With the diversification, intelligence, and high precision of modern detection technologies, the combination of laser, visible light, infrared (IR), and microwave detection technologies has begun to be widely used. This has led to higher requirements for stealth technologies, especially for multispectral stealth, where radar-IR compatible stealth research is gaining significant attention. The challenges in this field are evident: radar stealth requires materials with low reflectivity and high absorptivity, whereas IR stealth demands low emissivity and high reflectivity, creating a fundamental contradiction between the two. Furthermore, with the rapid development of AI detection technologies, weaponry faces increased threats; spectrally tunable stealth technologies have become an inexorable trend in the future. The rapid development of spectrally tunable materials, such as tunable metamaterials [1–3], phase change materials [4–6], and electrochromic/thermochromic devices [7–11], has made dynamically tunable radar-IR compatible stealth materials a current research hotspot.

Research on radar-IR compatible stealth primarily focuses on two approaches: one is to design multilayer structures with high metal surface coverage or specific frequency selective surfaces (FSSs) combined with radar-absorbing materials; the other is to coat specific low-emissivity layers on radar stealth materials. For example, Zhang et al. [12] designed a multilayer metamaterial structure that achieves radar-IR stealth compatibility through a metal base layer, an annular ITO resistive layer, and a high-coverage metal copper sheet layer. Li et al. [13] proposed a bilayer structure comprising an FSS layer, an isolation layer, a resistive periodic surface layer, and a dielectric substrate. Xu et al. [14] proposed an optically transparent ITO/dielectric/ITO sandwich structure that achieves transparency, low IR emissivity, and microwave absorption. Pang et al. [15] designed a coding metasurface that reduces microwave band reflections through signal interference cancellation and IR emissivity control using a high-filling-ratio metal FSS structure on the top layer. Meng et al. [16] replaced metal materials with ITO to create a transparent coding metasurface for radar signal diffuse reflection. Liu et al. [17] designed a radar-IR compatible stealth metamaterial comprising an IR shielding layer (ISL) and a microwave abnormal reflection layer (MARL), achieving both low IR emissivity and low microwave reflectivity. Cui et al. [18] designed an optically transparent microwave-absorbing material with IR camouflage functionality, adjusting emissivity by changing the filling ratio of the IR shielding layer. Sun et al. [19] proposed a noninvasive inset-integrated meta-atom for achieving a single-layer metasurface simultaneously with coding microwave reflectivity and digitalized IR emissivity, but the performance of the microwave and IR should be further improved.

Most current research on metamaterial-based radar-IR compatible stealth typically achieves this by adding an IR shielding layer (IRSL) on top of a radar stealth layer (RSL). Although this design allows for independent control of IR and radar stealth, resolving the conflicts between traditional stealth materials, it also leads to increased structural thickness and processing complexity. In practical applications, there are relatively few researches that integrate these two functionalities into a unified design, limiting the potential performance of metamaterials. Moreover, relying solely on low-emission materials to reduce IR emissivity is insufficient to achieve effective IR camouflage.

In this paper, we propose a bispectral camouflage metasurface with microwave diffuse reflection and tunable IR emissivity. This metasurface integrates an IRSL with an RSL, achieving both radar stealth and spatially tunable IR emissivity, thereby enabling effective radar-IR compatible multispectral camouflage. The functional layer of the structure is realized by etching indium-tin-oxide (ITO) units on a polyethylene terephthalate (PET) substrate, forming an encoding metasurface for integrated design. ITO encoding controls electromagnetic (EM) wave reflection, achieving microwave diffuse reflection. By adjusting the ITO fill ratio, IR emissivity can be tuned while maintaining microwave stealth. Additionally, microwave diffuse reflection effectively prevents the increase in thermal load caused by microwave absorption, and the thermal insulation provided by the aerogel felt effectively reduces surface radiation due to thermal conduction. The innovation and practicality of this integrated design offer significant application prospects in the field of multispectral compatible stealth.

In this paper, we propose an integrated metasurface design to achieve RCS reduction and IR emissivity tunability. The structure consists of the following layers from bottom to top: aerogel felt/PET/ITO/air/PET/ITO, as shown in Fig. 1(a). The side length of the patch is a, and the sheet resistance of the ITO layers is 5 Ω/sq. The substrate material is PET with a thickness of dp, the air layer has a thickness of da, and the aerogel felt has a thickness of dq. In the “0” unit, the outer ring, inner ring, and patch spacing are b, t, and c, respectively. For the “1” unit, the outer ring, inner ring, and patch spacing are 3b, 3t, and 3c, respectively. The bottom layer of the aerogel felt is made of silica (SiO2) and offers excellent thermal insulation, with a heat resistance of up to approximately 300°C. Its primary function is to provide insulation, reduce heat transfer, and lower surface temperature.

For radar stealth design, we employed the method of reducing the RCS using anomalous reflection metasurfaces. This principle is based on phase cancellation of reflected EM waves by altering the geometric structure and dimensions of the unit. We designed two units with opposite phase differences and controlled the reflection direction of EM waves by combining specific encoding sequences, thus achieving RCS reduction [16,20–22]. When a plane wave is incident perpendicularly, the far-field directivity function of the metasurface is as shown in Eq. (1) [23]: {g(θ,φ)=ge(θ,φ)∑i=1n∑i=1nexp{j{φ/(m,n)+kDsinθ[(m−12)cosφ+(n−12)sinφ]}},Dir(θ,φ)=4π|g(θ,φ)|2/∫02π∫0π/2|g(θ,φ)|2sinθdθdφ,(1)where θ and φ are the azimuth and elevation angles in any direction, respectively. g(θ,φ) is the directivity function of the grid, and Dir(θ,φ) is the directivity coefficient of the metasurface. Using Eq. (1), the far-field scattering pattern of the coding metasurface corresponding to any coding sequence can be calculated, allowing the design to achieve the desired scattering mode.

For the design of adjustable IR emissivity, we utilized the synergistic effect of controlling surface temperature and surface emissivity of metamaterials. Surface temperature control relies on the heat transfer inhibition of thermal insulation materials, while surface emissivity control is achieved through the adjustable duty cycle of the ITO array. The IR emissivity of the metasurface can be calculated as follows [24]: ε=ε1f1+ε2f2,(2)where ε, ε1, and ε2 represent the emissivity of the unit structure, material 1, and material 2, respectively, and f1 and f2 are the occupancy ratios of material 1 and material 2. In this paper, the unit structure is the designed metasurface, material 1 is ITO, and material 2 is PET. From the above formula, it can be observed that adjusting the occupancy ratios of the two materials can regulate the IR emissivity of the structure surface. Specifically, altering the occupancy ratio of surface ITO can modulate the IR emissivity of the structure surface, thereby achieving emissivity adjustment by varying different occupancy ratios.

The essence of achieving spatially adjustable IR emissivity is to realize IR camouflage. The principle of IR camouflage lies in coordinating the radiative energy of materials with the surrounding environment, allowing the materials to blend effectively into the IR background. According to the Stefan-Boltzmann law, the IR radiation intensity of an object is related to its own temperature and IR emissivity, expressed in the following relationship: P=εσT4,(3)where P is the IR radiation intensity, ε is the target’s IR emissivity, σ is the Stefan-Boltzmann constant, and T is the surface temperature of the target. According to Eq. (3), the IR radiation intensity is related to the surface IR emissivity and surface temperature of the target. Therefore, two technical approaches can be employed to achieve IR camouflage: first, by controlling the surface temperature of the target to reduce the temperature difference with the background environment; second, by adjusting the surface IR emissivity to minimize the difference in radiation intensity between the target and the background.

First, an analysis of the temperature factor is carried out. The surface temperature is primarily influenced by heat transfer from the environmental temperature, thermal interactions of the protected target, and the absorption of external electromagnetic waves. The approach taken in this paper is to control the surface temperature so that it is primarily influenced by heat transfer from the environmental temperature, thereby reducing the difference from the surrounding temperature. Consequently, microwave stealth mainly utilizes a diffuse reflection mechanism to mitigate surface temperature increase caused by absorbed electromagnetic waves, while the use of underlying aerogel materials helps reduce the impact of heat conduction from the protected target to its surface. Second, an analysis of the adjustable IR emissivity is conducted. Regarding the camouflage effect of spatially adjustable IR emissivity, it can be analogized to the optical camouflage effect. Among them, different IR emissivity regions correspond to the “color patches” in the optical spectrum and are called IR “color blocks”. In specific IR backgrounds, these IR “color blocks” can be strategically combined to effectively conceal the protected target. This analysis is based on the perspective of IR emissivity.

To design a metasurface that simultaneously achieves tunable IR emissivity and RCS reduction, we first implemented microwave diffuse reflection via a coding metasurface to achieve RCS reduction. Additionally, we incorporated tunable IR emissivity and thermal insulation into the designed metasurface.

To meet the aforementioned requirements, we first designed two structural units with opposite phase differences, referred to as the “0” unit and the “1” unit. Figures 1(d) and 1(e) show the schematic diagrams of the designed “0” and “1” units. We adopted a checkerboard arrangement to place the “0” and “1” units on the metasurface to achieve a reduction in RCS. To alleviate the EM coupling issue between adjacent encoding particles caused by the geometric differences between the “0” and “1” units, we used a supercell structure composed of 3×3 identical encoding particles for encoding. Figure 1(b) shows the designed encoding metasurface, which consists of 6×6 supercells arranged in a checkerboard pattern, with the coding sequence in both the x and y directions being “010101…”. To achieve tunable IR emissivity, we divided the metasurface into four regions (parts A, B, C, and D), and varied the ring size to constuct different ITO filling ratios for each region. We arranged the “0” and “1” units in a checkerboard pattern on the metasurface to achieve RCS reduction.

Considering the different emissivity resulted from material ITO’s filling ratio on the surface of the “0” and “1” units, we used CST software to simulate and calculate the response of the two units. The optimized structural parameters are as follows: dp=0.22mm, da=2.22mm, dq=2.22mm, a=16.67mm, b=5.0mm, c=0.56mm, tA=0mm, tB=1.43mm, tC=0.77mm, tD=2.22mm.

Figure 2(a) shows the variation in reflection phase and amplitude response of the “0” and “1” units in the four parts within the 4–14 GHz frequency range. From Fig. 2, it can be observed that within the 4.5–11 GHz range, there are two units in each case with a phase difference of about 180°±30°, and within this range, the amplitude of the “0” and “1” units reaches a minimum. This indicates that there is an absorption peak for the “0” and “1” units within this frequency range, which is highly beneficial for RCS reduction.

To better understand the roles that absorption and scattering play in the reduction of RCS, we can use EM theory to calculate the percentages of reflection, absorption, and scattering. These percentages can be computed using the following formulas [25]: f={|ka0exp(jφ0)+(1−k)a1exp(jφ1)|2reflection1−|ka0+(1−k)a1|2absorption|ka0+(1−k)a1|2−|ka0exp(jφ0)+(1−k)a1exp(jφ1)|2diffusion,(4)where a0, φ0, a1, and φ1 represent the reflection amplitude and phase of the “0” and “1” units, respectively, and k and 1−k are the proportions of the “0” and “1” units in the overall structure.

During the calculation of the absorption, scattering, and reflection energy ratios of the designed metasurface in the 4–14 GHz range, the 180° phase difference between the “0” and “1” coding units can lead to cancellation effects in the theoretical calculation. However, in practice, when the phase difference falls within the range of 180°±30°, diffuse reflection can still be effectively controlled. In this case, due to the involvement of absolute value and square operations, the corresponding data will not completely cancel out; in fact, they may even increase. This discrepancy results in the theoretical values being higher than the actual measurements. Hence, the difference arises mainly from the variance between theoretical and practical conditions, and using actual measurement values for subsequent calculations provides a more accurate explanation of energy ratio in the range of 4–14 GHz, as shown in Fig. 2(b).

Figure 2(b) indicates that in the 4.5–10.3 GHz band, the reflectivity is less than 0.1, meaning that the proportion of EM wave energy being reflected is very small, with most of the energy being scattered. Figure 2(c) is a pie chart showing the integrated energy of absorption, scattering, and reflection in the 4.5–10.3 GHz band. It can be seen that 63% of the EM wave energy is scattered, 34% is absorbed, and only 3% is reflected. Figure 2(d) shows the top and side views of the surface current distribution of the metasurface at a frequency of 6 GHz. It can be seen that the directions of the surface currents on the upper and lower surfaces are opposite. These antiparallel currents form a circulating loop, resulting in a resonant effect that enables the metasurface to absorb EM waves. In summary, the metasurface effectively suppresses the backscattering of EM waves in the 4.5–10.3 GHz band, with most of the EM wave energy being scattered. Meantime, the reflection magnitudes and surface current distributions of the “0” and “1” units also indicate that the designed metasurface can absorb EM waves. Furthermore, the calculation results verify that the frequency band with the highest EM wave absorption corresponds to the positions where the amplitude of the “0” and “1” units reaches a minimum.

To verify the suppression effect of the designed metasurface on radar backscattering, we used CST to model and simulate the far-field scattering patterns. The results show that the designed metasurface exhibits good scattering performance around 4–10 GHz, with optimal radar backscattering suppression observed at 6 GHz. Figure 3 presents the far-field scattering patterns of the designed metasurface at 6, 8, and 10 GHz, compared to a metal plate, along with the reduction in radar cross-section (RCS) of the designed metasurface compared to the metal plate. Figures 3(a)–3(c) demonstrate that the reflected field energy from the designed metasurface is primarily distributed in four directions: 45°, 135°, 225°, and 315°. Analysis at 6 GHz, for example, shows that the energy in the incident direction for the designed metasurface is only −15.1dB(m2), while the energy reflected in the incident direction for the metal plate reaches 18.4dB(m2). Figure 3(d) is the simulation reflection curve of the designed metasurface reduced compared to the metal backplane at 2–18 GHz. The results show that the designed metasurface effectively inhibits radar backscattering and significantly reduces RCS.

Based on the unit parameters of the four parts of the surface, the ITO filling ratios for parts A, B, C, and D are 0.81, 0.79, 0.74, and 0.65, respectively. Under normal circumstances, the emissivity of ITO is 0.1 and that of PET is 0.9. According to Eq. (2), the corresponding emissivity is calculated as 0.25, 0.27, 0.31, and 0.38, respectively.

In summary, the designed metasurface achieves the functions of RCS reduction and IR emissivity tunability. However, it was observed that using simulated parameters resulted in a relatively narrow tunable range for the IR emissivity. Therefore, reducing the side length b of the ITO patches or increasing the side length of the inner ring t can decrease the ITO filling ratio and expand the tunable range. The corresponding far-field scattering simulation results are shown in Fig. 4.

In Fig. 4(a), with the specified parameters and only the ITO patch side length b changed to 4 mm, the tunable IR emissivity range extends from 0.25 to 0.48. In Fig. 4(b), with the parameters based on Fig. 4(a), the gradient increase in the inner ring side length t is implemented: tA=2.22mm, tB=2.56mm, tC=3mm, tD=3.33mm. Consequently, with this gradient change, the maximum tunable IR emissivity reaches 0.68. Therefore, by adjusting the gradient, the tunable IR emissivity range can be expanded to approximately 0.2–0.7.

IR emissivity is typically between 0.2 and 0.8 in most application scenarios. The structure designed in this paper has an IR emissivity range approximately within this interval, rather than being too low or too high, as such extremes might create a stark contrast with the background environment, thereby weakening the camouflage effect. Therefore, setting the target emissivity range between 0.2 and 0.7 enables effective camouflage combinations in most IR environments, ensuring the concealment of the protected target.

We processed the designed metasurface to validate the rationality of the theoretical calculations and simulation analyses. A 260-nm-thick layer of ITO (sheet resistance of 5 Ω/sq) was electroplated onto PET to obtain the reflective backing plate. Subsequently, the structure’s top layer was etched according to the designed pattern. Finally, all layers were bonded together. It is worth noting that there is an air layer within the structure. Here, we used foam with a thickness of 2.2 mm as the air layer. The simulation environment and sample images are shown in Fig. 5(a).

To verify the suppression effect of the designed metasurface on radar backscattering, we measured the reflection characteristics (S11) of the processed sample in the microwave band using the Agilent N5224A vector network analyzer. Figure 5(b) shows the comparison between the measured reflection curve and the simulated reflection curve. From the figure, it can be seen that the measured reflection curve is basically consistent with the simulated reflection curve. The slightly higher measured reflection curve may be due to factors such as slightly lower horn power or reduced receiver gain in the experimental environment, but the overall trend remains consistent. In addition, the reflection curve remains below the dashed line at −10dB in the 4–10 GHz range, indicating that the microwave reflection rate of the sample in this frequency range is less than 10%, consistent with the simulated results. This confirms that the designed metasurface can effectively suppress radar backscattering.

To verify the accuracy of the far-field simulations, we conducted far-field reflection tests on the samples. Figure 5(c) shows the far-field reflection test results of the samples at 6, 8, and 10 GHz. The test results align well with the simulation results, indicating that the fabricated samples perform well.

To evaluate the IR stealth performance of this encoding metasurface, we employed an IR emissivity tester to measure the IR emissivity of the four regions of the sample. The emissivity of the four parts is respectively 0.25, 0.27, 0.31, and 0.38, as shown in Fig. 6(a). Additionally, we utilized an FTIR spectrometer to measure the emissivity of the sample in the 3–14 μm wavelength range, with the measurement results presented in Fig. 6(b). It can be observed that the results from both tests are close and consistent with the theoretical calculation results.

To visually demonstrate the IR stealth capability of the designed metasurface, we captured IR thermal images of different samples at various temperatures for comparison. The IR thermal imager used was the G120EX model, operating in the 8–14 μm wavelength range. Figure 6(c) shows the heating platform used. Figure 6(d) presents IR thermal images of “0” and “1” coding units with four different filling ratios after heating at 26°C, 70°C, 120°C, and 170°C for 10 min. It can be observed that the “0” and “1” coding units with different filling ratios exhibit distinct emissivity characteristics, demonstrating successful tunability of the IR emissivity by the designed metasurface. Furthermore, after heating to 70°C, 120°C, and 170°C, the designed metasurface shows significant decreases in displayed temperature, indirectly indicating the effective role of aerogel felt and diffuse reflection in reducing thermal load.

In summary, the bispectral camouflage metasurface with microwave diffuse emission and tunable IR emissivity has achieved the goal of integrated design and successfully achieved significant reduction of RCS and tunability of IR emissivity. Experimental tests have shown that the results are consistent with simulations within a certain error range, indicating the correctness of the design. In essence, this metasurface not only achieves radar-IR multi-spectral compatibility for stealth, effectively shielding targets from IR radiation, but also allows for camouflage adjustment, offering a new avenue for superior concealment. Its immense potential for future applications will inject new vitality and possibilities into the fields of military defense, stealth technology, and multi-frequency compatibility.