96.TiSe2 Advtsim
96.TiSe2 Advtsim
www.advtheorysimul.com
As a consequence, the graphene/Si heterojunction photodetector (5.13-8.11) × 10−6 K−1 .[28] This material is highly favored as a
exhibits diminished light-to-dark current distinction of ≈104 BSF layer in heterojunction solar cells due to its compatibility in
and a specific detectivity of 7.68 × 109 Jones.[16] To diminish the thermal expansion coefficient with the absorber layer, ensuring
reverse saturation current and accent absorption of photon a self- structural integrity during thermal fluctuations.[29] Additionally,
powered SnS0.25 Se0.75 /n-Si detector has been prepared that has the TiSe2 absorber layer’s PV performances can be improved by
achieved an outstanding responsivity of 377 mA·W−1 , a detectiv- the WSe2 BSF material.
ity of approximately ≈1011 Jones, and an I_light/I_dark ratio of In this work, numerical simulations using the 1D SCAPS-
about ≈4.5 × 102 .[17] Additionally, A near-infrared photodetector 1D software have been conducted to investigate and analyze
using a SnTe/Bi2 Se3 heterostructure shows ultrafast response an n-ZnSe/p-TiSe2 /p+ -WSe2 hetero-junction photodetector. This
to 1550 nm light with a high responsivity of 145.74 mA·W−1 structure is esteemed not only for its innovativeness but also for
and a maximum detectivity of 1.15 × 1010 Jones, highlighting its high sensitivity and responsiveness in the field, underscoring
its potential for low-power optical applications.[18] Besides, an its value in research environments. Numerical modeling is em-
𝛼-In2 Se3 /Ta2 NiSe5 heterojunction photodetector has been devel- ployed to scrutinize the influence of parameters such as breadth,
oped achieving a wide spectral response (405–1550 nm) and high defects, and the density of shallow acceptors or donors on the op-
performance with a photoresponsivity of 533 A W−1 at 520 nm erational efficacy of photodetectors. By using the photodetector’s
under a slight bias voltage (0.1 V) and detectivity exceeding experimental characteristics, the output photo-response, respon-
8.2 × 1013 Jones, highlighting its potential in the emerging field sivity (R), and detectivity (D* ) are calculated. Lastly, a comparison
of 2D optoelectronics.[19] However, the performance parameters is made between the current photodetector’s performance and
of those chalcogenide group PDs are not on the mark. the previously published one.
However, this article is presenting a highly detective and re-
sponsive PD based on TiSe2 material. TiSe2 is a TMD com-
pound that exhibits semi-metallic as well as semiconducting be- 2. Structure of the PD and Simulation Process
havior in the 2H and 1T phases, respectively, at different times.
Figure 1a shows the design of an n-ZnSe/p-TiSe2 /p+ -WSe2 dual
Furthermore, the charge density wave (CDW) transition tem-
hetero-structure PD, engineered for the near-infrared (NIR) spec-
perature of monolayer 1T-TiSe2 is closer to room temperature
trum, with molybdenum (Mo) and titanium (Ti) serving as the
compared to other monolayer TMD materials. Consequently,
anode and cathode, respectively. In this architecture, TiSe2 is uti-
monolayer 1T-TiSe2 offers an ideal platform for achieving the
lized as the absorber layer due to its narrow bandgap properties.
CDW phase at higher transition temperatures, even above room
Furthermore, n-ZnSe and p+ -WSe2 are incorporated as the win-
temperature.[21] In addition, TiSe2 is a van der Waals material dis-
dow layer and the BSF, respectively, within the photo-detector
tinguished by its layered crystalline structure. The weak bonding
structure. Figure 1b outlines the band structure progression of
in the van der Waals gaps allows for the exfoliation of films of
each layer. In this band schematic, the valence band maxima (EV )
different thicknesses from the bulk crystals.[22] However, there is
of the WSe2 and TiSe2 are 5.24 and 4.87 eV, in turn. The dispar-
a contradiction whether its electronic structure is semi-metallic
ity in the valence band maxima ∆EV between the two materials
or semiconducting. One study on 1T-TiSe2 reveals that it is pre-
is quantified at 0.37 eV. The WSe2 layer, characterized by a high
dominantly a semiconductor, with theoretical examination of the
carrier concentration exceeding 1017 cm−3 , induces a substantial
interactions involving electrons, holes, and phonons demonstrat-
built-in potential at the TiSe2 /WSe2 interface. This potential re-
ing semiconducting properties and part of the spectrum extend-
sults in an elevation of the WSe2 ’s EV in the upward direction,
ing beyond the Fermi level.[23] TiSe2 exhibits thermodynamic sta-
culminating in a slender energy spike at the interface as depicted
bility, good mobility, and thermal conductivity. It can be applied to
in the figure and therefore holes can easily pass form the absorber
various fields, including charge density waves, photonic devices,
to BSF and get picked up by the anode.[30]
and storage device electrode material.[21–23]
The SCAPS-1D (version 3.3.07), a sophisticated solar cell sim-
Better electron and hole mobilities and a wide range of light
ulation software, has been used in this study to numerically sim-
transmission are just two benefits of using ZnSe as a buffer
ulate the research. The program has been used extensively to
layer. Its energy gap of 2.7 eV, better electron and hole mobili-
model thin-film solar cells for the purpose of studying the physics
ties are considered as positive outcomes. Additionally, they ex-
involved, as well as their optical and electrical properties. The ex-
hibit a high electric field strength of breakdown, approximately
perimental and simulated data from SCAPS show good agree-
≈1 MVcm−1 ,[24] and demonstrate considerable resistance to in-
ment with reported previous publications. The SCAPS-1D soft-
tense UV and X-ray radiation. While thermal evaporation is a
ware has been used in a number of studies in recent years to ex-
more conventional method, ZnSe can also be grown by molecular
plore its possible uses in the creation of extremely efficient pho-
beam epitaxy, chemical vapor deposition with metal-organic pro-
tovoltaic devices.[31,32]
cedure, and deposition technique by electrochemical way.[25–27]
Poisson’s, electron continuity, and hole continuity equations
Thus, ZnSe can be a top-notch selection for a window layer.
are the three sets of equations that SCAPS-1D essentially solves
Additionally, the addition of BSF layer has increased the TiSe2
while maintaining boundary constraints.[28]
absorber material’s efficiency. WSe2 , a BSF layer, has a great
deal of potential in solar cells and electrochemical cells. It is ( ) 𝜕2 Ψ q[
earth-abundant, non-toxic material, and its carrier mobility is Poisson′ s equation + p (x) − n (x) + ND − NA
𝜕x2 𝜀
high. Additionally, it is discovered that the material exhibits a ]
notable thermal expansion coefficient, measured at a value of + 𝜌p − 𝜌n = 0 (1)
Figure 1. a) Schematic drawing and b) energy band scheme of n-ZnSe/p-TiSe2 /p+ -WSe2 heterostructure photodetector.
( ) 1 𝜕Jp saturation current is the place of origin of noise, below are the
Hole continuity equation = Gop − R (x) (2) equations for responsivity and detectivity.[34,35]
q 𝜕x
( ) 1 𝜕Jn Llight
Electron continuity equation = −Gop + R (x) (3) R= (6)
q 𝜕x Pin S
velocity of holes and electrons in WSe2 , TiSe2 & ZnSe semicon- Mobility of electron (cm2 Vs−1 ) 100 5 50
ductors at optimum temperature is fixed at 107 cm s−1 . Volume Mobility of hole (cm2 Vs−1 ) 500 22 20
defects employed in the simulation of the p+ -WSe2 /p-TiSe2 /n- Donor density, ND (cm−3 ) 0 0 1.00 × 1018
ZnSe hetero-junction device are set at 1015 cm−3 for all the lay- Acceptor density, NA (cm−3 ) 1.00 × 1019 1.00 × 1019 0
ers. In this setup, AM 1.5G light condition is consistently em- Defects at various interfaces:
ployed to refine the simulation studies and achieve the optimal Heterointerfaces Defect density [cm−2 ]
results. Photoresponsivity (R) and photo-detectivity (D* ) are cru- p+ -WSe2 / p-TiSe2 1.00 × 1011
cial parameters used to evaluate the performance and sensitivity
p-TiSe2 / n-ZnSe 1.00 × 1011
of a photosensor device. Assuming that noise collected from the
Figure 2. a) Current (J)-voltage (V) sketches and b) Responsiveness and Detection capability with the fluctuations of thickness of WSe2 ; c) Current
(J)-voltage (V) plots and d) Responsiveness and Detection capability with undulation of TiSe2 thickness; e) Current (J)-voltage (V) plots and f) Respon-
siveness and Detection capability with alteration of ZnSe layer thickness.
in short circuit photocurrent while changing the width of WSe2 in the JSC and VOC , which remain steady at 37.30 mA cm−2 and
layer. The short circuit photocurrent remains constant with a 0.795 V, respectively. Responsivity remains constant at a value of
value of 37.30 mA cm−2 as shown in Figure 2a. In Figure 2b, it 0.67 A W−1 , though the independent variable accelerated, as de-
is evident that the responsivity stays so as the width of the WSe2 picted in Figure 2f. There is no significant change in detectiv-
layer is raised from 50 to 350 nm with a responsivity value of 0.67 ity with the thickness of ZnSe layer, the detectivity almost con-
A W−1 . The detectivity of WSe2 layer also remains constant with stant value of 12.9 × 1014 Jones also as also shown in Figure 2f.
a detectivity value of 12.9 × 1014 Jones for the thickness rang- However, even with saturation in light absorption, increasing the
ing from 50 to 350 nm as shown in Figure 2b. The R and D* thickness further does not result in noticeable changes in per-
shows unchanged nature because of the constant photo-current. formance metrics. This is due to the long propagation path of
The width of the WSe2 BSF layer is selected at 200 nm as the photo-generated carriers, which increases the recombination rate
optimization and further calculations. within the ZnSe portion. The lifetime of minority carriers and
Figure 2c illustrates how the VOC and JSC change with the thick- the diffusion length are relatively greater than the thicknesses of
ness of the TiSe2 layer, ranging from 0.4 to 1 μm. Specifically, the WSe2 and ZnSe layers. Consequently, carrier recombination
the VOC measures 0.796 V. remains constant and that the JSC within the layer sheds little light on the total performance of the
is increased with the increasing thickness of TiSe2 layer from device.[40] For the optimization, the breadth of the window is se-
0.4 to 1 μm, and corresponding values of the JSC are 35 and lected to 100 nm for further calculations.
37.90 mA cm−2 , respectively. The short circuit current density
experiences an increase due to the enhanced photon absorption
from incident sunlight. This increment is attributed to the thicker 3.2. Effect of Doping of Device Layers
absorber layer that facilitates the generation of more Electron-
Hole Pairs (EHPs)[28] . The fluctuations of responsivity and de- The shallow acceptor density (NA ) of WSe2 BSF undoubtedly
tectivity concerning the same independent variable is considered keeps a drastic impact on the PD performance. Figure 3a demon-
in Figure 2d. Both responsivity and detectivity show an upward strates how the JSC and VOC vary as the doping concentration of
trend with the upliftment of TiSe2 thickness. The value of R and the WSe2 layer ranges from 1016 to 1021 cm−3 . It is observed that
D* enhances from 0.56 to 0.70 A W−1 and from 10.6 × 1014 Jones the JSC remains unchanged having a value of 37.30 mA cm−2 .
to 13.5 × 1014 Jones, respectively. The further calculation is done Figure 3b showcases the fluctuations in R and D* as the doping
by choosing the breadth of absorber of 700 nm. of WSe2 layer varies from 1016 to 1021 cm−3 . When the concentra-
In the light of Figure 2e, the thickness of the ZnSe layer ranges tion of WSe2 layer doping is increased to 1017 cm−3 , both the R
from 25 to 175 nm. Notably, there are no fluctuations observed and the D* increase as well. Following that, responsivity and de-
Figure 3. a) Current (J)-voltage (V) plots and b) R and D* fluctuations with changing acceptor level of WSe2 ; c) Current (J)-voltage (V) features and d)
Responsiveness and Detection capability fluctuation with carrier level of TiSe2 compound; e) Current (J)-voltage (V) plots and f) Responsiveness and
Detection capability switching with carriers of ZnSe.
tectivity remain unchanged at 0.67 A W−1 and 12.9 × 1014 Jones, mance. Within blemish concentration range of 1012 –1017 cm−3
respectively, even as the doping concentration increases further. of WSe2 layer, Figure 4a exhibits that the JSC remains constant at
As can be seen in Figure 3c, as the TiSe2 layer’s carrier den- 37.30 mA cm−3 . The VOC , which is 0.795 V, likewise stays con-
sity is between 1014 and 1019 cm−3 , the short-circuit photocurrent stant when the defects of WSe2 has the same range as shown
stays constant at 37.30 mA cm−2 and VOC of the device increases. in the figure. The R depicted in Figure 4b exhibits a steadfast
As can be seen in Figure 3d, responsivity slightly decreases be- constancy maintaining itself at 0.67 A W−1 , within the spectrum
tween doping concentration 1014 and 1019 cm−3 , with a respon- of WSe2 layer imperfection concentration spanning from 1012 to
sivity value of ≈0.67 A W−1 . The figure also illustrates how the 1017 cm−3 . Nevertheless, an intriguing shift occurs beyond this
detectivity increases from 8.61 × 1014 Jones to 12.9 × 1014 Jones specified threshold, where the R experiences a discernible de-
as the TiSe2 layer’s doping concentration rises from 1014 to 1019 scent, coinciding with an escalation in the defect density within
cm−3 . This phenomenon is occurred because the VOC enhances the WSe2 layer. From Figure 4b, it is seen that while the WSe2
as well as the dark current J0 reduces and the J0 is directly con- layer’s defect density varies between 1012 and 1017 cm−3 , there
nected to the D* .[40] is no discernible change in D and it has a value of 12.9 × 1014
In this investigation, the carrier volume of ZnSe portion is sys- Jones.
tematically changed, ranging from 1014 to 1018 cm−3 , maintain- In Figure 4c, defect density in TiSe2 layers ranges from 1012 to
ing the other features constant, as shown in Figure 3e. As seen in 1016 cm−3 and the JSC varies slightly from 37.5 to 37.28 mA cm−2 .
the figure, the short circuit photocurrent remains constant hav- The JSC then declines as the defect density rises, and so on. The
ing a value of 37.30 mA cm−2 while the VOC of the device slightly VOC , as also shown in the figure, stays constant at 0.795 V be-
falls. It is observed in Figure 3f that both the R and D* remain tween defect densities of 1012 and 1015 cm−3 . When the TiSe2
constant with the values of 0.67 A W−1 and 12.9 × 1014 Jones, layer’s defect density increases beyond this point, the VOC also
respectively even as the doping concentration increases. reduces. The impact of reducing the current and voltage is oc-
curred because of the increasing concentration of defects within
the absorber layer results in significant recombination of photo-
3.3. Effect of Total Defect Density of Device Layers carriers, leading to a substantial reduction in the performance of
the photovoltaic device.[40] Figure 4d illustrates a consistent re-
The total defect density in each layer also affects the device’s per- sponse in terms of R and D when the TiSe2 layer’s defect density
formance. An increment in defects causes more produced car- falls between 1012 and 1016 cm−3 . In this defect range, the R and
riers to recombine that significantly lessens the device’s perfor- D values stay nearly constant at 0.67 A W−1 and 12.9 × 1014 Jones,
Figure 4. a) Current (J)-voltage (V) plots and b) R and D* alteration for varying defects of WSe2 ; c) Current (J)-voltage (V) sketches and d) R and D*
alteration for varying defects of TiSe2 ; e) Current (J)-voltage (V) features, f) R and D* alteration for varying defects of ZnSe.
in turn. Beyond this point, R and D begin to decline as the TiSe2 3.4. Fluctuations in R and D* with the Presence and the Absence
layer’s defect density rises. of BSF Layer
Figure 4e elucidates the intricate interplay between the defects
for the ZnSe layer and the VOC and JSC . Notably, the graph reveals Herein, the wavelength-dependent spectral sensitivity of the PD
a steadfast response, with the photocurrent, JSC , and voltage, VOC has been quantified. The introduction of the BSF layer leads to
consistently holding its values at 37.30 mA cm−2 and 0.795 V, in an enhancement in spectral sensitivity, escalating with the wave-
turn exhibiting an unwavering stability even as the defect den- length until it attains its top value at 0.67 A W−1 at 920 nm. In the
sity spans the range from 1014 to 1018 cm−3 . Figure 4f unveils a absence of the WSe2 back surface, the spectral sensitivity exhibits
remarkable scenario wherein responsivity and detectivity exhibit a mounting trajectory with increasing light wavelength reaching
an unwavering resilience, maintaining steadfast values of 0.67 its pinnacle at 0.49 A W−1 at 850 nm. Following this peak, the re-
A W−1 and 12.9 × 1014 Jones, respectively. This constancy per- sponse diminishes as the wavelength further increases, becom-
sists seamlessly across a wide spectrum of defect densities span- ing negligible beyond 1030 nm, both in configurations with and
ning from 1014 to 1018 cm−3 . Defect state density can significantly without the BSF layer. This behavior is embellished in Figure 5a.
impact carrier recombination through various mechanisms. De- Figure 5b embellishes that the wavelength of higher detectiv-
fect trapping states can influence carrier recombination dynam- ity increases to 920 nm upon addition of the BSF layer, there
ics by exhibiting a super-linear increase in photoluminescence is an observed upward trend. The maximum detectivity value of
(PL) intensity at low excitation levels because of the sluggish rate 12.90 × 1014 Jones is reached in this instance. The detectivity, on
at which defect states depopulate.[41] This correlation depends on the other hand, increases with wavelength up to 850 nm and apex
both the density of defects and the rate at which carriers relax at 8.54 × 1014 Jones when the BSF layer is missed. Regardless of
in these defects.[41] Additionally, the state filling effect reduces the BSF layer’s presence or absence, it reaches this maximum,
the energy barrier required for recombination as carrier density begins to decrease as the wavelength lengthens, and ultimately
increases, resulting in significant recombination associated with becomes insignificant after 1030 nm.
defects, which is superlinearly dependent on carrier density and
can result in efficiency droop in optoelectronic devices.[42] Trap-
mediated recombination occurs when trapped electrons combine 3.5. Wavelength Dependence of EQE and Responsivity
with unbound holes within microseconds, unexpectedly easing
the restriction on achieving VOC under these circumstances.[43] Figure 6 shows the external quantum efficiency (EQE) and re-
That is why the defects for all the three layers are chosen at sposivity of the TiSe2 photodetector. The EQE values of the pho-
1015 cm−3 for further calculations. todetector decreases with the increase of the wavelength of the
Figure 5. Fluctuations of a) spectral response and b) detection level with the presence and the absence of the BSF with respect to photon wavelength.
layer. The doping concentrations are 1019 cm−3 for the absorber, [13] L. Q. Yao, L. M. Lin, Z. P. Huang, Y. Mao, H. Li, W. W. Lin, S. Y. Chen,
1019 cm−3 for the BSF, and 1018 cm−3 for the window layer. The Z. G. Huang, J. M. Li, G. L. Chen, Nano Energy 2023, 106, 108064.
device photocurrent is 37.30 mA cm−2 as well as the voltage is [14] Y. Mao, L. Huang, W. G. Zeng, F. Y. Wu, L. Q. Yao, L. M. Lin, J. M.
≈0.795 V. The simulated photodetector has a responsiveness of Zhang, J. M. Li, G. L. Chen, Chem. Eng. J. 2022, 446, 137400.
[15] P. Tang, Z. H. Huang, Y. X. Chen, H. Li, L. Q. Yao, H. Li, L. M. Lin, J. R.
0.67 A W−1 and a detection capability of 12.90 × 1014 Jones at
Cai, Y. L. Zhan, D. Wei, S. Y. Chen, Chem. Eng. J. 2023, 477, 146722.
920 nm of light wavelength. Also, a better photo-response be-
[16] X. Li, M. Zhu, M. Du, Z. Lv, L. Zhang, Y. Li, Y. Yang, T. Yang, X. Li, K.
tween 760 and 1010 nm is displayed by the PD with optimal Wang, H. Zhu, Small 2016, 12, 595.
parameters that proves this device an NIR detector. Finally, this [17] W. Gao, Z. Zheng, L. Huang, J. Yao, Y. Zhao, Y. Xiao, J. Li, ACS Appl.
structure has proven to be a leading contender in research en- Mater. Interfaces 2019, 11, 40222.
vironments, particularly when compared with traditional photo- [18] H. Zhang, Z. Song, D. Li, Y. Xu, J. Li, C. Bai, B. Man, Appl. Surf. Sci.
detectors. Its high sensitivity and responsiveness suggest promis- 2020, 509, 145290.
ing future prospects for advancing photo-detection technology. [19] H. Li, K. Zhang, X. Li, B. Liu, L. Li, Z. Mei, T. Chen, Q. Liu, W. Yu, J.
Yuan, H. Mu, Mater. Des. 2023, 227, 111799.
[20] W. Lin, W.-T. Guo, L. Yao, J. Li, L. Lin, J.-M. Zhang, S. Chen, G. Chen,
ACS Appl. Mater. Interfaces 2021, 13, 45726.
Acknowledgements [21] M. J. Wei, W. J. Lu, R. C. Xiao, H. Y. Lv, P. Tong, W. H. Song, Y. P. Sun,
The authors are highly indebted to Prof. Dr. Marc Burgelman, University Phys. Rev. B 2017, 96, 165404.
of Gent, Belgium, for supplying SCAPS 1D simulator. [22] P. Goli, J. Khan, D. Wickramaratne, R. K. Lake, A. A. Balandin, Nano
Lett. 2012, 12, 5941.
[23] G. Monney, C. Monney, B. Hildebrand, P. Aebi, H. Beck, Phys. Rev.
Lett. 2015, 114, 086402.
Conflict of Interest [24] J. Hossain, J. Phys. Commun. 2021, 5, 085008.
[25] G. S. Murugan, J. S. Wilkinson, N. P. Sessions, V. Mittal, Opt. Mater.
The authors declare no conflict of interest.
Express 2017, 7, 712.
[26] C. T. Hsu, Y. K. Su, T. S. Wu, M. Yokoyama, M. Takahashi, T. Nakada,
Y. Hashimoto, Appl. Surf. Sci. 1993, 65–66, 831.
Data Availability Statement [27] A. P. Samantilleke, M. H. Boyle, J. Young, I. M. Dharmadasa, J. Mater.
Sci. Mater. Electron. 1998, 9, 231.
The data that support the findings of this study are available from the cor- [28] M. I. R. Ebon, M. H. Ali, M. D. Haque, A. Z. M. Touhidul Islam, Eng.
responding author upon reasonable request.
Res. Express 2023, 5, 045072.
[29] J. Hossain, B. K. Mondal, S. K. Mostaque, Semicond. Sci. Technol.
2022, 37, 015008.
Keywords [30] J. Hossain, M. Rahman, M. M. A. Moon, B. K. Mondal, M. F. Rahman,
M. H. K. Rubel, Eng. Res. Express 2020, 2, 045019.
detectivity, photodetector, responsivity, SCAPS 1D, TiSe2 , WSe2 [31] M. Burgelman, P. Nollet, S. Degrave, Thin Solid Films 2000, 361, 527.
[32] K. Decock, P. Zabierowski, M. Burgelman, J. Appl. Phys. 2012, 111,
Received: April 24, 2024 43703.
Revised: June 24, 2024 [33] A. T. Abir, B. K. Mondal, J. Hossain;, Phys. Scr. 2023, 98, 125940.
Published online: [34] W. Choi, M. Y. Cho, A. Konar, J. H. Lee, G. B. Cha, S. C. Hong, S. Kim,
Y. Kim, D. Jena, J. Joo, S. Kim, Adv. Mater. 2012, 24, 5832.
[35] M. Binda, A. Iacchetti, D. Natali, L. Beverina, M. Sassi, M. Sampietro,
Appl. Phys. Lett. 2011, 98, 073303.
[1] J. Lu, Q. Ye, C. Ma, Z. Zheng, J. Yao, G. Yang, ACS Nano 2022, 16, [36] H. W. Myron, A. J. Freeman, Phys. Rev. B 1974, 9, 481.
12852. [37] H. Yang, E. Xu, C. Wu, J. Li, B. Liu, F. Hong, Y. Jiang, ACS Appl. Energy
[2] M. Ma, H. Chen, K. Zhou, C. Xie, Y. Liang, L. Wang, C. Wu, W. Yang, Mater. 2022, 5, 8254.
J. Guo, L. Luo, J. Mater. Chem. C 2021, 9, 2823. [38] R. H. Friend, D. Jerome, A. D. Yoffe, J. Phys. C: Solid State Phys. 1982,
[3] B. Wang, S. P. Zhong, Z. B. Zhang, Z. Q. Zheng, Y. P. Zhang, H. Zhang, 15, 2183.
Mater. Today 2019, 15, 115. [39] J. M. Bok, J. Hwang, H. Y. Choi, Phys. Rev. B 2021, 103, 205108.
[4] M. Mello, A. Scarascia, S. D. Guido, D. Altamura, V. Tasco, M. D. [40] M. I. R. Ebon, M. A. H. Pappu, S. N. Shiddique, J. Hossain, Opt. Mater.
Vittorio, A. Passaseo, Sensors 2019, 2008, 1588. Express 2024, 14, 907.
[5] T. C. Wei, D. S. Tsai, P. Ravadgar, J. J. Ke, M. Tsai, D. Lien, C. Huang, R. [41] X. Wen, Y. Feng, S. Huang, F. Huang, Y. B. Cheng, M. Green, A. Ho-
Horng, J. He, IEEE J. Sel. Top. Quantum Electron. 2014, 20, 3802006. Baillie, J. Mater. Chem. C 2016, 4, 793.
[6] M. Long, P. Wang, H. Fang, W. Hu, Adv. Funct. Mater. 2019, 29, [42] F. Yang, C. Zhang, C. Shi, M. J. Park, J. S. Kwak, S. Jung, Y. H. Choi, X.
1803807. Wu, X. Wang, M. Xiao, Appl. Phys. Lett. 2013, 103, 123506.
[7] J. Kang, L. Zhang, S. H. Wei, J. Phys. Chem. Lett. 2016, 7, 597. [43] T. Leijtens, G. E. Eperon, A. J. Barker, G. Grancini, W. Zhang, J. M. Ball,
[8] L. Lin, Y. Xu, S. Zhang, I. M. Ross, A. C. M. Ong, D. A. Allwood, ACS A. R. Kandada, H. J. Snaith, A. Petrozza, Energy Environ. Sci. 2016, 9,
Nano 2013, 7, 8214. 3472.
[9] S. Jo, N. Ubrig, H. Berger, A. B. Kuzmenko, A. F. Morpurgo, Nano [44] X. Wang, Z. Wu, J. Zhu, Y. Kang, M. Cai, Y. Xia, H. Deng, Opt. Mater.
Lett. 2014, 14, 2019. 2024, 147, 114594.
[10] R. Ahmad, R. Srivastava, S. Yadav, D. Singh, G. Gupta, S. Chand, S. [45] H. Yao, L. Liu, Nanomaterials 2022, 12, 325.
Sapra, J. Phys. Chem. Lett. 2017, 8, 1729. [46] Z. Zhu, H. Liu, Y. Du, S. Wen, J. Zhao, S. Wang, Phys. Scr. 2024, 99,
[11] M. Bernardi, M. Palummo, J. C. Grossman, Nano Lett. 2013, 13, 3664. 035918.
[12] A. Förster, S. Gemming, G. Seifert, D. Tomanek, ACS Nano 2017, 11, [47] S. N. Shiddique, M. I. R. Ebon, M. A. H. Pappu, M. C. Islam, J.
9989. Hossain, Heliyon 2024, 10, e32247.