WO1987000632A1 - Method and apparatus for the characterization and control of substances, materials and objects - Google Patents

Abstract

Method and apparatus for the characterization and/or control of substances, materials and objects, based upon means (2) for thermal excitation on the object (1) and a detection system (3 - 6) adapted to measure the resulting change in thermal radiation from the object. There is included means (11, 12) for providing a relative mouvement between the thermal excitation means (2), respectively the detection system (3 - 6) and the object (1). The excitation means (2) is adapted to deliver a continuous thermal excitation which can be heating or cooling by mechanical contact, with gas or a flame, ultrasound or an electron beam or possibly another particle beam or electromagnetic radiation.

Description

Method and apparatus for the characterization and control of substances, materials and objects.

This invention relates to a method and an apparatus for the characterization and control of substances and materials as well as factors of physical and chemical nature being associated therewith. By excitation of transient thermal waves in an object and measurement of the resulting changes in ther¬ mal radiation from the object, the invention, thus, in its widest aspect will make possible a completely contact free characterization_. and control as mentioned, with significant improvements in relation to comparable existing techniques.

All bodies and objects having a certain temperature emit thermal, electromagnetic radiation. For ideal blackbodies the emitted power per unit area within a wavelength interval dλ at wavelength λ is given by Planck's law of radiation:

J_ __ W( \)d\ = - E - (_β ^kTλ - i) dλ , ⁽1⁾ λ^s

in which T is the temperature of the body, h is Planck's con¬ stant, k is Boltzmann's constant and c is the speed of light; W(λ) is termed the spectral radiantexcitationof the body. For bodies at room temperature (T = 300 K) equation (1) gives an emission spectrum having a maximum at approximately 10 μm wavelength in the middle infrared spectral range. If the tem¬ perature is increased, the spectral distribution will change according to (1) , and the maximum point of the spectrum will be displaced towards shorter wavelengths; for T > 4000

* λ is close to or within the visible range. This displace- max

_* ment is described to a good approximation by Wien's displace¬ ment law:

T -λ __ = 2897,9 K μm, (2) max which can be derived from (1) .

S BSTIT T Bodies not being ideally black may to a good approxima¬ tion be described by multiplying (λ) given by (1) by an effective emissivity ε(T,λ) 1. Power emitted per unit area from a body not being black, within the wavelength interval λi λ λ₂ is then given by λ₂ yπ_ λ₂ - fe (T, λ^,) W(λ')dλ'. (3)

For a small temperature change Δ the change in radiated power per unit area will then be

>Wλ_.λ₂; &T. (4)

Because ε(T,λ) in general is a very complex function which also -depends upon the geometry of the body, equation (4) usually may not be expressed analytically.

However, both W(λχ;λ_) and δ (λι;λ₂) may be measured by means of radiation detectors being sensitive within the spec¬ tral range concerned. This range is usually selected so as to comprise also the wavelength for maximum emission, λ , but specific advantages and effects may be obtained by selecting other spectral ranges for the detection. Possible changes measured in W(λι;λ₂) from an object then will be directly re¬ lated to inherent variations in temperature and/or emissivity of the object. This is utilized today in standard measurement techniques, and commercial equipment has been available for a long time. The method is contact free and therefore does not interfere with the measurement object. However, because it is passive, it is restricted to the measurement of those contrasts in temperature and emissivity which occur per se in the object; this setting limits as to information which may be retrievedfrom the measurement object by the method. The sensi¬ tivity is also relatively low, and typical limits to tempera¬ ture contrasts which may be observed, are found in the range

SUBSTITUTE SHEET In the recent years there has been developed (Ref. 1) an active method of measurement designated photothermal radiometry (PTR) . By means of external thermal excitation, for example by illumination, there are induced temperature changes in the object, and the magnitude, phase or time vari¬ ation of the resulting thermal radiation from the illuminated point at the object, are measured. Initially it was used for the direct measurement of the temperature increase at the illu¬ minated point under the influence of strong radiation (Ref. 2) , but the possibilities of measurement technology have now been significantly expanded. In all known variants of PTR the illumination has always been pulse shaped, mostly in the form of a continuous pulse train. For the purposes of analysis it is usually presumed that the intensity of the illumination is varied harmonically, I=_I (1 + e^{1 π} ) , wherein f is the fre¬ quency, t is time and I is the amplitude. It is then also usual to define a thermal diffusion length μ = (k/πpfC) = (κ/τrf) , wherein k is thermal conductivity, p is density, C is specific heat and < is the diffusivity of the object material. μ is that distance in the object to which the heat is able to spread during one pulse period. There is also de¬ fined an optical absorption length α(λ) , wherein α(λ) is the spectral absorption coefficient of the object material at wavelength λ. From this it is then possible to calculate the induced temperature change δT in the object for different types of objects and illumination (Ref. 3) . Consider as an example the case in which α(λ) << the object thickness; thus the object is optically opaque. There are then two possible situations given by μ > (λ) and μ < α(λ)

For μ < α(λ)~ the object is photothermally transparent. The illumination at wavelength λ penetrates deeper into the material than one diffusion length, and it is found that the temperature at the outer surface is

δT = iI₀(2τrfpC)^"1α(λ) (5)

SUBSTITUTESHEET The temperature oscillations at frequency f corresponding to the illumination are then roportional to the absorption co¬ efficient α(λ) of the object material. Measurement of the thermal radiation at the pulse frequency f according to (4) makes possible the spectral characterization of the surface of the object by varying the spectral content of the illumina¬ tion (Ref..4) , and this has resulted in a new spectroscopic method.

For μ > α(λ) the object is photothermally opaque. All incident radiation is then absorbed in a surface layer being thinner than the thermal diffusion length, and it is found that the surface temperature is

δT = _I (2πfkpC)~*. (6)

The induced temperature variations will then be independent of the absorption coefficient of the object; and measurement of radiant exitance at frequency f will give information about the parameters k,p, C and/or ε. This also applies when the changes in k, p or C take place within the object and less than one thermal diffusion length from the surface. Internal structures in the object not being visible at the surface will then give changes both in the amplitude and the phase of δW (Ref. 5) . In other words this makes it possible to "look" into the object to a certain depth μ. A particularly inte¬ resting case is when the temperature variations are measured at the back sideof the object, as a result of illumination of the front side thereof. Structures within the object having deviating values of k, p and/or C will then have influ¬ ence on the heat propagation through the object and give changes in amplitude and phase of the temperature oscillations at the back side. This makes it possible to investigate the interior of opaque objects being significantly thicker than one thermal diffusion length (Ref. 6) .

The induced temperature oscillations may be very small, the detection limit being ', iinn tthhee rraannggee ooff 1100^"" KK.. In order to

SUBSTITUTESHEET be able to measure δ with a good signal-to-noise ratio it is therefore necessary to correlate the measured thermal emission with the illumination, so that only those thermal oscillations may be identified, which are synchroneous with the pulsations of the illumination. This is performed as a matter of routine by means of electronic lock-in amplifiers which essentially filter away all signals outside a narrow bandwidth Δf around the pulse frequency f; Δf being most often smaller than 1 Hz. For obtaining representative measure¬ ments the time for each measurement point must be at least equal to the inverse of the bandwidth, so that steady-state conditions apply. The measurements will then be slow and time consuming; and by sweeping across objects typical velo¬ cities are often lower than 1 mm/s (Refs. 5, 6) . Although such sweep techniques have been demonstrated for use in scientific analysises of surfaces and internal structures in objects, the method has in principle restrictions which in practice render it insuitable for real time measurements. Thus, the method is not suited for a number of tasks, being for example in connection with the continuous characterization and control of various industrial products and processes (chemical substances and compositions, semi-conductor com¬ ponents, surface coatings, film and layer formed materials, curing processes and so on) . The same applies to the possi¬ bility of quickly forming corresponding two-dimensional (and possibly also three-dimensional) photothermal images of objects with repeated sweeps. In all such situations each measurement point should be attended to at less than one millisecond. The steady-state photothermal radiometry method therefore is more than three orders of magnitude too slow to be useful for the purposes mentioned.

The heat propagation takes place by diffusion. This is a chaotic process which in the one dimensional case is de¬ scribed by the equation

9 T(x,t) - _1_ 3T (x,t) _ _{0 (7)}

3x' at

SUBSTITUTESHEET in which x is the coordinate, t is the time and T(x,t) is the temperature of the object in point x at the instant t. The complete solution of (7) is found in each individual case depending upon start and boundary conditions. For a harmoni¬ cally varying surface temperature T(0,t) = T . _e ^1*2τrf'fc o imposed to an object having an extent or dimension of

0 < x < ∞ it is found that (Ref. 7)

-^£- % ^■ ~ i( τcft - l) 2T₀ ^{2 (K} _r ^t} i- 2 .f(t - - J -_ξ2

^T(x. ^t) -^{" e} - — J ^{e 4κζ} e dζ (8)

The last term given by integral expression is a transient thermal disturbance in the object, due to the start of the temperature oscillations on the surface. This transient dies away when t —> ∞ , and there only remains the steady- state time variable solution given by the first term. This term describes the temperature oscillations being utilized in steady-state photothermal radiometry as described above. The solution has the form of damped waves the amplitude of which is reduced by a factor e over a distance equal to the thermal diffusion length μ. The wave velocity s = dx/dt is found by prescribing a constant phase term 2πft - x/μ, which gives

s = 2πfμ = 2 ^ = 2<ιr<f) *. (9)

From this it is seen that the thermal waves are strongly dis¬ persive; waves of higher frequency propagating more quickly. At the same time the high frequency thermal waves are more strongly attenuated with x because of the factor e^"~ ^μ. The wavelength of a thermal wave of frequency f according to (9) is Λ = 2πμ.

If it shall be possible to carry out faster measurements than what can be obtained with the above utilization of the steady-state thermal waves, it is accordingly necessary to employ the transient solutions of (8) . This requires a more

SUBSTITUTESHEET explicit calculation of the heat flow in the object. Conside then an individual pulse which instantaneously heats the sur¬ face of an object to the temperature T . At the depth x the object will experience a transient temperature excursion whic rises to a maximum and then decreases towards a stationary value. The evolution of this temperature front is given to a good approximation by

wherein A is a constant depending upon K, C and the thermal excitation on the surface. By calculating this function it is found (Ref. 8) that the temperature at point x has attaine the half of its maximum value at the instant

. _ 1,37 x² . (11)

-, 2

⁵ π <

To a good approximation this also applies to heat flow in three dimensions. Thus upon heating of the surface with an instantaneous point source the transient heat wave in a point on the surface at a distance x from the point of exitation, will have reached half of its maximum excursion approximately after the time t,.

From this there may be defined a thermal time constant ή ² τ = — 7— , (12) π K

which corresponds to the time taken until the temperature in a point at a distance d from an instantaneous exitation point has reached about 1/3 of its maximum excursion. From this follows that the effective thermal diffusion velocity u(d) = d/τ for diffusion across distance d is

u(d) = 1 *- ^• <¹³>

SUBSTITUTE SHξET In other words the effective thermal diffusion velocity across a certain length is inversely proportional to the length, which implies that the diffusion time onto a point increases with the square of the distance from the exitation point. This is related to the dispersion of the steady-state thermal waves given by (9) . If putting d = Λ = 2τrμ, it is found that the effective diffusion velocity of a transient thermal wave over a distance d equal to the wavelength of a steady-state thermal wave of frequency f is u(2πμ) = πκ/2μ = s, where s is the corresponding steady-state wave velocity given by (9) .

Equation (11) is employed in a standard pulse method for determining thediffusivity < in solid substances (Ref. 9) : By measuring t, on the basis of the temperature variation at the back side of an object having a thickness I upon heating of the front side with pulses of duration << t,, it is possible to determine K from

- < = ^ 2 . (14) ^"

During the last two or three years the pulse method has been developed further to a pulse type photothermal radiometry (Ref. 10) , wherein by recording the exact curve shape of the temperature variation at the front or back side of an object upon pulse heating of the front side, it is possible to de¬ termine absorption coefficients, thermal diffusivity and the thickness of both the object itself and of possible sur¬ face layers. In the principle an individual pulse is suffi¬ cient for determining all these parameters. In that respect the pulse method is faster than the steady-state techniques discussed above, and this by a factor being approximately corresponding to the number of periods f/Δf required for each steady-state measurement at frequency f and bandwidth Δf. If it is desired, however, to use the pulse method for the characterization of for example a complete object, a material stream or the like, one would in such case have to record correspondingly detailed temperature variations or curves at all points of interest, and then extract the relevant data by an individual analysis of each individual time curve. This recording and the subsequent signal processing thereby will take so long time that also the pulse method will be too slow for real time measurements. No such use of the pulse method has been demonstrated. The reason is that the pulse method acquires too many data, most of which is without significance for the measurements to be made. This will be explained more- closely in connection with the following presentation of the invention.

The present invention takes as a basis the fact that as a rule one is not interested in the complete temperature curve or history at each individual point of the object. Most often it is sufficient to record possible variations of the tempera¬ ture from point to point at the surface of the object, thereby to survey and identify possible dissimilarities in chemical and physical conditions at and beneath the surface. Thus, in its most general aspect as stated in claim 1, the invention consists in a particular utilization of the transient thermal waves in the object, making it possible continuously and in real time to exclusively record the thermal radiation from a series of object points after a time delay corresponding to thermal diffusion over a selectable length or depth d in the object. In short the invention provides for giving the object a relative movement in relation to a source for continuous (and possibly constant) thermalexcitation, whereby the rela¬ tive velocity between object and source, v. , is higher than the effective thermal diffusion velocity u(d) for the distance concerned in the object. This combination of continuous thermal exitation and physical movement may be compared to a consecutive series of independent, instantaneous thermal sources being displaced over the object. The thermal diffu¬ sion conditions for object points at a larger distance than d along the path of movement, will be independent of each other because the source is displaced faster than the transient

SUBSTITUTE SHE thermal wave front: After a time τ = d2 I tt2K the transient wave front from each individualexcitation point will have pro¬ pagated a distance d in the object, at the same time as the excitation point has moved in relation to the source along a distance L = v, • τ > d. At this position then the object passes through the field of view of a detector of thermal radiation, in such a manner that also the relative velocity v_ between the detector field of view and the object is higher than u(d) . Thus the detector will sense the thermal radiation from each individualexcited object point independently of each other and delayed by a time τ with respect to the initial thermalexcitation at the point. Possible inhomogeneties in the object within a distance d from each individual excita¬ tion point will then as explained above, result in correspondin variations of the delayed radiated thermal power. In this way it will be possible to utilize the transient thermal wave in a selective probing of physical parameters of the object within distances d from^" each individual excitation point. Of course the method also permits of recording and control of physical and chemical structures and patterns exposed at the surface of the object, for example by spectrally selective thermal excitation. The delay time or length may then if de¬ sired be reduced to an absolute minimum in order to better distinguish the maximum temperature excursions at the surface from the thermal effects being due to those transients ther¬ mal waves which are capable of propagating farther out from the excitation points.

The invention will be described more closely below with reference to the drawings, in which:

Figure 1 schematically shows an arrangement for thermal exci¬ tation of the front side (upper side) of the object. Figure 2 schematically shows detection of thermal radiation from the front side (fully drawn lines) or from the back side (dotted lines) of the object. Figure 3 schematically shows examples of systematic and characteristic structures and patterns which may have been included in materials and objects.

SUBSTITUTE SHEET Figure 4 schematically shows thermal excitation and detec¬ tion at a distance L from each other,

Figure 5 schematically shows a space filter in connection with the detection system.

Figures 6, 7 and 8 illustrate alternative forms of thermal excitation of the object, and

Figures 9 - 12 illustrate some interesting configurations of excitation devices and detectors shown schematically in relation to a part of an elongate object seen from above.

In Figure 1 there is shown a thermal excitation from a source 2 onto an object 1 being moved at a velocity v, with respect to the thermal excitation. The excitation has been illustrated here in an idealized manner as point shaped, whereas in practice it will have a certain extension a in the direction of movement. The thermal excitation can take place by a slip contact with a body having good thermal con¬ ductivity, possibly with a "heat pipe" in communication with _. heated or cooled thermal reservoir, but it may also take place without any mechanical contact by means of cold or hot gas, electromagnetic radiation, electron or other particle radiation or also by acoustic waves. Gas and contact excita¬ tion will position the initial heating or cooling at the sur¬ face of the object. However, as described above, electromag¬ netic radiation will penetrate into the object to a typical distance of α(λ)-^ equal to the inverse of the absorption co¬ efficient for radiation at the wavelength concerned. Thus by the choice of λ the thermal excitation can to some degree be adjusted to the situation of interest, and then particu¬ larly in connection with the detection of substances being exposed to the surface.

With electromagnetic radiation at two or more different wavelengths λ adapted to be applied at a certain mutual spacing transversally to the direction of movement and with separate detectors for each wavelength, there is also a poss bility of a more detailed and comprehensive spectral analysi

SUBSTITUTE SHEET of the object.

Particle radiation penetrates in a corresponding manner into the object depending upon the energy of the particles as well as the material properties and the com¬ position of elements in the object, and this may also be utilized for adapting the excitation in the volume of the object. Acoustic waves will penetrate into and propagate in the object depending upon the modulus of elasticity E there¬ of, and this may be able to heat selectively internal struc¬ tures having an E causing the waves to be absorbed. This gives altogether an arsenal of different thermal excitation modes, from which there may be selected a thermal excitation being particularly suited for each individual use of the invention.

It is also possible to combine two or more different thermal excitation modes in order to attain specific advan¬ tages. For example the object may be heated or cooled re¬ spectively, along the same path or along two parallel paths, with separate detection for each of these. The temperature excursions will then be directed mutually opposite, which gives complementary signals in the detection system. This can be used for example for distinguishing variations in emissivity from the thermally related signals. Detection along a neutral path, i.e. without thermal excitation, will in addition give a reference signal from a thermally undisturbed object.

Figure 2 shows detection of the resulting thermal radi¬ ation both from the same side (the front side) (a) as the thermal excitation, and from the opposite side (back side) (b) . As a rule in a given measurement situation only one of the alternatives will be used, but in many cases it will be of interest to have several detectors operating in parallel, for example in the form of an array. The field of view β of the detection system has a velocity v_ in relation to the object 1. The detection system consists of a device 3 which collects the thermal radiation from the object, and in the

SUBSTITUTE SHEET figure has been symbolized as a lens. Other devices of inte¬ rest for collecting thermal radiation may be mirror arrange¬ ments, optical fibers, optical waveguides and others. The device 3 collects the thermal radiation towards a detector 4 which converts the radiation into electrical signals. Such detectors are often made of semiconductor materials and may be adapted for different spectral ranges of the thermal radi¬ ation, depending inter alia upon the temperature of the objec but there are also thermal detectors having a more flat spectral characteristic. These are usually not so fast actin and sensitive as the semiconductor detectors. The signals from the detector are amplified in an amplifier 5 and can further be subjected to a more specific analysis in a unit 6 which comprises an electronic signal processing system. As mentioned below the electric bandwidth of units 4, 5 and 6 must be so related to the velocity v- and to the dimension of the structures to be identified, that equation (27) is satis¬ fied; otherwise there are no particular requirements to these units. From the signal processing system there may then be extracted information to be used for effecting the control and monitoring functions and so on which in each case is made possible by the invention, which is trivial and not in¬ cluded in the figure. In front of the device 3 there is additionally in position (b) shown how an optical filter 7 can be located in relation to the detection system, both for further definition of the spectral range for thermal detecti and for selective discrimination against thermal and other electromagnetic radiation in other spectral ranges (for example from the thermal excitation) . As indicated in the figure the majority of factors are identical when detecting thermal radiation from the front or back side. The choice of one of the alternatives depends essentially on the shape of the object and of the purpose of the measurement. Both configurations will often give corresponding information. Detection from the back side is of interest and particularly

SUBSTITUTE SHEET advantageous when thin objects and materials are concerned, whereby there may be the question of identifying internal structures or possibly variations in thickness and other object parameters being characteristic to the whole volume of the object. The same applies to objects and materials in which the structures of interest lie at^' a depth d which makes the diffusion determined resolution at the front side ,y d) unreasonably poor. Investigation of the surface layer of objects in which the structures of interest are exposed to the surface and of internal structures in objects being so thick that thermal diffusion to the back side takes un¬ reasonably long time and results in a corresponding reduction in the resolution, on the contrary is done best by detection from the front side.

Figure 3 shows examples of objects and materials contai¬ ning systematic and characteristic structures and patterns either at the surface (3a) , within the volume (3b) or as variations (for example in thickness as shown) of parameters which affect the whole cross-section of the object (3c) . Depending upon the nature and character of the structures and patterns there may be selected a suitable form of thermal excitation, as well as detection from the front or back side determined inter alia by the thickness and other properties of the object. When detecting the transient thermal evolu¬ tions delayed by a time τ with respect to the thermal excita¬ tion, and whereby the time τ is adjusted to the dimension d of the patterns and structures of interest, in accordance with equation (12) , the device 3 in figure 2 will provide signals containing information on these structures and patterns. With a priori knowledge of possible patterns and structures in the object, this may thus be utilized for characterizing each individual object in relation to a recording of such structures and patterns, and thereby also for checking that the object contains structures having a predetermined pattern. The dimension or extent d of the structures and patterns in the object is the important parameter in this connection, and

SUBSTITUTESHEET determines specifically the limits to the respective relative velocities of movement of the object as stated in claim 1. In a corresponding way the invention can also be employed for investigating and characterizing objects having random patterns or structures at the surface, in the volume thereof or through the whole cross-section of the object. This may serve many purposes, for example to identify objects in which the variation of such parameters are outside certain limits. This may also be utilized for establishing feed-back in industrial processes, so that these may be controlled in a manner which brings the parameters of interest to be kept within relevant limiting values.

Figure 4 shows the device for thermal excitation togethe with the detection system so arranged that the distance be¬ tween the excitation and the field of view of the detection system is given by the length L. If this length is kept fixed and constant the object has the same relative velocity v = v, = v- in relation to both the thermal excitation and the detection system. Then there will be a fixed delay time δt = L/v between the thermal excitation and detection, and there will be an opportunity to particularly investigate structures at a fixed distance d from the thermal excitation, whereby d satisfies (12) with τ = δt. The analyses explanations given above will then apply, this being in many cases also a practical embodiment of the invention.

When monitoring material flows and so on it is convenien to have stationary devices for the thermal excitation and detection, whereas t-he object is moved. For image-forming em¬ bodiments with objects (for example in medicine) it may be more practical with a common movement of excitation and detec¬ tion across a stationary object.

By varying v there may be performed a sweep over a larger or smaller interval of distances d = ττ(κ^»δt) . This may also be obtained by letting L be a variable, for example by

L = L_Q + L- • F(ωt) , (15)

SUBSTITUTESHEET in which L1, < Lo and F(ωt) is a periodic function with an amplitude equal to 1 and an angular frequency ω. The time delay between the thermal excitation and the detection may then be varied between the boundaries δt_.._j__n = (L - L,)/v and^' δt_ π,tc-ix,, = (L o + L- 1J /v. From ( 12 ) one will find the correspon- ding range of diffusion determined distances d which can be investigated. The necessary condition, however, is always that both v, and v„ > u(d) . This requires that the angular frequency ω in the periodic function F(ωt) is such that the velocity ωL, associated with this movement, satisfies v - ωL, > u(d) . (16)

The conditions given in claim 1 will then apply during the complete cyclic variation of L. This embodiment may be of interest when either the current values of d are not known, in cases where the distance d varies from object to object, or between parts of the object, or when the object has several structures with different characteristic distances d . On account of the electronic signal processing a reference sig¬ nal in such connections must be derived from a unit 8 which controls the distance L, so that the signal processing system may correct for the different diffusion lengths, delay times and signal frequencies which possibly will occur. Thus it will be most practical to vary the position of the thermal exc tation, whereas the position of the field of view of the detec tion system is maintained constant, in order that the velocity of the latter in relation to the object is also constant. This makes it easier to interpret the frequency spectrum of the thermal signals.

An arrangement for effecting movement of the object 1 is shown in figure 4 in the form of roller pairs 11 and 12 which by their rotation gives the object the velocity v.

Another embodiment may for example consist of one therma excitation with two or more detectors behind each other at distances from the excitation corresponding to different diffu sion times τ. Such an arrangement will then make it possible to carry out detailed investigations of structures, for

SUBSTITUTE SHEET example at certain depths in the object.

Figure 5 shows shielding of the detector and the detec¬ tion system by means of a spatial filter 9. This can often only be in the form of a shield which prevents the direct influence on the detector from the thermal excitation of the object. For example excitation by means of electromagnetic radiation can result in scattering into the detection system, and since this radiation will be orders of magnitude stronger than the resulting thermal radiation, even little scattering may disturb and mask the actual signals. The same can happen with excitation of for example hot or cold gas, which may lea into the area between the object and the detector and influ¬ ence the detection. Irrespective of the kind of thermal ex¬ citation chosen, it is as a rule also desirable always to shield the detector against sensing of the immediate maximum temperature excursion at the excited area of the object. Usually it is desired to sense the temperature excursion only after a certain time delay. Particularly in cases where this time delay is small, it is therefore important to have such a spatial filtering. The space filter is often -made as a housing completely enclosing the units 3 and 4, having only a small opening towards the object through which thermal radiation can enter.

Figures 6, 7 and 8 illustrate alternative forms of thermal excitation to the form shown in Fig. 1. A slip contact form of excitation is shown in Fig. 6, in which a moving object 61 is contacted by a heat conducting foot¬ like member 63 being supplied with heat from a heat source 62 and being pressed to engagement with the surface of object 61 by means of a spring 64. The contact portion of member 63 will engage the object surface at a small area the dimension of which is indicated with a in Fig. 6.

Fig. 7 shows another form of mechanical contact for thermal excitation. A rod shaped member is adapted to con¬ duct heat from a heat source 72 to a rounded end or head portion 73 engaging the surface 71 of the object. A spring 74 is provided in order to secure a safe contact.

In Fig. 8 there is shown a form of thermal excitation

SUBSTITUTESHEET based upon a flame 83 obtained by the combustion of a suit¬ able gas supplied from a source 82 through a tube and nozzle 84. Here also the heated or excited area is indicated by dimension a on the surface 81 of the object.

As in the arrangements shown in the above Figs. 6, 7 and 8, the following figures 9, 10, 11 and 12 include an elongate or continuous object moving in a direction towards the left of this sheet of drawings, as indicated with arrow V in Fig. 9 (and in Fig. 6) .

In the configuration illustrated in Fig. 9 an object 91 is moved (arrow V) in relation to a point or area of exci¬ tation 92 and two detectors 94 and 95 respectively, so as to establish a path of relative movement indicated at 90. Thus, there are two detectors 94 and 95 of which one detector 94 senses the surface of object 91 upstream of excitation 92 thereby recording a signal from the thermally unexcited object, this signal being intended to be used as a reference signal. The signal recorded by means of detector 95 is a result of the excitation at 92.

Fig. 10 illustrates an arrangement in which an object has two parallel paths of relative movement with respect to excitation point 102A with an accompanying detector 104A and excitation point 102B with detector 104B respectively. Excitations at 102A and 102B can be by means of separate electromagnetic beams at two different wavelengths and de¬ tectors 104A and 104B respectively, being adapted to respond to the respective wavelengths. Alternatively, excitation at 102A may be some form of heat excitation whereas excitation at 102B may be in the form of cooling, as briefly suggested in connection with Fig. 1 above. Instead of a parallel arrangement of two separate movement paths as in Fig. 10, the dual arrangement of excitation and detectors can be related to one and the same path as illustrated in Fig. 11. Thus, i-n Fig. 11 one path 110 is common to two excitation points 112 and 116 as well as two detectors 114 and 118. Excitation at 112 is detected at 114 and excitation at 116

SUBSTITUTESHEET is detected at 118. Excitation at 112 can be cooling and excitation at 116 can be heating or both may be heating for example by electromagnetic radiation at different wave¬ lengths.

Finally, Fig. 12 illustrates a configuration in which an excitation point 122 and a corresponding detector 124A are provided along one path 120A, whereas only a detector 124B (without preceding excitation) is provided along path 120B. Thus, the latter detector serves to detect thermal radiation from this object along a path being adjacent and parallel to the excited path 120A, so that the signal from detector 124B can be used as a reference signal.

It is emphasized that this invention is distinguished both in principle and practically from previously demonstrate stationary as well as pulse based PTR-techniques. The therma excitation of the object in contrast to the known techniques, is continuous and for- most purposes constant in time. Further however, it is important to note that the relative movements of the object serve at least six different purposes:

I: Firstly the relative movement between object and the thermal source, with v.>u(d) , will transform the continu¬ ous (and possibly constant) thermal excitation into a consecu tive series of transient thermal excitation over the object, each having its own transient time development. In other wor the object movement makes it possible to utilize the advan¬ tages of the transient thermal waves without having to resort to pulse excitation.

II: Secondly the object movement in relation to the source provides for the initiation of steadily new transient thermal waves during the fiffusion time τ, while preceding transient waves still are underway. The greater v, is in re¬ lation to u(d) the more such waves (i.e. measurement points) will there be time for during the time τ, whereas the pulse methods as a maximum make possible one measurement per time τ because the whole time evolution shall be recorded. Here

SUBSTITUTE SHEET 20

lies the inventive significant advantage in rate of measure¬ ment compared to the pulse method: If more measurements per time unit are needed, this is obtained just by increasing ^vl'

III: Thirdly the displacement of the thermally excited areas of the object by a distance L = v. •τ (in relation to the source) will selectively pick out for recording the ther¬ mal radiation at one and the same instant at the transient temperature evolutions from each individual object point being excited. In contrast to the pulse methods the object movement in other words involve that only one point on the temperature curve is recorded for each individual ^■ excitation point.

IV: Fourthly the relative movement between the object and the field of view of the detection system implies that the thermal signals corresponding to the definite instant τ on the respective temperature evolutions independently of each-other (because v,, > u(d)) by the detector are converted into consecutive time variable electrical signals with an unambiguous correspondence between the electrical signal at a certain time, and a certainexcitation point on the object.

V: Fifthly it will be possible just because of the relative movement between the object and the field of view in the detection system, to adjust the relative velocity v_ such that all the structural details to be investigated, give signal variations which in frequency fall within the electrical bandwidth of the detection system. Then these electrical signals can be amplified and processed in a conventional manner.

VI: Sixthly the relative movement means that the thermal excitationand the detection take place atphysically separated locations, which in a simple way separates the detection from the thermalexcitation so that theexcitation is effected out¬ side the field of view of the detection system. This involves inter alia that the same electromagnetic wavelength can be employed both for the thermal excitation and for the detection without interference (mutual disturbance) .

SUBSTITUTE SHEET 21

The above description states the necessary conditions for the invention to be able to function. In order to simp¬ lify the discussion it has been implicitly understood that the instantaneous thermal excitation takes place only in a thin surface layer of the object, much thinner than the cur¬ rent distance d , and that also the resulting thermal emission originates from a correspondingly thin surface layer. This corresponds in practice also to the most interesting materials and objects with which the invention involves ad¬ vantages compared to existing methods. As the method is de¬ scribed here, it will in the principle also be useful with materials being partly transparent at least with respect to excitation and in part also to emission, because spectral or thermal inhomogenities in the object both will give different start conditions for the thermal waves and also influence the evolution thereof and their re-radiation, differently. However, such cases may be more difficult to interpret and it is therefore an advantage to select spectral ranges for the detection as stated in (4) , where the object material has a maximum of opacity.

Depending upon the design of the equipment used for implementing the invention, there are additionally a few con¬ ditions to be satisfied:

1) : Thermal waves from an excited point on the object propagate a distance d = τ*u(d) along the surface (as well as in the volume) during the diffusion time τ. Therefore the distance d also determines the highest longitudinal resolution (i.e. in the direction of movement of the object in relation to the thermal excitation) which it is possible to obtain with this method. If the thermal excitation has an extent a > d on the object as seen along the direction of the relative movement, the relative velocity v. according¬ ly must be so high that a/v.. < τ, so that

v- > f u(d).^' <¹⁷>

SUBSTITUTESHEET The thermal excitation at each individual point of the object will then as before have a duration < τ, which implies that points at a distance > d still may be considered as independent of each other for times < τ from the diffusion point of view.

If the detection system shall be able to give this re¬ solution in the longitudinal direction, the geometric dimen¬ sion or extent β of the field of view on the object, according to common optical criteria must satisfy β _ d. This requires that the relative velocity v_ between the object and the field of view of the detection system satisfies

V- > u(d) ⁽18⁾

so that areas of extent d on the object pass through the field of view within times <τ. This is implicit with the invention according to item IV above; the detection can then also be regarded as instantaneous in relation to the thermal diffusion processes.

If β > d the effective longitudinal resolution is given by the extent β of the field of view on the object. In order to attain this resolution also from the thermal waves, it will be sufficient to require

v- _ u(β), ₍₁₉₎

provided that β < a; for d < a < β it is only required that v. _: u(β) . At the same time

v₂ ≥ u(β) , _{20)

must be satisfied in order t© have the same longitudinal reso¬ lution in the detection. The effective diffusion length in this case is equal to β, and the effective diffusion velocity is

2

,.. f : u(β) = -g— (21)

SUBSTITUTE SHEET 23

2 2 and the corresponding diffusion time is τ_β = β /π K . Thus far the longitudinal resolution.

2) : A main point of the invention is to implement a high resolution transversally to the direction of movement of the object. Assume that structures of interest having a dimension d transversally to the direction of movement shall be identified with a resolution δd << d. ( This may for exampl apply to the determination of thickness variations ≥δd of the object or of a surface layer having an average thickness d.) For propagating this additional distance the thermal wave front needs a time δτ which follows from differentiating (12)

^{δτ = 2}uTd). • <²²>

In order that structures at a mutual transverse spacing δd shall be able to result in detectable changes^* in thermal evo¬ lutions at the excited points on the object, there is accor¬ dingly a requirement that the object shall move with respect to the thermal excitation, by a distance d (equal to the resolution element) during the time δτ (at the same time as also (17)' must apply). This will make it possible marginally to detect differences in the thermal radiation at the times τ and τ + δτ. From v. • δτ _ d then follows

^vl ⁼ 2δd (d) (23)

An ideal optical detection of the thermal radiation has been assumed above, i.e. with β _ d. If β ≥ d, it will be required that the thermal excitation passes over the distance β during a time < τ. Then from v- • δτ _ β it follows in ana¬ logy with the above that

v- _ u(d) . (24;

2δd SUBSTITUTE SHEET 7/0

24

The longitudinal geometrical resolution is then given by 3. This requires at the same time also that

The field of view with an extent β will then be passed through by each individual point on the object during time < τ, and the detection can still be considered as instantaneous in relation to the thermal diffusion processes. Even with a longitudinal resolution restricted by the field of view β of the detection system and not by the diffusion processes, it will in other words still be possible to obtain the same transversal resolution, determined by the thermal waves, but then at higher relative velocities.

3) : If the largest structures in the object to be characterized and/or controlled have a dimension D, these will pass through the field of view in the detection system during a time D/v The inverse of this time corresponds to a frequency which must be higher than the lower cut-off fre¬ q^uency of the detection system fmm. , so that

Good detectors for heat radiation only exceptionally main¬ tain their detecting capability at frequencies lower than

100 Hz. With for example D = 1cm and fmi.n = 100 Hz it follows from (24) that v- > 100 cm/s in order that signals corres¬ ponding to structures of size 1 cm shall be sensed by the electronic part of the detection system. Since typical thermal diffusion velocities over distances of 1 cm vary from

_2 10 cm/s for materials having a low thermal diffusivity

(plastic materials, paper and so on) to 10 cm/s for good metallic heat conductors, the lower limit of the relative velocity v₂ between the object and the field of view of the detection system, will often be determined by (26) , and with values which can far exceed the diffusion velocities.

SUBSTITUTESHEET 25

Correspondingly the highest frequencies in the signal will be. given by v_/g, assuming a diffusion limited longitudinal re¬ solution g equal to the dimension of the smallest structure in the object to be identified. The relative velocity be¬ tween the object and the field of view in the detection sys¬ tem must therefore be chosen so that all frequencies within the range

v.

≤ f < (27)

lie within the electrical bandwidth of the detection system. As the upper cut-off frequency of heat radiation detectors is often of the order of magnitude of MHz or more, (27) provides for wide limits for the choice of v- .

It is emphasized that v.. and v- in the principle are unequal and this is also often the case in practice. Nor do they have to be directed along the same path in relation to the object in order that the invention shall be useful, as long as the conditions stated above are satisfied. For example v₁ and v- can be directed along the same path as the object, while for example v- is constant and v, varies periodically about an average value. The distance between the thermal excitationand the detection given by the diffu¬ sion time τ will then also vary periodically, so that there is essentially effected a scanning of the object over a range of diffusion lengths d corresponding to the variety of different diffusion times τ. This may be of interest in cases where the relevant distance d is not known or varies, possibly also when investigating a number of structures each having its own characteristic length d, by traversing a spectrum of diffusion lengths.

A few examples will serve to present the invention more concretely. Assume in the following that there is used a detection sy ^Jstem with fmm. = 100 Hz.

SUBSTITUTE SV gV 87/00632

26 a) . Investigation of internal structures in semi-con¬ ductors may for example be of interest for depths of the order' of magnitude 0,1 mm. Thermal diffusivity is K = 0,3 cm /s

(as for Ge) , which from (13) gives u = 3 m/s with d = 0,1 mm.

-2 With a resolution δd = 10 mm it follows then from (23) that v, ≥ 15 m/s, which is based upon a thermal excitation of dimension a < 0,5 mm (from (17) ) and a well focused detection system with 3 < 0,1 mm (from ( 24)) . If the largest dimension to be recorded is D = 1 mm, it follows from (26) that v.,>10 cm/

Since this is a lower value of the relative velocity than what is given by the diffusion velocity u, in this case v_ will be determined by the general requirement set by the invention as stated in claim 1, v~ > u ( = 3 m/s) . b) . Protective coatings on surfaces is another important field of use of the invention. Assuming a coating of thick-

-2 2 ness 50 μm and K = 10 cm /s (glass and the like) , then u = 20 cm/s. With a thermal excitation of dimension a = 0,5 mm, a field of view β = 0,2 mm and a resolution requirement δd = 5 μm during control of the thickness, it follows from (15) v_χ > 2 πi/s, whereas (24) gives v. > 4 m/s. Then the latter value will determine v, . In order to test for possible delamination over larger areas, it may be of interest to have D = 10 cm which from (26) gives v_ ≥ 10 m/s, and corresponding¬ ly (25) gives v- ≥ 0,8 m/s. The velocity of the object in relation to the detection system in this case will be the dimension determining the choice of the relative movements, c) . Measurement of thickness of for example aluminium products and identification of possible flaws in the material is another relevant example. The thickness can be d = 5 mm,

2 thermal^"diffusivity is < - 1 cm /s from which follows u = 5 cm/s. With a resolution δd = 0,1 mm it follows from

(23) that v, ≥ 1,25 m/s. Here it will be easy to make both a and β < d. If the largest structures sought have D = 1 cm, must v- ≥ 1 m/s in accordance with (26) . The requirements to the two relative velocities will then in other words be

SUBSTITUTE SHEET practically identical. d) . Paper and synthetic materials are industrial pro¬ ducts representing a need for monitoring the thickness during production. Assume that d = 0,1 mm and K = 10 -3cm2/s, which gives u = 1 cm/s. With a = 1 mm, β = 0,5 mm and δd = 5 μm it follows from (17) that v, > 20 cm/s, whereas (24) gives v- ≥ 50 cm/s. If defects, irregularities etc. of a dimension D = 1 cm is to be recorded, (26) as above gives v_ > 1 m/s and for comparison (25) gives v_ ≥ 5 cm/s. e) . Chemical substances exposed to the surface of the object can be identified by means of heating with electromag¬ netic radiation at the absorption line characteristic to the substances. Assume that the absorption coefficient is (λ) = 10 cm which gives a depth of penetration α(λ)~ = 10 cm. This gives the thickness of the surface layer in which the major part of the thermal excitation will be disposed. With d = 10 -3cm and < = 10-2 cm2/s it follows that u(d) = 1 m/s. If a = 0,5 mm and β = 0,2 mm the longi¬ tudinal resolution is given by β = 0,2 mm. If it is desired to monitor the object only to a depth equal to the absorp¬ tion length α , it is required that v, ≥ 50 m/s (from (17 ) and v- ≥ 20 m/s (from (25)). In such cases, however, these will be unnecessary high requirements to the relative velo¬ cities. Since here it may hardly be of interest to have a transverse resolution of the order of magnitude 10 cm, it may be sufficient to define a diffusion velocity u(β) which implies that the resolution is given by the longitudinal resolution β. From (21) it is found that u(β) = 5 cm/s. (19) and (20) then give v.. ≥ 12,5 cm/s and v- ≥ 5 cm/s re¬ spectively. The largest structure to be recorded may be for example D = 2 mm (text, line drawings and the like) , which gives v_ > 20 cm/s from (26) .

Depending upon the measurement situation concerned it is accordingly possible to carry out more^' exact calculations of the limits to the relative velocities of the object in relation to the thermal excitation and in relation to the field

SUBSTITUTESHEET of view of the detection system. The fundamental prerequisites and limiting values which define the method will anyhow be given by claim 1, which must always be fulfilled.

The above examples only indicate a little of the variety of uses of the invention: Inspection of semi-conductor compo¬ nents and surface coatings, thickness control and investigations of homogenety of thin metal materials, foil materials, paper etc., reading of text, patterns, chromatograms and so on, spectral 'thermography for medical analysis of skin and other organs, checking of the composition of chemical products and the like, testing and control of curing processes and other chemical processes and reactions, hot and glowing objects and so forth as well as liquid films. In all such cases the invention seems to offer new possibilities of measurements, as the method in its most general aspect is completely con¬ tact free and does not interfere with the object. The method may be made very fast and thereby can be adapted to industrial process and manufacturing requirements, the measurements are very sensitive and there is employed technical equipment and components being already well established.

It will also be possible to utilize the invention for recognizing characteristic structures and/or patterns existing in materials and objects. These may exist inherently or they may be imposed deliberately. Such patterns and structures will provide characteristic signals when they pass through the field of view of the detection system, and by means of an electronic signal processing system being particularly adapted to recognize just the signals concerned, the respective structμres and patterns can be identified among all possible other structures and patterns. For example this can be used for seeking objects having specific signatures which can be visually unobservable, it can be employed for characterizing a group of objects according to certain criteria and so forth. Another use consists in checking that objects have a set of properties which warrants that they are genuine, for example

SUBSTITUTE SHEET in connection with stocks and shares, bank notes, other securi¬ ties and the like.

For example the objects can include characteristic structures of different materials being for example included at a certain depth in the object (as in semi-conductor ma¬ terials) . By employing the method described above it will then be possible to bring about temperature variations at the surface which reflect these internal patterns, and then by means of the signal processing system it can be checked that they have the correct form. The same applies if the object contains systematic and characteristic mass changes not being visible at the surface, for example holes, inclusions and the like, systematic variations in thickness, ror example watermarks in paper, characteristic patterns and structures of chemical nature, for example dye stuffs having pronounced spectral absorption lines arranged as text, pictures and the like. Inall the cases mentioned here the invention can be utilized for recog¬ nizing, controlling and/or characterizing objects through the systematic and characteristic structures discussed, by thermal excitation at one side of the object and detection of resulting thermal radiation from the same or the opposite side of the object.

Finally it is emphasized that the invention is based upon the excitation and utilization of transient thermal waves in a way which has not previously been demonstrated. As stated in claim 1 a necessary condition for this is that both the source of thermalexcitation and the field of view of the detection system have relative velocities in relation to the object which are higher than the thermal diffusion velocity u(d) , as given by (13) , corresponding to the instant τ at which the thermal wave front at a distance d from the thermal excitation has reached approximately 1/3 of its maximum thermal excursion. The relationships (12) and (13) for the thermal time constant τ and the thermal diffusion velocity u(d) , respectively, therefore per se are incidental; since thermal diffusion is a chaoticprocess neither u(d)

SUBSTITUTESHEET nor ~_. can be given an exact definition. Thus, for many uses of the invention it will be an advantage to define a higher value of u(d) on the basis of a time at which the temperature excursion has reached less than 1/3 of its maximum value, since this can contribute to increasing the thermal contrast between adjacent structures. In other situ¬ ations it will be more reasonable to delay the thermal detec¬ tion until the temperature excursion has reached closer to 100 % of its maximum. This gives more powerful thermal signals The corresponding thermal diffusion velocity will then be perceived as a lower velocity. The choice made in that re¬ spect essentially determines the definition of u(d) and has a direct influence on the limits of the relative movement velocities as these are stated in claim 1. Therefore claim 1 must be understood in the light of what is said here about the definition of u(d). The limiting values of the relative velocities accordingly must be taken as. an approximation to the same degree as u(d) only is defined as an approximation, with allowance for practical adaptions in line with the above. In most cases it will only be question of small numeric corrections of the values of u(d) which can be calculated from (13) . When designing apparatuses for implementing the invention the above can be handled most simply by defining u(d) as in equation (13) , and by choosing the time delay δt between thermalexcitation and detection, greater or smaller than τ in equation (12) , as stated in claims 16 and 17. This will also take care of the case in which the object is porous, which implies that the thermal excitation will pene¬ trate deeper than in a corresponding homogenous material, at the same time as also the thermal radiation can originate from deeper layers in the object than what is contemplated in the analysis above. Both these effects contribute to an increase of the apparent thermal diffusion velocity.

SUBSTITUTE SHEET Literature references.

(1) Nordal, P.-E. and Kanstad, S.O., Physica Scripta 2 Q_ 659

(1979) .

(2) Hendler, E. , Crosbie, R. and Hardy, J.D., J.appl.Physiol. T2 177 (1958) .

(3) Rosencwaig, A. and Gersho, A., J. appl. Phys . 47_ 64 (1976) .

(4) Nordal, P.-E. and Kanstad, S.O., Appl.Phys.Lett. _3JL 486

(1981) .

(5) Nordal, P.-E. and Kanstad, S.O., in Scanned Image Micro¬ scopy (E.Ash, Ed.) , p. 341. Academic Press, London (1980) .

(6) Busse, G., Infrared Physics 2fJ ⁴¹⁹ (1980) .

(7) Carslaw. H.S. and Jaeger, J.C. Conduction of heat in Solids, Oxford University Press, Oxford (1959) , p. 65.

(8) Parker, W.J. , Jenkins, R.J., Butler, C.P. and Abbott, G.L. , J.appl. Phys. 3__. ¹⁶⁷⁹ (1961) .

(9) Deem, H.W. and Wood, W.D., Rev. scient .Instrum. _3 1107 (1962) .

(10) Tarn, A.C., Infrared Physics 2_5 305 (1985) .

-0-

SUBST1TUTE SHEET

Claims

C L I M S :

1. Method for the characterization and/or control of sub¬ stances, materials and objects by imposing temperature changes therein (in the object) and detecting the resulting change in thermal radiation from the object, characterized in that there is provided for a relative movement between the object and a device for external thermal excitation, that this excita¬ tion is continuous and preferably constant with time and is imposed on the object as a consecutive heating or cooling along a path defined by the relative movement of the object, that there is provided for a relative movement between the object and the field of view of a detection system adapted for mea¬ suring the resulting change in thermal radiation from the object, and that the velocity of movement of the object in re¬ lation to the field of view in the detection system and in re¬ lation to the external thermal excitation is higher than, and preferably much higher than the effective thermal diffusion velocity in the object.

2. Method according to claim 1, characterized in that the effective thermal diffusion velocity u(d) is determined by

2

wherein < is the thermal diffusivity of the object material and d represents the geometrical dimension of the structures, lengths and/or thicknesses of the object to be characterized and/or controlled.

3. Method according to claim 1 or 2, characterized in that the movement velocity v_ of the object in relation to the field of view of the detection system is such that all fre¬ quencies f satisfying the relationship

_^v2 < f < —^V2

SUBSTITUTE SHEET fall within the elctrical bandwidth of the detection system, wherein D and g represent the geometrical dimension of the largest, respectively smallest structure, length or thickness- of the object to be characterized and/or controlled.

4. Method according to any one of claims 1- 3, characterized in that the thermal excitation is effected by mechanical contact between the object and means for thermal heating or cooling.

5. Method according to any one of claims 1 - 3, characterized in that the thermal excitation is effected by means of hot or cold gas.

6. Method according to any one of claims 1 - 3, characterized in that the thermal excitation is effected by means of a flame.

7. Method according to any one of claims 1 - 3, characterized in that the thermal excitation is effected by means of ultrasound adapted to be absorbed by constituents of the object.

8. Method according to any one of claims 1 - 3, characterized in that the thermal excitation is effected by means of an electron beam or another particle beam.

9. Method according to any one of claims 1 - 3, characterized in that the thermal excitation is effected by electromagnetic radiation having a spectrum selected so as to be absorbed by constituents of the object.

10. Apparatus for carrying out the method according to any one of claims 1 - 9, comprising means for thermal excitation on the object and a detection system adapted to measure the resulting change in thermal radiation from the object, characterized by means for providing a relative movement be¬ tween the thermal excitation means, respectively the detection system, and the object, said excitation means being adapted to deliver a continuous thermal excitation.

SUBSTITUTE SHEET

11. Apparatus according to claim 10, characterized in that the excitation means and the detection system have a mutual spacing along the path for said relative movement.

12. Apparatus according to claim 10 or 11, characterized in that the movement means is adapted to provide for movement of the object and that the excitation means, respectively the detection system are substantially stationary.

13. Apparatus according to claim 10, 11 or 12, characterized in that the movement means is adapted to provide for the same relative movement velocity of the object in relation to the excitation means and to the detection system.

14. Apparatus according to any one of claims 10 - 13, characterized in that the excitation means is adapted to apply the thermal excitation at one side (the front side) of the object and that the detection system is adapted to measure the resulting thermal radiation from the same side of the object.

15. Apparatus according to any one of claims 10 - 13, characterized in that the excitation means is adapted to apply the thermal excitation at one side (the front side) ^• of the object and that the detection system is adapted to measure the resulting thermal radiation from the other side (the back side) of the object.

16. Apparatus according to any one of claims 10 - 15, characterized in that the distance L between the means for thermal excitation and the field of view of the detection system is fixed and given by

L = vδt,

2 2 where δt > 0,5 d (τ K) and v > u(d) .

17. Apparatus according to any one of claims 10 - 15, characterized in that the distance L between the means for thermal excitation and the field of view of the detection system is variable in accordance with the relationship:

SUBSTITUTE SHEET

2 2 wherein L = vδt, with δt > 0,5 d / (π K) and v > u(d) ,

L-_^ < L_Q and F(ωt) is a periodic function having an amplitude equal to 1 and an angular frequency ω which satisfies

ω < γ- (v - u(d)) .

^Ll

18. Apparatus according to any one of claims 10 - 17 for use with objects or products being provided with systematic and characteristic structures and/or patterns of physical and/or chemical nature, characterized by comprising an elec¬ trical signal processing system being adapted to recognize the corresponding systematic and characteristic thermal sig¬ nals generated when said structures pass through the field of view of the detection, system.

19. Apparatus according to any one of claims 10 -^' 18, characterized in that the detection system comprises a spec¬ tral filter for defining the spectral range of the thermal detection and discriminating against electromagnetic radiation imposed on the object from the means for thermal excitation.

20. Apparatus according to any one of claims 10 - 19, characterized in that the detection system comprises at least two detectors located at a spacing along the path of said relative movement.

21. Apparatus according to claim 20, characterized in that one the detectors is located before said means of thermal exication thereby recording a signal from the thermally unexcited object, said signal being used as a reference signal.

22. Apparatus according to any one of claims 10 - 21, characterized in that the excitation means is adapted to deliver separate electromagnetic beams at two or more different wavelengths and with positions on the object at a certain space transversally to the direction of movement, and that the detection system comprises corresponding separate detec¬ tors for each wavelength.

SUBSTITUTESHEET

23. Apparatus according to any one of claims 10 - 22, characterized in that the excitation means is adapted to delive a first excitation and a second excitation being independent of the first excitation, in parallel to or coinciding with the path thereof, that the detection system comprises a corresponding first and second detector respectively, and that one of the excitations is in the form of heating, whereas the other excitation is in the form of cooling.

24. Apparatus according to any one of claims 10-23, characterized in that in addition to the detector(s) for detecting radiation caused by said excitation(s) , there is provided at least one detector for detecting thermal radiation from the object along a path being adjacent and parallel to the excitation path(s) , the signal(s) from said at least one detection being used as reference signal(s) .

25. Apparatus according to any one of claims 10 - 24, characterized in that the movement means is adapted to pro¬ vide for a periodically variable relative movement velocity.

26. Product adapted to be subjected to the method" according to any one of claims 1 - 9, characterized in that it is pro¬ vided with systematic and characteristic structures and/or patterns of physical and/or chemical nature which represent a signature adapted to be detected upon a continuous thermal excitation and relative movement of the product in relation to the points at which thermal excitation and detection respectively, take place.

27. Product according to claim 26, characterized in that said systematic and characteristic structures and/or patterns consist in localized physical thickness, material and/or mass changes.

2 8. Product according to claim 26 or 27, characterized in that said systematic and characteristic structures and/or patterns consist of chemical substances having strong and pronounced spectral absorption lines in parts of the electro¬ magnetic spectrum.

SUBSTITUTE SHEET