Influence of the bulkiness of the substituent on the aggregation and magnetic properties of poly(3-alkylthiophene)s

Article abstract of DOI:10.1002/pola.26973

A series of poly(3-alkylthiophene)s (P3ATs) (P1-P5) has been synthesized via a Ni(dppp)-mediated polymerization, varying the bulkiness of the alkyl side chains in order to investigate the influence of the bulkiness of the alkyl substituent on the aggregation and magnetic properties of P3ATs. UV-Vis spectroscopy, performed in solution as well as in film, shows that the stacking of the polymers becomes more complicated as the bulkiness of the side chains increases. Both the π-interactions and the planarization of the polymer chains are diminished. While aggregation is absent in poor solvent for the polymer with the most bulky side chains, aggregation was present in film, albeit slowed down. This behavior was also confirmed by X-ray diffraction (XRD) and differential scanning calorimetry (DSC) experiments. Electron spin resonance (ESR) measurements, performed at 300 K on powders, confirmed the trend of decreasing supramolecular order with increasing bulkiness of the side-chain. Magnetization measurements, performed at 5 and 300 K, are in line with our hypothesis on the influence of π-interactions and the fraction of planar polymer chains on the coercivity and saturation magnetization, respectively.

Full text of DOI:10.1002/pola.26973

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
Influence of the Bulkiness of the Substituent on the Aggregation and
Magnetic Properties of Poly(3-alkylthiophene)s
1
2
2
3
4
ꢀ
Helmuth Peeters, Mihaela Jivanescu, Andre Stesmans, Lino M. C. Pereira, Leander Dillemans,
4
4
5
1
ꢀ
Jean-Pierre Locquet, Margriet J. Van Bael, Andre Persoons, Guy Koeckelberghs
1Laboratory for Polymer Synthesis, KU Leuven, Celestijnenlaan 200F, 3001 Heverlee, Belgium
2Semiconductor Physics Laboratory, KU Leuven, Celestijnenlaan 200D, 3001 Leuven, Belgium
3Nuclear and Radiation Physics Section, KU Leuven, Celestijnenlaan 200D, 3001 Leuven, Belgium
4Laboratory of Solid State Physics and Magnetism, KU Leuven, Celestijnenlaan 200D, 3001 Leuven, Belgium
5Laboratory of Molecular Electronics and Photonics, KU Leuven, Celestijnenlaan 200D, 3001 Heverlee, Belgium
Correspondence to: G. Koeckelberghs (E-mail: guy.koeckelberghs@chem.kuleuven.be)
Received 12 June 2013; accepted 4 October 2013; published online 29 October 2013
DOI: 10.1002/pola.26973
ABSTRACT: A series of poly(3-alkylthiophene)s (P3ATs) (P1–P5)
has been synthesized via a Ni(dppp)-mediated polymerization,
varying the bulkiness of the alkyl side chains in order to inves-
tigate the influence of the bulkiness of the alkyl substituent on
the aggregation and magnetic properties of P3ATs. UV–Vis
spectroscopy, performed in solution as well as in film, shows
that the stacking of the polymers becomes more complicated
as the bulkiness of the side chains increases. Both the p-
interactions and the planarization of the polymer chains are
diminished. While aggregation is absent in poor solvent for the
polymer with the most bulky side chains, aggregation was
present in film, albeit slowed down. This behavior was also
confirmed by X-ray diffraction (XRD) and differential scanning
calorimetry (DSC) experiments. Electron spin resonance (ESR)
measurements, performed at 300 K on powders, confirmed the
trend of decreasing supramolecular order with increasing bulk-
iness of the side-chain. Magnetization measurements, per-
formed at 5 and 300 K, are in line with our hypothesis on the
influence of p-interactions and the fraction of planar polymer
chains on the coercivity and saturation magnetization, respec-
C
V
tively.
2013 Wiley Periodicals, Inc. J. Polym. Sci., Part A:
Polym. Chem. 2014, 52, 76–86
KEYWORDS: conjugated polymers; ESR; p-interactions; poly(3-
alkylthiophene); magnetism; SQUID
substitution pattern and the connectivity are crucial parame-
ters for a (ferro)magnetic spin coupling causing spin align-
ment.6 Quite some “high-spin” molecules,7,8 oligomers,9 and
polymers10–13 have been reported based on this approach
and remarkable progress has been achieved.
INTRODUCTION Conjugated polymers have been explored
intensively for many years. Their optical, electrical, and elec-
tronic properties1 have been studied in detail, which has
resulted in many applications, such as solar cells and light-
emitting diodes. In contrast, the magnetic behavior has
received far less attention. The main reason is that ferromag-
netism in organic compounds requires the presence of
unpaired electrons (spins) and that spins in organic mole-
cules are typically paired. Nevertheless, unpaired spins can
be present in organic molecules, in which case they are often
introduced as radicals, carbenes, or by oxidation.2–4 A second
requirement is a pathway by which unpaired spins interact
(spin coupling).5 Within one molecule a conjugated p-system
is ideal to act as a (ferro)magnetic linker between different
spins. When stacked, there may also be inter-polymer (mole-
cule) interaction between occurring spins. Therefore, conju-
gated polymers are in fact particularly suited as organic
material possibly exhibiting (ferro)magnetic behavior. The
Besides the “high-spin” molecules, also some conjugated poly-
mers were reported as showing magnetic behavior under
very specific conditions. Typical examples are poly(aniline)s,14–17
poly(pyrrole),18
regio-irregular
poly(3-alkylthiophene)s
(P3ATs),19–23 and (substituted) poly(acetylene)s.24 In all
these cases, the materials were chemically or electrochemi-
cally doped, introducing spins (polarons). In contrast to
these results, we have previously reported on the magnetic
behavior of neutral (undoped) poly(thiophene)s carrying an
alkyl, alkoxy, or alkylthio sidechain.25,26 The samples were
examined by ESR spectroscopy and SQUID magnetometry.
The measurements reveal the presence of (at least) two
Additional Supporting Information may be found in the online version of this article.
C
V
2013 Wiley Periodicals, Inc.
76
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
spin systems. A first electron spin system gives rise to a par-
amagnetic behavior. A second spin system results in a ferro-
magnetic behavior with zero coercivity and remanence at
300 K and (for most polymers) with significant coercivity
and remanence at 5 K. The origin of the latter spin system is
unknown, but different from the ESR-active one. Although
the origin of the spin systems and the way they interact is
not fully understood, some hypotheses can nevertheless be
made. First, the amount of Bohr-magnetons measured by
SQUID magnetometry seems to be dependent on the number
of planar polymer chains. This might suggest that conjuga-
tion within one polymer chain is important. Second, concern-
ing the coercivity, it was found that polymers in which
strong p-interactions between stacked polymer chains are
present show the highest coercivity. Therefore, it can be
hypothesized that these p-interactions determine the ferro-
magnetic behavior (coercivity).
In this article, these hypotheses are tested by investigating a
series of P3ATs (Fig. 1). P3ATs were preferred since they can
be prepared using a controlled chain-growth mechanism27,28
providing optimal control on the molecular structure. More-
over, driven by p-interactions, P3ATs with linear side-chains
have a strong tendency to stack. The polymers studied are
all regio-regular (head-to-tail coupled) and have the same
molar mass, but differ in the bulkiness of the substituent. In
particular, a hexyl substituent was used in all cases, but an
increasingly longer branched side chain was employed. The
increase of the bulkiness should result in a larger distance
between the stacked P3ATs, weakening the p-interactions,
but leaving all other parameters unchanged. This should
therefore allow the verification of the influence of p-
interactions on coercivity. Moreover, in cases where the p-
stacking is prevented by a bulky alkyl substituent, it might
also affect the planarization of the polymer chains. These
polymers thus not only allow the study of the influence of
the p-interactions on the coercivity, but also the influence of
the fraction of planar polymer chains on the number of Bohr
magnetons.
FIGURE 1 Structure of P3ATs with increasingly more bulky
side chains.
2h scans were performed on a PANalytical X’pert Pro X-ray
diffractometer using Cu Ka radiation with a wavelength of
1.54 Å. An amount of powder was placed without additional
fixation on the horizontal sample holder plate with the
incoming X-ray beam impinging in a fixed direction parallel
to this horizontal plane while the detector was scanned.
The powders were weighted and stored under inert (argon)
atmosphere.
Conventional continuous wave slow-passage X-band (ꢀ9.2
GHz) ESR measurements were carried out at room tempera-
ture using a Jeol FA100 spectrometer. Low microwave power
(Pm ꢁ 1 mW) first derivative-absorption dPm/dB spectra were
recorded through applying sinusoidal modulation (ꢀ100 kHz;
amplitude Bm ꢀ0.35 G) of the externally applied magnetic field
B. Some of the observations were made in a locally built K-
band (ꢀ20.6 GHz) setup, as described elsewhere.28 A co-
mounted calibrated MgO:Mn21 reference sample was used for
absolute g-factor and spin density (spin S5 1/2) calibration,
with the latter performed through double numerical integra-
tion of the detected dPm/dB spectra. At X band, the g values of
the third and fourth hyperfine-split lines of the Mn21 (55Mn;
100% natural abundance; nuclear spin I5 5/2) sextet were
calibrated as 2.0338 6 0.0001 and 1.9807 6 0.0001, respec-
tively. The attained absolute and relative accuracy is estimated
at ꢀ20% and ꢀ5%, respectively. Signal averaging (typically
ꢀ50 scans) was routinely applied to enhance spectral quality.
Therefore, a first part of this article investigates how the bulki-
ness of the alkyl substituent influences the aggregation behav-
ior of P3AT. Next, the magnetic properties of the polymers are
studied and the aforementioned hypotheses are tested.
EXPERIMENTAL
Gel permeation chromatography (GPC) measurements were
done with a Shimadzu 10A apparatus with a tunable absorb-
ance detector and a differential refractometer in tetrahydro-
furan (THF) as eluent toward polystyrene standards.
UV–vis spectra were recorded with a Varian Cary 400. The DSC
experiments were performed on a DSC 7 from Perkin Elmer.
SQUID-based magnetometry is carried out in a MPMS-XL
magnetometer (Quantum Design). The powders are weighted
and a typical amount of 15 mg is fixed between small pieces
of cotton wool inside the non-magnetic plastic transparent
sample tube. The magnetic signal from the plastic sample
tube with cotton wool (without the powder sample) is
Films were prepared by spin coating from chloroform solu-
tions (1 mg/0.1 mL) on a glass substrate (spinning speed:
1500 rpm; spinning time: 30 s). A Mettler-Toledo FP900
Thermosystem was used for sample annealing.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86
77

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
measured separately. Magnetization was measured at differ-
ent fixed temperatures as a function of magnetic field in RSO
(reciprocating sample option) operation mode.
(dd, J 5 3.1 Hz, J 5 4.9 Hz, 1H, Ar H), 6.93 (m, 2H, Ar H),
2.61 (t, 2H, CH2), 1.59–1.54 (m, 2H, CH2), 1.28 (m, 9H, CH2,
CH), 0.89 (t, 6H, CH3). 13C NMR (CDCl3): d 5 143.7, 128.4,
125.2, 119.7, 36.8, 36.0, 34.62, 27.6, 19.9, 14.7. MS: 211
(MH1) (calc. 211.4).
When measuring weak magnetic responses, one needs to be
very cautious for (magnetic) contaminations.29 Therefore,
working conditions were kept as clean as possible and any
contact of the sample powders with metals (spatula, etc.)
was avoided. Moreover, control ESR experiments were per-
formed on several samples before and after the SQUID mag-
netization measurements, in which no traces of known
ferromagnetic metals (Fe, Co, and Ni) where detected in
atomic/ionic paramagnetic form.
3-(2-Butyllhexyl)thiophene (2e). The used reagents were 1e
(10.6 mmol; 2.34 g), 3-bromothiophene (9.71 mmol; 1.58 g),
Mg (11.5 mmol; 281 mg), and Ni(dppp)Cl2 (46.1 mmol; 25.0
mg). Yield: 428 mg (20%). 1H NMR (CDCl3): d 5 7.22 (dd,
J 5 3.0 Hz, J 5 4.9 Hz, 1H, Ar H), 6.90 (m, 2H, Ar H), 2.56 (d,
2H, CH2), 1.59 (m, 1H, CH), 1.25 (m, 12H, CH2), 0.87 (t, 6H,
CH3). 13C NMR (CDCl3): d 5 142.0, 12.0, 124.9, 120.8, 39.1,
34.9, 33.2, 29.0, 23.2, 14.3. MS: 225 (MH1) (calc. 225.4).
Compound 1c,30 1d,31 1e,32 and 4a33 were synthesized as
described in literature.
General Procedure for the Bromination of the
Alkylthiophenes
NBS (1 eq) was added to a solution of alkylthiophene (1 eq)
in dichloromethane (0.2 M) and shielded from light. The
reaction was stirred overnight under argon atmosphere.
Afterwards, the reaction mixture was neutralized with a
10% Na2S2O3 solution, extracted with dichloromethane, and
washed with a saturated NaCl solution. The organic layer
was dried with MgSO4 and the solvent was removed under
reduced pressure. The crude product was further purified
with column chromatography (SiO2; eluent: petroleum ether)
to obtain a colorless oil.
Synthesis of P1–P5
General Procedure for the Synthesis of 3-Alkylthiophene
A solution of the alkyl bromide (1.1 eq) in dry diethylether
(2 M) was added to a suspension of Mg (1.2 eq) in dry
diethyl ether (2 M) under an argon atmosphere. After addi-
tion of the alkyl bromide, the reaction mixture was refluxed
for 1 h and then allowed to cool to room temperature. Sub-
sequently, the mixture was added to a solution of 3-
bromothiophene (1 eq) and Ni(dppp)Cl2 (0.01 eq) in dry
diethyl ether (1 M) and under an argon atmosphere an
refluxed overnight. After cooling to room temperature, the
mixture was poured into a 1 M HCl solution and extracted
with diethyl ether. The organic layer was washed with a sat-
urated NaCl solution, dried over MgSO4 and concentrated in
vacuo. The crude product was further purified by column
chromatography (SiO2; eluent: petroleum ether), affording a
colorless oil.
2-Bromo-3-(5-methylhexyl)thiophene (3b). The used
reagents were 2b (5.00 mmol; 912 mg) and NBS (5.00
mmol; 890 mg). Yield: 1.15 g (88%). 1H NMR (CDCl3):
d 5 7.18 (d, J 5 5.7 Hz, 1H, Ar H), 6.79, (d, J 5 5.7 Hz, 1H, Ar
H), 2.56 (t, 2H, CH2), 1.58–1.54 (m, 3H, CH2, CH), 1.32 (m,
2H, CH2), 1.21 (m, 2H, CH2), 0.87 (d, 6H, CH3). 13C NMR
(CDCl3): d 5142.1, 128.4, 125.3, 108.9, 38.9, 30.1, 29.6, 28.1,
27.2, 22.8. MS: 262 (M1) (calc. 261.2).
3-(5-Methylhexyl)thiophene (2b). The used reagents were
1b (27.9 mmol; 5.00 g), 3-bromothiophene (24.3 mmol; 3.96
g), Mg (27.9 mmol; 678 mg), and Ni(dppp)Cl2 (244 mmol;
2-Bromo-3-(4-ethylhexyl)thiophene (3c). The used reagents
were 2c (2.50 mmol; 491 mg) and NBS (2.50 mmol; 445
mg). Yield: 434 mg (63%). 1H NMR (CDCl3): d 5 7.18 (d,
J 5 5.6 Hz, 1H, Ar H), 6.80, (d, J 5 5.6 Hz, 1H, Ar H), 2.54 (t,
2H, CH2), 1.54 (m, 2H, CH2), 1.27 (m, 7H, CH2, CH2, 0.83 (t,
6H, CH3). 13C NMR (CDCl3): d 5 142.1, 128.4, 125.3, 108.9,
40.3, 32.4, 29.9, 27.1, 25.5, 11.0. MS: 275 (M1) (calc. 275.2).
1
132 mg). Yield: 1.39 g (31%). H NMR (CDCl3): d 5 7.23 (m,
1H, Ar H), 6.92 (m, 2H, Ar H), 2.62 (t, 2H, CH2), 1.63–1.58
(m, 3H, CH2, CH), 1.33 (m, 2H, CH2), 1.21 (m, 2H, CH2), 0.86
(d, 6H, CH3). 13C NMR (CDCl3): d 5 143.4, 128.4, 125.2,
119.9, 39.0, 31.0, 30.5, 28.07, 27.3, 22.8. MS: 182 (M1) (calc.
182.3).
2-Bromo-3-(3-propylhexyl)thiophene (3d). The used
reagents were 2d (5.95 mmol; 1.25 g) and NBS (5.95 mmol;
1.06 g). Yield: 702 mg (41%). 1H NMR (CDCl3): d 5 7.18 (d,
J 5 5.6 Hz, 1H, Ar H), 6.79, (d, J 5 5.6 Hz, 1H, Ar H), 2.54 (t,
2H, CH2), 1.54 (m, 2H, CH2), 1.27 (m, 9H, CH2,CH), 0.89 (t,
6H, CH3). 13C NMR (CDCl3): d 5 142.4, 128.3, 125.3, 108.8,
36.8, 36.0, 33.7, 26.8, 19.9, 14.7. MS: 289 (MH1) (calc.
290.3).
3-(4-Ethylhexyl)thiophene (2c). The used reagents were 1c
(8.22 mmol; 1.59 g), 3-bromothiophene (7.50 mmol; 1.22 g),
Mg (8.50 mmol; 207 mg), and Ni(dppp)Cl2 (74.8 mmol; 40.6
1
mg). Yield: 707 mg (48%). H NMR (CDCl3): d 5 7.24 (m, 1H,
Ar H), 6.93 (m, 2H, Ar H), 2.61 (t, 2H, CH2), 1.62–1.57 (m,
2H, CH2), 1.28–1.26 (m, 7H, CH2, CH), 0.83 (t, 6H, CH3). 13C
NMR (CDCl3): d 5 143.5, 128.4, 125.2, 119.89, 40.4, 32.59,
30.9, 27.9, 25.6, 11.1. MS: 196 (M1) (calc. 196.4).
2-Bromo-3-(2-butylhexyl)thiophene
(3e).
The
used
3-(3-Propylhexyl)thiophene (2d).. The used reagents were
1d (18.9 mmol; 3.73 g), 3-bromothiophene (16.3 mmol; 2.66
g), Mg (19.6 mmol; 477 mg), and Ni(dppp)Cl2 (88.5 mmol;
48.0 mg). Yield: 1.25 g (36%). 1H NMR (CDCl3): d 5 7.24
reagents were 2e (1.91 mmol; 428 mg) and NBS (1.91
mmol; 340 mg) and the reaction mixture was heated to 40
ꢂC. Yield: 460 mg (79%). 1H NMR (CDCl3): d 5 7.18 (d,
J 5 5.6 Hz, 1H, Ar H), 6.76, (d, J 5 5.6 Hz, 1H, Ar H), 2.49 (d,
78
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
2H, CH2), 1.64 (m, 1H, CH), 1.26 (m, 12H, CH2), 0.88 (t, 6H,
CH3). 13C NMR (CDCl3): d 5 141.3, 129.0, 125.1, 109.6, 38.7,
34.2, 33.2, 28.9, 23.2, 14.3. MS: 304 (M1) (calc. 303.3).
suspension of Ni(dppp)Cl2 (2.5 mol %) in dry THF under an
argon atmosphere. After stirring the solution overnight at
room temperature, the reaction mixture was quenched with
a 1 M HCl solution and precipitated in MeOH. The filtrate
was further extracted with MeOH and CHCl3 (P1–P3, P5) or
chlorobenzene (P4). The CHCl3 (or chlorobenezene) soluble
fraction was precipitated in MeOH, filtered off, and dried.
General Procedure for the Iodination of the 2-Bromo-3-
alkylthiophenes
A solution of 2-bromo-3-alkylthiophene in THF (0.2 M)
under argon atmosphere was shielded from light and iodine
(0.5 eq) and iodobenzene diacetate (0.5 eq) were added. The
reaction was stirred at room temperature for 8 h. Afterwards
the reaction mixture was washed with a 10% Na2S2O3 solu-
tion, extracted with Et2O and washed with a saturated NaCl
solution. The organic layer was dried with MgSO4, the sol-
vent and iodobenzene were removed under reduced pres-
sure and finally, the crude product was purified with column
chromatography (SiO2; eluent: petroleum ether), affording a
colorless oil.
Synthesis of P1. The used reagents were 4a (1.00 mmol;
373 mg), i-PrMgClÁLiCl (1.10 M in THF; 1.00 mmol; 910 mL),
and Ni(dppp)Cl2 (25.0 mmol, 13.6 mg). Yield CHCl3 fraction:
126 mg (76%).
Synthesis of P2. The used reagents were 4b (1.00 mmol;
387 mg), i-PrMgClÁLiCl (1.10 M in THF; 1.00 mmol; 910 mL),
and Ni(dppp)Cl2 (25.0 mmol, 13.6 mg). Yield CHCl3 fraction:
128 mg (71%).
Synthesis of P3. The used reagents were 4c (675 mmol; 271
mg), i-PrMgClÁLiCl (1.10 M in THF; 675 mmol; 615 mL), and
Ni(dppp)Cl2 (16.9 mmol, 9.10 mg). Yield CHCl3 fraction: 90.5
mg (69%).
2-Bromo-3-(5-methylhexyl)25-iodothiophene (4b). The
used reagents were 3b (2.00 mmol; 522 mg), iodobenzene
diacetate (1.00 mmol; 322 mg), and I2 (1.00 mmol; 254 mg).
Yield: 599 mg (77%). 1H NMR (CDCl3): d 5 6.96 (s, 1H, Ar
H), 2.52 (t, 2H, CH2), 1.54–1.52 (m, 3H, CH2,CH), 1.31 (m,
2H, CH2), 1.19 (m, 2H, CH2), 0.86 (d, 6H, CH3). 13C NMR
(CDCl3): d 5 144.4, 138.1, 111.8, 71.2, 38.8, 30.1, 29.4, 28.0,
27.1, 22.8. MS: 388 (M1) (calc. 387.1).
Synthesis of P4. The used reagents were 4d (716 mmol; 297
mg), i-PrMgClÁLiCl (1.10 M in THF; 716 mmol; 651 mL), and
Ni(dppp)Cl2 (17.9 mmol, 9.70 mg). Yield chlorobenzene frac-
tion: 118 mg (79%).
2-Bromo-3-(4-ethylhexyl)25-iodothiophene (4c). The used
reagents were 3c (1.25 mmol; 344 mg), iodobenzene diace-
tate (625 mmol; 201 mg), and I2 (625 mmol; 159 mg). Yield:
Synthesis of P5. The used reagents were 4e (1.20 mmol;
516 mg), i-PrMgClÁLiCl (1.10 M in THF; 1.20 mmol; 1.10
mL), and Ni(dppp)Cl2 (30.0 mmol, 16.3 mg). Yield CHCl3 frac-
tion: 216 mg (81%).
1
439 mg (88%). H NMR (CDCl3): d 5 6.97 (s, 1H, Ar H), 2.50
(m, 2H, CH2), 1.52 (m, 2H, CH2), 1.27 (m, 7H,CH2, CH), 0.83
(t, 6H, CH3). 13C NMR (CDCl3): d 5 144.5, 138.1, 111.8, 71.1,
40.3, 32.4, 29.7, 27.0, 25.5, 11.0. MS: 402 (M1) (calc. 401.1).
RESULTS AND DISCUSSION
Polymer Synthesis
2-Bromo-3-(3-propylhexyl)25-iodothiophene (4d). The
used reagents were 3d (2.00 mmol; 579 mg), iodobenzene
diacetate (1.00 mmol; 322 mg), and I2 (1.00 mmol; 254 mg).
Yield: 789 mg (95%). 1H NMR (CDCl3): d 5 6.96 (s, 1H, Ar
H), 2.50 (t, 2H, CH2), 1.49-1.43 (m, 2H, CH2), 1.30–1.26 (m,
9H, CH2, CH), 0.89 (t, 6H, CH3). 13C NMR (CDCl3): d 5144.7,
138.1, 111.6, 71.2, 36.8, 35.9, 33.6, 26.7, 19.9, 14.6. MS: 416
(MH1) (calc. 416.2).
The polymers were prepared by the Ni(dppp)-mediated poly-
merization (dppp 5 1,3-diphenylphosphinopropane) of P3ATs
to yield regio-regular polymers.27,33 The synthesis of the
monomers starts with a Kumada-coupling of the respective
alkyl Grignard reagents 1b–e, prepared in situ from the cor-
responding bromides, and 3-bromothiophene (Scheme 1). In
the next step, 3-alkylthiophene 2b–e is brominated with NBS
in the 2-position, followed by an iodination at the 5-position
with iodine and iodobenzene diacetate resulting in 4a–e. The
polymerization of P1–P5 starts with the in situ GRIM
metathesis of 4a–e with i-PrMgCl.LiCl as Grignard reagent,
followed by a polymerization with Ni(dppp)Cl2 (2.5 mol %)
as catalyst to yield P1–P5. The use of i-PrMgClÁLiCl and not
i-PrMgCl as Grignard reagent is motivated by the fact that
we found that the former resulted in a lower degree of poly-
merization and lower yields for the more sterically hindered
monomers (4d and 4e). Indeed, Wu et al.34 and Lamps
et al.35 found that LiCl increases the reactivity of the mono-
mers in such way that HH couplings, which are not formed
in the absence of LiCl for steric reasons, become possible
using Ni(dppp) as catalytic moiety. Applied on our mono-
mers, we speculated that the presence of LiCl would also
tackle a possible slower polymerization of alkylthiophenes
2-Bromo-3-(2-butylhexyl)25-iodothiophene (4e). The used
reagents were 3e (1.52 mmol; 460 mg), iodobenzene diace-
tate (0.76 mmol; 245 mg), and I2 (0.76 mmol; 193 mg). Yield:
1
436 mg (67%). H NMR (CDCl3): d 5 6.92 (s, 1H, Ar H), 2.66
(d, 2H, CH2), 1.60–1.54 (m, 1H, CH), 1.26 (m, 12H, CH2), 0.88
(t, 6H, CH3). 13C NMR (CDCl3): d 5 143.7, 138.6, 112,5, 71.1,
38.7, 34.0, 33.1, 28.9, 23.2, 14.3. MS: 430 (MH1) (calc.
430.2).
General Procedure for the Polymerization of the
Poly(3-alkylthiophene)s
To a solution of 2-bromo-3-alkyl-5-iodothiophenes (1 eq) in
dry THF (0.1 M) was added 1 eq of i-PrMgClÁLiCl at room
temperature and under an argon atmosphere. The solution
was further stirred for 1 h and subsequently added to a
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86
79

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 1 Molar mass and DSC data of P1–P5
a
P3ATs
Mn (kg/mol)
Da
DPb
Tm (ꢂC)
DH (J/g)
P1
P2
P3
P4
6.2
8.7
8.0
8.4
1.1
1.4
1.4
1.1
23
29
23
26
201
197
159
65
18.1
17.3
16.4
11.9
a
Determined by GPC in THF towards poly(styrene) standards.
Relative integration of the inner and terminal methylene 1H NMR
b
resonances.
“thermal treatment” and not for, for example, a conventional
annealing, since this former treatment better reflects the
actual history of the samples prior to the magnetic measure-
ments. This revealed that in all samples except P1 DH
increases after standing for 3 days and that P5 showed a
clear melting peak. If the Tm and DH are considered for all
polymers, it is clear that the more the side-chain is
branched, the lower Tm and DH become. This already sug-
gests that the p-interactions, a driving force for the crystalli-
zation, weaken as more bulky substituents are used. It is
also in line with the behavior of poly(3-(2-ethylhexyl)thio-
phene): the bulkiness of the side-chains does not necessarily
impede the crystallization, it only slows down its speed and
reduces Tm and DH.
SCHEME 1 Synthesis of the monomers and polymers P1–P5.
Conditions: (i) (1) Mg, (2) Ni(dppp)Cl2, 3-bromothiophene, (ii)
NBS, (iii) I2, PhI(OAc)2, (iv) (1) i-PrMgClÁLiCl, (2) Ni(dppp)Cl2.
with a bulky substituent, as in P5, which turned out to be
the case. After polymerization, the crude polymer was pre-
cipitated in methanol, isolated and further purified via Soxh-
let extractions with methanol and chloroform. The
chloroform fraction was concentrated, precipitated in metha-
nol and dried in vacuo.
The crystalline nature of all samples, including P5, was con-
firmed by X-ray diffraction. The measurements were per-
formed on samples that were stored at room temperature
for many days, which allows crystallization. Figure 2 shows
the X-ray intensity versus 2h traces for P1–P3 and P5.
During its synthesis, P4 showed an anomalous behavior as it
visually precipitated during the polymerization and only
(slightly) dissolved in (warm) chlorobenzene. The synthesis
of P4 was repeated twice, but the same result was obtained.
Due to its deviant behavior and its rather insolubility, P4
was not withdrawn for further study.
The (100) peak reflects the distance between the planes con-
taining the stacked polythiophene backbones. For P1 with
the least bulky side chains, this distance is found to
1
The structure of P1–P3, P5 was confirmed by H NMR (Sup-
porting Information Fig. S25–S29) and the degree of poly-
merization was also calculated by relative integration of the
inner and terminal methylene protons (Table 1). The number
averaged molar mass (Mn ) and polydispersity (D) of the
polymers P1–P5 were also determined by GPC in THF
toward polystyrene standards (Table 1).
DSC and XRD Analysis
Next, the polymers were subjected to DSC analysis. Typically
the second heating scan was used. Initially, no melting was
observed for P5, pointing at the fact that P5 is amorphous
after a relative fast cooling (20 K/min). However, Boudouris
et al.36 have shown for HT-coupled poly(3-(2-ethylhexyl)thio-
phene), which also has a relative bulky, branched side-chain,
that the polymer is actually not amorphous, but that it
slowly crystallizes. Since the magnetic measurements are
performed on powders which were allowed to stand at room
temperature for a long time, the DSC experiments were
repeated after three days of standing. We opted for this
FIGURE 2 XRD spectra of P1–P3, P5. The curves are vertically
shifted for clarity.
80
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
be 16.07 6 0.02 Å. It is clear from Figure 2 that the distance
along the (100) direction between the stacked polythiophene
backbone planes changes when the side chains become
more bulky in P2, P3, and P5. For P2, the inter-backbone
distance increases to 18.05 6 0.02 Å. For P3 and P5, we
notice however an opposite trend with the distance reducing
to 16.82 6 0.02 Å and 15.35 6 0.02 Å, respectively. This
strongly suggests that there is a transition from a quasi-
orthorhombic unit cell to a monoclinic (or triclinic) one, in
which the thiophene monomers of neighboring backbones
are slightly shifted with respect to each other instead of
orthogonally opposite. Such a tilted unit cell would indeed
create more space to accommodate for the bulky side chains
even if it reduces the distance between neighboring planes
of stacked backbones. Based on the shifted (100) reflections,
we can obtain a lower limit for the monoclinic angle of the
tilted unit cell (i.e., the angle between the plane of the
stacked polymer chains and the plane containing the side
chains). Herefore, we reasonably assume that the distance
between the polythiophene backbones in the plane contain-
ing the side chains will not decrease with increasing bulki-
ness of the side chains in P3 and P5. For P5, we find a
lower limit for the deviation/tilt of 17.2 6 0.4ꢂ with respect
to the 90ꢂ orthorhombic angle.
This is in line with the findings of Boudouris et al.,36 who
demonstrated a large discrepancy of 90ꢂ and a triclinic unit
cell for the branched poly(3-(2-ethylhexyl)thiophene). As a
consequence, it was unfortunately not possible to determine
all unit cell parameters using XRD.
Optical Characterization
The optical behavior of P1–P3 and P5 was investigated both
in good solvent, upon addition of non-solvent and in film. In
good solvent, such as chloroform, the kmax of all polymers is
more or less equal; only P5, which has the most bulky side
chain, shows a slight blue shift in chloroform, meaning that
the increased bulkiness of the substituent of P5 results in a
more twisted conformation around the consecutive thio-
phene moieties [Fig. 3(A)]. Next, UV–vis spectra of P1–P3
and P5, with approximately the same concentration, were
recorded in different ratios of CHCl3/MeOH (Supporting
Information Fig. S30). Upon increasing the amount of metha-
nol, all polymers except P5 aggregate, as is evidenced by a
large red-shift and formation of fine structure. P5 shows
only a minor red-shift, which demonstrates that in this con-
dition, it planarizes to some extent, but does not stack.
Clearly, the stacking in P5 is prohibited by the bulky side
chains. The difference between P1–P3 is that the aggregation
starts at higher non-solvent concentration for the polymers
with more bulky side chains and that there is also a
decrease in fine structure in aggregated state upon increas-
ing the bulkiness of the side chains [Fig. 3(B)]. As mentioned
previously,25,26 the absorption spectrum of aggregated P3AT
consists of three chromophores. One band [Fig. 3(C)] is
related to the polymer in its disordered, random coil like
conformation (around 450 nm). The second transition comes
from the individual, planar, and stacked polymer chains
FIGURE 3 (A) UV–vis spectra of P1–P3, P5 in chloroform; cP1
5
1.69 3 1024 M; cP2 5 1.71 3 1024 M; cP3 5 1.73 3 1024 M; cP4
5 1.55 3 1024 M; cP5 5 1.82 3 1024 M; (B) UV–vis spectra of
P1–P5 in non-solvent; P1–P3, P5 CHCl3/MeOH (10/90); cP1
5
1.69 3 1024 M; cP2 5 1.71 3 1024 M; cP3 5 1.73 3 1024 M; cP5
5 1.82 3 1024 M. (C) Deconvolution of the spectrum of P1 in
CHCl3/MeOH (10/90).
(around 530 nm, with vibronic fine-structure). The last tran-
sition arises from a transition dipole moment, delocalized
over multiple, aggregated polymer chains (around 610 nm
for P3AT). The absorption spectra of P5 essentially consist
of only one band of coiled polymer chains. Deconvolution of
the UV–vis spectra in poor solvent using a linear program-
ming method [Fig. 3(C) for P1 and Supporting Information
Fig. S31 for P2, P3, and P5) reveals the different bands dis-
cussed above: band 2 corresponds to the polymer in its ran-
dom coil like conformation, band 3 and 4 to the individual,
planar, stacked polymer chains with exciton coupling and
band 5 to a transition dipole moment, which is delocalized
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86
81

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
TABLE 2 Relative intensity of the different “chromophores” of
4(B), it is clear that P5 does crystallize, which is in line with
the DSC results but different from the situation in poor solvent
and that the crystallization in thin films takes 30–45 min.
the polymers in the aggregated state
In poor solvent
Band 3 1 4% Band 5%
In film
Band 3 1 4%
Next, the spectra of the polymers were deconvoluted (Sup-
porting Information Fig. S34). The relative ratio of band
3 1 4, pointing at planar polymer chains and of band 5, indi-
cating p-stacking, is shown in Table 2. First, it is clear that
p-interactions are present in all polymers, also in P5, but
that increasing the bulkiness results in smaller p-
interactions. Second, the highest fraction of planar polymer
chains is also in film found for P3, followed by P1.
P3ATs
Band 5%
P1
P2
P3
P5
61.9
61.7
64.2
30.4
4.4
3.8
2.6
0
72.5
63.5
77.4
54.6
5.4
4.4
2.6
1.4
over multiple polymer chains and which therefore depends
on the p-interactions. The relative contribution of planar
polymer chains (sum of bands 3 1 4) and of the band due
to p-interactions (band 5) is shown in Table 2. Looking at
the low energy band (band 5), it is clear that upon increas-
ing the bulkiness of the side chain, a decrease in intensity
of the low energy band is observed, pointing at smaller p-
interactions. The amount of planar stacked chains in P5 is
considerably less than this for the other polymers, which
indicates that in P5, the planarization of the polymer
chains is more difficult due to its bulky side chains.
Surprisingly, the amount of planar polymer chains is the
highest in P3.
ESR Spectroscopy
The magnetic behavior of the polymers was first studied by
X-band ESR spectroscopy at 300 K (Fig. 5 for P1 and in Sup-
porting Information Fig. S35 for P2–P5). Prior to measure-
ment, the powder samples were treated with an alcoholic
hydrazine solution and dried under argon to avoid polaron
formation by oxidation in ambient air. As illustrated in Fig-
ure 5, only one signal, of symmetric shape, is observed with
g value in the range 2.0030–2.0045. The (inferred) spin den-
sities, g-values, and line widths and shapes are listed in
Table 3. Depending on the degree of supramolecular order,
the line shape of an ESR signal varies from Gaussian to
Also fluorescence spectroscopy was carried out in good
(CHCl3) as well as in poor solvent CHCl3/MeOH [1/9]). P1–
P3 were excited at 440 nm and P5 was excited at 425 nm.
From Supporting Information Figure S32, it is clear that
always the same emission is found in good solvent. Thus, the
emitting chromophore is essentially the same for P1–P5,
which means that all the polymers adopt the same random
coil like structure in good solvent. In non-solvent, in contrast
(Supporting Information Fig. S33), the bulkiness of the side
chain clearly influences the quenching of the emission. Upon
increasing the bulkiness, going from P1 to P3, which
increases the distance between the polymer chains, aggrega-
tion becomes less pronounced and the quenching of the fluo-
rescence is less efficient. This is expressed in an increase in
intensity of the signal, which originates from residual coiled
polymer chains. P5, which does not aggregate in non-
solvent, has a kem at higher wavelength, which indicates that
another, more planar chromophore is present in non-solvent
that weakly emits at higher wavelength.
Next, the experiments were repeated on spin coated films of
P1–P3 and P5. First, it was verified whether the polymers are
not trapped in a (partly) amorphous structure, as was the case
when the polymers were melted and cooled down relatively
fast (see “DSC and XRD analysis” section). Annealing did not
affect the spectra of P1–P3; the UV–vis spectrum of P5, how-
ever, changed dramatically. Clearly, melting the polymer and
subsequently cooling it down traps the polymer in its amor-
phous, coiled configuration, as shown by its UV–vis spectrum
and which is fully in line with the DSC measurements. Next, we
monitored the crystallization of the polymer at room tempera-
ture by recording UV–vis spectra at different times. From Figure
FIGURE 4 (A) UV–vis spectra (normalized) of the polymers
after annealing; (B) UV–vis spectra of P5 in function of time
after melting and fast cooling.
82
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
FIGURE 5 X-band (ꢀ9.2 GHz) first-derivative absorption spec-
trum measured on P1 at room temperature using Pm 5 0.5 mW
and Bm 5 0.35 G.
FIGURE 6 M(H) data of P1. Top: as measured. Bottom: after
subtraction of the diamagnetic component.
Lorentzian. A signal with a Lorentzian line shape is generally
an indication of homogeneous broadening as met when the
ESR-active centers are embedded in an environment of high
degree of order. This conclusion is supported by the smaller
line width of the signal from P1. An increase in line width
(DBpp) accompanied by a change in line shape from Lorent-
zian over Voigt to, finally, Gaussian points to gradually more
disorder being introduced into the system (e.g., g-spread
broadening). As such, the line shape provides crucial infor-
mation on the order present in the environment of the elec-
tron spins. The ESR data thus clearly confirm that increasing
the bulkiness of the side chain leads to a decrease in the
supramolecular order of the polymer samples, which is in
perfect agreement with the results obtained from the UV–vis
and DSC measurements.
that an additional spin system is observed in sample P3,
showing an angle dependent ESR signal of Lorentzian line
shape at g 5 2.44 (Supporting Information Fig. S36). But
recalling that the ESR experiments are carried out on pow-
der samples, the observed angle dependence of the signal is
to be ascribed to the overall shape of the ESR sample being
not a perfect cylinder (demagnetization effects related to dis-
tinct magnetic properties).
In summary, the ESR results confirm the supramolecular
behavior observed by UV–vis spectroscopy and DSC. The
data are also in line with our previous results obtained on
polythiophenes.
Looking at the number of spin centers for P1–P5, it is clear
that for all the samples the densities have the same order of
magnitude in agreement with previously obtained results for
P3ATs in the neutral state.25 The g values for P2–P5 are
equal within experimental accuracies, only for P1 a small
increase in g value is observed compared to the other poly-
mers studied here. It may indicate the signal to originate
from a spin system that is more localized than for the other
samples. For reasons of completeness, it should be added
SQUID Magnetometry
The magnetic properties of the polymers were next meas-
ured by SQUID magnetometry on the same powders as for
the ESR measurements. By comparing the magnetization
measurements performed with and without the polymer
sample (the latter corresponding to the contribution of the
sample holder, i.e., the plastic tube and the cotton), six differ-
ent components can be identified: diamagnetic-, paramag-
netic-, and ferromagnetic-like components originating from
the sample holder; diamagnetic-, paramagnetic-, and
ferromagnetic-like components of the polymer sample. The
two diamagnetic components correspond to the expected
diamagnetic behavior of the materials in question. Although
these diamagnetic components dominate the measured mag-
netization, since they are (to a good approximation) linear in
applied magnetic field and temperature-independent, they
can be easily separated from the paramagnetic and
ferromagnetic-like components. Figure 6 shows the magnet-
ization of sample P1 before and after subtraction of the dia-
magnetic component. The diamagnetism-corrected data is
TABLE 3 ESR spin densities (S 5 1/2), g-value, and line widths
(DBpp) observed on P1–P3, P5 P3AT materials
Spin density
g value
(1 3 1024
P3ATs
(31016 g21
)
)
DBpp (G)
Line shape
P1
P2
P3
P5
4.0 6 0.1
8.0 6 0.1
1.5 6 0.1
1.3 6 0.1
2.0045
2.0031
2.0032
2.0030
4.8 6 0.1
5.9 6 0.4
6.6 6 0.2
10.4 6 0.2
Lorentzian
Voigt
Voigt
Gaussian
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86
83

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
coercivity values are comparable to those of the cotton sam-
ple holder, and therefore cannot be reliably determined.
Before analyzing, in detail, the observed magnetic properties,
it is important to note that the ferromagnetic moments
detected in these polymers is very small (of the order 3 3
1024 emu/g, i.e., 3 3 1026 emu for a typical 10 mg sample).
This is, however, a regime where magnetic contamination
may play an important role. A moment of 3 3 1026 emu
corresponds, for example, to about 0.02 mg (i.e., about 2
ppm) of metallic Fe. This means that the observed magnetic
properties may originate, at least partially, from accidental
sample contamination with a few dust particles (typically
containing ferromagnetic Fe-containing compounds), for
example, during sample manipulation (cf. related discussion
in Ref. 29). Staying below such contamination levels is
extremely challenging, requiring that all sample preparation
and characterization is carried out under clean-room condi-
tions. Sample characterization with respect to such contami-
nants is also extremely challenging for several reasons: the
contaminant material is not necessarily uniformly distributed
in the sample, making any quantitative assessment very diffi-
cult; the number of contaminating particles may easily
increase or decrease between different characterization
steps, making it almost impossible to directly compare quan-
titative data from different techniques (e.g., magnetization vs.
amount of contaminant); even if the amount of potentially
magnetic elements (e.g., Fe, Co, etc.) could be precisely and
reproducibly determined, it is nearly impossible to deter-
mine what fraction of these elements is in a ferromagnetic
phase (e.g., ferromagnetic vs. non-ferromagnetic compounds).
Therefore, although the trends observed here (discussed
below) and in our previous work support an intrinsic origin,
we cannot completely exclude that the observed ferromag-
netism originates from magnetic contamination. In the fol-
lowing discussion, we assume that the observed
ferromagnetism is an intrinsic property of the synthesized
polymers.
FIGURE 7 M(H) data of P1, compared to the background from
the sample holder (mostly due to the cotton enclosing the
polymer).
given by M 2 vH, where H is the magnetic field and v is the
diamagnetic susceptibility DM/DH estimated at 300 K in the
high field (4–5 T) region. The paramagnetic component
observed in the sample holder can be ascribed to paramag-
netic contaminants present in the cotton (e.g., paramagnetic
metallic compounds or dilute transition-metal ions). Simi-
larly, the ferromagnetic-like component observed in the sam-
ple holder can be ascribed to ferromagnetic contaminants in
the cotton (e.g., ferromagnetic Fe-containing compounds,
such as metallic Fe and Fe oxides). The magnetization (para-
magnetic- and ferromagnetic-like components) originating
from the polymer samples can be inferred from comparing
the diamagnetism-corrected magnetization measured with
and without the polymer sample—direct subtraction has not
been performed since it introduces significant artifacts in the
magnetization curve, especially at the low field (low magnet-
ization) region of interest. The paramagnetic component of
the polymer samples is associated with the spin centers
identified in the ESR measurements. Figure 7 shows such a
comparison for sample P1. In fact, the paramagnetic compo-
nent is only visible at 5 K. At 300 K, the (Brillouin-like) mag-
netization is, to a good approximation, linear in applied
magnetic field (and orders of magnitude smaller compared
to that at 5 K), and is therefore also subtracted when
subtracting the diamagnetic component. Consequently, the
corrected magnetization, at 300 K, corresponds to the
ferromagnetic-like component. At 5 K, both the paramag-
netic- and ferromagnetic-like components are visible,
although the former dominates the magnetization. In general,
at room temperature, all polymers show ferromagnetic-like
behavior with negligible coercivity. At 5 K, only P1 shows a
coercivity value (350 Oe) which is significantly higher than
that of cotton (150 Oe). For the other polymers, the 5 K
As already indicated in the introduction, we hypothesize
that the saturation magnetization correlates with the num-
ber of planar polymer chains, while the coercivity is influ-
enced by p-interactions. Both entities—the number of
planar polymer chains and the p-interactions—can be quan-
tified from the UV–vis spectra. The results of the films that
were kept for 1 h at room temperature were used for this
evaluation, since this resembles best the powders used for
the SQUID measurements. If the coercivity is considered,
only P1 shows a value (350 Oe) that is significant higher
than that of cotton (150 Oe). Because P1 also shows the
highest integration of the band related with p-interactions,
this finding is in line with our hypothesis. Unfortunately,
since the low temperature coercivity of the other polymers
could not be determined with sufficient accuracy because it
is of the same order as the small coercivity of the cotton
sample, the other polymers cannot be used to further vali-
date our hypothesis. This requires polymers with stronger
p-interactions than P1, rather than weaker. Further
84
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86

JOURNAL OF
POLYMER SCIENCE
WWW.POLYMERCHEMISTRY.ORG
ARTICLE
ACKNOWLEDGMENTS
We are grateful to the Onderzoeksfonds KU Leuven/
Research Fund KU Leuven, the Flemish Concerted Action
(GOA), the Belgian Interuniversity Poles of Attraction (IAP,
FWO-Vlaanderen) and the Air Force Office of Scientific
Research.
REFERENCES AND NOTES
1 A. Skotheim, J. Reynolds, Eds.; Handbook of Conducting
Polymers, 3rd ed., CRC Press: Boca Raton, FL, 2007.
2 J. L. Bredas, G.B. Street, Acc. Chem. Res. 1985, 18, 309–315.
3 A. J. Fisher, W. Hayes, D. S. Wallace, J Phys: Condens Matter
1989, 1, 5567–5593.
4 T. Ikoma, S. Okada, H. Nakanishi, K. Akiyama, S. Tero-
FIGURE 8 Plot of the saturation magnetization at room temper-
ature and the integration of the bands related to planar poly-
mer chains [band 3 and 4, Fig. 2(C)]. The curve is added as a
guide for the eye.
€
Kubota, K. Mobius, S. Weber, Phys. Rev. B. 2002, 66, 014423.
5 A. A. Ovchinnikov, Theor Chim Acta 1978, 47, 297–304.
6 W. T. Borden, E. R. Davidson, J. Am. Chem. Soc. 1977, 99,
4587–4594.
€
7 G. Kothe, K.-H. Denkel, W. Summermann, Angew. Chem. Int.
Ed. 1970, 9, 906–907.
research in this respect is underway. In order to evaluate
the influence of the fraction of planar polymer chains on
the saturation magnetization, the relative integration of the
absorption bands 3 and 4 was plotted versus the saturation
magnetization (Fig. 8). Interestingly, the surprisingly higher
value for the integration of absorption bands 3 1 4 of P3
compared to P1 and P2 is also reflected in the magnetiza-
tion data. In general, it is observed that samples with larger
fractions of planarized chains also reveal larger values of
their saturation magnetization. Therefore, the SQUID results
are in line with our hypotheses on the magnetic properties
of neutral conjugated polymers.
8 A. Rajca, S. Rajca, J Chem Soc: Perkin Trans 1998, 2, 1077–
1082.
9 H. Tomioka, M. Hattori, K. Hirai, K. Sato, D. Shiomi, T. Takui,
K. Itoh, J. Am. Chem. Soc. 1998, 120, 1106–1107.
10 H. Murata, D. Miyajima, H. Nishide, Macromolecules 2006,
39, 6331–6335.
11 H. Nishide, T. Maeda, K. Oyaizu, E. Tsuchida, J. Org. Chem.
1999, 64, 7129–7134.
12 T. Kaneko, T. Makino, H. Miyaji, M. Teraguchi, T. Aoki, M.
Miyasaka, H. Nishide, J. Am. Chem. Soc. 2003, 125, 3554–3557.
13 A. Rajca, S. Rajca, J. Wongsriratanakul, J. Am. Chem. Soc.
1999, 121, 6308–6309.
14 P. K. Kahol, A. Raghunathan, B. J. McCormick, Synth. Met.
2004, 140, 261–267.
CONCLUSIONS
15 P. Dallas, D. Stamopoulos, N. Boukos, V. Tzitzios, D.
Niarchos, D. Petridis, Polymer 2007, 48, 3162–3169.
In conclusion, a series of regio-regular P3ATs with an increas-
ing degree of bulkiness of the alkyl side chain has been suc-
cessfully synthesized with Ni(dppp)Cl2 as a catalyst. UV–Vis
spectroscopy was used to study the supramolecular aggrega-
tion of the polymer both in poor solvent and in film. In partic-
ular, deconvolution of the spectra allowed to “quantify” the p-
interactions and planarization of the polymer chains. It was
shown that as the bulkiness of the polymers’ side chain
increases, the p-interactions and planarization diminishes. In
poor solvent, the polymer with the most bulky substituent
does not aggregate, while in film, this polymer does stack,
showing a band originating from p-interactions, but the aggre-
gate formation takes 30–45 min to occur. This delayed aggre-
gation was confirmed by DSC measurements; XRD
measurements confirmed that powders of all polymers are
semi-crystalline. ESR measurements, performed on powder
samples at 300 K, revealed a steady increase in line width
and change in shape of the singly observed signal as the poly-
mers’ substituent becomes more bulky, which denotes an
increase in disorder. Thus, the ESR data reflect the same trend
as previously found from the optical data. The analysis of the
magnetometry data, measured at 5 and 300 K, are in line
with our previously established hypotheses.
16 N. A. Zaidi, S. R. Giblin, I. Terry, A.P. Monkman, Polymer
2004, 45, 5683–5689.
17 Y. Long, Z. Chen, J. Shen, Z. Zhang, L. Zhang, H. Xiao, M.
Wan, J. L. Duvail, J. Phys. Chem. B 2006, 110, 23228–23233.
18 K. Mizoguchi, N. Kachi, H. Sakamoto, K. Kume, K. Yoshioka,
S. Masubuchi, S. Kazama, Synth. Met. 1997, 84, 695–698.
19 A. Konkin, H.-K. Roth, P. Scharff, A. Aganov, O. Ambacher,
S. Sensfuss, Solid State Commun 2009, 149, 893–897.
20 H. Nalwa, Phys. Rev. B 1989, 39, 5964–5974.
21 K. Sugiyama, T. Kojima, H. Fukuda, H. Yashiro, T. Matsuura,
Y. Shimoyama, Thin Solid Films 2008, 516, 2691–2694.
22 F. R. de Paula, L. Walmsley, E. C. Pereira, A. J. A. de
Oliveira, J Magn Magn Mater 2008, 320, e193–e195.
23 O. R. Nascimento, A. J. A. de Oliveira, E. C. Pereira, A. A.
Correa, L. Walmsley, J Phys: Condens Matter 2008, 20, 035214.
24 M. Nechtschein, F. Devreux, F. Genoud, M. Guglielmi, K.
Holczer, Phys. Rev. B 1983, 27, 61–78.
25 S. Vandeleene, M. Jivanescu, A. Stesmans, J. Cuppens,
M. J. Van Bael, H. Yamada, N. Sato, T. Verbiest, G.
Koeckelberghs, Macromolecules 2010, 43, 2910–2915.
26 S. Vandeleene, M. Jivanescu, A. Stesmans, J. Cuppens,
M. J. Van Bael, T. Verbiest, G. Koeckelberghs, Macromolecules
2011, 44, 4911–4919.
WWW.MATERIALSVIEWS.COM
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86
85

JOURNAL OF
POLYMER SCIENCE
ARTICLE
WWW.POLYMERCHEMISTRY.ORG
27 E. E. Sheina, J. Liu, M. C. Iovu, D. W. Laird, R. D.
Mccullough, Macromolecules 2004, 37, 3526–3528.
32 B. J. Meakin, F. R. Mumford, E.R. Ward, J Pharm Pharmacol
1959, 11, 540–547.
28 A. Yokoyama, R. Miyakoshi, T. Yokozawa, Macromolecules
2004, 37, 1169–1171.
33 R. Miyakoshi, A. Yokoyama, T. Yokozawa, J. Am. Chem.
Soc. 2005, 127, 17542–17547.
ꢀ
29 L. M. C. Pereira, J. P. Araujo, M. J. Van Bael, K. Temst, A.
34 S. Wu, L. Huang, H. Tian, Y. Geng, F. Wang, Macromole-
cules 2011, 44, 7558–7567.
Vantomme, J Phys D: Appl Phys 2011, 44, 215001.
30 F. S. Prout, J. Cason, J. Org. Chem. 1949, 14, 132–136.
35 J.-P. Lamps, J.-M. Catala, Macromolecules 2011, 44, 7962–7968.
^
31 J. Malthete, J. Canceill, J. Gabard, J. Jacques, Tetrahedron
36 B. W. Boudouris, V. Ho, L. H. Jimison, M. F. Toney, A.
Salleo, R. A. Segalman, Macromolecules 2011, 44, 6653–6658.
1981, 37, 2823–2828.
86
JOURNAL OF POLYMER SCIENCE, PART A: POLYMER CHEMISTRY 2014, 52, 76–86