Introduction

The introduction of perovskites to photovoltaics skyrocketed the efficiency within a short duration owing to their brilliant optical, electrical properties, easy, inexpensive, and tunable synthesis routes [1]. Therefore, perovskites have drawn attention of researchers in optoelectronic applications, particularly in photovoltaics [2]. Perovskites have opened the door for the next generation cost-effective and highly efficient photovoltaic technology [3]. The first perovskite solar cells/PSCs were introduced in 2009; since then, different research groups have been working on perovskite materials and their applications in photovoltaics [4,5,6]. Over a span of few years, PSCs have achieved 25.7% PCE and still there is big scope for the improvements of these devices [6].

The early PSCs were based on thin films of perovskites, which were deposited via different methods namely spin coating, sequential deposition of the perovskite solution and thermal evaporation [7]. Such devices have shown significant performance irrespective of few limitations like formation of defect states, extrinsic defects/grain boundaries, purity of materials and quality of films which severely affect device performance [8]. Besides good efficiency, the less environmental stability and hysteresis loss in PSCs are the big concerns [9]. Lead based perovskites (CH3NH3PbX3, where X = I, Cl, Br) are commonly used to fabricate PSCs, however, they may possess extrinsic defects (larger voids/pinholes) and toxicity [10]. These extrinsic defects can produce shunt-leakage path by connecting electron transport layer/ETL and hole transport layer/HTL directly resulting in a decrease in efficiency and stability of PSCs [11, 12].

In this context, optimization of perovskite absorber layer and use of inorganic charge transport layers may lead to the enhancement in the performance of PSCs [13, 14]. As perovskite are susceptible to external moisture and oxygen which leads to direct oxidation of absorber layer resulting in reduction in absorbance and hence degradation of device [15]. Therefore, it has become necessary to protect perovskite absorber layers from external moisture and oxygen. This problem of less stability may be overcome with optimization of absorber layer as well as charge transport layers [16]. In this regard, metal oxides, metal chalcogenides such as ZnS, ZnSe, CdS, CuInS2, TiO2, ZnO, NiOx, and SnO2, etc. may be beneficial as they have versatile properties [17,18,19,20,21,22,23]. Some polymers such as Polymethyl Methacrylate/PMMA, etc. may be applied as an anti-reflecting material which form an inter-cross-linked thin layer to protect photovoltaic devices from moisture and oxygen; hence, play an important role in passivation of surface trapping and suppression of hysteresis loss in devices [24, 25]. Therefore, the photovoltaic performance of PSCs may be improved by proper use or optimization of absorber layer and charge transport layer materials, surface/solvent engineering, and better device architectures [18, 26, 27].

In this study, we have focused on optimization of absorber layer and prepared ZnSe-perovskite composites (ZnSe:CH3NH3PbI3; 2, 4, and 6 mg/ml). Synthesis of these composites has been confirmed by XRD, UV–Visible spectroscopy, PL/TRPL spectroscopy, Contact angle analysis and SEM. SEM micrographs revealed significant decrease in extrinsic defects of perovskite films with the increase in the concentration of ZnSe into perovskite matrices. This decrease has clearly indicated an improvement in morphology resulting reduction in the extrinsic defects/grain boundary problem. The contact angle analysis showed a substantial decrease in the hydrophilicity as contact angle is increased by ~ 10° in comparison to that of pristine perovskite. Also, TRPL spectroscopy confirmed a small but gradual increase in the lifetimes indicating a reduction in the recombination rate within ZnSe:perovskite composites. Furthermore, as-fabricated composite materials have been deployed in the fabrication of solar cells. As-fabricated solar cells have demonstrated improved performance and have shown an efficiency of 5.98%, Voc of 0.931 V, JSC of 9.82 mA/cm2 and FF of 65.50%. Finally, the stability performance of as-fabricated solar cells has been studied. The best performing solar device retained ~ 79.6% of its actual efficiency even after 30 days exposure under ambient atmosphere.

Methodology

Synthesis of ZnSe nanoparticles/NPs

Initially, polyvinyl pyrrolidine/PVP (2 g) was dissolved into deionized water/DI (70 ml) and solution was stirred for 2 h at 70 °C. In the next step, selenium powder (0.4 mmol) was dissolved into ethylenediamine (80 ml) to get selenium ethylenediamine. Furthermore, solution of zinc nitrate (Zn(NO3)2) and selenium ethylenediamine was prepared under constant stirring and added to the PVP solution to obtain a homogeneous solution. Finally, solution was dried at 80 °C for 24 h and as-obtained powder was crushed and calcined at ~ 700 °C in a tube furnace for 90 min (under N2 atmosphere) [28].

Synthesis of CH3NH3PbI3 and ZnSe:CH3NH3PbI3 composites

Initially, Hydriodic acid/HI (dropwise) was added into methyl amine and solution was kept under stirring for 2–3 h at 0 °C (using ice-bath). Further, solution was dried at 60 °C for overnight to get precipitated and washed multiple times with di ethyl ether and ethyl alcohol. In the next step, precipitate annealed at 60 °C for overnight to get rid of impurities. Finally, white powder was obtained which indicated synthesis of methyl ammonium iodide/MAI [29].

Furthermore, 1 M of MAI and Lead Iodide/PbI2 were taken into N-N dimethyl pyrrolidone/NMP and stirred for 12 h at 60 °C to prepare Methyl ammonium lead Iodide/MALI solution (one step method) [30]. Similarly, ZnSe:CH3NH3PbI3 composites (2, 4, and 6 mg/ml) were prepared except the addition of different concentrations of ZnSe NPs during the preparation of composite solutions.

Fabrication of perovskite solar cells

The fabrication of solar cell was done in multiple steps; firstly, Fluorine doped Tin Oxide/FTO glass substrates were etched out and washed with soap, distilled water, and ethyl alcohol, and dried in the oven. In the next step, NiOX HTL was deposited onto FTO substrate and annealed at 110 °C for 2–3 min. Furthermore, photoactive layer of Methyl Ammonium Lead Iodide (CH3NH3PbI3) was deposited and annealed at 100 °C for 5 min. After this, ZnO ETL was deposited and annealed at 100 °C for 2–3 min. In the last step, bottom contact of silver/Ag metal was thermally deposited onto this substrate [31, 32]. Similar steps were followed to fabricate the ZnSe:CH3NH3PbI3 composites based devices except the deposition of ZnSe:CH3NH3PbI3 absorber layers instead of pristine CH3NH3PbI3 absorber layer.

Characterization techniques

All the prepared materials such as ZnSe NPs, MAI, MALI, MALI-Perovskite composites are confirmed using X-Ray Diffractometer (Rigaku SmartLab), UV–Visible Spectrophotometer (Perkin Elmer UV/Vis Lambda 365), Photoluminescence (PL) Spectrophotometer (Shimadzu RF-5301 PC Spectrofluorophotometer), Scanning Electron Microscope (FEI Nova Nanosem 450). Furthermore, as-fabricated solar devices based on ZnSe-perovskite composites are tested using Keithley Sourcemeter (2450 Sourcemeter) and PET Solar simulator (SS50AAA).

Results and discussion

Figure 1 (SEM micrograph of ZnSe) displays the nanoscale morphology (average particle size; 51 nm) of as-prepared ZnSe nanoparticles/NPs. Further, XRD pattern (Fig. 2) confirms the synthesis of ZnSe NPs and suggests the presence of (110), (220), (311), (400), (102), and (203) crystal planes [JCPDS-79–2205] [28]. Moreover, UV visible spectroscopy was used to record the absorbance–wavelength spectrum (Fig. 3). The absorbance spectrum shows a peak corresponding to 280 nm, further Tauc Plot was used to determine bandgap of ZnSe NPs (Fig. 4) which was found to be 4.43 eV which is an agreement with the reported results [33]. In the next step, we performed Photoluminescence (PL) spectroscopy to study the emission spectrum of ZnSe NPs. The (PL) emission spectrum (Fig. 5) shows a peak at 407 nm, whereas the peak at ~ 500 nm can be attributed to the trap states [34]. Therefore, the SEM micrographs, XRD patterns, (UV–Visible) absorbance spectrum, and (PL) emission spectrum of ZnSe evidently confirm the synthesis of ZnSe NPs. Furthermore, as-synthesized ZnSe NPs have been used to prepare different ZnSe-CH3NH3PbI3 composites which are deployed as absorber layers in the fabrication of peroskite solar cells/PSCs.

Fig. 1
figure 1

Scanning Electron Micrograph of ZnSe NPs

Full size image
Fig. 2
figure 2

XRD pattern of ZnSe NPs

Full size image
Fig. 3
figure 3

UV–Visible (absorbance) spectrum of ZnSe NPs

Full size image
Fig. 4
figure 4

Bandgap calculations of ZnSe NPs using Tauc plot

Full size image
Fig. 5
figure 5

PL spectrum of ZnSe NPs

Full size image

Figure 6 (a, b, c, and d) displays SEM micrographs of four different perovskite-ZnSe composites based absorber layers deployed in as-fabricated solar cells. Furthermore, using ImageJ (https://imagej.nih.gov/ij/) software, the grain size of perovskite-ZnSe composites based absorber layers are calculated. The average grain size of ZnSe:CH3NH3PbI3 composite (0, 2, 4, and 6 mg/ml, respectively) films are found to be increasing with the increase of ZnSe NPs into perovskite matrices (Fig. 6) which are 263, 738, 984, and 1364 nm, respectively. This increase in grain size may be owing to the prolonged crystallization/aggregation of perovskite film in the presence of ZnSe NPs [34] and due to the larger dipole moment (42 D) of ZnSe NPs in comparison to that of CH3NH3 (2.2 D) [35, 36]. ZnSe NPs have improved morphology of ZnSe:perovskite composite films by filling pinholes/voids. As morphology of films depends on coverage ratio, crystallinity, and crystal size, therefore, larger grain size of perovskite composite films increases optical performance resulting increase in the performance of the PSCs.

Fig. 6
figure 6

SEM of ZnSe-Perovskite composites (Top: L to R; 0 and 2 mg/ml, and Bottom L to R; 4 and 6 mg/ml)

Full size image

Figure 7 demonstrates XRD patterns of perovskite:ZnSe composites which clearly shows the role of the incorporation of ZnSe NPs into perovskite matrices. XRD patterns of CH3NH3PbI3:ZnSe composites (0, 2, 4, and 6 mg/ml, respectively) consist of CH3NH3PbI3 diffraction peaks corresponding to (110), (220), (202), (310), (400) and (404) planes [37, 38]. The positions of CH3NH3PbI3 diffraction peaks are unchanged even after incorporation of ZnSe NPs into perovskite matrices indicating that MALI has well-preserved crystal structure. Figure 7 does not show the diffraction peaks corresponding to ZnSe NPs which is probably due to the less amount of ZnSe NPs present in the films. Therefore, it is quite difficult to detect the diffraction peaks of ZnSe NPs present in CH3NH3PbI3:ZnSe composite films than that of the crystallite sites of CH3NH3PbI3 (Fig. 7). Furthermore, Table 1 shows a comparative study on FWHM (full width of half maximum) and diffraction peak intensities correspond to (110) and (220) diffraction peaks. It is obvious from Table 1 that the diffraction peak intensities increase with the increase in the concentration of ZnSe NPs into CH3NH3PbI3:ZnSe composite films resulting improvement in the morphology. Hence, the crystallinity of the perovskite phase has been improved drastically. Conversely, FWHM of diffraction peaks has been seen decreasing with the increase in the concentration of ZnSe NPs indicating the formation of larger crystalline domains of perovskite present in CH3NH3PbI3:ZnSe composites. The crystalline domain sizes (calculated from the Scherrer equation) of CH3NH3PbI3:ZnSe composites (0, 2, 4, and 6 mg/ml) are found to be 4.66 Å, 5.15 Å, 5.57 Å and 6.78 Å, respectively.

Fig. 7
figure 7

XRD patterns of CH3NH3PbI3:ZnSe (0, 2, 4, and 6 mg/ml, respectively) composite films deposited onto the FTO substrates

Full size image
Table 1 FWHM and Intensity plot
Full size table

UV–Visible spectroscopic technique (Fig. 8) is used to record optical absorbance of perovskite:ZnSe composites in the range of 400–800 nm. Figure 8 has shown that the incorporation of ZnSe NPs enhances the optical absorbance of the perovskite (composite) films. It is also clear from literature that ZnSe NPs absorb in the range less than 400 nm [38,39,40,41]. Therefore, higher absorbance of ZnSe:CH3NH3PbI3 composite films does not correspond to the absorbance of ZnSe NPs in the range of 400–800 nm, however, ZnSe NPs show high reflectance in this range [40]. Moreover, ZnSe NPs present in ZnSe:CH3NH3PbI3 composite films act as the network of reflection centers in the range 400–800 nm which further increases the light trapping in ZnSe:CH3NH3PbI3 composite films. Therefore, enhancement in the light trapping boosts the probability of light absorption by the perovskite and hence increases the absorbance of perovskite:ZnSe composite films [40].

Fig. 8
figure 8

The absorption spectra of CH3NH3PbI3:ZnSe (0, 2, 4, and 6 mg/ml, respectively) composite films deposited onto the FTO substrates

Full size image

The photo-physics of absorber layers has been studied using PL (emission) spectra of FTO/CH3NH3PbI3:ZnSe (0, 2, 4, 6 mg/ml, respectively) composite films (Fig. 9) which are peaked at 762 nm (Fig. 9) indicating that the optical bandgaps of the multilayers are being controlled by CH3NH3PbI3 layer [41]. An obvious quenching is observed for each composite films (Fig. 9) owing to the introduction of ZnSe. The maximum PL quenching is observed for the highest concentrated (6 mg/ml) perovskite:ZnSe composite film by improving morphology of perovskite:ZnSe composite films by forming larger grains which reduces extrinsic defects by filling the pinholes/voids. Mahmoudi et al. also reported that perovskite/NiO-graphene composites filled pinholes and voids of the perovskite films and hence improvement in morphology of perovskite films which further reduced charge recombination at interfaces. This reduction in charge recombination at interfaces increased overall performance of PSCs [42]. This study also shows reduction in extrinsic defects which indicates improvement in morphology of perovskite:ZnSe composite films by reducing the probability of charge recombination at the interfaces [43]. Hence, the drastic PL quenching in FTO/CH3NH3PbI3:ZnSe composite (2, 4, and 6 mg/ml, respectively) films is observed in comparison to that of pristine CH3NH3PbI3. This increase in the PL quenching with the increase in the concentration of ZnSe NPs into the perovskite NPs further indicates the increased charge recombination in photoactive layers owing to the aggregation of ZnSe NPs in CH3NH3PbI3:ZnSe composite films providing the required sites for charge recombination.

Fig. 9
figure 9

The steady-state PL spectra of FTO/CH3NH3PbI3, and FTO/CH3NH3PbI3:ZnSe (2, 4, and 6 mg/ml, respectively) composite films

Full size image

Figure 10 depicts an important physical property of the as-fabricated CH3NH3PbI3:ZnSe composites i.e., contact angle analysis which shows small contact angle and high surface wettability [36]. However, the incorporation of ZnSe NPs into perovskites reduces the hydrophilicity of the perovskite-ZnSe composites [44]. The results clearly demonstrate that the contact angle increases significantly as concentration of ZnSe NPs into perovskite matrices increases, and comes to be 33.92°, 35.03°, 40.31°, and 43.52° for perovskite-ZnSe composites (0, 2, 4, and 6 mg/ml, respectively). It is obvious from the obtained results that the pristine perovskite has a very high hydrophilicity [45]; nevertheless, this hydrophilicity considerably decreases with an increase in ZnSe NPs concentration into perovskite, indicating an improvement in environmental stability. Moreover, Fig. 11 (TRPL spectroscopy) revealed the charge carrier dynamics of perovskite-ZnSe composites. It is evident from Fig. 11 that the average lifetime of pure perovskite (deposited onto FTO substrate) is determined to be 21.49 ns; however, in case of FTO/NiOx/perovskite lifetime was found to be 21.115 ns. It is also found that the average lifetimes of perovskite-ZnSe composites (2, 4, and 6 mg/ml) are 21.206, 21.310, and 21.412 ns, respectively. The lifetimes of perovskite-ZnSe composites have shown a modest increase with the increase in the ZnSe NPs concentration into perovskite which may be due to the low concentration of ZnSe NPs into perovskite and carrier recombination process/charge carrier trapping at perovskite-NiOx interface as charge carriers are not transferred to glass substrate easily [46, 47]. However, from these findings it is obvious that the recombination rate decreased as ZnSe NPs concentration in perovskite increased, leading to an increase in the separation rate and, thus, an increase in charge carrier mobility [46, 47] (Fig. 12).

Fig. 10
figure 10

Contact Angle analysis of FTO/NiOx/ CH3NH3PbI3:ZnSe (Top: L to R: 0, and 2, and Bottom L to R: 4, and 6 mg/ml)/ZnO (from top left to bottom right), respectively

Full size image
Fig. 11
figure 11

The decay (average lifetime) plot of FTO/CH3NH3PbI3, and FTO/NiOx/CH3NH3PbI3:ZnSe (0, 2, 4, and 6 mg/ml) composite films, respectively

Full size image
Fig. 12
figure 12

(a) Schematic diagram, and (b) energy-level diagram of as-fabricated PSCs

Full size image

Figure 13 demonstrates J-V characteristics of the solar devices having the structures FTO/NiOx/CH3NH3PbI3:ZnSe (0, 2, 4, and 6 mg/ml, respectively). The schematic diagram as well as energy level diagram of as-fabricated solar devices are shown in Fig. 12 (a), and (b), respectively. Furthermore, Table 2 demonstrates photovoltaic parameters such as VOC, Jsc, PCE, FF, Series Resistance/Rs and Shunt Resistance/Rsh for as-fabricated devices. Table 2 also reveals that all the photovoltaic parameters (except Rsh) are found to be increasing with the increase in the concentration of ZnSe NPs into perovskite matrices owing to the improved optical properties and morphology [43]. The incorporation of ZnSe NPs facilitates crystallization and growth of bigger perovskite grains resulting in filling of the pinholes/voids by effectively reducing the charge recombination or charge trapping at the grain boundaries. This reduction in charge recombination helps in easy movement of charges towards the electrodes resulting increase in the Jsc of the devices. Moreover, larger grain size of the perovskite:ZnSe composites reduces series resistance/Rs and increases fill factor which is due to the less hindrance experienced by the charges [43]. The decrease in pinholes/voids and cracks increases the value of shunt resistance/Rsh [11, 12]. The conductivity and electron mobility are also improved with the increase in the concentration of ZnSe NPs into perovskite:ZnSe composite films by increasing Jsc and decreasing Rs of as-fabricated devices [11, 12].

Fig. 13
figure 13

The J-V curves of the photovoltaic devices having CH3NH3PbI3:ZnSe (0, 2, 4, and 6 mg/ml, respectively) composite films (as absorber layers)

Full size image
Table 2 Typical photovoltaic parameters of CH3NH3PbI3:ZnSe (0, 2, 4, and 6 mg/ml, respectively) composite films based solar devices
Full size table

Figure 14 displays stability responses of as-fabricated solar devices; the best device retained 79.6% of its original efficiency even after 30 days under ambient atmosphere. Whereas solar devices possessing pristine absorber layer has retained just 20.7% of the original efficiency after the same period of 30 days. Moreover, it has been supported by the literature that the use of inorganic charge transport, i.e., ZnO and NiOx enhances the stability of PSCs [13, 14]. In this work, we have deployed ZnO as ETL and NiOx as HTL (Fig. 12); this is because of the low stability of the PCBM (ETL) and PEDOT:PSS (HTL) based PSCs owing to the hygroscopic and acidic natures of PCBM and PEDOT:PSS [13]. Moreover, use of PCBM leads to the diffusion of electrode materials into the PCBM layer up to 10 nm during vacuum evaporation [48]. This diffusion of PCBM into electrode increases the probability of the exposure of electrode material with perovskite and hence increases the risk of chemical reaction between electrode and perovskite films under humid conditions which may trigger faster device degradation. In this context, inorganic charge transport layers such as highly air stable ZnO, and NiOx could be the good options. ZnO entirely separates the perovskite with electrode material [13] due to its scavenging effect [49, 50]. Conversely, highly stable NiOx may replace hydrophilic and acidic PEDOT:PSS [51, 52]. Therefore, use of ZnO and NiOx as ETL, and HTL, respectively improves the stability of PSCs. Likewise, incorporation of ZnSe NPs into perovskite matrices improves morphology of as-fabricated solar devices by filling pinholes/voids. Pinholes/voids leads to direct oxidation absorber layers resulting in rapid degradation of absorber layer [53,54,55,56]. Therefore, filling of pinholes/voids is important for the fabrication of stable and efficient solar cells. In this context, both the strategies (use of inorganic charge transport layers and incorporation of ZnSe NPs into perovskite) which were adopted by us have played a significant role which is further justified with the available literature [49, 50, 53,54,55,56].

Fig. 14
figure 14

Ambient air stability curves of the devices studied in the present work

Full size image

Conclusion

The incorporation of ZnSe NPs into perovskite matrices significantly enhanced overall performance of perovskite:ZnSe (CH3NH3PbI3:ZnSe; 0, 2, 4, and 6 mg/ml, respectively) composites based solar devices. Furthermore, such fabricated perovskite solar cells/PSCs demonstrated that employment of CH3NH3PbI3:ZnSe photoactive layers improved visible absorbance, and increased grain sizes and hence reduction in extrinsic defects of CH3NH3PbI3 films which further increased performance of PSCs. Moreover, highly stable inorganic carrier transport layers also played an important role in performance increase. As-fabricated PSCs have shown optimized performance; the best device has shown an efficiency of 5.98% and retained 79.6% of the original value even after 30 days under ambient atmosphere.