Many organisms in nature, such as chameleons, possess the remarkable ability to rapidly alter their skin color to blend into their surroundings, serving as an effective camouflage mechanism for protection or predation [1,2]. Similarly, in recent years, the pursuit of adaptive camouflage in the infrared (IR) spectrum has garnered significant attention, leading to the development of diverse materials and devices capable of dynamically regulating IR thermal radiation [3–5]. According to the Stefan–Boltzmann law, they can be classified into two main categories: temperature-based regulation and emissivity-based regulation. While temperature-based regulation strategies typically utilize microfluidics [6] or thermoelectric systems [7,8] to achieve IR radiation control, they suffer from high energy consumption and necessitate additional heating and cooling mechanisms. The use of thermochromic phase-change materials such as vanadium dioxide (VO2) [9–11] or Ge2Sb2Te5 (GST) [12,13], which exhibit changes in emissivity with temperature variations, has also been extensively reported for IR radiation modulation. However, when addressing the adaptive IR camouflage requirements that necessitate real-time adjustment of their own IR radiation characteristics based on background IR features, similar challenges to temperature-based modulation strategies are encountered, and stepless modulation of emissivity is difficult to achieve. In contrast, IR electrochromic materials or devices, due to their capacity for reversible dynamic changes in IR emissivity under external electrical stimuli, offering flexible regulation, rapid response times, simple structures, and low energy requirements, have emerged as a promising avenue [14–16]. IR electrochromic devices can be broadly classified into four main types based on different material systems: metal oxides [17,18], conducting polymers [19,20], carbon materials [4,21], and metals [3,22]. Among the crucial performance indicators for these materials, the IR emissivity regulation range (Δε) holds paramount importance. Benchmarking studies (see Appendix A) highlight that metal-based IR electrochromic devices selected in this work exhibit the most superior levels of emissivity regulation.

Nevertheless, it is abundantly clear that current electro-optical detection extends beyond a single frequency band, with detectors operating at different wavelengths integrated, rendering traditional single-band camouflage materials ineffective [23–27]. Radar detection is also an important means of detection in modern warfare. Therefore, the radar stealth compatibility issue of IR electrochromic devices cannot be ignored. However, the modulation of various IR electrochromic devices, including those based on metals, necessitates external electrical stimuli, leading to the inevitable presence of electrodes in the device structure, such as gold (Au) [4,17], indium tin oxide (ITO) [3], and graphene [4] (see Appendix A), and in order to achieve fast and uniform color change, these electrodes typically exhibit excellent conductivity. The strong microwave reflective properties inherent in the electrodes make them susceptible to radar detection, thereby compromising their camouflage effect, which poses a serious challenge for device applications. However, to the best of our knowledge, there have been no reported solutions to this challenge to date.

Frequency selective surfaces (FSSs) are a type of artificial electromagnetic functional materials that, through the design of subwavelength periodic structures, can exhibit unique electromagnetic characteristics not found in natural materials, and thus belong to a class of metamaterials [24,28]. FSSs made of lossy materials are widely used in absorber design, commonly known as circuit analog absorbers [29]. These absorbers typically feature a sandwich structure comprising an upper FSS structure, an intermediate dielectric layer, and a lower ground plane. Leveraging impedance matching and electromagnetic resonance mechanisms, FSS absorbers can achieve wideband and efficient absorption of incident electromagnetic waves to attenuate the return signal strength of the target, which is widely used in stealth technologies.

Inspired by the remarkable electromagnetic control capabilities of metamaterial, we propose the integration of FSS into IR electrochromic devices to address the challenge of device radar stealth compatibility. Specifically, we design and experimentally validate a monolithically integrated dual-functional metadevice based on a reversible metal electrodeposition (RME) IR electrochromic device, sharing a barium fluoride (BaF2) substrate and conductive electrodes. The rational and optimized design of the metadevice structure simultaneously achieves large and reversible electrical modulation of the IR emissivity and wideband high-efficiency microwave absorption, while minimizing mutual interference between these two functions. We believe that this design concept offers valuable insights for addressing multispectral compatibility challenges in various types of IR electrochromic devices and holds potential for application in the advancement of advanced adaptive multispectral camouflage systems.

Figures 1(a) and 1(b) illustrate the structure of the metadevice and its operation principle. Integrating FSS with Jerusalem cross (JC) structure with a metal-based IR electrochromic device constitutes our proposed metadevice structure. This integrated design achieves the function of a wideband microwave absorber while retaining the tunable IR emissivity characteristics of the electrochromic device, capable of countering both radar and IR detection. The RME IR electrochromic device employs a reversible copper (Cu) deposition system, serving as an electrochemical setup with a typical working electrode–electrolyte–counter electrode sandwich structure. Specifically, a nanoscopic platinum (Pt) film (∼4nm thick, with a sheet resistance of ∼150Ωsq−1) was evaporated on the IR-transparent BaF2 substrate as the working electrode, a Cu foil as the counter electrode, and a gel electrolyte containing Cu2+ ions sandwiched between the two electrodes. The metal-based IR electrochromic device operates on the electrochemical phase change between dissolved Cu2+ ions in the electrolyte and Cu films that plate onto the working electrode, thus realizing the modulation of the IR emissivity of the metadevice. An ITO FSS with JC structure is loaded on the BaF2 substrate, forming an absorber with a sandwich structure comprising the BaF2 substrate acting as the dielectric layer and the electrode serving as the ground plane. While our study primarily utilizes a Cu electrodeposition system for the design of the metadevice, it is noteworthy that this approach is equally applicable to other deposition systems such as silver, bismuth, and others. This versatility underscores the broader applicability and potential advantages of the methodology we propose. In addition to BaF2, common IR-transparent substrates include silicon (Si), polypropylene (PP), and polyethylene (PE). For this metadevice structure, BaF2 serves not only as an IR-transparent substrate but also as the dielectric layer in the absorber structure. Therefore, the microwave electromagnetic parameters and thickness of the IR-transparent substrate significantly influence the device’s microwave performance. Si, with a microwave dielectric constant of ∼11.9, presents challenges in achieving broadband impedance matching. PP and PE show promise for achieving large-area device fabrication but exhibit optimal IR transparency only in thin film states, thereby failing to meet the thickness requirements as a dielectric layer in the absorber. Hence, we have chosen BaF2 as the IR-transparent substrate here, despite the challenges of large-scale production and high cost. Developing low-cost, scalable IR-transparent substrates remains crucial.

To decouple the IR tunability and microwave absorption functions, two considerations were deliberately taken into account during the design of the metadevice. First, given the low emission characteristics of the ITO FSS structure with metallic properties, the electrochromic effect in the covered region is inevitably greatly reduced. Therefore, the FSS structure should be designed with the smallest possible filling ratio to minimize its impact on the IR electrochromic performance of the metal-based device. Second, for the microwave absorber, the electrodeposition process leads to a transformation of the ground plane from the counter electrode to the working electrode. In order to mitigate the influence of the electrodeposition process on the wideband microwave absorber, we controlled the thickness of the electrolyte layer to be as small as possible, with d2=0.1mm.

Further, we simulated and optimized the microwave absorption performance of the metadevice in CST Microwave Studio and obtained the optimized structural parameters as repeating period p=6mm; geometrical parameters of the FSS with JC structure l=3.5mm, s=1.25mm, and w=0.14mm; sheet resistance of the FSS with JC structure Rs−FSS=6Ωsq−1; and thickness of the BaF2 substrate d1=2.0mm. The corresponding simulated results are illustrated in Fig. 1(c). For comparison, the reflection spectra of the RME devices without the FSS structure were also simulated, and the results indicated that the RME devices exhibited high microwave reflection in both dissolved and deposited states due to the reflective properties of the electrodes. In contrast, the device with integrated FSS structure exhibited a reflection loss (RL) ≤−10dB (absorption ≥90%) in the frequency range of 9–18 GHz, which verifies the validity of our design. Meanwhile, the filling ratio of the FSS structure was calculated to be controlled below 4.45%, so the maximum IR emissivity regulation range of the designed metadevice should be able to reach more than 95% of the metal-based IR electrochromic device without FSS.

To investigate the dynamic IR regulation mechanism of the metadevices, the IR spectral response of the RME devices during the electrodeposition process was calculated and analyzed. We prepared samples of the Pt films with a thickness of ∼4nm on a Si wafer and samples of Cu films with a thickness of ∼40nm electrodeposited on Pt films on a Si wafer. Scanning electron microscopy (SEM) images of the samples are shown in Figs. 2(a) and 2(b), demonstrating that the metal films were dense and uniform. Further, the IR optical constants, refractive indices “n” and extinction coefficients “k” of the Pt film, the deposited Cu film, and the BaF2 substrate were fitted using ellipsometry [Figs. 2(c)–2(e); see Appendix C for ellipsometric parameters]. Pt films and deposited Cu films exhibit large extinction coefficients “k” that typically surpass their refractive indices “n,” with the extinction coefficients increasing as the wavelength increases. Using the simulated IR spectral response results of the Pt films with a thickness of ∼4nm as an illustration, it is obvious that nanoscale Pt films demonstrate high IR absorption due to phenomena such as multiple reflection and scattering as well as localized surface plasmon resonance [Fig. 2(f)] [3,30]. It should be noted that this absorption manifests prominently in a wide wavelength range, primarily attributed to the significant surface roughness of nanoscale Pt films prepared by electron beam evaporation, which facilitates multiple reflection and scattering in the IR wavelength range [31]. Additionally, the non-uniformity of the metal film surface may give rise to interactions among localized surface plasmons or multiple resonance modes [30,32,33]. Moreover, the presence of free electrons in the metal film is influenced by quantum size effects, restricting their motion and altering the band structure and density of states [34]. The combined effect of these phenomena leads to enhanced absorption over a broader spectral range. Further, the IR spectral response of Cu electrodeposited Pt films, formed during the electrodeposition process, was studied through simulations by varying the thickness of the deposited Cu films [see the inset in Fig. 2(g)]. The spectral evolution of the Cu electrodeposited Pt films at 8 μm, as illustrated in Fig. 2(g), shows that as the thickness of the deposited Cu film increases, the absorption and transmission decrease, which translates into an increasing reflection, and the variation gradually levels off when the thickness of the deposited Cu film reaches 20 nm. The analysis reveals that the deposited Cu film alters the ultrathin metallic morphology of the nanoscale Pt film, leading it to exhibit high IR reflection characteristics similar to that of a thick metal film, i.e., thus showcasing low IR emissivity. Based on the fitted IR optical constants, the reflection spectra of the electrodeposited Cu films with different thicknesses on the working electrode were calculated [Fig. 2(h), using TFCalc Thin Film Design Software]. Limited by the IR transmission band of BaF2, the devices showed regulation performance in the range of 2.5–11 μm. As the thickness of the electrodeposited Cu film increases (from 0 nm to 40 nm), the IR reflection of the device exhibits a nonlinear increase from ∼0.2 to ∼0.85, and the increase gradually levels off when the thickness of the deposited Cu film reaches 20 nm.

In the design of metadevices, in order to mitigate the potential negative impact of electrodeposition on the wideband microwave absorption performance of the metadevices, we strictly controlled the thickness of the electrolyte layer to 0.1 mm. Here, we simulated and analyzed the dependence of the microwave absorption performance of the devices in both deposited and dissolved states on the electrolyte layer thickness (d2) by parameter scanning. We observed a significant decrease in the efficient absorption bandwidth of the device in the dissolved state when the electrolyte layer thickness exceeded 0.2 mm, accompanied by a shift of the absorption peak towards lower frequencies as the thickness increased [Fig. 3(a)]. This shift is primarily attributed to changes in resonance frequency resulting from variations in thickness. However, the thickness variation had a minimal impact on the absorption performance of the device in the deposited state, showing only slight attenuation around 13.5 GHz [Fig. 3(b)]. This is due to the fact that the conductivity of the Cu electrodeposited Pt film in the deposited state is already excellent enough to act as a ground plate in the absorber. This analysis underscores the necessity of controlling the electrolyte layer thickness to circumvent the effect of the electrodeposition process on the absorption performance of the metadevice. In addition, the dependence of the reflection spectra of the metadevice on the incident wave’s polarization angle and incidence angle, as well as on the main structural parameters, was also demonstrated (Appendix D).

To explore the wideband microwave absorption mechanism of metadevices, both impedance matching and loss mechanisms have been investigated. Generally, if the impedance of an absorber (Z=Zre+Zimi) matches the free-space impedance (Z0=1+0i), that is, the real part Zre is close to one and the imaginary part Zim is close to zero, it implies that the incident electromagnetic wave can penetrate into the interior of the absorber successfully rather than being reflected at the surface. Here, we choose the metadevice in the deposited state for analysis, and the impedance results of the metadevice are calculated using the S-parameters obtained from the simulation, which shows a better impedance matching in the wideband range of 9–18.2 GHz, especially at the two absorption peaks [see Fig. 3(c)]. In order to analyze the absorption mechanism and better illustrate the function of the FSS structure, the distributions of electric field, magnetic field, surface current, and power loss density of the metadevice in the deposited state at the resonant frequency of 10.28 GHz were monitored [Figs. 3(d)–3(g), using CST Microwave Studio]. The FSS structure has a strong electromagnetic field distribution, exhibiting enhanced surface currents, which in turn enhances the ohmic losses in the FSS etched by the ITO film. Figures 3(h)–3(k) provide the corresponding distribution at the resonant frequency of 17.2 GHz, which exhibits similar absorption mechanisms at both resonance frequencies. Therefore, the combined effect of wideband impedance matching and strong ohmic losses achieves the wideband absorption properties of the metadevice.

To examine the dynamic IR regulation performance of the metadevice, a sample with dimensions of 70mm×70mm was fabricated, as shown in Fig. 4(c). In order to improve the long-range conductivity of the working electrode to ensure fast and uniform tinting of the large-area device, an Au grid was evaporated on the BaF2 substrate prior to the evaporation of the platinum electrode (see Section 4 for details). Initially, the “real-time” IR reflection spectra of the device were measured by controlling the deposition time [see Fig. 8(a) in Appendix E], and the “real-time” IR emissivity spectra at 3–14 μm were calculated [Fig. 4(a); see Appendix B for the calculation details]. By integrating the spectral data, we calculated the maximum IR emissivity regulation range of the metadevice (Δε3−5μm=0.57, Δε7.5−13μm=0.48; see Appendix B for the calculation details). For comparison, the maximum IR emissivity regulation range of the RME IR electrochromic device without FSS was also calculated [Δε3−5μm=0.69, Δε7.5−13μm=0.53; see Fig. 8(b) in Appendix E for the IR emissivity spectra]. The incorporation of the FSS structure evidently limited the high emissivity of the metadevice in the dissolved state and had no significant effect on the low emissivity in the deposited state, which ultimately led to a decrease of ∼0.1 in the emissivity tunability of the metadevice compared to that of the RME device, which was slightly higher than the calculated value based on the FSS filling ratio. It was analyzed that this could be due to factors such as the thickness tolerance of the Pt electrodes and the added Au grid. Furthermore, thermal video of the metadevice samples during electrodeposition and dissolution was captured using an IR camera (see Visualization 1), and selected frames from the thermal video are depicted in Fig. 4(b). The metadevices exhibit a relatively uniform tinting during electrodeposition, as well as a decent switching speed (∼10s).

To evaluate the wideband microwave absorption performance of the metadevice, a sample with dimensions of 180mm×180mm was fabricated, as depicted in Figs. 4(f) and 4(g). It is noted that due to challenges and high costs in obtaining large-area BaF2 substrates, glass was utilized instead as the substrate for the working electrodes. Since glass and BaF2 share similar dielectric constants in the microwave frequency range, this substitution had minimal impact on the microwave test results. A micrograph of the etched FSS with JC structure is illustrated in Fig. 4(e). The reflection spectra of the device in the deposited and dissolved states were measured in a microwave anechoic chamber using the NRL-arc method [Fig. 4(d)]. As illustrated in Fig. 4(h), the device exhibited a certain degree of microwave absorption characteristics in both states. However, it was evident that the test results did not perfectly agree with the simulation results, and the electrodeposition process significantly influenced the reflection spectra. Further analysis indicated that these deviations primarily resulted from imperfections in sample fabrication, particularly the inadequate control of the electrolyte layer thickness (see Appendix F for detailed analysis). Nevertheless, despite the imperfect agreement between measured and simulated results, it still provided sufficient evidence of the feasibility of achieving wideband microwave absorption through this metadevice structure design.

To show the application potential of metadevices in the field of adaptive IR camouflage, a demonstration was carried out under laboratory conditions. The demonstration system was as follows. An ITO glass (15Ω/sq) and a glass plate were placed on a 50°C hot plate to generate cold and hot IR backgrounds, respectively. The metadevice was placed in the center on the background. A power supply was used to control the switching of device states. An IR thermal imager was used to record the thermal video. Visualization 2 and Visualization 3 show the active fusion process of the metadevice with cold and hot backgrounds, respectively. Figures 5(a) and 5(c), selected frames from the thermal video, show the IR images of the metadevice in different backgrounds exhibiting the uncamouflaged and camouflaged effects. The apparent temperature time traces in Figs. 5(b) and 5(d) quantify the fusion process of the device with cold and hot backgrounds, respectively, and also demonstrate the good switching speed of the device (<10s). Going forward, we can pixelate the device and further integrate the sensing and control system to achieve adaptive camouflage that does not require human intervention and is capable of facing complex backgrounds [35,36].

Furthermore, the IR emissivity tunability and wideband microwave absorption of previously reported IR stealth materials or devices (including IR electrochromic devices [3,4,15,17,20] and IR-radar compatible stealth materials [24,37–39]) are compared with our designed metadevice (Table 1). Compared with other IR camouflage materials, our designed metadevice achieves the dual functions of IR emissivity tunability and wideband microwave absorption simultaneously in a single device. In addition, even though the integration of FSS leads to a slight impact on the IR tunability of the metadevice, it still exhibits large and consistent emissivity tunability in 3–5 μm and 7.5–13 μm atmospheric transmission windows compared to IR electrochromic devices based on other material systems. Compared with IR-radar compatible stealth materials, which typically pursue fixed low IR emissivities, our designed metadevice offers emissivity tunability, and it also has low emissivity values in the deposited state, measuring 0.21 at 3–5 μm and 0.31 at 7.5–13 μm, respectively. In conclusion, the combination of both IR emissivity tunability and wideband microwave absorption expands the application scenarios of the device and shows unique potential for the development of adaptive multispectral camouflage systems.Table 1.

Comparison of the Proposed Metadevices with Previously Reported IR Stealth Materials or Devices

In conclusion, inspired by the extraordinary electromagnetic wave manipulation capability of metamaterials and their extensive applications in stealth technology, the idea of integrating metamaterial FSS designs into IR electrochromic devices is proposed to address their multispectral compatibility challenges with radar stealth. By integrating the FSS design into the metal-based IR electrochromic device, a metadevice is proposed that simultaneously achieves electrically tunable IR emissivity and wideband microwave absorption, enabling compatibility with dynamic IR and radar stealth. This novel design idea provides a feasible technical path for the multispectral compatibility problem of emerging electrochromic devices and provides valuable experience for the development of adaptive multispectral camouflage systems. Looking ahead, we expect to develop stable large-area flexible multispectral compatible metadevices by adapting to flexible substrates and optimizing the electrodeposition system, etc., and expect that the technology can be widely applied.

Simulations: Full-wave simulations of the metadevice were carried out using the frequency-domain solver in CST Microwave Studio 2022. The incident plane waves were normally incident upon the devices from the +Z direction. The boundaries along the x and y directions were set as a unit cell.

Working electrode preparation: A BaF2 substrate with a thickness of 2 mm was used as the working electrode substrates. An ∼300nm thick ITO film was evaporated on one side of the BaF2 substrate and then etched with a laser etcher (SC-K750) to form the FSS structure. On the other side of the BaF2 substrate, an Au grid with a line thickness of 200 nm, a line width of 300 μm, and a line spacing of 12 mm was first evaporated with the aid of a stainless steel mask plate, and then a Pt film with a thickness of 4 nm was evaporated. The nominal thicknesses of the Au grid and Pt film were determined by extrapolation of the deposition rate and then calibrated using a quartz crystal oscillator.

Gel electrolyte preparation: The gel electrolyte was prepared by mixing 0.08 mol/L CuCl2 (Aladdin), 0.25 mol/L LiClO4 (Aladdin), 0.6 mmol/L KI (Aladdin), and 10% (mass fraction) PVB (Sinopharm Chemical Reagent) in DMSO (Aladdin) solution.

Device assembly: Cu foil with a thickness of 0.05 mm was used as the counter electrode. The perimeters of both 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. The edges of the electrodes were sealed using double-sided tape, functioning as the device’s frame, and further secured with silicone adhesive sealant. Finally, the electrolyte was injected into the device.

Characterization: The sheet resistance of the Pt films was measured using a four-probe resistivity measurement system (RTS-9). The IR optical constants of the BaF2 substrate, Pt film, and deposited Cu film were measured and fitted using a spectroscopic ellipsometer (IR-VASE MARK II Ellipsometer; J. A. Woollam Co.). 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 a field-emission scanning electron microscope (MIRA3 AMU). The IR reflection spectra (2.5–20 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 of 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) and a DC stabilized power supply (UTP1306-II) were used for device performance testing and demonstration.