Vapor Annealing Controlled Crystal Growth and Photovoltaic

Subscriber access provided by UNIV OF DURHAM

Functional Inorganic Materials and Devices

Vapour Annealing Controlled Crystal Growth and Photovoltaic Performance of Bismuth Triiodide Embedded in Mesostructured Configurations Ashish Kulkarni, Trilok Singh, Ajay Kumar Jena, Peerathat Pinpithak, Masashi Ikegami, and Tsutomu Miyasaka ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b00430 • Publication Date (Web): 06 Mar 2018 Downloaded from http://pubs.acs.org on March 7, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Vapour Annealing Controlled Crystal Growth and Photovoltaic Performance of Bismuth Triiodide Embedded in Mesostructured Configurations Ashish Kulkarni*, Trilok Singh, Ajay K. Jena, Peerathat Pinpithak, Masashi Ikegami, Tsutomu Miyasaka* Graduate School of Engineering, Toin University of Yokohama, 1614 Kurogane-cho, Aoba, Yokohama, Kanagawa, Japan. ABSTRACT Low stability of organic-inorganic lead halide perovskite and toxicity of lead (Pb) still remains a concern. Therefore, there is a constant quest for alternative non-toxic and stable light absorbing materials with promising optoelectronic properties. Herein, we report about non-toxic bismuth triiodide (BiI3) photovoltaic device prepared using TiO2 mesoporous film and spiro-OMeTAD as electron and hole transporting materials, respectively. Effect of annealing methods (e.g. thermal annealing-TA, solvent vapour annealing-SVA, Petri dish covered recycled vapour annealing-PRVA) and different annealing temperatures (90, 120, 150 and 180 OC for PR-VA) on BiI3 film morphology have been investigated. As found in the study, grain size increased and film uniformity improved with temperature raised from 90 to 150 OC. The photovoltaic devices based on BiI3 films processed at 150 OC with PR-VA treatment showed power conversion efficiency (PCE) of 0.5% with high reproducibility, which is, so far, the best PCE reported for BiI3 photovoltaic device employing organic HTM, owing to increase in grain size and uniform morphology of BiI3 film. These devices showed stable performance even after 30 days of exposure to 50% relative humidity, and after 100 OC heat stress and 20 min light soaking test. More importantly, the study reveals many challenges and rooms (discussed in the details) for further development of the BiI3 photovoltaic devices.

KEYWORDS: Bismuth iodide, Lead-free, Petri dish covered recycled vapour annealing, Annealing temperature, Mesostructured architecture, Stability. ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

INTRODUCTION Organic-inorganic lead halide perovskite solar cells (PSCs) have shown a rapid rise in power conversion efficiency (PCE),1, 2, 3 going beyond 22%, which is approaching the theoretical efficiency limit of the single junction solar cell. However, the cells have been facing a challenge of long-term stability4 because most of the widely used perovskites with organic cations degrade to PbI2 on exposure to humidity,5 UV-light6, 7 and heat.4 It has been reported that PbI2, formed after degradation of perovskite, is highly toxic to human reproductive and nervous systems.8 Hence, to address the Pb toxicity issue, tin (Sn), 9 germanium (Ge), 10 bismuth (Bi), 11 , 12 and antimony (Sb)13 based perovskite materials have been explored as alternate lead-free materials. Sn and Ge perovskites suffer from structural instabilities due to spontaneous oxidation when exposed to the ambient atmosphere14 and very recently, toxicity concern of Sn-perovskite has been also raised.8 Bi based perovskites, such as (CH3NH3)3Bi2I9 and Cs3Bi2I9 have not shown impressive performance due to relatively high band gap (~2eV), poor surface morphology, high intrinsic carrier densities and high exciton binding energy.11, 12, 15, 16 Even after several attempts by various groups, highest efficiency achieved so far is 1.64%.11, 12, 15, 16, 17, 18 Double perovskites (having a 3D structure) formed by combining trivalent metal halide (Bi3+, Sb3+) with a monovalent metal halide (Ag+, Cu+) have also been demonstrated with promising carrier lifetime19 but Savory et al. pointed out limitations of working with these materials because of high effective masses and large indirect bandgaps and suggested to expand the search beyond Ag-Bi double perovskites.20 Hence, finding an alternate eco-friendly material with lower band gap and promising optical properties suitable for the efficient photovoltaic system is critically important.

ACS Paragon Plus Environment

Page 2 of 27

Page 3 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


BiI3 has been used as one of the precursor components for synthesizing bismuth perovskite materials.11, 12, 15, 18, 19 However, interestingly, BiI3 itself belongs to a layered heavy metal semiconductors group and possesses interesting optical properties,21 such as higher optical absorption (>105 cm-1) than Si and GaAs,22, 23 electron diffusion length of 4.9 µm, large static dielectric constant etc.21 Besides, BiI3 has an outer shell electronic configuration of 6s2 which leads to dispersed valence band, high dielectric constant, shallow intrinsic point defects all of these are serviceable properties of defect-tolerant material.21 Moreover, room temperature photoluminescence (PL) of BiI3 at ~530 nm, both in vapor and solution-processed thin films,21 endorse its promising application in optoelectronic devices. From the application point of view, BiI3 has been previously investigated intensively in gamma-ray detectors24 and X-Ray radiation25 and recently, theoretical21 and experimental studies26, 27 have shown its potential to be used as a photovoltaic absorber. However, so far, only a few attempts have been made to integrate BiI3 into photovoltaic devices. Its first use in PV was as an HTM in organic solar cells with fullerenebased light absorber as an active layer.28 Recently, Lehner et al. and Hamdeh et al. employed it in TiO2 compact layer (CL) based planar structure solar cell and reported an efficiency of 0.3% (with solution processed organic HTM)26 and 1% (with vapour processed inorganic HTM),27 respectively. To best of our knowledge integration of BiI3 in mesoporous architecture and its morphological evolution with annealing conditions and its effect on the device performance, which is important for evaluation of actual potential of BiI3, have not been reported. Herein, for the first time, we report about incorporation of BiI3 in TiO2 mesostructured architecture and changes in its morphology at various annealing treatment such as thermal annealing, solvent vapour annealing (SVA), 29 and Petri dish covered recycled vapour annealing (PR-VA). 30 Additionally, effect of different annealing temperatures (with PR-VA treatment) on the



morphology, crystal orientation and device performance was investigated. The major difference between SVA and PR-VA treatment is that in SVA the spin-coated substrates are covered with Petri dish with few microliters of solvent drops within the inner boundary wall of Petri dish during the heating step while in PR-VA the active-layer coated substrates are covered with Petri dish (without any solvent drops) to maintain the solvent atmosphere (which is evaporated from the spin-coated substrates). EXPERIMENTAL SECTION: Device fabrication: Cleaning of FTO substrate: F-doped SnO2 (FTO) conductive glass substrates were washed by ultrasonic treatment (for 15 minutes) sequentially with detergent (2% Hellmanex in water), deionized water (DI), and ethanol. The substrates were given UV-ozone treatment for 20 min before coating TiO2 compact layer. TiO2 compact layer coating: Solution of titanium diisopropoxide bis(acetylacetonate) (Ti acac) (75 wt% in isopropanol, Sigma-Aldrich) in ethanol (99.99%, Wako) was spray-coated on to cleaned FTO coated glass substrates at 500 OC to obtain ~50 nm thick TiO2 compact layer (CL) and allowed the substrates to remain at the same temperature for 30 min followed by cooling down naturally to room temperature. TiO2 Mesoporous layer coating: Anatase TiO2 mesoporous layer was spin-coated (3000 rpm/30 sec) using a commercially available TiO2 paste (18NR-T, Dyesol, particle size ~20 nm) diluted in ethanol (weight ratio of 1:4), which was followed by sintering at 500 OC for 1 hour in a muffle furnace. After cooling down to room temperature, the mesostructured substrates were given 40 mM TiCl4 treatment at 70 OC for 1 hour. The substrates were rinsed and cleaned with


Page 4 of 27

Page 5 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


distilled water and ethanol and were again sintered at 500 OC for 30 min. TiO2 mesoporous coated substrates were given UV-ozone treatment for 10 min prior to the BiI3 layer deposition. Active layer and HTM deposition: 1 M bismuth triiodide (BiI3) solution in DMF was prepared by stirring on a hot plate at 50 OC for 1 hour prior to the coating. The BiI3 solution was spincoated on TiO2 mesoporous coated substrates at 1500 rpm for 30 sec. After spin coating, the substrates were transferred on to the hot plates which were already set at 90 OC, 120 OC, 150 OC and 180 OC. The films were annealed for 1 hour with 5 µL DMF solvent at the inner boundary wall of Petri dish covering the film for SVA treatment and without any solvent for PR-VA treatment. It is to be noted that for SVA and PR-VA treatment, the Petri dish cover was used during the entire heating step (i.e. 1 hour). The diameter and height of Petri dish used were 45 mm, and 18 mm respectively. For thermal annealing treatment, the substrates were heated (at same temperature) without any Petri dish cover for 1 hour. 8 wt % spiro-OMeTAD solution in chlorobenzene containing additives of lithium bis(trifluoromethanesulfonyl) imide and 4-tertbutylpyridine was spin-coated on BiI3 coated substrates at 3000 rpm for 30 sec. The HTM coated substrates were kept overnight in the air for oxidation of spiro-OMeTAD. Thermal deposition of Au metal electrode was done to complete the cell. To note, all the chemicals were used as received without any further purification Characterization: Solar cell characteristics (current-voltage measurement) of all devices were measured using a Peccell Technologies PEC-L01 solar simulator with a Keithley 2400 source meter under 1 sun illumination (AM 1.5G, 100 mW/cm2). The EQE spectra of the device were measured with Peccell Technologies, PEC-S20 action spectrum measurement setup. The optical, morphological



and structural analysis were carried out using UV-vis spectrophotometer (UV-vis 1800, Shimadzu), scanning electron microscope (SU8000, HITACHI), and X-ray diffractometer (D8 Discover, Bruker) with Cu Kα radiation source, respectively. The active areas of the cells were 0.09 cm2. RESULTS AND DISCUSSION In order to estimate the band gap energy of BiI3, we measured UV-vis absorption spectrum (Fig. 1a) of the optimized BiI3 film. From the Tauc plot 31 (Fig. 1b), assuming an indirect band gap,21 we obtained the Eg of ~1.81 eV, which is in agreement with prior reports.21, 26

As morphology of the film influences performance of the devices significantly, we first tried to

see the effect of different annealing conditions like TA, SVA (inspired from the work of Hamdeh et al.27) and PR-VA on the morphology of BiI3 film. The substrates were covered with petri dish during the heating step for SVA (with 5 µL DMF solvent drops within the inner boundary wall of Petri dish) and PR-VA (without any solvent drops) while for TA treatment, BiI3 spin coated substrates were heated on a hot plate without any Petri dish cover. Fig. 2a depicts the schematic presentation of BiI3 films processed by TA, SVA and PR-VA method. In all the cases, colour of the BiI3 films transitioned from orange to grayish black (Fig. S1a) during the heating process, indicating the decomposition of DMF-BiI3 complex and formation of BiI3 in the final film.27 To confirm this, we measured the UV-Vis spectra of the as-prepared BiI3 film (orange color), which showed absorption band at around ~500 nm (Fig. S1b) corresponding to the solvated intermediate BiI3-DMF complex. The absorption band of BiI3 thin film (Fig. 1a) was strikingly different from that of the BiI3-DMF complex. In case of SVA and PR-VA process, slow decomposition was observed due to slow evaporation of solvent DMF (due to Petri dish cover and solvent vapour atmosphere). Top surface scanning electron micrograph (SEM) image of TA


Page 6 of 27

Page 7 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


- processed (without petri dish cover) BiI3 film (Fig. 2b) shows apparently small grains and large pinholes. Although DMF-SVA processed BiI3 film showed improvement in grain size (Fig. 2c), indicating that DMF vapour facilitates migration and promotes grain growth, we still observed a large number of pinholes in the film (Fig. 2c), which is in contrast to the previous report.27 Moreover, lower magnification SEM image of TA and SVA processed films (Fig. S2a, b) shows large variation in grain size with a high degree of non-uniformity (BiI3 crystals accumulate, forming a non-homogeneous layer) and large gaps throughout the film. It is well known that rapid evaporation of solvent from polymeric thin films can cause kinetically constrained growth resulting in small grains32 and non-uniform growth of these grains can end up in the non-uniform film. This might be the case when the film was TA-processed (rapid solvent evaporation (Fig. 2a, S2a)), which resulted in the non-homogeneous layer. Even though many previous reports have shown that SVA treatment improves the morphology and performance of lead halide perovskite29 and as well as bismuth halide based solar cells,27 our group recently reported poor reproducibility of perovskite formation and cell performance with SVA treatment, owing to non-uniform solvent vapour atmosphere and excess vapor concentration (during SVA treatment) (Fig. 2a).30 This can be the reason for non-uniform grain distribution and enormous BiI3 grain growth observed (Fig. 2 c and S2b) in our study. Additionally, in comparison to the previous report,27 the difference in morphology in our case can also be due to the use of a different solvent (DMF instead of THF) to dissolve BiI3. Nevertheless, on the contrary to TA and SVA, we observed improvement in morphology and crystal growth with better uniformity (Fig. 2d) in case of PR-VA treatment, that is, confining BiI3 coated substrates in a Petri dish without additional solvent drops during the heating step. The pinholes obtained from PR-VA treatment were relatively small in size (Fig. 2d) compared to BiI3 layer processed by TA and SVA treatment (Fig. 2 b, c). The merit of PR-VA



treatment is that the solvent evaporated from the spin-coated BiI3 film gets uniformly distributed (due to Petri dish cover) and in the presence of recycled solvent atmosphere (within the optimum time), the grains grow uniformly, resulting in non-coalescence of BiI3 grains (Fig. 2d, S2c). Such phenomenon has been observed in lead perovskite system by Numata et al.30 As PR-VA method resulted in more uniform and large grains, we performed our further studies on BiI3 film processed by this method and finally compared the photovoltaic performance (which will be discussed later) with TA and SVA processed BiI3 film.

Figure 1: (a) UV-Vis absorption spectrum and (b) corresponding Tauc plot of the optimized BiI3 thin film.


Page 8 of 27

Page 9 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2: (a) Schematic presentation and top surface SEM image of BiI3 films processed at 150 OC with (b) thermal annealing (TA), (c) solvent vapour annealing (SVA) and (d) Petri dish covered recycled vapour annealing (PR-VA) treatment.

As it is well known that pinholes in the layer cause loss in photovoltage and charge transport, further optimization of PR-VA method was needed to obtain pinhole-free BiI3 film. In the above study for comparison of TA, SVA, and PR-VA method, we annealed the BiI3 film at 150 OC, which is very close to the boiling point of DMF (153 OC). This processing temperature can cause rapid evaporation of solvent resulting in pinholes (Fig. 2). To overcome this, adopting the PR-VA method, we annealed the films at various temperatures (90 OC, 120 OC, 150 OC and 180 OC). Fig. 3 shows top surface and cross-section SEM images of BiI3 films investigated. As



expected, we observed variation in the morphology of BiI3 with the change in temperature. When annealed at 90 OC, the grains grew randomly (discontinuous and non-uniform) and therefore, the film contained both small grains with no defined grain boundaries and enormous grains at places (Fig. 3a, b, S3a). Such grain growth indicates progress of solvent vapour-assisted Ostwald ripening29 i.e. large grains grow at the cost of small grains in the presence of recycled DMF vapour.27 Additionally, some rod-like morphology with several cracks was also observed in some other region of the film when annealed at 90 OC (Fig. S4). Such small grains with indistinct grain boundaries (Fig. 3a, S3a) and rod-like morphology (Fig. S4) indicates incomplete crystallization of BiI3, which is also reflected as lesser intense peaks in the XRD pattern (Fig. 4). More details about XRD pattern is discussed later. The origin of different morphology (rod and spherical) within the same film is still unknown but it implies that polydispersity grain growth is happening in the film, which usually results in formation of amorphous material along with large number of pinholes.27 Besides, a trace amount of solvent might still remain within the matrix, making it an amorphous material. With increase in the temperature up to 150 OC, we observed change in morphology; an increase in grain size (Fig. 3 c-f) and improvement in uniformity (Fig. S3b, c). This change/improvement in morphology can be due to complete evaporation of DMF solvent i.e. complete crystallization of BiI3 films (as clear grain boundaries can be seen in Fig. 3c-f) and due to uniform grain growth in the presence of recycled DMF vapour within the Petri dish. We quantified the relation between annealing temperature and the domain size based on the surface SEM (Fig. 3) of the BiI3 film, as shown in Fig. S5. In comparison to 90 OC (which showed average grain size of ~250 nm), grains became larger (~350 nm) and uniform in the film annealed at 120 OC and they grew to largest size (~580 nm) in 150 OC case (Fig. S5). This confirmed that the 150 OC annealing condition is optimum for such uniform grain growth, which


Page 10 of 27

Page 11 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


must have resulted from uniform distribution and optimum time of exposure of the DMF vapour inside the Petri dish. Although 150 OC was the best among all cases examined here, it was not perfect as the BiI3 layer showed both pinhole-rich and pinhole free regions within the film (Fig. S6). With further increase in temperature, from 150 OC to 180 OC, the grain size decreased (~350 nm) (Fig. 3g, h, S5) with non-uniform growth (Fig. S3d). This is most likely due to rapid evaporation of the solvent (above the boiling point of DMF), causing rapid crystallization with less time for BiI3 grains to grow. In addition, visible partial loss of the BiI3 layer (a brown layer formed on the inner upper wall of Petri dish during heating process) was observed in the films annealed at 180 OC (Fig. S7). This can also be the reason for reduction in grain size. However, detail investigation is required to understand which component is lost from BiI3 at 180 OC. Despite of several attempts, pinholes were still present but less number of pinholes (due to large grain size and no pinholes at some other region of the film (Fig. 3e, S6)) was observed for BiI3 films processed at 150 OC. It is reported that BiI3 is a very soft material with a Vickers Hardness ranging from 12 to 15 (ref. 25) and when the material is heated or cooled down in adhesion to its neighboring layers, cracks and pinholes are easily formed.21 We suspect that such property of BiI3 may be the reason for the cracks (Fig. S4) and pinholes (Fig. 3, S3) observed in our study.



Figure 3: Top surface and cross-sectional SEM images of BiI3 films processed with PR-VA treatment at (a, b) 90 OC, (c, d) 120 OC, (e, f) 150 OC and (g, h) 180 OC.


Page 12 of 27

Page 13 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


We performed X-ray diffraction (XRD) measurements to study the crystal structure and crystal plane orientation of BiI3 deposited on the mesoporous layer. The XRD pattern was resolved into R3̅ space group and was in well agreement with previous reports.21 Additionally, we observed that, in comparison to the TA method, the intensity of BiI3 XRD peaks improved with PR-VA method (Fig. S8), implying higher crystallinity in the latter. Hence, we employed PR-VA method to further study the effect of temperature on XRD pattern of the BiI3 thin film. From Fig. 4a, we can see that BiI3 films annealed at 90 OC showed major characteristic peaks at ~12O ~12.7O, 27O, and 41O which are assigned to (003), (101), (113) and (300) crystal planes of BiI3 respectively.21 Fig. 4b and c show the zoomed in (101) and (113) diffraction peak at ~12O and ~27O respectively. The intensity of these two peaks gradually increased with the annealing temperature while intensity of the peak at 12O corresponding to (003) remained same for varying temperature. The peak at ~42O, corresponding to (300) showed slight enhancement in peak intensity with the annealing temperature from 90 OC to 120 OC and remained constant thereafter. This evidences that, with an increase in processing temperature, the BiI3 grain growth is more preferred along (101), (113), and (300) directions. Additionally, few other minor diffraction peaks appeared at around 14O, and 25.6O assigned to (012) and (021) crystal planes21 which showed very slight increment with increase in temperature. The weaker intensity at low processing temperature (90 OC) can be due to incomplete crystallization (small grains without clear grain boundaries (Fig. 3a, S3a) and rod-like morphology (Fig. S4) at some places). Even though no diffraction peak was observed from the solvent-mediated BiI3 we believe that a part of solvent DMF may exist within the matrix but as an amorphous phase. Previously, Hamdeh et al. also made similar observation, that is, less intense XRD peak attributing to the presence of THF



solvent embedded within the BiI3 matrix.27 With increase in temperature, the DMF gets evaporated from the matrix, leading to better crystallinity and hence intense XRD peaks.

Figure 4: (a) XRD pattern of the BiI3 films processed at various temperatures, its corresponding zoomed (b) (101) oriented peak and (c) (113) oriented peak. * and # corresponds to FTO and TiO2 respectively. Changes made in (a) (012) and (c) (021)

Solar cells using BiI3 as an active layer were fabricated with widely used spiro-OMeTAD with dopants as HTM. Fig. 5a depicts the energy level diagram26 and the cross-sectional SEM image of a BiI3 solar cell showing stacks of layers in the structured configuration of FTO/TiO2 (CL + mesoporous)/BiI3/Spiro-OMeTAD/Au. The effect of PR-VA treatment at various annealing temperatures on the device performance was investigated. J-V curves of best performing devices of all cases and the average photovoltaic parameters (obtained from 6 devices of each case) are shown in Fig. 5b and in Table 1, respectively. The statistics of the photovoltaic parameters obtained from 5 batches (50 devices of each kind) are shown in Fig. S9. From the statistics plot, we can see that the short-circuit current density (Jsc) increased with the increase in temperature (from 90 OC to 150 OC) (Fig. S9a), which can be credited to the larger grains (Fig. 3) and preferred crystallographic orientation along (101), (113), and (300) of BiI3 crystals (Fig. 4) as the thickness in all the cases studied are same (Fig. 3). We think that grain growth in (101), (113) and (300) directions favors improvement in Jsc and thereby, the device


Page 14 of 27

Page 15 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


performance. However, a detail investigation is required to confirm the effect of such preferred crystallographic orientation on Jsc. In Fig. 5b and S9, we can see that devices with BiI3 films processed at 180 OC, despite partial degradation, showed higher Jsc compared to that processed at 90 OC and 120 OC. This also makes us believe that the preferred crystallographic orientation along (101), (113), and (300) enhances the carrier transport, resulting in higher Jsc. However, in comparison to 150 OC, Jsc, Voc and FF of the cells (processed at 180 OC) were low, resulting in overall performance. From Fig. S9d, it is evident that among all the cases studied, BiI3 films processed at 150 OC demonstrated the best device performance; minimum, average and maximum PCE were 0.44%, 0.48%, and 0.5% respectively (PCE histogram plot in Fig. S10). It is to be noted that the obtained PCE (0.5%) in this study is the highest value obtained so far in BiI3 solar cells employing solution processed organic HTM and results were reproducible, as can be seen from the histogram plot (Fig. S10). Fig. 5c shows the forward and reverse scan J-V curves (with negligible hysteresis) of the best-performing device (processed at 150 OC). We also fabricated devices with TA and SVA processed film and as expected, the device showed poor performance in comparison to PR-VA processed BiI3 device (Fig. S10 and 11 and Table S1). Fig. 5d shows the incident photon to current conversion efficiency (IPCE) or external quantum efficiency (EQE) of the best performing device. As can be seen in the Fig. 5d, the spectrum covers the visible region with sharp absorption onset around ~720 nm and is in good agreement with previous report.27 Integration of IPCE spectrum yielded short-circuit photo-current density of 3.43 mA/cm2 which matches well with the Jsc value (3.6 mA/cm2) obtained from J-V measurement. Table 1: Photovoltaic parameters of BiI3 device J-V curves obtained with different processing conditions. The standard deviations are obtained from average of six devices.



Page 16 of 27

Processing Temperature (OC)

Jsc (mA/cm2)

Voc (V)

FF

PCE (%)

90

1.32 ± 0.03

0.3 ± 0.01

0.3 ± 0.03

0.12 ± 0.02

120

1.93 ± 0.05

0.3 ± 0.01

0.35 ± 0.02

0.22 ± 0.02

150

3.4 ± 0.02

0.31 ± 0.01

0.4 ± 0.01

0.49 ± 0.01

180

2.4 ± 0.03

0.22 ± 0.01

0.32 ± 0.03

0.21 ± 0.02

Figure 5: (a) Energy level diagram and device cross-section image of the BiI3 photovoltaic device, (b) best performing BiI3 device J-V curves processed with different annealing temperature, (c) Champion device J-V curve with forward and reverse scan and its corresponding (d) EQE spectrum. The PCE of the cells was highest for the 150 OC PR-VA case essentially because of significantly higher Jsc and slightly improved FF. The films processed at 90 OC showed lowest PCE, as a result of lowest Jsc and FF. In this case, incomplete crystallization and large variation in the grain size (different kinds of morphology within the same film) might be responsible for


Page 17 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


lower Jsc and FF. With the increase in temperature to 120 OC and to 150 OC, the FF slightly improved (Fig. S9) due to improvement in the uniformity of grain growth which can form better interfacial contact with neighboring charge collecting layers. However, it can be noted that even though the Jsc improved effectively with an increase in processing temperature, FF didn’t improve much (for BiI3 films processed at 120 and 150 OC). Additionally, in all cases (90, 120, 150 OC) except for 180 OC, we observed low Voc (0.3 V), which showed no change with improvement in morphology. Voc degraded significantly in case of 180 OC (Fig. S9), which might be due to partial loss of BiI3. In general, there can be several reasons for lower Voc and FF in BiI3 cells; BiI3 has a lot of iodine vacancies33 and surface defects resulting in shallow traps33 which makes it much easier for charge carrier recombination (non-radiative recombination). Additionally, electron diffusion length of BiI3 is much less than lead perovskites and also the electron mobility is reported to be 30 times higher than the hole mobility.21 Such unbalanced charge mobilities and low diffusion length can cause inefficient charge transport. Moreover, BiI3 itself possesses high resistivity up to 108 to 109 Ω-cm, resulting in high series resistance. 34 Such high resistivity is one of the reasons responsible for short carrier lifetime (~200 ps). Large atomic mass of Bi3+ results in the formation of deeper 6s orbitals and its less contribution to valence band results in high hole effective masses.21 On the other hand, from the EQE spectrum (Fig. 5d), we can see relatively poor extraction of higher energy photons, which was attributed to the presence of defect states at TiO2/BiI3 interface, same as observed by Hamdeh et al.27 Additionally, we suspect that the presence of impurities in BiI3 can also cause such loss in high energy photons. Therefore, we believe, obtaining phase pure BiI3 film (by simple solution process) and choosing suitable n-type heterojunction layer are important and immediate steps for further improvement. In accordance



with prior reports26 and as shown in the schematic illustration (Fig. 5a), poor alignment of valence band maximum (VBM) between spiro-OMeTAD and BiI3 also limits efficient charge transport. All the above-mentioned reasons can be responsible for poor Voc and FF of the device. Hence, there are lots of rooms for further improvement of performance of BiI3 cells. We also investigated the stability of the best performing BiI3 devices. As shown in Fig. 6, the PCE remained above 0.45% for over 30 days after exposing them to the ambient atmosphere with a relative humidity of 50%. This result highlights promising stability of BiI3 material against exposure to moisture. Additionally, BiI3 thin films were quite stable against heat stress (100 OC) (Fig. S12a) for two hours. However, the device showed slight initial degradation and thereafter remained constant as shown in Fig. S 12b. Such initial loss in performance can be due to change in the composition of Spiro-OMeTAD during the heat stress at 100 OC.35 We also performed light soaking test for 20 min and found that the device showed stable performance (Fig. S 12c). The performance decreased after 20 min, which can possibly be due to the formation of bismuth oxide (Bi2O3) layer in the presence of light.27

Figure 6: PCE of BiI3 device after exposing to ambient atmosphere (relative humidity 50%).


Page 18 of 27

Page 19 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Based on our results, we presume that PCE of BiI3 photovoltaic devices can be further enhanced by mitigating intrinsic high resistive property of BiI3 by external doping with antimony (Sb),25 which is expected to reduce the series resistance and recombination rate. Thin devices based on CdTe with comparable carrier lifetime with that of BiI3 have demonstrated PCE of 8 to 9 %,36 which also suggests that improvement of BiI3 properties by eliminating impurities such as bismuth interstitial sites and vacancies can help to enhance the efficiency.33 Furthermore, the difference between carrier lifetime of single crystal and polycrystalline thin film of BiI3 prospects the route to enhance carrier lifetime of the polycrystalline film by obtaining phase pure BiI3 layer and/or by eliminating intragranular structural defects.21 In addition to tuning of the intrinsic properties of BiI3, device architecture needs major modification for efficient charge carrier extraction. The loss of higher energy photons evinces amendment of TiO2/BiI3 interface and/or choice of the suitable electron collecting layer.27 Choice of suitable HTM is also important as the VBM of BiI3 and widely used spiro-OMeTAD is highly mismatching. As mentioned earlier, BiI3 is a very soft material which presents challenges to device fabrication.25 Resolution of the above-mentioned issues can pave a path to enhancement of the PCE of BiI3 solar cells. CONCLUSION In summary, mesostructured BiI3 films were prepared by simple solution process. Among different annealing methods (TA, SVA, and PR-VA) investigated, the PR-VA method (during the heating step at 150 OC) showed better morphology compared to SVA and TA method. The PR-VA method followed at different annealing temperature; 90, 120, 150 and 180 OC displayed remarkable change in BiI3 layer morphology and crystal orientation. Among all the processing temperatures, BiI3 films processed at 150 OC showed the best performance (PCE = 0.5%), which



was enabled by an enhancement in Jsc. Larger grain size and uniform grain growth in the case of 150 OC annealing condition apparently improved the Jsc. However, Voc and FF were not highly affected with the annealing temperature (from 90 to 150 OC). Further increase in temperature to 180 OC resulted in the visible loss of the BiI3 layer, resulting in lowering of device performance. It was, however, interesting to find that BiI3 devices were stable for up to 30 days of exposure to 50% relative humidity, and were stable against 100 OC heat stress for two hours and 20 min of light soaking. This study demonstrates the initial efficiency of 0.5% of BiI3 devices in TiO2 mesostructured architecture employing widely used spiro-OMeTAD as HTM. We expect more enhancements of BiI3 device performance by extrinsic doping, eliminating the defects, choice of suitable charge collecting partners and engineering of solution process for ensuring good heterojunction interfaces.

ASSOCIATED CONTENT Supporting Information. Substrate visual images, UV-Vis spectra, top surface SEM, images of Petri dish showing visible degradation, grain size histogram plot, XRD pattern, PCE histogram plot and J-V curves, this material is available free of charge via the …. AUTHOR INFORMATION Corresponding Authors * [email protected] and [email protected] Present Addresses


Page 20 of 27

Page 21 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


† Graduate School of Engineering, Toin University of Yokohama, 1614 Kuroganecho, Aoba, Yokohama 225-8503, Japan. Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ACKNOWLEDGEMENTS A.K. would like to acknowledge the Japan Society for the Promotion of Science (JSPS) for the JSPS DC2 doctoral fellowship. This study was supported by Japan Science and Technology Agency (JST). T.M. acknowledges financial support from JSPS Grant-in-Aid for Scientific Research B Grant No. 26289265 and New Energy and Industrial Development Organization (NEDO). The author thanks, Prof. Hiroshi Segawa for allowing access to research facilities at Research Center for Advanced Science and Technology (RCAST), University of Tokyo. A.K. thanks Dr. Sudhir Gupta, Faculty of Chemical Sciences, Shri Ramswaroop Memorial University, India, and Miss Bhumika Chaudhary, ERI@N, Nanyang Technological University, Singapore, for their valuable discussion. REFERENCES

1 Kojima, A.; Teshima, K.; Shirai, Y.; Miyasaka, T. Organometal halide perovskites as visiblelight sensitizers for photovoltaic cells. J. Am. Chem. Soc. 2009, 131, 6050-6051, 2 Miyasaka, T. Perovskite Photovoltaics: Rare Functions of Organo Lead Halide in Solar Cells and Optoelectronic Devices. Chem. Lett., 2015, 44, 720-729.



Page 22 of 27

3 Yang, W. S.; Park, W. B.; Jung, E. H.; Jeon, N. J.; Kim, Y. C.; Lee, D. U.; Shin, S. S.; Seo, J.; Kim, E. K.; Noh, J. H.; Seok, S. I. Iodine Management in Formamidinium-lead-halide-based Perovskite Layers for Efficient Solar Cells. Science, 2017, 356, 1376-1379. 4 Legitens, T.; Bush, K.; Cheacharoen, R.; Beal, R.; Bowring, A.; McGehee, M. D. Towards enabling stable lead halide perovskite solar cells; interplay between structural, environmental, and thermal stability. J. Mater. Chem. A, 2017, 5, 11483-11500. 5 Chaudhary, B.; Kulkarni, A.; Jena, A. J.; Ikegami, M.; Udagawa, Y.; Kunugita, H.; Ema, K.; Miyasaka, T. Poly(4-vinylpyridine)-based Interfacial Passivation to Enhance Voltage and Moisture

Stability

of

Lead

Halide

Perovskite

Solar

Cells.

ChemSusChem,

2017,

DOI: 10.1002/cssc.201700271. 6 Kulkarni, A.; Jena, A. J.; Chen, H.-W.; Sanehira, Y.; Ikegami, M.; Miyasaka, T. Revealing and reducing the possible recombination loss within TiO2 compact layer by incorporating MgO layer in perovskite solar cells. Solar Energy, 2016, 136, 379-384. 7 Pinpithak, P.; Chen, H.-W.; Kulkarni, A.; Sanehira, Y.; Ikegami, M.; Miyasaka, T. Lowtemperature and Ambient Air Processes of Amorphous SnOx-based Mixed Halide Perovskite Planar Solar Cell. Chem. Lett., 2017, 46 (3), 382-384. 8 Babayigit, A.; Ethirajan, A.; Muller, M.; Conings, B. Toxicity of organometal halide perovskite solar cells. Nature Materials, 2016, 15, 247-251. 9 Konstantakou, M.; Stergiopoulous, T. A critical review on tin halide perovskite solar cells. J. Mater. Chem. A, 2017, 5, 11518-11549.


Page 23 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10 Krishnamoorthy, T.; Ding, H.; Yan, C.; Leong, W. L.; Baikie, T.; Zhang, Z.; Sherburne, M.; Li, S.; Asta, M.; Mathews, N.; Mhaisalkar, S. G. Lead-free germanium iodide perovskite materials for photovoltaic applications. J. Mater. Chem. A, 2015, 3, 23839-23832. 11 Singh, T.; Kulkarni, A.; Ikegami, M.; Miyasaka, T. Effect of Electron Transporting Layer on Bismuth-Based Lead-Free Perovskite (CH3NH3)3Bi2I9 for Photovoltaic Applications. ACS Appl. Mater. Interfaces, 2016, 8 (23), 14542–14547 12 Kulkarni, A.; Singh, T.; Ikegami, M.; Miyasaka, T. Photovoltaic enhancement of bismuth halide hybrid perovskite by N-methyl pyrrolidone-assisted morphology conversion. RSC Adv., 2017, 7, 9456-9460. 13 Hebig, J.-C.; Kuhn, I.; Flohre, J.; Kirchartz, T. Optoelectronic properties of (CH3NH3)3Sb2I9 thin films for photovoltaic applications. ACS Energy Lett., 2016, 1 (1), 309-314. 14 Shi, Z.; Guo, J.; Chen, Y.; Li, Q.; Pan, Y.; Zhang, H.; Xia, Y.; Huang, W. Lead-free organicinorganic hybrid perovskites for photovoltaic applications: recent advances and perspectives. Adv. Mater., 2017, 29 (16), 1605055. 15 Park, B.-W.; Philippe, B.; Zhang, X.; Ransmo, H.; Boschloo, G.; Johansson, E. M. J. Bismuth based hybrid perovskites A3Bi2I9 (A: methylammonium or cesium) for solar cell applications. Adv. Mater., 2015, 27, 6806-6813. 16 Ran, C.; Wu, Z.; Xi, J.; Yuan, F.; Dong, H.; Lei, T.; He, X.; Hou, X. Construction of Compact Methylammonium Bismuth Iodide Film Promoting Lead-Free Inverted Planar Heterojunction Organohalide Solar Cells with Open-Circuit Voltage over 0.8V. J. Phys. Chem. Lett., 2017, 8 (2), 394–400.



17 Mali, S. S.; Kim, H.; Kim, D.-H.; Hong, C. K. Anti-Solvent Assisted Crystallization Processed Methylammonium Bismuth Iodide Cuboids Towards Highly Stable Lead-Free Perovskite Solar Cells. Chemistry Select, 2017, 2(4), 1578-1585. 18 Zhang, Z.; Li, X.; Xia, X.; Wang, Z.; Huang, Z.; Lei, B.; Gao, Y. High quality (CH3NH3)3Bi2I9 film based solar cells: pushing efficiency up to 1.64%. J. Phys. Chem. Lett., 2017, 8 (17), 4300-4307. 19 Slavney, A. H.; Hu, T.; Lindenberg, A. M.; Karunadasa, H. I. A bismuth-halide double perovskite with long carrier recombination lifetime for photovoltaic applications. J. Am. Chem. Soc., 2016, 138 (7), 2138-2141. 20 Savory, C. N.; Walsh, A.; Scanlon, D. O. Can Pb-free halide double perovskites support highefficiency solar cells? ACS Energy Lett., 2016, 1 (5), 949-955. 21 Brandt, R. E.; Kurchin, R. C.; Hoye, R. L. Z.; Poindexter, J. R.; Wilson, M. W. B.; Sulekar S.; Lenahan, F.; Yen, P. X. T.; Stevanovic, V.; Nino, J. C.; Bawendi, M. G.; Buonassisi, T. Investigation of bismuth halide (BiI3) for photovoltaic applications. J. Phys. Chem. Lett., 2015, 6, 4297-4302. 22 Green, M. A.; Keevers, M. J. Optical properties of intrinsic silicon at 300 K. Prog. Photovoltaics, 1995, 3, 189−192. 23 Blakemore, J. S. Semiconducting and other major properties of gallium arsenide. J. Appl. Phys., 1982, 53, R123−R181. 24 Baciak, J. E.; Nino, J. C.; Qiu, W.; Lintereur, A. Single-Crystal Bismuth Iodide Gamma-Ray Spectrometers; DTRA-TR-12-70; DTIC Document, 2013.


Page 24 of 27

Page 25 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


25 Lintereur, A. T.; Qiu, W.; Nino, J. C.; Baciak, J. Characterization of Bismuth Tri-Iodide Single Crystals for Wide Band-Gap Semiconductor Radiation Detectors. Nucl. Instrum. Methods Phys. Res., Sect. A 2011, 652, 166−169. 26 Lehner, A. J.; Wang, H.; Fabini, D. H.; Liman, C. D.; Hebert, C.-A.; Perry, E. E.; Wang, M.; Bazan, G. C.; Chabinyc, M. L.; Seshadri, R. Electronic Structure and Photovoltaic Application of BiI3. Appl. Phys. Lett., 2015, 107, 131109. 27 Hamdeh, U. M.; Nelson, R. D.; Ryan, B. J.; Bhattacharjee, U.; Petrich, J. W.; Panthani, M. G. Solution-Processed BiI3 Thin Films for Photovoltaic Applications: Improved Carrier Collection via Solvent Annealing. Chem. Mater., 2016, 28 (18), 6567-6574. 28 Boopathi, K. M.; Raman, S.; Mohanraman, R.; Chou, F.-C.; Chen, Y.-Y.; Lee, C.-H.; Chang, F.-C.; Chu, C.-W. Solution-processable bismuth iodide nanosheets as hole transport layers for organic solar cells. Sol. Energy Mater Sol. Cells, 2014, 121, 35-41. 29 Xiao, Z.; Dong, Q.; Bi, C.; Shao, Y.; Yuan, Y.; Huang, J. Solvent annealing of perovskiteinduced crystal growth for photovoltaic-device efficiency enhancement. Adv. Mater., 2014, 26, 6503-6509. 30 Numata, Y.; Kogo, A.; Udagawa, Y.; Kunugita, H.; Ema, K.; Sanehira, Y.; Miyasaka, T. Controlled crystal grain growth in mixed cation-halide perovskite by evaporated solvent vapor recycling method for high efficiency solar cells. ACS Appl. Mater. Interfaces, 2017, 9, 1873918747. 31 Tauc, J.; Grigorovici, R.; Vancu, A. Optical Properties and Electronic Structure of Amorphous Germanium. Phys. Status Solidi B, 1966, 15, 627-637.



32 Sinturel, C.; Vayer, M.; Morris, M.; Hillmyer, M. A. Solvent Vapor Annealing of Block Polymer Thin Films. Macromolecules, 2013, 46, 5399−5415. 33 Curtis, B. J.; Brunner, H. R. The crystal growth of bismuth iodide. Materials Research Bulletin, 1974, 9, 715-720. 34 Han, H.; Hong, M.; Gokhale, S. S.; Sinnott, S. B.; Jordan, K.; Baciak, J. E.; Nino, J. C. Defect Engineering of BiI3 Single Crystals: Enhanced Electrical and Radiation Performance for Room Temperature Gamma-Ray Detection. J. Phys. Chem. C, 2014, 118, 3244−3250. 35 Jena, A. K.; Numata, Y.; Ikegami, M.; Miyasaka, T. Role of spiro-OMeTAD in performance deterioration of perovskite solar cells at high temperature and reuse of the perovskite films to avoid Pb-waste. J. Mater. Chem. A, 2018, 6, 2219-2230. 36 Moutinho, H. R.; Dhere, R. G.; Al-Jassim, M. M.; Ballif, C.; Levi, D. H.; Swartzlander, A. B.; Young, M. R.; Kazmerski, L. L. Study of CdTe/CdS Solar Cells Using CSS CdTe Deposited at Low Temperature. Photovoltaic Specialist Conference (PVSC), 2000 IEEE 28th 2000, 646−649.


Page 26 of 27

Page 27 of 27 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Graphical abstract


Vapor Annealing Controlled Crystal Growth and Photovoltaic

Recommend Documents