Infrared (IR) camouflage plays a critical role in modern defense systems, ensuring the safety and security of facilities, vehicles, and individuals [1–4]. IR camouflage, particularly in the commonly used long-wave infrared (LWIR) band (8–14 μm), aims to reduce the IR radiation of a target object to match the radiation of its surrounding environment [5–7]. IR camouflage achieves this blending effect by adjusting either the target’s temperature or its surface emissivity. Traditional thermal camouflage techniques often use low-emissivity coatings to reduce an object’s radiation temperature, thereby blending it with the background [8–12]. However, static emissivity adjustments may not be sufficient to adapt to dynamic environmental changes, which is where dynamic thermal camouflage, enabling real-time emissivity modulation, presents a promising solution [13–18]. Several methods for dynamic thermal camouflage have been explored, including electrically tunable materials like quantum wells [19–21] and two-dimensional materials [22–25], as well as mechanical approaches utilizing high thermal expansion materials integrated into microelectromechanical systems (MEMS) [26–28]. Phase-changing materials (PCMs), which exhibit significant variations in the complex refractive index during phase transitions, have been explored for dynamic emissivity control and show considerable promise [29–33]. However, when external heat sources are positioned at the specular angle relative to the detector, they can substantially elevate the radiative temperature of the target, thereby increasing the likelihood of detection. Consequently, advanced technology for dynamic camouflage still needs to be explored to ensure effectiveness in complex environments.

Metasurfaces constructed from artificial resonators can scatter reflected signals in multiple directions, achieving low visibility for laser active detection [34–39]. However, multidirectional scattering still faces detectability challenges due to scattering characteristics that are dependent on specific angles. Additionally, their practical application is hindered by challenges in scalability and cost-effective fabrication. Low-emissivity diffuse reflectors offer a potential solution by scattering incident thermal radiation across a hemisphere, reducing detection likelihood. Such diffuse reflectors have been applied in IR camouflage devices based on Lambertian surfaces [40–42]. Rough porous medium surfaces with low emissivity and omnidirectional low reflectivity have been utilized for decades [43–46]; however, their application in practical dynamic IR camouflage remains to be further explored and developed. For practical applications, developing surfaces that integrate dynamic emissivity control, low specular reflectance, and scalable fabrication processes remains an urgent challenge. The application of diffuse reflectors in dynamic camouflage presents a promising approach to mitigating external thermal interference. This study aims to overcome these limitations by advancing dynamic thermal radiation control within the LWIR band, combined with broadband scattering to effectively mitigate the effects of external thermal interference.

In this paper, we experimentally demonstrate a scattering-enhanced and tunable thermal emitter (SETTE) camouflage device composed of porous substrate, metallic film, and Ge2Sb2Te5 (GST). By leveraging the porous medium, the device achieves low specular reflectance and dynamic infrared camouflage through the GST phase transition. The device utilizes a porous medium to achieve low specular reflectance and exhibits camouflage capability across varying background temperatures through the GST phase transition. Experimental results show that SETTE maintains effective thermal camouflage as the background temperature varies from 35°C to 45°C, with specular reflectance to surrounding heat sources below 1%. Moreover, the camouflage performance remains robust over a wide observation angle range of 0° to 60°. The fabrication process employs straightforward film deposition techniques, which enhances scalability and suitability for practical real-world application. These results establish that diffuse reflectors, when combined with dynamic phase-change materials, hold considerable promise for next-generation infrared stealth and thermal management technologies.

Sample characterization. SETTE utilizes a 1 μm grain-sized 316 stainless steel powder-sintered substrate with a thickness of 1 mm and a diameter of 5 cm. The metal (Ag) was deposited by electron beam evaporation under a high vacuum (∼10−6Torr). The Ag film was deposited at a rate of 2 Å/s (monitored by a quartz crystal microbalance-based film thickness monitor) to a thickness of 800 nm that was sufficient to form a metal reflector with an electron beam current of 50 mA. The GST layer is then deposited on the silver film by magnetron sputtering, in which germanium is DC sputtered and the other two components (stibium and tellurium) are radio frequency (RF) sputtered with a 2:2:5 deposition rate ratio. The as-deposited GST is in an amorphous phase and is annealed at 200°C on a hot plate for 3 min to obtain the face-centered cubic GST (FCC-GST). Further annealing at 300°C for an additional 3 min transforms the material into the hexagonal GST (HEX-GST).

Optical measurements. IR photographs are acquired using an FLIR TG165 IR camera, which has a spectral range of 7.5–13 μm and a spatial resolution of 500 μm. The emission spectra of the blackbody and the SETTE samples were collected using a Bruker Vertex 70 FTIR spectrometer equipped with a DTGS detector. The samples were mounted on a temperature controller, and their emitted radiation was directed into the FTIR for detection by the DTGS. To ensure sufficiently high emissive power and minimize noise, both the blackbody and the SETTE samples were maintained above 100°C. Each measurement was repeated 15 times to further reduce noise. The measured emissive power of the samples was compared with that of a reference blackbody source (SR-200, CI Systems, Israel). The spectral emissivity of a 1.3cm×1.3cm specimen was calculated using the following relation: Ssample(λ,T)Sblackbody(λ,T)=εsample(λ,T)εblackbody(λ,T),(1)where S represents the signal magnitude. Oblique reflectance measurements are conducted using a Bruker Vertex 70 FTIR spectrometer equipped with a DTGS detector and an A513 attachment.

Numerical simulations. The simulation was conducted using COMSOL Multiphysics 5.2a with the Wave Optics Module and the finite element method. The unit cell of the SETTE structure [Fig. 1(a)] was modeled without the substrate, as the metallic ground layer acts as an ideal reflector. To simplify computation, the surface topography of the porous medium was approximated as a randomly varying rough surface, described by the following equation: f(x,y)=∑−MM∑−NNa(m,n)cos(2π(mx+ny)+φ(m,n)),(2)where x and y are spatial coordinates, m and n represent spatial frequencies, M and N denote their respective maximum integer values, a(m,n) is the amplitude, and φ(m,n) is the phase angle. Along the boundary in the Y-direction, a perfectly matched layer (PML) was implemented to absorb all reflected light, while a probe was utilized to calculate total reflectivity. Periodic boundary conditions are applied in the X-direction to ensure the computational domain is sufficiently large. The relative permittivity of silver was derived from Ref. [47], and the relative permittivities of GST are obtained experimentally from Ref. [15].

Measurement of the morphology of the SETTE surface. The morphology of the VIS-IR scattering surface is measured by a white light optical profiler (Rtec LAMBDA). The SEM image is measured by a field emission scanning electron microscope (ZEISS, SUPRA40).

Figure 1(a) illustrates the principle of detecting reflected heat sources with an IR camera. On a low-emissivity (low-e) smooth surface, specular reflection often dominates. Consequently, if the camera (detector) is positioned within the specular reflection region of an external heat source [highlighted by the green circle in Fig. 1(a)], the target becomes exposed, leading to a significantly higher detected temperature relative to the surroundings. The thermal camouflage principle for varying background temperatures using the SETTE is illustrated in Fig. 1(b). As the background temperature gradually increases from a lower to a higher range, the thermal images captured transition from representing a low radiation temperature to a high radiation temperature. Meanwhile, the SETTE, maintained at a fixed temperature above the background, can be made to merge with its surroundings by modulating its thermal emission through different GST phases. As the GST layer transitions from amorphous GST (aGST) to FCC-GST and ultimately to HEX-GST, the drop in reflectance (and thus increase in emissivity) elevates the SETTE radiative temperature, enabling effective thermal camouflage across a wide range of ambient conditions.

A schematic of the SETTE is shown in Fig. 1(c). The device consists of a sintered 316 stainless steel powder substrate, onto which silver and GST layers are deposited. The device includes an 800 nm thick silver layer at the bottom, which ensures that the film is optically thick, effectively minimizing transmission and providing the substrate with sufficiently low emissivity. If needed, the Ag layer can be replaced with alternative metals, such as gold or aluminum, to meet different fabrication requirements. A 450 nm thick GST layer is deposited on top, serving as the active medium for dynamic thermal emission control. Surface morphology measurements [Fig. 1(d)] reveal a Lambertian surface with a peak-to-peak height of 25.6 μm and randomly distributed pores. This rough surface structure, widely reported to enhance scattering, likely contributes to the diffuse reflection behavior of the SETTE. Figures 1(e) and 1(f) present SEM images of both the substrate and the SETTE surface. Notably, after Ag and GST deposition, the overall surface morphology remains largely unchanged. The inherent roughness of the structure imparts diffuse reflection characteristics to the SETTE.

Figure 1(g) presents the integrated and specular reflectance of the SETTE measured from 7.5 to 13 μm, showing significant contrast covering broadband range. For the aGST, FCC-GST, and HEX-GST, the average integrated reflectance over this spectral range is 0.60, 0.48, and 0.28, respectively. Meanwhile, the specular reflectance remains below 0.10 for aGST, FCC-GST, and HEX-GST. The radiation temperature of the SETTE also increases when SETTE is changed from aGST to HEX-GST due to the increased emissivity. Low specular reflectance can prevent external heat sources from being detected by sensors positioned in the specular direction. The GST exhibits distinct optical properties in its aGST, FCC-GST, and HEX-GST phases due to variations in atomic ordering. In the amorphous state, the random arrangement of atoms leads to a relatively large optical bandgap (approximately 0.7 eV [48]), resulting in negligible absorption across the mid-infrared region. Upon annealing at around 200°C, the aGST transitions to the metastable FCC-GST, where the increased crystallinity and partial delocalization of electronic states reduce the bandgap to ∼0.5eV [48], thereby enhancing mid-infrared absorption. Further heating to approximately 300°C induces a transition to the HEX-GST phase, which is thermodynamically the most stable. Due to its robust absorption and thermal stability, HEX-GST is particularly well suited for dynamic IR camouflage designs requiring tunable emissivity and reliable performance. The aGST itself exhibits minimal ohmic loss. Thermal radiation primarily originates from resistive losses in the silver layer from absorption by the randomly rough surface. In contrast, the aGST, HEX-GST, and Ag experience resistive losses, along with surface absorption, leading to higher overall thermal emission. Consequently, the reflectivity of the aGST-based SETTE is higher than that of the FCC-GST-based SETTE [Fig. 1(g)], while the HEX-GST-based SETTE exhibits an even lower reflectivity than its FCC-GST counterpart. Figures 1(h) and 1(i) illustrate the distribution of the electric field for aGST and HEX-GST under 10.6 μm at an incidence angle of 45°. As shown in Figs. 1(h) and 1(i), the SETTE redistributes the reflected light across the entire hemispherical space; the specular pattern is undetectable, which indicates strong scattering. In practical applications, redirecting scattered waves away from the line of sight of surveillance radar serves as an effective strategy for laser stealth. This behavior stems from the SETTE surface morphology, which promotes wide-angle scattering and a more uniform distribution of reflected energy, preventing the formation of strong specular peaks.

To evaluate the IR camouflage capability of the SETTE device, an IR camera was used to simultaneously record the temperature profiles of both the target and the background. The experimental setup is illustrated in Fig. 2(a), where the blackbody-like background varies from 35°C to 45°C, while the SETTE (representing the target) is maintained at 50°C. In this study, high-emissivity black tape, exhibiting a nearly constant emissivity of 0.97 across wavelengths, is used as background material. The target is annealed at 300°C for 23 min to ensure complete crystallization of the GST. This configuration provides a straightforward and effective approach for assessing the real-time thermal camouflage performance of the SETTE under controlled temperature conditions.

IR images clearly demonstrate the thermal camouflage capability of the SETTE under various background temperatures. As shown in Fig. 2(b), when the background temperature was maintained at 35°C, both the target and the background exhibited a radiative temperature of approximately 35°C, rendering the target nearly indistinguishable in the IR image. To simulate a wider range of natural environments, the background temperature is increased from 35°C to 45°C, thereby altering its radiation profile. As a result, the colder target, which remained at approximately 35°C, became distinctly visible against the hotter background. To investigate how annealing conditions influence the target’s emissivity, aGST is annealed at 200°C and 300°C for 23 min each [Fig. 2(f)]. The results indicate that at 200°C, the emissivity increased from 0.41 to 0.52, stabilizing after approximately 18 min. In contrast, when annealed at 300°C for the same duration, the emissivity significantly increased from 0.41 to 0.72 and stabilized within the first 3 min, indicating a faster crystallization rate for the HEX-GST phase. Higher annealing temperatures and longer annealing durations led to an increased average emissivity within the IR camera’s detection range (7.5–13 μm), elevating the target’s radiative temperature and enhancing its thermal camouflage effectiveness. Notably, when the target was annealed at 300°C for only 3 min, the HEX-GST radiative temperature matched that of the 45°C background [Fig. 2(e)], ensuring robust thermal camouflage across a background temperature range from 35°C to 45°C.

The average emitted power of the background, aGST-based SETTE device, and HEX-GST-based SETTE device is shown in Fig. 2(g). The radiative power was obtained by integrating the experimentally measured spectral emission over the 7.5–13 μm range and then taking the average. If the background’s planar radiative power falls within the regulatory range of aGST and HEX-GST, adjusting the GST phase in the SETTE device can align its radiative power with that of the background, thereby achieving thermal camouflage. As the actual temperature increases, both aGST and HEX-GST exhibit higher average radiative powers, which also amplifies the difference between them. Consequently, raising the target’s temperature expands the range of background temperatures that can be camouflaged. For instance, when the target is at 50°C, perfect concealment is achieved in backgrounds ranging from 35°C to 45°C, whereas a 90°C target can be effectively camouflaged in backgrounds between 50°C and 75°C.

To further investigate the IR scattering characteristics of the SETTE device, we conducted indoor experiments using an IR camera. A low-emissivity, smooth surface was also tested for comparison. As illustrated in Fig. 3(a), the experimental setup included two heating stages, each covered with high-emissivity tape. One stage was maintained at 100°C to serve as the external heat source, while the other was held at 45°C to simulate a 45°C background temperature. The external heat source was oriented at an incidence angle of 30°, and the IR camera was positioned at the corresponding specular reflection angle, enabling direct acquisition of specular reflection images.

As illustrated in Figs. 3(b) and 3(c), the temperatures of aGST and HEX-GST are set to 70°C and 50°C, respectively, to achieve perfect camouflage against a background temperature of 45°C. In the IR camouflage scenario without an external heat source [Fig. 3(b)], the IR signal detected by the IR camera originates primarily from the thermal radiation of the devices. Both aGST and HEX-GST exhibit radiative temperatures that closely match the background temperature, effectively rendering them camouflaged. In contrast, the low-emissivity smooth surface demonstrates a radiative temperature below the background due to its inherently low emissivity. Consequently, the cold target is easily distinguishable from the warm background in the IR images.

When an external heat source (100°C) is present, the IR signal detected by the IR camera comprises contributions from both the thermal radiation of the devices and the reflection of the external heat source [Fig. 3(c)]. The results indicate that camouflaging either a high-temperature target (aGST) or a low-temperature target (HEX-GST) near the external heat source produces radiative temperatures that closely match the background temperature. Conversely, the low-emissivity smooth surface exhibits significantly higher radiative temperatures due to pronounced specular reflection, causing the hot target to be readily identifiable against the cooler background in the IR images. These results demonstrate that the SETTE can maintain effective thermal camouflage under external heat sources. The ability of SETTE to minimize reflected IR signals ensures enhanced stealth performance, making it a superior choice for dynamic IR camouflage applications.

The average emitted power with the external heat source of the background, aGST-based SETTE device, and HEX-GST-based SETTE device is shown in Fig. 3(d). Compared to Fig. 2(f), the temperature range over which the target is effectively concealed is nearly identical with and without the external heat source. This indicates that effective camouflage can be achieved over a wide temperature range, even in the presence of an external heat source in the specular direction of the detector. The contributions of thermal radiation and reflection from aGST and HEX-GST to the detected power are further analyzed, as illustrated in Figs. 3(e) and 3(f). For the case with a heat source, the thermal radiation (yellow) and reflection of the heat source are considered (green). Due to the inherently low emissivity of aGST, its radiative contribution is significantly lower than that of HEX-GST. When the external heat source is positioned at the specular reflection angle, the diffuse reflection characteristics of the SETTE device ensure that the reflected power remains nearly constant, with the reflection contribution maintaining a value below 0.01. The minimal reflection contribution further enhances the device’s ability to maintain thermal camouflage by reducing the detectability of reflected IR signals.

Given that targets can be detected from a wide range of observation angles, the robustness of the thermal camouflage device to angular variations is crucial for ensuring target survivability. The robustness of thermal camouflage in SETTE with aGST and HEX-GST, in the presence of an external heat source at the specular position, is demonstrated across a wide observation angle range from 20° to 60° in Figs. 4(a) and 4(b). As the observation angle increases from 20° to 60°, the measured radiation temperature of the object remains nearly identical to that of the background, demonstrating the effective performance of the thermal camouflage device across a wide range of observation angles. Therefore, this device holds significant practical applications. The thermal images of the device exhibit non-uniformity, with the central temperature higher than that at the edges, which is attributed to uneven heating.

To evaluate the angular robustness of thermal camouflage, it is essential to consider the two primary factors that govern IR detection: the intrinsic thermal emission of the target (emissivity) and the reflected radiation from external heat sources (reflectivity). The emissivities of SETTE at different observation angles are measured to validate the robustness of the thermal camouflage over different observation angles [Figs. 4(c) and 4(d)]. Figures 4(c) and 4(d) show that the emissivities remain consistent across the observation angle range of 0° to 70°. This finding corroborates the results obtained from the infrared images. Figure 4(e) shows the specular reflectance of aGST and HEX-GST from 15° to 60°. As the incidence angle increases from 15° to 60°, the reflectivity remains below 0.1 across the infrared camera detection range (7.5–13 μm). This indicates that in the presence of an external heat source, SETTE exhibits minimal reflection over a wide range of observation angles. Nevertheless, SETTE exhibits minimal sensitivity to variations in the viewing angle, thereby maintaining effective camouflage performance even at large observation angles. In addition, while HEX-GST exhibits higher overall emissivity—and consequently lower integrated reflectance—than aGST, its specular reflectance is slightly higher. This could be attributed to the HEX-GST relatively lower surface roughness compared to aGST, which reduces diffuse scattering and enhances specular reflection due to the more ordered crystalline structure.

In this study, we present an SETTE device that adapts to varying background temperatures while maintaining low specular reflectance against external heat sources. The device incorporates a porous metallic substrate to achieve low specular reflection (<0.1) and GST, which transitions from aGST to HEX-GST, enabling tunable emissivity. It is demonstrated that the SETTE device achieves perfect thermal camouflage when the background temperature varies between 35°C and 45°C, while the target temperature remains fixed at 50°C. Furthermore, even in the presence of an external heat source in the specular direction, SETTE with temperatures ranging from 50°C to 70°C can effectively camouflage in a 45°C background. The thermal camouflage is robust over a large range of observation angles, from 20° to 60°. By integrating dynamic emission control with strong scattering capabilities, this GST-based design significantly advances thermal camouflage technology and is well-suited for large-scale production and deployment.