1. Introduction
The damage inflicted on optical films represents a significant constraint on the performance of laser systems. Hence, it is imperative to enhance the laser-induced damage threshold (LIDT) of such films for the advancement of high-power lasers [
1]. Hafnium oxide (HfO
2) emerges as a pivotal material for high-refractive-index films owing to its exceptional attributes, including a high melting point (3031 K), a wide bandgap (5.8 eV), an ultra-wide transparent band (220 nm to 12 μm), a high LIDT, and excellent mechanical properties. Consequently, HfO
2 films are highly suitable for the fabrication of various high-performance laser coatings [
2,
3,
4,
5]. Andre et al. successfully synthesized HfO
2 films with an LIDT of 8.0 J/cm
2 (at 1064 nm, 3 ns) using the ion beam-assisted deposition method [
6]. Jena et al. employed electron beam evaporation to produce HfO
2 films, achieving an LIDT of 9.8 J/cm
2 at 1064 nm and 3 ns [
7]. Grosso et al., utilizing the sol–gel method with Hf (OPr)
4 as a precursor, prepared HfO
2 films with an LIDT reaching 14.6 J/cm
2 (at 1064 nm, 12 ns) [
8]. In our previous research employing the sol–gel technique, we achieved an LIDT of 31.6 J/cm
2 (at 1064 nm, 12 ns) for HfO
2 films [
9].
The expanding applications of laser systems in diverse and extreme environments, including high temperatures, pressures, and radiation, necessitate a comprehensive exploration of the yet unknown damage characteristics of thin-film devices under these extreme conditions [
10,
11,
12]. For the studies of films at different temperatures, Mikami et al. found that the LIDT at 1064 nm and 4 ns of SiO
2 and HfO
2 films decreased with the increase in temperature from 123 K to 473 K [
13]. Xu et al. deposited Ta
2O
5 films by a dual ion beam sputtering method and found that the LIDT at 633 K (10.1 J/cm
2) decreased by 32% in reference to that measured at 298 K [
14]. Zhu et al. prepared sol–gel ZrO
2 films by using ZrOCl
2·8H
2O as a precursor and found that the LIDT at 523 K (23.9 J/cm
2) was 22.7% lower than that at normal temperature [
15]. Investigations focusing on films at different temperatures have revealed that, compared to normal temperatures, films exhibit heightened susceptibility to laser damage under high temperatures, ultimately impacting the overall performance of laser systems. Furthermore, it is vital to recognize a notable disparity between films employed at room temperature and those utilized in high-temperature environments. Under harsh conditions, the physical and chemical properties of the film are more prone to alteration. While films typically exhibit relatively stable properties at room temperature, exposure to high temperatures can induce the formation of surface and internal defects over time, thereby significantly impairing their resistance to laser damage [
16]. Consequently, when investigating the effects of high-temperature environments on film damage, it is crucial to evaluate the progressive evolution of damage over time. However, the current body of literature lacks substantial research in this specific area, highlighting the need for further exploration and understanding.
This study focused on the exploration of laser damage characteristics at high temperatures in sol–gel HfO2 films prepared using HfCl4 as a precursor, HNO3 as a hydrolysis inhibitor, and a copolymer of silicone and polyaldoxyl ether (CSPE) as an additive. The investigation involved examining temporal changes in the viscosity of the prepared HfO2 sol, as well as variations in its optical properties, microstructure, surface topography, and absorption following annealing at 423 K. To assess the laser damage resistance, an in situ high-temperature LIDT testing platform was employed to investigate the LIDT of the films under various heating times at 423 K. Additionally, a comprehensive analysis was conducted to elucidate the temporal evolution characteristics and underlying mechanisms of laser damage at this elevated temperature.
2. Experimental Details
To prepare the HfO2 sol, firstly, 2 g HfCl4 (99.5%, Aladdin, Shanghai, China) was dissolved in 40 mL ethanol (99.8%, Sinopharm Chemical Reagent, Shanghai, China) under a nitrogen atmosphere and stirred for 0.5 h; then, the additives 2.16 mL HNO3 (99.5%, Aladdin, Shanghai, China), 0.65 mL CSPE (Dow Corning, Shanghai, China), and 0.45 mL water were slowly added to the above solution every 15 min in order and stirred continuously for 1 h. Finally, the solution was aged at 276 K to form HfO2 sol.
Before preparing the films, BK7 substrates were carefully cleaned by ultrasonication with ethanol. HfO2 films were prepared by the dip-coating method using the sol, aged for 5 days, at a rate of 60 mm/min, with the room temperature controlled at 298 K and the humidity at less than 40%. After each coating, the films were heated at 353 K for 10 min in air. The prepared films were baked at 353 K for 1 h to completely evaporate the solvent ethanol. The annealing process occurred in air, and the temperature of the films increased to 423 K with heat preservation for 0.5 h, 2 h, 24 h, 7 d, and 15 d, respectively.
Sol viscosity was measured by a glass capillary viscometer at a relative humidity of 40% and a temperature of 293 K. The transmittance of the films was measured using a Lambda 750S spectrophotometer (Perkin Elmer, Shelton, CT, USA). The microstructure of the films was determined by a D8 Advance X-ray diffractometer (XRD) (Bruker, Billerica, MA, USA). The surface topography was measured using Dimension V atomic force microscopy (AFM) (Bruker, Billerica, MA, USA), followed by calculating the root mean square (RMS) roughness. The surface thermal lensing (STL) technique was applied to characterize the film absorption [
17]. A single-mode, 1064 nm YAG laser served as the pump source. The probe beam was a 20 mW He-Ne laser at a wavelength of 632.8 nm with a diameter of 500 μm. The sensitivity of the measurement was 1 ppm. The surface thermal lensing (STL) technique was applied to characterize the film absorption [
17]. The LIDT of the films was tested following the “1-on-1” regime according to ISO standard 11254-1, using a 1064 nm and 12 ns Nd:YAG laser in single longitudinal mode with up to a 5 Hz repetition rate [
18]. The experimental set-up of the in situ high-temperature LIDT testing platform is shown schematically in Ref. [
19]. The Q-switched Nd:YAG was focused to provide a far-field circular Gaussian beam with a diameter of 0.306 mm at 1/e
2 of the maximum intensity. The sample was placed inside a temperature-controlled chamber and was driven by a stepper motor. Before the laser damage test, the sample was heated to 423 K and kept there for 0.5 h, 2 h, 24 h, 7 d, and 15 d, respectively. It is worth mentioning that the heating rate remained the same as that in the annealing treatment in this experiment. The LIDT was defined as the incident pulse energy density when damage occurred at a 0% damage possibility. The total error was about 12% in the LIDT measurement. The damage morphology after laser irradiation was evaluated by a Sirion 200 field emission scanning electron microscope (FESEM) (JEOL, Tokyo, Japan).
3. Results
As shown in
Figure 1, the sol viscosity gradually increased as the aging time was prolonged. However, the overall increase in viscosity was relatively slow, at a mere 0.2 mm
2·s
−1 increment over a span of 15 days. This observation indicates the relatively stable properties of the prepared sol. The gradual increase in sol viscosity could be attributed to continuous hydrolysis and polycondensation reactions occurring within the sol, leading to an incremental rise in the degree of polymerization [
20]. Moreover,
Figure 1 demonstrates that the viscosity experienced rapid growth after 6 days of sol aging. As a result, film coating was typically performed during the 5–7-day aging period, when the sol’s viscosity reached a moderate level, ensuring favorable surface characteristics and optical properties of the prepared films.
It is evident in
Figure 2 that the transmittance of the films gradually decreased with longer annealing times. Interestingly, even after a 15-day annealing period, the film maintained a relatively higher transmittance overall, albeit with a further decrease. Notably, the films exhibited enhanced transmittance in the visible light range following different annealing times, which can be attributed to the reduced absorption and improved thermal stability facilitated by CSPE. Additionally, all films demonstrated a significant drop in transmittance below a wavelength of 370 nm, primarily due to the band gap absorption inherent in HfO
2 materials [
21].
Figure 3 reveals that even after a 15-day annealing period, the film retained an amorphous structure without any discernible diffraction peaks. This could be attributed to the high phase-transition temperature (>623 K) of HfO
2 [
22]. At an annealing temperature of 423 K, which is significantly lower than the phase-transition point of HfO
2, the film remained in an amorphous state even after prolonged annealing. Typically, the crystallization of films leads to the formation of structural defects, such as grain boundaries, which could diminish their LIDT [
23,
24,
25]. Thus, in this experiment, the film maintained its amorphous structure even after extended annealing, contributing to its enhanced resistance against laser damage.
The root mean square (RMS) roughness of the films increased gradually after annealing for 0.5 h, 2 h, and 24 h, reaching values of 0.45 nm, 0.48 nm, and 0.68 nm, respectively (
Figure 4). These relatively short annealing times resulted in a relatively flat film surface. However, when the annealing time was extended to 7 d, the film surface exhibited a convex shape, and its roughness increased to 1.33 nm. Subsequently, with a further extension of the annealing time to 15 d, the RMS roughness of the film sharply increased to 2.02 nm, accompanied by noticeable surface irregularities. This significant increase in roughness could be attributed to the evaporation of a considerable amount of organic matter from the film during the prolonged annealing process.
Figure 5 illustrates the absorption characteristics of the films after annealing for various durations, with 12 measurement points selected for each sample. The average absorption values of the films after different annealing times were 22.1 ppm, 22.6 ppm, 22.8 ppm, 23.1 ppm, and 23.3 ppm, respectively. A gradual increase in absorption was observed with the prolongation of the annealing time. This phenomenon could be attributed to two factors. Firstly, the extended annealing time led to the evaporation or carbonization of residual organic matter within the film, resulting in the formation of structural or impurity defects that contributed to increased absorption. Secondly, as shown in
Figure 4, the RMS roughness of the film increased with the annealing time, which further contributed to enhanced absorption [
26]. It is important to note that despite the increase in absorption after annealing, the absorption level remained significantly lower than that of films prepared by physical methods [
27].
It should also be noted that although an LIDT test under ambient environment after high-temperature annealing to some extent reflects the film’s resistance to laser damage at a high temperature, it is still different from tests in the actual high-temperature conditions. Correspondingly, in situ high-temperature LIDT tests were carried out on the films. The LIDT values decreased over time: 23.7 J/cm
2 after 0.5 h, 22.6 J/cm
2 after 2 h, and 17.3 J/cm
2 after 24 h (
Figure 6). The decrease was 4.6% from 0.5 h to 2 h, which is not significant considering the test error. However, after 24 h, there was a 27.0% reduction compared to 0.5 h (17.3 J/cm
2). After 7 days, the LIDT further decreased to 14.3 J/cm
2, a 39.7% reduction from 0.5 h. Interestingly, extending the heating time to 15 days showed minimal change, with the LIDT remaining at 12.9 J/cm
2, the same as after 7 days.
The damage to films under nanosecond laser irradiation was primarily due to the thermal effects caused by defects or impurities.
Figure 7a,c,e,g,i show that film damage followed a defect-induced pattern [
28]. Each damage center contained at least one defect point, where high absorption coefficients led to a rapid temperature increase and film destruction [
29]. The elevated temperature also caused organic matter within the film to continue evaporating. Comparing
Figure 7a,c,e,g,i revealed that with prolonged heating time, the annular damage intensified, blurring the textured border around the macula. Additionally, evaporation at the center of the damaged spot decreased over time (
Figure 7b,d,f,h,j). Notably, after 24 h of heating, white circular defects of approximately a 5 μm diameter appeared (
Figure 7e). As the heating time increased, the circular defects became more pronounced (
Figure 7g,i), with an expanding diameter that covered the film surface. These defects resulted from the prolonged high-temperature evaporation of organic matter within the sol–gel film and non-uniform film shrinkage. Moreover, the white defects transformed into small damage-induced centers with visible cracking after laser irradiation (
Figure 7i,j).
4. Discussion
This study demonstrated the effects of long-term high-temperature exposure on the optical properties, surface morphology, absorption, and LIDT of sol–gel HfO2 films. We discuss the mechanisms of these findings from several aspects, as follows:
(1) Increasing the duration of high-temperature (or annealing) exposure resulted in a decrease in the optical properties of the film, along with an increase in absorption and surface roughness. Although the film retained its amorphous structure, the high temperature caused the evaporation of residual organic matter in the sol–gel film. This evaporation led to the formation of additional tiny defects within the film and on its surface.
Figure 7e,g,i visually demonstrate the gradual growth and expansion of these surface micro-defects. Apart from adversely affecting the film’s optical properties, these defects also lowered the film’s resistance to laser damage to some extent.
(2) In this experiment, the film’s LIDT was tested at a high temperature instead of room temperature, providing a more accurate evaluation of its laser damage resistance under high-temperature conditions. This approach realistically reflected the film’s performance in such environments. This study found that the film’s LIDT gradually decreased with increasing heating time at 423 K. This decrease could be attributed to three factors: increased defects within and on the film’s surface due to long-term high-temperature exposure, enhanced film absorption, and the possible degradation or destruction of the film’s three-dimensional network structure [
15,
19].
(3) The decline rate of the film’s LIDT gradually slowed down with increasing heating time. Further fitting the LIDT values in
Figure 6 yields
Figure 8, revealing that the LIDT (
y) of the film follows an exponential decay relationship with the heating time (
t):
where
A is related to the initial LIDT of the film. A higher value of
A corresponds to a higher initial LIDT of the film, indicating better resistance to laser damage at room temperature.
B is associated with the range of variation in the LIDT curve. A smaller value of
B indicates a slower decline in LIDT, indicating that the film exhibits good durability at high temperatures and maintains excellent resistance to laser damage even after prolonged exposure to elevated temperatures. An intriguing question arises: Why does the LIDT in this experiment exhibit exponential decay over time? To our knowledge, there is no specific theory or model describing laser damage to films at high temperatures. An existing model for nanosecond laser damage is based on the heating diffusion equation, which can calculate the temperature rise process in the laser damage area, but it is not applicable to this study [
30]. Fortunately, previous research has shown that the evaporation of the liquid component within the composite follows a power function decrease with time under high-temperature conditions [
31]. In this experiment, the evaporation of residual organic matter at a high temperature led to the formation of defects within and on the film’s surface. As the evaporation time increased, the number of such defects exhibited a power function growth. Given that the LIDT under nanosecond laser irradiation was inversely proportional to the defect density [
32], the LIDT demonstrates an exponential decay relationship with time. Furthermore, these findings suggested that, compared to the increase in film absorption and the degradation of the film’s three-dimensional network structure caused by high temperatures, the accumulation of defects inside and on the film’s surface played a more significant role in reducing the LIDT in this experiment. These defects might also enhance the volume scattering and surface scattering irregularities and reduce the film transmission to some extent. Additionally, this study established a general law and formula for understanding the laser damage of sol–gel films at high temperatures over time. These results provided crucial reference values for future investigations on the laser damage behavior of sol–gel films subjected to high-temperature conditions.
5. Conclusions
This paper presented a novel method for synthesizing high LIDT sol–gel HfO2 films using HfCl4 as a precursor and CSPE as an additive. The study comprehensively investigated the optical properties, phase structure, surface morphology, absorption characteristics, and in situ laser-induced damage evolution mechanisms of these films at elevated temperatures. Remarkably, the experimental film achieved an outstanding LIDT of 23.7 J/cm2 at 423 K. Even under prolonged heating at 423 K, the film maintained desirable features such as high transmittance, low absorption, and a superior LIDT. However, at this elevated temperature, the LIDT of the films exhibited a declining trend as the heating time increased. Notably, the LIDT of the films follows an exponential decay pattern with an extension in the heating time. This intriguing observation could be attributed to the power function-driven accumulation of defects within and on the film’s surface as a result of prolonged exposure to high temperatures. Additionally, the film retained a significant LIDT of 12.9 J/cm2, even after a lengthy heating time of 15 days. These findings underscored the remarkable stability of the sol–gel film against laser damage and highlighted its promising potential for long-term applications in high-temperature environments.