Materials
Iron(Ⅲ) chloride hexahydrate (FeCl3∙6H2O, 99%), urea (CH4N2O, 99.5%), gelatin (from bovine skin, Type B), sodium dihydrogen phosphate dehydrate (NaH2PO4∙2H2O, 99%), and sodium hydroxide (NaOH, 98%)were purchased from Sigma Aldrich. All chemical reagents were used as purchased and without further purification.
Hydrochloric acid (HCl, 35%), ethyl alcohol (EtOH, 99.5%), and acetone (99.5%) were purchased from DaeJung Chemicals & Metals (Republic of Korea, ROK). Tin (IV) oxide (SnO2 15% in H2O colloidal dispersion) and potassium chloride (KCl, 99.995%) were purchased from Alfa Aesar (United States of America, USA). Lead iodide (PbI2, 99.99%), and lead bromide (PbBr2, 99.99%) were purchased from Tokyo Chemical Industry Chemicals (Japan). Formamidinium iodide (FAI, 99.9%), methylammonium bromide (MABr, 99.9%), and methylammonium chloride (MACl, 98%) were purchased from GreatCell Solar (Australia). Hellmanex Ⅲ (detergent), deionized water (DI-water), cesium iodide (CsI, 99.9%), N,N-dimethylformamide (DMF, anhydrous, 99.8%), dimethyl sulfoxide (DMSO, anhydrous, 99.9%), anisole (anhydrous, 99.7%), oleylamine (OLA, technical grade 70%), toluene (anhydrous, 99.8%), chlorobenzene (CB, anhydrous, 99.8%), 4-tert‐butyl pyridine (tBP, 98%), acetonitrile (anhydrous, 99.8%), bis(trifluoromethane)sulfonimide lithium salt (Li-TFSI, 99.95%), iron(Ⅲ) chloride hexahydrate (FeCl3∙6H2O, 99%), urea (CH4N2O, 99.5%), gelatin (from bovine skin, Type B), sodium dihydrogen phosphate dehydrate (NaH2PO4∙2H2O, 99%), and sodium hydroxide (NaOH, 98%) were purchased from Sigma-Aldrich (USA). Spiro-OMeTAD (2,2′,7,7′-tetrakis[N,N-di(4-methoxyphenyl)-amino]-9,9′-spirobifluorene, 99.5%) was purchased from Lumtec. All chemicals were used without any further purification.
Method
Synthesis of hematite nanoparticles
The HT NPs were synthesized by simple precipitation.29,30 Typically, 10 mmol dihydrogen phosphate dehydrate, 100 mmol urea, and 0.5 g of gelatin were dissolved in 250 mL deionized (DI) water, and mixed with 250 mL aqueous solution containing 10 mmol iron(Ⅲ) chloride hexahydrate, sequentially. Then, the 1M NaOH solution was rapidly added to adjust pH as 12, resulting reddish precipitation, and the obtained solution kept at 95oC for 24 h in an oven. Finally, the precipitate was collected by centrifugation and washed with DI water several times, followed by dry in freeze-drying apparatus.
One-step Dissolution Process
Specific number of PSCs (a substrate size of 7 x 7 cm2) were immersed in DMF solvent bath of 50 mL. The recovered DMF solvent after dissolving the PSCs were filtered or not filtered by hematite nanoparticles which are to adsorb heavy metal ions. The filtered (denoted as one-step DMF) and unfiltered DMF solvents were used for each perovskite precursor solution. Insoluble substances of PSCs such as metal electrode and substrates in the DMF bath were collected and cleaned to recycle.
Two-step Dissolution Process
1st step: Specific number of PSCs were immersed in CB solvent bath of 50 mL. The recovered CB solvent after dissolving the PSCs was used in Sp precursor solutions without any purification. Also, the selectively dissolved PSCs by CB were dried in a vacuum chamber. The insoluble metal electrodes of PSCs in the CB bath were collected and cleaned to recycle.
2nd step: The dried PSCs were immersed in DMF solvent bath of 50 mL. (Note: EtOH rinsing of the dried PSCs was conducted to obtain residual organic materials-free DMF solvent before immersing in DMF bath for the best performing devices.) Then, the recovered DMF solvent from 2nd step was filtered or not filtered by the hematite nanoparticles. The filtered (denoted as two-step DMF) and unfiltered DMF solvents were used for each perovskite precursor solution. The insoluble substrates of PSCs in the DMF bath were collected and cleaned to recycle.
Preparation Precursor Solutions
SnO2 solution for electron transport layer: Commercial SnO2 solution was diluted in DI-water with a 1:3 volume ratio and added KCl of 10 mM. After mixing 30 min, it was filtered by polyvinylidene fluoride filter with a pore size of 0.45 µm.
Perovskite precursor solutions for light-absorbing layer: Total 26 kinds of perovskite precursor solutions were prepared in this work. CsI of 0.09 M, MABr of 0.09 M, PbBr2 of 0.09 M, FAI of 1.62 M, PbI2 of 1.8 M, and additive MACl of 0.54 M were dissolved in an experimental DMF and a fresh DMSO solvents with a 7:3 volume ratio. The 26 experimental DMF solvents are follows; A fresh DMF solvent (control), 7 recovered DMF solvents from the one-step dissolution processes which each for 1, 5, 10, 20, 30, 40, and 50 PSCs were dissolved, 5 DMF solvents (the recovered DMF solvents which each dissolved 1, 10, 20, 40, and 50 PSCs) filtered by the hematite nanoparticles, 5 FA-rich DMF solvents with specific concentrations of FA (corresponding to the theoretical concentrations according to each dissolved 1, 5, 10, 20, and 40 PSCs in DMF), 5 spiro-OMeTAD-rich DMF solvents with specific concentrations of Sp (corresponding to the theoretical concentrations according to each dissolved 1, 5, 10, 20, and 40 PSCs in DMF), a recovered DMF solvent from the two-step dissolution process which 20 PSCs were dissolved, a DMF solvent from the recovered DMF solvent in the two-step dissolution process filtered by the hematite nanoparticles, and a residual organic and heavy metal materials-free DMF solvent from the two-step dissolution process with EtOH rinsing and hematite nanoparticles filtering. All perovskite precursor solution was mixed for 30 min and filtered by a polytetrafluoroethylene (PTFE) filter with a pore size of 0.20 µm just before use.
Spiro-OMeTAD precursor solutions for hole transport layer: 4 kinds of Sp precursor solutions were prepared in this work. Sp of 72.1 mg, tBP of 28.8 µL, and Li-TFSI solution (Li-TSFI in acetonitrile, 720 mg/mL) of 17.6 µL were mixed into an experimental CB of 1 mL. The 4 experimental CB solvents are follows; A fresh CB solvent (control), and 3 recovered CB solvents from the two-step dissolution processes which each for 10, 20, and 40 PSCs were dissolved.
Perovskite Solar Cells (PSCs) Fabrication
Patterned FTO glass substrates (Asahi VU glass) with a size of 2 x 2 cm2 were sequentially washed with 2% Hellmanex aqueous solution, deionized (DI) water, acetone, and ethanol each for 10 min in an ultrasonic bath. After drying of the washed substrates by air-blowing, those were treated by an ultraviolet (UV)-ozone cleaner for 30 min. Then, the prepared SnO2 solution was spin-coated on the substrates, and the SnO2-coated substrates were annealed on a hot-plate at 100℃ for 10 min. Above processes were conducted at 20–25℃, with a relative humidity of 40–60%. Subsequent processes were conducted in a dry-ambient air condition (at 20℃, with a relative humidity of 0.5-3%). A prepared perovskite precursor solution was spread on the glass/FTO/SnO2 substrate, then spin-coated at 5000 rpm for 25 s. After 18 s from start of the spin-coating, anisole of 0.55 mL was poured on the rotating substrate in 2 s. Subsequently, the glass/FTO/SnO2/pre-perovskite film substrate was annealed on a hot-plate at 150℃ for 15 min. After cooling down of the annealed glass/FTO/SnO2/perovskite substrate to 20℃ naturally, a prepared spiro-OMeTAD precursor solution was spin-coated on the cooled substrate at 3000 rpm for 30 s. (Note, for best performing devices, vacuum-assisted post-treatment was applied on the perovskite film before the spin-coating of spiro-OMeTAD precursor solution. Briefly, OLA in toluene solution (8 mg/mL) was spin-coated on the cooled perovskite film at 2000 rpm for 30 s, then the OLA-coated substrate was placed in a vacuum chamber for 5 min. Details are in our previous work34.). The Au electrode was deposited with thickness over 50 nm on the glass/FTO/SnO2/perovskite/ spiro-OMeTAD film using a thermal evaporator. Finally, an anti-reflection film (MgF2) was attached on glass side of the completed PSCs before PV performance measurement.
Characterization
TEM (JEOL, JEM ARM 200F) at 200kV and FE-SEM (JEOL, JSM7600F). The XRD patterns of all the samples were obtained using a D8 Discover instrument (Bruker) with CuKα radiation (1.5405 Å) at room temperature and H1 NMR (500MHz) analysis was carried out using Unity Inova (Varian Technology). The heavy metal and anion concentration was analyzed by inductively coupled plasma atomic emission spectrometry (ICP-AES, Perkin-Elmer, OPTIMA 8300) and ion chromatography (IC, 882 Compact IC Plus, Metrohm), respectively.
Adsorption Experiments
The adsorption experiments were conducted to investigate the adsorption performance hematite nanoparticles under DMF solvent. The heavy metal ion concentration before and after adsorption was determined by inductively coupled plasma atomic emission spectrometry (ICP-AES, Perkin-Elmer, OPTIMA 8300). The adsorption capacities (the amount of Pb ions adsorbed per specific amount of adsorbents, qt) was calculated using the following Eq. (1):
(1)\({q}_{t}=\frac{\left({C}_{0}-{C}_{t}\right)V}{m}\)
Where C0 is the initial Pb concentration, Ct is remaining Pb concentration, V is the volume of Pb solution, and m is the mass of hematite nanoparticles.
For adsorption isotherm, as-obtained data according to initial Pb concentration was fitted using two different isotherm models Langmuir and Freundlich by the following equation:
(2)\({q}_{e}=\frac{{q}_{m}{K}_{L}{C}_{e}}{1+{K}_{L}{C}_{e}}\)
Where qe is the amount of adsorbed Pb per specific amount of adsorbent, qm is the maximum amount of Pb ions required to form a monolayer, Ce is the equilibrium concentration of Pb solution, KL is the Langmuir constant. Among these, the value of qm and KL can be obtained from the linear plot of Ce/qe versus Ce using following equation:
(3)\(\frac{{C}_{e}}{{q}_{e}}=\frac{1}{{K}_{L}{q}_{e}}+\frac{1}{{q}_{m}}{C}_{e}\)
The Freundlich isotherm is represented by the following equation:
(4)\({q}_{e}={K}_{F}{C}_{e}^{\frac{1}{n}}\)
Where qe is the amount of adsorbed Pb at equilibrium, KF is the Freundlich constant or an approximate indicator of the adsorption capacity, Ce is the final Pb concentration after adsorption, 1/n is a function of the strength of the adsorption (n is the adsorption intensity). The KF and 1/n can be evaluated from following equation of the linear plot:
(5)\({\text{log}q}_{e}={\text{log}K}_{F}+\frac{1}{n}{\text{log}C}_{e}\)
For adsorption kinetics, the experimented data were fitted with two kinetics models.
The pseudo-first order (PFO) kinetics is given by the following equation:
(6)\(\text{log}\left({q}_{e}-{q}_{t}\right)=\text{log}{q}_{e}-\frac{{k}_{1}}{2.303}\)
Where qe is the adsorption capacity at the equilibrium, qt is the adsorption capacity at contact time t, k1 is the PFO constant. This equation may also be rewritten as:
(7)\(q\left(t\right)={q}_{e}[1-\text{exp}\left(-{k}_{1}t\right)]\)
The pseudo-second order (PSO) kinetics is defined by the following equation:
(8)\(\frac{t}{{q}_{t}}=\frac{1}{{k}_{2}{q}_{e}^{2}}+\left(\frac{1}{{q}_{e}}\right)t\)
Where qe is the adsorption capacity at the equilibrium, qt is the adsorption capacity at contact time t, k2 is the PSO constant. The Eq. (8) may also be written in the following equation:
(9)\(q\left(t\right)={q}_{e}\frac{{k}_{2}{q}_{e}t}{1+{k}_{2}{q}_{e}t}\)
Perovskite films characterization
The plan-view and cross-sectional scanning electron microscopy (SEM) images were captured using a field emission SEM (JSM-7600F, JEOL Ltd., Japan). The X-ray diffraction patterns were analyzed using a SmartLab SE (Rigaku) at a scan speed of 6°/min. The absorbance spectra were obtained using an UV–visible spectroscopy (Lambda 35, PerkinElmer, USA). The steady-state and time-resolved photoluminescence (PL) spectra were recorded using PL a spectrometer (Quantaurus, Japan). The I-V curves for the space-charge-limited current (SCLC) analysis of the perovskite films (control, FA-20, and Sp-20) was obtained using a potentiostat (CHI660D, C H Instruments Inc., USA) with a voltage range from − 0.1 to 1.5V under dark state.
PSCs characterization
J–V curves were measured using the potentiostat for reverse (forward) scans under AM 1.5 G sun illumination condition using a solar simulator with a 450 W xenon lamp (94023A, Newport, USA). The reverse (forward) scans were conducted from 1.25 (-0.05) to -0.05 (1.25) V with a scan rate of 500 mV/s. The 1 sun illumination condition was calibrated using a standard Si solar cell (Oriel VLSI standards, Newport, USA). Aperture area was fixed using a thin metal mask with an area of 0.14 cm2. Stabilized power output was calculated by multiplying output current and bias voltage. The output current was measured using the potentiostat at a bias voltage corresponding to the initial maximum power point of each PSC under continuous 1sun illumination. Mott-Schottky plots were converted from capacitance-voltage curves measured by using the potentiostat in a forward direction from − 0.1 to 1.2 V at a fixed frequency of 10 kHz under dark state. Humidity and heat stability tests of PSCs were conducted without any encapsulation. For the humidity stability test, 8 cells per case of PSCs (control, R2, and R1) were placed in a humidity control chamber (relative humidity (RH) of ~ 55%) at 20℃ for specific time. For the heat stability test, 8 cells per case of PSCs (control, R2, and R1) were placed on a hot-plate set at 85℃ inside of an argon glovebox. After the PCE measurement of the PSCs during each stability test, the PSCs were stored back under the specific conditions.