GB2054136A - Apparatus for Measuring the Thickness of Thin Layers - Google Patents

Abstract

Apparatus for measuring the thickness of thin layers 29 of strip- or wire-like material comprises a beta emitter (in tubes 17, 22) for directing radiation toward the layer, and a detector 11 for receiving radiation scattered back from the layer and operable to produce a count in response thereto. The arrangement is such that the curve showing the relationship between the counting rate and the distance of the emitter from the layer has a plateau region in which variation of the distance causes substantially no change in the counting rate, and the distance between the emitter and the layer is in this plateau region. The relationship is preferably obtained by arranging for the ratio of a characteristic cross- sectional dimension (D) of a window of the detector to the diameter of the intersection area between the layer and the radiation from the beta emitter to be from 15 to 2, and preferably from 10 to 4. <IMAGE>

Description

SPECIFICATION Apparatus for Measuring the Thickness of Thin Layers This invention relates to apparatus for measuring the thickness of thin layers, and in particular to such apparatus which uses a beta radiator which directs beta radiation to the to the layer, and a detector for detecting beta radiation scattered back from the layer.

Such apparatus is known from U.S. Patent Specification No. 4,115,690. This apparatus has, inter alia, the following disadvantages: 1. The strip-shaped material under measurement is bent over repeatedly in the vicinity of the measuring instrument. The usually metallic material may be permanently deformed as a result. In this case the material is bent over repeatedly so that there is not only bending in one direction, but bending back and forth in both directions. This is also disadvantageous. If the material is then rewound onto a reel with torsion after measurement, detrimental forces can occur as is known from twine or bobbins.

2. The measuring instrument rotates continuously. Therefore, the necessary inputs and outputs must be provided via slip rings.

3. The measuring time is limited. As beta radiation back scatter involves atomic statistics, the measuring precision can be increased by a sufficiently long measuring time. However, this is in principle impossible with the known instrument since the measuring time cannot commence until the material reaches a rotatable drum and must be terminated when the material is moved away from the measuring drum again. Therefore, in practice measurement is only possible over a time of about 2700 of drum rotation and absolutely impossible for about 900 of the time.

4. Synchronizers must be provided to allow the measuring time to begin and end at the correct time.

5. In order that sufficient measured results may also be obtained, a plurality of beta emitters must rotate with the drum. However, to ensure that one beta emitter does not measure for a longer period than the other, the same length of measuring time determines that the beta emitters must be staggered through 1 800 when there are two such devices, by 1 200 when there are three beta emitters, etc.

6. In practice, at least two beta radiators must be provided to prevent the reading being merely a random test measurement.

7. The subsequently used computer must be equipped to process overlapping measuring times. In this respect also there are higher requirements than if only one beta radiator were to be used.

8. A really continuous reading is in this case merely replaced by a random test measurement, depending on the system. This method of measurement is not adapted to the measuring problems If it is assumed, for example, that the instrument is intended for measuring copperplated objects coming from a copper bath or gilded objects coming from a gilding bath, these units generally change their properties only at a slow rate, the time constants being in the region of at least a quarter of an hour. However, it is also characteristic for the coating to be suddenly defective at some point, but completely flawless to the left and right of said point. The known apparatus is not easily adapted for solving these problems.

9. If the beta radiator has detected a flaw, it must be determined which of the. if necessary numerous beta radiators has detected the flaw.

1 0. The drum must have a comparatively large diameter to ensure careful treatment of the material. This creates design problems and, even when the instrument is installed in continuous production systems, this design results in disadvantages as there is usually very restricted space available in these systems.

11. If it is desired to measure strips or wires with the instrument, the known apparatus may be suitable as these materials are uniformly homogeneous. However, if it is desired to measure electric contacts which are in the form of punched-out strip, there is naturally sometimes one contact at one point and none at another point. It must therefore be ensured that the beta emitter is exactly opposite the contacts. This means that a quite special outer face of the drum is required for each measuring problem.

12. The known apparatus uses the known annular diaphragms or outer rings between the beta emitter and the material. Although these rings are made of a very hard material, they still wear. Moreover, particular care must be taken to ensure that the object to be measured is properly in contact with the annular diaphragm otherwise erroneous readings are obtained.

1 3. Because the material is guided on to the drum e.g. at 10 o'clock and guided off the drum at 7 o'clock, the first linearly advancing strip is not in alignment with the strip which is subsequently moved linearly off the drum. This staggered arrangement is in itself disadvantageous.

However, as the material is under tension, this means that a component of a force is exerted on the apparatus at the same time.

It is desirable to provide an apparatus which avoids at least some of these disadvantages, and which is cheap and the measuring principles of which can be applied to materials having different characteristics.

In accordance with the invention there is provided apparatus for measuring the thickness of thin layers on elongate material, comprising a beta emitter directed toward the layer, a guide for guiding the material so that it can be moved past the beta emitter at a predetermined distance therefrom, and a beta radiation detector disposed to the rear of the beta emitter for receiving beta radiation scattered back from the layer as the material is being moved in order to produce a count in response thereto, wherein the emitter and detector are so arranged that a graph showing the relationship between the distance of the emitter from the layer and the counting rate includes a plateau region in which variation of said distance causes substantially no change in the counting rate, and wherein said predetermined distance between the layer and the beta emitter is in said plateau region.

Preferred features of the invention are set out in the subordinate claims.

In the preferred embodiment, there is provided stationary apparatus particularly suited for measuring the thickness of thin layers of stripshaped, wire-shaped or like material. The relationship mentioned above may be achieved at least in part by arranging for a characteristic cross-sectional dimension (D) of a window through which the detector receives the backscattered radiation, and the diameter (d) of the intersection between the intensity distribution curve of the beta emitter and the layer, to have the relationship D from 15to2, d and preferably from 10 to 4.

The way in which D is measured depends on the shape of the window; for a circular window, D would be the diameter, and for a square window D would be the length of an edge of the square. In any event, D should be representative of the size of the window. As an alternative, the ratio of the areas of the window and of the intersection between the layer and the intensity distribution curve can be set to an appropriate value, instead of the ratio D/d, the appropriate ranges for the ratio of areas being obtained by squaring the figures mentioned above.

The value d need not be exact, and various approximations can be used to establish this value.

The number of backscattered beta particles absorbed by the counting tube pass back through the diaphragm or outer ring. However, this is of no consequence with this kind of measuring problem since the material is in any case continuously coated and since moreover the characteristics of the coating systems have high time constants, e.g. the thickness of the coating varies only slowly so that long measuring times can be used. One great advantage of this arrangement is also the fact that the curves for different materials are parallel. For example, the curves in respect of gold, nickel, platinum, etc. are now parallel for practical purposes. Therefore, one curve can be derived from the other merely by changing an additive factor.

The features of claim 4 make it possible for the plateau effect to be further intensified and prevent shading or dimming by the small tube.

Also, on account of the features of claim 4 large beta radiation detectors can also be used. In the case of Geiger-Muller tubes, for example, the large tubes are cheaper than the small tubes. No annular diaphragm or outer ring is required for the small tubes, but they are still costly.

Claim 4 shows a possible way of using large, but cheap beta radiation detectors. With semiconductor beta radiation detectors it is exactly the opposite to G-M tubes. In this case the smaller ones are cheap and may be generally used without an outer ring, while the larger detectors are more costly and therefore require an outer ring. It is a similar case with scintiilation counters.

The features of claim 5 increase the plateau effect and reduce the measuring time. Although only one tube need be used, if there are two optimization is achieved with regard to expenditure and result obtained. Even a large number of tubes would not bring about a substantial improvement by an order of magnitude (i.e., a factor of 10).

The features of claim 6 simplify the design and possibly also mathematical treatment or usage.

Through the features of claim 6 the tube is prevented from dimming backscattered beta particles, thereby also improving the plateau effect.

The features of claim 7 make it possible for the plateau effect to be improved, the layer or coating to be exposed to the maximum number of beta particles and at the same time for the maximum number of beta particles to be scattered back to the flaw detector tube.

Also, the features of claim 8 simplify the design and the possibly mathematical processing. The features of claim 8 make it possible to achieve a spatially identical staggered arrangement and therefore uniform exposure to radiation and better conditions for data processing.

Uniform measuring principles are achieved by the feature of claim 9. Also, the angle sizes according to claim 9 improve the plateau effect.

The features of claim 10 can also achieve the plateau effect, the optimum in terms of expenditure/result being obtained by using two tubes. A simple construction, a more uniform irradiation and a configuration which is simpler for data processing can also be achieved by these features.

The features of claim 11 have the same effect.

The plateau effect in these embodiments may also be improved by the features of claim 1 2.

The features of claim 13 enable the measuring time to be reduced and the plateau to be broadened somewhat.

The features of claim 14 enable the diaphragm or outer ring to be used at the same time as a carrier for the radionuclides.

Preferred embodiments of the invention will now be described by way of example with reference to the accompanying drawings, in which: Fig. 1 is a diagrammatic side view showing an apparatus according to the invention, in partial section, and a coated material, Fig. 2 is a view of the annular diaphragm in the direction of arrow A shown in Fig. 1, with tubes indicated by dot-and-dash lines, Fig. 3 is a substantially enlarged side view of a tube, showing a hole and an inserted radionuclide, the intensity distribution curve and intersection with the coated material, Fig. 4 shows the curve path with a slightly broadened plateau region which can be achieved with the invention, Fig. 5 is a view, in the direction of arrow A, of tubes in a three-dimensional arrangement, but without an annular diaphragm and counting tube, Fig. 6 is a view similar to Fig. 1, but showing a surface emitter combined with the outer ring, Fig. 7 is a side view similar to Fig. 1, but without showing the material or coating when only one tube is used, Fig. 8 is a view similar to Fig. 7, but showing two tubes, in a staggered arrangement.

A conventional type of G-M tube 11 (Geiger Müllertube) comprises an output 12 which generates a pulse in a known manner whenever beta particles pass from the left into the G-M tube 11 in the direction of arrow A, for example. The G M tube 11 is arranged coaxially with a geometric longitudinal axis 13. An annular diaphragm 14 made of beta radiation-absorbing material is provided in front of the window of the G-M tube 11. The annular diaphragm 14 has a coaxial and preferably circular aperture and is rigidly connected to the G-M tube 11.

A tube 1 7 is made of a beta radiationabsorbing material and has a blind hole, in the bottom of which is situated a radionuclide 1 9. The beta rays can be emitted from the aperture 21, but because of the geometry of this beta radiation device only a very thin pencil of rays is emitted.

The tube 17 is rigidly mounted (in a manner not shown) together with the G-M tube 11 and the diaphragm 14, that is, so that it is inclined at an angle of approximately 450 to the longitudinal axis 13, allowing the rays to be emitted at the bottom left end of the tube.

A second identical tube 22 is arranged below the first tube 17, lying at an angle of -450 to the longitudinal axis 1 3 and accordingly emitting radiation from the top left end. The tubes 1 7 and 22 lie in the plane of the drawing in Fig. 1 and have the same mirror-image geometry relative to the longitudinal axis 13 and therefore also to the ring 16 and G-M tube 11.

A guide 23, 24 is rigidly connected to the aforementioned parts (in a manner not shown) and, like these, is also stationary; the guide can, for example, be made from a section of plastic material and symmetrically arranged relative to the geometric longitudinal axis 13 at least in the vicinity thereof. An intermediate free space 26 is left between the guides 23, 24. The radionuclides of the tubes 1 7, 22 emit rays into part of the free space 26.

To the left of the guide 23, 24 there is provided a strip 27 made of a material 28 provided with a coating 29. The coating or layer 29 is in contact with the guide 23, 24 and is irradiated with beta rays from the tubes, 17, 22. The strip 27 is moved in a straight line in the direction of arrow B. It could also be moved antiparallel to arrow B or it could be moved in either of the directions perpendicular to the plane of the drawing in Fig.

1. The strip 27 can be solid or it can also take the form of punched-out parts not yet separated from the strip, such as e.g. contact springs or the like.

The strip 27 can also be in the form of a wire which can have a circular, oval or the like cross section. The material 28 can be, for example, copper and the coating 29 gold, for example. The material 28 does not always have to be thicker than the coating 29. On the contrary, the material 28 can have an infinitely variable thickness in practice in view of the backscatter properties of the coating 29 or alternatively it can have zero thickness.

There is a distance a between the layer 29 and the top of tube 17 or 22. As shown in Fig. 1, this distance is measured perpendicular to the layer 29. It is immaterial whether the distance a is measured from the right or the left bounding surface of the layer 29 since the layer thickness in relation to the distance a is negligible.

This distance a is plotted as the abscissa a in Fig. 4 and the counting rate X as the ordinate. The counting rate is the number of pulses obtained from the output 12 after a predetermined time.

Cure 31 is associated with a layer of platinum, curve 32 with a layer of gold and curve 33 with a layer of nickel. Each of these curves 31, 32, 33 has an ascending branch 34, extending to the right into a plateau 36 and leading even further to the right into a descending branch 37. For the sake of clarity the plateaus 36 are shown somewhat broader. Therefore, when the layers are of equal thickness, the counting rate for platinum is higher than that for gold because platinum has a higher atomic weight. The same applies to the ratio nickel to gold, on the one hand, and nickel to platinum, on the other.

If this diagram is compared with Fig. 8 from German Patent Specification No. 20 13 270, corresponding to U.S. Patent Specification No.

3 714436 and British Patent No. 1,323,906, the first difference to be noticed is that there are plateaus 36 in the diagram according to the invention and the second difference is that the curves 31, 32, 33 can be derived from one another by parallel translation.

The distance a 0 lies at the centre of the plateau region 36. If the layer 29 now varies by delta a half, this has no effect on the counting rate X. This characteristic is important because in many practical cases the layer 29 does not always have the distance a 0. Variations can occur because, for example, the layer 29 has a surface pattern, but the guide 23, 24 guides the layer 29, as it were, over the mountain peaks.

However, variations in distance can also be due to the fact that the stamped and/or punched-out parts are not always in exact alignment, when seen in the direction of arrow B. These variations in distance can be considerably greater than the thickness of the layer. However, the distance variation is not inconvenient as a result of the plateaus 36.

With practical embodiments the distance variations delta a can be in the region of 0.4 to 0.6 mm. In practical cases the value of a0 is 0.1 to 1.5 mm.

It is naturally also very convenient for subsequent mathematical calculation of the counting rate X if the curves 31,32,33 are substantially parallel displaced curves, and particularly in the delta a region in question. The counting rate contains information as to the thickness of the layer 29.

Fig. 5 shows diagrammatically how tubes 38, 39, 41 would have to be arranged, if the number of tubes =3, in order to obtain the simplest possible conditions for construction and data analysis. Irregular angular distances are naturally also possible. They are 1200 in the illustrated case. The angle of inclination of the tubes 38, 39, 41 is in this case also 450. However, the tubes could also be arranged at more acute or more obtuse angles.

According to Fig. 6 there is provided an annular diaphragm or outer ring 42 which also has a coaxial, circular aperture 43 and partially shields the G-M tube 11 on the left. In this case a coaxial V-shaped groove 46 is provided in the left front side 44. The outer edge 47 of this groove is in this case coated with radionuclide material 48 which detects the free space 26 and can emit radiation in the direction thereof. In this case also the diaphragm or ring 42 is rigidly connected to the G-M tube 11 and the guide 23, 24.

Better results are obtained both with the arrangements shown in Fig. 5 and in Fig. 6 than with the arrangement shown in Fig. 1. However, this improvement does not amount to an order of magnitude, particularly in view of the fact that the invention permits long measuring times.

A single tube 49 is arranged coaxially in the embodiment shown in Fig. 7. However, the tube 49 does not have to be coaxial in all cases. It can also be arranged at an angle, like the tube 17, for example.

In this connection it should be stressed once more that the free space 26 is in no way equivalent to the previously known diaphragms or outer rings which, according to the above mentioned reference, for example, limit the measuring surface. In the case of the invention there is no limitation of the measuring surface which would be comparable with known measuring surface limitation.

As shown in Fig. 8, the plateau effect and parallelization may also be achieved by two tubes 51, 52 arranged parallel to the longitudinal axis, but staggered at intervals. It can be imagined that this figure originated from Fig. 1, the angle of the tubes 17, 22 to the longitudinal axis being reduced to zero and in addition the tubes 1 7, 22 being in a staggered arrangement.

Another variation can be derived from Fig. in this case also the angle to the longitudinal axis is zero and the three tubes are staggered at three different intervals. Even then the plateau effect and parallelization are achieved.

A characteristic path according to Fig. 4 is obtained even when one or several emitters are used, but when the diaphragm or ring 14 is completely omitted and small G-M tubes or small semiconductor detectors are used.

The thin pencil or rays mentioned at the beginning can be designated as the intensity distribution curve 25 which is rotationally symmetrical with respect to the geometric longitudinal axis illustrated. This curve is therefore a space curve. It is club-shaped similar to the lobes which are used for aerial navigation, are known from the V.H.F. range, etc. In a stable state both the broken and unbroken parts indicate the intensity distribution curve. However, the intensity distribution curve 25 intersects the layer 29 in the embodiment shown in Fig. 1 at an angle of 45 0.

The intersection eliipse has a diameter d. The aperture 1 6 has a diameter D, and these are the values as claimed in Claims 1 and 2. The value of d is not quite exact. It would only be exact if the layer 29 were perpendicular to the geometric longitudinal axis. Also, the intensity distribution curve 25 is not a precisely plottable curve on account of the statistical processes required with emitters.

By way of approximation the value d can also be replaced by the value d 1, as indicated in Fig.

5. In this case it is the diameter of the circle which can be described through the apertures 21 of all the tubes. If two tubes lying opposite one another were used, the value d I could be determined in the same way.

Alternatively, the value d 2 can be adopted by approximation, as indicated in Fig. 7. This is the diameter of the free space 26.

In a practical embodiment dis equal to 3 mm, and D 8 mm, a G-M tube with a window diameter of 18 mm being used.

In practice 0 to 70% will be shielded depending on the size of the G-M tube. In the simplest example the aperture 16 will be circular.

However, it could also be square, and the characteristic cross-sectional dimension D would then be the edge length of the square. The aperture 1 6 and the free space 26 are usually circular because this aperture is easy to produce.

The outer ring or diaphragm will be provided as close as possible to the beta radiation detector.

The best arrangement is for the ring to be attached directly to the detector (G-M tube).

According to past experience the optimum distance between the G-M tube/outer ring unit and the layer 29 is approximately 1 to 4 mm.

In practice aha=0.4 to 0.6 mm. In practice a0 is approximately 0.1 to 1.5 mm. This can mean that the left branch in the curve path of Fig. 4 ceases to exist.

Claims (14)

Claims

1. Apparatus for measuring the thickness of thin layers on elongate material, comprising a beta emitter directed toward the layer, a guide for guiding the material so that is can be moved past the beta emitter at a predetermined distance therefrom, and a beta emitter for receiving beta radiation scattered back from the layer as the material is being moved in order to produce a count in response thereto, wherein the emitter and detector are so arranged that a graph showing the relationship between the distance of the emitter from the layer and the counting rate includes a plateau region in which variation of said distance causes substantially no change in the counting rate, and wherein said predetermined distance between the layer and the beta emitter is in said plateau region.

2. Apparatus as claimed in claim 1, wherein a characteristic cross-sectional dimension (D) of a window through which said detector receives said backscattered radiation, and the diameter (d) of the intersection between the intensity distribution curve of the beta emitter and said layer are such that D/d is from 15 to 2.

3. Apparatus as claimed in claim 2, wherein D/d is from 10 to 4.

4. Apparatus as claimed in any preceding claim, wherein said window is provided by an outer ring or diaphragm disposed in front of the detector, which has a larger window.

5. Apparatus as claimed in any preceding claim, characterised in that the beta emitter comprises an absorbing tube which is open at one end for emitting radiation.

6. Apparatus as claimed in claim 5, characterised in that the tube is inclined relative to an axis of the detector and is directed at the axis and the layer.

7. Apparatus as claimed in claim 6, characterised in that n tubes are provided, where n is greater than 1, the tubes being displaced relative to one another.

8. Apparatus as claimed in claim 7, characterised in that the relative displacement of the tubes is 3600/n.

9. Apparatus as claimed in claim 7 or 8, characterised in that the angle of inclination to said axis is from 600 to 300, and preferably from 400 to 500.

10. Apparatus as claimed in claim 9, characterised in that the angles of all the tubes are equal.

11. Apparatus as claimed in any one of claims 7 to 10, characterised in that the emitters are all of the same type and the tubes all have the same parameters.

12. Apparatus as claimed in claim 5, characterised in that a plurality of tubes are arranged parallel to and around said axis, said tubes being spaced at varied intervals from said layer.

1 3. Apparatus as claimed in any one of claims 1 to 4, characterised in that the beta emitter comprises a surface emitter disposed around an axis of the apparatus with its radiation emitting surface directed at said axis and the layer.

14. Apparatus as claimed in claim 13 as appendent to claim 4, wherein said outer ring or diaphragm forms a carrier for said surface emitter.

1 5. Apparatus for measuring the thickness of thin layers on elongate material, substantially as herein described with reference to Figures 1 to 4, optionally as modified by Figure 5, Figure 6, Figure 7 or Figure 8, of the accompanying drawings.