- Email: [email protected]
CuSCN hole-transport bilayer with improved stability
Solar Energy 171 (2018) 652–657
Contents lists available at ScienceDirect
Solar Energy journal homepage: www.elsevier.com/locate/solener
Low temperature processed inverted planar perovskite solar cells by r-GO/ CuSCN hole-transport bilayer with improved stability
T
⁎
Towhid H. Chowdhurya,b,c, Md. Akhtaruzzamanb, , Md. Emrul Kayesha, Ryuji Kanekoa, ⁎ Takeshi Nodaa, Jae-Joon Leec, Ashraful Islama, a
Photovoltaic Materials Group, Center for Green Research on Energy and Environmental Materials, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba 305-0047, Japan b Solar Energy Research Institute, Universiti Kebangsaan Malaysia, Bangi 43600, Selangor Darul Ehsan, Malaysia c Department of Energy & Materials Engineering & Research Center for Photoenergy Harvesting and Conversion Technology, Dongguk University, Seoul 04620, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Copper (I) thiocyanate Hole-transport bilayer Perovskite solar cell Reduced graphene oxide Stability
Low temperature processed Perovskite solar cells (PSCs) are popular due to their potential for scalable production. In this work, we report reduced Graphene Oxide (r-GO)/copper (I) thiocyanate (CuSCN) as an efficient bilayer hole transport layer (HTL) for low temperature processed inverted planar PSCs. We have systematically optimized the thickness of CuSCN interlayer at the r-GO/MAPbI3 interface resulting in bilayer HTL structure to enhance the stability and photovoltaic performance of low temperature processed r-GO HTL based PSCs with a standard surface area of 1.02 cm2. With matched valence band energy level, the r-GO/CuSCN bilayer HTL based PSCs showed high power conversion efficiency of 14.28%, thanks to the improved open circuit voltage (VOC) compared to the only r-GO based PSC. Moreover, enhanced stability has been observed for the r-GO/CuSCN based PSCs which retained over 90% of its initial efficiency after 100 h light soaking measured under continuous AM 1.5 sun illumination.
1. Introduction PSCs has been identified as the most promising solar technology for energy harvesting due to their potential high efficiency and easy fabrication process (Park et al., 2016; Correa-Baena et al., 2017; Green and Ho-Baillie, 2017; Seok et al., 2018). Since the inception of PSCs, two main structures such as- mesoporous and inverted planar structures have been under intense investigation. Until now, the highest power conversion efficiency (PCE) of 22.7% has been observed with mesoporous PSCs (Green et al., 2018). However, in the mesoporous structure, the fabrication of electron transport layer requires high temperature (500 °C) sintering process, which hinders their potential aspects for commercialization and potential application in flexible electronic devices. In this aspect, low temperature processed PSCs are more anticipated. Inverted planar structured PSCs with their total low temperature fabrication compatibility is steadily attracting considerable attention. Generally inverted planar PSCs are fabricated with a p-type semiconducting material which acts as a hole transport layer (HTL) is deposited on top of the transparent conductive oxide glass followed by
⁎
perovskite compound as the absorber and n-type semiconducting material as the electron transport material (ETM). The first inverted planar PSCs were originated from the concept of organic photovoltaics where, PEDOT:PSS was used as the hole transport material (Jeng et al., 2013). The low temperature processed PSC showed impressive photovoltaic performances but resulted in rapid degradation and induced decomposition of the perovskite absorber due to the acidic behaviour of the PEDOT: PSS (Chen et al., 2015). To address this issue, recently we have reported on low temperature processed stable inverted planar PSCs with NiOx nanoparticles (Chowdhury et al., 2018). Alternate low temperature processed HTLs such as Cu2O (Chatterjee and Pal, 2016), CuI (Sun et al., 2016), CuOx (Yu et al., 2017), CoOx (Shalan et al., 2016), CuSCN (Zhao et al., 2015; Ye et al., 2015) has been implemented in PSCs and showed promising aspects but failed to provide stable performances. In stability concern, carbon derivatives such as Graphene (Agresti et al., 2016), Graphene Oxide (GO) (Yang et al., 2017; Liu et al., 2014) and reduced Graphene Oxide (r-GO) (Yeo et al., 2015) are popular choice of HTLs in PSCs. With the capability of higher absorption in the near UV region, GO based PSCs can provide long term stable performance in inverted planar PSCs (Yang et al., 2017). However, for
Corresponding authors. E-mail address: [email protected] (A. Islam).
https://doi.org/10.1016/j.solener.2018.07.022 Received 24 April 2018; Received in revised form 28 June 2018; Accepted 7 July 2018 0038-092X/ © 2018 Elsevier Ltd. All rights reserved.
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
2. Experimental
obtaining high photovoltaic response these GO based HTLs require high sintering temperature or vacuum based evaporation techniques (Acik and Darling, 2016). It has been occasionally revealed that, deposition of low temperature processed r-GO fails to cover the whole surface of the transparent conductive glass eventually providing opportunity for large defects at the interface with the perovskite absorber. These defects at the surface of r-GO eventually leads to potential recombination centres creating instability of the overall PSC. Moreover, the valence band of low temperature processed r-GO is −4.97 eV which has a huge energy mismatch with the perovskite compound valence band (−5.4 eV). This huge mismatch can hinder the potential collection of holes at the r-GO/ perovskite interface and limit the photovoltaic performance of the overall PSC. To overcome these issues, we have recently reported with bilayer structured HTLs with r-GO and organic polymer, poly(triarylamine) (PTAA) and shown promising results (Zhou et al., 2017). However, in the long run, organic compounds as HTLs might not be a suitable choice due to their vulnerable behaviour in ambient environment. With an objective to obtain higher photovoltaic performance and potential commercialization aspect, insertion of an inorganic HTL with higher valence band than r-GO may increase the hole collection. CuSCN, a p-type material with a valence band of −5.30 eV and bandgap of 3.6 eV (Jung et al., 2015) can be considered as one of the potential interfacial material to transport hole faster in this aspect. However, to achieve maximum transparency, an ultrathin layer of CuSCN layer should be beneficial. The potential high bandgap with matched valence band level with the corresponding perovskite absorber can provide sufficient transparency across the UV–Vis/NIR wavelength region and lower loss caused by parasitic absorption. Here in this work, we introduce an ultrathin CuSCN layer as an efficient interlayer in low temperature processed r-GO based inverted planar PSCs. We have systematically optimized the thickness of CuSCN interlayer at the r-GO/Perovskite interface to enhance the photovoltaic performance of the PSC with a surface area of 1.02 cm2. With matched energy level, the r-GO/CuSCN bilayer HTM successfully prevented the recombination at the interface with the MAPbI3 absorber layer. Simultaneously, the r-GO/CuSCN bilayer HTL leads to faster hole extraction as confirmed by the photoluminescence study. The r-GO/ CuSCN bilayer HTL based PSC showed a high power conversion efficiency (PCE) of 14.28% with photovoltaic parameters of open circuit voltage (VOC) = 1.031 V, short circuit current density (JSC) = 18.21 mA cm-2, fill factor (FF) = 0.761. The PCE of the r-GO/ CuSCN bilayer HTL based PSCs were higher than that of the individual r-GO based PSCs (PCE = 9.52%). Additionally, enhanced stability has been observed for the r-GO/CuSCN bilayer HTL based PSCs retaining over the 90% initial efficiency after 100 h light soaking.
2.1. Materials The following chemicals were obtained from commercial suppliers and used as received: PbI2 (99%, Sigma–Aldrich), MAI (> 98%, Tokyo Chemical Industry Co., Japan), PCBM (99.5%, Lumtec Co., Taiwan), bathocuproine (Wako), Graphene oxide (2 mg mL−1, Sigma-Aldrich), Super dehydrated dimethylsulfoxide (DMSO), gamma-Butyrolactone (GBL), toluene, and methanol were purchased from Wako, Japan. All the chemicals were used as received without further purification. 2.2. Device fabrication The r-GO film was prepared on top of the ITO glass from a 0.5 mg mL-1 aqueous solution and annealed at 150 °C for 90 min to form a 2.5 nm thick r-GO film. After cooling down the r-GO films to room temperature, a 10 nm uniform CuSCN layer (from a solution of 10 mg mL-1 in diethyl sulfide) was obtained by spin-coating at 6000 rpm for 60 s on top of it and was annealed for 30 min in air. The thickness of the CuSCN layer between 10 nm and 40 nm was obtained by controlling the rotation speed of the spin coater between 3000 and 6000 rpm (see Table S2). Then the r-GO/CuSCN coated ITO glasses were cooled down to room temperature and transferred to a N2 glovebox. A MAPbI3 precursor solution (100 μL) consisting of PbI2 (922 mg) and CH3NH3I (318 mg) dissolved in 2 mL of 3:7 (v/v) DMSO/GBL was spread over the r-GO/CuSCN film and spin-coated with two steps of spin-coating, first at 1000 rpm for 12 s and then 5000 rpm for 30 s. 100 μL of toluene was dropped onto the perovskite-coated r-GO/CuSCN film 10 s prior to the start of second stage of spin-coating at 5000 rpm. Finally, the film was annealed at 100 °C for 30 min resulting in a 400 nm thick MAPbI3 layer. All the MAPbI3 layers has been fabricated with similar thickness regardless in the thickness variation of HTL layers. After cooling down to room temperature, a solution of PCBM in chlorobenzene (20 mg mL−1) was spin-coated on top of the film at a rotation speed of 1000 rpm for 30 s. A saturated methanol solution of bathocuproine (140 μL) was spincoated onto the PCBM coated film with spin rotation of 6000 rpm for 30 s. The PSC fabrication was completed by thermal evaporation of a 100 nm thick film of Ag as the cathode. Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.solener.2018.07.022. 2.3. Characterization The thickness of the fabricated r-GO, CuSCN and r-GO/CuSCN films were measured with J.A. Woollam (M-2000X) elipsometer. The angel of incident light was set at 45° for reflectance measurement of thickness of all the films. FESEM images were obtained with a Hitachi S-4800 field emission scanning electron microscope. The atomic force microscopy images were obtained with JSPM-5200 scanning probe microscope.
Fig. 1. SEM surface images of (a) ITO/r-GO and (b) ITO/r-GO/CuSCN. 653
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
compact nature of the film. Generally in inverted planar PSCs, a compact under layer with low surface roughness can suppress the charge recombination by preventing direct contact between the perovskite and ITO surface (Wu et al., 2016; Chen et al., 2015; Yu et al., 2018). Hence, it was expected that our compact CuSCN layer on top of r-GO can perform similarly and improve the performance of the overall PSC. Upon illumination, light transmits through the ITO layer to activate the electron-hole pairs in the perovskite absorber layer. We have measured the transmittance and absorbance spectra by means of UV–Vis spectroscopy. Fig. 3(a) shows the transmittance spectra of ITO, ITO/r-GO, ITO/CuSCN, ITO/r-GO/CuSCN. The ITO glass shows 85% transmittance in 450–1000 nm region. Spin coating the r-GO on top of ITO glass increases the transmittance up to 87% in the same region. However, spin-coating of CuSCN on top of the ITO glass reduces the transmittance to 80% in the region, which might facilitate enough light transportation to the perovskite absorber. With high transparency provided by the r-GO/CuSCN layer, high light harvesting efficiency at the perovskite absorber is expected. Furthermore, the high absorption provided by the r-GO layer at the UV region may increase the light soaking stability of the r-GO/CuSCN based PSCs. Fig. 3(b) shows the schematic diagram structure of the fabricated PSC and Fig. 3(c) shows the energy level diagram of the fabricated device. The valence band of r-GO, CuSCN and MAPbI3 was observed at −4.97 eV, −5.30 eV and −5.4 eV respectively. It is noticeable that, the valence bad mismatch between r-GO and MAPbI3 absorber is 0.43 eV, which results in loss of hole collection at the ITO and eventually affect the VOC of the overall PSC. Insertion of the CuSCN layer can reduce this mismatch up to 0.1 eV, thereby reducing the potential loss of hole collection at the interface between r-GO and MAPbI3. We were unable to deposit the MAPbI3 layer from the conventional precursor of DMF: DMSO solution. The DMF: DMSO solvent based CH3NH3PbI3 precursor solution did not have any adhesion on top of the r-GO/CuSCN film. By replacing the DMSO with GBL and changing the ratio to 3:7 (DMSO: GBL), we have successfully formed MAPbI3 absorber layer on top of the r-GO and r-GO/CuSCN layers. We have further observed the UV–Vis absorption spectra of the MAPbI3 films on rGO and r-GO/CuSCN substrates Fig. 3(d)). The MAPbI3 film fabricated on top of r-GO/CuSCN bilayer HTL shows slightly higher absorption in comparison to MAPbI3 film on top of the r-GO layer. We have observed the surface morphology of MAPbI3 films fabricated on top of r-GO and r-GO/CuSCN films by SEM and shown in Fig. 4(a) and (b). The MAPbI3 films fabricated on top of the r-GO/CuSCN was more uniform and compact with larger grain sizes than the MAPbI3 films fabricated on top of the r-GO underlayer. The average grain sizes of the MAPbI3 perovskite films with r-GO and r-GO/CuSCN HTL were
UV–Vis spectra were measured on a Shimadzu UV–Vis 3600 spectrophotometer. Steady-state PL spectra were measured with an Edinburgh FLS 920 fluorescence spectrometer (Edinburgh). XRD patterns were collected on a Rigaku RINT-2500 powder X-ray diffractometer equipped with a Cu Kα radiation source. The energies of the highest occupied molecular orbitals of the films were estimated by photoelectron yield spectroscopy with an AC-3e spectrometer (Riken Keiki). Current–voltage (J-V) characteristics were measured by means of AM 1.5 illumination at 100 mW cm−2 with a solar simulator (WXS-155S-10, Wacom Denso) under ambient conditions with a delay time of 100 ms. J–V curves for all devices were measured by masking the devices with a metal mask with an area of 1.02 cm2. Monochromatic incident photonto-current conversion efficiency spectra were measured with monochromatic incident light (1 × 1016 photons cm−2) in direct-current mode (CEP-2000BX, Bunko- Keiki). The light intensity of the solar simulator was calibrated with a standard silicon solar cell (PV Measurements). Light-soaking stability was tested on a solar cell light resistance test system (BIR-50, Bunko- Keiki), and a Keithley was used for automatically recording J–V curves. 3. Resuts and discussion Recently we have reported on low temperature processed conversion method of GO to r-GO with low resistance. Briefly, A GO film was spin coated on ITO from a 0.5 mg mL−1 GO solution and annealed for 90 min at 150 °C to form a ∼2.5 nm thick r-GO film (Zhou et al., 2017). Fig. 1 shows the top view surface image of the 2.5 nm thick r-GO film on top of ITO glass substrate and 10 nm thick CuSCN interlayer deposited by spin coating on top of the r-GO coated ITO glass substrate obtained by Field emission scanning electron microscopy (FESEM). The thickness measurement parameters have been shown in Table S1. We have observed a low coverage of r-GO film on top of the ITO glass. The uncovered areas of r-GO can facilitate leakage current paths from the perovskite layer to the ITO and limit the overall performance of the PSC. Hence, we have introduced an ultrathin compact CuSCN layer on top of the r-GO layer to cover the uncovered areas of r-GO. The 10 nm CuSCN layer was deposited by a one-step spin coating process from a diethyl sulphide solution (15 mg mL-1) and annealed at 100 °C for 30 min. In Fig. 1 (b) the full coverage of the uncovered areas of r-GO by CuSCN can be observed. Fig. 2 shows the images obtained by atomic force microscopy of the r-GO layer and the optimized r-GO/CuSCN layer deposited on top of ITO glass respectively. The r-GO films showed a root mean square (r.m.s) surface roughness of 3.57 nm and the r-GO/ CuSCN layer had r.m.s. roughness of 3.16 nm. The reduced surface roughness provided by the r-GO/CuSCN layer surely indicates the
Fig. 2. Surface morphology of the (a) r-GO layer on top of ITO glass (b) CuSCN layer deposited on top of r-GO/ITO obtained by atomic force microscopy. 654
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
Fig. 3. (a) Transmittance spectra of ITO, ITO/r-GO, ITO/CuSCN, ITO/r-GO/CuSCN (b) Schematic diagram structure of the fabricated PSC (c) Energy level diagram of the fabricated PSC (d) Absorbance spectra of MAPbI3 deposited on top of r-GO and r-GO/CuSCN.
Fig. 4. SEM top view images of MAPbI3 films on (a) ITO/r-GO and (b) ITO/r-GO/CuSCN. Cross section images of PSCs based on (c) r-GO and (d) r-GO/CuSCN HTLs. 655
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
Table 1 Photovoltaic performance of the fabricated inverted planar PSCs with r-GO and r-GO/CuSCN HTMs.
ITO/r-GO/CuSCN/MAPbI3
Intensity (a.u.)
ITO/r-GO/MAPbI3 MAPbI3
725
750
780
810
HTL
Thickness of HTL
JSC (mA cm-2)
VOC (V)
FF
PCE (%)
RS (Ω)
r-GO r-GO/CuSCN
2.5 nm 12.5 nm
17.45 18.21
0.826 1.031
0.660 0.761
9.52 14.28
9.7 4.9
CuSCN film by comparing the performance of r-GO/CuSCN based PSCs with variation of thickness of the CuSCN layer between 10 nm and 40 nm (see Table S2). As the thickness of the CuSCN layer was increased a higher RS was observed for the corresponding PSCs (see Table S2). The successful optimization of 10 nm CuSCN layer between the r-GO and MAPbI3 significantly reduced the RS to 4.9 Ω, which is half fold in comparison to the RS of the r-GO based PSC (see Table 1). This significant reduction can be co-related for facilitating faster hole transportation Fig. 5. As a result, with the increased VOC up to 1.031 V, JSC of 18.21 mA cm-2 and fill factor of 0.785, the PSC showed PCE up to 14.28%. The details photovoltaic parameters of the fabricated PSCs with the corresponding HTLs have been shown in Table 1. The r-GO/ CuSCN HTL based PSCs showed high reproducibility (Fig. S3). Fig. 6(b) shows the incident photon to current conversion efficiency (IPCE) of the fabricated PSCS with r-GO HTL and r-GO/CuSCN bilayer HTL. The r-GO HTL based PSC shows 70% IPCE over the whole visible region. The integrated JSC for the r-GO based PSC was recorded at 17.13 mA cm-2. The r-GO/CuSCN bilayer HTL recovers the loss of IPCE in the UV region and boosts up to 80% in the whole visible spectra. The integrated JSC for the r-GO/CuSCN based PSC was recorded at 18.09 mA cm-2. We have evaluated the light soaking stability of the best performing PSCs under continuous illumination at 100 mW cm-2 under short circuit condition for 100 h (Fig. 7). The PCE of the r-GO HTL based PSC showed stable photovoltaic response by decreasing only 10% from the start of the light soaking. We have fabricated a PSC with 10 nm thin CuSCN layer and observed PCE of 11.77%. The CuSCN based PSC showed rapid reduction in photovoltaic performance and resulted in 5.54% PCE after 100 h continuous light soaking. However, the r-GO/ CuSCN based PSC showed steady photovoltaic performance reducing 10% of its initial PCE after light soaking in similar period of time. It is eminent that inclusion of r-GO HTL between the ITO and CuSCN certainly upholds the stability of the overall PSC. After 100 h light soaking the PCE of r-GO/CuSCN bilayer HTL based PSC was observed with 12.84%.
825
Wavelength (nm) Fig. 5. Steady state PL measurement of the CH3NH3PbI3 on top of glass, ITO/rGO and ITO/r-GO/CuSCN.
estimated from the grains observed from the SEM surface morphology Fig. 4 with an average of 229.74 nm and 577.20 nm respectively (Fig. S1). Two strong peaks of the MAPbI3 films corresponding to 14.11° and 28.41° were assigned to the (1 1 0) and (2 2 0) planes respectively (Fig. S2). The perovskite films grown on r-GO/CuSCN showed sharper peaks in comparison to the MAPbI3 fabricated on top of r-GO film. Fig. 4(c) and (d) shows the cross section of the corresponding PSCs. whereas, the r-GO/CuSCN bilayer HTL based PSC was denser in comparison to the only r-GO based PSC. To evaluate the charge dynamics of the hole extraction and transport between the of MAPbI3 and the corresponding HTLs we have studied steady state photoluminescence as shown in Fig. 5. The emission peak for the MAPbI3 film was recorded at ∼770 nm deposited on glass, ITO/r-GO and ITO/r-GO/CuSCN substrates respectively. The r-GO and r-GO/CuSCN layer rigorously quenched the MAPbI3 emission. For the r-GO/CuSCN bilayer HTL, the quenching was strongest which confirms faster hole extraction and transportation from the MAPbI3. Hence, the r-GO/CuSCN bilayer HTL effectively hinders the charge recombination at the corresponding interface with the MAPbI3 Fig. 6(a) shows the current density-voltage (J-V) curves of the fabricated PSCs with r-GO, CuSCN and r-GO/CuSCN HTLs measured under AM 1.5 simulation (100 mW cm-2 illumination) in forward bias mode. All the corresponding PSCs were fabricated with an active area of 1.02 cm2. The r-GO HTL based PSCs showed PCE of 9.52% with a VOC of 0.826 V, JSC of 17.45 mA cm-2 and fill factor of 0.66. r-GO HTL based PSC shows high RS value of 9.7 Ω which can be correlated with the poor surface coverage of the r-GO HTL. With non-uniform coverage of r-GO Fig. 1(a), it may lead to increased shunt paths between the MAPbI3 absorber and the ITO layer. We have optimized the thickness of the
4. Conclusion In this work, we have introduced a new bilayer hole transport
(b) 80
2
Current density (mA/cm )
(a) 20
IPCE (%)
15 10 5 0 0.0
0.4
40 20
r-GO r-GO/CuSCN 0.2
60
0.6
Voltage (V)
0.8
1.0 1.1
0 300
r-GO/CuSCN r-GO 450
600
Wavelength (nm)
Fig. 6. (a) J-V curves and (b) IPCE spectra of PSCs fabricated with r-GO and r-GO/CuSCN. 656
750
820
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
15
to PEDOT: PSS hole transport layer for efficient and stable inverted planar heterojunction perovskite solar cells. J. Mater. Chem. A 3 (38), 19353–19359. Chen, W., Wu, Y., Yue, Y., Liu, J., Zhang, W., Yang, X., Chen, H., Bi, E., Ashraful, I., Grätzel, M., 2015. Efficient and stable large-area perovskite solar cells with inorganic charge extraction layers. Science 350 (6263), 944–948. Chowdhury, T.H., Kaneko, R., Kayesh, M.E., Akhtaruzzaman, M., Sopian, K.B., Lee, J.-J., Islam, A., 2018. Nanostructured NiOx as hole transport material for low temperature processed stable perovskite solar cells. Mater. Lett. 223, 109–111. Correa-Baena, J.-P., Abate, A., Saliba, M., Tress, W., Jacobsson, T.J., Grätzel, M., Hagfeldt, A., 2017. The rapid evolution of highly efficient perovskite solar cells. Energy Environ. Sci. 10 (3), 710–727. Green, M.A., Hishikawa, Y., Dunlop, E.D., Levi, D.H., Hohl-Ebinger, J., Ho-Baillie, A.W.Y., 2018. Solar cell efficiency tables (version 51). Progr. Photovolt.: Res. Appl. 26, 3–12. Green, M.A., Ho-Baillie, A., 2017. Perovskite solar cells: the birth of a new era in photovoltaics. ACS Energy Lett. 2 (4), 822–830. Jeng, J.-Y., Chiang, Y.-F., Lee, M.-H., Peng, S.-R., Guo, T.-F., Chen, P., Wen, T.-C., 2013. CH3NH3PbI3 perovskite/fullerene planar-heterojunction hybrid solar cells. Adv. Mater. 25 (27), 3727–3732. Jung, J.W., Chueh, C.C., Jen, A.K.Y., 2015. High-performance semitransparent perovskite solar cells with 10% power conversion efficiency and 25% average visible transmittance based on transparent CuSCN as the hole-transporting material. Adv. Energy Mater. 5 (17). Liu, J., Kim, G.-H., Xue, Y., Kim, J.Y., Baek, J.-B., Durstock, M., Dai, L., 2014. Graphene oxide nanoribbon as hole extraction layer to enhance efficiency and stability of polymer solar cells. Adv. Mater. 26 (5), 786–790. Park, N.-G., Grätzel, M., Miyasaka, T., Zhu, K., Emery, K., 2016. Towards stable and commercially available perovskite solar cells. Nat. Energy 1 (11), 16152. Seok, S.I., Grätzel, M., Park, N.G., 2018. Methodologies toward highly efficient perovskite solar cells. Small 14 (20), 1704177. Shalan, A.E., Oshikiri, T., Narra, S., Elshanawany, M.M., Ueno, K., Wu, H.-P., Nakamura, K., Shi, X., Diau, E.W.-G., Misawa, H., 2016. Cobalt oxide (CoOx) as an efficient holeextracting layer for high-performance inverted planar perovskite solar cells. ACS Appl. Mater. Interfaces 8 (49), 33592–33600. Sun, W., Ye, S., Rao, H., Li, Y., Liu, Z., Xiao, L., Chen, Z., Bian, Z., Huang, C., 2016. Roomtemperature and solution-processed copper iodide as the hole transport layer for inverted planar perovskite solar cells. Nanoscale 8 (35), 15954–15960. Wu, Y., Yang, X., Chen, W., Yue, Y., Cai, M., Xie, F., Bi, E., Islam, A., Han, L., 2016. Perovskite solar cells with 18.21% efficiency and area over 1 cm2 fabricated by heterojunction engineering. Nat. Energy 1, 16148. Yang, Q.-D., Li, J., Cheng, Y., Li, H.-W., Guan, Z., Yu, B., Tsang, S.-W., 2017. Graphene oxide as an efficient hole-transporting material for high-performance perovskite solar cells with enhanced stability. J. Mater. Chem. A 5 (20), 9852–9858. Ye, S., Sun, W., Li, Y., Yan, W., Peng, H., Bian, Z., Liu, Z., Huang, C., 2015. CuSCN-based inverted planar perovskite solar cell with an average PCE of 15.6%. Nano Lett. 15 (6), 3723–3728. Yeo, J.-S., Kang, R., Lee, S., Jeon, Y.-J., Myoung, N., Lee, C.-L., Kim, D.-Y., Yun, J.-M., Seo, Y.-H., Kim, S.-S., 2015. Highly efficient and stable planar perovskite solar cells with reduced graphene oxide nanosheets as electrode interlayer. Nano Energy 12, 96–104. Yu, Z.-K., Fu, W.-F., Liu, W.-Q., Zhang, Z.-Q., Liu, Y.-J., Yan, J.-L., Ye, T., Yang, W.-T., Li, H.-Y., Chen, H.-Z., 2017. Solution-processed CuOx as an efficient hole-extraction layer for inverted planar heterojunction perovskite solar cells. Chin. Chem. Lett. 28 (1), 13–18. Yu, J.C., Hong, J.A., Jung, E.D., Kim, D.B., Baek, S.-M., Lee, S., Cho, S., Park, S.S., Choi, K.J., Song, M.H., 2018. Highly efficient and stable inverted perovskite solar cell employing PEDOT:GO composite layer as a hole transport layer. Sci. Reports 8 (1), 1070. Zhao, K., Munir, R., Yan, B., Yang, Y., Kim, T., Amassian, A., 2015. Solution-processed inorganic copper (I) thiocyanate (CuSCN) hole transporting layers for efficient p–i–n perovskite solar cells. J. Mater. Chem. A 3 (41), 20554–20559. Zhou, Z., Li, X., Cai, M., Xie, F., Wu, Y., Lan, Z., Yang, X., Qiang, Y., Islam, A., Han, L., 2017. Stable inverted planar perovskite solar cells with low-temperature-processed hole-transport bilayer. Adv. Energy Mater. 7 (22), 1700763.
PCE (%)
12 9 6 r-GO CuSCN r-GO/CuSCN
3 0
0
20
40
60
80
100
Time (h) Fig. 7. Light soaking stability of the unsealed PSCs fabricated by r-GO, CuSCN and r-GO/CuSCN HTLs.
structure with inorganic CuSCN HTL for low temperature processed rGO based inverted planar PSCs. The r-GO/CuSCN bilayer HTL provided a platform for the growth of perovskite crystals with larger grain sizes resulting in higher PCE compared to the individual r-GO based HTLs. The bilayer structure reduced the energy-level gradient and successfully supressed charge recombination at HTL/perovskite interface. The PCE of the r-GO/CuSCN based hole transport bilayer was 14.28% with improved light soaking stability up to 100 h. However, further optimization must be carried out to increase the PCE of the r-GO/CuSCN bilayer structured inverted planar PSCs. This work surely highlights the viability of bilayer hole transport layers with inorganic compounds for low temperature processed PSCs with a standard surface area and facilitates the up-scaling of PSCs for roll-to-roll production one step closer. Acknowledgements M.A Acknowledges the financial support from the FRGS/1/2017/ TK07/UKM/02/9 grant. A.I acknowledges the support from JSPS KAKENHI with grant No. 18H02079. J-J.L. acknowledges the support from NRF-2016M1A2A2940912. References Acik, M., Darling, S.B., 2016. Graphene in perovskite solar cells: device design, characterization and implementation. J. Mater. Chem. A 4 (17), 6185–6235. Agresti, A., Pescetelli, S., Taheri, B., Del Rio Castillo, A.E., Cinà, L., Bonaccorso, F., Di Carlo, A., 2016. Graphene-perovskite solar cells exceed 18 % efficiency: a stability study. ChemSusChem 9 (18), 2609–2619. Chatterjee, S., Pal, A.J., 2016. Introducing Cu2O thin films as a hole-transport layer in efficient planar perovskite solar cell structures. J. Phys. Chem. C 120 (3), 1428–1437. Chen, W.-Y., Deng, L.-L., Dai, S.-M., Wang, X., Tian, C.-B., Zhan, X.-X., Xie, S.-Y., Huang, R.-B., Zheng, L.-S., 2015. Low-cost solution-processed copper iodide as an alternative
657
Contents lists available at ScienceDirect
Solar Energy journal homepage: www.elsevier.com/locate/solener
Low temperature processed inverted planar perovskite solar cells by r-GO/ CuSCN hole-transport bilayer with improved stability
T
⁎
Towhid H. Chowdhurya,b,c, Md. Akhtaruzzamanb, , Md. Emrul Kayesha, Ryuji Kanekoa, ⁎ Takeshi Nodaa, Jae-Joon Leec, Ashraful Islama, a
Photovoltaic Materials Group, Center for Green Research on Energy and Environmental Materials, National Institute for Materials Science (NIMS), 1-2-1 Sengen, Tsukuba 305-0047, Japan b Solar Energy Research Institute, Universiti Kebangsaan Malaysia, Bangi 43600, Selangor Darul Ehsan, Malaysia c Department of Energy & Materials Engineering & Research Center for Photoenergy Harvesting and Conversion Technology, Dongguk University, Seoul 04620, Republic of Korea
A R T I C LE I N FO
A B S T R A C T
Keywords: Copper (I) thiocyanate Hole-transport bilayer Perovskite solar cell Reduced graphene oxide Stability
Low temperature processed Perovskite solar cells (PSCs) are popular due to their potential for scalable production. In this work, we report reduced Graphene Oxide (r-GO)/copper (I) thiocyanate (CuSCN) as an efficient bilayer hole transport layer (HTL) for low temperature processed inverted planar PSCs. We have systematically optimized the thickness of CuSCN interlayer at the r-GO/MAPbI3 interface resulting in bilayer HTL structure to enhance the stability and photovoltaic performance of low temperature processed r-GO HTL based PSCs with a standard surface area of 1.02 cm2. With matched valence band energy level, the r-GO/CuSCN bilayer HTL based PSCs showed high power conversion efficiency of 14.28%, thanks to the improved open circuit voltage (VOC) compared to the only r-GO based PSC. Moreover, enhanced stability has been observed for the r-GO/CuSCN based PSCs which retained over 90% of its initial efficiency after 100 h light soaking measured under continuous AM 1.5 sun illumination.
1. Introduction PSCs has been identified as the most promising solar technology for energy harvesting due to their potential high efficiency and easy fabrication process (Park et al., 2016; Correa-Baena et al., 2017; Green and Ho-Baillie, 2017; Seok et al., 2018). Since the inception of PSCs, two main structures such as- mesoporous and inverted planar structures have been under intense investigation. Until now, the highest power conversion efficiency (PCE) of 22.7% has been observed with mesoporous PSCs (Green et al., 2018). However, in the mesoporous structure, the fabrication of electron transport layer requires high temperature (500 °C) sintering process, which hinders their potential aspects for commercialization and potential application in flexible electronic devices. In this aspect, low temperature processed PSCs are more anticipated. Inverted planar structured PSCs with their total low temperature fabrication compatibility is steadily attracting considerable attention. Generally inverted planar PSCs are fabricated with a p-type semiconducting material which acts as a hole transport layer (HTL) is deposited on top of the transparent conductive oxide glass followed by
⁎
perovskite compound as the absorber and n-type semiconducting material as the electron transport material (ETM). The first inverted planar PSCs were originated from the concept of organic photovoltaics where, PEDOT:PSS was used as the hole transport material (Jeng et al., 2013). The low temperature processed PSC showed impressive photovoltaic performances but resulted in rapid degradation and induced decomposition of the perovskite absorber due to the acidic behaviour of the PEDOT: PSS (Chen et al., 2015). To address this issue, recently we have reported on low temperature processed stable inverted planar PSCs with NiOx nanoparticles (Chowdhury et al., 2018). Alternate low temperature processed HTLs such as Cu2O (Chatterjee and Pal, 2016), CuI (Sun et al., 2016), CuOx (Yu et al., 2017), CoOx (Shalan et al., 2016), CuSCN (Zhao et al., 2015; Ye et al., 2015) has been implemented in PSCs and showed promising aspects but failed to provide stable performances. In stability concern, carbon derivatives such as Graphene (Agresti et al., 2016), Graphene Oxide (GO) (Yang et al., 2017; Liu et al., 2014) and reduced Graphene Oxide (r-GO) (Yeo et al., 2015) are popular choice of HTLs in PSCs. With the capability of higher absorption in the near UV region, GO based PSCs can provide long term stable performance in inverted planar PSCs (Yang et al., 2017). However, for
Corresponding authors. E-mail address: [email protected] (A. Islam).
https://doi.org/10.1016/j.solener.2018.07.022 Received 24 April 2018; Received in revised form 28 June 2018; Accepted 7 July 2018 0038-092X/ © 2018 Elsevier Ltd. All rights reserved.
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
2. Experimental
obtaining high photovoltaic response these GO based HTLs require high sintering temperature or vacuum based evaporation techniques (Acik and Darling, 2016). It has been occasionally revealed that, deposition of low temperature processed r-GO fails to cover the whole surface of the transparent conductive glass eventually providing opportunity for large defects at the interface with the perovskite absorber. These defects at the surface of r-GO eventually leads to potential recombination centres creating instability of the overall PSC. Moreover, the valence band of low temperature processed r-GO is −4.97 eV which has a huge energy mismatch with the perovskite compound valence band (−5.4 eV). This huge mismatch can hinder the potential collection of holes at the r-GO/ perovskite interface and limit the photovoltaic performance of the overall PSC. To overcome these issues, we have recently reported with bilayer structured HTLs with r-GO and organic polymer, poly(triarylamine) (PTAA) and shown promising results (Zhou et al., 2017). However, in the long run, organic compounds as HTLs might not be a suitable choice due to their vulnerable behaviour in ambient environment. With an objective to obtain higher photovoltaic performance and potential commercialization aspect, insertion of an inorganic HTL with higher valence band than r-GO may increase the hole collection. CuSCN, a p-type material with a valence band of −5.30 eV and bandgap of 3.6 eV (Jung et al., 2015) can be considered as one of the potential interfacial material to transport hole faster in this aspect. However, to achieve maximum transparency, an ultrathin layer of CuSCN layer should be beneficial. The potential high bandgap with matched valence band level with the corresponding perovskite absorber can provide sufficient transparency across the UV–Vis/NIR wavelength region and lower loss caused by parasitic absorption. Here in this work, we introduce an ultrathin CuSCN layer as an efficient interlayer in low temperature processed r-GO based inverted planar PSCs. We have systematically optimized the thickness of CuSCN interlayer at the r-GO/Perovskite interface to enhance the photovoltaic performance of the PSC with a surface area of 1.02 cm2. With matched energy level, the r-GO/CuSCN bilayer HTM successfully prevented the recombination at the interface with the MAPbI3 absorber layer. Simultaneously, the r-GO/CuSCN bilayer HTL leads to faster hole extraction as confirmed by the photoluminescence study. The r-GO/ CuSCN bilayer HTL based PSC showed a high power conversion efficiency (PCE) of 14.28% with photovoltaic parameters of open circuit voltage (VOC) = 1.031 V, short circuit current density (JSC) = 18.21 mA cm-2, fill factor (FF) = 0.761. The PCE of the r-GO/ CuSCN bilayer HTL based PSCs were higher than that of the individual r-GO based PSCs (PCE = 9.52%). Additionally, enhanced stability has been observed for the r-GO/CuSCN bilayer HTL based PSCs retaining over the 90% initial efficiency after 100 h light soaking.
2.1. Materials The following chemicals were obtained from commercial suppliers and used as received: PbI2 (99%, Sigma–Aldrich), MAI (> 98%, Tokyo Chemical Industry Co., Japan), PCBM (99.5%, Lumtec Co., Taiwan), bathocuproine (Wako), Graphene oxide (2 mg mL−1, Sigma-Aldrich), Super dehydrated dimethylsulfoxide (DMSO), gamma-Butyrolactone (GBL), toluene, and methanol were purchased from Wako, Japan. All the chemicals were used as received without further purification. 2.2. Device fabrication The r-GO film was prepared on top of the ITO glass from a 0.5 mg mL-1 aqueous solution and annealed at 150 °C for 90 min to form a 2.5 nm thick r-GO film. After cooling down the r-GO films to room temperature, a 10 nm uniform CuSCN layer (from a solution of 10 mg mL-1 in diethyl sulfide) was obtained by spin-coating at 6000 rpm for 60 s on top of it and was annealed for 30 min in air. The thickness of the CuSCN layer between 10 nm and 40 nm was obtained by controlling the rotation speed of the spin coater between 3000 and 6000 rpm (see Table S2). Then the r-GO/CuSCN coated ITO glasses were cooled down to room temperature and transferred to a N2 glovebox. A MAPbI3 precursor solution (100 μL) consisting of PbI2 (922 mg) and CH3NH3I (318 mg) dissolved in 2 mL of 3:7 (v/v) DMSO/GBL was spread over the r-GO/CuSCN film and spin-coated with two steps of spin-coating, first at 1000 rpm for 12 s and then 5000 rpm for 30 s. 100 μL of toluene was dropped onto the perovskite-coated r-GO/CuSCN film 10 s prior to the start of second stage of spin-coating at 5000 rpm. Finally, the film was annealed at 100 °C for 30 min resulting in a 400 nm thick MAPbI3 layer. All the MAPbI3 layers has been fabricated with similar thickness regardless in the thickness variation of HTL layers. After cooling down to room temperature, a solution of PCBM in chlorobenzene (20 mg mL−1) was spin-coated on top of the film at a rotation speed of 1000 rpm for 30 s. A saturated methanol solution of bathocuproine (140 μL) was spincoated onto the PCBM coated film with spin rotation of 6000 rpm for 30 s. The PSC fabrication was completed by thermal evaporation of a 100 nm thick film of Ag as the cathode. Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.solener.2018.07.022. 2.3. Characterization The thickness of the fabricated r-GO, CuSCN and r-GO/CuSCN films were measured with J.A. Woollam (M-2000X) elipsometer. The angel of incident light was set at 45° for reflectance measurement of thickness of all the films. FESEM images were obtained with a Hitachi S-4800 field emission scanning electron microscope. The atomic force microscopy images were obtained with JSPM-5200 scanning probe microscope.
Fig. 1. SEM surface images of (a) ITO/r-GO and (b) ITO/r-GO/CuSCN. 653
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
compact nature of the film. Generally in inverted planar PSCs, a compact under layer with low surface roughness can suppress the charge recombination by preventing direct contact between the perovskite and ITO surface (Wu et al., 2016; Chen et al., 2015; Yu et al., 2018). Hence, it was expected that our compact CuSCN layer on top of r-GO can perform similarly and improve the performance of the overall PSC. Upon illumination, light transmits through the ITO layer to activate the electron-hole pairs in the perovskite absorber layer. We have measured the transmittance and absorbance spectra by means of UV–Vis spectroscopy. Fig. 3(a) shows the transmittance spectra of ITO, ITO/r-GO, ITO/CuSCN, ITO/r-GO/CuSCN. The ITO glass shows 85% transmittance in 450–1000 nm region. Spin coating the r-GO on top of ITO glass increases the transmittance up to 87% in the same region. However, spin-coating of CuSCN on top of the ITO glass reduces the transmittance to 80% in the region, which might facilitate enough light transportation to the perovskite absorber. With high transparency provided by the r-GO/CuSCN layer, high light harvesting efficiency at the perovskite absorber is expected. Furthermore, the high absorption provided by the r-GO layer at the UV region may increase the light soaking stability of the r-GO/CuSCN based PSCs. Fig. 3(b) shows the schematic diagram structure of the fabricated PSC and Fig. 3(c) shows the energy level diagram of the fabricated device. The valence band of r-GO, CuSCN and MAPbI3 was observed at −4.97 eV, −5.30 eV and −5.4 eV respectively. It is noticeable that, the valence bad mismatch between r-GO and MAPbI3 absorber is 0.43 eV, which results in loss of hole collection at the ITO and eventually affect the VOC of the overall PSC. Insertion of the CuSCN layer can reduce this mismatch up to 0.1 eV, thereby reducing the potential loss of hole collection at the interface between r-GO and MAPbI3. We were unable to deposit the MAPbI3 layer from the conventional precursor of DMF: DMSO solution. The DMF: DMSO solvent based CH3NH3PbI3 precursor solution did not have any adhesion on top of the r-GO/CuSCN film. By replacing the DMSO with GBL and changing the ratio to 3:7 (DMSO: GBL), we have successfully formed MAPbI3 absorber layer on top of the r-GO and r-GO/CuSCN layers. We have further observed the UV–Vis absorption spectra of the MAPbI3 films on rGO and r-GO/CuSCN substrates Fig. 3(d)). The MAPbI3 film fabricated on top of r-GO/CuSCN bilayer HTL shows slightly higher absorption in comparison to MAPbI3 film on top of the r-GO layer. We have observed the surface morphology of MAPbI3 films fabricated on top of r-GO and r-GO/CuSCN films by SEM and shown in Fig. 4(a) and (b). The MAPbI3 films fabricated on top of the r-GO/CuSCN was more uniform and compact with larger grain sizes than the MAPbI3 films fabricated on top of the r-GO underlayer. The average grain sizes of the MAPbI3 perovskite films with r-GO and r-GO/CuSCN HTL were
UV–Vis spectra were measured on a Shimadzu UV–Vis 3600 spectrophotometer. Steady-state PL spectra were measured with an Edinburgh FLS 920 fluorescence spectrometer (Edinburgh). XRD patterns were collected on a Rigaku RINT-2500 powder X-ray diffractometer equipped with a Cu Kα radiation source. The energies of the highest occupied molecular orbitals of the films were estimated by photoelectron yield spectroscopy with an AC-3e spectrometer (Riken Keiki). Current–voltage (J-V) characteristics were measured by means of AM 1.5 illumination at 100 mW cm−2 with a solar simulator (WXS-155S-10, Wacom Denso) under ambient conditions with a delay time of 100 ms. J–V curves for all devices were measured by masking the devices with a metal mask with an area of 1.02 cm2. Monochromatic incident photonto-current conversion efficiency spectra were measured with monochromatic incident light (1 × 1016 photons cm−2) in direct-current mode (CEP-2000BX, Bunko- Keiki). The light intensity of the solar simulator was calibrated with a standard silicon solar cell (PV Measurements). Light-soaking stability was tested on a solar cell light resistance test system (BIR-50, Bunko- Keiki), and a Keithley was used for automatically recording J–V curves. 3. Resuts and discussion Recently we have reported on low temperature processed conversion method of GO to r-GO with low resistance. Briefly, A GO film was spin coated on ITO from a 0.5 mg mL−1 GO solution and annealed for 90 min at 150 °C to form a ∼2.5 nm thick r-GO film (Zhou et al., 2017). Fig. 1 shows the top view surface image of the 2.5 nm thick r-GO film on top of ITO glass substrate and 10 nm thick CuSCN interlayer deposited by spin coating on top of the r-GO coated ITO glass substrate obtained by Field emission scanning electron microscopy (FESEM). The thickness measurement parameters have been shown in Table S1. We have observed a low coverage of r-GO film on top of the ITO glass. The uncovered areas of r-GO can facilitate leakage current paths from the perovskite layer to the ITO and limit the overall performance of the PSC. Hence, we have introduced an ultrathin compact CuSCN layer on top of the r-GO layer to cover the uncovered areas of r-GO. The 10 nm CuSCN layer was deposited by a one-step spin coating process from a diethyl sulphide solution (15 mg mL-1) and annealed at 100 °C for 30 min. In Fig. 1 (b) the full coverage of the uncovered areas of r-GO by CuSCN can be observed. Fig. 2 shows the images obtained by atomic force microscopy of the r-GO layer and the optimized r-GO/CuSCN layer deposited on top of ITO glass respectively. The r-GO films showed a root mean square (r.m.s) surface roughness of 3.57 nm and the r-GO/ CuSCN layer had r.m.s. roughness of 3.16 nm. The reduced surface roughness provided by the r-GO/CuSCN layer surely indicates the
Fig. 2. Surface morphology of the (a) r-GO layer on top of ITO glass (b) CuSCN layer deposited on top of r-GO/ITO obtained by atomic force microscopy. 654
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
Fig. 3. (a) Transmittance spectra of ITO, ITO/r-GO, ITO/CuSCN, ITO/r-GO/CuSCN (b) Schematic diagram structure of the fabricated PSC (c) Energy level diagram of the fabricated PSC (d) Absorbance spectra of MAPbI3 deposited on top of r-GO and r-GO/CuSCN.
Fig. 4. SEM top view images of MAPbI3 films on (a) ITO/r-GO and (b) ITO/r-GO/CuSCN. Cross section images of PSCs based on (c) r-GO and (d) r-GO/CuSCN HTLs. 655
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
Table 1 Photovoltaic performance of the fabricated inverted planar PSCs with r-GO and r-GO/CuSCN HTMs.
ITO/r-GO/CuSCN/MAPbI3
Intensity (a.u.)
ITO/r-GO/MAPbI3 MAPbI3
725
750
780
810
HTL
Thickness of HTL
JSC (mA cm-2)
VOC (V)
FF
PCE (%)
RS (Ω)
r-GO r-GO/CuSCN
2.5 nm 12.5 nm
17.45 18.21
0.826 1.031
0.660 0.761
9.52 14.28
9.7 4.9
CuSCN film by comparing the performance of r-GO/CuSCN based PSCs with variation of thickness of the CuSCN layer between 10 nm and 40 nm (see Table S2). As the thickness of the CuSCN layer was increased a higher RS was observed for the corresponding PSCs (see Table S2). The successful optimization of 10 nm CuSCN layer between the r-GO and MAPbI3 significantly reduced the RS to 4.9 Ω, which is half fold in comparison to the RS of the r-GO based PSC (see Table 1). This significant reduction can be co-related for facilitating faster hole transportation Fig. 5. As a result, with the increased VOC up to 1.031 V, JSC of 18.21 mA cm-2 and fill factor of 0.785, the PSC showed PCE up to 14.28%. The details photovoltaic parameters of the fabricated PSCs with the corresponding HTLs have been shown in Table 1. The r-GO/ CuSCN HTL based PSCs showed high reproducibility (Fig. S3). Fig. 6(b) shows the incident photon to current conversion efficiency (IPCE) of the fabricated PSCS with r-GO HTL and r-GO/CuSCN bilayer HTL. The r-GO HTL based PSC shows 70% IPCE over the whole visible region. The integrated JSC for the r-GO based PSC was recorded at 17.13 mA cm-2. The r-GO/CuSCN bilayer HTL recovers the loss of IPCE in the UV region and boosts up to 80% in the whole visible spectra. The integrated JSC for the r-GO/CuSCN based PSC was recorded at 18.09 mA cm-2. We have evaluated the light soaking stability of the best performing PSCs under continuous illumination at 100 mW cm-2 under short circuit condition for 100 h (Fig. 7). The PCE of the r-GO HTL based PSC showed stable photovoltaic response by decreasing only 10% from the start of the light soaking. We have fabricated a PSC with 10 nm thin CuSCN layer and observed PCE of 11.77%. The CuSCN based PSC showed rapid reduction in photovoltaic performance and resulted in 5.54% PCE after 100 h continuous light soaking. However, the r-GO/ CuSCN based PSC showed steady photovoltaic performance reducing 10% of its initial PCE after light soaking in similar period of time. It is eminent that inclusion of r-GO HTL between the ITO and CuSCN certainly upholds the stability of the overall PSC. After 100 h light soaking the PCE of r-GO/CuSCN bilayer HTL based PSC was observed with 12.84%.
825
Wavelength (nm) Fig. 5. Steady state PL measurement of the CH3NH3PbI3 on top of glass, ITO/rGO and ITO/r-GO/CuSCN.
estimated from the grains observed from the SEM surface morphology Fig. 4 with an average of 229.74 nm and 577.20 nm respectively (Fig. S1). Two strong peaks of the MAPbI3 films corresponding to 14.11° and 28.41° were assigned to the (1 1 0) and (2 2 0) planes respectively (Fig. S2). The perovskite films grown on r-GO/CuSCN showed sharper peaks in comparison to the MAPbI3 fabricated on top of r-GO film. Fig. 4(c) and (d) shows the cross section of the corresponding PSCs. whereas, the r-GO/CuSCN bilayer HTL based PSC was denser in comparison to the only r-GO based PSC. To evaluate the charge dynamics of the hole extraction and transport between the of MAPbI3 and the corresponding HTLs we have studied steady state photoluminescence as shown in Fig. 5. The emission peak for the MAPbI3 film was recorded at ∼770 nm deposited on glass, ITO/r-GO and ITO/r-GO/CuSCN substrates respectively. The r-GO and r-GO/CuSCN layer rigorously quenched the MAPbI3 emission. For the r-GO/CuSCN bilayer HTL, the quenching was strongest which confirms faster hole extraction and transportation from the MAPbI3. Hence, the r-GO/CuSCN bilayer HTL effectively hinders the charge recombination at the corresponding interface with the MAPbI3 Fig. 6(a) shows the current density-voltage (J-V) curves of the fabricated PSCs with r-GO, CuSCN and r-GO/CuSCN HTLs measured under AM 1.5 simulation (100 mW cm-2 illumination) in forward bias mode. All the corresponding PSCs were fabricated with an active area of 1.02 cm2. The r-GO HTL based PSCs showed PCE of 9.52% with a VOC of 0.826 V, JSC of 17.45 mA cm-2 and fill factor of 0.66. r-GO HTL based PSC shows high RS value of 9.7 Ω which can be correlated with the poor surface coverage of the r-GO HTL. With non-uniform coverage of r-GO Fig. 1(a), it may lead to increased shunt paths between the MAPbI3 absorber and the ITO layer. We have optimized the thickness of the
4. Conclusion In this work, we have introduced a new bilayer hole transport
(b) 80
2
Current density (mA/cm )
(a) 20
IPCE (%)
15 10 5 0 0.0
0.4
40 20
r-GO r-GO/CuSCN 0.2
60
0.6
Voltage (V)
0.8
1.0 1.1
0 300
r-GO/CuSCN r-GO 450
600
Wavelength (nm)
Fig. 6. (a) J-V curves and (b) IPCE spectra of PSCs fabricated with r-GO and r-GO/CuSCN. 656
750
820
Solar Energy 171 (2018) 652–657
T.H. Chowdhury et al.
15
to PEDOT: PSS hole transport layer for efficient and stable inverted planar heterojunction perovskite solar cells. J. Mater. Chem. A 3 (38), 19353–19359. Chen, W., Wu, Y., Yue, Y., Liu, J., Zhang, W., Yang, X., Chen, H., Bi, E., Ashraful, I., Grätzel, M., 2015. Efficient and stable large-area perovskite solar cells with inorganic charge extraction layers. Science 350 (6263), 944–948. Chowdhury, T.H., Kaneko, R., Kayesh, M.E., Akhtaruzzaman, M., Sopian, K.B., Lee, J.-J., Islam, A., 2018. Nanostructured NiOx as hole transport material for low temperature processed stable perovskite solar cells. Mater. Lett. 223, 109–111. Correa-Baena, J.-P., Abate, A., Saliba, M., Tress, W., Jacobsson, T.J., Grätzel, M., Hagfeldt, A., 2017. The rapid evolution of highly efficient perovskite solar cells. Energy Environ. Sci. 10 (3), 710–727. Green, M.A., Hishikawa, Y., Dunlop, E.D., Levi, D.H., Hohl-Ebinger, J., Ho-Baillie, A.W.Y., 2018. Solar cell efficiency tables (version 51). Progr. Photovolt.: Res. Appl. 26, 3–12. Green, M.A., Ho-Baillie, A., 2017. Perovskite solar cells: the birth of a new era in photovoltaics. ACS Energy Lett. 2 (4), 822–830. Jeng, J.-Y., Chiang, Y.-F., Lee, M.-H., Peng, S.-R., Guo, T.-F., Chen, P., Wen, T.-C., 2013. CH3NH3PbI3 perovskite/fullerene planar-heterojunction hybrid solar cells. Adv. Mater. 25 (27), 3727–3732. Jung, J.W., Chueh, C.C., Jen, A.K.Y., 2015. High-performance semitransparent perovskite solar cells with 10% power conversion efficiency and 25% average visible transmittance based on transparent CuSCN as the hole-transporting material. Adv. Energy Mater. 5 (17). Liu, J., Kim, G.-H., Xue, Y., Kim, J.Y., Baek, J.-B., Durstock, M., Dai, L., 2014. Graphene oxide nanoribbon as hole extraction layer to enhance efficiency and stability of polymer solar cells. Adv. Mater. 26 (5), 786–790. Park, N.-G., Grätzel, M., Miyasaka, T., Zhu, K., Emery, K., 2016. Towards stable and commercially available perovskite solar cells. Nat. Energy 1 (11), 16152. Seok, S.I., Grätzel, M., Park, N.G., 2018. Methodologies toward highly efficient perovskite solar cells. Small 14 (20), 1704177. Shalan, A.E., Oshikiri, T., Narra, S., Elshanawany, M.M., Ueno, K., Wu, H.-P., Nakamura, K., Shi, X., Diau, E.W.-G., Misawa, H., 2016. Cobalt oxide (CoOx) as an efficient holeextracting layer for high-performance inverted planar perovskite solar cells. ACS Appl. Mater. Interfaces 8 (49), 33592–33600. Sun, W., Ye, S., Rao, H., Li, Y., Liu, Z., Xiao, L., Chen, Z., Bian, Z., Huang, C., 2016. Roomtemperature and solution-processed copper iodide as the hole transport layer for inverted planar perovskite solar cells. Nanoscale 8 (35), 15954–15960. Wu, Y., Yang, X., Chen, W., Yue, Y., Cai, M., Xie, F., Bi, E., Islam, A., Han, L., 2016. Perovskite solar cells with 18.21% efficiency and area over 1 cm2 fabricated by heterojunction engineering. Nat. Energy 1, 16148. Yang, Q.-D., Li, J., Cheng, Y., Li, H.-W., Guan, Z., Yu, B., Tsang, S.-W., 2017. Graphene oxide as an efficient hole-transporting material for high-performance perovskite solar cells with enhanced stability. J. Mater. Chem. A 5 (20), 9852–9858. Ye, S., Sun, W., Li, Y., Yan, W., Peng, H., Bian, Z., Liu, Z., Huang, C., 2015. CuSCN-based inverted planar perovskite solar cell with an average PCE of 15.6%. Nano Lett. 15 (6), 3723–3728. Yeo, J.-S., Kang, R., Lee, S., Jeon, Y.-J., Myoung, N., Lee, C.-L., Kim, D.-Y., Yun, J.-M., Seo, Y.-H., Kim, S.-S., 2015. Highly efficient and stable planar perovskite solar cells with reduced graphene oxide nanosheets as electrode interlayer. Nano Energy 12, 96–104. Yu, Z.-K., Fu, W.-F., Liu, W.-Q., Zhang, Z.-Q., Liu, Y.-J., Yan, J.-L., Ye, T., Yang, W.-T., Li, H.-Y., Chen, H.-Z., 2017. Solution-processed CuOx as an efficient hole-extraction layer for inverted planar heterojunction perovskite solar cells. Chin. Chem. Lett. 28 (1), 13–18. Yu, J.C., Hong, J.A., Jung, E.D., Kim, D.B., Baek, S.-M., Lee, S., Cho, S., Park, S.S., Choi, K.J., Song, M.H., 2018. Highly efficient and stable inverted perovskite solar cell employing PEDOT:GO composite layer as a hole transport layer. Sci. Reports 8 (1), 1070. Zhao, K., Munir, R., Yan, B., Yang, Y., Kim, T., Amassian, A., 2015. Solution-processed inorganic copper (I) thiocyanate (CuSCN) hole transporting layers for efficient p–i–n perovskite solar cells. J. Mater. Chem. A 3 (41), 20554–20559. Zhou, Z., Li, X., Cai, M., Xie, F., Wu, Y., Lan, Z., Yang, X., Qiang, Y., Islam, A., Han, L., 2017. Stable inverted planar perovskite solar cells with low-temperature-processed hole-transport bilayer. Adv. Energy Mater. 7 (22), 1700763.
PCE (%)
12 9 6 r-GO CuSCN r-GO/CuSCN
3 0
0
20
40
60
80
100
Time (h) Fig. 7. Light soaking stability of the unsealed PSCs fabricated by r-GO, CuSCN and r-GO/CuSCN HTLs.
structure with inorganic CuSCN HTL for low temperature processed rGO based inverted planar PSCs. The r-GO/CuSCN bilayer HTL provided a platform for the growth of perovskite crystals with larger grain sizes resulting in higher PCE compared to the individual r-GO based HTLs. The bilayer structure reduced the energy-level gradient and successfully supressed charge recombination at HTL/perovskite interface. The PCE of the r-GO/CuSCN based hole transport bilayer was 14.28% with improved light soaking stability up to 100 h. However, further optimization must be carried out to increase the PCE of the r-GO/CuSCN bilayer structured inverted planar PSCs. This work surely highlights the viability of bilayer hole transport layers with inorganic compounds for low temperature processed PSCs with a standard surface area and facilitates the up-scaling of PSCs for roll-to-roll production one step closer. Acknowledgements M.A Acknowledges the financial support from the FRGS/1/2017/ TK07/UKM/02/9 grant. A.I acknowledges the support from JSPS KAKENHI with grant No. 18H02079. J-J.L. acknowledges the support from NRF-2016M1A2A2940912. References Acik, M., Darling, S.B., 2016. Graphene in perovskite solar cells: device design, characterization and implementation. J. Mater. Chem. A 4 (17), 6185–6235. Agresti, A., Pescetelli, S., Taheri, B., Del Rio Castillo, A.E., Cinà, L., Bonaccorso, F., Di Carlo, A., 2016. Graphene-perovskite solar cells exceed 18 % efficiency: a stability study. ChemSusChem 9 (18), 2609–2619. Chatterjee, S., Pal, A.J., 2016. Introducing Cu2O thin films as a hole-transport layer in efficient planar perovskite solar cell structures. J. Phys. Chem. C 120 (3), 1428–1437. Chen, W.-Y., Deng, L.-L., Dai, S.-M., Wang, X., Tian, C.-B., Zhan, X.-X., Xie, S.-Y., Huang, R.-B., Zheng, L.-S., 2015. Low-cost solution-processed copper iodide as an alternative
657