Project Description

1.    Overview and Objectives

Nonlinear optical (NLO) materials provide versatile means for manipulating not only the amplitude and the phase, but also the frequency of coherent light via strong light-matter interaction.  Traditional oxide-based NLO crystals are available for use in the range of soft ultraviolet to near infrared (IR).  However, they become very inefficient as the wavelength approaches mid-IR and longer because of low nonlinearity and poor transparency, which is typical of oxide materials.  Although NLO crystals such as AgGaSe₂ and LiNbO₃ are commercially used at IR, the performance efficiency is low and the working range is also limited.  Several approaches have been employed for targeting improved IR NLO materials based on chalcogenides chemistry [1], rendering high optical nonlinearity.  However, these materials are plagued by propensity to optical damage.  Indeed, there exists a severe barrier in exploiting an NLO material essentially arising from the mutual exclusiveness of “optical nonlinearity” and “damage threshold”, which are two critical parameters affecting the nonlinear figure of merit.

For instance, AMQ₂ (A = Ag, Li; M = Ga, In; Q = S, Se) are noncentrosymmetric three-dimensional (3D) NLO chalcogenides, having nonzero second-order susceptibilities, c⁽²⁾.  Unfortunately, the practical use of these materials are limited for NLO applications either by a low c⁽²⁾ value or by a low damage threshold.  Namely, a large bandgap results in a high damage threshold at the expense of c⁽²⁾, whereas a small bandgap yields a high c⁽²⁾ value but with a low damage threshold (Fig. 1).  Some researchers describe this dilemma as a “balance” [2], and as such, simultaneously targeting a large optical nonlinearity and a high damage threshold within a single material remains an outstanding problem.  Nevertheless, next-generation IR NLO materials must overcome this limit in order to realize novel applications in the emerging age of IR science and technology.  High-performance IR NLO devices will enable high-density-information transfer in telecommunications, noninvasive imaging of biological structures and mechanisms, and remote sensor materials for environmental, military, and medical applications [3-6].

In order to address the challenge, we propose a unique program to discover novel two-dimensional (2D) hybrid halide perovskites that overcome the ultimate NLO limit through systematic synthesis and characterization of their structural and optical properties over a broad IR range (1 mm–25 mm and possibly further).  Our innovative claim is to explore the unusual impact of dimensional reduction on third-order susceptibility, c⁽³⁾, in this highly promising material class, as first demonstrated by our group [7].  In the 2D lead iodide perovskite series, the degree of quantum confinement can be tuned toward the simultaneous improvement of optical nonlinearity and damage threshold upon thinning down the perovskite layer number from the 3D perovskite (CH₃NH₃PbI₃: n = ¥) to n = 4, 3, 2, and 1: This generates layered van der Waals crystals repeating the 2D units shown in Fig. 2.  While this unusual nonlinearity is truly remarkable, it seemingly violates the Kramers-Kronig relation in the nonlinear regime [8].  The initial but presumably most critical step is to clarify the underlying mechanism for the anomaly unique to this 2D system.  Therefore, we will carry out comprehensive experiments on high-quality 2D perovskites prepared into single crystals to measure both real and imaginary parts of c⁽³⁾ as a function of wavelength and polarization by fully utilizing wavelength-dependent Z-scan nonlinear spectroscopy (WDZNS) [9].  As developed by the PI, WDZNS can characterize coexisting NLO effects arising from different nonlinear orders and their complicated interplay over a broad IR range.

Our next task is to investigate the feasibility of Pb-free perovskites prepared using Ge and Sn, which crystallize into noncentrosymmetric and centrosymmetric phases, thereby ideal for improving c⁽²⁾ nonlinearity and c⁽³⁾ nonlinearity, respectively, in the unique 2D landscape.  These perovskites are less explored due to the stability issue but eminently important to realize eco-friendly devices.  Since both Ge- and Sn-based perovskites are available in 3D [10,11], we assume that they can be engineered into the 2D series by using proper organic cations as the perovskite-layer spacer upon slow cooling [12] and/or antisolvent vapor-assisted crystallization (AVC) [13].  We will identify the optimized synthesis route for ensuring large single crystals of macroscopic size (> 10 mm³) from which we will demonstrate concept devices for high-efficiency/high-power wavelength conversion, self-focusing, and electro-optic modulation at IR.

Fig. 2: Schematics and photos of bulk (n = ¥) and layered (n = 1~4) perovskites: (BA)₂(MA)_n-1Pb_nI_3n+1 [BA = CH₃(CH₂)₃NH₃ and MA = CH₃NH₃].  Quantum confinement is evident from the color change.

The overall goal of the project is 1) to understand the impact of dimensional reduction on both c⁽²⁾ and c⁽³⁾ of the highly promising 2D hybrid perovskites and 2) to realize Pb-free devices for high-performance NLO applications working at IR by optimizing the perovskite configuration, which is predominantly determined by the organic cation size and the layer number.  Here we emphasize that high-quality, large-area single crystals will be pursued because thin films are unstable in the ambient condition.  Over the three years, we will test the central hypotheses and accomplish the objectives of this proposal by pursuing the above two aims, each of which includes several subtasks as detailed in Sec. 3.  During the project, the series of NLO data will be systematically analyzed to establish a rationale for structure-property correlations, specific to the hybrid perovskites for maximizing the overall nonlinear figure of merit.  Successful accomplishment of the project will lead to NLO perovskites by design with the added advantage of cost-effective fabrication.  Considering that NLO materials have been available from either solely organic or solely inorganic perspective, hybrid NLO perovskites can provide unique opportunities since the combination of their organic-inorganic properties can be further engineered.

2.    Description of Background, Previous Research, and Motivations

2.1  Basics of nonlinear optical (NLO) materials

Every NLO phenomenon occurs as a consequence of the change in optical properties of a host material under strong electromagnetic perturbation, which in turn induces the modification of the light field itself.  The corresponding NLO efficiency thus critically depends on the material properties that determine the frequency-dependent NLO coefficients, such as c⁽²⁾(w) and c⁽³⁾(w), which are complex quantities in general.  The broadband capability of the WDZNS technique [9] is ideal, as it can characterize both real and imaginary parts of the NLO coefficients as a function of frequency (or wavelength) which in turn determines the working range of the perovskites.

The second-order processes are important for generating a new frequency of radiation via coherent wave mixing in an NLO crystal lacking inversion symmetry (i.e., noncentrosymmetric).  For example, second harmonic generation (SHG) results in frequency doubling (w→2w), where an input wave generates a wave having twice the frequency (or half the wavelength).  The SHG efficiency is proportional to the absolute square of c⁽²⁾, where both real and imaginary parts of c⁽²⁾ contribute.  Most of all, potential c⁽²⁾ materials must ensure “phase matching” [1] since efficient frequency conversion requires coherent beam overlap between input (w) and SHG (2w) radiation throughout the NLO medium, i.e., the two beam should propagate with the same phase velocity.

The narrow-gap chalcogenide class lacking inversion symmetry has been the main paradigm for the discovery of new c⁽²⁾ materials simply driven by an empirical relation, c⁽²⁾ ∝ E_g^–3/2, where E_g is the bandgap [1].  However, the narrow bandgap severely limits their performance because of low damage thresholds that typically scale with E_g [14].  Recently, several wide-gap quaternary chalcogenides were synthesized to improve the damage threshold, but the resulting c⁽²⁾ values are not high as expected, again facing with the NLO dilemma between c⁽²⁾ and damage threshold.  A goal of the project is to utilize quantum confinement in Ge-based noncentrosymmetric perovskites that would simultaneously improve c⁽²⁾ and damage threshold in the device form, working at IR.

Without any symmetry argument, c⁽³⁾ is a complex quantity that every material possesses where its real and imaginary parts are related by the Kramers-Kronig relation [8].  The former is related to the nonlinear refractive index, n₂, and the latter to the two-photon absorption (2PA) coefficient, b, respectively.  Either n₂ or b can be utilized for specific purposes, but their coexistence in a single c⁽³⁾ material often causes undesirable effects that limit the performance.  For instance, a large n₂ value is required for self-focusing applications but inherently related large b promotes 2PA that eventually leads to optical damage.  This is the c⁽³⁾ version of the above-mentioned NLO dilemma.  However, our recent result of quantum confinement in the 2D perovskites seems to avoid this problem by only enhancing n₂, while leaving b unaffected [7].  In this project, we will investigate the essential mechanism for this surprising anomaly and further maximize the effect in large-area single crystals toward high-efficiency/high-power c⁽³⁾ applications.

2.2  Basics of hybrid halide perovskites and synthesis of single crystals

Over the past few years, the research community has shown considerable interest in perovskites triggered by a remarkable breakthrough in the solar power conversion efficiency (PCE).   Especially, 3D organic-inorganic hybrid halide perovskites, ABX₃, (A⁺ = organic cation, B²⁺ = Ge²⁺, Sn²⁺, or Pb²⁺, and X^– = halide anion; Fig. 3a) have outstanding potential for the photovoltaic technology due to the combination of useful hybrid properties such as high optical absorption, efficient charge transport, plasticity, and cost-effectiveness for fabrication [15].  While the device architecture is a critical factor in terms of technology evolution, the basic understanding of light-matter interaction in the perovskites is unarguably important for furthering the PCE as well as discovering other optoelectronic properties such as radiation detection, thermoelectric, lasing, light-emitting diodes, and nonlinear optics (Fig. 3b), where the latter is the main focus of this proposal.

Fig. 3: (a) 3D perovskite structure.  (b) Multi-functionality of hybrid halide perovskites.

Fig. 4: Methods for the growth of hybrid perovskite single crystals achieved in our lab.

These breakthroughs in the numerous application areas initiated efforts toward the synthesis of hybrid halide perovskites prepared into single crystals.  A single crystal is an ideal material type not only to fabricate high-performance devices, but also to understand intrinsic material properties because it is high quality with no grain boundary, compared with solution-processed films that typically contain a lot of grain boundaries and defects.  Large perovskite single crystals are especially useful for NLO applications because the interaction volume for nonlinear light-matter interaction is much larger compared with thin films.

The traditional synthesis method for ionic compound crystals is slow cooling from saturated precursor solution.  When applied to hybrid halide perovskites, this method has drawbacks such as long crystallization time and high trap density on the surface because of different solubility between the organic cation (A⁺) and the inorganic metal cation (B²⁺).  Crystallization can be expedited by employing the inverse temperature crystallization (ITC) method [16] and the crystal quality can be improved using antisolvent vapor-assisted crystallization (AVC) [13] (see Fig 4 and Table 1 for comparison of the three growth methods).

Table 1: Pros and cons of the crystallization methods

Fig. 5: (a) Trigonally distorted crystal structure of MAGeI₃ with no center of inversion.  Wavelength-dependent SHG responses at IR (red peaks) from (b) CsGeI₃ and (c) MAGeI₃, respectively.  The black trace corresponds to the linear absorption data, indicative of the bandgap.

Based on our preliminary results, we found that the ITC method is not ideal for the growth of 2D perovskites since fast crystallization does not commensurate with van der Waals bonding present in the 2D layered structures.  In fact, a recent study indicates that the AVC method can yield 2D perovskite single crystals with a few millimeter size [17].  To the best of our knowledge, however, layer-number control (n = 1, 2, 3…) in large single crystals has not been realized yet.  In this project, we will establish the best synthesis route for preparing high-quality, large-area 2D perovskite single crystals with active control on the layer number in order to study confinement-induced NLO effects as a function of n (layer number).

2.3  Hybrid halide perovskites from the perspective of nonlinear optics

Nonlinear optical (NLO) effects in the perovskites are relatively less explored but potentially important, and they have indeed gained renewed interest where we contributed to the field at the top notch as proven by our high-impact publication record [7,10,18,19].  For instance, we showed that the Ge-based hybrid iodide perovskite, MAGeI₃ (MA⁺ = CH₃NH₃⁺), exhibits strong “phase-matched” second harmonic generation (SHG) [10] owing to its noncentrosymmetric crystal structure (Fig. 5a).  Most remarkably, Figs. 5b and 5c show that the SHG efficiency of MAGeI₃ is notably higher than that of the all-inorganic counterpart, CsGeI₃, because of the additional polar character in the MA⁺ cation oriented within the perovskite structure.

Considering that the majority of halide perovskites adopt centrosymmetric crystal structures, noncentrosymmetric MAGeI₃ is truly unique as it can be utilized for c⁽²⁾ applications.  In this project, the 2D layered series (n = 1–4) derivatized from MAGeI₃ will be synthesized using a proper organic cation as the perovskite spacer to confirm the main hypothesis, i.e., the concomitant increase of c⁽²⁾ and damage threshold by lowering n.  One aim is to significantly boost the damage threshold via large bandgap blueshift particularly in the lowest n member such that the beam-mixing efficiency at IR is entirely free of two-photon absorption (2PA) as detailed in Sec. 3.

Fig. 6: (a) Schematic of two-photon-induced PL emission (green) by IR excitation in the 3D perovskite.  (b) 2PA coefficient vs bandgap of single crystalline MAPbX₃, fit by the two-band model [8] (red line).

We also reported on c⁽³⁾ nonlinearity of 3D perovskite single crystals (MAPbX₃) by measuring the 2PA coefficient, b µ Im[c⁽³⁾], within the bandgap-engineering range achieved by varying the halide component, X = I→Br→Cl [18].  The relative b values of MAPbX₃ were crosschecked with the relative brightness of 2PA-induced PL intensity (Fig. 6a).  Being essentially 3D materials, however, the absolute magnitude of b and its bandgap dependence (~ E_g^–3) are well explained by a conventional two-band model (Fig. 6b).  Although a very large c⁽³⁾ nonlinearity was reported in 3D perovskites [20], it is significant only when the input photon energy is resonant with excitonic or sub-gap state, which is certainly not the overall enhancement effect.  After all, any novel NLO effect is not anticipated from the 3D perovskite system.

In contrary, we showed that the nonlinear refractive index, n₂ µ Re[c⁽³⁾], increases as the layer number of the 3D perovskite is progressively reduced toward the 2D series (n = 1–4) [7].  This arises most likely from the enhancement of dipole matrix elements associated with c⁽³⁾ under quantum confinement as indicated by the dashed line in Fig. 7a.  We emphasize that dimensional reduction significantly scales up n₂ over the entire frequency range throughout mid-IR as confirmed by our own WDZNS technique.  However, it turns out that the corresponding Kramers-Kronig conjugate, b µ Im[c⁽³⁾], is not affected by the confinement effect (Fig. 7b).  This is a very intriguing observation because 2D confinement typically affects both n₂ and b at the same time [8].  Since the 2D perovskites were available as micro-crystallites at the time, we were not able to conduct a more comprehensive study to understand the origin of “the selectiveness”.  More importantly, our experimental result of the selective enhancement of n₂ without any increase in b is quite exceptional, where the latter improves the damage threshold.  Apparently, simultaneous increase in n₂ and damage threshold resolves the long-standing NLO dilemma.  The goal of the project is to clarify this anomalous effect using large-area single crystals and to maximize the effect for the realization of Sn-based c⁽³⁾ devices efficiently working at the broad IR range.

Fig. 7: Impact of dimensional reduction on (a) n₂ and (b) b in the 2D perovskite series.  Inset: micro-crystallites encapsulated in capillary tubes: bandgap blueshift by confinement is evident from the color change of the samples.  The bold red line is the case without any quantum enhancement effect.

3.    Description of Proposed Research

3.1  Synthesis of 2D perovskite single crystals (Throughout 3 years)

We have excellent crystal growth and characterization facilities where we can grow large perovskite single crystals (Fig. 4).  To move from 3D to 2D perovskites, the small MA⁺ cation should be periodically replaced by a much bulkier organic cation, thus confining the perovskite in the layered configuration.  As a follow up of our work on Pb-based 2D perovskites, we will use butylammonium (BA⁺) as a perovskite spacer that successfully generated the 2D series with n varying from 4 to 1 (Fig. 2).  However, we anticipate that the perovskites may have to be modified in composition by using a bulkier spacer such as benzylammonium (BeA⁺) or phenethylammonium (PEA⁺) cation to understand property trends and for further NLO property enhancements.  In this project, 2D bromide perovskite series will be pursued because they typically yields better optical properties and structural stability [19].  Moreover, using a smaller Br^– anion is highly desirable for boosting the damage threshold because it yields wide-gap perovskites compared with the iodide counterparts.

Pb-based 2D perovskites: In order to investigate the unusual 2D NLO effects, we will first focus on the controlled synthesis of the Pb-based 2D series, as they are much more stable against oxidation compared with Sn- and Ge-based ones.  In fact, large single crystals of the n = 1 member can be synthesized using the AVC method at room temperature [17].  In order to achieve other members up to n = 4, we will prepare perovskite precursor solution by dissolving the starting materials in accordance with the stoichiometry, (BA)₂(MA)_n-1Pb_nI_3n+1, (n = 1–4).  The precursor solution will be purified and placed in a glass container where diffusion of antisolvent initiates crystallization (Fig. 8a).  By carefully controlling vapor-flow rate and solution concentration, we will grow millimeter-sized 2D single crystals with the layer number varying from n = 1 to 4.

Fig. 8: Proposed growth methods for (a) Pb-containing and (b) Pb-free 2D perovskite single crystals.

Pb-free 2D perovskites: It is rather tricky to synthesize Ge- and Sn-based perovskites as these inorganic cations are prone to oxidation in the Ge⁴⁺ or Sn⁴⁺ state.  For the Pb-free compounds, we will grow the proposed 2D perovskite single crystals by combining advantages of slow cooling and AVC methods (Fig. 8b).  Our novel idea is to first obtain seed crystals through slow cooling in the Ar environment, and then the seed-containing solution will be subsequently crystallized with the AVC method.  Here, in order to avoid oxidation and to expedite the formation of seed crystals, strong acidic reduction agent such as hypophosphorous acid will be mixed in the solution so that inorganic cations exist in the Ge²⁺ or Sn²⁺ state during crystallization.  Various growth parameters such as starting temperature and cooling rate for slow cooling and the above-mentioned AVC parameters will be systematically optimized for the growth of large perovskite single crystals with stoichiometry control.

Our aim is to establish a reproducible synthesis method to yield layered single crystals available with the 2D series members of n = 1–4, whose size is larger than 10 mm³.  High-quality specimens will be screened by basic characterization that includes X-ray, Raman, PL, and FTIR spectroscopy, together with AFM and SEM imaging.  We will measure both linear refractive index and extinction coefficient, as they are required for the precise assessment of NLO parameters.  The IR transmittance will be assessed using the FTIR spectra, thereby identifying the working range of the NLO devices derived from the 2D perovskites.

3.2  Impact of dimensional reduction on optical nonlinearity (Year 1)

Pb-based 2D single crystals with smooth surface will be prepared to have the same thickness to consistently study the anomalous c⁽³⁾ effects in the 2D series (n = 1–4).  We will employ the WDZNS technique to precisely determine the absolute values of n₂ and b of each 2D member using closed- and open-aperture schemes, respectively.  Note that the broad wavelength tunability of our technique will yield a series of (n₂, b) values as a function of band dispersion, x, which is defined by the input photon energy per the bandgap energy [9].  The extensive experimental data set of frequency-dependent n₂(x) and b(x) will be plotted and directly compared with a simple two-band model (Fig. 9a).  Because of complications arising from the presence of sub-gap states associated with excitons, defects, and vibrational modes of organic cations [7], however, the experimental data may significantly deviate from the theory.

We will identify such resonance regimes by directly comparing with the linear absorption spectra throughout IR (Fig. 9b).  Here, the key issue is to check whether sub-gap resonance affects n₂ and b consistently or selectively and to check whether such selectiveness is localized or extends to other wavelengths if the anomaly is indeed observed.  This systematic approach will lead us to understand any possible origin for the violation of the Kramers-Kronig relation, which may arise from breakdown of analyticity and/or time-reversal symmetry of the nonlinear response depending on the type of the sub-gap state.

On the other hand, our recent study on 3D perovskite single crystals indicated that b can have a significant polarization dependence (Fig. 9c) due to the two-photon selection rules reflecting the underlying crystal symmetry [18,19].  Although the Kramers-Kronig conjugates typically follow the same polarization dependence in 3D materials, it may have a different story for the 2D perovskite series.  In fact, it is theoretically predicted that quantum confinement can increase or even decrease b depending on the polarization direction of input light with respect to the confinement direction of the 2D layer, whereas n₂ typically increases in semiconducting quantum structures.  Therefore, we will also measure the polarization dependence of n₂ and b in order to examine possible anisotropic effects, causing the anomalous NLO behavior.

Fig. 9: (a) Theoretical dispersion of n₂ and b based on the two-band model.  (b) Absorption spectra of the Pb-based 2D perovskite series [7], showing band edges in the visible and various sub-gap transitions in the IR.  (c) Polarization dependence of b in MAPbBr₃ single crystal.

Fig. 10: (a) Intensity-dependent SHG from 3D MAGeI₃.  The black trace corresponds to the ideal SHG.  Schematics for (b) c⁽²⁾ application (color-tunable DFG setup) and (c) c⁽³⁾ application (E-O setup).

Our aim is to identify the nature of anomalous NLO effects arising from quantum confinement in the 2D series and to establish a correlation chart between the perovskite-layer number (n) and the enhancement factor for n₂.  This approach will provide crucial information about the best perovskite configuration for the preparation of actual NLO devices in terms of the maximized nonlinear figure of merit and structural stability.

3.3  Strong second-order effects in Ge-based 2D perovskites and applications (Year 2)

Although 3D MAGeI₃ (n = ¥) has very promising second-order NLO properties such as a large c⁽²⁾ value (~161 pm/V) and phase matchability at IR [10], the low damage threshold (<< 1 GW/cm²) greatly limits its potential for high-power applications.  This problem is well demonstrated by power-dependent SHG (Fig. 10a), where the actual SHG counts are seriously lower than the ideal case (black curve) at high intensities.  We will resolve this critical problem by applying the concept verified from Sec. 3.2 to the Ge-based perovskites.  Pushing the perovskite bandgap beyond the two-photon energy of typical laser lines (2.3 eV for Nd: YAG laser and 3.1 eV for Ti-Sapp laser) is extremely helpful for boosting the damage threshold, since 2PA-induced damage is entirely absent.  We will achieve this condition by exploiting the lower n members.

The absolute c⁽²⁾ values of the 2D single crystals will be assessed based on the spectral Maker fringe method within the WDZNS scheme [9].  We will confirm the frequency-dependent phase-matching behavior by broadband polarization-dependent SHG measurement throughout the IR.  In order to assess the precise value of damage threshold, each 2D crystal will be exposed to the femtosecond pulse train of a high repetition rate while damage on the crystal surface is in-situ monitored.  We will determine the best perovskite configuration by simultaneously maximizing c⁽²⁾ and damage threshold for the next step described below.

For the actual generation and detection of picosecond IR radiation, we will build a pump-probe-type two-beam difference frequency generation (DFG) setup (Fig. 10b).  The pump pulse (arrows) is split into two paths by a beam splitter (bs).  The pump beam via a delay stage is focused onto the 2D perovskite crystal and mixed with the tunable idler beam (red arrow) from an optical parametric oscillator available at our lab.  The generated IR beam (blue arrows) is then focused onto a commercially available DAST film and mixed with the probe beam in a co-propagating geometry.  This will gate a detector via the polarization change of the probe beam by electro-optic (E-O) modulation in the presence of the generated IR through a quarter-wave (l/4) plate.  The response is proportional to the field strength of IR light generated, which indeed corresponds to the nonlinear figure of merit for DFG.  The working efficiency of the 2D perovskite crystal will be directly compared with that of AgGaSe₂ crystal, which is the benchmark IR NLO material.  The proposed method will generate spectrally pure and continuously tunable IR radiation owing to the variable idler wavelength.  We will determine the working range of the 2D perovskite crystal by varying the idler wavelength and by tuning the corresponding phase-matching angle.

Our aim is to realize a high-performance c⁽²⁾ device for IR generation over the broad range of 1 mm–25 mm, whose nonlinear figure of merit is 100 times higher than that of a commercially available AgGaSe₂ crystal.  We will also build a one-beam scheme using a femtosecond laser for THz generation that relies on optical rectification within a single broad spectral band.  We will therefore determine the low-frequency limit achievable from our novel Ge-based 2D perovskite device.  In order to confirm the stability of the perovskite against oxidation, we will measure the performance efficiency at the ambient over an extended time.  If problematic, we will employ proper capping where the capping material is transparent at IR.

3.4  Strong third-order effects in Sn-based 2D perovskites and applications (Year 3)

Although there exists slight difference in the crystal structure between MASnI₃ (a phase) and MAPbI₃ (b phase) at room temperature [11], they are all centrosymmetric from the NLO perspective.  Therefore, much of the essential NLO properties of Pb-based perovskites discovered from Sec. 3.2 are likely applicable to the Sn-based perovskites.  In fact, the success of this task is more contingent on the crystal quality of Sn-based perovskites achieved by our novel synthesis method, which we will optimize throughout the project years.  We assume that large-area Sn-based 2D perovskites are available and propose a particular example in order to demonstrate the high-efficiency/high-power c⁽³⁾ applications.  As also mentioned in Sec. 3.3, capping can improve the durability of the Sn-based perovskites.

Many of state of the art NLO devices utilize c⁽³⁾ nonlinearity based on the optical Kerr effect (self-focusing), which is directly related to n₂.  Here we propose a pump-probe-type electro-optic (E-O) modulator derived from the best Sn-series that efficiently changes the polarization state of IR light.  Considering that the currently available E-O modulators poorly work at IR, the successful accomplishment of the project will enable all-optical switching, thereby significantly advancing the IR technology in terms of information science.

For example, Fig. 10c illustrates the schematic for the proposed E-O setup where Kerr-effect-induced birefringence would rotate the polarization of the probe signal in the presence of a strong pump pulse.  Active switching behavior will be examined as we vary the pump intensity with an analyzer in the transmission side set to a proper angle.  Since the origin of c⁽³⁾ nonlinearity is basically electronic, we expect that the response time may directly depend on the pulse width.  Therefore, we will use a femtosecond laser for ensuring a faster response time and a stronger Kerr effect as well.

Our aim is to realize a high-performance E-O modulator prepared with eco-friendly 2D perovskite crystals working at IR, where we will determine the transparency window from the FTIR data.  We will target a high E-O coefficient, as comparable to the commercially available LiNbO₃-based E-O modulator, but with a much better switching behavior at IR.  As the proposed program matures, we will gradually seek broader collaboration for the investigation of novel NLO phenomena uniquely offered by the 2D halide perovskite system and stabilization of the Pb-free NLO devices.

4.    Summary of Project, Significance, and Broader Impacts

The proposed research is unique because it involves the investigation of an entirely new concept for maximizing the nonlinear figure of merit by defying the long-standing NLO dilemma in the emerging 2D hybrid system, which is very promising for high-performance NLO devices working over the broad IR.  The program consists of both fundamental and technological aspects of hybrid perovskite materials, which will contribute to a number of other areas across science and technology as already described in Sec. 2.2.  We aim to generate at least 5 SCI publications and 2 patents from the project.  The intellectual challenge revolves around how to maximize the NLO performance by systematically studying fundamental confinement effects in the 2D series.

The societal benefits of this activity are tangible and significant as we ultimately aim at eco-friendly, high-performance devices.  Successful achievement of the goals addressed in the project will serve as a foundation from which to develop future NLO devices and other optoelectronic components efficiently working at IR.  Therefore, the program will broadly benefit military, medical, commercial, and industrial sectors.

The synthesis methods to be developed in this project can be readily extended to exploration of advanced hybrid devices as the 2D layered perovskites can be exfoliated to combine with other 2D materials such as graphene, h-BN, black phosphorous, and topological insulators.  These may generate novel 2D heterostructures and scaffolds where new science and technological concepts likely emerge.  The long-term significance of the project is therefore clear and may resolve national grand challenges especially related to environment/cost issues in terms of highly efficient nonlinear light-matter interaction by using cost-effective, Pb-free materials, thereby fulfilling the mission of Samsung.

Biographical Sketch

Our research team will be composed of two professors, one postdoc, and four graduate students with complementary skills to carry out the project.  The postdoc will conduct precision experiments and analyses of the 2D NLO perovskites by working with two students at Sogang under PI’s supervision.  Under the guidance of the Co-PI, two students from Sungkyunkwan will participate in synthesizing and characterizing the 2D NLO perovskites.  Synergistic feedback loops between the two groups will be assured to pursue the grand goal of the project during the project years.

I. Principal Investigator

□ Name: Joon Ik Jang

□ Current Position / Title and Affiliation:

   Assistant Professor, Department of Physics, Sogang University (03/2017 ~ present)

□ Education:

□ Experience (starting with latest experience):

1. Assistant Professor (09/2010 ~ 02/2017)

Department of Physics, State University of New York (SUNY) at Binghamton, USA

Description of experience: (Teaching & Research) The PI developed his own experimental technique (wavelength-dependent Z-scan nonlinear spectroscopy) to investigate a broad set of optical properties of novel semiconducting materials and structures.  These include NLO chalcogenides, halide perovskites, multiferroics, and transition metal dichalcogenides.  The number of SCI publications during this period is 35.

2. Postdoctoral Research Associate (10/2005 ~ 08/2010)

Department of Physics and Astronomy, Northwestern University, USA

Description of experience: (Research) The PI studied fundamental exciton physics with a special emphasis on polaritons in a dipole-forbidden semiconductor, Cu₂O, towards BEC.  The PI also initiated a collaboration with Prof. M. G. Kanatzidis to investigate strong NLO properties of alkali metal chalcogenides.  The number of SCI publication during this period is 33.

□ Personal Statement:

In a research career spanning 20 years, I have studied electronic and optical properties of various semiconductors and their low-dimensional structures that host rich diversity of interesting physics and novel applications.  My current research interest encompasses key light-matter interactions in emerging systems such as 2D transitional metal dichalcogenides, low-dimensional NLO materials, and hybrid halide perovskites.  In particular, I have been at the frontier in exploring these novel materials from a unique perspective of nonlinear light-matter interactions.  I will direct the proposed program where the Co-PI’s group provides novel 2D NLO perovskites, all specifically custom-designed by my own request: I therefore actively participate in synthesis and design efforts.