Ferroelectric materials hold significant potential for ultralow-energy-consuming oxide electronics and have recently been pointed out as a suitable platform for next-generation neuromorphic and reservoir computing schemes. We provide a brief overview of the progress in engineering electric dipole textures of epitaxial ferroelectric oxide thin films, with an emphasis on the technologically relevant ultrathin regime. In epitaxial films that are only a few unit-cells thick, surface chemistry and interfacial electrostatics are commonly considered limiting factors in ferroelectric device integration, as they may suppress the net ferroelectric behavior. Here, we highlight how nanoscale lattice chemistry control, including off-stoichiometry and layer polarization in oxides, can, in fact, emerge as powerful tools for engineering ferroelectricity in thin films. We also discuss the potential of such an approach in the context of recent trends in the field, such as the design of ferroelectric freestanding membranes and the optical control of polarization in thin films. Hence, with our Perspective article, we aim to provide key insights on the use of lattice chemistry for ferroelectricity engineering in thin films to facilitate exciting developments in ferroelectric-based applications.
I. INTRODUCTION
Ferroelectric materials exhibit a nonvolatile electrical polarization, which can be manipulated with an external electric field. The ultralow-energy-consuming, voltage-controllable functionality of ferroelectrics motivates their expanded insertion into various applications,1,2 such as transducers,3 high-performance capacitor,4 and in thin-film form,5 memory elements,6 and electro-optical devices.7–9 The realization of nanoscale ferroelectrics became especially relevant in the search for alternatives to complementary metal-oxide-semiconductor (CMOS)-based electronics.10–12 The current transistor architecture miniaturization is reaching its fundamental limits, and ferroelectric materials appear as promising candidates for enabling the next generation of computing paradigms.13 In particular, the complex and non-linear response of ferroelectric systems to different types of stimuli makes these materials suitable platforms for reservoir computing, the most recently proposed framework for computations.14
The widespread integration of ferroelectric thin films into application-relevant devices has, however, remained elusive due to the detrimental size effects on the final polarization of the ferroelectric materials.15,16 The bound-charge accumulation accompanying the emergence of spontaneous polarization results in the onset of an electric field that is oppositely oriented to the polarization.17–19 In ultrathin films, i.e., films of only a few unit cells in thickness, the contribution of this so-called “depolarizing field”20 dominates and suppresses the ferroelectric behavior. In most-debated perovskite ferroelectric systems, the ferroelectric polarization vanishes in films below a critical thickness of 4–5 unit cells.16,21–23 Hence, modern research has focused on understanding and maintaining the ferroelectric functionality at the nanoscale. The polarization discontinuity at the surface and the corresponding unfavorable electrostatic environment for a robust ferroelectric response, initially designated as an insurmountable limitation, have since emerged as an exciting opportunity for the design of ferroelectric and polar textures with technological relevance.24
Following the advancement of the processing techniques and deposition monitoring tools for the growth of complex oxide thin films, the engineering of elastic and electrostatic boundary conditions at atomically sharp interfaces in ferroelectric-based heterostructures has enabled the stabilization of ferroelectricity from the very first unit cell deposited.25–27 Furthermore, the ability to tune the depolarizing field contribution by controlling the physics of the interfaces in advanced oxide heterostructures unleashed hitherto unforeseen capacity to design complex polar textures with promising functionalities.24,28–33 Some of the most conventional ways to tune the ferroelectric thin films by using interface physics are listed in Fig. 1.
Beyond the control of the physical properties of the interfaces34 in the vicinity of the ferroelectric materials, chemistry engineering of the film surfaces and interfaces emerges as an alternative tool for polarization control in thin films. The chemical manipulation of the electrostatic boundary conditions may, for instance, consist of the controlled distribution of charged defects in the films, the design of buffer layers with atomically sharp surface termination functioning as a charged plane, or the engineering of charged surface off-stoichiometric layers (Fig. 1). Within this Perspective article, we will cover the state-of-the-art use of surfaces and interfaces with engineered chemistry to tune ferroelectricity in perovskite oxide thin films. For non-perovskite ferroelectrics, such as hafnia, we direct the reader to the other sources.35–39 In our discussion, we will highlight that, in contrast to the approach consisting of solely engineering the physical properties of the ferroelectric interfaces, the control of lattice chemistry and charged-defect distribution offers exciting avenues for post-growth reconfiguration of polarization states and beyond-binary responses in the films at will. This may be key in the insertion of ferroelectric materials into catalytic applications40,41 and into the next generation of beyond-CMOS electronics.10
II. ENGINEERING FERROELECTRIC HETEROSTRUCTURES THROUGH LATTICE CHEMISTRY CONTROL
The depolarizing field tuning approach for the design of well-defined electric dipole configurations involves adjusting the physical properties of interfaces in ferroelectric heterostructures. In this section, we will highlight an emerging alternative, namely, lattice chemistry engineering. In particular, we will describe how controlling the chemistry at oxide interfaces opens new avenues for engineering the boundary conditions of ferroelectric thin films. In a first step, we will consider the concept of layer-polarization, first coined in the early 2000s42–44 and has been recently subjected to a renewed interest in ferroelectric thin films, triggered by the work of Spaldin et al.45 Next, we will discuss the most recent results on engineering the off-stoichiometry to achieve controlled ferroelectric polarization states in thin films. Finally, we will explore the consequences of the combined effects, emphasizing how the interplay of off-stoichiometry and layer polarization can induce robust ferroelectric behavior in the ultrathin regime.
A. Approaches using interfacial layer charges
The progress in thin film design capacity and the increasing use of various deposition monitoring tools have enabled the growth of films with well-defined atomic termination.73,74 In ferroelectric oxide thin-film-based heterostructures, this brings new degrees of freedom for the engineering of electric dipole textures. The ionic nature of atomic bonds in ceramic crystal structures may lead to a finite electronic charge density along specific crystal planes.44 This so-called “layer polarization”42,43,45,75 is best visualized using the prototypical ABO3 perovskite structure as an example, where the A cations occupy the corners and B cations occupy the center of the idealized perovskite cubic unit cell. Along the 100 directions, the lattice consists of a stacking of AO and BO2 atomic planes. Thus, when the A- and B-sites are occupied by cations with valence states deviating from the and , i.e., the II-IV configuration, a net charge density exists at the {001} planes. Among those materials, oxides with III-III configuration include members of ferrites (AFeO3), aluminates (AAlO3), gallates (AGaO3), manganites (AMnO3), nickelates (ANiO3), and vanadates (AVO3), and I–V configuration can exist in certain members of vanadates (AVO3), niobates (ANbO3), and tantalates (ATaO3), depending on the oxidation states of the cations. The stacking of oppositely charged planes within these crystals induces a macroscopic polarization throughout the structure. While being compensated in bulk materials, the layer polarization results in surface charges in thin films due to the truncated nature of their crystal structures.44,45,75–78 The layer polarization direction is set by the atomic plane sequencing and cannot be reversed upon external stimuli, in strong contrast to the spontaneous ferroelectric polarization. It enables, however, intriguing physics and charge compensation mechanisms at oxide interfaces,45 as it has been elegantly exemplified with the notorious polar catastrophe at the interface between the LaAlO3 thin film (III–III) and SrTiO3 substrate (II–IV), where a two-dimensional electron gas (2DEG) forms.79,80 This discovery has catalyzed a significant interest in researching thin film systems with charged layers.81
In this section, we will discuss experimental studies employing a layer-polarization-based approach to engineer polarization states in ferroelectric thin films. For comprehensive coverage of theoretical and phenomenological aspects, we direct readers to other sources.45,75 Note that a net ionic charge density may form not only across {00l}-planes but also various other crystal planes.44 For the sake of simplicity and relevance to thin film growth, we concentrate on (001)-oriented ABO3 perovskite oxides, the most common thin film growth orientation. In ferroelectric thin films, the sign of the bound charges, which accumulate at the surfaces, is determined by the direction of the macroscopic spontaneous ferroelectric polarization. Hence, in ferroelectric thin films and heterostructures of non-II–IV perovskites, in which the direction of layer polarization is set by the AO or BO2 termination, the interplay between the layer and ferroelectric polarization may enable the control of the final ferroelectric polarization state. This interaction can occur on two levels: at the interfaces between ferroelectric thin films and a buffer exhibiting a layer polarization (A.1) or intrinsically within the ferroelectric thin film itself (A.2).
1. Interface control of bulk ferroelectric polarization
The net charge at the surface of a substrate or buffer layer exhibiting a layer polarization can be employed to alter the electrostatic boundary conditions for the ferroelectric material grown on top. For instance, changing the atomic termination of the buffer layer between AO and BO2 affects the sign of the charge accumulation at the interface and, hence, drives the spontaneous polarization direction in the ferroelectric constituent. The most relevant example involves the use of (La,Sr)O- and MnO2-terminated La0.7 MnO3 (LSMO) buffer. Here, the (MnO2) and (La,SrO) result in negative and positive surface charge accumulation, respectively.82 This dictates the favorable sign for the bound charge accumulation at the interface with the ferroelectric constituent, hence the out-of-plane component of the ferroelectric polarization. The control of ferroelectric polarization direction by using the layer charges at the interface has been achieved so far in various ferroelectric model systems such as PbTiO3 (PTO),83,84 PbZr0.2Ti0.8O3 (PZT),82,83,85 and BaTiO3 (BTO).86 This deterministic control on the pristine polarization state enables the design of films with well-defined properties and surface charge, which is of high relevance for catalysis functionality where the absence of a top electrode prevents poling. It may as well facilitate reliable ferroelectric device operation, by eliminating pre-poling steps. Such an interface-termination-driven polarization control is not restricted to the use of LSMO buffer and can be achieved, in principle, with any non-II-IV perovskite oxide that is epitaxially compatible. For convenience, we suggest a non-exhaustive list of relevant buffer and substrate material systems exhibiting layer polarization and present their effective (pseudo)cubic lattice parameters in Fig. 2.
2. Coexistence of ferroelectric and layer polarization
Having discussed the control of the net out-of-plane polarization component direction in ferroelectric thin films through the manipulation of the atomic termination of buffers exhibiting charged planes, we now move onto systems, in which the ferroelectric layer itself consists of charged layers. A very good example of a ferroelectric with layer polarization is BiFeO3 (BFO), the most debated room-temperature ferroelectric, multiferroic material.87–89 The intrinsic layer polarization in BFO enables film growth with a well-defined polarization state without the need for a buffer exhibiting charged planes. Here, the buffer surface termination solely sets the plane sequence, which then ultimately dictates the layer polarization direction in BFO. The latter one finally orients the out-of-plane component of the ferroelectric polarization.22,45,82,90 Strikingly, in BFO, the magnitude of (001)-surface-charge induced by the (III-III) charged plane stacking equals the one of bound charges resulting from the ferroelectric polarization.45,90 This interplay enables electrically compensated surfaces in certain ferroelectric polarization directions. Such a control can be, in principle, applied to many other perovskite ferroelectric materials with intrinsic layer charges such as KNbO391 and LiNbO3, as listed in Fig. 2. Finally, the impact of the layer polarization on the improper ferroelectricity11 in hexagonal manganites92–94 thin films has also been recently reported.95
Hence, there is a wide range of available materials with layer charges. They may be ferroelectric, metallic, or insulating (see Fig. 2), bringing up an immense potential for the design of novel functionalities. The control of termination96 at the interface of materials exhibiting layer polarization may enable the creation of polar discontinuities and charge compensation mechanisms involving conducting interfaces.97,98 It may even trigger the onset of ferroelectricity in intrinsically non-ferroelectric materials in the bulk form, as predicted in the case of KTaO3 by Gattinoni et al.99 Thus, tuning the surface termination of oxide interface in epitaxial thin film design adds a degree of freedom to the already existing epitaxial-strain-enabled polarization orientation control techniques.100,101 Given the widespread use of etching or thermal treatment processes to achieve single-terminated substrates,96,102 we anticipate an increasing number of studies to take such interfacial chemistry considerations into account as it may reveal instrumental in setting well-defined electrostatic boundary conditions for the design of complex polar textures in ferroelectric oxide multilayers.
B. Approaches based on off-stoichiometry
Having highlighted the role of interface chemistry for the deterministic design of a polarization state, we now move onto the interplay of ferroelectric polarization and defect chemistry103 in thin films. Off-stoichiometry can be present in many different forms. Here, we restrict ourselves to oxygen (Sec. II B 1) and cation (Sec. II B 2) off-stoichiometry gradients in the bulk of the films as well as adatoms and parasitic phases on the surface of thin films. In this section, we will scrutinize the use of such off-stoichiometry for engineering and controlling the ferroelectric properties of oxide heterostructures.
1. Oxygen off-stoichiometry
Being the most prevalent type of point-charge defects in single-crystalline oxide films,104 oxygen vacancies contribute to the thin-film electrostatics immensely, and their interplay with the spontaneous polarization in ferroelectric thin films can have a strong influence on the polarization magnitude, direction, and domain configuration. They also play a major role in switching dynamics, leakage current,105 photoresponse,106,107 and ferroelectric fatigue performance.108 The density of oxygen vacancies can be tuned by changing the growth conditions and applying post-growth processes such as thermal annealing.109 For detailed reviews on oxygen vacancy engineering, we refer the reader to the dedicated literature.103,110–112 Here, we focus on their use to manipulate the polarization state in epitaxial layers. As epitaxial films grow most commonly in the ferroelectric phase, the control of the growth environment, in particular the oxygen partial pressure and temperature, which affect the oxygen vacancy formation and migration, is instrumental in setting the final polarization state in the films. Experimental reports revealed the deterministic control of the out-of-plane polarization direction in thin films using solely the oxygen partial pressure as a control parameter in BTO,86,113,114 PTO,115,116 and PZT85 thin films. Due to their mobility and ability to diffuse throughout the volume of the film, oxygen vacancies may not only control the polarization direction but also stabilize energetically unfavorable domain configurations and topological polar objects. They are key in the stabilization of charged-domain-walls117 and polar topological configurations with swirling electric dipoles.118
2. Cation off-stoichiometry
The energy minimization and the charge neutrality in oxide systems lead to an intricate correlation between the above-mentioned oxygen vacancy formation and cation off-stoichiometry. For simplicity, here, we treat defect formations and their respective effects on the polarization state separately. The most common cause of cation off-stoichiometry in ABO3 ferroelectric thin films is the inherently high volatility of A-site elements such as Pb, K, and Bi. In most oxide thin film processing, an A-cation-excess material is provided during the growth to compensate for the re-evaporation of the material. This may lead to an A-cation gradient over the thickness of the film or an excess of A-cation at the surface upon growth completion. In PTO and PZT, a positively charged (Pb) -excess surface layer has been identified.83,84 For Bi-containing compounds such as BFO, an outgrown Bi2O3119 surface layer can appear on the surface of the film.120–122 Although it has been perceived negatively in the past, here, we highlight that the off-stoichiometry in thin films is now emerging as a potential tool for engineering polarization. Such charged defect accumulation at the surface can be used as a tool to design electric dipole textures. For instance, the positively charged Pb2+-excess surface layer preferentially screens negative bound charges and, therefore, can be used to drive a downwards-oriented polarization state in the films from the top surface.83 In the case of Bi2O3-layer at the surface of BFO, the highly localized negative charge density in the oxygen ionic plane121,123 locally polarizes the film along the out-of-plane axis and stabilizes a super-tetragonal phase.122,124,125
C. Interplay of charged layers and off-stoichiometry in the same system
As mentioned earlier, one can control the polarization from the bottom interface by growing a buffer layer with a specific charged layer termination. Strikingly, the additional tuning of the surface off-stoichiometry may enable the control of the polarization state from the top interface of the film and, hence, opens the way for manipulating the delicate balance between top and bottom interface contribution for optimal control on the final polarization state. For instance, if we consider a Pb-containing ferroelectric film with Pb2+ surface adatoms favoring a downward-oriented polarization, flipping the surface termination of the bottom buffer layer (see Sec. II A 1) allows us to design two distinct configurations of interface electrostatic contributions. On the one hand, when each of these contributors favors opposite polarization directions, the resulting competing interface contributions can lead to a net suppression of polarization, which occurs via the formation of a multidomain configuration or reduction of the magnitudes of dipole moments. On the other hand, a cooperative configuration is achieved when both the surface off-stoichiometric layer and bottom interface charged surface termination stabilize an identical polarization direction. The latter has proved to be instrumental in setting a robust single-domain polarization, even in extreme electrostatic configurations, i.e., when the film is capped by a dielectric SrTiO3 (STO) layer,83 see Fig. 3(a).
Furthermore, by combining the different degrees of freedom described earlier, one may envision the design of ferroelectric multilayers with oppositely oriented ferroelectric polarization components. This has been so far unfeasible in the classical, depolarizing-field-tuning approach in epitaxial systems. For instance, by stacking a III-III ferroelectric constituent, where polarization is controlled through interface termination (see Sec. II A 2), with a II-IV ferroelectric compound, where off-stoichiometry dominates the polarization state (Sec. II B), one can create tail-to-tail or head-to-head polarization configurations.126 These multilayers would emulate the highly desired functionalities of ferroelectric domain walls.127,128 The combined effect of surface termination control and selective interfacial charge screening enabled by oxygen vacancy accumulation in BFO (III-III) and PTO (II-IV) epitaxial bilayers has been demonstrated by Liu et al. in a series of studies.129,130 Showcasing the stabilization of head-to-head and tail-to-tail polarization configurations129 as well as a charged interface,130 these studies offer new pathways for manipulating the diverse range of electric dipole textures in epitaxial systems.
Finally, let us consider ferroelectric systems beyond the realm of classical ABO3 perovskites structure and have a closer look at the Aurivillius compounds. Recently, the epitaxial deposition of twin-free, single-crystalline, highly anisotropic Aurivillius layered ferroelectrics with complex unit cells has been reported.25,131–133 Aurivillius Bin+1Fen−3Ti3O3n+3 (BFTO) structures consist of n-number of perovskite unit cells periodically interleaved with highly charged fluorite-like (Bi2O2) sheets. When the n is even, such as n = 4 (Bi5FeTi3O15), the layers exhibit a net uniaxial in-plane polarization.131,134,135 An out-of-plane polarization component is forbidden by symmetry in even-n compounds. It has been demonstrated that in-plane-polarized BFTO n = 4 films can be used as a buffer layer to grow other out-of-plane polarized ferroelectric perovskites on top. In these systems, the existence of the highly localized charge density in fluorite-like Bi2O2 layer (as in the Bi2O3 layer, see Sec. II B 2) combined with a robust uniaxial in-plane polarization in BFTO leads to the suppression of the polarization discontinuity at the interface between out-of-plane polarized-perovskite and in-plane-polarized Aurivillius layer. The resulting heterostructure with artificial flux-closure configuration defeats the depolarizing field effects on the out-of-plane polarized ferroelectric, enabling the emergence of spontaneous polarization from the very first unit cell in BTO and BFO thin films grown on top25 [Fig. 3(b)]. In such heterostructures, the combination of epitaxial strain and electrostatic contributions at the interface induces a uniaxial ferroelectric anisotropy in the top, downward polarized BFO layer. Strikingly, this interface stabilizes non-Ising ferroelectric domain wall configurations,22 and out-of-plane polarization components can be tracked between in-plane polarized domains, see Fig. 3(c). The continuous rotation sense of the polarization across the tail-to-tail and head-to-head domains in the BFO layer further reveals the onset of highly relevant homochiral polar textures,25 see Fig. 3(d).
III. OUTLOOK: EMERGING STRATEGIES AND OPPORTUNITIES FOR ULTRATHIN FERROELECTRICITY
With this perspective, we have put emphasis on the new design capacity on the polarization state enabled by the improved control of local chemistry and electrostatic boundary conditions in ferroelectric oxide thin films and heterostructures. Let us now consider how the interface chemistry control might be relevant in the context of the newest trends in the field of ferroelectric thin films. We discuss here the recent boost for the design of freestanding ferroelectric membranes (Sec. III A) and the push for optical poling of ferroelectric layers (Sec. III B).
A. Freestanding ferroelectric thin films
The recent advances in the controlled wet etching of oxide sacrificial buffer layers for the design of freestanding thin films have opened up new possibilities for ferroelectricity engineering.136–150 The freestanding membranes offer a unique platform for investigating the mechanical properties of ultrathin ferroelectric membranes as well as manipulating the intricate crystalline structure or even domain configuration of ferroelectrics. The design of ferroelectric membranes, which can be later transferred and stacked into functional heterostructures, shows potential for solving some of the current limitations of epitaxial growth of heterostructures on a substrate. For instance, it eliminates the restrictions caused by the limited number of lattice-matching substrates for a given ferroelectric layer. Furthermore, when films are stacked after the release from their original substrates, the van der Waals-type interaction dominates at the created interface. This may suppress constraining structural requirements for constituent layers in classical epitaxial heterostructure. Water-soluble sacrificial layers stand out as they promise simple processing of freestanding films. We highlight here the seminal work of Ji et al.,136 in which the successful design of freestanding BFO films grown on the water-soluble Sr3Al2O6 (SAO) buffer layer137 has been demonstrated. The high crystalline quality of the BFO membrane crystal structure, down to a monolayer thickness, was maintained. Remarkably, the robustness of the ferroelectric order in ultrathin BFO films151 is further evidenced with the report of the rhombohedral-like to tetragonal-like phase transition with decreasing thicknesses in freestanding layers [Fig. 4(a)]. Moving forward, it is possible to enhance the ferroelectric response, control the ferroelectric domain structure, or even trigger the ferroelectricity via inducing macroscopic deformation in the films. Such compelling possibilities underscore the imperative need for innovative experimental probes to comprehend the complex electrostatic interactions during the release and transfer of single-crystalline membranes. The field of freestanding oxide thin films is undeniably revolutionizing the ferroelectrics community and bringing new challenges. In light of the aforementioned discussions, we highlight here the so far under-explored consequences of surface chemistry during the creation of such quasi-two-dimensional objects. For instance, in BFO thin films, the interaction of the ferroelectric surface with water molecules can drive reversible polarization switching attributed to polarization-dependent chemical bonds at the water-film interface.90,152–157 This calls for the investigation of the possible influence of the polarization state on the thin film release processing and reciprocally, of the impact of the release on the initial polarization state. Such studies may be enabled by the development of noninvasive probes for ferroelectrics such as optical second harmonic generation73,158–162 and Raman spectroscopy,163–166 which both have proved to be efficient for the characterization of various ferroelectric and multiferroics thin film systems. Beyond the study of the thin film release mechanism, we can expect the surface chemistry to play an increasing role in so-called van der Waals epitaxial structures and oxide Moiré structures.143
B. Optical poling of ferroelectric films
Recent research endeavors have also shed light on the next-generation control over polarization in ferroelectric thin films. This follows the strong demand for utilizing light not only as a probe for ferroelectric polarization and domain architecture but also as a stimulus for active control over the polarization states through light–matter interactions. We focus here on the use of above-bandgap UV-light illumination. UV-light has been employed to manipulate the polarization and domain structure in ferroelectric thin films.106,167–171 Recent operando investigations revealed that UV illumination modifies the charge screening at the Schottky interface of ferroelectric thin films and, thus, leads to a transient polarization-orientation-dependent enhancement or suppression [Fig. 4(b)].106 It has been observed that the interface chemistry and off-stoichiometry plays a major role in this process. Thus, such an optical handle on electrostatics can be combined with thermal annealing or lattice chemistry engineering. Specifically, the interface atomic termination can influence the band alignment and Schottky interface characteristics.172,173 Moreover, the influence of the above-mentioned off-stoichiometric charged defects in the photoresponse is notorious.106,107 Charged defects such as oxygen vacancies interact with the photo-excited charged carriers and may serve as a tuning parameter for achieving optical control of polarization. The ability to manipulate the density and location of oxygen vacancies using photo-excited charge carriers has so far remained elusive. However, mastering this could enable the development of beyond-binary photoresponses in ferroelectric-based systems, key for the next generation of oxide-based optoelectronics.
IV. CONCLUSION
In this Perspective, we have highlighted the engineering of lattice chemistry as a complementary tool to design new polarization states in ferroelectric thin films. We have shown that among different lattice chemistry contributors, the layer polarization and off-stoichiometry gradient are the key factors to influence the polarization state. We have discussed how lattice chemistry control can be achieved for several technologically relevant ferroelectric thin films, including BFO, BTO, PTO, and PZT. A lot of progress has been made in recent years to achieve deterministic control of polarization by tuning the lattice chemistry; however, several aspects have remained under-explored. We point out the lack of studies investigating the systems with polarity conflict at the interfaces of III-III and I-V materials, which offer great potential for the design of conducting oxide interfaces. In addition, we highlight that the polarizing effect of layer charges is still under-estimated. It may become instrumental for enforcing a polarization in otherwise non-polar functional oxides and, thus, in expanding further the wide variety of ferroelectric and multiferroic oxide systems. Finally, we expect major developments in lattice-chemistry-engineered freestanding thin films as well as charged-defect-controlled layers for beyond-binary photoresponse and optical manipulation of polarization states.
ACKNOWLEDGMENTS
I.E., B.Y., and M.T. acknowledge the Swiss National Science Foundation under Project No. 200021_188414 and the ETH Zurich Research Grant funding under reference 22-2 ETH-016. All authors thank Elzbieta Gradauskaite for proofreading the manuscript.
AUTHOR DECLARATIONS
Conflict of Interest
The authors have no conflicts to disclose.
Author Contributions
Ipek Efe: Writing – original draft (equal); Writing – review & editing (equal). Bixin Yan: Writing – original draft (equal); Writing – review & editing (equal). Morgan Trassin: Conceptualization (equal); Writing – original draft (equal); Writing – review & editing (equal).
DATA AVAILABILITY
The data that support the findings of this study are available from the corresponding authors upon reasonable request.