Experimental Evidence for the Quasi-“Living” Nature of the Grignard
Metathesis Method for the Synthesis of Regioregular
Poly(3-alkylthiophenes)
Mihaela Corina Iovu, Elena E. Sheina, Roberto R. Gil, and
Richard D. McCullough*
Department of Chemistry, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213
Received May 31, 2005; Revised Manuscript Received August 15, 2005
ABSTRACT: The Grignard metathesis (GRIM) polymerization of 3-alkylthiophenes proceeds by a quasi-
“living” chain growth mechanism, not by a step growth process. Kinetic studies of the Grignard metathesis
polymeriza tion of 2,5-dibrom o-3-alkylth iophenes showed th at the molecular weight o f poly(3-alkyl-
thiophenes) is a function of the molar ratio of the monomer to nickel initiator, and conducting polymers
with predetermined molecular weights and relatively narrow molecular weight distributions (PDIs )
1.2-1.5) can be made. Sequential monomer addition resulted in new block copolymers containing different
poly(3-alkylthiophene) segments.
Introduction
Conventional design of advanced organic materials
that display a variety of desirable properties in a
controllable way continues to be one of the great
challenges of the contemporary polymer research. One
class of these advanced materials is organic conducting
polymers. Since the initial discovery of these polymers
in the late 1970s, various applications of these materials
have been explored due to their exceptional electronic
and photonic properties.
1-3
The straightforward synthesis of polythiophene de-
rivatives (PT) generates soluble and processable poly-
mers with a wide range of practical and potential
applications, including rechargeable batteries,
1
electro-
chromic devices,
1
chemical and optical sensors,
1
light-
emitting diodes,
4-6
and field-effect transistors.
7
While
traditional approaches to synthesizing PTs derivatives
via electrochemical or oxidative chemical polymerization
methods yield polymers with various degrees of regio-
regularity,
3
the regioselective synthesis of poly(3-alkyl-
thiophenes) (PATs) results in almost exclusively head-
to-tail (HT) couplings.
The synthesis of regioregular PATs, first discovered
by McCullough et al.
8,9
and shortly followed by Rieke,
10
results in the formation of defect-free, structurally
homogeneous HT-PATs that have greatly improved
electronic and photonic properties compared to regio-
random analogues,
11,12
including significantly enhanced
conductivity. One drawback with the McCullough and
Riecke methods lies in their use of cryogenic tempera-
tures.
8
The discovery of the Grignard metathesis (GRIM)
method allows the polymerization to occur at room
temperature or at reflux, hence leading to a quick and
cost-effective technique for the large-scale synthesis of
high molecular weight, regioregular PATs.
13,14
All of the aforementioned polymerizations are transi-
tion-metal-catalyzed cross-coupling reactions.
15
The gen-
erally acce pted mechanism for these Ni(II) catalyzed
cross-coupling reactions involves a catalytic cycle of
three consecutive steps: oxidative addition, transmeta-
lation, and reductive elimination.
16-26
The course of the
catalytic reaction has been extensively studied and
proven to be affected by both the ligand structure and
the choice of the metal.
15
It has been recently reported
that the nickel-initiated cross-coupling polymerization
proceeds via a chain-growth mechanism.
27,28
Further-
more, the addition of various Grignard reagents (R′MgX)
at the end of polymerization results in the end-capping
of regioregular PATs with an R′ end group.
29
Here we show that the Grignard metathesis polym-
erization generates regioregular poly(3-alkylthiophenes)
with precise molecular weights and very narrow poly-
dispersities. Furthermore, by the sequential addition of
different 3-alkylthiophene monomers, this method leads
to the synthesis of block copolymers. The near “living”
nature of the polymerization also presented and allows
for the development of new architectures and the ability
to create regioregular poly(3-alkylthiophenes) with spe-
cific functionalities.
Experimental Part
Materials. Synthesis of 2,5-dibromo-3-hexylthiophene and
2,5-dibromo-3-dodecylthiophene were performed according to
the literature.
13,14
Tetrahydrofuran (THF) was dried over
K/benzophenone under nitrogen and freshly distilled prior to
use. [1,3-Bis(diphenylphosphino)propane]dichloronickel(II) (98%)
(Ni(dppp)Cl
2
), tert-butylmagnesium chloride (2 M in diethyl
ether) and p-dimethoxybenzene (98%) were purchased from
Aldrich Chemical Co., Inc., and used without further purifica-
tion.
Polymerization Kinetic Experiments. In a typical ex-
periment, a dry 100 mL three-neck round-bottom flask was
flushed with N
2
and charged with 2,5-dibromo-3-hexylt hio-
phene (1.6 g, 5 mmol), p-dimethoxybenzene (internal standard)
(0.2 g), and anhydrous THF (50 mL). A 2 M solution of tert-
butylmagnesium chloride (2.5 mL, 5 mmol) in diethyl ether
(Et
2
O) was added via a deoxygenated syringe, and the reaction
mixture was gently refluxed for 2 h. At this time an aliquot
(0.5 mL) was taken out and quenched with water. The organic
phase was extracted in Et
2
O analyzed by GC-MS to determine
* Corresponding author. E-mail: rm5g@andrew.cmu.edu.
8649Macromolecules 2005, 38, 8649-8656
10.1021/ma051122k CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/24/2005
the composition of the reaction mixture. The main components
of the reaction mixture were 2-bromo-5-chloromagnesium-3-
hexylthiophene and 5-bromo-2-chloromagnesium-3-hexylthio-
phene regioisomers. Usually less than 5% of unreacted 2,5-
dibromo-3-hexylthiophene was detected by GC-MS analysis.
The concentration of 2-bromo-5-chloromagnesium-3-hexylthio-
phene isomer was considered as the initial monomer concen-
tration. The oil bath was then removed, and the reaction
mixture was allowed to cool to 23-25 °C, at which time
Ni(dppp)Cl
2
(0.04 g, 0.075 mmol) was added as a suspension
in 1 mL of anhydrous THF. After addition of Ni(dppp)Cl
2
,
aliquots (1 mL) were taken at different time intervals, and
each was precipitated in methanol (5 mL). For each aliquot a
sample was prepared in Et
2
O (2 mL) and analyzed by GC-MS
for the determination of concentration of unreacted monomer.
After filtration through PTFE filters (0.45 µm), the molecular
weight of the pristine polymer samples was measured by GPC.
Chain Extensi on Experiment for th e Synthesis of
Poly(3-hexylthiophene)-b-poly(3-dodecylthiophene). A
dry 250 mL three-neck round-bottom flask (A) was charged
with 2,5-dibromo-3-hexylthiophene (1.6 g, 5 mmol), p-dimethoxy-
benzene (internal standard) (0.3 g), and anhydrous THF (165
mL). A 2 M solution of tert-butylmagnesium chloride (2.5 mL,
5 mmol) in diethyl ether (Et
2
O) was added via a deoxygenated
syringe, and the reaction mixture was gently refluxed for 2 h.
After the consumption of 2,5-dibromo-3-hexylthiophene the
reaction mixture was cooled at 20-22 °C. The concentration
of unreacted 2-bromo-5-chloromagnesium-3-hexylthiophene
was determined by GC-MS (more than 90% of monomer was
consumed in 2 h). Ni(dppp)Cl
2
(0.05 g, 0.1 mmol) was added
as a suspension in 1 mL of anhydrous THF. The polymeriza-
tion continued for 3 h before addition of 2-bromo-5-chloromag-
nesium-3-dodecylthiophene (prepared as described below).
A dry 50 mL three-neck round-bottom flask (B) flushed with
N
2
was charged with 2,5-dibromo-3-dodecylthiophene (4.1 g,
10 mmol) and anhydrous THF (10 mL). A 2 M solution of tert-
butylmagnesium chloride (5 mL, 10 mmol) in diethyl ether
(Et
2
O) was added via a deoxygenated syringe, and the reaction
mixture was gently refluxed for 2 h. The concentration of
2-bromo-5-chloromagnesium-3-dodecylthiophene was deter-
mined by GC-MS.
Analyses. GC-MS analysis was performed on a Hewlett-
Packard Agilent 6890-5973 GC-MS workstation. The GC
column was a Hewlett-Packard fused silica capillary column
cross-linked with 5% phenylmethylsiloxane. Helium was the
carrier gas. The following conditions were used for all GC-MS
analyses: injector temperature, 250 °C; initial temperature,
70 °C; temperature ramp, 10 °C/min; final temperature, 300
°C. Gel permeation chromatography (GPC) measurements
were performed on a Waters 2690 separations module ap-
paratus and a Waters 2487 dual λ absorbance detector with
chloroform as the eluent (flow rate 1 mL/min, 35 °C, λ ) 254
nm) and a series of three Styragel columns (10
4
, 500, 100 Å;
Polymer Standard Services). Toluene was used as an internal
standard, and calibration based on polystyrene standards was
applied for determination of molecular weights.
1
H NMR
spectra of the polymer solutions in CDCl
3
were recorded on a
Bruker Avance 500 MHz spectrometer. UV-vis-NIR spectra
were measured on polymer solutions in anhydrous chloroform
or polymer thin films cast onto 22 mm square cover glass using
a UV-vis-NIR spectrophotometer (Varian Cary 5000). Elec-
trical conductivity measurements were performed by a stan-
dard spring-loaded pressure contact Signatone S-301-4 four-
point probe, which was connected to a Hewlett-Packard 6632A
system dc power supply, a Hewlett-Packard 3457 A multimeter
(for voltage measurements), and a Keithley model 196 system
DMM (for current measurements). Polymer films for conduc-
tivity measurements were prepared by drop-casting from
toluene solutions. The casting solutions were prepared by
dissolving 5 mg of polymer in 1 mL of dry toluene and
sonicating for 5 min, followed by the filtration of clear solutions
through PTFE 0.45 µm filters. Films were drop-cast onto 22
mm square cover glass that were washed with chromic acid
solution, rinsed several times with acetone and hexanes, and
dried before drop-casting. Conductivities were measured for
films of the polymers oxidized with I
2
for different times (1 h
is the usual doping time). The film thickness was measured
by scanning electron microscopy (SEM) and the conductivity
calculated according to the following equation:
where R is the resistance (R ) V/I) and l is the film thickness.
The film thickness (cross section) was measured by SEM
using a Hitachi S-2460N electron microscope.
Results and Discussion
The discovery of nickel-catalyzed aryl-aryl bond
formations of Grignard reagents with organohalides by
Kumada
16,17
and Corriu
30
has led to a significant
develop ment in the synthesis of various types of
thiophenes. Consequently, the Kumada reaction has
been applied to the synthesis of oligothiophenes and
polythiophenes. Despite its common use, the mechanism
of the nickel-catalyzed cross-coupling polymerization
has not been fully understood. Historically, three dif-
ferent mechanistic pathway s ha ve been independ-
ently proposed.
16-26
We will focus on the mechanistic
pathway suggested by Negishi,
18,19
Yamamoto,
20-22
and
Parshall,
23
which was later extended to cross-coupling
polycondensation.
31,32
It has been proposed that the
reductive elimination and oxidative addition were step-
wise processes, which involved formation of a “free”
Ni(0) intermediate, with the transmetalation as the
rate-determining step. The experimental observations
from both McCullough and Grignard metathesis meth-
ods has led to a new mechanistic proposal for the nickel-
catalyzed cross-coupling polymerization for the synthe-
sis of regioregular poly(3-alkylthiophenes).
16-26
We have recently proposed the mechanism of the
nickel-in itiated cross-cou pling polymerizat ion for
the synthesis of regioregular poly(3-alkylthiophene)
(McCullough method).
27
The proposed mechanism of
Grignard metathesis is shown in Scheme 1. In the first
reaction of Scheme 1, treatment of 2,5-dibromo-3-
alkylthiophene with 1 equiv of RMgCl (R ) alkyl)
results in magnesium-bromine exchange reaction, also
referred to as Grignard metathesis (GRIM). This reac-
Scheme 1. Mechanism of Grignard Metathesis
Method for the Synthesis of Regioregular
Poly(3-alkylthiophene)
σ )
1
4.53Rl
8650 Iovu et al. Macromolecules, Vol. 38, No. 21, 2005
tion proceeds with a moderate degree of regioselectivity,
leading to a distribution of regiochemical isomers 1 and
1′ of 85:15 to 75:25. Lowering the temperature during
magnesium haloge n exchange reaction results in the
formation of isomer 1 in higher proportion ([1]:[1′] )
85:15) when the reaction is performed at room tem-
perature. To verify the incorporation of isomer 1, we
needed to analyze the reaction mixture. The GC-MS
analysis of the quenched reaction mixture, after the
addition of Ni(dppp)Cl
2
, indicated that only isomer 1 is
incorporated into the polymer, while isomer 1′ was not
consumed.
The first step in the mechanism is the reaction of 2
equiv of 2-bromo-5-chloromagnesium-3-alkylthiophene
monomer (1) with Ni(dppp)Cl
2
generating a bis(organo)-
nickel compound (2), which undergoes reductive elimi-
nation resulting in the formation of an associated pair
of the 5,5′-dibromobithienyl (tail-to-tail coupling) and
Ni(0) [3-4]. We propose that the associated pair is
formed by coordination of 1,3-bis(diphenylphosphino)-
propane-nickel(0) to the thiophene ring in a η
2
- or η
4
-
bonded fashion. A similar type of Ni(0)-η
2
arene
complex was previously reported.
33-35
The dimer un-
dergoes a fast oxidative addition to the nickel(0) center,
generating a new organonickel compound. Growth of the
polymer chain occurs by insertion of one monomer at a
time, in which the Ni(dppp) moiety is incorporated into
the polymer chain as an end group (polymer 5 in
Scheme 1). According to the proposed mechanism, only
one structural defect (one tail-to-tail coupling) per
polymer chain is generated during the proposed cata-
lytic cycle.
In the Grignard metathesis method, po ly(3-alkyl -
thiophenes) with relatively high molecular weight are
produced early in the reaction.
27
This contradicts the
accepted step growth polymerization mechanism pro-
posed for nickel-catalyzed cross-coupling polymerization,
in which one expects the fast disappearance of the
monomer and increase of the molecular weight toward
the end of the reaction. Therefore, we explored the
kinetics of the polymerization reaction to achieve a near
“living” system.
Influence of Ni(dppp)Cl
2
Concentration. Several
experiments were performed at various Ni(dppp)Cl
2
concentrations with a constant monomer 1 concentra-
tion. As shown in Figure 1, the reaction rates increased
with the increase in the Ni(dppp)Cl
2
concentration. The
plots indicate the polymerization is over within just few
minutes. The semilogarithmic kinetic plots are linear
up to ∼60% conversion (Figure 2). The nonlinearity in
the semilogarithmic kinetic plots in the late stage of the
reaction indicates the presence of termination, which
may be due to the aggregation of polymer chains.
36
If
the reaction medium becomes heterogeneous due to the
formation of polymer aggregates, the active centers are
not accessible for further insertion of the monomer.
Because of the nonlinearity of the semilogarithmic
kinetic plots, further optimization in the reaction condi-
tions is needed to result in a true “living” GRIM
polymerization.
Molecular weight vs conversion plot (Figure 3) and
the GPC traces (Figure 4) show the increase of molec-
ular weight with conversion, which supports a “living”
chain growth mechanism for nickel-initiated cross-
coupling polymerization. The molecular weight of the
Figure 1. Conversion (filled symbols) and logarithm of
monomer concentration (open symbols) vs time plots for
2-bromo-5-chloromagnesium-3-hexylthiophene Grignard me-
tathesis polymerization at variable Ni(dppp)Cl
2
concentrations;
[M]
0
) 0.07 mol/L; THF; 23-25 °C.
Figure 2. Logarithm of monomer concentration vs time plot
for 2-bromo-5-chloromagnesium-3-hexylthiophene Grignard
metathesis polymerization; [M]
0
) 0.07 mol/L; THF; 23-25
°C.
Figure 3. Dependence of molecular weights and polydisper-
sities on conversion for 2-bromo-5-chloromagnesium-3-hexyl-
thiophene Grignard metathesis polymerization at variable
Ni(dppp)Cl
2
concentrations; [M]
0
) 0.07 mol/L; THF; 23-25
°C.
Macromolecules, Vol. 38, No. 21, 2005 Regioregular Poly(3-alkylthiophenes) 8651
polymers is a function of the molar ratio of monomer 1
to Ni(dppp)Cl
2
initiator (Figure 3). Furthermore, the
number-average molecular weight of the polymers can
be predicted by the following formula:
According to our proposed mechanism, Ni(dppp)Cl
2
acts
as an initiator rather than a catalyst, and the Ni(dppp)
moiety is incorporated in the polymer as an end group
(polymer 2, Scheme 1). Poly(3-hexylthiophenes) with
relatively narrow polydispersities (as low as PDI ) 1.2)
were obtained for the experiments performed at higher
concentration of Ni(dppp)Cl
2
(Figure 3).
Influence of Monomer Concentration. To comple-
ment the previous experiments, another series of ex-
periments were conducted at a constant Ni(dppp)Cl
2
concentra tion, while varying the 2-bromo-5-chloro -
magnesium-3-hexylthiophene (monomer) concentration.
These experiments were performed in order to deter-
mine the reaction order with respect to the monomer.
The polymerizations were conducted at low tempera-
tures (0-2 °C) to slow the rate of reaction and conserve
the linearity of the semilogarithmic kinetic plots. The
latter allowed for a more accurate determination of the
initial polymerization rate. As shown in Figure 5, the
reaction rates increased with the increase in the mono-
mer concentration.
Figure 6 shows the increase of molecular weight with
conversion. Poly(3-hexylthiophenes) with relatively nar-
row polydispersities (PDI < 1.5) were synthesized.
A value of ∼1 for the reaction order with respect to
the monomer was obtained from the slope of the plot of
the logarithm of the initial rate of polymerization vs the
logarithm of the monomer concentration (Figure 7).
End-Group Analysis of Regioregular Poly(3-
hexylthiophene). Grignard metathesis method for the
synthesis of poly(3-alkylthiophenes) results in the for-
mation of regioregular polymers (∼98% head-to-tail
couplings). Quenching of nickel terminated poly(3-
alkylthiophene) (polymer 2, Scheme 1) with methanol/
HCl mixture results in the formation of H/Br terminated
polymer. According to the mechanism proposed here,
only one tail-to-tail coupling should occur in the regio-
regular polymerization of 3-alkylthiophenes. Here we
Figure 4. Gel permeation chromatography traces (UV detec-
tor) for 2-bromo-5-chloromagnesium-3-hexylthiophene Grig-
nard metathesis polymerization [Ni(II)]
0
) 1.5 × 10
-3
mol/L;
[M]
0
) 0.07 mol/L; THF; 23-25 °C.
Figure 5. Conversion (filled symbols) and logarithm of
monomer concentration (open symbols) vs time plots for
Grignard metathesis polymerization at different 2-bromo-5-
chloromagnesium-3-hexylthiophene concentrations; [Ni(II)]
0
)
1.5 × 10
-3
mol/L; THF; 0-2 °C.
DP
n
)
∆[M]
t
[Ni(dppp)Cl
2
]
0
Figure 6. Dependence of molecular weights and polydisper-
sities on conversion for Grignard metathesis polymerization
at different 2-bromo-5-chloromagnesium-3-hexylthiophene con-
centrations; [Ni(II)]
0
) 1.5 × 10
-3
mol/L; THF; 0-2 °C.
Figure 7. Plot of the logarithm of the initial rate of polym-
erization vs logarithm of monomer concentration for Grignard
metathesis polymerization of 2-bromo-5-chloromagnesium-3-
hexylthiophene; [Ni(II)]
0
) 1.5 × 10
-3
mol/L; THF; 0-2 °C.
8652 Iovu et al. Macromolecules, Vol. 38, No. 21, 2005
investigate the structures of the resulting polymers by
NMR spectroscopy.
As an example, the full 500 MHz
1
H NMR spectrum
of a moderate molecular weight regioregular poly(3-
hexylthiophene) (rr-PHT) is presented in Figure 8a. The
main absorption signals of rr-PHT are assigned as
shown. Two small triplets at δ ∼ 2.6 ppm of the same
intensity for H/Br terminated rr-PHT can be assigned
to the methylene protons on the first carbon substituent
(h and h′) on the end units. Furthermore, the appear-
ance of the two separate triplet signals at different
resonance frequencies is due to different chemical
environment around h and h′ (Figure 8c). When the
H/Br terminated polymer is subjected to a magnesium
halogen exchange reaction (Scheme 2) and quenched
with an acidic methanol/water mixture, a pristine H/H
terminated rr-PHT is formed. Consequently, the signal
generated by the methylene protons h′ is shifted down-
field with the two groups (h and h′) resonating at the
same frequency (Figure 8b). Table 1 presents the
integration values from the
1
H NMR (500 MHz) spec-
trum of rr-PHT-H/Br and rr-PHT-H/H terminated poly-
mers. Note that the intensity of the h peak is doubled
in the absence of the bromine atom relative to the main
peak (b) of first β-substituent methylene protons. These
observations indicate that NMR analysis can distin-
guish between the two different types of coupling (e.g.,
head-to-tail (HT) and tail-to-tail (TT)), when rr-PHT
H/H terminated contains only one structural defect per
polymer chain. This finding is in full support of the
mechanism in Scheme 1 showing that the first step is
a tail-to-tail coupling, followed by Ni-initiated polym-
erization. The NMR cannot distinguish the H/H termi-
nated methylene resonances. However, it allows a
relatively accurate determination of molecular weight
from the integration of end-group resonances relative
to the bulk polymer. For instance, DP
n
for the afore-
mentioned polymer is equal to the ratio of b to h and
results in 50 monomer units corresponding to M
n
)
8300.
Chain Extension by Sequential Monomer Addi-
tion. Synthesis of Poly(3-hexylthiophene)-b-poly-
(3-dodecylthiophene). A previous report from our
group showed that the addition of various Grignards
reagents to the nickel termin ated poly(3-alk ylthio-
phene) results in the formation of end-functional poly-
mers.
29
Addition of a new portion of 2-bromo-5-chloro-
magnesium-3-alkylthiophene monomer at the end of the
polymerization also resulted in the further increase of
the molecular weight of the final polymer. Both experi-
ments strongly indicate the “living” nature of the nickel
terminated poly(3-alkylthiophene).
The main focus of this section is the chain exten-
sion of rr-PATs through the sequential addition of a
different monomer. An example of chain extension is
based on the synthesis of poly(3-hexylthiophene)-b-poly-
(3-dodecylthiophene) (PHT-b-PDDT) block copolymer
and is described herein for the first time.
The synthetic strategy used for the synthesis of PHT-
b-PDDT is outlined in Scheme 3.
The first step involves preparation of rr-PHT with
well-defined end group and structural homogeneity. The
reaction parameters were chosen with a special consid-
Figure 8.
1
H NMR (500 MHz) spectra of (a) rr-poly(3-hexylthiophene) H/Br terminated; (b) expansion of rr-poly(3-hexylthiophene)
H/H terminated; and (c) expansion of rr-poly(3-hexylthiophene) H/Br terminated.
Scheme 2. Magnesium Halogen Exchange of H/Br
Terminated Regioregular Poly(3-hexylthiophene)
Table 1. Integration Values from the
1
H NMR (500 MHz)
Spectrum of rr-Poly(3-hexylthiophene) (PHT) H/Br and
H/H Terminated
peak PHT-H/Br PHT-H/H
b 1 1
h 0.02 0.04
h′ 0.02
Macromolecules, Vol. 38, No. 21, 2005 Regioregular Poly(3-alkylthiophenes) 8653
![Figure 9. Molecular weight vs conversion plot for Grignard metathesis polymerization of 2-bromo-5-chloromagnesium-3dodecylthiophene initiated by nickel terminated poly(3-hexylthiophene); [MHT]0 ) 0.02 mol/L; [Ni(II)]0 ) 0.6 × 10-3 mol/L; [MDT]0 ) 0.04 mol/L; THF; 18-20 °C.](/figures/figure-9-molecular-weight-vs-conversion-plot-for-grignard-39txzf9s.webp)
