Recent progress of infrared photodetectors based on lead chalcogenide colloidal quantum dots

Abstract

Commercial photodetectors based on silicon are extensively applied in numerous fields. Except for their high performance, their maximum absorption wavelength is not over than 1100 nm and incident light with longer wavelengths cannot be detected; in addition, their cost is high and their manufacturing process is complex. Therefore, it is meaningful and significant to extend absorption wavelength, to decrease cost, and to simplify the manufacturing process while maintaining high performance for photodetectors. Due to the properties of size-dependent bandgap tunability, low cost, facile processing, and substrate compatibility, solution-processed colloidal quantum dots (CQDs) have recently gained significant attention and become one of the most competitive and promising candidates for optoelectronic devices. Among these CQDs, lead chalcogenide CQDs are getting very prominent and are widely investigated. In this paper, the recent progress of infrared (IR) photodetectors based on lead sulfide (PbS), lead selenide (PbSe), and ternary PbS x Se 1-x CQDs, and their underlying concepts, breakthroughs, and remaining challenges are reviewed, thus providing guidance for designing high-performance quantum-dot IR photodetectors.

Full text

Recent progress of infrared photodetectors based on lead chalcogenide colloidal quantum dots
Jinming Hu(胡津铭), Yuansheng Shi(史源盛), Zhenheng Zhang(张珍衡), Ruonan Zhi(智若楠), Shengyi Yang(杨盛谊),
Bingsuo Zou(邹炳锁)
Citation:Chin. Phys. B . 2019, 28(2): 020701. doi: 10.1088/1674-1056/28/2/020701
Journal homepage: http://cpb.iphy.ac.cn; http://iopscience.iop.org/cpb
What follows is a list of articles you may be interested in
Performance enhancement of ZnO nanowires/PbS quantum dot depleted bulk heterojunction
solar cells with an ultrathin Al2O3 interlayer
Shuaipu Zang(臧帅普), Yinglin Wang(王莹琳), Meiying Li(李美莹), Wei Su(苏蔚), Meiqi An(安美琦), Xintong Zhang(张昕
彤), Yichun Liu(刘益春)
Chin. Phys. B . 2018, 27(1): 018503. doi: 10.1088/1674-1056/27/1/018503
Experimental research on spectrum and imaging of continuous-wave terahertz radiation based
on interferometry
Tie-Lin Lu(卢铁林), Hui Yuan(袁慧), Ling-Qin Kong(孔令琴), Yue-Jin Zhao(赵跃进), Liang-Liang Zhang(张亮亮), Cun-Lin
Zhang(张存林)
Chin. Phys. B . 2016, 25(8): 080702. doi: 10.1088/1674-1056/25/8/080702
Band gap anomaly and topological properties in lead chalcogenides
Simin Nie(聂思敏), Xiao Yan Xu(许霄琰), Gang Xu(徐刚), Zhong Fang(方忠)
Chin. Phys. B . 2016, 25(3): 037311. doi: 10.1088/1674-1056/25/3/037311
Nb5N6 microbolometer array for terahertz detection
Tu Xue-Cou, Kang Lin, Liu Xin-Hua, Mao Qing-Kai, Wan Chao, Chen Jian, Jin Biao-Bing, Ji Zheng-Ming, Xu Wei-Wei, Wu
Pei-Heng
Chin. Phys. B . 2013, 22(4): 040701. doi: 10.1088/1674-1056/22/4/040701
Enhancement of terahertz coupling efficiency by improved antenna design in GaN/AlGaN high
electron mobility transistor detectors
Sun Yun-Fei, Sun Jan-Dong, Zhang Xiao-Yu, Qin Hua, Zhang Bao-Shun, Wu Dong-Min
Chin. Phys. B . 2012, 21(10): 108504. doi: 10.1088/1674-1056/21/10/108504
--------------------------------------------------------------------------------------------------------------------

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
TOPICAL REVIEW — Photodetector: materials, physics, and applications
Recent progress of infrared photodetectors based on lead
chalcogenide colloidal quantum dots∗
Jinming Hu(胡津铭)1,2, Yuansheng Shi(史源盛)1,2, Zhenheng Zhang(张珍衡)1,2,
Ruonan Zhi(智若楠)1,2, Shengyi Yang(杨盛谊)1,2,†, and Bingsuo Zou(邹炳锁)1,2,‡
1Beijing Key Laboratory of Nanophotonics and Ultrafine Optoelectronic Systems, School of Physics,
Beijing Institute of Technology, Beijing 100081, China
2Key Laboratory of Advanced Optoelectronic Quantum Design and Measurement, Ministry of Education,
Beijing Institute of Technology, Beijing 100081, China
(Received 28 August 2018; revised manuscript received 19 December 2018; published online 25 January 2019)
Commercial photodetectors based on silicon are extensively applied in numerous fields. Except for their high perfor-
mance, their maximum absorption wavelength is not over than 1100 nm and incident light with longer wavelengths cannot
be detected; in addition, their cost is high and their manufacturing process is complex. Therefore, it is meaningful and signif-
icant to extend absorption wavelength, to decrease cost, and to simplify the manufacturing process while maintaining high
performance for photodetectors. Due to the properties of size-dependent bandgap tunability, low cost, facile processing,
and substrate compatibility, solution–processed colloidal quantum dots (CQDs) have recently gained significant attention
and become one of the most competitive and promising candidates for optoelectronic devices. Among these CQDs, lead
chalcogenide CQDs are getting very prominent and are widely investigated. In this paper, the recent progress of infrared
(IR) photodetectors based on lead sulfide (PbS), lead selenide (PbSe), and ternary PbSxSe1−xCQDs, and their underlying
concepts, breakthroughs, and remaining challenges are reviewed, thus providing guidance for designing high-performance
quantum-dot IR photodetectors.
Keywords: colloidal quantum dots, lead chalcogenide, infrared photodetectors, nanocrystals
PACS: 07.57.Kp, 85.35.Be, 73.50.Pz, 85.60.Dw DOI: 10.1088/1674-1056/28/2/020701
1. Introduction
As we know, the fabrication of photodetectors with con-
ventional nanocrystalline semiconductors, such as silicon and
InGaAs, requires high-temperature processing under high-
vacuum condition, and it is associated with expensive facil-
ities. In this way, there is no doubt that the entire cost and
complexity will increase. The growth of semiconductors re-
lies on the crystalline substrates and it follows the principle of
lattice matching, and such a kind of system cannot be applied
on flexible platforms. Additionally, there are many challenges
in extending absorption, decreasing volume, and decreasing
high voltage for a photomultiplier tube in infrared (IR) detec-
tion applications.
During the past decades, among the candidate mate-
rials studied,[1]lead chalcogenide colloidal quantum dots
(CQDs), i.e., nanocrystals (NCs), have recently reaped quanti-
ties of attention owing to their properties of highly efficient
multiple exciton generation (MEG),[2]low cost, and facile
implementation for varieties of optoelectronic applications,
such as solar cells,[3–8]light-emitting diodes, [9–12]field-effect
transistors,[13–16]and photodetectors. [3,17–19]A primary ben-
efit of lead chalcogenide CQDs originates in their solution-
processed techniques: it promotes integration with substrates,
thus CQDs are almost unlimited by the types of substrates,
and they can be readily deposited by spin-coating, spray-
casting, and inkjet printing techniques. But beyond that, lead
chalcogenide CQDs’ optical absorption and emission spectra
are widely tunable by exploiting the quantum size effect,[20]
thereby they exhibit a broad spectral response from the ultravi-
olet to the near-infrared region. Furthermore, spectral tunabil-
ity eliminates the need for optical filters to select the spectrum
of interest, leading to the decrement of cost and complexity,
and avoiding loss of signal when the filtering and transmit-
ting are implemented. Among lead chalcogenide CQDs, PbS,
PbSe, and PbSxSe1−xCQDs are most widely studied and ap-
plied in the IR detection.
In this review, firstly a brief introduction to PbS, PbSe,
and PbSxSe1−xCQDs and their optoelectronic properties are
presented. And then the fundamental operating principle, in-
dividual merits and drawbacks, and relative trade-offs of dif-
ferent types of photodetectors are discussed. The terminology
and figures of merit applied in photodetectors which evalu-
∗Project supported by the Fund from the State Key Laboratory of Transducer Technology, China (Grant No. SKT1404) and the Fund from the Key Laboratory
of Photoelectronic Imaging Technology and System (Grant No. 2017OEIOF02) at Beijing Institute of Technology, Ministry of Education of China.
†Corresponding author. E-mail: syyang@bit.edu.cn
‡Corresponding author. E-mail: zoubs@bit.edu.cn
© 2019 Chinese Physical Society and IOP Publishing Ltd http://iopscience.iop.org/cpb http://cpb.iphy.ac.cn
020701-1

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
ate the performance of a photodetector are also summarized.
Then, an overview on recent progress in photodetectors based
on these three CQDs is presented in Table 1. Finally, a novel
integrating device, i.e., IR detecting and display devices (IR-
DDD), which directly integrates lead chalcogenide CQDs IR
photodetector with light-emitting diodes, is reviewed, as well
as tandem photodetectors.
2. Lead chalcogenide CQDs for light detection
The bandgap of bulk PbS and PbSe is ∼0.41 eV and
∼0.28 eV, respectively, and the considerably large average
distance between the electron and the corresponding created
hole under the effect of Coulomb binding is ∼18 nm[21]and
∼46 nm[22](i.e., bulk exciton Bohr radius). To decrease the
dimensions of bulk materials to the de Broglie wavelength of
these charge carriers, or approximately equal to a CQD di-
ameter of less than twice the bulk exciton Bohr radius, there
emerges the quantum size effect, resulting in CQDs’ size-
dependent bandgap tunability. Ternary PbSxSe1−xCQDs are
alloyed NCs, combining PbS NCs and PbSe NCs, and their
bandgap can be tuned by changing the S/Se atomic ratio.
For most optoelectronic applications, highly monodis-
perse and narrow size-distribution are critically required for
the synthesized CQDs. Typically, the formation of monodis-
perse CQDs consists of two steps: a rapid nucleation fol-
lowed by a slow growth process.[23,24]Several factors to con-
trol the rate of CQDs’ nucleation are temperature, interfacial
tension, and degree of supersaturation in solution.[25]The nu-
cleation of CQDs can be terminated by decreasing the con-
centration below a corresponding level or rapidly reducing the
reaction temperature by fast injecting the precursor(s) into the
hot mixed solution (hot-injection synthesis).[26]After nucle-
ation, there are two reciprocally competing behaviors during
the growth process: focusing and defocusing. The focusing
behavior is based on the diffusion-controlled growth model.
Compared to small particles, large particles grow more slowly
since the particles’ growth rate is inversely proportional to
their radii. In addition, provided that the secondary nucleation
is avoided and all particles grow, narrower size-distribution
can be gained.[27]The defocusing behavior is based on the
solubility of particles and depends on their sizes; smaller par-
ticles dissolve and larger particles continue to grow (Ostwald
ripening).[28]During a typical synthesis, the focusing behav-
ior prevails firstly, and then a great number of particles emerge
and form a near-symmetrical size-distribution. The defocusing
behavior predominates when the average radius of particles
exceeds a critical figure, and the number of particles decreases
and a broader and asymmetrical size-distribution forms.
Wavelength/nm
Energy/eV
(e)
(d)
(a) (c)
Absorbance/a.u.
(b)
Fig. 1. (a) Schematic diagram of hot-injection synthesis;[23](b) schematic diagram of heating-up method; [29](c) size-dependent optical
absorption spectra supplied by PbS CQDs;[19](d) transmission electron microscopy (TEM) images of a two-dimensional self-assembled PbS
CQD and a three-dimensional self-assembled PbS CQD superlattice;[30]and (e) PbS CQD conduction band and valence band energy level
tuning via capping with different ligands. Starting from Br−and moving clockwise in the left panel, the corresponding band-edge tuning is
exhibited from left-side to right-side in the right panel.[31]
There are varieties of methods to synthesize CQDs. How-
ever, most size- and shape-controlled CQDs are obtained by
utilizing thermal decomposition method due to its low cost and
facile implementation.[23,26,32]With regard to the segregation
of nucleation and growth procedures by thermal decomposi-
tion, it is competent either by the hot-injection synthesis or
020701-2

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
the heating-up method. Figure 1(a) illustrates a schematic di-
agram of the hot-injection. The nucleation occurs upon the in-
jection of a room-temperature reagent into a high-temperature
surfactant (which is also known as ligand), followed by re-
action cooling for going into the growth stage.[33]For the
heating-up method (see Fig. 1(b)), this two-step mechanism
can be actualized by a period of stable heating of the mixture
comprising precursor(s) and ligands.[29]
For lead chalcogenide CQDs, the hot-injection synthesis
is the most widely used method. By controlling the related
precursors’ concentration, ligands, temperature, and reaction
time, CQDs of desired size and shape can be obtained. Fig-
ure 1(c) shows the absorption tunability of PbS CQDs via al-
tering the CQD size from 4.3 nm to 8.4 nm, and the excitonic
maxima ranging from 1000 nm to 1800 nm can be realized,
corresponding to bandgap from 1.25 eV to 0.70 eV.[19]In prac-
tice, the following equation (1) exhibits the explicit relation-
ship between the CQD size diameter (din unit nm) and the
first exciton absorption peak (Egin unit eV, corresponding to
1Sh–1Setransition):[30]
Eg=0.41 +1
0.0392d2+0.114d.(1)
Lead chalcogenide CQDs can also be self-assembled into
regular arrays or superlattices (see Fig. 1(d)).[30]In most cases,
surfactant molecules such as oleic acid (OA) are coated on the
surface of lead chalcogenide CQDs, enabling CQDs to dis-
perse uniformly in varieties of solvents and restrain CQDs
from aggregation. The majority of surfactant molecules ap-
plied in CQDs’ synthesis contain two parts: a long hydro-
carbon tail and a polar head. The selection of surfactant
molecules depends on the coordination chemistry with CQD
surface atoms. According to the Lewis’ theory of acids and
bases, a hard Lewis acid prefers to bond to a hard Lewis base
and this theory can also be applied to the soft Lewis acid and
soft Lewis base.[34]Moreover, via capping with different lig-
ands, the position of conduction band (CB) and valence band
(VB) in CQDs can be tuned whilst maintaining a constant
bandgap.[31]Therefore, the energy level of lead chalcogenide
CQD can be tuned to optimize the energy level alignment with
types of commonly used charge extraction layers. The effec-
tive integration with these layers can obviously enhance the
device performances.
3. Fundamentals of CQD photodetectors
3.1. Types of CQD photodetectors
Basically, there are three kinds of CQD photodetectors:
photoconductors, phototransistors, and photodiodes.
A typical CQD photoconductor (see Fig. 2(a)) is com-
prised of a photoactive semiconducting CQD layer and two
ohmic contacts integrated with metal to form a two-port op-
toelectronic device. It works on the principle of a tempo-
rary transformation of carrier mobility or density or both in
the semiconducting CQD layer when irradiated under inci-
dent illumination. The enhancement of conductivity under
illumination is most typically ascribed to considerable exci-
ton generation. The mechanism of conductivity in bulk lead
chalcogenide film has been extensively studied.[38]Here, we
will briefly retrospect the concept of photoconductive gain and
the role of trap states in photodetectors based on lead chalco-
genide CQD films to interpret the mechanism of photoconduc-
tive detectors.
An initial current is generated resulting from the absorp-
tion of photons, followed by the extraction of photo-generated
carriers, which dominates the external quantum efficiency
(EQE) of the whole device. This initial photocurrent is the
principal source of the photoconductivity. It may also orig-
inate from a second photocurrent generated by the injection
and the transit of carriers from the electrodes. This second
photocurrent dominates the photoconductive gain G, which is
expressed as
G=τlt
τtt
=τltµV
l2τtt =l
µF=l2
µV,(2)
where τlt is the lifetime of the majority carriers, τtt is the transit
time of the majority carriers, µis the mobility of the majority
carriers, Vis the bias voltage, lis the distance between the two
electrodes, and Fis the electric field arising from bias voltage
(F=V/l). Obviously, a gain of G>1 can be obtained if
the device satisfies the condition of τlt >τtt. Generally, car-
rier lifetime of a material is adjusted by the existence of hole-
or electron-trapping states within bandgap. The trap states in
semiconducting materials can be divided into two categories:
trapping centers and recombination centers. The former are
the traps located in the position close to the band edges, so
that the trapped carriers have a great probability to CB or VB
under thermal excitation. The latter are the traps located in the
position close to the middle of the bandgap, so that the trapped
carriers have a great probability to recombine with the oppo-
site charge carriers.[39]Recombination centers can be classi-
fied into two kinds by their functions: one can shorten the car-
rier lifetime (type I centers), while the other one can prolong
it (type II or sensitizing centers). The lifetime of a charge car-
rier in a material with n-type trap states per unit volume with
capture cross-section Scan be evinced by[39]
τtrap =1
vSN exp−
∆E
kT ,(3)
where v=p(3kT /m∗
e)is the thermal velocity of the charge
carriers, ∆Eis the energy distance between the trap states and
020701-3

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
the relative band edge, kis the Boltzmann constant, and Tis
the temperature. For ligand-exchanged PbS CQD films, the
capture coefficient of electrons vS is ∼400 times that of holes,
thus the electrons are more likely to be captured.[40]There-
fore, the photoconductivity in PbS CQD films is predominated
by holes. For recombination centers, ∆EkT . Additionally,
by using the reported figures (v≈106cm·s−1,S≈10−16 cm2,
and N≈1014 cm−3) and Eq. (3), τtrap can be figured out as
∼0.1 ms.[41–43]Consequently, the mid-gap trap states origi-
nated from ligand-exchange act as sensitizing centers, and it
can prolong the charge carriers’ lifetime which depends on the
density of traps and the capture cross-section. Based on this
mechanism, a net positive or negative charge is left in the pho-
todetector under instantaneous condition, because the majority
carriers can be extracted by electrodes but the minority carri-
ers are captured in the trap states. Then, the majority carriers
will be injected from electrodes to maintain the neutrality of
charges, bringing about the flow of a secondary photocurrent
until the whole device has achieved charge balance or recom-
bination occurrence. Systematic and detailed discussion of the
implementation of this mechanism on the photoconductors has
been investigated.[44–46]
(e) (f)
(d)
(b)
(a) (c)
Fig. 2. Schematic images of different device architectures and configurations. (a) A typical photoconductor;[19](b) a bottom-gate phototransistor;[19]
(c) a Schottky photodiode;[35](d) a heterojunction photodiode;[36](e) a p–n-type photodiode;[18]and (f) a p–i–n-type photodiode. [37]
Phototransistors have three metal contacts, as shown in
Fig. 2(b). It provides a better control over the conductiv-
ity of the semiconducting material, on account of the ability
to change the gate voltage, the source-to-drain voltage, and
the intensity of illumination. For a p-type lead chalcogenide
CQD, considerable holes will be induced by the negative ap-
plied bias, thereby increasing the conductivity.[40]The gate
electrode acts as a switch which can adjust the position of
the Fermi level, leading to the increment or reduction of trap
states.
Photodiodes are divided into several categories by the
types of junction: Schottky junction (Fig. 2(c)), heterojunc-
tion (Fig. 2(d)), p–n junction (Fig. 2(e)), and p-i-n junction
(Fig. 2(f)). However, the operating principle for each case
is basically the same: separating and collecting the photo-
generated electron–hole pairs by building a built-in electric
field in junction. During the process of charge extraction, these
photo-generated charge carriers need to traverse fairly con-
siderable distances to reach the corresponding electrodes and
then to be collected. Therefore, the device thickness is a crit-
ical parameter for photodiodes and the lifetime of the charge
carriers must be longer than their transit time. The diffusion
length of charge carriers in lead chalcogenide CQD films is
relatively long and thus can improve the performance of the
device. The diffusion length of charge carriers is dominated by
not only their mobility but also the average spacing among the
recombination centers.[47]Photodiodes can be operated under
zero-bias condition (photovoltaic mode), but usually under re-
verse bias condition (photoconductive mode) which can sup-
ply greater bandwidth and wider linear dynamic range. Gain
mechanism is also applied to photodiodes by introducing trap-
rich material in photoactive region or building heterojunction
to capture one type of charge carriers.[48–51]
3.2. Figures of merit
A series of figures of merit are formulated to evaluate the
performance of various photodetectors.[52]Firstly, responsi-
bility R(λ)is employed to quantify the electric signal output
per optical signal input for photodetectors, and it is defined as
the ratio of photogenerated current to optical power of incident
illumination on the photodetectors at a specific wavelength λ,
thus it is measured in units of A/W. The expression of respon-
sibility is given by
R(λ) = EQE(λ)qλ
hc
1
√1+ω2τ2G,(4)
where qis the electron charge, his the Planck’s constant, cis
the speed of light, ωis the modulation frequency, τis the time
constant, EQE(λ) is the external quantum efficiency, and Gis
the photoconductive gain.
020701-4

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
Another critical parameter which affects the photode-
tectors’ sensitivity is noise. The total noise current in a
photodetector consists of four main sources: thermal noise,
generation–recombination G–Rnoise, shot noise, and low-
frequency flicker 1/fnoise.
For photodetectors, the current signal level produced by
incident optical power must be stronger than the noise signal
level for the sake of being detected. Considering both the
responsibility and the noise, noise equivalent power NEP(λ)
is introduced to estimate the sensitivity of photodetectors in
noise, and it is defined as the optical power at which the signal-
to-noise ratio SNR is unity. NEP(λ) is the minimum optical
power that the photodetector can discern per square root of
bandwidth. SNR and NEP(λ) are expressed as
SNR =R(λ)P.qI2
n,NEP(λ) = qI2
n.R(λ),(5)
where Inis the total noise current density, including all the
noise sources, and here qI2
npresents its root-mean-squared
value. R(λ)is the responsibility and Pis the incident optical
power.
The normalized detectivity or specific detectivity D∗(λ)
is the most critical figure of merit for each photodetector, al-
lowing comparison among photodetectors with various config-
urations and areas.[53]It signifies the SNR when the incident
optical power is 1 W, the noise bandwidth is normalized to
1 Hz, and the detector area is 1 cm2.D∗(λ)is expressed in the
unit of cm·Hz1/2·W−1(Jones) and it is obtained by using the
following equation:
D∗(λ) = √A.NEP(λ) = √A.qI2
nR(λ),(6)
where Ais the active area of the photodetector. As D∗(λ)is
directly proportional to R(λ), it is also an indirect function of
applied bias, temperature, modulation frequency, and wave-
length. The description of D∗(λ)for a photodetector should
be followed by the measurement condition.
Dynamic range DR(λ) of a photodetector indicates the
range over which the photocurrent increases with the incident
optical power. DR(λ) is the range of optical power over which
the detector can be utilized to detect the incident light. Ideally,
R(λ)should maintain constant with the enhancement of opti-
cal intensity. DR(λ) is usually reported in the unit of decibel
(dB) and expressed as
DR(λ) = 20 log P(λ)max
P(λ)min ,(7)
where P(λ)max is the maximum incident optical power when
the photocurrent saturation arises, and P(λ)min is the mini-
mum detectable optical power or NEP(λ). The linear dynamic
range implies that the photocurrent increases linearly with the
incident optical power.
4. Lead chalcogenide CQDs
In the past decade, photodetectors based on lead chalco-
genide CQDs have been widely studied, focusing on PbS,
PbSe, and PbSxSe1−xCQDs. As compared to early work
of PbS CQD photodiodes pioneered by Sargent group,[54,55]
there has been significant promotion in many aspects. The
progress of these IR photodetectors is summarized in Table 1.
Table 1. Progress in lead chalcogenide CQD-based photodetectors. PC: photoconductor; PT: phototransistor; PD: photodiode; NC: nanocrystal;
NP: nanoparticle; NR: nanorod; NW: nanowire; MPA: mercaptopropionic acid; EDT: ethanedithiol; OA: oleic acid; CTAB: cetyltrimethylammonium
bromide; TGL/DTG: thioglycerol/dithioglycerol; TBAI: tetrabutylammonium iodide; PCBM: [6,6]-phenyl-C61-butyric acid methyl ester; P3HT: Poly(3-
hexylthiophene); PVK: Poly(9-vinylcarbazole); MoS2: molybdenum disulphide; MoO3: molybdenum trioxide; ZnO: zinc oxide; TiO2: titanium
dioxide; NiO: nickelous oxide; TAPC: 1,1-bis[(di-4-tolylamino)phenyl]cyclohexane.
Photoactive Type Ligands Active Spectral Responsibility Detectivity Bandwidth Ref.
material area/mm2range/nm /[(A/W)/EQE/%] /Jones
PbS CQD PC Butylamine 0.015 800–1500 2700 A/W 1.8×1013 18 Hz [54]
PbS CQD PC As2S30.08 900–1550 200 A/W 1.2×1013 – [69]
PbS CQD PC OH−/S2−– 2100–2400 50 A/W 3.4×10840 Hz [70]
PbS CQD:Ag NC PC MPA 1.5 350–800 4 ×10−3A/W 7.1×1010 9.4 Hz [75]
PbS CQD:Ag NP PC EDT 0.015 400–1700 5 A/W 2.5×1011 200 Hz [73]
PbS CQD:PCBM PC EDT 0.59 800-1400 57% 4.4×107330 kHz [76]
PbS CQD:PCBM PC OA – 700–1400 0.32 A/W 2.5×1010 – [77]
PbS CQD:C60 crystals PC Carboxylate 25 400–1350 7 ×10−4A/W 3.2×1010 – [78]
PbS CQD:C60 NR PC OA 25 350–1100 0.125 A/W 2.3×109– [79]
PbS CQD:Au NC PC CTAB – 350–1000 0.0016 A/W 1.1×1010 61.2 Hz [86]
PbSe CQD:P3HT PT Cl−– Vis/NIR 500 A/W 5.02 ×1012 – [58]
PbSe CQD:PVK PT Cl−4 Vis/NIR 2.93 A/W 1.24 ×1012 – [87]
PbS0.4Se0.6CQD:P3HT PT EDT – Vis/NIR 5.6×10−2A/W 1.02 ×1010 – [88]
PbS CQD:graphene PT Pyridine 0.2 NIR 17A/W – – [89]
PbS CQD:graphene PT EDT – 600–1600 17A/W 7 ×1013 10 Hz [90]
PbS CQD:graphene PT TGL/DTG 6 ×10−5700–1250 19A/W – – [91]
020701-5

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
Table 1. (Continued).
Photoactive Type Ligands Active Spectral Responsibility Detectivity Bandwidth Ref.
material area/mm2range/nm /[(A/W)/EQE/%] /Jones
PbS CQD:MoS2PT EDT 1.5×10−5550–1150 6 ×105A/W 5 ×1011 – [92]
PbS CQD:MoO3PT TBAI 0.0075 400–1100 4 A/W 2 ×1010 – [96]
PbS CQD PD Benzenedithiol – 400–1800 17% 1 ×1012 3 MHz [35]
PbS CQD:P3HT NW PD OA 1.2×10−5365–940 100 A/W 2.1×1012 – [97]
PbS CQD:P3HT:PCBM PD OA 4 100–1850 0.5 A/W/51% 2.3×1092.5 kHz [98]
PbS CQD:P3HT:PCBM:ZnO CQD PD Butylamine 6.25 300–1100 1.24 A/W/1624% 2.2×1011 – [99]
PbS CQD:ZnO/TiO2PD EDT 9 300–1100 80% 1.45 ×1012 – [18]
PbS0.4Se0.6CQD:ZnO PD EDT 1.767 800–1600 25.8 A/W 1.3×1013 – [59]
PbS CQD:TiO2PD MPA – 400–1100 0.3 A/W 2.4×1013 1.2 MHz [100]
NiO:PbS CQD:ZnO PD Benzenedithiol 4.5 400–1300 0.2 A/W 1.1×1012 36 kHz [102]
ZnO:PbS CQD:TAPC PD Benzenedithiol 4 700–1600 18700% 7 ×1013 – [37]
4.1. Film deposition and ligand exchange
Lead chalcogenide CQDs-based photodetectors are fabri-
cated by spin-coating, dip-coating, or drop-casting techniques
on a rigid or flexible substrate, followed by completing a
metal electrode. The film conductivity is directly governed
by the capping ligands. The carrier transport in CQD films
origins from the thermally activated hopping mechanism.[56]
The inter-nanocrystal distance dominates the transport, and
it depends on the CQD monodispersion and the length of
capping ligands. CQDs are typically capped with long and
insulated ligands, such as oleic acid (carbon chain length
∼2.5 nm)[57–59]and oleylamine (carbon chain length ∼
2.1 nm).[60]These ligands are essential for CQDs to maintain
colloidal stability in organic solvent with low boiling point,
but significantly restrain the effective transmission of charge
carriers in CQD solids; also, they extend spacing between in-
terfacing dots and lead to low density of QDs.[61]The long, in-
sulated ligands should be exchanged by short, conductive lig-
ands to decrease spacing, and promote film density and elec-
tron transport between dots. Another function of capping lig-
ands is to passivate the surface states to reduce the trap state
density. It is generally known that there are substantial trap
states on the surface of bulk materials due to the presence of
dangling bonds. For CQDs, the large surface-to-volume ra-
tios aggravate this effect. Up to date, the most effective and
widely-used strategies comprise purely atomic ligands, such
as iodine[62–64]and bromine, [4]or a hybrid of atomic ligands
and organic ligands with short carbon chain, such as mercap-
topropionic acid (MPA)[6]or ethanedithiol (EDT). [65]
Ligand exchange can be divided into two categories:
solid-state ligand exchange[7,65,66]and solution-phase ligand
exchange.[32,66,67]Figures 3(a) and 3(b) illustrate these two
ligand exchange processes. For the solid-state ligand ex-
change, CQDs capped with long-chain ligands are spin-coated
onto the substrate, then a solution containing short-chain lig-
ands such as MPA or EDT is fabricated on the CQD film. In or-
der to wash off exchanged ligands and superfluous short-chain
ligands, the as-deposited film is rinsed by a protic solvent fol-
lowed by a short soaking time and the removal of the solu-
tion. Then, the device will be self-assembled by layer-by-layer
technique. To ensure the completeness of ligand exchange, the
thickness of each CQD layer needs to be controlled by adjust-
ing the concentration of CQD solution.
As for solution-phase ligand exchange, long-chain lig-
ands are exchanged by short-chain ligands in two orthogo-
nal solvents, then one-step method is employed to deposit
ligand-exchanged CQD film. Comparing to the former, the
solution-phase process is more facile. However, many factors
are needed to consider: the binding among solvent molecules,
CQD surface atoms and ligands, and the concentration of
CQDs and ligands. Therefore, each part of solvent, CQDs,
and ligands needs to be carefully selected and accurately con-
trolled.
(a)
(b)
(c)
Fig. 3. (a) Schematic diagram of solid-state ligand exchange and film deposition;[66](b) schematic diagram of solution-phase ligand exchange
and film deposition;[66]and (c) alterable surface dipoles on PbS CQDs via applying various ligands, causing controlled p- or n-type behavior.[68]
020701-6

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
Various types of ligands can considerably affect the elec-
trical properties of the CQD film, sequentially affecting the
whole performance of photodetectors.[54,69–71]Also, the se-
lection of ligand will directly determine whether the CQDs
perform p- or n-type behavior.[7,72]As shown in Fig. 3(c), PbS
CQDs capped with EDT show p-type, but they show n-type
behavior when they are capped with tetrabutylammonium io-
dide (TBAI). As described by Milliron,[68]the selection of lig-
and induces surface dipoles which shift the CB and VB energy
level of CQDs, causing them to show p- or n-type behavior,
which is also in accordance with Fig. 1(e).
4.2. Lead chalcogenide CQD photodetectors
4.2.1. Lead chalcogenide CQD photoconductors
High Gand low noise level are two preconditions for
high-sensitivity photoconductors. From Eq. (2), obviously
one can see that the majority-carrier lifetime and the majority-
carrier transit time play a key role on G. The majority-carrier
lifetime can be increased by introducing recombination cen-
ters (mid-gap trap states). Other scenarios irrelevant to recom-
bination centers or traps to enhance Gare possible; for in-
stance, one kind of the photogenerated carriers are transferred
to the high mobility acceptor material while the other kind of
carriers stay in the donor material. Sometimes, one kind of
charge carriers are trapped in the acceptor material, leaving
the other kind of charge carriers to transport, which plays the
same effect of arising G. Therefore, the devices can show high
Gunder high electric field if the electrodes replenish the ex-
tracted carriers promptly.
Dark current density Jdalso plays an important role on
device performance since it is a key factor to determine NEP.
It mainly depends on the applied electric field, the charge car-
rier mobility, the type of ligands, CQD dispersity and its size
distribution, the band alignment between the metal contacts,
and the CB of the photoactive material. Jdcan be expressed
by[73]
Jd=NhqµF
l,(8)
where Nhis the majority carrier (hole) concentration, qis the
electron charge, µis the carrier mobility, and Fis the elec-
tric field. Most lead chalcogenide photoconductors are fab-
ricated with gold contacts (work function ϕ=−5.1 eV); in
this case, Au can inject holes into the VB of PbS CQDs as
its diameter is larger than ∼6 nm, and their energy levels are
matched. Thus, Nhis considerably increased and the ability
of p-type transportation can be enhanced. For PbS CQDs with
smaller diameters, only a lower potential barrier for hole in-
jection exists.[32,74]
Ag NPs:PbS QD blend ratio
N/cm-3
Dark current density/mASm-2
102
101
100
10-1
10-2
10-3
Applied bias/V
(a) (b)
(c) (d)
Fig. 4. (a) Architecture of a PbS-CQD:Ag NP photoconductor;[73](b) Ag NPs act as an electron donor, which would be capable to apply
electrons to system and thus to modify the density of charge carriers;[73](c) a gradual change from p-type to intrinsic and n-type behavior with
increasing Ag NPs (w/w) concentration;[73](d) the dark current density for different PbS-CQDs:Ag NP blends.[73]
Efforts on suppressing dark current have been made for
lead chalcogenide CQD photoconductors. The dark current
density can be reduced by two orders of magnitude via blend-
ing appropriate proportion of silver nanoparticles (NPs) with
PbS CQDs.[73]As shown in Fig. 4, this reduction is due to
the introduction of Ag NPs (work function ϕ=−4.3 eV) as
they act as an electron donor and fill almost all shallow elec-
tron traps, thus restraining p-type conductivity. In a similar
study, electrons are transferred from CB of PbS CQDs to Ag
NCs which act as electron traps, prolonging the carrier life-
020701-7

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
time and enhancing the photocurrent up to three orders of
magnitude.[75]
Usually, the as-synthesized PbS CQDs are p-type. There-
fore, electron–hole pairs are generated in the PbS CQD film
under illumination and the majority of photocurrent is holes.
As shown in Fig. 5(a), there are two different pathways for
excited electrons: they are directly excited to mid-gap states
(Mxand M1s) or directly to CB states (1Shto 1Se). Also, the
trapped electrons in mid-gap states can be further excited to
CB states. The majority of photocurrent originates from the
transitions of Mxand M1s, which are more dominant than the
intrinsic band edge transitions (see Fig. 5(b)). The intrinsic
band edge transitions can be seen at longer wavelengths ac-
cording to the excitation energy of mid-gap states. For ligand-
exchanged lead chalcogenide CQD photodetectors, R(λ)is
dominated by the occupation of mid-gap states, and it de-
pends on the incident light intensity. Trap states dominate un-
der low illuminating intensity, and their density is higher than
the photo-generated charge carriers; in this way, holes transit
while electrons are trapped or partly fill mid-gap states, thus
leading to high G. On the contrary, the trap states will be grad-
ually filled as the incident light intensity increases, resulting to
the increment of electrons’ quasi-Fermi level, and the gain will
be lower. From Eq. (2), it is obvious to see that Gis propor-
tional to the applied bias, and an unlimited Gcan be obtained
for a high applied bias (provided that the bias is lower than
the device breakdown voltage). However, there is a critical
limit: the fast recombination of charge carriers is induced by
the growing density of charge carriers filled in the trap states;
in this way, the carrier lifetime will decrease and then Gwill
be reduced. Therefore, the maximum Gis related not only to
the incident light intensity, traps density, and position, but also
to the applied bias.
As described above, there is a conflict between Gand the
transient response, and there is a fundamental trade-off be-
tween Gand bandwidth. A feasible approach to this challenge
is to introduce a bimolecular interfacial recombination process
in devices.[76]This photoconductor was fabricated by employ-
ing a [6,6]-phenyl-C61-butyric acid methyl ester (PCBM) layer
on top of PbS CQD layer. When the device is illuminated,
photo-generated electrons transfer from PbS CQD to PCBM,
and then recombine with holes at PbS-CQD/PCBM interface.
As the hole mobility in PbS film is low (10−4–1 cm2·V−1·s−1),
this bi-layer configuration decreases the carrier lifetime of PbS
CQDs and avoids the trap-assisted recombination. A band-
width >300 kHz can be realized under the incident light in-
tensity of 57 mW/cm2.
Type-II heterojunction based organic-inorganic hybrid
photoconductors were reported via blending lead chalco-
genide CQDs with fullerene acceptors, such as PCBM,[77]C60
crystals,[78]and C60 nanorods (NRs). [79]A highlighted merit
of hybrid devices is that they can operate under the condition
of CQDs capped with long-chain ligand, such as OA. In this
case, the photo-generated charge carriers transport through the
doping acceptor materials, despite that the long-chain ligands
can suppress the transport of electrons. Both types of charge
carriers can be extracted by electrodes in these devices, de-
pending on the type of ligands and the charge percolation
pathways. Supposing the electron transportation occurs from
CQDs to PCBM, then the diameter of the synthesized CQDs
needs to be strictly controlled to make the CB level of CQDs
consistently higher than the lowest unoccupied molecular or-
bital (LUMO) of PCBM (3.7 eV–4.3 eV). [80]Therefore, the
application of PCBM for wavelength shorter than ∼1400 nm
is limited.[81]As a substitution of PCBM, C60 was used due
to its lower LUMO (∼4.5 eV), and it can accept electrons
from larger diameter CQDs. It has been confirmed that R(λ)
of devices employing C60 responses up to ∼2000 nm.[82]Re-
lated studies with various organic and inorganic charge accep-
tor materials have been presented.[83]Moreover, CQD-based
photoconductors incorporating with 1D nanomaterials were
fabricated,[79]which benefit from large donor/acceptor inter-
faces. Also, devices by making good use of surface plasmons
have been carried out.[84–86]
Photon energy/eV
Photon energy/eV
A/arb. units
Number of photoelectrons
per incident photon/%
slow
fast
Eg
(a)
(b)
Fig. 5. (a) PbS CQD energy level diagram depicting the transitions be-
tween the quantized energy levels and mid-gap band (MGB) trap states (red
arrows);[40]the red/blank circles represent electrons/holes, the wavy black
arrows represent the relaxation pathways, and the blue arrow marks the po-
sition of Fermi level; (b) EQE as a function of incident photon energy under
different gate biases;[40]the energy spacing between two highlighted features
M1s and Mxis 0.6 eV, corresponding to half of the energy spacing between
the 1S and Xfeatures in the absorption spectrum (inset).
020701-8

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
4.2.2. Lead chalcogenide CQD phototransistors
From Eq. (2), one can see that Gis directly propor-
tional to the mobility of charge carriers, thereby the hy-
brid phototransistors are generally fabricated by integrating
lead chalcogenide CQDs with high mobility materials, such
as polymers[57,58,87,88]or two-dimensional materials [89–92]or
both.[93]
Poly(3-hexylthiophene) (P3HT) is the most widely used
p-type polymer which possesses the highest hole mobility
(0.1 cm2·V−1·s−1) among the reported polymers.[94,95]Ac-
cordingly, it is reasonable to blend lead chalcogenide CQDs
with P3HT to promote hole transport. Wang et al.[58]reported
a hybrid PbSe-CQD:P3HT (2:1 wt%)-based phototransistor
with R(λ)of 500 A/W and D∗(λ)of 5.02 ×1012 Jones. Song
et al.[88]reported a hybrid PbS0.4Se0.6-CQD:P3HT (2:1 wt%)-
based phototransistor with R(λ)of 5.6×10−2A/W and D∗(λ)
of 1.02 ×1010 Jones. Under illumination, photo-generated
holes transfer from CQDs to P3HT while photo-generated
electrons are trapped in CQDs. The transfer of holes is rapid
under the applied electric field due to the high mobility of
holes in P3HT, which significantly decreases the transit time
of holes, leading to high G. Konstantatos et al.[90]have intro-
duced mechanically exfoliated single/bilayer graphene flakes
into PbS CQD-based phototransistor, realizing relatively high
R(λ)and D∗(λ)of 107A/W and 7 ×1017 Jones, respectively.
The effective detectable optical powers for this hybrid device
can fall to ∼10 fW, exhibiting a slow rise time of ∼10 ms
and a bi-exponential decay time of ∼100 ms and ∼2 s. The
decay time can be decreased by applying a gate voltage to re-
duce the potential barrier at PbS-CQD/graphene interface by
maintaining the electrons captured in the CQDs, and its R(λ)
can be enhanced by ligand exchange techniques: an ultra-
high R(λ)∼109A/W can be achieved by applying thioglyc-
erol (TGL)/dithioglycerol (DTG) ligands.[91]It is worthwhile
to mention that the trade-off between Gand bandwidth has
been conquered via establishing a photo-junction field-effect
transistor (photo-JFET).[96]Such a kind of device has been
fabricated by depositing a transparent conductive molybde-
num trioxide (MoO3) layer on top of a moderately n-type PbS
CQD layer to form a rectifying junction. The photo-JFET can
achieve Gof 10 at modulation frequency of 100 kHz, whereas
those photoconductors employing PbS CQDs generally ex-
hibit Gbelow 100 Hz. Gof lead chalcogenide CQD:two-
dimensional material hybrid phototransistors have exhibited
several orders of magnitude larger than that of CQD-only
based photoconductors. As compared with photoconductors,
phototransistors focus on decreasing the transit time of charge
carriers instead of prolonging their lifetime, although both can
enlarge G.
4.2.3. Lead chalcogenide CQD photodiodes
As mentioned above, photodiodes can be divided into
several categories according to the types of junction. The
Schottky-type photodiodes are fabricated by constructing a
rectifying junction at proper metal/semiconductor interface,
building an built-in electric field to extract the charge carri-
ers efficiently. Clifford et al.[35]reported a Schottky-type pho-
todiode with a configuration of indium-tin oxide (ITO)/PbS-
CQDs/aluminum (Al). The PbS-CQDs form ohmic con-
tacts with both ITO and Al. This device exhibited a band-
width of 3 dB at modulation frequency >1 MHz. More-
over, CQD:polymer nanocomposites have also been applied in
this type of photodiodes. A PbS-CQD:P3HT single nanowire
(NW) hybrid device has been fabricated with a competitive
D∗(λ)of 2.1×1012 Jones, using Schottky contacts and OA-
capped PbS CQDs as the active layer.[97]
The heterojunction photodiodes are fabricated by con-
structing a semiconductor/semiconductor junction to extract
the charge carriers efficiently; it is typically structured by
means of cooperating lead chalcogenide CQDs with n-type
semiconductor, and π-conjugated polymers are also feasible
in some cases. The selection of materials should satisfy the
band alignment of the devices, i.e., it can promote the extrac-
tion of both holes and electrons at their corresponding con-
tacts. An organic-inorganic hybrid photodiode was reported
by blending OA-capped PbS CQDs with P3HT and PCBM.[98]
The introduction of P3HT and PCBM prompts efficient trans-
port of holes and electrons from CQDs, respectively, exhibit-
ing an EQE of 51% and D∗(λ)of 2.3×109Jones. Dong et
al.[99]introduced zinc oxide (ZnO) nanoparticles into the PbS-
CQDs:P3HT:PCBM blend film, and the device showed an
EQE of 1624% and D∗(λ)of 2.2×1011 Jones. The high trap
density in metal oxides can availably trap the photo-generated
electrons, triggering holes injection via electrodes from the ex-
ternal circuit.
The p–n junction photodiodes are fabricated by integrat-
ing p-type lead chalcogenide CQDs with wide-bandgap n-
type metal oxides to form a p–n junction. The most com-
monly adopted metal oxides are ZnO[59]and titanium diox-
ide (TiO2).[100]ZnO/TiO2as an electron-transporting layer
was introduced by Pal et al.[18]The dark current of this de-
vice is two orders of magnitude lower than that of Schottky-
type photodiodes. Sulaman et al. [59]reported a photodi-
ode employing ZnO nanoparticles as an interlayer between
ITO and PbS0.4Se0.6CQD layer, and it displayed a compet-
itive D∗(λ)of 1.3×1013 Jones under 100 µW/cm2980 nm
laser illumination. Afterwards, Sulaman et al.[101]reported a
broadband photodiode by epitaxially blending CH3NH3PbBr3
CQDs with PbS0.4Se0.6CQDs as the active layer, and it
exhibited a maximum R(λ)and D∗(λ)of 21.48 A/W and
3.59 ×1013 Jones, 22.16 A/W and 3.70 ×1013 Jones at room
020701-9

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
temperature under 49.8 µW/cm2532-nm laser and 62 µW/cm2
980-nm laser illumination, respectively.
Also, the p–i–n-type photodiodes have been reported with
an ultra-high G(maximum EQE(λ)≈18700%) and a high
D∗(λ)of 7 ×1013 Jones in these years.[37,102]This type of
photodiode integrates both electron- and hole-blocking layer
in one device to inhibit the charge carrier injection via the elec-
trodes from the external circuit, as shown in Fig. 6. The Fermi
level of PbS CQDs will be reduced and close to the intrinsic
level after ligand treatment with 1,3-benzenedithiol. There-
fore, PbS CQD layer can be treated as an intrinsic semicon-
ductor layer and it is sandwiched between p-type layer (1,1-
bis[(di-4-tolylamino)phenyl]cyclohexane (TAPC)) and n-type
layer (ZnO). It is worth noting that EQE(λ) surpasses 100%,
and it results from not only the primary photocurrent but also
the secondary photocurrent. A photoconductive process oc-
curs due to the charge carrier injection from electrodes un-
der illumination and applied bias. Therefore, the transient re-
sponse of p–i–n-type photodiodes is slow, and it is usually in
ms (similar to that of photoconductors). The high D∗(λ)in
these devices is due to the generation of secondary photocur-
rent (instead of a low noise level). However, for Schottky-type
photodiodes and p–n-type photodiodes, a low noise level is
expected which allows the device to be operated under zero-
bias condition, although a slight reverse bias will obviously
elevate both the EQE(λ) and bandwidth. Further efforts can
be done via optimizing the thickness of active layer to make
it adequately depleted, and meanwhile, constructing potential
barriers with proper height to efficiently inhibit the charge car-
rier injection from electrodes.
105
104
103
102
101
100
10-1
Wavelength/nm Wavelength/nm
Power density/mWScm-2 Power density/mWScm-2
EQE/104 % EQE/%
Detectivity/1013 Jones Detectivity/Jones
(a) (b)
(d)
(c)
Fig. 6. (a) Schematic cross-section view of a p–i–n type photodiode;[37](b) schematic energy band diagrams of a p–i–n type photodiode in the
dark (left panel) and under IR illumination (right panel);[37](c) EQE(λ) values as function of wavelength under reverse bias (left panel), and
D∗(λ)values as function of wavelength under reverse bias (right panel);[37](d) light intensity dependence of EQE(λ) (left panel) and D∗(λ)
(right panel) at a constant bias of 10 V. [37]
4.2.4. Tandem photodetector
To further enhance the performance of photodiodes, tan-
dem structure was introduced to suppress the high dark cur-
rent. Compared with single-layer architecture, the unique ad-
vantage of tandem architecture is that the intermediate layer
functions as a valid energy barrier which suppresses the dark
current; meanwhile, photocurrent is harvested through effec-
tive charge carrier recombination at the intermediate layer.
The dominant electrical transport mechanism is altered from
020701-10

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
carriers hopping between the nearest CQD units to the charge
carrier recombination at the intermediate layer, and thus the
CQDs are no longer the dominant limiting factor for the per-
formance of photodiodes. In 2015, a PbSe CQDs-based tan-
dem photodetector glass/ITO/PEDOT:PSS/PbSe/ZnO/ poly-
TPD/PbSe/ZnO/Al was reported by Jiang et al.,[103]in which
ZnO as a hole-blocking layer (HBL) and poly-TPD as an
electon-blocking layer (EBL) were employed as the interme-
diate layers (see Fig. 7(a)) to serve as the recombination cen-
ters for electrons from one sub-detector and holes from the
other sub-detector. The measured dark current of tandem pho-
todetector showed a significant reduction of more than three
orders of magnitude as compared with that of the photode-
tector with single-layer architecture, as shown in Fig. 7(b).
And the tandem photodetector exhibited quite high D∗(λ)
of 4.7×1013 Jones and 8.1×1013 Jones under 34 µW/cm2
1100-nm laser illumination, respectively, at 275 K and 100 K.
Recently, our research group presented a high-performance
tandem photodetector via employing blends of PbS0.4Se0.6
CQDs and CsPbBr3CQDs,[101]and found the dark cur-
rent can be further reduced by introducing a poly(methyl
methacrylate) (PMMA) dielectric layer between the two sub-
detectors, i.e., ITO/PEDOT:PSS/CsPbBr3:PbS0.4Se0.6/ZnO/
PMMA/PVK/CsPbBr3:PbS0.4Se0.6/ZnO/Au, and it exhibited
a high R(λ)of 27.72 A W−1and an ultra-high D∗(λ)1.32 ×
1014 Jones under 57.8 µW/cm2980-nm laser illumination.
Also, other crucial parameters of the tandem photodetector
with PMMA interlayer are summarized in Table 2.
Fig. 7. (a) Device configuration and band diagrams of single-layer photodetector and tandem photodetector; (b) current density–voltage (J–V) characteristics
in dark (Jd) and under illumination (Jph) of single-layer photodetector and tandem photodetector.[103]
Table 2. Parameters obtained for tandem photodetector with PMMA interlayer.[101]
Laser wavelength/nm Illumination intensity Responsibility/(A/W) Detectivity/Jones
/µW/cm2
520 43.5 20.61 9.87 ×1013
650 49.8 18.87 9.04 ×1013
780 50 17.33 8.30 ×1013
980 57.8 27.72 1.32 ×1014
1050 55 18.95 9.07 ×1013
4.2.5. Lead chalcogenide CQD-based IR detecting and
display devices (IR-DDD)
Based on the development of IR detectors, IR imaging
devices have recently harvested quantities of attention owing
to their wide range of applications, such as night vision,[104]
national defense,[105]and biomedical research. [106]The con-
ventional IR imaging system is generally realized by intercon-
necting IR focal plane detector arrays with read-out circuits
via indium bumps.[107]However, there are a series of limits on
this technology, i.e., high cost, time-consuming, complicated
techniques, and restricted detectable IR wavelength range. To
solve these challenges, optical up-conversion technology has
been proposed and developed as an alternative method,[108]
and we have called it as IR-DDDs for such a kind of up-
conversion devices.[108]
An IR-DDD commonly consists of two categories: with
light-emitting diode (LED) and without LED. The former de-
vice integrates a detecting unit (photodetector) with a display
unit (LED or organic LED (OLED)). The free charge carriers
020701-11

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
are generated by the incident light in the active layer of pho-
todetector, subsequently these charge carriers inject into dis-
play unit and ultimately emit light in the light-emitting layer,
of which the emission wavelength is shorter than the incident
light.[110,111]The latter device possesses a nonlinear optical
process, in which the device absorbs two or more incident pho-
tons via the intermediate long-lived energy states, and emit
light with a shorter wavelength.[112]Compared to the latter,
the emission wavelength in visible region can be more easily
chosen in the former device, as the emission wavelength of the
former is independent of the incident wavelength.
Fig. 8. IR-DDD configuration consisting of a PbS CQD-based VFEPT
and a phosphorescent OLED, in which a porous ITO layer is used as the
source electrode.[109]
With the rapid development of organic electronics,
OLEDs have drawn widespread interest due to their unique
advantages: low cost, easy to realize large-area fabrication
and flexibility; in addition, they are no longer restricted by the
properties of substrates and lattice matching principle. There-
fore, OLEDs have been increasingly employed as the display
units in IR-DDDs.[113–115]
Lead chalcogenide CQD-based IR-DDD generally inter-
connects a lead chalcogenide CQD-based photodetector as an
IR detecting unit and a high-efficiency phosphorescent OLED
as a display unit. It is worth mentioning that the configu-
ration of photodetector in IR-DDD transforms from p–i–n-
type photodiode[115]to heterojunction field-effect phototran-
sistor (HFEPT)[116]to further enhance the performance of IR-
DDD, as it exists high internal Gand can realize the photocur-
rent amplification. In 2011, So et al. [117]firstly reported an
IR-DDD integrated with PbSe CQD sensitizing layer and a
phosphorescent OLED. The photon-to-photon conversion ef-
ficiency for this device (from 1300 nm to 520 nm) is about
1.3%. They adopted a broad-bandgap ZnO layer as the hole-
blocking layer, and it increased the threshold voltage in the
dark and decreased the dark current obviously, thus enhanc-
ing the SNR. However, its response range in IR region is only
from 1000 nm to 1500 nm and significantly limits its detection
range. In 2016, So et al. [109]reported their latest results on
PbS CQD-based vertical field-effect phototransistor (VFEPT)
(see Fig. 8), in which a perforated metallic source electrode
was employed. This VFEPT showed an EQE up to 1 ×105%
with a D∗(λ)of 1.23 ×1013 Jones under 2.4 µW/cm21042-
nm laser illumination. After incorporating a phosphorescent
OLED into the VFEPT, an IR-DDD is constructed and its av-
erage photon-to-photon conversion efficiency (from 1043 nm
to 550 nm) is 1597% at an IR power density of 0.53 µW/cm2.
In a word, IR-DDDs represent a promising subject when
it gets a high inner-gain, and might lay the foundation for new
emerging optoelectronic devices for infrared imaging.
5. Conclusion
A great success has been achieved by introducing lead
chalcogenide CQDs for IR photodetectors. During the past
decade, it is fairly promising for lead chalcogenide CQD-
based IR photodetectors on account of facile implementation
at room temperature, low cost, flexible substrate compatibility,
and high figures of merit. In order to further promote device
performance and reach commercialization, there are still sev-
eral challenges that need to be solved. For example, control-
ling the synthesis process of CQDs to enhance the stability
of CQDs, understanding how the CQD size affects the op-
toelectronic properties of the devices, CQD ligand exchange
techniques, surface charge transfer between CQDs, and com-
prehensive diagram of charge carrier transport along with the
manipulation of the trap states or charges in CQD films. With
the recent progress of advanced technologies, these problems
would be solved. Although currently there are some chal-
lenges, we still firmly believe that lead chalcogenide CQD-
based photodetectors possess further potential in the continu-
ous improvement for each figure of merits.
References
[1] Konstantatos G and Sargent E H 2010 Nat. Nanotechnol. 5391
[2] Ellingson R J, Beard M C, Johnson J C, Yu P, Micic O I, Nozik A J,
Shabaev A and Efros A L 2005 Nano Lett. 5865
[3] McDonald S A, Konstantatos G, Zhang S, Cyr P W, Klem E J, Levina
L and Sargent E H 2005 Nat. Mater. 4138
[4] Tang J, Kemp K W, Hoogl, S, Jeong K S, Liu H, Levina L, Furukawa
M, Wang X, Debnath R, Cha D, Chou K W, Fischer A, Amassian A,
Asbury J B and Sargent E H 2011 Nat. Mater. 10 765
[5] Sargent E H 2012 Nat. Photon. 6133
[6] Ip A H, Thon S M, Hoogl, S, Voznyy O, Zhitomirsky D, Debnath R,
Levina L, Rollny L R, Carey G H, Fischer A, Kemp K W, Kramer I J,
Ning Z, Labelle A J, Chou K W, Amassian A and Sargent E H 2012
Nat. Nanotechnol. 7577
020701-12

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
[7] Ning Z, Voznyy O, Pan J, Hoogl, S, Adinolfi V, Xu J, Li M, Kirmani A
R, Sun J P, Minor J, Kemp K W, Dong H, Rollny L, Labelle A, Carey
G, Sutherl, B, Hill I, Amassian A, Liu H, Tang J, Bakr O M and Sargent
EH2014 Nat. Mater. 13 822
[8] Chuang C H, Brown P R, Bulovic V and Bawendi M G 2014 Nat.
Mater. 13 796
[9] Mashford B S, Stevenson M, Popovic Z, Hamilton C, Zhou Z Q, Breen
C, Steckel J, Bulovic V, Bawendi M, Coe-Sullivan S and Kazlas P T
2013 Nat. Photon. 7407
[10] Shirasaki Y, Supran G J, Bawendi M G and Bulovic V 2013 Nat. Pho-
ton. 713
[11] Dai X, Zhang Z, Jin Y, Niu Y, Cao H, Liang X, Chen L, Wang J and
Peng X 2014 Nature 515 96
[12] Gong X W, Yang Z Y, Walters G, Comin R, Ning Z J, Beauregard E,
Adinolfi V, Voznyy O and Sargent E H 2016 Nat. Photon. 10 253
[13] Osedach T P, Zhao N, Andrew T L, Brown P R, Wanger D D, Strasfeld
D B, Chang L Y, Bawendi M G and Bulovic V 2012 ACS Nano 63121
[14] Liu Y, Tolentino J, Gibbs M, Ihly R, Perkins C L, Liu Y, Crawford N,
Hemminger J C and Law M 2013 Nano Lett. 13 1578
[15] Hetsch F, Zhao N, Kershaw S V and Rogach A L 2013 Mater. Today
16 312
[16] Adinolfi V and Sargent E H 2017 Nature 542 324
[17] Sukhovatkin V, Hinds S, Brzozowski L and Sargent E H 2009 Science
324 1542
[18] Pal B N, Robel I, Mohite A, Laocharoensuk R, Werder D J and Klimov
VI2012 Adv. Funct. Mater. 22 1741
[19] Saran R and Curry R J 2016 Nat. Photon. 10 81
[20] Efros A L and Rosen M 2000 Ann. Rev. Mater. Sci. 30 475
[21] Hines M A and Scholes G D 2003 Adv. Mater. 15 1844
[22] Yang Z, Wang M Q, Shi Y H, Song X H, Lin Z H, Ren Z Y and Bai J
T2012 J. Mater. Chem. 22 21009
[23] Park J, Joo J, Kwon S G, Jang Y and Hyeon T 2007 Angew. Chem. Int.
Ed. Engl. 46 4630
[24] Thanh N T, Maclean N and Mahiddine S 2014 Chem. Rev. 114 7610
[25] Kumar S and Nann T 2006 Small 2316
[26] Burda C, Chen X, Narayanan R and El-Sayed M A 2005 Chem. Rev.
105 1025
[27] Reiss H 1951 J. Chem. Phys. 19 482
[28] De Smet Y, Deriemaeker L and Finsy R 1997 Langmuir 13 6884
[29] Kwon S G, Piao Y, Park J, Angappane S, Jo Y, Hwang N M, Park J G
and Hyeon T 2007 J. Am. Chem. Soc. 129 12571
[30] Weidman M C, Beck M E, Hoffman R S, Prins F and Tisdale W A 2014
ACS Nano 86363
[31] Brown P R, Kim D, Lunt R R, Zhao N, Bawendi M G, Grossman J C
and Bulovic V 2014 ACS Nano 85863
[32] Talapin D V, Lee J S, Kovalenko M V and Shevchenko E V 2010 Chem.
Rev. 110 389
[33] Carey G H, Abdelhady A L, Ning Z, Thon S M, Bakr O M and Sargent
EH2015 Chem. Rev. 115 12732
[34] de Mello Donega C 2011 Chem. Soc. Rev. 40 1512
[35] Clifford J P, Konstantatos G, Johnston K W, Hoogl, S, Levina L and
Sargent E H 2009 Nat. Nanotechnol. 440
[36] Dong R, Bi C, Dong Q, Guo F, Yuan Y, Fang Y, Xiao Z and Huang J
2014 Adv. Opt. Mater. 2549
[37] Lee J W, Kim D Y and So F 2015 Adv. Funct. Mater. 25 1233
[38] Petritz R L 1956 Phys. Rev. 104 1508
[39] Bube R H 1975 J. Electron. Mater. 4991
[40] Nagpal P and Klimov V I 2011 Nat. Commun. 2486
[41] Konstantatos G and Sargent E H 2007 Appl. Phys. Lett. 91 173505
[42] Bozyigit D, Volk S, Yarema O and Wood V 2013 Nano Lett. 13 5284
[43] Bozyigit D, Jakob M, Yarema O and Wood V 2013 ACS Appl. Mater.
Inter. 52915
[44] Rose A 1963 Concept in Photoconductivity and Allied Problems (New
York: Interscience Publishers) Vol. 19
[45] Bube R H and Ho C T 1966 J. Appl. Phys. 37 4132
[46] Espevik S, Wu C H and Bube R H 1971 J. Appl. Phys. 42 3513
[47] Zhitomirsky D, Voznyy O, Levina L, Hoogl, S, Kemp K W, Ip A H,
Thon S M and Sargent E H 2014 Nat. Commun. 53803
[48] Chen H Y, Lo M K, Yang G, Monbouquette H G and Yang Y 2008 Nat.
Nanotechnol. 3543
[49] Guo F, Yang B, Yuan Y, Xiao Z, Dong Q, Bi Y and Huang J 2012 Nat.
Nanotechnol. 7798
[50] Kim D Y, Ryu J, Manders J, Lee J and So F 2014 ACS Appl. Mater.
Inter. 61370
[51] Wei H, Fang Y, Yuan Y, Shen L and Huang J 2015 Adv. Mater. 27 4975
[52] Dereniak E L and Boreman G D 1996 Infrared Detectors and Systems
(New York: Wiley)
[53] Jones R C 1953 Rev. Sci. Instrum. 24 1035
[54] Konstantatos G, Howard I, Fischer A, Hoogl, S, Clifford J, Klem E,
Levina L and Sargent E H 2006 Nature 442 180
[55] McDonald S A, Cyr P W, Levina L and Sargent E H 2004 Appl. Phys.
Lett. 85 2089
[56] Yu D, Wang C, Wehrenberg B L and Guyot-Sionnest P 2004 Phys. Rev.
Lett. 92 216802
[57] Yang S, Zhao N, Zhang L, Zhong H, Liu R and Zou B 2012 Nanotech-
nology 23 255203
[58] Wang H W, Li Z X, Fu C J, Yang D, Zhang L, Yang S Y and Zou B S
2015 IEEE Photon. Technol. Lett. 27 612
[59] Sulaman M, Yang S Y, Bukhtiar A, Fu C J, Song T J, Wang H W, Wang
Y S, Bo H, Tang Y and Zou B S 2016 RSC Adv. 687730
[60] Mourdikoudis S and Liz-Marz´
anLM2013 Chem. Mater. 25 1465
[61] Remacle F 2000 J. Phys. Chem. A 104 4739
[62] Wang H W, Li W L, Huang Y L, Wang Y S, Yang S Y and Zou B S
2017 Mater. Lett. 187 136
[63] Wang H W, Yang S Y, Wang Y S, Xu J F, Huang Y L, Li W L, He B,
Muhammad S, Jiang Y R, Tang Y and Zou B S 2017 Org. Electron. 42
309
[64] Xu J F, Wang H W, Wang Y S, Yang S Y, Ni G Q and Zou B S 2018
Org. Electron. 58 270
[65] Wang H, Wang Y, He B, Li W, Sulaman M, Xu J, Yang S, Tang Y and
Zou B 2016 ACS Appl. Mater. Inter. 818526
[66] Wang R L, Shang Y Q, Kanjanaboos P, Zhou W J, Ning Z J and Sargent
EH2016 Energ. Environ. Sci. 91130
[67] Kovalenko M V, Scheele M and Talapin D V 2009 Science 324 1417
[68] Milliron D J 2014 Nat. Mater. 13 772
[69] Yakunin S, Dirin D N, Protesescu L, Sytnyk M, Tollabimazraehno S,
Humer M, Hackl F, Fromherz T, Bodnarchuk M I, Kovalenko M V and
Heiss W 2014 ACS Nano 812883
[70] Hu C, Gassenq A, Justo Y, Devloo-Casier K, Chen H T, Detavernier C,
Hens Z and Roelkens G 2014 Appl. Phys. Lett. 105 171110
[71] Gao Y, Aerts M, Sandeep C S, Talgorn E, Savenije T J, Kinge S,
Siebbeles L D and Houtepen A J 2012 ACS Nano 69606
[72] Zhitomirsky D, Furukawa M, Tang J, Stadler P, Hoogl, S, Voznyy O,
Liu H and Sargent E H 2012 Adv. Mater. 24 6181
[73] Garcia de Arquer F P, Lasanta T, Bernechea M and Konstantatos G
2015 Small 11 2636
[74] Kim Y, Cho S, Jeong S, Ko D H, Ko H, You N, Chang M, Reichmanis
E, Park J Y, Park S Y, Lee J S, Yang H, In I and Park B 2014 Chem.
Mater. 26 6393
[75] He J G, Qiao K K, Gao L, Song H S, Hu L, Jiang S L, Zhong J and
Tang J 2014 ACS Photon. 1936
[76] Osedach T P, Zhao N, Geyer S M, Chang L Y, Wanger D D, Arango A
C, Bawendi M C and Bulovic V 2010 Adv. Mater. 22 5250
[77] Szendrei K, Cordella F, Kovalenko M V, Boberl M, Hesser G, Yarema
M, Jarzab D, Mikhnenko O V, Gocalinska A, Saba M, Quochi F, Mura
A, Bongiovanni G, Blom P W M, Heiss W G and Loi M A 2009 Adv.
Mater. 21 683
[78] Saran R, Nordin M N and Curry R J 2013 Adv. Funct. Mater. 23 4149
[79] Saran R, Stolojan V and Curry R J 2015 Sci. Rep. 45041
[80] Gocalinska A, Saba M, Quochi F, Marceddu M, Szendrei K, Gao J, Loi
M A, Yarema M, Seyrkammer R, Heiss W, Mura A and Bongiovanni
G2010 J. Phys. Chem. Lett. 11149
[81] El-Ballouli A O, Alarousu E, Bernardi M, Aly S M, Lagrow A P, Bakr
O M and Mohammed O F 2014 J. Am. Chem. Soc. 136 6952
[82] Klem E, Lewis J, Gregory C, Cunningham G, Temple D, D’Souza A,
Robinson E, Wijewarnasuriya P S and Dhar N SPIE Optical Engineer-
ing + Applications p. 6
[83] Noone K M, Subramaniyan S, Zhang Q F, Cao G Z, Jenekhe S A and
Ginger D S 2011 J. Phys. Chem. C 115 24403
[84] Pelayo Garc´
ıa de Arquer F, Beck F J, Bernechea M and Konstantatos
G2012 Appl. Phys. Lett. 100 043101
[85] Beck F J, de Arquer F P G, Bernechea M and Konstantatos G 2012
Appl. Phys. Lett. 101 041103
[86] He J G, Luo M, Hu L, Zhou Y L, Jiang S L, Song H S, Ye R, Chen J,
Gao L and Tang J 2014 J. Alloys Compd. 596 73
020701-13

Chin. Phys. B Vol. 28, No. 2 (2019) 020701
[87] Sulaman M, Yang S Y, Bukhtiar A, Fu C J, Song T J, Wang H W, Wang
Y S, Bo H, Tang Y and Zou B S 2016 RSC Adv. 644514
[88] Song T, Cheng H, Fu C, He B, Li W, Xu J, Tang Y, Yang S and Zou B
2016 Nanotechnology 27 165202
[89] Sun Z, Liu Z, Li J, Tai G A, Lau S P and Yan F 2012 Adv. Mater. 24
5878
[90] Konstantatos G, Badioli M, Gaudreau L, Osmond J, Bernechea M, de
Arquer F P G, Gatti F and Koppens F H L 2012 Nat. Nanotechnol. 7
363
[91] Turyanska L, Makarovsky O, Svatek S A, Beton P H, Mellor C J,
Patane A, Eaves L, Thomas N R, Fay M W, Marsden A J and Wilson
NR2015 Adv. Electron. Mater. 11500062
[92] Kufer D, Nikitskiy I, Lasanta T, Navickaite G, Koppens F H and Kon-
stantatos G 2015 Adv. Mater. 27 176
[93] Song X X, Zhang Y T, Zhang H T, Yu Y, Cao M X, Che Y L, Wang J
L, Yang J B, Ding X and Yao J Q 2016 Mater. Lett. 178 52
[94] Sirringhaus H, Tessler N and Friend R H 1998 Science 280 1741
[95] Dodabalapur A, Bao Z, Makhija A, Laquindanum J G, Raju V R, Feng
Y, Katz H E and Rogers J 1998 Appl. Phys. Lett. 73 142
[96] Adinolfi V, Kramer I J, Labelle A J, Sutherl, B R, Hoogl, S and Sargent
EH2015 ACS Nano 9356
[97] Yoo J, Jeong S, Kim S and Je J H 2015 Adv. Mater. 27 1712
[98] Rauch T, Boberl M, Tedde S F, Furst J, Kovalenko M V, Hesser G N,
Lemmer U, Heiss W and Hayden O 2009 Nat. Photon. 3332
[99] Dong R, Bi C, Dong Q F, Guo F W, Yuan Y B, Fang Y J, Xiao Z G and
Huang J S 2014 Adv. Opt. Mater. 2549
[100] Kim J Y, Adinolfi V, Sutherl, B R, Voznyy O, Kwon S J, Kim T W,
Kim J, Ihee H, Kemp K, Adachi M, Yuan M, Kramer I, Zhitomirsky D,
Hoogl, S and Sargent E H 2015 Nat. Commun. 67772
[101] Sulaman M, Yang S, Jiang Y, Tang Y and Zou B (unpublished)
[102] Manders J R, Lai T H, An Y B, Xu W K, Lee J, Kim D Y, Bosman G
and So F 2014 Adv. Funct. Mater. 24 7205
[103] Jiang Z, Hu W, Mo C, Liu Y, Zhang W, You G, Wang L, Atalla M R,
Zhang Y, Liu J, Kurhade K K and Xu J 2015 Nanoscale 716195
[104] Abbas M M, Kostiuk T and Ogilvie K W 1976 Appl. Opt. 15 961
[105] Tan M C, Al-Baroudi L and Riman R E 2011 ACS Appl. Mater. Inter. 3
3910
[106] Arnaoutakis G E and Richards B S 2018 Opt. Mater. 83 47
[107] Hsieh C C, Wu C Y, Jih F W and Sun T P 1997 IEEE T. Circ. Syst. Vid.
7594
[108] Challenor M, Gong P, Lorenser D, Fitzgerald M, Dunlop S, Sampson
DDandIyerKS2013 ACS Appl. Mater. Inter. 57875
[109] Yu H, Kim D, Lee J, Baek S, Lee J, Singh R and So F 2016 Nat. Pho-
ton. 10 129
[110] Chen J, Ban D, Helander M G, Lu Z H and Poole P 2010 Adv. Mater.
22 4900
[111] Chu X, Guan M, Niu L, Zeng Y, Li Y, Zhang Y, Zhu Z and Wang B
2014 ACS Appl. Mater. Inter. 619011
[112] Li X M, Zhang F and Zhao D Y 2013 Nano Today 8643
[113] Guan M, Li L S, Cao G H, Zhang Y, Wang B Q, Chu X B, Zhu Z P and
Zeng Y P 2011 Org. Electron. 12 2090
[114] Chen J, Tao J, Ban D, Helander M G, Wang Z, Qiu J and Lu Z 2012
Adv. Mater. 24 3138
[115] Chen J, Ban D Y, Feng X D, Lu Z H, Fathololoumi S, SpringThorpe A
JandLiuHC2008 J. Appl. Phys. 103 103112
[116] Chen J, Ban D Y, Helander M G, Lu Z H, Graf M, Poole P and Liu H
C2009 IEEE Photon. Technol. Lett. 21 1447
[117] Kim D Y, Choudhury K R, Lee J W, Song D W, Sarasqueta G and So
F2011 Nano Lett. 11 2109
020701-14