Multifunctional Flipping‐Based Mechanical Metasurface Enabling Optical‐Microwave Dual‐Modal Reconfigurability

ABSTRACT

Dynamically and energy‐efficient control of electromagnetic (EM) waves is highly desirable for next‐generation wireless communication and sensing. However, most existing intelligent metasurfaces rely on power‐hungry electronic circuits and complex fabrication, limiting their scalability, and deployment. Here we introduce MetaScreen, a mechanically reconfigurable metasurface that exploits a unique flipping mechanism to alternate between meta‐atoms with distinct EM responses and visual colors, thereby enabling simultaneous wavefront manipulation and optical display. Each bistable flip element integrates a permanent magnet and is actuated by a microcontroller‐driven magnetic‐control module to convert a short electrical pulse into rapid and nonvolatile mechanical switching. This design achieves low‐power and stable EM wave manipulation and endows the metasurface with intrinsic visual programmability through color‐coded coatings on the meta‐atom surfaces, thus supporting microwave wavefront control and visible information display in the same platform. Dynamic beam steering for wireless communication, adaptive EM wave focusing for contactless respiration monitoring, bidirectional human‐machine interaction, and microwave holography are demonstrated, highlighting the versatile EM manipulation and multi‐domain programmability of the platform. Owing to its energy efficiency, low cost, and long‐term stability, MetaScreen offers a practical route toward sustainable and scalable applications in wireless communication, smart internet of things (IoT), and optical‐EM camouflage.

A multifunctional flipping‐based mechanical metasurface enables visible display and microwave wavefront control in the same platform. Electrical pulses trigger bistable, nonvolatile state switching, allowing optical display, wireless communication, respiration sensing, and microwave holography while maintaining low power consumption, long‐term stability, and scalable operation. The study points toward sustainable applications in wireless communication, smart IOTs, and optical‐EM camouflage.

1. Introduction

Programmable metasurfaces have emerged as transformative platforms to reconfigure electromagnetic (EM) fields, underpinning future applications in wireless communication, sensing, and intelligent environments [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13]. A variety of reconfigurable mechanisms have been explored to achieve tunability, such as phase‐changing materials, optically, electrically, and mechanically tunable structures [14, 15, 16, 17, 18, 19, 20, 21, 22, 23]. Among these, electrically tunable metasurfaces based on PIN diodes and varactors have attracted particular attention for real‐time wavefront modulation, owing to their compact integration, sub‐millisecond response, and compatibility with existing communications infrastructures [24, 25, 26, 27, 28]. These characteristics position them as a promising alternative to conventional multi‐antenna and relaying technologies for high‐capacity data processing, with envisioned applications in intelligent transportation systems, internet of things (IoTs), and wireless power transfer [29, 30, 31, 32, 33]. Furthermore, their strong integrability also establishes a new paradigm for multimodal information interaction [34]. Despite these impressive advances, electrically reconfigurable metasurfaces suffer from inherent limitations. Their reconfiguration mechanism is inherently volatile, requiring continuous electrical biasing that introduces elevated thermal loads and spectral drift. Meanwhile, the active meta‐atoms that integrate PIN diodes or varactors with dense bias‐routing networks impose substantial fabrication complexity and increase per‐element cost. In addition, the field‐programmable gate arrays (FPGAs)‐based centralized control architectures commonly employed in such systems are expensive and difficult to cascade, thereby impeding large‐scale deployment and limiting scalability compared with simpler microcontroller unit (MCU)‐based solutions. Furthermore, the tuning states and functional configurations are typically invisible during operation, and fault diagnosis of defective elements often requires complex procedures [35, 36, 37]. These challenges have sparked growing interest in alternative reconfiguration strategies that combine nonvolatile, reconfigurability, reduced energy demand, and simplified system integration. Here, nonvolatility refers to the ability to retain programmed metasurface states without continuous power consumption [38].

Thermal metasurfaces have recently emerged as a powerful platform for achieving reconfigurable EM functionalities [39, 40, 41]. However, most implementations rely on thermally driven phase transitions that require substantial heating power and elevated operating temperatures. By contrast, mechanically reconfigurable metasurfaces, which translate external mechanical stimuli into tunable EM responses, have attracted considerable attention across multiple research disciplines [42, 43, 44, 45, 46, 47]. By eliminating active electronic and thermal components, mechanical reconfiguration provides an intrinsically energy‐efficient, and nonvolatile mechanism for dynamic wavefront control. Power is consumed only during brief actuation events, enabling negligible static power consumption and stable long‐term state retention. The resulting energy‐efficiency advantage is therefore most pronounced in low‐update‐rate and long‐retention regimes, where the elimination of static power compensates for the finite switching energy required for occasional actuation. A detailed analysis is provided in Note S5. These characteristics render mechanical reconfiguration particularly attractive for applications that do not require rapid modulation but place a premium on long‐term stability, robustness, and minimal static power consumption.

To realize mechanical reconfiguration, various strategies have been explored such as microelectromechanical (MEM) system [21, 48, 49, 50, 51, 52], microfluidics [53], magic cube configuration [54], mechanically stretchable [55], and kirigami/origami‐based metasurfaces [56, 57, 58]. Among these approaches, MEM‐based mechanically reconfigurable metasurfaces have been most actively explored in wireless communication and intelligent environment. These representative studies validate the feasibility of mechanical actuation across microwave and millimeter‐wave regimes and highlight the remarkable potential for adaptive and programmable EM systems. However, these techniques maintain several challenges unaddressed. Most existing platforms offer insufficient wavefront control capability and limited system robust [57]. Moreover, the lack of large‐scale, multifunctional implementation has hindered the deployment in practical scenarios. These considerations motivate the development of a mechanically reconfigurable metasurface system that integrates scalability, multifunctionality, broad applicability, mechanical robustness, and precise wavefront control across diverse application domains.

In this article, we present MetaScreen, a multifunctional, mechanically reconfigurable metasurface composed of an array of bistable flip elements. Each element integrates two elaborate meta‐atoms for EM wave manipulation together with a centrally embedded magnet. A brief electrical pulse activates the magnetic‐control module positioned behind the perfect electric conductor (PEC) base, converting electrical energy into nonvolatile, low‐power mechanical modulation that overcomes the energy barrier between two stable states. This mechanism allows reversible switching between the distinct configurations, thereby realizing bistability. By applying color‐specific coatings to the meta‐atom surfaces, the same mechanism extends reconfigurability from microwave wavefront manipulation to macroscopic appearance modulation.

Building on this mechanism, MetaScreen establishes a system‐level platform that unifies EM and optical functionalities, enabling dual‐modal reconfigurability. Specifically, the mechanical flipping process simultaneously modifies both the EM scattering response and the visible optical appearance of each bistable element. This paradigm enables programmable EM wavefront control (modality 1) in the microwave regime and visual information delivery (modality 2) in the optical regime, thereby forming a complete innovation pipeline from concept to platform realization and application deployment. A detailed explanation of this intrinsic dual‐modal reconfiguration mechanism is provided in Note S4. With an integrated visual sensing module, MetaScreen further responds to environmental stimuli, supporting bidirectional human–machine interaction. Moreover, the mechanical architecture ensures that each element can be individually addressed, replaced, or reconfigured without affecting the rest of the array, providing exceptional robustness and long‐term operational stability. Importantly, because power is only consumed during state transitions, the system achieves inherently low‐power performance suitable for sustainable, large‐scale, and long‐duration deployments. In this work, MetaScreen is conceived from the outset as a system‐level, multifunctional reconfigurable metasurface platform that integrates EM wavefront control and optical information delivery within a unified architecture. To substantiate this capability, we experimentally demonstrate beamforming, dynamic EM focusing, and microwave holography, validating the advanced EM wavefront control abilities of the platform. Simultaneously, programmable optical patterns illustrate visual communication and interactive potential. By combining powerful, energy‐efficient wavefront manipulation with unique optical functionality, MetaScreen substantially broadens the functionality and application scope of intelligent metasurfaces, unlocking opportunities in smart architecture, adaptive sensing, as well as IoT, 5G/6G wireless communication, and active multi‐domain camouflage [59]. Such AI‐driven paradigms further indicate that the MetaScreen platform could evolve toward next‐generation embodied intelligence [60] and homeostatic neuro‐metasurfaces [51].

2. Results and Discussions

2.1. Framework of MetaScreen

To illustrate the concept and multifunctionality of the MetaScreen, Figure 1 presents a schematic diagram highlighting the basic optical reconfigurability and EM control capability alongside the corresponding applications. The multifunctional system comprises an array of mechanically bistable flip elements, a visual sensing module, and an MCU for addressing and actuation. A host computer generates and manages coding patterns, collectively enabling dual‐modal reconfigurability across both optical and EM regimes. Each flip element is mounted on a rectangular PEC base and laminated with two meta‐atoms designed for producing distinct EM responses. Each surface is overlaid with a color‐specific optical coating, so that mechanical flipping simultaneously toggles the microwave scattering behavior and the visible appearance, coupling optical display to microwave functionality.

FIGURE 1.

Conceptual illustration of the mechanically reconfigurable MetaScreen. (a) MetaScreen as a dual‐domain platform for IoT, enabling wireless communication, and contactless sensing. (b) Mechanism and architecture of the flip elements. Each element integrates two meta‐atoms with opposite optical coatings (code‐0 and code‐1), actuated by a compact MCU‐driven magnetic‐control module that provides nonvolatile reconfiguration. (c) Demonstration of dual‐modal information delivery achieved by encoding microwave holographs and projecting programmable optical patterns, offering scalable, energy‐efficient, and bidirectional reconfigurability.

Reconfiguration is achieved through an integrated MCU‐driven magnetic‐control module, in which brief electrical pulses from the MCU directly convert electrical energy into the rapid mechanical flipping of a magnetized element. This approach enables energy‐efficient, nonvolatile state switching, with experimentally measured reconfiguration times as short as 50 ms. Moreover, all flip elements are independently addressable and physically replaceable, combining precise control with exceptional system robustness. Localized malfunction of the elements does not compromise the rest of the array, and defective elements can be visually identified and replaced without disassembling the entire system, facilitating rapid fault diagnosis and ensuring long‐term operational stability. Importantly, the modular architecture also provides a practical foundation for large‐scale deployment by enabling module‐level assembly and distributed control, with further scalability considerations provided in Note S20.

Mechanically actuated flipping confers MetaScreen with intrinsic dual‐modal reconfigurability by synchronously modulating both the EM response and the optical appearance of each element. Consequently, a single structural switching mechanism simultaneously triggers functional transitions across the two modalities. In the optical domain, MetaScreen generates reconfigurable high‐contrast patterns for visual information delivery, whereas in the microwave domain, the same flipping operation enables programmable wave manipulation, supporting beam steering, dynamic EM wave focusing, and near‐field holography. When integrated with the visual sensing module to realize real‐time detection, localization, and posture estimation, the platform can further achieve the bidirectional human–machine interaction and multimodal information delivery.

2.2. Design and Applications of MetaScreen

The platform comprises an array of mechanically bistable flip elements, each integrating two meta‐atoms engineered for producing distinct EM responses. The two surfaces of the elements are coated with color‐specific coating (white for code‐1 and black for code‐0), ensuring that a single flip simultaneously reconfigures the microwave scattering state and the optical appearance. As shown in Figure 2a, each flip element features a multilayered architecture with a centrally embedded magnet, which cooperates with the magnetic‐control module to enable rapid, power‐on switching between two functional states. The structural details of the meta‐atoms are provided in Supporting Information. The mechanical reconfiguration process of the MetaScreen is depicted in Figure 2b, where the state of each flip element can be individually addressed by modulating the current supplied to the corresponding magnetic‐control module. This mechanism allows for programmable, reversible reconfiguration across the entire MetaScreen.

FIGURE 2.

Mechanically reconfigurable MetaScreen enabling multifunctional application. (a) Structure of a single flip element, consisting of multilayered architecture layer to construct binary coding state with different EM responds (code‐0 and code‐1) enabled by mechanical flipping. (b) Programmable MetaScreen composed of an array of reconfigurable meta‐atoms with a 26 mm lattice period, each individually actuated via a magnetic‐control module driven by a single chip microcomputer controller. (c) Measured reflection amplitude and phase response for the code‐0 and code‐1 states, the green‐shaded areas indicate a neighborhood of the operational frequency (5.4 GHz). (d) Demonstration of MetasScreen‐based visual display (Function 1), where pixel patterns can be dynamically reconfigured by mechanical flipping(t = t ₁). (e) Time‐sequential functions enabled by programmable MetaScreen: beam steering for wireless communication (Function 2, t = t ₂), EM focusing for noncontact respiratory sensing at (Function 3, t = t ₃), and dynamic holographic display (Function 4, t = t ₄).

We further experimentally validate the EM performance of the platform through measuring its microwave reflection coefficient, as shown in Figure 2c. For this proof‐of‐concept demonstration, a small‐scale prototype containing 14  ×  16 meta‐atoms is fabricated. It could be observed that the two states exhibit a reflection phase difference of approximately 180°, while maintaining high reflection amplitudes across the operational band. Measurements reveal a stable response from 5.35 to 5.45 GHz, encompassing representative communication and sensing frequencies. Beyond the normal‐incidence characterization, we further perform oblique‐incidence full‐wave simulations, as detailed in the Note S1. The results confirm that the cross‐shaped meta‐atoms retain high reflection amplitudes and a pronounced binary phase contrast from 0° to 45°, supporting stable wavefront control under practical incidence conditions. It should be emphasized that the operating frequency and bandwidth reflect application‐driven choices tailored to WiFi‐based wireless communication and sensing, rather than intrinsic limitations of the proposed platform. Notably, the flipping mechanism itself imposes no fundamental bandwidth constraint. The current bandwidth arises solely from the resonant characteristics of the selected meta‐atom geometry and can be readily extended through multi‐resonant element designs or Pancharatnam–Berry (PB) phase engineering.

Unlike conventional electronic modulators or mechanically tunable metasurfaces, the MetaScreen integrates optical display functionality by leveraging color‐specific coatings. By assigning specific coding patterns, the system is capable of producing high‐resolution outputs in the optical regime. As illustrated in Figure 2d, we demonstrate this functionality by encoding the readable text pattern “META screen”, which operates as an interactive visual display. The platform exhibits precise mechanical control across the entire array, enabling diverse visual patterns to be dynamically displayed through straightforward modification of the coding matrix. This vision‐based operation introduces a novel mechanism for real‐time optical information transmission and underscores the potential of MetaScreen as a multifunctional platform.

Moreover, owing to the meta‐atoms on opposite sides of the flip element exhibiting distinct EM reflection responses, the MetaScreen can achieve a variety of microwave functionalities (Figure 2e). At the initial time step, a periodic binary pattern of “00001111” is programmed (Figure 2e, top), enabling beamforming for wireless communication applications. Then, a concentric binary pattern resembling a circular zone plate is encoded (Figure 2e, middle), enabling near‐field energy focusing for contactless respiratory monitoring. In this configuration, the coding patterns depend on the chest motion detected by the visual sensing module, enabling the real‐time, noninvasive monitoring suitable for smart‐home and clinical healthcare scenarios. Furthermore, the MetaScreen is also capable of reconstructing holographic images (Figure 2e, bottom). Collectively, these demonstrations establish the MetaScreen as a dual‐modal reconfigurable platform that seamlessly integrates real‐time optical display with reconfigurable microwave wavefront control, highlighting the potential for wireless communication, contactless sensing, bidirectional interaction, and multimodal information transmission.

2.3. MetaScreen for Wireless Communication and Contactless Human Sensing

Two typical EM functions, including wireless communication and contactless human sensing, have been implemented to demonstrate the versatility of MetaScreen in the EM wave manipulation. In the first experiment, MetaScreen is configured as a reconfigurable wireless relay to dynamically control spatial communication channels. The experimental setup is shown in Figure 3a, where the MetaScreen is connected with a control computer for both real‐time coding pattern switching and synchronization between two software‐defined radio (SDR) systems: a WiFi‐SDR transmitter and two universal software radio peripheral (USRP) receiver. A high‐gain horn antenna transmitted 16‐QAM WiFi signals toward MetaScreen, while four spatially separated horn antennas connected to different USRP receive chains that capture signals at predefined azimuth angles (0°, 15°, 30°, and 45°). By mechanically flipping elements to change the binary coding pattern, MetaScreen performed dynamic beamforming, directing the incident signal toward a specific receiver while effectively suppressing all other channels (Figure 3b). The applied coding matrix and the corresponding 2D far‐field pattern are presented in Figure 3c, with further experimental details available in Supporting Information. Additional far‐field measurements confirm that the fabricated mechanical MetaScreen preserves a dominant reflected beam over a steering range from 10° to 60°, with measured angular deviations below 3° despite realistic mechanical imperfections.

FIGURE 3.

MetaScreen‐enabled wireless communication and contactless respiration sensing. (a) Experiment setup for space‐division wireless communication using MetaScreen. (b) Schematic of the communication process: a transmitter located at 45° is redirected by MetaScreen to form a spatial channel at 30°, enabling only receivers positioned near 30° to successfully decode the signal. (c) Spatial coding pattern of MetaScreen and the corresponding 2D normalized far‐field radiation pattern in polar coordinates. (d) Transmission of a test image from the transmitter at 45° and its recovered image at the 30° receiving position, alongside the measured constellation diagram. (e) Example of a transmission attempt to a non‐targeted receiver at 45°, showing severe image distortion and constellation degradation, indicating spatial selectivity of the channel. (f) Experimental setup for noncontact vital‐sign sensing using MetaScreen‐enhanced field focusing. (g) Real‐time field focusing onto targeted human chest positions at three consecutive time intervals (t  =  Δt,  2Δt,  3Δt), with corresponding coding patterns and measured electric‐field distributions. A visual sensing module tracks target and dynamically updates the coding pattern to maintain focus on the chest region. (h) Continuous respiration monitoring enabled by MetaScreen‐assisted sensing, with comparison to a wearable respiratory sensor (WRS) and corresponding error rate analysis.

Experimental results confirm that the intended receiver at 30°achieves error‐free 16‐QAM transmission with a data rate of 43.3 Mbps, faithfully reconstructing the transmitted image (Figure 3d). In contrast, receivers positioned at other angles (e.g., 45°) exhibit severe signal degradation due to intentional phase mismatches (Figure 3e), resulting in a serious reduction in signal‐to‐noise (SNR) relative to the target channel. This angular selectivity enables real‐time channel switching among multiple users within 50 ms reconfiguration cycles, maintaining low bit‐error rates during active transmission. Measured channel response closely matches theoretical predictions, demonstrating the fast, precise, and robust beamforming capabilities of the platform for programmable spatial wireless communication. Meanwhile, since the proposed MetaScreen functions purely as a passive relay for beamforming, the system's data throughput is primarily determined by the WiFi transceiver rather than by the metasurface. The MetaScreen prototype supports an operational bandwidth of approximately 100 MHz, substantially surpassing the 20‐MHz bandwidth of the WiFi transceiver and underscoring the platform's potential for much higher communication throughput. Future developments can therefore focus on further broadening the operating bandwidth and enhancing the EM wave manipulation accuracy of the metasurface, which enables MetaScreen‐based wireless communication systems to support substantially higher data rates.

In the second experiment, we demonstrate the capability of MetaScreen for adaptive, contactless respiration monitoring. The experimental configuration (Figure 3f) comprises two high‐gain antennas directly connected to a USRP and a WiFi‐SDR separately, a wearable respiration sensor (WRS) as the benchmark, and MetaScreen integrated with a visual sensing module. The sensing module incorporates target recognition, localization, and posture detection algorithms, endowing the system with the ability to identify the thoracic position in real time. An excitation antenna linked to the WiFi‐SDR transmits EM waves toward MetaScreen, which dynamically focuses the energy onto the target thoracic region through rapid and precise wavefront manipulation. Breathing‐induced chest displacement modulates the focused EM field, which will be subsequently captured by the receiving antenna and routed to the USRP for further processing. Then vital‐sign parameters such as respiration rate can be extracted, enabling noncontact monitoring.

By combining precise spatial localization with EM wave focusing, MetaScreen acts as an intelligent EM wave concentrator that adaptively concentrates EM energy on the chest of the target individual. As illustrated in Figure 3g (top), the system detects the human target at varying times and positions within the field of view. The coordinates relative to MetaScreen are then computed to generate the corresponding coding sequences in real time (Figure 3g, middle), thereby producing the focused EM wave. The measured near‐field distributions for each coding pattern are shown in Figure 3g (bottom). The real‐time reconfigurable EM wave focusing significantly improves the SNR of respiration detection, producing a respiration rate curve that closely tracks the WRS reference throughout 120s recording (Figure 3h). The system achieves consistently low respiration rate errors, typically within 2 respiration per minute (RPM), and exhibits stable performance with no noticeable drift over the observation period. Together, these experiments confirm that MetaScreen operates not only as a reconfigurable communication relay, but also as a sensing system for noninvasive physiological monitoring. By integrating adaptive wireless communication with contactless sensing, MetaScreen emerges as a promising platform for future applications in smart‐home, healthcare, and ambient intelligence.

2.4. Event‐Triggered Bidirectional Interaction and Multimodal Information Delivery

In addition to the advanced EM wavefront control, Metascreen incorporates intelligent optical display capabilities through color‐specific coatings on its flip elements. By reconstructing the coding sequence, distinct visual patterns can be formed, thereby extending the functionality of the platform into the visible spectrum. This dual‐modal reconfigurability, spanning both EM and optical regimes, enables seamless multimodal information delivery. When combined with a visual sensing module, the platform acquires the ability to perceive environmental information and respond optically and dynamically, thereby establishing a foundation for real‐time, bidirectional human–machine interaction.

The experimental configuration for the optical information delivery and interactive functions is illustrated in Figure 4a. The integrated visual sensing module is responsible for human detection and localization, continuously analyzing the surrounding environment to identify relevant stimuli. Once the environment state is determined, the system triggers the selection of scenario‐specific coding patterns. In the absence of a human target, the system enters an “ambient information” mode, rendering real‐time clock patterns directly through the optical flip interface (Figure 4b), effectively functioning as an environmental information board. Upon detecting a human target, the sensing module captures the target and transmits the processed image to MetaScreen, which mechanically flips its elements to render a silhouette in real time. These visual‐display modes operate at low refresh rates (hertz level), and each update typically changes only a subset of elements. The programmed element states are retained without static power between updates, consistent with the energy‐efficient operating regime of MetaScreen. This event‐triggered dynamic display exemplifies the capacity for fundamental bidirectional interaction: MetaScreen actively acquires external information through the visual sensing module while simultaneously communicating information back to the environment via direct optical display.

FIGURE 4.

Dual‐modal MetaScreen for intelligent optical interaction and reconfigurable microwave holography. (a) Experimental setup for multimodal demonstration, where the MetaScreen functions as a reconfigurable interface bridging optical display and microwave holography. (b) Event‐triggered dynamic display: in the absence of human target, the MetsScreen presents a real‐time clock; upon detection, it switches to a silhouette rendering of the individual and reverts to the clock when the person leaves. (c) Control‐flow diagram of mode selection: the system first detects in function selection (FS) mode; upon this mode, MetaScreen performs continuous optical display (e.g., real‐time person silhouette). Otherwise, the system generates programmable microwave holography. (d) Optical rendering of the alphanumeric characters “M”, “E”, “T”, and “A”. (e) Corresponding 1‐bit coding patterns for microwave holography (top), and the experimentally near‐field results (bottom), demonstrating direct optical‐to‐microwave information delivery within the same physical platform.

To further demonstrate the performance of multimodal operation, the system incorporates a function selection (FS) module (Figure 4c) that switches between continuous optical display and reconfigurable EM holography. In optical display mode, the system presents dynamic optical display. Otherwise, the system switches to the reconfigurable microwave holographic mode. In the proof‐of‐concept experiment (Figure 4d), four optical characters (“M,” “E,” “T,” and “A”) are selected as target images, and the Gerchberg–Saxton (GS) algorithm is employed to generate the corresponding coding patterns for the four holographs (Figure 4e, top). The measured near‐field holographic results (Figure 4e, bottom) closely match their optical counterparts, demonstrating the high‐fidelity wavefront control and multimodal display capability from the optical to the microwave domain.

MetaScreen unifies visible and microwave control within a single hardware platform, functioning simultaneously as an interactive visual terminal and a programmable microwave‐holographic interface. This dual‐modal reconfigurability not only enables seamless multimodal information delivery in microwave and optical regimes but also realizes the bidirectional interaction between the MetaScreen and environment. Furthermore, this coupled optical‐microwave operation can be adapted to scenarios requiring partial functional independence through spatial multiplexing, temporal multiplexing, or multi‐objective optimization, as discussed in Note S19. By bridging the visible and invisible spectrum, this capability provides a versatile foundation for adaptive communication systems, with far‐reaching implications for smart human‐machine interfaces, immersive IoT networks, and integrated optical‐EM defense technologies.

3. Conclusion

Mechanically reconfigurable platforms that unify EM wave manipulation with visual information delivery remain largely unexplored. Here, we introduce MetaScreen, a bistable metasurface that bridges programmable microwave wavefront control with optical pattern encoding, unlocking dual‐modal reconfigurability for diverse functions. Each flip element incorporates two meta‐atoms with distinct EM responses, while the corresponding color‐specific coating directly maps these states into visible patterns. This dual‐function architecture allows precise microwave scattering control alongside multimodal information delivery. Experiments encompassing intelligent spatial wireless communication, contactless respiratory monitoring, event‐triggered dynamic display, and microwave holography collectively demonstrate the powerful capability of MetaScreen in wavefront shaping and multimodal information delivery. Collectively, these results demonstrate optical‐microwave dual‐modal reconfigurability of the Metascreen, offering a scalable, low‐power platform for multifunctional smart‐surface applications.

Compared with the conventional electrically reconfigurable metasurfaces that operate as unidirectional EM control interfaces, our platform exhibits decisive advantages. The mechanical flipping mechanism eliminates the need for sustaining bias to preserve the state. This design ensures intrinsic nonvolatility, robustness, simple fabrication, low cost, and energy‐efficient while suppressing thermal effects during prolonged operation. This makes the platform particularly suitable for large‐scale, long‐term deployments. Additionally, MetaScreen adopts a modular and independently detachable architecture, whereby each flip element can be rapidly diagnosed and physically replaced without disturbing the surrounding array. This unique feature substantially enhances system robustness and maintainability. Furthermore, the integration of visually encoded coatings with functional meta‐atoms extends reconfigurability into both the microwave and visible regimes, enabling adaptive multimodal information delivery that bridges human perception and EM communication. Although the current platform operates in the sub‐6 GHz due to application‐driven considerations tailored to indoor WiFi‐based environments, the underlying mechanical reconfiguration mechanism is intrinsically scalable to higher frequencies, including the millimeter‐wave regime. This can be achieved by proportionally reducing the meta‐atom dimensions and miniaturizing the magnetic‐control module. Collectively, these attributes highlight the practical scalability and measurable benefits of the proposed MetaScreen. Building on this foundation, the integrated optical display introduces an additional modality and unlocks a unique set of intended applications. It enables real‐time verification of the metasurface's spatial coding distribution, thereby improving configuration reliability and reducing calibration overhead. Moreover, when coupled with vision‐based sensing and the reasoning capabilities of large language models (LLMs), the color display provides a physical substrate for emerging metamaterial‐agent frameworks. Mechanically reconfigurable color patterns can also induce controlled visual confusion enabling adaptive optical camouflage. Together, these potential functionalities demonstrate that the optical display is not merely decorative but serves as a functional and system‐level component that substantially enhances the versatility and applicability of MetaScreen in practical environments.

In summary, MetaScreen embodies a new paradigm of mechanical reconfigurability that unites nonvolatility, energy efficiency, and dual‐modal functionality within a scalable and robust architecture. These advances directly overcome the intrinsic bottlenecks of conventional reconfigurable metasurfaces and unlock opportunities for long‐term, large‐scale, and energy‐efficient deployments. We anticipate that the distinctive capabilities of MetaScreen will accelerate advances in sixth generation (6G) wireless networks, IoT, and noninvasive sensing, while also inspiring next‐generation platforms that unify microwave–optical camouflage, next‐generation metaAgents, and multimodal communication technologies.

4. Methods

4.1. Sample Design and Fabrication

The MetsScreen prototype comprises 28 × 32 flip elements (lattice constant a = 26 mm), yielding an overall panel size of 832 mm × 728 mm. Each flip element is realized by multilayer PCB‐lamination process and is mounted on a rectangular PEC base. Each flip element adopts a multilayer laminated architecture, consisting of (from top to bottom) a color‐specific solder‐resist coating, a cross‐shaped copper resonator on a dielectric substrate with a metallic backing, plastic support layers with a centrally embedded magnet, and a second metallic backing and dielectric layer carrying the complementary resonator with bottom‐side color coating. This arrangement simultaneously provides stable EM performance, mechanical flipping capability, and visible contrast. The resonators are printed on F4B dielectric sheets (ε _r = 2.65, tanδ  =  0.001, thickness  =  0.508 mm) using standard copper metallization. Geometric parameters for the resonators are R = 25.0 mm, L ₂ = 18.9 mm, W ₂ = 6.4 mm for code‐1, and r = 5.0 mm, L ₁ = 18.5 mm, W ₁ = 6.4 mm for code‐0. This laminated construction and material choice provide stable microwave performance, reconfigurable flipping dynamics and straightforward fabrication. To ensure fabrication consistency, the flip elements are fabricated using mature commercial processes, including standardized CNC milling of commercial dielectric substrates, established PCB etching, and commercial solder‐resist coating. These processes help maintain consistent meta‐atom geometries, mechanical assembly accuracy, stable EM responses, and visible optical contrast across different elements. More details about layer drawings, fabrication tolerances, and device‐to‐device variability are provided in the Supporting Information.

Actuation of each flip element is achieved through a base‐mounted magnetic‐control module under MCU control (STM32F103). The magnetic‐control modules are batch‐fabricated and optimized through multiple design iterations, providing high fabrication yield and reliable actuation performance across the array. This high yield ensures reliable actuation across the array and mitigates the impact of device‐to‐device variations in magnetic actuation strength. Short current pulses from the MCU energize EM coils wound around a U‐shaped magnetizable core (DT4), converting an instantaneous electrical pulse into a transient magnetic field that couples with the permanent magnet embedded in each flip element. The resulting torque induces a repeatable, bistable rotation, flipping the element between its two functional faces. As power is required only during these brief switching progress (with pulse durations of approximately 50 ms), the scheme delivers energy‐efficient, power‐on actuation while preserving a nonvolatile state between transitions. The experimentally measured reconfiguration time is conservatively reported as 50 ms, which corresponds to the complete switching process, including mechanical flipping, and subsequent stabilization of the bistable element. During the measurement, the number of successful switching events completed within a fixed time interval was recorded, and the average reconfiguration time was determined from the ratio of the total measurement duration to the number of successful state transitions.

4.2. Numerical Simulations

EM numerical simulations were performed using CST Microwave Studio 2023. The reflection coefficients (S ₁₁) of the flip element (comprising the rectangular PEC base and two meta‐atoms on opposite sides) were numerically simulated with the frequency‐domain solver. Simulations were conducted with unit cell boundary conditions in the x–y plane with open‐space boundary conditions along the z axis. Meanwhile, the time‐domain solver was employed to simulate the beamforming, EM wave focusing and holography functionalities of the MetaScreen, with open boundary conditions in all three axes (x, y, and z). It is noted that the EM performance reported in Figure 2 was evaluated at an operating frequency of 5.4 GHz. Additional details regarding the model geometry and simulation results are provided in Supporting Information.

4.3. Measurement Setups

To validate the multifunctionality of the MetaScreen, four representative experiments were conducted in the article. The wireless communication experiment shown in Figure 3a was performed in a large indoor open‐space environment. The transmission chain comprised a linearly polarized broadband horn antenna (GJ‐WRDHA‐U8/08‐N) driven by an open‐source WiFi SDR stack (ZedBoard FPGA with an AD9361 RF front end). On the receiving side, four high‐gain Vivaldi antennas were deployed at predefined azimuth angles (0°, 15°, 30°, and 45°) and connected to two USRP devices, providing four parallel receive channels for real‐time capture and demodulation.

The second experiment was conducted in an indoor environment to evaluate the ability of adaptive contactless respiration sensing. The same WiFi SDR and USRP were employed to generate and acquire the sensing waveform at 5.4 GHz. A Vivaldi antenna and a broadband horn antenna were employed, with one used for transmission to illuminate the MetaScreen that focuses the EM wave onto the human thoracic region, and the other used to receive the field scattered from the human thoracic region. A visual sensing module (Zed 2i Stereo Camera) provided real‐time 3D localization of the chest. The video stream was processed on a control computer to detect human presence, after which the corresponding coding patterns were generated and transmitted to achieve real‐time EM wave focusing. The variational mode decomposition (VMD) algorithm was employed to extract respiratory signals based on the periodicity of physiological movements. A WRS (HKH‐11C) provided ground‐truth measurements for accuracy benchmarking.

The third experiment involves event‐triggered bidirectional interaction, using the same visual sensing module to capture environmental information. The camera can not only determine the position of a person but also perform real‐time detection and posture estimation. The holograph experiment employed near‐field scanning microwave microscopy (NSMM) to characterize EM wavefront manipulation. The measurement setup consisted of a vector network analyzer (Agilent N5230C) and two phase‐stable coaxial cables connected to the analyzer's ports: one linked to the same high‐gain linearly horn antenna served as the excitation source, and the other to a coaxial probe mounted on a movable stage for detecting spatial electric‐field distributions. By adjusting the probe's orientation in the x, y, and z directions, all vector components of the electric field (E_x ,E_y ,E_z ) were measured. The probe, positioned 200 mm above the sample surface, was mounted on a scanning platform allowing point‐by‐point measurements across a measuring 900 × 800 mm² region to reconstruct the spatial electric‐field pattern.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: adma74169‐sup‐0001‐SuppMat.docx.

Data Availability Statement

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