摘要:

JOURNAL OF ANALAYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL. 7 1029 Applications of Laser-induced Emission Spectral Analysis for Industrial Process and Quality Control* Claus J. Lorenzen Christoph Carlhoff Ulrich Hahn and Martin Jogwich Krupp Forschungsinstitut GmbH Postfach 70 22 52 0-4300 Essen 7 Germany Laser-induced emission spectral analysis (LIESA a registered trademark of instruments developed by Krupp) better known in the literature as laser microanalysis or laser-induced breakdown spectroscopy is a suitable method for the direct in-process measurement of elemental concentrations in various solid and liquid materials. This method has been developed recently by Krupp for in-process quality assurance and process control in different industrial branches such as steel production ahd plant making.As a result several LIESA instruments have already been developed or are under development for marketing. In all cases on-line and in-process elemental analysis of materials at various stages of production yield information on the quality of the material and the fabrication process. The beam of a pulsed high-power laser (irradiance 1 x 108-5 x lo9 W cm-2) focused onto the solid or liquid sample surface in an ambient gas atmosphere of normal pressure (focus areazablation area 0.1 -6 mm2) produces a hot bright plasma (early electron temperatures 20 000-30 000 K). The emitted plasma light is observed end-on and passes by way of an optical fibre bundle to a spectrometer where it is detected in the focal plane by means of an optical multichannel analyser with high time resolution (on the microsecond scale).A fast computer evaluates the measured spectra and calculates the element concentrations via calibration procedures. Relative detection limits of between 10 and 100 ppm can be achieved for most of the detectable elements in various matrices (steel rubber rock and glass). Procedures are available to convert relative measurements with relative standard deviations of between 1 and 2% into absolute concentration values with relative accuracies of about 3%. Keywords In-process laser microanalysis; remote surface analysis; depth profiling; laser ablation; optical emission spectrometry Laser ablation of solid or liquid samples and optical emission spectrometry of the microplasma produced is a simple and fast technique for direct elemental analysis.Time-consuming sample preparation can be omitted. Bulk and surface analysis in addition to depth profiling are possible with the same apparatus. Tight focusing of the laser beam even allows microanalysis of sample surfaces. Re- cently two have been published briefly describ- ing the historical background and the present status of scientific investigations and applications. In the period 1989- 1992 Niemax and c o - ~ o r k e r s ~ - ~ published a series of papers presenting systematic investigations of the spatial and temporal evolution of laser-produced microplasmas and ablation of the related material. By using a Q-switched Nd:YAG laser operating at its fundamental wavelength ( 1 064 nm) Niemax and co-workers found the following experimental conditions and parameters to be best-suited to laser-induced emission spectral analysis (i) use of a buffer gas at reduced pressure (most appropriate argon at 140 hPa); (ii) use of reduced laser irradiance (no electrical breakdown in the surrounding atmosphere when the sample is removed); (iii) use of long delay times ( 2 3 0 p s ) between laser pulse and detector gate pulse; and (iv) use of analyte and reference spectral lines with comparable excitation energies.A buffer gas plasma formed simultaneously with the ablation process serves as an energy reservoir for the atomization of ablated droplets and particles and for the excitation of atoms and ions. The heating of the buffer gas plasma by inverse bremsstrahlung is most effective with argon.Since a buffer gas keeps ablated material in the observation region for a longer time and a larger fraction of this material can penetrate into the buffer gas plasma at reduced pressure best analytical results were achieved with argon at 140 hPa. Recombination of the buffer gas plasma *Presented at the 1992 Winter Conference on Plasma Spectro- chemistry San Diego CA USA January 6-1 1 1992. is completed after approximately 1 ps. At this time ablated material from the sample surface reaches the centre of the buffer gas plasma at almost supersonic velocities. Atomiza- tion and excitation processes start to develop. Complete atomization is achieved after 20-30 ps with number densities of free atoms and ions and plasma temperatures remaining high (>6000 K).The buffer gas-sample material plasma can be described by partial local thermo- dynamic equilibrium verified by plasma temperature mea- surements using Boltzmann plots. Internal standardization can be applied and matrix independent measurements (common calibration graphs) can be performed when delay times greater than 30 ps and analyte and reference spectral lines of comparable excitation energies are used in optical emission spectrometry. In 1986 Krupp started with basic investigations of the laser analysis method applied to solid and liquid steel samples (Carlhoff et A pilot system for on-line and in- process carbon and temperature monitoring was developed and successfully operated in a Krupp steel plant (Carlhoff and Kirchhoff'O). This was the first successful attempt worldwide to install a laser-induced emission spectral analysis (LIESA) system in the harsh environment of a production steel plant.Meanwhile Krupp started the development of a laboratory prototype for the in-process monitoring of element distributions in polymeric materials (Lorenzen et a/."). Further activities in this field are the on- line control of geological raw materials on conveyor belts and investigations of depth profiling of multi-layers on metallic substrates. The following sections describe the basics and some details of the different industrial LIESA applications at Krupp. Emphasis is placed mainly on instrument develop- ment rather than on fundamental scientific investigations. Instrumentation The general set-up of all LIESA instruments is shown in Fig.I . In some applications where elemen1 distribution measurements are performed a linear scanner system is1030 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL. 7 Table 1 Hardware components and ranges of operating conditions of LIESA instruments Q-switched Nd YAG laser- Wavelengthhm Mode Pulse energy/mJ Pulse widthlns Repetition rate/Hz or Excimer laser- Wavelengthhm Pulse energy/mJ Pulse durationh Repetition rate/Hz Optical multichannel analyser- 1064 or 355 Multimode 25-300 (on sample surface) 6-9 10-25 248 (IGrF) 50-1 10 (on sample surface) 26 15-50 Detector Gated MCP-intensified diode array (1024 pixel UV enhanced) Sensitive wavelength rangehm 150-900 Wavelength capture rangehm (see spectrometer) 14-42 (in the UV) Gate widthdps 0.2-10 Delay times (laser gate)/ps 0-30 Data transfer speedspectra per second Maximum 50 Spectrometer- Mounting Focal lengthlm Gratinglrules per mm Reciprocal dispersionhm mm-I Entrance slit-widthlpm Linear scanner system- Scan range/m Scan velocity/m s-' Scan acceleration/m s-* Load/kg Czerny-Turner 0.5 cf/4) 3 600 or 2 400 0.6 or 1.7 (in the UV) =s 80 0-1.2 0-2 0-6 20 Pu Laser plasma (material ablation.1pr light emission) Sample S ,P ipectrom Grating Computer Array detector Fig. 1 General set-up of LIESA instruments employed to scan the laser beam across the sample surface. The specifications for the hardware components and the operating conditions are summarized in Table 1. Since the operating parameters are dependent on the application of the instrument parameter ranges are given in most in- stances for reasons of simplicity.When designing a LIESA instrument for a specific application special attention has to be paid to the selection of the appropriate laser system. Commercial Nd;YAG and excimer lasers were found to be most suitable for operation in industrial environments owing to their sturdiness and ease of operation. High pulse-to-pulse stability of both lasers [about 2% relative standard deviation (RSD)] and long lifetimes of the flashlamp of the Nd:YAG laser and of the gas fills and resonator mirrors of the excimer laser (up to 5 180 181 182 183 184 185 186 z Wavelengthhm Fig. 2 Comparison of detector signals in the spectral region 180-186 nm obtained with use of conventional optics (upper trace) and 3 m fibre optics (lower trace) from a rubber sample.The sulfur concentration is 1.12% 1 x lo8 laser pulses in both cases) are additional features making these lasers useful for industrial applications. In the next step the appropriate operating laser wave- length is most important to obtain the best analytical results. It was found that the 1024 nm Nd:YAG laser produces plasmas with strong emission on metallic and glass surfaces. Ultraviolet (UV) excimer laser wavelengths must be employed on polymeric surfaces since only then are sharp and regular ablation patterns produced without any thermal side-effects. For geological samples infraredJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL. 7 1031 iX1o7 1x106 iX1o5 iX1o4 A cn C 4- .- = 1000 L.I 4? 100 4- .- I u ~ 2 0.2 0.5 1 2 5 10 % cn C a) c .- 4- .E 1x106 Y 0) Q C -I a 5x105 .- 2x105 iX1o5 5x1 0' 2x10' De I a y ti me/p s C 0.1 1 10 Gate wi dt h/p s 100 Fig. 3 (a) Double-logarithmic plot of peak intensities of the lines A Si I 288.2 nm; B Ca I1 393.4 nm; and C Al I 396.2 nm versus delay time. The gate width is always 200 ns. A rubber sample was used as the target. (b) Double-logarithmic plot of peak intensities of the lines A Mg I1 279.6 nm; B Mg I 285.2 nm; and C Si I 288.2 nm versus gate width. The delay time is always 200 ns. A rubber sample was used as the target (IR) as well as UV laser light give comparable analytical results. Another key component in the LIESA instruments are the optical fibres used for guidance of the plasma emission light to the spectrometer system thus greatly simplifying the opto-mechanical design and adjustment of the apparatus.The fibre output is directly coupled to the entrance slit of a 0.5 m spectrometer. It can be seen from Fig. 2 that the transmission of the fibre guide near the sulfur triplet rapidly decreases with decreasing wavelength (lower trace of Fig. 2) compared with the spectrum taken with direct lens imaging (upper trace of Fig. 2). The beam paths and spectrometer are purged with Ar and N2 respectively. In addition the cut-off transmission wavelength at about 180 nm is also governed by limited detector window transmission and detector sensitivity in addition to limited transmission through residual oxygen in the purged beam paths. Transmission of nanosecond laser pulses through optical fibres with high peak power densities (>0.1 GW cm-2) causes problems even with core diameters of up to 2 mm.As described below in some instances element distribu- tion measurements are performed. Here the laser beam is scanned across the sample surface by means of an optical system carried by a high speed scanning machine (single axis linear positioning stage). This industrial device is designed for long life under high duty cycle and continuous operating conditions. Linear drive actuation is effected by a brushless linear d.c. servomotor. Precision Dosition feed- back is achieved by an optical incremental linear encoder (for specifications see Table 1). Data Acquisition and Evaluation The main component from the analytical point of view is the detection system connected to a host computer.The plasma emission light is detected with high time resolution. It is necessary to create a trigger signal coincident with the nanosecond laser pulse in order to start the delay and gate generator in the optical multichannel analyser. In the case of the Nd:YAG laser this trigger signal is directly available from the Q-switch electronics of the laser power supply whereas in the excimer laser a fast UV sensitive photodiode mounted behind a laser beam aperture creates the trigger signal. A constant delay due to electrical propagation differences in the detector electronics always has to be taken into account when calculating the true delay time. As the laser analysis is restricted to atmospheric pressures in most instances only delay times much shorter than 3Ops can be used.In general the spectral line intensities decrease by at least two orders of magnitude within 3 ps after the nanosecond laser pulse [Fig. 3(a)]. Therefore delay times must be kept short in order to obtain detector signals (count rates) with sufficient signal-to-noise ratio particularly when taking spectra of single laser shots. In general the width of the detector gate pulse should not be greater than 10 ps [see Fig. 3(b)]. Only two data acquisition modes are used throughout the measurements. Either the spectra are taken from each single laser plasma and transferred to the host computer or a certain number of spectra are taken from successive laser plasmas and then accumulated into one resulting spectrum.The first mode is used for measurements with a scanning laser beam while the second mode is used for measurements at a fixed sample position When a certain loss of spatial resolution is acceptable during element distribution mea- surements spectra accumulation can also be employed. For the determination of net signals (central peak intensities of spectral lines) a mathematical procedure was developed. This model takes into account the underlying continuum radiation and the contributions of the wings of neighbouring spectral lines. Concentrations of analytes are determined via internal standardization. The ratio of net peak intensities of an analyte line to a reference line is found. The corresponding concentration ratio of the analyte to the internal standard is then calculated by using a calibration graph.However owing to the limited wave- length range captured simultaneously by the detector array 0.30 5 0.25 0 .- r 4- .- g 0.20 a .c 0.15 4- Y rn - 0.10 If 1 0.05 u 0 a) 0 200 400 600 800 Copper concentration/ppm Fig. 4 Calibration graph of copper (Cu I 327.4 nm) to iron (Fe I 327.1 nm) (in solid steel) peak intensity ratio versus copper concentration. Copper varies between 30 and 700 ppm in the certified reference samples1032 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL. 7 Table 2 Selection of prominent analyte and reference lines used with LIESA for a variety of sample matrices Solid/liquid steel in laboratory- Analyte line Reference line Reference element concentration Liquid steel in steel plant- Analyte line Reference line Reference element concentration Rubber mixtures- Analyte line Reference line Reference element concentration Geological samples- Analyte line Reference line Si 1288.2 nm Mn 1403.3 nm Cr 1425.4 nm Ni 1341.5 nm Cu I 327.4 nm C I 193.1 nm Mo I1 281.6 nm Fe 1439.4 nm >9Oo/o iron C I 193.1 nm Fe I1 193.2 nm Fe I 196.0 nm Fe I1 201.1 nm >9Oo/o iron Si I 212.4 nm Zn 1208.7 nm S I 182.6 nm Co I1 228.6 nm C I 193.1 nm C I 247.9 nm > 80% carbon Si 1288.2 nm Al I 308.2 nm Mg I 285.2 nm Fe I 302.0 nm Na I 589.0 nm Ca I1 3 15.9 nm and lines of other elements a presumption of internal standardization i.e.comparable excitation energies of analyte and reference lines cannot be fulfilled in many cases. As an example the linear calibration graph of the Cu I 327.4 nm:Fe I 327.1 nm line intensity ratio versus the copper concentration in the range between 30 and 700 ppm is shown in Fig.4. The relative detection limit is around 10 PPm. To achieve accurate absolute concentration measure- ments the most abundant element with a high concentra- tion should serve as the internal standard. The prominent analyte and refe,rence lines and the internal standards used with the different LIESA instruments are sum- marized in Table 2. The analytical figures of merit as a result of many experimental investigations are described in Table 3. In order to determine absolute analyte concentrations either the concentration of the reference element must be known or all elements must be measured relative to the reference element with unknown concentration.The latter can only be achieved when the major constituents of the sample are detected simultaneously in one spectrum. Then the sum of all concentrations is near to 100%. High concentration of the most abundant reference element significantly improves the accuracy of the absolute analyte concentration. In such cases the relative error of the analyte to reference concentration ratio is mainly governed by the relative error of the analyte concentration; relative accura- cies of 3% are achievable when measuring absolute element contents (see Table 3). All measurements should ideally be performed under constant operating conditions. However in reality changes in system parameters may occur from time to time resulting in different plasma conditions (e.g.temperatures and number densities). In such cases the measured calibra- tion graphs are no longer valid. To overcome these problems and to allow for longer time periods between system recalibrations the plasma conditions can frequently be monitored spectroscopically. In all spectral regions of interest atomic and ionic spectral lines of one element can be simultaneously detected. The intensity ratios of such atomic/ionic line pairs are a sensitive monitor of the averaged plasma temperature. By permanent observation of these intensity ratios changes in plasma temperatures can be easily detected and a correction factor can be derived to calculate back to the plasma conditions for which the calibration graphs were measured (Carlhoff et ~ 1 .' ~ ) . Applications Direct Analysis of Liquid Steel Steelmakers have long been searching for ways to obtain data on the condition of molten metal throughout pro- duction. Since conventional methods such as lance sampl- ing lead to an interruption of the process there is a strong need for on-line control. By using laser-induced emission spectrometry it is possible to determine directly the concentration of chemical elements in the melt. A LIESA instrument has been adapted to an 80-t AOD (argon- oxygen decarburization) converter at a Krupp steel plant in Germany. This system measures the carbon content (by laser analysis) and the temperature of the melt' (by pyrome- try). Additional laboratory measurements were carried out at the University of Madrid. Table 3 Analytical figures of merit.In most cases ranges are given to cover all investigated applications of laser-induced emission spectral analysis On-line capability In-process capability Micro/macro sampling (Nd:YAG excimer laser) Relative detection limit Absolute detection limit RSD value Relative accuracy Dynamic range Detection transfer and partial evaluation of a maximum of 50 single spectra per second (1 spectrum per 'laser shot for 1024 14-bit integers) multi-element analysis no sample preparation access to targets in harsh and hazardous industrial environments Ablated masses of 0.1-7 ,ug per laser shot ablation areas of 0.1-6 mm2 linear scanning of maximum 2.5 m s-I scan velocity Direct elemental analysis simultaneous 10-100 ,ug g-1 1-100 pg I-2% 3 %o Linearity over 1-3 orders of magnitudeJOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL.7 1033 Fig. 5 Water-cooled box at converter wall containing laser head optics and control electronics The system design for the converter is as follows the beam of a high-power Nd:YAG laser (wavelength 1064 nm) is focused onto the surface of the melt through a narrow channel in the side wall of the converter. Pre-heated argon gas (T>800 "C) is injected to prevent the steel from escaping and the melt at the end of the channel from solidifying. The laser beam produces a hot plasma. The light generated passes by way of a 7.5 m long optical fibre to the spectrometer where it is detected with an optical multichannel analyser (see above). A computer is used to calculate the concentrations of the elements from the detected spectral lines.Melt tempera- tures are determined by a pyrometer. The results are transferred to the process computer for optimum pro- duct ion control. To achieve correct alignment of the laser beam through the converter channel it was necessary to attach laser head and beam guiding optics directly to the converter wall. These parts of the system are installed in a closed water- cooled box (Fig. 5 ) while the remaining parts (host computer electronic control units spectrometer laser power supply and gas handling system) are contained in a housing on the converter platform. During many runs the LIESA system proved that it can sustain the extreme environmental conditions of a steel production furnace. The system has been operated at the converter for several months.A set of spectra from liquid steel in the near vacuum UV spectral region with the 193.1 nm carbon analyte line at different carbon concentrations is shown in Fig. 6. All remaining lines are atomic and ionic iron lines. By accumulating the optical emission of 200 laser shots within 10 s repeatability is significantly improved. Despite the very small observation angle and the long optical path of the plasma light which has to pass several mirrors lenses Cl193.1 nm Fe II 201.0 nm I z Wavelengthlnm Fig. 6 Spectra of liquid steel with increasing carbon content. The signal intensities are normalized to the Fe I1 193.2 nm line. The carbon analyte line at 193.1 nm and one of the iron reference lines at 20 1 .O nm are indicated I I ' I 1 I 0.01 0.02 0.05 0.1 0.2 0.5 1 Carbon concentration (%) Fig.7 Calibration graph of carbon (C I 193 1 nm) to iron (Fe I1 193.2 nm) (in liquid steel) peak intensity ratio versus carbon concentration and the optical fibre resolution and signal-to-noise ratio of the resulting spectra are excellent. The concentration of carbon can be evaluated by calculat- ing the net line peak intensity ratio of the carbon line to an isolated iron line as internal standard. A calibration graph for the decarburization as a result of several melts is shown in Fig. 7. After lance sampling carbon concentrations could be measured off-line in the analytical laboratory with X-ray fluorescence. At very low carbon contents the graph flattens because of coincidence with a small iron line.The carbon detection limit from this graph is less than 200 ppm. It will be possible in the near future to avoid occasional solidifica- tion of metal in the converter channel by optimizing the argon gas pre-heating process. The application of LIESA to the converter will permit optimum control of the melt process. Thus tap-to-tap times can be shortened reducing consumption of gases fluxes and refractories. This will enhance cost efficiency in steel- making. More accurate adjustment of the composition will improve steel quality. The LIESA system appears to be suitable not only for the converter process but also for other components in a steel production line. Homogeneity Measurements of Tyre Rubber in the Mixing Shop The European tyre industry is today faced with two major trends a constant growth in raw materials being processed and a decrease in profit compared with previous years.1034 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL.7 Since the market rigorously demands low price and high quality there are various approaches that can be used to improve profitability. The reduction of material costs is limited owing to their close relation to product perform- ance. However processing techniques offer a number of possibilities for higher economy. Increased productivity and improved consistency of quality can be achieved by better preparation and performance during the compound- ing stages. The tolerances in compounding and mixing must be kept as tight as possible so that the rejection rate is decreased to a minimum.Additionally automation must be increased in order to reduce the human influence on the result. For industrial rubber mixing processes it is desirable to evaluate on-line the dispersion of the different ingredients in the polymer matrix (i.e. compound homogeneity). Therefore Krupp Pirelli (as a tyre manufacturer) and King's College London (as a scientific supporter) started the development of a laboratory system capable of in-process monitoring of the homogeneity of rubber slabs in the open mill and other locations in the early stage of tyre pro- duction. By scanning the focused beam of a pulsed excimer laser across the surface of rubber slabs in a rubber mixing line spatial element distributions can be measured and evalu- ated on-line.Thus the actual element composition and homogeneity of the rubber material ' can be monitored frequently during the mixing and forming process. Each laser pulse produces a hot bright plasma on the rubber surface in a buffer gas flow of constant flow rate (argon at normal pressure). The optical plasma emission. is detected and the spectra are evaluated as described above. Systematic investigations of the optical plasma emission were performed in the wavelength region from 180 to 800 nm. All measured spectral features in that range could be identified and assigned to appropriate elements and mole- cules. Atomic and ionic spectral lines of C H s Si Zn Co Mg Al Ca Na and K and molecular bands of C2 and CN have been detected. The temporal behaviour of the laser- induced plasma under different experimental conditions has been studied.It has been observed with tyre rubber that spectral line peak intensities exhibit roughly the same behaviour as the lines of all the other materials analysed [see Fig. 3(a) and (h)]. The interaction of the UV laser with rubber was also investigated in detail. As expected the UV laser ablation pattern on the rubber surface exhibits sharp edges without thermal side-effects. This is not the case with the IR Nd:YAG laser beam. Excimer laser peak powers as low as 1 x loE W cm-2 were sufficient to produce plasmas with strong spectral emission. Ablation depths of 2 pm per laser pulse with an ablation area of 3 mm2 have been achieved with this irradiance employing the KrF excimer laser at 248 nm.A typical calibration graph is shown in Fig. 8 where the line peak intensity ratio of a zinc to a carbon line is plotted versus the zinc oxide to carbon concentration ratio. Such calibration graphs reveal RSD values of between 0.5 and 2%. The corresponding spectra are shown in Fig. 9. Since each spectrum is normalized to the C I 193.1 nm line variations in the sulfur and silicon (oxide) content of the rubber samples can also be seen. Element Distribution Measurments of Geological Materials on Conveyor Belts I There is a great need in industry for the on-line control of continuous mass streams of geological raw materials. By permanent monitoring of element distributions raw ma- terials with narrow composition tolerances can be available ahead of further processing. In the case of bad or incorrect I I 0.08 - 0 .- i- F .g 0.06 - C P) C m a 4- .- y 0.04 - t.l 0.02 - 15 - I 0 - J 0 0.010 0.020 0.030 0.040 0.050 0.060 0,070 czn, cc Fig.8 Calibration graph of zinc oxide in solid tyre rubber. The peak intensity ratio of the Zn I 208.7 to the C I 193.1 nrn line is plotted versus the zinc oxide to carbon concentration ratio CI193.lnm Zn n I ZnI208.7nm 80 185 190 195 200 205 210 215 0 Wavelengt h/nm z Fig. 9 Spectra of solid tyre rubber with increasing zinc oxide content. The signal intensities are normalized to the C I 193. I nm line. The zinc analyte line at 208.7 nm the carbon reference line at 193.1 nm and spectral lines of S and Si are indicated ".L 0.01 0.1 1 10 cslol Ccao Fig. 10 Calibration graph of silicon oxide in geological samples.The peak intensity ratio of the Si I 288.2 to the Ca 11 3 17.9 nm line is plotted versus the silicon oxide to calcium oxide concentration ratio compositions measures will be taken to reject or divert the materials that are outside of the range of tolerance. Preliminary studies of different samples led to encourag- ing results. A calibration graph of Si02 versus CaO employing the Si I 288.2 nm analyte and Ca I1 317.9 nm reference line is shown in Fig. 10. Several data points lying close together at a given concentration ratio demonstrate the very good precision of the measurements.JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY. SEPTEMBER 1992. VOL. 7 1035 1 1 I 1 0 5 10 15 Track position/mm Fig. I1 along a linear track on the sample surface Thickness profile of TIN coated on a flat metallic substrate 8x103 6x103 4x103 2x103 v) c.I 1 I I 200 400 600 800 1000 Wavelength (channel No.) Fig. 12 ( a ) Spectrum of Ti in the UV spectral region emitted after the tenth laser pulse and (h) spectrum of the metallic substrate in the same spectral region as in ( a ) emitted after the hundredth laser pulse Depth Profiling of Muilti-layers on Metallic Substrates Intensive investigations at Krupp showed that laser analysis is not only suitable for elemental analysis but also for the depth profiling of coated substrates. During laser-matter interaction material is ablated in a defined way. The ablation pattern and ablation depth per laser pulse can be controlled via the laser parameters (spatial intensity distribution of the laser beam pulse energy pulse repetition rate and shape and size of laser focus).If each laser plasma is analysed a depth profile of the substrate layer(s) is created. The change in a layer is indicated by the change in the detected spectral pattern. A thickness profile (single TIN layer on a metallic substrate) along a straight line on the flat surface of the coated sample is shown in Fig. 1 1. The ablation depth of the pulsed laser was 0.1 pm per pulse. Corresponding spectra are presented in Fig. 12(a) and ( 6 ) (same wavelength region in the UV). A Ti spectrum emitted after the tenth laser pulse is shown in Fig. I2(a) while a metal spectrum emitted after the hundredth laser pulse deep in the substrate can be seen in Fig. 12(b). The measured profile was in excellent agreement with results of off-line standard measurement techniques.Conclusions Laser-induced emission spectral analysis appears to be a very promising method for on-line analysis in those industrial fields where element distribution measurements of materials at all stages of production yield information on the quality of material and production process. At present Krupp is developing a whole family of LIESA instruments to cover the needs of different industrial branches. Financial support for the development of the LIESA systems by the German Minister of Research and Techno- logy (BiMFT) by the European Communities for Steel and Coal (ECSC) and by the BRITE/EURAM programme of the Commission of the European Communities is gratefully acknowledged. 1 2 3 4 5 6 7 8 9 10 1 1 12 References Moenke-Blankenburg L. Laser Microanal.vsis Wiley New York 1989. Laser-Induced Plasmas and Applications eds. Radziemski L. J. and Cremers D. A. Marcel Dekker New York 1989 pp. KO J. B. Sdorra W. and Niemax K. Fresenius’ 2. Anal. Chem. 1989 335 648. Leis F. Sdorra W. KO J. B. and Niemax K. Mikrochirn. Acta 1989 11 185. Sdorra W. and Niemax K. Spectrochim. Acta Part B 1990 45 9 17. Niemax K. and Sdorra W. Appl. Opt. 1990 29 5000. Sdorra W. and Niemax K. Mikrochirn. Acta 1992 107 319. Sdorra W. Brust J. and Niemax K. Mikrochim. Acta 1992 in the press. Carlhoff C. Lorenzen C.-J. Nick K.-P. and Siebeneck H.-J. in In-Process Optical Measurements ed. Spring K. H. Proc. SPIE SPIE Publications Bellingham vol. 101 2 1989 pp. Carlhoff C. and Kirchhoff S. Laser Optoelektronik I99 I 23 50. Lorenzen C.-J. Jogwich M. Burge R. E. Michette A. G. Daghooghi R. and Nahmias M. paper presented at the BRITE/EURAM-EUREKA Workshop on Polymer Techno- logy and Industrial Applications Ferrara Italy May 13-1 5 1991 Book of Abstracts 2 pp. 122-125. Carlhoff C. Lorenzen C.-J. and Nick K.-P. German patent No. DE 39 1 1 965 C2; date of application April 12 1989. 295-346. 194- 1 96. Paper 2/00 7130 Received February I I I992 Accepted April 13 I992

ISSN:0267-9477    

DOI:10.1039/JA9920701029

出版商:RSC    

年代:1992

数据来源: RSC

摘要:

JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL. 7 1037 Abell Ian 53 Abollino Ornella I9 Adamik Miklbs 707 Adams Freddy C. 987 Ajayi Olubode O. 689 Akatsuka Kunihiko 889 Akman Suleyman I87 Alimonti A. 859 Aller A. Javier 753 Al-Maawali Sabah 579 AL-Rashdan Amel 55 1 Amarasiriwaradena Chitra J. Anderson Mats 26 1 Anglov Thomas 329 Ansari Tariq M. 689 Argentine Mark D. 91 5 Badini Raul G. 481 Baeten Wilhelmina L. M. Bagdi Gyula 769 Bai Jian 35 425 Barba Maria F. 869 877 Barnes Ramon M. 653 825 833 839 845 851 915 1013 845 851 915 1013 599 Barrett Jon F. R. 109 Baxter Douglas C. 141 405 Beauchemin Diane 937 965 Bekkers Mirjan H. J. 599 Bergamin Filho Henrique Berglund Michael 14 1 46 1 Berman Shier S. 889 Bhattacharyya Shuvendu S. Bielski Bradley A. 899 Biffi Claudio 409 Bolshov Mikhail A.1 99 Bosch Mossi F. 47 Bosch Reig F. 47 Boutron Claude F. 99 Bozsai Gabor 505 Braverman Diane S. 43 Bremier Phillipe 8 19 Brenner Isaac B. 8 19 Browner Richard F. 8 I3 Brumme Mathias 28 1 Bulska Ewa 20 1 405 Byrne John P. 371 579 Cabon J. Y. 383 Caetano Manuel 1007 Campbell Michael J. 6 17 Carlhoff Christoph 1029 Carlini Enzo 19 Caroli Sergio 859 CarrC Martine 791 Carroll John 533 Caruso Joseph A. 551 807 Castillano Theresa M. 807 Chakrabarti Chuni L. 371 579 Chaudry Muhammad Mansha 29 701 Chekalin Nikolai V. 225 Chenery Simon 647 Chen Zhongxing 905 Chiba Koichi I 1 5 Christensen Jytte Molin 329 Cobo Isabel Gutierrez 247 461 727 865 417 461 899 929 971 993 CUMULATIVE AUTHOR INDEX FEBRUARY-SEPTEMBER 1992 Coedo Aurora Gdmez 247 Coles Barry J.587 Cornett Claus 629 Cox Rosamund J. 635 Craig Jane M. 937 Crain Jeffrey S. 605 Dams Richard 617 Das Arabinda K. 4 17 Dautheribes Jean-Luc 92 3 Dean John R. 229 Delle Femmine P. 859 Demeny Dezsb 545 707 DEdina Jiii 307 Docekal Bohumil 52 1 Dong Liping 293 439 Dorado Lopez M. Teresa 1 1 Duan Yixiang 7 Duckworth Douglas C. 71 I Duller,. Geoff A. T. 53 Ebdon Les 23 51 I 719 895 Eglington Timothy I. 979 Ehmann William D. 749 Emteborg HAkan 405 Ericson Inger 979 Escudero Baquero Esther Evans Hywel 929 Falconieri P. 859 Falk Heinz 255 Fang Duencheng 959 Fang Zhaolun 293 439 Farifias Juan C. 869 877 Farnsworth Paul B. 89 197 Feldmann Ingo 121 Fellows Craig S. 3 15 743 Feng Jianxing 171 Fernanda Gine Maria 865 Fischer Werner 239 Fisher Andrew S. 51 1 Foner Henry 845 851 915 Ford Michael J.7 19 Frech Wolfgang 141 405 Freedman Philip A. 57 1 Freire dos Reis Boaventura Fryer Brian J. 905 Fuge Ronald 53 595 61 1 Galley P. J. 69 Gallimore David L. 605 Garcia-Olalla Conception Gilmutdinov Albert Kh. 675 Glick M. 69 Gluodenis Thomas J. Jr. Golding Rafael E. 1007 Gomez Coedo Aurora 11 Gorlach Ursula 99 Grazhulene Svetlana S. 105 GrCgoire D. C. 371 579 Grosse-Wilde H. 343 Guntur Daru 239 Guo Tiezheng 667 Gutierrez Cobo Isabel 1 1 Guqer Seref 179 Guell Oscar A. 135 Gunther Detlef 25 I Hahn Ulrich 1029 Hakala Erkki 191 Han Heng-bin 447 247 247 865 753 30 1 247 247 Han Myung S. 641 Hanselman D. S. 69 Hansen Steen Honore 629 Harnly James M. 533 Harrison W. W. 75 Hartley James 23 895 Haschke Michael 28 1 Haug Hermann O. 451 Heckel Joachim 28 1 Heitkemper Douglas T.55 1 Hermann Gerd M. 457 Hernandez Cordoba Manuel Hieftje Gary M. 69 335 Hill Steve J. 23 51 1 71 9 Hinds Michael W. 685 Hoffmann Erwin 727 Holcombe James A. 135 Houk R. S. 799 Huang Benli 287 Huie Carmen W. 353 Huneke John C. 943 Hunt Andrew 647 Hutton Robert C. 623 943 Hutsch Bruno 1 Inamoto Isamu 115 Ivanov V. P. 675 !warnoto Etsuro 42 1 Ince Hurrem 187 Jacksier Tracey 653 839 Jahl Matthias J. 653 825 Jakubowski Norbert I2 1 Jansen Elisabeth B. M. 127 Jian An-bei 5 15 Jiang Gui-bin 447 Jimenet Seco Jose L. 1 1 Jin Qinhan 7 Jogwich Martin I029 Jose Krug Francisco 865 Katoh Takashi 539 Kato Takunori 15 KBntor Tibor 219 Key Edgar A. 1007 Khvostikov Vladimir A. 105 Kibble Helen A. B. 315 Kishimoto T. 343 Kitagawa Kuniyuki 539 Kleiner Joachim 433 Klockenkamper Reinhold Knipscheer Joop H.127 Koloshnikov Vsevolod G. 99 Kong Xiangxing 7 Kovacic Nada 999 Koklu Unel 187 Krishna Prabhu R. 565 Krivan Viliam 155 52 1 Kruchevska Antoaneta 845 Kujirai Osamu 661 Kumamaru Takahiro 42 1 Kumpulainen Jorma 165 Kuss Heinz-Martin 25 1 Kuzua Mikio 493 Lajunen Lauri H. J. 735 Lakatos Istvan 769 Lakatos Janos 769 Lam Joseph W. 889 Lamoureux Marc 371 579 529 783 895 833 95 1 273 851 915 Larkins Peter L. 265 Larsen Erik H. 629 Lasnitschka George F. 457 Le Bihan A. 383 Lee Kwang W. 641 Lee Milton L. 197 Lei Zhu 425 Lemarchand Allain 8 19 Liang Zhongwen 10 1 9 Li Ang 447 Li Bingwei 425 Li Ke 141 Li Qinguan 131 Li Wenchong 131 Li Yongquan 425 Li Zhikun 425 Lin Fan 175 Lin Yuehe 287 Littlejohn David 29 533 689 695 701 727 Liu Jun 7 Liu Mingzhong 667 Lobinski Ryszard 987 Lonardo Robert F.10 19 Longerich Henry P. 905 Lopez Garcia Ignacio 529 Lopez M. Teresa Dorado Lorenzen Claus J. 1029 Louie Honway 557 Luan Shen 799 Luecke Werner 765 Ludke Christian 727 Lupke G. 343 Ly T. 371 Mahalingam T. R. 565 Majidi Vahid 749 Marawi Isam 899 Marchante Gayon Juan M. Marcus R. Kenneth 71 1 Markesbery William R. 749 Marowsky G. 343 Marshall John 229 Martines Laura 845 851 Mathews C. K. 565 Ma Yizai 35 425 McLaren James W. 889 McLeod Cameron W. 66 1 Meeks Frank R. 899 Menditto A. 859 Menotti A. 859 Mentasti Edoardo I9 Mermet Jean-Michel 79 1 Michel Robert G. 1019 Mikami Osamu 493 Miller-Ihli Nancy J. 533 Moder Ralph 457 Moenke-Blankenburg Lieselotte 25 1 Mohammad Bashir 695 Morimoto Satoru 21 1 Morisi G.859 Morita Masatoshi 15 Mouillere Delphine 70 I Mudakavi Jayateerth R. 499 Mu Huiling 175 Nagahiro Tohru I83 Naghmush Abdulmagid M. Nagtegaal Mario 127 Nakahara Taketoshi 21 1 Naoumidis Aristidis 239 247 743 3231038 JOURNAL OF ANALYTICAL ATOMIC SPECTROMETRY SEPTEMBER 1992 VOL. 7 Nichol Robbin 727 Nickel Hubertus. 239 Ni Zhe-ming 447 5 15 Ofley Stephen G. 315 O’Gram Samantha J. 229 O’Haver Thomas C. 533 Ohlsson K. E. Anders 357 Okochi Haruno 661 Olbrych-Sleszynska Ewa 323 Olson Lisa K. 993 Omenetto Nicolo 89 Ottaway Barbara J. 701 Pang Ho-ming 799 Park Chang J. 641 Park Sang R. 641 Pastor Garcia A. 47 Patterson Clair C. 99 Paul Michael 251 Pearce Nicholas J. G. 53 Peramaki Paavo. 735 Peris Martinez V. 47 Perkins William T. 53 595 Perry Bruce J.883 Peters Gregory R. 965 Petrucci F. 859 Petrucci Guiseppe A. 48 1 Perez Parajon Juan 743 Pickford Christopher J. 635 Porta Valerio 19 Poulsen Otto Melchior 329 Poussel Emmanuelle 79 1 Preli Francis R. Jr. 1019 Pritzl Gunnar 629 Provitina Olivier 923 Puri Be1 Krishan 183 Pyy,. Lauri 19 1 Qian Haowen 13 1 Quevauviller Philippe 6 17 Rademeyer C. J. 347 595 61 1 61 1 RadiC-PeriC Jelena 235 Radziuk Bernard 389 397 Raith Angelika 623 943 Ramsey Michael H. 587 Ramus Terry L. 999 Rankin Andrew H. 587 Revy Daniel 923 Riglet Chantal 923 Rosen Arne 261 Rudnev Serge1 N. 1 99 Saarela Kjell-Erik 165 Sack Brigitte 121 Saeki Masao 1 1 5 Sanz-Medel Alfredo 743 Sarzanini Corrado 19 Satake Masatada I83 Schindler Rolf 28 1 Schlemmer Gerhard 499 Schrader Werner 667 Schumann Thomas 25 1 Seare Nichola J.315 Seegopaul Purnesh 959 Seeley Jeffrey A. 979 Senofonte O. 859 Shan Xiao-quan 394 447 Shi Huiming 175 Shimazu Hiromichi 42 1 Shum Sam C. K. 799 Sieverdes F. 343 Sinemus Hans-Werner 433 Skole Jochen 727 Slaveykova Vera I. 147 Smith Benjamin W. 89 Soo Susan Yoke-Peng 557 Sorokin Mikhail V. 105 Speller D. V. 883 Sperling Michael 505 Stabel Hans-Henning 433 433 505 76 1 365 Starn Timothy K. 335 Story W. Charles 807 Stuewer Dietmar 12 1 95 1 Sturgeon Ralph E. 339 Sun Di-jun 35 Sychra Vaclav 389 Szardening Thomas W. 457 Sziics Laszlo 707 Takahashi Junichi 10 19 Tan Jingyuan 131 Tao Keyi 171 Tarr Matthew A. 8 13 Thomassen Yngvar 397 Thompson Michael 635 647 Tittarelli Paolo 409 Tomlinson William R. 229 Trojanowicz Marek 323 Tsalev Dimiter L.147 365 Tyson Julian F. 301 3 15 IJden Peter C. 979 IJehiro Takashi 15 IJre Allan M. 695 van de Weijer Peter 599 Van Grieken Rene 81 Van Loon Jon C. 883 Vela Nohora P. 551 807 Vermaak I. 347 Vermeir Gerda 6 17 Vieth Wojciech 943 Vijayalakshmi S. 565 Viiias Pilar 529 Violante N. 859 Vlasov Igor I. 225 Yoloshin A. V. 675 von Bohlen Alex 273 Vullings Peter J. M. G. 599 Vykhristenko Nina N. 105 Walder Andrew J. 57 I Wang Jiansheng 929 97 1 499 505 Wang Jiazhen 425 Wang Wen 761 Wang Xiaoru 287 Wang Xinsheng 175 Wasa Tamotsu 21 1 Wei Jizhong 175 Welz Bernhard 307 389 Wen Bei 761 Wenzel N. 343 Whitley John E. 29 Whittaker Paul G. 109 Wilkinson Jamie J. 587 Williams John G. 109 Willie Scott N. 339 Winefordner James D. 89 Wu Mingin 197 Wu Nian 353 Xhoffer C. 81 Xiao Jian 131 Xu Fu-chun 5 15 Xu Ning 749 Xu Peiqing 775 Xu Shukun 293 Yamada Kei 661 Yaman Mehmet 179 Yang Pengyuan 287 515 Yang Seok R.641 Yasuhara Akio 15 Yokota Kayoko 421 Yuan Dongxing 287 Zakharov Yu. A. 675 Zeng Yadi 979 Zhang Baogui 17 1 Zhang Hanqi 7 Zhang Li 447 Zhang Zhanxia I3 1 Zheng Hui 425 Zhu Guangxuan 813 Zhu Lei 425 Zhuang Zhi-xia 287 5 I5 48 I

ISSN:0267-9477    

DOI:10.1039/JA9920701037

出版商:RSC    

年代:1992

数据来源: RSC