1. Introduction
HfO2 coatings have excellent photoelectric properties and high dielectric constant [1]-[6]. It is an ideal material to replace semiconductor gate material SiO2 [7]. At the same time, because of low absorption coefficient and high transmittance from ultraviolet to infrared, HfO2 layers are commonly used optical coatings [8] [9]. High refractive index, wide bandgap and high laser damage threshold, HfO2 layers are usually combined with SiO2 layers to prepare laser anti-reflection mirrors and high-reflection mirrors [10] [11]. The laser damage problem of HfO2 coatings has always been a research hotspot [12] [13] [14] [15] [16].
Commonly used methods for preparing HfO2 laser thin films are: electron beam evaporation, magnetron sputtering and ion beam sputtering [17] [18] [19]. Generally, nodule defects are produced in coatings evaporated by electron beams [20] [21]. Current research shows that nodule defects are the main reason for the laser damage of electron beam evaporated coatings [22] [23]. Ion beam sputtering and magnetron sputtering methods are relatively denser than electron beam evaporation, and have fewer nodule defects, and the cause of laser damage is usually attributed to the nano-level absorbent precursor [18] [24] [25] [26] [27]. Research in the past two years has shown that more argon components will accumulate in coatings deposited by ion beams, and argon may have an impact on the laser damage threshold of the coatings [18] [28]. But there is no detailed study on the process parameters, argon content and damage threshold. However, the relationship between process parameters, argon content and laser damage threshold has not been studied in detail.
This paper is the first to study the influence of the argon content change of the ion beam sputtering HfO2 coatings and the laser damage threshold under different process parameters. Studies have shown that changes in argon gas flux have the greatest impact on the damage threshold. Oxygen gas flux and beam voltage (screen grid voltage) have less impact on the damage threshold of the film. The relative change in argon content has a greater impact than the change in absolute content. The damage threshold is related to the O/Hf ratio. At the same time, the changes of spectrum and surface morphology under different process parameters are studied. This study provides a reference for the preparation of high-threshold sputtering coatings.
2. Experiments
2.1. Sample Preparation
The HfO2 single layer is prepared by ion beam sputtering deposition (IBS) method. The coating equipment is VEECO dual ion beam equipment, including a 16 cm radio frequency (RF) ion source for sputtering the target to deposit on the substrate, and a 12 cm ion source for auxiliary deposition to make the coatings denser and more closely integrated with the substrate. The substrate is fused silica substrate (ultraviolet grade) with a diameter of 50 mm and a thickness of 5 mm. The vacuum degree is 2 × 10−6 Torr. The target material is Hf metal, and the purity is higher than 99.9%. Argon is used as the working gas of the ion source, and oxygen is the reaction gas. The thickness of the six groups of different HfO2 single layers is 800 nm, the deposition speed is 0.04 nm/s, the ion beam current is 600 mA, and the substrate temperature is 80 centigrade. The changed process parameters are argon flux, oxygen flux, beam voltage (screen grid voltage), as shown in Table 1.
2.2. Laser Damage Experiments
Perform 1-on-1 laser damage test according to ISO-21254 standard [29]. A Nd:YAG
![]()
Table 1. Six different HfO2 single layers process parameters.
laser with a wavelength of 1064 nm and a pulse width of 12 ns is used. The incident angle is 0˚. The area of the laser energy 1/e2 range on the sample is 0.068 mm2. The laser energy fluctuation range in the test is less than 5%. The laser energy is continuously adjustable by the combination of half-wave plate and polarizer. There are 20 sampling points for each energy step. The real-time display of 5 times optical zoom CCD detects the laser irradiation area.
3. Results and Discussion
The phases of six groups of HfO2 single layers are measured by X-ray Diffraction (XRD, Panalytical Empyrean). The X-ray source is Cu-Kα (λ = 0.15418 nm), and the 2θ range is 10˚ to 90˚. The result is the broad scattering bugle. Therefore, the HfO2 coatings are all amorphous phases. The XRD diffraction spectrum of typical HfO2 single layers is shown in Figure 1.
A spectrophotometer (Lambda 1050, Perkin Elmer) is used to test the transmittance spectra of different groups of HfO2 single layers. The spectra of the two samples with varying argon flux (8 sccm and 28 sccm) are shown in Figure 2. It can be seen that the overall transmittance of 28 sccm sample is slightly higher than 8 sccm sample. The surface morphologies of different single layers are tested by atomic force microscope (AFM, Dimension 3100, Veeco). The surface morphologies of the two samples with varying argon gas flux are shown in Figure 3. It can be seen that both samples show dense void-like structures, and 28 sccm sample has more voids than 8 sccm sample. Therefore, the reason why the transmittance of 28 sccm sample is higher than 8 sccm sample may be that the cavity structure of 28 sccm sample has better anti-reflection effect than that of 8 sccm sample.
The transmittance spectra and morphologies of different oxygen flux (15 sccm and 40 sccm) are shown in Figure 4 and Figure 5, respectively. It can be seen that the peak transmittance of the 15 sccm sample is higher than 40 sccm, and the shortwave absorption of the 15 sccm sample is lower than the 40 sccm sample. It can be seen from the surface morphology that both samples show a porous structure, with more pores of 40 sccm sample. The reason for the difference in the transmittance of the two film samples may be related to the difference in surface structure and the absorption coefficient of the film.
The beam voltage (screen grid voltage) determines the energy of the discharge ions. Changing the beam voltage will change the deposition density, stress and other parameters of the coatings. The transmittance spectra and surface morphology of different beam voltages (950 V and 1250 V) are shown in Figure 6
![]()
Figure 1. Typical XRD diffraction spectrum of HfO2 single layers.
![]()
Figure 2. Transmittance spectra of HfO2 single layers under different argon gas flux: 8 sccm (black line) and 28 sccm (red line).
![]()
Figure 3. Surface morphologies of HfO2 single layers under different argon gas flux: 8 (sccm) Ar (a) and 28 (sccm) Ar (b).
and Figure 7, respectively. It can be seen that in the medium and long wave region, the peak transmittance is similar, but in the short-wave region, the 1250 V sample has a higher transmittance than the 950 V sample. So, 950 V single-layer sample has higher absorption. Both samples show porous surface morphologies.
![]()
Figure 4. Transmittance spectra of HfO2 single layers under different oxygen gas flux: 15 sccm (black line) and 40 sccm (red line).
![]()
Figure 5. Surface morphologies of HfO2 single layers under different oxygen gas flux: 15 sccm (a) and 40 sccm (b).
![]()
Figure 6. Transmittance spectra of HfO2 single layers under different beam voltages (screen grid voltages): 950 V (black line) and 1250 V (red line).
The size and number of pores in the 1250 V sample are higher than the 950 V sample, which is due to the high-energy sputtering deposition process with high beam voltage.
![]()
Figure 7. Surface morphologies of HfO2 single layers under different beam voltages: 950 V (a) and 1250 V (b).
X-ray photoelectron spectroscopy (XPS; Thermo Scientific) is used to characterize the composition of different HfO2 single layers. The diameter of the test sampling range is 400 μm. The test depth is about 1 nm. The content of argon (Ar (%)) in the layers and the ratio of the content of oxygen to hafnium (O/Hf ratio) are shown in Table 2. The argon content in the sample is higher than 4%, and the argon content is much higher than the solubility of argon in solid materials, which is about 1% [30]. It can be seen that the sample with 28 sccm argon flux contains more argon than the sample with 8 sccm, and the sample with 28 sccm argon flux has an O/Hf ratio lower than 8 sccm. This is due to the difference in argon flux. The argon content in samples with more argon flux is higher, and the hafnium metal is not fully oxidized, resulting in lower O/Hf ratio. The argon content in the sample with 40 sccm oxygen flux is slightly lower than the sample with 15 sccm oxygen flux, and the O/Hf ratio is slightly higher than the sample with 15 sccm oxygen flux. This is because the hafnium metal in the sample with high oxygen flux is fully oxidized, resulting in a high O/Hf ratio, and at the same time, an increase in the proportion of oxygen leads to a lower argon content. In the sample with a beam voltage of 1250 V, the argon content is higher than the sample with 950 V, and the O/Hf is lower than the sample with 950 V. During the sputtering deposition of the coatings, ions will be deposited on the surface of the substrate with high energy, which will cause the working gas argon to be wrapped in the coatings. Therefore, a higher beam voltage will cause an increase in the argon component of the coatings, and at the same time, the deposition of inclusions will become more and more dense, resulting in insufficient reaction, and therefore the non-stoichiometric defects will increase.
The laser-induced damage thresholds (LIDT) of different samples are shown in Figure 8 (1064 nm, 12 ns, 1-on-1). It can be seen that the HfO2 single layers with 28 sccm argon flux has the lowest damage threshold, followed by the sample with a beam voltage of 1250 V. The thresholds of other samples are not much
![]()
Figure 8. Laser damage thresholds of different HfO2 single layers (1064 nm, 12 ns, 1-on-1).
![]()
Table 2. Surface roughness, argon content and O/Hf ratio of different HfO2 single layers.
different, and the sample with 8 sccm argon flux has the highest damage threshold. It can be seen that the laser damage threshold is related to the argon inclusions and non-stoichiometric defects. Under different process parameters, higher O/Hf ratio and lower argon content, HfO2 single layers have higher LIDTs. Among them, the argon flux and beam voltage are important process parameters that affect the O/Hf ratio and the argon content in the coating.
4. Conclusion
The main process parameters (argon flux, oxygen flux and beam voltage) of the ion beam sputtering preparation of HfO2 single layers, the transmittance spectrum, surface morphology and laser damage characteristics are studied. The prepared HfO2 single layers are all amorphous structures. The surface of the HfO2 single layers all showed a porous morphology. Due to the absorption of the coatings with different process parameters and the difference in surface morphology, the transmittance spectrum of the HfO2 single layers is different. In sputtering deposition, the high ion energy will wrap the argon into the coatings, resulting in a higher content of argon in the coatings than coatings prepared by other methods. The laser damage thresholds of the HfO2 single layers are related to the argon inclusions and non-stoichiometric defects. The changes in argon flux and beam voltage parameters have the greatest impact on the O/Hf ratio and argon content. When the O/Hf ratio is high and the argon content is low, the laser damage threshold of the HfO2 single layers is high. A more detailed study will be conducted on specific process parameters later. This thesis is of great significance for the preparation of sputtered coatings with high laser damage threshold.
Acknowledgements
This work is supported by the Youth Innovation Fund of the Ji Hua Laboratory, China (Grant No. X201191XA200), the GuangDong Basic and Applied Basic Research Foundation, China (Grant No. 2020A1515110714), and the Key Area Research and Development Program of Guangdong province, China (Grant No. 2019B010144001).