Sacrificial additive-assisted film growth endows self-powered CsPbBr₃ photodetectors with ultra-low dark current and high sensitivity

Abstract

Inorganic halide perovskite CsPbBr₃ is promising for stable, high-performance self-powered photodetectors (PDs). However, solution-processed polycrystalline CsPbBr₃ films generally exhibit inferior photodetection performance in contrast to single-crystal and micro-/nano-structured counterparts, primarily because of their low carrier mobility and high defect density. We propose herein a sacrificial additive-assisted CsPbBr₃ film growth strategy, in which a 2-phenylethanamine iodide (PEAI) additive was firstly introduced into a CsPbBr₃ precursor film and then was extracted during CsPbBr₃ film crystallization by high-temperature annealing. The as-obtained CsPbBr₃ film exhibits multiple advantages of full coverage, pure phase, micro-sized grains, higher crystallinity, fewer defects, and particularly a CsBr-rich surface with less conductivity, compared with the control film prepared without PEAI sacrificial additive. The resulting self-powered PD with a simple configuration of FTO/TiO₂/CsPbBr₃/carbon yields an ultra-low dark current of 8.05 × 10⁻¹¹ A cm⁻², a high R of 0.35 A W⁻¹, a superior D* of 3.83 × 10¹³ Jones, and a fast response time of 1.46 μs. These figures of merit are far beyond those of the one prepared with the control film and even most of self-powered CsPbBr₃ PDs reported previously. Furthermore, the optimized PD exhibits excellent operational reliability and long-term stability in an ambient air atmosphere.

---

1. Introduction

Organic–inorganic hybrid perovskites are demonstrated with many appealing optoelectronic properties combined with low-temperature solution processibility,^1–4 which makes them promising for high-performance, low-cost solar cells which are known as perovskite solar cells.^5,6 Over the past decade, the efficiency of perovskite solar cells has soared from 3.8% to the certified 25.2%.^7,8 Such success motivates the side-by-side use of hybrid perovskites for other optoelectronic devices, such as photodetectors (PDs).^9–12 PDs that could convert light signals into electrical ones are fundamental and essential devices in various fields, including optical communication, automatic control, imaging, etc.¹³ Since PDs usually need to transduce signals repeatedly over long periods, their operational stability and reliability become vital particularly.^13–15 But the intrinsic instability of hybrid perovskites against oxygen, heat, and/or moisture brings severe stability and reliability obstacles to the resulting PDs.^16,17

In comparison with hybrid perovskites, CsPbBr₃, typically as the inorganic counterpart, possesses upgraded stability besides its superb characteristics inherited from the perovskite structure, hence drawing ever-increasing interest in the field of PDs.^18–32 In this case, both metal–semiconductor–metal and photodiode-type CsPbBr₃ PDs have been explored. The latter ones hold the peculiarity of self-powered functionality because of the formation of the built-in electric field, which makes them independent of the external power supply.^33–35 Meanwhile, they could offer a higher sensitivity and a faster response speed.^14,27 Hence, tremendous effort is devoted to building photodiode-type ones, namely self-powered CsPbBr₃ PDs. Very recently, there has been significant progress in self-powered PDs based on CsPbBr₃ single crystals and micro-/nano-structures.^{18,23,27,29,30,34–40} For instance, M. Bakr et al.²³ demonstrated the one with a low-temperature, solution-processed CsPbBr₃ single crystal, which yielded a high on/off ratio of 10⁵. Fang et al.³⁰ reported a CsPbrB₃ single-crystal/CuI heterojunction-based self-powered PD with a good wavelength selectivity of 525–565 nm, a high photocurrent of ∼100 nA, a large on/off ratio of 1.5 × 10³, and a fast response speed of 0.04/2.96 ms. Yan et al.^27,38,40 developed various PDs based on CsPbBr₃ micro-/nano-structures, such as microcrystals and nanowires, which delivered a superior responsivity (R) of over 0.3 A W⁻¹, on/off ratios of up to 10⁶, a specific detectivity (D*) of 10¹³, and a linear dynamic range (LDR) as high as 123.5 dB. Besides, Zhao et al.³⁵ fabricated a self-powered PD based on the sandwich structure of GaN/CsPbBr₃ microcrystals/ZnO, in which a D* of 10¹⁴ Jones, an on/off ratio of 10⁵, and a R peak of 89.5 mA W⁻¹ were realized.

Compared with CsPbBr₃ single crystals and micro-/nano-structures, solution-processed polycrystalline CsPbBr₃ films retain outstanding advantages in terms of preparation simplicity, controllability, and adjustability,^31,32,41 which endow them with great potential for large-scale industrialization in future optoelectronic devices. However, as for self-powered PDs based on solution-processed CsPbBr₃ polycrystalline films, their performance parameters are far behind those fabricated with CsPbBr₃ single crystals and micro-/nano-structures, even though some innovative interfacial modification strategies and heterojunction technologies are developed (Al₂O₃ and PEIE).^14,36 Such a dilemma is primarily caused by inferior carrier mobility and high-density defects of polycrystalline CsPbBr₃ films, because of low solubility of the CsBr component in common solvents.^31,32,42,43 Therefore, it is of great significance to improve the electronic quality of solution-processed polycrystalline CsPbBr₃ films to promote further development of self-powered PDs.

In this work, we report a sacrificial additive-assisted CsPbBr₃ film growth strategy for self-powered PDs, wherein the 2-phenylethanamine iodide (PEAI) additive was firstly introduced into the CsPbBr₃ precursor film and then was extracted during its crystallization at a high temperature of 300 °C. Such a strategy endows the full-coverage, pure-phased CsPbBr₃ film with micro-sized grains, high-crystallinity, greatly reduced defects, and particularly a CsBr-rich surface that has a relatively low conductivity. The resultant self-powered PDs with a simple configuration of FTO/TiO₂/CsPbBr₃/carbon exhibit an ultralow dark current of 8.05 × 10⁻¹¹ A cm⁻², a superior peak R of 0.35 A W⁻¹, an outstanding peak D* of 3.83 × 10¹³ Jones, and a fast response time of 1.46 μs, along with excellent operational reliability and long-term stability in ambient air atmosphere. These performance parameters are much better than the one prepared without PEAI sacrificial additive, and even beyond most of self-powered CsPbBr₃ PDs reported previously. In addition, the mechanism that accounts for the superior performance of self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive is explored preliminarily.

2. Results and discussion

The main steps in preparing the CsPbBr₃ film by a sacrificial additive-assisted strategy are illustrated in Fig. 1(a). Firstly, PEAI species was loaded on the FTO/TiO₂ substrate by spin-coating PEAI methanol solution. Then, the CsPbBr₃ precursor film was deposited via an intermediate phase halide exchange method, as demonstrated previously.³¹ During this process, the PEAI species on the FTO/TiO₂ substrate could be dissolved in dimethyl sulfoxide (DMSO) solvent of the CsPbBr₃ precursor and they deposit on the CsPbBr₃ precursor film, because of the soluble nature of the PEAI molecules in DMSO.⁴⁴ Finally, the sample was annealed at 300 °C for 30 min to promote CsPbBr₃ film crystallization, during which the escape of the PEAI species occurs simultaneously, as discussed below. The detailed preparation procedures can be found in the Experimental section.

Fig. 1 (a) Schematic illustration of the experimental procedures in the PEAI sacrificial additive-assisted CsPbBr₃ film growth strategy. (b) XPS survey spectra and (c) core-level N 1s XPS spectra of the FTO/TiO₂ substrate with loading of PEAI species, the CsPbBr₃ film prepared without and with PEAI species, and the dissolved CsPbBr₃ film formed with PEAI species, respectively. The N 1s peak in panel (b) is marked by a red dotted circle.

X-ray photoelectron spectroscopy (XPS) measurements were conducted to investigate the FTO/TiO₂ substrate with loading of PEAI species. As shown in Fig. 1(b) and (c), apart from those of C, Ti, and O elements, an extra peak located at 401.8 eV was detected, which can be assigned to N 1s of PEAI species.^45,46 This result verifies the successful loading of PEAI species onto the FTO/TiO₂ substrate. Meanwhile, the compositions of CsPbBr₃ films synthesized without and with PEAI species were investigated. Clear XPS peaks corresponding to Cs, Pb, and Br elements were detected for both films, revealing that they are composed of CsPbBr₃ materials.^47,48 Interestingly, there is no detectable signal that relates to PEAI species for the CsPbBr₃ film deposited on the FTO/TiO₂ substrate with loading of PEAI species, similar to that of the control sample without loading of PEAI species. This fact is also indicated by the core-level N 1s XPS spectra in Fig. 1(c), because the typical core-level N 1s characteristic peak of PEAI species is absent for both CsPbBr₃ films.

It is noted that XPS can only detect the surficial composition information of the samples. To probe the bulk composition of the CsPbBr₃ film deposited on the FTO/TiO₂ substrate with loading of PEAI species, the film was firstly dissolved in DMSO solvent and then was annealed at 100 °C for 5 min to evaporate the solvent. As shown in Fig. 1(b), the additional Ti 2p peaks that come from the FTO/TiO₂ substrate appear for the sample after DMSO treatment, disclosing the full dissolution of the CsPbBr₃ film and therefore the elimination of the possible non-uniform composition distribution in the film or at its interface. In this case, XPS technology can study the bulk composition of the CsPbBr₃ film reasonably. Once again, as shown in Fig. 1(c), the N 1s peak is still absent for the dissolved CsPbBr₃ film deposited on the FTO/TiO₂ substrate containing PEAI species, further indicating that there is indeed no detectable PEAI species in the resulting CsPbBr₃ film. We attribute such an interesting fact to the decomposition of PEAI species during CsPbBr₃ film crystallization under high-temperature annealing conditions at 300 °C, since PEAI powder has a decomposition temperature of ∼280 °C in a N₂ atmosphere, as demonstrated by Ma et al.⁴⁹ This is also supported by the thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) results of PEAI powder in Fig. S1(a) and (b) (ESI†). It can be seen that PEAI powder undergoes 100% weight loss in one step with an onset temperature around 200 °C. Moreover, the TGA curve indicates that the typical decomposition temperature of the PEAI powder is about ∼282 °C. Overall, one can make sense that PEAI species are incorporated before CsPbBr₃ film crystallization and then are extracted after its crystallization. They act as a sacrificial additive during CsPbBr₃ film growth. Hereafter, the PEAI species are defined as a sacrificial additive to clarify our discussion.

Next, the fundamental physical properties of CsPbBr₃ films prepared without and with PEAI sacrificial additive are investigated. Fig. 2(a) shows the XRD results, wherein all the diffraction peaks well coincide those of CsPbBr₃ materials with a cubic perovskite phase,^32,47 revealing the pure phase of both CsPbBr₃ films. By comparison, the diffraction peak intensities of the (100) and (200) crystal planes are somewhat stronger for the CsPbBr₃ film prepared with PEAI sacrificial additive, indicating its better crystallinity with fewer inter- and/or intra-grains defects. Fig. 2(b) shows the ultraviolet-visible (UV-vis) absorption spectra of the concerned CsPbBr₃ films. Both films exhibit a similar absorption onset of ∼540 nm and an exciton absorption resonance at ∼530 nm, which agree with those of the CsPbBr₃ films reported previously.^26,43,50 The slightly higher absorption can be identified for the CsPbBr₃ film prepared with PEAI sacrificial additive, which is probably related to its higher crystallinity and/or different surficial morphology (as discussed below). Fig. 2(c) shows the time-resolved photoluminescence (TRPL) results of the obtained CsPbBr₃ films on insulating glass substrates. The average lifetime (τ_ave) values of the minority carriers are estimated to be 3.12 and 5.82 ns for the CsPbBr₃ films prepared without and with PEAI sacrificial additive, respectively. The longer τ_ave for the latter is indicative of the weaker non-radiative recombination of charge carriers, because of fewer defects in it. This inference is further supported by the space-charge-limited current (SCLC) analysis of hole-only dark current density versus voltage (J–V) curves of the prepared CsPbBr₃ films,^51,52 as shown in Fig. 2(d). Obviously, the CsPbBr₃ film synthesized with PEAI sacrificial additive exhibits a much smaller trap-filled limit voltage (V_TFL) of 0.67 V, compared with the control sample with a V_TFL of 1.02 V. In general, V_TFL is connected with trap density (N_t) by the relationship of V_TFL = eN_tL²/(2εε₀), where ε and ε₀ are the relative dielectric constant of CsPbBr₃ and vacuum permittivity, L is the thickness of the CsPbBr₃ film, and e is the elementary charge of the electron, respectively. Therefore, the much smaller hole trap density can be identified for the CsPbBr₃ film prepared with PEAI sacrificial additive, which suggests substantially fewer defects in it in turn. In addition, one can see that the CsPbBr₃ film prepared with PEAI sacrificial additive exhibits lower J values under bias voltages below 2.5 V, indicative of its lower conductivity.

Fig. 2 (a) XRD patterns, (b) UV-vis absorption spectra, (c) TRPL curves, and (d) SCLC analysis of J–V curves for hole-only CsPbBr₃ devices prepared without and with PEAI sacrificial additive, respectively. The diamond in panel (a) indicates the signal from the FTO/TiO₂ substrate.

A scanning electron microscope (SEM) and an atomic force microscope (AFM) were employed to analyze the surficial morphology of the as-prepared CsPbBr₃ films. As shown in Fig. 3(a) and (b) and Fig. S2(a) and (b) (ESI†), both films have full surface coverage, regardless of being prepared without and with PEAI sacrificial additive. Moreover, they consist of micro-sized crystal grains with trip-junction grain boundaries and similar average sizes of ∼0.75 μm (Fig. S3, ESI†). However, there seems to be some extra inclusions in the adjoining grains of the CsPbBr₃ film formed with PEAI sacrificial additive, since they show relatively bright contrast, possibly because of their less conductive nature.^53–55 As with the SEM results, the AFM images in Fig. 3(c) and (d) show similar characters of the CsPbBr₃ films. Meanwhile, the root-mean-square (RMS) roughness values were estimated to be 24.3 and 20.6 nm for the CsPbBr₃ film prepared without and with PEAI sacrificial additive, respectively. That is to say, the less-conductive inclusions in the CsPbBr₃ film prepared with PEAI sacrificial additive do not increase the film's surface roughness, but decrease it. Such a result further indicates that the inclusions cover conformally on the CsPbBr₃ grains.

Fig. 3 (a and b) SEM and (c and d) AFM images of CsPbBr₃ films prepared (a and c) without and (b and d) with PEAI sacrificial additive, respectively.

Since XRD does not detect any impurity phase in the CsPbBr₃ film prepared with PEAI sacrificial additive, the XPS test is further adopted to gain insight into the less-conductive inclusions observed in Fig. 3(b) and (d). The control sample is also investigated for comparison. All the XPS results are calibrated with the C 1s peak with a binding energy of 284.8 eV, as indicated in Fig. 4(a). Fig. 4(b)–(d) show the corresponding core-level XPS spectra of Cs 3d, Pb 4f, and Br 3d, respectively. These XPS peaks are basically in accordance with previously reported results of CsPbBr₃ materials.^47,48 By contrast, they shift slightly towards lower binding energies for the CsPbBr₃ film prepared with PEAI sacrificial additive, which is probably caused by the less-conductivity inclusions in it. In addition, the atomic ratios of Cs and Br atoms relative to Pb atoms for both CsPbBr₃ films are calculated, as shown in Fig. 4(e). The CsPbBr₃ film prepared without PEAI sacrificial additive exhibits Cs : Pb : Br atomic ratios of 0.98 : 1.00 : 2.75. However the one formed with PEAI sacrificial additive possesses values of 1.45 : 1.00 : 3.36. These results reveal that there are certain CsBr-rich CsPbBr₃ crystals in the surface region of the CsPbBr₃ film prepared with PEAI sacrificial additive. Therefore, the less-conductive inclusions in the adjoining grains of the CsPbBr₃ film are ascribed to CsBr-rich CsPbBr₃, as schematically shown in Fig. 4(f). As a whole, the above investigations reveal that PEAI sacrificial additive can induce the formation of the desired CsPbBr₃ film with full coverage, pure phase, micro-sized grains, higher crystallinity, fewer defects, and particularly a CsBr-rich surface with lower conductivity.

Fig. 4 (a) C 1s, (b) Cs 3d, (c) Pb 4f, and (d) Br 3d core-level XPS spectra for CsPbBr₃ films prepared without and with PEAI sacrificial additive, respectively. (e) Atomic ratios of Cs and Br relative to Pb derived from the XPS spectra for the same CsPbBr₃ films, respectively. (f) Schematic illustration of the surficial composition difference between CsPbBr₃ films prepared without and with PEAI sacrificial additive, respectively.

Furthermore, CsPbBr₃ films prepared without and with PEAI sacrificial additive are used to build self-powered PDs with the layer stacking of FTO/TiO₂/CsPbBr₃/carbon. This configuration could enable simplified fabrication procedures and improved stability of the PDs, along with their substantially reduced cost, due to avoiding the use of high-cost, unstable hole transport materials and noble metal electrodes.^50,51,56Fig. 5(a) shows the cross-sectional SEM images of typical PDs prepared without and with PEAI sacrificial additive, from which each functional layer can be identified. In particular, the thicknesses of the CsPbBr₃ films are estimated to be ∼410 nm similarly. Moreover, one can see that nearly all the individual grains in each CsPbBr₃ film can cross the entire film and form vertically oriented grain boundaries, wherein the interfacial voids may be induced by air bubbles in the precursor solution. Fig. 5(b) shows the working mechanism of self-powered CsPbBr₃ PD prepared with PEAI sacrificial additive, for which the band energy of CsPbBr₃ is determined by the ultraviolet photoelectron spectroscopy (UPS) spectrum in Fig. S4 (ESI†), considering its bandgap of 2.3 eV. Upon light illumination, photo-generated electrons are extracted by the TiO₂ layer and then are collected by the FTO electrode, while photo-generated holes are extracted and collected by the carbon electrode, due to the favored energy alignments. Thus, the optical signals are converted into electrical signals. It is noted that the PEAI solution at different concentrations of 5, 10, and 15 mg mL⁻¹ are employed in this section, aiming to optimize the performance of PDs fabricated with PEAI sacrificial additive.

Fig. 5 (a) Cross-sectional SEM images of self-powered CsPbBr₃ PDs with the configuration of FTO/TiO₂/CsPbBr₃/carbon fabricated without and with PEAI sacrificial additive, respectively. (b) Schematic working mechanism of self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive. (c) Dark J–V curves, (d) light J–V curves, (E) EQE spectra, (f) R plots, (g) D* curves, and (h) TPC curves for self-powered CsPbBr₃ PDs fabricated without and with different concentrations of PEAI solution, respectively.

Fig. 5(c) provides dark J–V curves of all the fabricated CsPbBr₃ PDs. The PDs prepared with PEAI sacrificial additive show much smaller dark current densities, in particular the one prepared with 10 mg mL⁻¹ PEAI solution. For example, at 0.5 V bias voltage, the PD prepared without PEAI sacrificial additive exhibits a dark current density of 1.40 × 10⁻⁴ A cm⁻², while the value is reduced significantly to 1.87 × 10⁻⁶ A cm⁻² for the one fabricated with PEAI sacrificial additive (10 mg mL⁻¹). Meanwhile, the latter exhibits a minimum dark current density of 8.05 × 10⁻¹¹ A cm⁻², which is lower than almost all of previously reported self-powered CsPbBr₃ PDs.^{14,18,23,27,31,36,38,40} Such a low dark current can be primarily ascribed to the unique CsBr-rich surface for CsPbBr₃ prepared with PEAI sacrificial additive, which has low conductivity. A lower dark current is generally coupled with a smaller shot noise and a lower thermal noise, which could endow a PD with higher sensitivity.^14,31 Under illumination and a bias voltage of 0 V, all the PDs yield greatly enlarged current densities (Fig. 5(d)), indicating their functionality of self-powered photoresponse. The open-circuit voltage (V_oc) for self-powered CsPbBr₃ PD fabricated without PEAI sacrificial additive is estimated to be 1.00 V. By contrast, such values are boosted greatly to 1.12, 1.24, and 1.16 V for the ones obtained with PEAI sacrificial additive, when the concentrations of the PEAI solution are 5, 10, and 15 mg mL⁻¹, respectively. The improved V_oc indicates the recued non-radiative recombination of charge carriers in the self-powered CsPbBr₃ PDs.^51,57Fig. 5(e) shows the external quantum efficiency (EQE) spectra of all the fabricated PDs. Their photocurrent generation starts at ∼540 nm, disclosing that CsPbBr₃ films govern their photoelectric conversion. Furthermore, the R spectra and D* curves of all the PDs obtained without and with PEAI sacrificial additive are calculated and provided in Fig. 5(f) and (g). There is an obvious fact that self-powered CsPbBr₃ PDs prepared with PEAI sacrificial additive exhibit much higher R and D* values in contrast to the one fabricated without that. Moreover, when the concentration of the PEAI solution is optimized to 10 mg mL⁻¹, the PD yields the best R and D* values with a maximum R of 0.35 A W⁻¹ and a D* of 3.83 × 10¹³ Jones, respectively. Fig. 5(h) shows the transient photocurrent (TPC) curves for all the prepared self-powered CsPbBr₃ PDs. Their response time values can be obtained by fitting the curves with a single-exponential decay function of time.^14,58 By this way, the response time of the PD fabricated without PEAI sacrificial additive is estimated to be 2.74 μs. The values decrease to 2.20, 1.46, and 1.59 μs for the ones fabricated with 5, 10, and 15 mg mL⁻¹ PEAI solution, respectively. Therefore, the fast response of self-powered PDs fabricated with PEAI sacrificial additive can be identified, particularly when 10 mg mL⁻¹ PEAI solution is adopted. Overall, the PEAI sacrificial additive-assisted film growth strategy can endow self-powered CsPbBr₃ PD with a minimum dark current density of 8.05 × 10⁻¹¹ A cm⁻², a maximum R of 0.35 A W⁻¹, a D* peak of 3.83 × 10¹³ Jones, and a response time of 1.46 μs, most of which are better than self-powered CsPbBr₃ PDs reported previously,^{14,18,23,27,29–31,34–36,38,40} as summarized in Table 1.

Table 1 Summary of the performance parameters for self-powered CsPbBr₃ PDs reported currently

PD configuration	Dark current @ bias	R (A W^−1)	D* (Jones)	Falling time	Ref.
FTO/TiO2/CsPbBr3/carbon	8.05 × 10^−11 A cm^−2 @ 0 V	0.35	3.83 × 10^13	1.46 μs	This work
ITO/CsPbBr3:ZnO/Ag	0.240 A cm^−2 @ 0 V	0.0115	—	0.0179 s	18
FTO/ZnO/CsPbBr3/MoO3/Au	—	0.3	1.15 × 10^13	500 ms	34
FTO/Al2O3/CsPbBr3/TiO2/Au	10 nA	0.44	1.88 × 10^13	270 μs	36
FTO/TiO2/CsPbBr3/carbon	8.7 × 10^−8 A cm^−2 @ −0.3 V	0.35	1.94 × 10^13	0.58 μs	31
ITO/PTAA + PEIE/CsPbBr3/PCBM/BCP/Ag	4.8 × 10^−8 A cm^−2 @ 0.3 V	0.13	6.0 × 10^12	62 ns	14
Au/CsPbBr3/Pt	10 pA @ 0 V	0.028	1.7 × 10^11	60 ms	23
Ag/CsPbBr3/CuI/Ag	—	0.0014	6.2 × 10^10	2.96 ms	30
ITO/SnO2/CsPbBr3/Spiro-OMeTAD/Au	<10 nA	0.172	4.8 × 10^12	0.12 ms	40
ITO/SnO2/CsPbBr3/PTAA/Au	2.5 × 10^−9 A cm^−2	0.206	7.23 × 10^12	39 μs	38
GaN/CsPbBr3/ZnO	1 × 10^−9 A cm^−2	0.0895	1.03 × 10^14	140 μs	35
ITO/SnO2/PMMA + CsPbBr3/PTAA/Au	4.0 × 10^−9 A cm^−2 @ 0 V	0.3	1 × 10^13	0.43 ms	27
In/SnO2/CsPbBr3/In	2 pA	0.01	2.6 × 10^11	1.94 ms	29

---

In addition, carrier dynamics in self-powered CsPbBr₃ PDs fabricated without and with PEAI sacrificial additive are investigated to understand the better performance of the latter in depth. Fig. 6(a) shows the re-plotted Mott–Schottky (M–S) results of two typical PDs, from which the built-in potential (V_bi) values are estimated to be 1.38 and 1.51 V for the PDs fabricated without and with PEAI sacrificial additive, respectively.^51,56,59 The larger V_bi could promote more effective extraction and separation of photo-generated carriers and thus reduces the carrier recombination in the latter one. Fig. 6(b) shows the corresponding Nyquist plots under a bias voltage of 1 V and dark conditions. The obvious arc is ascribed to the recombination of photo-generated carriers, reflecting the recombination resistance (R_rec), as indicated by the equivalent circuit model in Fig. S5 (ESI†).^51,56,59 The PD fabricated with PEAI sacrificial additive yields a larger R_rec than that without PEAI, indicative of a weaker carrier recombination. Therefore, we can infer that the superior performance of the self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive is largely related to the reduced recombination of photo-generated carriers in it, which originally benefits from the desired features of the CsPbBr₃ film, in particular its higher crystallinity, fewer defects, and a CsBr-rich surface with lower conductivity.

Fig. 6 (a) C⁻²–V results under dark conditions and (b) Nyquist plots for the optimized self-powered CsPbBr₃ PD fabricated without and with PEAI sacrificial additive, respectively.

Finally, the operational reliability and long-term stability of the optimized self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive are investigated primarily, which are carried out on an unencapsulated device in ambient air atmosphere with a relative humidity (RH) of 40–50% and a temperature of ∼25 °C. Fig. 7(a) shows its response to a 405 nm on/off modulated laser. One can see that both the dark current and the photocurrent show little variations over the whole test period of ∼5 h, indicating the excellent operational reliability of self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive. Fig. 7(b) shows time-dependent variations of normalized photocurrents for the same PD. One can make sense that it delivers a stable photocurrent over ∼1000 h. About 96% of its initial photocurrent can be maintained after this test period. The above results demonstrate the excellent operational reliability and long-term stability of the self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive.

Fig. 7 (a) Response of the optimized self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive to a 405 nm on/off modulated laser in a test period of ∼5 h. The left and right panels show the enlarged details of the responses at the first and last 70 s, respectively. (b) Time-dependent variations of the normalized photocurrent for the same PD under ambient air conditions without encapsulation. All the measurements are conducted in ambient air atmosphere with a RH of 40–50% and a temperature of ∼25 °C.

3. Conclusions

In summary, we report for the first time the preparation of the CsPbBr₃ film by a PEAI sacrificial additive-assisted strategy for self-powered PDs. The CsPbBr₃ film possesses full coverage, a pure phase, micro-sized grains, higher crystallinity, fewer defects, and particularly a CsBr-richer surface that has a lower conductivity, in contrast with that prepared without PEAI sacrificial additive. The resulting self-powered PD with a simple configuration of FTO/TiO₂/CsPbBr₃/carbon exhibits a minimum dark current density of 8.05 × 10⁻¹¹ A cm⁻², a peak R of 0.35 A W⁻¹, a peak D* of 3.83 × 10¹³ Jones, and a response time of 1.46 μs. They are much better than the one prepared without PEAI sacrificial additive and are even beyond most self-powered CsPbBr₃ PDs reported previously. The mechanism underlying the superior performance of the self-powered CsPbBr₃ PD fabricated with PEAI sacrificial additive is explored and mainly attributed to the much reduced recombination of photo-generated carriers. In addition, we demonstrate the excellent operational reliability and long-term stability of the self-powered CsPbBr₃ PD in an ambient air atmosphere. Hence, our study suggests a simple and robust protocol for developing high-performance, stable CsPbBr₃ PDs for potential practical applications.

4. Experimental section

Reagents and materials

Ultradry PbBr₂ (99.999%, Alfa Aesar), ultradry CsI (99.998%, Alfa-Aesar), PEAI (≥99.5%, Xi'an Polymer Light Technology Corp.), CH₃NH₃Br (≥99.5%, Xi'an Polymer Light Technology Corp.), anhydrous methanol (99.9%, Alfa-Aesar), DMSO (ACS reagent, ≥99.9%, Sigma-Aldrich), conductive carbon paste (ZF-G03-04-2, Shanghai MaterWin New Materials Co., Ltd), and FTO glass (TEC8, 8 Ω sq⁻¹, Pilkington TEC Glass) were purchased and used as received unless otherwise noted.

Preparation of the CsPbBr₃ film

100 μL of TiO₂ sol was spin-coated onto a pre-cleaned FTO substrate at 3500 rpm for 30 s, which was further annealed at 500 °C for 60 min in ambient air. After cooling down to room temperature naturally, the FTO/TiO₂ substrate can be prepared. The CsPbBr₃ film was then coated on it by an intermediate phase halide exchange method. To be specific, 100 μL precursor solution obtained by pre-dissolving 370 mg PbBr₂ and 260 mg CsI into 1 mL DMSO was dipped onto the FTO/TiO₂ substrate, which was followed by spin-coating at 3500 rpm for 30 s and 5000 rpm for 120 s. After this, 60 μL CH₃NH₃Br methanol solution with a concentration of 20 mg mL⁻¹ was spin-coated onto it at 5000 rpm for 30 s. Thus, the CsPbBr₃ precursor film can be obtained. After being annealed at 300 °C for 30 min, the crystalline CsPbBr₃ film was achieved.

Fabrication of self-powered CsPbBr₃ PD

A carbon back-electrode with an active area of 0.085 cm² was loaded onto the CsPbBr₃ film using screen-printing commercial conductive carbon paste. After thermal treatment at 120 °C for 15 min, a self-powered CsPbBr₃ PD can be obtained.

Characterization

The XPS/UPS spectra were acquired using a photoelectron spectrometer (VG Multilab 2000×, Thermo Electron Corp., Waltham, MA, USA). The TGA and DSC curves were recorded using a STA 409 PC Luxx thermal analyser (NETZSCH, Germany) with a heating rate of 10 °C min⁻¹. The XRD patterns were tested using a Bruker D8 Advance powder X-ray diffractometer and Cu K_α radiation (1.5406 Å). The UV-vis absorption spectra were recorded using a spectrophotometer (U-4100, Hitachi). The TRPL spectra were obtained using a time-resolved single photon-counting setup (FluoTime 300, PicoQuant). The SEM images were obtained by high-resolution SEM (JEOL-7100F). The AFM images were tested using a Bruker Dimension FastScan AFM. The J–V curves were measured using a Keithley 2636 source meter. The EQE spectra were collected using a Zolix DSR-101-UV system equipped with a Si PD as reference. The R curves of the PDs can be obtained using the following formula: , where λ is the wavelength, h is the Planck constant, c is the light speed, and e is the elementary charge. D* spectra can be calculated using the followed equation: , where A is the device working area, e is the elementary charge and I_d is the measured minimum dark current. The TPC curves were recorded using a digital oscilloscope (Tektronix, MSO5204B) with an input resistance of 50 Ω, where the device was illuminated with a 520 nm pulse laser (MDL-NS-520). EIS measurements were performed on an electrochemical workstation (CHI600E, Shanghai Chenhua) with 10 mV amplitude perturbation. M–S plots were obtained using the same system with an AC excitation amplitude of 30 mV at a frequency of 5 kHz. Stability curves of PDs were recorded using a Keithley 2636B source meter, during which the device was illuminated with a 405 nm CW laser (MDL-III-405) modulated by a square signal (100 mHz, 10 s) generated by an arbitrary function generator (Tektronix, AFG3051C). All the measurements of PD performance were conducted in an ambient air atmosphere with a RH of 40–50% and a temperature of ∼25 °C.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

All the authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (61804113, 61704128, and 61874083), the Initiative Postdocs Supporting Program (BX20190261), and the National Natural Science Foundation of Shaanxi Province (2018ZDCXL-GY-08-02-02 and 2017JM6049).

References

1. J. S. Manser, J. A. Christians and P. V. Kamat, Chem. Rev., 2016, 116, 12956–13008.
2. W. J. Yin, T. Shi and Y. Yan, Adv. Mater., 2014, 26, 4653–4658.
3. K. Chen, S. Schünemann, S. Song and H. Tüysüz, Chem. Soc. Rev., 2018, 47, 7045–7077.
4. M. Saliba, J. P. Correa-Baena, M. Grätzel, A. Hagfeldt and A. Abate, Angew. Chem., Int. Ed., 2018, 57, 2554–2569.
5. Z. Li, T. R. Klein, D. H. Kim, M. Yang, J. J. Berry, M. F. van Hest and K. Zhu, Nat. Rev. Mater., 2018, 3, 18017.
6. Y. Rong, Y. Hu, A. Mei, H. Tan, M. I. Saidaminov, S. I. Seok, M. D. McGehee, E. H. Sargent and H. Han, Science, 2018, 361, eaat8235.
7. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051.
8. Cell efficiency chart, https://www.nrel.gov/pv/cell-efficiency.html, accessed: August 2019.
9. L. Dou, Y. M. Yang, J. You, Z. Hong, W.-H. Chang, G. Li and Y. Yang, Nat. Commun., 2014, 5, 5404.
10. H. Wang and D. H. Kim, Chem. Soc. Rev., 2017, 46, 5204–5236.
11. M. Ahmadi, T. Wu and B. Hu, Adv. Mater., 2017, 29, 1605242.
12. Y. Zhao, C. Li and L. Shen, InfoMat, 2019, 1, 164–182.
13. Y. Yan, Q. Wu, Y. Zhao, S. Chen, S. Hu, J. Zhu, J. Huang and Z. Liang, Small, 2018, 14, 1802764.
14. C. Bao, J. Yang, S. Bai, W. Xu, Z. Yan, Q. Xu, J. Liu, W. Zhang and F. Gao, Adv. Mater., 2018, 30, 1803422.
15. J. Deng, J. Li, Z. Yang and M. Wang, J. Mater. Chem. C, 2019, 7, 12415–12440.
16. A. Fakharuddin, U. Shabbir, W. Qiu, T. Iqbal, M. Sultan, P. Heremans and L. Schmidt-Mende, Adv. Mater., 2019, 1807095.
17. Z. Wang, Z. Shi, T. Li, Y. Chen and W. Huang, Angew. Chem., Int. Ed., 2017, 56, 1190–1212.
18. C. Li, C. Han, Y. Zhang, Z. Zang, M. Wang, X. Tang and J. Du, Sol. Energy Mater. Sol. Cells, 2017, 172, 341–346.
19. M. Shoaib, X. Zhang, X. Wang, H. Zhou, T. Xu, X. Wang, X. Hu, H. Liu, X. Fan and W. Zheng, J. Am. Chem. Soc., 2017, 139, 15592–15595.
20. J. Ding, S. Du, Z. Zuo, Y. Zhao, H. Cui and X. Zhan, J. Phys. Chem. C, 2017, 121, 4917–4923.
21. Y. Dong, Y. Gu, Y. Zou, J. Song, L. Xu, J. Li, J. Xue, X. Li and H. Zeng, Small, 2016, 12, 5622–5632.
22. Y. Li, Z.-F. Shi, S. Li, L.-Z. Lei, H.-F. Ji, D. Wu, T.-T. Xu, Y.-T. Tian and X.-J. Li, J. Phys. Chem. C, 2017, 5, 8355–8360.
23. M. I. Saidaminov, M. A. Haque, J. Almutlaq, S. Sarmah, X. H. Miao, R. Begum, A. A. Zhumekenov, I. Dursun, N. Cho and B. Murali, Adv. Opt. Mater., 2017, 5, 1600704.
24. Z. Yang, Q. Xu, X. Wang, J. Lu, H. Wang, F. Li, L. Zhang, G. Hu and C. Pan, Adv. Mater., 2018, 30, 1802110.
25. Y. Li, Z. Shi, L. Lei, F. Zhang, Z. Ma, D. Wu, T. Xu, Y. Tian, Y. Zhang and G. Du, Chem. Mater., 2018, 30, 6744–6755.
26. G. Tong, H. Li, D. Li, Z. Zhu, E. Xu, G. Li, L. Yu, J. Xu and Y. Jiang, Small, 2018, 14, 1702523.
27. H. Zhou, Z. Song, C. R. Grice, C. Chen, J. Zhang, Y. Zhu, R. Liu, H. Wang and Y. Yan, Nano Energy, 2018, 53, 880–886.
28. Y. Shen, C. Wei, L. Ma, S. Wang, X. Wang, X. Xu and H. Zeng, J. Mater. Chem. C, 2018, 6, 12164–12169.
29. Y. Zhang, W. Xu, X. Xu, J. Cai, W. Yang and X. Fang, J. Phys. Chem. Lett., 2019, 10, 836–841.
30. Y. Zhang, S. Li, W. Yang, M. K. Joshi and X. Fang, J. Phys. Chem. Lett., 2019, 10, 2400–2407.
31. W. Zhu, M. Deng, Z. Zhang, D. Chen, H. Xi, J. Chang, J. Zhang, C. Zhang and Y. Hao, ACS Appl. Mater. Interfaces, 2019, 11, 22543–22549.
32. J. Zeng, X. Li, Y. Wu, D. Yang, Z. Sun, Z. Song, H. Wang and H. Zeng, Adv. Funct. Mater., 2018, 28, 1804394.
33. Z. L. Wang, Adv. Mater., 2012, 24, 280–285.
34. M. Xue, H. Zhou, G. Ma, L. Yang, Z. Song, J. Zhang and H. Wang, Sol. Energy Mater. Sol. Cells, 2018, 187, 69–75.
35. C. Tian, F. Wang, Y. Wang, Z. Yang, X. Chen, J. Mei, H. Liu and D. Zhao, ACS Appl. Mater. Interfaces, 2019, 11, 15804–15812.
36. G. Cen, Y. Liu, C. Zhao, G. Wang, Y. Fu, G. Yan, Y. Yuan, C. Su, Z. Zhao and W. Mai, Small, 2019, 1902135.
37. K. Shen, X. Li, H. Xu, M. Wang, X. Dai, J. Guo, T. Zhang, S. Li, G. Zou and K.-L. Choy, J. Mater. Chem. A, 2019, 7, 6134–6142.
38. H. Zhou, Z. Song, C. R. Grice, C. Chen, X. Yang, H. Wang and Y. Yan, J. Phys. Chem. Lett., 2018, 9, 4714–4719.
39. H. Zhou, L. Fan, G. He, C. Yuan, Y. Wang, S. Shi, N. Sui, B. Chen, Y. Zhang and Q. Yao, RSC Adv., 2018, 8, 29089–29095.
40. H. Zhou, J. Zeng, Z. Song, C. R. Grice, C. Chen, Z. Song, D. Zhao, H. Wang and Y. Yan, J. Phys. Chem. Lett., 2018, 9, 2043–2048.
41. B. Yang, F. Zhang, J. Chen, S. Yang, X. Xia, T. Pullerits, W. Deng and K. Han, Adv. Mater., 2017, 29, 1703758.
42. F. Jin, B. Zhao, B. Chu, H. Zhao, Z. Su, W. Li and F. Zhu, J. Mater. Chem. C, 2018, 6, 1573–1578.
43. L. Zhang, F. Yuan, H. Dong, B. Jiao, W. Zhang, X. Hou, S. Wang, Q. Gong and Z. Wu, ACS Appl. Mater. Interfaces, 2018, 10, 40661–40671.
44. K. Wang, Z. Jin, L. Liang, H. Bian, D. Bai, H. Wang, J. Zhang, Q. Wang and L. Shengzhong, Nat. Commun., 2018, 9, 4544.
45. Y. Wang, T. Zhang, M. Kan, Y. Li, T. Wang and Y. Zhao, Joule, 2018, 2, 2065–2075.
46. Q. Jiang, Y. Zhao, X. Zhang, X. Yang, Y. Chen, Z. Chu, Q. Ye, X. Li, Z. Yin and J. You, Nat. Photonics, 2019, 13, 460–466.
47. Y. Li, Z.-F. Shi, S. Li, L.-Z. Lei, H.-F. Ji, D. Wu, T.-T. Xu, Y.-T. Tian and X.-J. Li, J. Mater. Chem. C, 2017, 5, 8355–8360.
48. X. Zhang, B. Xu, J. Zhang, Y. Gao, Y. Zheng, K. Wang and X. W. Sun, Adv. Funct. Mater., 2016, 26, 4595–4600.
49. D. Thrithamarassery Gangadharan, Y. Han, A. Dubey, X. Gao, B. Sun, Q. Qiao, R. Izquierdo and D. Ma, Sol. RRL, 2018, 2, 1700215.
50. J. Duan, Y. Zhao, B. He and Q. Tang, Angew. Chem., Int. Ed., 2018, 57, 3787–3791.
51. Q. Zhang, W. Zhu, D. Chen, Z. Zhang, Z. Lin, J. Chang, J. Zhang, C. Zhang and Y. Hao, ACS Appl. Mater. Interfaces, 2018, 11, 2997–3005.
52. W. Zhu, L. Kang, T. Yu, B. Lv, Y. Wang, X. Chen, X. Wang, Y. Zhou and Z. Zou, ACS Appl. Mater. Interfaces, 2017, 9, 6104–6113.
53. Q. Chen, H. Zhou, T.-B. Song, S. Luo, Z. Hong, H.-S. Duan, L. Dou, Y. Liu and Y. Yang, Nano Lett., 2014, 14, 4158–4163.
54. Q. Ma, S. Huang, S. Chen, M. Zhang, C. F. J. Lau, M. N. Lockrey, H. K. Mulmudi, Y. Shan, J. Yao and J. Zheng, J. Phys. Chem. C, 2017, 121, 19642–19649.
55. L. A. Frolova, Q. Chang, S. Y. Luchkin, D. Zhao, A. F. Akbulatov, N. N. Dremova, A. V. Ivanov, E. E. Chia, K. J. Stevenson and P. A. Troshin, J. Mater. Chem. C, 2019, 7, 5314–5323.
56. W. Zhu, Q. Zhang, D. Chen, Z. Zhang, Z. Lin, J. Chang, J. Zhang, C. Zhang and Y. Hao, Adv. Energy Mater., 2018, 8, 1802080.
57. J. Chen and N. G. Park, Adv. Mater., 2018, 1803019.
58. J. Yang, C. Bao, W. Ning, B. Wu, F. Ji, Z. Yan, Y. Tao, J. M. Liu, T. C. Sum and S. Bai, Adv. Opt. Mater., 2019, 1801732.
59. F. Cao, L. Meng, M. Wang, W. Tian and L. Li, Adv. Mater., 2019, 31, 1806725.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9tc05403k

‡ The authors contributed equally to this work.

---