In the micro-electro-mechanical system, inertial confinement fusion, high power laser, and other fields, the problem of thermal failure or even damage of thin film components has become increasingly prominent, which has raised the concern on the thermal characteristics of thin-film devices. Therefore, it is necessary to study the generation and diffusion of heat in these devices. In addition, if necesssary, the forced heat dissipation is considered. Thermal diffusivity is a very important thermal parameter. The mechanism of laser damage, especially under long pulse or continuous laser irradiation, is thermal melting or thermal-mechanical coupling. Under certain heat sources, thermal diffusivity determines the temperature field distribution, thus determining the damage form and threshold. Therefore, the laser damage threshold of thin-film elements can be improved by improving thermal diffusivity. However, compared with bulk materials, the thermophysical properties of thin films are more specific. Different microstructures, impurities, and defects of thin films lead to great differences in thermal diffusivity. At present, there are not many high-precision measurement methods for micro-nano thin-film materials, and the existing methods have some disadvantages. Therefore, it is necessary to study a thermal diffusivity test method with high accuracy that is suitable for micro-nano thin-film materials, without letting the measurement results getting easily affected by the environment.
In this paper, a thermal diffusivity measurement method based on surface thermal lens (STL) technology is proposed. Pulsed pump light is used to heat the sample and a temperature field is formed by conduction along the film layer. A surface thermal bulge forms and the probe light is modulated to form an STL effect because of the thermal expansion in the temperature rise area. As the heat gradually diffuses from the pump spot to the surrounding area, the radius of the heat bulge gradually becomes large, but the time for the heat bulge reaching its peak height is different at different positions. The larger the distance from the pump spot, the more delayed the time for forming the heat bulge. Therefore, there exist phase differences among heat bulges generated at different positions along the radial direction, and this phase differences reflect the time required for heat to diffuse from the center to the surrounding area, which is affected by thermal diffusivity. The phase of the STL signal is found to be linearly related to the detection distance, and its proportional coefficient is related to the thermal diffusion length. The thermal diffusivity can be calculated using the diffusion length and the modulation frequency of the pump light. In the experiment, the phase data of the STL signal can be directly measured by the lock-in amplifier and the measurement distance data can be read on the displacement platform. By calibrating the magnification of the measurement distance and the detection distance, the detection distance can be calculated. Following this, the slope can be calculated and then thermal diffusivity can be calculated by drawing the relationship diagram between the phase and the detection distance.
The thermal diffusivity of chromium film samples with a film thickness of 150 nm is measured by the proposed method and photothermal deflection spectroscopy (PDS) (Fig.5). For the STL method, the wavelength of pump light is 1064 nm, the power before modulation is 50 mW, and the wavelength of probe light is 632.8 nm. The measurement result by STL method is 36.9 mm2/s and the measurement error is only 0.8% compared with that by PDS. To verify the applicability of this method, measurements are also conducted for carbon, high-reflection, and antireflective films. The data of the two methods are close enough to prove the effectiveness of this method.
Compared with PDS, the STL method has two major advantages. First, it is convenient to adjust and the detection light is conveniently aligned with a large pump spot. Moreover, the influence of the adjustment of the distance between the probe beam and the sample surface on the measurement results is avoided, which can save a lot of tedious device adjustment work, and errors caused by inadequate adjustment are avoided. Second, compared with PDS, which often uses air as the transmission medium, STL uses the probe beam as the medium to transmit thermal diffusion information, which can avoid signal fluctuation and measurement error caused by air flow disturbance around the sample. The device can be placed in a vacuum environment to reduce heat dissipation via air convection and improve measurement accuracy. The proposed method has advantages of relatively high measurement accuracy, capability of measuring micro-nano thin-film materials, device simplicity, and less environmental influence. Moreover, it can be further optimized to improve the measurement accuracies.
