Using bis(pinacolato)diboron to improve the quality of regioregular conjugated co-polymers

Article abstract of DOI:10.1039/c0jm02359k

We demonstrate the use of bis(pinacolato)diboron to directly polymerize symmetric, bisbromo, thiophene-based monomers via a Suzuki homo-polymerization to form co-polymers in less steps than the corresponding co-polymerization. We compare this method to the commonly used Stille co-polymerization by preparing four thiophene-based co-polymers using both methods. We use MALDI-TOF mass spectrometry to show that this new method produces high-quality, uniform polymers with narrow distributions of end-groups. By varying the electronegativity of the monomers, we demonstrate rudimentary control over these end-groups, forming either bis-H-, mono-H-mono-Br-, or bis-Br-terminated polymers in order of increasing electronegativity.

Full text of DOI:10.1039/c0jm02359k

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/materials | Journal of Materials Chemistry
Using bis(pinacolato)diboron to improve the quality of regioregular
conjugated co-polymers†
*
Frank Brouwer, Jan Alma, Hennie Valkenier, Thomas P. Voortman, Jorrit Hillebrand, Ryan C. Chiechi
*
and Jan C. Hummelen
Received 22nd July 2010, Accepted 11th October 2010
DOI: 10.1039/c0jm02359k
We demonstrate the use of bis(pinacolato)diboron to directly polymerize symmetric, bisbromo,
thiophene-based monomers via a Suzuki homo-polymerization to form co-polymers in less steps than
the corresponding co-polymerization. We compare this method to the commonly used Stille
co-polymerization by preparing four thiophene-based co-polymers using both methods. We use
MALDI-TOF mass spectrometry to show that this new method produces high-quality, uniform
polymers with narrow distributions of end-groups. By varying the electronegativity of the monomers,
we demonstrate rudimentary control over these end-groups, forming either bis-H-, mono-H-mono-Br-,
or bis-Br-terminated polymers in order of increasing electronegativity.
during the polymerization reaction.12,13 These and other side-
reactions introduce defects and terminate chains uncontrollably,
negatively affecting the electronic properties of the polymer.
Janssen et al. demonstrated that polymers prepared by Suzuki
co-polymerization can be terminated in 11 different ways,
producing 11 unique polymer chains per repeating unit.14–17 This
is an inescapable consequence of co-polymerizations which, even
in the absence of side-reactions, can initiate from and be termi-
nated by either monomer.
1. Introduction
This paper describes the highly reproducible synthesis of regio-
regular, thiophene-based, conjugated co-polymers using bis-
(pinacolato)diboron (BiPi) to generate the active species for
Suzuki homo-polymerizations in situ, obviating the need for
co-polymerization reactions. We used MALDI-TOF mass
spectrometry to demonstrate that this approach produces well-
defined polymer chains and generally increases the quality of
co-polymers over those synthesized by the commonly used Stille
co-polymerization by reducing the number of different types of
polymer chains that form.
As rrP3HT has shown, if these complex co-polymers are to see
commercial success,18–24 reproducible polymerization techniques
that produce well-defined polymers are needed.25–27 The non-
toxic, mild conditions of the Suzuki polymerization make it
preferable to the Stille.28 However, the electron-rich nature of
thiophene-based monomers, which are a popular choice because
they are easily functionalized to fine-tune their properties,28,29
slows the catalytic cycle of the Suzuki method. This requires high
temperatures and long reaction times, leading to low molecular-
weight polymers with poor electronic properties. Additionally,
the requisite boronate-functionalized thiophene monomers are
notoriously difficult to purify. For both Stille and Suzuki
polymerizations, copolymers can be synthesized either by
co-polymerizing a bis-halogen, X–ArA–X, with a bis-boronate or
bis-stanane, Y–ArB–Y, or homo-polymerizing an asymmetric
co-monomer, X–ArAArB–Y; both will produce the same co-
polymer, (ArAArB)n. The co-monomer approach is, in theory,
preferable because, even though it may require more synthetic
steps, it produces a higher degree of polymerization and leads to
less variation in end-groups than the equivalent co-polymeriza-
tion because the chains can only initiate from and terminate with
the same monomer. In reality, however, the syntheses and, in
particular, the purification of these co-monomers are often very
difficult.
Conjugated polymers of thiophene moieties are synthesized by
metal-mediated crosscoupling or, less frequently, by oxidative
polymerization (which we will not discuss). The reactive inter-
mediates of metal-mediated polymerizations are thiophene–
metal complexes, formed in situ from a halogen, and a thiophene
bearing a Sn (Stille), Mg (GRIM), Br (Yamamoto), or B
(Suzuki).1–10 Each of these methods is suited to a particular
system; GRIM and Yamamoto are typically used for homo-
polymerizations and random co-polymerizations of simple
monomers. For example, regio-regular poly(3-hexylthiophene)
(rrP3HT) is easily prepared using GRIM, which reproducibly
leads to high-quality polymers—a major reason for the wide-
spread use of rrP3HT. For alternating co-polymers or polymers
with labile functional groups, Suzuki or Stille must be used.
Unlike GRIM, these two reactions proceed by a step-growth
mechanism in which the molecular-weight and polydispersity
index (PDI) of the polymer are extremely sensitive to the purity
and stoichiometry of the monomers, hindering reproducibility.11
Homo-coupling and side-reactions exacerbate this problem by
uncontrollably affecting the stoichiometry of the monomers
We synthesized a series of thiophene-based compounds and
polymerized them directly from the bis-bromide by generating the
active boronate species for a Suzuki polymerization in situ from
BiPi via the Miyaura reaction.30–35 We also synthesized identical
polymers by the commonly used Stille co-polymerization method
Stratingh Institute for Chemistry and Zernike Institute for Advanced
Materials, University of Groningen, Nijenborgh 4, Groningen, 9747AG,
The Netherlands. E-mail: r.c.chiechi@rug.nl; j.c.hummelen@rug.nl
† Electronic supplementary information (ESI) available: Additional
syntheses and further discussion of co-polymerizations. See DOI:
10.1039/c0jm02359k
1582 | J. Mater. Chem., 2011, 21, 1582–1592
This journal is ª The Royal Society of Chemistry 2011

View Online
for comparison. These reactions are summarized in Scheme 2.
This ‘‘BiPi method’’ combines the more facile syntheses of the
co-polymerization route with the reproducibility and higher-
quality polymers of the co-monomer route by enabling the direct
homopolymerization of a symmetric co-monomer36,37 to form
a co-polymer. This method obviates the synthesis of Sn- or B-
containing monomers and retains the versatility and tolerance
for functional groups of the standard Suzuki polymerization. It
produces higher-quality polymers in fewer synthetic steps and
with less variation in end-groups than the same co-polymers
prepared by Stille co-polymerizations.
Thieno[2,3-b]thiophene,38 2,5-bis(trimethylstannyl)thiophene,39
2,5-bis(trimethylstannyl)thieno[2,3-b]thiophene,40 2,7-dibromo-
9,9-dihexyl-9H-fluorene,41 3-decylthiophene,42 2-bromo-3-decyl-
thiophene,43 1,5-diiodonapthalene,44 and 4,7-dibromo-2,1,3-
benzothiadiazole45 were synthesized according to the literature
procedures; we obtained 1H-NMR and 13C-NMR spectra that are
identical to the reported spectra.
2.2 Monomers
General bromination procedure. The bromination resulting in
the co-monomers 1, 2, 3, 4, 5 and 6 was performed using
a modified literature procedure under a nitrogen atmosphere in
the dark.46 To a cooled (0 ꢀC) mixture of co-monomer (1.0 g) in
50 mL of THF was added drop-wise a solution of NBS (2.10–
2.25 equivalents) in 15 mL of THF. The resulting mixture was
allowed to warm up to RT overnight and poured into 200 mL of
0.1 M HCl followed by extraction with CH2Cl2. The combined
organic layers were washed with brine and dried over Na2SO4.
The solvent was removed by rotary evaporation and the crude
product purified by column chromatography (silica, hexane or
heptane), followed by recrystallization from isopropanol/
methanol (3 : 1).
2. Experimental
2.1 General
All reagents and solvents were purchased from commercial
sources and used without further purification unless otherwise
indicated. NMR were measured using a Varian VXR-300
(300 MHz) or a Varian Gemini-200 (200 MHz) instrument at
25 ꢀC (in CDCl3 with the exception of polymers with overlapping
peaks which were performed in CD2Cl2). ATR-IR spectra were
recorded on a Bruker IFS88. GPC measurements were done on
a Spectra Physics AS 1000 series machine equipped with a Vis-
kotek H-502 viscometer and a Shodex RI-71 refractive index
detector. The columns (PLGel 5m mixed-C) (Polymer Labora-
tories) were calibrated using narrow disperse polystyrene stan-
dards (Polymer Laboratories). Samples were made in chloroform
at a concentration of 1 mg mLꢁ1. MALDI-TOF measurements
were performed on a Biosystems Voyager apparatus. Samples
were prepared by mixing the matrix (terthiophene or dithranol,
20 mg mLꢁ1 in CHCl3) and the sample (1 mg per 5 mL in CHCl3)
in a 1 : 1 ratio. All the samples were measured in negative ion
mode.
3,3%-Didecyl-2,20:50,200:500,2%-quaterthiophene. 3,3%-Didecyl-
2,20:50,200:500,2%-quaterthiophene was prepared according to
modified literature procedures.47,48 2-Bromo-3-decyl thiophene
(4.0 g, 13 mmol) was dissolved in 15 mL dry THF and added
drop-wise to activated magnesium (380 mg, 15 mmol) at reflux.
After complete addition the mixture was refluxed for an addi-
tional 5 h. The mixture was cooled to RT and cannulated to
a second flask containing 5,50-dibromo-2,20-bithiophene (1.5 g,
5 mmol), Ni(dppp)Cl2 (46 mg), and 40 mL of dry THF/toluene
(4 : 3) before refluxing overnight. The reaction was cooled to RT,
quenched with saturated NH4Cl, extracted with CH2Cl2, and
washed with brine. The combined organic layers were dried over
Kumada, Stille and Suzuki coupling reactions were performed
under dry conditions and nitrogen atmosphere.
Scheme 1
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1582–1592 | 1583

View Online
Na2SO4 and the solvent removed by rotary evaporation. The
crude product purified by column chromatography (silica) using
hexanes as the eluent, followed by recrystallization from iso-
propanol/methanol (3 : 1) to give 3,3%-didecyl-2,20:50,200:500,2%-
quaterthiophene (1.87 g, 60%) as a fluorescent-yellow solid. dH
(300 MHz, CDCl3) 7.16 (d, J ¼ 5.2, 2H), 7.11 (d, J ¼ 3.7, 2H),
7.00 (d, J ¼ 3.8, 2H), 6.92 (d, J ¼ 5.2, 2H), 2.85–2.67 (m, 4H),
1.62 (d, J ¼ 7.6, 4H), 1.24 (s, 28H), 0.85 (t, J ¼ 6.6, 6H). dC
(50 MHz, CDCl3) 140.08, 136.99, 135.50, 130.51, 130.30, 126.72,
124.06, 124.02, 32.14, 30.89, 29.85, 29.83, 29.75, 29.68, 29.57,
29.48, 22.92, 14.36. HRMS (APCI) calculated for [M + H]+
611.2868, found 611.2866.
5 mmol) and 2-thiopheneboronic acid (1.29 g, 10 mmol). To this
was added crushed and dried K3PO4 (2.14 g, 10 mmol) to form
a white suspension. The mixture was sparged with dry nitrogen
three times, followed by the addition of Pd(PPh3)4 (23 mg, 0.02
mmol) and additional sparging. The reaction mixture was stirred
at 110 ꢀC overnight in the absence of light, cooled to RT, poured
into ice water, and extracted with CH2Cl2. The combined organic
layers were washed with brine, dried over Na2SO4, and removed
by rotary evaporation. The crude product was purified by short
column chromatography (silica) using petroleum ether/CH2Cl2
(10 : 1) as the eluent and recrystallized from isopropanol/meth-
anol (4 : 1) to give 2-[9,9-dihexyl-7-(2-thienyl)-9H-fluoren-
2-yl]thiophene (2.0 g, 84%) as green crystals. dH (300 MHz,
CDCl3) 7.67 (dd, 4H, J ¼ 7.7, J ¼ 15.4), 7.56 (s, 2H), 7.38 (d, 2H,
J ¼ 2.2), 7.29 (d, 2H, J ¼ 5.1), 7.12 (t, 2H, J ¼ 3.6), 2.02 (m, 4H),
1.10 (m, 12H), 0.73 (m, 10H). dC (200 MHz, CDCl3) 151.66,
145.14, 140.17, 133.23, 128.04, 124.95, 124.51, 122.87, 120.11,
120.06, 55.26, 40.42, 31.44, 29.65, 23.70, 22.56, 13.99. HRMS
(APCI) calculated for [M + H]+ 499.2488, found 499.2479.
5,5%-Dibromo-3,3%-didecyl-2,20:50,200:500,2%-quaterthiophene 1.
3,3%-Didecyl-2,20:50,200:500,2%-quaterthiophene (1.0 g, 1.6 mmol)
was brominated with NBS (637 mg, 3.6 mmol) according to the
general bromination procedure to give 1 (1.1 g, 88%). dH (200
MHz, CDCl3) 7.08 (d, J ¼ 3.7, 2H), 6.94 (d, J ¼ 3.8, 2H), 6.88 (s,
2H), 2.69 (t, J ¼ 7.9, 4H), 1.53 (s, 6H), 1.24 (s, 28H), 0.84 (t, J ¼
6.5, 6H). dC (50 MHz, CDCl3) 140.74, 137.25, 134.26, 132.92,
131.89, 127.17, 124.20, 110.87, 32.13, 30.74, 29.83, 29.78, 29.62,
29.56, 29.41, 22.91, 14.36. HRMS (APCI) calculated for [M +
H]+ 769.1057, found 769.1058.
2-bromo-5-[7-(5-bromo-2-thienyl)-9,9-dihexyl-9H-fluoren-2-
yl]thiophene 3. 2-[9,9-Dihexyl-7-(2-thienyl)-9H-fluoren-2-yl]-
thiophene (900 mg, 1.8 mmol) was brominated with NBS
(705 mg, 4.0 mmol) according to the general bromination
procedure to give 3 (1.724 g, 88%) as yellow, slightly green
crystals; mp 100–110 ꢀC. dH (300 MHz, CDCl3) 7.66 (dd, 4H, J ¼
7.7, J ¼ 42.1), 7.45 (s, 2H), 7.12 (dd, 4H, J ¼ 3.7, J ¼ 15.0), 1.99
(m, 4H), 1.12 (m, 12H), 0.75 (t, 6H, J ¼ 6.6), 0.65 (b, 4H). dC (200
MHz, CDCl3) 151.82, 146.50, 140.39, 132.61, 130.85, 124.67,
123.04, 120.28, 119.79, 111.11, 73.72, 55.32, 40.35, 31.43, 29.62,
23.70, 22.55, 13.99. HRMS (APCI) calculated for [M + H]+
657.0678, found 657.0685.
3,300-Didecyl-2,20:50,200-terthiophene.
3,300-Didecyl-2,20:50,200-
terthiophene was prepared according to modified literature
procedures.49,50 To a mixture of 120 mL dry toluene and 40 mL
dry DMF in a three-necked flask were added 2,5-bis(trimethyl-
stannyl)thiophene (1.7 g, 4.1 mmol), 2-bromo-3-decylthiophene
(2.5 g, 8.1 mmol), and Pd(PPh3)4 (150 mg, 0.13 mmol). The
mixture was sparged with dry nitrogen, heated to 110 ꢀC for 16 h,
and then cooled to RT, poured over ice, and extracted with
CHCl3. The combined organic layers were washed with 2 M HCl
solution, brine, and dried over Na2SO4 followed by removal of
the solvent by rotary evaporation. The crude product was puri-
fied byꢀcolumn chromatography (silica) with petroleum ether
(40–60 C) as the eluent followed by recrystallization from iso-
propanol/methanol (3 : 1) to give 3,300-didecyl-2,20:50,200-terthio-
phene (1.4 g, 65%) as a yellow oil. dH (300 MHz, CDCl3) 7.15 (d,
J ¼ 5.2, 2H), 7.03 (s, 2H), 6.92 (d, J ¼ 5.2, 2H), 2.76 (t, J ¼ 7.6,
4H), 1.71–1.56 (m, 4H), 1.24 (s, 28H), 0.85 (t, J ¼ 6.6, 6H). dC
(50 MHz, CDCl3) 139.92, 136.26, 130.61, 130.29, 126.25, 123.93,
32.14, 30.99, 29.86, 29.80, 29.73, 29.58, 29.52, 22.92, 14.36.
2,5-Bis(3-decylthiophen-2-yl)thieno[2,3-b]thiophene.
To
a mixture of 120 mL dry toluene and 40 mL dry DMF in a dry
three-necked flask were added 2,5-bis(trimethylstannyl)-
thieno[2,3-b]thiophene (1.9 g, 4.1 mmol), 2-bromo-3-decylth-
iophene (2.5 g, 8.1 mmol), and Pd(PPh3)4 (150 mg, 0.13 mmol)
under an atmosphere of dry nitrogen. The mixture was sparged
with dry nitrogen, heated to 110 ꢀC for 17 h, and then cooled to
RT, poured over ice, and extracted with CHCl3. The combined
organic layers were washed with 2 M HCl, brine, and then dried
over Na2SO4 and followed by removal of the solvent by rotary
evaporation. The crude product was purified by column chro-
matography (silica) with petroleum ether as the eluent, followed
by recrystallization from isopropanol/methanol (3 : 1) to give the
product (1.3 g, 51%) as a white powder. dH (300 MHz, CDCl3)
7.20 (d, J ¼ 6.1, 4H), 6.94 (d, J ¼ 5.2, 2H), 2.85–2.66 (m, 4H),
1.63 (s, 4H), 1.24 (s, 28H), 0.85 (t, J ¼ 6.6, 6H). dC (50 MHz,
CDCl3)146.54, 140.64, 138.67, 130.75, 130.70, 130.08, 124.60,
119.21, 32.13, 31.12, 29.83, 29.76, 29.70, 29.57, 29.36, 22.92,
14.36. HRMS (APCI) calculated for [M + H]+ 585.2712, found
585.2712. Calc. for C34H48S4: C, 69.81; H, 8.27; S, 21.92. Found:
C, 69.52; H, 8.26; S, 22.22%.
5,500-Dibromo-3,300-didecyl-2,20:50,200-terthiophene 2. 3,300-
Didecyl-2,20:50,200-terthiophene (1.0 g, 1.9 mmol) was brominated
with NBS (707 mg, 3.9 mmol) according to the general bromi-
nation procedure affording 2 (940 mg, 74%) as a light yellow
powder. dH (200 MHz, C2D2Cl4) 6.96 (s, 2H), 6.88 (s, 2H), 2.78–
2.52 (m, 4H), 1.71–1.41 (m, 4H), 1.41–1.03 (s, 28H), 0.82 (t, J ¼
6.7, 6H). dC (50 MHz, C2D2Cl4) 140.42, 134.88, 132.81, 131.46,
126.31, 110.41, 31.82, 30.46, 29.55, 29.51, 29.36, 29.35, 29.27,
29.19, 29.11, 22.65, 14.18. HRMS (APCI) calculated for [M +
H]+ 687.1181, found 687.1180.
2-[9,9-Dihexyl-7-(2-thienyl)-9H-fluoren-2-yl]thiophene. 2-[9,9-
Dihexyl-7-(2-thienyl)-9H-fluoren-2-yl]thiophene was prepared
according to modified literature procedures.51 To 40 mL of dry
DMF were added 2,7-dibromo-9,9-dihexyl-9H-fluorene (2.5 g,
2,5-Bis(5-bromo-3-decylthiophen-2-yl)thieno[2,3-b]thiophene 4.
2,5-Bis(3-decylthiophen-2-yl)thieno[2,3-b]thiophene (1 g, 1.7
mmol) was brominated with NBS (640 mg, 3.6 mmol) according
to the general bromination procedure affording 4 (807 mg, 65%)
1584 | J. Mater. Chem., 2011, 21, 1582–1592
This journal is ª The Royal Society of Chemistry 2011

View Online
as an off-white powder. dH (200 MHz, CDCl3) 7.13 (s, 2H), 6.91
(s, 2H), 2.64 (t, J ¼ 7.7, 4H), 1.76–1.42 (m, 4H), 1.42–1.00 (m,
28H), 0.83 (t, J ¼ 6.6, 6H). dC (50 MHz, CDCl3) 146.02, 141.17,
137.14, 132.58 (2ꢂ), 131.59, 119.32, 111.03, 31.82, 30.62, 29.52,
29.47, 29.33, 29.25, 29.01, 22.65, 14.18. HRMS (APCI) calcu-
lated for [M + H]+ 743.0901, found 743.0898. Calc. for
C34H46Br2S4: C, 54.98; H, 6.24; S, 17.27. Found: C, 54.49; H,
6.23; S, 17.39%.
(b, 28H), 1.01–0.72 (b, 6H). ATR-IR (cmꢁ1): 3062, 2918, 2849,
1495, 1456, 1375, 1189, 1067, 821, 780, 720.
Poly(3,30-didecyl-terthiophene) PTBT-BiPi. K3PO4 (223 mg,
1.5 mmol), 2 (150 mg, 0.22 mmol), BiPi (56 mg, 0.22 mmol),
Pd(dppf)Cl2 (5 mg, 0.006 mmol) were reacted in 20 mL of dry
DMF according to the general BiPi polymerization procedure.
Pure PTBT-BiPi (56 mg, 48%) was obtained as a dark red
powder. dH (300 MHz, CD2Cl2) 7.45–6.70 (b, 4H), 3.04–2.56(b,
4H), 1.89–1.52 (b, 4H), 1.51–1.11 (b, 28H), 1.04–0.73 (b, 6H).
ATR-IR (cmꢁ1): 3062, 2952, 2918, 2849, 1497, 1456, 1376, 1187,
1065, 821, 782, 720.
2.3 Polymers
General BiPi polymerization procedure. All reactions were
performed under anhydrous conditions under a nitrogen atmo-
sphere. To a three-necked flask of stirring solvent were added
equal molar quantities of monomer and bis(pinacolato)diboron.
The solution was then sparged with dry nitrogen for 10 min
before the addition of 5 mol% 1,10-bis(diphenylphosphino)-
ferrocenepalladium(II)dichloride dichloromethane complex
(Pd(dppf)Cl2; Fig. S4†) and 5 eq. of crushed K3PO4 and an
additional 10 min of sparging. The mixture was heated to 110 ꢀC
and stirred for 24 h, cooled to RT, and the solvent removed by
rotary evaporation. The remaining residue was dissolved in
a minimal amount of CHCl3 and precipitated by pouring slowly
into 1 L of CH3OH. To facilitate the precipitation, 1 mL of
concentrated HCl was then added. The resulting slurry was
stirred for an hour and the precipitate collected by centrifugation
at 4000 rpm for 10 min and dried in vacuo. The crude polymer
was then purified in a Soxhlet extractor using CH3OH, acetone,
and CHCl3, in which the purified polymer dissolved before being
re-precipitated into cold CH3OH and dried in vacuo.
Poly[2,7-(9,9-dihexylfluorene)-alt-bithiophene]
PFBT-BiPi.
K3PO4 (806 mg, 3.8 mmol), 3 (500 mg, 0.76 mmol), BiPi (193 mg,
0.76 mmol), Pd(dppf)Cl2 (14 mg, 0.017 mmol) were reacted in
25 mL of dry DMF according to the general BiPi polymerization
procedure. Pure PFBT-BiPi (279 mg, 74%) was obtained as
a yellow/green powder. dH (300 MHz, CD2Cl2) 8.02–7.47 (b, 6H),
7.47–6.98 (b, 4H), 2.38–1.74 (b, 4H), 1.40–0.91 (b, 12H), 0.91–
0.45 (b, 10H). ATR-IR (cmꢁ1): 3064, 2922, 2850, 1888, 1748,
1606, 1464, 1416, 1375, 1257, 1196, 1134, 1065, 1005, 878, 815,
788, 742, 722.
Poly(2,5-bis(3-decylthiophen-2-yl)thieno[2,3-b]thiophene)
PDTT-BiPi. K3PO4 (551 mg, 2.6 mmol), 4 (400 mg, 0.53 mmol),
BiPi (134 mg, 0.53 mmol), Pd(dppf)Cl2 (7 mg, 0.008 mmol) were
reacted in 25 mL of dry DMF according to the general BiPi
polymerization procedure. Pure PDTT-BiPi (208 mg, 67%) was
obtained as a red powder. dH (300 MHz, CD2Cl2) 7.31–7.12 (b,
2H), 7.12–6.87 (b, 2H), 2.96–2.52 (b, 4H), 1.87–1.52 (b, 4H),
1.52–0.98 (b, 28H), 0.98–0.68 (b, 6H). ATR-IR (cmꢁ1): 3068,
2951, 2917, 2848, 1628, 1512, 1455, 1376, 1260, 1158, 821, 719.
General Stille polymerization procedure.52 All reactions were
performed under anhydrous conditions under a nitrogen atmo-
sphere. To a three-necked flask of stirring solvent were added
equal molar quantities of both monomers. The solution was then
sparged with dry nitrogen for 10 min before the addition of
5 mol% Pd(PPh3)4 and an additional 10 min of sparging. The
mixture was heated to reflux and stirred for 24 h, cooled to RT,
and the solvent removed by rotary evaporation. The remaining
residue was treated identically as the general BiPi polymerization
procedure.
Poly(3,30-didecyl-quaterthiophene)
PQT-Stille.
5,50-Bis-
(trimethylstannyl)-2,20-bithiophene (492 mg, 1.0 mmol), 5,50-
dibromo-4,40-didecyl-2,20-bithiophene (605 mg, 1.0 mmol),
Pd(PPh3)4 (58 mg, 0.05 mmol) were reacted in a mixture of 17 mL
dry DMF and 80 mL dry toluene according to the general Stille
polymerization procedure. Pure PQT-Stille (552 mg, 90%) was
obtained as a dark red powder. dH (300 MHz, CD2Cl2) 7.85–6.61
(b, 6H), 3.18–2.31 (b, 4H), 1.91–1.52 (b, 4H), 1.52–1.04 (b, 28H),
1.00–0.72 (b, 6H). ATR-IR (cmꢁ1): 3061, 2951, 2917, 2848, 1493,
1455, 1435, 1375, 1309, 1158, 1120, 820, 779, 722.
Poly(3,30-didecyl-quaterthiophene) PQT-BiPi. K3PO4 (552 mg,
2.6 mmol), 1 (400 mg, 0.52 mmol), BiPi (132 mg, 0.52 mmol),
Pd(dppf)Cl2 (7 mg, 0.008 mmol) were reacted in 25 mL of dry
DMF according to the general BiPi polymerization procedure.
Pure PQT-BiPi (128 mg, 37%) was obtained as a dark red
powder. dH (300 MHz, CD2Cl2) 7.45–6.73 (b, 6H), 2.98–2.60 (b,
4H), 1.93–1.52 (b, 4H), 1.52–1.03 (b, 28H), 1.03–0.72 (b, 6H).
ATR-IR (cmꢁ1): 3062, 2951, 2918, 2848, 1495, 1455, 1375, 1260,
1065, 821, 780, 720.
Poly(3,30-didecyl-terthiophene)
PTBT-Stille.
2,5-Bis-
(trimethylstannyl)thiophene (410 mg, 1.0 mmol), 5,50-dibromo-
4,40-didecyl-2,20-bithiophene (605 mg, 1.0 mmol), Pd(PPh3)4
(59 mg, 0.05 mmol) were reacted in a mixture of 16 mL dry DMF
and 80 mL dry toluene according to the general Stille polymer-
ization procedure. Pure PTBT-Stille (160 mg, 30%) was obtained
as a dark red powder. dH (300 MHz, CD2Cl2) 6.68–6.15 (b, 4H),
2.42–1.61 (b, 4H), 1.20–0.84 (b, 4H), 0.84–0.33 (b, 28H), 0.33–
0.02 (b, 6H). ATR-IR (cmꢁ1): 3061, 2951, 2916, 2848, 1497, 1456,
1434, 1375, 1260, 1183, 1066, 817, 776, 719.
Poly(3,30-didecyl-quaterthiophene) PQT-BiPi-2. K3PO4 (550
mg, 2.6 mmol), 1 (400 mg, 0.52 mmol), BiPi (132 mg, 0.52 mmol),
Pd(dppf)Cl2 (7 mg, 0.008 mmol) were reacted in a mixture of
8 mL dry DMF and 17 mL dry toluene according to the general
BiPi polymerization procedure. Pure PQT-BiPi-2 (268 mg, 77%)
was obtained as a dark red powder. dH (300 MHz, CD2Cl2) 7.35–
6.85 (b, 6H), 2.97–2.56 (b, 4H), 1.87–1.53 (b, 4H), 1.53–1.05
Poly[2,7-(9,9-dihexylfluorene)-alt-bithiophene]
5,50-Bis(trimethylstannyl)-2,20-bithiophene (491 mg, 1.0 mmol),
PFBT-Stille.
2,7-dibromo-9,9-dihexyl-9H-fluorene (492 mg, 1.0 mmol),
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1582–1592 | 1585

View Online
Pd(PPh3)4 (58 mg, 0.05 mmol) were reacted in a mixture of 17 mL
dry DMF and 83 mL dry toluene according to the general Stille
polymerization procedure. Pure PFBT-Stille (356 mg, 71%) was
obtained as a dark orange powder. dH (300 MHz, CD2Cl2) 8.29–
6.62 (b, 10H), 3.31–2.60 (b, 2H), 2.63–1.78 (b, 2H), 1.49–0.93 (b,
12H), 0.93–0.21 (b, 10H). ATR-IR (cmꢁ1): 3064, 2922, 2850,
1888, 1748, 1602, 1455, 1375, 1256, 1195, 1134, 1066, 1005, 979,
877, 816, 788, 752.
a
palladium catalyst produces boronated products, while
a stronger base (e.g., K2CO3) produces symmetric biaryls.53 This
result indicates that the cross-coupling step is the rate-deter-
mining step overall in the BiPi method (i.e., that the Suzuki cycle
is slower than the Miyaura) because bases specifically activate the
transmetallation step of the Suzuki reaction, meaning that the
polymerization is sensitive to the stoichiometry of BiPi.
However, unlike the ex situ generation of co-monomers dis-
cussed above, BiPi is easy to obtain in high purity, as well as the
bis-bromo co-monomers.
Poly(2,5-bis(3-decylthiophen-2-yl)thieno[2,3-b]thiophene) PDTT-
Stille. 2,5-Bis(trimethylstannyl)thieno[2,3-b]thiophene (600 mg,
1.0 mmol), 5,50-dibromo-4,40-didecyl-2,20-bithiophene (774 mg,
1.0 mmol), Pd(PPh3)4 (75 mg, 0.06 mmol) were reacted in a mixture
of 25 mL dry DMF and 75 mL dry toluene according to the general
Stille polymerization procedure. Pure PDTT-Stille (659 mg, 87%)
was obtained as a dark orange powder. dH (300 MHz, CD2Cl2)
7.42–7.14 (b, 2H), 7.14–6.91 (b, 2H), 3.01–2.41 (b, 4H), 1.98–1.52
(b, 4H), 1.52–1.13 (b, 28H), 1.05–0.73 (b, 6H). ATR-IR (cmꢁ1):
3068, 2918, 2849, 1648, 1536, 1456, 1375, 1302, 1159, 1100, 1063,
963, 898, 818, 720.
The presence of a strong base also promotes deboronation,
where B(OR)2 is lost and then replaced by H during the work-up.
Regardless of the exact mechanism of deboronation, we make
the assumption that the longer the boronated compounds persist
in the reaction mixture, the more likely they are to deboronate.
Thus, if the Suzuki cycle is sufficiently fast, the equilibrium
concentration of BrXB will remain low because it will be quickly
captured by the Suzuki cycle and either initiate or propagate
a polymer chain. If, however, the Suzuki reaction is decelerated,
BrXB will persist (and some BXB may form) and we will observe
more deboronated products. Conversely, if the Suzuki is accel-
erated, we will observe more Br-terminated products. These
pathways are summarized in Scheme 1.
3. Results and discussion
3.1 Nomenclature
For brevity, we use ‘‘X’’ to denote a generic monomer or repeat-
unit, omitting H end-groups. Thus, Xn refers to bis-H-terminated
polymer chains and, for example, SnXn, and BrXnBr refer to
mono-Sn-terminated and bis-Br-terminated polymer chains,
respectively. When necessary, the two monomers of a co-poly-
merization are distinguished with superscripts, e.g., XA and XB;
in these cases Xn is equivalent to (XAXB)n. We refer to all boro-
nate end-groups as ‘‘B’’ (boronate esters are hydrolyzed during
the work-up). We use ‘‘Y’’ to refer to variable end-groups (e.g.,
either B or Br). We abbreviate ‘‘MALDI-TOF’’ as ‘‘MT.’’ We
abbreviate molecular-weight as ‘‘MW’’ (not to be confused with
Mw).
3.3 Comparison of the BiPi and Stille GPC data
To compare the BiPi method to the Stille method, we divided the
‘‘quality’’ of a polymer into four observables: (i) MW, (ii) PDI
(Mw/Mn), (iii) yield, and (iv) uniformity (i.e., distribution of
endgroups). We then synthesized the four thiophene-based,
symmetric co-monomers shown in Scheme 2. We prepared two
sets of four polymers: one by directly polymerizing these
bis-bromo comonomers using the BiPi method (homo-poly-
merization) and the other by co-polymerizing the appropriate
bis-bromo and bis-stannyl monomers (Scheme 2). We chose the
Stille copolymerization as a fair comparison to the BiPi homo-
polymerization for three reasons: (i) the well known tendency of
thiophene-based monomers to deboronate under standard
Suzuki copolymerization conditions, (ii) to compare the
composition of polymers from tin-based crosscouplings to that
of boron-based, and (iii) because Stille co-polymerizations are
more commonly used for thiophene-based monomers.54 We
chose Pd(PPh3)4 over more sophisticated (and possibly better) Pd
3.2 The proposed mechanism of the BiPi method
The likely mechanism of the BiPi method is that of two palla-
dium-mediated catalytic cycles (Fig. S5†). In one cycle
(Miyaura), the active Pd0 species oxidatively inserts into a C–Br
ꢁ
bond and exchanges Brꢁ for K2PO4 (from the base). The
resulting K2PO4PdXBr complex then transmetallates with BiPi
and the PdII reductively eliminates, forming the asymmetric
BXBr monomer in situ (and inert K2PO4B). This monomer can
then either undergo a second palladium-mediated bromine–
boron exchange or enter the other catalytic cycle (Suzuki) and
react with another YXBr (where Y ¼ Br, B, or a growing
polymer chain). The rate-limiting steps for entry and exit in both
cycles are oxidative insertion and transmetallation, respectively.
It does not matter which specific monomers react (e.g., BrXBr
with BrXB, BXB with BrXB, etc.) because the absolute stoichi-
ometry is set by the co-monomer and the BiPi. However, since
oxidative insertion is slower for BrXB than for BrXBr, we
assume that BXB is not present in significant quantities. The
equilibrium stoichiometry will be dictated by the relative rates of
transmetallation in the two cycles. It has been demonstrated that
treating aryl halides with KOAc and BiPi in the presence of
catalysts because it is the most widely used. We adapted the
55
€
specific conditions for the BiPi method from Brase et al. and
Wu et al.56 We characterized the resulting polymers by NMR,
FT-IR, GPC, and MT mass-spectrometry.6,9 The NMR data (see
Experimental) confirmed the connectivity of the polymers, but
were not sensitive enough to differentiate the end-groups. Simi-
larly, FT-IR data confirmed the presence/absence of expected/
unexpected functional groups in the polymers. Neither revealed
any abnormalities with either method. AlthoughGPCisknownto
over-estimate the MW of conjugated polymers,6 we are, in prin-
ciple, comparing structurally identical polymers that differ only by
how they were prepared. Table 1 summarizes these GPC data.
The BiPi method, in all cases, gave a slightly (within one unit)
higher PDI than the Stille method. This may be because of the
relatively slow propagation step (which is limited by trans-
metallation) of the Suzuki cycle, leading to a higher frequency of
1586 | J. Mater. Chem., 2011, 21, 1582–1592
This journal is ª The Royal Society of Chemistry 2011

View Online
Scheme 2
chain-terminating (and monomer-consuming) deboronation
reactions. Monomers 1 and 2 both gave approximately the same
Mn for the Stille and BiPi methods (PQT-BiPi/PQT-Stille), while
the Mw was somewhat higher for the BiPi method. This means
that the BiPi method produces, on average, higher MW chains
than the Stille (presumably because it is a homo-polymerization
chain-terminating reactions due to imbalances in stoichiometry).
Initially, PQT-BiPi was also obtained in a significantly lower
yield than PQT-Stille, suggesting that not all of 1 was converted
to the active co-monomer (i.e., BrXB) in situ, leading to fewer,
longer polymer chains. We thought that solubility may be
playing a role and polymerized 1 by the BiPi method, but using
the same solvents as the Stille, producing PQT-BiPi-2 not only in
a similar yield to PQT-Stille, but also with almost an identical
Mn, Mw, Mz, and PDI. The same did not hold true for 2, which,
despite the structural similarity to 1, produced PTBT-BiPi in
a higher yield, Mn, Mw, Mz, and PDI than PTBT-Stille.
and
does
not
suffer
from
Table 1 GPC data and yields of polymers comparing BiPi and Stille
Monomer/
reaction
Yield
(%)
Polymer Solvent
Mn
Mw
Mz
PDI
The polymerization of 3 using the BiPi method produced
PFBT-BiPi in about the same yield as PFBT-Stille, but with
a much higher Mn and almost the same PDI. This is likely
because fluorenes are more electron-withdrawing than thio-
phenes and therefore increase the rate of the Suzuki cycle.
(Reynolds et al. recently showed that fluorene-based polymers,
lacking thiophenes, can be prepared in very high yields by
a fluoride-promoted BiPi homo-polymerization in the absence of
base.34) We found exactly the opposite for the cross-conjugated
monomer 4, obtaining PDTT-BiPi in lower yield, Mn, Mw, Mz,
and higher PDI than PDTT-Stille. This means that either the
Suzuki or Miyaura cycle (or both) is inhibited significantly for 4.
(Thienothiophene is also more electron-withdrawing than thio-
phene.57) To directly probe the relative rates of the two cycles, we
prepared the BXB co-monomer of 4 and reacted it with 4 under
the same conditions, but encountered difficulties in isolating and
purifying BXB and were unable to isolate any polymer.
1/BiPi
1/BiPi
PQT-
BiPi
PQT-
BiPi-
2
PQT-
DMF
37
77
6709 17 557 49 153 2.62
5204 9555 17 262 1.84
Tol/DMF
(3 : 1)
1/Stille
2/BiPi
2/Stille
3/BiPi
3/Stille
4/BiPi
4/Stille
Tol/DMF
(3 : 1)
90
48
30
74
71
67
87
5875 9488 14 184 1.61
8921 22 870 63 272 2.56
8733 16 142 25 991 1.85
9464 25 800 79 697 2.73
3976 8758 15 760 2.20
5718 16 349 44 590 2.86
10 092 20 169 33 811 1.99
Stille
PTBT- DMF
BiPi
PTBT- Tol/DMF
Stille
(3 : 1)
PFBT- DMF
BiPi
PFBT- Tol/DMF
Stille
(3 : 1)
PDTT- DMF
BiPi
PDTT- Tol/DMF
Stille
(3 : 1)
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1582–1592 | 1587

View Online
3.4 Comparison of BiPi and Stille MALDI-TOF data
Of the four measures of quality, the most telling is the unifor-
mity, which we probed by MT analysis. Unlike GPC, MT tends
to underestimate the MW of polydisperse polymers because
higher MW chains do not desorb easily.58,59 Effective ionization
and desorption are necessary to obtain high quality MT spectra
and the choice of matrix and optimized sample preparation for
each individual type of polymer is critical.60 This includes the
manual adjustment of laser power, affecting the ability to resolve
higher MW chains. We partially offset this using a low-mass gate
(LMG), which blocks low-mass ions from reaching the sensor,
improving the signal-to-noise ratio (SNR).61 What MT does
provide is an accurate picture of the distributions of chains in
terms of end-groups (i.e., uniformity).6
In analyzing MT data, we must take into account the possi-
bility of side-reactions (i.e., reactions that do not form the desired
polymer structure). For example, although the use of Stille co-
polymerizations to form thiophene-based polymers is well
established, side-reactions are prevalent.12,14,49 The homo-
coupling of stannanes is the most common, leading to defects in
the backbone of the polymer and a stoichiometric imbalance of
monomers. Less pronounced is the transfer of methyl groups
(from the tin center) and phenyl groups (from the Pd center) to
the polymer and destannylation by trace acids, all of which
compete with the co-polymerization. The side reactions of the
Suzuki cycle are similar:16,62 homo-coupling, phenyl transfer, and
deboronation. This can not only deplete monomers and termi-
nate growing chains, but also leads to more uniform endgroups.
Homocoupled products cannot be distinguished from hetero-
coupled products because we used symmetric monomers. We did
not observe any products of phenyl-transfer. Direct debromi-
nation is not likely under the BiPi conditions, and it is more likely
that Br is lost via deboronation after the Miyaura cycle. (And in
the transmetallation step of the Stille.63) This means that for the
BiPi method, the three most probable end-groups are B(OH)2
(boronate esters are hydrolyzed during the workup), Br, and H.
The analysis of the possible end-groups produced by Stille co-
polymerizations is much more complicated. As discussed above,
Janssen et al. found that standard Suzuki co-polymerizations
produce 11 combinations of end-groups, excluding side reac-
tions.14–17 We found approximately the same number for Stille
co-polymerizations. However, because the BiPi method is
a homo-polymerization of symmetric co-monomers, it can
produce at most six. An important difference between the BiPi
and Stille methods is that the chains of the copolymers prepared
using the BiPi method initiate and terminate with the same
monomer, while those from the Stille method can initiate or
terminate with either of the two monomers (Fig. S2†). The
distribution of these end-groups will appear as clusters around
n ꢂ M where n is the number of monomers in the chain and M is
the mass-to-charge ratio (m/z) of the most prevalent chain.
Fig. 1 is a comparison of the MT spectra for polymers 1–4,
comparing the BiPi and Stille methods. The calculated mass of
the Xn chains is indicated. The two most obvious features of the
polymers prepared by the BiPi method are that they contain
higher MW chains (with the exception of PDTT) and much more
clearly defined distributions of these chains. By contrast, all of
the polymer chains prepared by the Stille method contain three
Fig. 1 A comparison of the MALDI-TOF spectra for polymers
prepared using the BiPi and Stille methods from monomers 1 (A and B), 2
(C and D), 3 (E and F), and 4 (G and H). The vertical lines are the
calculated masses for polymer chains that are H-terminated (where X is
the repeat unit).
pronounced peaks corresponding to XBXn, Xn, and XnXA—
a result of co-polymerizing two monomers. The distribution of
smaller peaks comprises various combinations of these chains
terminated with Br, H, CH3, and Sn endgroups. We did not
observe any boronate-terminated chains. (They are likely lost
through deboronation after the polymerization is complete and
then converted to H during the work-up.) Thus, the BiPi method
generates only three of the six possible types of chains: Xn, BrXn,
and BrXnBr.
1588 | J. Mater. Chem., 2011, 21, 1582–1592
This journal is ª The Royal Society of Chemistry 2011

View Online
The MT spectrum of PQT-BiPi (Fig. 1B), which was obtained
in a much lower yield than PQT-Stille (Fig. 1A), is composed
almost entirely of Xn chains. In addition to these chains, there are
two smaller peaks corresponding to the BrXn and BrXnBr chains
in decreasing intensity (Fig. 2B). We observed the same trend for
PQT-BiPi-2 (Fig. 2C). Although PQT-BiPi-2 has a lower Mw
than PQT-BiPi, it is composed almost entirely of Xn chains,
indicating deboronation as a likely cause of the lower MW. The
only difference between PQT-BiPi an PQT-BiPi-2 is that the
former was prepared in pure DMF and the latter in 3 : 1 toluene/
DMF. It is known that DMF accelerates standard Suzuki
polymerizations,49,64 but that toluene is a better solvent for the
resulting polymer. Thus, assuming no direct debromination, the
toluene is likely decreasing the relative rate of the Suzuki cycle,
which allows more time for chain-terminating deboronation
reactions. The yield of PQT-BiPi-2, however, is double that of
PQT-BiPi. These data imply that the dominant pathways in the
PQT-BiPi polymerization are rapid chain-growth followed by
deboronation and debromination immediately after initiation,
while the dominant pathway for PQT-BiPi-2 is (slower) chain-
growth followed by deboronation, resulting in a lower Mw and
PDI, but higher yield for PQTD-BiPi-2 than PQTD-BiPi.
The MT spectrum of PQT-Stille, the GPC data for which are
almost identical to those of PQT-BiPi-2, reveals a strikingly
different MT spectrum comprising a relatively even distribution
of end-groups. The two prominent peaks at higher m/z than Xn
correspond to the XBXn and XnXA chains (the MW of XA and XB
are different) and are much more prevalent than the BrXn and
BrXnBr chains are for PQT-BiPi and PQT-BiPi-2 (Fig. 2B). The
small peaks near the baseline comprise the remaining possible
combinations of end-groups (e.g., BrXn, XnXABr, XnSn, etc.).
The MT spectrum of PTBT-BiPi (Fig. 1D) again shows Xn
chains, but no chains above n ¼ 9 are visible. While it is expected
that the intensity of peaks will decay as a function of MW, these
data do not agree with the GPC data, which show a higher Mw
for PTBT-BiPi than PTBT-Stille. The MT spectrum of PTBT-
Stille (Fig. 1C) clearly shows higher MW chains, even though
both spectra were acquired using the same conditions. In order to
see the higher MW peaks for PTBT-BiPi, we had to adjust the
LMG. These data are summarized in Fig. 3. With the LMG set at
4000 m/z, chains up to n ¼ 14 are clearly visible (Fig. 3A), and
chains up to n ¼ 27 are visible (Fig. 3B), though with a much
lower SNR. Once again, a blow-up of the (un-gated) spectrum
Fig. 3 The MALDI-TOF spectrum of PTBT-BiPi with the low-mass
gate set at 4000 showing n ¼ 9–14 (A), a blow-up of this spectrum
comprising the region n ¼ 20–27 (B), and a blow-up of n ¼ 3–4 (C). The
peaks corresponding to bis-H-terminated (Xn) and Br-terminated chains
(XnBr and BrXnBr) are labeled.
shows that there are very few BrXn chains and almost no BrXnBr
chains present (Fig. 3C). By contrast, the spectrum of PTBT-
Stille clearly resolves peaks up to n ¼ 12, but there is much less
uniformity in the chains (Fig. 1C). Peaks corresponding to XBXn
and XnXA chains are clearly visible, in addition to the various
Sn-, Br-, etc. terminated chains. Qualitatively, PTBT-BiPi is
similar to PQT-BiPi and is much more uniform than PTBT-
Stille, though we cannot explain why it was necessary to set the
LMG to 4000 to see chains of n > 9. It may be that PTBT-BiPi is
more crystalline than PTBT-Stille and therefore more difficult to
desorb, or simply fluctuations in the laser in the time (days)
between measurements.
The MT spectrum of PFBT-BiPi (Fig. 1F) shows a higher
degree of BrXn chains than for PQT-BiPi, PQT-BiPi-2, and
PTBT-BiPi, but essentially no BrXnBr (Fig. 4). Given the rela-
tively high yield of PFBT-BiPi, it is not likely that the fluorene
unit is decelerating the Suzuki cycle, causing deboronation of the
monomer (thus leaving unreacted BrXn chains). We postulate
that the Suzuki cycle is accelerated for growing chains containing
fluorene, meaning that BXBr is coupled to BXnBr as soon as it is
formed, while for the growth of PQT-BiPi chains, BrXnB persists
longer, allowing it to be converted to BXnB, leading to a higher
percentage of Xn. It is also possible that 1 and 2 form BXB
monomers that couple with YXnBr. In either case, fluorene is
accelerating the Suzuki cycle. What is clear is that, in agreement
with the GPC data, PFBT-Stille is composed of lower MW
chains comprising a nearly equal mixture of every possible
combination of end-groups (Fig. 1E).
The MT spectrum of PDTT-BiPi (Fig. 1H) shows that the
predominant chains are BrXn and BrXnBr (Fig. 5D), suggesting
Fig. 2 The MALDI-TOF spectrum of PQT-BiPi-2 (A), a blow-up of the
spectrum of PQT-BiPi (B), and a blow-up of the same region of the
spectrum of PQT-BiPi-2 (C). The peaks corresponding to bis-H-termi-
nated (Xn) and Br-terminated chains (XnBr and BrXnBr) are labeled.
Fig. 4 A blow-up of the MALDI-TOF spectrum of PFBT-BiPi with the
peaks corresponding to bis-H-terminated (Xn) and mono-Br-terminated
chains (XnBr) labeled.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1582–1592 | 1589

View Online
due to terthiophene, we performed two additional MT
measurements, one without a matrix (Fig. 5B) and one in
a dithranol matrix (Fig. 5C). We also measured 4 in a terthio-
phene matrix and found that it reacts almost completely with
terthiophene (Fig. 5A).
3.5 Other polymers prepared using BiPi
To help understand the different behaviors of the thiophene-
monomers 1 and 2, versus the fluorene-monomer 3 and thieno-
thiophene-monomer 4 towards the BiPi polymerization, we
prepared two additional monomers and polymerized them using
the BiPi method, forming PTN-BiPi (naphthalene-containing)
and PTB-BiPi (benzothiadiazole-containing; Scheme 3).65
Monomer 5 is expected to behave similarly to 3 because naph-
thalene is more similar to fluorene in terms of reactivity and
electronegativity than it is to thiophene. Likewise, monomer 6 is
expected to behave similarly to 4 because benzothiadiazole is
much more electron-withdrawing and reactive than thiophene,
fluorene, or naphthalene. Indeed, the MT spectrum of PTN-BiPi
is almost identical to that of PFBT-BiPi—the predominant
chains are Xn, with a significant amount of BrXn, but a complete
lack of BrXnBr (ESI†). The MT spectrum of PTB-BiPi is also
very similar to that of PDTT-BiPi. The predominant chains are
BrXnBr, with a small amount of BrXn and almost no Xn (ESI†).
Like PDTT-BiPi, PTB-BiPi (and 6) reacted with the terthio-
phene matrix (PTB-BiPi also incorporated O2) and was obtained
in a lower yield than PTN-BiPi. The increasing amount of Br-
terminated chains going from PTN-BiPi to PTB-BiPi again
suggests that the Suzuki cycle is activated by electron-with-
drawing groups. However, the decreasing yield suggests that at
a certain point, this electronegativity accelerates deboronation as
well—i.e., that the active monomer, BrXB is unstable.
Fig. 5 The MALDI-TOF spectrum of: 4 using a terthiophene matrix
(A), PDTT-BiPi without a matrix (B), PDTT-BiPi in a dithranol matrix
and (C), and a blow-up of the spectrum of PDTT-BiPi in a terthiophene
matrix (D). The peaks corresponding to bis-H-terminated (Xn) and Br-
terminated chains (XnBr and BrXnBr) are labeled. Peaks labeled +88 and
+167 correspond to the addition of one and two terthiophenes, respec-
tively.
that the Suzuki cycle is accelerated. However, PDTT-Stille is
formed in higher yields and MW than PDTT-BiPi—the opposite
of what was observed for PFBT-BiPi and PFBT-Stille. This
means that, either the Miyaura cycle for 4 is so slow that the
reaction time is not sufficient to convert all of it to BrXB, or that
BrXB deboronates too rapidly to efficiently polymerize—we
suspect the latter. The lack of Xn means that BrXnB and BrXB
are less reactive towards the Miyaura cycle than for PFBT-BiPi.
The abundance of BrXnBr (the dominant product) means that
chain-growth occurs mainly between BrXnBr and BrXB. Thus,
the net effect is that entry into the Suzuki cycle is significantly
accelerated (i.e., BrXB enters the Suzuki as soon as it is formed),
but that deboronation is also disproportionately accelerated
compared to PFBT-BiPi. We postulate that the active monomer,
BrXB is simply unstable such that deboronation (or decompo-
sition) competes with the Suzuki cycle. This could be because
thienothiophene groups activate an alternate pathway for
deboronation or slow the exit from the Suzuki cycle (i.e., trans-
metallation). Aside from being more electronegative than thio-
phene, thienothiophene is crossconjugated. It is possible that the
extended conjugation formed by growing polymer chains
accelerates the Suzuki cycle and that this effect is absent for
PDTT-BiPi.
4. Conclusion
The BiPi method enables the syntheses of thiophene-containing
co-polymers via the homopolymerization of symmetric Br-
terminated monomers. These polymers are of a higher quality
and are prepared in less steps than via the equivalent Stille co-
polymerization. These polymers comprise chains that are
initiated and terminated by the same monomer and contain only
two different endgroups: Br or H. The relative proportion of
BrXnBr, BrXn, and Xn polymers can be controlled by
An unexpected feature of the MT of PDTT-BiPi are extra
peaks at Xn +88 and +167 m/z, corresponding to the addition of
one and two terthiophenes from the matrix, respectively, and in
the same relative ratios as BrXn and BrXnBr. This is probably
due to the higher UV cross-section of PDTT-BiPi compared to
the other polymers. To ensure that these extra peaks were in fact
Scheme 3
1590 | J. Mater. Chem., 2011, 21, 1582–1592
This journal is ª The Royal Society of Chemistry 2011

View Online
5 J. Liu, R. S. Loewe and R. D. McCullough, Macromolecules, 1999, 32,
5777.
6 J. Liu and R. D. McCullough, Macromolecules, 2002, 35, 9882.
ꢀ
7 M. Jeffries-El, G. Sauve and R. McCullough, Adv. Mater., 2004, 16,
1017.
8 T. D. McCarley, C. O. Noble, C. J. DuBois and R. L. McCarley,
Macromolecules, 2001, 34, 7999.
9 M. Jayakannan, X. Lou, J. L. J. V. Dongen and R. A. J. Janssen, J.
Polym. Sci., Part A: Polym. Chem., 2005, 43, 1454.
10 I. Adachi, R. Miyakoshi, A. Yokoyama and T. Yokozawa,
Macromolecules, 2006, 39, 7793.
incorporating different groups into the monomers. Generally,
thiophenes give predominantly Xn, aromatic hydrocarbons give
a higher fraction of BrXn, and highly reactive and/or electron
withdrawing groups give predominantly BrXnBr. We interpret
these observations as follows: the Miyaura cycle generates the
active BrXB monomer, the Suzuki grows the polymer, and
deboronation terminates growing chains. The transmetallation
steps of the Miyaura and Suzuki involve breaking a B–B bond
and a C–B bond respectively, thus the Miyaura is generally
faster. Ideally, this leads to the complete conversion of BrXBr to
BrXB, followed by the homo-polymerization of BrXB to form
BrXn. Electron-withdrawing groups accelerate the oxidative
insertion of Pd, balancing the two cycles and favoring the
formation of BrXn in high yields, which we observed for PFBT-
BiPi and PTN-BiPi. If, however, the rate of one cycle is affected
disproportionately, the dominant products will be BrXnBr (faster
Suzuki) or Xn (faster Miyaura). Groups that are too electron-
withdrawing also accelerate deboronation (either by simulta-
neously increasing the oxidative insertion and slowing the
transmetallation steps of the Suzuki cycle or creating unstable
monomers), leading to lower MWs and yields. This appears to be
the case for PDTT-BiPi and PTB-BiPi, which accelerate the
Suzuki cycle disproportionately, forming predominantly
BrXnBr, but in low yields. PQT-BiPi and PTBT-BiPi, on the
other hand, decelerate the Suzuki cycle, forming predominantly
Xn in moderate yields. Thus, assuming that the deboronation
takes place primarily in the Suzuki cycle and that BrXB is
reasonably stable, we conclude that the most relevant factors are
the relative rates that monomers/growing-chains enter the Suzuki
cycle (limited by oxidative insertion) and exit it (limited by
transmetallation).
€
€
11 J. Frahn, B. Karakaya, A. Schafer and A. Schluter, Tetrahedron,
1997, 53, 15459–15467.
12 F. C. Krebs and H. Spanggaard, Sol. Energy Mater. Sol. Cells, 2005,
88, 363.
13 G. Sotgiu, M. Zambianchi, G. Barbarella and C. Botta, Tetrahedron,
2002, 58, 2245.
14 A. Dhanabalan, J. L. J. van Dongen, J. K. J. van Duren,
H. M. Janssen, P. A. van Hal and R. A. J. Janssen,
Macromolecules, 2001, 34, 2495.
15 A. Dhanabalan, P. M. S. Ferreira, A. M. Barros, J. L. J. van Dongen,
P. Dynarowicz-Latka and R. A. J. Janssen, Colloids Surf., A, 2002,
198–200, 313.
16 M. Jayakannan, J. L. J. van Dongen and R. A. J. Janssen,
Macromolecules, 2001, 34, 5386.
17 S. V. Mello, P. Dynarowicz-Latka, A. Dhanabalan, R. F. Bianchi,
R. Onmori, R. A. J. Janssen and O. N. Oliveira, Colloids Surf., A,
2002, 198–200, 45.
18 H. Yan, Z. Chen, Y. Zheng, C. Newman, J. R. Quinn, F. Dotz,
M. Kastler and A. Facchetti, Nature, 2009, 457, 679.
19 S. R. Forrest and M. E. Thompson, Chem. Rev., 2007, 107, 923.
20 J. R. Heath, Annu. Rev. Mater. Res., 2009, 39, 1.
21 A. Dodabalapur, Mater. Today (Oxford, U. K.), 2006, 9, 24.
22 V. Saxena and B. D. Malhotra, Curr. Appl. Phys., 2003, 3, 293.
23 H. E. Katz and J. Huang, Annu. Rev. Mater. Res., 2009, 39, 71.
24 H. Choi and C. C. Mody, Soc. Stud. Sci., 2009, 39, 11.
25 X. Lu, L. Majewski and A. Song, Org. Electron., 2008, 9, 473.
26 S. Tiwari and N. Greenham, Opt. Quantum Electron., 2009, 41, 69.
27 S. Tseng, Y. Chen, H. Meng, H. Lai, C. Yeh, S. Horng, H. Liao and
C. Hsu, Synth. Met., 2009, 159, 137.
This means that not only does the BiPi method enable the
synthesis of otherwise inaccessible co-polymers, but that the
endgroups can be tuned for specific purposes. We hope to further
improve this method by better understanding the effect on the
relative rates of the Suzuki and Miyaura cycles. All of the
molecules we studied were polymerized through thiophenes, with
only the adjacent aryl groups differing. From this, we can
conclude that relatively small changes in the electronic properties
of the monomers used for the BiPi polymerization influence the
relative rates of the two catalytic cycles for the monomers as well
as the growing chains, leading to substantial changes in the
distribution of the end-groups in the resulting polymers.
28 G. Barbarella, M. Melucci and G. Sotgiu, Adv. Mater., 2005, 17, 1581.
€
29 A. Mishra, C. Ma and P. Bauerle, Chem. Rev., 2009, 109, 1141.
30 A. Izumi, R. Nomura and T. Masuda, Chem. Lett., 2000, 728.
31 N. Ma, Z. Zhu and Y. Wu, Tetrahedron, 2007, 63, 4625.
€
32 C. F. Nising, U. K. Schmid, M. Nieger and S. Brase, J. Org. Chem.,
2004, 69, 6830.
33 S. R. Piettre and S. Baltzer, Tetrahedron Lett., 1997, 38, 1197.
34 R. M. Walczak, R. N. Brookins, A. M. Savage, E. M. van der Aa and
J. R. Reynolds, Macromolecules, 2009, 42, 1445.
35 T. Ishiyama, K. Ishida and N. Miyaura, Tetrahedron, 2001, 57, 9813.
36 N. Blouin and M. Leclerc, Acc. Chem. Res., 2008, 41, 1110.
37 I. Osaka and R. D. McCullough, ChemInform Abstract: Advances in
Molecular Design and Synthesis of Regioregular Polythiophenes,
Acc. Chem. Res., 2008, 41, 1202–1214.
38 A. Comel and G. Kirsch, J. Heterocycl. Chem., 2001, 38, 1167.
39 D. E. Seitz, S. Lee, R. N. Hanson and J. C. Bottaro, Synth. Commun.,
1983, 13, 121.
40 M. Heeney, C. Bailey, K. Genevicius, M. Shkunov, D. Sparrowe,
S. Tierney and I. McCulloch, J. Am. Chem. Soc., 2005, 127, 1078–
1079.
41 Z. Zhang, Z. Zheng, L. Qiang, Z. Ma, S. Xie, B. Peng, W. Huang and
W. Wei, Macromol. Rapid Commun., 2006, 27, 1317–1322.
Acknowledgements
H.V. would like to thank NanoNed, funded by the Dutch
Ministry of Economic Affairs (project GMM.9648), for financial
support.
€
42 P. Bauerle, F. Pfau, H. Schlupp, F. Wurthner, K. Gaudl, M. B. Caro
and P. Fischer, J. Chem. Soc., Perkin Trans. 2, 1993, 489–494.
43 R. S. Loewe, P. C. Ewbank, J. Liu, L. Zhai and R. D. McCullough,
Macromolecules, 2001, 34, 4324–4333.
ꢀ
44 J. G. Rodrıguez and J. L. Tejedor, J. Org. Chem., 2002, 67, 7631.
45 J. Li, H. Lee, S. Lee, K. Zong, S. Jin and Y. Lee, Synth. Met., 2009,
159, 201.
46 V. Promarak, A. Pankvuang and S. Ruchirawat, Tetrahedron Lett.,
2007, 48, 1151.
Reference
1 R. Miyakoshi, A. Yokoyama and T. Yokozawa, J. Am. Chem. Soc.,
2005, 127, 17542.
2 A. Yokoyama, H. Suzuki, Y. Kubota, K. Ohuchi, H. Higashimura
and T. Yokozawa, J. Am. Chem. Soc., 2007, 129, 7236.
3 T. A. Skotheim and J. R. Reynolds, Conjugated Polymers, CRC Press,
2006.
€
47 R. Azumi, G. Gotz, T. Debaerdemaeker and P. Bauerle, Chem.–Eur.
€
J., 2000, 6, 735.
48 H. Kanato, K. Takimiya, T. Otsubo, Y. Aso, T. Nakamura, Y. Araki
and O. Ito, J. Org. Chem., 2004, 69, 7183.
4 R. Miyakoshi, A. Yokoyama and T. Yokozawa, J. Polym. Sci., Part
A: Polym. Chem., 2008, 46, 753.
This journal is ª The Royal Society of Chemistry 2011
J. Mater. Chem., 2011, 21, 1582–1592 | 1591

View Online
49 X. Zhang, Y. Qu, L. Bu, H. Tian, J. Zhang, L. Wang, Y. Geng and
F. Wang, Chem.–Eur. J., 2007, 13, 6238.
50 The standard Kumada coupling resulted in mainly asymmetric mono-
coupled product, probably due to steric hindrance. We obtained
better results using a Stille coupling.
56 M. Ning, Z. Zhu and Y. Wu, Tetrahedron, 2007, 63, 4625.
ꢀ
57 B. Milian Medina, A. Van Vooren, P. Brocorens, J. Gierschner,
M. Shkunov, M. Heeney, I. McCulloch, R. Lazzaroni and
J. Cornil, Chem. Mater., 2007, 19, 4949.
58 R. S. Loewe and R. D. McCullough, Chem. Mater., 2000, 12, 3214.
59 F. Babudri, D. Colangiuli, G. Farinola and F. Naso, Eur. J. Org.
Chem., 2002, 2785.
51 C. Querner, A. Benedetto, R. Demadrille, P. Rannou and P. Reiss,
Chem. Mater., 2006, 18, 4817.
ꢀ
52 A. R. Murphy, J. Liu, C. Luscombe, D. Kavulak, J. M. J. Frechet,
60 O. Kazuo, R&D Review of Toyota CRDL, 2006, 41, 29.
61 W. E. Wallace, C. M. Guttman and J. M. Antonucci, Polymer, 2000,
41, 2219.
62 J. Murage, J. W. Eddy, J. R. Zimbalist, T. B. McIntyre, Z. R. Wagner
and F. E. Goodson, Macromolecules, 2008, 41, 7330.
63 Q. Liu, Y. Lan, J. Liu, G. Li, Y. Wu and A. Lei, J. Am. Chem. Soc.,
2009, 131, 10201.
64 Z. Bao, W. K. Chan and L. Yu, J. Am. Chem. Soc., 1995, 117, 12426.
65 The experimental procedures differ slightly from the general methods
reported in the Experimental: see ESI†.
R. J. Kline and M. D. McGehee, Chem. Mater., 2005, 17, 4892.
53 T. Ishiyama, M. Murata and N. Miyaura, Tetrahedron Lett., 1997, 38,
3447.
54 We attempted the syntheses of the requisite asymmetric co-monomers
to directly compare to the Suzuki method as well, but isolating them
from the mixtures of boronates without decomposition was not
possible.
€
55 C. F. Nising, U. K. Schmid, M. Nieger and S. Brase, J. Org. Chem.,
2004, 69, 6830.
1592 | J. Mater. Chem., 2011, 21, 1582–1592
This journal is ª The Royal Society of Chemistry 2011