betaFeedback

WorksPhysics and Astronomy

Intense and stable second-harmonic generation in robust MePEA-based quasi-2D halide perovskite crystals

Abstract

Quasi-2D halide perovskites offer a unique platform for nonlinear optics applications, where structural and optical properties are tuned by the variation of organic spacer cations. Notably, quasi-2D perovskite single crystals can demonstrate intense and stable second-harmonic generation signals when a proper composition is selected. Here, the influence of phenethylammonium and 4-methyl-2-phenylethylammonium spacers on the crystal symmetry and second-harmonic signal response of quasi-2D perovskite phases with a varying number of single-octahedron [PbX6_6]4−^{4-} layers (nn) is investigated. High-quality perovskite crystals with lateral dimensions of several hundred micrometers are fabricated using a cavity-confined crystallization synthesis method. Optical characterization of the crystals shows strong and temporally stable second-harmonic generation. Power-dependent and polarization-resolved measurements confirm the second-order nonlinear origin of the observed response and reveal anisotropic emission patterns. Our results demonstrate a direct correlation between spacer-induced center-symmetry breaking and nonlinear optical activity. These results establish quasi-2D halide perovskites as a promising class of symmetry-engineered materials for efficient frequency conversion and integrated nonlinear photonics.

Manuscript.pdf

Introduction

Layered quasi-two-dimensional (quasi-2D) halide perovskites represent a promising class of hybrid organic-inorganic semiconductor materials for next-generation optoelectronics [1, 2] and nonlinear photonics [3, 4]. They consist of halide perovskite layers separated by insulating organic spacer layers. The band gap of the latter is larger than that of the former, which results in the localization of the excited charge carriers in the perovskite layers, i.e. reproducing multiple quantum wells [5]. Therefore, optical and electronic properties of these materials are dependent on the quantum confinement and dielectric screening effects, which, in turn, are governed by the number of single-octahedron [MX6]4−[\mathrm{MX}_{6}]^{4-} (M=Pb2+,Sn2+\mathrm{M}=\mathrm{Pb}^{2+}, \mathrm{Sn}^{2+}; X=Cl−\mathrm{X}=\mathrm{Cl}^{-}, Br−\mathrm{Br}^{-}, I−\mathrm{I}^{-}) layers (nn) in A2′An−1MnX3n+1\mathrm{A}_{2}^{\prime}\mathrm{A}_{n-1}\mathrm{M}_{n}\mathrm{X}_{3n+1} (A′\mathrm{A}^{\prime} – organic spacer; A=MA+\mathrm{A}=\mathrm{MA}^{+}, FA+\mathrm{FA}^{+}, Cs+\mathrm{Cs}^{+}) structure [6, 7].

With increasing nn, the exciton energy decreases due to weaker quantum confinement, while the electron-hole binding energy also decreases as a result of enhanced dielectric screening [8]. Consequently, the optical absorption edge undergoes a redshift, the exciton diffusion length increases, and the fraction of free charge carriers contributing to electrical conductivity gradually rises. In addition to the aforementioned tunability, quasi-2D perovskites have been demonstrated to exhibit enhanced moisture stability, suppressed ion migration, and lower structural defect densities compared with their bulk counterparts [5, 9]. Together, these characteristics position quasi-2D perovskites as a reliable bridge between strictly 2D and bulk semiconductors.

Beyond their linear optical properties, quasi-2D perovskites are attracting increasing attention as nonlinear optical (NLO) materials [10, 11]. Among various NLO processes, second-harmonic generation (SHG) is of particular importance due to its wide range of photonic applications, including frequency conversion in laser systems [12], high-resolution imaging [13, 14], and optical sensing [15]. SHG is governed by the second-order nonlinear susceptibility tensor χ(2)\chi^{(2)}. Within the electric-dipole approximation, this tensor is strictly forbidden in centrosymmetric crystals [16]. Consequently, efficient SHG can only occur in materials with a noncentrosymmetric crystal structure.

In layered halide perovskites, the incorporation of organic spacers can introduce structural asymmetry that is absent in bulk perovskites, resulting from octahedral distortions and spacer ordering. Hence, these materials are highly attractive for symmetry-engineered nonlinear optics due to proper spacer selection [17, 18]. Until now, most SHG-active quasi-2D perovskites have been leveraged linear alkyl spacers such as butylammonium (BA) halide [19, 20] or aromatic spacers like phenethylammonium (PEA) halide [21]. Nevertheless, BA- and PEA-based quasi-2D perovskites demonstrate poor long-term stability and low SHG conversion efficiency, which limits their potential practical application [21, 22]. Alternatively, chiral organic spacers provide another promising way to form noncentrosymmetric crystal structures with strong SHG [23, 24, 25]. However, the synthesis complexity of chiral ligands and their frequently limited structural and operational stability is considered as a significant issue for scalable nonlinear photonics [26, 27]. Consequently, rational spacer design strategies is critical for achieving simultaneously high SHG efficiency and structural robustness [28, 29]. In this context, methyl-substituted phenethylammonium spacers provide an attractive option for symmetry engineering without introducing molecular chirality. Specifically, the additional methyl group in 4-methyl-2-phenylethylammonium (MePEA) can modify molecular packing, induce lattice distortions, and afford noncentrosymmetry in quasi-2D perovskite structures. Although MePEA ligand has already demonstrated an excellent performance as a spacer cation in perovskite solar cells [30, 31, 32] and blue light-emitting diodes [33], its potential to enable nonlinear optical functionality in quasi-2D perovskites remains unexplored.

Here, we systematically study bromide quasi-2D perovskite single crystals with n=1−3n = 1-3 PEA and MePEA spacer cations. Bromide compositions were selected due to their enhanced resistance to laser-induced degradation and improved stability compared to iodide-based crystals [34, 35, 36]. We obtain high-quality single crystals via the cavity-confined crystallization method and directly study centrosymmetric and noncentrosymmetric perovskite phases. Through power-dependent and polarization-resolved SHG measurements, we establish a clear structure-property relationship between spacer-induced structural diversity and nonlinear optical response and demonstrate spacer-driven crystal symmetry engineering as an effective route toward robust second-order nonlinear functionality in quasi-2D halide perovskites.

Results and Discussion

Growth and characterization of quasi-2D perovskite microcrystals

To fabricate quasi-2D perovskite crystals, we employed a cavity-confined crystallization method. As schematically illustrated in Figure 1a, the precursor solution was confined within a cavity formed by two glass substrates and subsequently annealed at 60°C to induce crystal growth, as described in the Materials and Methods section. This approach enables the growth of high-quality quasi-2D perovskite crystals with lateral dimensions of several hundred micrometers and thicknesses of up to several micrometers. In this work, we focused on quasi-2D perovskites incorporating phenethylammonium (PEA) and 4-methylphenethylammonium (MePEA) as the organic spacer cations. The molecular structures of these spacer cations differ only by the presence of an additional methyl substituent at the para (4-) position of the phenyl ring in MePEA, as illustrated in Figure 1b.

Fabrication and characterization of quasi-2D perovskite crystals

Figure 1. Fabrication and characterization of quasi-2D perovskite crystals. (a) Schematic illustration of the cavity-confined crystallization process. (b) Chemical structures of the PEA and MePEA organic spacer cations. (c,d) PL spectra of MePEA2_2MAn−1_{n-1}Pbn_nBr3n+1_{3n+1} (n=1n = 1, 2, and 3) and PEA2_2MAn−1_{n-1}Pbn_nBr3n+1_{3n+1} (n=1n = 1 and 2) crystals, respectively. (e–i) Bright-field optical microscopy images (lower panel), dark-field (upper-left panel) and PL (upper-right panel) images of MePEA2_2PbBr4_4 (e), MePEA2_2MAPb2_2Br7_7 (f), MePEA2_2MA2_2Pb3_3Br10_{10} (g), PEA2_2PbBr4_4 (h), and PEA2_2MAPb2_2Br7_7 (i) crystals. Scale bars, 50 µm. (j–n) AFM images of the regions highlighted in (e–i).

We synthesized five quasi-2D perovskite compositions using this method: MePEA2PbBr4\mathrm{MePEA}_{2}\mathrm{PbBr}_{4}, MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7}, MePEA2MA2Pb3Br10\mathrm{MePEA}_{2}\mathrm{MA}_{2}\mathrm{Pb}_{3}\mathrm{Br}_{10}, PEA2PbBr4\mathrm{PEA}_{2}\mathrm{PbBr}_{4}, and PEA2MAPb2Br7\mathrm{PEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7}, corresponding to MePEA-based phases with n=1n = 1, 2, 3 and PEA-based phases with n=1n = 1, 2, respectively. The photoluminescence (PL) spectra of these crystals (Figure 1c and Figure 1d) exhibit well-defined emission peaks that systematically blueshift with decreasing nn, reflecting the stronger quantum confinement in lower-nn quasi-2D perovskites. The absence of additional emission bands confirms the formation of predominantly phase-pure quasi-2D structures, whereas the weak shoulder observed for the MePEA-based n=3n = 3 crystal is attributed to a minor amount of the n=4n = 4 phase, whose formation at the crystal edges and surface is promoted by humid environmental conditions [37, 38, 39]. Bright-field (lower panel), dark-field (upper-left panel), and PL images (upper-right panel) in Figure 1e–Figure 1i demonstrate well-defined crystal morphology and spatially uniform photoluminescence, indicative of high crystal quality. AFM measurements further reveal smooth crystal surfaces with a root-mean-square (RMS) roughness of around 4 nm and crystal thicknesses ranging from 0.3 to 2.1 µm (Figure 1j–Figure 1n). Representative crystals investigated in this work have thicknesses of 2.1 µm for MePEA2_2PbBr4_4, 1.1 µm for MePEA2_2MAPb2_2Br7_7, 1.5 µm for MePEA2_2MA2_2Pb3_3Br10_{10}, 0.4 µm for PEA2_2PbBr4_4, and 0.8 µm for PEA2_2MAPb2_2Br7_7.

Although the investigated crystals exhibit high structural quality and sufficient thickness for SHG measurements, crystal symmetry remains the key factor governing the nonlinear optical response. PEAbased quasi-2D perovskites, including PEA2_2PbBr4_4 and PEA2_2MAPb2_2Br7_7, are reported to crystallize in the centrosymmetric P1ˉP\bar{1} space group [40, 41] and are therefore not expected to exhibit measurable SHG.

We therefore investigated the effect of methyl substitution on the crystal structure using X-ray diffraction (XRD). The XRD patterns of PEA2_2PbBr4_4 and MePEA2_2PbBr4_4 are similar, with the diffraction peaks of MePEA2_2PbBr4_4 systematically shifted toward lower diffraction angles, indicating an increase in the interlayer spacing between adjacent inorganic layers (Figure 2a). Specifically, the (002) reflection shifts from 5.37° for PEA2_2PbBr4_4 to 5.07° for MePEA2_2PbBr4_4, corresponding to an increase in the interlayer spacing from 16.43 to 17.43 Å. Despite the incorporation of two additional methyl groups, each with an approximate size of 2 Å [42], this increase amounts to only about 1 Å. This observation suggests a reorientation of the organic spacer molecules, resulting in modified molecular packing and associated structural distortions that may alter the crystal symmetry.

Structural characterization of quasi-2D perovskite crystals

Figure 2. Structural characterization of quasi-2D perovskite crystals. (a) XRD pattern of PEA2_2PbBr4_4 and MePEA2_2PbBr4_4 crystals. Insets: enlarged view of the (002) reflections and optical images of the PEA2_2PbBr4_4 and MePEA2_2PbBr4_4 crystals. Scale bars, 50 μ\mum. (b,c) SAED patterns of the MePEA2_2PbBr4_4 and MePEA2_2MAPb2_2Br7_7 crystals, respectively.

To examine the crystal symmetry of the MePEA-based quasi-2D perovskites, we performed selected area electron diffraction (SAED) measurements. The SAED patterns of MePEA2_2PbBr4_4 and MePEA2_2MAPb2_2Br7_7 single crystals are shown in Figure 2b and Figure 2c, respectively. The SAED pattern of MePEA2_2PbBr4_4 reveals a centered rectangular 2D Bravais lattice with primitive vectors a≠ba \ne b (a=2.57a = 2.57 Å, b=4.27b = 4.27 Å) and γ=90∘\gamma= 90^\circ (a∥ha \parallel h and b∥kb \parallel k) [43]. Such a lattice is compatible with several possible space groups. Considering that related layered perovskites are commonly reported to crystallize in the centrosymmetric monoclinic space group C2/mC2/m [44], MePEA2_2PbBr4_4 is most likely centrosymmetric, consistent with the behavior of most n=1n = 1 layered perovskites. In contrast, the SAED pattern of MePEA2_2MAPb2_2Br7_7 reveals an oblique 2D Bravais lattice with a≠ba \ne b (a=7.97a = 7.97 Å, b=8.10b = 8.10 Å) and γ=86∘\gamma= 86^\circ. This lattice geometry is compatible with low-symmetry space groups such as P1P1 and P2P2. The observed diffraction symmetry indicates the presence of a twofold rotational axis, supporting the assignment to the non-centrosymmetric space group P2P2 and indicating the absence of inversion symmetry.

Second-harmonic generation in quasi-2D perovskites

The structural characterization suggests that methyl substitution of the PEA spacer can promote structural distortions and inversion-symmetry breaking in quasi-2D perovskites. To investigate the impact of spacer-cation engineering on the second-order nonlinear optical response, we performed a systematic SHG study of all five quasi-2D perovskite compositions investigated in this work. The SHG measurements were performed in a reflection geometry using femtosecond laser excitation at a wavelength of 1045 nm (see Section S1 of the Supporting Information for experimental details).

In agreement with the structural characterization, the nonlinear optical responses of the MePEA-based quasi-2D perovskites differ markedly. As depicted in Figure 3a, MePEA2_2PbBr4_4 exhibits multi-photon pho-

SHG spectra, pump-power dependence, and polar plots for MePEA-based quasi-2D perovskite crystals

Figure 3. SHG in MePEA-based quasi-2D perovskite crystals under excitation at λexc=1045\lambda_{\mathrm{exc}} = 1045 nm. (a–c) Nonlinear emission spectra of MePEA2PbBr4\mathrm{MePEA}_{2}\mathrm{PbBr}_{4}, MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7}, and MePEA2MA2Pb3Br10\mathrm{MePEA}_{2}\mathrm{MA}_{2}\mathrm{Pb}_{3}\mathrm{Br}_{10}, respectively. (d) Pump-power dependence of the SHG and mPPL intensities of MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7}. (e,f) Polar plots of the total and polarization-resolved SHG intensities as functions of the incident polarization angle. Dots represent the experimental data, and solid lines represent the theoretical fits.

Second-harmonic generation in quasi-2D perovskites

toluminescence (mPPL), whereas no detectable SHG is observed prior to laser-induced damage, consistent with its likely centrosymmetric structure. Likewise, no measurable SHG is observed for PEA2PbBr4\mathrm{PEA}_{2}\mathrm{PbBr}_{4} or PEA2MAPb2Br7\mathrm{PEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} (Figure S3, Supporting Information), in accordance with previous reports assigning both compositions to the centrosymmetric Pˉ1\bar{P}1 space group.

Nonlinear emission spectra of PEA2PbBr4 and PEA2MAPb2Br7 crystals

Figure S3. Nonlinear emission spectra of PEA2PbBr4\mathrm{PEA}_{2}\mathrm{PbBr}_{4} and PEA2MAPb2Br7\mathrm{PEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} crystals.

In contrast, under similar excitation conditions, a pronounced SHG signal emerges from MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} alongside the mPPL emission, as shown in Figure 3b. The corresponding effective second-order nonlinear susceptibility, estimated using crystalline urea as a reference, is approximately χ(2)=0.88\chi^{(2)} = 0.88 pm/V (see Section S2 of the Supporting Information for details). This value is comparable to that of KDP [45] and is nearly twice as large as the value reported for BA2MAPb2Br7\mathrm{BA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} [19].

SHG remains observable in MePEA2MA2Pb3Br10\mathrm{MePEA}_{2}\mathrm{MA}_{2}\mathrm{Pb}_{3}\mathrm{Br}_{10} (Figure 3c), although with substantially lower intensity than in MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7}. The weaker SHG response may be associated with changes in local structural distortions and dipole ordering as the inorganic layer thickness increases, potentially modifying the degree of inversion-symmetry breaking. A similar decrease in SHG intensity with increasing nn was previously reported for (BA)2(MA)n−1PbnI3n+1(\mathrm{BA})_{2}(\mathrm{MA})_{n-1}\mathrm{Pb}_{n}\mathrm{I}_{3n+1} perovskites, where the observed SHG response was attributed to local symmetry-breaking effects [20]. Interestingly, a nonmonotonic dependence on nn was also observed for the third-order nonlinear susceptibility of these perovskites, highlighting the influence of quantum confinement on their nonlinear optical properties [46]. These findings underscore the complex relationship between inorganic layer thickness, crystal symmetry, and nonlinear optical response, which warrants further investigation.

We next focus on MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7}, which exhibits the strongest SHG response among the investigated compositions. To verify that the observed signal originates from a second-order nonlinear optical process, we measured the pump-power dependence of the SHG intensity, as shown in Figure 3d. The SHG signal remains clearly observable over a broad pump-power range and scales nearly quadratically with the incident pump power below 80 mW, yielding a fitted slope of approximately 1.9. At higher pump powers, the SHG intensity deviates from the quadratic dependence and gradually approaches saturation. Meanwhile, the mPPL pump-power dependence reveals a slope of 3.6, consistent with three-photon absorption followed predominantly by radiative excitonic recombination, with an additional contribution from free-carrier recombination [47].

To further examine the crystal symmetry, we performed polarization-resolved SHG measurements on a MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} crystal. The excitation beam was incident normal to the crystal surface, while the in-plane crystal orientation was controlled by rotating the sample (see Section S1 and Figure S7 of the Supporting Information). Figure 3e presents the total SHG intensity (without an analyzer in the detection path) as a function of the excitation polarization, characterized by a twofold angular dependence. The SHG signal was subsequently measured with the analyzer oriented parallel (p-polarized) or perpendicular (s-polarized) to the laboratory X-axis. As shown in Figure 3f, the p- and s-polarized SHG exhibit four- and two-lobe patterns, respectively, revealing a pronounced in-plane anisotropy of the nonlinear optical response. The polarization patterns are well reproduced by the theoretical model based on twofold rotational symmetry (see Section S3 of the Supporting Information for details), supporting the P2P2 space-group assignment inferred from the SAED measurements.

Coordinate system

Figure S7. Coordinate system.

Spectral characteristics and stability of SHG

We next investigate the SHG response of MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} over a broad range of excitation wavelengths (Figure 4a). As the excitation wavelength is varied from 810 to 1410 nm, the SHG peak shifts accordingly, appearing at half of the excitation wavelength. In general, the SHG intensity is expected to increase as the generated second-harmonic wavelength approaches an excitonic resonance, where the second-order nonlinear susceptibility is enhanced while absorption of the fundamental excitation remains relatively low. Accordingly, the SHG intensity increases as the second-harmonic wavelength approaches the excitonic absorption peak of the n=2n = 2 MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} crystal at approximately 440 nm. However, the maximum SHG intensity is observed at a wavelength of 525 nm, corresponding to a fundamental excitation wavelength of 1050 nm. The optimum SHG response therefore occurs at a longer wavelength than the excitonic absorption peak, as previously reported for quasi-2D perovskites [48]. This behavior likely reflects a balance between resonance enhancement of the second-order nonlinear susceptibility and competing effects, including pump absorption, self-absorption of the generated SHG, phase mismatch, and laser-induced heating [3, 49].

Wavelength dependence and stability of SHG in MePEA2MAPb2Br7 crystal

Figure 4. Wavelength dependence and stability of SHG in MePEA2_2MAPb2_2Br7_7 crystal. (a) SHG spectra recorded under different excitation wavelengths. (b) Temporal evolution of the SHG intensity under continuous excitation at λexc=1045\lambda_{\mathrm{exc}} = 1045 nm and a pump intensity of 9.59.5 GW/cm2^2, measured at relative humidity levels of 10%, 50%, and 90%.

The temporal stability of the SHG response was initially evaluated under ambient conditions (50% relative humidity, RH) with continuous femtosecond excitation at 1045 nm and an incident optical intensity of 9.5 GW/cm2^{2} (Figure 4b). After 60 min of irradiation, the SHG intensity retained approximately 86% of its initial value, demonstrating sustained nonlinear optical response under prolonged excitation. By comparison, previously reported chiral (R- and S-β\beta-methylphenethylammonium)BAPbBr4_4 perovskites exhibited similar SHG retention of approximately 85% after only 20 min of irradiation at a slightly lower pump intensity of 7.86 GW/cm2^{2} [23]. Furthermore, a MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} crystal stored under ambient conditions for approximately one year continued to exhibit a pronounced SHG response that remained stable under intense femtosecond excitation (see Figure S8 of the Supporting Information). The effect of environmental humidity was also examined at 10% and 90% relative humidity using a custom-built humidity-controlled chamber equipped with a humidity sensor (see Section S1, Supporting Information). At 10% RH, the SHG intensity retained approximately 84% of its initial value after 60 min, whereas at 90% RH, the response remained nearly unchanged throughout the measurement (Figure 4b). This remarkable stability under high humidity may be associated with the increased hydrophobicity of the methyl-substituted organic spacer. Methyl substitution is known to increase the lipophilicity of aromatic molecules [50], potentially reducing moisture penetration into the perovskite structure.

Pump-power dependence and temporal evolution of SHG intensity after ambient storage

Figure S8. (a) Pump-power dependence of the SHG intensity of MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} after storage under ambient conditions for 12 months. (b) Temporal evolution of the SHG intensity under continuous excitation at λexc=1045\lambda_{\mathrm{exc}} = 1045 nm and a pump intensity of 9.5 GW/cm2\mathrm{GW}/\mathrm{cm}^{2}, measured at a relative humidity of 50%. Insets: Optical images of the SHG and combined SHG+mPPL signals.

To further assess the resistance of MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} to intense optical excitation, we determined its laser damage threshold (LDT). The LDT was estimated from the pump intensity at which the SHG response deviates from the quadratic power dependence [51]. As shown in Figure 3d, this deviation occurs at a pump power of approximately 80 mW, corresponding to an LDT of 107.84 GW/cm2^{2} (16.17 mJ/cm2^{2}).

This value exceeds those reported for most quasi-2D and other low-dimensional halide perovskites with organic spacers under near-infrared excitation (800–1100 nm), as summarized in Table 1. Under prolonged high-intensity laser irradiation, these materials can undergo photoinduced degradation associated with photooxidation [52] and migration or loss of organic spacer cations [53], which can progressively disrupt the layered structure and reduce the nonlinear optical response [5]. Resistance to such degradation is strongly influenced by the molecular structure and packing arrangement of the organic spacer. In particular, structural complexity and steric hindrance of spacer ligands can introduce additional lattice disorder and disrupt ligand packing [54, 55, 56, 57], potentially compromising structural robustness under intense optical excitation. The high LDT of MePEA2_2MAPb2_2Br7_7 may therefore be related to the improved spacer packing induced by methyl substitution, which simultaneously promotes inversion-symmetry breaking and improves resistance against laser-induced degradation, further supporting their potential for nonlinear photonic applications.

Materialλex\lambda_{\mathrm{ex}} [nm]τ\tau [fs]ff [MHz]dd [μ\mum]Power [mW]LDT [GW cm−2^{-2}]χ(2)\chi^{(2)} [pm V−1^{-1}]Ref.
(R/S-BrsBA)2_2PbBr4_484013076368.59b^{\mathrm{b}}-[25]
(S-2-C5_5H14_{14}N2_2)2_2PbI6_6960100802074529.66-[58]
(S-2-MPD)PbBr3_3980100802071428.4229.9[59]
(R-MBA)CuBr2_294010082--1724.7[60]
(S-FMBA)2_2SnCl6_68601008020100039.8118.84[61]
(R-MPEA)6_6InCl9_9800100802096138.26-[62]
DMA4_4[InCl6_6]Br1040100802080031.85-[63]
R-2-mpipSnBr6_610001008020118547.17-[64]
(R-3-APD)SbCl5_51020100802096139.4-[65]
MPEA2_2SnBr6_6840100802042016.720.33[66]
(C4_4H10_{10}NO)PbBr3_380014080216.47b^{\mathrm{b}}46.820.32[67]
(S-NEA)3_3Ge2_2I7⋅_7\cdotH2_2O88010080686.99b^{\mathrm{b}}38.460.86[68]
(L-Pro)PbI3⋅_3\cdotH2_2O920100802070027.879.5[69]
(S-APD)PbBr4_48001008020149159.360.4[51]
(R-2-EP)PbBr4_48601008020136254.220.33[70]
(R-BrMBA)2_2GeI4_41010100806157.769.75b^{\mathrm{b}}0.4[71]
MePEA2_2MAPb2_2Br7_7104515070380107.840.88This work

Table 1. Comparison of the second-order NLO performance of different low-dimensional perovskite materials.

Conclusion

We demonstrated that molecular engineering of an organic spacer cation provides an effective strategy for achieving robust second-harmonic generation (SHG) in quasi-2D bromide perovskites. While PEA-based and the n=1n = 1 MePEA-based microcrystals exhibited no detectable SHG, methyl substitution enabled a pronounced nonlinear optical response in the n=2n = 2 MePEA2_2MAPb2_2Br7_7 crystals, with a weaker response observed for the n=3n = 3 counterparts. Polarization-resolved SHG measurements supported the assignment of the n=2n = 2 phase to the noncentrosymmetric P2P2 space group, highlighting the role of spacer-induced inversion-symmetry breaking. Notably, MePEA2_2MAPb2_2Br7_7 exhibited an effective second-order nonlinear susceptibility of χ(2)=0.88\chi^{(2)} = 0.88 pm/V, broadband SHG under excitation from 810 to 1410 nm, and an estimated laser damage threshold of 107.84 GW/cm2^2. Its nonlinear response remains stable under prolonged femtosecond excitation at 90% relative humidity, while pronounced SHG persists after 12 months of ambient storage. Our findings reveals how a subtle molecular modification - the introduction of a single methyl group into the organic spacer cation - significantly influences crystal symmetry, nonlinear optical response, and environmental stability of quasi-2D halide perovskites. This structure-property relationship highlights organic-spacer engineering as a promising strategy for developing stable and efficient quasi-2D perovskites for optical frequency conversion and integrated nonlinear photonics.

a^{\mathrm{a}} λex\lambda_{\mathrm{ex}} denotes the excitation wavelength, τ\tau the pulse duration, ff the repetition rate, and dd the laser spot diameter.

b^{\mathrm{b}} Calculated from the reported parameters using LDT=4P/(fπd2τ)LDT = 4P/(f\pi d^2\tau), where PP is the incident power at the damage threshold.

Experimental Section

Quasi-2D perovskite solution preparation

Quasi-2D perovskite precursor solutions of MePEA2_2MAPb2_2Br7_7 and PEA2_2MAPb2_2Br7_7 were prepared by dissolving 0.3 mmol PbBr2_2 (≥99.9%, Lanhit), 0.15 mmol MABr (≥99.99%, Greatcell Solar Materials), and 0.3 mmol MePEABr (≥98%, Aladdin Scientific) or PEABr (≥98%, Greatcell Solar Materials) in anhydrous dimethyl sulfoxide (DMSO).

The MePEA2_2PbBr4_4 and PEA2_2PbBr4_4 solutions were prepared by dissolving 0.3 mmol PbBr2_2 (≥99.9%, Lanhit) and 0.6 mmol MePEABr (≥98%, Aladdin Scientific) or PEABr (≥98%, Greatcell Solar Materials) in anhydrous DMSO. For MePEA2_2MA2_2Pb3_3Br10_{10}, 0.3 mmol PbBr2_2 (≥99.9%, Lanhit), 0.2 mmol MABr (≥99.99%, Greatcell Solar Materials), and 0.2 mmol MePEABr (≥98%, Aladdin Scientific) were dissolved in anhydrous DMSO.

Cavity-confined crystallization of perovskite microcrystals

Cavity-confined crystallization (a modified space-confined crystallization method [72]) was used to obtain crystals up to about 100 μ\mum in size on the glass substrate. In this method, two glass substrates were attached together using a U-shaped metal bracket with a screw, the latter fitted with a cup nut, and 3 μ\muL of the stoichiometric quasi-2D perovskite DMSO solution was then deposited between the two glass substrates, as illustrated in Figure 1a. The uniform spreading of the solution is attributed to the capillary effect. The metal bracket with substrates inside was annealed in a laboratory oven at 60°C for 48 hours to induce crystal growth.

Structural and optical measurements

To characterize the crystal structure of the quasi-two-dimensional halide perovskite with the composition MePEAn_nMAn−1_{n-1}Pbn_nBr3n+1_{3n+1} (phases n=1n = 1 and n=2n = 2), selected-area electron diffraction patterns were recorded along a specific crystallographic zone using a ZEISS LIBRA 200FE transmission electron microscope.

For diffraction analysis, individual perovskite flakes were transferred onto 300-mesh copper grids with a carbon film. These flakes were prepared via mechanical exfoliation using Kapton tape, then transferred onto a polydimethylsiloxane (PDMS) lens. The lens was subsequently dissolved in hexane, releasing the flakes into suspension, which was then drop-cast onto the copper grids. Suitable samples for further investigation were identified on the grids using an Axio Imager A2m optical microscope (Carl Zeiss SMT) equipped with a 350 nm UV excitation source, coupled with an Ocean Optics QE PRO spectrometer.

X-ray diffraction patterns were collected using a Bruker D8 Advance diffractometer (Bruker AXS, Germany) equipped with Cu Kα radiation (λ=1.5406A˚\lambda= 1.5406 \mathring{\mathrm{A}}), operating at 40 kV and 40 mA.

Dark-field, bright-field, and photoluminescence images of perovskite crystals were obtained on a Karl Zeiss AxioImager A2m microscope coupled with halogen and mercury lamps utilizing 50x and 20x magnification objectives. PL spectra were measured using a QE Pro Ocean Optics spectrometer connected to the Karl Zeiss AxioImager A2m optical microscope with optical fiber.

A detailed description of the SHG measurements, including a schematic illustration of the optical setup, is provided in Section S1 of the Supporting Information.

Supporting Information

Supporting Information is available from the Wiley Online Library or from the author.

Acknowledgements

This work was supported by the Russian Science Foundation (project No. 25-22-00278, https://rscf.ru/project/25-22-00278/). The SAED investigations were conducted at the ”Interdisciplinary Resource Center for Nanotechnology” of the Research Park of St. Petersburg State University. The authors thank D.V. Danilov for performing the SAED measurements.

Conflict of Interest

The authors declare no conflict of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

SI.pdf

Nonlinear optical measurements

All optical experiments were performed on a self-made setup, presented in Figure S1. The studied sample was placed on a three-axis stage and rotated at different angles around its axis. The femtosecond laser emission (1045 nm, 150 fs, 70 MHz) from the Yb-doped laser (TeMa, Avesta Project) was focused onto the sample to a spot size of 3 μm3\,\mu\mathrm{m} using 50×50\times NIR objective (Mitutoyo Plan APO, NA=0.42\mathrm{NA}=0.42). The generated second harmonic signal was collected through the same objective and sent to the camera and Horiba LabRam HR spectrometer coupled with a detector (CCD, Andor DU 420A-OE 325). The incident beam and emitted light were separated with a dichroic mirror (DMSP 900, Thorlabs). Fundamental frequency was additionally blocked on the way to the spectrometer by a shortpass filter (FESH 750, Thorlabs). The excitation beam power was adjusted via attenuator (Glan-Taylor prism and half-wavelength plate) and measured with a power meter (Ophir photonics StarBright). Polarization of the incident light was tuned through the rotation of another half-wave plate. The polarization of the SHG signal was selected by an analyzer. The pump

SHG measurements setup

Figure S1. SHG measurements setup.

wavelength-dependent measurements were conducted under the femtosecond laser excitation (810-1410 nm, 200 fs, 80 MHz) provided by the optical parametric amplifier (TOPOL-C, Avesta) at a fixed pump power of 10 mW. The laser beam was passing through the dichroic mirror (DMLP 800, Thorlabs) and focused on the sample using the 50×50 \times NIR objective. The generated signal was transferred to the spectrometer (Andor Shamrock 500i).

SHG measurements at different relative humidity (RH) levels were performed using a custom-built chamber equipped with a humidity sensor. The chamber featured a threaded opening for sample placement and a transparent glass window integrated into its top surface. The desired humidity level was established by placing the chamber in an environment with the target RH until the internal humidity stabilized. The chamber was then sealed with Parafilm to preserve these conditions during SHG measurements. The excitation beam was directed onto the sample through the glass window, and the generated SHG signal was collected in reflection geometry. The RH inside the chamber was monitored using the humidity sensor (RH measurement error 0.1 %).

Nonlinear optical characterization

Additionally, in this study, we investigated the relative SHG efficiency of MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} crystals. Crystalline urea was employed as a standard for SHG. S1 The SHG efficiency of the MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} crystals is found to be 0.15 ×\times crystalline urea, based on statistics, which were collected for approximately 20 perovskite and urea crystals at varying pumping intensities. The SHG signal from perovskite and urea crystals was measured in the reflection configuration under the same experimental conditions at λexc=1045\lambda_{\mathrm{exc}} = 1045 nm (Figure S2). The size and shape of the perovskite and urea crystals were selected to be comparable. Based on these measurements, we estimated the second-order susceptibility χ(2)\chi^{(2)} of MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7}. from the following expression:

SHG intensity spectra of MePEA2MAPb2Br7 crystal and crystalline urea

Figure S2. SHG intensity of MePEA2_2MAPb2_2Br7_7 crystal and crystalline urea at 2.4 mJ/cm2^2 fluence.

χS(2)=χR(2)∣IS(2ω)IR(2ω)∣1/2,(1)\chi_{\mathrm{S}}^{(2)}=\chi_{\mathrm{R}}^{(2)}\left|\frac{I_{\mathrm{S}}(2\omega)}{I_{\mathrm{R}}(2\omega)}\right|^{1/2}, \tag*{(1)}

where χR(2)\chi_{\mathrm{R}}^{(2)} is SHG coefficient of the urea χR(2)∼2.4\chi_{\mathrm{R}}^{(2)} \sim2.4 pm/V,S2 and IS(2ω)I_{\mathrm{S}}(2\omega) and IR(2ω)I_{\mathrm{R}}(2\omega) are the SHG intensities of the sample and urea reference, respectively. Thus, the absolute second-order SHG coefficient of MePEA2_2MAPb2_2Br7_7 at 1045 nm can be estimated as χS(2)∼0.9\chi_{\mathrm{S}}^{(2)} \sim0.9 pm/V. This obtained χ(2)\chi^{(2)} value exceeds that of α\alpha-quartz (χα-quartz(2)=0.67\chi_{\alpha\text{-quartz}}^{(2)} = 0.67 pm/V) and is comparable with KDP (χKDP(2)=0.88\chi_{\mathrm{KDP}}^{(2)} = 0.88 pm/V).S3 Moreover, it is almost twice as large as the values reported for BA2_2MAPb2_2Br7_7 (χ(2)=0.56\chi^{(2)} = 0.56 pm/V).S4

Figure S3 shows the mPPL spectra of PEA2_2PbBr4_4 and PEA2_2MAPb2_2Br7_7 crystals under 1045-nm excitation at a pump power of approximately 10 mW, with no detectable SHG signal.

Figures S4, S5 and S6 present polar plots of the measured SHG response as a function of the sample rotation angle in the plane perpendicular to the incident field, acquired with 15∘^\circ steps for unpolarized, p-polarized, and s-polarized SHG signals, respectively. The observed evolution of the polar plot shape with sample orientation indicates an anisotropy in the MePEA-based quasi-2D perovskite crystals.

SHG theoretical analysis

The quasi-2D perovskite crystal MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} is characterized by monoclinic symmetry belonging to the P2P2 space group with twofold rotation symmetry about the cc axis. In the case of the C2C_{2} point group, the second-order nonlinear polarization (P2ωP^{2\omega}) can be written in matrix notation as:

Polar plots of unpolarized SHG intensity for different $\theta$ angles

Figure S4. Polar plots of unpolarized SHG intensity as a function of incident polarization angle for different θ\theta angles, corresponding to substrate rotation in the plane perpendicular to the incident field with a 15∘15^{\circ} step for MePEA2_2MAPb2_2Br7_7 crystal.

\begin{pmatrix} P_{x}^{2\omega} \\ P_{y}^{2\omega} \\ P_{z}^{2\omega} \end{pmatrix} =2\varepsilon_{0} \begin{pmatrix} 0 & 0 & 0 & d_{14} & 0 & d_{16} \\ d_{21} & d_{22} & d_{23} & 0 & d_{25} & 0 \\ 0 & 0 & 0 & d_{34} & 0 & d_{36} \end{pmatrix} \begin{pmatrix} E_{x}^{\omega}E_{x}^{\omega} \\ E_{y}^{\omega}E_{y}^{\omega} \\ E_{z}^{\omega}E_{z}^{\omega} \\ 2E_{y}^{\omega}E_{z}^{\omega} \\ 2E_{x}^{\omega}E_{z}^{\omega} \\ 2E_{x}^{\omega}E_{y}^{\omega} \end{pmatrix}, \tag*{(2)}\end{pmatrix}

where ε0\varepsilon_{0} is the permittivity of free space and xx, yy and zz are the crystal coordinates. Here we have used a convenient contracted notation dimd_{im} of the second-order nonlinear susceptibility tensor χijk(2)\chi_{ijk}^{(2)}, where the first subscript refers to 1: xx, 2: yy, 3: zz and the second to 1: xxxx, 2: yyyy, 3: zzzz, 4: yzyz, zyzy, 5: xzxz, zxzx, 6: xyxy, yxyx.

In the experiment, the laser is normally incident on the crystal (Ez=0E_{z}=0) and linearly polarized in the sample plane, at an angle φ\varphi with respect to the X laboratory axis (Figure S7).

The orientation of the excitation linear polarization is varied by rotating the half-waveplate.

Thus, the incident field in laboratory coordinates can be expressed as

(EXωEYω)=(E0cos⁡φE0sin⁡φ)\begin{pmatrix} E_{X}^{\omega} \\ E_{Y}^{\omega} \end{pmatrix} = \begin{pmatrix} E_{0}\cos\varphi\\ E_{0}\sin\varphi\end{pmatrix}

where E0E_{0} is the amplitude of the electric field. We then express the excitation field in crystal coordinates by multiplying with the rotation matrix:

(ExωEyω)=(cos⁡(θ)sin⁡(θ)−sin⁡(θ)cos⁡(θ))(EXωEYω)=(E0cos⁡(φ−θ)E0sin⁡(φ−θ))(3)\begin{pmatrix} E_{x}^{\omega} \\ E_{y}^{\omega} \end{pmatrix} = \begin{pmatrix} \cos(\theta) & \sin(\theta) \\ -\sin(\theta) & \cos(\theta) \end{pmatrix} \begin{pmatrix} E_{X}^{\omega} \\ E_{Y}^{\omega} \end{pmatrix} = \begin{pmatrix} E_{0}\cos(\varphi-\theta) \\ E_{0}\sin(\varphi-\theta) \tag*{(3)} \end{pmatrix}

where the crystal xx axis is taken at an angle θ\theta from the laboratory XX axis (Figure S7).

Considering the propagation of the fundamental and SHG signals along the ZZ-axis, (2) reduces to:

(Px2ωPy2ω)=2ε0E02(d16sin⁡(2(φ−θ))d21cos⁡2(φ−θ)+d22sin⁡2(φ−θ))(4)\begin{pmatrix} P_{x}^{2\omega} \\ P_{y}^{2\omega} \end{pmatrix} = 2\varepsilon_{0}E_{0}^{2}\begin{pmatrix} d_{16}\sin\left(2(\varphi-\theta)\right) \\ d_{21}\cos^{2}(\varphi-\theta)+d_{22}\sin^{2}(\varphi-\theta) \tag*{(4)} \end{pmatrix}

We then transform Equation (4) back to laboratory coordinates by multiplying with the rotation matrix (cos⁡(θ)−sin⁡(θ)sin⁡(θ)cos⁡(θ))\begin{pmatrix} \cos(\theta) & -\sin(\theta) \\ \sin(\theta) & \cos(\theta) \end{pmatrix}, yielding the following expressions for the intensity of the SHG field.

When the analyzer is oriented parallel to the X-axis, the SHG intensity is given by:

I∥2ω∼∣PX∣2∼116(−2(d21+d22)sin⁡(θ)+(−2d16+d21−d22)sin⁡(θ−2φ)+(−2d16−d21+d22)sin⁡(3θ−2φ))2(5)I_{\parallel}^{2\omega} \sim|P_X|^2 \sim\frac{1}{16}\left(-2(d_{21}+d_{22})\sin(\theta)+(-2d_{16}+d_{21}-d_{22})\sin(\theta-2\varphi)+(-2d_{16}-d_{21}+d_{22})\sin(3\theta-2\varphi)\right)^2 \tag*{(5)}

For the case of the analyzer perpendicular to the X-axis:

I⊥2ω∼∣PY∣2∼116(2(d21+d22)cos⁡(θ)+(−2d16+d21−d22)cos⁡(θ−2φ)+(2d16+d21−d22)cos⁡(3θ−2φ))2(6)I_{\perp}^{2\omega} \sim|P_Y|^2 \sim\frac{1}{16}\left(2(d_{21}+d_{22})\cos(\theta)+(-2d_{16}+d_{21}-d_{22})\cos(\theta-2\varphi)+(2d_{16}+d_{21}-d_{22})\cos(3\theta-2\varphi)\right)^2 \tag*{(6)}

The SH intensities I∥2ωI_{\parallel}^{2\omega} and I⊥2ωI_{\perp}^{2\omega} (or IpI_p and IsI_s) can also be expressed in terms of dimensionless ratios of the χ(2)\chi^{(2)} tensor elements in order to fit the experimental data:

I∥2ω=A(−2(B+1)sin⁡(θ)+(−2C+B−1)sin⁡(θ−2φ)+(−2C−B+1)sin⁡(3θ−2φ))2I⊥2ω=A(2(B+1)cos⁡(θ)+(−2C+B−1)cos⁡(θ−2φ)+(2C+B−1)cos⁡(3θ−2φ))2(7)\begin{aligned} I_{\parallel}^{2\omega} &= A\left(-2(B+1)\sin(\theta)+(-2C+B-1)\sin(\theta-2\varphi)+(-2C-B+1)\sin(3\theta-2\varphi)\right)^2 \\ I_{\perp}^{2\omega} &= A\left(2(B+1)\cos(\theta)+(-2C+B-1)\cos(\theta-2\varphi)+(2C+B-1)\cos(3\theta-2\varphi)\right)^2 \tag*{(7)} \end{aligned}

where A is a multiplication factor and parameters B and C depend on ratios of the second-order susceptibility tensor.

We fit the experimental I∥2ωI_{\parallel}^{2\omega} and I⊥2ωI_{\perp}^{2\omega} using Equation (7) over a wide range of θ\theta and obtained perfect agreement between the theoretical and measured dependencies (Figure S5 and Figure S6).

Polar plots of p-polarized SHG intensity for different $\theta$ angles

Figure S5. Polar plots of p-polarized SHG intensity as a function of incident polarization angle for different θ\theta angles, corresponding to substrate rotation in the plane perpendicular to the incident field with a 15∘15^{\circ} step for MePEA2_2MAPb2_2Br7_7 crystal. Dots represent the experiment data, and solid lines represent fitting.

Polar plots of s-polarized SHG intensity as a function of incident polarization angle for different $\theta$ angles

Figure S6. Polar plots of s-polarized SHG intensity as a function of incident polarization angle for different θ\theta angles, corresponding to substrate rotation in the plane perpendicular to the incident field with a 15° step for MePEA2_2MAPb2_2Br7_7 crystal. Dots represent the experiment data, and solid lines represent fitting.

For the case of θ=0\theta=0, Equation (7) can be further simplified. The total SHG intensity (without an analyzer) then reads I2ω=I∥2ω+I⊥2ω∝(2Csin⁡(2φ))2+(B+1+(B−1)cos⁡(2φ))2I^{2\omega}=I_{\parallel}^{2\omega}+I_{\perp}^{2\omega}\propto(2C\sin(2\varphi))^2+(B+1+(B-1)\cos(2\varphi))^2. The calculated polar dependencies of I2ωI^{2\omega}, I∥2ωI_{\parallel}^{2\omega}, and I⊥2ωI_{\perp}^{2\omega} for θ=0\theta=0 are shown in Figure 3e and Figure 3f. of the main text. The extracted relative ratios of the tensor components are estimated as ∣d22/d21∣∼2.3|d_{22}/d_{21}|\sim2.3 and ∣d22/d16∣∼1.16|d_{22}/d_{16}|\sim1.16.

SHG after long-term ambient storage

To evaluate the SHG response after long-term storage, a MePEA2MAPb2Br7\mathrm{MePEA}_{2}\mathrm{MAPb}_{2}\mathrm{Br}_{7} crystal stored under ambient conditions for 12 months was investigated. The crystal exhibited a pronounced SHG signal with a nearly quadratic pump-power dependence with fitted slope of 1.9, as shown in Figure S8. Furthermore, the SHG intensity remained stable during 60 min of continuous femtosecond excitation at 1045 nm with a pump intensity of 9.5 GW/cm2\mathrm{GW}/\mathrm{cm}^{2} (Figure S8).

References

  1. [1](S1) Kurtz, S.; Perry, T. A powder technique for the evaluation of nonlinear optical materials. Journal of applied physics 1968, 39, 3798–3813.
  2. [2](S2) Lin, Z.; Wang, Z.; Chen, C.; Lee, M.-H. Mechanism of linear and nonlinear optical effects of KDP and urea crystals. The Journal of chemical physics 2003, 118, 2349–2356.DOI
  3. [3](S3) Fleming, J. W.; Weber, M. J.; Day, G. W.; Feldman, A.; Chai, B. H.; Kuzyk, M. G.; Holland, W. R.; Rapp, C. F.; Minden, M.; Moore, D. T.; others Handbook of optical materials; CRC press, 2018.
  4. [4](S4) Li, L.; Liu, X.; Li, Y.; Xu, Z.; Wu, Z.; Han, S.; Tao, K.; Hong, M.; Luo, J.; Sun, Z. Two-dimensional hybrid perovskite-type ferroelectric for highly polarization-sensitive shortwave photodetection. Journal of the American Chemical Society 2019, 141, 2623–2629.DOI

Paper details

vlnc:W_f6qfmesn
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 6

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents