Slow Response of Carrier Dynamics in Perovskite Interface upon

Subscriber access provided by Saint Louis University Libraries

Energy, Environmental, and Catalysis Applications

Slow Response of Carrier Dynamics in Perovskite Interface upon Illumination Fei Zheng, Xiaoming Wen, Tongle Bu, Sheng Chen, Jianfeng Yang, Weijian Chen, Fuzhi Huang, Yi-Bing Cheng, and Baohua Jia ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b13932 • Publication Date (Web): 28 Aug 2018 Downloaded from http://pubs.acs.org on August 28, 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 23 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

Slow Response of Carrier Dynamics in Perovskite Interface upon Illumination Fei Zheng,† Xiaoming Wen,*, †, ‡ Tongle Bu,§ Sheng Chen,‡ Jianfeng Yang,‡ Weijian Chen,‡ Fuzhi Huang,§ Yibing Cheng,§ Baohua Jia*,† †

Center for Micro-Photonics, Swinburne University of Technology, Hawthorn, VIC3122, Australia

‡

School of Photovoltaics and Renewable Energy Engineering, University of New South Wales, Kensington, NSW2052, Australia §

State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Wuhan University of Technology, Wuhan 430070, P. R. China * X. Wen: [email protected] * B. Jia: [email protected]

Abstract The current-voltage hysteresis, as well as performance instability of perovskite solar cells (PSCs) under working condition, is serving as the major obstacles toward their commercialization, while the exact fundamental mechanisms to these issues are still in debate. In this study, we investigated the slow variation of photogenerated carrier dynamics in (FAPbI3)0.85(MAPbBr3)0.15 perovskite interface under continuous illumination. Different response behaviours of carrier dynamics in perovskite interfaces with and without hole transporting layer, Spiro-OMeTAD (Spiro), were systematically studied by time-dependent steady-state and time-resolved photoluminescence (PL). It was demonstrated that lightinduced defect curing process is dominantly responsible for the carrier dynamic evolution for perovskite interface without Spiro, while both defect curing process and mobile ions migration should be account for the dynamic response of perovskite interface contact with Spiro. When contacted with Spiro, energy band curvature evolution in perovskite interface induced by ion migration would decrease the hole transfer rate from perovskite to Spiro upon illumination. This research work can faithfully highlight the strong correlation of slow photoresponse behaviours of perovskite interface with both light-induced defect curing and ion migration process, providing novel implications into the physical mechanism for the slow variation of PSCs performance under working condition.

Keywords: slow response, carrier dynamics, photoluminescence, interface, transfer 1 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

1. Introduction. Organic-inorganic lead halide perovskites have emerged as excellent light absorber materials for perovskite solar cells (PSCs) with their power conversion efficiency (PCE) developed from 3.8% to more than 22.1% in the past few years.1 The great success is mainly attributed to the appealing long charge carrier lifetimes, large charge carrier mobility and high absorption coefficients of perovskite materials.2-3 However, there are still many obstacles impairing the ultimate performance and hindering the commercialization of PSCs, such as current-voltage (J-V) hysteresis,4 as well as performance (i.e. photocurrent, power conversion efficiency) instability under standard working condition.5 Many strategies have been identified to pursue highly efficient and long-term stable PSCs, including composition engineering,6 device architecture modification,7 and interface manipulation8. Therefore, a deeper insight into the carrier dynamics in perovskite phase and contact interface under the working condition of PSCs are significant for the device optimization Particularly, the influence of light illumination on the performance of PSCs concerning the underlying device physics under working condition are intensively studied.9-10 Critical roles of light illumination in performance enhancement of PSCs owing to defect curing and structural modification have also received growing attentions. Huang et al.5 revealed a “fatigue” behaviour for planar PSCs by light on/off diurnal cycling wherein the efficiency of device degraded when stored in dark but gradually recovered after daylight illumination, raising concerns about the light induced performance altering for PSCs. While the correlation between PSC performance instability phenomena and light illumination is still unclear. Defect curing process for perovskite phase by external illumination has been proposed to explain the improved film qualities observed for perovskite films, including enhanced PL intensity and enlarged PL decay time, which are directly resulted from the decreased defect-induced carrier recombination.11-12 However, to what extent defect curing impact the carrier dynamics in perovskite layers and PSC performance remains an unsolved question. Hysteresis during current-voltage (J-V) characterization of PSCs is disadvantageous to the accurate determination of devices performance, which is also known as the symptom of device instability.13 Although the origins of hysteresis are still under debate, 2 ACS Paragon Plus Environment

Page 2 of 23

Page 3 of 23 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


migration, and accumulation of mobile ions in perovskite films has been widely accepted to be one of the major causes of hysteresis.13-17 Meanwhile, the activation energy or migration barrier of mobile ions effectively decreases under light illumination,18 which gives implications for the strong correlation between the light illumination and hysteresis. Photoluminescence (PL) spectroscopy measurement is a powerful route for probing the carrier recombination behaviours inside perovskite films and at contact interface, which can directly reflect the ion migration and accumulation behaviours.19 Xu et al.20-21 observed a hysteric PL intensity responding to scanning voltage for MAPbI3 film in solar cell architecture, revealing the effect of ion migration on charge drifting and recombination processes in PSCs. However, a complete understanding of the role of perovskite contact interface with charge selecting layer on the charge recombination and its correlation to hysteresis is still to be done. In this work, we systematically investigated the carrier recombination and extraction dynamics of the perovskite film composed of mixed cations and mixed halides, (FAPbI3)0.85(MAPbBr3)0.15,6 with and without (w/o) hole extraction layer, SpiroOMeTAD (Spiro), under continuous light illumination. Time-dependent steady-state PL and time-resolved PL (TRPL) spectroscopies were performed to analyse the PL intensity evolution and transient PL decay profile variations in timescale of seconds and minutes, from which different carrier dynamics were deduced for perovskite interface contact with and without Spiro layer. Wavelength-dependent investigations were carried out by using 470 nm and 640 nm illumination for comparison, offering the capability to probe carrier dynamics in perovskite interface with different penetration depth to further identify the carrier dynamics evolution behaviours.

2. Results and discussions. Figure 1a shows the structure of PSC device used for this study with the configuration of FTO/SnO2/ (FAPbI3)0.85(MAPbBr3)0.15/Spiro/Au, showing the thickness of perovskite and Spiro layer to be around 400 nm and 150 nm, respectively. To eliminate the impact of oxygen and humidity, a 200 nm transparent poly(methyl methacrylate) (PMMA) layer is spin-casted for encapsulation for PL and TRPL measurements. The uniform surface morphology of the prepared perovskite film was demonstrated by the top-view scanning electron microscopy (SEM) images as shown in Figure 1b, from which a grain size in the scale of around 500 nm can be seen. The high quality of this prepared 3 ACS Paragon Plus Environment


mixed perovskite film was identified by the low density of carrier trapping defect sites estimated by the space-charge-limited-current (SCLC) technique detailed in the Supplementary Information (Figure S1). The electron trap density in the mixed perovskite film sample is calculated to be 5.9×1015 cm-3, which is comparable to literature value for high performance PSCs.17 The J-V curves under standard AM 1.5G sun-light illumination (0.1 W/cm2) shown in Figure 1c and the corresponding performance parameters are listed in the inset. The power conversion efficiency (PCE) of 19.35% for PSC device derived from reverse scan (from 1.2 V to -0.1 V) J-V curve is superior to the previously reported results with the same perovskite material,22 demonstrating the high qualities of the perovskite film and solar cell device in this work. Figure 1d shows the external quantum efficiency (EQE) spectrum of PSC measured at short circuit and the corresponding photocurrent integration curve. A broad EQE plateau (>80%) range from 400 nm to 760 nm indicates the efficient photon to electron conversion under visible light illumination. The short-circuit current integrated from EQE (JSC=22.6 mA/cm2) is in good agreement with the scan measured one (22.6 mA/cm2), reflecting the accuracy of device performance measurements. Evident hysteresis can be seen from the higher power conversion efficiency (PCE), fill factor (FF) and open circuit voltage (VOC) for reverse scan in comparison to forward scan (from -0.1V to 1.2 V). Therefore, hysteresis correlated ionic process should be taken into account when study carrier dynamics in perovskite interface contacted with Spiro layer. The steady-state photocurrent of the PSC measured at the maximum power point (Figure 1e) shows a stable output along prolonged working time. However, flowing the initial rapid response of photocurrent to the switch-on illumination, a slow response component in the tens of second time scale appears before reaching stabilized power output, as shown in the rectangle in Figure 1e. Thus, it is worthwhile further exploring the carrier dynamics of this perovskite layer under light illumination to find out the mechanism regulating its slow photo-response and corresponding performance-response behaviours under working condition. UV-vis absorption spectra of perovskite film deposited in SnO2 covered FTO substrate is depicted in Figure 2a, showing apparent absorption coefficient (α) for visible light ranging from 770 nm to the UV light, which is consisted with the EQE spectrum of PSC device. The intensity profile of the incident light along the depth (x) is expressed as = ∙ exp − ∙ according to Beer-Lambert Law, from which the absorption 4 ACS Paragon Plus Environment

Page 4 of 23

Page 5 of 23 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


profiles can be deduced as ∆ = = ∙ exp − ∙ ∙ , as depicted in the inset

of Figure 2a. Thus, penetration depth ( = 1⁄ ) of illumination light is calculated to be 78 nm and 180 nm for 470 nm and 640 nm, which are smaller than the perovskite film thickness (400 nm), ensuring the collected signals are dominantly from the surface of perovskite film in the following PL measurements. The PL spectra of perovskite layer without the charge transport layer were measured from the back side (contacting with the glass substrate) and the front side (contacting with the PMMA encapsulating layer) with 405 nm laser excitation, as shown in Figure S2 (Supplementary Information). The much lower PL intensity from the front side compared to the back side indicates the higher density of defect traps in the front side, which causes apparent PL quenching. The nearly identical PL peak positions (773 nm and 774 nm) for back side and front side layers indicate the uniform bandgap throughout the perovskite film, which is estimated to be 1.61 eV according to Eg=h·c/λPL. The PL spectrum of the perovskite film deposited on SnO2 layer excited by 405 nm laser exhibits a similar peak centred at 775 nm (Figure 2b), which consists with literature reports.23 After the deposition of Spiro layer, the PL intensity of the perovskite under the same excitation condition is largely quenched. While no discernible changes of PL peak and Full-WidthHalf-Maximum (FWHM) value are observed for perovskite interface with Spiro, indicating the unaltered film crystallinity and interior structure for perovskite after Spiro coverage. Thus, any PL change caused by film quality alerting can be eliminated in this situation. The observed PL quenching is only ascribed to the deposited Spiro layer upon perovskite layer, which serves as a hole extracting layer to effectively decrease the photogenerated carriers (holes) number in perovskite.2 The PL intensity was measured as a function of time (PL time traces) under continuous illumination (excitation at 470 nm) for perovskite interfaces w/o and with Spiro layer (Figure 2c). PL intensity enhancement is observed for both samples. Visible light illumination can effectively annihilate defect trap states in perovskite (MAPbI3), often referred as defect curing, leading to the increased weight ratio of radiative recombination and thus increasing the PL intensity.12, 24-25 Hence, it is reasonable to infer that defect curing by light illumination is responsible for the PL enhancement in both cases. Figure 2d shows the PL time traces divided (normalized) by initial values (t=0 s), which reflects the relative increment of PL intensity. Both curves can be well fitted by the 5 ACS Paragon Plus Environment


saturation function, = 1 + ∙ ⁄1 + ⁄ ,26 wherein is the saturation time describing the increase speed of PL intensity (Figure 2d). The extracted saturation time from fitting curves for perovskite interfaces w/o and with Spiro is 413.2 s and 76.0 s, respectively, showing a more pronounced PL enhancement for perovskite interface after Spiro coverage. We speculate that such phenomenon is caused by the additional hole transfer process in addition to defect curing and the detailed correlation will be elucidated below. Upon photo-excitation, the electrons are excited to high energy levels and undergo thermalization to the conduction band (CB) edge, leaving holes in the valance band (VB). A large number of active phonons are generated by excess energy.27 Free electrons and holes (charge carriers) are dominant due to the small exciton binding energy.28 The carrier density in perovskite can be described by – d!⁄d = "# ∙ ! + "$ ∙ !$ + "% ∙ !% , where terms in the right represent Shockley–Read–Hall (SRH) recombination via defect trapping (first order term, "# ∙ ! ), radiative recombination of free electron-hole pairs (second order term, "$ ∙ !$ ) and Auger recombination (third order term, "% ∙ !% ).29 Radiative electron-hole recombination is responsible for PL emission, while the other two routes are non-radiative. Auger recombination plays a role wherein the carrier concentration is extremely high, i.e. when semiconductor is heavily doped or strongly excited, thus could be ignored in this experiment since the light illumination intensity (0.4 W/cm2) is relatively low.30 Therefore, only defect trapping and free electron-hole recombination need to be considered for perovskite interface w/o Spiro, as depicted in the inset of Figure 3a. For perovskite/Spiro sample, an additional carrier relaxation route, hole transfer from perovskite to Spiro, dominates the relaxation process and results in PL quenching (Figure 3b, inset). To obtain the detailed insight of photogenerated carrier dynamics, time-resolved PL (TRPL) measurements of perovskite interface w/o Spiro were consecutively measured under continuous 470 nm laser illumination (excitation), as shown in Figure 3a. All decay curves can be well fitted by a bi-exponential decay function, & = # exp− ⁄'# + $ exp −/'$ , giving the fast decay lifetime '# and the slow decay lifetime '$ with their weight ratio of # , $ . These fitting parameters together with the calculated effective lifetimes, ') = # ∙ '# + $ ∙ '$ , are summarized in Table 1. The ') for perovskite with 0 s, 60 s and 120 s illumination are 182.6 ns, 217.8 ns and 236.7 ns, which is increased by

6 ACS Paragon Plus Environment

Page 6 of 23

Page 7 of 23 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


29.7% over the whole time range. This result is consistent with the enhancement of PL intensity under same illumination condition as described above as strong PL intensity corresponds to long PL lifetime.31 '# is correlated to carrier decay time by defect trapping while '$ to the radiative electron-hole recombination.32-33 Upon continuous illumination, the effective lifetime of carrier increases from 183 ns to 218 ns, then to 237 ns upon continuous illumination, which is mainly attributed to the decreased ratio of fast decay component (# and increased ratio of the slow radiative decay component ($ (Figure 3c).34 This result gives an implication that the defect trapping ratio of carriers are dramatically decreased upon prolonged illumination due to defect curing process. The TRPL was measured for perovskite interface with Spiro under the identical condition (Figure 3b). All the three decay traces corresponding to 0 s, 60 s, 120 s illumination can be well fitted by a tri-exponential function, & = # exp− ⁄'# + $ exp− ⁄'$ + % exp− ⁄'% , with the fitting parameters listed in Table 1. The increase of effective PL lifetime, '* = # ∙ '# + $ ∙ '$ + % ∙ '%, by 35.0% from 0 s to 120 s is in line with the aforementioned PL enhancement. In this case, the fast component '# and '$ are positively correlated to hole transfer and defect trapping respectively while the slow decay time '% is related to the radiative recombination of carriers in the perovskite phase.35-36 The plots of corresponding weight ratio of each decay component over time (Figure 3d) depict that the gradual proportion increase of radiative recombination component ( % ) accompanied by the decrease of non-radiative hole transfer (# and defect trapping ($ ). Thus, beside defect curing process, the decreased ratio of hole transfer process is also the partial reason for the effective PL lifetime increase in perovskite/Spiro interface upon illumination, giving a more pronounced PL enhancement. To identify the divergent roles of defect curing and hole transfer variation on the slow response behaviours of carrier dynamics, we measured TRPL of perovskite interface w/o and with Spiro layer under continuous 640 nm illumination (excitation) for 0 s, 60 s and 120 s (Figure 4a, 4b).

The fitting parameters are summarized in Table 2.

Interestingly, the decay traces of perovskite interface w/o Spiro (Figure 4a) undergo less notable evolution over illumination time compared to 470 nm illumination case, which was further confirmed by nearly constant ratio-time plots depicted in Figure 4c. The effective PL lifetime increased about 8.5% after 120 s illumination, which is obviously

7 ACS Paragon Plus Environment


Page 8 of 23

smaller than that under 470 nm illumination (29.7%). This result reveals that the defect curing process is not prominent in perovskite interface under 640 nm illumination compared to 470 nm. In contrast, the PL decay trace of perovskite interface with Spiro (Figure 4b) as well as decay time weight ratio (A1, A2, A3) (Figure 4d) still evolve over time evidently. The effective PL lifetime increase ratio of 26.6% is comparable to that under 470 nm illumination (35.0%), despite the small contribution of defect curing process. Therefore, it can be inferred that variation of hole transfer process becomes more predominant for the carrier dynamics evolution in perovskite/Spiro interface under 640 nm illumination. In order to make a comprehensive interpretation of the above phenomena, surface recombination mechanism should be deeply analysed. Note the existence of liquid/vapour face during the preparation of perovskite film, defects are mainly distributed at the front surface, i.e. the perovskite interface in our experiment,12 which can be supported by the PL measurements from two sides presented in Figure S2. Thus majority of SRH recombination via defect trapping occurs at the surface, leading to the flow of carriers toward the surface with the velocity of S, which is defined as surface recombination velocity.37 Upon illumination, the photogenerated carriers are exponentially distributed along the film thickness (coordinate +) inside the perovskite phase, which can be written as: !+, 0 = ! exp[− )/ ∙ +]

(1)

where ! is the initial carrier density at front surface and )/ is the absorption coefficient of perovskite material at the excitation wavelength. These initially distributed carriers will undergo recombination and diffusion described by the equation: 123,4 14

=5

1 6 23,4 13 6

−

23,4 78

(2)

in which 5 is the carrier diffusion coefficient and '9 refers to the bulk lifetime of pristine perovskite material without defect. When neglect the SRH recombination at the back surface of perovskite film, the boundary conditions are: : ∙ !0, = 5

12,4 14

; 0 = 5

12

Slow Response of Carrier Dynamics in Perovskite Interface upon

Recommend Documents