A nanocomposite ultraviolet photodetector based
on interfacial trap-controlled charge injection
Fawen Guo, Bin Yang, Yongbo Yuan, Zhengguo Xiao, Qingfeng Dong, Yu Bi and Jinsong Huang
*
Ultraviolet photodetectors have applications in fields such as
medicine, communications and defence
1
, and are typically
made from single-crystalline silicon, silicon carbide or gallium
nitride p–n junction photodiodes. However, such inorganic
photodetectors are unsuitable for certain applications
because of their high cost and low responsivity (<0.2 AW
21
)
2
.
Solution-processed photodetectors based on organic materials
and/or nanomaterials could be significantly cheaper to manu-
facture, but their performance so far has been limited
2–7
.
Here, we show that a solution-processed ultraviolet photo-
detector with a nanocomposite active layer composed of ZnO
nanoparticles blended with semiconducting polymers can
significantly outperform inorganic photodetectors. As a result
of interfacial trap-controlled charge injection, the photodetec-
tor transitions from a photodiode with a rectifying Schottky
contact in the dark, to a photoconductor with an ohmic
contact under illumination, and therefore combines the low
dark current of a photodiode and the high responsivity of a
photoconductor (∼721–1,001 A W
21
). Under a bias of <10 V,
our device provides a detectivity of 3.4 3 10
15
Jones at
360 nm at room temperature, which is two to three orders of
magnitude higher than that of existing inorganic semiconductor
ultraviolet photodetectors.
The use of solution-processed thin films of colloidal inorganic
semiconductor nanoparticles or colloidal quantum dots as photo-
conductors has been a critical step in the quest to fabricate low-
cost photodetectors. Several types of nanomaterials, including PbS
colloidal quantum dots and ZnO nanoparticles, have been used to
measure different response spectra ranging from the near-infrared
to the ultraviolet
2–4,6,8
. These photodetectors demonstrate remark-
ably high responsivities—greater than 1,000 A W
21
in the near-
infrared range
3,6
and 61 A W
21
in the ultraviolet
2
. However, the
lateral structure adopted in these photoconductive photodetectors
inevitably leads to large driving voltages and slow responses. A
large lateral electrode spacing (.5 mm) is required to increase the
shunt resistance and reduce the dark current, so, to maintain high
gain, these photodetectors need a very high driving voltage of
100 V, which cannot be provided by commercially available
thin-film transistors
2,3,7
. The response speed of lateral-structure
photoconductive photodetectors is also sacrificed to maintain this
high gain. Gain is determined by the ratio of the lifetime of the
free charge and the transit time
9
. A large electrode spacing is
needed to allow more light to be absorbed, both for high gain and
to reduce the dark current, but this increases the transit time of
the charges across the electrodes. A long trapping time (on the
scale of minutes) is therefore needed, which inevitably limits the
applications of these photodetectors
2,3,6
.
A vertical-structure photoconductive photodetector, with its
much smaller electrode spacing, has a much shorter carrier transit
length, therefore providing high gain as well as a quick response.
A photoactive layer thickness of 500 nm is generally thick
enough to absorb most of the ultraviolet radiation, as a result of
the high absorption coefficient of many types of nanoparticles in
the ultraviolet range
10
. One challenge in making a vertical-structure
photoconductive photodetector with such a thin absorber film is
dealing with the large dark current injected from its much larger
electrode contact area than in lateral structures. Ohmic contact
with at least one of the electrodes is required to take advantage of
the high gain of the photoconductors
9
. However, the large charge
injection, combined with the relatively smaller shunt resistance
(due to the much shorter charge transit length than in the lateral
structure), will result in a large dark current, which can ruin the
detectivity of the photodetector
11
. The large dark current problem
is exacerbated when using non-dense nanoparticle films, which
are susceptible to the penetration of metal atoms by diffusion
during metallization and subsequent thermal treatment.
We have developed a highly sensitive ultraviolet photodetector
with a vertical device structure combining the gain of a photocon-
ductor and the low noise of a diode. The active-layer materials are
nanocomposites composed of ZnO nanoparticles blended with
semiconducting polymers. ZnO is emerging as a potential
alternative to GaN or SiC as an ultraviolet absorber due to its
wide bandgap of 3.4 eV, low-cost material and variable
synthetic strategies
2,12–19
. Our nanocomposite photodetectors were
fabricated using a low-cost spin-coating process, which is
compatible with complementary metal–oxide–semiconductor
(CMOS) readout circuits
4
.
The structure and operating principle of our photodetector are
presented in Fig. 1. The simple device structure comprises a
polymer:ZnO nanocomposite layer sandwiched between a transpar-
ent indium tin oxide (ITO) anode and an aluminium cathode (for
chemical structures of the materials see Supplementary Fig. S1).
This device structure is essentially the same as those of polymer:na-
noparticle hybrid solar cells
20–22
or bistable memories
23
, except that
the nanoparticles work as charge traps in our photodetector instead
of as charge conductors (as is the case in hybrid solar cells). Two
types of hole-conducting semiconducting polymers were used to
obtain different response spectra (Fig. 1b): poly-3(hexylthiophene)
(P3HT, optical bandgap of 1.9 eV) for UV–vis detection and poly-
vinylcarbazole (PVK, bandgap of 3.5 eV) for ultraviolet detection.
PVK is used because of its reasonably high hole mobility and very
high bandgap, although it is not a conjugated polymer. To minimize
the dark current, a thin layer of 4,4
′
-bis[(p-trichlorosilylpropylphe-
nyl)phenylamino]-biphenyl (TPD-Si
2
) and PVK blend was inserted
between poly(3,4-ethylenedioxythiophene) poly(styrenesulphonate)
(PEDOT:PSS) and the nanocomposite layer as an electron-
blocking/hole-conducting layer. This material blend combines the
hole-injection and hole-transport capabilities of TPD-Si
2
24
with
the electron-blocking capability of PVK and has been shown to
reduce the dark current by two to three orders of magnitude in
our devices. The hole-transport layer was crosslinked by annealing
TPD-Si
2
in air so that the photoactive layer coating step that
Department of Mechanical and Materials Engineering, University of Nebraska-Lincoln, Lincoln, Nebraska 68588, USA.
*
e-mail: jhuang2@unl.edu
follows would not wash it away. Similarly, a hole-blocking/electron-
conducting layer—2,9-dimethyl-4,7-diphenyl-1,10-phenanthroline
(BCP)—was also deposited by thermal evaporation on the
cathode side.
The hybrid photodetector works as follows. First, both the nano-
particles and polymers absorb incident photons and generate
Frenkel excitons (Fig. 1c, 1). The Frenkel excitons then diffuse to
the polymer/nanoparticles interface and the electrons transfer
from the nanoparticles and semiconducting polymer, as shown in
the energy diagram (Fig. 1d, 2). Holes are transported in the semi-
conducting polymer under the applied reverse bias/electric field,
while the electrons remain trapped in the nanoparticles due to
the lack of a percolation network for electrons and the strong
quantum confinement effect in nanoparticles (Fig. 1c, 3). The
strong electron trapping effect is demonstrated by the very small
electron current in the electron-only devices, which is three to
four orders of magnitude lower than the hole current in hole-only
devices using the same nanocomposite layers as the carrier transport
layer (Supplementary Fig. S2). The electron trapping effect of the
ZnO nanoparticles was also directly observed by electrostatic force
microscopy (EFM) (Supplementary Fig. S3). In the absence of illu-
mination, the dark current is small because of the very large charge
injection barrier (.0.6 eV) under reverse bias, as illustrated in
Fig. 1e. Under illumination, the trapped electrons quickly shift the
lowest unoccupied molecular orbital (LUMO) of the polymer down-
wards and align the Fermi energy of the nanocomposite with that of
the cathode. The electron traps are predominantly located close to
the cathode because of the formation of vertical phase separation
in the nanocomposite, with ZnO nanoparticles segregated to the
cathode side, as observed by cross-sectional scanning electron
microscopy (SEM; Supplementary Fig. S4). This phase separation
(both lateral and vertical) has been observed widely in P3HT:ZnO
nanocomposites and was purposely promoted by the slow drying
of the film taking advantage of the different surface affinity of
ZnO and P3HT for the substrate
20,25–27
. The hole-injection barrier
on the cathode side becomes so thin that the holes can easily
tunnel through it with a small reverse bias (Fig. 1f). Accordingly,
the nanocomposite/aluminium interface acts as a photon-address-
able optoelectronics ‘valve’ for hole injection, and incident photons
can switch on this ‘valve’. The average energy barrier change, D
F
,
is a linear function of trapped electron density n
t
, and the injection
current follows an exponential relationship with the energy barrier
change according to the Richardson–Dushman equation:
J / exp −
D
F
kT
/ exp
n
t
kT
where k is the Boltzmann constant and T is the temperature. The
gain of a photodetector is the ratio of the measured photocurrent
(carriers) and the number of incident photons. If the injected hole
number exceeds the absorbed photon number, there is gain as a
result of the exponential dependence of the injected holes on
incident photons.
To characterize the wavelength-dependent gain of the photode-
tectors, external quantum efficiency (EQE) versus wavelength was
measured using an incident photon-to-current efficiency (IPCE)
system at different reverse biases (Fig. 2a,b). The EQE curves also
agree with the absorption curves of the nanocomposites. EQE
values exceed 100% at a bias of 23 V for PVK:ZnO devices and
21 V for P3HT:ZnO devices, and increase quickly with increasing
negative bias, especially at reverse biases above 28 V. The rapid
increase of EQE is consistent with the rapid increase of photocur-
rent, as shown in Fig. 2c. For a bias of 29 V (the highest voltage
output of our light bias amplifier) at 360 nm, the peak EQE
values are 245,300% and 340,600% for the PVK:ZnO and
a
de f
bc
300 350 400 450 500 550 600 650 700
0
20
40
60
80
100
Absorption (%)
Wavelength (nm)
P3HT:ZnO
PVK:ZnO
P3HT
−
NP
Ligand
E field
1
2
3
3
1
LUMO
CB
VB
HOMO
−
3.2
5.1
2
4.3
7.65
3.2
5.1
X
−
−
−
−
−
−
−
−
−
−
−
−
−−
−
−
+
+
+
+
+
+
+
+
+
−
−
−
−
−
−
+
+
+
+
+
+
AI
BCP
PVK:TPD-Si
2
PEDOT:PSS
ITO
Glass substrate
−
Figure 1 | Device structure and working principle of the photodetector. a, Schematic layout of the photodetector structure. b, Absorption sp ectra of the
P3HT:Z nO and PVK:ZnO nanocomposite films. c, Illustra tion of electron–hole pair genera tion (1), splitting (2), hole transport and electron trapping process (3)
in the nanocomposite. d, Energy diagram of the nanoparticle with the su rr ounding polymer tha t also shows the steps 1–3. CB, conduction band; VB, valence
band; LUMO, lowest unoccupied molecular orbital; HOMO, highest occupied molecular orbital. e,f, Energy diagram of the de vice in the dark (e)andunder
illumination (f).
P3HT:ZnO devices, respectively. The corresponding responsivity (R
in A W
21
)—that is, the ratio of photocurrent to intensity of incident
light—can be calculated from the EQE using
R = EQE/hn
where hn is the energy of the incident photon in electronvolts. The
peak responsivities at an illumination light intensity of
1.25 mWcm
22
are 721 A W
21
for the PVK:ZnO device and
1,001 A W
21
for the P3HT:ZnO device at 360 nm; these values
are more than three orders of magnitude larger than those of com-
mercial GaN or SiC photodetectors (,0.2 A W
21
). Indeed, they are
the highest reported responsivities of all types of solid-state ultra-
violet photodetectors
2,5,6
.
Our nanocomposite photodetector devices show a transition in
operation from a photodiode in the dark with a rectifying
Schottky contact, to a photoconductor under illumination with an
ohmic contact, as shown by the dark current and photocurrent
traces in Fig. 2c. The combination of the low dark current of the
photodiode and large gain of the photoconductor is expected to
yield a new type of photodetector with high sensitivity. The figure
of merit for a photodetector is the specific detectivity that character-
izes the capability of a photodetector to detect the weakest light
signal
9
. In addition to responsivity, the other factor that limits the
specific detectivity of a photodetector is the noise current. The
dark current of our device is as low as 6.8 nA at 29 V because of
the blocking contact both at the anode and cathode sides under
dark conditions (Fig. 2c), which provides a very low shot noise.
To include other possible noise, such as flicker noise and thermal
noise, the total noise current of the photodetector was directly
measured with an SR830 lock-in amplifier under different con-
ditions of dark current density and frequency
3,4
. As shown in
Fig. 3a,b, the measured total noise current was found to be domi-
nated by the shot noise within the frequency range 1 Hz to 5 kHz.
The specific detectivities (D*) of a photodetector are given by
7,9
D
∗
=
AB()
1/2
NEP
cm Hz
1/2
W
−1
or Jones
NEP =
i
2
n
1/2
R
W
()
where A is the device area, B is the bandwidth, NEP is the noise
equivalent power,
i
2
n
1/2
is the measured noise current, and R is the
responsivity. The detectivities of our nanocomposite photodetector
were calculated at different wavelengths with the measured noise
current and responsivity at 29 V bias, and the results are plotted
in Fig. 3c. At an illumination light intensity of 1.25 mWcm
22
,
the specific detectivities at 360 nm were 3.4 × 10
15
Jones for
PVK:ZnO devices and 2.5 × 10
14
Jones for P3HT:ZnO devices.
The specific detectivities in the ultraviolet range were two to three
orders of magnitude larger than those of silicon and GaN ultraviolet
photodetectors. The specific detectivity of a P3HT:ZnO device
within the visible light range was also more than ten times better
than that of silicon photodetectors.
Another important parameter of photodetectors is their response
speed. The temporal response of our nanocomposite photodetector
was characterized using a chopper-generated short light pulse.
Figure 4 shows the transient photocurrent of the P3HT:ZnO
device measured under a bias of 29 V at a light intensity of
1 mWcm
22
. The transient response result shows a rise time
(output signal changing from 10% to 90% of the peak output
value) of 25 ms, which was limited by the rising edge of the light
pulse from the optical chopper. The decay of the photocurrent
after switching off the ultraviolet pulse has a fast component of
142 ms and a slow component of 558 ms, which indicates the exist-
ence of two channels for the recombination of holes. The response
speed is among the highest reported in any nanoparticle- or col-
loidal quantum dot-based photodetector
2–4,6,8
. The 3 dB bandwidth
is 9.4 kHz. The devices provide an improvement by a factor of over
1 × 10
5
in gain–bandwidth product in solution-processed ZnO
ultraviolet photodetection relative to previous reports
2
. The mul-
tiple-exponential decay time can arise from the existence of electron
traps with different trap depths due to the non-uniform nature of
ZnO nanoparticles or aggregates in the present hybrid devices.
Deeper traps have longer charge release times and thus result in a
slower device response speed. It should be mentioned that the
photodetector response speed is related to trap occupancy, which
depends on light intensity. At lower light intensity, the photocurrent
decay is expected to be dominated by the slower process of 558 ms,
because deeper traps are easier to be filled. The response speed of
hybrid devices at a light intensity of ,1 mWcm
22
was not
measured, because the lower light intensity could not provide suffi-
cient signal for the present measurement system. Increasing the
response speed at lower light levels by improving the uniformity
of the ZnO nanoparticles is still under investigation. The photocon-
ductive gain is the ratio between hole recombination time, or device
switch-off time, and the transit time during which holes sweep
through the nanocomposite film to the ITO. The gain of the
P3HT:ZnO device calculated from the measured hole mobility
and hole recombination time is 3,798, which is very close to the
gain measured by IPCE (see Supplementary Information).
A photodetector needs to have a large linear dynamic range so as
to measure both strong and weak light. The linear dynamic range of
the nanocomposite photodetector was also characterized by
measuring the photocurrent at a fixed frequency of 35 Hz but with
varied light intensity from 1 × 10
21
Wcm
22
to 1 × 10
212
Wcm
22
.
a
bc
300 400 500 600 700
10
−1
10
0
10
1
10
2
10
3
10
4
10
5
10
−1
10
−2
10
0
10
1
10
2
10
3
10
4
10
5
−9 V
−9 V
EQE (%)
EQE (%)
Wavelength (nm)
0 V
300 325 350 375 400 425 450 475
Wavelength (nm)
−10 −8 −6 −4 −2 0 2 4 6 8 10
10
−8
10
−7
10
−6
10
−5
10
−4
10
−3
10
−2
10
−1
10
0
Dark current
J (A cm
−2
)
Voltage (V)
Photocurrent
0 V
Figure 2 | Performance of the photodetector. a,b, EQEs of the P3HT:ZnO devic e under reverse bias with a voltage step of 1 V (a) and the PVK:ZnO device
with bias of 0 V, 21V,23V,25V,27V,28V,28.5 V and 29 V fr om bottom to top (b). c, Photocurrent and dark current density of the PVK:ZnO device.
As shown in Fig. 5, the PVK:ZnO photodetector shows a linear
response within the incident light intensity range from
1 × 10
21
Wcm
22
to 1 × 10
29
Wcm
22
, corresponding to a linear
dynamic range of 80 dB. This is among the highest reported
linear dynamic ranges for both inorganic and organic ultraviolet
photodetectors. The responsivity remains almost constant in this
light intensity range, despite a slight (10%) drop at high light
levels (Fig. 5, inset). This slight sublinear response at high light
intensities is possibly caused by electron trapping saturation
and/or limitation of hole mobility in the nanocomposite layers.
The device begins to lose its linearity when the incident light inten-
sity is below 1 × 10
29
Wcm
22
. The responsivity drops to
52 A W
21
and the specific detectivity accordingly drops to
2.45 × 10
14
Jones at a light intensity of 1.25 × 10
212
Wcm
22
.
The sublinearity and reduced detectivity at low light levels is a
disadvantage, because high gain at low light intensity is desired
for weak light detection; however, it can be improved by tuning
the morphology of the nanocomposite layer. In principle, we
expect a constant responsivity down to the lowest detectable inci-
dent light level (NEP) if the automatic transition from the
Schottky junction to ohmic contact occurs at such a low incident
light level. This is possible, because the incident photons can
cause band bending in the local environment surrounding the
light-absorbing ZnO nanoparticles, which induces strong local
hole injection. However, the degree of local band bending varies
10
−12
10
−13
10
−11
10
−9
10
−7
10
−5
10
−3
10
−1
10
−10
10
−8
10
−6
10
−4
10
−2
10
0
10
2
J (A cm
−2
)
Irradiance (W cm
−2
)
10
−13
10
−11
10
−9
10
−7
10
−5
10
−3
10
−1
10
1
10
2
10
3
R (A W
−1
)
Irradiance (W cm
−2
)
Figu re 5 | Dynamic range of the PV K:ZnO photodetector . The device
shows a linear response within the incident light intensity range, which
corresponds to a linear dynamic range of 80 dB. Inset: responsivities under
different illumination intensities.
0.0 0.5 1.0 1.5 2.0
0
1
Photocurrent (a.u)
Time (ms)
τ
1
= 142 μs
τ
2
= 558 μs
Figure 4 | Transient photocurrent of the P3HT:ZnO device. Photocurrent is
measured under a bias of 29 V with a light inte nsity of 1
m
Wcm
22
,andan
optical chopper is used to provide short light pulses. After switching of f the
ultraviolet pulse, the decay of the photocurr ent has a fast com ponent of
142
m
s and a slow component of 558
m
s, which indicates that there are two
channels for the recombination of holes.
a
b
c
60 80 100 120 140
10
−2
10
−1
10
−2
10
−3
10
15
10
14
10
13
10
−1
10
0
10
0
10
1
10
2
10
3
10
−1
10
0
10
1
Noise current (pA Hz
−1/2
)
Noise current (pA Hz
−1/2
)
Dark current (nA)
Shot noise limit
Thermal noise limit
Noise current
(pA Hz
−1/2
)
Noise current
(pA Hz
−1/2
)
Frequency (Hz)
10
0
10
1
10
2
10
3
Frequency (Hz)
P3HT:ZnO
345678
10
−2
10
−1
10
0
10
1
Dark current (nA)
Shot noise limit
Thermal noise limit
PVK:ZnO
300 600 900
D* (Jones)
Wavelength (nm)
PVK:ZnO
P3HT:ZnO
Silicon
GaN
Figure 3 | Noise characteristics and specific detectivity of the
photodetector. a,b, Noise current of the P3HT:ZnO (a) and PVK:ZnO (b)
devices under different dark curr ents. Insets: frequency-dependent noise
current at 29V.c, Specific detectivities of the photodetector a t
different wavelengths.
with the morphology of the nanocomposite layer. If there is an
aggregation of ZnO nanoparticles, which is very likely to occur in
our material system, the local average trapped electron density will
be reduced and the induced charge injection will be weakened. In
addition, there are still many ZnO nanoparticles located in the
middle of the nanocomposite layers or at the anode side, despite
the higher concentration of ZnO nanoparticles at the cathode side
due to the TPD-Si
2
interface-induced vertical phase separation.
Light absorption by these ZnO nanoparticles located far away
from the cathode will not induce as much of a Schottky junction-
narrowing effect as those close to the cathode side. This non-ideal
morphology might increase the lowest light intensity detectable by
the nanocomposite photodetector. The influence of morphology
on the lowest detectable light intensity is still under investigation,
and we expect to see a lower limit of detectable light intensity and
a better linear response by pushing more ZnO nanoparticles
closer to the cathode side.
We have reported a new type of hybrid photodetector that has a
Schottky contact in the dark and an ohmic contact under illumina-
tion, enabled by interfacial trap-controlled charge injection. Its
specific detectivity of 3.4 × 10
15
Jones is tens to hundreds of times
better than that of inorganic semiconductor photodetectors. It
should be noted that ZnO nanomaterial-based ultraviolet photode-
tectors have been extensively explored over the past decade, with
most effort focusing on single nanowires because of their quick
response, which arises from their large carrier mobility
2,12–19
.
However, for several reasons, these devices have not shown
comparable performance or any advantage over inorganic
ultraviolet photodetectors. First, the photodetectors have been
constructed from a single nanowire and have not been scalable to
large areas using current synthesis techniques. Second, the ZnO
nanowires are directly connected to two electrodes, leading to
high dark current. In contrast to these earlier photodetectors,
which were constructed with material systems similar to ours, our
ultraviolet photodetectors can be made at low cost, and can be
scaled up easily to large areas. They also have very high gain and
very low dark current because of the absence of ohmic contact in
the dark, and can respond quickly. Our nanocomposite photo-
detectors have great potential for replacing inorganic ultraviolet
photodetectors and for opening avenues to new applications.
Methods
ZnO nanoparticles were prepared using a hydrolysis method in methanol with some
modifications. EFM was used to characterize the topography and electron trap
distribution in the nanocomposite films. The cross-sectional morphology of the
P3HT:ZnO nanoparticle nanocomposite was measured by SEM. EQE was measured
with a Newport QE measurement kit by focusing a monochromatic beam of light
onto the devices. For the transient response measurement, an optical chopper was
used to provide the light pulse, and an oscilloscope (LeCroy WaveRunner) was used
to record the voltage variation of the resistor. The absorption spectra of the
photoactive layers were measured with a PerkinElmer Lambda 900 spectrometer.
Film thickness was measured with an AMBIOS XP-2 stylus profilometer. Noise
current was directly measured with a lock-in amplifier SR830.
Received 10 May 2012; accepted 24 September 2012;
published online 11 November 2012
References
1. Li, W. D. & Chou, S. Y. Solar-blind deep-UV band-pass filter (250–350 nm)
consisting of a metal nano-grid fabricated by nanoimprint lithography.
Opt. Express 18, 931–937 (2010).
2. Jin, Y. Z., Wang, J. P., Sun, B. Q., Blakesley, J. C. & Greenham, N. C. Solution-
processed ultraviolet photodetedtors based on colloidal ZnO nanoparticles.
Nano Lett. 8, 1649–1653 (2008).
3. Konstantatos, G. et al. Ultrasensitive solution-cast quantum dot photodetectors.
Nature 442, 180–183 (2006).
4. Chen, H., Lo, M., Yang, G., Monbouquette, H. & Yang, Y. Nanoparticle-assisted
high photoconductive gain in composites of polymer and fullerene. Nature
Nanotech. 3, 543–547 (2008).
5. Sukhovatkin, V., Hinds, S., Brzozowski, L. & Sargent, E. H. Colloidal
quantum-dot photodetectors exploiting multiexciton generation. Science
324, 1542–1544 (2009).
6. Konstantatos, G. & Sargent, E. H. Nanostructured materials for photon
detection. Nature Nanotech. 5, 391–400 (2010).
7. Gong, X. et al. High-detectivity polymer photodetectors with spectral response
from 300 nm to 1450 nm. Science 325, 1665–1667 (2009).
8. Arnold, M. S. et al. Broad spectral response using carbon nanotube/organic
semiconductor/C60 photodetectors. Nano Lett. 9, 3354–3358 (2009).
9. Liu, J-M. Photonic Devices 960–986 (Cambridge Univ. Press, 2005).
10. Yoshikawa, H. & Adachi, S. Optical constants of ZnO. Jpn. J. Appl. Phys. 36,
6237–6243 (1997).
11. Redington, R. W. Maximum performance of high-resistivity photoconductors.
J. Appl. Phys. 29, 189–193 (1958).
12. Chen, K. J., Hung, F. Y., Chang, S. J. & Young, S. J. Optoelectronic characteristics
of UV photodetector based on ZnO nanowire thin films. J. Alloy Comp. 479,
674–677 (2009).
13. Li, Y., Della Valle, F., Simonnet, M., Yamada, I. & Delaunay, J-J. High-
performance UV detector made of ultra-long ZnO bridging nanowires.
Nanotechnology 20, 045501 (2009).
14. Kind, H., Yan, H. Q., Messer, B., Law, M. & Yang, P. D. Nanowire ultraviolet
photodetectors and optical switches. Adv. Mater. 14, 158–160 (2002).
15. Soci, C. et al. ZnO nanowire UV photodetectors with high internal gain.
Nano Lett. 7, 1003–1009 (2007).
16. Liang, S. et al. ZnO Schottky ultraviolet photodetectors. J. Cryst. Growth 225,
110–113 (2001).
17. Lao, C. S. et al. Giant enhancement in UV response of ZnO nanobelts by
polymer surface-functionalization. J. Am. Chem. Soc. 129, 12096–12097 (2007).
18. Das, S. N. et al. ZnO single nanowire-based UV detectors. Appl. Phys. Lett.
97, 022103 (2010).
19. Basak, D., Amin, G., Mallik, B., Paul, G. K. & Sen, S. K. Photoconductive UV
detectors
on sol–gel-synthesized ZnO films. J. Cryst. Growth 256, 73–77 (2003).
20. Oosterhout, S. D. et al. The effect of three-dimensional morphology on the
efficiency of hybrid polymer solar cells. Nature Mater. 8, 818–824 (2009).
21. Sun, B. Q., Marx, E. & Greenham, N. C. Photovoltaic devices using
blends of branched CdSe nanoparticles and conjugated polymers.
Nano Lett. 3, 961–963 (2003).
22. Huynh, W. U., Dittmer, J. J. & Alivisatos, A. P. Hybrid nanorod–polymer
solar cells. Science 295, 2425–2427 (2002).
23. Verbakel, F., Meskers, S. C. J. & Janssen, R. A. J. Electronic memory
effects in diodes from a zinc oxide nanoparticle–polystyrene hybrid material.
Appl. Phys. Lett. 89, 102103 (2006).
24. Huang, Q. et al. Covalently bound hole-injecting nanostructures. Systematics
of molecular architecture, thickness, saturation, and electron-blocking
characteristics on organic light-emitting diode luminance, turn-on voltage,
and quantum efficiency. J. Am. Chem. Soc. 127, 10227–10242 (2005).
25. Beek, W. J. E., Wienk, M. M., Kemerink, M., Yang, X. N. & Janssen, R. A. J.
Hybrid zinc oxide conjugated polymer bulk heterojunction solar cells. J. Phys.
Chem. B 109, 9505–9516 (2005).
26. Campoy-Quiles, M. et al. Morphology evolution via self-organization and lateral
and vertical diffusion in polymer: fullerene solar cell blends. Nature Mater. 7,
158–164 (2008).
27. Baoquan, S., Snaith, H. J., Dhoot, A. S., Westenhoff, S. & Greenham, N. C.
Vertically segregated hybrid blends for photovoltaic devices with improved
efficiency. J. Appl. Phys. 97, 014914 (2005).
Acknowledgements
This work was supported by the Office of Naval Research (ONR, grant no.
N000141210556), a Defense Threat Reduction Agency (DTRA) Young Investigator
Award (HDTRA1-10-1-0098) and the University of Nebraska–Lincoln.
Author contributions
J.H. conceived the idea. J.H. and F.G. designed the experiments and analysed the data.
F.G. carried out the fabrication of devices, measurements and data analysis. Y.B., B.Y. and
Q.D. synthesized ZnO nanoparticles. B.Y. and Z.X. fabricated the single carrier devices.
J.H. and F.G. wrote the manuscript.
Additional information
Supplementary information is available in the online version of the paper.
Reprints and
permission information is available online at http://www.nature.com/reprints. Correspondence
and requests for materials should be addressed to J.H.
Competing financial interests
The authors declare no competing financial interests.
