DE102004042155B4 - Method for monitoring the layer thickness and the depth profile of the chemical composition of a coating of moving workpieces - Google Patents

Abstract

Method for monitoring the layer thickness of a coating (3a) of moving workpieces (3) by means of laser emission spectrometry, in which - a sequence of laser bursts is directed onto a surface of the coating (3a) with which at different points on the surface a plasma (3b) is generated in the coating (3a) with laser ablation, burst parameters being set for each laser burst of the sequence so that each laser burst generates a crater of a depth in the coating (3a) that corresponds to the Layer thickness exceeds, - an optical emission of the plasmas (3b) is detected and to determine a ratio of one or more spectral lines of elements and / or element compounds of the elements and / or element compounds of the coating (3a) is spectrally evaluated Ratio compared to a comparison value to detect a change in the layer thickness of the coating.

Description

Technical application

The present invention relates to a method for monitoring the layer thickness and / or the depth profile of the chemical composition of a coating of moving workpieces by means of laser emission spectrometry, which is particularly suitable for on-line monitoring of moving strip materials.

The online monitoring of the layer thickness and the elemental composition of a coating plays an important role, especially in the quality control in coil coating systems. An example is the process of hot-dip galvanizing steel sheets, in which both the maintenance of a given layer thickness of the zinc layer produced and the maintenance of a certain aluminum content on the surface of the zinc layer and in the transition region to the steel sheet an important role for the function and thus the Play quality of galvanized steel sheet. The results obtained with an online monitoring procedure can be used for optimized process control in the coating plant in order to meet the required quality criteria.

State of the art

In the field of coating thickness measurement and depth profile analysis, numerous different measuring methods are known, which can be divided into non-destructive and destructive methods.

Non-destructive optical methods based on diffraction, interference or ellipsometry can be used to characterize optically transparent coatings. Thickness determination is possible with these methods on single and multiple coatings with resolutions in the sub-micron range. By refractive index fluctuations or in combination with optical spectrometry, the chemical composition of a transparent coating can be determined within certain limits.

For layer thickness measurement and depth profile analysis of non-transparent layers, for example of metallic layers, techniques of X-ray fluorescence and X-ray diffraction are used in particular. However, the X-ray fluorescence technique is limited to high atomic number materials. A depth profile of the coating can not be determined with this technique. The X-ray diffraction technique makes use of X-ray diffraction at the crystal planes of the coating and the workpiece. By focusing the X-ray radiation on different depth ranges, it is thus also possible to achieve a depth resolution in the determination of the element concentrations.

In addition to these non-destructive techniques, destructive methods such as gravimetry, glow discharge-based optical emission spectrometry (GDOES) or laser emission spectrometry (LIES: Laser Induced Breakdown Spectrometry) are also known.

In the method of gravimetry, the coated workpiece is placed in a liquid solvent which dissolves the layer but does not attack the workpiece. Then the liquid is weighed. A comparison of the mass before and after the dissolution allows for a known size of the workpiece and density of the coating material to determine the average thickness. In addition, the average chemical composition of the coating can be determined. However, the method can only be used on a laboratory scale and is very time-consuming.

Glow-discharge emission spectrometry uses electrical gas discharges under reduced pressure in a noble gas atmosphere. To generate the glow discharge, a high voltage is applied between a hollow cylindrical electrode as the anode and the sample to be analyzed as the cathode. Electrons, which are triggered by the high voltage from the sample surface and accelerated in the direction of the anode, collide with the noble gas atoms and ionize them. The noble gas ions hit the sample surface and sputter off the surface of the sample. The ablated atoms or ions are excited by this ablation process and can be detected spectroscopically by their element-specific line radiation. With a known removal rate, both a chemical depth profile and the layer thickness can be determined. So far, however, this technique can also be used only in the laboratory.

Laser emission spectrometry represents another method for determining the layer thickness and the depth profile analysis of coated workpieces. In this technique, the coating is locally removed with the aid of laser pulses and the optical emission of the plasma generated on the surface of the coating is recorded and evaluated. As a rule, the element-specific line radiation of the plasma is detected individually for each individual pulse. Since the laser removes a small amount of material with each pulse, the laser penetrates ever deeper into the material, so that a chemical depth profile of the coating can be obtained. The amount of pulse energy and the size of the focus diameter can be used to set the ablation rate per pulse, which is on the order of 1 nm to 100 nm. At layer thicknesses of several micrometers, typically several 100 pulses of a pulsed laser are applied to a sample site. This procedure is therefore limited only to static samples, since a tracking of the laser beam with moving samples, especially in belt coating systems with belt speeds of a few m / s, technically can not be realized with the required accuracy. The focus and detection optics would have to be carried over several meters with micrometer accuracy.

From R. Sattmann et al. "Laser-Induced Breakdown Spectroscopy of Steel Samples Using Multiple Q-Switch Nd: YAG Laser Pulses" J. Phys. D: Appl. Phys. 28 (1995), 2181-2187, a technique of laser emission spectrometry is known in which laser bursts are directed instead of single pulses on a sample to be examined to produce the plasma. With this technique, an increased material removal compared to the use of single pulses is achieved.

The DE 41 38 157 A1 describes two variants of the method for determining the thickness of a coating by means of laser emission spectrometry. In the first variant of the method, laser pulses of the same energy are directed to the same point of the coating. Each laser pulse deepens the recess created by ablation at the measuring point. Upon detection of a significant change in the emission spectrum, the thickness of the coating is calculated from the number of previously irradiated laser pulses and their energy. This corresponds to a procedure as already explained in the penultimate section. In another variant of the method, in particular in moving workpieces, successive laser pulses whose pulse energy increases or decreases over time, are directed to different locations of the moving coating. Upon detection of a significant change in the emission spectrum, the energy of the laser pulse at which this change was detected is detected. From this energy, the thickness of the coating can be determined by comparison values. For larger layer thicknesses, however, this technique leads to a very large diameter of Ablationskraters due to the high pulse energies required, which is no longer tolerable for many applications.

From Stepputat et al., "Online Detection of Heavy Metals and Brominated Flame Retardants in Technical Polymers with Laser-Induced Breakdown Spectrometry", Appl. Optics 42, no. 30 (2003), 6210-6220, an application of laser emission spectrometry for the sorting of plastic waste or metal scrap is known. This is neither a determination of a layer thickness nor the detection of a depth profile of the chemical composition but only the detection of certain elements in an object. The laser is directed via an autofocus system onto the surface of the objects moving on a conveyor belt, the distance for the correct laser focusing being determined by means of a laser triangulation device.

The US 2003/0095266 A1 discloses a method for determining the layer thickness and for detecting the depth profile of the chemical composition of a coating of a heterogeneous material by means of laser emission spectroscopy. In this method, a sequence of laser pulses is applied to this site to determine a depth profile at a location of the workpiece and each optical emissions of the resulting plasma spectrographically analyzed. By recording such depth profiles at several points of the surface of the workpiece, a three-dimensional detection of the element distribution in the near-surface region of the workpiece is possible.

The US 6,532,068 B2 shows a method for detecting the depth profile of the chemical composition of a coating of workpieces by means of laser emission spectroscopy. In the method, a crater is first produced in the workpiece with a first laser pulse or laser burst. Subsequently, a plasma is generated with a second laser pulse or laser burst of smaller diameter by laser ablation of the crater floor, whose optical emissions are detected and analyzed. To detect a depth profile of the chemical composition of a coating, this procedure is repeated often at the same point of the coating.

The object of the present invention is to provide a method for monitoring the layer thickness and / or the depth profile of the chemical composition of a coating of moving workpieces, which also allows the monitoring of thicker coatings without significant ablation damage.

Presentation of the invention

The object is achieved by the method according to claims 1 and 2. Advantageous embodiments of the method are the subject of the dependent claims or can be found in the following description and the exemplary embodiments.

In the method for monitoring the layer thickness of a coating of moving workpieces by means of laser emission spectrometry, a sequence of laser bursts is directed onto a surface of the coating with which a plasma is produced at different points on the surface of the coating by laser ablation. The different locations arise due to the movement of the workpieces relative to the laser beam. In the method, the burst parameters for each laser burst of the sequence are adjusted so that each laser burst produces a crater of depth in the coating, which in each case is greater than the thickness of the coating. The optical emission of the plasmas generated by the different laser bursts is detected and evaluated spectrally to determine a ratio of one or more spectral lines of elements and / or element compounds of the workpiece to one or more spectral lines of elements and / or element compounds of the coating Change of this ratio compared to a comparison value to detect a change in the layer thickness of the coating.

In this case, the comparison value may be a value determined in advance with identical burst parameters at a reference workpiece or the value determined with a laser burst having identical burst parameters with identical burst parameters in chronological order.

Preferably, the burst parameters for all laser bursts of the sequence for the layer thickness monitoring are selected to be identical. In this case, even without a comparison value determined in advance on a reference workpiece, a change in the layer thickness during the method can be detected. For a quantitative determination of the layer thickness, however, comparative values are required, which must be determined in advance experimentally with the selected burst parameters at different known layer thicknesses of the reference workpiece.

In a further embodiment of this method, different burst parameters can also be set as a consequence for at least some of the laser bursts. In this case, for each adjustment of the burst parameters corresponding reference values of a reference workpiece with the desired layer thickness are required in order to be able to detect a change in the layer thickness during the process by deviation from the respective comparison value. The same applies to the aforementioned embodiment of identical burst parameters, if the deviation from a desired layer thickness should also be detected there.

In the present method for monitoring the layer thickness and / or, as set forth in any of the following sections, for monitoring the depth profile of the chemical composition of a coating of moving workpieces, the monitoring or analysis by means of a series of individual laser bursts, the on different locations of the coating are applied. In the present patent application, a laser burst is a sequence of n short consecutive individual pulses, where n ≥ 2, which does not exceed an overall time length of 1 ms. The burst energy E _B is distributed over the n individual pulses with pulse durations between 1 fs and 100 μs. The distance between the individual laser pulses within a laser burst is selected such that the gas-dynamic state generated by the plasma from laser pulse to laser pulse persists within the laser burst. Such laser bursts can be generated in particular with Q-switched, pulsed lasers in which the laser bursts are emitted at the repetition rate of the laser.

By optimizing the burst parameters, in particular the number of pulses of a laser burst, the focusing and the burst energy, the depth of the laser burst generated crater in the coating is adjustable. By adjusting the depth, the analysis range, in particular when monitoring the depth profile of the chemical composition, can thus also be selected. During a measurement cycle comprising a sequence of laser bursts, the burst parameters become either, depending on the application quasi-continuously changed or kept constant or different constant burst parameters are selected.

Thus, for monitoring the layer thickness of a coating of moving workpieces, in particular of moving strip material of a strip coater, individual bursts of a fixed energy or of a set of discrete burst energies are used. The burst parameters are adapted to the respective coating material so that the crater depth created by a single burst has at least one location which is greater than the maximum layer thickness value to be measured. In this way, the plasma generated by the burst always contains a proportion of spectral lines of the elements and / or element compounds of the workpiece and a proportion of spectral lines of the elements and / or element compounds of the coating. If the layer thickness remains constant from laser burst to laser burst, the intensity ratio of the spectral lines of the coating and of the workpiece does not change with the same burst parameters. However, if the layer thickness changes, this is manifested by an increase or decrease in the intensity of the spectral lines of elements and / or element connections of the workpiece compared to those of the coating. By monitoring this ratio, a change in the layer thickness of the coating can thus be detected immediately.

The detection of a change in the layer thickness can of course also be done by comparing the intensities of one or more spectral lines and / or one or more ratios of spectral lines with comparison or calibration data, which were determined in advance on the basis of a coated reference workpiece.

In the method variant for monitoring the depth profile of the chemical composition of a coating of the moving workpieces by means of laser emission spectrometry, a series of laser bursts is likewise directed onto the surface of the coating in order to apply plasma to the surface of the coating under laser ablation produce. The burst parameters of the laser bursts are adjusted so that at least some of the laser bursts of the sequence produce craters of different depths in the coating. Again, the optical emission of the plasmas is detected and evaluated spectrally this time to determine one or more portions of elements and / or element compounds in different depths of the coating, which are determined by the different depth of the respective craters and thus the different burst parameters.

In this process variant, differently deep craters are produced by laser bursts with different burst parameters, so that the measured values obtained from each deliver different depth information about the elements or element connections contained therein. By a quasi-continuous increase or decrease of the crater depth over the sequence of laser bursts by appropriate adjustment of the burst parameters, a complete depth profile of the chemical composition can thus be obtained. The method also makes it possible to monitor not a complete depth profile but only the elemental composition at particular depths by comparison with corresponding comparison values during the process.

Here, too, comparison or calibration data are preferably used, from which the crater depth can be derived for predetermined burst parameters and optionally element compositions.

By using laser bursts instead of single pulses, a high aspect ratio of the craters is achieved at a high crater depth. In this way, larger layer thicknesses can be analyzed or monitored with the present methods for smaller crater diameters than in the case of the single-pulse technique cited in the introduction to the description.

The two process variants of the present method can also be advantageously combined with each other, so that a simultaneous determination of deviations in chemical depth profiles and in the layer thickness for on-line monitoring of coating systems is achieved. In this case, for example, laser bursts may be alternately directed to achieve sufficient depth of crater for layer thickness monitoring and laser bursts for lower crater depths to monitor the depth profile on the surface of the coating.

The present method is particularly suitable for the on-line control of the layer thickness and the elemental composition in a coil coating plant, in particular for nontransparent layers. In particular, coatings in which the process-related microscopic changes, ie the generated craters, the function of the coating or the coated workpiece or only slightly influence. An example of this is the zinc layer on galvanized steel sheets, in which microscopic damage due to the cathodic action at a distance does not lead to any impairment of the corrosion protection. A reduction in quality can also be avoided by carrying out the measurement in areas which are not included in any subsequent processing of the material. The results obtained online can be advantageously used to control the coating process in the coating system in order to ensure defined, uniform layer thicknesses and a defined chemical depth profile of the layers.

Brief description of the drawings

The present method will be explained again by way of example with reference to embodiments in conjunction with the drawings without limiting the scope defined by the claims. Hereby show:

1 an example of a structure for performing the laser emission spectrometry;

2 an intensity-time diagram for illustrating a laser burst according to the present invention;

3 an example of the layer structure of a hot-dip galvanized steel sheet;

4 an example of a depth profile of a hot-dip galvanized steel sheet;

5 a representation for illustrating the monitoring of the layer thickness according to the present method;

6 a representation to illustrate the choice of Ablationstiefe;

7 an example of a calibration curve for layer thickness calibration;

8th a representation for illustrating the depth profile analysis according to the present method;

9 an exemplary representation of the integral N (E _B );

10 an example of possible Al depth profiles in the zinc layer of a hot dip galvanized sheet;

11 an example of the qualitative behavior of the ratio I _{Ist, Al} / I _{Soll, Al} as a function of the burst energy E _B ; and

12 an exemplary representation of a measuring device for carrying out the present method.

Ways to carry out the invention

1 shows an example of a typical structure for performing the laser emission spectrometry on a workpiece 3 , In the measurement, the laser beam 2 a pulsed laser 8th For example, a Q-switched Nd: YAG laser, via a dichroic mirror 4 (highly reflective for the laser wavelength, transparent for the element-specific line radiation) with the focusing optics 1 on the surface of the coating 3a of the workpiece 3 focused. With a suitable choice of the focus diameter and the energy of the laser pulses is on the surface of the coating 3a a plasma 3b generated. The element-specific line radiation of the plasma passes through the dichroic mirror 4 and comes with an imaging optics 5 on a waveguide 6 imaged the plasma light to the entrance slit of the spectrometer 7 forwards. In this way, for each laser pulse or laser burst plasma emission with the spectrometer 7 be evaluated. Different pulse energies make it possible to achieve different crater depths of the laser at the surface of the coating 3a or the workpiece 3 produce generated craters, so that at different depths optionally different material compositions contribute to the line spectrum.

2 shows an example of a laser burst, as used in the present method. This laser burst is understood to mean a sequence of successive, closely spaced laser pulses P ₁ ... P _n , which have a distance of ≦ 1 ms from each other and a pulse duration of 1 fs to 100 μs each. The total duration of the laser burst does not exceed 1 ms. This can thus be a pulse train of numerous laser pulses or just a double pulse. The individual laser pulses of a laser burst are indicated in the figure by P ₁ to P _n . They can have different intensities and pulse durations within the laser burst. Such a laser burst can be generated, for example, by a flash-lamp-pumped Nd: YAG laser, wherein such a laser burst is emitted by the flash lamp during each pumping operation. A particular characteristic of the laser bursts used in the present method is that the plasma generated by the temporally first pulse P _{1 of} the laser burst generates a gas-dynamic state that persists between the individual pulses of the laser burst, so that the plasma during the total duration of the laser burst is maintained.

In the following, the present method for on-line layer thickness measurement and online depth profile analysis of coated strip material will be explained in more detail on the basis of the industrially significant application of the production of hot-dip galvanized steel sheets. 3 shows the schematic structure of the zinc layer of a hot-dip galvanized steel sheet 3 , In hot-dip galvanizing, in addition to the main element zinc, low-concentration aluminum (<1%) is also added to the zinc bath. This forms between the steel substrate 3 and the zinc layer, an approximately 100 nm thick intermediate layer of Al ₂ Fe ₅ , which prevents the diffusion of iron into the zinc layer and ensures the adhesion of the zinc layer. At the same time, a thin Al ₂ O ₃ layer forms on the surface of the zinc layer, as shown in the 3 can be seen.

4 schematically shows a depth profile of this layer system, as it can be obtained with the present method. Evident is the increased Al concentration at the surface and in the transition region between layer and substrate. The layer thickness d _s is the point at which the zinc concentration has fallen to 50% of the initial value or the iron concentration has reached 50% of its final value. The layer thickness d _s is typically in the range of 5 to 15 microns for industrially manufactured sheets.

5 shows a representation for illustrating a process variant of the present method for on-line monitoring of the layer thickness of the zinc layer of a hot-dip galvanized sheet. By varying the burst parameters, such as the burst energy generated by the laser 8th emitted laser bursts, the diameter and depth of the craters produced in the coating can be set in a wide range. This makes it possible to produce craters with a high aspect ratio, ie a small diameter at great depth. Compared to single pulses, it is thus possible to increase the depth by more than a factor of 10 while reducing the crater diameter. Compared to a process procedure in which the laser ablation and plasma generation takes place by individual pulses spaced far apart from one another, between which the plasma extinguishes, the use of the laser bursts greatly reduces the damage occurring on the strip material in the present process. The use of laser bursts thus enables measurements on coatings of several micrometers thick.

The burst parameters are selected to monitor the layer thickness so that the ablation depth per laser burst is greater than the thickness of the layer to be monitored. This ensures that the plasma emission, hereinafter also referred to as LIBS signal, in addition to line emissions of the layer elements zinc and aluminum and iron emissions from the steel substrate contains. The burst parameters here are, for example, the pulse number of the burst, the burst energy, the energy distribution within the burst or the focusing. If the burst parameters are kept constant, the layer thickness changes, as in the 5 indicated, so also the intensity of the element lines of the layer or substrate material changes. With a larger layer thickness results in a smaller intensity of the substrate and a greater intensity of the laminar element lines. With smaller layer thickness results in the reverse relationship. The ratio of at least one substrate line intensity I _Subs to a line intensity I _layer of the coating is therefore a measure for the layer thickness d _s.

When setting the burst parameters, it should be noted how the volume of the ablated layer material V ₁ relates to the volume of the ablated substrate material V ₂ . V ₂ must be greater than 0 so that the LIBS signal always contains line emissions from the substrate. However, V ₂ must not be chosen too small, since otherwise, with a larger increase in the layer thickness, only signals from the layer may be measured. On the other hand, V ₂ must not be too large to be smaller variations in the Be able to dissolve layer thickness. Examples of different ratios between V ₁ and V ₂ are in the 6 illustrated.

In addition, the burst parameters should be optimized so that the smallest possible crater diameters 2R and at the same time sufficient measurement signal are obtained with regard to possible impairments of the layer. R denotes the radius of the crater on the layer surface. In order to fulfill these criteria, V ₂ should be chosen approximately in the range 0 <V ₂ ≦ V ₁ and the depth of the crater h in the range d _s <h ≦ 2 d _s .

By averaging over several measurements and involving several element lines, statistical fluctuations in the measuring process can be reduced. When using a plurality of element lines, for example, the sum of a plurality of line intensities of the substrate I _{subs, i can be} referenced to the sum of a plurality of line intensities of the coating I _{layer, j} . By introducing weighting factors or exponentiating certain line intensities, a different weighting of individual element lines can be achieved so that the referenced substrate signal F is, in the general case, a ratio of two functions: F = f _Subs (I _{Subs, I} ... I _{Subs, n} ) / f _Layer (I _{Layer, l} ... I _{Layer, m} ) (1)

At the same time, it is possible to set the number and the distance of the measuring points on which length scale is averaged. To determine the absolute layer thickness, a prior calibration of F with reference samples of different layer thicknesses is necessary, ie the dependence of the referenced substrate signal on the layer thickness must be determined beforehand with the specified burst parameters. This calibration curve is integrated into the evaluation process so that the layer thickness can be determined directly from the line ratio measured online. The qualitative course of such a calibration curve is in 7 shown. The result is a monotone decreasing curve with increasing layer thickness.

Furthermore, it is possible to use several sets of discrete burst parameters, which can be used either to determine the layer thickness from a set of individual calibration curves or to include several parameter sets simultaneously in the calibration model using multivariate methods.

In the continuous coating of strip materials, in addition to monitoring the layer thickness, it is also of great interest to control the chemical composition or the depth profile of the layer. For example, in the case of hot-dip galvanized sheet steel, the aluminum content determines the weldability of the sheet and the thickness of the Al ₂ Fe ₅ layer determines the adhesion of the coating. With the present method changes in the depth profile and in the chemical composition can be detected. In this case, the ablation depth is changed quasi-continuously during a measurement cycle by suitably changing the burst parameters, as is shown by the illustration of FIG 8th is illustrated with double pulses. If, for example, the burst energy is increased from laser burst to laser burst, the LIBS signal increasingly receives more contributions from greater depth. By comparing the line intensities measured as a function of the burst energy with those of a desired curve, which was determined in advance on a reference sample, deviations in the composition can be recognized immediately.

• – I_Element,i (E_B) ist näherungsweise proportional zu der im ablatierten Kratervolumen V(E_B) vorhandenen Zahl der Atome des Elementes i N_Element,i(E_B). Um Änderungen der Plasmaparameter, wie z. B. die Größe des Plasmas oder die Plasmatemperatur, in Abhängigkeit der Burstenergie mit zu berücksichtigen, wird eine energieabhängige Korrekturfunktion f_Korr(E_B) anstelle einer Konstanten eingeführt, die I_Element,i(E_B) mit N_Element,i(E_B) verknüpft: I_Element,i(E_B) = f_Korr(E_B)·N_Element,i(E_B) (2)
• – Die Abhängigkeit der Querschnittsfläche A(E_B, z) des Kraters parallel zur Schichtoberfläche von der Burstenergie und der Tiefe z sind bekannt oder werden vor oder nach der LIBS-Messung beispielsweise mit einem Weißlichtinterferometer an einer Referenzprobe experimentell bestimmt. Insbesondere sind dadurch der Verlauf der Kratertiefe h(E_B) und des Radius R(E_B) des Kraters an der Schichtoberfläche als Funktion von E_B ebenfalls bekannt.
• – Die Teilchendichte des Elements i hängt nur von der Tiefe z ab und nicht von der x- oder y-Koordinate. Somit lässt sich die Teilchendichte n_i des Elements i schreiben als: n_i = n_i(z)

The relationship between the depth profile and the LIBS signal is explained in more detail below. First of all, it shall be shown which depth information contains the line intensity I _{element, i} (E _B ) of an atomic species i measured as a function of the burst energy. The following assumptions are made:

• - I _{element, i} (E _B) is approximately proportional to the volume V in the ablated crater (E _B) existing number of atoms of element i N _{element i} (E _B). To change the plasma parameters, such. As the size of the plasma or the plasma temperature to be considered in dependence of the burst energy, an energy-dependent correction function f _Korr (E _B ) is introduced instead of a constant, the I _{element, i} (E _B ) with N _{element, i} (E _B ) connected: I _{element, i} (E _B ) = f _corr (E _B ) · N _{element, i} (E _B ) (2)
• The dependence of the cross-sectional area A (E _B , z) of the crater parallel to the layer surface of the burst energy and the depth z are known or are determined experimentally before or after the LIBS measurement, for example with a white light interferometer on a reference sample. In particular, the course of the crater depth h (E _B ) and of the radius R (E _B ) of the crater on the layer surface are also known as a function of E _B.
• - The particle density of the element i depends only on the depth z and not on the x- or y-coordinate. Thus, the particle density n _{i of} the element i can be written as: n _i = n _i (z)

Im Integral steht das Produkt aus der Funktion A(E_B, z), die der Querschnittsfläche des Kraters an der Stelle z entspricht, mit dem Tiefenprofil n₁(z) des Elements i. Under these assumptions, the number of atoms of an element in the ablated crater volume can be written as a function of the burst energy as:

In the integral represents the product of the function A (E _B, z) corresponding to the cross-sectional area of the crater at the point z, with the depth profile n ₁ (z) of the element i.

In the event that the Kratervolumen V (E _B ) is approximately conical (see. 6 ), which can be assumed when using lasers with Gaussian beam profile, A (E _B , z) results in:

In 9 By way of example, the depth profile n _Al (z) of Al in a hot-dip galvanized sheet and the course of A (E _B , z) for two different sized burst energies in conical crater geometry are shown. It can be seen that N _{element, Al} (E _B ) increases monotonically with increasing E _B , because on the one hand the crater diameter 2R (E _B ) increases and on the other hand the Al of the Al ₂ Fe ₅ layer also contributes to higher burst energies.

The line intensity I _{element, i} (E _B ) is linked via equation (2) with the particle number N _{element, i} (E _B ). Substituting (3) in (2) provides:

The measured profile of I _{element, i} (E _B ) thus contains the chemical depth profile n _i (z) of the element i. The profile of I _{element, i} (E _B ) determined on a reference sample having a known depth profile n _{setpoint, i} is referred to below as setpoint curve I _{setpoint, i} (E _B ). From n _{Soll, i} (z), the z. As can be determined by means of GDOES-measurement, and A (E _B, z), can be combined with (3) N _{reference, i} (E _B) Calculate _B for each E and with the help of (2) the correction function f Determine _corr (E _B ).

The aim of the online control is to detect deviations in the depth profile n _{actual, i} (z). If the chemical composition changes, ie n _{ist, i} (z), the course of I _{ist, i} (E _B ) also changes.

bildet und diese Integralgleichung numerisch nach n_{Ist , i}(z) auflöst.Assuming that the ablation behavior does not change due to the changed chemical composition of the layer, the depth profile of n _{Ist, i} (z) can be determined by

and this integral equation numerically resolves to n _{Ist , i} (z).

In order to be able to recognize qualitative changes in the depth profile, however, it suffices to consider the quotients I _{actual, i} (E _B ) / I _{nominal, i} (E _B ) as a function of E _B. Thus, in the case of hot-dip galvanized sheets, it is often sufficient to decide whether the Al ₂ Fe ₅ layer is present or not. 10 shows qualitatively three possible courses of n _Al (z). The solid line shows the target profile of n _Al (z), the dotted line the case that the thickness of the Al ₂ Fe ₅ layer increases, and the dotted line the case that the Al ₂ Fe ₅ layer did not form. A thicker Al ₂ Fe ₅ layer causes the ratio I _{actual, i} (E _B ) / I _{target, i} (E _B ) to bend upward as the layer energy E _{B increases} , while if Al ₂ does not exist Fe _5- layer bend down the curve, see 11 ,

When performing the measurement, it is advisable to use a burst sequence as in 8th to use shown, ie each perform a set of individual measurements with the same burst energy at different positions and to determine an average value for this burst energy. This is in 8th , shown lower left for times t ₁₁ to t _1i . The setting of the burst energy can be via an external energy attenuator ( 9 ), the z. B. consists of a combination of a Pockels cell with a polarizer. The averaging range, ie the distance that the tape material travels during a measurement cycle, is determined by the tape speed, the laser's repetition rate, and the number of measurement bursts. By adjusting the repetition rate and the number of measurement bursts, the averaging range can be set variably depending on the measurement task.

Variations in the measuring process can be reduced by suitable referencing. Thus, in the example presented here, it is appropriate to refer I _{to Al} on a Zn line of the main element of the layer. If, in addition to the line intensities of the layer elements, those of the substrate are also measured in parallel, the layer thickness can be determined simultaneously, as already described above.

12 finally shows an example of an embodiment of a measuring apparatus for carrying out the present method. Belt materials are usually moved in industrial manufacturing processes on different pulleys, whereby the belt performs vibration movements. For a static design of the focusing optics 1 according to 1 the focus position is subject to statistical fluctuations relative to the band surface. In order to compensate for these fluctuations, in the present measuring apparatus according to 12 an autofocus system integrated. The measuring arrangement consists of a LIBS structure with an autofocus system based on a distance measurement by means of a laser triangulation sensor 10 and a refocusing by means of a telescope 11 works with variable lens spacing. The laser beam 14 the triangulation sensor 10 is through the two dichroic mirrors 4 . 4a adjusted so that it is central and vertical by the short focal length focusing optics 1 runs. On the surface of the layer, a scattered leak with a Gaussian intensity distribution is created on the CCD chip of the triangulation sensor 10 is shown. From the position of the image by triangulation of the distance of the belt to the triangulation sensor 10 certainly. The second component of the autofocus system consists of the telescope 11 , whose diverging lens can be adjusted motorized, whereby the beam divergence and thus the focus position can be readjusted.

During the LIBS measurement, the distance from the lens to the tape is continuously measured and the value to a PC 12 transfer. A control software calculates the extent to which the diverging lens of the telescope is calculated by means of a previously performed distance calibration 11 has to be moved in order to keep the focal position constant relative to the layer surface.

Additional fluctuations in the LIBS signal may result from variations in the laser burst energy. In order to detect these fluctuations and not to be included in the analysis or measurement too strong deviation from a target value, a monitoring of the single-burst energy can take place. For example, a photodiode is available for this purpose 13 behind a semitransparent mirror in the beam path of the laser 8th is placed.

LIST OF REFERENCE NUMBERS

1

focusing optics

2

laser beam

3

workpiece

3a

coating

3b

plasma

4, 4a

dichroic mirror

4

imaging optics

5

optical fiber

6

spectrometer

7

laser

8th

Energieabschwächer

9

laser triangulation sensor

10

telescope

11

PC

12

photodiode

13

Laser beam of the laser triangulation sensor

P _i

Pulse of a laser burst

Claims (8)

Method for monitoring the layer thickness of a coating ( 3a ) of moving workpieces ( 3 ) by means of laser emission spectrometry, in which - a sequence of laser bursts on a surface of the coating ( 3a ) with which at different points on the surface of the coating ( 3a ) under laser ablation a plasma ( 3b ) - where burst parameters are set for each laser burst of the sequence so that each laser burst will have a crater of a depth in the coating ( 3a ), which exceeds the layer thickness, - an optical emission of the plasmas ( 3b ) and for determining a ratio of one or more spectral lines of elements and / or element connections of the workpiece ( 3 ) to one or more spectral lines of elements and / or element compounds of the coating ( 3a ) is spectrally evaluated, - in order to detect a change in the coating layer thickness when the ratio changes compared to a comparison value. Method for monitoring the depth profile of the chemical composition of a coating ( 3a ) of moving workpieces ( 3 ) by means of laser emission spectrometry, in which - a sequence of laser bursts on a surface of the coating ( 3a ) with which at different points on the surface of the coating ( 3a ) under laser ablation a plasma ( 3b ) - where burst parameters of the laser bursts are adjusted so that at least some of the laser bursts of the sequence have craters of different depths in the coating ( 3a ), and - an optical emission of the plasmas ( 3b ) and for determining one or more fractions of elements and / or element compounds of the coating ( 3a ) in different depths of the coating determined by the depth of the respective craters ( 3a ) is spectrally evaluated. A method according to claim 2, characterized in that the burst parameters of the laser bursts are adjusted so that at least some of the laser bursts of the sequence have craters of a depth in the coating ( 3a ), which exceeds the layer thickness, whereby the optical emission of the plasmas ( 3b ) also for determining a ratio of one or more spectral lines of elements and / or element connections of the workpiece ( 3 ) to one or more spectral lines of elements and / or element compounds of the coating ( 3a ) is spectrally evaluated in order to detect a change in the layer thickness of the coating when the ratio changes compared with a comparison value. A method according to claim 2 or 3, characterized in that for monitoring the depth profile deviations from comparison values are detected, which are obtained from calibration measurements on reference workpieces. A method according to claim 2 or 3, characterized in that to monitor the depth profile deviations from comparison values are detected, which are obtained from results of previous laser bursts with identical burst parameters. A method according to claim 1, characterized in that the comparison value is obtained from a calibration measurement on a reference workpiece. Method according to Claim 1, characterized in that the comparison value is obtained from the results of previous laser bursts having identical burst parameters. A method according to claim 1, 6 or 7, characterized in that the burst parameters are set identically for each laser burst.

Images