1 Introduction

Infrared transparency, as a fundamental property in infrared optics, has been of great interest in the development of infrared optical technology and directly affects the progress in many fields. The application of infrared transparency in extreme environments is highly active and increasingly interdisciplinary, as shown in Fig. 1. However, theoretical and experimental studies on infrared transparency in extreme environments are still in the early stages, and several aspects need further exploration in the future.

Fig. 1
figure 1

The schematic representing the guideline of this review on Infrared transparency and application scenarios

When an aircraft flies at high speed, the air in front of it becomes strongly compressed, generating friction with the aircraft. Most of the kinetic energy from compression and friction is converted into heat energy, leading to a significant increase in air temperature. As flight speed increases, the aerodynamic heating phenomenon becomes more pronounced, causing a sharp rise in window temperature. This phenomenon is known as the “window thermal barrier” problem. Additionally, viscous interference significantly influences the aircraft’s surface pressure, affecting lift, resistance, and stability, thus generating aerodynamic forces. A more critical issue arises from the combination of intense impact and thermal energy, causing the infrared (IR) window to break and become damaged.

The complex electromagnetic environment both inside and outside an aircraft result from a combination of natural and man-made sources of electromagnetic radiation, along with the presence of electrical and electronic equipment on board. An ideal infrared optoelectronic system relies on a large front aperture window for detecting infrared signals. Currently, commonly used infrared fairings for weapons and equipment consist mainly of materials such as sapphire and zinc sulfide. These materials lack radar stealth characteristics, allowing electromagnetic radiation to easily penetrate the window, leading to electromagnetic interference with sensitive infrared detectors. Moreover, radar electromagnetic waves undergo multiple reflections within the fairing, resulting in conspicuous radar reflection signals.

Sand and rain erosion is a complex phenomenon occurring when solid particles, like sand, dust, or ice, impact the surface of an optical window at high speeds in the presence of rain or suspended liquid droplets. These impacts cause mechanical damage, leading to cracking, corrosion, and deformation. Reduced IR transmittance and material strength negatively affect the detection and guidance performance of such systems.

Laser damage to thin film materials can be categorized as thermal, mechanical, or radiation damage. Thermal damage primarily involves the thermal ablation effect of lasers on materials, while mechanical damage includes shockwave effects and thermal stress. Furthermore, human eyes, optoelectronic detection devices, and optical systems exposed to intense laser environments are susceptible to laser attacks, leading to failure and damage to target materials and electronic optical components.

For window materials, it is insufficient to focus solely on high transparency (T). Robust mechanical and other properties are necessary to enable utilization in demanding application environments. However, high hardness creates a paradoxical situation of lower transparency. Simultaneously, the conflict between high transparency and low infrared thermal radiation presents a challenging dilemma. Additionally, carrier effects impact both electrical conductivity and intraband transition absorption, resulting in a contradiction between high electrical conductivity and high infrared transmittance. Balancing the pursuit of high transparency with the development of a solution to maintain optical limiting functionality becomes an essential endeavor.

The outline of this review is shown in Fig. 1. Firstly, we discuss the types of typical extreme environments and their impact on infrared transparency. Then, we summarized strategies to improve infrared transparency performance. Finally, we predicted new challenges for the future.

2 Extreme environment

2.1 Aerothermal and aerodynamic environment

Supersonic and hypersonic are the most important development directions of today’s infrared-guided aircrafts. When an aircraft flies at high speed, the air in front of it is strongly compressed and generates friction with the aircraft. Most of the kinetic energy of compression and friction is converted into heat energy, resulting in a sharp increase in the air temperature. There is a huge temperature difference between the gas and aircraft surface, and the temperature is quickly transferred to the aircraft via convective heat transfer, which causes the surface temperature of the aircraft to increase and generates aerodynamic heating. With the increase of flight speed, the aerodynamic heating phenomenon becomes more and more obvious, and the window temperature rises sharply. The high-temperature infrared radiation of the optical window increases the background brightness of the infrared image of the infrared imaging system, reduces the ability to detect and track the target, and even causes the infrared detector to saturate in severe cases, resulting in a so-called window thermal barrier problem. In addition, in the direction perpendicular to the aircraft surface, the pressure from the boundary layer to the aircraft does not change. Viscous interference has an important influence on the surface pressure of the aircraft, affecting the lift, resistance, and stability and thereby generating aerodynamic forces. The more serious problem is the powerful impact coupled with thermal energy that causes the IR window to break and be damaged.

2.2 Sand and rain erosions

The front protection components of infrared detection and guidance systems of aircraft, including fairings and side windows of infrared guided missiles, as well as airborne infrared search and tracking systems (IRST), are critical components that must possess high transmittance in the operating band and exceptional mechanical properties. These components must be able to withstand harsh environmental conditions, including high-speed impacts of sand, dust, ice particles, and other solid materials. The presence of raindrops or suspended liquid droplets exacerbates the damage caused by these particles, leading to cracks, corrosion, and deformation on the outer surface of the IR window cover. The reduction of IR transmittance and material strength negatively impacts the detection and guidance performance of these systems.

Rain erosion is a complex phenomenon that occurs when solid particles, such as sand, dust, or ice, impact the surface of optical window at high speeds in the presence of rain or suspended liquid droplets. The impact of these particles causes mechanical damage to the surface, which can lead to cracking, corrosion, and deformation. The extent of damage depends on several factors, such as particle size, impact angle, velocity, and surface material properties. Rain erosion can have a significant impact on various applications, such as aircraft, where the front protection components of infrared detection and guidance systems, including fairings, side windows, and infrared window coverings, can be particularly vulnerable to this phenomenon. In addition to the erosion of raindrops that the aircraft needs to face, there will be solid particles such as sand and dust impacting the window area at high speed. When the optical window surface is impacted by sand and dust moving at high speed, plastic penetration occurs on its outer surface, forming a nearly hemispherical plastic zone. The surrounding area is still elastic. There is residual tension stress at the boundary of the plastic zone, and the residual stress field expands outward, which will form two forms of microcracks. Radial cracks and transverse cracks, with the transverse cracks parallel to the surface. If the incident sand velocity is large enough, the volume surrounded by the transverse cracks will break and fall off, which creates damage to the surface of the optical fairing. Such damage will degrade the optical transmittance and thus affect the detection and guidance performance of the detector.

2.3 Complex electromagnetic interference environment

The complex electromagnetic environment inside and outside of an aircraft is formed by a combination of natural and man-made sources of electromagnetic radiation, as well as by the presence of electrical and electronic equipment on board the aircraft. Natural sources of electromagnetic radiation, such as cosmic rays and solar radiation, can affect the electromagnetic environment of an aircraft at high altitudes. In addition to these sources, the electrical and electronic equipment on board an aircraft can also generate electromagnetic radiation. This includes communication systems, navigation equipment, and airborne radar equipment. The ideal infrared optoelectronic system is to rely on the front large aperture window to detect infrared signals, at present, weapons and equipment commonly used infrared fairings are mostly sapphire, zinc sulfide, and other materials, which itself does not have the characteristics of radar stealth, electromagnetic radiation can easily pass through the window so that sensitive infrared detectors are subject to electromagnetic interference. Secondly, the radar electromagnetic waves are reflected many times inside the fairing, causing the radar reflection signal to be obvious. To ensure the safe and reliable operation of an aircraft in this complex electromagnetic environment, aircraft designers must take into account the potential effects of electromagnetic interference (EMI) on the operation of electronic equipment.

2.4 Extreme opto environment

The research focus of extreme opto environments is laser interference. due to its high precision and strong anti-interference capabilities, laser interference has become a highly researched technological field worldwide. In practical applications, when materials are irradiated by lasers, they absorb the energy, leading to heating, melting, vaporization, and the formation of plasmas. At the same time, human eyes, optoelectronic detection devices, and optical systems exposed to intense laser environments are vulnerable to laser attacks, resulting in failure. The damage caused by lasers to thin film materials can be classified into thermal damage, mechanical damage, and radiation damage. Thermal damage primarily manifests as the thermal ablation effect of lasers on materials. Mechanical damage includes shockwave effects and thermal stress. When thin film materials are irradiated by high-energy lasers, the gaseous substances will be ionized into plasma clouds. On one hand, the plasma cloud can shield the laser, but it also emits ultraviolet and X-ray radiation, causing damage to the target material and electronic optical components.

3 Infrared anti-reflective protective film

The infrared optical detection has been widely applied in various fields such as remote sensing, infrared guidance, infrared imaging, and gas detection. It has demonstrated significant value in modern military applications. Moreover, the infrared spectrum serves as the “fingerprint” of volatile organic compounds (VOCs) gases, making it a commonly employed means of rapid detection and analysis in industrial settings to monitor gas leaks and pollution. In practical applications, to enhance the clarity of infrared imaging and increase detection sensitivity, optical systems including camera modules, detector optical windows, and displays often incorporate anti-reflective thin films to minimize energy loss during propagation. However, the majority of infrared anti-reflective coating materials, such as ZnS and MgF2, exhibit poor mechanical properties, making it challenging to meet the requirements of extreme operating environments, as shown in Table 1. Therefore, the development of infrared anti-reflective protective thin films that possess a wide spectral range, high transmittance, and high hardness concurrently is of paramount importance.

Table 1 Properties of infrared optical materials

When light travels from a medium with refractive index n0 to another medium with refractive index n1, reflection of light occurs at the interface between the two media. If the medium does not absorb light and the interface is an optical surface, when the light is incident perpendicularly, the reflectance R is given by R = (n0-n1)2/(n0 + n1)2, and the transmittance T is T = 1-R. In fact, as long as n1 < n2, RF < RS and TF > TS, a single-layer thin film can reduce the surface reflectance, which is called an anti-reflection coating. If n1 < n2 and n1d1cosθ1 = (2 m + 1)λ0/4, then Rmin(λ0) = \(\frac{{({\eta }_{0}{\eta }_{S}-{\eta }_{1}^{2})}^{2}}{{({\eta }_{0}{\eta }_{S}+{\eta }_{1}^{2})}^{2}}\). Clearly, the condition for R(λ0) = 0 is \({n}_{1}=\sqrt{{n}_{0}{n}_{s}}\). For commonly used glass substrates, it is difficult to find a material that satisfies \({n}_{1}=\sqrt{{n}_{0}{n}_{s}}\), making it challenging to achieve zero reflection. To overcome this problem, bilayer and multilayer anti-reflection coatings have emerged. Taking a bilayer film stack as an example, to achieve R(λ0) = 0, it is not enough to only rely on designs that satisfy \({n}_{1}=\sqrt{{n}_{0}{n}_{s}}\). The overall value of R(λ0) = n12/ns can also be altered to approach n0. For example, by adding an additional film between n1 and ns, changing the reflection film system from G|L|A to G|HL|A, Y(λ0) = nL2nS/nH2. As long as nL2nS/nH2 = n0/nS, R(λ0) = 0 can be achieved.

The infrared anti-reflection protective film system needs to balance low reflectance and high hardness, which are contradictory physical properties. It is highly challenging to achieve both functions in a single thin film. Typically, the overall hardness of the film system is enhanced by adding a layer of hard protective film on the outermost surface of the traditional anti-reflection film. However, achieving broadband light transmission requires a more complex film system with additional layers and thickness, leading to prominent mechanical issues at the interfaces. A lower number of layers reduces the number of interfaces, resulting in better stability in extreme environments. In practical scenarios, the hardness of the outermost thin film determines the overall hardness of the anti-reflection film system. Generally, hard thin films have higher refractive indices, which are unfavorable for improving the anti-reflection performance of the film system. Therefore, in the design of the film system, the thickness of the protective layer should be appropriately reduced while meeting the mechanical performance requirements. To further minimize the impact of high-hardness film layers on the anti-reflection performance, a cross-combination design of high-hardness and high-transmittance film layers can be employed. Protective film layers typically have higher refractive indices and stress, and they often operate in high-density aerothermal, aerodynamic impact, and high-temperature environments. The residual stress in the film preparation process can cause wrinkling, cracking, or even delamination of brittle optical films during service. When designing thin films, it is necessary to minimize the physical property differences between each layer to reduce internal stresses. There are generally three methods to reduce stress: 1) adding transition layers between the conventional anti-reflection film system and the protective film to enhance the bonding between the film and the substrate; 2) inserting relatively softer low-stress layers into hard multilayer films to release the stress from the high-stress layers; 3) eliminating film defects and releasing residual stress through heat treatment.

ZnS and sapphire are the most commonly used materials for IR windows. ZnS has superior IR transmittance in the broadband range, while sapphire has higher transmittance than ZnS in the visible to the mid-IR range [1,2,3,4,5,6,7,8]. However, ZnS has lower hardness and transmittance than sapphire, which makes it difficult to withstand high-speed aerodynamic heating and forces. On the other hand, sapphire’s high-temperature emissivity limits its applicability, and its emissivity tends to increase with temperature, leading to background radiation that obscures the infrared target signal [9,10,11,12,13,14]. Researchers have increasingly focused on developing anti-reflective and protective film materials to overcome these problems. By applying functional films to the window material, the disadvantages of the material can be compensated for, and the application requirements can be met. Developing anti-reflective and protective films has attracted the attention of developed countries [15,16,17,18,19]. For example, the Lewis Research Center has investigated amorphous C protective films for infrared ZnS windows [20]. At the same time, the University of Texas Research Institute has a long history of research into anti-reflective and protective films for infrared windows [21].

To this day, anti-reflection protective coatings (ARPCs) have developed into various material systems, such as diamond films [22], diamond-like C (DLC) films [23], Ge films [24], Ge carbide (GexC1-x) films [25], B phosphide (BP) and galladium phosphide (GaP) films [26], oxide (yttrium oxide (Y2O3), aluminum oxide (Al2O3)) films [27], fluoride (yttrium fluoride (YF3) [28], magnesium fluoride (MgF2)) films [29]. ARPC materials such as DLC, BP, GexC1-x, and others have high hardness and elastic modulus, which can withstand erosion by solid particles during high-speed flight. However, these materials have poor stability at high temperatures and cannot resist the effects of aerodynamic heating and forces. Moreover, due to the opacity of visible light, it is difficult to match the multispectral transmission ranges of substrates such as multispectral ZnS or sapphire. The spectral transmission ranges of some anti-reflective and protective film materials, such as oxides and fluorides, are comparable to those of the substrate materials, and their high-temperature stability is better than that of DLC. However, their low hardness limits their applicability. Therefore, developing an ARPC with high hardness, stability at high temperatures, and resistance to the effects of aerodynamic heating and forces is necessary.

In the early 1970s, The Space Science Department of Whittaker Company in the United States used superheated C ion beams to deposit thin film materials with infrared transparency, insulation, and high hardness, like a diamond. However, the amorphous structure of this film is DLC [30,31,32]. DLC film has high optical T (especially in the infrared range) and a smooth microscopic surface [33]. DLC is an ideal protective coating with high hardness and good abrasion and corrosion resistance [34]. Due to its excellent properties, DLC film has great application value. In recent decades, various forms of DLC have been prepared via different processes [35,36,37]. Compared with hydrogenated DLC films, non-hydrogenated DLC films have advantages such as higher performance indices, better thermal stability, and lower preparation temperatures, which have gradually become the mainstream research on and application of DLC [38,39,40,41].

Film hardness is a comprehensive property, which is equal to the ratio of the positive force acting on the diamond indenter to the projected area of indentation deformation [42]. Experiments have shown that the hardness and elastic modulus of DLC are closely related to the film’s tetracoordinate hybridization content. Within a certain range, the film hardness and elastic modulus are almost linearly related to the sp3 hybridization content [43, 44]. When the incident wavelength is short, the film’s refractive index changes considerably with the wavelength.

The application of DLC depends on the combination of film and substrate. Due to the stress produced by the deposition process, the maximum adhesion thickness of the DLC on the substrate is often limited. Consequently, researchers have improved the interface between the film and substrate by modifying, introducing an intermediate layer, and other technologies. Effective combinations of DLC with different substrates, such as silicon, glass, piezoelectric crystal, and aluminum alloy, have been achieved. However, the poor combination with infrared window materials, such as ZnS, remains a significant problem. For example, filtered cathode vacuum arc deposition technology can produce high-performance ta-C thin films at high deposition rates near room temperature. However, high compressive stress remains in the sp3-rich thin film formed by the shallow implantation of the deposited ions and local densification, which can easily cause the film layer to fall off the substrate. Therefore, the high internal stress of the film limits the application of ta-C within the field of infrared window protection.

BP is a group III-V compound composed of elements P and B. The main existing crystal form is cubic, which is denoted as c-BP and has a Zn-blended structure [45]. Compared with other films, BP film has the best rain erosion resistance, which can greatly improve the substrate’s damage threshold velocity (DTV) [46]. Compared with GexC1-x, BP has better mechanical properties. The Knoop hardness of BP is 47 GPa, which is more than twice that of sapphire, and Young’s modulus can be as high as 270 GPa [47]. However, in most cases, BP films with polycrystalline and amorphous structures are obtained, and the hardness is lower than that of the single crystal. However, the infrared T of BP films is not greater than 80% [48]. It is difficult to satisfy the requirements of anti-rain erosion protection and anti-reflection simultaneously because BP has a high refractive index (n ≈ 3) and a large absorption ratio. As shown in Fig. 2b, Tu et al. [5] designed and fabricated a double-layer coated structure on the surface of the ms-ZnS substrate. The outer surface of the ms-ZnS was coated with DLC/BP as a high performance protective and anti-reflective coating with a maximum transmittance of about 93.3% and an average transmittance of about 87.9% in the 8–12 μm band. The nano-hardness of the ms-ZnS sample increased from 2.5 to 19.5 GPa.

Fig. 2
figure 2

a Diamond-like carbon films coated on ZnS substrate [49], b DLC/BP and ZnS/YbF3 double-layer protective and antireflective coatings [5], and c Diamond-like carbon thin films as antireflective and protective coatings [50]

Like DLC, GexC1-x film has many unique properties and has attracted considerable research attention. It is an ideal infrared anti-reflective and protective film [25, 51]. GexC1-x film has a sphalerite-like structure, which is well combined with most infrared window materials such as ZnS and Ge, making it appropriate as the intermediate transition layer [6, 52, 53]. Liu et al. [54] show that the refractive index of GexC1-x films can be varied between 1.8 and 4.3 by changing the gas flow ratio, the hardness values of the films grew with the augmentation of the amount of carbon content in the films, and were higher than those of IR optical material Ge and ZnS (Fig. 3).

Fig. 3
figure 3

a Refractive index, b Vickers microhardness of GexC1-x films as a function of the gas flow ratio [54], and c Infrared transmittance of the multilayer antireflection coating [55]

Barr-Stround et al. show a classic work [56]. After GexC1-x ARC deposition on a ZnS substrate, the average infrared T was 91%, and the average R was only 2.5% in the 8–11.5 μm. The average infrared T of the products was far better than those obtained in other GexC1-x studies. DLC and GexC1-x inevitably exhibit serious performance degradation under service conditions, leading the optical properties of the films to decline. Even though GexC1-x has low absorption and low stress, it is difficult for the transparency in the infrared wavelength to meet the application requirements.

Alumina (Al2O3) has superior mechanical and optical properties and is a very important mid-infrared transparent material [57]. Its hardness of more than 15 GPa is higher than those of most oxide infrared anti-reflection films. Al2O3 film has good visible T [58,59,60,61,62,63,64,65,66,67]. Al2O3 is an amphoteric oxide with good chemical stability. Many metastable structures of Al2O3 can be obtained from room temperature to 1000 °C. It is difficult to control the preparation process to obtain a specific form of Al2O3, which makes the preparation of and research on Al2O3 more complex. The mechanical properties of amorphous Al2O3 film cannot meet the requirements. However, the methods of preparing high-quality crystalline Al2O3 film have many problems, such as high deposition temperature, complexity, and high cost, which seriously limit its application as an anti-reflective and protection material in the mid-infrared range [68,69,70,71,72,73]. Sadaf Bashir Khan et al. [74] describe the design and fabrication of amphiphobic AR coatings with reduced reflectance and enhanced mechanical resilience (Fig. 4b). The fabricated film showed better thermal stability (up to 300 °C) with stable AR efficiency, when an ultrathin dense coat of Al2O3 was deposited via atomic layer deposition (ALD) on the polymer-based bilayer SiO2 antireflective coating (P-BSAR). J. Gottmann et al. [58] uses pulsed lasers to deposit thin films of alumina and zirconia on polymers and glass as optical and protective coatings, the 400 nm thick Al2O3 films deposited at 1–3 Pa on PMMA substrates show a hardness of 800–900 HV and a comparatively small elastic modulus of 40–50 GPa (Fig. 5a).

Fig. 4
figure 4

a Panorama view of the nanoscratches on the coatings. b Graphical representation of critical failure of as-deposited and annealed coatings under progressive loading [74]. c Effect of a 100 nm thick alumina coating on the buckling load of a 2 μm-high, 1 μm-wide bare Si cone on a (001) wafer [75] and d Hardness and elastic modulus of Al2O3 films deposited on glass and PMMA [58]

Fig. 5
figure 5

a Pulsed laser deposition of alumina and zirconia thin films on polymers and glass as optical and protective coatings [58], b The Effect of Surface Anti reflection Layer on Transmittance and Hardness of Transparent Conductive Thin Films [62], and c Optical and mechanical properties of reactive sputtered alumina–silica mixtures [76]

Y2O3 has excellent high-temperature physical and chemical properties, such as mechanical properties (hardness, modulus, and corrosion resistance), optical properties (wide wave band, optical matching), and dielectric properties [77,78,79,80,81,82,83,84,85]. Because of their high-temperature stability and excellent mechanical and optical performances, Y2O3 films have been creatively applied in ARPC of infrared window materials [86]. Among the numerous infrared coatings, Y2O3 film is one of the preferred materials for infrared windows due to its resistance to aerodynamic thermal failure and stable high-temperature optical and mechanical properties [87, 88]. The refractive index of Y2O3 films varies from 1.6 to 2.3 according to the preparation method, substrate, and conditions [89,90,91,92,93]. In addition, the bandgap of Y2O3 film is greater than 5.5 eV, which means that the film is transparent (T > 80%) in the visible range [94]. The film also has good T in the infrared range, and the plasma wavelength is 12 μm [95]. Liu et al. [96] shows low absorption in the far-infrared range and low stress, making it suitable for thin film fabrication, as shown in Fig. 6.

Fig. 6
figure 6

a Picture of infrared window surface after thermal shock [96]. b The transmittance of the Y2O3/ZnS/Y2O3. c Hardness and modulus of yttrium oxide films prepared under different conditions on ZnS substrates

Y2O3 film deposited on ZnS is widely utilized for anti-reflective and transmitting windows [97]. ZnS has low hardness, which cannot meet the requirements for use in harsh service environments, such as robustness against rain and sand erosion without fracture failure. Y2O3 film deposited on ZnS can compensate for its shortcoming of low hardness and improve the T of ZnS windows. The minimum hardness of Y2O3 film is greater than 6 GPa, which is greater than the hardness of ZnS (3.5 GPa) [98].

It is worth mentioning that Prof. Zhu has invented the Anti-Aerodynamic Heating and Force Failure Infrared Anti-Reflection Film [89, 92, 93, 99,100,101,102,103,104,105], which is based on Y2O3 thin film material, prepared by magnetron sputtering technology, and achieved high-quality bonding with ZnS infrared window through interfacial modulation. The window has an average transmittance of over 90% in the 8–12 µm band, a twofold increase in hardness, and can resist oxygen-propane flame impacts with a heat flux of 563.8 kW m−2. The invention breaks through the bottleneck of high hardness and wideband transmission compatible film material, brings the anti-reflection protection film from the visible and near-infrared bands into the long infrared band, and solves the failure problem of the long infrared window under the impact of high heat flow density.

4 Infrared transparent conductive films

Transparent conductive films require not only good transparency in the visible light range, but also excellent optical transmittance in the infrared range. Therefore, higher performance requirements need to be proposed for existing mature transparent conductive films [106,107,108,109,110,111,112,113,114,115,116]. However, there is a trade-off between the electrical conductivity and optical transmittance of transparent conductive films. Generally, films that simultaneously exhibit good transmittance in the visible and infrared ranges and good electrical conductivity are called infrared transparent conductive films (ITCF). According to the Drude [117, 118] free electron theory, high concentrations of free electrons will absorb and reflect photons in both the visible and infrared ranges, thus reducing the transmittance in the visible range and the transmittance range in the infrared range (plasma wavelength blue-shift). If the concentration of carriers is reduced to improve the optical transmittance, this will have a negative impact on the electrical conductivity of the film. Therefore, it is difficult to balance the wide-band high transmittance in the infrared range and the required electrical conductivity for infrared transparent conductive films.

Due to the interaction between photons and electrons, it is difficult to achieve both high transmittance and high electrical conductivity in the wide-band infrared range for infrared transparent conductive films. According to the Drude free electron theory, the interaction between photons and electrons can be used to explain the significant difference in reflectance and transmittance between metal or transparent conductive film layers. To better explain the complex interaction between the optical and electrical properties of transparent conductive films, the Maxwell equations are combined with the Lorentz oscillator model. By using the film’s dielectric constant ε and plasma wavelength λp, the relationship between the electrical conductivity and transmittance of the film can be expressed using the real and imaginary parts of the complex dielectric constant as \({\varepsilon }_{r}={\varepsilon }_{\infty }(1-\frac{1}{{\omega }^{2}{\lambda }_{p}^{2}})\) and \({\varepsilon }_{i}=\frac{{\varepsilon }_{\infty }}{{{\omega }^{3}\lambda }_{p}^{2}\tau }\).

τ is the relaxation time of the free electron, ε is the high frequency permittivity, ε0 is the free space permittivity (8.8542 × 10–12 F/m), ɛr is the real part of the complex permittivity, ɛi is the imaginary part of the complex permittivity, and ω is the vibrational frequency of light. From these two equations, it can be seen that for the plasma wavelength λp, when the vibration frequency of photons is less than the vibration frequency of photons at the plasma wavelength (ω < ωp), the real part of the complex dielectric constant ɛr can be calculated to be negative, while the imaginary part of the complex dielectric constant ɛi is calculated to be positive at this time, at which time the transparent conductive film becomes more reflective at this wavelength. When the vibration frequency of photons is greater than the vibration frequency of photons at plasma wavelengths (ω > ωp), the real part of the dielectric constant ɛr≈ɛ, while the imaginary part of the dielectric constant ɛi≈0. At this time, the vibration frequency of electrons cannot be synchronized with the vibration frequency of photons, and only a small number of free electrons absorb photons, so the photoelectric film shows high permeability. When the real part of the dielectric constant is zero ɛr = 0, i.e. ω = ωp, the optical transmittance of the film undergoes a significant abrupt change at this time, and the wavelength corresponding to this position is the plasma wavelength of the film, such as λp in Fig. 7, the plasma wavelength of the film can be expressed as \({\lambda }_{p}=2\pi {c}_{0}\sqrt{\frac{{m}^{*}{\varepsilon }_{0}{\varepsilon }_{\infty }}{n{e}^{2}}}\), Where σ0 is the free space conductivity, c0 is the speed of light, n is the carrier concentration and m* is the effective carrier mass. It is clear from this equation that the plasma wavelength is inversely proportional to the concentration of free electrons, meaning that at high concentrations of free electrons the film’s IR band transmission range will become smaller. If the carrier concentration of the film needs to be reduced in order to obtain a wider transmission range, the conductivity of the film will be negatively affected, resulting in a difficult reconciliation between the optoelectronic properties. At the same time, the plasma wavelength is proportional to the effective mass m* of the carriers in the film, so an alternative approach can be taken to adjust the plasma wavelength of the film by changing the effective mass of the carriers in the film. A review of the literature shows that the effective mass of free electrons in conventional n-type transparent conductive oxide films is small, about 0.5me, while the effective mass of holes in p-type transparent conductive oxide films is much larger than that of electrons, about 3me [119].

Fig. 7
figure 7

The transmittance of transparent conductive films as a function of wavelengths [120]

The difficulty of reconciling the photovoltaic properties of transparent conducting films is the main scientific problem facing the field. Although the problem of small plasma wavelengths due to high carrier concentrations can be solved by reducing the effective mass of free electrons, the effective mass of carriers is an intrinsic property for a particular type of transparent conducting film and it is difficult to change its effective mass.

The focus, then, can only be temporarily on the carrier concentration, increasing the plasma wavelength by reducing the content of free electrons in the film. In order to reduce the carrier concentration of the film without affecting the conductivity, other factors affecting the conductivity, such as the thickness of the film and the free electron mobility, need to be explored to reduce the negative effects of a smaller carrier concentration by adding another variable.

Transparent conductive oxide films have excellent electrical conductivity due to the presence of significant vacancy defects or elemental substitution defects. In contrast to metallic films, the large forbidden band width allows for good high transmission properties in the visible to near-infrared wavelengths. With the development of technology, there are more and more ways to prepare oxide transparent conductive films, and their excellent properties have led to a significant development of these films, as shown in Table 2.

Table 2 Summary on the physical quantities related to infrared transparent conductive materials

The earliest transparent conductive films were oxides, with CdO and PbO transparent conductive oxide films prepared by Bädeker in the early part of the last century, when the resistivity of the films could reach 10–3 Ω−cm [189]. With the development of deposition techniques and theories, a variety of systems have emerged so far for transparent conductive oxide thin film materials. Depending on the type of element, they can be divided into binary material systems (e.g. CdO, In2O3, ZnO and SnO2) and ternary doped material systems (In2O3: Sn, In2O3: Ge, In2O3: Mo, In2O3: Ti, CdO: In, CdO: Sn, CdO: Al, SnO2: F, SnO2: Sb, ZnO: Al (ZnO: B, ZnO: In and ZnO: Y, etc.). Most of these films are amorphous and polycrystalline materials, and only a few researchers have obtained single-crystal films by heterogeneous epitaxy or homogeneous epitaxy techniques in order to improve the performance of the films. As the name implies two important properties of transparent conductive films are their transparency and conductivity, and different material systems have different optoelectronic properties.

The carrier concentration and mobility in oxide transparent conducting films vary significantly at different temperatures, from which it can be inferred that the mobility will also vary significantly at different free electron concentrations. Therefore, the carrier and mobility of different oxide transparent conducting films were reviewed to explore the coupling between the two, as shown in Fig. 8 [190,191,192,193,194,195,196,197]. From the figure, it can be seen that the mobility of free electrons decreases with increasing carrier. When the concentration of free electrons in the film is greater than 1021 cm−3, the mobility starts to decrease rapidly. For the doped ZnO film, the maximum concentration of free electrons is 1.9 × 1021 cm−3, at which point the mobility in the film is only 11 cm2 V−1 s−1, and the doped elements in this case account for only 3.7 at% of all elements, which shows that the impurity defects formed by doping can have a significant effect on the mobility The impurity defects formed by doping can have a significant effect on the mobility [194, 198]. Higher carrier concentrations can be obtained for heavily doped films, although they produce a higher content of scattering centres, leading to a decrease in mobility. The problem of low mobility can be improved by improving the crystalline quality of transparent conductive films and reducing the defect concentration using processes such as substrate thermal assist or post-annealing [192, 198].

Fig. 8
figure 8

Free electron mobility as s function of carrier concentration [199]

The carrier concentration in a film can have a significant effect on the optical transmittance of the film and its value is a major factor in determining the conductivity of the film, so the conductivity and optical transmittance of the film can be regulated by adjusting the carrier concentration. However, if the conductivity and transmittance of a film are only regulated by changing the carrier concentration, this is not sufficient to resolve the contradiction between the optoelectronic properties of the film, so it is necessary to combine other influencing factors to regulate the properties. In order to better analyse and resolve the contradiction between the transparent conductive and optoelectronic properties, the effect of mobility on the plasma wavelength is indirectly demonstrated by using the product and ratio of mobility and carrier concentration, as shown in Figs. 9 and 10 [131, 200,201,202,203,204].

Fig. 9
figure 9

The plasma wavelength as a function of carrier concentration [199]

Fig. 10
figure 10

The relationship between carrier concentration, mobility and plasma wavelength of different thin film materials [199]

As can be seen from Fig. 10, when the product of the two is a constant value, i.e. when the resistivity is a constant value, this corresponds to two plasma wavelengths, one large and one small. As can be seen from the above analysis, a larger value of carrier concentration will result in a smaller plasma wavelength, which means that the mobility is a smaller value at this point. When the plasma wavelength corresponds to a larger value, the carrier concentration in the film is smaller and a larger value of mobility is required to ensure that the product is the same (same resistivity). Therefore, it can be concluded that a large plasma wavelength can be obtained by increasing the mobility by decreasing the carrier concentration, thus obtaining excellent optical transmission properties without negatively affecting the electrical conductivity of the film [205].

Koffyberg [206] at Brock University, Canada, has theoretically calculated that by modulating the carrier concentration and mobility, it is possible to achieve high transmission in the infrared band of single-crystal CdO crystals with high performance conductivity. Using molecular beam epitaxy and simulation calculations, Sachet [207] et al. at North Carolina State University (USA) have prepared dysprosium-doped single-crystal CdO transparent conducting films, achieving high transmission and high conductivity in the mid-infrared band. As the ionic radius of Dy3+ is almost equal to that of Cd2+, it can be assumed that no scattering centre is generated when Dy replaces Cd, resulting in a high mobility. The doping of the Dy element also suppresses the formation of oxygen vacancies, further reducing the formation of defects in the film, so that the mobility of CdO: Dy films can reach 500 cm2 V−1 s−1. Figure 11 shows the real versus imaginary parts of the dielectric constant and the carrier concentration versus mobility for CdO: Dy films When the real part of the dielectric constant is zero, the corresponding wavelength is the plasma wavelength of the film, which is the wavelength at which ɛr is zero in the graph. When the carrier concentration is less than 1020 cm−3 and the mobility is close to 500 cm2 V−1 s−1, the plasma wavelength of the film can reach 5 μm and the resistivity can also reach 1.33 × 10–4 Ω−cm, which makes the CdO: Dy film have both high mid infrared wavelengths and high electrical conductivity.

Fig. 11
figure 11

Calculated real (solid lines) and imaginary (dashed lines) dielectric functions for selected CdO: Dy films [199]

The analysis of the optoelectronic properties of infrared transparent conductive films shows that the contradiction between high transmission and high conductivity in the broad band is mainly caused by the carrier concentration, which is also in accordance with the Drude free electron theory. This contradiction is mainly reflected in the fact that a high concentration of free electrons will reflect infrared light, causing a blue shift in the plasma wavelength and reducing the optical transmittance of the film. If the carrier concentration is reduced, this will destroy the conductivity of the film. There is therefore some debate as to whether it is possible to achieve IR transparency and conductivity at the same time.

Here, we summarise some strategies for improving the optoelectronic properties of IR-transparent conducting materials. For N-type oxides, the main strategy is to maintain a small carrier concentration, and researchers have commonly used controlled oxygen vacancies and the introduction of metallic/non-metallic impurities to achieve these aims.

As illustrated in Fig. 12, Xu [123, 198, 208, 209] revealed the basic phase structure and property control mechanism of SnOx films, solved the controllable growth of preferentially oriented tin oxide films, and revealed the inherent relationship between crystal orientation, carrier effective mass, and optoelectronic properties. By using the in-situ plasma layer-by-layer annealing technology for film performance enhancement, a new approach combining atomic layer deposition and in-situ annealing technology was developed for the preparation of transparent conductive films. The SnOx film Hall mobility decreased from 18.25 cm2/Vs to 12.06 cm2/Vs, the carrier concentration increased from 5.12 × 1019 cm−3 to 4.425 × 1020 cm−3, and the resistivity was 1.117 × 10−3 Ω·cm. Hydrogen-doped In2O3:H infrared transparent conductive films were grown by atomic layer deposition, and the form of hydrogen in the indium oxide lattice was determined using secondary ion mass spectrometry and X-ray photoelectron spectroscopy. The mechanism of mobility change was mastered, providing ideas for the preparation of high-quality infrared transparent conductive films.

Fig. 12
figure 12

a Reaction mode-controlled crystal structure and optical and electrical properties of SnOx infrared transparent conducting films [123], b Plasma treatment to tailor growth and photoelectric performance of plasmaenhanced atomic layer deposition SnOx infrared transparent conductive thin films [198], and c Simulation of the brain exposed to electromagnetic radiation from a radar antenna [208]

As shown in Fig. 13, Guo [126, 210] Using high power impulse magnetron sputtering (HiPIMS) technology to create a high-density oriented plasma by constructing a pre-magnetic field, the selective growth of In2O3 transparent conductive thin film with columnar crystals was achieved. The resistivity was as low as 1.28 × 10–4 Ω•cm and the average transmittance of visible to near-infrared light was over 80%. By introducing In-N bonds and O-N bonds to effectively dope nitrogen, p-type In2O3-N infrared transparent conductive thin film with nitrogen-containing holes was obtained, achieving a long-wavelength redshift. The average transmittance of visible to mid-infrared light reached 80%, and the electromagnetic shielding performance reached 10 dB. In addition, Guo et al. [158, 211,212,213] investigated the effects of different annealing temperatures and annealing times on the optical and electrical properties of Mg-C-O–H films, with the highest mobility of 58.38 cm2V−1 s−1.

Fig. 13
figure 13

a Modulation of optical and electrical properties of In2O3 films deposited by high power impulse magnetron sputtering by controlling the flow rate of oxygen [126]; b High-quality transparent conductive indium oxide film deposition by reactive pulsed magnetron sputtering, Cross-sectional TEM images of In2O3 film deposited on a sapphire substrate at temperature of 400 C, Transmittance and Electrical resistivity of In2O3 films deposited on quartz glass [210]; c Mg(OH)2 film with nanoflake structure [211], and d Visible transmittance of hydrated magnesium-carbon films on quartz glass as a function of substrate temperature. The thickness of the hydrated magnesium-carbon films is approximately 1.5 μm(left); Transmittance of films formed on ZnS substrates as a function of substrate temperature treatment. Typical film thickness is approximately 1.5 μm(right) [212]

P-type transparent conductive films are also considered as one of the alternative materials for electromagnetic shielding layers. With the development of transparent conductive films, p-type films are mainly divided into copper-based, tin-based, and nickel-based oxide films. The holes in the p-type film are multi-carrier, and the reflection of infrared light will be smaller than that of free electrons. At this time, the contradiction between optics and electronics is not obvious, so this type of film has good transmission performance in the infrared light band. However, compared with n-type films, the hole concentration in p-type transparent conductive films is lower, resulting in poor conductivity and unsatisfactory electromagnetic shielding results. Since the bandgap of most p-type films is less than 3 eV, the transmittance in the visible light band is relatively low. The main strategy for p-type oxides is to reduce the effective mass.

Geng [187, 214, 215] prepared acceptor-doped CuI thin films using both physical and chemical methods, and achieved controllability of the S doping concentration, enhancing the p-type conductivity of CuI thin films. Geng also revealed the formation mechanism and model of (111) single-oriented growth. By constructing p orbitals of I-5p and S-3p anions, insensitive to amorphous structure, a hole transport pathway was obtained, resulting in amorphous films and achieving an extremely high hole conductivity of p-type semiconductor of 5000–1000 S/cm, which is 100 times higher than that of common p-type semiconductors. In Gao et al. [216] previouse work, a P-type mid-infrared transparent conductive LaSe2 film is obtained by a two-step method combining the RF sputtering and a selenized annealing process (Fig. 14d). the LaSe2 film has a higher conductivity (about 3.6 S/cm) than that of traditional P-type mid-infrared TCMs, which is attributed to its ten times higher carrier concentration (about 1019 cm−3) and a much smaller hole effective mass (about 0.34 me) compared to the conventional P-type mid-infrared TCMs. The transmittance (about 75%) is at the same level as that of the traditional P-type mid-infrared TCMs.

Fig. 14
figure 14

a Transmittance spectra (T) of CuI films fabricated at different temperatures of iodine vapor (the inset shows photographs of CuI), b Comparison of the FOM values of CuI films prepared in this work, and the traditional iodination method by one-step process in a previous report [214, 217], Mid-infrared photoelectric properties of LaSe2 films compared with those of the conventional P-type mid-infrared transparent conductive films. c Conductivity and transmittance. d Carrier mobility and carrier concentration [216]

In addition to the transparent conductive oxide films mentioned above that can be used as electromagnetic shielding layers, electromagnetic shielding layers can also be classified into non-oxide film systems such as metal films, metal grid, and new carbon materials, depending on the material type. Different types of films have their own advantages and disadvantages and can be chosen according to different application scenarios.

Metal films have excellent conductivity due to their high carrier concentration. However, they also contain a large number of free electrons, resulting in a high absorption coefficient while being highly conductive. For metal films, their bandgap is small (Eg < 3 eV), and when light shines on the surface of the metal film, electrons receive the photon energy and jump to an empty energy level above the Fermi level. Therefore, the metal film can absorb all the energy of the incident photons, resulting in poor light transmission performance. Aleksandar and others used the Brendel-Bormann model to calculate the reflectivity of four metal films, Au, Ag, Pt, and Al, with a thickness of 100 nm on a GaAs substrate. The calculation results showed high reflectivity. Although the optical transmittance of metal films is poor, it can be achieved by reducing the thickness of the metal film. Guillén and others used magnetron sputtering to deposit 8 nm Ag and Cu films on a calcium-aluminum glass substrate, and the transmittance of the two films in visible light was close to 60%. However, due to the reflection of free electrons, the transmittance of the two films in the near-infrared region was less than 20%, showing strong reflection. Since the film needs to be deposited thinly to achieve a certain degree of optical transmittance, the metal film will produce an island structure during preparation, which reduces the conductivity and increases light reflection. Metals are highly active, which leads to poor chemical stability, and their low hardness further limits the application of metal films.

Metal grid is a unit structure arranged periodically on the substrate layer (usually metal). The periodic arrangement of the array ensures the high conductivity of the grid and achieves electromagnetic shielding, and the existence of periodic structures can also achieve high transmittance of special wavelength bands. Ulrich of Germany first published a simple grid structure film with optical transmittance and electromagnetic shielding performance. The optoelectronic properties of the metal grid are related to parameters such as linewidth and period, and there is a trade-off between the two. Suitable linewidth and period need to be simultaneously adjusted to achieve good transmittance and electromagnetic shielding properties in the required wavelength band, thus obtaining a compromise solution. Due to the periodic structure of the grid, the film will produce a certain diffraction order, except for the zero-order diffraction imaging, other diffractions are interference signals. Therefore, the grid will reduce signal strength, reduce light flux, and increase signal-to-noise ratio.

New transparent conductive carbon films mainly include graphene and carbon nanotube transparent conductive films, which have received much attention due to their unique structure, optoelectronics, mechanics, and thermodynamics properties. Due to their special optoelectronic properties, carbon nanotubes and graphene can achieve good electromagnetic shielding and optical transparency. Compared with metal films and metal grids, carbon films have advantages such as high transparency, flexibility, and chemical stability. However, carbon films also have disadvantages such as low conductivity and low mechanical strength.

Recently, based on the intrinsic relationship between material structure and optoelectronic properties, Zhu and Hu et al. [218] proposed and preliminarily verified two potential infrared transparent conductive material systems: heavy metal chalcogenides with high dielectric constants (Fig. 15) and topological semimetals with low carrier concentration and high relaxation time (Fig. 16).

Fig. 15
figure 15

Prediction and verification of heavy metal sulfides with high dielectric constants [218]

Fig. 16
figure 16

Prediction and verification of topological semimetals with low carrier concentration and high relaxation time [218]

With the rapid development of laser equipment and experimental analysis techniques, researchers have been investigating and identifying the primary destructive effects of lasers on thin film materials. As a result, efforts have been made to study thin film materials that can be used for laser protection, and significant progress has been achieved. Laser protection technology can be categorized into two types based on the working principles: linear optical-based and nonlinear optical (NLO)-based. Additionally, there are also thermal phase-change thin film materials and liquid crystal thin film materials.

5 Optical limiting film

As shown in Fig. 17, Wang et al. [219]. successfully synthesized a significant amount of graphene monolayers and a small quantity of few-layer graphene by dispersing graphite in an inorganic solvent for the first time. Subsequently, the graphene samples were characterized using transmission electron microscopy (TEM) and Raman spectroscopy, confirming the presence of unoxidized and defect-free graphene. It was observed that the graphene dispersion exhibited a significant NOL response to 532 and 1064 nm nanosecond laser pulses, indicating its potential for broadband optical limiting applications. Dong et al. [220] investigated the NLO behavior of MoS2, MoSe2, WS2, WSe2 and graphene nanosheets at nanosecond laser pulses. Both TMDC and graphene exhibit strong optical limiting responses originated from NLS under the excitation of near infrared laser, while TMDC show a joint effect of SA and NLS at 532 nm. The fact that TMDC possess better optical limiting responses at higher pulse energies than graphene makes these 2D nanomaterials to be promising candidates for broadband optical limiters. Shi et al. [221] used few-layer BP as an electron donor and C60 as an electron acceptor to prepare BP:C60 blends (weight ratio: 1:1), which were embedded into a PMMA matrix to produce BP:C60/PMMA films with good optical quality. Upon illumination with a 450 nm laser, the BP:C60 blends exhibited an EPR signal with a g value of 2.0030 and a ΔHpp of 1.27 G. The signal intensity of the BP:C60 blends was found to be highly reduced when compared to the blends before illumination due to the photoinduced charge transfer from BP to C60 producing BP˙ and C+60. In contrast to C60, BP and the non-annealed BP:C60 blends, the annealed BP:C60 blends show significantly enhanced NLO and OL responses. The observed nonlinear coefficient changed from 39.39 cm GW−1 before annealing to 241.73 cm GW−1 after annealing at the pulse energy of 400 μJ. The most widely researched phase transition materials presently encompass various vanadium oxides, notably vanadium dioxide (VO2) and vanadium pentoxide (V2O5). VO2, a thermochromic material, exhibits a monoclinic structure near room temperature, displaying semiconductor characteristics with a bandgap width of 0.7 eV, corresponding to a cutoff wavelength of approximately 1800 nm. It possesses high migration rate and transmittance. At a temperature of 68 °C, VO2 undergoes a transition to an orthorhombic structure, assuming a metallic state with reflective properties towards light. The transition time is on the order of nanoseconds. Owing to VO2’s lower phase transition temperature and rapid responsivity, it has gained significant favor in fields such as laser protection.

Fig. 17
figure 17

a Broadband Nonlinear Optical Response of Graphene Dispersions, b Optical Limiting and Theoretical Modelling of Layered Transition Metal Dichalcogenide Nanosheets, and c Donor–acceptor type blends composed of black phosphorus and C60 for solid-state optical limiters

As shown in Fig. 18, Xia et al. [222] prepared V2O5 on a sapphire substrate through radio frequency magnetron sputtering and investigated the influence of ambient atmosphere on the structure, optical properties, and morphology of the annealed thin films. They discovered that the V2O5 films undergo four distinct transformation behaviors during the annealing process. When annealed in 0.1% O2/Ar, the film transformed into VO2(B), resulting in a decrease in infrared transmittance. In the case of 0.01% O2/Ar, before the complete transformation of V2O5 into VO2, metastable phases VO2(B) and VO2(R) were observed as intermediate crystalline phases (M), leading to periodic changes in infrared transmittance. When annealed in pure argon atmosphere, V2O5 transformed into VO2(R) and VO2(M) without the formation of VO2(B) phase. This study provides a more detailed understanding of the thermal transformation process and data reference for the preparation technique and laser protection applications of vanadium oxide materials. Huang et al. [223] reported a nanostructured VO2 thin film fabricated by direct current magnetron sputtering. Both the electrical and IR transmission tests show that the films exhibit phase transition at a low temperature, around 45 °C. In addition, the transmission spectra of VO2 thin film reveal that the thin film possesses excellent switching property in IR range at critical transition temperature, with switching efficiency of 52% at 3.39 μm. Besides, the characteristic transition obtained from the radiated VO2 thin film reveals that the fabricated VO2 thin film exhibits excellent high power resist property, thus makes it a promising material for smart window and laser protection application. Wang et al. [224] utilized reactive ion beam deposition (RIBD) technique and post-annealing method to fabricate vanadium dioxide (VO2) thin films containing nanoparticles on quartz glass substrates. The VO2 films exhibited a first-order reversible metal–insulator transition (MIT) at 68℃. They also demonstrated photo-induced MIT, where a quasi-metallic phase of monoclinic VO2 appeared during this process. The exceptional laser protection performance of the VO2 films was comprehensively verified. Laser protection experiments conducted on silicon photovoltaic cells showed that the VO2 thin films improved the system’s anti-interference capability by approximately 2.6 times, demonstrating the suitability of VO2 films in laser protection systems. The threshold power density for MIT transition of a 100 nm-thick VO2 film at room temperature was investigated using a continuous-wave laser with a wavelength of 1.08 μm and a continuous tunable system, yielding a value of 4.35 W/cm2. Furthermore, the VO2 film exhibited a very high-power density damage threshold (404 W/cm2). The low MIT transition threshold and high damage threshold of the VO2 film further confirmed its suitability as a key component in laser protection systems. The VO2 film possesses high switching efficiency and a high damage threshold, making it suitable for applications such as optical switches, smart windows, and optoelectronic devices.

Fig. 18
figure 18

a Thermal transition behaviors of vanadium pentoxide film during post-deposition annealing [222], b Infrared characteristics of VO2 thin films for smart window and laser protection applications [223], and c Infrared laser protection of multi-wavelength with high optical switching efficiency VO2 film [224]

6 Summary and challenge

We have reviewed the current progress in infrared transparency under opto Electrical thermo and mechanical four environments.

For application under extreme thermo and mechanical environments, High hardness film materials can be used as protective films. Still, the higher refractive index can limit their application in antireflection, and infrared antireflection and protective films are important for optical components, especially windows. With continued research, current anti-reflective protective film materials have the hardness and elastic modulus to withstand solid particles during high-speed flight. However, the principles of synergy between relevant optical and mechanical properties are still worth exploring, and the main challenge in the future is to improve the resistance of thin films to ultra-high temperatures and thermal shocks. In extreme environments, coatings must have high surface hardness, stability, and adhesion. Infrared windows must also have high transmittance at long wavelengths. Antireflection and protective coatings can resist erosion by sand and raindrops, prevent scratching, and enhance the mechanical properties of materials. However, on one hand, the stability of infrared optical materials at high temperatures is poor, and they cannot withstand aerodynamic heating and forces. The physical mechanisms that simultaneously achieve high infrared transmittance and excellent mechanical performance need to be further explored.

For application under extreme electro environment, the preparation, structure, transparency, conductivity properties, and material design principles of N and P-type compounds are highlighted. The analysis of IR transparent conducting films according to the Drude free electron theory shows that the contradiction between the IR transmission properties and the electrical conductivity of the films is mainly reflected in the fact that a high carrier concentration (n) increases the electrical conductivity (σ), but decreases the plasma wavelength (λp) and the IR transmittance (T). Currently, N-type metal compounds generally suffer from high σ but low T, but P-type metal compounds have high T and large λp but low σ in the infrared band. Although some progress has been made in material preparation, overall transparency and conductivity still need to be improved. Therefore, resolving the paradox that infrared transparent conductivity is hardly compatible is an important challenge for the future. Although some progress has been made in these studies, it is difficult to meet practical applications. Therefore, strategies such as developing new materials and optimizing growth methods are still needed in the future to synergize the contradictions between the two. Interpreting the complex relationship between microstructure and optoelectronic properties is crucial and challenging. The establishment of a synergistic mechanism between infrared transparency and conductivity requires further research in the quantitative analysis of optical and electrical properties. Furthermore, as the development of information technology and aerospace progresses, the service environments for materials are becoming increasingly complex, and multifunctional integration is a fundamental trend. In the future, it is necessary to develop thin film materials that are transparent and conductive in the far-infrared region. Additionally, with the expansion of usage environments, it is important to develop infrared-transparent and conductive thin film materials with good stability in heating and oxidizing environments.

For application under extreme opto environment, we have reviewed the basic concepts of laser protection technology and several mechanisms by which thin film technology is applied to laser protection technology (optical limiting technology). Typical two-dimensional semiconductor nonlinear optical materials and vanadium oxide are considered suitable choices for laser protection applications. In the future, laser protection technology will develop towards high protection efficiency and ultra-fast response, while mitigating the thermal radiation interference caused by laser absorption on detection components. Firstly, laser protection technology for detection components needs to balance transmission and protection requirements within the detection wavelength range. It should achieve a perfect switch between high transmission in the detection state and low transmission in the protection state, with high optical modulation depth and ultra-fast response characteristics. However, current mechanical protection technology achieves high protection efficiency but has a slow response speed, while protection technology based on phase-change materials still has relatively low single-layer film protection efficiency. Therefore, the development of laser protection technology with ultra-fast response and high protection efficiency is of great significance. Secondly, some laser protection technologies exhibit light absorption effects and even utilize light absorption for optical limiting. Light absorption inevitably leads to thermal effects and thermal radiation. Infrared detection components based on infrared technology are highly sensitive to infrared signals, and the thermal effects caused by light absorption may interfere with the detection components, leading to their failure. Upgrading technology to weaken or even eliminate the thermal radiation interference caused by light absorption during the protection process will greatly enhance the compatibility of protection technology.

We anticipate that this review will serve as a valuable resource for both experts in the fields of physics, materials, and related disciplines, as well as beginners seeking guidance. Its comprehensive coverage is expected to inspire further research and exploration in the realms of advanced materials, information technology, energy systems, and advanced optics applications.