Published on Web 08/19/2009
Externally Initiated Regioregular P3HT with Controlled Molecular Weight and
Narrow Polydispersity
Hugo A. Bronstein and Christine K. Luscombe*
Materials Science and Engineering Department, UniVersity of Washington, Seattle, Washington 98195-2120
Received July 8, 2009; E-mail: luscombe@u.washington.edu
Poly(3-hexylthiophene) (P3HT) remains the most commonly used
polymer in organic photovoltaics because of its desirable electronic
properties.1 Both the Yokozawa and McCullough groups have
developed methods for the synthesis of highly regioregular P3HT
with controlled molecular weights and narrow molecular weight
distributions using the 1,3-bis(diphenylphosphino)propanenickel(II)
chloride [NiCl2(dppp)]-catalyzed polymerization of Grignard-type
monomers.2,3 The polymerization control is thought to originate
from either intramolecular transfer of the Ni catalyst3 or an
associated pair formed by the growing polymer chain and the Ni
catalyst.4 The drawback of this synthetic methodology is that it is
Figure 1. 31P{H} NMR spectra of nickel(II) complexes in chlorobenzene:
(top) trans-chloro(phenyl)bis(triphenylphosphine)nickel(II) 2a; (bottom) cis-
chloro(phenyl)(dppp)nickel(II) 3a.
not possible to initiate the polymerization from an external moiety,
as has been demonstrated for the chain-growth polymerization of
polyfluorene.5 This is necessary for the synthesis of more complex
polymer architectures such as brushes and star and block copoly-
mers. There are a small number of examples of externally initiated
P3HT,6,7 including Senkovskyy et al.,6 who demonstrated chain-
growth polymerization of P3HT from both small-molecule and
surface initiators. We have also recently investigated methods for
the external initiation of P3HT, including the effect of varying the
initiating aryl halide.8 However, these examples all employed the
more reactive tetrakis(triphenylphosphine)nickel(0) [Ni(PPh3)4]
catalyst, and while the regioregularity of the synthesized polymer
was high, the molecular weight distribution was broad. Furthermore,
the polymer end groups were not uniform, which would significantly
impair the synthesis of a block copolymer. Consequently, the
externally initiated polymerization of P3HT with high regioregu-
larity and narrow polydispersity (PDI) remains a highly desirable
goal. We believed that in order to realize this, it would be necessary
to employ dppp as the spectrator ligand on the Ni catalyst, as it
has been established that its use leads to the greatest degree of
polymerization control in nonexternally initiated P3HT synthesis.3
Addition of 1.5 equiv of dppp resulted in rapid ligand substitu-
tion, which reached completion within 2 h to form the thermody-
namically favored cis-chloro(aryl)(dppp)nickel(II) complexes 3a and
3b (Scheme 1). The cis geometry was confirmed by the appearance
of two sets of doublets in the 31P NMR spectra at ∼20 and -6
ppm (JP-P ) 44 Hz) (Figure 1b). The excess dppp was observed at
-18 ppm, and the liberated PPh3 peak became sharp, indicating
the absence of exchange reactions. In the case of 3a, a small amount
(∼5%) of what is believed to be Ni(dppp)2 was observed; this could
be formed by chelation of the residual Ni(PPh3)4 by dppp.10 This
species was not observed in the synthesis of 3b.
Polymerization was initiated by addition of the crude complex
mixture to a solution of 2-bromo-5-chloromagnesio-3-hexylthio-
phene in THF at 0 °C followed by stirring at room temperature
(Scheme 2). The reaction mixture was quenched with 5 M HCl
and precipitated into methanol, after which the polymer was isolated
by filtration and washed with methanol and cold hexane.
Addition of complex 3a (1.43 mol %) to the Grignard-type
monomer 4 afforded P3HT with an Mn of 11.2 kDa and a PDI
(i.e., Mw/Mn) of 1.1 versus polystyrene standards (Figure 2a). H
Scheme 1
1
NMR analysis revealed the polymer to be fully regioregular, with
the presence of peaks corresponding to the initiating groups [Figures
5 and 6 in the Supporting Information (SI)]. Integration of the
initiating peaks with respect to the polymer backbone signals
indicated a degree of polymerization of ∼63 (conversion ) 90%).
MALDI-TOF analysis indicated the majority of the peaks to be
Ph/H terminated, alongside a small amount of Ph/Br (<5%) and
H/H (<1%) termination (Figure 2b and SI Figure 7). The presence
of Ph/Br-terminated polymer chains is believed to be due to the
partial precipitation of the polymer toward the end of the reaction
because of the decreasing solubility of the polymer chain, which
subsequently undergoes quenching processes. The H/H-terminated
P3HT likely originates from the small amount of Ni(dppp)2 complex
that was observed in the crude initiator mixture. Furthermore, as a
result of the similarity in the molecular weights of a phenyl group
and a bromine atom, it is possible that the major peaks observed
In order to synthesize the desired initiating complexes, Ni(PPh3)4
was added to neat aryl chlorides 1a and 1b at room temperature
and left overnight (Scheme 1). Sharp peaks at ∼21 ppm were
observed in the corresponding 31P NMR spectra (Figure 1a), which
indicated the formation of the trans-chloro(aryl)bis(triphenylphos-
phine)nickel(II) complexes 2a and 2b.9 Broad peaks were observed
at -6 ppm due to the liberated triphenylphosphine. The broadness
indicates exchange processes on the NMR time scale, which can
be attributed to residual Ni(PPh3)4, resulting in rapid ligand
exchange with the free triphenylphosphine.
9
12894 J. AM. CHEM. SOC. 2009, 131, 12894–12895
10.1021/ja9054977 CCC: $40.75 2009 American Chemical Society
Bronstein, Hugo A.
