Ryozo Kato

Thermal diffusivity measurement of thin films by means of an ac calorimetric method

A new method to measure thermal diffusivity of a thin sample was developed using a light‐irradiated ac calorimetric technique. The experimental conditions and the fundamental equations for the measurement are discussed. In principle, this method can be applied no matter how thin a sample may be. This method was tested for samples of nickel, silicon, stainless steel, and alumina in the range from 50 to 300 μm in thickness. The measured thermal diffusivities coincide satisfactorily with the values reported for bulk materials. It is concluded that this method is useful in the measurement for thin materials with a variety of thermal diffusivities, and the sample mounting is easy in comparison with other methods.

Thermal conductivity measurement of submicron-thick films deposited on substrates by modified ac calorimetry (laser-heating Ångstrom method)

Technological development, especially in microelectronics, necessitates the development of new and improved methods for measuring the thermal properties of materials, especially in the form of ultrathin films. Previously, modified ac calorimetry (laser-heating Ångstrom method) using a scanning laser as the energy source was developed and shown to provide accurate values of thermal diffusivity and derived thermal conductivity for a broad range of materials in the form of free-standing thin sheets or films, wires including fiber bundles, and some films on substrates. This paper describes further applications of the modified ac-calorimetry technique for measurements of the thermal conductivity of thin films deposited on substrates. It was used to measure successfully the thermal conductivities of 1000- to 3000-Å-thick aluminum nitride films, aluminum oxide films, etc., which were deposited on a glass substrate. It was also shown to be suitable for developmental measurements on submicron-thick chromatic films deposited on a PET substrate, which are photothermal recording layers, used in the media of CD-R drives of computer systems.

Thermal conductivity of ZnO thin film produced by reactive sputtering

ZnO thin films have been produced by reactive sputtering with different oxygen contents in the sputtering gas. As a result of transmission electronic microscopy observation, each film consist of two layers: an interfacial layer close to the substrate, with a thickness of about 200 nm, composed of very fine crystal grains and an upper layer above the interfacial layer, composed of column-shaped grains aligned along the out-plane direction. The grain diameter ranges from 35 to 100 nm depending on the oxygen partial pressure. The in-plane and out-plane thermal conductivity have been measured at room temperature. The out-plane thermal conductivity of the interfacial layer is 2.3 W m−1 K−1, independent on the oxygen partial pressure. The out-plane thermal conductivity of the upper layer is 5.4, 7.1, and 4.0 W m−1 K−1, and the in-plane thermal conductivity 4.86, 6.01, and 2.66 W m−1 K−1, for the O2 30%, 60%, and 90% ZnO film, respectively. Both out-plane and in-plane thermal conductivity decrease with the decre...

Development of ac Calorimetric Method for Thermal Diffusivity Measurement. II. : Sample Dimension Required for the Measurement

In an ac light-irradiated method for thermal diffusivity measurements of rectangular plate-like samples partly shadowed by a mask, the required dimensions are considered. The contour length around the cross-section (given by the thickness times the width) should be so long that the sample has a proper amount of heat capacity; then, the external thermal relaxation time becomes sufficiently long so as not to dissipate ac thermal energy. The thickness, however, should be thin enough so as to obtain a uniform temperature across the sample during oscillations (as discussed in detail previously). For the sample length, the distance between the thermocouple and the sample boundary on which ac light is irradiated should be sufficiently long so that the thermal waves reflected at the boundary decay. However, there is no restriction on the other bourtdary shadowed by the mask, i.e., the thermocouple can be placed at the boundary.

Effect of microstructure on Au/sapphire interfacial thermal resistance

We deposit Au films on single crystal sapphire substrates by sputtering and evaporation methods. The microstructure characteristics such as crystal textures, grain sizes, and fraction of contacted area of the films are examined by x-ray diffraction and transmission electron microscopy. The sputtered films have an average grain size of about 200 nm and perfectly attach to the substrates; the as-evaporated films partially attach to the substrate; the grain size varies from 10 to 30 nm, and after annealing, increases to 50 nm. Au2Al phase is observed in the annealed samples. The interfacial thermal resistance is measured by a frequency domain thermoreflectance method. The thermal resistance of the sputtered Au/sapphire interfaces is 35.5×10−9 m2 K W−1, and those of the evaporated samples are up to three times as large as this value. The change in interfacial thermal resistance is explained by the effect of detachment using a parallel arranged thermal resistance model, the effect of grain size, and the influe...

Thermal Diffusivity by Modified ac Calorimetry Using a Modulated Laser Beam Energy Source

Modified ac calorimetry, a variation of the Angstrom method, has been shown to be a precise tool for measuring the in-plane thermal diffusivity of thin films (thickness less than 300 μm) of a wide variety of materials and layered composites. The property is determined from an analysis of the decay curve of the ac temperature waves generated by irradiation of a specimen using uniform chopped light (at frequencies from 1 to 20 Hz) from a halogen lamp source. To address certain limiting factors, especially to improve the signal-to-noise ratio and to eliminate heat losses, an improved form of measurement instrument has been developed. It is based on the use of a modulated laser beam heating to provide a higher intensity energy source plus a special optical system to ensure that one-dimensional ac temperature wave propagation is obtained. Measurements can now be made using frequencies in the range of 0.01 to 10 Hz, i.e., 10 times lower than in the traditional method. The performance of the improved measurement instrument will be illustrated by results on various materials of known thermal properties such as nickel and stainless steel, proposed reference materials such as a glassy carbon and alumina, plus a comparison of results obtained on CVD diamond films used in an international round-robin series with those obtained by the traditional technique.

Phonons with long mean free paths in a-Si and a-Ge

We investigated phonons with long mean free paths (MFPs) in amorphous Si (a-Si) and amorphous Ge (a-Ge). The thermal conductivity of a-Si and a-Ge thin films prepared by magnetron sputtering was found to depend on film thickness and deposition temperature. From the film thickness dependence, we conclude that phonons with MFPs longer than 100 nm contribute to heat transport in a-Si and a-Ge. Also, as deposition temperature was increased, phonons with MFPs ranging from 100 to 250 nm in a-Si and from 15 to 250 nm in a-Ge increased.

Development of ac Calorimetric Method for Thermal Diffusivity Measurement I. Contribution of Thermocouple Attachment in a Thin Sample

In a very thin sample, the contribution of addenda, thermocouple, glue etc., to the precise measurement is considered especially in two typical cases, in which the thermocouple is attached to the sample surface using glue. In the one case, glue is painted over the whole one-sided region starting from the thermocouple on the surface and in the other, a small amount of glue is used to fix the thermocouple at a particular point on the surface. Consistent with the theoretical consideration, it is experimentally verified in both cases that the addenda do not affect determination of the thermal diffusivity.

Thermal conductivity of sputtered amorphous Ge films

We measured the thermal conductivity of amorphous Ge films prepared by magnetron sputtering. The thermal conductivity was significantly higher than the value predicted by the minimum thermal conductivity model and increased with deposition temperature. We found that variations in sound velocity and Ge film density were not the main factors in the high thermal conductivity. Fast Fourier transform patterns of transmission electron micrographs revealed that short-range order in the Ge films was responsible for their high thermal conductivity. The results provide experimental evidences to understand the underlying nature of the variation of phonon mean free path in amorphous solids.

Measurements of the Seebeck coefficient of thermoelectric materials by an ac method

An ac method for measurement of the Seebeck coefficient was developed. Specimens were heated periodically at frequencies in the range 0.2 10 Hz using a semiconductor laser. The small temperature increase and the resultant thermoelectric power were measured with a Pt Pt 13% Rh thermocouple (25 μm in diameter) through a lock-in amplifier. The Seebeck coefficient of a Pt90Rh10 foil measured by the ac method was in agreement with that obtained from the standard table. The optimum frequency and specimen thickness for the ac method were 0.2 Hz and 0.1 0.2 mm, respectively. The Seebeck coefficients of silicon single crystal and several thermoelectric semiconductors (Si80Ge20, PbTe, FeSi2, SiB14) measured by the ac method agreed with those measured by a conventional de method in the temperature range between room temperature and 1200 K. The time needed for each measurement was less than a few tens of minutes, significantly shorter than that for a conventional de method.