimono, K.; Yokoyama, A.; Yokozawa, T. J. Am.
Chem. Soc. 2006, 128, 16012–16013.
(33) Knochel, P.; Dohle, W.; Gommermann, N.; Kneisel Florian, F.;
Kopp, F.; Korn, T.; Sapountzis, I.; Vu Viet, A. Angew. Chem., Int.
Ed. 2003, 42, 4302–4320.
(34) Bonrath, W.; Poerschke, K. R.; Wilke, G.; Angermund, K.;
Krueger, C. Angew. Chem. 1988, 100, 853–855.
(35) Tsou, T. T.; Kochi, J. K. J. Am. Chem. Soc. 1979, 101, 6319–6332.
(36) Doubina, N.; Stoddard, M.; Bronstein, H. A.; Jen, A. K. Y.;
Luscombe, C. K. Macromol. Chem. Phys. 2009, 210, 1966–1972.
(37) Xu, Z.; Horowitz, G.; Garnier, F. J. Electroanal. Chem. Interfacial
Electrochem. 1988, 246, 467–472.
(38) Shi, L.; Chu, Y.; Knochel, P.; Mayr, H. Org. Lett. 2009, 11, 3502–
3505.
(39) Kovacic, P.; Jones, M. B. Chem. Rev. 1987, 87, 357–379.
(40) Tieke, B.; Bubeck, C.; Lieser, G. Makromol. Chem., Rapid Com-
mun. 1982, 3, 261–268.
(41) Froyer, G.; Pelous, Y.; Maurice, F.; Petit, M. A.; Digua, A.;
Fauvarque, J. F. Synth. Met. 1987, 21, 241-247.
(42) Lee, C. H.; Kang, G. W.; Jeon, J. W.; Song, W. J.; Seoul, C. Thin
Solid Films 2000, 363, 306–309.
(1) McQuade, D.; Pullen, A.; Swager, T. Chem. Rev. 2000, 100, 2537–
2574.
(2) Coakley, K.; McGehee, M. Chem. Mater. 2004, 16, 4533–4542.
(3) Gunes, S.; Neugebauer, H.; Sariciftci, N. Chem. Rev. 2007, 107,
1324–1338.
(4) Jhaveri, S. B.; Carter, K. R. Langmuir 2007, 23, 8288–8290.
(5) Ocafrain, M.; Tran, T. K.; Blanchard, P.; Lenfant, S.; Godey, S.;
Vuillaume, D.; Roncali, J. Adv. Funct. Mater. 2008, 18, 2163–2171.
(6) Willner, B.; Katz, E.; Willner, I. Curr. Opin. Biotechnol. 2006, 17,
589–596.
(7) Liu, J.; Tanaka, T.; Sivula, K.; Alivisatos, A. P.; Frechet, J. M. J.
J. Am. Chem. Soc. 2004, 126, 6550–6551.
(8) Zhang, Q.; Russell, T. P.; Emrick, T. Chem. Mater. 2007, 19, 3712–
3716.
(9) Xu, J.; Wang, J.; Mitchell, M.; Mukherjee, P.; Jeffries-El, M.;
Petrich, J. W.; Lin, Z. J. Am. Chem. Soc. 2007, 129, 12828–12833.
(10) Skaff, H.; Sill, K.; Emrick, T. J. Am. Chem. Soc. 2004, 126, 11322–
11325.
(11) Liu, S.; Tok, J. B. H.; Bao, Z. Nano Lett. 2005, 5, 1071–1076.
(12) Stefan, M. C.; Javier, A. E.; Osaka, I.; McCullough, R. D.
Macromolecules 2008, 42, 30–32.
(13) Iovu, M.; Sheina, E.; Gil, R.; McCullough, R. Macromolecules
2005, 38, 8649–8656.
(14) Miyakoshi, R.; Yokoyama, A.; Yokozawa, T. J. Am. Chem. Soc.
2005, 127, 17542–17547.
(15) Yokozawa, T.; Adachi, I.; Miyakoshi, R.; Yokoyama, A. High
Perform. Polym. 2007, 19, 684–699.
(16) Jeffries-El, M.; Sauve, G.; McCullough, R. D. Adv. Mater. 2004,
16, 1017–1019.
(17) Boudouris, B. W.; Molins, F.; Blank, D. A.; Frisbie, C. D.;
Hillmyer, M. A. Macromolecules 2009, 42, 4118–4126.
(18) Palaniappan, K.; Murphy, J. W.; Khanam, N.; Horvath, J.;
Alshareef, H.; Quevedo-Lopez, M.; Biewer, M. C.; Park, S. Y.;
Kim, M. J.; Gnade, B. E.; Stefan, M. C. Macromolecules 2009, 42,
3845–3848.
(19) Jeffries-El, M.; Sauve, G.; McCullough, R. D. Macromolecules
2005, 38, 10346–10352.
(20) Senkovskyy, V.; Khanduyeva, N.; Komber, H.; Oertel, U.; Stamm,
M.; Kuckling, D.; Kiriy, A. J. Am. Chem. Soc. 2007, 129, 6626–
6632.
(43) Aeiyach, S.; Soubiran, P.; Lacaze, P. C.; Froyer, G.; Pelous, Y.
Synth. Met. 1995, 68, 213–219.
(44) Grem, G.; Leditzky, G.; Ullrich, B.; Leising, G. Adv. Mater. 1992,
4, 36–37.
(45) Gouri Sankar, P.; Pranjol Jyoti, S.; Parameswar Krishnan, I.;
Pratima, A. Macromol. Chem. Phys. 2008, 209, 417–423.
(46) Yu, B. Z.; Li, M. K.; Lu, M.; Li, H. L. Appl. Phys. A: Mater. Sci.
Process. 2003, 76, 593–598.
ꢀ
ꢀ
(47) Choy de Martinez, M.; Marquez, O. P.; Marquez, J.; Hahn, F.;
Beden, B.; Crouigneau, P.; Rakotondrainibe, A.; Lamy, C. Synth.
Met. 1997, 88, 187–196.
(21) Sontag, S. K.; Marshall, N.; Locklin, J. Chem. Commun. 2009,
3354–3356.
(22) Yokoyama, A.; Kato, A.; Miyakoshi, R.; Yokozawa, T. Macro-
(48) Lanni, E. L.; McNeil, A. J. J. Am. Chem. Soc. 2009, 131, 16573–
16579.
(49) Krasovskiy, A.; Knochel, P. Angew. Chem., Int. Ed. 2004, 43, 3333–
3336.
(50) Sheina, E. E.; Liu, J.; Iovu, M. C.; Laird, D. W.; McCullough,
R. D. Macromolecules 2004, 37, 3526–3528.
(51) Achord, B. C.; Rawlins, J. W. Macromolecules 2009, 42, 8634–
8639.
(52) Senkovskyy, V.; Tkachov, R.; Beryozkina, T.; Komber, H.; Oertel,
U.; Horecha, M.; Bocharova, V.; Stamm, M.; Gevorgyan, S. A.;
Krebs, F. C.; Kiriy, A. J. Am. Chem. Soc. 2009, 131, 16445–16453.
(53) Boyd, S. D.; Jen, A. K. Y.; Luscombe, C. K. Macromolecules 2009,
42, 9387–9389.
molecules 2008, 41, 7271–7273.
(23) Yokoyama, A.; Suzuki, H.; Kubota, Y.; Ohuchi, K.; Higashimura,
H.; Yokozawa, T. J. Am. Chem. Soc. 2007, 129, 7236–7237.
(24) Beryozkina, T.; Boyko, K.; Khanduyeva, N.; Senkovskyy, V.;
Horecha, M.; Oertel, U.; Simon, F.; Stamm, M.; Kiriy, A. Angew.
Chem., Int. Ed. 2009, 48, 2695–2698.
(25) Diederich, F.; Stang, P.; Wipf, P. Metal-Catalyzed Cross-Coupling
Reactions; Wiley-VCH: Weinheim, 1998.
(26) Crabtree, R. The Organometallic Chemistry of the Transition
Metals; Wiley-Blackwell: Hoboken, NJ, 2009.