materials
Article
Thermally Stable Solution Processed Vanadium
Oxide as a Hole Extraction Layer in Organic
Solar Cells
Abdullah Alsulami 1 , Jonathan Griffin 1 , Rania Alqurashi 1 , Hunan Yi 2 , Ahmed Iraqi 2 ,
David Lidzey 1 and Alastair Buckley 1, *
1
2
*
Department of Physics & Astronomy, University of Sheffield, Hicks Building, Hounsfield Rd., Sheffield,
South Yorkshire S3 7RH, UK; [email protected] (A.A.); [email protected] (J.G.);
[email protected] (R.A.); [email protected] (D.L.)
Department of Chemistry, University of Sheffield, Sheffield, South Yorkshire S3 7HF, UK;
[email protected] (H.Y.); [email protected] (A.I.)
Correspondence: [email protected]; Tel.: +44-0114-22-23597
Academic Editor: Lioz Etgar
Received: 16 February 2016; Accepted: 22 March 2016; Published: 25 March 2016
Abstract: Low-temperature solution-processable vanadium oxide (V2 Ox ) thin films have been
employed as hole extraction layers (HELs) in polymer bulk heterojunction solar cells. V2 Ox films
were fabricated in air by spin-coating vanadium(V) oxytriisopropoxide (s-V2 Ox ) at room temperature
without the need for further thermal annealing. The deposited vanadium(V) oxytriisopropoxide film
undergoes hydrolysis in air, converting to V2 Ox with optical and electronic properties comparable to
vacuum-deposited V2 O5 . When s-V2 Ox thin films were annealed in air at temperatures of 100 ˝ C and
200 ˝ C, OPV devices showed similar results with good thermal stability and better light transparency.
Annealing at 300 ˝ C and 400 ˝ C resulted in a power conversion efficiency (PCE) of 5% with a
decrement approximately 15% lower than that of unannealed films; this is due to the relative decrease
in the shunt resistance (Rsh ) and an increase in the series resistance (Rs ) related to changes in the
oxidation state of vanadium.
Keywords: organic photovoltaic; vanadium oxide; thermal stability; solution processing;
photoelectron spectroscopy
1. Introduction
Currently, the high cost of commercial inorganic photovoltaics remains an obstacle for the
wide-scale installation in both residential and commercial settings [1]. In order to overcome this
high cost, researchers have studied numerous materials as alternatives, with organic polymer
solar cells being a promising candidate [2,3]. This is due to organic polymers exhibiting several
advantageous properties, such as solution-processability, mechanical-flexibility, and thin device
architectures allowing for light-weight devices [4–6]. This combination of factors allows organic
solar to scale up manufacture via roll-to-roll- or sheet-to-sheet-based deposition techniques leading
to reduced fabrication costs. Recent advances in polymer synthesis and device processing have
pushed the power conversion efficiency of single junction cells as high as 10.8% [7]. These efficiencies
are above the 10% mark often quoted as the point at which polymer photovoltaics will become
commercially viable. However, currently, the lifetime of devices incorporating the highest
performing materials are below the 10 year mark required; one method to improve the lifetime
of devices is to replace the commonly-used hole extraction interfacial layer (HELs), polyethylene
dioxythiophene:polystyrenesulfonate (PEDOT:PSS) [8–10]. This is because, despite the positive aspects
of aqueous PEDOT:PSS, the residual moisture and the acidic nature of PEDOT:PSS can cause
Materials 2016, 9, 235; doi:10.3390/ma9040235
www.mdpi.com/journal/materials
Materials 2016, 9, 235
Materials 2016, 9, 235 2 of 12
2 of 11 degradation of the electrode and organic films and, therefore, reduce the devices operational
degradation of the electrode and organic films and, therefore, reduce the devices operational lifetime lifetime [11–14]. To overcome some of these issues thin metal oxides 3such
MoO3 , WO
NiO,
[11–14]. To overcome some of these issues thin metal oxides such as MoO
, WOas
3, NiO, and V
2O
3 ,5 have and V2 O5 have been used; these materials have been shown to exhibit performances that are similar
been used; these materials have been shown to exhibit performances that are similar to or better than to or better than OPVs with an interfacial PEDOT:PSS layer [15–18]. Not only are these materials of
OPVs with an interfacial PEDOT:PSS layer [15–18]. Not only are these materials of interest in organic interest in organic photovoltaics because of their unique electronic properties and chemical stability,
photovoltaics because of their unique electronic properties and chemical stability, they are also being they are also being used in other photovoltaic technologies such as CiGS, and CdTe [19–22]. However,
used in other photovoltaic technologies such as CiGS, and CdTe [19–22]. However, like PEDOT:PSS like presence PEDOT:PSS
the presence
of water
can lead
to changes
at between the interface
between
oxides
the of water can lead to changes at the interface metal oxides metal
and organic and organic semiconductors,
due to the intercalation
of water
within
metal
oxideleading clustersto leading
to
semiconductors, due to the intercalation of water within metal oxide clusters changes changesthe within
the level energy
level structure
[23].makes This makes
the improvement
of processing
conditions
within energy structure [23]. This the improvement of processing conditions and and reduction in the cost and complexity of depositing these interfacial layers highly desirable.
reduction in the cost and complexity of depositing these interfacial layers highly desirable. Solution-processing of of the the metal metal oxide oxide interfacial interfacial layers layers has has shown shown significant significant promise promise for for
Solution‐processing reducing the the complexity
Previous studies studies have have shown shown that that
reducing complexity of
of the
the deposition
deposition of
of these
these materials.
materials. Previous vanadium oxide can be deposited from solution at low temperature in air without the need for any high
vanadium oxide can be deposited from solution at low temperature in air without the need for any temperature
post-deposition
treatments
[24–28]. On
the other
some optoelectronic
devices which
high temperature post‐deposition treatments [24–28]. On hand,
the other hand, some optoelectronic are fabricated at high temperatures require developed metal oxides which must be able to keep their
devices which are fabricated at high temperatures require developed metal oxides which must be properties under high temperature fabrication processes. Whilst the optical and chemical properties of
able to keep their properties under high temperature fabrication processes. Whilst the optical and V2 Ox have
previously
after high-temperature
[29–31], its performance
chemical properties of been
V2Oxinvestigated
have previously been investigated annealing
after high‐temperature annealing with
OPVs
has
not.
Therefore,
our
motivation
for
this
work
is
to
explore
the
impact
of
thermal
heating
[29–31], its performance with OPVs has not. Therefore, our motivation for this work is to explore the on V2 Ox thin films as HELs in organic
solar cells.
impact of thermal heating on V
2Ox thin films as HELs in organic solar cells. In this
paper
we incorporated
V2 Ox thinVfilms
deposited
from
a vanadium(V)
In this paper we incorporated 2Ox thin films deposited from oxytriisopropoxide
a vanadium(V) precursor into PFDT2BT-8:PC70 BM OPV devices70achieving
a power conversion efficiency of 6%,
oxytriisopropoxide precursor into PFDT2BT‐8:PC
BM OPV devices achieving a power conversion comparable
PEDOT:PSS
and to vacuum
deposited
fabricated
some OPV
efficiency of to6%, comparable PEDOT:PSS and MoO
vacuum deposited we
MoO
3. Furthermore, we 3 . Furthermore,
devices with
annealed
V2 Ox with layersannealed at high V
temperatures
air temperatures before spin coating
the active
fabricated some OPV devices 2Ox layers at in
high in air before spin layer to study their thermal stability. We demonstrated that OPV devices show PCE ě 5% with
coating the active layer to study their thermal stability. We demonstrated that OPV devices show thermally-annealed
V2 Ox at 400 ˝ C. Absorption
spectroscopy,
X-ray
photoelectronX‐ray spectroscopy
(XPS),
PCE ≥ 5% with thermally‐annealed V2Ox at 400 °C. Absorption spectroscopy, photoelectron and
ultraviolet
photoelectron
spectroscopy
(UPS)
are
combined
to
explain
the
J–V
characteristics
spectroscopy (XPS), and ultraviolet photoelectron spectroscopy (UPS) are combined to explain the J–
of OPVs.
V characteristics of OPVs. 2. Results and Discussion
2. Results and Discussion 2.1. PFDT2BT-8 Structure
2.1. PFDT2BT‐8 Structure The chemical structure of our donor polymer used in this work is shown in Figure 1a [32].
The chemical structure of our donor polymer used in this work is shown in Figure 1a [32]. The The highest occupied molecular orbital (HOMO) energy level and lowest unoccupied molecular
highest occupied molecular orbital (HOMO) energy level and lowest unoccupied molecular orbital orbital (LUMO) of PFDT2BT-8 are ´5.33 eV and ´3.34 eV, respectively, as determined from a cyclic
(LUMO) of PFDT2BT‐8 are −5.33 eV and −3.34 eV, respectively, as determined from a cyclic voltammetry. The energy band diagram and work function of the relative materials in our study are
voltammetry. The energy band diagram and work function of the relative materials in our study are presented in Figure 1b [32,33].
presented in Figure 1b [32,33]. (a) (b)
Figure 1. (a) The chemical structure of PFDT2BT‐8; and (b) energy levels of electrodes, HELs, and Figure 1.
(a) The chemical structure of PFDT2BT-8; and (b) energy levels of electrodes, HELs, and active
active layer materials used in this work. layer materials used in this work.
2.2. s‐V2Ox as a Hole Extraction Layer To assess the performance of PFDT2BT‐8 polymer with the s‐V2Ox interlayer, we fabricated sets of PFDT2BT‐8:PC70BM devices with variable thicknesses of s‐V2Ox. The extracted data suggests an Temperature State State Function (eV) Band (eV) (eV) 400 5.2 5.0 ± 0.17 0.83 10.1 ± 0.12 59.6 ± 0.9 17.6 ± 0.4 933 ± 84 Unannealing 15% 85% 5.26 2.53 2.5 200 °C 22.5% 77.5% 5.22 2.53 3.78 It can be seen that the performance of devices annealed at temperatures 100 or 200 °C is quite 300 °C similar to those prepared without heat treatment. The slight increase in PCE of devices annealed up 27% 73% 5.23 2.50 2.68 400 °C to 200 °C can be ascribed to the relative increase in the shunt resistance R
33% 67% 5.19 2.50 2.42 sh and decrease in the series 16.216.2
16.116.1
16.016.0
15.915.9
(eV/cm) 2

h x 10
10
Intensity (counts)
15.815.8
0.9
0.5
0.2
5
0.2
5
0.4
0.02
2.52.0
4
(eV/cm) 2
0.00.5
0.0
6
6 0.4
2.5
10
0.6
2.01.5
0.04
0.02
1.51.0
1.00.5
0.50.0
0.0

0.6
h x 10
0.7
10
0.7
0.04

0.8
h x 10
0.8
(eV/cm) 2
0.9
Transmittance
-4
-4
-6
-6
-8
-8
-10
-10
Transmittance
Intensity
(counts)
Intensity (counts)
Intensity (counts)
Intensity (counts)
2
2
Current
density
(mA/cm
) )
Current
density
(mA/cm
Intensity (counts)
Transmittance
2
Currenst
density
(mA/cm
) 2)
Currenst
density
(mA/cm
resistance R
Materials
2016, 9,s. In contrast, as the film annealing temperature is raised to 400 °C, OPV devices show a 235
3 of 12
2.6. Ultraviolet Photoelectron Spectroscopy PCE of 5% with a decrement approximately 15% lower than that of those with unannealed films. It is clear from Table 1 that the relatively weak photovoltaic response can be attributed to the decrease in Figure 7 plots the UPS spectra of V
2Ox films that have been annealed at different temperatures. open‐circuit voltage, due to the change in chemical structure as will be discussed based on the XPS The work function and valence band edge determined by the UPS are also summarised in Table 1. 2.2. s-V2 Ox as a Hole Extraction Layer
data on. Furthermore, the shunt resistance falls annealing above 300 °C As shown in Figure 7a, later the position of the secondary electron cut‐off for considerably unannealed on film was reaching a minimum R
sh of 933 Ω∙cm
determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
To assess the performance
of2 on annealing at 400 °C due to the recombination process at the PFDT2BT-8 polymer with the work s-V2 Ox interlayer, we fabricated sets
anode interface. This caused a decrease in hole density at the anode interface leading to a decrease in function of 5.26 eV. Compared to previous studies, the work function of our s‐V
2Ox films is similar to Materials 2016, 9, 235 3 of 11 an
of PFDT2BT-8:PC
70 BM devices with variable thicknesses of s-V2 Ox . The extracted data suggests
those prepared by the internal electric field [35,36]. different processes [34]. On annealing, the secondary electron cut‐off shows a Materials 2016, 9, 235 3 of 11 optimum s-V2 Ox layer thickness below 10 nm. Therefore, we selected a layer thickness of 5 nm for all
slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the optimum s‐V2Ox layer thickness below 10 nm. Therefore, we selected a layer thickness of 5 nm for all Ox devices.
work function s-V
we 22.3. Optical Properties can’t these changes directly to the processing of the s‐V2Ox due to the optimum s‐V
2Ox layer thickness below 10 nm. Therefore, we selected a layer thickness of 5 nm for all s‐Vassign 2Ox devices. possible influence of adsorbates. Figure 7b displays the onset of the s‐V
2Oxour
valence band to be 2.5 eV Figure
2
shows
the J–V characteristics of
OPV devices and those of the most widely used
s‐VTo understand the changes that we observed in the photovoltaic response as a function of the 2O
x devices. Figure 2 shows the J–V characteristics of our OPV devices and those of the most widely used below the Fermi level which is identical with all samples.
We did not notice any shift for the valence annealing temperature, we investigated the optical properties of annealed x films. Figure It
4 can be seen that
HELs,HELs, PEDOT:PSS and thermally evaporated MoO
PEDOT:PSS
and thermally evaporated MoO33 under AM 1.5 G illumination.
under
AM 1.5s‐V
G 2O
illumination.
Figure 2 shows the J–V characteristics of our OPV devices and those of the most widely used It can be seen that band maximum with increase annealing temperature. Careful observation of the valence band presents the optical transmittance of the films as a function of the wavelength for the as‐deposited usingHELs, PEDOT:PSS and thermally evaporated MoO
PEDOT:PSS and s-V2 O2xOlayers
as HEL showed
a similar photovoltaic response
(PCE = 6.5%)
3 under AM 1.5 G illumination.
It can be seen that using PEDOT:PSS and s‐V
x layers as HEL showed a similar photovoltaic response (PCE = 6.5%) region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below and annealed films. All s‐V2Ox films show high transmission for wavelengths above 500 nm covering Materials 2016, 9, 235 4 of 11 that
Materials 2016, 9, 235 4 of 11 using PEDOT:PSS and s‐V
2
O
x
layers as HEL showed a similar photovoltaic response (PCE = 6.5%) which
is
better
than
devices
fabricated
with
MoO
interlayer
(PCE
=
6.3%).
These
results
indicate
which is better than devices fabricated with MoO
the Fermi level as shown in the inset of Figure 7b. This suggests that oxygen vacancies shift the 33 interlayer (PCE = 6.3%). These results indicate that the majority of the solar spectrum. The transmittance in the visible wavelength range, from 400 to which is better than devices fabricated with MoO
3
interlayer (PCE = 6.3%). These results indicate that Fermi level up atoward the conduction band leaving some occupied states within the band gap. a high performance can be achieved for PFDT2BT‐8 devices by using untreated s‐V
2Oxx films which high
performance
can be1. achieved
for
PFDT2BT-8
devices
by
using
untreated
s-V
500 nm, increases slightly by annealing up to °C which is with attributed to change of annealed the films 2annealed Table Summary of solar cell parameters s‐V
2Ox s‐V
buffer at films
different Table 1. Summary of 300 solar cell parameters with 2Ox layer buffer layer at which
different are
Therefore, introducing oxygen vacancies can act as n‐type dopants resulting a decrease the a high performance can be achieved for PFDT2BT‐8 devices by using untreated s‐V
2Ox films which are in good agreement with the previous studies in literature [25,27,34]. refractive index related to chemical structure changes. On in annealing to in 400 °C, the film exhibits in good
agreement
with
the
previous
studies
in literature
[25,27,34].
temperatures for different periods of time. temperatures for different periods of time. work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms are in good agreement with the previous studies in literature [25,27,34]. absorption peak at 415 nm which could be ascribed to small polaron absorption [37,38].
This effect Ann. 2OAnn. Maximum
Voc Voc Jsc
FF 7 of 11 Rs Maximum
Average Jsc
FF between Rs Rsh Rsh that the high‐temperature annealing of s‐V
x film produces a change in the chemical structure of results from disordering defects in V2OAverage x structure leading to transferring charges Materials 2016, 9, 235 Materials 2016, 9, 235 7 of 11 2
−2) 2) 2) −2) 2) Temp. PCE (%) PCE
(Ω∙cm
(Ω∙cm
(av) (%) (V) (mA∙cm
(%) Temp. PCE (%) PCE
(Ω∙cm
(Ω∙cm2) (av) (%) (V) (mA∙cm
(%) the vanadium oxide layer and, thus, increases slightly the hole‐electron recombination at the neighbouring sites with a significant spread in energy. 2 that Non 6.0 5.8 ± 0.14 0.86 10.0 ± 0.22 67.1 ± 1.2 11.6 ± 0.9 1264 ± 91 Non 6.0 5.8 ± 0.14 0.86 10.0 ± 0.22 67.1 ± 1.2 11.6 ± 0.9 1264 ± 91 interface. Nevertheless, our experiments demonstrate s‐V
2O
xO
film can be used to replace Table Summary solar parameters with s‐V
2
x
buffer layer annealed at different Table 2. 2. Summary of of solar cell cell parameters with s‐V
2
O
x
buffer layer annealed at different 0
100 100 6.0 0.86 0.86 10.1 ± 0.13 68.3 ± 1.7 10.5 ± 1.0 1282 ± 215 5.9 ± 0.14 10.1 ± 0.13 68.3 ± 1.7 10.5 ± 1.0 1282 ± 215 evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. temperatures for different periods of time. temperatures for different periods of time. 0 6.0 5.9 ± 0.14 1.0 6.3 200 200 0.87 0.87 10.2 ± 0.14 67.1 ± 2.6 10.3 ± 0.7 1346 ± 246 6.3 5.9 ± 0.26 5.9 ± 0.26 10.2 ± 0.14 67.1 ± 2.6 10.3 ± 0.7 1346 ± 246 -2
4+
5+
4+
5+
Annealing V V
Oxidation V VOxidation Oxidation Work Valence Annealing Oxidation Work Valence 300 300 5.6 0.83 0.83 9.9 ± 0.41 64.3 ± 2.7 12.6 ± 1.7 1095 ± 238 5.6 5.3 ± 0.28 5.3 ± 0.28 9.9 ± 0.41 64.3 ± 2.7 12.6 ± 1.7 1095 ± 238 -2
(eV) Temperature State 0.9 -4
State Function (eV) Band (eV) Band (eV) (eV) Temperature State State Function (eV) 400 400 5.2 0.83 0.83 10.1 ± 0.12 59.6 ± 0.9 17.6 ± 0.4 933 ± 84 5.2 5.0 ± 0.17 5.0 ± 0.17 10.1 ± 0.12 59.6 ± 0.9 17.6 ± 0.4 933 ± 84 Unannealing 15% 85% 5.26 2.53 Unannealing 15% 85% 2.53 2.5 2.5 0.04 5.26 0.8 -4
200 °C 22.5% 77.5% 5.22 2.53 3.78 200 °C 22.5% 77.5% 5.22 2.53 3.78 -6
It can be seen that the performance of devices annealed at temperatures 100 or 200 °C is quite It can be seen that the performance of devices annealed at temperatures 100 or 200 °C is quite 300 °C 27% 73% 5.23 2.50 2.68 300 °C 27% 73% 5.23 2.50 2.68 -6
0.7
similar to those prepared without heat treatment. The slight increase in PCE of devices annealed up similar to those prepared without heat treatment. The slight increase in PCE of devices annealed up 0.02
-867% 400 °C 33% 67% 5.19 2.50 2.42 sh and decrease in the series 400 °C 33% 5.19 2.50 2.42 to 200 °C can be ascribed to the relative increase in the shunt resistance R
to 200 °C can be ascribed to the relative increase in the shunt resistance R
sh and decrease in the series 2.5
2.0
1.5
1.0
0.5
0.0
0.6 -8
resistance R
s. In contrast, as the film annealing temperature is raised to 400 °C, OPV devices show a resistance R
s. In contrast, as the film annealing temperature is raised to 400 °C, OPV devices show a -10
2.6. Ultraviolet Photoelectron Spectroscopy 2.6. Ultraviolet Photoelectron Spectroscopy 0.00
PCE of 5% with a decrement approximately 15% lower than that of those with unannealed films. It is PCE of 5% with a decrement approximately 15% lower than that of those with unannealed films. It is 0.5 -10
2.2 2.4 2.6 2.8 3.0 3.2
clear from Table 1 that the relatively weak photovoltaic response can be attributed to the decrease in clear from Table 1 that the relatively weak photovoltaic response can be attributed to the decrease in E (eV)
Figure 7 plots the UPS spectra of V
Ox films that have been annealed at different temperatures. Figure 7 plots the UPS spectra of V
2Ox2 films that have been annealed at different temperatures. 16.2
16.1
16.0
15.9
15.80.4
0.0
0.2
0.4
0.8
1.0
6
5
4
3
2
1 0.6 0
open‐circuit voltage, due to the change in chemical structure as will be discussed based on the XPS open‐circuit voltage, due to the change in chemical structure as will be discussed based on the XPS The work function and valence band edge determined by the UPS are also summarised in Table 1. The work function and valence band edge determined by the UPS are also summarised in Table 1. Binding
energy
eV 7000.6
400 0.0 500 0.2 600
800
Binding Energy eV
0.4
0.8
1.0
Voltage
(V)
data later Furthermore, the shunt resistance falls considerably on annealing above above 300 °C 300 °C data on. later on. Furthermore, the shunt resistance considerably on annealing falls film shown Figure the position of the secondary electron cut‐off for unannealed film was As As shown in in Figure 7a, 7a, the position of the secondary electron cut‐off for unannealed was Wavelength
(nm)
2 on annealing at 400 °C due to the recombination process at the 2 on annealing at 400 °C due to the recombination process at the (a) (b)
work Voltage
(V)
reaching a minimum R
sh of 933 Ω∙cm
reaching a minimum R
sh of 933 Ω∙cm
determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
work determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
□) solar cell with (2)
anode interface. This caused a decrease in hole density at the anode interface leading to a decrease in anode interface. This caused a decrease in hole density at the anode interface leading to a decrease in Figure
2. The
density-voltage
characteristics
of PFDT2BT-8:PC70BM
based
function of 5.26 eV. Compared to previous studies, the work function of our s‐V
Ox films is similar to function of 5.26 eV. Compared to previous studies, the work function of our s‐V
2O
x2 films is similar to Figure 2. The current density‐voltage characteristics of PFDT2BT‐8:PC70BM based solar cell with (□) Figure 4. Optical transmission spectra of the s‐V
x films with different Annealing temperature; (
Figure 7. UPS measurements for Vcurrent
2Ox films Span coated on ITO in 2Oambient conditions with no the internal electric field [35,36]. the internal electric field [35,36]. those prepared by different [34]. secondary cut‐off shows those prepared by different On On the shows a a unannealed, (
) 100 °C, (
) 200 °C, (
) 300 °C, and (
) 400 °C. The insert shows the absorption annealing and annealed, (□) unannealed, ( ) [34]. 200 °C, ) 300 °C, (secondary ) 400 °C. electron (a) electron shows cut‐off the Figure 2. The current density‐voltage characteristics of PFDT2BT‐8:PC70BM based solar cell with (□) . the PEDOT:PSS, (
) s‐V2Ox, and (
) MoO
PEDOT:PSS,
(processes )processes s-V2Ox,
and
( annealing, )annealing, MoO
3 .3and 2 as a function of the photon energy. slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the coefficient (αhν)
secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert PEDOT:PSS, (
) s‐V2Ox, and ( ) MoO3. 2.3. Optical Properties 2.3. Optical Properties work function can’t assign these changes directly processing x due work function we we can’t assign these changes directly to to the the processing of of the the s‐Vs‐V
2Ox2 O
due to to the the shows the density of gap states formed about 1 eV below the Fermi level. To explore the effect of thermal annealing (≥100 °C) on the photovoltaic response of our devices, ˝ C) on the photovoltaic response of our devices,
To explore the effect of thermal annealing (ě100
possible influence of adsorbates. Figure 7b displays the onset of the s‐V
Ox valence band to be 2.5 eV possible influence of adsorbates. Figure 7b displays the onset of the s‐V
2Ox2 valence band to be 2.5 eV To understand the changes that we observed in the photovoltaic response as a function of the To understand the changes that we observed in the photovoltaic response as a function of the To explore the effect of thermal annealing (≥100 °C) on the photovoltaic response of our devices, s‐V2O
x films were annealed in air at temperatures of 100, 200, 300, and 400 °C for 30 min before spin ˝ C for 30 min before spin
below the Fermi level which is identical with all samples.
We did not notice any shift for the valence s-V2 Os‐V
were
annealed
intemperature, air atwe temperatures
ofoptical 100,
300,
and
400
below the Fermi level which is identical with all samples.
We did not notice any shift for the valence x films
temperature, investigated the properties of annealed s‐V
2Ox s‐V
films. 4 annealing we investigated the 200,
optical properties of of annealed Ox Figure films. in Figure 4 2Ox films were annealed in air at temperatures of 100, 200, 300, and 400 °C for 30 min before spin coating the annealing active layer. The corresponding current‐voltage characteristic OPVs is 2shown band maximum increase annealing temperature.
Careful observation of valence band band maximum with increase annealing temperature.
Careful observation of characteristic
the the valence band presents the optical transmittance of the films as a function of the wavelength for the as‐deposited presents the optical transmittance of the films as a function of the wavelength for the as‐deposited coating
thewith active
layer.
The corresponding
current-voltage
of
OPVs
is shown
in Figure
3.
coating the active layer. The corresponding current‐voltage characteristic of which OPVs is shown in Figure 3. The photovoltaic parameters obtained are summarised in Table 1 represent the region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below and annealed films. All s‐V
2Ox films show high transmission for wavelengths above 500 nm covering and annealed films. All s‐V
2Ox films show high transmission for wavelengths above 500 nm covering The photovoltaic
parameters
obtained
areobtained summarised
in Table 1 in which
represent
the average
Figure 3. The photovoltaic parameters are summarised Table 1 which represent the of at
average of at least 12 pixels from 18 pixels defined on three separate substrates. The errors quoted Fermi level shown in the inset Figure This suggests that oxygen vacancies shift the the Fermi level as as shown in the inset of of Figure 7b. 7b. This suggests that oxygen vacancies shift the the the majority of the solar spectrum. The transmittance in the visible wavelength range, from 400 to the majority of the solar spectrum. The transmittance in the visible wavelength range, from 400 to least
12
pixels
from
18
pixels
defined
on
three
separate
substrates.
The
errors
quoted
are
defined
by
average of at least 12 pixels from 18 pixels defined on three separate substrates. The errors quoted are defined by the standard deviation about the mean. Fermi level toward conduction band leaving occupied states within the band gap. Fermi level up up toward the the conduction band leaving some occupied states within the band gap. 500 nm, increases slightly by some annealing up to 300 °C which is attributed to change of the of films 500 nm, increases slightly by annealing up to 300 °C which is attributed to change the films are defined by the standard deviation about the mean. Therefore, introducing oxygen vacancies can act as n‐type dopants resulting in a decrease in the theintroducing standard oxygen deviation
about
the
Therefore, vacancies can act mean.
as n‐type resulting in changes. a decrease in annealing the to 400 to refractive index related to chemical structure changes. On annealing °C, 400 the °C, film exhibits refractive index related to dopants chemical structure On the film exhibits 2
work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms absorption peak at 415 nm which could be ascribed to small polaron absorption [37,38].
This effect absorption peak at 415 nm which could be ascribed to small polaron absorption [37,38].
This effect that the high‐temperature annealing of s‐V
O
film produces a change in the chemical structure of 2 xdisordering that the high‐temperature annealing of s‐V
2Ox2 film produces a change in the chemical structure of results results from disordering defects defects in V2Oin x structure leading to transferring charges between from V2Ox structure leading to transferring charges between 0 slightly vanadium oxide layer and, thus, increases slightly hole‐electron recombination the the vanadium oxide layer and, thus, increases the the hole‐electron recombination at at the the neighbouring sites with a significant spread in energy. neighbouring sites with a significant spread in energy. 0
interface. Nevertheless, our experiments demonstrate that x film can used replace interface. Nevertheless, our experiments demonstrate that s‐Vs‐V
2Ox2 O
film can be be used to to replace -2
evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. -2
1.0
1.0
0.00
0.00
0.4 2.2 0.6
1.03.0
2.4 2.22.6 2.40.8
2.8 2.63.0 2.83.2
E (eV) E (eV)
0.4
0.6
0.8
1.0
4Voltage
3 3 (V)
2 2
1 1
0 0
3.2
Binding
eV600
eV
400Binding
500energy
400energy
500
600700 700800
800
Binding
Energy
Binding
Energy
eVeV
Voltage
(V)
Wavelength
(nm)
Wavelength
(nm)
2Ox Figure 3. The current density‐voltage characteristics of organic solar cell with (□) unannealed s‐V
(a)(a)
(b)(b)
2Ox Figure 3. The current density‐voltage characteristics of organic solar cell with (□) unannealed s‐V
interlayer and annealed at ( ) 100 °C, ( ) 200 °C, ( ) 300 °C, and ( ) 400 °C for 30 min. 3. The current
characteristics
of organic
solar
cell
Figure 4. Optical transmission spectra of the s‐V
2O
x films with different Annealing temperature; (
□) 2 Ox □) Figure 4. Optical transmission spectra of the s‐V
2O
x films with different Annealing temperature; (
Figure 7. UPS measurements V
x films Span coated ITO ambient conditions with no (2) unannealed s-V
Figure 7. Figure
UPS measurements for for V2density-voltage
O
x2 O
films Span coated on on ITO in in ambient conditions with no with
interlayer and annealed at ( ) 100 °C, (
) 200 °C, (
) 300 °C, and (
) 400 °C for 30 min. ˝) 100 °C, (
˝and ˝ C,
unannealed, (
) 300 °C, and (
) 400 °C. The insert shows the absorption unannealed, (
) 200 °C, (
annealing and annealed, (□) unannealed, ) 200 °C, ) °C, ) ) 300 °C, and (
400 °C. (a) shows the ˝ C for 30 min.
annealing and annealed, (□) annealed
unannealed, °C, ( () ) 200 °C, (
300 °C, 400 °C. (a) shows interlayer
and
at(( () ) 100 °C, (
)200 100
C,
) 300 200
C,and ( ()) 300
and
( ) 400 °C. The insert shows the absorption )the 400
2 as a function of the photon energy. 2 as a function of the photon energy. coefficient (αhν)
coefficient (αhν)
secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert shows the density of gap states formed about 1 eV below the Fermi level. shows the density of gap states formed about 1 eV below the Fermi level. Materials 2016, 9, 235
4 of 12
Materials 2016, 9, 235 4 of 11 Table 1. Summary of solar cell parameters with s-V2 Ox buffer layer annealed at different temperatures
forTable different
periods of of time.
1. Summary solar cell parameters with s‐V2Ox buffer layer annealed at different temperatures for different periods of time. Ann.
Maximum
Average
Voc
Ann. PCE
Maximum
Voc(V)
Temp.
(%)
PCEAverage (av) (%)
Jsc
Jsc cm´2 )
(mA¨
Rs
Rsh
FF (%)
s cm2 )
sh cm2 )
FF
R
R(Ω¨
(Ω¨
−2
2
2
Temp. PCE (%) PCE
(av) (%) (V) ) (%) ˘ 1.2 (Ω∙cm
) 0.9 (Ω∙cm
Non
6.0
5.8
˘ 0.14
0.86 (mA∙cm
10.0 ˘ 0.22
67.1
11.6 ˘
1264) ˘ 91
Non 0.86 100
6.06.0 5.95.8 ± 0.14 ˘ 0.14
0.86 10.0 ± 0.22 10.1 ˘ 0.13 67.1 ± 1.2 68.3 ˘ 1.7 11.6 ± 0.9 10.5 ˘ 1.0 1264 ± 91 1282 ˘ 215
200100 6.36.0 5.95.9 ± 0.14 ˘ 0.26
0.87 10.1 ± 0.13 10.2 ˘ 0.14 68.3 ± 1.7 67.1 ˘ 2.6 10.5 ± 1.0 10.3 ˘ 0.7 1282 ± 215 1346 ˘ 246
0.86 300
5.6
5.3
˘
0.28
0.83
9.9
˘
0.41
64.3
˘
2.7
12.6
˘
1.7
1095
˘ 238
Materials 2016, 9, 235 4 of 11 200 Materials 2016, 9, 235 6.3 5.9 ± 0.26 0.87 10.2 ± 0.14 67.1 ± 2.6 10.3 ± 0.7 4 of 11 1346 ± 246 400300 5.25.6 5.05.3 ± 0.28 ˘ 0.17
0.83
10.1
˘
0.12
59.6
˘
0.9
17.6
˘
0.4
933
˘ 84
0.83 9.9 ± 0.41 64.3 ± 2.7 12.6 ± 1.7 1095 ± 238 Table 1. Summary of solar parameters with s‐V
2Ox s‐V
buffer at different Table 1. Summary of cell solar cell parameters with 2Ox layer buffer annealed layer annealed at different 400 temperatures for different periods of time. 5.2 5.0 ± 0.17 0.83 10.1 ± 0.12 59.6 ± 0.9 17.6 ± 0.4 933 ± 84 temperatures for different periods of time. It can be seen that the performance of devices annealed at temperatures 100 or 200 ˝ C is quite
Ann. Ann. Maximum
Average Voc Voc Jsc
FF 7 of 11 Maximum
Average Jsc
FF Rs Rs Rsh Rsh Materials 2016, 9, 235 similarIt can be seen that the performance of devices annealed at temperatures 100 or 200 °C is quite Materials 2016, 9, 235 7 of 11 to Temp. thoseTemp. prepared
without
heat(av)treatment.
increase
in PCE
of(Ω∙cm
devices
annealed
up to
−2) slight
2) 2) −2) 2) 2) PCE (%) PCE(av) (%) (V) (mA∙cm
(%) PCE (%) PCE
(Ω∙cm
(Ω∙cm
(%) (V) The
(mA∙cm
(%) (Ω∙cm
˝ C can be ascribed to the relative increase in the shunt resistance R and decrease in the series
similar to those prepared without heat treatment. The slight increase in PCE of devices annealed up 200
Non 6.0 with 5.8 ± 0.14 0.86 10.0 ± 0.22 67.1 ± 1.2 11.6 ± 0.9 1264 ± 91 Non 6.0 5.8 ± 0.14 0.86 10.0 ± 0.22 67.1 ± 1.2 11.6 ± 0.9 1264 ± 91 sh
Table Summary solar cell parameters with x buffer layer annealed different Table 2. 2. Summary of of solar cell parameters s‐Vs‐V
2Ox2 O
buffer layer annealed at at different ˝ C, OPV
to 200 °C can be ascribed to the relative increase in the shunt resistance R
sh and decrease in the series 100 6.0 0.86 0.86 10.1 ± 0.13 68.3 ± 1.7 10.5 ± 1.0 6.0 5.9 ± 0.14 10.1 ± 0.13 1282 ± 215 resistance
R
contrast,
as 5.9 ± 0.14 the film
annealing
temperature
is68.3 ± 1.7 raised
to 10.5 ± 1.0 400 1282 ± 215 devices show a
temperatures for different periods of time. temperatures for different periods of time. s . In 100 resistance R
s. In contrast, as the film annealing temperature is raised to 400 °C, OPV devices show a 200 6.3 0.87 15%
10.2 ± 0.14 67.1 ± 2.6 10.3 ± 0.7 1346 ± 246 200 6.3 5.9 ± 0.26 5.9 ± 0.26 0.87 10.2 ± 0.14 67.1 ± 2.6 10.3 ± 0.7 1346 ± 246 films. It is
PCE
of
5%
with
a
decrement
approximately
lower
than
that
of
those
with
unannealed
4+
5+
4+
5+
Annealing V V
Oxidation V VOxidation Oxidation Work Valence Annealing Oxidation Work Valence 300 300 5.6 0.83 0.83 9.9 ± 0.41 64.3 ± 2.7 12.6 ± 1.7 1095 ± 238 5.6 5.3 ± 0.28 5.3 ± 0.28 9.9 ± 0.41 64.3 ± 2.7 12.6 ± 1.7 1095 ± 238 PCE of 5% with a decrement approximately 15% lower than that of those with unannealed films. It is (eV) can be attributed to the decrease in
clear
from
Table 400 1 that
the
relatively
weak
photovoltaic
response
Temperature State State Function (eV) Band (eV) (eV) Temperature State State Function (eV) Band (eV) 400 5.2 5.0 ± 0.17 0.83 10.1 ± 0.12 59.6 ± 0.9 17.6 ± 0.4 933 ± 84 5.2 5.0 ± 0.17 0.83 10.1 ± 0.12 59.6 ± 0.9 17.6 ± 0.4 933 ± 84 clear from Table 1 that the relatively weak photovoltaic response can be attributed to the decrease in Unannealing 15% voltage, due
85% 5.26 2.53 Unannealing 15% 85% 5.26 2.53 2.5 2.5 as will be discussed based on the XPS
open-circuit
to the change
in chemical
structure
200 °C open‐circuit voltage, due to the change in chemical structure as will be discussed based on the XPS 22.5% 77.5% 5.22 2.53 3.78 200 °C 22.5% 77.5% 2.53 3.78 It can be seen that the performance of devices annealed at temperatures 100 or 200 °C is quite It can be seen that the performance of devices annealed at temperatures 100 or 200 °C is quite data later
on.
Furthermore,
the shunt5.22 resistance
falls
considerably
on annealing above 300 ˝ C reaching
300 °C data similar to those prepared without heat treatment. The slight increase in PCE of devices annealed up 27% 73% 5.23 resistance 2.50 2.68 300 °C 27% 73% 5.23 2.50 2.68 later on. Furthermore, the shunt falls considerably on annealing above 300 °C similar to those prepared without heat treatment. The slight increase in PCE of devices annealed up ˝ C due to the recombination
Rsh of 93367% Ω¨
cm2 on annealing
process at the anode
400 °C a minimum
33% 67% 5.19 at 4002.50 2.50 2.42 sh and decrease in the series 2 on annealing at 400 °C due to the recombination process at the 400 °C 33% 2.42 reaching a minimum R
sh
of 933 Ω∙cm5.19 to 200 °C can be ascribed to the relative increase in the shunt resistance R
to 200 °C can be ascribed to the relative increase in the shunt resistance R
sh and decrease in the series interface.
This
caused
a
decrease
in
hole
density
at
the
anode
interface
leading
to a decrease in the
resistance R
s. In contrast, as the film annealing temperature is raised to 400 °C, OPV devices show a resistance R
s. In contrast, as the film annealing temperature is raised to 400 °C, OPV devices show a anode interface. This caused a decrease in hole density at the anode interface leading to a decrease in 2.6. Ultraviolet Photoelectron Spectroscopy 2.6. Ultraviolet Photoelectron Spectroscopy internal
electric
field
[35,36].
PCE of 5% with a decrement approximately 15% lower than that of those with unannealed films. It is PCE of 5% with a decrement approximately 15% lower than that of those with unannealed films. It is the internal electric field [35,36]. 16.116.1
16.016.0
15.915.9
Binding
Energy
Binding
Energy
eVeV
(a)(a)
15.815.8
6
6 0.4
5
5400
4
0.4
0.02
2.52.0
0.00
(eV/cm) 2
h x 10
Intensity (counts)
Intensity (counts)
0.4
16.216.2
(eV/cm) 2
0.5
10
0.5
2.5
0.02
10
0.6
0.04
2.01.5
0.04
0.02
1.51.0
0.00
1.00.5
0.50.0
0.0

0.6
0.5
h x 10
0.7
(eV/cm) 2
0.7
0.6
0.04

0.8
10
0.7
0.8

0.9
Transmittance
0.9
h x 10
0.8
Transmittance
Intensity
(counts)
Intensity (counts)
Transmittance
clear from Table 1 that the relatively weak photovoltaic response can be attributed to the decrease in clear from Table 1 that the relatively weak photovoltaic response can be attributed to the decrease in Figure 7 plots the UPS spectra of V
Ox films that have been annealed at different temperatures. Figure 7 plots the UPS spectra of V
2Ox2 films that have been annealed at different temperatures. open‐circuit voltage, due to the change in chemical structure as will be discussed based on the XPS open‐circuit voltage, due to the change in chemical structure as will be discussed based on the XPS 2.3. Optical
Properties
The work function and valence band edge determined by the UPS are also summarised in Table 1. The work function and valence band edge determined by the UPS are also summarised in Table 1. 2.3. Optical Properties data later Furthermore, the shunt resistance falls considerably on annealing above above 300 °C 300 °C data on. later on. Furthermore, the shunt resistance falls film considerably on annealing shown Figure the position of the secondary electron cut‐off for unannealed film was As As shown in in Figure 7a, 7a, the position of the secondary electron cut‐off for unannealed was 2 on annealing at 400 °C due to the recombination process at the Toreaching a minimum R
understand
the changes
that2 on annealing at 400 °C due to the recombination process at the we observed
in the photovoltaic
response as a function of the
sh of 933 Ω∙cm
reaching a minimum R
sh of 933 Ω∙cm
determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
work determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
work To understand the changes that we observed in the photovoltaic response as a function of the anode interface. This caused a decrease in hole density at the anode interface leading to a decrease in anode interface. This caused a decrease in hole density at the anode interface leading to a decrease in function of 5.26 eV. Compared to previous studies, the work function of our s‐V
Ox films is similar to annealing
temperature,
properties
of annealed annealeds‐V
s-V2O
Figure
function of 5.26 eV. Compared to previous studies, the work function of our s‐V
2Ox2 films is similar to x films.
2O
annealing temperature, we
we investigated
investigated the
the optical
optical properties of x films. Figure 4 4
the internal electric field [35,36]. the internal electric field [35,36]. those prepared different processes [34]. On annealing, the secondary electron cut‐off shows a those prepared by by different processes [34]. On annealing, the secondary electron cut‐off shows a presents
the optical transmittance of the films as a function of the wavelength for the as-deposited
presents the optical transmittance of the films as a function of the wavelength for the as‐deposited slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the and
annealed
films. All s-V2 O
films show high transmission for wavelengths above 500 nm covering
and annealed films. All s‐V
2Oxx films show high transmission for wavelengths above 500 nm covering 2.3. Optical Properties 2.3. Optical Properties work function can’t assign these changes directly the processing x due the work function we we can’t assign these changes directly to to the processing of of the the s‐Vs‐V
2Ox2 O
due to to the the
majority
of
the
solar
spectrum.
The transmittance
in the visible wavelength range, from 400 to
the majority of the solar spectrum. The transmittance in the visible wavelength range, from 400 to possible influence of adsorbates. Figure 7b displays the onset of the s‐V
2 valence band to be 2.5 eV O
x valence band to be 2.5 eV possible influence of adsorbates. Figure 7b displays the onset of the s‐V
2
O
x
To understand the changes that we observed in the photovoltaic response as a function of the To understand the changes that we observed in the photovoltaic response as a function of the ˝ C which is attributed to change of the films refractive
below the Fermi level which is identical with all samples.
We did not notice any shift for the valence 500500 nm,
increases
slightly
by
annealing
upthe to
300
below the Fermi level which is identical with all samples.
We did not notice any shift for the valence nm, increases slightly by annealing up to °C properties which is of attributed to 2Ochange of the 4 films annealing temperature, we investigated optical properties of annealed s‐V2Ox s‐V
films. 4 annealing temperature, we investigated the 300 optical annealed x Figure films. Figure band maximum with increase annealing temperature.
Careful observation valence band maximum with increase annealing temperature.
Careful changes.
observation of of the the valence band presents the optical transmittance of the films as a function of the wavelength for the as‐deposited presents the optical transmittance of the films as a function of the wavelength for the as‐deposited index
related
to chemical
structure
annealing
toband 400 ˝ C,to the
film
refractive index related to chemical structure On
changes. On annealing 400 °C, exhibits
the film absorption
exhibits region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below and annealed films. All s‐V
2Ox films show high transmission for wavelengths above 500 nm covering and annealed films. All s‐V
2Ox films show high transmission for wavelengths above 500 nm covering absorption peak at 415 nm which could be ascribed to small polaron absorption [37,38].
This effect peak at 415 nm which could be ascribed to small polaron absorption [37,38]. This effect results from
the Fermi level shown in the inset Figure This suggests that oxygen vacancies shift the the Fermi level as as shown in the inset of of Figure 7b. 7b. This suggests that oxygen vacancies shift the the majority of the solar spectrum. The transmittance in the visible wavelength range, from 400 to the majority of the solar spectrum. The transmittance in the visible wavelength range, from 400 to results from disordering defects in V2states Ostates x to
structure leading to transferring charges between disordering
defects
in
V
Ox slightly structure
leading
transferring
charges
between
neighbouring
sites with
2leaving Fermi level toward conduction band occupied within the gap. Fermi level up up toward the the conduction band leaving some occupied within the band gap. 500 nm, increases slightly by some annealing up to 300 °C which is band attributed to change of the of films 500 nm, increases by annealing up to 300 °C which is attributed to change the films neighbouring sites with a significant spread in energy. Therefore, introducing oxygen vacancies can act as n‐type dopants resulting in a decrease in the a significant
spread
in
energy.
Therefore, introducing oxygen vacancies can act as n‐type dopants resulting in a decrease in the refractive index related to chemical structure changes. On annealing to 400 to °C, 400 the °C, film exhibits refractive index related to chemical structure changes. On annealing the film exhibits work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms absorption peak at 415 nm which could be ascribed to small polaron absorption [37,38].
This effect absorption peak at 415 nm which could be ascribed to small polaron absorption [37,38].
This effect that the high‐temperature annealing of s‐V
Oxdisordering film produces a change in the chemical structure of that the high‐temperature annealing of s‐V
2Ox2 film produces a change in the chemical structure of results results from disordering defects defects in V2Oin x structure leading leading to transferring charges between from V2Ox structure to transferring charges between 1.0 the the the vanadium oxide layer and, thus, increases slightly hole‐electron recombination the the vanadium oxide layer and, thus, increases slightly hole‐electron recombination at at the neighbouring sites with a significant spread in energy. neighbouring sites with a significant spread in energy. interface. Nevertheless, our experiments demonstrate that x film can used replace interface. Nevertheless, our experiments demonstrate that s‐Vs‐V
2Ox2 Ofilm can be be used to to replace 0.9
evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. 1.0
1.0
2.2
2.2
0.00
2.4
2.6
2.8
E (eV)
3.0
2.4 2.22.6 2.42.8 2.63.0 2.83.2 3.0
E (eV) E (eV)
3.2
3.2
700
4
1
3 35002 2
1 600
0 0
Binding
eV
eV
400Binding
500energy
600 600
700 700800
400energy
500
Wavelength
(nm)
Wavelength
(nm) (nm)
Wavelength
(b)(b)
800
800
Figure 4. Optical transmission spectra of the s‐V2Ox films with different Annealing temperature; (□) FigureFigure 4. Optical transmission spectra of the s‐V
4.for transmission
spectra
of
the
films
with
2O
xs-V
films with different Annealing temperature; (
□) temperature;
Oxx films with different Annealing temperature; (
□) 22O
Figure 7. UPS measurements VFigure 4. Optical transmission spectra of the s‐V
x films Span coated ITO ambient conditions with no different Annealing
Figure 7. UPS measurements for VOptical
2O
x2 O
films Span coated on on ITO in in ambient conditions with no ˝
˝
˝
˝
unannealed, (
) 100 °C, (
) 200 °C, (
) 300 °C, and (
) 400 °C. The insert shows the absorption unannealed, (
) 300 °C, and (
) 400 °C. The insert shows the absorption unannealed, (
) 200 °C, (
annealing and annealed, (□) unannealed, ) 200 °C, ) °C, ) ) 300 °C, and (
400 °C. (a) shows the C. The insert shows the absorption
annealing and annealed, unannealed, (( () ) 100 °C, (
°C, ) 100 °C, (
( () ) 200 °C, (
300 °C, and 400 °C. (a) shows (2)(□) unannealed,
)200 100
C,
) 300 200
C,and ( ()) 300
C,
and
( ) 400 °C. The insert shows the absorption )the 400
2 as a function of the photon energy. 2 as a function of the photon energy. as a function of the photon energy. coefficient (αhν)
2 2as
coefficient (αhν)
coefficient (αhν)
secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert coefficient
(αhν)
a function
of the photon energy.
shows the density of gap states formed about 1 eV below the Fermi level. shows the density of gap states formed about 1 eV below the Fermi level. Materials 2016, 9, 235
Materials 2016, 9, 235 55 of 11 of 12
Using Tauc’s Law, the optical band gap of s‐V2Ox films was determined [25,27]. The relationship Using Tauc’s Law, the optical band gap of s-V2 Ox films was determined [25,27]. The relationship
between the band gap energy ( ) and the absorption coefficient (α) can be represented as: between the band gap energy (Eg ) and the absorption coefficient (α) can be represented as:
(1) α
∝“
‰1{ N α phνq 9 hν ´ Eg
(1)
where represents the photon energy and is equal to 2. As can be seen in Figure 4, the optical absorption coefficient is plotted as a function of the incident photon energy. By extrapolating where
hν represents theαphoton
energy and N is equal to 2. As can be seen in Figure 4, the optical
from the straight‐line portion of the asplots to the ofenergy‐axis, determined that of the 2
absorption coefficient pαhνq is plotted
a function
the incident we photon
energy. By
extrapolating
unannealed film lies at 2.5 eV which is very close to those reported in literature for from the straight-line portion of the plots to the energy-axis, we determined that Eg of the
solution‐processed samples [24,27,39].
Several studies have reported various values of V
2Ox band unannealed film lies at 2.5 eV which is very close to those reported in literature for solution-processed
gap due to the different processing conditions or the measurement method. For instance, electron samples [24,27,39]. Several studies have reported various values of V2 Ox band gap due to the different
spectroscopy measurements of V2Ox layer spin For
coated in electron
air from isopropanol solution of processing
conditions
or the measurement
method.
instance,
spectroscopy
measurements
vanadium(V) oxitriisopropoxide showed a band gap of
of vanadium(V)
3.6 eV [24]. oxitriisopropoxide
In contrast, the showed
optical of V2 Ox layer spin
coated in air from
isopropanol
solution
absorption experiments of the same precursor revealed a lower band gap of 2.3 eV [25]. In addition, a band gap of 3.6 eV [24]. In contrast, the optical absorption experiments of the same precursor
impact the processing conditions on Inthe V2Ox band gap has processing
been reported by other authors revealedof a lower
band gap of
2.3 eV [25].
addition,
impact
of the
conditions
on the
V2 Ox
[24,27,39–41]. For example, V
2Ox samples spin coated and annealed under N2 atmosphere had a band band gap has been reported by other authors [24,27,39–41]. For example, V2 Ox samples spin coated
gap of 3.2 eV under
while N
fabrication of the films in air showed lower band gaps of 2.5 eV [40]. On and annealed
2 atmosphere had a band gap of 3.2 eV while fabrication of the films in air
annealing s‐V
2Ox films to 200 °C, the band gap energy rises to 2.78 eV. On annealing, showed lower band gaps of 2.5 eV [40]. On annealing s-V2 Ox films to 200 ˝ C,further the band
gap energy reduces again to 2.42 eV at an annealing temperature of 400 °C.
rises to 2.78 eV. On further annealing, Eg reduces again to 2.42 Looking at the individual values of eV at an annealing temperature of
J400
sc in Table 1 it can be individual
observed values
that the in 1 the band gap is independent of the inOPV ˝ C.
Looking
at the
of variation Jsc in Table
it can
be observed
that the variation
the
performance. band gap is independent of the OPV performance.
2.4. Atomic Force Microscopy 2.4. Atomic Force Microscopy
AFM performed to investigate the impact of
of thermal
thermal annealing
annealing on
on surface
surface AFM scanning scanning was was performed
to investigate
the impact
topography of s‐V
2
O
x
. Figure 5 shows 2 μm × 2 μm AFM scans for an unannealed film (a), and films topography of s-V2 x . Figure 5 shows 2 µm ˆ 2 µm AFM scans for an unannealed film (a), and films
˝ C for thermally‐annealed ((b) and
and (c),
(c), respectively).
respectively). RMS of thermally-annealed at at 200 200 ˝°Cand C and 400 400 °C for 30 30 min min ((b)
RMS roughness roughness of
films decreased slightly upon annealing, from 2.5 nm for an as‐cast precursor film to 1.9 nm for films films decreased slightly upon annealing, from 2.5 nm for an as-cast precursor film to 1.9 nm for
˝ C, with no local crystallisation structure observed. The low RMS
annealed at 400 °C, with no local crystallisation structure observed. The low RMS of V
2Ox annealed films annealed at 400
of V2 Ox
samples short circuit into OPV devices as shown in annealed avoided samples avoided
shortfaults circuitwhen faults they whenwere they incorporated were incorporated
into OPV
devices
as shown
Table 1. in Table 1.
(a) (b)
(c) Figure 5. AFM topography (2 μm × 2 μm) of a 10 nm thick s‐V2Ox layer deposited on top of Si with Figure 5. AFM topography (2 µm ˆ 2 µm) of a 10 nm thick s-V2 Ox layer deposited on top of Si with
native oxide; (a) as deposited; (b) annealed at 200 °C; (c) annealed at 400 °C. The average grain size native oxide; (a) as deposited; (b) annealed at 200 ˝ C; (c) annealed at 400 ˝ C. The average grain size
was 21.3 nm, 16.8 nm, and 19.6 nm, respectively. was 21.3 nm, 16.8 nm, and 19.6 nm, respectively.
2.5. X‐ray Photoelectron Spectroscopy 2.5. X-ray Photoelectron Spectroscopy
We investigated the chemical composition and electronic structure of s‐V2Ox thin films by We investigated the chemical composition and electronic structure of s-V2 Ox thin films by
photoelectron spectroscopy (UPS and XPS). Figure 6a shows the XPS spectrum of an as‐cast photoelectron spectroscopy (UPS and XPS). Figure 6a shows the XPS spectrum of an as-cast precursor
precursor film prepared in ambient conditions. The O1s signal, determined at 530.0 eV as reported in film prepared in ambient conditions. The O1s signal, determined at 530.0 eV as reported in
literature [42], was used as a binding energy reference. Therefore, we find V2p1/2, and V2p3/2 lines literature [42], was used as a binding energy reference. Therefore, we find V2p1/2 , and V2p3/2
correspond to 517.1 eV and 524.6 eV, respectively, which is in good agreement with the commonly lines correspond to 517.1 eV and 524.6 eV, respectively, which is in good agreement with the commonly
reported values [42–46]. reported values [42–46].
Materials 2016, 9, 235
6 of 12
Materials 2016, 9, 235 6 of 11 (a) (b)
Figure 6. XPS spectra of s‐V2Ox thin film deposited in air. (a) The solid line represents the Figure 6. XPS spectra of s-V2 Ox thin film deposited in air. (a) The solid line represents the experimental
5+ peaks are experimental XPS spectra; the dashed lines are decomposed XPS; the binding energy of V
XPS spectra; the dashed lines are decomposed XPS; the binding energy of V5+ peaks are higher than
higher than V4+ peaks (b) Photoelectron spectra of (i) unannealed film and those annealed at (ii) 200 V4+ peaks (b) Photoelectron spectra of (i) unannealed film and those annealed at (ii) 200 ˝ C; (iii) 300 ˝ C,
°C; (iii) 300 °C, and (iv) 400 °C. and
(iv) 400 ˝ C.
5+ and Decomposition analysis reveals that V2p3/2 peak consists of two different species (i.e., V
5+ and
Decomposition
analysis
reveals
that
V2p
peak
consists
of
two
different
species
(i.e.,
V
V4+ oxidation state) with a ratio of 5.7:1. The 3/2
peak position Of V4+ and V5+ in V2p3/2 spectrum was V4+ oxidation state) with a ratio of 5.7:1. The peak position Of V4+ and V5+ in V2p3/24+ spectrum was
identified by using L‐G fitting to be 515.8 eV and 517.1 eV, respectively, in which V oxidation state identified by using L-G fitting to be 515.8 eV and 517.1 eV, respectively, in which V4+ oxidation state
arises as a low‐energy shoulder in the core level of V2p3/2. The calculated composition analyses arises as a low-energy shoulder in the core level of V2p3/2 . The calculated composition analyses
indicate that a small amount of oxygen vacancy exists in unannealed s‐V2Ox films. Fabrication indicate that a small amount of oxygen vacancy exists in unannealed s-V2 Ox films. Fabrication4+
conditions such as atmospheric gases and thermal annealing can increase the concentration of V
conditions such as atmospheric gases and 5+thermal
can increase the concentration of V4+
4+. annealing
species leading to the reduction of the V
to V
For instance, H
2O molecules in air facilitate the species
leading to the reduction of the V5+ to V4+ . For instance, H2 O molecules in air facilitate the
oxygen removal such that oxygen‐hydrogen interaction weakens the binding energy of the oxygen oxygen
removal such that oxygen-hydrogen interaction weakens the binding energy of the oxygen
atom with the neighbouring vanadium atoms [47,48]. Moreover, a previous study demonstrated that atom
with
the neighbouring vanadium atoms [47,48]. Moreover, a previous study demonstrated
that
V2O5 can be reduced during XPS measuring in which X‐rays create vanadyl oxygen atoms pV
“
Oq
Vvacancies [49]. O
can
be
reduced
during
XPS
measuring
in
which
X-rays
create
vanadyl
oxygen
atoms
2 5
vacancies
[49].
Figure 6b exhibits a comparison between the XPS spectra of (i) unannealed s‐V2Ox films and Figure
6b exhibits
comparison
between
the XPS spectra
of (i)°C, unannealed
s-Vand and°C. those
2 Ox films
those annealed in air aat three different temperatures (ii) 200 (iii) 300 °C, (iv) 400 As ˝
˝
˝
annealed
in
air
at
three
different
temperatures
(ii)
200
C,
(iii)
300
C,
and
(iv)
400
C.
As
observed,
observed, the binding energies of the V2p3/2 and O1s core levels do not change appreciably with the
binding energies of the V2p3/2 and O1s core levels do not change appreciably with3/2
increasing
increasing temperature. Furthermore, as the annealing temperature increases, the V 2p
spectrum temperature.
Furthermore,
as
the
annealing
temperature
increases,
the
V
2p
spectrum
becomes
3/2
becomes broader at low binding energies due to the increase of V4+ species. Table 1 summarises the 4+
broader
energies
due
to the increase
of V species.
Table 1 which summarises
the change
change at
of low
the binding
oxidation state of vanadium with annealing temperature is consistent with of
the
oxidation
state
of
vanadium
with
annealing
temperature
which
is
consistent
with
previous
previous studies [50,51]. By deconvolution of V2p3/2 lines, we did not observe any shift in the studies
[50,51]. By deconvolution
of V2p3/2 lines, we
did not observe any shift in the4+binding
energy
4+ and V5+ peaks in any cases.
binding energy of V
In addition, it is evident that V
oxidation state 4+
5+
4+
of
V steadily and V with peaksincreasing in any cases.
In addition,
it is evident
that V Voxidation
state rises steadily
rises annealing temperatures. Therefore, 2Ox thin films are gradually with
increasing
annealing
temperatures.
Therefore,
V
O
thin
films
are
gradually
reduced
owing
2 x
reduced owing to the generation of more oxygen vacancies. This partial reduction leads to an to
the
generation
of
more
oxygen
vacancies.
This
partial
reduction
leads
to
an
increase
in
the
V3d
increase in the V3d electron density [52]. Furthermore, oxygen vacancies in V2Ox can cause electron
density
[52].
Furthermore,
oxygen
vacancies
in
V
O
can
cause
localisation
excess
electrons
in
x
2
localisation excess electrons in unfilled 3d orbitals which can explain the absorption peak near to the unfilled
3d
orbitals
which
can
explain
the
absorption
peak
near
to
the
ultraviolet
band
observed
in
ultraviolet band observed in Figure 4 [37]. From Table 2 we conclude that vanadium oxide can be Figure
4 [37]. From Table 2 we conclude that vanadium oxide can be partially
reduced by a thermal
partially reduced by a thermal annealing process in air for 30 min from V
2O4.93 to V6O14. Increasing annealing
process
in
air
for
30
min
from
V
O
to
V
O
.
Increasing
the
annealing
temperature can
2 4.93
6 14
the annealing temperature can result in further reduction as reported previously [50]. resultIn the literature, there have been detailed investigations on the physical properties and gradual in further reduction as reported previously [50].
reduction of vanadium oxide by heating at high temperatures [29,50]. Materials 2016, 9, 235
7 of 12
Table 2. Summary of solar cell parameters with s-V2 Ox buffer layer annealed at different temperatures
for different periods of time.
Materials 2016, 9, 235 7 of 11 Work Function
Valence Band
Annealing
V4+ Oxidation
V5+ Oxidation
Eg (eV)
(eV)
(eV)
Temperature
State
State
Table 2. Summary of solar cell parameters with s‐V2Ox buffer layer annealed at different Unannealing
15%
85%
5.26
2.53
2.5
temperatures for different periods of time. 200 ˝ C
22.5%
77.5%
5.22
2.53
3.78
Materials 2016, 9, 235 5+
300 ˝ C
73%Oxidation 5.23 Work 2.50
Annealing V4+27%
Oxidation V
Valence 2.68
˝C
400
33%
67% State 5.19
(eV) Temperature State Function (eV) 2.50
Band (eV) 2.42
4 of 11 Table 1. Summary of solar cell parameters with s‐V2Ox buffer layer annealed at different Unannealing 15% temperatures for different periods of time. 85% 5.26 2.53 2.5 200 °C there have
22.5% 77.5% Average onVoc
5.22 physical
2.53 3.78 In the literature,
been
detailed
investigations
the
properties
and
gradual
Ann. Maximum
Jsc
FF
Rs Rsh Materials 2016, 9, 235 7 of 11 Materials 2016, 9, 235 7 of 11 27% 5.23 (mA∙cm−2) 2.50 2.68 Temp. atPCE (%) PCE(av) (%) [29,50].
(Ω∙cm2) (Ω∙cm2) (V) (%) reduction of300 °C vanadium oxide by
heating
high73% temperatures
Non 6.0 5.8 ± 0.14 10.0 ± 0.22 67.1 ± 1.2 11.6 ± 0.9 1264 ± 91 400 °C 33% 67% with 5.19 2.50 2.42 Table 2. Summary of solar cell cell parameters s‐V2O
x 2buffer layer annealed at different Table 2. Summary of solar parameters with s‐V
Ox0.86 buffer layer annealed at different 100 6.0 5.9 ± 0.14 0.86 10.1 ± 0.13 68.3 ± 1.7 10.5 ± 1.0 1282 ± 215 temperatures for different periods of time. temperatures for different periods of time. 2.6. Ultraviolet Photoelectron
Spectroscopy
200 6.3 5.9 ± 0.26 0.87 10.2 ± 0.14 67.1 ± 2.6 10.3 ± 0.7 1346 ± 246 2.6. Ultraviolet Photoelectron Spectroscopy 5+ Oxidation Annealing V5+ Oxidation Work Valence Annealing V4+ Oxidation V4+ Oxidation Work Valence 5.6 V that
5.3 ± 0.28 0.83 9.9 ± 0.41 64.3 ± 2.7 12.6 ± 1.7 1095 ± 238 Figure 7 plots the UPS spectra 300 of V2 Ox films
have been
annealed
at different
temperatures.
(eV) Temperature State State Function (eV) Band (eV) (eV) Temperature State State Function (eV) Band (eV) 400 5.2 5.0 ± 0.17 0.83 10.1 ± 0.12 59.6 ± 0.9 17.6 ± 0.4 933 ± 84 Figure 7 plots the UPS spectra of V
2
O
x
films that have been annealed at different temperatures. The work function
and valence band
edge determined
are also 2.53 summarised
Unannealing 15% 15% 85% 85% by the UPS
5.26 5.26 Unannealing 2.53 2.5 2.5 in Table 1.
The work function and valence band edge determined by the UPS are also summarised in Table 1. As shown in Figure
7a,
the position
of the secondary
cut-off for2.53 unannealed
200 °C 22.5% 77.5% 5.22 5.22 200 °C 22.5% 77.5% electron
2.53 3.78 3.78 film was
It can be seen that the performance of devices annealed at temperatures 100 or 200 °C is quite As shown 7a, the similar to those prepared without heat treatment. The slight increase in PCE of devices annealed up position of 15.94
the secondary cut‐off unannealed film 27% 27% 73% 5.23 5.23 2.50 300 °C 73% electron 2.50 2.68 2.68 determined
toin beFigure at300 °C the
binding
energy
of
eV
below
the Fermi
level,for corresponding
to a was work
400 °C 33% 67% 5.19 2.50 2.42 400 °C 33% 67% 5.19 2.50 2.42 determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
work to 200 °C can be ascribed to the relative increase in the shunt resistance R
sh
and decrease in the series function of 5.26 eV. Compared to previous studies, the work function of our s-V2 Ox films is similar
function of 5.26 eV. Compared to previous studies, the work function of our s‐V
2Ox films is similar to resistance Rs. In contrast, as the film annealing temperature is raised to 400 °C, OPV devices show a to those
prepared
by different processes
[34]. On annealing, the secondary electron
cut-off shows a
2.6. Ultraviolet Photoelectron Spectroscopy 2.6. Ultraviolet Photoelectron Spectroscopy PCE of 5% with a decrement approximately 15% lower than that of those with unannealed films. It is those prepared by different processes [34]. On annealing, the secondary electron cut‐off shows a slight shift up toFigure 7 plots the UPS spectra of V
0.07 eV towardclear from Table 1 that the relatively weak photovoltaic response can be attributed to the decrease in the higher binding
energy.
While
we
see
a
0.07
eV
change
in the work
2Ox films that have been annealed at different temperatures. Figure 7 plots the UPS spectra of V
2Ox films that have been annealed at different temperatures. slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the function weThe work function and valence band edge determined by the UPS are also summarised in Table 1. can’t
assign these open‐circuit voltage, due to the change in chemical structure as will be discussed based on the XPS changes directly to the processing of the s-V2 Ox due to the possible
The work function and valence band edge determined by the UPS are also summarised in Table 1. work function we can’t assign these changes directly to the processing of the s‐V2Ox due to the data later on. of Furthermore, shunt falls considerably on annealing shown in Figure 7a, the displays
position the secondary electron for unannealed film was As shown in Figure 7a, the position of onset
the secondary electron cut‐off for unannealed film was influence ofAs adsorbates.
Figure
7b
the
ofthe the
s-V
Ocut‐off band
to be
2.5
eV
belowabove 300 °C x valence
2resistance possible influence of adsorbates. Figure 7b displays the onset of the s‐V
2Ox valence band to be 2.5 eV reaching a minimum Rsh of 933 Ω∙cm2 on annealing at 400 °C due to the recombination process at the determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
work determined to be at the binding energy of 15.94 eV below the Fermi level, corresponding to a
work the Fermi level which is identical with all samples. We did not notice any shift for the valence band
below the Fermi level which is identical with all samples.
We did not notice any shift for the valence anode interface. This caused a decrease in hole density at the anode interface leading to a decrease in function of 5.26 eV. Compared to previous studies, the work function of our s‐V
2Ox films is similar to function of 5.26 eV. Compared to previous studies, the work function of our s‐V
2Ox films is similar to maximum
with
increase
annealing
temperature.
Careful
observation
ofelectron the valence
band a region of
the internal electric field [35,36]. band maximum with by increase annealing temperature.
Careful observation of the valence those prepared different processes [34]. On On annealing, the the secondary cut‐off shows those prepared by different processes [34]. annealing, secondary electron cut‐off shows band a ˝ C and 400 ˝ C reveals the formation of more gap states at 1 eV below the Fermi
films
annealed
at
300
slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the slight shift up to 0.07 eV toward the higher binding energy. While we see a 0.07 eV change in the region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below 2.3. Optical Properties work we of
can’t assign these changes directly to oxygen
the processing the s‐V2s‐V
Oxthe
2due to the work function we can’t assign these changes to the processing of shift
the Ox Fermi
due to the the level
asFermi shown
infunction the
inset
Figure
7b.
This
suggests
that
vacancies
level
up
the level as shown in the inset of Figure 7b. directly This suggests that of oxygen vacancies shift possible influence of adsorbates. Figure 7b displays the onset of the s‐V
2Ox valence band to be 2.5 eV possible influence of adsorbates. Figure 7b displays the onset of the s‐V
2
O
x
valence band to be 2.5 eV To understand the changes that we observed in the photovoltaic response as a function of the toward
the
conduction
band
leaving
some
occupied
states
within
the
band
gap.
Therefore,
introducing
Fermi level up toward the conduction band leaving some occupied states within the band gap. below the Fermi level which is identical with all samples.
We did not notice any shift for the valence below the Fermi level which is identical with all samples.
We did not notice any shift for the valence annealing temperature, we in
investigated the optical properties of annealed s‐V2Ox films. Figure 4 oxygen
vacancies
can
act as
n-type
dopants
resulting
a decrease
inobservation the
work
function
of
vanadium
Therefore, introducing oxygen vacancies can act as n‐type dopants resulting in the a decrease in the band maximum with increase annealing temperature.
Careful observation of the valence band band maximum with increase annealing temperature.
Careful of valence band presents the optical transmittance of the films as a function of the wavelength for the as‐deposited oxide
[53]. Consequently,
photoelectron
spectroscopy analysis confirms that the high-temperature
work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below region of films annealed at 300 °C and 400 °C reveals the formation of more gap states at 1 eV below and annealed films. All s‐V2Ox films show high transmission for wavelengths above 500 nm covering Fermi level as shown in the inset of Figure 7b. 7b. This suggests that oxygen vacancies shift the the Fermi level as shown in the inset of Figure This suggests vacancies shift the and,
annealing
ofthe s-V
O
film
produces
a
change
in2O
the
chemical
structure
ofthat theoxygen vanadium
oxide
layer
that the high‐temperature annealing of s‐V
x film produces a change in the chemical structure of the majority of the solar spectrum. The transmittance in the visible wavelength range, from 400 to 2 x
Fermi level up toward the conduction band leaving some occupied states within the band gap. Fermi level up toward the conduction band leaving some occupied states within the band gap. 500 nm, increases slightly by annealing up to 300 °C which is attributed to change the vanadium oxide layer and, thus, increases slightly the hole‐electron recombination at the of the films thus, increases slightly the hole-electron recombination at the interface. Nevertheless, our experiments
Therefore, introducing oxygen vacancies can can act as n‐type dopants resulting in On a in decrease in to the Therefore, introducing oxygen vacancies act as n‐type dopants resulting a decrease in the °C, the film exhibits refractive index related to chemical structure changes. annealing 400 interface. that
Nevertheless, our demonstrate that s‐V
2Ox oxides
film can be used to replace demonstrate
s-V2 Ox film
canexperiments be used to replace
evaporated
metal
in optoelectronic
devices
work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms work function of vanadium oxide [53]. Consequently, photoelectron spectroscopy analysis confirms absorption peak at 415 nm which could be ascribed to small polaron absorption [37,38]. This effect evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. which
are fabricated
at
high
temperatures.
that the high‐temperature annealing of s‐V
2Ox film produces a change in the chemical structure of that the high‐temperature annealing of s‐V
2Ox film produces a change in the chemical structure of results from disordering defects in V2Ox structure leading to transferring charges between Intensity
Intensity
(counts)(counts)
Intensity (counts)
the the vanadium oxide layer and, thus, increases slightly the the hole‐electron recombination at the vanadium oxide layer and, thus, increases slightly hole‐electron recombination at the neighbouring sites with a significant spread in energy. interface. Nevertheless, our our experiments demonstrate that that s‐V2s‐V
Ox 2film can can be used to replace interface. Nevertheless, experiments demonstrate Ox film be used to replace evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. evaporated metal oxides in optoelectronic devices which are fabricated at high temperatures. 1.0
16.1
15.9
16.2 Binding
16.2 16.1 Energy
16.1 16.0eV
16.0
15.8
15.9 15.9
Binding
Energy
eV eV
Energy
(a)
Binding
(a) (a) 15.8 15.8
(eV/cm) 2
Intensity (counts)
10
0.7
0.6
0.5
6
0.46
0.04
2.5
2.0
1.5
1.0
0.5
0.0
0.02
2.5
2.0
2.5
1.5
2.0
1.0
1.5
0.5
1.0
0.0
0.5
0.0

Transmittance
16.0
0.8
h x 10
16.2
Intensity (counts)
Intensity (counts)
0.9
5
6 5
400
4
0.00
2.2
3
2.4
22.6
2.8 1 3.0
2 1
1 0
E (eV)
Binding energy eV
5 4
4 3
3 2
Binding
eV
Binding
500energy
600eV
(b)energy
700
3.20
0
800
Wavelength
(nm)
(b) (b)
Figure 7. UPS measurements for V2Ox films Span coated on ITO in ambient conditions with no Figure 7. UPS
measurements
forFigure 4. Optical transmission spectra of the s‐V
V2for OxVfor coated
onITO ITO
conditions
2ambient Ox films with different Annealing temperature; (
□) Figure 7. UPS measurements 2films
OV
x 2films Span coated on in in
ambient conditions with with no with
Figure 7. UPS measurements Ox Span
films Span coated on ITO in ambient
conditions no no
annealing and annealed, (□) unannealed, ( ) 200 ˝°C, ( ) 300 °C, and ( ) 400 °C. (a) shows the ˝ C,
˝ C.
unannealed, (
) 300 °C, and (
) 400 °C. The insert shows the absorption annealing and and annealed, (□) unannealed, () 200
) (200 °C, () 200 °C, (
(°C, ) °C, and ( ) ( 400 °C. (a) shows the annealing annealed, (□) unannealed, ) 200 ) 300 °C, and
and °C. (a) (a)
shows the the
annealing and
annealed,
(2)
unannealed,
( ) 100 °C, (
C,
)(300 300
)) 400 400
shows
secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert 2 as a function of the photon energy. secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert secondary electron cut‐off region; and (b) shows the expanded region near the Fermi level; the insert secondary electron
cut-off region;coefficient (αhν)
and (b) shows
the expanded region near the Fermi level; the insert
shows the density of gap states formed about 1 eV below the Fermi level. shows the density of gap states formed about 1 eV below the Fermi level. shows the density of gap states formed about 1 eV below the Fermi level. shows the density of gap states formed about 1 eV below the Fermi level.
Materials 2016, 9, 235
8 of 12
3. Materials and Methods
3.1. Materials
Vanadium(V) oxytriisopropoxide was purchased from Sigma-Aldrich (Dorset, UK) and was
mixed with iso-propanol (99.5%) at a 1:250 volume ratio to obtain a solution with a concentration of
4 mg¨ mL´1 . PEDOT:PSS was purchased from Ossila Ltd., Sheffield, UK, MoO3 (99.95%) was purchased
from Testbourne Ltd (Basingstoke, UK), aluminium (99.99%) and calcium (99%) were purchased from
Sigma-Aldrich (Dorset, UK). The donor polymer PFDT2BT-8 was synthesised in the Department of
Chemistry at the University of Sheffield via a previously reported method, and had a molecular weight
of 91.6 kDa and a polydispersity index (PDI) of 1.47 [32]. PC70 BM was purchased from Ossila Ltd
with a purity of 95% (5% PC60 BM). The active layer solution was prepared by mixing PFDT2BT-8 and
PC70 BM at a weight ratio of 1:4 in chloroform with an overall concentration of 20 mg¨ mL´1 .
3.2. OPV Device Fabrication
All OPV devices were fabricated onto pre-patterned ITO coated glass substrates purchased from
Ossila Ltd (Sheffield, UK). Prior to use, the substrates were sonicated in a warm cleaning solution of
Hellmanex (2 wt %) for 10 min at 70 ˝ C. After sonication, they were washed with the de-ionized water.
They were then placed in isopropanol and sonicated for 10 min at 70 ˝ C. Finally, they were dried
with nitrogen gas. Thin films (~5 nm) of s-V2 Ox were deposited via spin coating onto cleaned ITO
substrates in ambient conditions and were then transferred into a dry nitrogen glove box. A thermal
evaporated MoO3 film (10 nm) was deposited at a rate of 3 Ũ s´1 and at a pressure of ~10´7 mbar in
a high-vacuum system in the glove box. PEDOT:PSS film with (30 nm ˘ 3 nm) thick was spin cast
in ambient conditions and was then annealed at 130 ˝ C for an hour in the glove box environment.
Semiconducting thin films were prepared by spin casting the solution onto a substrate at a spin speed
of 3000 rpm in order to obtain an active layer with a thickness of 70 nm ˘ 4 nm. The substrates were
then transferred within the glove box to a high-vacuum system (10´7 mbar) to thermally evaporate
the cathodes. The bi-layer cathodes of Ca (5 nm) and Al (100 nm) were evaporated at a rate of 3 Ũ s´1
and 10 Ũ s´1 , respectively. The final step in device fabrication was encapsulation of the central area of
each substrate by using a glass slide and light-curable epoxy, to extend the lifetime for measurement
and storage.
3.3. Current-Density Characterisation
OPV devices were measured under ambient conditions using a Keithley 2400 source meter
(Tektronix Ltd., Bracknell, UK) and a Newport 92251A-1000 AM1.5 solar simulator (Newport Co.,LTD,
Didcot, UK). An NREL calibrated silicon diode was used to calibrate the power output at 1000 W¨ cm´2 .
Each device had an area of approximately 2.6 mm2 as defined by a shadow mask.
3.4. Atomic Force Microscopy
A Veeco Instruments Dimension 3100 was used to perform the AFM. Aluminium-coated silicon
tips from Budget Sensors (Tap 300 Al-G) with a resonance frequency of 300 kHz and a spring constant
of 40 N¨ m´1 were used throughout. The obtained data were processed using the Gwyddion.
3.5. Photoelectron Spectroscopy
The UPS and XPS measurements were carried out using KratosUltra AXIS photoelectron
spectroscopy (Kratos Analytical Ltd., Manchester, UK). XPS measurements were taken using the
Al Kα radiation with an energy of 1486.6 eV as the excitation source, a band pass energy of 10 eV,
a step size of 0.025 eV, and a dwell time of 250 ms. UPS measurements were taken using the He (I)
emission line (21.2 eV), a band pass energy of 10 eV, a step size of 0.025 eV, and a dwell time of 250 ms.
Materials 2016, 9, 235
9 of 12
3.6. Absorption Spectroscopy
Absorption measurements were conducted under ambient conditions using a Horiba Fluoromax
4 spectrometer (HORIBA Ltd., Stanmore Middlesex, UK) with a xenon arc lamp (200 to 950 nm).
Transmittance and absorbance of all the samples in this work were measured with the range 300–900 nm
in increments of 2 nm and a slit size of 2 nm. Transmittance spectra were normalised against the
power output recorded by the reference detector to account for any potential variations in light output
between the separate measurements.
4. Conclusions
In contrast to previous studies, we have demonstrated that thermal annealing of V2 Ox thin films
deposited from a vanadium(V) oxytriisopropoxide does not significantly enhance the performance
of OPV devices. However, the efficiency of OPV devices is reduced by 15% after annealing s-V2 Ox
layers at high temperatures (i.e., 400 ˝ C) due to the decrease of Voc and shunt resistance. Absorption
spectroscopy studies reveal that OPV performance is independent of the variable optical band gap
of s-V2 Ox . XPS analysis shows that annealing of vanadium oxide films for 30 min in air resulted in a
reduction of V2 O4.93 to V6 O14 as a result of the generation of more oxygen vacancies. This reduction
causes a decrease of Voc and, thus, the PCE of OPV device. Nevertheless, OPV results confirm that
s-V2 Ox thin film is suitable for optoelectronic devices which are fabricated at high temperatures, and is
promising for efficient OPV with a low cost manufacturing process.
Acknowledgments: We thank the UK EPSRC for funding the project “Polymer fullerene photovoltaic devices:
New materials and innovative processes for high-volume manufacture” (EP/I02864/1) and “Photovoltaics
for Future Societies” (EP/I032541/1). A. Alsulami thanks the Saudi Cultural Bureau in London, UK,
and King Abdulaziz City for Science and Technology (KACST) in Riyadh, Saudi Arabia, for the provision
of a PhD scholarship.
Author Contributions: Abdullah Alsulami, Jonathan Griffin and Rania Alqurashi fabricated the OPV devices,
performed the experiments and contributed to manuscript writing. Ahmed Iraqi and Hunan Yi synthesised
PFDT2BT-8, measured its electronic structure and commented on various aspects during the manuscript
preparation. Alastair Buckley and David Lidzey conceived and designed the overall research concept, analyzed
the data, contributed to manuscript writing and supervised the work.
Conflicts of Interest: The authors declare no conflict of interest. The founding sponsors had no role in the design
of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the
decision to publish the results.
References
1.
2.
3.
4.
5.
6.
7.
8.
Timilsina, G.R.; Kurdgelashvili, L.; Narbel, P.A. Solar energy: Markets, economics and policies.
Renew. Sustain. Energy Rev. 2012, 16, 449–465. [CrossRef]
Nielsen, T.D.; Cruickshank, C.; Foged, S.; Thorsen, J.; Krebs, F.C. Business, market and intellectual property
analysis of polymer solar cells. Sol. Energy Mater. Sol. Cells 2010, 94, 1553–1571. [CrossRef]
Li, G.; Zhu, R.; Yang, Y. Polymer solar cells. Nat. Photonics 2012, 6, 153–161. [CrossRef]
Shah, A.; Torres, P.; Tscharner, R.; Wyrsch, N.; Keppner, H. Photovoltaic technology: The case for thin-film
solar cells. Science 1999, 285, 692–698. [CrossRef] [PubMed]
Guenes, S.; Neugebauer, H.; Sariciftci, N.S. Conjugated polymer-based organic solar cells. Chem. Rev. 2007,
107, 1324–1338. [CrossRef] [PubMed]
Delgado, J.L.; Bouit, P.A.; Filippone, S.; Herranz, M.A.; Martin, N. Organic photovoltaics: A chemical
approach. Chem. Commun. 2010, 46, 4853–4865. [CrossRef] [PubMed]
Liu, Y.; Zhao, J.; Li, Z.; Mu, C.; Ma, W.; Hu, H.; Jiang, K.; Lin, H.; Ade, H.; Yan, H. Aggregation and
morphology control enables multiple cases of high-efficiency polymer solar cells. Nat. Commun. 2014, 5.
[CrossRef] [PubMed]
Steim, R.; Kogler, F.R.; Brabec, C.J. Interface materials for organic solar cells. J. Mater. Chem. 2010,
20, 2499–2512. [CrossRef]
Materials 2016, 9, 235
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
10 of 12
Arias, A.C.; Granstrom, M.; Thomas, D.S.; Petritsch, K.; Friend, R.H. Doped conducting-polymer
-semiconducting-polymer interfaces: Their use in organic photovoltaic devices. Phys. Rev. B 1999,
60, 1854–1860. [CrossRef]
Ko, C.J.; Lin, Y.K.; Chen, F.C.; Chu, C.W. Modified buffer layers for polymer photovoltaic devices.
Appl. Phys. Lett. 2007, 90. [CrossRef]
Lee, K.; Kim, J.Y.; Park, S.H.; Kim, S.H.; Cho, S.; Heeger, A.J. Air-stable polymer electronic devices. Adv. Mater.
2007, 19. [CrossRef]
Norrman, K.; Madsen, M.V.; Gevorgyan, S.A.; Krebs, F.C. Degradation patterns in water and oxygen of an
inverted polymer solar cell. J. Am. Chem. Soc. 2010, 132, 16883–16892. [CrossRef] [PubMed]
Ratcliff, E.L.; Zacher, B.; Armstrong, N.R. Selective inter layers and contacts in organic photovoltaic cells.
J. Phys. Chem. Lett. 2011, 2, 1337–1350. [CrossRef] [PubMed]
De Jong, M.P.; van Ijzendoorn, L.J.; de Voigt, M.J.A. Stability of the interface between indium-tin-oxide and
poly(3,4-ethylenedioxythiophene)/poly(styrenesulfonate) in polymer light-emitting diodes. Appl. Phys. Lett.
2000, 77, 2255–2257. [CrossRef]
Shrotriya, V.; Li, G.; Yao, Y.; Chu, C.W.; Yang, Y. Transition metal oxides as the buffer layer for polymer
photovoltaic cells. Appl. Phys. Lett. 2006, 88. [CrossRef]
Park, J.H.; Lee, T.W.; Chin, B.D.; Wang, D.H.; Park, O.O. Roles of interlayers in efficient organic photovoltaic
devices. Macromol. Rapid Commun. 2010, 31, 2095–2108. [CrossRef] [PubMed]
Chen, S.; Manders, J.R.; Tsang, S.W.; So, F. Metal oxides for interface engineering in polymer solar cells.
J. Mater. Chem. 2012, 22, 24202–24212. [CrossRef]
Wang, H.Q.; Li, N.; Guldal, N.S.; Brabec, C.J. Nanocrystal V2 O5 thin film as hole-extraction layer in normal
architecture organic solar cells. Org. Electron. 2012, 13, 3014–3021. [CrossRef]
Chambers, B.A.; MacDonald, B.I.; Ionescu, M.; Deslandes, A.; Quinton, J.S.; Jasieniak, J.J.; Andersson, G.G.
Examining the role of ultra-thin atomic layer deposited metal oxide barrier layers on cdte/ito interface
stability during the fabrication of solution processed nanocrystalline solar cells. Sol. Energy Mater. Sol. Cells
2014, 125, 164–169. [CrossRef]
Townsend, T.K.; Yoon, W.; Foos, E.E.; Tischler, J.G. Impact of nanocrystal spray deposition on inorganic solar
cells. ACS Appl. Mater. Interfaces 2014, 6, 7902–7909. [CrossRef] [PubMed]
Kim, A.; Won, Y.; Woo, K.; Jeong, S.; Moon, J. All-solution-processed indium-free transparent composite
electrodes based on Ag nanowire and metal oxide for thin- film solar cells. Adv. Funct. Mater. 2014,
24, 2462–2471. [CrossRef]
Song, S.H.; Aydil, E.S.; Campbell, S.A. Metal-oxide broken-gap tunnel junction for copper indium gallium
diselenide tandem solar cells. Sol. Energy Mater. Sol. Cells 2015, 133, 133–142. [CrossRef]
Gwinner, M.C.; di Pietro, R.; Vaynzof, Y.; Greenberg, K.J.; Ho, P.K.H.; Friend, R.H.; Sirringhaus, H. Doping
of organic semiconductors using molybdenum trioxide: A quantitative time-dependent electrical and
spectroscopic study. Adv. Funct. Mater. 2011, 21, 1432–1441. [CrossRef]
Zilberberg, K.; Trost, S.; Meyer, J.; Kahn, A.; Behrendt, A.; Luetzenkirchen-Hecht, D.; Frahm, R.; Riedl, T.
Inverted organic solar cells with sol-gel processed high work-function vanadium oxide hole-extraction
layers. Adv. Funct. Mater. 2011, 21, 4776–4783. [CrossRef]
Zilberberg, K.; Trost, S.; Schmidt, H.; Riedl, T. Solution processed vanadium pentoxide as charge extraction
layer for organic solar cells. Adv. Energy Mater. 2011, 1, 377–381. [CrossRef]
Tan, Z.A.; Zhang, W.Q.; Cui, C.H.; Ding, Y.Q.; Qian, D.P.; Xu, Q.; Li, L.J.; Li, S.S.; Li, Y.F. Solution-processed
vanadium oxide as a hole collection layer on an ito electrode for high-performance polymer solar cells.
Phys. Chem. Chem. Phys. 2012, 14, 14589–14595. [CrossRef] [PubMed]
Xie, F.X.; Choy, W.C.H.; Wang, C.D.; Li, X.C.; Zhang, S.Q.; Hou, J.H. Low-temperature solution-processed
hydrogen molybdenum and vanadium bronzes for an efficient hole-transport layer in organic electronics.
Adv. Mater. 2013, 25, 2051–2055. [CrossRef] [PubMed]
Jin, S.; Jung, B.J.; Song, C.K.; Kwak, J. Room-temperature and solution-processed vanadium oxide buffer
layer for efficient charge injection in bottom-contact organic field-effect transistors. Curr. Appl. Phys. 2014,
14, 1809–1812. [CrossRef]
Materials 2016, 9, 235
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
11 of 12
Oksuzoglu, R.M.; Bilgic, P.; Yildirim, M.; Deniz, O. Influence of post-annealing on electrical, structural and
optical properties of vanadium oxide thin films. Opt. Laser Technol. 2013, 48, 102–109. [CrossRef]
Sahana, M.B.; Sudakar, C.; Thapa, C.; Lawes, G.; Naik, V.M.; Baird, R.J.; Auner, G.W.; Naik, R.;
Padmanabhan, K.R. Electrochemical propertiesof V2 O5 thin films deposited by spin coating. Mater. Sci.
Eng. B 2007, 143, 42–50. [CrossRef]
Haber, J.; Witko, M.; Tokarz, R. Vanadium pentoxide.1. Structures and properties. Appl. Catal. A Gen. 1997,
157, 3–22. [CrossRef]
Watters, D.C.; Yi, H.; Pearson, A.J.; Kingsley, J.; Iraqi, A.; Lidzey, D. Fluorene-based co-polymer with high
hole mobility and device performance in bulk heterojunction organic solar cells. Macromol. Rapid Commun.
2013, 34, 1157–1162. [CrossRef] [PubMed]
Lu, L.; Xu, T.; Chen, W.; Landry, E.S.; Yui, L. Ternary blend polymer solar cells with enhanced power
conversion efficiency. Nat. Photonics 2014, 8, 716–722. [CrossRef]
Zilberberg, K.; Meyer, J.; Riedl, T. Solution processed metal-oxides for organic electronic devices.
J. Mater. Chem. C 2013, 1, 4796–4815. [CrossRef]
Wagenpfahl, A.; Rauh, D.; Binder, M.; Deibel, C.; Dyakonov, V. S-shaped current-voltage characteristics of
organic solar devices. Phys. Rev. B 2010, 82, 115306. [CrossRef]
Kim, J.; Kim, H.; Kim, G.; Back, H.; Lee, K. Soluble transition metal oxide/polymeric acid composites for
efficient hole-transport layers in polymer solar cells. ACS Appl. Mater. Interfaces 2014, 6, 951–957. [CrossRef]
[PubMed]
Talledo, A.; Granqvist, C.G. Electrochromic vanadium-pentoxide-based films—Structural, electrochemical,
and optical-properties. J. Appl. Phys. 1995, 77, 4655–4666. [CrossRef]
Bullot, J.; Cordier, P.; Gallais, O.; Gauthier, M.; Babonneau, F. Thin-layers deposited from V2 O5 gels 2.
An optical-absorption study. J. Non Cryst. Solids 1984, 68, 135–146. [CrossRef]
Meyer, J.; Zilberberg, K.; Riedl, T.; Kahn, A. Electronic structure of vanadium pentoxide: An efficient hole
injector for organic electronic materials. J. Appl. Phys. 2011, 110, 033710–033715. [CrossRef]
Hancox, I.; Rochford, L.A.; Clare, D.; Walker, M.; Mudd, J.J.; Sullivan, P.; Schumann, S.; McConville, C.F.;
Jones, T.S. Optimization of a high work function solution processed vanadium oxide hole-extracting layer
for small molecule and polymer organic photovoltaic cells. J. Phys. Chem. C 2013, 117, 49–57. [CrossRef]
Negreira, A.S.; Aboud, S.; Wilcox, J. Surface reactivity of V2 O5 (001): Effects of vacancies, protonation,
hydroxylation, and chlorination. Phys. Rev. B 2011, 83. [CrossRef]
Silversmit, G.; Depla, D.; Poelman, H.; Marin, G.B.; de Gryse, R. Determination of the V2 p XPS binding
energies for different vanadium oxidation states (V5+ to V0+ ). J. Electron Spectrosc. Relat. Phenom. 2004,
135, 167–175. [CrossRef]
Sawatzky, G.A.; Post, D. X-ray photoelectron and auger-spectroscopy study of some vanadium-oxides.
Phys. Rev. B 1979, 20, 1546–1555. [CrossRef]
Demeter, M.; Neumann, M.; Reichelt, W. Mixed-valence vanadium oxides studied by XPS. Surf. Sci. 2000,
454–456, 41–44. [CrossRef]
Coulston, G.W.; Thompson, E.A.; Herron, N. Characterization of vpo catalysts by X-ray photoelectron
spectroscopy. J. Catal. 1996, 163, 122–129. [CrossRef]
Suchorski, Y.; Rihko-Struckmann, L.; Klose, F.; Ye, Y.; Alandjiyska, M.; Sundmacher, K.; Weiss, H. Evolution
of oxidation states in vanadium-based catalysts under conventional XPS conditions. Appl. Surf. Sci. 2005,
249, 231–237. [CrossRef]
Hermann, K.; Witko, M.; Druzinic, R.; Tokarz, R. Hydrogen assisted oxygen desorption from the V2 O5 (010)
surface. Top. Catal. 2000, 11, 67–75. [CrossRef]
Toledano, D.S.; Henrich, V.E.; Metcalf, P. Surface reduction of Cr-V2 O3 by Co. J. Vacuum Sci. Technol. A 2000,
18, 1906–1914. [CrossRef]
Ganduglia-Pirovano, M.V.; Sauer, J. Stability of reduced V2 O5 (001) surfaces. Phys. Rev. B 2004, 70. [CrossRef]
Yuan, N.Y.; Li, J.H.; Lin, C.L. Valence reduction process from sol-gel V2 O5 to VO2 thin films. Appl. Surf. Sci.
2002, 191, 176–180.
Materials 2016, 9, 235
51.
52.
53.
12 of 12
Surnev, S.; Ramsey, M.G.; Netzer, F.P. Vanadium oxide surface studies. Prog. Surf. Sci. 2003, 73, 117–165.
[CrossRef]
Bermudez, V.M.; Williams, R.T.; Long, J.P.; Reed, R.K.; Klein, P.H. Photoemission-study of hydrogen
adsorption on vanadium dioxide near the semiconductor-metal phase-transition. Phys. Rev. B 1992,
45, 9266–9271. [CrossRef]
Greiner, M.T.; Helander, M.G.; Tang, W.M.; Wang, Z.B.; Qiu, J.; Lu, Z.H. Universal energy-level alignment of
molecules on metal oxides. Nat. Mater. 2012, 11, 76–81. [CrossRef] [PubMed]
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons by Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).