Figure 1 presents the XRD analysis of the LiCl-LDHs composite microspheres, revealing the characteristic diffraction peaks of the Mg–Al LDHs in the sample’s XRD pattern20. Specifically, the peaks near 11° and 23° aligned with the (003) and (006) crystal planes of the Mg–Al LDHs, demonstrating the quintessential peaks of hydrotalcite. Additionally, the absence of diffraction peaks from extraneous substances suggested that LiCl was mechanically intermixed with Mg–Al LDHs in an amorphous state, rather than reacting with them21. The incorporation of a suitable quantity of LiCl was found to enhance the crystallinity of the LDHs, achieving optimal crystallinity at a LiCl composition of 10%. However, an increase in LiCl content beyond this point led to a diminished intensity of the Mg–Al LDHs diffraction peaks. The superfluous amorphous LiCl diminished the relative concentration of the Mg–Al LDHs in the composite. Then, the smaller-sized Li+ ions possibly infiltrated the lamellar structure of hydrotalcite, thus disturbing the crystal lattice22.Figure 1XRD pattern of LiCl-LHDs composite microspheres with different LiCl content.Figure 2 presents the SEM images of the LiCl-LDHs composite microspheres with different LiCl concentrations, demonstrating that post-spray drying, samples with minimal LiCl content possessed spherical particles characterized by a uniform morphology and smooth exteriors. The particle size exhibited a broad distribution, with the dominance of smaller particles. As the LiCl concentration increased, a marked increase was observed in the larger particles, which was attributed to the spray granulation process transitioning from atomized droplets to solid particles. The water evaporation rate from the droplets dictated the ultimate size and shape of the resultant particles. Given its hygroscopic nature as a salt filler, LiCl extended the evaporation duration of the water within the droplets during spray drying, resulting in enlarged particle dimensions. At excessive LiCl ratios (26%), incomplete drying of the particle exteriors could occur, resulting in interparticle adhesion. Moreover, superabundant LiCl modified the surface tension of the aqueous solution, diminishing the capillary forces that acted upon the particle exteriors and compromising their sphericity.Figure 2SEM morphology of LiCl-LDHs composite microspheres with different LiCl content [(a) 2% LiCl-LDHs; (b) 10% LiCl-LDHs; (c) 18% LiCl-LDHs; (d) 26% LiCl-LDHs].Moreover, higher LiCl concentrations precipitated on the particle surfaces, influencing the morphological integrity of the composite particles during the drying phase. As shown in Fig. 2, the LiCl-LDHs composite microspheres containing a minimal LiCl fraction (2%) exhibited smooth exteriors with an absence of discernible LiCl precipitates, which was indicative of the intraspherical distribution of lithium salts23. Increasing LiCl ratios resulted in an incremental roughening of the composite microsphere surfaces, which was particularly noticeable at an 18% LiCl-LDHs concentration. At a critical threshold of 26% LiCl, the microsphere surfaces were noticeably enriched with LiCl particulates. Concurrently, the composite particles displayed an uneven morphology, with a significantly increased tendency for agglomeration24.Figure 3 presents the particle size distribution and specific surface area measurements of the LiCl-LDHs composite microspheres. Particle sizes resulting from spray granulation ranged from 10 to 90 μm, demonstrating a wide distribution. The particle size distributions of the samples showed minor variations in response to different LiCl concentrations. Analysis of the median particle size (D50) data, as illustrated in Fig. 3b, suggested that LiCl incorporation decreased the water evaporation rate in spray drying, resulting in a slight increase in microsphere size with higher LiCl levels. At an 18% LiCl concentration, the median particle size (D50) of the sample increased to 26.9 μm. However, further increases in LiCl content led to a significant reduction in particle size, which could be attributed to the compromised spherical morphology.Figure 3Particle size distribution (a), D50 particle size (b), specific surface area (c) and average pore size (d) of LiCl-LDHS composite microspheres with different LiCl content.The internal architecture of the composite microspheres was also examined using specific surface area analyses. Figure 3c demonstrates a significant decrease in the specific surface area of the composite microspheres as the LiCl content increased, particularly at higher concentrations. When correlated with the SEM images and particle size distribution data, it was clear that variances in external size and morphology were minor across the range of 2–18% LiCl content. Consequently, the substantial difference observed in the specific surface area suggested that changes occurred within the internal structures of the microspheres. These findings indicated a considerable effect of LiCl incorporation on the inner structure of the composite microspheres.Building on established research25, this study confirmed that the Mg–Al layered hydroxide materials, synthesized via co-precipitation, consisted of nanoparticle flakes. This morphology resulted in the formation of numerous internal voids during the subsequent spray drying stages. The data presented in Fig. 3d indicate a trend of both the average pore size and porosity of the composite microspheres gradually decreasing with increasing LiCl content. This trend likely originated from the precipitation of lithium salts during the drying phase, which occupied the voids within the microspheres26. Specifically, at a lower LiCl threshold (≤ 10%), the microspheres maintained high porosity levels (> 0.39 cm3/g) with pore sizes exceeding 35 nm. Conversely, at an LiCl concentration of 26%, an excess of lithium salt resulted in nearly closed internal pores, thus reducing the porosity to 0.016 cm3/g and decreasing the pore diameter to approximately 11.98 nm.The FT-IR spectra of the LiCl-LDHs composite microspheres, recorded over a 400–4000 cm−1 range, are illustrated in Fig. 4. The peak approximately at 3455 cm−1 was attributed to the O–H stretching vibrations, suggesting the presence of surface-adsorbed and interlayer water within the samples27. The broadening phenomenon of this peak, correlated with an increase in LiCl content, indicated an enhanced water retention capacity that was attributable to the hygroscopic nature of LiCl. A peak emerged at 1637 cm−1 due to the bending vibrations of crystalline water28. The band at 1500 cm−1 possibly resulted from interactions between OH− ions and Li+29. Moreover, the distinct peak proximate to 1359 cm−1 could be attributed to the C–O stretching vibrations, which indicated the intercalation of CO32− in the LDH layers, consisting of a byproduct of atmospheric CO2 exposure during sample handling30. Upon addition of LiCl, the vibrational peak at 1359 cm−1 disappears and transforms into a small shoulder at 1400 cm−1, indicating a decrease in the symmetry of the carbonate anion. This is most likely due to some interaction between the free carbonate anion and its surroundings31. The spectral region ranging from 500 to 790 cm−1 included the translational and deformation modes of Al–OH bonds, and the band observed near 445 cm−1 corresponded to the O–M–O (where M = Mg or Al) lattice vibrations within the octahedral layers of Mg–Al LDHs32. The maintenance of these signature infrared features across the spectrum confirmed that the core chemical structure of the Mg–Al LDHs material remained intact despite the incorporation of LiCl.Figure 4FT-IR curves of LiCl-LDHs composite microspheres with different LiCl content.To clarify the influence of LiCl content on the moisture uptake efficacy of Mg–Al LDHs, the hygroscopic profiles of different LiCl-LDHs composite microspheres were assessed under conditions of constant temperature and elevated humidity. As shown in Fig. 5a, a direct correlation was evident between the LiCl proportion in the composite and its corresponding water-loading capacity. Significantly, the sample without LiCl exhibited the lowest saturation moisture uptake ratio, registering only 0.32 g/g. The homogeneous distribution of LiCl, even in modest amounts, in the microsphere pores notably enhanced the moisture absorption capacity of the composite material, thus simultaneously improving the water-loading capacity of the internal micro-pores and interfaces. During the initial hygroscopic stage, adsorbed water was present in the form of solid crystalline hydrates. With increasing water content, a saline solution progressively formed and permeated the internal pores of the microspheres, endowing the material with the ability to chemically absorb moisture33. According to this hygroscopic mechanism, the greater the LiCl content in the composite microsphere, the more pronounced its hygroscopic ability. However, with the inclusion of 26% LiCl, the maximum saturation moisture uptake ratio reached 3.4 g/g. Nevertheless, the absorbed water was not limited to the internal pores of the microspheres but also formed a thin film on the surface of the particles, indicating a deliquescence phenomenon, which led to powder agglomeration.Figure 5Water absorption ratio curves and spectral curves of LiCl-LDHs composite microsphere with different LiCl content [(a) hygroscopic curves; (b) spectral curves of dry powders; (c) spectral curves after saturation moisture absorption; (d) continuum removal spectrum after saturation moisture absorption].To examine the effect of LiCl content on the optical properties of the composite microspheres in greater detail, the visible and near-infrared reflectance spectra of the LiCl-LDHs composite microspheres before and after moisture absorption were analyzed. Figure 5b shows the visible and near-infrared spectral curves of the LiCl-LDHs samples in a desiccated state. Although the pure Mg–Al LDHs microspheres displayed spectral characteristics analogous to green leaves, the overall spectral reflectance was substantially higher than that of natural foliage. Moreover, the water absorption peak intensity of the desiccated microspheres was noticeably subdued, indicating significantly lower water content relative to natural plants. When a minimal quantity of LiCl (≤ 10%) was introduced to the microspheres, the spectral shape of the desiccated samples was largely retained. However, due to the high reflectance of LiCl, the spectral reflectance of the composite microspheres experienced a minor elevation. With increasing LiCl concentrations (≥ 18%), the position of the red edge was found to gradually shift toward longer wavelengths, spanning from 630 to 750 nm. This shift likely resulted from the precipitation of LiCl particles onto the surfaces of the microspheres with elevated LiCl content, resulting in a light scattering effect that intensified light absorption in the 700–760 nm range. Furthermore, an enhanced water peak intensity in the near-infrared band was indicative of the tendency of the microspheres to absorb moisture from the surrounding air.Figure 5c presents the spectral curves of the LiCl-LDHs composite microspheres following complete moisture absorption. Each sample was saturated with moisture after 74 h. For LDHs samples without LiCl, the spectral shape and reflectance exhibited negligible variations due to weak hygroscopicity. By contrast, incorporating LiCl into the LDHs materials resulted in increased water content, which altered the material’s refractive index and markedly reduced the spectral reflectance across the entire visible and near-infrared band. Using the sample with 2% LiCl content as an example, the average spectral reflectance at 400–2500 nm was found to decrease from 66 to 58% after hygroscopic absorption. With increasing LiCl content, the water content in the sample also increased, leading to a further reduction in spectral reflectance. Notably, the red edge position of the sample experienced only a 20 nm redshift following hygroscopic absorption, and the spectral shape became more akin to that of a green leaf. Specifically, the near-infrared reflectance of the sample was slightly lower than that of natural plants, which was attributed to the increased water content in the sample after hygroscopic absorption. The absorbed water dissolved some of LiCl, diminishing its influence on spectral characteristics. However, water also modified the refractive index of the composite particles to a degree, thus enhancing the scattering and absorption of incident light. Ultimately, the visible light spectra of the composite microsphere samples closely matched that of plant leaves, while the reflectance of the near-infrared spectrum remained noticeably lower than that of plant leaves.To facilitate the visual comparison of the intensity variations among characteristic absorption peaks, the reflectance spectral curve of the sample was subjected to envelope removal. As shown in Fig. 5d, at lower LiCl concentrations, the intensity of the water absorption peaks near 1450 and 1940 nm correlated with the concentration of LiCl. The water peak intensities for the 10% and 18% LiCl-LDHs composite microspheres were found to be relatively high. However, with further increases in LiCl content, the intensities of the water absorption peaks diminished. This phenomenon could be attributed to excess lithium salt, which caused a significant amount of brine solution to form not only within the microporous material but also on the surface of the powder34. However, to prevent instrument contamination during spectral testing, the removal of water on the coating surface was imperative. As a result, the water content indicated by the spectral test results was substantially lower than the actual water content in the sample. Along with the analysis of the microsphere morphology and structure, the comprehensive performance of the 10% LiCl-LDHs samples proved to be superior, demonstrating the closest resemblance to the visible and near-infrared spectra of natural plants. At this point, the spectral similarity index, rxy, achieved a value of 95.58%.Subsequently, coating samples were fabricated utilizing x% LiCl-LDHs microsphere pigments, and their hygroscopic curves and reflectance spectra were evaluated. As illustrated in Fig. 6a, the coating fabricated with a 2% LiCl-LDHs pigment exhibited a saturation moisture uptake ratio of 0.34 g/g. Compared with the coating without LiCl (0.26 g/g), the moisture uptake was marginally enhanced. The coating sample incorporating 10% LiCl-LDHs pigment demonstrated the highest moisture uptake capability, with a maximum saturation moisture uptake ratio of 0.75 g/g. With further increases in LiCl content, the moisture uptake capacity of the coating diminished. The saturation moisture uptake ratio for the coating with 26% LiCl-LDHs was only 0.18 g/g. The water absorption ratio of the coating was noticeably lower than that of the LiCl-LDHs powder because the composite microsphere particles were uniformly embedded in the resin, and their surfaces were entirely encased by acrylic resin. The microsphere particles could only interact with atmospheric moisture via confined pathways, thus significantly constraining the hygroscopic process.Figure 6Water absorption ratio curves and spectral curves of composite coating with different LiCl content [(a) hygroscopic curves; (b) spectral curves of dry coating; (c) spectral curves after saturation moisture absorption; (d) continuum removal spectrum after saturation moisture absorption].When the coating encompassed an optimal amount of LiCl, water could permeate the coating through the pores and could be subsequently absorbed by the composite microspheres and retained within their structure. However, when the LiCl concentration exceeded a certain threshold, an excessive quantity of lithium salt became intermixed with the LDHs material and could also coat the surface of the microspheres. Upon water absorption, the lithium salt progressively dissolved, and excess salt solution could not be sustained at the powder-resin interface, necessitating its migration toward the coating surface through the pores. Throughout this process, the significant depletion of lithium salt diminished the effective LiCl concentration in the coating, consequently leading to a reduction in its water absorption capacity. In addition, the migration of LiCl may also block the internal channels of atmospheric moisture into the coating, thereby reducing the water absorption capacity of composite particles and pores. The relevant influence mechanisms are shown in Fig. 7.Figure 7Schematic diagram of hygroscopic mechanism (a) LiCl-LDHs composite microspheres; (b) the coating before moisture absorption; (c,d) the coating after moisture absorption).As indicated by the spectral curve in Fig. 6, the influence of LiCl on the spectral properties of the dry coating was the same as that the composite particles. With an increase in LiCl content, a gradual reduction in spectral reflectance was observed, accompanied by a pronounced redshift in the spectral characteristics of the red edge. All the coated samples were subjected to moisture absorption for 24 h and all the samples were saturated with moisture absorption rate. After subjecting the coating material to saturated moisture absorption, the spectral reflectance diminished further, aligning with the change in spectral curve with the water absorption characteristics. As illustrated in Fig. 6d, when the LiCl concentration in the coating was 10%, the peak of spectral water absorption was most pronounced, and the spectral profile more closely resembled that of natural vegetation, with a spectral curve similarity reaching an impressive value of 97.41%.To determine the cyclic hygroscopicity of the composite coating, a series of 10 drying and moisture absorption cycles was performed on the 10% LiCl-LDHs coating samples. Figure 8a presents the mass fluctuation curve of the sample following repeated drying and hygroscopic cycle treatment. Despite observed fluctuations in the mass of the coating, it retained a relatively consistent moisture absorption capacity, and the water absorption ratio was generally sustained at approximately 0.77 g/g. As indicated by the spectral curve in Fig. 8b, repeated drying and moisture absorption treatment did not significantly change the visible–near-infrared reflectance spectrum of the coating. The spectral congruence between the samples and natural vegetation remained at approximately 95%. This suggested that the composite material exhibited robust resilience to drastic changes in ambient temperature and humidity, demonstrating considerable environmental adaptability.Figure 8Mass change data (a) and reflection spectral curves (b) of 10% LiCl-LDHs composite coating after n times of drying and moisture absorption cycle experiment.Finally, the camouflage efficacy of the coatings in the context of hyperspectral imaging was illustrated, as shown in Fig. 9. Attempts were made to detect the coatings positioned against a foliar background employing a hyperspectral camera. Figure 9c shows the near-infrared spectral curves (900–2500 nm) derived from the hyperspectral image. Analysis revealed that the near-infrared spectrum of the 10% LiCl-LDH coatings demonstrated the closest resemblance to that of natural vegetation in terms of both spectral shape and intensity.Figure 9Concealment of LiCl-LDHs coatings in the hyperspectral image [(a) the optical images of LiCl-LDHs powders and coatings; (b) the optical images of LiCl-LDHs coatings in the leaf background; (c) NIR spectral curves extracted from the hyperspectral image; (d) SAM classification results of camouflage coatings in leaf background with different thresholds; (e) MD classification results of camouflage coatings in leaf background with different threshold].To advance the verification of the spectral discrepancy effects on near-infrared hyperspectral camouflage, spectral angle mapping (SAM) and Mahalanobis distance (MD) mapping classifications were employed to assess the similarity between the coatings and the plant backdrop. SAM constituted a physically grounded spectral classification method that used an n-dimensional angle for aligning pixels with the reference spectra35. By setting a cosine similarity threshold, precise mapping of the areas with a higher cosine similarity to the reference spectra could be achieved. This technique focused on discerning the spectral shape disparities between the target and the backdrop within hyperspectral images. In contrast to SAM, MD was characterized by calculating the MD to quantify the disparity between the sample spectrum and background reference spectrum36. Through the application of a defined maximum distance threshold, targets with spectral distances exceeding this margin could be distinguished in the hyperspectral images, which primarily captured variances in reflectance between the target and background spectra.As shown in Fig. 9d, setting the threshold at 96.0% resulted in an accurate classification of most genuine leaves, where all four samples were concurrently mistaken for foliage. An increase in the threshold value decreased the extent of the mask, resulting in a more exact classification. At a threshold of 99%, the 0% LiCl-LDHs, 2% LiCl-LDHs, and 26% LiCl-LDHs samples were effectively differentiated from the mask, thus standing out against the background. Complete isolation of the 10% LiCl-LDHs sample required the threshold to be raised to 99.7%. However, this high threshold introduced an overfitting issue, where a significant number of green leaves were misclassified as false, alongside the 10% LiCl-LDHs sample, thereby rendering this criterion untenable for distinguishing between the camouflage coating and real leaves.The classification results obtained via the MD are illustrated in Fig. 9e. When the maximum distance error threshold was set to 65, all samples remained indistinguishable against the background. Due to the most pronounced spectral reflectance difference between the 0% LiCl-LDHs sample and natural leaves, the former was partially detectable at a maximum distance error of 55, while the latter three samples continued to be classified as plant leaves. After reducing the maximum distance error to 45, the 2% LiCl-LDHs and 26% LiCl-LDHs samples became distinguishable from the background, while the 10%LiCl-LDHs sample continued to blend with the background. The 10% LiCl-LDHs sample remained undetected against the background until the threshold was further lowered to 35.In summary, the superior hyperspectral camouflage capabilities of the 10% LiCl-LDHs coating were corroborated in practical scenarios. Unlike traditional optical camouflage materials, which may be readily detectable, an optimal concentration of LiCl enhanced the water content of the coating, which contributed to the augmented scattering and absorption effects of near-infrared light. The intensity and shape of the spectrum more closely resembled those of natural leaves, indicating significant potential for application in hyperspectral camouflage.
