The influence of experimental conditions and intermolecular interaction on the band gap determination. Case study of perylene diimide and carbazole-fluorene derivatives.

Article abstract of DOI:10.1016/j.electacta.2016.09.046

The perylene diimide and carbazole-fluorene derivatives are intensively investigated in organic electronic and photovoltaic applications. However, they intermolecular interactions vary considerably. Herein, we present a systematic UV–vis spectroscopy and electrochemical study of intermolecular interaction influence on determination of HOMO, LUMO and band gap. The influence of this type of interactions on photophysical properties with the aid of NMR spectroscopy and DFT calculations was examined. Effect of the electrode type was investigated as well. The cyclic voltammetry results obtained at different species concentrations (including solid state measurements) were compared. We have found that when interactions are strong differences in determined bang gap are significant. Contrary to half wave potential, peak onset was found to be concentration independent. Finally, we have observed that comparison of values from solid state measurement and carried out for diluted species can introduce high inconsistencies in determination of structure/property relationships.

Full text of DOI:10.1016/j.electacta.2016.09.046

Electrochimica Acta 216 (2016) 449–456
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
The influence of experimental conditions and intermolecular
interaction on the band gap determination. Case study of perylene
diimide and carbazole-fluorene derivatives.
*
Michał Filapek , Marek Matussek, Agata Szlapa, Slawomir Kula, Michał Paja˛k
Institute of Chemistry, Faculty of Mathematics, Physics and Chemistry, University of Silesia, Szkolna 9, 40-007 Katowice, Poland
A R T I C L E I N F O
A B S T R A C T
Article history:
The perylene diimide and carbazole-fluorene derivatives are intensively investigated in organic
electronic and photovoltaic applications. However, they intermolecular interactions vary considerably.
Herein, we present a systematic UV–vis spectroscopy and electrochemical study of intermolecular
interaction influence on determination of HOMO, LUMO and band gap. The influence of this type of
interactions on photophysical properties with the aid of NMR spectroscopy and DFT calculations was
examined. Effect of the electrode type was investigated as well. The cyclic voltammetry results obtained
at different species concentrations (including solid state measurements) were compared. We have found
that when interactions are strong differences in determined bang gap are significant. Contrary to half
wave potential, peak onset was found to be concentration independent. Finally, we have observed that
comparison of values from solid state measurement and carried out for diluted species can introduce
high inconsistencies in determination of structure/property relationships.
Received 11 April 2016
Received in revised form 7 September 2016
Accepted 7 September 2016
Available online 9 September 2016
Keywords:
Cyclic voltammetry
intermolecular interactions
arylene bisimide
carbazole
UV-Vis spectroscopy
ã 2016 Elsevier Ltd. All rights reserved.
1. INTRODUCTION
energy which has paired electrons (formally it would entail adding
two electrons to the neutral molecule). This does not reflect the
Organic electronic and photovoltaic are being developed
extremely fast. Progress in this field is possible thanks to structure
$ properties feedback ꢀ in each step it is better known how the
structure affects physical and chemical characteristics of the
molecule and by that device performance. With that knowledge it
is easier to design and synthesize compounds which have
exaggerated most desired properties. However, sometimes it is
difficult to compare results from different sources, especially
properties related to energy levels in the molecule (HOMO and
LUMO energies). This is due to the inability to measure their
absolute value experimentally. On the other hand, energy values
calculated by quantum chemistry methods depends on the base
and functional which were used (and also by HF/DFT ratio in hybrid
functional [1]). That makes those values very often only estimated.
Differences between calculated and experimental values of LUMO
energy are predominately especially different [2]. In contrast to
HOMO, LUMO is unreal (it is empty in the real molecule which is in
ground state). Furthermore, by definition of MO, that is an orbital
experience ꢀ electrochemical reduction the most often is a single-
electron process [3], photoexcitation processes involved one
electron as well. That is why sometimes radicals are analyzed
using unrestricted DFT approach [4,5] The ionization energy (IE)
and electron affinity (EA) can be directly determined by UV-photo
electron spectroscopy (UPS) and inverse photo electron spectros-
copy (IPES). On the other hand, the most convenient (the fastest,
the cheapest and very small amount of sample is needed) methods
of the EA and IE determining are the electrochemical methods.
Generally speaking, UPS/IPES values are larger than the values
derived from CV. The offset is about 0.5–0.8 eV in case of EA and
0.4–1.0 eV for IE [6,7]. Moreover, when comparing values deter-
mined on the physical (eV) and electrode potential scales (V vs.
reference electrode) in solution the potential drop across an
electrical double layer (the Helmholtz layer) needs to be
considered. Usually zeta potential is used for estimating the
degree of this surface ꢀ solution interactions (a typical value is
approximately 25–100 mV). But still there is the problem of
reference calculation results to the experimental values which has
been described in detail by Jean-Luc Bredas [8], P. Bujak [9], or A.
Pron [10]. As mentioned before the inability to measure the
absolute value of MO is the biggest problem. Because of that those
* Corresponding author at: Institute of Chemistry, University of Silesia, 9 Szkolna
Str 40-006 Katowice, Poland.
E-mail address: michal.filapek@us.edu.pl (M. Filapek).
http://dx.doi.org/10.1016/j.electacta.2016.09.046
0013-4686/ã 2016 Elsevier Ltd. All rights reserved.

450
M. Filapek et al. / Electrochimica Acta 216 (2016) 449–456
Fig. 1. Synthetic pathway for 1. (i) 2-ethylhexylamine, DMF, reflux, 8 h.
values have to be showed in relation to a standard, for example SCE,
NHE or ferrocene [11]. It is worth to be mentioned that in literature
there are different values of ionization energy of ferrocene (-4.8 eV
[12], ꢀ4.88 eV [13], ꢀ 5.1 eV [9]). Taking into consideration fact that
this value is directly use to calculate HOMO and LUMO levels this
inconsistency may hinder comparison of data from different
sources.
Moreover, intermolecular interaction can also affect determi-
nation of values connected with energetic levels. Thus, we have
choose compounds differing greatly in intermolecular interactions
coil, and silver wire were used as auxiliary and reference electrode,
respectively. Potentials are referenced with respect to ferrocene
(Fc), which was used as the internal standard. Cyclic and
differential pulse voltammetry experiments were conducted in a
standard one-compartment cell, in CH2Cl2 (Carlo Erba, HPLC
grade), under argon. Bu4NPF6 (Aldrich; 0.2 M, 99%) was used as the
supporting electrolyte. The quantum theoretical calculations were
performed with use of density functional theory (DFT), with an
exchange correlation hybrid functional B3LYP and the basis 6–
311 + G(d,p) for all atoms. The calculations were carried out with
use of Gaussian 09 program.
i.e. perylene diimide (exhibiting strong
p
ꢀ interaction between
the flat aromatic sections [6]) and carbazole-fluorene (which is an
amorphous substance). This paper is an attempt to systematize
measurement techniques and the compare electrochemistry in
solution and in the solid state with results of theoretical
calculations and obtained by NMR and UV–vis spectroscopy. In
addition, we examined the electrode type, scan rate and whether
the sample (whether the sample is in form of solid or solution) on
the obtained empirical value.
3. RESULTS
For the measurement, two compounds with very different
physical and chemical properties have been chosen, although they
both belong to the intensively investigated for use in organic
electronics groups of compounds. Compound
1 (carbazole
2. EXPERIMENTAL
2.1. General Methods
All chemicals and starting materials were commercially
available and were used without further purification. Solvents
were distilled as per the standard methods and purged with
nitrogen before use. Column chromatography was carried out on
Merck silica gel. Thin-layer chromatography (TLC) was performed
on silica gel (Merck TLC Silica Gel 60). NMR spectra were recorded
with a Bruker Avance 400 MHz instrument by using CDCl3 as
solvent. 2,7-diiodofluorene [14], 2,7-diiodo-9,9-dioctylfluorene
[15], 2-iodo-7-(9H-carbazole-9-yl)-9,9-dioctylfluorene [16], N,N’-
bis(2-ethylheksyl)-3,4,9,10-perylene diimide [17] were prepared
to the literature procedures. Synthesis details for 2-(trimethylsi-
lylethynyl)-7-(9H-carbazol-9-yl)-9,9-dioctylfluorene (1) can be
found in SI. Both synthesis pathways are summarize in Figs. 1
and 2 for compounds (1) and (2), respectivelly. UV/Vis spectra were
recorded with a Hewlett–Packard model 8453 UV/Vis spectropho-
tometer in dichloromethane solution. Electrochemical measure-
ments were carried out with an Eco Chemie Autolab PGSTAT128n
potentiostat using glassy carbon, gold, platinum (all with diam.
2 mm) or Indium tin oxide(ITO, with 10
V per square) as working
electrode. For thin film preparation, substances were spin-coated
on the cleaned electrode surface from fresh solution (2 mg/cm3 in
methylene chloride (DCM)) and baked 30 min at 60 ꢁC. Platinum
Fig. 2. Synthetic pathway for 2.

M. Filapek et al. / Electrochimica Acta 216 (2016) 449–456
451
Fig. 3. The structures of the examined compounds together with photos of 1 an 2.
derivatives) is an amorphous compound with high viscosity
(Fig. 3). It does not crystallize even after cooling, this suggests a
very weak intermolecular interactions. Compound 2 i.e. perylene
diimide (PDI) derivatives is selected from the group of compounds
between the behavior of both substances during measurements by
other techniques (CV and UV- vis) should be expected.
In a further step the effect of the concentration on UV–vis
spectrum has been examined. In the case of 1 the difference
between the absorbance onsets in a solid and in a solution is low (a
few nanometers) i.e. 360 nm (what can be recalculated to 3.44 eV
using Eg = 1240/labs formula)in solution (2e ꢀ6 mol/L) and 363 nm
(3.41 eV) in a thin layer (Fig. 2S in supporting information). As has
been previously demonstrated with a NMR technique orthogonal
rings prevent the creation of strong intermolecular interactions,
thus concentration has negligible influence. In case of 2 significant
decreasing energy gap when the concentration increases was
observed ꢀ from 536 nm (2.31 eV; for a 2e ꢀ6 mol/L solution) to
608 nm (2.04 eV for thin layer) ꢀ see Fig. 5.
with the tendency to form a strong
p
ꢀ interaction between the
flat aromatic sections [6]. This feature makes this type of
compounds crystalline solids what was also observed in the case
of 2 (Fig. 3).
In a first step, a series of NMR analysis of the solutions with
different concentrations of the 1 and 2 have been performed. As
shown in Fig. 4 aromatic hydrogen in PDI with increasing
concentration become increasingly shielded. In other words, the
lines of magnetic force (generated by the electrons swirling motion
within the aromatic ring) are arranged in a way those hydrogens
are increasingly obscured as the concentration increases (the
paramagnetic effect) (Fig. 4) and thus, hydrogen atoms absorb
wave of increasing energy as the concentration increases. As has
been reported earlier [18] the planes of the aromatic rings tend to
set in parallel (if there are no steric obstacles). This results the
A similar relation was already described for compounds having
in their structure PDI moiety [6,19] ꢀ aggregation always causes an
increase of absorbed light wavelength (i.e. photons with smaller
energy). It should be empathizing that during UV–vis spectra
measurement generation of “Frenkel exciton” occurs ꢀ excited
electron remains in a Coulomb contact with a hole. Bearing in mind
stronger interactions induced dipole
ꢀ
induced dipole and
consequently lowers the energy of the system. On the other hand,
in the case of 1 completely reverse phenomenon has been
observed (see Fig. 1S in supplementary materials). This is due to its
structure ꢀ that is, nearly orthogonal two aromatic rings ꢀ
carbazole and fluorene. Thus, it is impossible for the molecule to
set in parallel ꢀ and this increases diamagnetic effect as the
concentration increases ꢀ lines of the magnetic field are arranged
in such way that counteract external magnetic field.
strong p-p interaction in PDI, holes become more stabilized by a
resonance with increasing concentration. This fundamentally
differentiates this technique from electrochemical measurement.
It should be always concerned that during the electrochemical
reduction electron is delivered “to the system” (from the electrode
surface). Similarly, during the oxidation of molecule the electron is
“derived” from the electrode. Therefore questionable is the method
used sometimes in the literature when for the HOMO energy
calculation equation: Eg (determined from uv–vis spectroscopy)
ꢀLUMO (determined from cyclic voltammetry) is used (or,
similarly, for LUMO: Eg (determined from uv–vis spectroscopy)
It is worth remembering that NMR technique directly measures
the resonance frequency of the nuclei of hydrogen, but it is directly
dependent on the ambient electron (electron density). Differences
Fig. 4. 1H NMR spectrum (depending on the concentration in CDCl3 solution) and paramagnetic effect of 2.

452
M. Filapek et al. / Electrochimica Acta 216 (2016) 449–456
Fig. 5. UV–vis spectrum at various concentration (top) and luminescence of compound 2, concentrations in CH2Cl2 solution (bottom): 1) 10ꢀ2 M; 2) 10ꢀ3 M; 3) 10ꢀ4 M; 4)
10ꢀ5 M; 5) 10ꢀ6 M; 6) 10ꢀ7 M (Left: under visible light, right: under irradiation with light at 366 nm).
ꢀꢀHOMO (determined from cyclic voltammetry)). Moreover, some
transitions are either symmetry forbidden or has negligible
oscillator strength, becomes spectroscopically inactive. On the
other hand, cyclic voltammetery don’t share these restrictions ꢀ
electrochemical measurement is based on thermodynamics lows
deriving directly from a chemical potential change and electrical
work:
Hsiao et al. [21] was recorded, what is also in agreement with our
previous research [20,22,23]. As it is shown on voltamograms
above (Fig. 6) the first (i.e. PDI/PDIꢀ) and the second (PDIꢀ/PDI2ꢀ
)
reductions were reversible and well detectable (except for the case
0
dGi = RT dln ai = We = ꢀzFE01 E = E0 + (RT/zF)ln(aox/ared
)
(1)
The last equation (1) is of course the Nernst equation. However,
we have to remember that in the equation the activities of the
substances are used (ai = f * c), not simply the concentration (a = c
ꢀonly for ideal solutions or ideal, dilute solutions). Secondly, the
effect of electrolyte and solvent has to be considered. This issue is
substantially reduced by the use of an internal standard i.e.
ferrocene (assuming similar environmental shift of oxidation/
reduction onsets) and using the extrapolation of the linear part of
the obtained curve (for which Nerst equation is strictly fulfilled).
Additionally, our experience with measurements of amide systems
clearly shows that working electrode type is also important,
especially in the case of measuring systems with multistage redox
processes (eg. imide copolymers) [20]. Thus, electrochemical
research has been begun by analyzing the impact of electrode
material on the result.
For each diimide (DI) a reduction process towards the formation
of an anion radical according to mechanism put forward by S.-H.
Fig. 6. Cyclic voltammograms obtained for 2 for different working electrodes;
arrow shows the initial direction of current flow; c = 10ꢀ5 mol/dm3 sweep rate
,
n
= 100 mV/s, 0.1 M Bu4NPF6 in CH2Cl2.

M. Filapek et al. / Electrochimica Acta 216 (2016) 449–456
453
Table 1
Comparison of electrochemical and theoretical data.
Code
IEa (CV)
[eV]
EAb (CV)
[eV]
HOMO (DFT)
[eV]
LUMO (DFT)
[eV]
IEc (DFT) [eV]
EAd (DFT) [eV]
Eg (DFT)
[eV]
Eg (CV)
[eV]
Eg (opt)
[eV]
e
f
e
f
f
f
1
2
5.85e (5.80)f
6.29 (6.41)
ꢀ2.75 (-2.68)
ꢀ5.79
ꢀ6.14
ꢀ1.86
ꢀ3.68
5.52
5.94
ꢀ2.15
ꢀ3.92
3.93g (3.37)h
3.10 (3.22)
3.44e (3.41)
2.26 (2.04)
e
f
e
f
h
e
e
ꢀ4.10 (-3.96)
2.46g (2.02)
2.19 (2.45)f
a
IE(eV) = |e|(Eox(onset) + 5.1) measured using GC.
b
c
d
e
f
EA(eV) = ꢀ|e|(Ered(onset) + 5.1) measured using GC.
IE is energy change in the process M ꢀ eꢀ ! Mꢃ+
.
EA is energy change in the process M + eꢀ ! Mꢃꢀ (negative values indicate exothermicity upon reduction of molecule.
determined for solution.
determined for solid state.
from HOMO-LUMO.
from |IE|-|EA|.
g
h
where ITO were used as working electrode). First reducion onsets
measured for the 2 on gold and GC electrodes are almost identical
(-0.99 V and ꢀ1.00 V, respectively). On the other hand, in the case
of Pt electrode onset is lower by about 70 mV (i.e. ꢀ1.07 V). What is
also important difference between onsets E1red and E2red registered
on the Au and GC electrodes is as follows: 240 mV and 220 mV (so
very close in value to each other), while on Pt electrode is almost
300 mV, which suggests difficulties with the charge transfer.
Probably, in this case platinum complex is formed on a surface with
the imide (being O-donating ligand) which impedes the flow of
charge. However, in each case of the conducting electrodes
processes are fully reversible from a formal point of view. As
can be seen at Fig. 6 for ITO (which is semiconducting electrode)
even in the solution under identical conditions processes are
quasi-reversible. Moreover, onsets of the first stage of the
reduction are evidently lower (E1red = ꢀ1.12 V) and quasi-reversible
while second become to be irreversible. For the oxidation we also
observed some variations in onsets (see Table 1S in supp. inf.). But
what is important in the case of platinum and (especially) ITO
electrodes the oxidation onsets were much less apparent, and this
explains the fact that sometimes it is difficult to determine the
oxidation potential of those electrodes (see Fig. 3S in supp. inf.). On
the other hand on GC and Au electrodes are well defined, so if there
are difficulties with electrochemical measuring it should be
decided to use a different type of electrode.
previously the measured potential is dependent on the activity of a
examined substance. Therefore, changing the activity (by changing
the concentration) must affect on the observed potential (it
changes in logarithmic segment). We have to remember that
difference between the cathodic and anodic peaks |Epc-Epa| (peak-
to-peak separation) should be 56 mV (approximately) for one
electron, reversible process. Additionally, the average of this two
peaks (E1/2) is taken as excellent approximation of E0' [11]and is
commonly used by many researchers. On the other hand peak
onset value is also commonly used [24,25]
As shown on the voltamograms above (Fig. 7) with increasing
concentration the maximum of Epa is shifted by 50 mV concen-
trating solution 10-times (in case of 2 at 1 ꢂ10ꢀ2 mol/L precipitates
of measured species occurs). But Epc for the concerned process is
nearly independent from this factor. In our opinion, this
phenomenon is caused by the fact that in each case concentration
of reduced form generated during measurement is approximately
the same (and aredas consequence). But aox is strongly depended
from concentration, thus aox/ared ratio is different in each time. This
implies, that from the thermodynamicall point of view only at low
concentration process is pure reversible (|Epc-Epa| = 56 mV). At
higher concentration process could be mistakenly assigned as
quasi-reversible (for example |Epc-Epa| = 150 mV at 1 ꢂ10ꢀ3 mol/L).
However, in all cases (even in 1 ꢂ10ꢀ2 mol/L solution) peak onset is
constant.
We have studied also the concentration affects on a band gap
determined electrochemically. As it has been already reminded
We have also compared the behavior of the substance in the
form of solutions and the thin layer applied on the surface of the
electrode. As shown at Fig. 8, the oxidation of the carbazole
derivative takes place a bit easier for a solid. Furthermore, in the
solution characteristic irreversible oxidation of carbazole can be
registered, not fully shaped in a solid.
Much bigger differences have been observed making a similar
comparison for perylene derivative (Fig. 9).
It can be easily observed that both oxidation and reduction, in
this case take place much harder in the solid than in solution. In
addition, the reduction is no longer a reversible process (in the
pure sense thermodynamics), and become quasireversible (similar
behavior was reported by A.O. Aleshinloye et al. for similar PDI
derivative [26,27]. This also results in a significant diminution in
“energy gap” determined electrochemically. Moreover, the differ-
ence is surprisingly high ꢀ Eg in the solution is 2.1 eV, while in the
solid state is 2.4 eV. Importantly, the results are almost identical,
regardless of the material of the electrode. What is interesting that
the ratio of current of redox processes changes. In the case of
solutions the reduce peak current (E1red) is comparable to the
current which flows during the oxidation (even higher during
oxidation). In the case of solids the peak of oxidation is almost
unnoticed when the same scale as for reduction is used. So how
this difference and quasi-reversible nature of the reduction should
be explained? It must be remembered that the perylene derivative
is an n-type semiconductor [28]. That is why during the oxidation
Fig. 7. Selected cyclic voltammograms obtained for 2 at different concentration, GC
as working electrode, arrow shows the initial direction of current flow; sweep rate
n
= 100 mV/s, 0.1 M Bu4NPF6 in CH2Cl2.

454
M. Filapek et al. / Electrochimica Acta 216 (2016) 449–456
Fig. 8. Selected cyclic voltammograms obtained for 1; arrow shows the initial direction of current flow, GC as working electrode; c = 10ꢀ5 mol/dm3 (for solution), sweep rate
= 100 mV/s, 0.1 M Bu4NPF6 in CH2Cl2.
n
Fig. 9. Selected cyclic voltammograms obtained for 2. Reduction (left on GC electrode) and on various electrode (right), arrow shows the initial direction of current flow;
c = 10ꢀ5 mol/dm3 sweep rate
= 100 mV/s, 0.1 M Bu4NPF6 in CH2Cl2.
n
,
(or p-doping) after the oxidation the layer in direct contact with
the surface of the electrode the further flow of the charge is
considerably limited. However, during a similar situation when the
reduction is performed, electrons are more easily transported into
a solid. Thus, in solid state work function is rather determined
(instead of IE or HOMO level).
We have also investigated the influence of scan rate. As can be
seen at Fig. 10 at higher scan rate band gap determined from peak
onset is slightly lowered. On the other hand, at higher scan rate
peak onset is more pronounced (especially when we look at
oxidation for 2) (similar as observed by C. Yang et al. [29]).
Previously, we have also reported in details the influence of pH
[30], type of acid [31] or nanoparticles [32] on optical and
electrochemical properties. Additionally, very recently M. Pasini
et al. have shown that the CV response may be also affected by the
used electrolyte [33].
Finally, we have compared experimental data with DFT
calculations (see Table 1) using equation:
IE(eV) = |e|(Eox(onset) + 5.1) and EA(eV) = ꢀ|e|(Ered(onset) + 5.1) [7]
As can be seen in Table 2 in case of 1 HOMO is laying on both
aromatic moieties with small contribution from acetylene group.
On the other hand, LUMO is located only at fluorine and acetylene
part of molecule. For 2 both HOMO and LUMO are located
exclusively on polycyclic aromatic moieties. However, in both cases
H/L transition is allowed.
In both cases band gap determined for diluted solutions by UV–
vis spectroscopy is higher than those from cyclic voltammetry
technique (230 mV for 1 and 160 mV for 2). It is probably caused by
previously mentioned additional hole-electron binding energy.
Additionally, in 2 HOMO and LUMO are located in the same
Fig. 10. Cyclic voltammograms obtained for 2 at different scan rate (as solid on GC
electrode), arrow shows the initial direction of current flow; c = 10ꢀ5 mol/dm3; 0.1 M
Bu4NPF6 in CH2Cl2.

M. Filapek et al. / Electrochimica Acta 216 (2016) 449–456
455
Table 2
HOMO, LUMO contours of the 1 and 2 compounds.
HOMO
LUMO
1
E = ꢀ5.79 eV
E = ꢀ1.86 eV
2
E = ꢀ6.14 eV
E = ꢀ3.68 eV
molecule region, while for 1 they are partially separated ꢀ it could
causing difficulties in electron transfer (some energy is consumed
for electron transport within molecule). However, both band gap
values determined experimentally are significantly lower than
calculated from HOMO(DFT) ꢀ LUMO(DFT), which is a common
observation [1,25]. It is worth noting when we use |IE|-|EA|
formula, is in the better agreement with experimentally one (see
Table 1). But still, calculation in each case lowered the energy level
for about 0.3 eV approximately, what is also commonly obtained
result [28,34,35]. Thus, DFT is a helpful tool, allowing deeper
photophysical properties understanding, however they must be
used with caution.
Appendix A. Supplementary data
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.
electacta.2016.09.046.
References
[1] M.C. Ruiz Delgado, E.-G. Kim, D.A. da Silva Filho, J.-L. Bredas, Tuning the
Charge-Transport Parameters of Perylene Diimide Single Crystals via End and/
or Core Functionalization: A Density Functional Theory Investigation, J. Am.
Chem. Soc. 132 (2010) 3375–3387, doi:http://dx.doi.org/10.1021/ja908173x.
[2] A. Tomkeviciene, J.V. Grazulevicius, D. Volyniuk, V. Jankauskas, G. Sini,
Structure?properties relationship of carbazole and fluorene hybrid trimers:
experimental and theoretical approaches, Phys Chem Chem Phys 16 (2014)
13932–13942, doi:http://dx.doi.org/10.1039/C4CP00302K.
4. CONCLUSIONS
[3] N. Sakai, J. Mareda, E. Vauthey, S. Matile, Core-substituted
naphthalenediimides, Chem. Commun. 46 (2010) 4225–4237, doi:http://dx.
doi.org/10.1039/C0CC00078G.
[4] M. Grucela-Zajac, M. Filapek, L. Skorka, J. Gasiorowski, E.D. Glowacki, H.
Neugebauer, E. Schab-Balcerzak, Thermal, optical, electrochemical, and
electrochromic characteristics of novel polyimides bearing the Acridine
Yellow moiety, Mater, Chem. Phys. 137 (2012) 221–234, doi:http://dx.doi.org/
10.1016/j.matchemphys.2012.09.011.
[5] S.S. Naghavi, T. Gruhn, V. Alijani, G.H. Fecher, C. Felser, K. Medjanik, D.
Kutnyakhov, S.A. Nepijko, G. Schönhense, R. Rieger, M. Baumgarten, K. Müllen,
Theoretical study of new acceptor and donor molecules based on polycyclic
aromatic hydrocarbons, J. Mol. Spectrosc. 265 (2011) 95–101, doi:http://dx.doi.
org/10.1016/j.jms.2010.12.004.
[6] C. Arantes, M. Scholz, R. Schmidt, V. Dehm, M.L.M. Rocco, A. Schöll, F. Reinert, F.
Würthner, Comparative analysis of the energy levels of planar and core-
twisted perylene bisimides in solution and solid state by UV/VIS, CV, and UPS/
IPES, Appl Phys A 108 (2012) 629–637, doi:http://dx.doi.org/10.1007/s00339-
012-6941-3.
To summarize, we carried out NMR, CV and UV–vis spectros-
copy studies (with the aid of DFT calculations) of the selected
semiconducting PDI and carbazole derivatives. We have observed
that working electrode type influence must be considered,
especially when there is possibility of chemical surface-com-
pounds interaction or during the measurement of multistep
reduction/oxidation. Additionally, concentration of measured
species is crucial for obtained value estimation. We have also
observed that NMR spectra obtained at various concentrations can
be used to estimate intermolecular interaction degree. In the case
of small interaction (1) differences between value obtained for
solution and solid state are small. However, when interactions are
strong (2) differences are significant. Our study also revealed that
theoretically determined band gap from |IP|-|EA| equation better
fits to experimentall data than determined from HOMO-LUMO
equation. However, DFT overestimated IE and EA value, which
needs further investigation. Generally, many variables mentioned
above affected the resulted value, thus in each cases all
experimental conditions should be precisely defined in order to
be repeated or compared with data from different source.
[7] R. Rybakiewicz, P. Gawrys, D. Tsikritzis, K. Emmanouil, S. Kennou, M. Zagorska,
A. Pron, Electronic properties of semiconducting naphthalene bisimide
derivatives—Ultraviolet photoelectron spectroscopy versus electrochemistry,
Electrochim. Acta 96 (2013) 13–17, doi:http://dx.doi.org/10.1016/j.
electacta.2013.02.041.
[8] J.-L. Bredas, Mind the gap!, Mater. Horiz. 1 (2014) 17–19, doi:http://dx.doi.org/
10.1039/C3MH00098B.
[9] P. Bujak, I. Kulszewicz-Bajer, M. Zagorska, V. Maurel, I. Wielgus, A. Pron,
Polymers for electronics and spintronics, Chem. Soc. Rev. 42 (2013) 8895–
8999, doi:http://dx.doi.org/10.1039/C3CS60257E.
[10] A. Pron, P. Gawrys, M. Zagorska, D. Djurado, R. Demadrille, Electroactive
materials for organic electronics: preparation strategies, structural aspects
and characterization techniques, Chem, Soc. Rev. 39 (2010) 2577–2632, doi:
http://dx.doi.org/10.1039/B907999H.
[11] C.M. Cardona, W. Li, A.E. Kaifer, D. Stockdale, G.C. Bazan, Electrochemical
Considerations for Determining Absolute Frontier Orbital Energy Levels of
Conjugated Polymers for Solar Cell Applications, Adv. Mater. 23 (2011) 2367–
2371, doi:http://dx.doi.org/10.1002/adma.201004554.
Acknowledgments
This work was supported by National Science Centre of Poland
Grant No. DEC-2015/17/N/ST5/03892 and NCBiR Grant No. PBS2/
A5/40/2014. Calculations have been carried out in Wroclaw Centre
for Networking and Supercomputing (http://www.wcss.pl), grant
No. 18.
[12] W. Yue, Y. Zhao, H. Tian, D. Song, Z. Xie, D. Yan, Y. Geng, F. Wang, Poly
(oligothiophene-alt-benzothiadiazole)s: Tuning the Structures of

456
M. Filapek et al. / Electrochimica Acta 216 (2016) 449–456
Oligothiophene Units toward High-Mobility Black Conjugated Polymers,
Macromolecules 42 (2009) 6510–6518, doi:http://dx.doi.org/10.1021/
ma900906t.
[26] A.O. Aleshinloye, J.B. Bodapati, H. Icil, Synthesis, characterization, optical and
electrochemical properties of a new chiral multichromophoric system based
on perylene and naphthalene diimides, J.Photoch. Photobio. A 300 (2015) 27–
37, doi:http://dx.doi.org/10.1016/j.jphotochem.2014.12.007.
[27] E. Kozma, D. Kotowski, M. Catellani, S. Luzzati, A. Famulari, F. Bertini, Synthesis
and characterization of new electron acceptor perylene diimide molecules for
photovoltaic applications, Dyes Pigments 99 (2013) 329–338, doi:http://dx.
doi.org/10.1016/j.dyepig.2013.05.011.
[28] J. Vura-Weis, M.A. Ratner, M.R. Wasielewski, Geometry and Electronic
Coupling in Perylenediimide Stacks: Mapping Structure-Charge Transport
Relationships, J. Am. Chem. Soc. 132 (2010) 1738–1739, doi:http://dx.doi.org/
10.1021/ja907761e.
[29] C. Yang, A. Tang, X. Li, K. Jiang, F. Teng, Electrochemical evaluation of the
frontier orbitals of organic dyes in aqueous electrolyte, Electrochimica Acta
102 (2013) 108–112, doi:http://dx.doi.org/10.1016/j.electacta.2013.04.005.
[30] A. Iwan, B. Boharewicz, I. Tazbir, M. Filapek, Enhanced power conversion
efficiency in bulk heterojunction solar cell based on new polyazomethine with
vinylene moieties and [66]-phenyl C61 butyric acid methyl ester by adding 10-
camphorsulfonic acid, Electrochimica Acta 159 (2015) 81–92, doi:http://dx.
doi.org/10.1016/j.electacta.2015.01.215.
[13] C.-W. Chang, H.-Y. Tsai, K.-Y. Chen, Green Perylene Bisimide Dyes: Synthesis,
Photophysical and Electrochemical Properties, Materials 7 (2014) 5488–5506,
doi:http://dx.doi.org/10.3390/ma7085488.
[14] S.H. Lee, T. Nakamura, T. Tsutsui, Synthesis and Characterization of Oligo(9 9-
dihexyl-2,7-fluorene ethynylene)s: For Application as Blue Light-Emitting
Diode, Org. Lett. 3 (2001) 2005–2007, doi:http://dx.doi.org/10.1021/ol010069r.
[15] X. Ma, E.A. Azeem, X. Liu, Y. Cheng, C. Zhu, Synthesis and tunable chiroptical
properties of chiral BODIPY-based D?P?A conjugated polymers, J Mater Chem
C 2 (2014) 1076–1084, doi:http://dx.doi.org/10.1039/C3TC32029D.
[16] Y. Lin, Y. Chen, T.-L. Ye, Z.-K. Chen, Y.-F. Dai, D.-G. Ma, Oligofluorene-based
push-pull type functional materials for blue light-emitting diodes, J. Photoch.
Photobio. A 230 (2012) 55–64, doi:http://dx.doi.org/10.1016/j.
jphotochem.2011.12.008.
[17] Y. Wang, L. Zhang, G. Zhang, Y. Wu, S. Wu, J. Yu, L. Wang, A new colorimetric and
fluorescent bifunctional probe for Cu2+and Fꢀ ions based on perylene bisimide
derivatives, Tetrahedron Lett. 55 (2014) 3218–3222, doi:http://dx.doi.org/
10.1016/j.tetlet.2014.03.137.
[18] W. Wang, L.-S. Li, G. Helms, H.-H. Zhou, A.D.Q. Li, To Fold or to Assemble?, J, Am.
Chem. Soc. 125 (2003) 1120–1121, doi:http://dx.doi.org/10.1021/ja027186h.
[19] P.-O. Schwartz, L. Biniek, E. Zaborova, B. Heinrich, M. Brinkmann, N. Leclerc, S.
Méry, Perylenediimide-Based Donor–Acceptor Dyads and Triads: Impact of
Molecular Architecture on Self-Assembling Properties, J. Am. Chem. Soc. 136
(2014) 5981–5992, doi:http://dx.doi.org/10.1021/ja4129108.
[31] A. Iwan, M. Palewicz, I. Tazbir, B. Boharewicz, R. Pietruszka, M. Filapek, J.
Wojtkiewicz, B.S. Witkowski, F. Granek, M. Godlewski, Influence of ZnO:Al,
MoO3 and PEDOT:PSS on efficiency in standard and inverted polymer solar
cells based on polyazomethine and poly(3-hexylthiophene), Electrochimica
Acta 191 (2016) 784–794, doi:http://dx.doi.org/10.1016/j.
electacta.2016.01.107.
[20] M. Grucela-Zajac, M. Filapek, L. Skorka, K. Bijak, K. Smolarek, S. Mackowski, E.
Schab-Balcerzak, Photophysical, electrochemical and thermal properties of
new (co)polyimides incorporating oxadiazole moieties, Synthetic Met. 188
(2014) 161–174, doi:http://dx.doi.org/10.1016/j.synthmet.2013.12.012.
[21] Y.-C. Kung, S.-H. Hsiao, Solution-processable, high-Tg, ambipolar polyimide
electrochromics bearing pyrenylamine units, J.Mater. Chem. 21 (2011) 1746–
1754, doi:http://dx.doi.org/10.1039/C0JM02642E.
[22] K. Bijak, M. Filapek, M. Wiacek, H. Janeczek, M. Grucela, K. Smolarek, S.
Mackowski Mater, Chem. Phys. 171 (2016) 97–108, doi:http://dx.doi.org/
10.1016/j.matchemphys.2015.12.005c.
[32] A. Iwan, B. Boharewicz, I. Tazbir, M. Filapek, K.P. Korona, P. Wróbel, T. Stefaniuk,
A. Ciesielski, J. Wojtkiewicz, A.A. Wronkowska, A. Wronkowski, B.
Zboromirska-Wnukiewicz, S. Grankowska-Ciechanowicz, M. Kaminska, T.
Szoplik, How do 10-camphorsulfonic acid, silver or aluminum nanoparticles
influence optical, electrochemical, electrochromic and photovoltaic properties
of air and thermally stable triphenylamine-based polyazomethine with
carbazole moieties? Electrochimica Acta 185 (2015) 198–210, doi:http://dx.
doi.org/10.1016/j.electacta.2015.10.110.
[33] M. Pasini, B. Vercellib, G. Zotti, A. Berlin, Solid-state Effects in the
Cyclovoltammetric HOMO-LUMO Determination: The case of
[23] M. Grucela-Zajac, K. Bijak, S. Kula, M. Filapek, M. Wiacek, H. Janeczek, L. Skorka,
J. Gasiorowski, K. Hingerl, N.S. Sariciftci, N. Nosidlak, G. Lewinska, J. Sanetra, E.
Schab-Balcerzak, (Photo)physical Properties of New Molecular Glasses End-
Capped with Thiophene Rings Composed of Diimide and Imine Units, J. Chem.
Phys.C 118 (2014) 13070–13086, doi:http://dx.doi.org/10.1021/jp501168b.
[24] P. Gawrys, A. Zoltowska, M. Zagorska, A. Pron, Alternating copolymers of
oligoarylenes and naphthalene bisimides as low band gap semiconductors:
Synthesis, electrochemical and spectroelectrochemical behavior,
Electrochimica Acta 56 (2011) 10464–10472, doi:http://dx.doi.org/10.1016/j.
electacta.2011.03.016.
Dinitrophenyl-hydrazone a, v-substituted Oligothiophenes, Electrochimica
Acta 193 (2016) 261–267, doi:http://dx.doi.org/10.1016/j.
electacta.2016.02.035.
[34] A. Szłapa, S. Kula, U. Błaszkiewicz, M. Grucela, E. Schab-Balcerzak, M. Filapek,
Simple donor-P-acceptor derivatives exhibiting aggregation-induced
emission characteristics for use as emitting layer in OLED, Dyes and Pigments
129 (2016) 80–89, doi:http://dx.doi.org/10.1016/j.dyepig.2016.02.002.
[35] S.A. Kula Szlapa, J.G. Malecki, A. Maron, M. Matussek, E. Schab-Balcerzak, M.
Siwy, M. Domanski, M. Sojka, W. Danikiewicz, S. Krompiec, M. Filapek,
Synthesis and photophysical properties of novel multisubstituted benzene and
[25] M. Czichy, P. Wagner, L. Grza˛dziel, M. Krzywiecki, A. Szwajca, M. Łapkowski, J.
naphthalene derivatives with high 2D-
(2015) 118–128, doi:http://dx.doi.org/10.1016/j.optmat.2015.07.011.
P-conjugation, Optical Materials 47
_
Zak, D.L. Officer, Electrochemical and photoelectronic studies on C60-
pyrrolidine- functionalised poly(terthiophene), Electrochimica Acta 141
(2014) 51–60, doi:http://dx.doi.org/10.1016/j.electacta.2014.06.100.