Defect Passivation Using Trichloromelamine for Highly Efficient and Stable Perovskite Solar Cells

Nonradiative recombination losses caused by defects in the perovskite layer seriously affects the efficiency and stability of perovskite solar cells (PSCs). Hence, defect passivation is an effective way to improve the performance of PSCs. In this work, trichloromelamine (TCM) was used as a defects passivator by adding it into the perovskite precursor solution. The experimental results show that the power conversion efficiency (PCE) of PSC increased from 18.87 to 20.15% after the addition of TCM. What’s more, the environmental stability of PSCs was also improved. The working mechanism of TCM was thoroughly investigated, which can be ascribed to the interaction between the –NH– group and uncoordinated lead ions in the perovskite. This work provides a promising strategy for achieving highly efficient and stable PSCs.


Introduction
Organic-inorganic hybrid perovskite has demonstrated its application potential in the photovoltaic field due to its unique properties, such as its strong light absorption capacity, long charge carrier diffusion length, high charge carrier mobility, and adjustable band gap [1][2][3][4][5]. The power conversion efficiency (PCE) of perovskite solar cells (PSC) has increased from 3.8% to 25.5%, which is comparable to the commercial Si-based solar cell [6]. Although great progress has been made with respect to the performance of PSC, there are still several challenges which remain. One of the key problems is defects in the perovskite films, such as uncoordinated sites (lead and halide ions) [7], lead clusters [8], and vacancy defects [9], which will lead to non-radiative recombination losses. The interaction of these defects with moisture and oxygen will also cause the degradation of the perovskite [10][11][12].
The use of additives for defect passivation is an effective strategy to improve the PCE and stability of PSCs [13,14]. Many types of additives, such as polymers [15], fullerene and its derivatives [16], inorganic acid [17], organic halide salt [18], metal halide salt [19], and nanoparticles [20] were applied to achieve grain boundary (GBs) passivation through morphology control or passivation of crystal defects [21][22][23][24]. Among them, selecting organic molecules with specific functional groups is very effective to improve the PCE and stability of PSCs. For example, the introduction of 3-aminopropyltrimethoxysilane (APMS) inhibited ion migration and passivated defects in perovskite layer because of the interaction between the amino-group and Pb or I ions, leading to the morphology improvement of perovskite film [25]. The PCE of PSC increased from 18.85 to 20.72% [25]. After adding maleimide undecanoic acid (11MA) to the perovskite precursor solution, the density of trap states in perovskite layer reduced due to the strong coordination interaction between 11MA and Pb 2+ , resulting in the increase of PCE from 18.24 to 23.34% [26]. The additive molecules with π-conjugated structure are more likely to gather at the perovskite GBs, inducing the interconnection of the perovskite grains and therefore the more stable PSC [27].
In this work, a low-price organic molecular, trichloromelamine (TCM) was used as a defects passivator in the MAPbI 3 light-absorbing layer by adding it into the perovskite precursor solution. The chlorine substituted amino groups with Lewis base characteristics in the molecular structure of TCM will simultaneously improve the morphology and passivate defects of the perovskite film. MAPbI 3 film with improved crystallinity and reduced defect density was obtained. With 0.05 wt% TCM, the PCE of the PSC increased from 18.87% of control device to 20.15%. In addition, after 48 h aging in the dark with 80% humidity at room temperature, the encapsulated device with TCM retained its initial PCE of 77.31%, while the value for the control device is 31.2%. The mechanism was thoroughly investigated, which can be ascribed to the interaction between the -NH-group in TCM and lead ions in MAPbI 3 .

Preparation of NiOx Nanoparticles
First, under magnetic stirring, 12.885 g of NiCl 2 ·6H 2 O was dissolved in 100 mL of deionized water. Then, 10 M NaOH solution was added dropwise to the solution until the pH value reached 10. Following, the obtained green solution was centrifuged. After being washed twice with deionized water, the resulting precipitate was dried at 80 • C overnight, and then annealed at 270 • C for 2 h.

Preparation for Characterization
Solutions for NMR test were prepared by adding 0.05 wt/% trichloromelamine into PbI 2 solution (1 mol/L) in deuterated DMSO.
Perovskite films for XRD, SEM, UV, PL, UPS and XPS tests were all prepared by spincoating perovskite precursor solution on ITO substrate with a concentration of 1 mol/L. It was spun-coated at 4000 rpm for 30 s. Then the films were thermal annealed on a hot stage at 100 • C for 10 min.
For FT-IR measurement, the above-mentioned perovskite films were scraped from ITO substrates and then blended with spectrum grade KBr. The mixed powder was pressed into pieces before using.

Device Fabrication
Before using, indium tin oxide (ITO) glass substrate (7 Ω, RS −1 ) was sequentially ultrasonically cleaned with detergent, deionized water, acetone, and absolute ethanol for 20 min each. Then, the substrate was cleaned with plasma for 4 min after being dried with a nitrogen stream. Follow that, NiOx nanoparticles dispersion (20 mg/mL in deionized water) was spin-coated at 4000 rpm for 30 s and then annealed at 135 • C for 10 min under atmospheric conditions. Then perovskite precursor solution was spin-coated at 4000 rpm for 30 s in a N 2 -filled glovebox to form perovskite layer, which was heated on a hot plate at 100 • C for 20 min. After that, PCBM film was deposited by spin-coating at 1200 rpm for 30 s. Finally, 5 nm-thick BCP and 100 nm-thick Ag were evaporated as the interface layer and the top metal electrode, under a pressure of 9 × 10 −5 Pa.

Characterization
Fourier transform infrared (FT-IR) spectroscopy measurements were conducted on a Fourier transform infrared spectrometer (model: IRPrestige-21, range 4000-1000 cm −1 ). Ultraviolet-visible (UV-vis) absorption measurements were measured on a Lamba 35 spectrophotometer (Perkin-Elmer, Waltham, MA, USA). The X-ray diffraction (XRD) patterns of the films were obtained by a Bruker D8 ADVANCE X-ray diffractometer (Bruker Corp, Berlin, Germany) under the operation conditions of 40 kV and 40 mA. The morphology of perovskite films was obtained by field emission scanning electron microscopy (FESEM, S4800 microscope, Hitachi Ltd., Tokyo, Japan). The transient-state photoluminescence (PL) was measure by FLSP920 spectrometer (Edinburgh Instruments Ltd., Livingston, UK). H nuclear magnetic resonance (NMR) spectra were collected by using Bruker DELL PC1 equipment. X-ray photoelectron spectroscopy (XPS) was studied using a PHI Quantera SXM (ULVAC-PHI Inc., Tokyo, Japan). The current density-voltage (J-V) curves of the devices were measured by a Keithley 2400 Source Meter under an illumination of 1 sun (100 mW/cm 2 AM 1.5 G, generated by a solar simulator Oriel Sol3A, Newport Corp., Irvine, CA, USA), which was calibrated with a standard Si photodiode. The active area was 0.096 cm 2 .

Film Properties
The morphology of the MAPbI 3 film has great influence on the performance of PSCs. Thus, scanning electron microscopy (SEM) images of MAPbI 3 films were demonstrated, as shown in Figure 1. The crystal sizes were estimated by using Nano Measurer 1.2 software. For MAPbI 3 films doped with 0, 0.01, 0.05, and 0.1 wt% TCM, the average grain sizes are 246, 260, 308, and 221 nm, respectively ( Figure 1e). Compared with the pristine film, when the doping concentration of TCM increased from 0.01 wt% to 0.1 wt%, the grain size of MAPbI 3 increased first and then decreased. The MAPbI 3 film with 0.05 wt% TCM (hereinafter refer to as 0.05 TCM film) has the largest grain size. The effect of TCM on the crystallinity of MAPbI 3 was studied by X-ray diffraction (XRD) measurement ( Figure 1f). All the XRD patterns show significant peaks at 13.90 • and 28.17 • , which are corresponding to the (110) and (220) planes of MAPbI 3 , respectively. It indicates that all MAPbI 3 films have orthorhombic crystal structure. Meanwhile, the XRD peak intensity obviously varied after doping TCM, which increased first and then decreased with the concentration increased from 0.01 wt% and 0.05 wt% to 0.1 wt%. The perovskite film with 0.05 wt% TCM has the strongest peak intensity, indicating the best crystallinity. Cross-sectional SEM images of PSCs further proves the crystallinity improvement after adding TCM, as shown in Figure S1. The average thickness of the pristine perovskite layer is 243 nm. It is 260 nm with 0.05 wt% TCM, indicating the larger grain size.

Charge Carrier Dynamic
To investigate the optical properties of MAPbI 3 films, UV-Vis absorption spectra were measured, as shown in Figure 2a. The same shape of UV-Vis absorption spectra reflects that the crystal structure of MAPbI 3 film did not change, which is consistent with the XRD results. The absorbance intensity slightly increased at the wavelength lower than 500 nm after adding 0.05 wt% TCM, which can be ascribed to the larger grain size. In addition, for semiconductors, the absorbing edge is called the Urbach tail, which is related to Urbach energy (E u ). Generally, the Urbach energy is the tail width of the local defect state in the band gap, which can be calculated by fitting the exponential part of the Urbach tail according to Equation (1) [28]: where α is the absorption coefficient and E is the photon energy [29,30]. E u value can be calculated by plotting Ln(α) against E, as shown in Figure 2b. The calculated E u values are 50.9 meV and 39.96 meV for the pristine and 0.05 TCM MAPbI 3 films, respectively. The lower E u value indicated the reduced defect density. It means that the adding of 0.05 wt% TCM reduced the defect density of MAPbI 3 film. It was further proved by measuring the steady-state photoluminescence (PL) spectra, as shown in Figure 2c. The PL intensity of MAPbI 3 film is significantly enhanced with the addition of 0.05 wt% TCM, indicating the reduction of nonradiative recombination losses (which is always caused by defects). In addition, after the addition of TCM, the PL peak blue-shifted from 796.5 nm to 795.9 nm, which can be ascribed to the reduction of surface defects [31]. Therefore, the surface defects of MAPbI 3 film were passivated by adding TCM, which can also be proved by the Ultraviolet Photoelectron Spectroscopy (UPS) data, as shown in Figure 2d. The Fermi energy level (E F ) with respect to the valance band maximum (VBM) of perovskite films shift from 1.06 to 0.96 eV (Figure 2d) after the introduction of TCM, indicating that it is closer to the center of the bandgap (E g : 1.55 eV for MAPbI 3 ). Thus, the perovskite is more like the intrinsic semiconductor after the passivation of surface defects by TCM [32]. The above experimental results show that the defects of the perovskite film reduced after adding 0.05 wt% TCM.

Charge Carrier Dynamic
To investigate the optical properties of MAPbI3 films, UV-Vis absorption spectra were measured, as shown in Figure 2a. The same shape of UV-Vis absorption spectra reflects that the crystal structure of MAPbI3 film did not change, which is consistent with the XRD results. The absorbance intensity slightly increased at the wavelength lower than 500 nm after adding 0.05 wt% TCM, which can be ascribed to the larger grain size. In addition, for semiconductors, the absorbing edge is called the Urbach tail, which is related to Urbach energy (Eu). Generally, the Urbach energy is the tail width of the local defect intrinsic semiconductor after the passivation of surface defects by TCM [32]. The above experimental results show that the defects of the perovskite film reduced after adding 0.05 wt% TCM. In order to explore the work mechanism of defect passivation, 1 H nuclear magnetic resonance (NMR) and Fourier transform infrared spectroscopy (FT-IR) tests were conducted. Figure 3a shows the 1 H NMR spectra of pure TCM and the mixture of PbI2 and TCM. After adding PbI2, the chemical shift of the hydrogen peak corresponding to the -NH-group of TCM shifted from 7.81 to 7.71 ppm. It indicates that Pb 2+ interacted with the -NH-group in TCM. FT-IR was also used to detect the interaction between the -NHgroup of TCM and Pb 2+ (Figures 3b and S2). For pure TCM, the characteristic peak of the stretching vibration of the -NH-group located at 1660 cm −1 , which is 1651 cm −1 for the mixture of TCM and PbI2. The shift of the peak position indicates the interaction between the -NH-group and Pb 2+ . X-ray photoelectron spectroscopic (XPS) pattern (Figure 3c) shows that the Pb 4f7/2 and Pb 4f5/2 signals shift from 137.8 eV and 142.7 eV to 137.6 eV and 142.5 eV, respectively, after the addition of TCM. It demonstrated that the binding energy of the Pb 4f peak reduced by 0.2 eV, which indicates the decreased cationic charge of under-coordinated Pb 2+ [33].
To estimate the defect of density (Ndefects) of perovskite films, electron-only and holeonly devices with the structure of ITO/SnO2/MAPbI3/PCBM/Ag and ITO/NiOx/MAPbI3/PTAA/Ag were fabricated, respectively. The J-V curves are shown in Figure 4. The density of the defects can be calculated in the space-charge-limited current region by Equation (2): (2) In order to explore the work mechanism of defect passivation, 1 H nuclear magnetic resonance (NMR) and Fourier transform infrared spectroscopy (FT-IR) tests were conducted. Figure 3a shows the 1 H NMR spectra of pure TCM and the mixture of PbI 2 and TCM. After adding PbI 2 , the chemical shift of the hydrogen peak corresponding to the -NH-group of TCM shifted from 7.81 to 7.71 ppm. It indicates that Pb 2+ interacted with the -NH-group in TCM. FT-IR was also used to detect the interaction between the -NH-group of TCM and Pb 2+ (Figure 3b and Figure S2). For pure TCM, the characteristic peak of the stretching vibration of the -NH-group located at 1660 cm −1 , which is 1651 cm −1 for the mixture of TCM and PbI 2 . The shift of the peak position indicates the interaction between the -NH-group and Pb 2+ . X-ray photoelectron spectroscopic (XPS) pattern (Figure 3c) shows that the Pb 4f 7/2 and Pb 4f 5/2 signals shift from 137.8 eV and 142.7 eV to 137.6 eV and 142.5 eV, respectively, after the addition of TCM. It demonstrated that the binding energy of the Pb 4f peak reduced by 0.2 eV, which indicates the decreased cationic charge of under-coordinated Pb 2+ [33].
To estimate the defect of density (N defects ) of perovskite films, electron-only and holeonly devices with the structure of ITO/SnO 2 /MAPbI 3 /PCBM/Ag and ITO/NiOx/MAPbI 3 /PTAA/Ag were fabricated, respectively. The J-V curves are shown in Figure 4. The density of the defects can be calculated in the space-charge-limited current region by Equation (2): where ε is the relative dielectric constant [34], ε 0 is the vacuum permittivity, L is the thickness of the perovskite film, e is the unit charge, and V TFL is the trap-filled-limit (TFL) voltage. V TFL refers to the voltage at the kink point from linear region to the TFL region. The N defects and V TFL values are summarized in Table 1. It can be seen that the density of the electronic defect of the control device is about twice that of the device with 0.05 wt% TCM. The density of the hole defect states are only slightly reduced (from 2.8 × 10 16 to 2.2 × 10 16 cm −3 ). Combined with the results of UPS, this means that the addition of TCM effectively passivated the n-type undercoordinated Pb 2+ defects on the surface of the perovskite film. Therefore, after the addition of TCM, the surface of the perovskite film is more intrinsic. This is consistent with the passivation mechanism of TCM doping described above.
where ε is the relative dielectric constant [34], ε0 is the vacuum permittivity, L is the thickness of the perovskite film, e is the unit charge, and VTFL is the trap-filled-limit (TFL) voltage. VTFL refers to the voltage at the kink point from linear region to the TFL region. The Ndefects and VTFL values are summarized in Table 1. It can be seen that the density of the electronic defect of the control device is about twice that of the device with 0.05 wt% TCM. The density of the hole defect states are only slightly reduced (from 2.8 × 10 16 to 2.2 × 10 16 cm −3 ). Combined with the results of UPS, this means that the addition of TCM effectively passivated the n-type undercoordinated Pb 2+ defects on the surface of the perovskite film. Therefore, after the addition of TCM, the surface of the perovskite film is more intrinsic. This is consistent with the passivation mechanism of TCM doping described above.      Table 2. The experimental error values were obtained by subtracting the average value and then divided the results by two. The control device shows a short circuit current density (J SC ) of 22.63 ± 0.43 mA cm −2 , an open circuit voltage (V OC ) of 1.067 V and a fill factor (FF) of 78.14 ± 0.72%, resulting in a PCE of 18.87 ± 0.46%. With the addition of TCM, the performance of PSCs depended on the ratio of TCM. As the weight ratio of TCM increased (V OC ), FF and PCE both increased and reached 0.05 wt%. Then, when it continued to increase to 0.1%, J SC had a significant drop. Thus, 0.05 wt% is the optimized weight ratio of TCM, based on which PSC with a V OC of 1.085 V, a J SC of 22.68 ± 0.28 mA cm −2 , a FF of 81.81 ± 1.66%, and therefore a PCE of 20.15% ± 0.41% was obtained. Obviously, the increase of PCE is mainly caused by the enhancement of V OC and FF, which may benefit from the film quality improvement of the perovskite layer.

Device Characterization
The statistics of the photovoltaic performance parameters of PSC (Figure 5b) shows that the 0.05% TCM device not only has higher photovoltaic parameters than the control device but also shows that the distribution is more concentrated. It indicates that after adding TCM, the repeatability of the device is also improved. In addition, as shown in Figure S3, for the control device and device with 0.05 wt% TCM, the forward scanned J-V curve matched well with that from reverse scan, indicating a negligible hysteresis effect.   Figure 5a show the J-V curves of the PSCs, ITO/NiOx/MAPbI3/PCBM/Ag ( Figure S4) based on MAPbI3 with/without TCM. The corresponding performance values are summarized in Table 2. The experimental error values were obtained by subtracting the average value and then divided the results by two. The control device shows a short circuit current density (JSC) of 22.63 ± 0.43 mA cm −2 , an open circuit voltage (VOC) of 1.067 V and a fill factor (FF) of 78.14 ± 0.72%, resulting in a PCE of 18.87 ± 0.46%. With the addition of TCM, the performance of PSCs depended on the ratio of TCM. As the weight ratio of TCM increased (VOC), FF and PCE both increased and reached 0.05 wt%. Then, when it continued to increase to 0.1%, JSC had a significant drop. Thus, 0.05 wt% is the optimized weight ratio of TCM, based on which PSC with a VOC of 1.085 V, a JSC of 22.68 ± 0.28 mA cm −2 , a FF of 81.81 ± 1.66%, and therefore a PCE of 20.15% ± 0.41% was obtained. Obviously, the increase of PCE is mainly caused by the enhancement of VOC and FF, which may benefit from the film quality improvement of the perovskite layer.   The statistics of the photovoltaic performance parameters of PSC (Figure 5b) shows that the 0.05% TCM device not only has higher photovoltaic parameters than the control device but also shows that the distribution is more concentrated. It indicates that after adding TCM, the repeatability of the device is also improved. In addition, as shown in Figure S3, for the control device and device with 0.05 wt% TCM, the forward scanned J-V curve matched well with that from reverse scan, indicating a negligible hysteresis effect.

Device Characterization
The current density of the 0.05 wt% TCM PSC under dark decreases at low voltage, as shown in Figure 5c. It further proves that the defects density of perovskite film with 0.05 wt% TCM reduced [35], which is conducive to the transfer of charge. Therefore, the series resistance (Rs) of the 0.05 wt% TCM device decreased from 50.28 Ω·cm −2 of the control device to 29.57 Ω·cm −2 , as shown in Table 2. The reduction of Rs value is also responsible for the increase in the FF of PSC. The current density of the 0.05 wt% TCM PSC under dark decreases at low voltage, as shown in Figure 5c. It further proves that the defects density of perovskite film with 0.05 wt% TCM reduced [35], which is conducive to the transfer of charge. Therefore, the series resistance (Rs) of the 0.05 wt% TCM device decreased from 50.28 Ω·cm −2 of the control device to 29.57 Ω·cm −2 , as shown in Table 2. The reduction of Rs value is also responsible for the increase in the FF of PSC.
In order to further explore the electrical and optical properties of TCM doped perovskite films, the light intensity dependence of V OC had been investigated. It can provide more detailed information about the reorganization process under open circuit conditions according to the following Equation (3) [36,37] δV oc = n k B T e Ln(I) + constant where n is the ideality factor, k B is the Boltzmann constant, q is the elementary charge, T is the absolute temperature, and I is the light intensity. The 0.05 wt% TCM device exhibits a higher V OC than the control device under the same light intensity (Figure 5d). A plot of V OC as a function of logarithmic light intensity Ln(I) is linearly fitted to evaluate the slope, n (k B T/e), which represents the recombination process caused by the trap states in the optoelectronic device. The slope value of the 0.05 TCM device is 1.14 k B T/e, which is 1.704 k B T/e for the control device. The decrease in the slope value means that TCM doping can effectively reduce the trap-assisted carrier recombination, which is consistent with the above results.
In order to evaluate the environmental stability of PSCs, the PCE of PSCs were measured during storage. Figure 6a shows the normalized PCE against time. After being stored under 80% humidity in the dark at room temperature for 50 h, the PCE of the 0.05 wt% TCM device retained 77.3% of its initial value, while the value for the control device is 31.2%. Obviously, the 0.05 wt% TCM device is more stable than the control device. XRD was also used to detect the stability of perovskite film after being exposed to ambient air with a humidity of 80% in the dark for a week. As shown in Figure 6b, the peak at 12.8 • is observed for both film, which is corresponding to the (100) plane of PbI 2 . Meanwhile, in the XRD pattern of the newly fabricated MAPbI 3 film (Figure 1f), there was no peak at 12.8 • . The precipitation of lead iodide after storage indicates the decomposition of MAPbI 3 . Meanwhile, the peak intensity at 12.8 • for the 0.05 TCM MAPbI 3 film is much lower than that of the pristine film, indicating less MAPbI 3 decomposed. It means that TCM additive effectively protected the perovskite film from moisture. The contact angle of water on the perovskite film was measured, as shown in Figure 6c-d. It shows that the 0.05 TCM MAPbI 3 film is more hydrophobic than the pristine film (54.2 • vs. 40.9 • ), which is also responsible for the improved environment stability of the perovskite film and PSCs.

Discussion
In conclusion, TCM was used as an additive in the MAPbI3 film by doping it in the perovskite precursor solution. The experimental results show that TCM passivated the defects at the surface of the MAPbI3 film, resulting in the PCE of PSC increasing from 18.87% to 20.15% due to the enhancement of VOC and FF. With TCM, both the PSC and MAPbI3 film show improved environmental stability. Passivation is caused by the interaction between the uncoordinated lead ion and the -NH-group in TCM. This work provides an effective method to improve the efficiency and stability of PSCs.
Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1. Figure S1: Cross-sectional SEM images of perovskite devices: (a) control and (b) 0.05 TCM additive; Figure S2: FT-IR spectrum of pure TCM and PbI2 mixed with TCM; Figure S3: J-V curve of forward and reverse voltage scanning of the best control device and 0.05 TCM device; Figure S4

Discussion
In conclusion, TCM was used as an additive in the MAPbI 3 film by doping it in the perovskite precursor solution. The experimental results show that TCM passivated the defects at the surface of the MAPbI 3 film, resulting in the PCE of PSC increasing from 18.87% to 20.15% due to the enhancement of V OC and FF. With TCM, both the PSC and MAPbI 3 film show improved environmental stability. Passivation is caused by the interaction between the uncoordinated lead ion and the -NH-group in TCM. This work provides an effective method to improve the efficiency and stability of PSCs.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10.3390/ polym14030398/s1. Figure S1: Cross-sectional SEM images of perovskite devices: (a) control and (b) 0.05 TCM additive; Figure S2: FT-IR spectrum of pure TCM and PbI 2 mixed with TCM; Figure S3: J-V curve of forward and reverse voltage scanning of the best control device and 0.05 TCM device; Figure S4: Device configuration of inverted perovskite solar cells.