[Laser Technology Sharing] Research on Ultrafast Laser Damage Threshold Issues (1)

Ultrafast lasers, especially femtosecond lasers, have very short pulse widths and generally have the advantage of high peak power. This is one of the reasons why many scientific researchers choose femtosecond lasers as research tools. During use, users often find that the optical components used with femtosecond lasers, including mirrors, beam splitters, filters, etc., are very easy to break. This is of course also due to the high peak power of femtosecond lasers. In order to safely use femtosecond lasers and protect optical components, the first thing everyone considers is to confirm the damage threshold of the optical components. When we look at the damage thresholds of many optical components, we will find that manufacturers often provide the following two damage threshold parameters:

1. Continuous laser damage threshold parameters (common unit: W/cm2) Most manufacturers can provide this type of damage threshold parameters.

2. Pulse laser damage threshold parameters (common units are: J/cm2, with repetition frequency and pulse width also indicated)

Some component manufacturers can also provide pulse laser damage threshold parameters while giving continuous laser damage threshold parameters. However, the test conditions given are generally from a few nanoseconds to tens of nanoseconds, and the repetition frequency is tens of Hz. This is because many component manufacturers do not have femtosecond laser testing capabilities and therefore cannot provide effective femtosecond pulse damage threshold parameters. Generally speaking, there are three main forms of damage to optical components caused by optical radiation: one is the thermal effect caused by light absorption; the other is the dielectric ionization breakdown caused by short-pulse laser radiation; the third is the destruction of chemical bonds of substances directly caused by the extremely high peak power of ultra-short pulse lasers.

The mechanism of optical element damage caused by continuous laser obviously belongs to the first category, while the mechanism of optical element damage caused by ultrashort pulse laser should be classified into the second and third categories. It can be found that the mechanisms of injury caused by the two types of lasers are completely different. Therefore, the damage threshold parameters of continuous lasers and long pulse lasers provided by manufacturers do not have sufficient reference value when we use femtosecond lasers for experiments. What is gratifying is that some researchers dedicated to materials and coating research have conducted a series of laser destructive experiments with reference value, and the experimental data provided can be used as an important reference for everyone in device selection and experimental design. In this article, we will mainly introduce the experimental process and results of the damage threshold test brought by KYLE R. P. KAFKA’s experimental group from the Department of Physics of The Ohio State University in the United States.

1. Experimental design

The process of this test experiment is actually very simple. The experimenter focuses the laser pulse, and then places different optical components at the focus for testing. While adjusting the laser energy, the damage to the optical components is observed. Finally, the corresponding experimental data are given.

Figure 1. Ultrafast laser shock damage experimental device diagram

As shown in Figure 1, the light source used by this experimental group is a set of 3mJ, 500Hz, 35fs ultrafast lasers. The experimenter injected a 0.5mJ laser pulse into a hollow-core fiber filled with inert gas for spectrum broadening. After compression by the chirped mirror group, a periodic femtosecond laser with a pulse energy of 0.35mJ was obtained.

The experimental group conducted experiments in both vacuum and air. Since the laser pulse used in the experiment was very short, the experimenter pre-compensated the dispersion of the vacuum window before focusing during the experiment in vacuum. In the end, the pulse width at the sample was maintained at about 5-6fs. The element used for focusing is a curved silver mirror with a focal length of 150mm, and the beam waist radius at the focus is 15 microns. The damage at the focal spot is sampled by a continuous He-Ne laser and introduced into the detection equipment for observation.

2. Experimental samples

Here is a list of the samples they tested:

Figure 2. Test sample list

The first five products are all from the Spectra-Physics FemtoOptics product line, including broad-spectrum chirped mirrors, broad-spectrum reflective silver mirrors, and broad-spectrum beam splitters of different proportions.

The final fused silica substrate is from MTI Corp.

3. Test methods

The methods used in the test process are single-shot test F1 and 1000-shot test F1000. Under each energy condition, the experimental group independently selected 10-20 points on the component and used the test laser to impact a single or a thousand times. The damage probability was obtained by dividing the number of damage caused by the total number of experiments. The picture below shows two examples from the test. The measurement results of some components are as shown in the picture on the right. The damage possibility P only differs between 0 and 1, and the dividing line is easier to judge. The test results of some other components are as shown in the picture on the left. The damage possibility P is a gradual curve as the energy density increases. Therefore, the experimental group uniformly selected the position where damage begins to appear as the dividing line. The dividing line between the two situations is used as the damage threshold result of the test component under this condition.

Figure 3. Test example

Because the fused quartz sheet itself is not coated, the evaluation standards are different from other coated components. The damage is judged based on the depth of the pits caused by the laser on the surface of the fused quartz sheet. The experimental team used a variety of instruments to judge the damage on the surface of the test components for different test components, including:

1. Optical microscope

2. Interferometric surface profiler

3. Atomic force microscope (AFM)

4. Scanning electron microscope（SEM）

4. Damage threshold results (I know everyone wants to see this part the most)

Figure 4. Experimental results

Note: The damage threshold of fused quartz here is the result divided by 10 times, mainly to put it together with the results of other components.

From the chart of the test results, we can see that the damage threshold F1000 of 1000 rounds is significantly lower than the damage threshold F1 of a single round, and the difference for different test components is about 20%-50%. For fused quartz sheets, F1000/F1 is around 0.57 regardless of vacuum or air conditions. Under the same test conditions, generally speaking, the higher the reflectivity of the device, the lower the damage threshold.

However, the performance of the chirped mirror and the silver mirror here is relatively abnormal. This may be due to the small difference in their reflectivity (both >98%), but the large difference in coating design. For example, the refractive index of the first coating on the outer surface of a DE-Ag I type silver mirror is higher than that of a DE-Ag II type silver mirror, and the corresponding damage threshold of materials with a high refractive index will be lower.

This is because the cause of device damage under this test condition is dielectric ionization. Whether the test laser is a periodic pulse is also a key influencing factor. For non-periodic femtosecond lasers, the damage threshold of optical devices is increased by 1/2 to 1/3 (depending on different devices). In this issue, we have briefly introduced the experimental process and numerical results of KYLE R.

P. KAFKA and others' research on the damage threshold of optical components under periodic laser pulse impact. In the future, we will continue to bring their image analysis of the damage of optical components. After all, AFM and SEM are used, and the result analysis is quite interesting.

PS: Follow the backend reply of this public account: "Research on Ultrafast Laser Damage Threshold Issues (1) Literature" to obtain the reference information of this article.