Skip to main content

Lead-free perovskite solar cells using Sb and Bi-based A3B2X9 and A3BX6 crystals with normal and inverse cell structures

Abstract

Research of CH3NH3PbI3 perovskite solar cells had significant attention as the candidate of new future energy. Due to the toxicity, however, lead (Pb) free photon harvesting layer should be discovered to replace the present CH3NH3PbI3 perovskite. In place of lead, we have tried antimony (Sb) and bismuth (Bi) with organic and metal monovalent cations (CH3NH3 +, Ag+ and Cu+). Therefore, in this work, lead-free photo-absorber layers of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, (CH3NH3)3SbBiI9, Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6 were processed by solution deposition way to be solar cells. About the structure of solar cells, we have compared the normal (n-i-p: TiO2-perovskite-spiro OMeTAD) and inverted (p-i-n: NiO-perovskite-PCBM) structures. The normal (n-i-p)-structured solar cells performed better conversion efficiencies, basically. But, these environmental friendly photon absorber layers showed the uneven surface morphology with a particular grow pattern depend on the substrate (TiO2 or NiO). We have considered that the unevenness of surface morphology can deteriorate the photovoltaic performance and can hinder future prospect of these lead-free photon harvesting layers. However, we found new interesting finding about the progress of devices by the interface of NiO/Sb3+ and TiO2/Cu3BiI6, which should be addressed in the future study.

1 Introduction

Recent advancements of organic–inorganic perovskite (CH3NH3PbI3) thin-film solar cells, which can be fabricated by economical-promising solution process, have marked significant achievements of photoconversion efficiency (PCE) from the initial efficiency of 3.8% to over 22% [1,2,3]. The PCE progress has been realized due to the excellent ability of CH3NH3PbI3 about light harvesting, charge separation and charge transportation. These excellent properties also realize to fabricate perovskite solar cells in different structural designs as normal (n-i-p) mesoscopic and inverted (p-i-n) planar configurations. However, the working instability had been a profound problem [4,5,6,7]. Moreover, the toxicity of lead (Pb) shows big impact on the environment and human being, and then, the quest for such lead-free thin film solar cell has been highly demanded [8, 9]. The first lead-free perovskite solar cells were initially fabricated with the substitution of lead to tin (Sn) for the mesoscopic structure solar cells as <FTO glass/cp-TiO2/CH3NH3SnI3/Spiro-OMeTAD/Au> [10, 11]. A moderate PCE of 6% was reported, but the rapid oxidation of Sn2+ to be Sn4+ at the ambient condition arised concern. Afterward, however, it was found that tin cation shows higher toxicity than lead [12]. Hence, another elements have been considered for the further ongoing lead-free solar cells.

In order to substitute the Pb, another potential, germanium (Ge2+) cation was implemented into the organo-metal-halide crystal as a harvesting layer, which has the same oxidation state and lower electronegativity and resulted in 0.2% PCE [13]. As another divalent, earth abundant and non-toxic element, copper cation was utilized in mesoscopic perovskite solar cell structure with (CH3NH3)2CuClxBry as a photon harvesting layer [14]. The low PCE of 0.017% was observed due to the low absorption coefficients, the high effective mass of holes and the low intrinsic conductivity of the employed perovskite layer.

The other viable opinion towards lead cation (Pb2+) substitution is adaptation of heterovalent (not divalent) cation into the organic-metal-halide crystals. These substitutions have to follow charge neutrality and can alter perovskite structure (ABX3) to another crystal (A: organic ion; B: metal ion; X: halide ion). Recently, trivalent cations of antimony (Sb) and bismuth (Bi) based organic-metal-halide crystals (A3B2X9 or A3BX6) got significant interest as a photon harvesting layer due to their environmentally friendly nature and available inactive outer shell s orbital [15, 16]. Öz et al. proposed (CH3NH3)3Bi2I9 based inverted solar cells as <ITO/PEDOT:PSS/(CH3NH3)3Bi2I9/PCBM/Ca/Al. However, the energy mismatch limited the charge extraction, resulting in the low PCE to 0.07%. Hebig et al. proposed <ITO/PEDOT:PSS/(CH3NH3)3Sb2I9)/PCBM/ZnO/Al> structures and reported 0.49% PCE [15, 16]. The trivalent bismuth (Bi3+) has utilized for the fabrication of (CH3NH3)3Bi2I9 based organic-metal-halide layer, which showed optimal absorption coefficient [17, 18]. This Bi-based organic-metal-halide solar cells were reported 0.33% of the low PCE, which was limited by its high exciton binding energy of 300 meV. It was founded that the morphology of (CH3NH3)3Bi2I9 was changed by the selection of electron transporting layers (ETL), and that the small amount addition of NMP (N-methyl-2-pyrrolidone) solvent could also attain to grow uniform surface morphology [18, 19].

In order to improve the Bi-based A3BX6-structured solar cells, recently, Ag and Cs were composed as A-site monovalent cations to be Cs2AgBiX6 structure, due to better optical and electrical properties and its ambient stability [20,21,22,23]. The substitution of methylammonium ion with silver can extend the dimension of all inorganic active material maintaining its 3d structure, which can be more suitable for photo harvesting due to its uniform nature. Ag found application in various structural compounds including AgBiI4 and (CH3NH3)2AgBiBr6 although, only optical properties are established and the PCE is yet to be demonstrated [24, 25]. Moreover, the Chemical materials Evaluation and REserarch BAse (CEREBA) reported the incorporation of the silver based A-site cation by replacing the methylammonium (CH3NH3 +) ion preserving the Bi and iodine (Ag3BiI6) and reported 4.3% PCE [23]. Hence, it can be assumed that the Ag–Bi based photo absorber can perform the lead free organo-metal-halide solar cells with better PCE.

In order to improve the photovoltaics of lead free A3B2X9 and A3BX6 solar cells, the further morphological and structural study of Sb and Bi-based metal-halide layers and the effect on PCE should be studied. In this study, we have tried A3B2X9 and A3BX6-structured crystals for the photo absorber layers of solar cells with several combinations of A site cations (CH3NH3 +, Ag+ and Cu+) and B site cations (Bi3+ and Sb3+), which were (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, (CH3NH3)3SbBiI9, Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6. The crystal configurations of A3B2X9 and A3BX6 were tried to be controlled by the amount of elements in the solutions of source materials. These crystal layers were implemented in normal (n-i-p) and inverted (p-i-n) solar cell architectures (Fig. 1): the normal mesoscopic solar cell <FTO glass/cp TiO2/mp TiO2/(Pb-free A3B2X9 and A3BX6 layer)/spiro-OMeTAD/Au> and the inverted structure <FTO glass/NiO layer/(Pb-free A3B2X9 and A3BX6 layer)/PCBM/BCP/Ag>. It was confirmed that the adjacent substrate to A3B2X9 crystal (TiO2 electron transporting layer (ETL) and NiO hole transporting layer (HTL)) affected on the lead-free A3B2X9 and A3BX6 morphology and its effect on PCE are studied.

Fig. 1
figure 1

Structures of fabricated solution processed lead-free A3B2X9 and A3BX6 crystal (A: monovalent cation; B: trivalent cation; X: halogen anion) solar cells: normal mesoscopic (n-i-p) (a) and inverted planar (p-i-n) (b) structures

2 Results and discussion

2.1 A3B2X9 crystals with methyl ammonium cation for the A site

The XRD patterns of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, and (CH3NH3)3SbBiI9 thin films are similar and possesses the strong preferential growth in c axis direction (Fig. 2a). The observed XRD peaks of thin (CH3NH3)3Bi2I9 film at 8.16, 16.34 and 24.62 are well matched to the literature and correspond to the indexed planes (002), (004) and (006) respectively in a P63/mmc hexagonal space group where preferential growth direction is in c axis direction [15]. The observed peak of (CH3NH3)3Sb2I9 film at 8.22, 16.54, 24.88 and 50.98 could be indexed by planes (001), (002), (003) and (402) respectively in P63/mmc hexagonal space group and possesses the strong preferential growth in c axis direction similar to the (CH3NH3)3Bi2I9 film [16]. The XRD pattern of (CH3NH3)3SbBiI9 film containing Bi and Sb also shows preference to have the same c axis orientation growth. Due to the multi cation at B site in the crystal, we could not define the exact element at each position. Hence, we should not put the exact index at the peaks. The details analysis is ongoing for the next publication as the future works.

Fig. 2
figure 2

XRD patterns (a) and UV–vis absorption spectra (b) of grown solution processed (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9 and (CH3NH3)3SbBiI9 films on a glass substrate. The insets were the pictures of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9 and (CH3NH3)3SbBiI9 films

Figure 2b shows UV–vis spectra and photographs of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, and (CH3NH3)3SbBiI9 thin films. The (CH3NH3)3Bi2I9 film absorbs with the peak appearing around 500 nm. On the other hand, The UV–vis spectrum of (CH3NH3)3Sb2I9 shows a shoulder peak around 460 nm and no clear peak. The casted (CH3NH3)3SbBiI9 film shows higher absorption around 500 nm than (CH3NH3)3Bi2I9 and (CH3NH3)3Sb2I9 films. The exact absorption coefficients were not shown in the figure due to the inhomogeneous structures of films, which will be shown later.

The (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, and (CH3NH3)3SbBiI9 layers processed on different substrates (TiO2 or NiO) were observed with scanning electron microscopy (SEM) images (Fig. 3). Due to the crystal structures with hexagonal space group detected by XRD (Fig. 2a), the morphologies became also hexagonal shapes, basically. The (CH3NH3)3Bi2I9 film growth morphology on TiO2 layer is hexagonal (Fig. 3a), but were irregular hexagon and star shape on NiO film (Fig. 3b). Apparently, this observed variation of morphology pattern is attributed to the interface between metal oxides (TiO2 and/or NiO) and A3B2X9 organic-metal-halide crystals. On the other hand, the (CH3NH3)3Sb2I9 (Fig. 3c, d) crystals were hexagonal simply, irrespective of substrate (TiO2 and/or NiO). Additionally, on the NiO substrate, the appearing irregular hexagon shows its polycrystalline nature. About the (CH3NH3)3SbBiI9 film (Fig. 3e, f), these observed SEM images provide the information of the surface coverage, suggesting polycrystalline growth of hexagonal and irregular hexagonal in nature. Essentially, the inhomogeneity is involved in the growth of film, and the homogeneous surface coverage is difficult. This characteristics may be attributed to the poor PCE and with polycrystalline growth nature [26].

Fig. 3
figure 3

SEM images of A3B2X9 films [(CH3NH3)3Bi2I9: (a, b); (CH3NH3)3Sb2I9: (c, d); (CH3NH3)3SbBiI9: (e, f)] grown on different substrates [TiO2: (a, c, e); NiO: (b, d, f)]

Figure 4 shows the SEM images of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, and (CH3NH3)3SbBiI9 crystals casted on TiO2 and NiO substrates for the elemental distribution by EDX analysis. In order to reduce the electron charge accumulation on the surface during the EDX measurements, the gold (Au) thin layer was deposited by sputtering system beforehand. Hence the component contribution from Au was observed. The sum of all elemental contribution was set to be 100. The points of elemental analysis were chosen at various particular spot locations and are numbered in Fig. 4. The results of elemental analysis are summarized in Tables 1, 2, 3, 4, 5, 6, which are projected by the numbers in Fig. 4. About (CH3NH3)3Bi2I9 film elemental analysis on TiO2 substrates (Fig. 4a, Table 1), the small amounts of Bi and I out of grain were observed (points 4 and 5). The points on grain (point 1, 2, and 3) show the large amount of Bi and I than those out of grain (points 4 and 5). Hence, It can be confirmed that there was also the thin layer of (CH3NH3)3SbBiI9 at the points out of the large grain. The ratios of I/Bi at the points 1, 2, 3, 4, and 5 in Fig. 4a were 4.21, 3.91, 3.76, 4.08 and 4.25, respectively. Thinking about the I/Bi stoichiometry ratio (= 4.5) in the (CH3NH3)3Bi2I9 crystal, the results of EDX show the reduction of iodide element in the material. Although Bi was distributed on all points of the TiO2 surface (Table 1), there are missing points of Bi on NiO layer (at the points 2 and 3 in Table 2). The points 2 and 3 out of grain on NiO may possess only CH3NH3I. The ratios of I/Bi at the points 1, 4, and 5 in Fig. 4b and Table 2 are 4.92, 4.13, and 4.02, respectively, which are larger than those of grains in Fig. 4a and Table 1. Possibly, the solution of (CH3NH3)3Bi2I9 was repelled from the surface of NiO and crystallized on NiO close to the stoichiometry ratio. About (CH3NH3)3Sb2I9, it was surprising that there are missing points of Sb not only on NiO, but also on TiO2 (Fig. 4c, d, Tables 3, 4). Therefore, the anchoring strength of Sb on to the oxides (TiO2 or NiO) would be weaker than that of Bi. The elemental ratios (I/Sb) of crystals were 6.17 and 4.79 on TiO2 (at the points 1 and 3 in Fig. 4c and Table 3) and 3.98, 3.65 and 4.03 on NiO (at the points 1, 3 and 4 in Fig. 4d and Table 4), respectively. The reason was not clear, but the elemental ratios (I/Sb) of crystals were higher on TiO2 than NiO. About (CH3NH3)3SbBiI9 (Fig. 4e, f, Tables 5, 6), Bi was not observed out of the crystal as (CH3NH3)3Sb2I9 (Fig. 4c, d, Tables 3, 4), which was predicted by the weak anchoring strength between Sb and metal oxides as above. Although the material ratio in the solution of (CH3NH3)3SbBiI9 was in the stoichiometry, the elemental ratio of Sb/Bi was changed very much to the crystals, which would be due to the segregation to (CH3NH3)3SbI9 and (CH3NH3)3BiI9. The elemental ratios (I/(Sb + Bi)) of (CH3NH3)3SbBiI9 crystals were 4.11, 3.92, and 3.95 on TiO2 (at points 1, 2, and 3 in Fig. 4e and Table 5) and 4.92, 7186, 3.92, and 4.52 on NiO (at points 1, 2, 4 and 5 in Fig. 4f and Table 6), respectively. The large variation of elemental ratio [I/(Sb + Bi)] would be attributed to the crystal segregations. Anyway, the SEM–EDX analysis suggests the nonuniformity of the crystals of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, and (CH3NH3)3SbBiI9. These all results are in contrast to the lab scale established CH3NH3PbI3 perovskite-based solar cells where high efficiencies are achieved with uniform and pin hole free surface morphology [27, 28].

Fig. 4
figure 4

SEM images observed for elemental analysis of A3B2X9 films [(CH3NH3)3Bi2I9: (a, b); (CH3NH3)3Sb2I9: (c, d); (CH3NH3)3SbBiI9: (e, f)] grown on different substrates [TiO2: (a, c, e); NiO: (b, d, f)]. The marked spots indicate the positions of elemental analysis, which are related to the results shown in Tables 1, 2, 3, 4, 5, and 6

Table 1 Elemental distribution (atomic %) of the individual elements observed on particular spot on (CH3NH3)3Bi2I9/TiO2 film shown in Fig. 4a
Table 2 Elemental distribution (atomic %) of the individual elements observed on a particular spot on (CH3NH3)3Bi2I9/NiO film shown in Fig. 4b
Table 3 Elemental distribution (atomic %) of the individual elements observed on particular spot on (CH3NH3)3Sb2I9/TiO2 film shown in Fig. 4c
Table 4 Elemental distribution (atomic %) of the individual elements observed on a particular spot on (CH3NH3)3Sb2I9/NiO substrate shown in Fig. 4d
Table 5 Elemental distribution (atomic %) of the individual elements observed on (CH3NH3)3SbBiI9/TiO2 substrate shown in Fig. 4e
Table 6 Elemental distribution (atomic %) of the individual elements observed on (CH3NH3)3SbBiI9/NiO substrate shown in Fig. 4f

Figure 5 shows the photo I–V characteristics of solar cells with best PCE using the lead-free photo absorbing materials ((CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, and (CH3NH3)3SbBiI9) with different structure (n-i-p and p-i-n, in Fig. 1), measured under the simulated sun light (AM 1.5, 100 mW cm−2). The observed PCE parameters are summerized in Table 7 and are significantly lower than the standard perovskite solar cells using CH3NH3PbI3 [1,2,3], which would be attributed to the uneven surface morphology of lead free perovskite solar cells. Basically, the cells using porous-TiO2 electrode gave better FFs, which is due to the prohibition of short circuiting by porous TiO2 layer [29]. Actually, the cells without A3B2I9 conformal layers on porous TiO2 [(CH3NH3)3Sb2I9 (Table 3) and (CH3NH3)3SbBiI9 (Table 5)] also provide better FFs. On the other hand, the cells using NiO provide small FFs. Specially, the cells using (CH3NH3)3Bi2I9 on NiO performed the short circuitting (Fig. 5b, Table 2). However, in spite of the missing points of (CH3NH3)3Sb2I9 and (CH3NH3)3SbBiI9 on NiO (Tables 4, 6), the I–V curves in Fig. 5d and f show FFs over 0.3. The prohibition of short circuiting would be due to the slight presence of Sb3+ on NiO surface, which was not detected by the EDX analysis. It was interesting that the hysteresis of A3B2X9 solar cells on NiO layer were smaller than those on TiO2 layer, as CH3NH3PbI3 solar cells [30].

Fig. 5
figure 5

Photo I–V curves of lead-free A3B2X9-crystal solar cells measured under simulated irradiation (AM 1.5, 100 mW cm−2) as normal mesoscopic and inverted architectures respectively. The photoabsorption layers were (CH3NH3)3Bi2I9 (a, b), (CH3NH3)3Sb2I9 (c, d), and (CH3NH3)3SbBiI9 (e, f) crystal films. The photovoltaic characteristics were summarized in Table 7

Table 7 Photovoltaic performance of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, (CH3NH3)3SbBiI9 photon harvesting layer based mesoscopic normal and inverted planar architecture solar cells

2.2 A3BX6 crystals with Ag or Cu cation for the A site

Copper and silver are non-toxic material and especially, Cu+ is a cheap and earth abundant one. The structure of A3BX6 was regulated by the ratio of elemental materials in the DMSO solution. Hence, CuI and BiI3 were mixed in 3:1 molar ratio to be Cu3BiI6. AgI and BiI3 were mixed in 3:1 molar ratio to be Ag3BiI6. In case of Ag3BiI3(SCN)3, AgSCN and BiI3 were mixed in 3:1 molar ratio. The XRD patterns of Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6 as casted films are shown in Fig. 6a. The film made by Ag3BiI6 signifies the trigonal structure with R \(\overline{3}\) m space group [31]. The space group determination of relatively new materials Ag3BiI3(SCN)3 and Cu3BiI6 could not be realized. Now, we are analyzing the exact crystal structures of Ag3BiI3(SCN)3 and Cu3BiI6 as the further research works. Figure 6b shows the UV–vis absorption spectra observed. The Ag3BiI6 and Cu3BiI6 films show higher absorption than the Ag3BiI3(SCN)3 one. Due to the absorbance onset of Ag3BiI6 and Cu3BiI6 at around 680 nm, the films were dark brown. On the other hand, Ag3BiI3(SCN)3 film was pale yellowish. The difference of Ag3BiI6 and Cu3BiI6 were the variation of spectra at around 580 nm. The exact absorption coefficients were not shown in the figure due to the inhomogeneous structures of films, which will be shown later.

Fig. 6
figure 6

XRD patterns (a) and UV–vis absorption spectra (b) of Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6 casted on the glass substrates. The insets were the pictures of Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6 films

Figure 7 shows the surface morphologies of Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6 films on porous-TiO2 and planer-NiO substrates which were observed by the SEM. Although the Ag3BiI6 films seems to be uniform observed by eyes, the nanoscale morphology in this SEM image scaled at 1 µm was relatively uniform (Fig. 7a). It was noticed that there were 1 μm-sized small grains of Ag3BiI6 on porous-TiO2 (Fig. 7a). However, Ag3BiI6 on planer NiO became the flakey layer with cracks, and exact crystals were not observed in the SEM image (Fig. 7b). For the other silver cation and mixed anion based Ag3BiI3(SCN)3 film, the SEM image with 10 μm scale shows the leaf-like crystal pattern on porous-TiO2 substrate and the grain structure on planer-NiO substrate with space between the crystals as the non-uniform substrate coverage resulting in the limitation in photon harvesting and charge collection (Fig. 7c, d). The morphologies of Cu3BiI6 were small grains with around 0.1–0.5 μm which were dispersed homogeneously on porous-TiO2 and planer-NiO substrates (Fig. 7e, f).

Fig. 7
figure 7

SEM images observed for elemental analysis of A3BX6 films [Ag3BiI6: (a, b); Ag3BiI3(SCN)3: (c, d); Cu3BiI6: (e, f)] grown on different substrates [TiO2: (a, c, e); NiO: (b, d, f)]. The marked spots indicate the positions of elemental analysis, which is related to the results shown in Tables 8, 9, 10, 11, 12 and 13

The individual elemental distribution (EDX analysis) involved of surface morphology of Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6 films are shown in Tables 8, 9, 10, 11, 12 and 13. The analyses spots were chosen randomly and are shown on its SEM images in Fig. 7. Although A3BX6 crystals of Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6 distributed relatively homogeneously than A3B2X9 ones using CH3NH3 + cations (Fig. 3 and Tables 1, 2, 3, 4, 5, 6), still the materials were dispersed with inhomogeneity about the elemental ratio (Tables 8, 9, 10, 11, 12, 13). It can be noticed that the amount of Bi was smaller than A site monovalent cations (Ag+ and Cu+). Specially, it was prominent that the amount of Ag+ was quite higher than that of Bi3+, which should be improved for the further progress.

Table 8 Elemental distribution (atomic %) of the individual elements observed on a particular spot on Ag3BiI6/TiO2 substrate shown in Fig. 8a
Table 9 Elemental distribution (atomic %) of the individual elements observed on a particular spot on Ag3BiI6/NiO substrate shown in Fig. 8b
Table 10 Elemental distribution (atomic %) of the individual elements observed on Ag3BiI3(SCN)3/TiO2 substrate shown in Fig. 8c
Table 11 Elemental distribution (atomic %) of the individual elements observed on Ag3BiI3(SCN)3/NiO substrate shown in Fig. 8d
Table 12 Elemental distribution (atomic %) of the individual elements observed on Cu3BiI6/TiO2 substrate shown in Fig. 8e
Table 13 Elemental distribution (atomic %) of the individual elements observed on Cu3BiI6/NiO substrate shown in Fig. 8f

The photo I–V characteristics of these A3BX6 (Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6) solar cells with different structure (n-i-p and p-i-n, in Fig. 1) under simulated irradiation of 1 SUN are shown in Fig. 8, which were the I–V curves for the best PCE in the series. The photovoltaic parameters were summarized in Table 14. Basically, Voc using porous-TiO2 layer were higher than using planer-NiO one. The effect of hysteresis on NiO layer was larger, which were different from A3B2X9 [(CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, and (CH3NH3)3SbBiI9] solar cells (Fig. 5). But, specially, the cells using Ag3BiI6 on NiO and Ag3BiI3(SCN)3 on TiO2 perform the strong hysteresis and large overshooting at the reverse voltage scanning, which is the overestimation of photovoltaic results. The best J SC and PCE in this work using trivalent B-site cation crystals are shown in Fig. 8a with the cell configuration of <FTO/cp-TiO2/mp-TiO2/Ag3BiI6/spiro-OMeTAD/Au>. Although there was the hysteresis, the variation was not large as other combinations. The observed PCE (reverse scan) of Ag3BiI6 based photon harvesting layer attained 1.08% (Fig. 8a, Table 14), which is comparable of reported PCE using AgBi2I7 based photo absorber [32]. This relatively-high PCE would be due to the pinhole-less surface morphology. However, the association of Ag3BiI6 layer in inverted architecture could result in only 0.32% PCE (Fig. 8b and Table 14). The IPCE of the Ag3BiI6 based mesoscopic solar cell is shown in Fig. 8g, the close matching of Jsc of 1.78 mA cm−2 is observed. We tried to measure the IPCE of other lead free solar cell, but IPCE measurements could not be managed due to the low J SC values. It was interesting that there was negligible hysteresis at the photo I–V measurements of <FTO/cp-TiO2/mp-TiO2/Cu3BiI6/spiro-OMeTAD/Au> cells, which would have the significance for the further research efforts in the future.

Fig. 8
figure 8

Photo I–V curves of lead-free A3BX6-crystal solar cells measured under simulated irradiation (AM 1.5, 100 mW cm−2) as normal mesoscopic and inverted architectures respectively. The photoabsorption layers are Ag3BiI6 (a, b), Ag3BiI3(SCN)3 (c, d), and Cu3BiI6 (e, f) crystal films, and IPCE spectrum of the mesoscopic structure with the Ag3BiI6 layer (g). The photovoltaic characteristics were summarized in Table 14

Table 14 Photovoltaic performance of Ag3BiI6, Ag3BiI3(SCN)3 photon harvesting layer based mesoscopic normal and inverted planar architecture solar cells (Fig. 8g)

3 Conclusions

In summary, we fabricated the lead-free hybrid solar cells in standard mesoscopic and inverted architecture incorporating various solution processed photon harvesting layer of (CH3NH3)3Bi2I9, (CH3NH3)3Sb2I9, (CH3NH3)3SbBiI9, Ag3BiI6, Ag3BiI3(SCN)3 and Cu3BiI6. The best PCE results was obtained using <FTO/cp-TiO2/mp-TiO2/Ag3BiI6/spiro-OMeTAD/Au> structure with 1.08% of power conversion efficiency (PCE). The grown morphology of A3B2X9 and A3BX6 crystals can be different on the substrates (porous TiO2 or planar NiO layers). The unevenness can be the hindrance for the improvement of photovoltaic effects. Hence, it is advisable to use lead-free element which can show uniform coverage of the substrate.

Other interesting findings in this report were the prohibition of short circuiting of solar cells using Sb deposition on NiO layer and the diminishment of photo I–V hysteresis using porous-TiO2/Cu3BiI6 combination, which can be considered for the further progresses.

4 Experimental details

All chemicals were of reagent grade quality and used without any further processing. Antimony iodide (SbI3) was purchased from Yanagishima Pharmaceutical Co. Ltd. Bismuth iodide (BiI3), Silver iodide (AgI), Nickel (II) acetylacetonate, Phenyl-C61-butyric acid methyl ester (PCBM) and bathocuproine (BCP) were purchased from Aldrich. Copper(I) iodide (CuI) from Kanto Chemical Ltd., Silver thiocyanate (AgSCN) from Wako Pure Chemical Industries Ltd., and Methylammonium iodide was purchased from TCI respectively. Dimethyl sulfoxide (DMSO), Chlorobenzene, and Methanol were purchased from Wako. γ-Butyrolactone (GBL) and Acetonitrile were procured from Chameleon Reagent and Kanto Chemical Co. Inc. respectively.

FTO glass (TEC-15, t = 2.1 mm), purchased from Pilkington were cut in an appropriate size and cleaned ultrasonically with detergent water, distilled water and ethanol, respectively each for 15 min. The cleaned FTO glass was passed through UV-O3 treatment for 10 min to remove the organic impurities. To fabricate the standard mesoscopic n-i-p solar cell, diluted TAA solution (Titanium di isopropoxide bis(acetylacetonate)/Sigma-Aldrich) of (200 µL in 7.5 mL ethanol) was aerosol spray coated maintaining the substrate temperature 500 °C. After arriving the ambient temperature the coated FTO glass was passed through the UV-O3 treatment and, diluted PST-30NRD TiO2 solution (1:3.5 wt/wt) was coated by spinning process at 5000 rpm for 30 s. The mesoporous TiO2 coated substrate was dried out at 120 °C for 5 min and again baked at 500 °C for 30 min in a furnace.

To make the perovskite layer, the substrate was treated for 5 min for UV-O3 treatment and transferred inside N2 filled glove box. To prepare the (CH3NH3)3Sb2I9 film, SbI3 and methylammonium iodide was mixed in DMSO:GBL (1:1) solvent and 0.25 M concentration of (CH3NH3)3Sb2I9 was maintained by keeping the molar ratio as CH3NH3I:SbI3 = 3:2 molar ratio. To prepare the (CH3NH3)3Bi2I9 film, BiI3 and methylammonium iodide was mixed to maintain the 0.25 M concentration of (CH3NH3)3Bi2I9 in DMSO:GBL (1:1) by keeping the molar ratio as CH3NH3I:BiI3 = 3:2. To prepare the (CH3NH3)3SbBiI9 film, SbI3, BiI3 and methylammonium iodide was mixed to maintain the 0.25 M concentration of (CH3NH3)3SbBiI9 in DMSO:GBL (1:1) by keeping the molar ratio as CH3NH3I:SbI3:BiI3 = 3:1:1. Semiconductor Ag3BiI6 layer overcoated TiO2 mp substrate was prepared by making a 0.5 M solution of AgI and BiI3 in DMSO solvent with 0.5 M concentration of Ag3BiI6 by keeping the molar ratio as AgI:BiI3 = 3:1. To fabricate and explore the photon harvesting properties of Ag3BiI3(SCN)3, BiI3 and Silver thiocyanate (AgSCN) was mixed in 1:3 molar ratio to maintain the 0.5 M concentration in DMSO solvent. Copper based photon harvesting layer (Cu3BiI6) was prepared by mixing the CuI and BiI3 in DMSO solvent with 0.4 M concentration of Cu3BiI6 by keeping the molar ratio as CuI:BiI3 = 3:1.

In all cases, the solution temperature was maintained at 80 °C and the heated solution was spun on mp TiO2 substrate with 2000 rpm for 30 s and preserved on a hot plate at 80 °C for 30 min to grow the respective photo absorber layer.

To collect the holes smoothly a hole selective layer spiro-OMeTAD (2,2,7,7-Tetrakis(N,Ndi-p-methoxy phenylamine)-9,9-spirobifluorene)of 28.9 mg in 400 µL chlorobenzene, with additives [11.5 µL t-butyl pyridine (Sigma-Aldrich), 7 µL Li-TFSI (520 mg in 1 mL acetonitrile) and 8.8 µL Co Complex (40 mg in 0.1 mL acetonitrile)] was spun on a cold substrate of lead-free prepared semiconducting layer at 4000 rpm for 20 s and stored for drying under dark. Finally, gold layer was thermally evaporated and deposited to maintain 80 nm thickness, which works as a metal back contact layer.

The inverted structure hybrid solar cell was fabricated by coating Nickel (II) acetylacetonate in acetonitrile (0.04 M) solution by spray process while maintaining the UV-O3 treated FTO glass temperature at 500 °C. After reaching the ambient temperature the coated NiO substrate was transferred to N2 filled glove box for lead-free semiconductor layer preparation. The preparation of individual photo harvesting layer was followed as described previously. On top of the lead-free photo absorber layer, a 20 mg/mL PCBM in chlorobenzene was spun at 1000 rpm for 60 s and kept under ambient N2 for 30 min to dry it out. Subsequently, a BCP solution of 1 mg/mL in methanol was spun at 1000 rpm for 60 s. Finally, a silver metal contact was prepared by thermally evaporation process maintaining its thickness 80 nm.

The I–V characteristics were measured by an AM 1.5G solar simulator equipped with a 500 W Xe lamp (YSS-80A, Yamashita Denso) by placing a black 0.09 cm2 mask area. A reference Si photodiode (Bunkou keiki co. ltd., Japan) was used to calibrate the power of the solar simulator light. The I–V characteristics were obtained by applying an external bias to the fabricated solar cell and measurement of generated photocurrent was performed with a DC voltage current source (Agilent, B2901A). SEM images were obtained by HITACHI Microscope TE3030 and EDX analysis was performed with Oxford Instruments, X-stream-2.

References

  1. A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc. 131, 6050 (2009)

    Article  Google Scholar 

  2. S. Ito, APL Mater. 4, 91504 (2016)

    Article  Google Scholar 

  3. W.S. Yang, B.-W. Park, E.H. Jung, N.J. Jeon, Y.C. Kim, D.U. Lee, S.S. Shin, J. Seo, E.K. Kim, J.H. Noh, S. Il Seok, Science 356, 1376 (2017)

    Article  Google Scholar 

  4. T.A.N. Peiris, A.K. Baranwal, H. Kanda, S. Fukumoto, S. Kanaya, L. Cojocaru, T. Bessho, T. Miyasaka, H. Segawa, S. Ito, Nanoscale 9, 5475 (2017)

    Article  Google Scholar 

  5. A.K. Baranwal, S. Kanaya, T.A.N. Peiris, G. Mizuta, T. Nishina, H. Kanda, T. Miyasaka, H. Segawa, S. Ito, Chemsuschem 9, 2604 (2016)

    Article  Google Scholar 

  6. S. Ito, G. Mizuta, S. Kanaya, H. Kanda, T. Nishina, S. Nakashima, H. Fujisawa, M. Shimizu, Y. Haruyama, H. Nishino, Phys. Chem. Chem. Phys. 18, 27102 (2016)

    Article  Google Scholar 

  7. S. Bae, S. Kim, S.-W. Lee, K.J. Cho, S. Park, S. Lee, Y. Kang, H.-S. Lee, D. Kim, J. Phys. Chem. Lett. 7, 3091 (2016)

    Article  Google Scholar 

  8. A. Babayigit, A. Ethirajan, M. Muller, B. Conings, Nat. Mater. 15, 247 (2016)

    Article  Google Scholar 

  9. S.F. Hoefler, G. Trimmel, T. Rath, Monatsh. Chem. 148, 795 (2017)

    Article  Google Scholar 

  10. E.S. Parrott, R.L. Milot, T. Stergiopoulos, H.J. Snaith, M.B. Johnston, L.M. Herz, J. Phys. Chem. Lett. 7, 1321 (2016)

    Article  Google Scholar 

  11. F. Hao, C.C. Stoumpos, D.H. Cao, R.P.H. Chang, M.G. Kanatzidis, Nat. Photonics 8, 489 (2014)

    Article  Google Scholar 

  12. A. Babayigit, D. Duy Thanh, A. Ethirajan, J. Manca, M. Muller, H.-G. Boyen, B. Conings, Sci. Rep. 6, 18721 (2016)

    Article  Google Scholar 

  13. T. Krishnamoorthy, H. Ding, C. Yan, W.L. Leong, T. Baikie, Z. Zhang, M. Sherburne, S. Li, M. Asta, N. Mathews, S.G. Mhaisalkar, J. Mater. Chem. A 3, 23829 (2015)

    Article  Google Scholar 

  14. D. Cortecchia, H.A. Dewi, J. Yin, A. Bruno, S. Chen, T. Baikie, P.P. Boix, M. Grätzel, S. Mhaisalkar, C. Soci, N. Mathews, Inorg. Chem. 55, 1044 (2016)

    Article  Google Scholar 

  15. S. Öz, J.-C. Hebig, E. Jung, T. Singh, A. Lepcha, S. Olthof, F. Jan, Y. Gao, R. German, P.H.M. van Loosdrecht, K. Meerholz, T. Kirchartz, S. Mathur, Sol. Energy Mater. Sol. Cells 158, 195 (2016)

    Article  Google Scholar 

  16. J.-C. Hebig, I. Kühn, J. Flohre, T. Kirchartz, A.C.S. Energ, Lett. 1, 309 (2016)

    Google Scholar 

  17. B.W. Park, B. Philippe, X. Zhang, H. Rensmo, G. Boschloo, E.M.J. Johansson, Adv. Mater. 27, 6806 (2015)

    Article  Google Scholar 

  18. A. Kulkarni, T. Singh, M. Ikegami, T. Miyasaka, RSC Adv. 7, 9456 (2017)

    Article  Google Scholar 

  19. T. Singh, Y. Udagawa, M. Ikegami, H. Kunugita, K. Ema, T. Miyasaka, APL Mater. 5, 16103 (2017)

    Article  Google Scholar 

  20. G. Volonakis, A.A. Haghighirad, R.L. Milot, W.H. Sio, M.R. Filip, B. Wenger, M.B. Johnston, L.M. Herz, H.J. Snaith, F. Giustino, J. Phys. Chem. Lett. 8, 772 (2017)

    Article  Google Scholar 

  21. G. Volonakis, M.R. Filip, A.A. Haghighirad, N. Sakai, B. Wenger, H.J. Snaith, F. Giustino, J. Phys. Chem. Lett. 7, 1254 (2016)

    Article  Google Scholar 

  22. E.T. McClure, M.R. Ball, W. Windl, P.M. Woodward, Chem. Mater. 28, 1348 (2016)

    Article  Google Scholar 

  23. I. Turkevych, S. Kazaoui, E. Ito, T. Urano, K. Yamada, H. Tomiyasu, H. Yamagishi, M. Kondo, S. Aramaki, Chemsuschem 8575, 1 (2017)

    Google Scholar 

  24. H.C. Sansom, G.F.S. Whitehead, M.S. Dyer, M. Zanella, T.D. Manning, M.J. Pitcher, T.J. Whittles, V.R. Dhanak, J. Alaria, J.B. Claridge, M.J. Rosseinsky, Chem. Mater. 29, 1538 (2017)

    Article  Google Scholar 

  25. F. Wei, Z. Deng, S. Sun, F. Zhang, D.M. Evans, G. Kieslich, S. Tominaka, M.A. Carpenter, J. Zhang, P.D. Bristowe, A.K. Cheetham, Chem. Mater. 29, 1089 (2017)

    Article  Google Scholar 

  26. H.-S. Kim, J.-Y. Seo, N.-G. Park, J. Phys. Chem. C 120, 27840 (2016)

    Article  Google Scholar 

  27. H.-S. Kim, J.-Y. Seo, N.-G. Park, Chemsuschem 9, 2528 (2016)

    Article  Google Scholar 

  28. M. Saliba, J.-P. Correa-Baena, M. Graetzel, A. Hagfeldt, A. Abate, Angew. Chemie Int. Ed. (2017). doi:10.1002/anie.201703226

    Google Scholar 

  29. G. Murugadoss, G. Mizuta, S. Tanaka, H. Nishino, T. Umeyama, H. Imahori, S. Ito, APL Mater. 2, 81511 (2014)

    Article  Google Scholar 

  30. H.-S. Kim, I.-H. Jang, N. Ahn, M. Choi, A. Guerrero, J. Bisquert, N.-G. Park, J. Phys. Chem. Lett. 6, 4633 (2015)

    Article  Google Scholar 

  31. T. Oldag, T. Aussieker, H.-L. Keller, C. Preitschaft, A. Pfitzner, ChemInform 36, 44227 (2005)

    Article  Google Scholar 

  32. Y. Kim, Z. Yang, A. Jain, O. Voznyy, G.-H. Kim, M. Liu, L.N. Quan, F.P. García de Arquer, R. Comin, J.Z. Fan, E.H. Sargent, Angew. Chemie Int. Ed. 55, 9586 (2016)

    Article  Google Scholar 

Download references

Authors’ contributions

HM, HS performed the experiments. AKB, HM, NS analysed the data. AKB, HK, NS, and SI designed the experiments. AKB, HM, HS, HK, SK, NS, YS, MI, YN, KY, TM, TU, and HI provided an equal contribution in useful discussion and modification. AKB and SI wrote the manuscript. All authors read and approved the final manuscript.

Competing interests

The authors declare that they have no competing interests.

Availability of data and materials

The authors have no any more data to share.

Consent to publication

Not applicable.

Ethics approval and consent to participate

Not applicable.

Funding and Acknowledgements

We would like to express sincere thanks to the Advanced Low Carbon Technology Research and Development Program (ALCA-JST), Japan.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Author information

Authors and Affiliations

Authors

Corresponding authors

Correspondence to Ajay Kumar Baranwal or Seigo Ito.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Baranwal, A.K., Masutani, H., Sugita, H. et al. Lead-free perovskite solar cells using Sb and Bi-based A3B2X9 and A3BX6 crystals with normal and inverse cell structures. Nano Convergence 4, 26 (2017). https://doi.org/10.1186/s40580-017-0120-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s40580-017-0120-3

Keywords