Toward Water-Resistant, Tunable Perovskite Absorbers Using Peptide Hydrogel Additives

Abstract

In recent years, hydrogels have been demonstrated as simple and cheap additives to improve the optical properties and material stability of organometal halide perovskites (OHPs), with most research centered on the use of hydrophilic, petrochemical-derived polymers. Here, we investigate the role of a peptide hydrogel in passivating defect sites and improving the stability of methylammonium lead iodide (MAPI, CH₃NH₃PbI₃) using closely controlled, in situ X-ray photoelectron spectroscopy (XPS) techniques under realistic pressures. Optical measurements reveal that a reduction in the density of defect sites is achieved by incorporating peptide into the precursor solution during the conventional one-step MAPI fabrication approach. Increasing the concentration of peptide is shown to reduce the MAPI crystallite size, attributed to a reduction in hydrogel pore size, and a concomitant increase in the optical bandgap is shown to be consistent with that expected due to quantum size effects. Encapsulation of MAPI crystallites is further evidenced by XPS quantification, which demonstrates that the surface stoichiometry differs little from the expected nominal values for a homogeneously mixed system. In situ XPS demonstrates that thermally induced degradation in a vacuum is reduced by the inclusion of peptide, and near-ambient pressure XPS (NAP-XPS) reveals that this enhancement is partially retained at 9 mbar water vapor pressure, with a reduced loss of methylammonium (MA⁺) from the surface following heating achieved using 3 wt % peptide loading. A maximum power conversion efficiency (PCE) of 16.6% was achieved with a peptide loading of 3 wt %, compared with 15.9% from a 0 wt % device, the former maintaining 81% of its best efficiency over 480 h storage at 35% relative humidity (RH), compared with 48% maintained by a 0 wt % device.

Figure 1

(a) Schematic illustration of the fabrication process for peptide-incorporated MAPI films. (b) Chemical formula of the peptide FEFKFEFK (F, phenylalanine; E, glutamic acid; K, lysine). (c) Crystal structure of MAPI.

Figure 2

C K-edge (left) and N K-edge (right) PEY and TEY NEXAFS spectra from 0 and 3 wt % samples with prominent features labeled. A DFT simulated FEFKFEFK spectrum is included for comparison.

Figure 3

(a) XRD (patterns for higher peptide loadings are shown in Figure S1 of the Supporting Information). (b) UV–vis–NIR (spectra for higher loadings are shown in Figure S6). (c) PL spectra (normalized to peak height for clarity–unnormalized spectra are shown in Figure S5 of SI) of 0, 1.5, 3, and 6 wt % FEFKFEFK peptide. (d) Time-resolved PL spectra of MAPI containing 0 and 3 wt % peptide. Time-resolved PL spectra fitted with a double-exponential decay (0 wt %; see Supporting Information and Figure S5).

Figure 4

SEM images of (a) 0 wt % (b) 3 wt % (c) 6 wt %, showing a decrease in particle size as the peptide loading increases. Lower panels show TEM of (d) FEFKFEFK fibers, (e) 0 wt % and (f) 3 wt % samples. The peptide fibers can be seen as long criss-crossed structures with the peptide crystallites between them. Again, a decrease in MAPI crystallite size is observed in the image following addition of the peptide.

Figure 5

(a) VB spectra obtained using a He II UV source. (b) VB difference spectrum calculated by subtracting at 0 wt % VB from the 3 wt % VB (bottom) and a DFT simulated VB for FE+FK for comparison (top).

Figure 6

SEED edge (left) and VB (right) spectra of 0 and 3 wt % samples measured using a He I UV source (hν = 21.2 eV). The black dashed lines indicate the linear fits used to determine the work functions and VBM of the samples.

Figure 7

(a) Schematic diagram of the PSC device architecture. (b) J–V curve of PSCs with active layers containing 0 and 3 wt % peptide. (c) Evolution with time of PCE of 0 and 3 wt % devices stored at 35% RH.

Figure 8

(a) Pb 4f_7/2, I 3d_5/2, and N 1s high-resolution core-level XPS spectra of 0 and 3 wt % samples, measured in UHV at RT, 100 °C, 150 °C, and 180 °C. Initial peak positions are indicated by the dashed lines, and the final shift is noted (as measured following cooling). (b) 0 wt % and (c) 3 wt % N 1s core level spectra at RT and 150 °C with fitted peaks labeled.

Figure 9

Surface stoichiometry of 0 and 3 wt % samples as a function of temperature, calculated from the areas of fitted XPS peaks. I, N_MAPI, and Pb⁰ concentrations with respect to Pb²⁺ are shown. Values are listed in Tables S2 and S3. The nominal stoichiometry of MAPI is 1:3:1 (Pb²⁺/I/N).

Figure 10

(a) Pb 4f_7/2, I 3d_5/2, and N 1s high-resolution core-level XPS spectra of 0 and 3 wt % samples, measured under UHV conditions at RT; then at RT, 100, and 150 °C under 9 mbar water vapor pressure; and finally under UHV following cooling to RT. Initial peak positions are indicated by the dashed lines, and the final shift is noted (as measured following cooling). (b) 0 wt % and (c) 3 wt % N 1s core level spectra before and after the heating regime at 9 mbar water vapor pressure, with fitted peaks labeled.

Figure 11

Schematic diagram of peptide-incorporated MAPI, showing MAPI crystallites encapsulated by a network of peptide chains. Proposed mechanisms for enhanced stability are included: (a) inhibited ion migration, (b) reduced water ingress, (c) passivation of surface defects by peptide termini (in this case an undercoordinated Pb²⁺ is passivated by a carboxyl anion terminus⁻), and (d) passivation of surface defects by amino acid functional groups (in this case an MA⁺ vacancy is passivated by a protonated lysine side chain).