Origin of the mixing ratio dependence of power conversion efficiency in bulk heterojunction organic solar cells with low donor concentration


Song, H-Jun.; Kim, J.Young.; Lee, D.; Song, J.; Ko, Y.; Kwak, J.; Lee, C.

Journal of Nanoscience and Nanotechnology 13(12): 7982-7987

2013


We studied the origin of the improvement in device performance of thermally evaporated bulk heterojunction organic photovoltaic devices (OPVs) with low donor concentration. Samples with three different donor-acceptor mixing ratios, 0:10 (C70-only), 1:9 (low-doped) and 3:7 (high-doped), were fabricated with 1,1-bis-(4-bis(4-methyl-phenyl)-amino-phenyl)-cyclohexane (TAPC):C70. The power conversion efficiencies (PCEs) of these samples were 1.14%, 2.74% and 0.69%, respectively. To determine why the low-doped device showed a high PCE, we measured various properties of the devices in terms of the effective energy band gap, activation energy, charge carrier mobility and recombination loss. We found that the activation energy for charge carrier transport was increased as we increased the TAPC concentration in the blends whereas the hole and electron mobilities became more balanced as the TAPC concentration was increased. Furthermore, the recombination loss parameter alpha (from the light intensity dependence) remained alpha to approximately 0.9 in the low-doped device, but it decreased to alpha to approximately 0.77 in the high-doped device, indicating a large recombination loss as a result of space charge. Therefore, the improved PCE of low-doped OPVs can be attributed to the balance between carrier mobilities with no increase in recombination loss.

ASP
Copyright
©
2013
American
Scientific
Publishers
All
rights
reserved
Printed
in
the
United
States
of
America
Journal
of
Nanoscience
and
Nanotechnology
Vol.
13,
7982-7987,
2013
RE
SEAR
CH
ARTI
CLE
Origin
of
the
Mixing
Ratio
Dependence
of
Power
Conversion
Efficiency
in
Bulk
Heterojunction
Organic
Solar
Cells
with
Low
Donor
Concentration
Hyung-Jun
Songl,
Jun
Young
Kiml,
Donggu
Leel,
Jiyun
Songl,
Youngjun
Kol,
Jeonghun
Kwak
2
,
and
Changhee
Leel
,
*
1Department
of
Electrical
and
Computer
Engineering,
Global
Frontier
Center
for
Multiscale
Energy
Systems,
Seoul
National
University,
Seoul
151-742,
Republic
of
Korea
2
Department
of
Electronic
Engineering,
Dong-A
University,
Busan
604-714,
Republic
of
Korea
We
studied
the
origin
of
the
improvement
in
device
performance
of
thermally
evaporated
bulk
het-
erojunction
organic
photovoltaic
devices
(OPVs)
with
low
donor
concentration.
Samples
with
three
different
donor—acceptor
mixing
ratios,
0:10
(C
70
-only),
1:9
(low-doped)
and
3:7
(high-doped),
were
fabricated
with
1,1-bis-(4-bis(4-methyl-phenyl)-amino-phenyl)-cyclohexane
(TAPC):C
70
.
The
power
conversion
efficiencies
(PCEs)
of
these
samples
were
1.14%,
2.74%
and
0.69%,
respectively.
To
determine
why
the
low-doped
device
showed
a
high
PCE,
we
measured
various
properties
of
the
devices
in
terms
of
the
effective
energy
band
gap,
activation
energy,
charge
carrier
mobility
and
recombination
loss.
We
found
that
the
activation
energy
for
charge
carrier
transport
was
increased
as
we
increased
the
TAPC
concentration
in
the
blends
whereas
the
hole
and
electron
mobilities
became
more
balanced
as
the
TAPC
concentration
was
increased.
Furthermore,
the
recombination
loss
parameter
a
(from
the
light
intensity
dependence)
remained
a
ti
0.9
in
the
low-doped
device,
but
it
decreased
to
a
0.77
in
the
high-doped
device,
indicating
a
large
recombination
loss
as
a
result
of
space
charge.
Therefore,
the
improved
PCE
of
low-doped
OPVs
can
be
attributed
to
the
balance
between
carrier
mobilities
with
no
increase
in
recombination
loss.
Keywords:
Organic
Photovoltaic
Cells,
Small
Molecule,
Bulk
Heterojunction,
Donor—Acceptor
Mixing
Ratio.
1.
INTRODUCTION
Thermally
evaporated
small-molecule-based
organic
pho-
tovoltaic
cells
(OPVs)
have
received
considerable
attention
because
of
their
advantages
over
their
polymeric
counter-
parts
such
as
high
purity,
well-ordered
film
structure
and
uniform
performance
without
batch-to-batch
variation."
In
terms
of
device
structure,
bulk
heterojunction
(Bill)
OPVs,
based
on
a
network
of
donor
and
acceptor
molecules
in
the
active
layer,
are
widely
used
because
of
their
remark-
ably
high
power
conversion
efficiency
(PCE).
3
To
increase
the
performance
of
BHJ-OPVs,
many
efforts
have
been
made,
such
as
an
alternative
thermal
deposition
technique,
substrate
heating
a
post-annealing
process
and
adding
co-
evaporant
molecules.
4-7
Recently,
Tang
et
al.
reported
that
BHJ-OPVs
with
a
low
ratio
of
donor
molecules
led
to
an
enhancement
in
the
PCE
*Author
to
whom
correspondence
should
be
addressed.
7982
J.
Nanosci.
Nanotechnol.
2013,
Vol.
13,
No.
12
by
dramatically
increasing
the
photocurrent.
8
By
contrast,
it
has
been
shown
that
the
PCE
of
devices
whose
donor
concentration
exceeds
the
proper
ratio
decreases
sharply.
Thus,
it
is
necessary
to
understand
how
donor
molecules
behave
in
BHJ-OPVs
as
a
function
of
their
concentration.
However,
to
date,
donor
molecules
have
simply
been
con-
sidered
paths
for
hole
carriers
and
the
influence
of
them
on
electrical
properties
of
devices
have
not
been
stud-
ied.
Furthermore,
the
origin
of
the
decrease
in
the
PCE
in
devices
with
high
donor
concentrations
has
not
been
clearly
elucidated.
In
this
study,
we
analyzed
the
origin
of
the
donor con-
centration
dependence
of
PCE
in
low-donor-doped
BHJ-
OPVs
by
measuring
the
electrical
characteristics
of
the
devices
at
different
temperatures.
The
results
show
that
improving
carrier
mobility
without
affecting
recombina-
tion
loss
in
low-donor-doped
devices
leads
to
an
enhance-
ment
in
the
PCE.
1533-4880/2013/13/7982/006
doi:10.1166/jnn.2013.8155
1111,11111111,111111111,1111111111
111,11,11,11
0—
C
70
-only
0—
Low-doped
High-doped
Al
(100nm)
/LiF
(0.5nm)
(b)
2
o
o
E
0)
—4
rz)
a)
—6
(.)
—8
—10
(a)
Bphen
(8nm)
Cx.:TAPC
(X%)
(45nm)
TAPC
(3nm)
/MoOx
(5nm)
<—
ITO
/Glass
Song
et
al.
Origin
of
the
Mixing
Ratio
Dependence
of
PCE
in
Bulk
Heterojunction
Organic
Solar
Cells
2.
EXPERIMENTAL
DETAILS
The
device
structure
employed
in
this
study
is
illustrated
in
Figure
1(a).
Devices
were
fabricated
on
cleaned
indium-
tin
oxide
(ITO)-coated
glass
substrates.
For
hole
extraction
and
to
make
a
Schottky
contact
with
the
active
layer,
5-nm
molybdenum
oxide
(MoO
3
)
and
3-nm
1,1
bis-(4-bis(4-
methyl-phenyl)-amino-phenyl)-cyclohexane
(TAPC)
layers
were
sequentially
deposited.9'
to
An
active
layer
with
a
thickness
of
45
nm
was
formed
by
co-evaporating
an
acceptor
material,
C70
(Electrical
Materials
Index
Corp.),
and
a
donor
material,
TAPC
(Luminescence
Technology
Corp.).
To
verify
the
effect
of
the
donor-acceptor
mixing
ratio
on
OPVs,
three
samples
with
different
TAPC
con-
centrations
were
prepared.
The
active
layer
of
the
first
one
was
pure
C
7
0
(C
70
-only).
The
second
one
contained
10%
(volume
ratio)
TAPC
relative
to
the
C
7
0
content
(low-
doped).
The
donor
concentration
of
the
third
device
was
30%
(high-doped).
Before
evaporating
a
lithium
fluoride
(LiF)/aluminum
(Al)
cathode,
a
10-nm
4,7-diphenyl-1,10-
phenanthroline
(Bphen)
layer
was
inserted
to
block
hole
carriers."
The
evaporation
rate
of
C70
was
fixed
at
1
A/s
for
all
samples,
but
°
that
of
TAPC
ranged
from
0.1
A/s
(low-doped)
to
0.3
A/s
(high-doped)
using
two
indepen-
dently
controlled
heating
sources.
The
rates
of
deposition
for
MoO
3
,
TAPC
buffer,
Bphen,
LiF
and
Al
were
1,
0.5,
1,
0.05
and
3
A/s,
respectively.
All
samples
were
fabricated
in
a
high-vacuum
system
(<
5
x10
-6
Torr)
and
encapsu-
lated
in
an
argon
atmosphere
to
prevent
the
devices
from
degradation.
The
size
of
the
devices
was
20
mm
2
.
To
measure
the
carrier
mobility
single-carrier
devices
were
also
fabricated.
To
eliminate
the
side
effects
from
dif-
ferent
injection
layers,
the
buffer
layers
in
the
devices
were
the
same
as
those
used
in
OPVs.
In
electron-only
devices,
Al
was
used
for
both
electrodes
to
match
the
work
func-
tion
with
the
lowest
unoccupied
molecular
orbital
(LUMO)
level
of
C70.
12,13
Moreover,
10-nm
Bphen
layers
were
inserted
between
the
100-nm
C
m
:TAPC
layer
and
both
electrodes
to
prevent
hole
injection
from
the
electrodes.
In
contrast,
the
electrodes
typically
used
for
hole-only
devices
were
ITO
and
gold,
whose
work
functions
are
higher
than
the
work
function
of
Al.'
Additionally,
10-nm
MoO
3
lay-
ers
were
used
as
hole
buffers
to
make
an
ohmic
contact.
The
deposition
rates
and
TAPC
doping
ratio
for
the
single-
carrier
devices
were
the
same
as
those
used
in
the
OPVs.
The
photocurrent-voltage
characteristics
of
the
cells
were
measured
at
temperatures
ranging
from 100
K
to
400
K
in
the
dark
and
under
irradiation
intensities
ranging
from
2
mW/cm
2
to
100
mW/cm
2
using
a
solar
simulator
(Newport
91160A,
Air
mass
(AM)
1.5
G
with
a
KG
5
fil-
ter).
All
measurements
were
taken
under
low
vacuum
by
using
a
Keithley
237
source
measurement
unit.
3.
RESULTS
AND
DISCUSSION
3.1.
Effective
Energy
Gap
The
current
density-voltage
curves
for
the
three
devices
with
different
donor-acceptor
mixing
ratios
at
room
temperature
are
shown
in
Figure
1(b).
The
PCEs
of
the
C
m
-only,
low-doped
and
high-doped
samples
were
1.14%,
2.74%
and
0.69%,
respectively.
The
PCE
of
the
low-doped
device
was
much
higher
than
the
PCE
of
the
other
devices,
which
is
consistent
with
the
results
of
a
previous
study.
8
To
investigate
the
main
sources
of
improvement
in
the
PCE,
we
measured
the
OPV
characteristics
in
terms
of
the
open-
circuit
voltage
(V
oc
),
short-circuit
current
density
(Jsc)
,
fill
factor
(FF)
and
PCE
by
changing
the
device
temperature.
The
temperature
dependence
of
17
0c
is
depicted
in
Figure
2(a).
The
17
0c
of
all
devices
decreased
linearly
with
increasing
temperature,
as
indicated
by
the
following
equa-
tion
for
a
p-n
junction
solar
cell,
14
E
g
k
B
T
Voc
=
C
(
1
)
where
E
g
is
the
energy
band
gap
between
the
LUMO
of
an
acceptor
and
the
highest
occupied
molecular
orbital
(HOMO)
of
a
donor,
C
is
a
constant,
k
B
is
the
Boltzmann
0
0
0.2
0.4
0.6
0.8
10
Voltage
(V)
Fig.
1.
(a)
Schematic
of
the
device
structure
and
(b)
current
density—voltage
characteristics
under
a
light
intensity
of
100
mW/cm
2
for
OPVs
with
different
donor
ratios.
J.
Nanosci.
NanotechnoL
13,
7982-7987,
2013
7983
1.0
1
1111
1
0.9
7
0.8
7
0.7
=
111 ,''II I'''','
0.6-
=
G
m
-only
0.5-
Low-doped
High-doped
-
A
1•
6
6
A A
-
1111111111111111111 111111111
1111111111111111
C7
only
_
Low-doped
-
High-doped
1111111111
m=
II
'
i....i..
C.
Tr
only
Low-doped
A
High-doped
11 ''I'll'
11111
.
1
111111111111111111
I
I
7
A
A
7
(d)
3.0
2.5
2.0
0
0
0.5
0.0
100 150
200
250
300
350
400
Temperature
(K)
Origin
of
the
Mixing
Ratio
Dependence
of
PCE
in
Bulk
Heterojunction
Organic
Solar
Cells
Song
et
al.
(a)
0.4
'
'
"
'
100
150
200
250
300
350
400
Temperature
(K)
0
100 150
200
250
300
350
400
Temperature
(K)
111
I
.
-
(b)
10
8
E
C
6
E
8
4
2
1
1 1 1
1
--
0.6
I
-
C
70
-only
-
Low-doped
0.5
7
High-doped
0.4—
I
I
0.2—
0.1
'
1111111111111111
11111111111
1
100
150
200
250
300
350
400
Temperature
(K)
Fig.
2.
Photovoltaic
performance
parameters
in
terms
of
(a)
V
ac
,
(b)
J„,
(c)
FF
and
(d)
PCE
for
C„-only,
low-doped
and
high-doped
devices
as
a
function
of
absolute
temperature
under
a
light
intensity
of
100
mW/cm
2
.
The
solid
line
in
(a)
represents
the
fit
using
Eq.
(1).
_
A A
0
(c)
constant,
T
is
the
absolute
temperature
and
e
is
the
charge
of
a
single
electron.
However,
in
this
case,
the
effective
E
g
can
be
considered
the
energy
band
gap
between
the
HOMO
of
the
buffer
layer
and
the
LUMO
of
the
active
layer
because
the
donor
doping
ratio
is
sufficiently
low.
8
The
effective
E
g
values,
deduced
from
Eq.
(1),
for
the
C
70
-only,
low-doped
and
high-doped
cells
were
0.98
eV,
0.90 eV
and
0.80
eV,
respectively.
These
values
indicate
that
the
effective
E
g
decreased
with
the
increasing
concen-
tration
of
donor
molecules
in
the
composite.
Because
the
HOMO
of
the
TAPC
buffer
layer
was
fixed,
the
decrease
in
the
effective
E
g
originated
from
the
drop
in
the
effec-
tive
LUMO
level
of
the
active
layer.
As
a
result,
the
V
oc
decreased
as
the
donor
concentration
increased."
However
the
variation
in
17
0c
is
not
sufficiently
large
to
explain
the
drastic
decrease
in
the
PCE
of
the
high-doped
device
and
contradicts
the
improvement
in
the
PCE
of
the
low-doped
device.
Hence,
the
17
0c
is
negatively
correlated
with
the
donor
concentration
due
to
the
change
in
the
effective
E
g
though
the
change
in
the
V
oc
has
limited
effects
on
the
PCE.
3.2.
Activation
Energy
By
contrast,
as
shown
in
Figures
2(b)
and
(c),
as
the
tem-
perature
was
increased,
a
substantial
enhancement
in
J
sc
and
FF
was
observed
in
the
low-doped
device
compared
to
the
high-doped
device.
The
change
in
these
two
factors,
J
sc
and
FF,
is
mainly
attributable
to
the
variation
in
the
PCE
(see
Fig.
2(d)).
In
the
C
70
-only
and
low-doped
devices,
the
PCE
showed
a
negative
parabolic
dependence
on
temperature,
which
was
optimized
at
room
temperature,
analogous
to
the
tem-
perature
dependence
of
FE
In
disordered
organic
films,
it
is
well
known
that
this
shape
is
due
to
the
combina-
tion
of
two
factors:
an
improvement
in
carrier
transport
by
hopping
as
the
temperature
increases
and
a
decrease
in
mobility
due
to
phonon
scattering
at
much
higher
temperatures."'
17
This
temperature
dependence
indicates
that
carriers
can
be
easily
extracted
by
the
electrodes
of
these
devices
at
room
temperature.
By
contrast,
in
the
high-doped
device,
the
PCE
as
well
as
the
FF
and
J
sc
were
continuously
improved
as
the
temperature
increased
up
to
400
K,
which
indicates
that
high
thermal
energy
helped
release
carriers
held
in
trap
sites.
Thus,
the
donor
concen-
tration
is
correlated
with
the
energy
depth
of
trap
sites.
To
determine
the
energy
level
of
traps,
In
J
sc
versus
inverse
temperature
(1000/T)
under
different
light
intensi-
ties
was
measured,
as
shown
in
Figure
3.
According
to
the
Arrhenius
equation,
J
sc
can
be
expressed
as
a
function
of
temperature,
light
intensity
and
activation
energy
(AE,
A),
Jsc(T
,
PLight)
=
J
0
(PLight)
x
eXP
)
(2)
7984
J.
Nanosci.
Nanotechnol.
13,
7982-7987,
2013
Song
et
al.
Origin
of
the
Mixing
Ratio
Dependence
of
PCE
in
Bulk
Heterojunction
Organic
Solar
Cells
(a)
"
-El
-
0MW/CM
2
C
70
-only
Low-doped
-
1
High-doped
----
_
,
_
---
fitting
-
2
2
4
6
8
10
12
(1000/T)
(1/K)
(b)
Ac
tiva
tr
ion
e
nergy
(
meV
12
-
10-
4-
8-
-
6-
2
-
0
0
70
-only
Low-doped
High-doped
A
I
1 1
1 1 1 1 1 1
A
1
10
Light
Intensity
(mW/cm
2
)
(c)
12
I
100
-
Measure
Fitting
10
C
n
-only
--,
-
Low-doped
N
E
A
High-doped
-C.).
8—
E
6—
The
AE
for
the
high-doped
(8.6-10.3
meV)
device
was
higher
than
that
for
the
C
m
-only
(2.5-3.5
meV)
and
low-doped
(5.9-6.7
meV)
devices
at
each
light
intensity,
as
shown
in
Figure
3(b).
The
AE
increased
in
proportion
to
the
concentration
of
donor
molecules
in
the
blend
layer.
The
AE
values
for
the
low-doped
and
the
C
m
-only
devices
were
not
sufficiently
high
to
allow
carriers
to
easily
move
to
the
electrode
with
small
losses
in
energy,
while
the
AE
for
the
high-doped
device
was
quite
high,
preventing
cap-
tured
carriers
from
easily
overcoming
the
energy
barrier.
In
other
words,
donor
molecules
act
as
trap
sites
that
hin-
der
the
transport
of
carriers
to
electrodes;
thus,
the
high
AE
led
to
a
decrease
in
the
PCE
of
the
highly
doped
OPVs.
3.3.
Carrier
Mobility
The
pre-exponential
factor
[J
o
(P
ught
)]
in
the
Arrhenius
equation
also
supports
the
enhancement
in
the
PCE
of
the
low-doped
device,
as
illustrated
in
Figure
3(c).
As
men-
tioned
previously,
the
factor
is
governed
by
two
electrical
properties:
carrier
mobility
and
the
number
of
carriers
that
are
involved
in
recombination.
Thus,
we
investigated
the
relationship
between
these
two
factors
and
the
donor con-
centration.
First,
Figure
4
shows
the
hole
and
electron
mobil-
ity
as
a
function
of
the
TAPC
ratio
in
the
single-carrier
devices.
The
mobility
was
obtained
by
considering
a
space-charge-limited
current
model
under
an
electric
field
of
1
x10
5
V/cm.
The
hole
mobility
of
the
low-doped
device
(2.8
x
10
-6
cm
2
/V
s)
was
one
order
of
magni-
tude
higher
than
that
of
the
pure
C
70
film
(2.7
x
10
-7
cm
2
/V
s),
while
the
value
of
the
high-doped
device
(4.6
x
10
-6
cm
2
/V
s)
did
not
significantly
increase
beyond
that
of
the
low-doped
one
Similar
results
were
previously
reported
for
OPVs
based
on
fullerene
derivative/polymer
blends
or
vacuum-evaporated
C
60
/donor
mixtures."'
21
Meanwhile,
the
electron
mobility
of
the
devices
remained
within
the
same
order
of
magnitude,
even
though
the
donor
concentration
reached
up
to
30%.
The
elec-
tron
mobility
for
the
low-doped
(3.9
x
10
-3
cm
2
/V
s)
2=
IN
100mW/cm
2
C„-only
Low-doped
A
High-doped
fitting
E
-
1 I
i
I
20
40
60
80
100
Light
Intensity
(mW/cm
2
)
Fig.
3.
(a)
Arrhenius
equation
plots
of
In
J
sc
versus
the
inverse
abso-
lute
temperature
under
different
incident
light
intensities
of
10
mW/cm
2
(open)
and
100
mW/cm
2
(closed).
The
solid
fitting
lines
are
for
100
mW/cm
2
and
the
dotted
ones
for
10
mW/cm
2
.
(b)
The
activation
energy
and
(c)
the
pre-exponential
factor
for
OPVs
with
different
donor:
acceptor
ratios.
where
P
ught
is
the
incident
light
intensity.
18
The
term
J
o
(P
ught
)
reflects
the
amount
of
photocurrent
associated
with
carrier
mobility,
the
number
of
carriers
and
electri-
cal
field,
while
the
exponential
term
represents
the
energy
depth
of
traps.
18
'
19
10
-1
_L
_L
Holes
Electrons
0
5
10
15
20
25
30
TAPC
concentration
(%)
Fig.
4.
Hole
and
electron
mobility
of
single-carrier
devices
as
a
func-
tion
of
TAPC
concentration.
The
error
bars
indicate
the
deviation
among
samples
with
the
same
doping
ratio.
a
4-
-
2-
0
1
0
J.
Nanosci.
NanotechnoL
13,
7982-7987,
2013
7985
I
I
I
a
0
90
100K
111
=
Measure
Fitting
0
-
100K
-
200K
A
-
300K
O
-
400K
a
400K
=0
91
a
400K
=0
'
90
cc
100K
=0
'
90
Measure
Fitting
O
-
100K
-
200K
-
300K
O
-
400K
Ili
a
400K
=0
'
84
100K
0
'
7 7
Measure
Fitting
O
-
100K
-
200K
-
A
-
300K
O
-
400K
I
I I I
I
10
100
Light
Intensity
(mW/cm
2
)
10
100
Light
Intensity
(mW/cm
2
)
Origin
of
the
Mixing
Ratio
Dependence
of
PCE
in
Bulk
Heterojunction
Organic
Solar
Cells
Song
et
al.
Table
I.
Electrical
properties
of
OPVs
with
different
donor
concentrations.
C„-only
Low-doped
High-doped
Effective
E
g
(eV)
0.98
0.90
0.80
AE
(meV)
2.5-3.5
5.9-6.7
8.6-10.3
Hole
mobility
(cm
2
/Vs)
2.7
x
10'
2.8
x
10
-6
4.6
x
10
-6
Electron
mobility
(cm
2
/Vs)
5.6
x
10
3
3.9
x
10
3
3.2
x
10
3
a
0.90-0.91
0.90
0.77-0.84
and
high-doped
(3.2
x
10
-3
cmW
s)
films
did
not
vary
much
compared
with
that
of
the
neat
C70
layer
(5.6
x
10
-3
cmW
s).
This
result
is
in
agreement
with
that
of
a
previous
study,
in
which
the
electron
mobility
remained
nearly
the
same
until
donor
molecules
were
added
to
BHJ
blends
in
concentrations
of
up
to
30%.
12
'
13
More-
over,
doping
reduced
the
gap
between
the
electron
and
the
hole
mobility
by
enhancing
hole
transport.
This
balanced
mobility
between
electron
and
hole
transport
consequently
improved
the
values
of
the
pre-exponential
factor
in
doped
devices,
resulting
in
the
improvement
of
the
PCE.
n
3.4.
Recombination
Loss
To
probe
the
recombination
characteristics
that
also
affect
the
pre-exponential
value,
J
sc
values
were
measured
under
light
intensities
ranging
from
2
mW/cm
2
to
100
mW/cm
2
.
Figure
5
shows
the
light-intensity
dependence
of
J
sc
(Jsc
=
'Tight)
for
devices
on
a
double-logarithmic
scale.
Here,
a
is
a
power-law
value
that
represents
recombination
properties.
Several
authors
have
reported
that
monomolec-
ular
recombination
dominates
the
recombination
properties
of
a
device
when
a
is
close
to
1.
Meanwhile,
when
a
=
0.75,
the
recombination
loss
of
a
device
is
controlled
by
space
charges
2
3-25
The
a
of
the
C
70
-only
and
low-doped
devices
ranges
from
0.90
to
0.91
throughout
the
entire
range
of
temperatures,
whereas
the
value
of
the
high-doped
one
increases
from
0.77
(T
=
100
K)
to
0.84
(T
=
400
K).
Therefore,
the
C
70
-only
and
low-doped
devices
are
in
the
monomolecular
recombination
region
and
the
high-doped
device
is
close
to
the
space-charge
limited
recombination
state.
The
reduction
in
the
photocurrent
of
a
device
in
the
monomolecular
recombination
region
is
generally
smaller
than
that
due
to
space
charges
because
photo-generated
carriers
are
quenched
only
by
intrinsic
impurities
in
this
region.
Thus,
the
recombination
loss
in
the
C
70
-only
and
low-doped
OPVs
was
low,
while
the
high-doped
one
exhibited
a
high
space-charge-limited
recombination
loss
due
to
deep-trapped
charges.
The
temperature
dependence
of
a
in
the
high-doped
cell
can
also
be
understood
by
considering
these
captured
carriers,
which
had
a
higher
probability
of
being
released
from
trapped
states
at
high
temperature.
Thus,
not
only
J
sc
but
also
FF
decreased
in
the
high-doped
cell
because
the
free-charge
extraction
effi-
ciency
was
suppressed
by
the
recombination
loss.'
There-
fore,
the
small
recombination
loss
in
the
low-doped
device
(a)
10
E
E
a)
0
2
C.)
0.1
(b)
10
E
NE
0.1
(c)
10
E
E
C
U
0
.
1
10
100
Light
Intensity
(mW/cm
2
)
Fig.
5.
The
J
sc
under
different
light
intensities
for
(a)
C
ho
-only,
(b)
low-
doped
and
(c)
high-doped
devices
at
different
temperatures.
The
power-
law-dependent
factor
a
at
100
K
and
400
K
is
marked
in
each
graph.
improved
J
sc
and
FF,
while
the
mobility-balanced
state
in
the
high-doped
sample
was
overwhelmed
by
recombina-
tion
loss.
4.
CONCLUSIONS
In
summary,
we
investigated
the
effect
of
donor
doping
on
BHJ-OPVs
to
assess
the
origin
of
PCE
enhancement
7986
J.
Nanosci.
Nanotechnol.
13,
7982-7987,
2013
Song
et
al.
Origin
of
the
Mixing
Ratio
Dependence
of
PCE
in
Bulk
Heterojunction
Organic
Solar
Cells
in
low-doped
cells.
The
doping
of
donor
molecules
in
the
active
layer
influenced
four
electrical
properties
of
the
devices:
the
effective
E
g
,
activation
energy,
carrier
mobility
and
recombination
loss.
Although
donor
doping
reduced
the
effective
E
g
and
produced
trap
sites,
the
photocurrent
increased
owing
to
the
facilitation
of
hole
carrier
transport
by
a
small
activation
energy
and
balanced
electron—hole
mobility
in
the
low-doped
device.
Furthermore,
at
small
doping
ratios,
the
recombination
loss
was
not
increased.
Therefore,
we
can
conclude
that
these
factors
are
impor-
tant
in
developing
highly
efficient
OPVs.
We
also
believe
that
the
analysis
and
results
reported
in
this
paper
will
aid
in
developing
and
further
improving
other
small-molecule-
based
BHJ-OPVs.
Acknowledgments:
This
work
was
supported
by
the
Human
Resources
Development
Program
of
the
Korea
Institute
of
Energy
Technology
Evaluation
and
Planning
(KETEP)
grant
funded
by
the
Korea
government
Ministry
of
Knowledge
Economy
(No.
20124010203170).
References
and
Notes
1.
B.
Walker,
A.
B.
Tamayo,
X.-D.
Dang,
P.
Zalar,
J.
H.
Seo,
A.
Garcia,
M.
Tantiwiwat,
and
T.-Q.
Nguyen,
Adv.
Funct.
Mater.
19,
3063
(2009).
2.
A.
Mishra
and
P.
Bauerie,
Angew.
Chem.
Int.
Ed
51,
2020
(2012).
3.
C.
J.
Brabec,
N.
S.
Sariciftci,
and
J.
C.
Hummelen,
Adv.
Funct.
Mater.
11,
15
(2001).
4.
J.
W.
Kim,
H.
J.
Kim,
H. H.
Lee,
T.
Kim,
and
J.-J.
Kim,
Adv.
Funct.
Mater.
21,
2067
(2011).
5.
D.
Wynands,
M.
Levichkova,
M.
Riede,
M.
Pfeiffer,
P.
Baeuerle,
R.
Rentenberger,
P.
Denner,
and
K.
Leo,
J.
Appl.
Phys.
107,
014517
(2010).
6.
J.
Y.
Kim,
J.
Kwak,
S.
Noh,
and
C.
Lee,
J.
Nanosci.
Nanotechnol.
12,
5724
(2012).
7.
T.
Kaji,
M.
Zhang,
S.
Nakao,
K.
Iketaki,
K.
Yokoyama,
C.
W.
Tang,
and
M.
Hiramoto,
Adv.
Mater.
23,
3320
(2011).
8.
M.
Zhang,
H.
Wang,
H.
Tian,
Y.
Geng,
and
C.
W.
Tang,
Adv.
Mater.
23,
4960
(2011).
9.
Wan,
M.
Zhang,
H.
Ding,
C.
W.
Tang,
and
Y.
Gao,
Org.
Electron.
12,
1588
(2011).
10.
P.
M.
Borsenberger,
L.
Pautmeier,
R.
Richert,
and
H.
Bassier,
J.
Chem.
Phys.
94,
8276
(1991).
11.
S.
Naka,
H.
Okada,
H.
Onnagawa,
and
T.
Tsutsui,
Appl.
Phys.
Lett.
76,
197
(2000).
12.
R.
Pandey,
A. A.
Gunawan,
K.
A.
Mkhoyan,
and
R.
J.
Holmes,
Adv.
Funct.
Mater.
22,
617
(2012).
13.
V.
D.
Mihailetchi,
L.
J.
A.
Koster,
P.
W.
M.
Blom,
C.
Meizer,
B.
de
Boer,
J.
K.
J.
van
Duren,
and
R.
A.
J.
Janssen,
Adv.
Funct.
Mater.
15,
795
(2005).
14.
M.
D.
Perez,
C.
Borek,
S.
R.
Forrest,
and
M.
E.
Thompson,
J.
Am.
Chem.
Soc.
131,
9281
(2009).
15.
G.
Chen,
H.
Sasabe,
Z.
Wang,
X.
F
Wang,
Z.
Hong,
Y.
Yang,
and
J.
Kido,
Adv.
Mater
24,
2768
(2012).
16.
D.
Chirvase,
Z.
Chiguvare,
M.
Knipper,
J.
Parisi,
V.
Dyakonov,
and
J.
C.
Hummelen,
J.
Appl.
Phys.
93,
3376
(2003).
17.
P.
M.
Borsenberger,
L.
Pautmeier,
and
H.
Bassier,
J.
Chem.
Phys.
94,
5447
(1991).
18.
I.
Riedel
and
V.
Dyakonov,
Phys.
Stat.
Sol.
(a)
201,
1332
(2004).
19.
J.
Schafferhans,
C.
Deibel,
and
V.
Dyakonov,
Adv.
Env:
Mater.
1,
655
(2011).
20.
A.
Opitz,
M.
Bronner,
J.
Wagner,
M.
Gotzenbrugger,
and
W.
Briitting,
Proc.
SPIE
7002,
70020J
(2008).
21.
S.
M.
Tuladhar,
D.
Poplayskyy,
S.
A.
Choulis,
J.
R.
Durrant,
D. D.
C.
Bradley,
and
J.
Nelson,
Adv.
Funct.
Mater.
15,
1171
(2005).
22.
J.
H.
Seo,
J.-R.
Koo,
S.
J.
Lee,
B.
M.
Seo,
and
Y.
K.
Kim,
J.
Nanosci.
Nanotechnol.
11,
7307
(2011).
23.
L.
J.
A.
Koster,
V.
D.
Mihailetchi,
H.
Xie,
and
P.
W.
M.
Blom,
Appl.
Phys.
Lett.
87,
203502
(2005).
24.
V.
D.
Mihailetchi,
J.
Wildeman,
and
P.
W.
M.
Blom,
Phys.
Rev.
Lett.
94,
126602
(2005).
25.
P.
Schilinsky,
C.
Waldauf,
and
C.
J.
Brabec,
Appl.
Phys.
Lett.
81,
3885
(2002).
26.
R.
Mauer,
I.
A.
Howard,
and
E
Laquai,
J.
Phys.
Chem.
Lett.
1,
3500
(2010).
313111:
IV
H31:
1V3S31:
1
Received:
1
May
2012.
Accepted:
20
December
2012.
J.
Nanosci.
Nanotechnol.
13,
7982-7987,
2013
7987