Multispectral metal-based electro-optical metadevices with infrared reversible tunability and microwave scattering reduction

2.1 Design of metadevice

Compared to achieving dynamic IR and microwave camouflage compatibility based on emissivity control, temperature control appears to be more readily achievable. There have been reports of integrating metamaterials for microwave camouflage into temperature control structures or devices based on microfluidic [27] or thermoelectric systems [28] to achieve this purpose. However, these approaches typically require additional heating, cooling, and pneumatic systems, inevitably leading to high energy consumption and limited flexibility in control methods. To the best of our knowledge, there are currently no reports on utilizing electrically controlled emissivity to achieve compatibility for dynamic IR and microwave camouflage. Here, we propose a realistic approach based on electrically controlled emissivity by integrating coding metamaterials into RME devices, enabling dynamic IR and microwave camouflage compatibility. Figure 1 illustrates the structure and operation principle of the proposed metadevice. The main part of the metadevice is a copper-based RME variable emissivity device, which, as an electrochemical device, features a typical sandwich structure. It utilizes a gold grid and evaporated ultra-thin platinum (Pt) film as the working electrode, and employs a copper (Cu) foil as the counter electrode, with electrolyte containing Cu²⁺ ions sandwiched between the working and counter electrodes. The electrochemical reactions involved in electrodeposition and dissolution of Cu layers in electrochromic devices can be briefly described as follows:

Barium fluoride (BaF₂) is employed as the IR-transparent substrate for the working electrode, onto which a JC metal structure, resembling a chessboard-like configuration, is loaded. This implies that BaF₂ also serves as the dielectric layer in coding metamaterials, while the electrode acts as the metallic backplane layer. The integrated design scheme of metadevices incorporating coding metamaterial structures in RME devices effectively controls the overall device thickness and avoids complex structures. It is worth noting that while BaF₂ was chosen as the IR-transparent substrate here, other IR-transparent substrates such as zinc selenide (ZnSe), silicon (Si), zinc sulfide (ZnS), and polypropylene (PP) films can also be considered. Additionally, the Cu deposition system can be adjusted to other deposition systems such as silver (Ag), bismuth (Bi), nickel (Ni), and zinc (Zn) according to requirements. However, during the design phase of metadevice structures, the following two aspects should be comprehensively considered. On the one hand, the loading of JC metal structures inevitably impacts the IR modulation performance of RME variable emissivity devices. To mitigate this impact, while ensuring outstanding microwave scattering reduction performance, efforts should be made to minimize the filling ratio of the JC metal structures on the device surface. On the other hand, the electrodeposition process unavoidably affects the device’s microwave response; hence the thickness of the electrolyte layer should be compressed to the minimum feasible extent during design to mitigate this influence (see Figure S1, Supplementary Material for detailed analysis).

The key to the design of coding metamaterials lies in creating two distinct unit cells with a 180° phase difference, namely the “0” and “1” elements, respectively. Many structures, such as JC, ring, patches, and split ring, are commonly used for the design of the “0” and “1” elements. We also compare the reflection phases of different structures in Figure S2 (Supplementary Material). After considering the effective modulation bandwidth and structural filling ratio, we ultimately selected the JC structure. Figure 2(a) illustrates the schematic diagram of the designed “0” and “1” elements, with their only difference being whether or not the subwavelength JC metal structure is loaded on the RME variable emissivity device. In CST Microwave Studio, we simulated the reflection phase and magnitude of the “0” and “1” elements and optimized their structural parameters. Specifically, we set the electrolyte layer thickness d ₂ to a feasible minimum of 0.1 mm, and its complex permittivity was measured by the coaxial probe method (measurement results see Figure S3, Supplementary Material), and further optimized the structural parameters as depicted in Figure 2(b) and (c). The optimization of structural parameters was primarily conducted through parameter sweeping, and the results for the main geometric parameters are depicted in Figure S4 (Supplementary Materials). The geometric parameters of the structure are optimized as follows: The periodicity of the “0” and “1” elements is p = 4 mm; the BaF₂ layer has a thickness of d ₁ = 2 mm, with a dielectric constant of 6.6 and loss tangent of 0.003; the JC metal structure layer has a thickness of t = 300 nm, with geometric parameters of l = 3.5 mm, s = 1.25 mm, and w = 0.12 mm. Additionally, it is noteworthy that the thickness of the ultra-thin Pt film is 4 nm, with a sheet resistance of ∼150 Ω/sq, and the sheet resistance of the Cu electrodeposited Pt film in the deposited state can be reduced to ∼10 Ω/sq. Based on this, we simulated the reflection phase and amplitude of the “0” and “1” elements in both the dissolved and deposited states, as illustrated in Figure 2(d) and (e). Since the Cu foil used as counter electrode has excellent conductivity and is continuous, the transmission is very small and negligible here. It can be observed that in both the dissolved and deposited states, the “0” and “1” elements exhibit an ∼180° phase difference in a wideband range of 10–20 GHz and have a high reflection amplitude. Additionally, it can be noted that the electrodeposition process does not significantly impact the microwave response of the device.

Figure 2:

Design of “0” and “1” elements. (a) Schematic diagram of the designed “0” and “1” elements. (b) Top view and geometrical parameter of the “1” element. (c) Side view and geometrical parameter of the “1” element. (d) Simulated reflection phases and amplitudes of the “0” and “1” elements in the dissolved and (e) deposited states under normal incidence.

After completing the design of “0” and “1” elements, we can manipulate electromagnetic waves by coding “0” and “1” elements with controlled coding sequences to achieve different functions [23]. In order to achieve the function of microwave scattering reduction, we adopted a common chessboard-like configuration here, where the coding sequence is “101,010 …” along both the x and y directions. Figure 3(a) depicts a schematic diagram of the proposed metadevices. To mitigate electromagnetic coupling between adjacent unit cells due to geometric variations, we employed a super cell consisting of 6 × 6 identical unit cells representing “0” or “1” elements. The metadevice is composed of a 5 × 5 super cell, i.e., 30 × 30 unit cells. We performed simulations to evaluate the microwave scattering reduction performance of metadevice. Figure 3(b) and (c) and Figure S5a, b (Supplementary Material) illustrate the simulated 3D scattering patterns of the metadevice in the dissolved and deposited states, respectively, under normal incidence at 12.0 and 19.3 GHz. For comparison, we also simulated the 3D scattering pattern of a metal plate with the same size (Figure 3(d) and Figure S5c, Supplementary Material). Video S1 records the simulated 3D scattering patterns of the metadevice in the deposited and dissolved states and of a metal plate with the same size at 12.0 GHz. It can be observed that the metadevice scatters incident electromagnetic waves relatively uniformly in four symmetric directions, namely 45°, 135°, 225°, and 315°, contrasting sharply with the strong backward reflection along the normal direction exhibited by the metallic plate. The simulated 2D scattering patterns in Figure 3(e) and (f) illustrate that the metadevice achieves a reflection reduction of >10 dB in the backward direction at 12.0 and 19.3 GHz compared to the metal plate. The microwave reflection spectra of the metadevice in the dissolved and deposited states are simulated as shown in Figure 3(g), and the results indicate that the metadevice maintains excellent wideband microwave scattering reduction performance in both states. Note that the bandwidth of the reduction less than −10 dB ranges approximately from 10 to 20 GHz, which is consistent with that for the phase difference. For comparison, the reflection spectra of device without loading JC metal structure, i.e., RME variable emissivity device only, were also simulated (dashed line in Figure 3(g)), and it can be observed that loading JC metal structure effectively reduces the reflection of the device. Additionally, by optimizing the arrangement of the coding sequences, a diffusion scattering effects can be achieved [29], for which further simulation analysis has been shown in Figure S6 (Supplementary Material). Furthermore, it would be interesting to explore the possibility of introducing diverse patterns for multi-state signal differentiation in the microwave wavelength range by designing different coding sequences.

Figure 3:

Design of metadevice. (a) Schematic diagram of the proposed metadevice with a chessboard-like configuration. (b) Simulated 3D scattering patterns under normal incidence of the metadevice in the dissolved and (c) deposited states at 12.0 GHz. (d) Simulated 3D scattering patterns under normal incidence of a metal plate with the same size as the metadevice at 12.0 GHz. (e) Simulated 2D scattering patterns under normal incidence in the 45° plane of the metadevice in the dissolved and deposited states, as well as a same-sized metal plate, at 12.0 GHz and (f) 19.3 GHz. (g) Simulated microwave reflection spectra of the metadevice in the dissolved and deposited states under normal incidence. The dashed line represents the simulated reflection spectra of the device without loading JC metal structure, i.e., RME variable emissivity device only.

2.2 Microwave scattering reduction performance

To verify the microwave scattering reduction performance, we fabricated a metadevice sample with dimensions of 180 × 180 mm² and conducted reflection testing in a microwave darkroom using the NRL-arc method (Figure 4(a)). Due to the high cost of large-area BaF₂, we used glass (with a dielectric constant of 6.6 and a loss tangent of 0.003) as a substitute for BaF₂ as the substrate for the top electrode. Since they have similar dielectric constants, this substitution has little impact on the results of the microwave reflection testing. The fabricated metadevice in the dissolved and deposited states are shown in Figure 4(b) and (c). The measured geometric parameters of the etched JC metal structure in the fabricated metadevice are shown in Figure S7 (Supplementary Material). During the measurement, the reflection from a metal plate of the same size as the sample was initially measured for normalization. The real-time microwave reflection spectra of the metadevice can then be measured by controlling the deposition time. Figure 4(d) shows the reflection spectra of the metadevice in the dissolved and deposited states, and the real-time reflection spectra of the metadevice during the electrodeposition process are shown in Figure S8 (Supplementary Material). It can be observed that throughout the electrodeposition process, the metadevice exhibits relatively stable wideband microwave scattering reduction performance in 10–20 GHz. The measurement results are relatively consistent with the simulation results, although minor discrepancies may arise due to imperfections in sample fabrication and differences between the measurement and simulation environments. Furthermore, the surface morphology of the Pt film and electrodeposited Cu film was characterized (Figure 4(e) and (f)) and it can be clearly observed that a dense and uniform Cu deposition layer was formed on the Pt film in the deposited state.

Figure 4:

Microwave scattering reduction performance. (a) Measurement setup in a microwave anechoic chamber. (b) Photographs and micrographs of the fabricated metadevice in the dissolved state. (c) Photographs of the fabricated metadevice in the deposited state. (d) Measured microwave reflection spectra of the fabricated metadevice in the dissolved and deposited states under normal incidence. The dashed lines are simulation results. (e) SEM image of the Pt film. (f) SEM image of the Cu film formed on the Pt film in the deposited state.

2.3 Dynamic IR performance

To verify the dynamic IR performance, we fabricated a metadevice sample with dimensions of 70 × 70 mm² and recorded IR video of the fabricated metadevice during the electrodeposition and dissolution process using an IR thermal imager, as shown in Video S2 (Supplementary Material), and captured a series of IR images from the video, as shown in Figure 5(a) and Figure S9 (Supplementary Material). Figure 5(b) illustrates the apparent temperature curves of the metadevice during the electrodeposition and dissolution process. The real temperature of the metadevice was controlled at 50 °C using a hot plate during the experiment. Therefore, the change in apparent temperature implies a change in the IR emissivity of the metadevice. When the deposition voltage was applied, Cu film was gradually electrodeposited on the working electrode, and as the deposition time increased, the Cu film became thicker, leading to a corresponding decrease in the apparent temperature of the sample. When the dissolution voltage was applied, the deposited Cu film gradually dissolved until the initial state was restored, demonstrating reversibility. Due to the long-distance conductivity provided by the gold grid, the device exhibited relatively uniform IR modulation capability during the electrodeposition and dissolution process. In addition, we measured the real-time IR reflection spectra of the “0” element and “1” element regions of the sample (Figure S10, Supplementary Material) and calculated the real-time IR emissivity spectra (Figure 5(c) and Figure S11, Supplementary Material). Although the surface of the “1” element region of the device is covered with a JC metal structure array, its relatively small filling rate results in a limited impact on the IR emissivity modulation performance compared to the “0” element region. By integrating the spectral data, we calculated the maximum emissivity modulation of the metadevice in mid-wave IR and long-wave IR atmospheric transmission windows to be 0.58 and 0.50, respectively, demonstrating a relatively consistent level of emissivity modulation. Notably, various electrochromic materials, including those based on RME, have been reported to achieve reversible tuning in the near-IR band, which is an important guide for further exploration to broaden the tuning band [30], [31], [32], [33].

Figure 5:

IR reversible tunability performance. (a) IR images of the fabricated metadevices during the electrodeposition process. (b) Apparent temperature curves (blue rectangular region) of the fabricated metadevices during the electrodeposition and dissolution process. (c) Real-time IR emissivity spectra of the “1” element region of the metadevice during the electrodeposition process. The general trend is increasing IR emissivity with increasing deposition time. (d–g) Adaptive IR camouflage demonstration. The device can adjust its IR emission to blend into (d) cold and (f) hot backgrounds. The graphs show the temporal evolution of the apparent temperature of the device during adaptation into (e) cold and (g) hot backgrounds with the red dashed line representing the apparent temperature of the background.

To showcase the prospects of the metadevice, we conducted a demonstration of its adaptive IR camouflage potential in a laboratory setting. Samples with different emissivity were placed on a hot plate at 50 °C to create distinct IR backgrounds. The metadevice was then placed on each background, and the process of the metadevice actively blending with the background was recorded using an IR thermal imager, as shown in Video S3 and Video S4 (Supplementary Material). Figure 5(d) shows IR images of the metadevice as it adapts its IR emission to cold backgrounds. Here, ITO glass with a sheet resistance of 15 Ω/sq was utilized as the cold background. In the dissolved state (initial state), the device exhibited a high IR emission state, resulting in an apparent temperature significantly higher than the background, making it prone to exposure. By applying the deposition voltage, the metadevice can switch to a low IR emission state, effectively blending into the cold background environment and achieving camouflage. Similarly, we utilized glass as the hot background, and presented IR images of the metadevice as it adapts its IR emission to hot backgrounds in Figure 5(f). The temporal evolution of the apparent temperatures of the metadevice and background, as shown in Figure 5(e), (g), not only quantifies the adaptation of the device’s IR emission to the background, but also reveals its high transition speed (∼4 s). In fact, by precisely controlling the deposition time, placing the metadevice in an intermediate emissivity state enables integration with more IR backgrounds. Furthermore, by incorporating perception and control systems into the metadevice, human intervention can be avoided, enhancing the device’s adaptive capabilities [34]. Considering the adoption of pixelated architecture design [35], [36], segmented addressing techniques [37], and intelligent control algorithms [26] to achieve pixelized addressing control in devices, is also crucial for the development of devices in fields such as IR digital camouflage and information display.

4.1 Simulations

Full-wave simulations of the metadevice were carried out using the time-domain solver in CST Microwave Studio 2022. The incident plane waves were normally incident upon the devices from +Z direction. The boundaries along x, y and z directions were set as open add space. Far-field monitor is set at 12.0 GHz and 19.3 GHz to get the scattering pattern.

4.2 JC metal structure preparation

A gold film with a thickness of 300 nm was initially evaporated on BaF₂ substrates using an electron-beam evaporation system (Kurt J. Lesker PVD 75) with a deposition rate of 0.08 nm s⁻¹ at room temperature, and then the gold film was etched using a laser etcher (SC-K750) to form the JC metal structure.

4.3 Working electrode preparation

The BaF₂ substrates with a thickness of 2 mm were used as the working electrode substrates. The gold grid with 200 nm line thickness, 300 µm line width, and 12 mm line spacing was directly evaporated on the BaF₂ substrates with the aid of a stainless steel mask plate using an electron-beam evaporation system (Kurt J. Lesker PVD 75) with a deposition rate of 0.08 nm s⁻¹ at room temperature, and then the Pt film with a thickness of 4 nm was further evaporated with a deposition rate of 0.02 nm s⁻¹ at room temperature. The nominal thicknesses of the gold grid and Pt film were determined by extrapolation of the deposition rate, in turn calibrated using a quartz crystal oscillator.

4.4 Gel electrolyte preparation

The gel electrolyte was prepared by mixing 0.08 M CuCl₂ (Aladdin), 0.25 M LiClO₄ (Aladdin), 0.6 mM KI (Aladdin), and 10 wt% PVB (Sinopharm Chemical Reagent) in DMSO (Aladdin) solution. All chemicals were mixed and stirred on a hot plate set to 60 °C and 900 rpm for at least 12 h.

4.5 Device assembly

The metadevice was constructed by using BaF₂ substrates with evaporated gold grid and Pt film as the working electrode. Cu foil with a thickness of 0.05 mm was used as the counter electrode. The perimeters of the working and counter electrodes were framed with Cu tape to provide uniform electrical contacts. Polyamide tapes were used to seal the Cu tape and prevent contact with the electrolyte. A double-sided tape with a thickness of 0.1 mm was used to seal the edges of the electrodes, which worked as the frame of the devices. Silicone adhesive sealant was applied to the edges of the devices. The electrolyte was subsequently injected into the device.

4.6 Characterization

The sheet resistance of the evaporated Pt films was measured using a four-probe resistivity measurement system (RTS-9). The complex permittivity of gel electrolyte was measured using a dielectric performance probe (Keysight N1501A) connected to a network analyzer (Agilent E8363C). The microwave reflection measurements were conducted using broadband standard gain horn antennas connected to a network analyzer (Agilent E8363C). The samples were photographed using a digital camera (Canon EOS M50 Mark II). The SEM images were obtained with field-emission scanning electron microscope (MIRA3 AMU). The IR reflection spectra (3–14 µm) were measured using a Fourier transform IR (FTIR) spectrometer (Bruker Vertex 70) equipped with a mid-IR integrating sphere (A562). The IR images were recorded using an IR thermal imager with a working range of 7.5–14 µm (FLIR T1050sc), with predefined emittances set to 0.95. The apparent temperature curves the devices in IR images were extracted by the box measurement tool in the FLIR software packages (FLIR Tools V 5.7). A PARSTAT 4000 Advanced Electrochemical System (Princeton Applied Research, USA) was used for electrochemical measurements and device performance demonstration.