Modification of NiOx hole transport layers with 4-bromobenzylphosphonic acid and its influence on the performance of lead halide perovskite solar cells
- 128 Downloads
Lead halide perovskites have proved to be exceptionally efficient absorber materials for photovoltaics. Besides improving the properties of the perovskite absorbers, device engineering and the optimization of interfaces will be equally important to further the advancement of this emerging solar cell technology. Herein, we report a successful modification of the interface between the NiOx hole transport layer and the perovskite absorber layer using 4-bromobenzylphosphonic acid based self-assembled monolayers leading to an improved photovoltaic performance. The modification of the NiOx layer is carried out by dip coating which allows sufficient time for the self-assembly. The change in the surface free energy and the non-polar nature of the resulting surface is corroborated by contact angle measurements. X-ray photoelectron spectroscopy confirms the presence of phosphor and bromine on the NiOx surface. Furthermore, the resultant solar cells reveal increased photovoltage. For typical devices without and with modification, the photovoltage improves from 0.978 V to 1.029 V. The champion VOC observed was 1.099 V. The increment in photovoltage leads to improved power conversion efficiencies for the modified cells. The maximum power point tracking measurements of the devices show stable power output of the solar cells.
Lead halide perovskite solar cells have reached certified power conversion efficiencies (PCEs) greater than 23% in a conventional n-i-p architecture [1, 2, 3]. Similarly, also with perovskite solar cells based on a planar inverted p-i-n architecture efficiencies greater than 20% have imposingly been realized [4, 5]. In addition, devices in inverted p-i-n architecture have piqued considerable interest as they offer the possibility to be processed at low temperature and also on flexible substrates [5, 6, 7]. The diminishing gap between the performance of both device architectures has been majorly possible due to optimizing the inorganic hole transport layers (HTLs). In particular, nickel oxide (NiOx) films proved to be promising, as they allow facile processing, good stability and efficient hole extraction [8, 9]. Furthermore, different doping procedures for NiOx such as incorporation of copper, cesium or silver ions have led to improved device performance and are opening up pathways for further optimisations [10, 11, 12].
In recent practice, the NiOx layers for the application in perovskite solar cells are deposited majorly via solution processes by using sol–gel methods or nanoparticle dispersions in water [10, 13, 14, 15, 16, 17, 18, 19]. Moreover, atomic layer deposition, flame spray synthesis, pulsed laser deposition, sputtering and electrodeposition have been applied for the formation of NiOx thin films [20, 21, 22, 23, 24, 25].
In photovoltaics, interface engineering between charge transport layers and the absorber layer is an important approach in device optimization. One of the primary concepts utilized to achieve such engineering is the application of self-assembled monolayers (SAMs) [26, 27, 28]. The formation of a monolayer and the effective dipole of the applied molecules can be explored for designing the properties at the aforementioned interface. The SAMs allow the manipulation of the work function and the wettability of the metal oxide films, and they can also passivate the interface [29, 30]. For example, Wang et al. reported on the effects of para-substituted benzoic acid SAMs on the NiOx and perovskite interface. They highlighted an improved open circuit voltage (VOC) due to the introduction of bromobenzoic acid (dipole moment of 2.1 D) as a monolayer on the NiOx film .
In this work, we discuss the influence of the modification of the NiOx HTL with 4-bromobenzylphosphonic acid (Br-BPA) on lead halide perovskite solar cells. We chose phosphonic acids as the interface modifier as they have so far been barely investigated in perovskite solar cells even though they have been thoroughly studied in organic field effect transistors as well as organic photovoltaics and possess beneficial properties for surface modification of metal oxides . In a study on the modification of zinc oxide (ZnO) with benzylphosphonic acids (BPA), Lange et al. reported that BPA SAMs have a preferential tridentate binding on a ZnO surface. A tridentate binding suggests a stronger chemisorption of the SAM molecules on the metal oxide surface in comparison to the mono- and bi-dentate binding [32, 33, 34, 35]. As absorber layer for the solar cells prepared in p-i-n architecture, a triple cation based perovskite with the composition Cs0.1(FA0.83MA0.17)0.9Pb(I0.83Br0.17)3, introduced by Saliba et al. , was selected.
2.1 Sample and solar cell preparation
All chemicals and solvents were used as purchased without any further purification. Nickel(II) nitrate hexahydrate and sodium hydroxide was purchased from Fluka and VWR, respectively. Lead iodide, lead bromide, formamidinium iodide and PC60BM were purchased from TCI, Alfa Aesar, GreatCell Solar and Solenne, respectively. The other chemicals used in this study, including the phosphonic acid molecule, and all the solvents were purchased from Merck (Sigma Aldrich).
2.2 Material synthesis
2.2.1 Synthesis and characterization of nickel oxide nanoparticles
The synthesis of the NiOx nanoparticles was performed according to previous reports [15, 37, 38, 39]. In brief, nickel (II) nitrate hexahydrate (NiNO3 · 6 H2O) (0.05 mol) was dispersed in 10 mL deionized water and stirred for 5 min. Afterwards sodium hydroxide (NaOH, 10 mmol/mL) was added dropwise to adjust a pH of 10, which results in a colour change from dark to light green. The colloidal precipitate was then washed with deionized water to remove side products. The light green residue was further dried at 80 °C for 6 h followed by calcination at 270 °C for 2 h. This resulted in non-stoichiometric black nickel (II) oxide nanoparticles. The X-ray diffraction (XRD) pattern of the NiOx nanoparticles is very similar to as reported in Ref.  and reveals reflections at 37.2°, 43.2°, 62.7°, 75.4° and 79.3° 2θ, which correspond to the (111), (200), (220), (311), and (222) lattice planes in a cubic crystal structure. An estimation of the primary crystallite size based on the peak broadening via Scherrer formula led to a value of approx. 8 nm. For the preparation of the NiOx ink, 20 mg/mL of the NiOx nanoparticles were dispersed in deionized water. The ink was put in an ultrasonic bath for 2 h and then filtered using a 0.45 µm PVDF syringe filter.
2.3 Solar cell preparation
2.3.1 Glass/ITO substrates
Pre-patterned glass/ITO substrates (15 × 15 × 1.1 mm3) (15 Ω/sq) from Luminescence Technology Corp. (Lumtec) were carefully wiped using acetone before putting them into an isopropanol bath. The bath was further subjected to an ultrasonic bath for 10 min. The substrates were then dried using in an N2 stream. Just before spin coating the NiOx nanoparticle ink, the substrates were plasma etched using oxygen plasma for 3 min.
2.3.2 Preparation of the NiOx films
The NiOx nanoparticle ink was spin coated onto the glass/ITO substrates at a speed of 1000 rpm. Before the determination of the layer thickness and solar cell preparation, the films were allowed to dry in ambient conditions for 3 days. The as-prepared films revealed an average thickness distribution of ~ 25 nm.
2.3.3 Surface modification of the NiOx films with Br-BPA
A 5 mmol/mL solution of Br-BPA was prepared in acetonitrile and was filtered using a 0.45 µm PVDF syringe filter. The application of the Br-BPA molecule was done by dip coating. In the dip coating method, the glass/ITO/NiOx substrates were dipped in the phosphonic acid solution for 1 min to allow adsorption. These substrates were then cleaned by dipping them in fresh acetonitrile solvent to allow removal of non-adsorbed phosphonic acid molecules from the modified surface. The substrates were then dried in an N2 stream. The deposition was carried out in ambient conditions.
2.3.4 Preparation of the perovskite absorber layers
The cesium, formamidinium, and methylammonium based triple cation lead halide perovskite absorber layer was adapted from Saliba et al. . The final composition chosen was Cs0.1(FA0.83MA0.17)0.9Pb(I0.83Br0.17)3. The precursor solution consisted of 1 mmol/mL formamidinium iodide (FAI), 1.1 mmol/mL PbI2, 0.2 mmol/mL methylammonium bromide (MABr), and 0.2 mmol/mL PbBr2 in a mixed dimethyl formamide (DMF)/dimethyl sulfoxide (DMSO) solvent with a 4:1 volume ratio. To this solution, 1.5 mmol/mL CsI in DMSO, was added to obtain a 10% Cs content. The final precursor solution was stirred overnight in inert conditions to allow sufficient reaction time. The solution was then filtered using a 0.45 µm PTFE syringe filter prior to spin coating. The perovskite absorber layer was spin coated on the non-modified and modified glass/ITO/NiOx substrates in a two-step spin coating process with 1000/6000 rpm for 10/20 s. In the last 5 s of spinning, chlorobenzene was dripped onto the substrate as an antisolvent. The substrates were annealed at 100 °C for 1 h. The thickness of the perovskite absorber layers was monitored using a Stylus profilometer and was found to be approximately 485 nm.
2.3.5 PC60BM electron transport layer and top electrode
A phenyl-C61-butyric acid methyl ester (PC60BM) solution in chlorobenzene having a concentration of 20 mg/mL was prepared. The solution was stirred overnight and filtered using a 0.45 µm PTFE syringe filter prior to spin coating at 4000 rpm for 20 s. In a last step of the solar cell preparation, a 120 nm thick silver layer was deposited by thermal evaporation on top of the PC60BM layer at an evaporation rate of 1–2 Å s−1 using a shadow mask (0.09 cm2).
2.4 Characterization techniques
XRD was performed on a PANalytical Empyrean system using Cu Kα radiation. Ultraviolet–visible (UV–Vis) spectroscopy measurements were done using the UV–Vis Spectrometer—Lambda 35 by Perkin Elmer. The layer thicknesses were measured by surface profilometry using a DektakXT device by Bruker and the surface morphology of the perovskite films was characterized by scanning electron microscopy (SEM) images acquired on a Zeiss-Supra 40 scanning electron microscope with an in-lens detector and 5 kV acceleration voltage.
Contact angle measurements of the NiOx films before and after the modification with Br-BPA were carried out on a Krüss DSA100 system using water and ethylene glycol as liquids. The surface free energy calculations were performed with the Owens-Wendt-Rabel & Kaelble method  using the Krüss Advance software.
X-ray photoelectron spectroscopy (XPS) measurements of the NiOx and NiOx/Br-BPA SAM films were recorded using a multiprobe surface analysis system (Omicron Nanotechnology) equipped with a DAR 400 X-ray source (Al Kα1 radiation, 1486.7 eV), an XM 500 quartz crystal monochromator (energy width: 0.15 eV), and an EA 125 hemispherical electron energy analyzer based on a 5-channel pulse counting channeltron.
The current density–voltage (JV) curves and maximum power point (MPP) tracking measurements of the solar cells were performed using a Keithley 2400 source meter and a LabView-based software inside a glove box (nitrogen atmosphere). For the JV curves, the scan rates were adjusted to 100 mVs−1 in the forward (fwd) direction (− 0.02 V to 1.2 V) and backward (bwd) direction (1.2 V to − 0.02 V) for both light and dark measurements. The illumination area was defined using a shadow mask (0.07 cm2) and the light was provided by a Dedolight DLH500 lamp calibrated to an intensity of 100 mWcm−2 using a pyranometer from Kipp & Zonen. The External Quantum Efficiency (EQE) spectra were acquired using a MuLTImode 4 monochromator (Amko) equipped with a 75 W xenon lamp (LPS 210-U, Amko), a lock-in amplifier (Stanford Research Systems, Model SR830), and a Keithley 2400 source meter. The monochromatic light was chopped at a frequency of 30 Hz and the measurement setup was spectrally calibrated with a silicon photodiode (Newport Corporation, 818-UV/DB).
3 Results and discussion
Results of the contact angle measurements of the non-modified and modified NiOx layers
Contact angle—water (°)
Contact angle–ethylene glycol (°)
Surface free energy (mN m−1)
Disperse part (mN m−1)
Polar part (mN m−1)
21.9 ± 0.3
15.8 ± 2.1
88.4 ± 0.4
0.47 ± 0.03
88.0 ± 0.4
93.1 ± 0.3
24.3 ± 0.4
126.5 ± 2.0
120.9 ± 1.6
5.6 ± 0.4
The corresponding high-resolution XPS core-level spectra are given in the Fig. 3d and e. Peak fits were performed by using a convolution of a Gaussian and a Lorentzian profile.
The inelastic mean free path of the photoelectrons (λ) is dependent on the kinetic energy of the photoelectrons and the layer material. The values of λ were determined from the Tanuma, Powell, and Penn TPP2 M formula  and the software QUASES written by Sven Tougaard. θ is the detection angle relative to the surface normal. For a given layer system, in our case ITO/NiOx/Br-BPA, the expected total photoelectron intensity of each element can be calculated by the signal sum over all atomic layers including the atomic density of the element in the layer. The layer model was optimized for the best matching of the calculated contribution of each element with the element quantification of the measured spectra at all detection angles. We found a good agreement between the measured and calculated data by assumption of a NiOx nanoparticle layer with an average thickness of around 5 nm on the ITO and a Br-BPA SAM-layer with a density of 4 molecules per nm2. This matches well with previously reported surface coverage densities of phosphonic acid molecules on metal oxides [35, 43]. Previous studies on NiOx HTLs for perovskite solar cells revealed that NiOx film thicknesses of around 25 nm are beneficial for the solar cell performance. Therefore, we adjusted the thickness of the NiOx films used in the solar cells in the further course of the study by optimizing the spin coating parameters [10, 11, 15, 31].
The absorption spectra of the perovskite films suggest a similar absorption onset (~ 770 nm) in agreement with a bandgap of ~ 1.61 eV [36, 45]. The slight differences in the intensity of the absorption spectrum stem from minor changes in the film thickness. The surface profile measurements revealed film thicknesses of the perovskite layers between 485 and 510 nm independent of the surface modification.
Characteristic parameters of typical perovskite solar cells prepared with non-modified and modified NiOx hole transport layers
JSC (mA cm−2)
The MPP tracking measurements (Fig. 6b) reveal a constant power output of the solar cells. While the voltage at the MPP (VMPP) is slightly decreasing at the beginning of the measurement, the current density at the MPP (JMPP) improves slightly leading to a constant power output and a PCE of 11.4% and 12.2% for the non-modified and the modified device after 10 min of continuous illumination, respectively.
The EQE spectra (Fig. 6c) for the representative devices show a typical shape of perovskite solar cells with an onset at 770 nm and a characteristic plateau at wavelengths below 750 nm. Moreover, the spectra remain almost unchanged upon the modification of the NiOx HTL. This is expected, as the JSC for the representative devices are rather similar. The integrated JSC calculated from the EQE spectrum sums up to 18.2 and 18.4 mA/cm2 and is within a few percent deviation to the values extracted from the JV curves.
In summary, we successfully functionalized solution processed NiOx HTLs by a dip coating procedure using a Br-BPA solution in acetonitrile. This is substantiated by the markedly increased contact angle of water, the surface free energies and the detection of P 2p and Br 3p peaks in the modified sample by XPS measurements. We did not observe any notable change in the optical properties and surface morphology of the perovskite layers. However, the JV curves reflect improved photovoltaic performance, particularly an increased VOC. A typical device shows an improved PCE from 11.2% (10.8%) to 12.5% (12.7%) due to an improved VOC from 0.978 V (0.978 V) to 1.019 V (1.029 V) in fwd (bwd) scan directions, respectively. We assume that the improvement in the VOC is largely due to the realignment of the energy levels based on the dipole moment of the Br-BPA SAM molecules. With devices having Br-BPA modifications, VOCs of up to 1.099 V could be obtained. Furthermore, MPP tracking measurements revealed a steady state PCE of 11.35% and 12.22% for typical devices without and with Br-BPA modified NiOx films.
Open access funding provided by Graz University of Technology. This work was carried out within the project “flex!PV_2.0” (FFG No. 853603) funded by the Austrian Climate and Energy Fund within the program Energy Emission Austria. Birgit Ehmann is gratefully acknowledged for experimental support.
- 15.S. Weber, T. Rath, J. Mangalam, B. Kunert, A.M. Coclite, M. Bauch, T. Dimopoulos, G. Trimmel, J. Mater. Sci.: Mater. Electron. 29, 1847–1855 (2018)Google Scholar
- 18.Z. Zhu, Y. Bai, T. Zhang, Z. Liu, X. Long, Z. Wei, Z. Wang, L. Zhang, J. Wang, F. Yan, S. Yang, Angew. Chem. Int. Ed. 53, 12571–12575 (2014)Google Scholar
- 24.Y. Hou, W. Chen, D. Baran, T. Stubhan, N.A. Luechinger, B. Hartmeier, M. Richter, J. Min, S. Chen, C.O.R. Quiroz, N. Li, H. Zhang, T. Heumueller, G.J. Matt, A. Osvet, K. Forberich, Z.-G. Zhang, Y. Li, B. Winter, P. Schweizer, E. Spiecker, C.J. Brabec, Adv. Mater. 28, 5112–5120 (2016)CrossRefGoogle Scholar
- 41.D. Briggs, M.P. Seah (eds.), Practical Surface Analysis (Second Edition) Volume 1: Auger and X-ray Photoelectron Spectroscopy (Wiley, Hoboken, 1990), p. 183Google Scholar
- 51.I.M. Dharmadasa, Y. Rahaq, A.A. Ojo, T.I. Alanazi, J. Mater. Sci.: Mater. Electron. 30, 1227–1235 (2019)Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.