Metal‐Doped MAPbBr3 Single Crystal p‐n Junction Photodiode for Self‐Powered Photodetection

Lead halide perovskites have emerged as the next‐generation materials for self‐powered photodetectors enabling operation without an external power source. In this study, a planar‐type photodetector based on metal‐doped p‐type MAPbBr3/n‐type MAPbBr3 single crystal showing excellent self‐powered photodetection properties is presented. The p‐n junction on the MAPbBr3 single crystal is formed by controlled epitaxial growth of Ag+‐doped MAPbBr3 SC (p‐type) on the facet of Sb3+‐doped MAPbBr3 SC (n‐type). The as‐integrated p‐n junction device with asymmetric electrodes shows a typical photovoltaic behavior with a high open circuit voltage of 0.95 V and great sensitivity to 530 nm illumination at zero bias with a responsivity of up to 0.41 A W−1 and a specific detectivity of 6.39 × 1011 Jones, which are among the highest values reported for MAPbBr3 single crystal‐based self‐powered photodetectors. In addition, the rise time and fall time of this device are as fast as 14 and 10 ms, respectively. These results pave the way for the fabrication of self‐powered perovskite‐based p‐n junction photodiode, which may find potential application in advanced photodiode and future optoelectronic devices.


Introduction
Photodetectors (PDs) which can directly convert various light wavelengths into detectable electrical signals have been the center of innovation for a long time due to their wide use in environmental monitoring, optical communication, and sensing. [1]OI: 10.1002/adom.202302032 The essential part of every PD is a semiconductor, which generates electron and hole pairs after the absorption of photons.Nowadays, a variety of organic and inorganic semiconductors such as carbon nanotubes, conjugated polymers, silicon, germanium, and indium gallium arsenide have been exploited in PDs. [2]Besides these materials, hybrid organic-inorganic halide perovskites in the form of single crystals (SCs) attracted the researcher's attention due to their low-temperature processing, broad flexibility of materials composition engineering, and easy bandgap tunability with outstanding optoelectronic properties. [3,4]16][17][18][19][20] To form a photoconductor, two metal contacts separated with a channel should be deposited on one plane of an SC surface. [20,21]owever, the photoelectric performance of MAPbBr 3 SC-based photoconductor still needs to be improved as it usually reveals a large dark current and a relatively slow response. [22,23]arious approaches such as passivation of the surface crystal defects, [24,25] A-site cation engineering, [26,27] depositing an asymmetrical electrode on a crystal surface [28] and selecting a (110) exposed facet [29,30] have been applied to optimize the optoelectronic properties of MAPbBr 3 SC and achieve efficient device.In spite of this progress, little work has been carried out to form photodiodes based on perovskite SC, which generally should exhibit reduced dark current and fast response speed.A photodiode has a depleted semiconductor region based on p-n, p-i-n, or Schottky junction with a high built-in electrical field at the interface, which efficiently extracts the photoexcited carriers. [31,32]Notably, these devices can also operate as self-powered PDs and convert the incident light into an electrical signal without an external power source. [33]ntentional doping of perovskite structure with various metal ions has appeared as a favorable method to form a perovskite SC-based photodiode and improve the optoelectronic properties without much change of the host crystal structure.For example, Liang et al. reported on the fabrication of a p-n homojunction photodiode by controlled doping of bulk MAPbBr 3 SC with Bi 3+ ions. [34]In this system, pristine MAPbBr 3 SC acts as a ptype semiconductor, while Bi 3+ -doped MAPbBr 3 SC shows n-type conductivity.Consequently, the as-assembled p-n homojunction SC-based photodiode demonstrated high responsivity and specific detectivity of the green light (520 nm) due to small lattice mismatch, continuous band alignment, and low carrier trap density.In another work, a p-n heterojunction photodiode was fabricated by epitaxially growing MAPbBr 3 SC (p-type) on Bi 3+ -doped MAPbCl 3 SC (n-type), showing low dark current density and favorable X-ray detection performance. [35]Recently, Wang et al. realized p-i-n photodiode for X-ray detection in which Ag + -doped MAPbBr 3 (p-type) and Bi 3+ -doped MAPbBr 3 (n-type) grow on the opposite faces of intrinsic MAPbBr 2.5 Cl 0.5 SC through epitaxial growth method. [36]In this photodiode system, p-type and n-type layers decrease the leakage of dark current by forming charge barriers.In the works mentioned above, the undoped perovskite SCs were used as a substrate for the controlled epitaxial growth of metal-doped perovskite SC on the face(s) of the substrate.However, to our best knowledge, a p-n junction photodiode based on either p-type or n-type metal-doped perovskite SC has not yet been reported.
Herein, we demonstrate for the first time a planar-type PD based on metal-doped p-type MAPbBr 3 /n-type MAPbBr 3 SC junction photodiode showing excellent self-powered photodetection properties.This device was achieved by the epitaxial growing of Ag + -doped MAPbBr 3 SC (p-type) on the facet of Sb 3+ -doped MAPbBr 3 SC (n-type), and depositing asymmetric electrodes on the top of the crystal surface.Optoelectronic analysis reveals that the self-driven p-n junction photodiode shows smaller dark current density and higher photocurrent compared to the undoped MAPbBr 3 and individual Ag + -or Sb 3+ -doped MAPbBr 3 SCs.As a result, a high responsivity (R) of 0.41 A W −1 and specific detectivity (D*) of 6.39 × 10 11 Jones was achieved at 530 nm and zero bias.In addition, the p-n junction photodiode exhibits an obvious photovoltaic effect with a high open circuit voltage (V OC ) of 0.95 V and short-circuit current (I SC ) of 5 μA.Meanwhile, it maintained good self-powered operation stability wherein the photocurrent maintained 80% of the initial performance after 12 h of continuous operation illumination in atmospheric conditions.

Results and Discussion
First, individual MAPbBr 3 SCs doped with Ag + and Sb 3+ were prepared using the inverse temperature crystallization method.As was previously reported by others, doping of perovskite structure with Ag + enhances its p-type conductivity, [37] while doping with Sb 3+ (similarly to Bi 3+ ) would move the Fermi level close to the conduction band (n-type). [38]Recent work by Haque et al. showed a negative effect of Bi 3+ doping on the photoresponse of MAPbBr 3 PD, [39] hence, in our study Sb 3+ was chosen as an n-type metal ion dopant.For the structural, optical, and morphological characterization of Ag + -and Sb 3+ -doped SCs (see Figures S1-S4 and Note S1, Supporting Information).Then, the selected SCs with a dimension of ≈4 × 4 × 2 mm 3 were used for the fabrication of planar-type photoconductors.The performance metrics of each device were collected under a green LED light pulse ( = 530 nm) with the irradiance power densities ranging from 0.1 to 50 mW cm −2 at a fixed bias of 2 V.The effect of p-and n-type metal doping on the electrical and photodetection properties of resulting PDs was thoroughly discussed.The fabrication of an individual p-n junction structure of Ag + -doped MAPbBr 3 /Sb 3+doped MAPbBr 3 SC was achieved by dipping a part of a Sb 3+doped MAPbBr 3 crystal into Ag + -doped MAPbBr 3 solution for 1 h followed by heating at 65 °C (Figure S5, Supporting Information).Finally, a self-powered p-n junction photodiode was fabricated by depositing asymmetric metal electrodes (Ag and Pt) on the top of the crystal surface, which creates a good asymmetric contact with a high built-in potential (Scheme 1).

Ag + Doping of MAPbBr 3 SC (p-Type)
To investigate the effect of Ag + doping at various concentrations (0.1%, 0.5%, and 1%) on the electronic properties of MAPbBr 3 SC, we first fabricated Pt/SC/Pt hole-only devices with the vertical geometry to calculate the trap density ( trap ), mobility (μ), and conductivity () (Figure S6, Supporting Information).The values of ,  trap , and μ can be determined using the Equations S1 and S2 (Supporting Information), respectively (see Note S2, Supporting Information).Space charge-limited current (SCLC) is the most common method to calculate these parameters for perovskite SCs. [40,41]However, it remains under debate whether SCLC measurement can accurately provide information about μ and  trap due to the strong ion migration in halide perovskite materials. [42]A recent study demonstrated that the pulsed voltage sweep SCLC method gives more accurate information about μ and  trap of halide perovskite materials by minimizing the ion migration effect during the measurement. [42,43]Therefore, we utilized this method to collect the dark I-V characteristics of the fabricated vertical devices from 0 to 10 V with 25 mV step voltage.A short time voltage pulse (20 ms) followed by a long rest time at 0 V (2 min) was used in this measurement as shown in Figure S6a (Supporting Information).
The dark I-V curves exhibit three distinct regions marked as Ohmic, trap-filling, and Child region (Figure S6, Supporting Information).To minimize the error of the measurement, six different SCs were tested for each composition.We observed a rapid increase in the  and μ with increasing the doping concentration of Ag + , reaching (7.86 ± 0.26) × 10 −4 Ω −1 cm −1 and 97 ± 9 cm 2 V −1 s −1 for 1% Ag-doped MAPbBr 3 SC as compared to the undoped MAPbBr 3 SC ((1.56 ± 0.26) × 10 −8 Ω −1 cm −1 and 27 ± 7 cm 2 V −1 s −1 ), respectively (Figure S7a,b, Supporting Information).The calculated  trap for all the SCs is shown in Figure S7c (Supporting Information).The average  trap of the undoped MAPbBr 3 SC is equal (2.96 ± 0.54) × 10 9 cm −3 , which is consistent with the recently reported works. [16,24]In the case of Ag + -doped samples,  trap reduces with increasing the doping concentration showing (2.16 ± 0.34) × 10 9 cm -3 for 1% Ag-doped MAPbBr 3 SC.In principle, a p-type dopant acts as an acceptor and creates more holes in the valence band of the semiconductor without acting as traps.Therefore, doping with Ag + creates more excess positive charge carriers in the MAPbBr 3 SC and changes the occupancy/behavior of existing traps leading to the reduction of trap density.Further analysis of PL spectra revealed a large increase in the PL intensity after increasing Ag + doping concentration (Figure S4, Supporting Information).This improvement could be ascribed to the reduced non-radiative recombination caused by the defects, which is consistent with the trap density results.Hence, we can expect higher photocurrent after Ag + doping due to the combined effect of low trap density and high conductivity with superior hole mobility.
Next, planar-type photoconductors based on the synthesized SCs were fabricated to elucidate the effect of Ag + doping concentration on the photodetection performance.Based on our previous study, [20] symmetric Pt electrodes rather than Au were deposited on the surface of SCs to achieve high-performed PD (Figure S8a, Supporting Information).We first measured the dark current (I d ) of our devices, which is an essential parameter for the characterization of PD.As shown in Figure S8b (Supporting Information), undoped MAPbBr 3 SC-based PD exhibits the lowest I d , which increases with increasing the Ag + concentration.It is important to note that I d of a semiconductor contains dark current due to the transport of holes (I dh ) and electrons (I de ).In a p-type MAPbBr 3 , electrons are the minority carriers, and the contribution of I de in the total dark current is negligible.On the other hand, the hole concentration increases with increasing the Ag + doping, which contributes more to the total dark current.In addition, the Fermi level shifts toward the valance band after p-type doping, and the additional acceptor (p-type dopant) creates more available states for electrons near the valence band edge, which helps to reduce the barrier between the Pt/MAPbBr 3 interface.Overall, a lower Schottky barrier height and decrease in depletion width due to higher hole density near the surface can trigger the overall I d of the device.Thus, the high I d of Ag + -doped devices can be attributed to the combination effect of enhanced charge mobility and conductivity with lower Schottky barrier height and depletion width.Figure S9 (Supporting Information) shows the chopped-light current-time characteristics of PDs with variable irradiance power densities ranging from 0.1 to 50 mW cm −2 .
Figure 1a shows the photocurrent density (J ph ) of all PDs as a function of incident light intensity.The value of J ph increases with increasing the intensity of incident light due to the increase in the charge generation rate.Notably, Ag + -doped MAPbBr 3 SC PDs generate almost 1000 times higher J ph than the undoped device.The value of J ph increases with increasing the Ag + doping concentration, which could be ascribed to the enhanced conductivity and major carriers concentration.The responsivity (R), external quantum efficiency (EQE), and specific detectivity (D*) of the fabricated PDs were calculated using the Equations S3-S5, respectively (see Note S3, Supporting Information).From Figure 1b,c, we can observe that 0.1% Ag + -doped device exhibits the highest values of R and EQE compared to other devices.The maximum R and EQE values for the undoped MAPbBr 3 SC PD are estimated to be 3.31 A W −1 and 773% under 0.1 mW cm −2 green light at a fixed 2 V.At the same conditions, the 0.1% Ag + -doped device exhibits the R of 3.95 A W −1 and EQE of 924%, which corresponds to an ≈20% increase compared to the undoped device.These parameters decrease with increasing the Ag + doping concentration, and for the 1% Ag +doped device R and EQE reach 2. Despite the highest photocurrent in 1% Ag + -doped PD, the highest values of R were observed for the 0.1% Ag + -doped device.We took a deeper look at this behavior by performing alternating current (AC) impedance spectroscopy (IS) in frequencies ranging from 100 mHz to 1 MHz and at 5 mW cm −2 green light.In addi-tion, a 2 V constant DC bias was applied during the IS measurement with 20 mV perturbation.The impedance responses in the form of Nyquist plots are shown in Figure S10 (Supporting Information).The shape of the Nyquist plot for the undoped MAPbBr 3 appears as a semicircle at the high frequencies followed by a small inductive loop at intermediate frequencies and a semicircle at low frequencies, which is similar to our recent study. [20]owever, the Nyquist plots for the Ag + -doped devices depict two semicircles without an inductive loop at the intermediate frequencies.A recent study shows that the absence/presence of an inductive feature could be directly related to the existence/lack of charge accumulation at the perovskite/contact interface. [44,45]herefore, the absence of the inductor loop in the Ag + -doped MAPbBr 3 samples indicates the less efficient charge separation compared to undoped MAPbBr 3 due to the higher amount of charge accumulation at the perovskite/Pt interface.In addition, this type of Nyquist plot is well established for the perovskitebased PDs and the arc at the high frequencies can be ascribed to the recombination process, while the arc at low frequencies is to the ion accumulation or migration. [13,20]As seen, the semicircle radius of the IS spectra rapidly reduces with increasing the Ag + doping due to the increase in conductivity (Figure S10b, Supporting Information).In addition, the recombination resistance (R rec ) also decreases after Ag + doping (Figure 2a).The lowest R rec was found for the device with 1% Ag + doping concentration, which suggests the highest recombination in this device.Figure 2b shows the normalized imaginary part of IS versus frequency for the undoped and Ag + -doped PDs under 10 mW cm −2 green light.The observed shift in the resonance frequency (f 0 ) toward the higher frequency range with increasing the Ag + doping concentration indicates the change in carrier recombination lifetime () (Figure 2b). [20,46]As shown in Figure 2c, the value of  (calculated from the resonance frequency ( = 1/f 0 )) slowly decreases with increasing the Ag + doping concentration.These results reconfirmed the presence of higher recombination of Ag + -doped devices compared to undoped SCs due to an increase in carrier concentration.On the other hand, the carrier mobility (μ) and carrier diffusion length (L D ) of a perovskite SC play a vital role in the device performance.It is well known that the L D depends on μ and , and it can be understood by the equation: where, L n , L p , e, k B , and T are the electron diffusion length, hole diffusion length, elementary charge, Boltzmann constant, and absolute temperature, respectively.We calculated  for all the PDs and found that the value of  is maximum for 0.1% Ag +doped PD (Figure 2d).These results indicate that the highest R and EQE for 0.1% Ag + -doped PD is due to the combined effect of the improved conductivity and optimum carrier diffusion length with superior hole mobility.Therefore, we chose 0.1% as the optimum Ag + doping concentration for further study on metal-doped self-powered PD (vide infra).

Sb 3+ Doping of MAPbBr 3 SC (n-Type)
Similarly to Ag + doping, a low concentration of Sb 3+ (0.1%) was used for doping MAPbBr 3 SC (for more details, see Experimental Section).Further increasing the doping concentration up to 0.5% leads to an increase in  S1 (Supporting Information).

p-n Junction Photodiode Based on Metal-Doped MAPbBr 3 SC
Encouraged by the improved electrical properties and photodetection performance of the individual p-and n-type metaldoped MAPbBr 3 SC-based PDs, we move forward to investigate the photodetection performance of a p-n junction photodiode with the configuration of Pt/Ag + -doped MAPbBr 3 /Sb 3+ -doped MAPbBr 3 /Ag.We deposited asymmetric electrodes on the surface of the p-n junction photodiode to enhance the responsivity under self-powered mode.Figure S13 (Supporting Information) shows the optical image of the as-fabricated device and top-view SEM of the as-growth p-n junction SC at the interface before asymmetric electrode deposition.
To evaluate the capability of the metal-doped p-n junction MAPbBr 3 photodiode for self-powered light detection, we also tested self-powered photodetection of the devices based on undoped MAPbBr 3 , individual 0.1% Ag + and 0.1% Sb 3+ -doped MAPbBr 3 SCs with the same asymmetric electrodes.The asymmetric Pt and Ag electrodes on the perovskite SCs form two different Schottky barriers at the interface between perovskite and metal electrodes due to the different work functions (Pt = −5.2eV; Ag = −4.3eV).Therefore, an electric field was built into the device, which helps to separate photogenerated charge carriers under self-powered conditions (Figure S14, Supporting Information).Figure 4a,b shows I−V characteristics of different devices under dark and green light illumination (50 mW cm −2 ), respectively (for all power-dependent I−V characteristics see Figure S15, Supporting Information).As seen, the p-n junction photodiode shows a lower dark current than both metal-doped MAPbBr 3 SC-based PDs under reverse bias conditions, which is beneficial for PD application.Under light illumination, a clear photovoltaic effect is observed for all devices, which can be ascribed to the built-in electric field due to asymmetric Schottky barrier formation (Figure 4c; Figure S16, Supporting Information).As expected, the p-n junction photodiode exhibits the highest open circuit voltage (V OC ) of 0.95 V with a significant short-circuit current (I SC ) of 5 μA under 50 mW cm −2 green light.This can be attributed to the combined effect of the electric field due to asymmetric Schottky barrier formation and additional built-in electric field in the p-n junction depletion region, which helps to collect the photogenerated carriers without external bias (Figure S16, Supporting Information).As shown in Figure 4d, the photovolt-age of all PDs is found to increase with increasing light intensity due to the increased number of photogenerated charges.In order to study the self-powered photodetection performance of the devices, the transient photoresponses were measured at zero bias and under green LED light pulse ( = 530 nm) with the irradiance power densities ranging from 0.1 to 50 mW cm −2 .The currenttime (I-t) characteristics of PDs at 1 mW cm −2 irradiance power are shown in Figure 4e.Due to the high conductivity, the photocurrent of the Ag + -doped device is very high as compared to other devices.On the other hand, the p-n junction photodiode generates almost 5 times higher I ph than the undoped MAPbBr 3 and Sb 3+ -doped MAPbBr 3 SC-based devices.Figure S17 (Supporting Information) shows the transient photoresponse of the p-n junction photodiode under light intensity ranging from 0.1 to 50 mW cm −2 .Figure 4f shows the logarithmic plot of J ph as a function of the irradiation intensity under self-power conditions for all devices.To reveal the recombination-dependent relationship between the photocurrent and irradiation power, we fitted these results with a power law (J ph ∝ P  , where P is the irradiation power and  is the recombination under illumination).The maximum value of the  exponent increases to 0.46 for the p-n junction photodiode compared to the undoped device ( = 0.40) at the low light intensity region, which indicates lower recombination and better charges collection of the p-n junction photodiode.The  exponent of the Ag + -doped device is almost 10 times lower than the p-n junction photodiode due to the high recombination.The low dark current of our p-n junction photodiode may indicate a low noise.The shot noise (i noise ) was determined by the dark current using the following expression with a bandwidth of 1 Hz [47] : where q is the elementary charge, I d is the dark current, and B is the bandwidth.Based on the dark current of the p-n junction photodiode at 0 V, the shot noise is calculated to be 300 fA Hz −1/2 .To shed more light on the operational performance of the fabricated self-powered PDs, we compared their photodetection parameters.From Figure 4g  perspective.The functional stability of the p-n junction photodiode was measured under a continuous 10 mW cm −2 green light pulse at zero bias under ambient conditions (RH = 30%, 25 °C).As shown in Figure 4i, the output of this PD increases by 10% after 1 h of operation and then decreases slowly.After 12 h of continuous operation, the device maintained 80% of its initial performance, which confirms the long-term operational stability of the as-fabricated p-n junction photodiode based on metal-doped MAPbBr 3 SC.Notably, the device restored 94% performance of its initial value after 36 h of storage in the dark.For further long-term stability tests, the p-n junction photodiode was remeasured after 90 days of storage in the dark and maintained ≈92% performance of its initial value, confirming the noticeable stability of the metal-doped p-n junction photodiode.

Conclusion
In summary, a high-performance PD based on metal-doped ptype MAPbBr 3 /n-type MAPbBr 3 SC junction photodiode was designed and fabricated.First, individual MAPbBr 3 SCs doped with Ag + (p-type) and Sb 3+ (n-type) were prepared, optimized, and characterized.The planar-type PD based on MAPbBr 3 SC with an optimum Ag + doping concentration (0.1%) shows the best photodetection properties of green light at 2 V with a peak responsivity of 3.95 A W −1 and EQE of 1093%, which was almost 20% higher than those of undoped device.The better performance of the Ag + -doped device was attributed to the improved conductivity and hole mobility.Similarly, n-type MAPbBr 3 SC with a low concentration of Sb 3+ (0.1%) shows similar improvement in PD performance due to the combined effect of high electron

Scheme 1 .
Scheme 1. Schematic illustration of the self-powered p-n junction photodiode based on metal-doped MAPbBr 3 SC with the asymmetric metal electrodes.

Figure 2 .
Figure 2. a) Recombination resistance (R rec ) and b) normalized complex impedance part.c) Carrier recombination lifetime () and d)  values of the undoped and Ag + -doped MAPbBr 3 SC-based PDs under 10 mW cm −2 green LED light ( = 530 nm).Data were collected on six different SCs.
26 A W −1 and 530%.The values of D* as a function of the intensity of light for all the studied PDs are shown in Figure 1d.As could be expected, the undoped MAPbBr 3 SC-based PD shows better D* than the Ag +doped devices due to the inverse correlation of D* with the dark current.
d and a decrease in R and D* values (FigureS11, Supporting Information).The values of ,  trap , and μ are calculated using the pulsed voltage sweep SCLC method and summarized in FigureS12(Supporting Information).As seen, the value of  slightly increases to (2.06 ± 0.26) × 10 −8 Ω −1 , while values of  trap and μ reduce to (3.16 ± 0.34) × 10 9 cm −3 and 12 ± 7 cm 2 V −1 s −1 after Sb 3+ doping, respectively.Next, a planar-type photoconductor with symmetric Ag electrodes was fabricated to elucidate the effect of n-type Sb 3+ doping on the photodetection performance.We chose an Ag electrode rather than Pt to make Schottky contact on the Sb 3+ -doped MAPbBr 3 as the work function (ϕ m ) of Ag is smaller than the work function of Sb 3+ -doped MAPbBr 3 .As shown in Figure3a, the I d and I ph of the MAPbBr 3 SC-based PD increases after Sb 3+ doping.In n-type MAPbBr 3 , electrons are the majority carriers, and the contribution of the hole is reduced after Sb 3+ doping.Therefore, depletion width and Schottky barrier height at the Ag/MAPbBr 3 interface decrease after Sb 3+ doping due to higher electron density near the surface.As a result, the I d and I ph of the MAPbBr 3 SCbased PD increase after Sb 3+ doping due to the combined effect of high electron concentration and low Schottky barrier height between Ag/MAPbBr 3 interface with lower depletion width compared to the undoped device.Next, we calculate all the performance parameters, and Figure3bshows the logarithmic plot of J ph as a function of the irradiation intensity under 2 V bias.The Sb 3+ -doped PD generates almost 30% higher I ph than the undoped device.From Figure3c,d, we can observe that R and EQE of MAPbBr 3 SC PD improve after 0.1% Sb 3+ doping.The maximum R and EQE values for the undoped PD are estimated to be 0.38 A W −1 and 90% under 0.1 mW cm −2 green light at a fixed bias voltage of 2 V.At the same conditions, the 0.1% Sb 3+ -doped PD exhibits R of 0.49 A W −1 and EQE of 115%, which corresponds to an ≈30% increase compared to the undoped device.The values of D* as a function of the intensity of light for both PDs are shown in Figure3c.The values of D* have no significant change after Sb 3+ doping due to the higher dark current compared to the undoped device.The performance parameters of our and other previously reported PDs based on MAPbBr 3 and metal-doped MAPbBr 3 SC are summarized in Table

Figure 4 .
Figure 4. I-V characteristics of the undoped and Ag + -doped MAPbBr 3 SC-based PDs under (a) dark and (b) 50 mW cm −2 green light illumination.c) I-V curves of the p-n junction photodiode under 530 nm light with various light intensities.d) The photovoltage of the self-powered PD under different light intensities at zero bias.e) Current-time (I-t) characteristics of self-powered PD at 1 mW cm −2 irradiance power.f) Responsivity (left axis), specific detectivity (right axis), and g) EQE of the self-powered PD. h) Functional stability of the p-n junction photodiode under continuous 10 mW cm −2 irradiation intensity for 12 h, after 36 h and 90 days of storage in the dark.
,h, we can observe that the p-n junction photodiode exhibits the highest values of R, EQE, and D* compared to other devices.The maximum R and EQE values are estimated to be 0.41 A W −1 and 96.6% under 0.1 mW cm −2 green light at zero bias, respectively.The rise-time and fall-time of the PDs were measured under periodically switched light irradiation.The response time of MAPbBr 3 SC-based PD increases after doping with metal ions due to enhanced conductivity and improved collection of the photogenerated charge carriers.The p-n junction photodiode exhibits a short response time ( on = 14 ms,  off = 10 ms), while the fastest on and off time ( on = 8 ms,  off = 5 ms) was observed for the Ag + -doped MAPbBr 3 PD due to the fastest movement and recombination of the photogenerated charge carrier.TableS2(Supporting Information) summarizes the performance parameters of our and other previously reported self-powered MAPbBr 3 SC-based devices.Compared with the reported results, the p-n junction photodiode based on metal-doped MAPbBr 3 SC exhibits superior photodetector parameters than all other SC-based Schottky junction and heterojunction PDs.The operational stability of PDs under continuous operation is an essential requirement from the commercialization