Identifying Carrier Behavior in Ultrathin Indirect-Bandgap CsPbX3 Nanocrystal Films for Use in UV/Visible-Blind High-Energy Detectors

Article abstract of DOI:10.1002/smll.202004513

High-energy radiation detectors such as X-ray detectors with low light photoresponse characteristics are used for several applications including, space, medical, and military devices. Here, an indirect bandgap inorganic perovskite-based X-ray detector is reported. The indirect bandgap nature of perovskite materials is revealed through optical characterizations, time-resolved photoluminescence (TRPL), and theoretical simulations, demonstrating that the differences in temperature-dependent carrier lifetime related to CsPbX₃ (X = Br, I) perovskite composition are due to the changes in the bandgap structure. TRPL, theoretical analyses, and X-ray radiation measurements reveal that the high response of the UV/visible-blind yellow-phase CsPbI₃ under high-energy X-ray exposure is attributed to the nature of the indirect bandgap structure of CsPbX₃. The yellow-phase CsPbI₃-based X-ray detector achieves a relatively high sensitivity of 83.6 μCGy_air^?1 cm^?2 (under 1.7 mGy_air s^?1 at an electron field of 0.17 V μm^?1 used for medical diagnostics) although the active layer is based solely on an ultrathin (≈6.6?μm) CsPbI₃ nanocrystal film, exceeding the values obtained for commercial X-ray detectors, and further confirming good material quality. This CsPbX₃ X-ray detector is sufficient for cost-effective device miniaturization based on a simple design.

Full text of DOI:10.1002/smll.202004513

Full PaPer
www.small-journal.com
Identifying Carrier Behavior in Ultrathin Indirect-Bandgap
CsPbX Nanocrystal Films for Use in UV/Visible-Blind
3
High-Energy Detectors
Bin Xin, Naresh Alaal, Somak Mitra, Ahmad Subahi, Yusin Pak, Dhaifallah Almalawi,
Norah Alwadai, Sergei Lopatin, and Iman S. Roqan*
and military devices. Among these applica-
High-energy radiation detectors such as X-ray detectors with low light
photoresponse characteristics are used for several applications including,
space, medical, and military devices. Here, an indirect bandgap inorganic
perovskite-based X-ray detector is reported. The indirect bandgap nature
of perovskite materials is revealed through optical characterizations,
time-resolved photoluminescence (TRPL), and theoretical simulations,
demonstrating that the differences in temperature-dependent carrier
tions, medical diagnostics is particularly
significant, but it presently requires direct
exposure to high doses of radiation, which is
harmful to human health, increasing cancer
[1,2]
risk, especially in children. Hence, X-ray
detectors possessing both high sensitivity
and high resolution with a low light inter-
[3,4]
face noise are required to minimize the
radiation exposure during routine medical
diagnostics. In the pertinent literature, use
of conventional semiconductors for direct
X-ray detection has been reported by sev-
lifetime related to CsPbX (X = Br, I) perovskite composition are due to the
3
changes in the bandgap structure. TRPL, theoretical analyses, and X-ray
radiation measurements reveal that the high response of the UV/visible-blind
yellow-phase CsPbI under high-energy X-ray exposure is attributed
[5]
3
eral groups, including amorphous Se,
[6]
[7]
[8]
[9,10]
to the nature of the indirect bandgap structure of CsPbX . The yellow-
crystalline Si,
and CdZnTe,
Ge, HgI2,
CdTe,
indicating that the most
effective materials are indirect bandgap
3
[9,10]
phase CsPbI -based X-ray detector achieves a relatively high sensitivity of
3
−
1
−2
−1
−1
8
3.6 μCGy
cm (under 1.7 mGyair
s
at an electron field of 0.17 V μm
air
[5–10]
semiconductors
that exhibit low light
used for medical diagnostics) although the active layer is based solely on an
interference noise. However, as such devices
possess low sensitivity due to the low atomic
ultrathin (≈6.6 μm) CsPbI nanocrystal film, exceeding the values obtained
3
for commercial X-ray detectors, and further confirming good material
[7]
number values of their materials, as well as
quality. This CsPbX X-ray detector is sufficient for cost-effective device
miniaturization based on a simple design.
they still require expensive fabrication and
complex processing methods, there is a high
industrial demand for cost-effective highly
sensitive X-ray detectors that are more suited
for mass production.
3
Ionic perovskite crystals can be processed from solution at
low temperatures, in contrast to covalent bond semiconductors,
which require a high-temperature crystallization process. Con-
sequently, lead halogen perovskite has emerged as a promising
candidate due to its cost-effectiveness, high crystal quality, high
absorption cross-section, high illumination, and ability to form
1
. Introduction
High-energy radiation detectors including X-ray detectors are uti-
lized in many fields, such as medical diagnostics, astronomy, secu-
rity inspection, and industrial monitoring. In particular, UV/visible
blind detectors are important for several applications, such as space
Dr. B. Xin, Dr. N. Alaal, Dr. S. Mitra, Dr. Y. Pak, D. Almalawi,
N. Alwadai, Prof. I. S. Roqan
Physical Sciences and Engineering Division
King Abdullah University of Science and Technology (KAUST)
Thuwal 23955-6900, Saudi Arabia
E-mail: iman.roqan@kaust.edu.sa
Prof. A. Subahi
College of Science and Health Professions
King Saud Bin Abdulaziz University for Health Sciences (KSAU-HS)
Jeddah 22384, Saudi Arabia
D. Almalawi
Physics Department
Faculty of Science
Taif University
P. O. Box 888, Taif 21974, Saudi Arabia
N. Alwadai
Department of Physics
College of Sciences
Princess Nourah bint Abdulrahman University (PNU)
Riyadh 11671, Saudi Arabia
Dr. S. Lopatin
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/smll.202004513.
Imaging and Characterization Core Laboratory
King Abdullah University of Science and Technology (KAUST)
Thuwal 23955-6900, Saudi Arabia
DOI: 10.1002/smll.202004513
Small 2020, 2004513
2004513 (1 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
quantum dots.[11] Moreover, these materials were conceived to
(NCs) by a modified room-temperature co-precipitation method
[
38]
act as an active layer for X-ray detectors due to the high absorp-
reported by Konstantatos and coworkers.
Even though the
tion arising from the high-Z atoms.[
12–19]
Owing to these ben-
nominal elemental ratio of the resulting NCs for CsPbBr I is
x 3−x
eficial characteristics, when perovskite materials are utilized
not very accurate, those four perovskite samples are denoted as
CsPbI , CsPbBrI , CsPbBr I, and CsPbBr , respectively, based
in devices based on direct X-ray detection mode,[
20–26]
sim-
3
2
2
3
pler system configuration compared to scintillators can be
achieved.[ However, commercial X-ray detectors rely mostly
on indirect bandgap materials to mitigate the light interfer-
ence, as medical diagnostics and other industrial applications
require high sensitivity, which can only be attained at a high
signal-to-noise ratio.[ Given that the high cross-sectional light
absorption of perovskites (such as CsPbBr and MAPbI ) and
on the molar ratio of the halogen element used during chem-
ical synthesis.
27]
As shown in the high-resolution transmission electron
microscopy (TEM) images presented in Figure S1a–d (Sup-
porting Information), the as-synthesized NCs with different
Br/I ratios exhibit markedly different shapes depending on the
weight of the halogen element. Generally, there are two types of
7]
3
3
their highly sensitive response to UV/visible light characteris-
CsPbX NC morphology—nanorod and nanosheet (Figure 1).
3
[
11,28–32]
tics
will increase the X-ray detector complexity as well
Figure 1a–c show the high resolution TEM (HR-TEM) image
and corresponding electron energy loss spectroscopy (EELS)
mapping as well as the NC profile, respectively, which confirm
as the production cost, there is a need for investigating X-ray
detectors based on indirect bandgap perovskite-related mate-
rials. In particular, CsPbI -based X-detectors remains insuffi-
the quality and the shape of the CsPbBr NCs. The average
3
3
ciently studied.
thickness of these NC was estimated to be ≈2 nm along the
z-direction, which is much smaller than the dimensions in
the xy-plane (tens to hundreds of nanometers), thus forming
Based on the X-ray detector fundamentals, semiconductor-
based X-ray detectors rely on two key features (low density
of mobile carriers in the absence of ionizing radiation such
as X-rays, and high density of generated carriers with a long
decay lifetime to ensure that the carriers can be collected
before recombination). Besides, the high performance of X-ray
detectors is indicative of high crystal quality.[ Therefore, the
2D NC structures. On the other hand, CsPbI NCs possess a
3
nanorod structure of 150–1000 nm length, as indicated in their
HR-TEM image, EELS map and the corresponding NC pro-
file (Figure 1d–f, respectively). Figure 1f shows that the CsPbI3
nanorod diameter is ≈15 nm. The HR-TEM images (shown in
the insets of Figure 1a,d confirm that both NC types exhibit
crystalline structure, suggesting high crystal quality.
7]
low quantum yield of the orthorhombic (δ) phase CsPbI is
3
not a measure of the material quality and X-ray detector per-
formance. Recently, X-ray detectors based on γ-phase CsPbI3
exhibiting very high response and sensitivity for X-ray radiation
have been reported, which was attributed to its high resistivity
and large (carrier mobility × lifetime) value.[ However, there
The dramatically different size of these two NC types is
determined by the colloidal dispersion in solution, which
ranges from weak (CsPbI ) to strong (CsPbBr ) as the Br/I
3
3
33]
ratio increases. The scanning electron microscope (SEM)
is a paucity of research on orthorhombic (δ) phase CsPbI , as
well as its optical properties and carrier mechanisms, even
though its properties can be good for X-ray detectors.
Furthermore, the correlation between the band structure of
images of films produced by drop-casting these CsPbX NCs
3
3
(Figure S1e–h in the Supporting Information) are consistent
with the corresponding TEM results (Figure S1a–d in the Sup-
porting Information). CsPbI3 films are formed by nanorod
stacking (Figure S1e in the Supporting Information), whereas
the CsPbBr3 film surface (Figure S1h, Supporting Infor-
mation) is much smoother due to the inherent nanosheet
[
34–37]
(
δ) phase CsPbI3 and its optical properties has never been
conducted.
In this work, we address this gap in extant research by per-
forming advanced optical, theoretical, and structural analyses.
Specifically, we used advanced optical time-resolved spectros-
copy analyses to explore the changes in the carrier lifetime
[
39]
structure, which is in line with our recent findings.
The
TEM and SEM images of the mixed halogen (CsPbBr I and
2
CsPbBrI ) perovskite reveal structures combining those two
2
stemming from the change in perovskite CsPbBr I composi-
morphology types.
x 3−x
tion. We found that such effects can be attained by engineering
To elucidate the chemical composition contribution to the
NC morphology, energy-dispersive X-ray spectroscopy (EDX)
measurements were carried out and the findings indicate that
the Cs/Pb/Br/I elemental ratios are about 1:1:0:2.3, 1:1:1:1.5,
1:1:2.3:0.65 and 0.6:1:3:0, for CsPbI , CsPbBr I, CsPbBrI , and
the band structure from direct bandgap perovskite CsPbBr to
3
indirect bandgap orthorhombic (δ) phase CsPbI . We also suc-
3
cessfully developed a UV/visible-blind X-ray detector based on
the indirect bandgap ultrathin CsPbI active layer with higher
3
3
2
2
device sensitivity compared to the commercial devices. As this
device demonstrates high response, it satisfies two key X-ray
detector prerequisites (negligible response to UV/visible light
in the absence of ionizing radiation).
CsPbBr , respectively, as shown in Figure S2 in the Supporting
3
Information. We attributed the reduced Cs ratio in the CsPbBr3
sample to the fact that the Cs ion surface is replaced by a sur-
[
40]
factant. The EDX mapping of the CsPbBr I sample (shown
2
in Figure 1g as an example) indicates that the color mapping
contrast (denoting elemental distribution) between Pb and
Br is much clearer than that pertaining to Cs, implying that
Cs drifts more easily than the bonded lead−halide framework
(PbBr42 tetrahedron).
2
. Results and Discussion
−
2
.1. Synthesis and Characterization of CsPbX Perovskite
3
Figure 1h shows the θ−2θ X-ray diffraction (XRD) scans
of the four perovskite samples. Using the ICSD-181287 and
ICSD-27979 XRD data, it can be determined that the CsPbBr3
For the present investigation, we prepared different inorganic
perovskites by synthesizing four CsPbX (X = I, Br) nanocrystals
3
Small 2020, 2004513
2004513 (2 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 1. a) HR-TEM image of nanosheet (CsPbBr NCs), with the HR-TEM of CsPbBr NCs shown in the inset. b) EELS mapping corresponding to the
3
3
TEM image in (a). c) The corresponding profile image of CsPbBr NCs demarcated by the dashed rectangle in (b), confirming the nanosheet feature
3
and the mean free path (l) estimates (using 300 kV) 60–70 nm x–y dimensions for CsPbBr NC. d) HR-TEM image of nanorods (CsPbI NCs), with the
3
3
HR-TEM of CsPbI NCs shown in the inset. e) EELS mapping corresponding to the TEM image in (a). f) The corresponding profile image of CsPbI
3
3
NCs demarcated by the dotted line shown in (e), indicating nanorod structure. g) EDX mapping for CsPbBr I, where the colored and black patterns
2
indicate the elemental distributions. h) XRD spectra for CsPbI , CsPbBrI , CsPbBr I, and CsPbBr , respectively.
3
2
2
3
structure has a cubic phase that belongs to the Pm-3m group,
stacking of NCs or superstructures corresponding to the
nanosheet morphology evident from the SEM and HR-TEM
images, as discussed in previous works. However, it should
whereas the CsPbI structure has orthorhombic δ-phase (non-
3
[
42]
perovskite yellow phase) that belongs to the Pnma group, in
line with the previously reported findings.[ The XRD meas-
urements suggest that both NC types have good crystal quality
without secondary phases, which is in line with the find-
ings yielded by the HR-TEM analysis. The peaks at 7.5° and
41]
be noted that both CsPbBrI2 and CsPbBr I comprise of a
2
mixture of cubic and yellow phases. Moreover, their lattice
constants can be increased by changing the I concentration,
o
as confirmed by the slight shift of the peaks located at 21.5
o
1
1.2° (noted by green arrows in Figure 1j) in the CsPbBrI2,
and 31.5 , which are indicated by the pink- and cyan-colored
CsPbBr I, and CsPbBr3 spectra are ascribed to the ordered
arrows in Figure 1h.
2
Small 2020, 2004513
2004513 (3 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
2
.2. Advanced Optical Analysis and the Band Structure
Consequently, as shown in the inset of Figure 2a, under UV
light illumination, the color brightness of the material declines
The four studied perovskite samples exhibit different optical
properties due to the chemical phase differences, as shown in
the normalized room temperature (RT) photoluminescence (PL)
and absorption spectra (Tauc plots) in Figure 2a,b. The PL peaks
rapidly, from significantly bright (CsPbBr ) to completely dark
3
(no emission from CsPbI ), as I concentration increases. On
3
the other hand, the inset of Figure 2a shows that the material
color becomes either yellow or orange under room light, corre-
sponding to the absorption spectra shown in Figure S3b in the
Supporting Information.
(excited by the λ = 400 nm line) located within the 500–525 nm
range exhibit a very slight shift as I and Br composition changes.
However, the PL integrated peak intensities of all samples shown
in Figure S3 in the Supporting Information indicate a signifi-
cant emission decline as the I composition increases, as the
To elucidate the differences in the CsPbBr optical properties
3
compared to those characterizing PCI samples incorporating I,
the band structure was revealed using density functional theory
[
50]
3
:17:100:2000 intensity ratio is obtained for the CsPbI , CsPbBrI ,
(DFT) calculations and the related carrier dynamics of the
studied materials were investigated carefully. Orthorhombic
δ-CsPbI (yellow phase), cubic CsPbBr , orthorhombic γ-CsPbI
3
2
CsPbBr I, and CsPbBr3 emission peaks. The CsPbI3 sample
2
produces very weak emission under the weak excitation of the
UV lamp source at RT, while the laser excitation induces a broad
peak emission from this sample at RT, as shown in Figure S3c
in the Supporting Information. A redshift from the absorption
edge is observed for the emission peaks of all materials, which is
3
3
3
(black phase), and cubic CsPbI were examined by applying
3
DFT. As can be seen from the band structure of yellow-phase
CsPbI shown in Figure 2c, it is an indirect bandgap material,
3
as its valence band maximum (VBM) is situated between the
highly symmetric Brillouin zone Γ and Y k-points, and its con-
duction band minimum (CBM) is located at a different point
indicative of the Stokes shift.[ The color change in CsPbX as
43]
3
a result of varying the halogen element composition ratio is not
observed, countering the previously reported findings.[
11]
(X k-point). However, cubic CsPbBr exhibits direct bandgap
3
It can also be observed that the absorption edge becomes
material properties, as both CBM and VBM are located at the
very broad as the I composition increases. CsPbBr NCs show
a much sharper absorption edge, indicating its direct bandgap
R k-point (Figure 2d). Unlike the yellow-phase CsPbI , γ-CsPbI
3
3
3
and cubic CsPbI3 (Figure S4a,b, Supporting Information)
behavior, whereas perovskite containing iodine (PCI) NCs
exhibit direct bandgap nature, similar to CsPbBr (shown in
3
deemed as indirect bandgap,[
44–46]
which is in line with the the
Figure 2c). For γ-CsPbI , both CBM and VBM occur at the Γ
3
broad absorption edge and emission peaks. The corresponding
point. These results align with the absorption measurements,
Tauc plot[
47–49]
as shown in Figure 2b reveal the samples
confirming that the yellow-phase CsPbI is an indirect bandgap
3
bandgap ranges from 2.15 eV (CsPbI ) to 2.46 eV (CsPbBr ).
material.
3
3
Figure 2. a) Normalized PL spectra and absorption of CsPbX samples. b) Tauc plots of CsPbX samples, the plots show the characteristics of indirect
3
3
bandgaps for CsPbI , CsPbBrI , and CsPbBr I, and a direct bandgap for CsPbBr . The band structure of c) yellow-phase CsPbI and d) cubic CsPbBr .
3
2
2
3
3
3
Small 2020, 2004513
2004513 (4 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
As typical indirect bandgap materials (such as CsPbI ) pro-
obtained for the CsPbI sample (0.26 nm). On the other hand,
3
3
duce very low emission at RT, whereas direct bandgap samples
produce much intense emission,[ we study the optical prop-
erties and carrier dynamics of both samples CsPbBr3 and
at low-temperature (5 K), the decay lifetime trend of the CsPbI3
47]
sample compared to that of CsPbBr reverses; the carrier life-
3
time of CsPbI at 5 K increased significantly and became much
3
CsPbI . Temperature-dependent PL measurements using laser
longer than that of CsPbBr as shown in the TRPL measure-
3
3
excitation were performed on CsPbI and CsPbBr , as shown
ments presented in Figure 3d. For CsPbI , the total lifetime
increased from 0.265 ns at RT to 0.748 × 10 ns at 5 K (an
increase of three orders of magnitude), as shown in Figure 3c,d,
respectively, whereas a minimal change in the total lifetime of
3
3
3
3
in Figure 3a,b, indicating that the CsPbBr peak remained rela-
3
tively intense and sharp as the temperature increased from 4 K
to RT, whereas the CsPbI peak was much broader and weaker.
3
In addition, as the temperature increased to RT, the emission
CsPbBr was observed as the temperature increased from 5 K
3
produced by CsPbI became weaker and exhibited a broader
(5.2 ns) to RT (9.7 ns). The TRPL parameters at 5K are reported
in Table S2.
3
spectrum. A similar PL signature has been observed for high-
quality indirect bandgap materials which indicated that the low
emission was due to the nature of the indirect bandgap and was
unrelated to the material quality.
To understand such carrier lifetime behavior in both mate-
rials, radiative and non-radiative components were ana-
[
51,54,55]
lyzed,
and the results are reported in Figure 3e,f. As
The carrier dynamics of CsPbI and CsPbBr samples were
shown in Figure 3d, the lifetime of CsPbBr at RT was domi-
3
3
3
investigated using time-resolved photoluminescence (TRPL)
measurements to further confirm our findings and better under-
stand the optical phenomena observed in Figure 3c,d (Table S1
in the Supporting Information shows the RT TRPL parameters).
The total decay lifetime was fitted with bi-exponential decay func-
nated by radiative recombination (as it is in the same ns range
of the total lifetime). By contrast, TRPL analysis of CsPbI NCs
3
(Figure 3f) revealed that the non-radiative recombination, τnr,
was 0.279 ns at RT, which is comparable to the total recombi-
nation value (0.265 ns), though the radiative recombination
3
tions (comprising both fast and slow decay components) due to
lifetime was very long (τ = 325 × 10 ns), suggesting more non-
r
the presence of multiple recombination centers,[
51,52]
as shown in
radiative recombination contribution that is usually observed
in the indirect bandgap material. Similar decay behavior has
been reported in pertinent literature for typical direct- and indi-
[
53]
Table S1. TRPL results indicated that the total lifetime was domi-
nated by the slow decay component, which is indicative of high-
quality materials. Based on the available literature, the fast decay
[
51,54]
rect bandgap materials.
Since radiative recombination is
component is dominant in materials characterized by low crystal
as it has been attributed to hot carrier−phonon inter-
actions and optical phonon decay.
independent of temperature, the total lifetime did not change
quality,[
53,54]
significantly for CsPbBr in contrast to that in CsPbI .
[42,53,56]
3
3
These findings revealed the reason behind the intense PL emis-
Notably, the RT decay lifetime of CsPbBr (τ = 9.7–10 ns)
sion and longer RT total lifetime decay of CsPbBr compared to
that of CsPbI3.
3
3
is almost two orders of magnitude greater than the values
Figure 3. Temperature-dependent PL spectra for a) CsPbBr and b) CsPbI excited by 400 nm laser line. c) Room temperature and d) low temperature
3
3
TRPL spectra of both CsPbBr and CsPbI NCs. The lifetimes at low temperature (5 K) and room temperature (300 K) for e) yellow-phase CsPbI and
3
3
3
f) cubic CsPbBr3.
Small 2020, 2004513
2004513 (5 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
At low temperatures, we observed that the carrier lifetime
The I−V curves of the PCI samples shown in Figure S5b,c in
the Supporting Information indicate that their photoresponse
produced under UV/visible excitation almost coincides with
that under dark conditions, which implies highly unfavorable
response to UV/visible illumination and is in line with the pre-
was considerably longer in indirect bandgap CsPbI materials
3
compared to that of direct bandgap CsPbBr materials. This
3
may be due to the contribution of phonon interactions in the
generation/recombination processes, allowing for momentum
conservation[ to transfer the carriers from the CBM at the X
k-point to the VBM at the k-point between Γ and Y, as shown
56]
[35]
viously reported findings. Such weak photoresponse is most
likely due to the nature of the indirect bandgap structure of the
PCI materials, rather than reflecting the material quality. We
attribute the poor photoresponse of the PCI samples and the
in the band structures of CsPbI in Figure 3a. This stands in
3
contrast to the recombination/generation processes in direct
bandgap CsPbBr3 that do not require phonon interaction.
Similar long lifetime behavior has been observed in several
indirect bandgap materials at low temperatures, though the
RT lifetime is very short as the phonon is highly dependent on
temperature.[ Therefore, the phonon interaction in the indi-
favorable photoresponse of CsPbBr to the fact that, unlike the
3
former, the latter is a direct bandgap material, characterized
by higher generated carrier density due to direct carrier transi-
tions from the VB to the CB. Conversely, in indirect bandgap
materials, the energy absorbed from the incident light is dis-
sipated through phonon interactions, leading to poor device
photoresponse. It is well known that indirect bandgap mate-
rials are not suitable for optoelectronic applications because
the generation and recombination processes of electron–hole
pairs require not only photon involvement, but also necessi-
tates momentum conservation via a phonon, which signifi-
cantly reduces the probability of photoelectric conversion.
53]
rect bandgap structure of CsPbI led to a shorter lifetime at RT
3
and a longer lifetime at low temperature compared to those of
direct bandgap materials.[
57,58]
2
.3. Detector Performance under Vis/UV/X-Ray Illumination
Four perovskite samples were prepared by drop-casting 100 ml
of colloid solution on interdigitated contact electrode (IDE)
(
Au/SiO /Si substrate). Au was used as an ohmic contact in the
2.4. Detector Performance under X-ray Exposure
2
horizontal metal−semiconductor−metal (MSM) structures, as
shown in Figure 4a. The energy level plotted in Figure S5a in
the Supporting Information is based on the 6.35 eV Pt work
The CsPbI X-ray absorption spectrum was compared to that
3
of traditional semiconductors over a broad range of photon
[
60]
function and the CsPbBr3 CBM and VBM energy levels of
energies using the photon cross-section database, as shown
59]
−
4.1 and −6.3 eV, respectively.[ The active layer was exposed
in Figure 5a. The remaining CsPbX structures exhibit similar
3
to the excitation source (visible light, UV light, or X-ray radia-
tion), which generated carriers, thereby enhancing the current
produced when a bias voltage is applied to the two electrodes.
Figure 4b shows the on/off current transient measurements
obtained when the photodetectors were exposed to visible light,
while the curve depicted in Figure 4c was produced by exposing
the devices to UV radiation at 1 V bias voltage. The visible light
was generated by a 53 mW cm− white LED (operating in the
absorption properties, as shown in Figure S6a in the Sup-
porting Information. The X-ray absorption coefficient is deter-
4
3
mined by α = Z /E , where Z is the atomic number and E is
the radiation energy. As Cs, Pb, I, and Br are heavy elements,
this results in efficient X-ray attenuation (Figure S6b in the
Supporting Information), as well as improved signal current,
as X-ray radiation is readily absorbed by the material in the
X-ray path. For example, the CsPbI3 absorption coefficient
across the entire examined energy range (excluding 3.7–5.2 kV
and 26.7–36 kV) is slightly higher than that of CdTe (shown
in the inset of Figure 5a). Its absorption coefficient is also
2
4
00–700 nm wavelength range), whereas a 244 nm, 158.5 mW
−2
cm UV laser served as the UV light source. CsPbBr produced
3
the photocurrent of 3 and 8.3 nA, under visible light and UV
light, respectively. CsPbI yielded a much lower photocurrent
superior to that reported for inorganic perovskite MAPbI and
3
3
(
0.2 nA), which is equivalent to ≈7% of the photocurrent
is much better than the performance of traditional Si-based
detectors.
obtained from CsPbBr under visible light exposure. Similarly,
3
0
.22 nA was obtained for CsPbI under UV light, corresponding
The performance of X-ray detectors based on the four studied
perovskites was investigated using a dark-environment clinical
3
to ≈2.5% of the value measured for CsPbBr3.
Figure 4. a) The detector structure, with the working energy level shown in the inset. b) The ON/OFF transient measurements of CsPbI and CsPbBr
3
3
−
2
−2
under 53 mW cm visible light illumination. c) The ON/OFF transient measurements of CsPbI and CsPbBr under 158.5 mW cm UV laser illumina-
3
3
tion at 244 nm.
Small 2020, 2004513
2004513 (6 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 5. a) Calculation of the mass absorption of Si, CdTe, MAPbI3, and CsPbI as a function of X-ray energy (the region marked by the dashed rectangle
3
is shown in the inset). b) The I–T curve of the CsPbBrI detector at different voltages when subjected to a 60 kV, 200 mA, 2 s X-ray pulse. c) Sensitivity of
2
devices based on CsPbI , CsPbBrI , CsPbBr I, and CsPbBr as a function of electric field. d) Sensitivity of devices based on CsPbI , CsPbBrI , CsPbBr I, and
3
2
2
3
3
2
2
CsPbBr as a function of X-ray dose. e) A schematic diagram of the X-ray carrier generation process. f) The I–T curve of the CsPbI detector as a function
3
3
of dose, which was varied by controlling the X-ray tube current, while the X-ray tube acceleration bias was maintained at 60 kV.
diagnostics tube, as shown in Figure S7 in the Supporting Infor-
in Table S3. Dose-dependence experiments were conducted
by varying the X-ray tube current. As shown in Figure 5d, the
sensitivity of all studied devices followed a very similar trend,
increasing with the dose rate reduction.
mation. The I–T curve obtained for CsPbBrI under different
2
bias voltages is shown in Figure 5b, as the remaining three
samples produced a similar response (Figure S8 in the Sup-
porting Information), confirming a high quality of PCI similar
−
1
−2
The highest sensitivity (up to 83.6 μCGy
cm ) was
air
to that of CsPbBr , which results in similar high-performance
obtained for the CsPbI (the indirect bandgap material) sample
3
3
−
1
−1
devices. The repeated on/off current cycles are characterized
by the same intensity and are capable of reversible and rapid
switching between the dark and under the X-ray exposure, indi-
cating stable device performance as shown in Figure S8 in the
Supporting Information. We found that the current increases as
voltage increases (linearly), suggesting good device stability. No
device degradation was observed when we repeated the on/off
measurements under different applied bias and several X-ray
exposure. Thus, we demonstrated the good performance with
high stability under different X-ray doses (Figure 5f). The on-
current exhibited much greater fluctuations than those induced
by visible light or UV light, which could be attributed to X-ray
tube ripple or random carrier generation in the perovskite
active layer.[ When the X-ray measurements were repeated
after a few weeks, similar I−V results were obtained to those
shown in Figure S5b,c in the Supporting Information. This
under 1.7 mGyair
s
at an electron field of 0.17 V μm (which
is within the range used in medical diagnostics, given that the
medical hospital system shown in Figure S7 in the Supporting
Information was used in the present investigation). It can
thus be concluded that the X-ray detector sensitivity increases
as the electric field increases and the dose rate decreases.
Device sensitivity can also be calculated via the linear fitting
[
18]
of the current density as a function of dose rate, as shown
in Figures S8 and S9 in the Supporting Information. For all
−
1
−2
devices, the average sensitivity of about 21 μCGy
cm was
air
derived from the slope of the fitted line. Moreover, the fitting
−
1
results indicate that −5.48 mGyair
s
is the lowest detectable
dose rate. This negative value could be attributed to the non-
linear relationship between the current density and the dose
rate, especially in the low dose rate region. We believe that the
lowest detectable dose rate calculated by this method is not of
practical significance, because most detectors are characterized
by a distinct signal-to-noise ratio.
16]
evidence confirms that inorganic CsPbX is sufficiently stable
3
under X-ray exposure.
By using the S = Q/AX sensitivity formula, where A is
the exposed area and X is the radiation exposure,[ it can be
shown that the sensitivity of the four studied devices can be
described by a linear function of the electric field strength, as
demonstrated in Figure 5c. Here, X-ray intensity was calcu-
lated using Rad Pro Calculator[ and the findings are reported
We expect that the X-ray absorption would be governed
by the core electrons in each of the constituent elements. As
the four samples contain both Cs and Pd, the X-ray response
is expected to arise primarily from Pd, as it is a much larger
atom, dominating the response of other elements (as con-
firmed by the elemental X-ray absorption analyses presented in
5]
61]
Small 2020, 2004513
2004513 (7 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure S6b in the Supporting Information). Figure S6a in the
Supporting Information represents the calculated X-ray absorp-
tion values of all four samples, which are equal, leading to the
conclusion that all devices exhibit comparable performance.
As shown in Figure 5e, when X-rays are incident on the active
atoms, photons eject electrons from the inner orbitals (i.e., the
L or K shell).[ These ejected electrons either become free elec-
trons or may be captured by other atoms. In either case, the
resulting holes in the lower orbitals will be filled by electrons
from an outer shell (M shell), each creating a hole that would
act as a carrier.
absorbing X-rays efficiently. Thus, the perovskite layer thickness
is a crucial determinant of device sensitivity. However, when
the perovskite layer is too thick, it will be challenging to obtain
high spatial resolution. Therefore, we fabricated devices based
on an ultrathin CsPbX active layer (≈3.6–9.4 µm), as shown in
3
Figure S11a-d in the Supporting Information. In all previous
reports, the active layer was up to 1000 times thicker than that
utilized in our study (a detailed comparison is given in Table S4
in the Supporting Information). In addition, the sensitivity of
our device is better than that currently achieved in commercial
devices in which the active layer thickness ranges from 100 µm to
62]
[15,18]
The data related to CsPbI , CsPbBrI , CsPbBr I, and CsPbBr
a few mm.
Even though it is expected that our device archi-
3
2
2
3
−2
−2
under visible light (53 mW cm ) and UV light (158.5 mW cm ),
tecture and ultrathin film thickness to result in lower sensitivity,
the response is still superior compared to that reported for other
−1
as well as under X-ray radiation (26.5 mGyair s ) confirm that
the device based on indirect bandgap perovskite has a high
X-ray response similar to that obtained from devices based on
direct bandgap perovskite (Figure S10 in the Supporting Infor-
mation), while also indicating the superior quality of the four
[15,18]
devices.
These characteristics can be beneficial for portable
cost-effective UV/visible blind X-ray detectors powered by solar
cells that require high sensitivity and simple design. Owing to the
high X-ray absorption of CsPbX , even a thin layer of this mate-
3
materials. However, the CsPbBr current signal produced by
rial could yield high sensitivity (Figure 5a). Further investigations
focusing on optimizing the active layer thickness and reducing
the dark current (i.e., by adding a hole-block layer) should, in
principle, significantly boost the performance of our devices.
3
vis/UV light is much higher than that obtained under X-ray
exposure. Although perovskite materials typically exhibit a
highly sensitive response to UV/visible light, such photore-
sponse will produce noise in X-ray detectors. On the other
hand, the photoresponse under light illumination is markedly
suppressed for the remaining three PCI samples. Hence, our
UV/visible blind devices based on PCI NC could operate under
a very high light exposure without any further modifications
to their structure or additional signal modulation. These find-
3
. Conclusions
We reported on a novel concept based on applying solution-
processed iodine CsPbX NCs for high-energy (mainly X-ray)
3
ings confirm that CsPbI is capable of eliminating solar light
interference to ensure that the detector would meet the criteria
imposed by its intended applications.
detection. The properties of these NCs have been studied by
HR-TEM and TRPL that revealed different carrier dynamics and
3
band structures of CsPbX when X changes. We showed that
3
Our experimental results also indicate that the detector
although the carrier lifetime of CsPbBr is much longer than
3
−1
was sensitive to a lower dose of 1.7 mGyair
s
with an on/off
that of CsPbI at RT, the lifetime trend reverses at 5K, which
3
ratio of 4, as shown in Figure 5e. This on/off ratio is similar
to that reported by Shrestha under a much higher dose
was attributed to their band structure nature. For the first time,
the fabrication of high sensitivity UV/visible blind X-ray detec-
[17]
−
1
−1
(
5.86 mGyair s ) and a higher electron field (0.2 V μm ), indi-
tors based on yellow phase PbCsI operating under ambient
3
cating superior characteristics of our devices. In addition, the
conditions was demonstrated. We showed that the X-ray detector
−
1
−2
−1
−2
highest sensitivity of the CsPbI device (83.6 μCGy
cm )
exhibits higher sensitivity (83.6 μCGy
cm ) compared to
3
air
air
is more than ten times higher than that reported for an amor-
commercial semiconductor-based X-ray devices, though a much
thinner active layer (≈6.6 μm) was adopted, confirming suffi-
cient evidence of the device’s excellent performance. Our find-
ings showed that the fabrication process of our X-ray device is
simple and cost-effective, it does not require the complex inte-
gration of several filter layers and can be used for producing
portable detectors, revealing the potential for adopting indirect
−1
−2 [6]
phous silicon detector (0.3 μCGy
cm ) as well as Cd(Zn)
air
Te single-crystal detectors and much higher than α-Se detec-
−
1
−2 [5]
tors (20 μCGy
cm ), while it is comparable to the results
air
[18]
reported by Wei and Pan (as shown in Table S4). Our PCI
devices exhibit a higher sensitivity compared to semiconductor
devices because the X-ray attenuation coefficient (α) or stop-
ping power of a material is very sensitive to its atomic number
bandgap iodine CsPbX materials for practical applications.
3
(Z). Therefore, although commercial single-crystalline semi-
conductor-based (e.g., Si, Se and Cd(Zn)Te) detectors possess
large (charge mobility × carrier lifetime) value, their sensitivity
remains low due to the small Z values.[
63]
4. Experimental Section
Chemicals: Lead(II) bromide (PbBr , 99.999% trace metals basis),
It is worth noting that this is the first report on an X-ray
2
detector based on ultrathin perovskite (about 6.6 µm thickness
lead(II) iodide (PbI , 99.999% trace metals basis), cesium acetate
2
for CsPbI device, as shown in Figure S11 in the Supporting
(CsAc, 99.99% trace metals basis), octylamine (OcAm, 99%), hexanoic
acid (HexAc, 99%), isopropanol (IPA, 99%), n-hexane (Hex, 99%) and
toluene (TOL, 99.8%) were purchased from Sigma-Aldrich. All chemicals
were used without any further purification.
3
Information). Recently, needle-like γ-phase CsPbI -based X-ray
3
detector was reported to exhibit high sensitivity, but this needle-
3
like CsPbI has a dimension of 5000 (length) × 250 × 250 µm
3
Synthesis and Purification of CsPbX3 Nanocrystals: CsPbX3 (CsPbI3,
CsPbBrI , CsPbBr I, and CsPbBr ) NCs were synthesized by a modified
[33]
(
two orders of magnitude thicker than our NC film).
2
2
3
Perovskite-based X-ray detector has a similar device structure
procedure reported by Konstantatos and coworkers[38] Typically, Cs
to photodetectors but requires a much thicker active layer for
precursor and PbX precursor were prepared separately, and the reaction
2
Small 2020, 2004513
2004513 (8 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
was initiated by injecting the latter into the former. First, Cs precursor
solution was prepared by dissolving 32 mg CsAc in 1 mL IPA in a 20 mL
vial under stirring in air at RT, followed by addition of 6 mL Hex and
UV laser, respectively. A filter was placed in the light path to obtain
visible or UV light.
X-Ray Tube Experiment: For practical purposes, this experiment was
conducted under a Definium 8000 system produced by GE Healthcare
with a Maxiray 100 X-ray tube, whereby 2 mm Al filtration (SIEMENS
AL01C II) acted as the collimator. In an X-ray tube, electrons are
accelerated in a vacuum by an electric field toward a metal target.
Because X-rays are emitted as the electrons decelerate in the metal, the
output signal consists of a continuous bremsstrahlung spectrum, with
additional sharp characteristic peaks associated with the atoms in the
target. For tungsten, an L-shell electron filling a K-shell vacancy results
in a characteristic X-ray energy: EK-shell − EL-shell = 69.5 keV − 10.2 keV =
59.3 keV. The shape of this continuous spectrum is approximately
described by Kramers’ law.
2
0
mL IPA. Next, PbX2 precursor solution was prepared by dissolving
.167 mmol PbX (307.5 mg PbI for CsPbI , 81.6 mg PbBr and 205 mg
2
2
3
2
PbI for CsPbBrI , 163.2 mg PbBr and 102.5 mg PbI for CsPbBr I, and
2
2
2
2
2
2
0
45 mg PbBr2 for CsPbBr ) into a mixture solution of 0.45 mL IPA,
.45 mL hexanoic acid, and 0.45 mL OcAm at 90 °C in air under vigorous
3
stirring. The hot PbX2 precursor was subsequently swiftly injected
into the Cs precursor under vigorous stirring at RT. The reaction was
completed in 120 s. High illumination was noted under UV exposure for
CsPbBr and CsPbBrI . The CsPbX NCs were isolated by centrifugation
3
2
3
at 3000 rpm for 180 s, and the pellet was dispersed into 3 mL of toluene.
As a result of centrifugation, CsPbBrI NCs lose the illumination ability,
2
as indicated by the phase change during dispensation.
Characterization: XRD measurements were conducted using a Bruker
D8 Advance X-ray diffractometer equipped with monochromatized
Cu Kα radiation (λ = 1.5418 Å). The samples were prepared by drop- Supporting Information
casting the nanosheet on a glass slide before drying the specimen
Supporting Information is available from the Wiley Online Library or
from the author.
at RT. HR-TEM was performed using a Titan CT (with 200 kV), while
Cary 6000i UV–Vis–NIR spectrophotometer was used for absorption
measurements. RT PL measurements were performed utilizing
Edinburgh FLS980 Photoluminescence Spectrometer attached to a 1000 W
Xenon lamp (Newport 66 923). ZEISS SEM (with 5 kV) was employed
for SEM, and EDX analysis was carried out using an X-ACT detector
from Oxford Instruments coupled with SEM (20 kV). The samples for
SEM measurements were prepared by drop-casting colloidal NCs on a Si
wafer and drying under ambient air conditions, with a 3 nm-thick iridium
layer serving as the conductive coating.
Acknowledgements
This work was financially supported by the King Abdullah University of
Science and Technology (KAUST) baseline funding (BAS/1/1319-01-01).
The authors thank the KAUST Medical Center (KMC) located at the
KAUST campus for providing their X-ray facilities for carrying out the
measurements.
Advanced Optical Characterizations: TRPL measurements were
performed using a mode-locked Ti:sapphire laser (Coherent Mira 900),
with a 1.90 W laser power at 800 nm. A second harmonic generator
(
APE) was used to excite the samples by 400 nm (150 fs pulse width and
6 MHz pulse repetition rate). Laser power on the sample was >1 mW Conflict of Interest
7
(60 µm spot diameter). Emission produced by the samples was detected
The authors declare no conflict of interest.
by the monochromator attached to a Hamamatsu C6860 streak camera
with a temporal resolution of 2 ps. The single sweep mode was used
for 5 K measurements, whereas the synchroscan mode was used for RT
measurements. For the single sweep mode, a pulse picker (PulseSelect
APE) was utilized to reduce the frequency from 76 to 2 MHz.
Keywords
ab-initio electronic structure calculations, indirect bandgap perovskites,
time-resolved spectroscopy (TRPL), UV–vis blind devices, X-ray detectors
Ab Initio Electronic Structure Calculations of Different Phase CsPbX3:
First principles-based DFT calculations were performed using VASP
software package as it is very good tool to obtain the electronic
properties and band structures[
64,65]
Received: July 26, 2020
Published online:
and other of the orthorhombic
δ-CsPbI3 (yellow phase), orthorhombic γ-CsPbI3 (black phase), cubic
CsPbI , and cubic CsPbBr . The electron–ion interaction was described
by projected augmented wave (PAW) method.
Ernzerhof functional
3
3
[66]
The Perdew−Burke−
was used as the exchange and correlation
[
67]
potential. A kinetic energy cut-off of 500 eV was chosen for the plane
wave basis sets. All atoms in the unit cell were relaxed until the force
on each atom declined below 0.01 eV Å , and 10 eV total energy
convergence was chosen for the electronic relaxation. A gamma-centered
k-point mesh of 6 × 6 × 6 and 8 × 8 × 6 dimensions was adopted for
cubic and orthorhombic crystal structures, respectively.
[1] D. J. Brenner, E. J. Hall, N. Engl. J. Med. 2007, 357, 2277.
[2] B. Sinnott, E. Ron, A. B. Schneider, Endocr. Rev. 2010, 31, 756.
[3] M. Yaffe, J. Rowlands, Phys. Med. Biol. 1997, 42, 1.
−
1
−5
[
4] A. Sakdinawat, D. Attwood, Nat. Photonics 2010, 4, 840.
[
[
5] S. O. Kasap, J. Phys. D: Appl. Phys. 2000, 33, 2853.
6] P. R. Granfors, R. Aufrichtig, Med. Phys. 2000, 27, 1324.
Detectors—Fabrication and Measurements: An 80 nm thick electrode
[
7] D. Pennicard, B. Pirard, O. Tolbanov, K. Iniewski, MRS Bull. 2017,
2, 445.
was deposited on SiO /Si substrates of 200 nm thickness by thermal
2
4
[68,69]
evaporation assisted with shadow masks.
The interdigitated
[
[
8] D. Wong, T. E. Schlesinger, R. B. J/ames, C. Ortale, L. van den Berg,
W. F. Schnepple, J. Appl. Phys. 1988, 64, 2049.
9] Y. Eisen, A. Shor, I. Mardor, IEEE Trans. Nucl. Sci. 2004, 51, 1191.
contact electrode (ICE) comprised of four close parallel branches
extending from two separate trunks. The channel length between the
branches and the branch width was 30 and 950 µm, respectively. The
CsPbX3 solution was drop-casted on the electrode. After the solution
had dried, the sample was placed on a hot plate for thermal annealing
in ambient air at 110 °C for 10 min. Keithley DC power supply was used
as the voltage source for I−V measurements, whereby the photocurrent
was recorded via LABVIEW program (current vs time frame) and ReRa
Solutions Tracer I−V curve software was employed for data acquisition.
For the experiments, white light/UV light and X-ray radiation sources
[
[
10] C. Szeles, Phys. Status Solidi B 2004, 241, 783.
11] L. Protesescu, S. Yakunin, M. I. Bodnarchuk, F. Krieg, R. Caputo,
[
C. H. Hendon, R. X. Yang, A. Walsh, M. V. Kovalenko, Nano Lett.
2
015, 15, 3692.
12] C. C. Stoumpos, C. D. Malliakas, J. A. Peters, Z. Liu, M. Sebastian,
J. Im, T. C. Chasapis, A. C. Wibowo, D. Y. Chung, A. J. Freeman,
B. W. Wessels, M. G. Kanatzidis, Cryst. Growth Des. 2013, 13,
2722.
−
2
−2
were supplied by a 53 mW cm visible light source and 158.5 mW cm
Small 2020, 2004513
2004513 (9 of 10)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
[
13] W. Wei, Y. Zhang, Q. Xu, H. Wei, Y. Fang, Q. Wang, Y. Deng, T. Li,
[42] J. Song, L. Xu, J. Li, J. Xue, Y. Dong, X. Li, H. Zeng, Adv. Mater. 2016,
28, 4861.
A. Gruverman, L. Cao, J. Huang, Nat. Photonics 2017, 11, 315.
[
14] J. A. Rowlands, Nature 2017, 550, 47.
[43] M. C. Brennan, J. E. Herr, T. S. Nguyen-Beck, J. Zinna, S. Draguta,
S. Rouvimov, J. Parkhill, M. Kuno, J. Am. Chem. Soc. 2017, 139,
12201.
[
15] Y. C. Kim, K. H. Kim, D. Y. Son, D. N. Jeong, J. Y. Seo, Y. S. Choi,
I. T. Han, S. Y. Lee, N. G. Park, Nature 2017, 550, 87.
[
16] W. Pan, H. Wu, J. Luo, Z. Deng, C. Ge, C. Chen, X. Jiang, W.-J. Yin,
G. Niu, L. Zhu, L. Yin, Y. Zhou, Q. Xie, X. Ke, M. Sui, J. Tang,
Nat. Photonics 2017, 11, 726.
17] S. Shrestha, R. Fischer, G. J. Matt, P. Feldner, T. Michel, A. Osvet,
I. Levchuk, B. Merle, S. Golkar, H. Chen, S. F. Tedde, O. Schmidt,
R. Hock, M. Rührig, M. Göken, W. Heiss, G. Anton, C. J. Brabec,
Nat. Photonics 2017, 11, 436.
[44] T. T. Tran, J. R. Panella, J. R. Chamorro, J. R. Morey, T. M. McQueen,
Mater. Horiz. 2017, 4, 688.
[45] V. Sarritzu, N. Sestu, D. Marongiu, X. Chang, Q. Wang, S. Masi,
S. Colella, A. Rizzo, A. Gocalinska, E. Pelucchi, M. L. Mercuri,
F. Quochi, M. Saba, A. Mura, G. Bongiovanni, Adv. Opt. Mater.
2018, 6, 1701254.
[46] A. Schmitz, L. L. Schaberg, S. Sirotinskaya, M. Pantaler,
D. C. Lupascu, N. Benson, G. Bacher, ACS Energy Lett. 2020, 5, 559.
[47] T. Yoshizawa, Handbook of Optical Metrology: Principles and Applica-
tions, CRC Press, Boca Raton, FL 2017.
[48] B. Vargas, E. Ramos, E. Pérez-Gutiérrez, J. C. Alonso, D. Solis-Ibarra,
J. Am. Chem. Soc. 2017, 139, 9116.
[
[
[
18] H. Wei, Y. Fang, P. Mulligan, W. Chuirazzi, H.-H. Fang, C. Wang,
B. R. Ecker, Y. Gao, M. A. Loi, L. Cao, J. Huang, Nat. Photonics 2016,
10, 333.
19] S. Yakunin, M. Sytnyk, D. Kriegner, S. Shrestha, M. Richter,
G. J. Matt, H. Azimi, C. J. Brabec, J. Stangl, M. V. Kovalenko,
W. Heiss, Nat. Photonics 2015, 9, 444.
[49] P. Makuła, M. Pacia, W. Macyk, J. Phys. Chem. Lett. 2018, 9, 6814.
[50] S. A. Aravindh, U. Schwingenschloegl, I. S. Roqan, J. Chem. Phys.
2015, 143, 224703.
[
20] M. D. Birowosuto, D. Cortecchia, W. Drozdowski, K. Brylew,
W. Lachmanski, A. Bruno, C. Soci, Sci. Rep. 2016, 6, 37254.
[
21] Q. Chen, J. Wu, X. Ou, B. Huang, J. Almutlaq, A. A. Zhumekenov,
X. Guan, S. Han, L. Liang, Z. Yi, J. Li, X. Xie, Y. Wang, Y. Li, D. Fan,
D. B. L. Teh, A. H. All, O. F. Mohammed, O. M. Bakr, T. Wu,
M. Bettinelli, H. Yang, W. Huang, X. Liu, Nature 2018, 561, 88.
22] P. Büchele, M. Richter, S. F. Tedde, G. J. Matt, G. N. Ankah,
R. Fischer, M. Biele, W. Metzger, S. Lilliu, O. Bikondoa, Nat.
Photonics 2015, 9, 843.
[51] I. A. Ajia, P. R. Edwards, Y. Pak, E. Belekov, M. A. Roldan, N. Wei,
Z. Liu, R. W. Martin, I. S. Roqan, ACS Photonics 2018, 5, 820.
[52] N. Alfaraj, S. Mitra, F. Wu, I. A. Ajia, B. Janjua, A. Prabaswara,
R. A. Aljefri, H. Sun, T. K. Ng, B. S. Ooi, I. S. Roqan, X. Li, Appl.
Phys. Lett. 2017, 110, 161110.
[53] J. Shah, Ultrafast Spectroscopy of Semiconductors and Semiconductor
Nanostructures, Vol. 115, Springer Science & Business Media, Berlin
1999.
[
[
23] F. Moretti, G. Patton, A. Belsky, M. Fasoli, A. Vedda, M. Trevisani,
M. Bettinelli, C. Dujardin, J. Phys. Chem. C 2014, 118, 9670.
24] V. Nagarkar, T. Gupta, S. Miller, Y. Klugerman, M. Squillante,
[54] I. A. Ajia, Y. Yamashita, K. Lorenz, M. M. Muhammed, L. Spasevski,
D. Almalawi, J. Xu, K. Iizuka, Y. Morishima, D. H. Anjum, N. Wei,
R. W. Martin, A. Kuramata, I. S. Roqan, Appl. Phys. Lett. 2018, 113,
082102.
[55] H. Sun, S. Mitra, R. Subedi, Y. Zhang, W. Guo, J. Ye, M. K. Shakfa,
T. K. Ng, B. S. Ooi, I. S. Roqan, Z. Zhang, J. Dai, C. Chen, S. Long,
Adv. Funct. Materials 2019, 29, 1905445.
[56] K. M. Groom, A. I. Tartakovskii, D. J. Mowbray, M. S. Skolnick,
P. M. Smowton, M. Hopkinson, G. Hill, Appl. Phys. Lett. 2002, 81, 1.
[57] J.-I. Shim, H. Kim, D.-P. Han, D.-S. Shin, K. S. Kim, Proc. SPIE 2014,
8986, 89861S, https://doi.org/10.1117/12.2037986.
[58] F. Wu, H. Sun, I. A. AJia, I. S. Roqan, D. Zhang, J. Dai, C. Chen,
Z. C. Feng, X. Li, J. Phys. D: Appl. Phys. 2017, 50, 245101.
[59] A. Subramanian, Z. Pan, Z. Zhang, I. Ahmad, J. Chen, M. Liu,
S. Cheng, Y. Xu, J. Wu, W. Lei, Q. Khan, Y. Zhang, ACS Appl. Mater.
Interfaces 2018, 10, 13236.
[
G. Entine, IEEE Trans. Nucl. Sci. 1998, 45, 492.
[
25] Y. Zhang, R. Sun, X. Ou, K. Fu, Q. Chen, Y. Ding, L.-J. Xu,
L. Liu, Y. Han, A. V. Malko, X. Liu, H. Yang, O. M. Bakr, H. Liu,
O. F. Mohammed, ACS Nano 2019, 13, 2520.
26] L. Basiricò, S. P. Senanayak, A. Ciavatti, M. Abdi-Jalebi, B. Fraboni,
H. Sirringhaus, Adv. Funct. Mater. 2019, 29, 1902346.
[
[
[
27] W. Heiss, C. Brabec, Nat. Photonics 2016, 10, 288.
28] Y. Zhang, M. I. Saidaminov, I. Dursun, H. Yang, B. Murali,
E. Alarousu, E. Yengel, B. A. Alshankiti, O. M. Bakr,
O. F. Mohammed, J. Phys. Chem. Lett. 2017, 8, 961.
29] S. Chen, G. Shi, Adv. Mater. 2017, 29, n/a.
30] Z. Tan, Y. Wu, H. Hong, J. Yin, J. Zhang, L. Lin, M. Wang, X. Sun,
[
[
L. Sun, Y. Huang, K. Liu, Z. Liu, H. Peng, J. Am. Chem. Soc. 2016,
138, 16612.
[
31] Y. Iso, T. Isobe, ECS J. Solid State Sci. Technol. 2018, 7, R3040.
[60] M. Berger, http://physicsnistgov/PhysRefData/Xcom/Text/XCOMhtml,
1998 (accessed: November 2010).
[
32] J. Liang, J. Liu, Z. Jin, Sol. RRL 2017, 1, 1700086.
[
33] B. Zhang, X. Liu, B. Xiao, A. B. Hafsia, K. Gao, Y. Xu, J. Zhou,
[61] Rad Pro Calculator, http://www.radprocalculator.com/Gamma.aspx
(accessed: 2011).
Y. Chen, J. Phys. Chem. Lett. 2020, 11, 432.
[
34] Q. Wang, X. Zheng, Y. Deng, J. Zhao, Z. Chen, J. Huang, Joule 2017,
, 371.
[62] Y. Su, W. Ma, Y. M. Yang, J. Semicond. 2020, 41, 051204.
[63] R. D. Evans, The Atomic Nucleus, Vol. 582, McGraw-Hill, New York
1955.
[64] Z. Zhang, U. Schwingenschlögl, I. S. Roqan, RSC Adv. 2014, 4,
50759.
1
[
35] M. Lai, Q. Kong, C. G. Bischak, Y. Yu, L. Dou, S. W. Eaton,
N. S. Ginsberg, P. Yang, Nano Res. 2017, 10, 1107.
36] Y. Hu, F. Bai, X. Liu, Q. Ji, X. Miao, T. Qiu, S. Zhang, ACS Energy
Lett. 2017, 2, 2219.
[
[65] Z. Zhang, U. Schwingenschlögl, I. S. Roqan, J. Appl. Phys. 2014, 116,
183905.
[
37] P. Luo, W. Xia, S. Zhou, L. Sun, J. Cheng, C. Xu, Y. Lu, J. Phys. Chem.
Lett. 2016, 7, 3603.
[66] a) G. Kresse, J. Furthmüller, Phys. Rev.
B 1996, 54, 11169;
[
38] F. Di Stasio, S. Christodoulou, N. Huo, G. Konstantatos, Chem.
Mater. 2017, 29, 7663.
39] B. Xin, Y. Pak, S. Mitra, D. Almalawi, N. Alwadai, Y. Zhang,
I. S. Roqan, ACS Appl. Mater. Interfaces 2019, 11, 5223.
b) G. Kresse, J. Hafner, Phys. Rev. B 1993, 47, 558.
[67] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
3865.
[68] Y. Pak, W. Park, S. Mitra, A. A. Sasikala Devi, K. Loganathan,
Y. Kumaresan, Y. Kim, B. Cho, G.-Y. Jung, M. M. Hussain,
I. S. Roqan, Small 2018, 14, 1703176.
[
[
40] V. K. Ravi, P. K. Santra, N. Joshi, J. Chugh, S. K. Singh, H. Rensmo,
P. Ghosh, A. Nag, J. Phys. Chem. Lett. 2017, 8, 4988.
[
41] R. J. Sutton, M. R. Filip, A. A. Haghighirad, N. Sakai, B. Wenger,
[69] S. Mitra, A. Aravindh, G. Das, Y. Pak, I. Ajia, K. Loganathan,
E. Di Fabrizio, I. S. Roqan, Nano Energy 2018, 48, 551.
F. Giustino, H. J. Snaith, ACS Energy Lett. 2018, 3, 1787.
Small 2020, 2004513
2004513 (10 of 10)
© 2020 Wiley-VCH GmbH