of nonvolatile ferroelectric random access memory. [11] Another example, given the optical anisotropy of ferroelectric crystals, the electro-optic performance of a ferroelectric-based modulator relays on the relationship between the domain orientation, the applied electric field, and the polarization of the optical signal. [12] Recent years, people have further found that the ferroelectric polarization can effectively tune the light emission from the doped metal ions, which offers an additional degree of freedom for designing luminescent materials and devices. [10,13] Control over ferroelectric polarization is of great importance from both scientific and technological viewpoints. To date, a static or pulsed electric field represents the most effective tuning knob. Electric field method, however, suffers from limitations associated with the requirement of circuitry access or the cumbersome and time-consuming process. [14] Furthermore, electrically controllable polarization modulation can be typically accomplished within nanoseconds. It is difficult to use electric fields for motivating and exploiting ultrafast phenomena on the picosecond time scale or less. [15] Strain engineering is now possible to render the polarization status and build artificial heterostructures engineered down to a unit cell. [16] Breakthroughs in synthesizing high-quality epitaxial ferroelectric thin films have provided more opportunities to explore strain effects on ferroelectric polarizations. [17] Practical implementations of strain control over ferroelectric polarization are limited to some particular systems. Ferroelectric oxides are generally brittle. However, their thin-film counterparts can tolerate lattice-mismatch-based biaxial strains within ±3%. [8,18] Also, mechanical strain cannot realize ultrafast modulation of ferroelectric polarization.Electromagnetic waves have offered an extremely versatile manner for controlling ferroelectric polarization. The interaction of light with ferroelectrics gives rise to several fascinating photosensitive and photoresponse phenomena such as anomalous photovoltaics, [19] photostriction, [20] and piezophototronics. [21] These fields were termed photoferroelectrics. [22] The recent renaissance of photoferroelectrics began with the interest on ferroelectric photovoltaics, especially after a giant photovoltaic effect observed in multiferroic BiFeO 3 (BFO) thin films. [23,24] The above bandgap open-circuit voltage represents the most unique feature of ferroelectric photovoltaics, [25] which is associated with remnant polarizations, domain walls, defects,

Among them, EML materials are capable of emitting reproducible luminescence through elastic deformation, and become the most investigated ML systems that promise extensive applications, such as stress sensing, structure crack monitoring, noncontact diagnosis, artificial skin and wearable devices. [2] To date, dozens of EML materials have been synthesized and developed. Multicolor ML has been realized by incorporating various lanthanide ions into host crystals, covering full spectrum from violet to nearinfrared range. [3] Much endeavor has been devoted to finding efficient host systems including wurtzite CaZnOS, [4,5] perovskite compound LiNbO 3 , [6] and stuffed tridymite-type compounds BaAl 2 O 4 , [7] etc. In addition to homogenous compounds, a series of heterostructures based on ZnS/ CaZnOS exhibit tunable and efficient EML process via band offsets at the heterojunction interface. [8] EML materials build a certain quantitative relationship between the mechanical stress and ML intensity in an in situ and real-time manner. [9] More intriguingly, EML can lead to self-powered devices without electric circuitry. This seamless and sustainable mechano-optical transduction gives rise to miscellaneous functionalities. [10,11] Besides visualization of stress distribution, EML particles incorporated into translucent Mechanoluminescence (ML) materials present widespread applications. Empirically, modulation for a given ML material is achieved by application of programmed mechanical actuation with different amplitude, repetition velocity and frequency. However, to date modulation on the ML is very limited within several to a few hundred hertz low-frequency actuation range, due to the paucity of high-frequency mechanical excitation apparatus. The universality of temporal behavior and frequency response is an important aspect of ML phenomena, and serves as the impetus for much of its applications. Here, we push the study on ML into high-frequency range (∼250 kHz) by combining with piezoelectric actuators. Two representative ML ZnS:Mn and ZnS:Cu, Al phosphors were chosen as the research objects. Time-resolved ML of ZnS:Mn and ZnS:Cu, Al shows unrevealed frequency-dependent saturation and quenching, which is associated with the dynamic processes of traps. From the point of applications, this study sets the cut-off frequency for ML sensing. Moreover, by in-situ tuning the strain frequency, ZnS:Mn exhibits reversible frequency-induced broad red-shift into near-infrared range. These findings offer keen insight into the photophysics nature of ML and also broaden the physical modulation of ML by locally adjusting the excitation frequency.

We describe a Si-integrated photochromic photomemory based on lanthanide-doped ferroelectric Na0.5Bi2.5Nb2O9:Er3+ (NBN:Er) thin films. We show that upconversion emission can be effectively modulated by up to 78% through the photochromic reaction. The coupling between lanthanide upconversion emission and the photochromic effect ensures rewritable and nondestructive readout characteristics. Moreover, integrating photochromic thin films with Si would benefit from its compatibility with the mature complementary metal-oxide semiconductor (CMOS) technique. These results demonstrate the opportunity to develop more compact photochromic photomemories and related photonic devices.

No abstract