Photon assisted-inversion of majority charge carriers in molecular semiconductor-based organic heterojunctions

Article abstract of DOI:10.1039/d0tc05828a

Ambipolar molecular materials hold great promise as a building block of next generation highly efficient, less complex and low cost electronic devices. In this endeavor, the present work reports the fabrication of organic heterojunction devices based on halogenated copper phthalocyanine (CuPc) and lutetium bisphthalocyanine (LuPc2) bilayers, investigates their structural and electrical properties and probes the ambipolar behavior by ammonia sensing. Microstructural analysis of the heterostructure thin films revealed compact and semicrystalline organization, depending on the number of halogen substituents in CuPc. The heterojunction devices reveal non-linear I(V) characteristics associated with an interfacial energy barrier which is tuned by the number and type of halogen substituents. The behavior of the devices towards NH3 exposure revealed n-type, p-type and ambipolar transport depending on 16, 0 and 8 fluorine substituents, respectively, present in the CuPc component of the heterostructure, demonstrating a correlation between the electronic effects of substituents and the thin film electrical properties. The trigger of the ambipolar transport regime in the heterojunction device is the visible light, such that the device conduction channel is dominated by holes in the dark and electrons under illumination. In depth charge transport studies by impedance spectroscopy explain the device electrical behavior and highlight the important role played by the interface in the heterostructures. Different charge transport parameters are extracted by fitting the experimental Nyquist curves, analysis of those confirms the presence of bulk and interfacial transport and provides an interpretation of the observed ambipolarity.

Full text of DOI:10.1039/d0tc05828a

Journal of
Materials Chemistry C
View Article Online
View Journal | View Issue
PAPER
Photon assisted-inversion of majority charge
carriers in molecular semiconductor-based
organic heterojunctions†
Cite this: J. Mater. Chem. C, 2021,
, 5008
9
a
a
b
Gracia Loma Kikobo, Abhishek Kumar, * Vaibhav Vibhu,
ac
a
a
a
Seydou Ouedraogo, Alix Deshotel, Mickael Mateos, ‡ Rita Meunier-Prest
¨
a
and Marcel Bouvet
*
Ambipolar molecular materials hold great promise as a building block of next generation highly efficient,
less complex and low cost electronic devices. In this endeavor, the present work reports the fabrication
of organic heterojunction devices based on halogenated copper phthalocyanine (CuPc) and lutetium
2
bisphthalocyanine (LuPc ) bilayers, investigates their structural and electrical properties and probes the
ambipolar behavior by ammonia sensing. Microstructural analysis of the heterostructure thin films
revealed compact and semicrystalline organization, depending on the number of halogen substituents in
CuPc. The heterojunction devices reveal non-linear I(V) characteristics associated with an interfacial
energy barrier which is tuned by the number and type of halogen substituents. The behavior of the
devices towards NH
3
exposure revealed n-type, p-type and ambipolar transport depending on 16, 0 and
8
fluorine substituents, respectively, present in the CuPc component of the heterostructure,
demonstrating a correlation between the electronic effects of substituents and the thin film electrical
properties. The trigger of the ambipolar transport regime in the heterojunction device is the visible light,
such that the device conduction channel is dominated by holes in the dark and electrons under
illumination. In depth charge transport studies by impedance spectroscopy explain the device electrical
behavior and highlight the important role played by the interface in the heterostructures. Different
charge transport parameters are extracted by fitting the experimental Nyquist curves, analysis of those
confirms the presence of bulk and interfacial transport and provides an interpretation of the observed
ambipolarity.
Received 11th December 2020,
Accepted 15th March 2021
DOI: 10.1039/d0tc05828a
rsc.li/materials-c
n-type and p-type transistors can be replaced by one ambipolar
transistor, enabling simpler circuit designs, reducing the power
Introduction
2
,3
Molecular semiconductors displaying ambipolar charge dissipation and improving the switching efficiency, as well as
1
4
5
transport properties have been the research focus in recent in high density data storage, photonic memory and artificial
6
years in the field of organic electronics. Such materials have synapses. Chemical sensors are another widely researched
demonstrated high prospects in CMOS-like inverters in which area inspiring numerous studies on organic semiconductors,
but only a few deal with ambipolar carrier transport despite
a
their advantages of dual mode operation (device operating at
negative and positive bias), bias-dependent selectivity,
Institut de Chimie Mol ´e culaire de l’Universit ´e de Bourgogne (ICMUB), UMR CNRS
302, Universit ´e Bourgogne Franche-Comt ´e , 9 av. Alain Savary, 21078 Dijon
7
6
Cedex, France. E-mail: abhishek.kumar@u-bourgogne.fr,
miniaturized multiplexed detection platforms and lower
fabrication cost compared to their inorganic counterparts.
marcel.bouvet@u-bourgogne.fr
8,9
b
c
Institute of Energy and Climate Research, Fundamental Electrochemistry (IEK-9),
Forschungszentrum J u¨ lich GmbH, 52425 J u¨ lich, Germany
In particular, organic macrocycles like metal phthalocyanines
(MPcs) are versatile molecular semiconductors in which the
Laboratoire de Chimie Mol ´e culaire et de Mat ´e riaux, Universit ´e Joseph Ki-Zerbo,
Ouagadougou, 03 BP 7021, Ouagadougou, Burkina Faso
ꢀ
+
dynamics of the charge carriers (e and h ) upon chemical
†
Electronic supplementary information (ESI) available: SEM images of all the
doping by redox gases, such as nitrogen dioxide, ammonia and
sublayers, EDX mapping of Cu(F16Pc), reliability factors of XRD data refinement
and I–V curves of CuPc/LuPc and Cu(Cl Pc)/LuPc heterojunctions and of the
LuPc resistor. See DOI: 10.1039/d0tc05828a
Permanent address: Institut de Chimie et des Mat ´e riaux Paris-Est, Universit ´e
Paris-Est Cr ´e teil Val de Marne, France.
10–12
ozone, can be exploited for chemosensing studies.
MPc
2
8
2
materials manifested a strong correlation between molecular
level electronic properties with thin film electrical behaviour and
thus the charge carriers in devices based on them can be
2
‡
5008
|
J. Mater. Chem. C, 2021, 9, 5008–5020
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry C
effectively tuned by molecular engineering by either changing Also, there are only a few reports on external triggers to invert the
1
3,14
peripheral ligands or central metal atoms.
For example, majority charge carriers in gas sensors. In OFET-based gas
chemiresistors based on p-type transition metal MPc displayed sensors, gate voltage modulation and chemical doping from
ꢀ
+
unipolar carrier transport in redox gas sensing, depicting current analyte gases are commonly used to switch between e or h
7
increase under electron acceptor gases and current decrease conduction channels. An ambipolar OFET operating alternatively
1
5,16
under donor gases.
On the other hand, chemiresistors based in n-channel and p-channel modes based on diketopyrrole was
on substituted lanthanide complexes of double or triple decker reported to selectively detect different isomers of xylene vapours
29
phthalocyanine exhibited current increase under NO
for NH exposure, clearly demonstrating an ambipolar humidity (rh) in the sensor’s ambient environment as an external
chemosensing regime.
2
as well as depending on the applied gate voltage. The role of relative
3
1
7,18
The carrier transport modulation in trigger for majority carrier inversion was demonstrated in our
MPc by changing the peripheral ligands, among fluoro, alkoxy studies on a MPcs bilayer heterojunction, which turns from a
and ester, and its implication on NH sensing were exemplified p-type to a n-type device on changing the rh value from 30% to
3
1
9
30
in our recent works based on an organic heterojunction device.
Moreover, the phenoxy and naphthoxy substitution in europium trigger to reverse conduction channel polarity in an inorganic
70%. Elsewhere, the chemical doping from NO2 acted as a
31
triple-decker phthalocyanine complexes generated very high, p–n junction diode based on tungsten disulfide. Despite
ꢀ
+
2
ꢀ1 ꢀ1
similar e and h carrier mobilities, both above 1 cm V
s
these advancements, the availability of a universal trigger for
in OFET devices and thus demonstrated air stable ambipolar majority carrier inversion, which can be applied in any device
2
0,21
carrier transport.
However, it must be noted that all the configuration with a minimum instrumentational complexity, is a
materials having both p- and n-channels did not reveal ambipolar pertaining challenge, limiting the wide scale use of semiconducting
ꢀ
+
sensing behaviour because of a large difference of e and h
concentrations or the lack of a suitable trigger (an external
materials with bipolar carrier transport in gas sensors.
Herein, MPcs based bilayer heterojunctions consisting of
stimulus) to shift the equilibrium concentration of either charge LuPc
carriers (e and h ). An example is a europium triple decker incorporated in an original device configuration, recently
2
as a top layer and substituted CuPc as a sublayer
ꢀ
+
3
2
phthalocyanine complex bearing thiohexyl groups, Eu
in chemiresistor configuration, exhibited n-type polarity with ambipolar NH
2
[Pc(S-hexyl)
8
]
3
,
developed and patented by us, have been investigated for
sensing by using visible light as an external
3
a current decrease under NO
2
, but was insensitive to NH
3
trigger for device polarity inversion. Our strategies involve
2
2
and H S.
peripheral substitution of variable numbers of fluorine or
2
ꢀ
Therefore, to achieve an ambipolar transport regime in chlorine substituents in CuPc to achieve equilibrium e and
organic electronic devices including gas sensors, there are
two main challenges; tuning the e and h concentrations in effects in order to enhance the bipolar carrier mobility and
+
h
concentrations, optimizing the organic heterojunction
ꢀ
+
ꢀ
+
the device near equilibrium to a sufficiently large level and an using low-energy photons to switch between e and h in the
efficient trigger for the majority charge carrier inversion. To device conduction channels. It is well known that CuPc and its
ꢀ
+
obtain high e and h densities in equilibrium in a molecular perfluorinated analogue are p-type and n-type semiconductors,
3
3
semiconductor like MPc, an appropriate combination of respectively, in air. Therefore, partially fluorinated CuPc, like
electron accepting and electron donating substituents are Cu(F Pc), has a high prospect of possessing bipolar carrier
8
3
4
grafted and metal centers of different electron affinities are mobility. Accordingly, heterojunction devices based on CuPc,
used. However, by adopting such a strategy, only a few Cu(F Pc), Cu(Cl Pc) and Cu(F Pc) as a sublayer and LuPc have
8
8
16
2
2 2 3
lanthanide phthalocyanines (LuPc and Eu Pc ) were synthesized been investigated to probe the evolution of ambipolar transport.
ꢀ
+
possessing sufficiently large e and h densities in equilibrium Different MPc thin films and related heterostructures have been
comparable to unipolar MPcs, and a narrow band gap such that characterized by XRD, EDX and SEM to investigate the structure,
their HOMO and LUMO levels aligned with the electrodes’ work- chemical purity and morphologies of the thin films and the
function to facilitate charge injection and reception by the nature of interfaces. Finally, in depth charge transport studies by
electrodes. Since the majority of MPcs are wide band gap semi- impedance spectroscopy in different heterostructures have been
conductors with low carrier mobility, a bilayer of n-type and p-type performed to assess the important role played by the interface,
MPc heterostructure is an alternative in which each component’s which has been further correlated with thin film structures and
workfunction could align with the electrode’s workfunction, electronic effects of substituents in MPc.
facilitating low-energy charge transfer at the electrode–MPc
23,24
interface.
Moreover, the associated organic heterojunction
effects enhance the carrier mobility at the interface by the
Results and discussion
ꢀ
+
accumulation of e and h , thereby filling the traps and
5,26 Heterojunction device structure and characterization
2
transforming discrete hopping into a band like transport.
In fact, the bilayer thin films of CuPc/Cu(F Pc) heterojunctions The device configuration (Scheme 1) consisted in a bilayer
16
have been extensively investigated in OFET like configuration and heterostructure of MPc sequentially coated on ITO electrodes
have shown an 8-order of magnitude higher conductivity than such that the sublayer is a relatively poor conducting CuPc or
2
7,28
individual MPc films and ambipolar carrier transport,
its halogenated derivatives while the top layer is the high
, which is an intrinsic radical
but none of these were investigated for gas sensing applications. conducting double decker LuPc
2
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 5008–5020 | 5009

View Article Online
Journal of Materials Chemistry C
Paper
films (50 nm) are shown in Fig. 1, displaying distinct
morphologies as a function of halide substituents. In general,
surfaces are dense with poor organization in which two different
phases comprising a flat surface and grains decorating the flat
surface can be noticed, which can be associated with the
amorphous and crystalline packing, respectively, of the MPc
molecules. However, the grain size changes with the degree of
halogenation in the CuPc. Perfluorinated CuPc depicts the bigger
8
grains as compared to CuPc, while Cu(F Pc) presents almost a flat
morphology and the grains are too small to be visualized on the
micrometric scale. In fact, such trends are in agreement with
previous work reporting larger crystallites in perfluorinated phtha-
2
Scheme 1 Scheme of organic heterojunction device having LuPc as the
top layer and Cu(X
to F, Cl and n = 0, 8 and 16.
n
Pc) as the sublayer coated on ITO electrodes. X refers locyanines compared to their unsubstituted counterpart, which is
34,38
attributed to stronger edge-on intermolecular interactions.
Interestingly, Cu(Cl Pc) thin films present a different arrange-
ment, consisting of relatively larger flacks instead of grains
semiconductor, matching its carriers density with the conventional (Fig. 1e), which are of variable sizes. Such surface evolution points
8
35
inorganic semiconductors. Such a large gradient of carrier to random agglomeration of the macrocycles which is also
concentration between the two layers facilitates higher charge consistent with previous studies mentioning that higher than
39
injection across the interface and thus enhanced accumulation tetrachlorination in CuPc produces mainly an amorphous film.
of dipoles at the interface because of the organic heterojunction LuPc thin films also present a highly dense structure over which
2
36,37
effects.
Additionally, low conductivity of the sublayer a few bigger size grains are randomly distributed. The dense
in thin film is advantageous for the hetero-
minimizes the bulk transport between the electrodes and favors packing of LuPc
2
the transport along the interface. On the other hand, high structure device because it is used as the top layer and thus acts as
conductivity of the top layer is preferred for redox gas sensing a kinetic barrier against oxygen and humidity diffusion into the
4
0
applications because gas molecules interact on the sensors top sublayer and thus protects the sublayer from oxidation.
ꢀ
+
surface and transfer e or h , which must be quickly transported Nanometric images obtained at higher magnifications of different
to the interface for a faster response. Four different devices based CuPc derivatives are shown in Fig. S1 (ESI†), in which crystallites
on relatively poor conducting CuPc, Cu(F16Pc), Cu(F
Cu(Cl Pc) sublayers were fabricated.
SEM and EDX studies of MPc films. Thin film microstructures and grains spaced far apart are noticed in Cu(F16Pc). Even the
and morphologies of the sub layers and the top layers strongly Cu(F Pc) films exhibit a few randomly distributed grains while
influence the heterostructure interface and accordingly the nanometer sized flacks are visualized in Cu(Cl Pc) films.
electrical properties of the heterojunction, which were studied Since MPc were coated through high vacuum sublimation, it
by SEM and XRD. The SEM micrographs of different MPc thin becomes imperative to assess the chemical purity of the film
8
Pc) and are more clearly seen. Smaller, closely spaced and homogeneously
8
distributed grains are evident in CuPc films while larger grains
8
8
Fig. 1 SEM images of thin films of Cu(F
8
Pc) (a), Cu(F16Pc) (b), CuPc (c), LuPc
2
(d) and Cu(Cl
8
Pc) (e) coated on glass substrate at room temperature.
5010
|
J. Mater. Chem. C, 2021, 9, 5008–5020
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry C
4
1
because high temperature processing may be destructive to the MPc, but small pockets of locally organized structures are
macrocycles stability. Therefore, EDX spectra of different MPc present also corroborated from the SEM images. Bilayer films
films were recorded and are shown in Fig. 2a. In the EDX mainly show the XRD peaks of the sublayer because the top
spectrum of each MPc film, characteristic peaks associated with layer is relatively low crystalline and adopts similar molecular
4
2
constituent elements are observed (marked in each spectrum). orientations as that of the sublayer. CuPc displays a diffrac-
For example, spectral peaks corresponding to the Ka band of tion peak at 6.971 associated with the (200) plane in its b
carbon (0.27 eV), nitrogen (0.39 eV) and fluorine (0.67 eV) polymorph which shifts to 6.781 in the CuPc/LuPc
atoms can be noticed. Additionally, the La band of copper well as increased intensity, indicating an enlargement of the
0.93 eV) and M band of lutetium (1.59 eV) are also observed. It crystallites size which is further supported by the lowering in
2
bilayer as
(
is also evident that the F/C peak ratio is larger in Cu(F16Pc) than FWHM (0.33 to 0.21). In contrast, XRD peaks of heterostructures
in Cu(F Pc), which is expected because of higher fluorine based on Cu(F Pc) (at 6.431) and Cu(F Pc) (at 6.431) exhibit a
8
8
16
content. These results confirm that there is no damage to the decrease in the peak intensity and an increase in FWHM (0.35 to
MPc during high temperature film processing. To further 0.41 for the former and 0.25 to 0.48 for the latter), indicating a
establish the chemical purity of the coated MPc film over a decrease in the crystallite sizes from the respective sublayer.
large area, EDX mapping was performed. As depicted in Fig. S2 On the other hand, Cu(Cl
ESI†), a homogeneous distribution of macrocycles is noticed amorphous heterostructure, which is also in agreement with
which is expected for the thin film prepared by high vacuum the SEM images of the sublayer and the previous structural
8 2
Pc)/LuPc presents an almost
(
3
9
sublimation.
studies.
To further understand the MPc molecular orientations in
Notably, in all these EDX spectra (except LuPc ), a peak
2
associated with the Ka band of oxygen is observed, which can the thin film, Fullprof refinement was performed on the
be attributed to ITO substrate and atmospheric oxygen experimental data and different phases were identified through
adsorption.
matching with the JCPDS-ICDD database. Fig. 3 depicts the
XRD studies of MPc films. The crystallinity of the MPc thin XRD patterns and their fitting over a large angle range, exhibiting
film microstructures was assessed by XRD, performed at room a sharp peak around 6–71 and a broad peak around 24–251. For
temperature. The diffraction pattern of different bilayer hetero- CuPc and its halogenated derivatives, monoclinic and triclinic
structures are compared with the constituent MPc in Fig. 2b, unit cells are proposed in P21/n and P1% space groups (SG),
exhibiting superposition of the characteristic reflections of MPc respectively (reliability factor given in the ESI†). Different lattice
at a low diffraction angle 6–71. Notably, films have low crystal- parameters including a, b, c, a, b, and g and unit cell volumes were
linity globally as expected for the film thickness r100 nm for estimated (Table 1).
Fig. 2 EDX spectra of different MPc films coated on glass substrate (a). Comparison of XRD pattern of bilayer films with the individual films (b).
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 5008–5020 | 5011

View Article Online
Journal of Materials Chemistry C
Paper
interface creating an energy barrier. Based on the simple
scheme given in the inset of Fig. 4a, evolution of I(V) curves
can be formalized as follows. Because of the poor conductivity
of the sublayer, the current remains small at lower bias as
charges injected from the electrode experience high resistance.
At higher bias, injected charges from the electrode have enough
energy to overcome the bulk resistance of the sublayer and
reach the highly conducting interface, resulting in higher
current. The apparent energy barrier
th
U , deduced by
extrapolating the tangent to the curve at high bias to abscissa
is estimated as 6.6 V and 6.3 V for Cu(F Pc)/LuPc and 1.6 V and
8
2
1
.0 V for Cu(F Pc)/LuPc , in the dark and under LED light,
16 2
respectively. It is evident that the energy barrier is much higher
2
in the Cu(F Pc)/LuPc heterojunction than Cu(F16Pc)/LuPc ,
8
2
which can be attributed to larger crystalline grains present in
the latter as revealed by the XRD and SEM. Moreover higher Uth
Fig. 3 XRD pattern of different MPc thin films recorded at room
of the Cu(F
the large difference in the energies of frontiers orbitals of LuPc
and Cu(F Pc), which makes it difficult for charges hopping
8 2
Pc)/LuPc heterojunction is also in agreement with
temperature and their refinement by Fullprof.
2
8
4
6
In these arrangements, molecules are staked in column
along the b-axis through strong p–p intermolecular interactions
in which the magnitude of b represents the distance between
two adjacent molecules in the column. Based on b B4.78 Å
for CuPc unit cells, it is evident that the thin film adopts
from LuPc orbitals to Cu(F Pc) at the interface. On the other
2
8
hand, the LUMO level of Cu(F Pc) aligns well with the SOMO
1
6
2
of LuPc , facilitating easier charge hopping at the interface.
Notably, in both heterojunctions, Uth decreases upon light
exposure, indicating that charge injection from the electrode
to the interface becomes easier. Such behavior is explained by
4
3
b-polymorph. Higher b values in halogenated derivatives of
CuPc are attributed to the presence of highly electronegative
peripheral substituents, accumulating the negative charge at
the periphery of the macrocycles which causes intermolecular
repulsion. The lattice parameters a, c and b represent
assemblies of molecular columns in the film. A highly different
value of these parameters for CuPc from its halogenated
derivatives can be associated with the herringbone and parallel
ꢀ
+
the generation of additional free carriers (e or h ), which are
4
7
otherwise held in the shallow traps, upon photon exposure.
Accordingly, device current increases under LED exposure
compared to dark current for both devices. Similar
measurements on other heterojunctions and LuPc
2
resistor
did not result in any significant change in the current values
(I(V) curves shown in Fig. S3, ESI†).
4
3,44
assembly, respectively.
Contrary to CuPc and its derivatives,
Effects of light on the ammonia sensing properties of
LuPc thin film, which is relatively low crystalline, presents a
2
heterojunctions. The behavior of the heterojunction devices
totally different crystal structure described by orthorhombic
towards ammonia was studied by alternate exposure under
unit cells and P21 space group. Based on the lattice parameters
estimation, the b parameter is highly larger in LuPc (50.42 Å).
2
2
4
0 ppm of NH3 for 1 min and recovery under pure air for
min in the dark as well as under LED illumination. For the
Cu(F
and increases during recovery cycle in the dark (Fig. 5a)
Effects of light exposure on I–V characteristics. The electrical indicating the p-state of the device taking into account the
properties of the bilayer heterojunction devices were electron donating nature of NH
investigated by recording current in a voltage sweep from
Interestingly, device polarity turns opposite into the n-state
10 V to 10 V in the dark and under LED illumination (red under illumination, i.e. current increases during exposure and
8 2 3
Pc)/LuPc device, current decreases under NH exposure
Electrical properties of organic heterojunction devices
3
.
ꢀ
light; l
= 642 nm). The I(V) curves (Fig. 4) obtained in the decreases during the recovery cycle. Such polarity inversion
max
dark and under LED exposure exhibit a typical non-linear under photon exposure, which has never been reported so far,
4
5
behavior as expected for an organic heterojunction device,
reveals the ambipolar carrier transport regime in the Cu(F
8
Pc)/
which is attributed to accumulation of opposite charges at the LuPc
2
heterojunction. Moreover, the device baseline current
Table 1 Lattice parameters of studied MPc determined by FULLPROF refinement
a
b
c
a
b
G
Volume
Space group
Cell
CuPc
CuF Pc
CuF16Pc
14.05
3.71
4.85
8.02
10.52
4.78
16.92
13.81
28.42
14.74
8.65
90
105.02
89.93
87.87
82.01
90
90
1097.71
594.23
1393.70
1571.46
4593.78
P21/n
P1%
Monoclinic
Triclinic
Triclinic
Triclinic
Orthorhombic
8
11.70
10.14
13.70
50.42
85.17
86.65
81.99
90
83.24
87.73
80.54
90
P1%
CuCl
LuPc
8
Pc
P1%
P212121
2
5012
|
J. Mater. Chem. C, 2021, 9, 5008–5020
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 4 Normalized I(V) characteristics of Cu(F16Pc)/LuPc
2
(a) and Cu(F
8
Pc)/LuPc
2
(b) heterojunctions, in the dark (black) and under LEDs illuminations
(red). Inset in (a) represents a simplified device scheme.
under its n-state (ca. 0.42 mA) is approximately double as for the Cu(F Pc)/LuPc (Fig. 5c) heterojunction and vice versa
1
6
2
observed in its p-state (ca. 0.22 mA) operation, which can be for CuPc/LuPc2 (5d), Cu(Cl Pc)/LuPc (5e) and LuPc2 resistor
8
2
attributed to the photogeneration of additional electrons. (5f), revealing an n-type conduction channel in Cu(F16Pc)/LuPc
Notably, this phenomenon was observed in a broad range of but p-type conduction channels in other devices. Notably,
bias voltage (1–8 V), and the device exhibited n-state and p-state Cu(F16Pc)/LuPc and Cu(Cl Pc)/LuPc devices experienced
under light and in the dark, respectively, as illustrated by the enhancement in the current under visible light illumination, while
positive relative response under light and negative relative the CuPc/LuPc heterojunction and LuPc resistor are insensitive
2
2
8
2
2
2
response in the dark (Fig. 5b). Therefore, it can be inferred to light. Such behavior can be explained by the stronger hetero-
that ambipolar transport is independent of bias voltage and junction effects in Cu(F Pc)/LuPc and Cu(Cl Pc)/LuPc devices.
16
2
8
2
depends only on light exposure in the Cu(F Pc)/LuPc device.
On the other hand, CuPc/LuPc2 has very low Uth and charge
The relative response (RR) as defined in eqn (1) varied carriers are discretely distributed in the bilayer similar to the
between 0.5% and 1.5% under illumination and between LuPc resistor and lacks any significant interfacial accumulation,
3% and ꢀ7% in the dark, depending on the applied bias. thus the charges produced by photoexcitation have a negligible
The maximum RR value was estimated at 4 and 5 V in the dark effect on the device current. Therefore, the unique electronic
and further decreases at higher bias because the contribution properties of Cu(F Pc) imparted by octa-fluorination in generating
8
2
2
ꢀ
8
ꢀ
+
of device background current becomes much higher than the an optimum amount of e and h to display ambipolar carrier
changes induced by NH . In contrast, below 3 V, the RR value transport is highlighted. To further understand the electrical
3
increases due to a smaller device background current (Iair). properties of these heterojunctions and in particular ambipolar
Under illumination, the ratio between maximum and minimum behavior displayed by the Cu(F
RR values is 3, with a maximum response between 2 and 5 V. transport studies by impedance spectroscopy were performed.
We already reported such dependence of RR on the bias, in the Charge transport at the heterojunction interface. Different
dark, for an n-type unipolar bilayer heterojunction based on heterojunction devices as well as LuPc resistors were studied
by impedance spectroscopy in a broad frequency range (10 Hz–
0 MHz) and at variable DC voltage in the range of 0–10 V. The
8 2
Pc)/LuPc device, in depth charge
2
4
8
perylene derivatives as the sublayer and an LuPc
2
top layer.
1
INH ꢀ Iair
3
RRð%Þ ¼
ꢁ 100
(1)
Nyquist plots of these measurements are shown in Fig. 6,
exhibiting two depressed semicircles for the heterojunction
Iair
2
Contrary to the photon exposure, extended exposure to NH3 devices while one semicircle for the LuPc resistor. Moreover,
under a humid atmosphere turns the device permanently into the semicircles at high frequency (HF) for heterojunction
the n-state, i.e. with positive response in the dark and under devices remain unaffected by the DC voltage variation while
illumination (Fig. S4, ESI†). Finally, it was possible to come back those at low frequency (LF) get smaller with increasing bias.
to the p-type state by heating at 150 1C under primary vacuum Clearly, it can be confirmed based on these observations that
for 8 h.
an interfacial energy barrier exists in heterojunction devices,
The effect of photoexcitation and its implication on NH3 which decreases as a function of increasing DC bias. To extract
response was also studied on other heterojunction devices as more physical information from the Nyquist curves, they were
well as LuPc resistor. Contrary to the heterojunction based on modelled through an impedance circuit, describing the electrical
2
Cu(F
exposure to photons for other devices. A current increase upon properties of the heterostructures films are distributed owing
NH exposure while decrease during clean air recovery is noted to crystalline and amorphous phases, producing grain
8 3
Pc), no ambipolarity was observed in NH response under nature of heterojunction devices. Since the microscopic
3
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 5008–5020 | 5013

View Article Online
Journal of Materials Chemistry C
Paper
Fig. 5 Response of Cu(F
8
Pc)/LuPc
2 2 2 8 2 2 3
(a), Cu(F16Pc)/LuPc (c), CuPc/LuPc (d), Cu(Cl Pc)/LuPc (e) heterojunctions and LuPc resistor (f) towards NH
(20 ppm), during 1 min and 4 min of exposure and recovery cycles, respectively, at 40% rh, under a bias of 4 V, in the dark and under light. Relative
8 2
response of the Cu(F Pc)/LuPc device in the dark (black curve) and in light (red curve) as a function of bias (b).
boundaries, organizational defects and roughness, resulting in the inset of Fig. 6b. The semicircles at HF are described by CPE1
non-smooth formation of interfaces. and R1 circuits while those at LF are described by CPE2 and R2.
Such systems can be modelled by constant phase elements As semicircles at HF are independent of bias, they correspond to
4
9
(
Z
CPE), which are imperfect capacitors, for which impedance the charge transport in the bulk while the semicircles at LF change
50
i
CPE is defined by eqn (2), where Q is the non-ideal capacitance, with bias, they manifest the interfacial charge transport.
o corresponds to frequency (2pf) and a is a coefficient varying
between 0 and 1.
Based on this analogy, it can also be said that there is only
bulk charge transport in the LuPc resistor as only one semi-
2
circle independent of bias is present in its Nyquist plots. To
further correlate Nyquist semicircles with the charge transport
in the device, a parameters should be analyzed. a represents the
1
ZCPE ¼
(2)
i
Qi ꢁ ð joÞai
The value of a at 0 and 1 corresponds to the CPE element as an nonideality of the capacitive element such that a = 1 indicates
ideal resistor and capacitor, respectively. Thus, the bilayer an ideal capacitor or homogeneous R–CPE elements (single
heterojunction can be modelled as a summation of two Ri–CPEi time constant). As semicircles get more depressed, a also
elements in series represented by the equivalent circuit shown in decreases and accordingly time constants in the distributed
5014 | J. Mater. Chem. C, 2021, 9, 5008–5020
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry C
Fig. 6 Nyquist plots of Cu(F
8 2 8 2 2 2 8 2
Pc)/LuPc – n type (a), Cu(F Pc)/LuPc – p type (b), Cu(F16Pc)/LuPc (c), CuPc/LuPc (d), and Cu(Cl Pc)/LuPc
(
e) heterojunctions and LuPc resistor (f), recorded at variable DC voltage in the range of 0–10 V.
2
elements increase, indicating enhancement of heterogeneity in multitudes of interconnected R–CPE circuits. The LF semi-
the R–CPE elements. In other words, capacitive contribution in circles collapsed as the bias increased. This occurs when the
the device electrical output decreases and conductive contribution bias becomes higher than the threshold voltage deduced from
increases at a constant resistance. Based on the depression in the the I(V) curves. Notably, the LF semicircles in the CuPc/LuPc
2
semicircles (estimated by Im(Z)/Re(Z) at maximum Im(Z); values device is very small compared to those at HF, which also
plotted in Fig. S5, ESI†), the ease of bulk charge transport is thus disappear at higher bias and the Nyquist curves become similar
in the order of Cu(F
8
Pc)/LuPc
2
– n-type 4 Cu(Cl
8
Pc)/LuPc
2
4
4
2
to those of LuPc . Such behaviour indicates the small interface
effects which is completely removed at higher bias because of
effectively overcoming the low energy barrier. This is attributed
Cu(F16Pc)/LuPc
LuPc2.
2
4 Cu(F Pc)/LuPc
8
2
– p-type 4 CuPc/LuPc
2
On the other hand, depression in the semicircles at LF to the similar energies of HOMO and SOMO orbitals of CuPc
increases with increasing bias (Fig. S6, ESI†), indicating a and LuPc , respectively, resulting in facile interorbital charge
2
5
0
decrease in the a values and thus enhancement in the R–CPE hopping. Different circuit parameters such as the resistances
elements heterogeneity in the device. In other words, at higher R1 and R2, a1 and a2 and the effective capacitances Ceff1
bias, injected charges from the electrodes can overcome the and Ceff2 (eqn (3)) were estimated through fitting of the
capacitive barrier and participate in the conduction through experimental Nyquist curves and their variation with the bias
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 5008–5020 | 5015

View Article Online
Journal of Materials Chemistry C
Paper
are shown in Fig. 7.
interfacial effects on the conductivity of the different materials.
8 2
R2 is very high for Cu(F Pc)/LuPc devices at low bias which is
in agreement with high Uth values determined from I–V curves,
which decreases with increasing bias, indicating that the inter-
1
ai
1
ai
ꢀ
1
Ceffi ¼ Qi ꢂ Ri
(3)
It is evident from Fig. 7a, c and e that most of the circuit facial energy barrier is overcome. In contrast, R2 remains lower
parameters associated with HF semicircles are almost than R1 for the CuPc heterojunction, indicating negligible
independent of the bias, which is expected for the bulk charge interfacial properties in this device, as expected from the linear
transport. Interestingly, R1 of Cu(F
8
Pc)/LuPc
2
– n-type is ca. I–V curves. It is also worthwhile noting that, for the Cu(F
– p-type, highlighting LuPc heterojunction, bulk conductivity is higher in its n-mode
Pc). while interfacial transport is dominant in its p-mode.
The a1 value for all heterojunction devices remains close to 1
8
Pc)/
40 times lower than R1 of Cu(F
8
Pc)/LuPc
2
2
ꢀ
the significant photogeneration of e in the bulk of Cu(F
The parameter R2 is very crucial to characterize the interfacial
8
energy barrier in the heterojunction devices. As depicted in and experiences a slight increase with rise in the bias voltage,
Fig. 7b, except for the CuPc heterojunction, R2 is always much indicating that the R–CPE elements are close to an ideal
larger than R1 in agreement with the prevalence of the capacitor. Therefore, heterogeneity in the systems is low and
Fig. 7 Variations of equivalent circuit resistances, R1 (a) and R2 (b), a1 (c) and a2 (d) and effective capacitances Ceff1 (e) and Ceff2 (f), as a function of bias
voltage (0–5 V) for different heterojunction devices.
5016
|
J. Mater. Chem. C, 2021, 9, 5008–5020
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry C
decreases at higher bias. Such transport is expected in the bulk
region of the heterojunction, where microstructures remain
homogeneous. In contrast, a2 experiences relatively large varia-
tions (Fig. 7d) and decreases as a function of bias (except CuPc/
LuPc ), such that the largest change is noted for Cu(F Pc)/
2
2
16
LuPc
. Such behaviour reveals enhancement in the hetero-
geneity, which can be correlated with the higher number of
the injected electrons arriving at the interface with increasing
bias causing an extension of the conducting R–CPE elements.
Based on the variations of a1 and a2, a broader conclusion
can be drawn as the bulk charge transport remains constant
Fig. 8 The charge transport scheme between frontier orbitals and trap
(
or decreases slightly) while interfacial charge transport
8 2
states in Cu(F Pc) and LuPc under photon exposure (a) and charge
increases with increasing bias, manifesting the key
phenomenon of organic heterojunction effects.
alignments at the interface in the dark (b) and in light (c).
To further confirm the interfacial charge accumulation and
implication of bias voltage over it, effective capacitances of bulk
and interface were estimated based on eqn (3) and their
variation with bias was probed. Based on the circuit elements
designation, Ceff1 corresponds to the bulk capacitance while
Ceff2 is associated with the interface. At first, very low value of
Ceff1 in pF while relatively higher value of Ceff2 in nF can be
remarked. Since capacitance is inversely proportional to the
layer thickness, it is confirmed that the interface is much
thinner than the bulk, which is also a typical case of organic
into the sublayer and excite charge carriers. In fact, the effect of
photoexcitation on the carrier concentration in the top LuPc
2
layer will be negligible considering its very high carrier density
and narrow band gap. But in the wide band gap and low
carrier density Cu(F Pc) sublayer possessing large amount of
8
trap states, photoexcitation can have a profound effect on the
ꢀ
+
relative amount of e and h carriers. It is well known that
oxygen adsorption on molecular materials causes their p-type
doping by creating a lot of donor like shallow traps providing a
5
1
+
55
heterojunction effects. As depicted in Fig. 7e, bulk capacitance
remains approximately constant for all heterojunction devices as
medium for
h
hopping.
Photoexcitation desorbs the
5
6
molecular oxygen from shallow traps near the HOMO of
Cu(F Pc), thereby suppressing the h concentration and con-
currently promotes e from an acceptor like shallow traps to
the LUMO level, increasing the e concentration in the sub-
+
the largest change is noticed in Cu(F Pc)/LuPc -n type in which it
8
2
8
ꢀ
decreases from 45 pF to 20 pF. On the other hand, interfacial
capacitance Ceff2 experiences an increase with the bias which can
be correlated with the thinning of the interface as a consequence
of the decrease in energy barrier. The largest and steepest rise in
ꢀ
layer. As a consequence, the Fermi level of Cu(F Pc) is lowered
8
enough to enable charge transfer through path 1 in Fig. 8a.
Because of that, interfacial charge alignment takes the form of
an accumulation heterojunction as shown in Fig. 8c. The
C
eff2 is noticed for the Cu(F16Pc)/LuPc
highlighting effective overcoming of the energy barrier. It can
also be noted that for the Cu(F Pc)/LuPc heterojunction, in its
2
device from 7 nF to 45 nF,
8
2
chemical doping from NH
promotes the e transport to the sublayer, also manifested by
in the top LuPc
layer further
3
2
ꢀ
p-type, Ceff2 is larger at all studied bias than that of n-type,
confirming a higher interfacial energy barrier in the latter.
3
a current increase under NH exposure.
Mechanism
The mechanism of such photon-triggered ambipolarity and the
Experimental
Reagents and materials synthesis
respective energy band position of Cu(F
explained by taking into account the e and h hopping
between the frontier orbitals of Cu(F Pc) and LuPc and sub- The precursors for MPc synthesis including 4,5-difluoro-1,2-
8 2
Pc) and LuPc can be
ꢀ
+
8
2
sequent alignment of charges at the heterojunction interface. dibromobenzene, copper cyanide, 4,5-dichlorophthalic
Some of us previously estimated the orbital energies of these anhydride, copper(II) acetate tetrahydrate, urea, ammonium
MPcs by Ultraviolet Photoelectron Spectroscopy (UPS) and based heptamolybdate, tetrafluorophthalonitrile, copper(II) chloride,
on those, LuPc
shown in Fig. 8a.
2
and Cu(F
8
Pc) orbitals can be positioned as phthalonitrile, and lutetium acetate trihydrate (Lu(OAc)
Since the workfunction of LuPc is higher were purchased from Sigma-Aldrich. CuPc was purchased
8
Pc), h transport takes place in steady state conditions directly from the same supplier and was used for vacuum
3
ꢂ3H
2
O)
5
2,53
2
+
than Cu(F
through path 2, resulting in h accumulation in Cu(F
depletion in LuPc near the interface. Such interfacial charge the solvothermal synthesis of phthalocyanines such as DMF,
redistribution generates a hole accumulation/depletion hetero- acetone, THF, CHCl , and H SO were of analytical grade and
+
+
8
Pc) and h
sublimation without any purification. The solvents used in
2
3
2
4
5
4
junction (Fig. 8b). NH interacts mainly with the top LuPc
were procured from Merck. Throughout the synthesis, Milli-Q
layer during a short exposure period and reduces the h hopping, ultrapure water (resistivity B18 MO cm) was used.
which is manifested by the decrease in device current. Synthesis of Cu(F Pc) was achieved through reaction of 4,5-
The effect of photon exposure is however different from difluoro-1,2-dibromobenzene with copper cyanide as described
3
2
+
8
5
7
3
chemical doping from NH because the photon can easily pass in detail previously in the ESI of our previous work. In brief,
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 5008–5020 | 5017

View Article Online
Journal of Materials Chemistry C
Paper
the mentioned precursors were refluxed in DMF at 150 1C for
2
2 4
hours followed by treatment with concentrated H SO and
washing with CHCl . Finally, the products were dried at 100 1C
3
under primary vacuum to yield the powder of Cu(F Pc), which
8
was confirmed by elemental analysis (calc. C, 53.38%; H,
1
.12%; F, 21.11%; Cu, 8.83%. Found: C, 52.74%; H, 1.13%;
F, 21.03%; Cu, 9.17%), (UV-vis (chloronaphtalene): Q band-
64 nm). The synthesis of Cu(Cl Pc) was performed by refluxing
6
8
a mixture of 4,5-dichlorophthalic anhydrid, copper(II) acetate
tetrahydrate, urea and ammonium heptamolybdate in nitro-
benzene for 5 h. The detailed synthesis procedure is reported in
5
8
our recent work. The obtained precipitate was filtered,
washed (with ethanol and water) and treated with 10% HCl. Scheme 2 Illustration of device testing workstation integrating gas
The resulting solid was successively washed with toluene, sources, fluidic circuits, electronic valves, humidity generator, test
chamber equipped with LEDs and electrometer.
ethanol, methanol and acetone to yield finally a blue coloured
8
powder of Cu(Cl Pc), and the elemental composition and
structure were confirmed (calc. C, 45.13; H, 0.95; N, 13.16;
found: C, 44.1; H, 0.87; N, 13.21), (UV-vis (DMF): Q band-632
controlled by electronic valves and the total flow was
ꢀ
1
and 701 nm, aggregates). Cu(F Pc) was synthesized by heating
1
6
maintained in the range of 0.5–0.55 NL min depending on
the ammonia concentration required in the sensors test
the reaction mixture of tetrafluorophthalonitrile and copper(II)
3
chloride at 250 1C for 7 hours based on the previously reported
chamber (of volume 8 cm ). The workstation was fully auto-
5
9
procedure (UV-vis (chloronaphtalene): Q band-691 nm). The
synthesis of LuPc was realized without any solvent between
phthalonitrile and Lu(OAc)
mated in which the gas flow, humidity, switching of the valves,
data transfer and initiation of new measurements were
controlled by homemade and customized software. Gas sensing
experiments were carried out in a dynamic way through
2
6
0
3
ꢂ3H
2
O in a 8 : 1 ratio at 300 1C.
The reaction product was washed with DMF and ethanol to
remove unreacted precursors and monophthalocyanines
alternate exposure to 20 ppm of NH and recovery under clean
3
followed by drying at 100 1C to obtain LuPc powder, and the
2
air for 1 min and 4 min, respectively, at rh of 40%. The sensors
test chamber was integrated with 6 standard LEDs (green,
yellow and red with maximum wavelength 565 nm, 594 nm
elemental composition and structure was confirmed (UV-vis
Soret band – 320 nm; Q band – 658 nm, FTIR peaks at 722,
ꢀ
1
1
123, 1317 and 1449 cm ).
642 nm, respectively), placed at the bottom of the test chamber
at a distance of approximately 2 cm from the sensor. Continuous
measurements were recorded for repeated exposure-recovery
cycles at first in the dark, then under LED illumination.
Device fabrication and electrical measurements
Heterojunction devices were fabricated on ITO interdigitated
electrodes (IDE), lithographically patterned on a (1 ꢁ 1) cm
2
glass substrate and separated by 75 mm. The bilayer films of Thin film characterization
5
0 nm each of MPc (except for Cu(Cl Pc): 30 nm) were sequentially
8
The MPc films coated on glass substrates were studied by X-ray
diffraction (XRD) at room temperature using a PANanalytical
X’pert MPD diffractometer with Cu-Ka incident radiation (l =
deposited on ITO IDE maintained at room temperature under
ꢀ
6
secondary vacuum (ca. 10 mbar) in a UNIVEX 250 thermal
evaporator (Oerlikon, Germany). The sublimation temperature of
MPcs were noted in the range of 400–480 1C. No post deposition
thermal or chemical treatment was performed. A similar sized glass
substrate was placed next to the IDE for further structural and
morphological studies of the films. Electrical measurements of the
fabricated devices were carried out using a Keithley electrometer at
a bias voltage from ꢀ10 V to 10 V in steps of 0.1 V in ambient air.
Impedance measurements of different heterojunction devices
were performed using a Solartron SI 1260 impedance analyzer.
The frequency range was 10 Hz to 10 MHz with a fixed ac
oscillation amplitude of 200 mV and a DC bias ranging from 0 V
to 10 V. Commercial software Zview from Ametek was used for
impedance data fitting and parameters extraction.
1
0
.5418 Å) in the 2y window of 5 to 801, using a scan step of
.051. To evaluate the phase purity and spatial orientations,
experimental data were fitted by profile matching in Fullprof
software. The morphology of the MPc films coated on glass
were analysed using a Scanning Electron Microscope (Quanta
FEG 650, FEI equipped with an EDS detector), operating at an
accelerating voltage of 3–5 kV, while performing a horizontal
scan of the surface. A thin gold layer was at first sputtered on
the film to establish electrical contact with the microscope and
also to avoid over-charging of the film. EDX spectra and
mapping of the different MPc films were also recorded.
The workstation used for electrical measurements and NH3
sensing at different rh is shown in Scheme 2, consisting of
Conclusion
fluidic circuits connected with commercial cylinders of In summary, we have demonstrated for the first time
synthetic air and ammonia gas (1000 ppm concentration, the application of visible wavelength photons produced by
purchased from Air Liquide France). The flow in the circuit is LED illumination, as an external trigger for the majority
5018
|
J. Mater. Chem. C, 2021, 9, 5008–5020
This journal is © The Royal Society of Chemistry 2021

View Article Online
Paper
Journal of Materials Chemistry C
charge carrier inversion in an ambipolar organic hetero- Acknowledgements
junction device. To achieve the ambipolar transport regime, a
judicious selection among p-type and n-type halogenated CuPc
has been made, which are combined with highly conductive
The authors acknowledge the Agence Nationale de la Recherche
OUTSMART ANR-2015-CE39-0004-03). This work was sup-
(
ported by the French ‘‘Investissements d’Avenir’’ program,
project ISITE BFC (contract ANR 15 IDEX 0003) ‘‘Master
T2MC’’ (G. L. K.) and CoMICS program 2019-2022: Chemistry
of Molecular Interactions – Catalysis and Sensors (A. K.).
Financial assistance from Conseil R ´e gional de Bourgogne and
European Union (FEDER) through the FABER and the CDEA
programs are gratefully acknowledged.
LuPc in a bilayer heterostructure configuration.
2
Thin film microstructural properties of the heterostructures
revealed a compact and semicrystalline organization having
predominantly an amorphous phase in which small pockets of
crystallites are present, whose sizes and orientations depend on
the halogen substitution in CuPc. The I(V) curves of the
heterojunction devices revealed a non-linear behavior consisting
of an energy barrier, which is largest for Cu(F Pc)/LuPc and
8
2
lowest for CuPc/LuPc . The energy barrier is slightly decreased
upon photon exposure which is attributed to generation of
additional charge carriers, which were otherwise held in shallow
2
Notes and references
1 J. Cornil, J. L. Br ´e das, J. Zaumseil and H. Sirringhaus, Adv.
Mater., 2007, 19, 1791–1799.
8
traps. Under exposure to ammonia, the Cu(F Pc) based hetero-
junction device exhibited a p-type and n-type response in the
dark and under LED illumination, respectively, while other
heterojunction devices displayed only either of the two states,
confirming an ambipolar transport regime in the Cu(F Pc)/LuPc
device, which is controlled uniquely by photons and
remains independent of the bias voltage. Charge transport
studies by impedance spectroscopy revealed a combination of
bulk and interfacial transport in the heterojunction devices
2 E. J. Meijer, D. M. de Leeuw, S. Setayesh, E. van Veenendaal,
B. H. Huisman, P. W. M. Blom, J. C. Hummelen, U. Scherf
and T. M. Klapwijk, Nat. Mater., 2003, 2, 678–682.
3 W.-J. Zeng, X.-Y. Zhou, X.-J. Pan, C.-L. Song and H.-L. Zhang,
AIP Adv., 2013, 3, 012101, DOI: 10.1063/1.4774287.
4 M. Kang, D. Khim, J. Kim, H. J. Lee, J. Y. Jo, K.-J. Baeg and
D.-Y. Kim, Org. Electron., 2018, 58, 12–17.
8
2
5 Y. Zhou, S.-T. Han, P. Sonar and V. A. L. Roy, Sci. Rep., 2013,
3, 2319, DOI: 10.1038/srep02319.
2
while only bulk transport exists in LuPc resistors. It is also
confirmed that interfacial transport increases as a function of
increasing bias voltage, which is associated with a decrease of
energy barrier at the interface. Fitting of the experimental
Nyquists plots with two components R–CPE equivalent
circuits allowed determination of different circuit parameters
which were finally correlated with the charge transport in the
devices. These parameters confirmed the generation of additional
6 X. Chen, J. Pan, J. Fu, X. Zhu, C. Zhang, L. Zhou, Y. Wang,
Z. Lv, Y. Zhou and S.-T. Han, Adv. Electron. Mater., 2018,
4, 1800444, DOI: 10.1002/aelm.201800444.
7 H. Kwon, H. Yoo, M. Nakano, K. Takimiya, J.-J. Kim and
J. K. Kim, RSC Adv., 2020, 10, 1910–1916.
8 Y. Wu, P. Ma, N. Wu, X. Kong, M. Bouvet, X. Li, Y. Chen and
J. Jiang, Adv. Mater. Interfaces, 2016, 3, 1600253, DOI:
10.1002/admi.201600253.
8 2
charge carriers in the bulk Cu(F Pc)/LuPc heterostructure upon
photon exposure, changing its semiconducting polarity from p to
n-type.
9 J. Gao, G. Lu, J. Kan, Y. Chen and M. Bouvet, Sens. Actuators,
B, 2012, 166-167, 500–507.
10 A. Kumar, R. Meunier-Prest and M. Bouvet, Sensors, 2020,
2
0, 4700, DOI: 10.3390/s20174700.
Author contributions
1
1
1 M. Bouvet, Anal. Bioanal. Chem., 2006, 384, 366–373.
2 G. Bengasi, R. Meunier-Prest, K. Baba, A. Kumar,
A. L. Pellegrino, N. D. Boscher and M. Bouvet, Adv. Electron.
Mater., 2020, 2000812, DOI: 10.1002/aelm.202000812.
3 R. D. Gould, Coord. Chem. Rev., 1996, 156, 237–274.
4 M. Grobosch, C. Schmidt, R. Kraus and M. Knupfer, Org.
Electron., 2010, 11, 1483–1488.
5 F. I. Bohrer, C. N. Colesniuc, J. Park, M. E. Ruidiaz,
I. K. Schuller, A. C. Kummel and W. C. Trogler, J. Am. Chem.
Soc., 2009, 131, 478–485.
Gracia Loma: investigation, methodology and visualization.
Abhishek Kumar: conceptualization, investigation, visualiza-
tion, data curation, software, formal analysis, writing – original
draft, writing – review & editing. Vaibhav Vibhu: investigation,
methodology, visualization, data curation, software, formal
analysis. Alix Deshotel: investigation, methodology and
visualization. Micka ¨e l Mateos: investigation, methodology and
visualization. Seydou Ouedraogo: investigation, methodology
and visualization. Rita Meunier-Prest: conceptualization,
formal analysis, investigation, methodology, data curation,
validation. Marcel Bouvet: conceptualization, formal analysis,
visualization, funding acquisition, writing – original draft, super-
vision, validation, project administration.
1
1
1
1
1
1
1
6 L. S. Chia, Y. H. Du, S. Palale and P. S. Lee, ACS Omega, 2019,
4, 10388–10395.
7 S. Liu, H. Wang, X. Wang, S. Li, H. Liu, Y. Chen and X. Li,
J. Mater. Chem. C, 2019, 7, 424–433.
8 X. Kong, Z. Dong, Y. Wu, X. Li, Y. Chen and J. Jiang, Chin.
J. Chem., 2016, 34, 975–982.
9 A. Wannebroucq, S. Ouedraogo, R. Meunier-Prest,
J.-M. Suisse, M. Bayo and M. Bouvet, Sens. Actuators, B,
2018, 258, 657–664.
Conflicts of interest
There are no conflicts to declare.
This journal is © The Royal Society of Chemistry 2021
J. Mater. Chem. C, 2021, 9, 5008–5020 | 5019

View Article Online
Journal of Materials Chemistry C
Paper
2
2
2
2
2
0 X. Kong, X. Zhang, D. Gao, D. Qi, Y. Chen and J. Jiang, Chem. 39 A. Sukhikh, D. Bonegardt, D. Klyamer, P. Krasnov and
Sci., 2015, 6, 1967–1972. T. Basova, Molecules, 2020, 25, 1620.
1 D. Gao, X. Zhang, X. Kong, Y. Chen and J. Jiang, ACS Appl. 40 J. T. E. Quinn, J. Zhu, X. Li, J. Wang and Y. Li, J. Mater.
Mater. Interfaces, 2015, 7, 2486–2493. Chem. C, 2017, 5, 8654–8681.
2 G. Lu, X. Kong, C. Wang, L. Zhao, D. Qi, Y. Jiang, S. Zhao, 41 S. Senthilarasu, Y. B. Hahn and S.-H. Lee, J. Appl. Phys.,
Y. Chen and J. Jiang, Dyes Pigm., 2019, 161, 240–246. 2007, 102, 043512, DOI: 10.1063/1.2771046.
3 A. Dodabalapur, H. E. Katz, L. Torsi and R. C. Haddon, 42 W. Chen, S. Chen, H. Huang, D. C. Qi, X. Y. Gao and
Science, 1995, 269, 1560–1562.
4 A. Dodabalapur, H. E. Katz, L. Torsi and R. C. Haddon, Appl.
Phys. Lett., 1996, 68, 1108–1110.
A. T. S. Wee, Appl. Phys. Lett., 2008, 92, 063308, DOI:
10.1063/1.2857460.
43 E. Jungyoon, S. Kim, E. Lim, K. Lee, D. Cha and
B. Friedman, Appl. Surf. Sci., 2003, 205, 274–279.
25 H. Wang and D. Yan, NPG Asia Mater., 2010, 2, 69–78.
26 A. Opitz, A. Wilke, P. Amsalem, M. Oehzelt, R.-P. Blum, 44 P. A. Pandey, L. A. Rochford, D. S. Keeble, J. P. Rourke,
J. P. Rabe, T. Mizokuro, U. H o¨ rmann, R. Hansson, E. Moons
and N. Koch, Sci. Rep., 2016, 6, 21291, DOI: 10.1038/
srep21291.
7 S. N ´e non, D. Kanehira, N. Yoshimoto, F. Fages and
C. Videlot-Ackermann, Synth. Met., 2011, 161, 1915–1920.
T. S. Jones, R. Beanland and N. R. Wilson, Chem. Mater.,
2012, 24, 1365–1370.
45 M. Mateos, M.-D. Tchanga ¨ı , R. Meunier-Prest, O. Heintz,
F. Herbst, J.-M. Suisse and M. Bouvet, ACS Sens., 2019, 4,
740–747.
2
2
2
8 K. Eguchi, M. M. Matsushita and K. Awaga, J. Phys. Chem. C, 46 M. Bouvet, V. Parra and J. M. Suisse, Eur. Phys. J.: Appl. Phys.,
018, 122, 26054–26060. 2011, 56, 34103, DOI: 10.1051/epjap/2011110220.
9 B. Wang, T.-P. Huynh, W. Wu, N. Hayek, T. T. Do, 47 A. Litke, E. J. M. Hensen and J. P. Hofmann, J. Phys. Chem.
J. C. Cancilla, J. S. Torrecilla, M. M. Nahid, J. M. Colwell, C, 2017, 121, 10153–10162.
O. M. Gazit, S. R. Puniredd, C. R. McNeill, P. Sonar and 48 P. Gaudillat, A. Wannebroucq, J.-M. Suisse and M. Bouvet,
H. Haick, Adv. Mater., 2016, 28, 4012–4018. Sens. Actuators, B, 2016, 222, 910–917.
0 S. Ouedraogo, R. Meunier-Prest, A. Kumar, M. BayoBangoura 49 B. Hirschorn, M. E. Orazem, B. Tribollet, V. Vivier, I. Frateur
and M. Bouvet, ACS Sens., 2020, 5, 1849–1857. and M. Musiani, Electrochim. Acta, 2010, 55, 6218–6227.
1 H. Tang, Y. Li, R. Sokolovskij, L. Sacco, H. Zheng, H. Ye, 50 K. A. Miller, R. D. Yang, M. J. Hale, J. Park, B. Fruhberger,
2
3
3
H. Yu, X. Fan, H. Tian, T.-L. Ren and G. Zhang, ACS Appl.
Mater. Interfaces, 2019, 11, 40850–40859.
C. N. Colesniuc, I. K. Schuller, A. C. Kummel and
W. C. Trogler, J. Phys. Chem. B, 2006, 110, 361–366.
32 V. Parra and M. Bouvet, Semiconductor Transducer and Its 51 K. M. Lau, J. X. Tang, H. Y. Sun, C. S. Lee, S. T. Lee and D. Yan,
Use in a Sensor for Detecting Electron-donor or Electron-
Acceptor Species, US Pat., 8450725 B2, 2013.
3 N. T. Boileau, O. A. Melville, B. Mirka, R. Cranston and
B. H. Lessard, RSC Adv., 2019, 9, 2133–2142.
Appl. Phys. Lett., 2006, 88, 173513, DOI: 10.1063/1.2198484.
52 R. Murdey, N. Sato and M. Bouvet, Mol. Cryst. Liq. Cryst.,
2006, 455, 211–218.
53 R. Murdey, M. Bouvet, M. Sumimoto, S. Sakaki and N. Sato,
Synth. Met., 2009, 159, 1677–1681.
3
3
4 H. Jiang, P. Hu, J. Ye, Y. Li, H. Li, X. Zhang, R. Li, H. Dong,
W. Hu and C. Kloc, Adv. Mater., 2017, 29, 1605053, DOI: 54 H. Wang and D. Yan, NPG Asia Mater., 2010, 2, 69–78.
1
0.1002/adma.201605053.
55 P. K. Nayak, R. Rosenberg, L. Barnea-Nehoshtan and
D. Cahen, Org. Electron., 2013, 14, 966–972.
56 X. Zhang, J. Jie, Z. Wang, C. Wu, L. Wang, Q. Peng, Y. Yu,
P. Jiang and C. Xie, J. Mater. Chem., 2011, 21, 6736–6741.
57 V. Parra, J. Brunet, A. Pauly and M. Bouvet, Analyst, 2009,
134, 1776–1778.
3
3
3
5 G. Guillaud, M. Al Sadoun, M. Maitrot, J. Simon and
M. Bouvet, Chem. Phys. Lett., 1990, 167, 503–506.
6 H. Alves, A. S. Molinari, H. Xie and A. F. Morpurgo, Nat.
Mater., 2008, 7, 574–580.
7 A. K. Debnath, A. Kumar, S. Samanta, R. Prasad, A. Singh,
A. K. Chauhan, P. Veerender, S. Singh, S. Basu, D. K. Aswal 58 S. Ouedraogo, T. Coulibaly, R. Meunier-Prest, M. Bayo-
and S. K. Gupta, Appl. Phys. Lett., 2012, 100(14), 142104,
DOI: 10.1063/1.3699272.
Bangoura and M. Bouvet, J. Porphyrins Phthalocyanines,
2019, 24, 750–757.
38 H. Jiang, J. Ye, P. Hu, F. Wei, K. Du, N. Wang, T. Ba, S. Feng 59 J. M. Birchall, R. N. Haszeldine and J. O. Morley, J. Chem.
and C. Kloc, Sci. Rep., 2014, 4(1), 7573, DOI: 10.1038/
Soc. C, 1970, 2667–2672, DOI: 10.1039/J39700002667.
srep07573.
60 G. Clarisse and M. T. Riou, Inorg. Chim. Acta, 1987, 130, 139–144.
5020
|
J. Mater. Chem. C, 2021, 9, 5008–5020
This journal is © The Royal Society of Chemistry 2021