Photodynamic response of a solution-processed organolead halide photodetector

Abstract

CH₃NH₃PbI₃ perovskite semiconductors have received intensive attention as a light absorbing material in high performance solar cells and photodetectors. Herein, we demonstrated a solution-processed photodetector on an indium tin oxide (ITO) substrate by a low-cost spin coating technique. The photodetector exhibits high responsivity, large detectivity and a broad spectrum response. In order to understand the operational mechanism of photodetectors, diode characteristics under dark and at various wavelength lights are performed. The band structure calculation and absorption spectra based on density functional theory (DFT) exhibits a direct band-gap value of 1.53 eV, which is in good agreement with the experimental results. The projected density of states (PDOS) and total density of states (DOS) analysis shows the bonding mechanism in the CH₃NH₃PbI₃. The performance of this perovskite-based photodetector markedly exceeds that of organic and hybrid photodetectors. The accumulation of charge carriers and the hysteresis mechanism was also addressed by a model based on ion migration. Our results provide a proper device design with a simpler approach that has the potential to be scaled up.

---

Introduction

Photodetectors having a wide spectral range sensitivity have a variety of applications from daily consumer electronics, like smoke detectors, remote controls, imaging in cameras and disc players, to more complicated applications like optical communication, defense sensing and environmental monitoring.^1–4 To get a reasonable detectivity and sensitivity, a photodetector requires a certain potential. For that, they can either be integrated with energy storage or energy harvesting units like batteries, bio-fuel and piezoelectric devices.^5–7 The utilization of these extra circuits makes the overall system complex and expensive. Therefore, the development of photodetectors that are energy efficient or self-powered is crucial for an independent and sustainable system. Development of this kind of photodetector is one of the aims of the present study.

In general, photodetectors are operated in photovoltaic and photo-conducting mode using the principles of photo-electrochemistry, photodiode action and photoconduction.^8–10 The nano-structured inorganic semiconductor-based photodetector that is currently in the market exhibits many attractive properties like high mobility, absorption co-efficient and stability; but there are still some drawbacks in fabrication and material properties such as high temperature processing, complex fabrication processes and the presence of internal and deep level of defects.^11,12 Recently, photodetectors based on organic semiconductors have attracted much attention due to their solution processability, wide range of band gaps, high absorption co-efficient, long carrier life time and diffusion lengths.^13–15 Shi et al.¹⁶ have studied and reported an anti-solvent vapor-assisted crystallization approach to synthesize sizable crack-free MAPbX₃ single crystals. Yang et al.¹⁷ have investigated the recombination dynamics in CH₃NH₃PbBr₃ and CH₃NH₃PbI₃ films in order to reveal the influence of exciton binding energy. The organic semiconductor material-based photodetector utilizes either an organic or organic–inorganic hybrid configuration. A hybrid photodetector employs semiconductors like TiO₂, SnO₂, ZnO and Al₂O₃, which act as a hole transporting layer (HTL) or electron transporting layer (ETL). But the performance of these devices is lacking due to excessive parasitic capacitance along with additional fabrication steps.^18–21 Therefore, organic photodetectors with high detectivity and sensitivity, with less processing steps, are highly desired.

CH₃NH₃PbI₃, an organic semiconductor, has a wide range of optoelectronic applications because of its many functionalities like non-vacuum processing at low temperature, presence of only shallow level defects, high diffusion length and long absorption wavelength range that makes CH₃NH₃PbI₃ the most preferred candidate for a self-powered and broad wavelength photodetector.^22,23 The fast photoconductive response in an organometallic halide perovskite was studied by Wang et al.¹ Hu et al. demonstrated a flexible broadband high performance photodetector based on CH₃NH₃Pbl₃.²⁴ A high gain and low driving voltage photodetector based on organolead tri-iodide perovskite was illustrated by Dong et al.²⁵ A hybrid perovskite photodetector with high detectivity using a solution-processable method was developed by Dou et al.²⁶ Zhou et al.²⁷ have investigated the performance of an ultraviolet-visible perovskite photodetector based on TiO₂ nanorods/CH₃NH₃PbI₃ heterojunctions. Moreover, Yu et al.²⁸ have fabricated a CH₃NH₃PbI₃/ZnO nanorod arrays heterostructure in order to study the performance parameters. Lee et al.²⁹ have also presented a perovskite-graphene hybrid photodetector that resulted in a photo-responsivity of 180 A/W and photo-detectivity of 10⁹ Jones. Zhang et al.³⁰ have prepared island-structure CH₃NH₃PbI₃ thin films and obtained a rapid response speed with good stability at temperatures up to 100 °C. Chen et al.³¹ have fabricated a flexible photodetector based on a perovskite/conjugated-polymer composite.

In the literature, different structural, morphological and optical strategies have been discussed for the development of perovskite-based photodetectors.^1,9–11 However, it still remains a challenge to develop high speed, self-powered and broad wavelength photodetectors using a simple and low cost route. Researchers have investigated different structures consisting of HTM and ETM in the form of thin films or nanorods, and good performance has been observed. However, this increases the complexity of structure, cost of the device and the more junctions present in such a device acts as a recombination zone. The structure presented in this article can lead to a low-cost, easily processable and high performance photodetector. Moreover, the detailed, in depth, photodetector diode characteristics are rarely discussed in the literature. In the present report, we synthesize the perovskite semiconductor by using a low-cost spin coating technique. The fabricated photodiode shows a zero bias and broad wavelength photo-detection with an impressive photo-responsivity that is comparable with efficient organic-based and inorganic-based photodiodes. Moreover, the present article also focuses on the following: (i) the diode characteristics of the photodetector that reveals detailed information about recombination and charge transportation. (ii) The photodynamic response under different wavelengths. The photodetector shows a photo-response to a wide range of wavelengths. (iii) The scan rate-dependent response of the photodetector. To the best of our knowledge, this is rarely discussed. Herein, we have successfully demonstrated a solution processable, high sensitivity, impressive, photodetector with a wide band performance along with in depth diode and transient response analysis.

Experimental section

The detailed steps involved in the synthesis of methylammonium iodide (CH₃NH₃I) have been discussed in detail in our previous reports.^32–34 However, in brief, the fabrication of the perovskite absorber layer was prepared by mixing CH₃NH₃I and PbI₂ (1.27 g, 99% Aldrich) in a 3 : 1 molar ratio in γ-butyrolactone (2 ml, 99% Aldrich) at 60 °C for 24 hours to get CH₃NH₃PbI₃. Indium-doped tin oxide glasses (ITO) (1 × 1 cm) were used as substrates and cleaned ultrasonically for 15 min in acetone, ethanol and deionized water, respectively. CH₃NH₃PbI₃ was spin coated onto ITO at 3500 rpm for 45 seconds and then annealed at 100 °C for 1 hour under open air conditions.

The morphological and the optical properties of the as prepared CH₃NH₃PbI₃ were investigated by X-ray diffraction (X'Pert Pro) in the 2θ range of 5–70° and UV-visible spectrometry (UV-VIS) on a Shimadzu UV 2600 spectrophotometer. The diode and hysteresis was tested under LED light (blue, green and red) and in the dark, respectively, at 25 °C using an Agilent 2772 source meter. The photodynamic measurements were carried out at room temperature using a CH Instrument 660D potentiostat. Impedance spectroscopy (IS) measurements at zero bias were done by applying a perturbation signal of 10 mV amplitude in the frequency range of 1 Hz to 100 kHz using a CH Instrument 660D potentiostat equipped with general purpose electrochemical system software.

Theoretical section

The calculations were performed using density functional theory (DFT) implemented in the Vienna ab initio simulation package (VASP). The gradient approximation (GGA) was employed for exchange effects along with the Perdew–Burke–Ernzerhof (PBE) and projected augmented wave (PAW) potential for the correlation effects. A CH₃NH₃PbI₃ structure was considered for structural optimization with a cutoff energy of 550 eV for plane-wave expansion of the electronic wave function. The Brillouin zone integration scheme was sampled with a 8 × 6 × 8 and 11 × 11 × 11 grid for geometry optimizations and static electronic structure calculations, respectively. All calculated equilibrium configurations were fully relaxed with residual forces smaller than 0.01 eV Å⁻¹ and the energy convergence criteria of 1 × 10⁻⁶ eV for each atom.

Results and discussion

Structural and optical properties of CH₃NH₃PbI₃

Fig. 1 depicts the X-ray diffraction (XRD) pattern of the as prepared, polycrystalline, thin film perovskite semiconductor on the ITO substrate. The characteristic peak at 2θ values of 15.23, 29.60, 41.59 and 44.23 corresponds to the 110, 310, 224 and 314 planes of tetragonal perovskite with a lattice parameter of a = 8.825 Å, b = 8.835 Å and c = 11.24 Å. The obtained pattern and corresponding lattice parameters are consistent with those reported in the literature.^35,36 The shown XRD pattern exhibits a high degree of crystallinity and orientation.

Fig. 1 (A) XRD spectra of the polycrystalline thin film CH₃NH₃PbI₃. (B) The UV-VIS absorption spectrum of the CH₃NH₃PbI₃. (C) Schematic diagram of the fabricated device. (D) Band structure of the tetragonal CH₃NH₃PbI₃ semiconductor. (E) PDOS analysis of CH₃NH₃PbI₃ using the PAW–GGA functional. (F) Calculated absorption spectra of CH₃NH₃PbI₃ under (x and y) parallel and perpendicular (z) polarizations.

The UV-VIS absorption spectrum of the perovskite material is shown in Fig. 1B. The absorbance spectrum exhibits a good light absorbing capacity over a wide range of wavelengths from ultraviolet to the visible range. The absorption spectrum at 700 nm corresponds to the transition of charge carriers between the valence and conduction band was used to calculate the band gap.^33,34 From the Tauc plot, the direct band gap value of 1.5 eV was estimated, which is commonly quoted by other authors.^37,38 Fig. 1C shows a schematic diagram of the fabricated device structure where the CH₃NH₃PbI₃ represents the photoactive material, and ITO and gold represent the electrical contacts for the collection of photo-generated electrons and holes, respectively. Compared to the conventional structure employed for photodetector applications, the presented archetype has the following advantages: (i) the presented architecture has not employed any electron transport and hole transport layer that causes an excess parasitic capacitance in the device and leads to a slower response; and (ii) a low temperature solution processable method was used compared to vacuum or high temperature techniques. The present approach provides a way of low cost deposition along with the additional advantage of lower current leakage. Leakage is generally observed in mesoporous oxide based photodetector.^33,34,38

To further understand the band structure and structural properties of CH₃NH₃PbI₃, theoretical calculations were performed based on the projected augmented wave (PAW) method. Fig. 1D shows the calculated band structure of the tetragonal CH₃NH₃PbI₃ semiconductor. To generate the theoretical band structure, the perovskite CH₃NH₃PbI₃ was initially optimized. For that, obtained structural parameters discussed in the context of Fig. 1A were taken as input parameters. The presented band structure (Fig. 1A) was quite similar, as shown by authors, which exhibits a direct band-gap value of 1.53 eV, in agreement with the experimental results in Fig. 1B.^39,40 The band structures of the valance band minimum (VBM) and conduction band minimum (CBM) (top level) were quite flat and broad, which corresponded to non-localized states. In practice, the transport property of the semiconductor is correlated with the relative broadness of the VBM and CBM, which explains charge carrier transport over a long distance in CH₃NH₃PbI₃. Moreover, a higher number of band levels in the bottom of the CBM can accelerate the transition and transport of photo-excited charge carriers in the conduction band.

In order to understand the bonding mechanism in the CH₃NH₃PbI₃ atoms, the projected density of states (PDOS) analysis was done. The total density of states (TDOS) using the GGA functional was also performed; but due to the small variation in the characteristic peak, an attempt using the PAW–GGA functional analysis of PDOS is shown in Fig. 1E. As shown in Fig. 1E, the strong hybridization in the orbital of I 5p and Pb 6s atoms mainly dominates the VBM, whereas the CBM is mainly dominated by the Pb atom.⁴¹ A minor contribution from CH₃NH₃⁺ cations in the VBM and CBM was observed. It is concluded that net resultant PDOS of CH₃NH₃PbI₃ was due to Pb–I bonding. Under illumination conditions, the photoelectrons are excited from the I 5p orbital level, which later becomes a photohole site, and the Pb 6p conduction band is occupied by the excited electrons.

Along with the special electrical structure and property, CH₃NH₃PbI₃ also possesses excellent absorption spectra, as calculated using the PBE functional. The absorption spectrum of CH₃NH₃PbI₃ is depicted in Fig. 1F, which consists of low frequency peaks and threshold absorption peaks. The absolute threshold absorption peaks were equal to the energy band gap. For both parallel and perpendicular polarization direction (E_∥x) and (E_⊥z), almost the same photon energy absorption peaks were observed. The first two absorption peaks were observed at energies of 3.49 eV and 6.98 eV (CH₃NH₃PbI₃), and the highest absorption peaks were at energies of 5.21 eV and 3.49 eV.

Diode characteristics

The net J–V and diode characteristic of the perovskite photodetector under dark conditions at room temperature is shown in Fig. 2A. The main objective of characterizing the diode parameter of the fabricated photodetector is to address the carrier recombination and transport mechanism that defines the detectivity and the speed of the photodiode. The fabricated device exhibits a knee voltage of 0.78 V, rectification ratio of 8 at ±1.3 V and reverse saturation current (J₀) of 1.5 × 10⁻⁸ A. A higher value of knee voltage in the present case than conventional and perovskite-based photodetectors signify a better contact formed between ITO and the perovskite material, whereas a much lower value of J₀ can lead to a high photo-detectivity. Fig. 2A shows the plot of √J vs. V, where in a low forward bias J ∝ Vⁿ, n = 1 follows ohm's law and at n > 2, space charge limited transport dominates. To further understand the dark J–V characteristics by considering the both Shockley–Read–Hall (SRH) and radiative recombination, it is observed that the obtained response was mainly dominated by SRH recombination. Agarwal et al. concluded a similar domination of SRH over radiative recombination for the perovskite-based solar cell.^42,43

Fig. 2 (A) The net J–V and diode characteristic of the perovskite photodetector under dark conditions. (B) Energy band diagram of the HTM-free perovskite solar cell. (C) Variation of ideality factor as a function of applied bias.

The energy band diagram shown in the Fig. 2B and based on our previous studies on the HTM-free perovskite solar cell, the following possible recombination/transport in different layers are hypothesized: (i) transport of electrons from ITO contact to gold, holes transport from gold to ITO contact, (ii) recombination within the perovskite material, (iii) recombination between HTM, ETL and the perovskite interface and (iv) recombination/transport mechanism in HTM and ETL. From the listed hypothesis, (i), (iii) and (iv) can be eliminated because it is known that transport in ITO and gold is high, and it has also been shown in literature that the diffusion length of electron and hole is high in perovskite. Therefore, the net current in the device is limited through a recombination/transport mechanism in the perovskite layer. This hypothesis is further confirmed by the √J vs. V plot, where a linear characteristic in the high forward bias region signifies the domination of SCLC (space charge leakage current).^42,43 The presented results support the findings of Agarwal et al. that the applied potential across the perovskite photodetector or solar cell is mostly dropped across the perovskite layer (at a low forward bias) due to the SRH recombination, irrespective of the ETL and HTL.^43,44 Authors have shown that the structure consisting of the ETL and HTL layer under higher bias, the further potential drop is processed by the HTL while the drop across the ETL is negligible under the entire probed bias region.^42–44 Moreover, this is not a case in the present study as we have not employed ETL and HTL layers.

From Fig. 2A, the mobility of the carriers was calculated using the Mott–Gurney relation for the ohmic and SCLC regions. In the lower bias voltage, the linear behavior of net current with respect to voltage can be given by J_ohm = qn_oμV/d, where V is the applied voltage, q is the electronic charge, n_o, d and μ represents the charge carrier density, thickness of the film and carrier mobility, respectively. By taking the experimental curve with the discussed expression, the value of μ was obtained as 1 cm² V⁻¹ s⁻¹. In the mid-bias range, where the slope of log J vs. V is described by the expression J = 9/8(εε_oμV²/d³), (ε is the dielectric constant of the material), using above expression, the value of μ was found to be 4 cm² V⁻¹ s⁻¹. The disparity in the value of μ in both of these cases may be due to the source of origin, where in the case of ohmic conduction, the density of thermionic emission of the electron in the bulk perovskite semiconductor is greater than the injection of electron from electrodes. Whereas in the Mott–Gurney region, the injected electron carriers exceed the bulk carrier density.⁴⁵ In the high forward bias region, log J tends to saturate and follows the power law, indicating the trap-limited SCLC region, and can be given through the expression V = qN_td/2ε. Considering the carrier lifetime of 10⁻⁷ s, the experimentally obtained mobility of the carrier diffusion length of a few hundred nm was obtained, which is commonly quoted for CH₃NH₃PbI₃ material.^46,47

In order to further elucidate properties of the fabricated photodiode, the ideality factor as a function of bias and contribution of resistive losses due to contacts are discussed below. The ideality factor can be further explained by a one diode model, and it is given as J = J₀[exp(qV/nkT) − 1], where n, k and T represent the diode ideality factor, Boltzmann's constant and device temperature, respectively. By rearranging the above expression, the variation of n as a function of applied bias is given by n = q/kT(∂ ln J/∂V)⁻¹, and as shown in Fig. 2C, the value of n ranges from a lower value of 1.1 to a maximum value of 5.2. Moreover, the value of n = 2 represents the domination of the SRH region. In general, n < 1 does not represent the ideality factor, but acts as an indicator to locate this region. A high forward bias that is mainly dominated by SCLC is represented and n > 2 is obtained, which is observed by other authors for perovskite-based photovoltaic devices; but the physical significance is still not clear. In the literature, the following postulates were discussed for n < 1 and n > 2 as: (i) n ≤ 1, the domination of charge carriers over the barrier is dominated. (ii) A significant voltage drop across the HTL and ETL layer leads to n > 2. (iii) Parasitic shunt resistance effect. In the present case, the series and shunt resistances of the fabricated diode is discussed further.

Photodynamic response

To investigate the photo response behavior of the fabricated photodiode in the visible spectrum of blue (455 nm), green (530 nm) and red (625 nm) illumination, investigations on the net terminal J–V characteristics were carried out and shown in Fig. 3A.

Fig. 3 (A) The net terminal J–V characteristics of blue (455 nm), green (530 nm) and red (625 nm) illumination. (B) Photodynamic response of the fabricated photodetector under red, green and blue light having the same intensity at room-temperature and zero bias. (C) The photodynamic response of the fabricated photodetector at zero bias under the illumination with blue. (D) The IS spectra recorded at zero bias for red, green and blue wavelengths at room temperature.

From the obtained curve, it seems that the lowest current and highest resistance is found under red light illumination; while under blue light, the current is increased many-fold. The correlation between photocurrent magnitudes with wavelength of illumination can easily be elaborated from the shown UV-VIS spectra in Fig. 1B. As shown in the absorbance curve, it can be easily seen that the maximum absorption peak is located in the vicinity of blue light. The minimum absorbance is seen near to the 700–750 nm wavelength region, which corresponds to red light and the bandgap of the CH₃NH₃PbI₃ semiconductor. Thus, the net photo-generated current mainly depends upon the wavelength of the light and the interfacial electric field. As the wavelength of light is more than the bandgap of the semiconductor, photo carriers are generated and can be transferred to different electrodes by energy band bending or alignment. The photodynamic response of the fabricated photodetector under red, green and blue light having the same intensity at room-temperature and zero bias is shown in Fig. 3B. Such properties are promising for the designing of self-bias broad wavelength range photodetectors. When the LED was in the OFF condition, the current through the photodetector was very low. While with the LED in the ON condition, the photocurrent was increased by 9-times for the red LED. Whereas in the case of green and blue light, the photocurrent increased by 13 and 25 times, respectively. Here, the generation of the photocurrent and difference in their magnitude correspond to different wavelengths attributed to the same UV-VIS absorption phenomena, and band to band transition is discussed in the context of Fig. 3A. Moreover, under the illumination of red, green and blue light, the photocurrent rises to its 90% of the maximum value in a rise time of 20.4, 27.1 and 61 ms, respectively. In the OFF condition, the photo-generated charge carrier decay towards the dark condition, due to progressive involvement of the charge carrier recombination, was processed in a time delay of 6.06, 18.19 and 40 ms, respectively. The photodiode exhibits a photosensitivity (S) (J_ph − J_d)/J_d of 4.55, 18.0 and 27, under red, green, and blue LED wavelengths, which is comparable with those reported for perovskite-based photodetector.^1,5,8 Another characteristic feature of the broad band photodetector is the spectral responsivity. This is given as R_λ = I_P/PA, where A is the active area and P is the power density. The responsivity calculated for red, green and blue light under zero bias and a power density of 0.2 μW cm⁻² was obtained to be 0.27, 0.42 and 0.79 A/W, respectively. The specific detectivity of the photodetector is given as , where B is the electrical bandwidth and NEP is the equivalent noise power. D* is also expressed using the expression . In the present study, the D* for red, green and blue light was obtained to be 0.52 × 10¹³, 1.071 × 10¹³ and 2.20 × 10¹³ Jones, respectively. In conclusion, it is shown that high specific detectivity, high ON/OFF ratio and high detection speed are consequences of the particular archetype of the device, which avoids the additional parasitic capacitance offered by ETL and HTL and lower diode recombination current. The analysis of the fabricated photodetector at different wavelengths shows the applicability of the device as a self-biased broad band photodetector. It is clearly seen in Fig. 3A that the maximum and minimum photocurrent is observed under blue and red light respectively. Impedance spectroscopy (IS) has been performed to get further insight into resistive and capacitive components of device at different wavelengths wherein, a small perturbation voltage signal is applied across the photodetector with an angular frequency (ω) in the steady state. It is to be mentioned here that very few reports on utilizing the IS technique for photodetectors have been reported.^32–34 The objective of utilizing the IS response in the present case is to get a view on the variation of resistive and capacitive parameters of the photodetector with changes in the wavelength of light. Fig. 3D represents the IS spectra at zero bias for red, green and blue wavelengths at room temperature. The left and right most intercepts of the imaginary impedance on the real axis represent the maximum frequency of 100 kHz and low frequency of 1 Hz for all the wavelengths. In order to extract the resistive and capacitive values, the obtained IS spectra was fitted with an electrical equivalent circuit consisting of parallel RC components in series with series resistance (R_S). The net impedance expression for the electrical equivalent circuit is given by the expression Z = Z′ + Z′′, where Z′ and Z′′ are the real and imaginary components of the impedance, respectively.³³ The obtained fitted values of the interfacial capacitance were 1.79, 2.78 and 4.8 pF for red, green and blue wavelengths, respectively. The high photosensitivity in the photodetector is only possible when the excited photo-generated charge carriers rapidly sweep across the terminal and high numbers of photo-carriers are generated and the values of interfacial resistance are found to be 14.9, 9.22, 4.8 MΩ for red, green and blue light respectively which matches well with the trend shown in Fig. 3A. A lower value of interfacial resistances for the blue wavelength signifies a minimum resistance was faced by the photo-generated charge carriers during transportation.

Scan rate dependent response

The photodetector presented in this study has demonstrated a significant wavelength-to-sensitivity response. However, the hysteresis behavior in perovskite devices together with the performance deciding parameter have emerged as critical issues. Although the accurate origin mechanism is still an open quest, several mechanisms are proposed based upon the ferroelectricity, trapping and slow ion migration in perovskite semiconductors.^48,49 Hence, it is necessary to first explore whether the device poses hysteresis effect or not and if it exists, then what will be the physical process affecting the photoresponse is discussed in more details below. Fig. 4 shows the hysteresis measurement by J–V at different scan rates for both forward and reverse sweep directions under the dark and illumination conditions at room temperature. The J–V characteristic curves at different scan rate under dark and illumination condition follows the same trend, but differ only in the magnitude, which may be due to the excess electron–hole pair generation under illumination. In the voltage range of 0.2 to 0.8 V at a higher sweep rate, the hysteresis effects are profound and follows the sweep direction; where as in the voltage range of V > 0.8 and at a low scan rate, the net current emerges to lower values and attains the equilibrium value of J₀. The details of origin and physical significance of J₀ are discussed in the context of Fig. 2.

Fig. 4 (A) Hysteresis measurement by J–V at different scan rates in both the forward and reverse sweep directions under dark conditions. (B) Plot of net capacitive current under dark conditions. (C) Linear variation of J_CAP vs. scan rate under dark conditions. (D) Hysteresis measurement by J–V at different scan rates in both the forward and reverse sweep directions under illumination conditions. (E) The plot of net capacitive current under illumination conditions. (F) Linear variation of J_CAP vs. scan rate under illumination conditions.

The excess current to the J₀ can be calculated by J_CAP = J − J₀ and is shown in Fig. 4(B and E) for dark and illumination conditions. The shown plot trend in Fig. 4(B and E) is similar to the results of Almora et al. reported for planar and mesoporous perovskite-based solar cells and those reported for electrochemical cyclic voltammetry methods, batteries and supercapacitors.^50,51 The value of capacitance can be derived from the Fig. 4(B and E) data and the discussed electrical equivalent circuit in Fig. 3. The net current has the form of J = J_r ± V_dcC′, where V_dc represents the scan rate, C′ is the scaled capacitance and V_dcC′ = J_CAP with an assumption of J → 0, so that J_r ≈ J_CAP and voltage independent capacitance is J_CAP = CdV/dt = C.⁴⁴ The linear variation of J_CAP vs. scan rate in dark and illumination is shown in Fig. 4(C and F) with an almost voltage independent current density, which exhibits a constant capacitance value of 4.8 and 8.3 nF in dark and illumination conditions, respectively. The obtained higher value of capacitance under illumination conditions support the findings of Almora et al. and Zarazua et al.,^50,52 and were calculated for the planar and mesoporous perovskite solar cell in dark and light illumination using the J–V curve method (employed in the present study) and capacitance vs. frequency method, respectively. The increase in the value of capacitance under light conditions was due to the accumulation of ionic defects at the interfaces.⁵⁰ The accumulation of charge carriers practically lowers the responsivity of the photodetector. The authors have explained the phenomena of charge carrier accumulation under illumination in terms of the band bending as follows: at equilibrium, a photovoltage is established at the ETL/perovskite interface with an increase in the photo-generated charge carriers moving the depletion region towards the flat band condition and the progressive uplift of the depletion region causes the creation of an accumulation layer.⁵⁰ From the above discussion, it has been concluded that, in the fabricated photodetector, the transient decay observed in the photodynamic response is due to the charge accumulation at the ITO/perovskite contact. The device or structural engineering that effectively suppress these effects are extremely important for future research directions. Moreover, the observed hysteresis response in our device under illumination and dark conditions are mainly due to the combined effect of ion migration and surface charge trapping. Edmands and Yu et al. have explained this phenomenon in detail using experimental data and theoretical calculations for a similar structured system to the present case and for planar and mesoporous structured solar cells.^49,53 These authors have shown that the accumulation of I_i and V_ṀA at the ETL/perovskite interface would restrict the extraction of the charge carriers, while the accumulation or presence of V_I close to the ETL increases the transport mechanism by band bending. In equilibrium, a Schottky contact formed between the ITO/perovskite interface leads to the formation of an electric field pointing towards the bulk of perovskite due to the difference in concentration. When the applied bias increased with the higher potential value from the equilibrium state, the space charge region compresses and the interfacial electric field reduces, which allows low current to flow across the junction. During the process, the slow detraining and redistribution of iodine ions inside the perovskite bulk could increase the conductivity of the material which in turn enhance the net current. In a similar way, when the applied bias decreases towards a lower potential value from the maximum potential, the space charge widens and the interfacial electric field increases. At the same time, iodine ions move towards the interface to the region where they may also trap at the ITO/perovskite interface, which would facilitate the higher current.

Conclusions

A perovskite semiconductor was synthesized using a low-cost spin coating technique for its application in photodiodes. The fabricated photodiode showed a zero bias and broad wavelength photo-detection with an impressive photo-responsivity that is comparable with efficient organic and inorganic-based photodiodes. The fabricated device exhibits a knee voltage of 0.78 V, rectification ratio of 8 at ±1.3 V and reverse saturation current (J₀) of 1.5 × 10⁻⁸ A. The net current in the device was limited through a recombination/transport mechanism in the perovskite layer. This hypothesis was further confirmed by the √J vs. V plot, where a linear characteristic in the high forward bias region signified domination by the space charge leakage current. The observed hysteresis response in our device under illumination and dark conditions were mainly due to the combined effect of the ion migration and surface charge trapping. The device or structural engineering that can effectively suppress these effects are extremely important for future research directions.

References

1. F. Wang, J. Mei, Y. Wang, L. Zhang, H. Zhao and D. Zhao, ACS Appl. Mater. Interfaces, 2016, 8, 2840–2846.
2. H. Wu, Y. Sun, D. Lin, R. Zhang, C. Zhang and W. Pan, Adv. Mater., 2009, 21, 227–231.
3. M. Chen, L. Hu, J. Xu, M. Liao, L. Wu and X. Fang, Small, 2011, 7, 2449–2453.
4. L. Wang, J. Jie, Z. Shao, Q. Zhang, X. Zhang, Y. Wang, Z. Sun and S.-T. Lee, Adv. Funct. Mater., 2015, 25, 2910–2919.
5. H. Lu, W. Tian, F. Cao, Y. Ma, B. Gu and L. Li, Adv. Funct. Mater., 2016, 26, 1296–1302.
6. X. Wang, B. Liu, R. Liu, Q. Wang, X. Hou, D. Chen, R. Wang and G. Shen, Angew. Chem., Int. Ed., 2014, 53, 1849–1853.
7. Q. Yang, Y. Liu, Z. Li, Z. Yang, X. Wang and Z. L. Wang, Angew. Chem., 2012, 124, 6549–6552.
8. Y. Zhang, J. Du, X. Wu, G. Zhang, Y. Chu, D. Liu, Y. Zhao, Z. Liang and J. Huang, ACS Appl. Mater. Interfaces, 2015, 7, 21634–21638.
9. A. Osinsky, S. Gangopadhyay, R. Gaska, B. Williams, M. A. Khan, D. Kuksenkov and H. Temkin, Appl. Phys. Lett., 1997, 71, 2334–2336.
10. Z. Wang, M. Safdar, C. Jiang and J. He, Nano Lett., 2012, 12, 4715–4721.
11. K. Das, S. Mukherjee, S. Manna, S. K. Ray and A. K. Raychaudhuri, Nanoscale, 2014, 6, 11232–11239.
12. Q. Hong, Y. Cao, J. Xu, H. Lu, J. He and J.-L. Sun, ACS Appl. Mater. Interfaces, 2014, 6, 20887–20894.
13. M. Schreier, L. Curvat, F. Giordano, L. Steier, A. Abate, S. M. Zakeeruddin, J. Luo, M. T. Mayer and M. Gratzel, Nat. Commun., 2015, 6, 7326.
14. Y. Zhao and K. Zhu, J. Phys. Chem. C, 2014, 118, 9412–9418.
15. M. M. Lee, J. Teuscher, T. Miyasaka, T. N. Murakami and H. J. Snaith, Science, 2012, 338, 643–647.
16. D. Shi, V. Adinolfi, R. Comin, M. Yuan, E. Alarousu, A. Buin, Y. Chen, S. Hoogland, A. Rothenberger, K. Katsiev, Y. Losovyj, X. Zhang, P. A. Dowben, O. F. Mohammed, E. H. Sargent and O. M. Bakr, Science, 2015, 347, 519–522.
17. Y. Yang, M. Yang, Z. Li, R. Crisp, K. Zhu and M. C. Beard, J. Phys. Chem. Lett., 2015, 6, 4688–4692.
18. F. Lin, S.-W. Chen, J. Meng, G. Tse, X.-W. Fu, F.-J. Xu, B. Shen, Z.-M. Liao and D.-P. Yu, Appl. Phys. Lett., 2014, 105, 073103.
19. L. Rigutti, M. Tchernycheva, A. De Luna Bugallo, G. Jacopin, F. H. Julien, L. F. Zagonel, K. March, O. Stephan, M. Kociak and R. Songmuang, Nano Lett., 2010, 10, 2939–2943.
20. C. Soci, A. Zhang, B. Xiang, S. A. Dayeh, D. P. R. Aplin, J. Park, X. Y. Bao, Y. H. Lo and D. Wang, Nano Lett., 2007, 7, 1003–1009.
21. S. R. Ali Raza, Y. T. Lee, S. H. Hosseini Shokouh, R. Ha, H.-J. Choi and S. Im, Nanoscale, 2113, 5, 10829–10834.
22. J. Kim, S.-H. Lee, J. H. Lee and K.-H. Hong, J. Phys. Chem. Lett., 2014, 5, 1312–1317.
23. W.-J. Yin, T. Shi and Y. Yan, Appl. Phys. Lett., 2014, 104, 063903.
24. X. Hu, X. Zhang, L. Liang, J. Bao, S. Li, W. Yang and Y. Xie, Adv. Funct. Mater., 2014, 24, 7373–7380.
25. R. Dong, Y. Fang, J. Chae, J. Dai, Z. Xiao, Q. Dong, Y. Yuan, A. Centrone, X. C. Zeng and J. Huang, Adv. Mater., 2015, 27, 1912–1918.
26. L. Dou, Y. Yang, J. You, Z. Hong, W.-H. Chang, G. Li and Y. Yang, Nat. Commun., 2014, 5, 6404.
27. H. Zhou, Z. Song, P. Tao, H. Lei, P. Gui, J. Mei, H. Wang and G. Fang, RSC Adv., 2016, 6, 6205–6208.
28. J. Yu, X. Chen, Y. Wang, H. Zhou, M. Xue, Y. Xu, Z. Li, C. Ye, J. Zhang, P. A. Van Aken, P. D. Lund and H. Wang, J. Mater. Chem. C, 2016, 4, 7302–7308.
29. Y. Lee, J. Kwon, E. Hwang, C.-H. Ra, W. J. Yoo, J.-H. Ahn, J. H. Park and J. H. Cho, Adv. Mater., 2015, 27, 41–46.
30. Y. Zhang, J. Du, X. Wu, G. Zhang, Y. Chu, D. Liu, Y. Zhao, Z. Liang and J. Huang, ACS Appl. Mater. Interfaces, 2015, 7, 21634–21638.
31. S. Chen, C. Teng, M. Zhang, Y. Li, D. Xie and G. Shi, Adv. Mater., 2016, 28, 5969–5974.
32. B. Tripathi, P. Bhatt, P. Chandra Kanth, P. Yadav, B. Desai, M. Kumar Pandey and M. Kumar, Sol. Energy Mater. Sol. Cells, 2015, 132, 615–622.
33. P. Bhatt, K. Pandey, P. Yadav, B. Tripathi, C. K. P. Kanth, M. K. Pandey and M. Kumar, Sol. Energy Mater. Sol. Cells, 2015, 140, 320–327.
34. P. Yadav, K. Pandey, P. Bhatt, D. Raval, B. Tripathi, C. P. Kanth, M. K. Pandey and M. Kumar, Sol. Energy, 2015, 122, 773–782.
35. C. Wehrenfennig, G. E. Eperon, M. B. Johnston, H. J. Snaith and L. M. Herz, Adv. Mater., 2014, 26, 1584–1589.
36. A. Marchioro, J. Teuscher, D. Friedrich, M. Kunst, R. van de Krol, T. Moehl, M. Gratzel and J.-E. Moser, Nat. Photonics, 2014, 8, 250–255.
37. B. Murali, M. I. Saidaminov, A. L. Abdelhady, W. Peng, J. Liu, J. Pan, O. M. Bakr and O. F. Mohammed, J. Mater. Chem. C, 2016, 4, 2545–2552.
38. M. Ibrahim Dar, M. Abdi-Jalebi, N. Arora, T. Moehl, M. Grätzel and M. K. Nazeeruddin, Adv. Mater., 2015, 27, 7221–7228.
39. Y. Wang, T. Gould, J. F. Dobson, H. Zhang, H. Yang, X. Yao and H. Zhao, Phys. Chem. Chem. Phys., 2014, 16, 1424–1429.
40. W. Geng, L. Zhang, Y.-N. Zhang, W.-M. Lau and L.-M. Liu, J. Phys. Chem. C, 2014, 118, 19565–19571.
41. A. Amat, E. Mosconi, E. Ronca, C. Quarti, P. Umari, M. K. Nazeeruddin, M. Grätzel and F. De Angelis, Nano Lett., 2014, 14, 3608–3616.
42. S. Agarwal and P. R. Nair, Appl. Phys. Lett., 2015, 107, 123901.
43. S. Agarwal, M. Seetharaman, N. K. Kumawat, A. S. Subbiah, S. K. Sarkar, D. Kabra, M. A. G. Namboothiry and P. R. Nair, J. Phys. Chem. Lett., 2014, 5, 4115–4121.
44. S. M. Sze and K. K. Ng, Physics of semiconductor devices, John wiley & sons, 2006.
45. R. W. Mott and N. F. Gurney, Electronic Processes in Ionic Crystals, Clarendon Press, Oxford, U.K., 1948.
46. S. D. Stranks, G. E. Eperon, G. Grancini, C. Menelaou, M. J. P. Alcocer, T. Leijtens, L. M. Herz, A. Petrozza and H. J. Snaith, Science, 2013, 342, 341–344.
47. Y. Yamada, T. Nakamura, M. Endo, A. Wakamiya and Y. Kanemitsu, J. Am. Chem. Soc., 2014, 136, 11610–11613.
48. H.-W. Chen, N. Sakai, M. Ikegami and T. Miyasaka, J. Phys. Chem. Lett., 2015, 6, 164–169.
49. H. Yu, H. Lu, F. Xie, S. Zhou and N. Zhao, Adv. Funct. Mater., 2016, 26, 1411–1419.
50. O. Almora, I. Zarazua, E. Mas-Marza, I. Mora-Sero, J. Bisquert and G. Garcia-Belmonte, J. Phys. Chem. Lett., 2015, 6, 1645–1652.
51. F. Fabregat-Santiago, I. Mora-Seró, G. Garcia-Belmonte and J. Bisquert, J. Phys. Chem. B, 2003, 107, 758–768.
52. I. Zarazua, J. Bisquert and G. Garcia-Belmonte, J. Phys. Chem. Lett., 2016, 7, 525–528.
53. J. Beilsten-Edmands, G. E. Eperon, R. D. Johnson, H. J. Snaith and P. G. Radaelli, Appl. Phys. Lett., 2015, 106, 173502.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra23668e

---