Thermophysical Properties Research Laboratory,
Inc.
3080 Kent Ave..
West Lafayette, In. 47906
voice: (765) 463 - 1581
fax : (765) 463-5235
Flash Diffusivity Method
With the recent improvements in rapid data acquisition and
laboratory equipment, transient measurement techniques have gained
markedly in popularity. In particular one technique, called the
flash method, has been used to measure materials whole diffusivities
range from 0.001 to 10 (a range of 10 ^ 4) over a temperature
range from 80 to 2500K (-315 to 4050 F). The method uses small
easy to fabricate samples and results can be obtained within seconds.
The thermal diffusivity (alpha) is not only
important in its own right, but it also offers a convenient,
economical and accurate method of determining the thermal conductivity
(lambda). The relationship between lambda and
alpha is given by 
where
Cp is the specific heat and d is the
density. The specific heat and density are relatively
structurally insensitive, obey well-known physical laws and can
be measured readily on small samples. It is possible to either
measure these properties on the same sample used for
diffusivity measurements or duplicate samples or even to calculate
their values based on the known values of the constituent
elements. Thus it is often much easier to measure alpha, Cp, and
d and calculate lambda, than it is to measure directly. Furthermore,
the accuracies are at least comparable to and often exceed
so-called "direct" lambda measurements. (Actually thermal
conductivity can not be measured directly. Experimentation measures
heat flux, temperature gradients, and sample geometries to calculate
thermal conductivity from steady state experiments.)
The flash
method was first described in 1960 by Parker, Butler, Jenkins, and
Abbott of the U.S. Navy Radiological Defense Laboratory. In this
method the front face of a small disk-shaped sample (often about
the size of a small coin), See Sample Sizes,
is subjected to a very short burst of radiant energy. The source of
the radiant energy is usually a laser or a xenon flash lamp and
irradiation times are of one millisecond or less. The resulting
temperature rise of the rear surface of the sample is measured
and thermal diffusivity values are computed from the temperature
rise versus time data. Often times of less than one second are
involved. The ambient temperature may be controlled by a small
furnace tube or chiller. The flash method is shown schematically in
Figure 4 using a laser as the energy source. The temperature
response of the rear face of the sample is also shown in Figure
6. This rear face temperature rise is typically 1 to 2 C.
The thermal mass of the system can be made quite small so that it is
possible to quickly change temperature and record data over large
temperature intervals rapidly. Thus a large amount of data can be
generated in a short period of time. This ability along with the
small sample size required caused the method to gain rapidly in
popularity. The method has been used to measure the thermal
diffusivity of metals, alloys, ceramics, semiconductors,
composites, liquid metals, and even amoebas. At the
Thermophysical Properties Research Laboratory this method has been used
to measure the diffusivity of carbon fibers, fiber reinforced
materials, individual layers of layered composites, thermal contact
conductances at interfaces, and dispersed composites in addition to
more routine measurements.
Diffusivity values may be
calculated from halftime (t1/2 , Figure 4) using the relation
where t1/2 is the time from
the initiation of the pulse until the rear face temperature rise
reaches one-half of its maximum value (Figure 4) and 1 is the
sample's thickness. Actually one may use any percent rise:
where Kx is a constant corresponding to x percent rise and tx
is the elapsed time to x percent rise. When one has a digital data
acquisition system, it is relatively easy to calculate alpha at a
number of percent rises. Table 1 gives values of alpha at various
percent (PER) rises for an experiment involving 316 stainless steel.
The actual computer output for a typical rise is given in Table 1.
Actually only one-fifth of the data are shown. The output of the
temperature detector is given along with the elapsed time in
microseconds that the data were taken. Also initial and cooling data
are given. It is possible to compare the experimental values with the
theoretical model. This is done by dividing the temperature
rise by the maximum rise, thus non-dimensionalizing the
ordinate. Times are divided by the halftime (Figure 4) to
non-dimensionalize the abscissa. An actual experimental result is
shown in Figure 6. The smooth line is the
theoretical model and the points are the experimental results.
In this case the experimental data followed the model extremely
well and the results are accurate within 1%. The current
state-of-the-art concerning thermal diffusivity measurements using the
flash technique is discussed in detail in Chapter 8 of the book
"Compendium of Thermophysical Property Measurement Methods", edited by
K.D. Maglic, A.Cezairliyan, and V.E. Peletsky, 1984 by Plenum Press .
It is possible to satisfactorily correct for radiation heat losses and
for situations in which the time duration of the energy pulse is not
negligible compared to the transient time. It is also possible to
correct for non-uniform heating (at least in selected cases). It
has been shown that very heterogeneous dispersed composites
can be measured. Techniques for layered samples have also been
developed and this has led to the ability to measure liquids,
thick films, and contact conductance between layers. This
technique also allows us to keep temperature excursions to less than
one degree by deliberately applying a layer of material whose
properties are known. Thus heat-sensitive materials can be measured
or measurements can be made very near phase transitions.
The flash method has been extended to two-dimensional heat flow
so that large samples can be measured and the diffusivity in both
the axial and radial directions in anisotropic materials can be
obtained.
Thermal diffusivity is the measure of the way heat flows through the
material to the other side. It can also be expressed as the rate of
change of temperature in a transient heat transfer process. The higher
the Thermal Diffusivity of a substance the higher the rate of
temperature propagation. This is very important to help prevent or to
make the heat flow is either hindered or aided. Thermal Diffusivity is
denoted by the symbol alpha and can be expressed by the equation :
Thermal
Conductivity is a physical property of a substance and characterises the
ability of the substance to transfer heat. The value of Thermal
Conductivity determines the quanitity of heat passing per unit of time
per unit area at a temperature drop of 1 degree C per unit length.
Thermal Conductivity differs with each substance and may depend upon on
structure, density, humidity, pressure and temperature. Thermal
Conductivity is denoted by the symbol lambda and can be expressed by the
equation : 
where Q is heat flux, L
is length, A is cross-sectional area and delta T/ deltaX is temperature
gradient.
Specific heat values are based on the quantity of heat requires to
raise one gram of water one degree celcius. This value is ONE. Specific
heat values of other materials are a ratio of the quantity of heat
required to raise one gram one degree relative to that required for
water.
The flash diffusivity requires the material to be the following
sizes.
0.500 ±0.005 inches in diameter or 0.500 ±0.005 inches
square. The thickness of the sample should be a UNIFORM
0.15 inches for metal, 0.10 inches for dense ceramics, and 0.080 inches
for some porous materials. The devience in the uniformity is specified
as ±0.001 inches. Shown below is the measurements for the sample
in cenetemeters.
Return to
Home Page
