JPS6351500B2 - - Google Patents

Description

Detailed Description of the Invention: One surface of a plate-shaped sample is instantaneously irradiated with radiation such as a laser beam, and the temperature rise on the other surface is observed, and the temperature rises by, for example, half of the maximum value. The thermal diffusivity can be measured based on the time taken to reach 1 and the thickness of the sample. In this type of flash method thermal diffusivity measurement, the radiation irradiation time must be increased in order to prevent the temperature of the sample surface from rising abnormally and to generate an appropriate temperature rise on the back surface. However, in this case, the waveform and the temperature on the back side of the sample are 2
It is necessary to correct the measured value depending on the time it takes to reach 1/200%. Therefore, this correction amount not only differs from device to device, but also from measurement to measurement, making the correction operation extremely complicated. The present invention eliminates these drawbacks and provides a method that does not require correction for each measurement, and will be described in detail below.

As shown in Figure 1, for example, one surface of a disk-shaped sample 1 is instantaneously irradiated with laser light with uniform intensity as indicated by arrow 2, and a thermocouple attached to the other surface. The temperature rise is measured by contact 3. FIG. 2 shows curves a and θ showing the relationship between time t, the intensity L of the laser beam, and the temperature T of the lower surface of the sample 1, and FIG. 3 shows the curve a enlarged. When the maximum value of the curve θ in FIG. 2 is p, the time when the curve reaches p/2 is t _1/2 ,
Further, it is known that when the thickness of the sample 1 is l and the constant is K ₀ , the thermal diffusivity α of the sample is expressed as α=K ₀ l ² /t _1/2 (1). Figure 4 is a constant
This shows K ₀ , where τ is the time width of the radiation that irradiates the sample, and t _c is l ² /α, and τ/t _c is plotted on the horizontal axis. Horizontal line A corresponds to the case where the time width τ of curve a in Fig. 2 is extremely small, and the above
K ₀ has a constant value of 0.1388 regardless of t _c . Curves B, C, D, and E are all obtained when the time width τ is relatively large, and the waveforms of radiation a are assumed to be triangular waves b, c, rectangular waves d, and triangular waves e, respectively, in Fig. 3. . In this way, the constant K ₀ changes almost linearly, regardless of the radiation waveform, and τ/t _c
When is 0, it has a value of 0.1388. Therefore, if the slope of the above straight line is q, then K ₀ = t _1/2 / t _c = 0.1388 + q・τ/t _c ...(2) holds, and if qτ is t _g , then from equation (1), α= 0.1388l ² /t _1/2 −t _g ...(3) is obtained.

Also, the waveform a of the radiation irradiating the sample is f(t')
Then, the temperature rise curve θ on the back side of the sample shown in Figure 2 is It is known that it is given by The origin of time t in this equation (4) is the rising point of the radiation waveform a, but if that origin is delayed by the above-mentioned time t _g , equation (4) becomes It is expressed as By Taylor expansion of this equation (5), t _g =∫〓／ _p t′(t′)dt′／∫〓／ _p f(t′)d
If we set t′...(6), we can see that the first-order term disappears. In other words, t _g given by equation (6) above is the waveform a
This corresponds to the position of the center of gravity. If we observe the time when the value of the curve θ becomes 1/2 of its maximum value p with such a position as the origin of the time axis, as is clear from equation (3) above, a constant value is obtained regardless of the shape of the waveform a.
We can always set K ₀ as 0.1388. Although some errors occur due to the quadratic and subsequent terms in equation (5),
In the range where t _1/2 is more than twice τ, the error is 1
Calculations have shown that it is ~2% or less.

As explained above, the present invention delays the origin of the time axis for observing the temperature rise curve on the back surface of the sample by the time t _g given by equation (6) from the rising point of the radiation irradiating the surface of the sample. This eliminates the need to make corrections for each measurement, and the constant K ₀ in equation (1) can always be a constant value, making measurement extremely easy. In addition, it is no longer necessary to significantly reduce the time width of the radiation that irradiates the sample, so by making the sample thicker and applying a sufficiently large amount of energy to the front surface, the temperature rise on the back surface is increased and precise measurements can be performed. Can be done. In addition, the temperature history curve on the back side of the sample may be used to correct heat loss due to radiation from the sample or the distribution of incident energy, but in such cases, the origin of the time axis should also be set as described above. This allows for precise correction.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing the configuration of an apparatus for implementing the method of the present invention, Fig. 2 is an example of the intensity curve of the radiation irradiating the sample and the temperature curve on the back side of the sample, and Fig. 3 is an example of the radiation intensity curve irradiating the sample. Line intensity curve, Figure 4 is a constant
It is a diagram showing K ₀ . In the figure, 1 is a sample, 2 is an arrow indicating a laser beam irradiating the surface of the sample, and 3 is a thermocouple contact.

Claims (1)

[Claims] 1 Irradiate one side of a plate-shaped sample with radiation whose intensity is given by a function f(t') of time t' and whose time width is τ,
From the start of irradiation ∫〓 ₀ t′f(t′)dt′／∫〓 ₀ f
(t′)
The thermal diffusivity is determined by the time it takes for the temperature on the other side of the sample to reach a constant ratio to its maximum value and the thickness of the sample, with the time given by dt' as the reference point. A thermal diffusivity measurement method characterized by determining .