Suzuki route to regioregular polyalkylthiophenes using Ir-catalysed borylation to make the monomer, and Pd complexes of bulky phosphanes as coupling catalysts for polymerisation

Article abstract of DOI:10.1016/j.tetlet.2006.05.063

We have applied palladium complexes of bulky, electron-rich phosphane ligands as catalysts for the Suzuki synthesis of highly head-to-tail regioregular polyalkylthiophenes from 2-(5-bromo-4-n-alkyl-thiophen-2-yl)-4,4,5,5-tetramethyl-[1,3,2]dioxaborolane. The monomer can be prepared in high yield by Ir-catalysed borylation of 2-bromo-3-hexylthiophene, without the need for organolithium reagents or strong bases.

Full text of DOI:10.1016/j.tetlet.2006.05.063

Tetrahedron Letters 47 (2006) 5143–5146
Suzuki route to regioregular polyalkylthiophenes using Ir-catalysed
borylation to make the monomer, and Pd complexes of bulky
phosphanes as coupling catalysts for polymerisation
Iain A. Liversedge,a Simon J. Higgins,a,* Mark Giles,b Martin Heeneyb
and Iain McCullochb
aDepartment of Chemistry, University of Liverpool, Crown Street, Liverpool L69 7ZD, UK
bMerck Chemicals, Chilworth Technical Centre, Southampton SO16 7QD, UK
Received 4 April 2006; revised 28 April 2006; accepted 11 May 2006
Available online 5 June 2006
Abstract—We have applied palladium complexes of bulky, electron-rich phosphane ligands as catalysts for the Suzuki synthesis of
highly head-to-tail regioregular polyalkylthiophenes from 2-(5-bromo-4-n-alkyl-thiophen-2-yl)-4,4,5,5-tetramethyl-[1,3,2]dioxa-
borolane. The monomer can be prepared in high yield by Ir-catalysed borylation of 2-bromo-3-hexylthiophene, without the need
for organolithium reagents or strong bases.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Polythiophenes have been widely studied as they have
potentially useful electronic and optical properties.1,2
Poly-(3-alkyl)thiophenes are soluble in organic solvents,
but early syntheses afforded polymers which had
essentially regiorandom relative orientation of the alkyl
chains, and these had poor performance in electronic
devices. In particular, the field-effect mobility, a key
parameter for thin-film transistor devices, was typically
To date, the most satisfactory route to regioregular
polyalkylthiophenes uses [NiCl2(dppp)]-mediated coup-
ling of the Grignard reagent generated by metathesis
of 2,5-dibromo-3-alkylthiophene with MeMgBr
(Scheme 1).9,10 Polymers with P98% regioregularity
and MW 25,000, MN 20,000 can be made reproducibly
in moderate yield after purification (60%). However,
(i) stoichiometry control in the Grignard metathesis
reaction is important to obtain high molar mass mate-
rial and (ii) Grignard reagents have limited functional
group tolerance.
610ꢀ5 cm2 V ꢀ1 sꢀ1. The threshold for a worthwhile
organic semiconductor material is P10ꢀ2 cm2 Vꢀ1 sꢀ1
.
3
Regioregular (head-to-tail coupled) polyalkylthiophenes
were first synthesised in the early 1990s.4–6 Interest in
these materials became intense following the discovery
of very high field-effect mobility for solvent-cast films
(ca. 0.05 cm2 V ꢀ1 sꢀ1).7 This was attributed to
self-assembly of polymer chains during solvent casting,
giving a lamellar structure with organised crystalline
domains, and unusually favourable p–p stacking inter-
actions between polymer chains that facilitated inter-
chain charge carrier hopping.8 For the best electrical
performance, it is crucial to have very high (>97%)
regioregularity and purity.
Functionalised polymers have been made using Stille or
Rieke chemistry.11,12 However, stoichiometry control
can be difficult when using Rieke chemistry. Although
Suzuki couplings have been used to prepare oligothioph-
enes, in particular using combinatorial methodology,13–15
*
Corresponding author. Tel.: +44 151 794 3512; fax: +44 151 794
3588; e-mail: shiggins@liv.ac.uk
Scheme 1.
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.05.063

5144
I. A. Liversedge et al. / Tetrahedron Letters 47 (2006) 5143–5146
polymerisations have proven more difficult. Where thio-
phene boronates have been used, deboronation was a
problem, as was the termination of polymerisation by
aryl group transfer from the PPh3 ligands employed.16
A phosphane-free route produced quite high molar mass
polymer, but with low regioregularity;17 in our hands,
the yield has also proved capricious.
As an initial test of possible polymerisation catalysts, we
investigated an analogous dimerisation reaction, the
coupling of 1 with 3 to give 4. For those ligands giving
good results, the optimised conditions were then
employed to polymerise monomer 2. The crude polymer
was fractionated using successive Soxhlet extraction,23
which removes by-products and shorter oligomers. The
fractions containing polymeric material (in order of
increasing MW: hexane, CH2Cl2, THF, CHCl3) were
characterised by NMR, GPC and (where possible)
MALDI mass spectrometry.
Recently, Pd complexes with bulky, electron-rich phos-
phane ligands have been found to be highly active in a
range of cross-coupling reactions,18,19 including Suzuki
couplings.20 These catalysts are extremely reactive
towards oxidative addition, so that commercially-avail-
able aryl chlorides become suitable substrates, whereas
traditional catalysts necessitate the use of more reactive
but less available aryl bromides or iodides. We reasoned
that since thiophene halides are also poor substrates for
oxidative addition (thiophenes are electron-rich aryls),
such catalysts might be suitable for the synthesis of poly-
thiophenes using Suzuki chemistry, if the deboronation
problem could be overcome. Accordingly, we have
investigated a range of these Pd catalysts for the synthe-
sis of regioregular polyalkylthiophenes by the Suzuki
reaction (Scheme 2).
Fu et al. reported that [Pd(tBu3P)2] was an effective
precatalyst for the Suzuki coupling of aryl chlorides.24
We employed this complex, generated in situ from
[Pd2(dba)3] (dba = dibenzylideneacetone) and the
ligand, as a catalyst for the coupling of 1 with 3. Optimis-
ing the conditions (solvent, base, temperature), we
found that the conditions previously used for couplings
with this system using unactivated aryl chloride sub-
strates (3 mol % Pd, 4.5 mol % ligand, 3.3 mol KF as
base, THF at room temperature, 24 h)24 worked best.
This gave the desired unsymmetrical bithiophene 4 in
excellent yield (90%) after chromatography. Prior exam-
1
ination of the crude product by H and 13C NMR spec-
We selected as monomer compound 2, based upon the
work of Janssen et al., who showed that, in polymeri-
sations using thiophene boronates, pinacolates gave
higher molar mass polymers than other boronate
esters.16 Deprotonation of 1 by LDA, reaction with
B(OMe)3, hydrolysis and subsequent reaction of the
thiopheneboronic acid in situ with pinacol, gave 2 in
74% yield.17
troscopy and mass spectrometry showed that little
deboronation had occurred, since unreacted 1 and 3
were the only detectable impurities.
We therefore employed the same reaction conditions in
the polymerisation of 2.25 After purification,23 the high-
est molar mass fraction, extracted by CH2Cl2, was
obtained in 25% yield (Table 1). This had MW 9900, MN
7600, PD 1.30 by GPC against polystyrene standards,
1
Although this preparation of 2 is satisfactory, the
increased functional group tolerance of Suzuki coupling
would clearly be of limited use for widening the range
of accessible functionalised polyalkylthiophenes if it is
necessary to use organolithium reagents to generate
the appropriate boronic acid or boronate ester. Re-
cently, there has been much interest in Ir-catalysed boryl-
ation of arenes with either (pin)B–B(pin) or HB(pin)
(pin = pinacolato), in which an arene C–H bond is acti-
vated and undergoes boronation directly.21 We found
that treatment of 1 with 1 equiv of (pin)B–B(pin) in
the presence of a catalytic amount (5 mol %) of [Ir2(l-
Cl)2(g4-COD)2] (COD = 1,5-cyclooctadiene) and 4,40-
(tBu)2-2,20-bipyridine (10 mol %) in refluxing THF for
16 h gave 2 in 97% yield after chromatography.22
and was 98% regioregular (HT) by H and 13C NMR
spectroscopy. The hexane fraction (35%) was also poly-
meric (Table 1). Although the degree of HT regioregu-
larity of the hexane fraction appeared low, GPC
exaggerates the molar mass of these materials since they
behave more like rigid rods than do the polystyrene cal-
ibration standards, and the high proportion of end
groups for shorter polymer chains will cause the ‘regio-
regularity’ as determined by integration of the arylCH2–
resonances26 to be artificially low (since the thiophene
end groups contribute to the integration of the ‘regio-
irregular’ resonances). A MALDI mass spectrum of
the hexane fraction showed three series of peaks, corre-
sponding to polymer chains terminated by 2H, HBr
and 2Br. The approximate ratio of these peaks was
35:45:20, respectively, at the peak of the distribution
(which was at 10 monomer units for all three series). It
is likely that the MALDI technique underestimates the
average molar mass by preferentially ionising the smal-
ler chains.26 The likely true mean chain length for this
fraction would therefore be somewhere between the 10
units found by MALDI and the 28 found by GPC.
Unfortunately, with available equipment we were un-
able to obtain meaningful MALDI spectra with the
higher molar mass fraction.
The MeOH fraction from this reaction consisted mainly
of unreacted 2, its deboronation product 1, and some
dimer. Therefore, we conclude that this reaction gives
Scheme 2.

I. A. Liversedge et al. / Tetrahedron Letters 47 (2006) 5143–5146
5145
Table 1. Summary of the results of the Pd-catalysed Suzuki polymerisation with electron-rich bulky phosphane ligands
Ligand
tBu3P
Conditions
Polymer fraction
Yield (%)
RRa (%)
MW
MN
7600
4600
3400
PD
1.5 mol % Pd2(dba)3, 4.5 mol % L,
KF, THF, 24 h, rt
As above, but CsF base
2 mol % Pd2(dba)3, 4 mol % L,
CsF, THF, rt, 24 h
2 mol % Pd(OAc)2, 5 mol % L,
K3PO4, THF, 24 h rt, then reflux 3 h
2 mol % Pd(OAc)2, 5 mol % L, K3PO4,
THF, 24 h rt, then reflux 3 h
CH2Cl2
Hexane
(No extraction)
CH2Cl2
Hexane
CH2Cl2
Hexane
THF
25
35
50b
<5
15
17
<10
20
98
88
97
80
<80
92
86
97
95
87
9900
1.30
1.20
3.30
5500
11,100
7600
4600
9100
tBu3P
(o-biphenyl)PtBu2
5
6
8000
1.14
4500
16,900
11,400
5100
15,300
10,000
4250
1.10
1.14
1.20
CH2Cl2
Hexane
21
35
a RR = regioregularity, determined by integration of ArCH2 resonances in the 1H NMR spectra.
b Reaction performed on 0.25 g monomer 2; all other reactions on a 2 g scale.
a good yield (60% total) of highly regioregular polymer,
albeit with rather low molar mass. Use of the more sol-
Acknowledgements
uble base CsF resulted in a small but significant
improvement in molar mass (MW 11,500 without Soxhlet
fractionation; 97% HT regioregularity). Use of isolated
[Pd(tBu3P)2], in place of Pd2dba3/1.5 equiv of ligand,
did not make an appreciable difference.
The authors thank EPSRC for a DTA studentship (IL),
Merck Chemicals for CASE support and Professor A. I.
Cooper for access to his group’s GPC facilities.
Supplementary data
t
Following our moderate success with Bu3P, we next
investigated some (o-biphenyl)PR2 ligands.27 The parent
(o-biphenyl)PtBu2 was unsuccessful, giving low yields of
low molar mass material on attempted polymerisation
of 2. More recently, it has been shown that phosphane
5 is successful in Pd-catalysed Suzuki coupling of boro-
nic acids with unactivated aryl sulfonates, which are
normally highly unreactive.28 We therefore employed
this ligand in an attempted polymerisation of 2, using
the published optimum conditions (Table 1). Although
polymer was obtained in low yield, the regioregularity
and molar mass were low (Table 1).
Scans of 1H and 13C NMR spectra and mass spectra for
1
compound 2, and of H and 13C NMR spectra for the
high molecular weight polymer fractions, are available.
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/j.tetlet.2006.
05.063.
References and notes
1. Handbook of Oligo- and Polythiophenes; Fichou, D., Ed.;
Wiley-VCH: Weinheim, 1999.
Further improvements in Suzuki coupling with this fam-
ily of ligands has recently been noted for 6.29,30 Accord-
ingly, we employed this ligand in a polymerisation of 2,
using the same conditions as for ligand 5.31 We were
pleased to find that this system gave excellent results.
Firstly, it was the only one to give appreciable amounts
(21% yield) of THF-extracted polymer. The latter frac-
tion contained >97% HT polymer with good molecular
weight (Table 1). The CH2Cl2 and hexane fractions were
also polymeric (Table 1).
2. Roncali, J. Chem. Rev. 1992, 92, 711.
3. Assadi, A.; Willander, M.; Svensson, C.; Hellberg, J.
Synth. Met. 1993, 58, 187.
4. McCullough, R. D.; Lowe, R. D. J. Chem. Soc., Chem.
Commun. 1992, 70.
5. Chen, T. A.; Rieke, R. D. J. Am. Chem. Soc. 1992, 114,
10087.
6. McCullough, R. D.; Lowe, R. D.; Jayaraman, M.;
Anderson, D. K. J. Org. Chem. 1993, 58, 904.
7. Bao, Z.; Dodabalapur, A.; Lovinger, A. J. Appl. Phys.
Lett. 1996, 69, 4108.
8. Sirringhaus, H.; Brown, P. J.; Friend, R. H.; Nielsen, M.
M.; Bechgaard, K.; Langeveld-Voss, B. M. W.; Spiering,
A. J. H.; Janssen, R. A. J.; Meijer, E. W.; Herwig, P.; de
Leeuw, D. M. Nature 1999, 401, 685.
9. Loewe, R. S.; Khersonsky, S. M.; McCullough, R. D. Adv.
Mater. 1999, 11, 250.
10. Loewe, R. S.; Ewbank, P. C.; Liu, J. S.; Zhai, L.;
McCullough, R. D. Macromolecules 2001, 34, 4324.
11. Ewbank, P. C.; Nuding, G.; Suenaga, H.; McCullough, R.
D.; Shinkai, S. Tetrahedron Lett. 2001, 42, 155.
12. Bjornholm, T.; Greve, D. R.; Reitzel, N.; Hassenkam, T.;
Kjaer, K.; Howes, P. B.; Larsen, N. B.; Bogelund, J.;
Jayaraman, M.; Ewbank, P. C.; McCullough, R. D.
J. Am. Chem. Soc. 1998, 120, 7643.
The catalyst utilising this ligand therefore gave a total of
73% yield of regioregular, polymeric product. Although
the molar mass, and hence the proportion of polymer
extracted into the later solvents, was lower than that
which we typically obtain for the optimised Grignard
coupling chemistry, it is nonetheless very significant that
the regioregularity and total yield approaches that of the
Grignard metathesis route, yet monomer 2 can be made
by Ir-catalysed borylation without the use of strongly-
basic or nucleophilic reagents, and can be purified by
column chromatography. We are now exploring the
application of this chemistry to the syntheses of regio-
regular polyalkylthiophene derivatives bearing func-
tional groups for various purposes.
13. Melucci, M.; Barbarella, G.; Sotgiu, G. J. Org. Chem.
2002, 67, 8877–8884.

5146
I. A. Liversedge et al. / Tetrahedron Letters 47 (2006) 5143–5146
14. Briehn, C. A.; Ba¨uerle, P. J. Comb. Chem. 2002, 4, 457.
15. Spivey, A. C.; Turner, D. J.; Turner, M. L.; Yeates, S.
Org. Lett. 2002, 4, 1899.
16. Jayakannan, M.; van Dongen, J. L. J.; Janssen, R. A. J.
Macromolecules 2001, 34, 5386; Jayakannan, M.; Lou, X.;
Van Dongen, J. L. J.; Janssen, R. A. J. J. Polym. Sci. A,
Polym. Chem. 2005, 43, 1454.
Soxhlet thimble. The precipitate was Soxhlet-extracted
with, successively, MeOH, acetone, hexane, CH2Cl2 and
THF. The hexane, CH2Cl2 and THF fractions were
separately evaporated to small volume and poured into
MeOH, and the precipitates were filtered off and dried
in vacuo. Yield (hexane fraction) 0.605 g, 35%, (CH2Cl2
fraction) 0.356 g, 21% (THF fraction) 0.04 g of regio-
regular poly-3-hexylthiophene. Found: C, 71.5; H, 8.00;
17. Guillerez, S.; Bidan, G. Synth. Met. 1998, 93, 123.
´
18. Hassan, J.; Sevignon, M.; Gozzi, C.; Schulz, E.; Lemaire,
S, 18.5; Br, 2.1. {C10H14S} requires C, 72.23; H, 8.49;
1
M. Chem. Rev. 2002, 102, 1359.
19. Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 2002, 41,
4176.
S, 19.28. Selected 1H NMR (400 MHz, CDCl3): d ppm
(J Hz): 6.98 (s, 1H, HT alkylthiophene 4-H), 2.79 (br t,
2H,
J
6.3, arylCH2–). 13C{1H} NMR (101 MHz,
20. Miyaura, N. Top. Curr. Chem. 2002, 219, 11.
21. Ishmaya, T.; Miyaura, N. J. Organomet. Chem. 2003,
680, 3.
CDCl3): d ppm: 139.0, 132.8, 129.6, 127.7 (thiophene
C, HT monomer unit). GPC: MW = 9900, MN = 7600,
PD = 1.30 (all data for CH2Cl2 fraction).
22. Into a flame dried three-neck flask equipped with a
condenser, 1 (0.5 g, 2.02 mmol), di-l-chloro-bis(g4-1,5-
cycloctadiene)diiridium (0.07 g, 0.101 mmol), 4,40-di-tert-
butyl-2,20-dipyridyl (0.054 g, 0.202 mmol) and 4,4,5,5-
tetramethyl-[1,3,2]-dioxaborane (0.30 cm3, 2.07 mmol)
were added and the flask was purged with N2. Anhydrous
THF (20 cm3) was added and the reaction was refluxed for
16 h. After cooling and quenching with water (15 cm3), the
product was extracted into diethyl ether (5 · 20 cm3). The
combined organic extracts were then washed with brine
(2 · 50 cm3), dried over anhydrous MgSO4, filtered, and
the solvent removed in vacuo. The pure product was
obtained after column chromatography (dichloromethane;
Et3N-treated silica; 0.747 g, 97%). Found: C, 51.59; H,
7.13; S, 8.34. C16H26SO2BBr requires C, 51.50; H, 7.02; S,
8.59. NMR data identical with lit.16
23. Trznadel, M.; Pron, A.; Zagorska, M.; Chrzaszez, R.;
Pielichowski, J. Macromolecules 1998, 31, 5051.
24. Littke, A. F.; Fu, G. C. Angew. Chem. Int. Ed. 1998, 37,
3387.
25. To KF (2.0 g, 34 mmol) and [Pd2(dba)3] (0.095 g,
0.103 mmol), in a Schlenk flask under argon, was added
2 (3.88 g, 10.4 mmol) in THF (20 cm3), and a solution
of PtBu3 in THF (2.2 cm3 of 0.12 M, 0.264 mmol). The
reaction was stirred for 24 h at room temperature. The
solution was then poured into MeOH (200 cm3) to yield
a purple precipitate, which was then filtered through a
26. Liu, J. S.; Loewe, R. S.; McCullough, R. D. Macromole-
cules 1999, 32, 5777.
27. Wolfe, J. P.; Singer, R. A.; Yang, B. H.; Buchwald, S. L. J.
Am. Chem. Soc. 1999, 121, 9550.
28. Nguyen, H. N.; Huang, X. H.; Buchwald, S. L. J. Am.
Chem. Soc. 2003, 125, 11818.
29. Walker, S. D.; Barder, T. E.; Martinelli, J. R.; Buchwald,
S. L. Angew. Chem., Int. Ed. 2004, 43, 1871.
30. Barder, T. E.; Walker, S. D.; Martinelli, J. R.; Buchwald,
S. L. J. Am. Chem. Soc. 2005, 127, 4685.
31. Into a flame-dried Schlenk flask was placed anhydrous
K3PO4 (3.4 g, 16.1 mmol), Pd(OAc)2 (0.024 g, 0.107 mmol),
2 (2.00 g, 5.36 mmol) and 2-dicyclohexylphosphino-20,60-
dimethoxybiphenyl (6; 0.044 g, 0.107 mmol) under Ar.
Anhyd THF (50 cm3) was added and the reaction was
stirred at room temperature for 16 h. It was then brought to
reflux for a further 3 hours, allowed to cool to room
temperature, and poured into MeOH (200 cm3). The
precipitate formed was filtered through a Soxhlet thimble,
and worked up as described above.25 Yield (CH2Cl2)
0.174 g, 21% (THF) 0.175 g, 20%, (hexane) 0.287 g, 33%.
1
From H NMR data (determined as above25), CH2Cl2
fraction: 95% HT regioregularity, GPC (in THF)
MW = 11400, MN = 10000, PD = 1.14. Data for THF
fraction: >97% HT regioregularity. Found: C, 71.2; H,
7.98; S, 19.1; Br, 1.9. {C10H14S} requires C, 72.23; H, 8.49;
1
S, 19.28. MW = 16900, MN = 15300, PD = 1.10.