Introduction - UPCommons

vainclamInternet και Εφαρμογές Web

14 Δεκ 2013 (πριν από 3 χρόνια και 9 μήνες)

98 εμφανίσεις


1

E
lectronic properties of

poly(thiophene
-
3
-
methyl acetate)


Alex L. Gomes,
1,2

Jordi Casanovas,
3

Oscar Bertran,
4

João Sinézio
de C. Campos,
2

Elaine Armelin,
1,5,*

and Carlos Alemán
1,5,*

1

Departament d’Enginyeria Química, E.T.S d’Enginyers Industrials de Bar
celona,
Uni
v
ersitat Politècnica de Catalunya, Diagonal 647, Barcelona E
-
08028, Spain

2

Departamento de Tecnologia de Polímeros, Faculdade de Engenharia Química,
Uni
v
ersidade Estadual de

Campinas, Avenida Albert Einstein nº 500, Barão Geraldo, CEP
13083
-
852
, Campinas
-
SP, Brazil

3

Departament de Química,

Escola Politècnica Superior, Uni
v
ersitat de Lleida, c/Jaume II
n
°
69, Lleida E
-
25001, Spain

4

Departament de Física Aplicada, E.U.E.T.I.I., Uni
v
ersitat Politècnica de Catalunya, Pça
Rei 15, Igualada 08700, Spa
in

5

Center for Research in Nano
-
Engineering, Uni
v
ersitat

Politècnica de Catalunya, Campus
Sud, Edifici C

, C/Pasqual i Vila s/n, Barcelona E
-
08028, Spain


*

elaine.armelin@upc.edu

and
carlos.aleman@upc.edu



2

Abstract

The electronic structure of poly(thiophene
-
3
-
methyl ac
etate) has been investigated
using UV
-
vis absorption spectroscopy and quantum mechanical calculations. Experimental
measures in chloroform solution indicate that the

-
conjugation length increases with the
polymer concentration, which is reflected by the r
ed shift of the absorbance peak of the


-

*

transition. On the other hand, the energy required for the

-

*

transition has been
found to decrease with the volatility of the solvent for concentrated polymer solutions, even
though the influence of the solve
nt is very small for dilute solutions. Quantum mechanical
calculations indicate that the interactions between the

-
conjugated backbone and the
methyl acetate side groups are very weak. On the other hand, the lowest energy transition
predicted for an infin
ite polymer chain that adopts the
anti
-
gauche

and all
-
anti

conformations is 2.8 and 1.9 eV, respectively. Finally, measurements on spin
-
casted
nanofilms reflect that the

-

*

transition energy increases with the thickness, which has
been attributed to the
distortion of the molecular conformation. In spite of this, the energy
gap obtained for the thinnest film (1.52 eV) is significantly smaller than that determined for
dilute and concentrated chloroform solutions (2.56 and 2.09 eV, respectively).


Keywords:

Conducting polymers; Polythiophene; UV
-
vis spectroscopy; Quantum
chemistry; Conjugation


3

Introduction

Organic semiconducting polymers have important applications in microelectronics,
electrochemical switching, photovoltaics, light
-
emitting diodes, field e
ffect transistors,
electrochromic and electromechanical devices, chemical and physical sensing, etc

[1
-
2
]
.
Among these materials, polythiophene (PTh) derivatives are the most commonly used due
to their high charge carrier mobility. In particular, polymers
bearing substituents into the 3
-
position of the thiophene are especially important because in some cases they overcome the
solubility and processability problems intrinsically associated to PTh. Thus, the
incorporation of long alkyl side chains increases t
he solubility in organic solvents

[
3
-
5
]
,
while hydrophilic substituents produce PTh derivatives soluble in water or polar solvents

[
6
]
.

PTh derivatives bearing carboxylic acid groups have been found to be soluble in a wide
variety of organic solvents

[
6
-
1
3
]
. Within this field, the work of Heeger and co
-
workers on
poly(thiophenesulfonate)s was a pioneering contribution to water
-
soluble semiconducting
polymers

[
6
]
. At the end of the nineties, electroactive polymers and copolymers with acetic
acid, propinion
ic acid and octanoic acid linked to the thiophene ring were also investigated

[
8,9
]
. More recently, we reported the synthesis, structure and electronic characterization of
PTh derivatives bearing malonic acid

[
10,11
]

and acrylic acid

[
12,13
]

as substituent
s.
Among all these functionalized, soluble and electroactive PTh derivatives, poly(thiophene
-
3
-
acetic acid), or PT3AA (Scheme 1), has attracted much attention because of its versatility
to prepare layer
-
by
-
layer (LBL) self
-
assembled systems. Specifically,
in the last few years
multilayered LBL thin films obtained by combining PT3AA with organic polycations [
e.g.

poly(sodium
-
2
-
(4
-
methyl
-
3
-
thiehyloxy)ethane sulfonate) and chitosan]

[1
4,15
]
, small

4

surfactants (
e.g.

dialkyldimethylammonium)

[1
6
]
, metallic and i
norganic particles (
e.g.

gold, TiO
2

and zeolites)

[
17,18
]
, other conducting polymers (
e.g.

polyaniline)

[1
9
]
,
conventional organic polymers [
e.g.

poly(4
-
vinylpyridine) and polyimide]

[
20
]

or even
peptides

[
21
]

have been reported.


Scheme 1

PT3AA presents
high surface resistivity as compared with other PTh derivatives. For
example, the surface resistivity reported by Zhang and Srinivasan for PT3AA
-
polyimide
composites ranged from 10
6

to 10
8


∙sq
-
1

depending on the preparation of the sample

[
22
]
.
Similarly,
the surface resistivity of the undoped forms of both PT3AA and poly(3
-
thiophene
-
acetic acid
-
co
-
3
-
hexylthiophene) was found to be

7∙10
10


∙sq
-
1

[
23
].

After
doping with iodine, the surface resistivity of the latter copolymer decreases to

10
5


∙sq
-
1
,
while
the homopolymer does not exhibit any change remaining almost identical to the value
measured before doping

[
23
]
. On the other hand, the electrical conductivity reported for
PT3AA
-
poly(ethyleneoxide) blends was also very low,
i.e.

around 10
-
6

S∙cm
-
1

[
24
]
.

T
he
high electrical resistivity of PT3AA was attributed to the presence of carboxylic side
groups close to the

-
conjugated backbone

[
23
]
. Thus, this polar substituent was proposed
to affect the electron transport through the

-
system of the polymer. Moreov
er, the poor
doping behavior of PT3AA was explained by the dense packing of the chains with
-

5

CH
2
COOH pendant groups that interact through hydrogen bonds, precluding the diffusion
of the dopant molecules through the polymeric matrix

[
23
]
. Furthermore, the e
lectronic
structure of PT3AA was examined by studying the pH and temperature dependence of the
UV
-
Vis absorption spectrum

[
8
]
. An abrupt reversible increase of

max

was found when the
pH varies from 5 to 6, which was attributed to a variation in the effect
ive electronic length.
Thus, the electrostatic repulsions between the dissociated carboxylic acid side groups were
suggested to induce conformational changes in the polymer chains from the aggregated
state, which is stabilized by hydrogen bonds, to the ext
ended state, giving rise to an
increased effective conjugation length

[
25
]
.

On the other hand, the

max

after the pH
transition (pH= 6.3) decreased with the increase in temperature, whereas that before the
transition (pH= 5.2) increases slightly with incre
asing temperature. These thermal effects
suggested that at the former pH the extended polymer chains evolve towards a disordered
state as the temperature increases

[
8
]
. In opposition, at pH= 5.2 the intramolecular
hydrogen bonds break upon the increase of
temperature, PT3AA chains becoming more
conjugated.

In this work we examine the electronic structure of a PT3AA derivative in which both
the ability to form hydrogen bonds and the pH dependence have been eliminated. More
specifically, the electronic proper
ties of poly(thiophene
-
3
-
methyl acetate), or PT3MA
(Scheme 2), have been investigated in detail by combining both absorption spectroscopy
and quantum mechanical calculations. Specifically, experimental investigations were
performed considering dilute and c
oncentrated PT3MA solutions in different organic
solvents as well as nanofilms prepared by spin
-
coating, while theoretical studies were

6

carried out using Density Functional Theory (DFT) calculations in both gas
-
phase and
solution environments.



Methods

M
aterials and Polymerization.

3
-
Thiophene acetic acid (T3AA) and anhydrous ferric
chloride (FeCl
3
) were purchased from Sigma
-
Aldrich Química S.A. and were employed
without further purification. All solvents were purchased from Panreac Química S.A. with
ACS
grade.

3
-
Thiophene methyl acetate (T3MA) was prepared and subsequently polymerized by
oxidative coupling following the procedure described by Kim
et al.

[
8
],

which is
summarized in Scheme 2. The purified T3MA monomer was obtained with 82% of yield,
while
the yield of PT3MA after remove the residual oxidant and oligomers was 25%.
Unfortunately, changes in the monomer:oxidant (T3MA:FeCl
3
) ratio did not improve the
low yield of the oxidative coupling polymerization. The characterization of PT3MA using
FTIR an
d
1
H
-
NMR was previously reported

[
8
]
, our results being in complete agreement
with them. Table 1 describes the solubility of the resulting PT3MA in different organic and
polar solvents at room (25 ºC) and high temperatures (60 ºC).



7


Scheme 2


Spin
-
Casting.

PT3MA solutions were prepared by dissolving the polymer in
chloroform, concentrations ranging from 0.01 mg/mL to 5.00 mg/mL being considered.
Nanofilms were obtained by spin
-
coating (WS
-
400B
-
NPP Laurell Technology Co.) abou
t
18
-
20 drops of solution deposited on indium
-
tin oxide (ITO) glass slides, previously
cleaned with ethanol, at 8000 rpm for 60 s with the acceleration set to be of 15 s. It is well
known that the film thickness depends on the concentration and the speed o
f spinning.
Previous studies on other polymeric systems reported film thickness ranging from 10 to 80
nm for concentrations similar to those used in this work

[2
6,27
]
.
Due to the poor
mechanical properties of P3TMA, the thickness of the nanofilms was estim
ated using a
calibration curve for 50:50 mixtures of this PTh derivative and poly(tetramethylene
succinate), the latter being added to impart mechanical consistency. The thickness of the
nanofilms obtained using
spinning
speeds ranging from 1500 to 12000 r
pm for 60 s were
determined by scratch AFM.

According to the curve thickness
versus

speed, the thickness
of the nanofilms obtained at 8000 rpm for 60 s was comprised between 20 and 30 nm.



Absorption Spectroscopy.

The absorption spectra were obtained with

a Shimadzu 3600
spectrophotometer equipped with a tungsten halogen visible source, a deuterium arc UV
source, a photomultiplier tube UV
-
vis detector, and a InGaAs photodiode and cooled PbS
photocell NIR detectors.
Spectra were recorded in the absorbance m
ode using the

8

integrating sphere accessory (model ISR
-
3100), the wavelength range being 200
-
1200 nm.
The interior of the integrating sphere is coated with a highly diffuse BaO reflectance
standard. Spectra were obtained for both polymer solutions and nanof
ilms. Single
-
scan
spectra were obtained at a scan speed of 60 nm/min using the UVProbe 2.31 software.


Quantum Mechanical Calculations.
Complete and partial geometry optimization of
oligomers containing
n

repeating units (
n
-
T3MA) with
n

ranging from 1 to 1
0 were
performed using Density Functional Theory (DFT) calculations at the B3LYP

[2
8,29
]

level
combined with the 6
-
31+G(d,p) basis set

[
30
]
. Calculations on
n
-
T3MA were performed
considering different arrangements for the methyl acetate side groups, which
were selected
from previous studies on polythiophene derivatives bearing carboxylic acid substituents

[
10,31
]
. The all
-
anti

conformation, in which all the inter
-
ring dihedral angles (

; defined by
the S
-
C
-
C
-
S sequence) adopt a value of 180.0º, was used as
starting structure in all cases.

The Koopman’s theorem

[
32
]
, which
according to Janak’s theorem can be applied to
DFT calculations

[
33
]
,

was used to estimate the ionization potentials (IPs) and electron
affinities (EAs). Accordingly, the IP and the EA were

taken as the negative of the highest
occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO)
energies, respectively,
i.e.

IP=
-
E
HOMO

and EA=
-
E
LUMO
. The IP and EA indicate if a given
acceptor (p
-
type dopant) and donor (n
-
type
dopant) is capable of ionizing, at least partially,
the molecules of the compound, respectively. The

-

*

lowest transition energy (E
g
) was
approximated as the difference between the HOMO and LUMO energies,
i.e.

E
g
= E
HOMO

-

E
LUMO
.
Levy and Nagy evidenced t
hat

g

can be rightly approximated using this procedure
in DFT calculations

[
34
]
.


9


Results and Discussion

Dilute Solutions (0.01 mg/mL to 0.05 mg/mL).

Dried PT3MA (5 mg) was used to
prepare a chloroform solution with concentration 2.6∙10
-
3

mol∙L
-
1
. In orde
r to achieve the
best solubility conditions, chloroform anhydrous than contains 0.5
-
1.0% ethanol was used
as solvent. This is because some residual oxidant particles coming from synthesis were
found to precipitate when no stabilizer was contained in the so
lvent, while such residuals
disappear in presence of a small fraction of ethanol (
i.e.

hereafter, chloroform refers to
chloroform stabilized with ethanol). Exact volumes (1
-
5 mL) of such solution were
transferred to 10 mL volumetric flasks and diluted with

chloroform to prepare dilute
solutions with various known concentrations.

Dilute solutions show an absorbance peak in the UV
-
vis range at

max
= 400 nm (Figure
1), which correspond to the

-

*

transition of the thiophene ring of PT3MA

[
35,36
]
.
Furthermore
, the absorbance of the solutions scaled with concentrations linearly. These
UV
-
vis absorbance (
A
) data (peak at 400 nm) were used to develop the Beer’s law
calibration curve (Figure 2), for which each measurement was repeated three times and the
average v
alues were recorded. The linear portion was plotted and the molar absorptivity (

)
was obtained from the slop of the straight line. The Beer’s law plot for PT3MA was
expressed as the equation:
A
= 6193.7
c

+ 0.1038. Accordingly, considering
A
=

∙b∙c
, where
b

is the path length of the sample (
b
= 1 cm for a standard cuvette), the molar absorptivity of
PT3MA (

max
= 400 nm) in chloroform is

= 6193.7 mol∙g
-
1
∙cm
-
1
, while 0.1038 is typically
attributed to systematic experimental errors (cuvettes, air, solvent, etc)
.



10

Concentrated Solutions (0.03 mg/mL to 2.5 mg/mL).

The evidence of chemical reaction
between the dopant and the polymer comes from the changes observed in the UV
-
vis
spectra. Figure 3 compares the UV
-
vis spectra of PT3MA in chloroform solution with
conc
entrations ranging from 0.03 mg/mL to 2.5 mg/mL. The

-

*

transition of the
thiophene ring shows a red shift with respect to the spectra recorded for dilute solutions
(Figure 1) indicating that the

-
conjugation length increases with the concentration of
p
olymer (
i.e.

the energy required for the

-

*

transition decreases). Thus, the

max

increases
from 400 nm (dilute solutions) to 496 nm (highest concentration in Figure 3). Furthermore,
the band observed at 744 nm, which is not present in dilute solutions,
can be attributed to
the formation of polarons and bipolarons in polymer chains evidencing the influence of the
dopant agent (
i.e.

FeCl
3
)

[
37
-
39
]
.


Effect of the solvent.

Changing from chloroform to more polar and less volatile solvents
in dilute solution
s, has a negligible effect in the

-

*

absorption band of PT3MA. Thus, the

max

is 400, 407 and 400 nm for a 0.01 mg/mL solution in chloroform, chlorobenzene and
cyclohexanone, respectively (spectra not shown). The optical

-

*

lowest transition energy
(E
g
) or band gap energy was determined from the onset wavelength (

onset
) of the UV
-
vis
spectra recorded in these environments,
i.e.

E
g
= 1240/

onset

eV

[4
0]
, the resulting values
being 2.56, 2.48 and 2.48 eV (Table 2), respectively. These results indicate tha
t the
influence of the solvent in the

-
electron delocalization is very small (<

0.1 eV) at the
molecular level. Thus,

although the dielectric constant of these solvents ranged from 3.7
(chloroform) to 15.6 (cyclohexanone)
, the nature of the non
-
specific s
olute∙∙∙solvent

11

interactions

(
i.e.

specific hydrogen bonds between these solvents and P3TMA are not
possible)

is relatively similar in all cases
minimizing the

effects.

In opposition, the solvent plays a very important role in the electronic structure of

concentrated solutions, which is reflected in Figure 4 for the 5.0 mg/mL solutions in the
same three solvents. As it can be seen, the solvent affects the peak position of the

-

*

transition, the

max

obtained for the

-

*

transition in chloroform, cycloh
exanone and
chlorobenzene being 508, 522 and 537 nm, respectively. Thus, red shift occurs when the
volatility of the solvent decreases, which results in a reduction of the energy required for
the

-

*

transition (Table 2). Thus, in chloroform (vapor pressu
re 160 mmHg at 20º) the
value of E
g

is 2.09 eV, decreasing to 1.91 and 1.89 eV in cyclohexanone (vapor pressure
3.4 mmHg at 20ºC) and chlorobenzene (vapor pressure 11.8 mmHg at 25ºC), respectively.
These results
indicate

that the
intermolecular
interaction
s
among

the
polymer

molecules are
affected by the
molecular
mobilit
y of the different

solvent
s in the medium, which in turn
vary

with the
ir

volatility.
Another interesting feature is that the oxidation band observed at
744 nm in chloroform solution is not
detected in chorobenzene and cyclohexanone
confirming that this band corresponds to the FeCl
3

complex.


Effect produced by the addition of other dopants.

Polystyrensulfonic acid (PSSA) and
dodecylbenzenesulphonic acid (DBSA) were dropped in dilute (0.001,
0.005 and 0.01
mg/mL) and concentrated (5.0 mg/mL) chloroform solutions of PT3MA. The spectra
absorption recorded for PT3MA doped with PSSA and DBSA were very similar to those
obtained with the FeCl
3

coming from polymerization (data not shown). For example
, the

max

and E
g

obtained for PSSA in chloroform solution (0.01 mg/mL) were 403 nm and 2.53

12

eV, respectively, whereas they were 402 nm and 2.54 eV for DBSA. The poor doping
behavior of PT3MA is fully consistent with that found for PT3AA

[23
]
, which was
at
tributed to the short distance between the conjugated chain and the quenching carboxylate
group.


Modeling of the electronic properties in the gas
-
phase: influence of the molecular
conformation.

Quantum mechanical calculations on 4
-
T3MA were performed to d
etermine
the conformational preferences of the methyl acetate side group. Thus, different
conformations, which only differ in the disposition of the methyl acetate side groups, were
constructed considering an
anti
arrangement for all the inter
-
ring dihedra
l angles (

= 180º).
All these structures were used as starting points for geometry optimizations at the
B3LYP/6
-
31+G(d,p) level. The two arrangements of lower energy, which are almost
isoenergetic, are depicted in Scheme 3. As it can be seen, in the lowest

energy one (
I
) the
side groups of all repeating units retain the same relative orientation, while in the second
one (
II
) the side groups show an alternated disposition, the latter being only 0.1 kcal/mol
less stable than the former. In both structures, al
l the inter
-
ring dihedral angles evolved
toward an
anti
-
gauche

conformation with


ranging from 130º to 132º.




13

Scheme 3


In order to evaluate the influence of the side groups in electronic properties (IP, EA and
E
g
) of PT3MA
, quantum mechanical calculations on
n
-
T3MA oligomers with
n

ranging
from 2 to 10 were performed considering the arrangements
I

and
II

for each value of
n
. The
most favored arrangement was
I

in all cases, the maximum destabilization obtained for
II

being o
nly 0.3 kcal/mol. Furthermore, the backbone adopted an
anti
-
gauche

conformation
in all cases with


varying between 128º and 140º. These results clearly indicate that,
independently of the number of repeating units, the interaction between the methylacetat
e
side groups and the polythiophene backbone is very weak. On the other hand, the all
-
anti

arrangement has been also considered for the modeling of the electronic properties. Thus,
each
n
-
T3MA oligomer was optimized considering initially the side groups as

in
arrangement
I
and fixing the inter
-
ring dihedral angles at 180º (restricted geometry
optimizations). As it was expected, the
all
-
anti

conformation was less stable than the
anti
-
gauche

one, this effect increasing with
n
,
i.e.
the relative energy increas
es from 1.9 to 12.8
kcal/mol when
n

grows from 2 to 10.

Figure 5 displays the variation of the IP, EA and E
g

calculated in the gas
-
phase with the
inverse of the number of repeating units for the
anti
-
gauche

(
I

and
II
) and all
-
anti

arrangements. Linear regr
ession analyses, which are also displayed in Figure 5, allowed
extrapolate the IP, EA and E
g

values for an infinite chain of PT3MA. As it was expected,
the IP predicted for two
anti
-
gauche

arrangements, which only differ in the relative
disposition of the
side groups, were similar due to the relatively poor influence of the
backbone


side chains interactions (Figure 5a),
i.e.

5.25 and 5.37 eV, respectively.

14

However, oxidation of PT3MA becomes an easier process when the thiophene rings are
obligated to adop
t a planar disposition, the IP for the all
-
anti

conformation being 4.74 eV.
The latter value is almost identical to those calculated for the all
-
anti

conformation of
unsubstituted polythiophene (4.75 eV)

[4
1]

and poly(2
-
thiophen
-
3
-
yl
-
malonic acid
dimethyl
ester) (PT3MDE, 4.71 eV)

[10
]
, a similar polythiophene derivative that contains
two carboxilate groups per repeating unit (Scheme 4), using similar theoretical methods.



Scheme 4


Figure 5b, which displays the variation of the EA against 1/
n

for the thre
e investigated
arrangements, indicates that energy of the LUMO is around 0.4 eV larger for the all
-
anti

than for
I

and
II
. The influence of the backbone conformation on the IP and EA has a
significant impact on the E
g
, the value predicted for an infinite o
f PT3MA being 2.78, 2.93
and 1.89 for the
I
,
II

and all
-
anti

conformations, respectively (Figure 5c). As it was
expected, the

g

value decreases when the planarity of the backbone increases, which must
be attributed to the enhancement of the

-
conjugation
produced when the inter
-
ring
dihedral angles evolve towards 180º. The E
g

value determined experimentally for a dilute
chloroform solution, which is the solvent with the lowest polarity among those used in this
work, is 2.56 eV indicating an overestimation
of 0.22 and 0.37 eV for arrangements
I

and

15

II
, respectively. In contrast, the E
g

of the all
-
anti

conformation is underestimated by 0.67
eV. Accordingly, the backbone of the PT3MA molecules in dilute chloroform solution
adopts an intermediate conformation b
etween the
anti
-
gauche

and the
all
-
anti
, even though
it is closer to the former than to latter. In contrast, the E
g

determined by UV
-
vis
spectroscopy for a concentrated chloroform solution (5 mg/mL) is closer to the E
g

predicted
for the all
-
anti

conformati
on than for the
anti
-
gauche

one,
i.e.

the difference is 0.20 and
0.69 eV, respectively. Accordingly, PT3MA chains tend to adopt planar arrangements in
concentrated solutions, which is fully consistent with the red
-
shift observed for

max

(
i.e.

the

-
conjug
ation length increases in polymer chains). Moreover, this planarity increases
when the volatility of the solvent decreases, the E
g

predicted for a concentrated
chlorobenzene solution being identical to that calculated for the ideal all
-
anti

conformation.

F
igure 6 shows the contour plots HOMO and LUMO computed for
n
-
T3MA with
n
= 3
and 10. As it can be seen, the HOMO presents a major bonding character delocalized along
the backbone. For the shortest oligomer such delocalization is uniform, whereas for the
lar
gest one this effect involves the eight central repeating units. Furthermore, in both cases
the HOMO resides in the conjugated backbone, no participation of the carbonyl side groups
being detected. This is fully consistent with the weak interaction found b
etween the
methylacetate side groups and the polythiophene backbone. Similar distributions are
predicted for the LUMO of 3
-
T3MA and 10
-
T3MA,
i.e.

complete and partial
delocalization, respectively, which mainly resides in the inter
-
ring


bonds.


Nanofilm
s.

Figure 7 shows the absorption spectra of ultra
-
thin films of PT3MA. Films
were prepared by spin
-
casting from chloroform solutions using the same speed and time of

16

spinning in all cases, their thickness being varied by changing the concentration of polym
er
in the solution (see Methods section). Although the absorption profiles recorded in the solid
state looks similar to those obtained in solution, the thickness of the films affects both the
peak position and the absorbance. A blue shift is detected with
increasing the concentration
of PT3MA in the solution used for the spin
-
casting, the

max

of the films cast using 0.01,
1.00 and 5.00 mg/mL in chloroform being 482, 433 and 419 nm, respectively. Furthermore,
the E
g

gap increases with the concentration, gro
wing from 1.52 to 2.17 eV when the
concentration increases from 0.01 to 5.00 mg/mL (Table 2). These results indicate that the
more effective packing of the

-
conjugated backbone is obtained for the nanofilm samples
prepared using the more dilute solutions.

Furthermore, the very low E
g

values determined
for nanofilms produced using dilute solutions indicate that intermolecular

-
conjugation
effects are very important in the solid state, as was recently predicted from sophisticated ab
initio calculations

[
4
2]
. Thus, the E
g

measured for the nanofilms casted from the 0.01
mg/mL solution is 0.37 eV smaller than that predicted for an isolated polymer chain
arranged in an ideal planar conformation.

On the other hand, it is worth noting that the E
g

measured for the
film cast from the
5.00 mg/mL solution (2.17 eV) is higher than those determined in the solution state for the
same concentration, independently of the solvent (1.89
-
2.09 eV, in Table 2). Thus, the
reduction of the gap produced by

-
stacking interactions i
s not enough to compensate the
negative conformational effects in the molecular

-
conjugation length. Accordingly, spin
-
casting of concentrated solutions produce distortions in the shape of the polymer chains,
which are not able to readapt their conformati
ons due to the packing of the neighboring
molecules.


17


Conclusions

The electronic properties of PT3MA have been examined in different environments
using UV
-
vis absorption spectroscopy and quantum mechanical calculations. The
absorption
of

the

thiophene


-

*

transition
is not affected by the chemical nature of the
solvent in

dilute polymer solutions. In contrast,

this

transition shows a red
-
shift for
concentrated solutions
,

which produces a reduction of the E
g
.
On the other hand, q
uantum
mechanical

calculat
ions indicate that
the

anti
-
gauche

arrangement
is

more stable in the gas
-
phase than the all
-
anti

by
3 kcal/mol per repeating unit
.

Comparison of the E
g

values
calculated for such two conformations

with those determined experimentally allow
s

conclude that i
n dilute solution PT3MA chains
are

closer to the
anti
-
gauche

than to the all
-
anti
, whereas

in concentrated solutions polymer chains
prefer the latter.

The E
g

values
measured

for nanofilms are significantly smaller than
that predicted for an ideally planar
molecule
,
the more effective packing of the chains
being

obtained for the films prepared
using the more dilute solutions.


Acknowledgements

This work has been supported by MICINN and FEDER (Grant MAT2009
-
09138), by
the Generalitat de Catalunya (research g
roup 2009 SGR 925 and XRQTC). Computer
resources were generously provided by the “Centre de Supercomputació de Catalunya”
(CESCA).
A.G. acknowledges financial support from the Euro Brasilian Windows agency
(Grant No.

41309
-
EM
-
1
-
2008
-
PT
-
ERAMUNDUS
-
ECW
-
L16) f
or his 6
-
month stay at the

18

UPC.

Support for the research of C.A. was received through the prize “ICREA Academia”
for excellence in research funded by the Generalitat de Catalunya.


References

1.

Skotheim

TA
, Reynolds

JR (2007).

Handbook of Conducting Polymer
s, 3
rd

ed.

CRC Press, Taylor and Francis Group, Boca Raton, FL.

2.

Wallace

GG,
Spinks

GM
, Kane
-
Maguire

LAP
, Teasdale

PR (2009).

Conductive
Electroactive Polymers,

3
rd

ed.

CRC Press, Taylor and Fra
ncis Group, Boca Raton,
FL
.

3.

Chen

TA
, Wu

X
, Rieke

RD (1995) J Am

Chem Soc 117:
233
-
244
.

4.

Yang

C
, Orfino

FP
, Holdcroft

S (1996)

Macromolecule
s 29:6510
-
6517
.

5.

McCullough

RD (1998)

Adv Mater 10:93
-
116
.

6.

Patil

OA
, Ikenoue

Y
, Wudl

F
, Heeger

AJ (1987) J Am Chem Soc 109:
1858
-
1859
.

7.

Chayer

M
, Faïd

K
, Leclerc

M (1997) Chem Mater
9:2902
-
2905.


8.

Kim

B
, Chen

L
, Gong

J
, Osada

Y (1999) Macromolecules

32:3964
-
3969
.

9.

Visy

K
, Kanlare

J
, Kriván

E (2000) Electrochim Acta 45:

3851
-
3864
.

10.

Armelin

E
, Bertran

O
, Estrany

F
, Salvatella

R
, Alemán

C (2009) Eur Polym J 45:

2211
-
2221
.

11.

Bertran

O
, Armeli
n

E
, Estrany

F
, Gomes

A
, Torras

J
, Alemán

C (2010) J Phys
Chem B 114:
6281
-
6290
.

12.

Bertran

O
, Pfeiffer

P
, Torras

J
, Armelin

E
, Estrany

F
, Alemán

C (2007) Polymer 48:

6955
-
6964
.


19

13.

Bertran

O
, Armelin

E
, Torras

J
, Estrany

F
, Codina

M
, Alemán

C (2008) Polymer
49:1
972
-
1980.

14.

Takeoka

Y
, Iguchi

M
, Rikukawa

M
, Sanui

K (2005) Synth Met 154:109
-
112.

15.

Constantine

CA
, Mello

SV
, Dupont

A
, Cao

X
, Santos Jr

D
, Oliveira Jr

ON,

Strixino

FT
, Pereira

EC
, Cheng

TC
, Defrank

JJ
, Leblanc

RM (2003) J Am Chem Soc 125:

1805
-
1809
.

16.

Yoon

YS
,

Park

K
-
H
, Lee

J
-
C (2009) Macromol Chem Phys 210:1510
-
1518
.

17.

Kim

Y
-
G
, Kim

J
, Ahn

H
, Kang

B
, Sung

C
, Samuelson

LA
, Kumar

J (2003) J
Macromol Sci Pure Appl Chem A40:
1307
-
1333
.


18.

Thuwachaowsoan

K
, Chotpattananont

D
, Sirivat

A
, Rujiravanit

R
, Shwank

JW
(2007) Ma
ter Sci Eng B 140:
23
-
30
.

19.


Zhang

Z
, Wang

L
, Deng

J
, Wan

M (2008) React Funct Polym 68:
1081
-
1087
.

20.

Jiang

Y
, Wu

P (2008) Appl Spectrosc 62:
207
-
.

21.

Nilsson

KP
, Rydberg

J
, Baltzer

L
, Inganäs

O (2004) Proc Natl Acad Sci 101:
11197
-
11202
.

22.

Zang

F
, Srinivivasan

MP (2
005) Thin Solid Films 479:
95
-
102
.

23.

Zang

F
, Srinivivasan

MP (2008) Macromol Chem Phys 112, 223
-
224.

24.

Hsieh

KH
, Ho

KS
, Wang

YZ
, Ko

SD
, Fu

SC (2001) Synth Met 123:
217
-
224
.

25.

Faid

K
, Leclerc

M (1998) J Am Chem Soc 120:
5274
-
5278
.

26.

Watanabe

H
, Kunitake

T (2007) Adv M
ater 19:9
09
-
912
.

27.

Jaczewska

J
, Budkowski

A
, Bernasik

A
, Moons

E
, Rysz

J (2008) Macromolecules
41:
4802
-
4810
.

28.

Becke

AD (1993) J Chem Phys 98:1372
-
1377.

29.

Lee

C
, Yang

W
, Parr

RG (1988) Phys Rev B 37:785
-
789.


20

30.

Fris
c
h

MJ
, Pople

JA
,
Binkley JS (1984) J Chem Phys 80
:3265
-
3269.

31.

Casanovas

J
, Zanuy

D
, Alemán

C (2005) Polymer 46:
9452
-
9460
.

32.

Koopmans

T (1934) Physica 1:104
-
113.

33.

Janak

JF (1978) Phys Rev B 18:7165
-
7168.

34.

Levy

M
, Nagy

A (1999) Phys Rev A 59:
1687
-
1689
.

35.

de Souza JM, Pereira EC (2001) Synth Met 118:167
-
170.

36.

Jiang

Y
, Shen

Y
, Wu

P (2008) J Col Interf Sci 319:398
-
405

37.

Aaron

JJ
, Fall

M (2000) Spectrochim Acta Part A 56:
1391
-
1397
.

38.

Calado

HDR
, Matencio

CL
, Donnci

LA
, Cury

LA
, Rieumont

J
, Pernaut

JM (2008)
Synth Met 158:1037
-
1042

39.

Alves

MRA
, Calado

HDR
, Donnici

CL
, Matenc
io

T (2010) Synth Met 160:22
-
27.

40.

Meng

H
, Zheng

J
, Lovinger

AJ
, Wang

B
-
C
, Van Patten

PG
, Bao

Z (2003) Chem
Mater 15:
1778
-
1787
.

41.

Casanovas

JC
, Alemán

C (2007) J Phys Chem C 111:
4823
-
4830
.

42.

Rodríguez
-
Ropero

F
, Casanovas

J
, Alemán

C (2008) J Comput Chem 29:69
-
,
J.
Comput. Chem.

2008
,
29
, 69
-
78
.


21

Table 1.

Basic solubility data of PT3MA in different solvents at room and high
temperature.


Solvent

25 ºC

60 ºC

Water (pH= 7.0)





乡佈⁡焮





Acetone

Δ

Δ

CH
2
Cl
2





CHCl
3





THF





DMSO

Δ



DMF

Δ



TFA





Xylene





Methanol





Ethanol





Toluene





Hexane





Acetonitrile





Diethyl ether



-

Ciclohexanone





Chlorobenzene







= soluble;
Δ

= partialy soluble;


= insoluble



22

Table 2.

Values of

max
,

onset

and E
g

determined experimentally
and theoretically for
PT3MA in different conditions.


Conditions


max

(nm)


onset

(nm)

E
g

(eV)

Dilute chloroform sol. (
0.01 mg/mL)

400

485

2.56

Dilute chlorobenzene sol. (
0.01 mg/mL)

407

500

2.48

Dilute cyclohexanone sol. (
0.01 mg/mL)

400

499

2.48

Con
centrated chloroform sol. (5.00 mg/mL)

508

592

2.09

Concentrated chlorobenzene sol. (5.00 mg/mL)

537

657

1.89

Concentrated cyclohexanone sol. (5.00 mg/mL)

522

647

1.92

DFT calculations, gas
-
phase, arrangement I

-

-

2.78

DFT calculations, gas
-
phase arra
ngement II

-

-

2.93

DFT calculations, gas
-
phase, all
-
anti

-

-

1.89

Nanofilm


獰楮⁣s獴s湧 E〮〱M⽭i)

㐸4

㠱8

ㄮ㔲

乡湯晩汭


獰楮⁣s獴s湧 Eㄮ〰1⽭i)

㐳4

㘲S

ㄮ㤸

乡湯晩汭


獰楮⁣s獴s湧 E㔮〰R⽭i)

㐱4

㔷R

㈮ㄷ



23

Captions to Figures


Figure 1.
U
V
-
vis absorption spectra of PT3MA in chloroform solutions with different
concentrations used to develop the calibration curve.

Figure 2.

Calibration curve used to develop the Beer’s law slope for PT3MA in dilute
chloroform solution.

Figure 3.
UV
-
vis absorp
tion spectra of concentrated PT3MA solutions in chloroform.

Figure 4.
UV
-
vis absorption spectra of concentrated PT3MA solutions (5.0 mg/mL) in
chloroform, cyclohexanone and chlorobenzene.

Figure 5.

Variation of the calculated IP (a), EA (b) and E
g

(c) ag
ainst 1/
n
, where
n

is
the number of repeat units for PT3MA arranged in the following conformations:
anti
-
gauche

with
I

(black line, black triangles),
anti
-
gauche

with
II
(grey line, grey squares) and
all
-
anti

(dashed line, empty circles). The lines corresp
ond to the linear regressions used to
obtain the values of these electronic properties for infinite chain systems.

Figure 6.

Comparison of the contour plots of the HOMO and LUMO for
n
-
T3MA with
(a)
n
= 3 and (b)
n
= 10.

Figure 7.

UV
-
vis absorption spectra of

PT3MA nanofilms obtained by the solution
spin
-
casting process using polymer solutions with concentrations ranging from 0.01 to 5.00
mg/mL. The values of

max

are indicated for each concentration.






24






























Figure 1



25






























Figure 2


26
































Figure 3



27

































Figure 4





28












































Figure 5


29









































Figure 6


30






























Figur
e 7



31

Text for the Table of Contents


The

-
conjugation length and the energy required for the

-

*

transition of poly(thiophene
-
3
-
methyl acetate) have been examined in different environments (
i.e.

gas
-
phase, dilute and
concentrated solutions considering s
olvents with different polarity and volatility, and spin
-
casted nanofilms) using a combination of UV
-
vis spectroscopy and quantum mechanical
calculations.


Graphic for the Abstract