Perovskite solar cells photovoltaic and structural evolution during film formation


Simultaneous in situ GIWAXS and operando current-voltage measurements on metal-halide perovskite interdigitated-back-contact (IBC) solar cells were used to reveal the remarkably clean semiconductor behaviour of perovskites films that emerges in the earliest phase of conversion from the as-coated precursor film.

  • Share

Perovskite photovoltaics (PV) is one of the fastest growing opto-electronic technologies with device efficiencies currently exceeding 22.1% and potential for low-cost production. To further improve device performance and stability, researchers are continuously exploring a variety of optimisation strategies, such as thermal engineering. For most solution-processed perovskite materials, annealing temperature and duration are critical factors for optimised conversion of the as-coated precursor material into a functional polycrystalline perovskite film. The optimisation is done by looking at how the photovoltaic performance of a perovskite solar cell evolves with time during the anneal. The fastest way to perform such a measurement is to record current-voltage (JV) sweeps of a complete device placed on a hotplate under simulated solar illumination during the anneal. While this investigation method has been applied to other third generation PV devices, its application to perovskite solar cells is not feasible because the perovskite material must be annealed before further layers can be deposited to complete the device. Therefore, to obtain analogous data for perovskite devices, one needs to pre-anneal and complete the device stack, then measure ex situ several solar cells for different annealing periods, which is extremely time consuming.

Importantly, structural data from synchrotron in situ grazing-incidence X-ray diffraction (GIXRD) are essential to interpret photovoltaic figures-of-merit of perovskite solar cells vs anneal time [1,2]. However, a direct correlation between the PV performance and the structural properties can only be established when such measurements are done on the same perovskite film of a working device. Since, as mentioned above, in situ annealing is not an option, one would need to fabricate and anneal several solar cells at different periods by using top layers that do not interfere with GIXRD measurements, and then measure their diffraction patterns ex situ. Apart from the time needed and the requirements of a large solar cell population for good statistics, top layers above the perovskite film would cause some degradation in the signal-to-noise ratio of the GIXRD measurement when compared to a bare perovskite film.

In this work, a new method was employed that can significantly reduce the workload required in the thermal engineering of perovskite solar cells and, at the same time, establish a direct correlation between their opto-electrical and structural properties during the anneal in situ. The investigation method is illustrated in Figure 1. Interdigitated back-contact (IBC) solar cells were the crucial element in this methodology. Their electron (TiO2) and hole (PEDOT) selective electrodes are co-positioned on the backside of the cell in an interdigitated fashion (Figure 1b) [3]. The main advantage of IBC perovskite solar cells is that the perovskite layer represents the final step of the device fabrication. As the perovskite film is unobstructed by a top layer, in situ annealing can be performed without compromising the film formation. At the same time, the perovskite film is directly accessible by an X-ray beam in GIWAXS geometry. This disposes IBC devices for simultaneous in situ opto-electrical and GIWAXS measurements during the anneal. With the setup developed at BM28 (the XMaS CRG beamline), a high throughput thermal engineering route has been demonstrated that can be used on a variety of perovskite materials, along with the possibility of establishing a direct correlation between the figures-of-merit of a solar cell and its structural properties.

Illustration of the experimental setup.

Figure 1. Illustration of the experimental setup. (a) Annealing chamber. (b) Illustration of ITO interdigitated substrate electrodeposited with PEDOT (electrodes 1-5) and TiO2 (electrodes 6-10). Electrode 11 is bare ITO.  (c) Illustration of the experimental setup for current-voltage sweeps performed in-situ without X-rays. (d) Illustration of the setup for diffraction pattern measurement (light off) with a 10 keV X-ray at grazing incidence.

Figure 2 shows a summary of the photovoltaic and structural measurements on CH3NH3PbI3 (MAPI3) perovskite IBC solar cells. The remarkably clean semiconductor behaviour of perovskites is evidenced by the high photovoltages measured at the first stages of perovskite conversion from precursors, at the percolation threshold for bulk conductance. Open circuit voltages (Voc) reach maximum value before the precursor has been fully converted into perovskite, when the fraction of precursor and perovskite crystals are comparable (cf. Figure 2c-d). Short circuit currents (Jsc) and power conversion efficiencies (PCE) follow a similar trend to that of the perovskite peak intensity extracted from the GIWAXS measurements.

Structural and opto-electrical parameters extracted from simultaneous GIWAXS diffraction patterns (under dark conditions) and current-voltage (under light conditions) measurements of a perovskite (CH3NH3PbI3) IBC solar cell

Figure 2. Structural and opto-electrical parameters extracted from simultaneous GIWAXS diffraction patterns (under dark conditions) and current-voltage (under light conditions) measurements of a perovskite (CH3NH3PbI3) IBC solar cell during in situ anneal at 88.2°C. The following measurement was repeated every 54.4 s: measure one GIWAXS pattern (1 s integration time, light off), forward JV sweep (18 s, light off), reverse JV sweep (18 s, light on). (a) Azimuthally integrated line profiles vs annealing time. (b) IBC solar cell current-voltage sweeps measured under light. (c) Integrated peak intensities of the second precursor peak, perovskite, and PbI2 vs annealing time. (d) Normalised figures-of-merit (FOM) vs annealing time.

The behaviour of the figures-of-merit versus annealing time of IBC devices compare well with analogous profiles of conventional planar heterojunction solar cells. These results can be generalised to other perovskite precursor routes, such as: MAPbI3 prepared from MAI: PbAc2, the triple mixed cation Cs0.5(MA0.17FA0.83)0.95Pb(I0.83Br0.17)3 (FA = formamidinium),  and the mixed cation (FAI)1.0(MABr)0.2(PbI2)1.1(PbBr2)0.20. The measurement strategy described here is not limited to perovskite materials and is of interest for any solution-processable photovoltaic technology requiring thermal annealing.


This experiment has been documented in video.


Principal publication and authors
In-situ simultaneous photovoltaic and structural evolution of perovskite solar cells during film formation, M. Alsari (a), O. Bikondoa (b), J. Bishop (c), M. Abdi-Jalebi  (a), L.Y. Ozer (d), M. Hampton (c), P. Thompson (f), M. Hoerantner (g), S. Mahesh (g), C. Greenland (c), J.E. Macdonald (e), G. Palmisano (d), H.J. Snaith (g), D.G. Lidzey (c), S.D. Stranks (a), R.H. Friend (a), S. Lilliu (c,h), Energy Environ. Sci. (2018); doi: 10.1039/c7ee03013d.
(a) Cavendish Laboratory, University of Cambridge (UK)
(b) Department of Physics, University of Warwick, Coventry (UK)
(c) Department of Physics and Astronomy, University of Sheffield (UK)
(d) Department of Chemical Engineering, Khalifa University of Science and Technology, Masdar Institute, Abu Dhabi (UAE)
(e) School of Physics and Astronomy, Cardiff University (UK)
(f) University of Liverpool (UK)
(g) Clarendon Laboratory, Department of Physics, University of Oxford (UK)
(h) The UAE Centre for Crystallography (UAE)


[1] L.M. Pazos-Outón et al., Science 351, 1430-1433 (2016).
[2] S. Lilliu et al., CrystEngComm 18, 5448-5455 (2016).
[3] A.T. Barrows et al., Adv. Funct. Mater. 26, 4934-4942 (2016).