Method for analyzing, by energy-dispersive X-ray spectroscopy, at least one chemical species in a molten pool or solidified bead produced by an additive manufacturing or welding process of a metallic alloy

Abstract

Method for analyzing by energy-dispersive X-ray spectroscopy of at least one chemical species in a melt pool or a solidified bead produced by an additive manufacturing or welding process of a metallic alloy. Method for analyzing by energy-dispersive X-ray spectroscopy of at least one chemical species in a melt pool (B) or a solidified bead (C) produced by a metallic additive manufacturing process or a welding process of a metallic alloy, the method of analysis comprising the following steps: Step a: emission of ionizing rays into at least a part of the melt pool or the solidified bead using at least one ionizing source (16), Step b: detection, using at least one multi-pixel spectral detector (35), respectively using at least two single-pixel spectral detectors (17;17a, 17b), of fluorescence photons emitted by said at least a portion of the fused bath (B) or the solidified bead (C) irradiated, and generation of at least one signal representing the fluorescence photon count rate by said multi-pixel spectral detector (35), respectively each of said single-pixel spectral detectors (17; 17a, 17b), in at least one measurement zone (Zm1, Zm2, Zm3, Zmk) given of the fused bath (B) or the solidified bead (C) and for at least one given fluorescence line, Step c: processing of said at least one signal in order to provide at least one piece of information on said at least one chemical species in the fused bath (B) or the solidified bead (C). Figure for the abstract: Fig. 2;

Description

Title of the invention: Method for analyzing, by energy-dispersive X-ray spectroscopy, at least one chemical species in a molten pool or a solidified bead produced by an additive manufacturing or welding process of a metallic alloy. Technical field

[0001] The present invention relates to a method of analysis by energy-dispersive X-ray spectroscopy, commonly referred to as "X-ray fluorescence analysis", of at least one chemical species, in particular its distribution, in a melt pool or a solidified bead produced by an additive manufacturing process or a welding process of a metallic alloy.

[0002] It also relates to a computer program product, a processing device, an analysis system and an additive manufacturing or welding installation.

[0003] The invention applies to the field of additive manufacturing or welding and more particularly to the characterization of the weld pool or solidified bead generated during or by the implementation of an additive manufacturing or welding process. Prior art

[0004] During the additive manufacturing or welding process, a molten pool is produced as a result of the interaction of the material, generally in the form of a wire or particles, with a heat source, for example, an electric arc, a laser, or an electron beam. Depending on the process parameters, such as the induced thermal power, the material feed rate, or the system geometry, the molten pool may exhibit thermal and chemical gradients or geometric variations, either at the surface or in depth.

[0005] An important physical factor is the distribution of the alloy's chemical species on the surface and within the volume of the bath, along with their potential migrations during the additive manufacturing or welding process. This can indeed affect the quality of the part produced by additive manufacturing or welding.

[0006] Two studies investigate the differential evaporation of titanium alloys. These are the study by Schwerdtfeger, Jan, and Carolin Kôrner entitled "Selective electron beam melting of Ti-48Al-2Nb-2Cr: Micro structure and aluminium loss." Intermetallics 49 (2014): 29-35 and that by Juechter, V., et al. entitled "Processing window and evaporation phenomena for Ti-6Al-4V produced by selective electron beam melting." Acta Materialia 76 (2014): 252-258.

[0007] A solution for analyzing the composition of the melting bath is proposed by Vasily N Lednev et al., in the publication "In situ elemental analysis and failure detection during additive manufacturing process utilizing laser induced breakdown spectroscopy," Optics Express 27.4 (2019): 4612-4628 and in the publication "In situ multi-elemental analysis by laser induced breakdown spectroscopy in additive manufacturing," Additive Manufacturing 25 (2019): 64-70, describe this solution. It involves performing laser-induced breakdown spectroscopy (LIBS) analysis on both the melt pool and the weld bead. This analysis can be performed during the additive manufacturing process.While this solution allows for elemental chemical analysis, it is not entirely satisfactory because the limitation of this approach is intrinsically linked to LIBS analysis, which remains surface or subsurface (limited to about ten micrometers below the surface of the melt pool).

[0008] To address this, a device for in-depth analysis of an oxidizable molten metal by LIBS technique is known from patent EP 3 146 314 but this solution is not entirely satisfactory because the device is intrusive.

[0009] There is a need for a method of analyzing the melt pool and / or the solidified bead that can be implemented during the additive manufacturing or welding process and that allows for the study, without intrusion, of at least one chemical species inside the melt pool or the solidified bead at a depth that can preferably be greater than 10 pm below the surface of the melt pool or the solidified bead. Description of the invention

[0010] The invention succeeds in meeting all or part of this need thanks to, according to one of its aspects, an energy-dispersive X-ray spectroscopy method for analyzing at least one chemical species, in particular its distribution, in a molten pool or a solidified bead produced by a metal additive manufacturing process or a metal alloy welding process, the analysis method comprising the following steps:

[0011] a) Step a: emission of ionizing radiation into at least a part of the molten bath or solidified bead using at least one ionizing source, preferably collimated,

[0012] b) Step b: detection, using at least one multi-pixel spectral detector, or at least two, in particular two, single-pixel spectral detectors, preferably collimated, of fluorescence photons emitted by said at least a portion of the irradiated melt bath or solidified bead, and generation of at least one signal representing the fluorescence photon count rate by said at least a multi-pixel spectral detector, or each of said single-pixel spectral detectors respectively, in at least one given measurement zone of the melt pool or solidified bead and for at least one given fluorescence line,

[0013] c) Step c: processing said at least one signal with a view to providing at least one piece of information on said at least one chemical species in the melt pool or solidified bead, in particular information on a representative value, for example a physico-chemical quantity such as the amount or concentration, of said at least one chemical species in the melt pool or solidified bead.

[0014] Step c may for example include processing said at least one signal in order to determine at least one physico-chemical quantity, in particular the quantity and / or concentration, of said at least one chemical species in said at least one given measurement zone, or even in all or part of the melt pool or solidified bead.

[0015] Step c may include the establishment of at least a partial, in particular quantitative, mapping representative of the distribution of said at least one chemical species in the melt pool or the solidified bead.

[0016] When the analytical method is implemented for the analysis of several chemical species present in the metal alloy, step c can establish a single map representing the distribution of the chemical species in the melt pool or the solidified bead or several at least partial maps, each being representative of the distribution of one of said chemical species in the melt pool or the solidified bead.

[0017] Preferably, step b is implemented for a plurality of given measurement zones of the molten pool or solidified bead. In this case, each generated signal can be processed in step c.

[0018] Step b can be implemented for several given fluorescence lines, characteristic of a single chemical species or of several chemical species. In the latter case, the analytical method according to the invention makes it possible to analyze several different chemical species of the alloy in the molten pool or the solidified bead.

[0019] Step b may include taking cross measurements by said at least one multi-pixel spectral detector or said at least two single-pixel spectral detectors.

[0020] Thanks to the invention, a method is provided that allows for non-invasive quantitative elemental analysis to be performed both on the surface and in the depth of the bath. Indeed, detection, whether using a multi-pixel spectral detector or at least two single-pixel spectral detectors, allows for a cross-measurement approach.

[0021] The method according to the invention is of great interest for controlling the process of additive manufacturing or welding, and also to provide reference points for validating simulation models of these processes. Indeed, knowing the migrations of one or more chemical species in a melt pool can allow for control of the additive manufacturing or welding process.

[0022] In addition, we benefit from an in situ analysis method implementing a non-intrusive device in the material.

[0023] Moreover, the method according to the invention allows for in-depth and unlimited analysis, like LIBS technology, to a sub-surface depth of approximately 1 Opm, except by the nature of the chemical element analyzed.

[0024] The method according to the invention also makes it possible to have real-time information on at least one chemical species present in the metal alloy, in particular its distribution and / or concentration, within the melt pool.

[0025] The analysis method according to the invention advantageously allows the distribution of one, preferably several, chemical species of the alloy considered within the melt pool or the solidified bead to be characterized.

[0026] The expression "in a molten bath" includes both the interior of the molten bath and the interface between the molten bath and the atmosphere. The analytical method according to the invention thus allows, optionally or additionally, the study of one or more chemical species on the surface of the molten bath and within the molten bath, below its surface.

[0027] The term "solidified bead" refers to material that has solidified upon cooling after passing through a liquid state via melting in additive manufacturing or welding. The analysis will preferably focus on an area of ​​the bead located near the weld pool, with no time limit for analyzing the bead.

[0028] Said metal alloy may be selected from the group consisting of iron alloys, copper alloys, in particular bronze or brass alloys, aluminum alloys, titanium alloys, nickel-based superalloys, in particular Inconel 625, Inconel 718, 316L, tungsten carbide alloys, low coefficient of expansion alloys, in particular those based on iron and nickel such as Invar®, and any other metal alloy.

[0029] A metallic alloy comprises at least two different chemical species, generally in different proportions.

[0030] Each chemical species, also called a chemical element, present in said alloy may have an atomic number Z between 12 and 92. The chemical species may, in particular, be chosen from the group consisting of: transition metals such as, among others, iron, cobalt, chromium, molybdenum, zirconium, lead, tungsten, nickel, and niobium; post-transition metals, in particular lead, zinc, and aluminum; alkaline earth metals such as magnesium; and any other element whose The atomic number is between 12 and 92.

[0031] The chemical species present in the metallic alloy are of course a function of said alloy, as is the mass proportion in which they are present.

[0032] Since a metallic alloy by definition comprises several types of metals and therefore several chemical species, the analytical method can be implemented for all or some of these. The chemical species targeted by the analytical method according to the invention will be designated as chemical species of interest, and their characterization will be achievable using the energy-dispersive X-ray spectroscopy method according to the invention.

[0033] The chemical species of interest will then be associated with one or more fluorescence lines of interest, characteristic of this or these chemical species.

[0034] Steps a and b are advantageously carried out simultaneously. They preferably take place during the additive manufacturing or welding process, in particular as soon as the metal alloy melts in at least one given measurement zone of the weld pool. Alternatively, they may take place after at least partial solidification of the weld pool into a solidified bead, in at least one given measurement zone of the solidification bead, during or after the additive manufacturing or welding process.

[0035] Said spectral detector(s) preferably include a collimation system. Such a collimation system may be chosen from the group consisting of a hole collimation system and a coded mask collimation system. Such a coded mask may be the Mura mask described by SR Gottesman and EE Fenimore in Appl. Opt. 28 (1989) 4344, "New family of binary arrays for coded aperture imaging," or the Barker mask, in 1D but generalizable to 2D, described by RH Barker in Communication Theory (Butterworth, 1953), p. 273. The collimation system may also be a slit collimation system.

[0036] During step a, the ionizing source preferably emits in an energy range between 1 keV and 300 keV, preferably between 2 keV and 50 keV, and more preferably between 5 keV and 30 keV. The choice of the ionizing source may, in particular, be guided by ease of implementation. In a preferred embodiment, the ionizing source implementing step a is an X-ray tube capable of emitting an X-ray beam.

[0037] Alternatively, the ionizing source is capable of emitting a gamma ray beam, a sufficiently accelerated electron or ion beam, or a neutron beam.

[0038] Said spectral detector(s) implementing step b are preferably configured to detect fluorescence photons emitted by one or more species chemical compounds present in the metal alloy within the molten pool or solidified bead.

[0039] The method may include a preliminary step, prior to step a, of theoretical determination relating, for example, to an evaluation of measurement sensitivity or, for example, to depth sensitivity. This preliminary step may be carried out by analytical or numerical modeling. By way of illustration, an evaluation of depth sensitivity may be carried out by numerical modeling of the radiation-matter interaction, in particular with regard to the absorption, by the weld pool or the weld bead, of fluorescence X-ray photons emitted at depth, for a given metallic alloy and a given additive manufacturing or welding process, for at least one given chemical species of said alloy.

[0040] By "depth sensitivity" is meant the sub-surface depth, i.e. below the surface of the melt bath, to which a given chemical species can be detected using the spectral detector(s), i.e. the depth to which the spectral detector(s) can detect the emission of fluorescence photons characteristic of that chemical species.

[0041] The depth sensitivity can be considered to correspond to a depth that allows a transmission rate of 5%, representing 5% of the transmitted fluorescence photon flux, through the bath or solidified bead, and thus reaching the detector(s). The remaining 95% of this fluorescence photon flux has been absorbed, for example, by the alloy itself. Naturally, the greater the subsurface thickness, the higher this absorption will be, thus reducing the transmission rate of fluorescence photons, or X-ray photons, to the spectral detector(s).

[0042] It is of course possible, without departing from the scope of the invention, to choose to equate the depth sensitivity to a transmission rate other than 5%, in particular a lower one, for example equal to 3%. This may depend on the measurement noise. Indeed, if there is 0.5% noise, a transmission rate of 3% may be suitable.

[0043] This depth sensitivity depends on the chemical species of interest, the composition of the metallic alloy in which this species is present and / or the additive manufacturing or welding process including the process parameters.

[0044] The method can thus allow, by this preliminary step, to determine theoretically the depth sensitivity of each chemical species of interest of the alloy, for the additive manufacturing or welding process implemented.

[0045] Step b and / or c preferably comprises determining the depth sensitivity of the chemical species associated with said fluorescence line, from the energy-centered count rate of a fluorescence line of a given chemical species, measured by the spectral detector(s), and taking into account the arrangement of the multi-pixel spectral detector or of the relative arrangement of the single-pixel spectral detectors during measurement.

[0046] The measurement performed is preferably averaged over a measurement surface that can be adjusted by collimation. Such a measurement surface may be between 0.007 mm² and 4 mm², for example, 0.03 mm².

[0047] The analysis method according to the invention can employ two single-pixel spectral detectors with a collimation system. In this case, the ratio A between the counting rates of the two spectral detectors is such that

[0048] [Math.l] |1- A| preferably greater than or equal to 5%.

[0049] Alternatively, the analysis method may implement at least one multi-pixel spectral detector with collimation system.

[0050] Step c may include at least one treatment such as a correction relating to detection efficiency, stacking phenomena or escape peak.

[0051] Step c may include a Gaussian fit produced on the fluorescence lines of interest allowing an estimation of the net count, i.e. a count measurement without the contribution of noise.

[0052] Step c may further include, in particular after estimation of the net count, the implementation of calibration sets in order to link the measurement of the net count to a physico-chemical quantity of the chemical species such as, for example, the concentration of the chemical species.

[0053] The invention also relates, according to another of its aspects, in combination with the above, to a computer program product comprising program code instructions which, when executed by a computer, implement at least step c of the analysis method as defined above.

[0054] The invention also relates, according to another aspect, whether in combination with the foregoing or not, to an analysis system for the analysis by energy-dispersive X-ray spectroscopy of at least one chemical species, in particular of the distribution of at least one or even several chemical species, in a molten pool or a solidified bead produced by an additive manufacturing process or a welding process of a metallic alloy, in particular an analysis system for implementing the analysis method as defined above, comprising:

[0055] - an ionizing source configured to emit ionizing radiation into at least a portion of the molten bath or solidified bead, - at least one multi-pixel spectral detector or at least two, in particular two, single-pixel spectral detectors configured to detect fluorescence photons emitted by said at least a part of the melting bath or the irradiated solidified bead, and to generate at least one signal representing the fluorescence photon count rate, in at least one given measurement area of ​​the melt pool or solidified bead and for at least one given fluorescence line and - a data processing system configured to provide, from said at least one signal from said or said spectral detectors, at least one piece of information on said at least one chemical species in the melt pool or solidified bead, including information on a representative value of said at least one chemical species in the melt pool or solidified bead.

[0056] The entire ionizing source and said spectral detector(s) is preferably attached to a heating and material supply device of an additive manufacturing or welding installation for the implementation of said additive manufacturing or welding process.

[0057] This assembly can be mechanically fixed following said heating and material supply device in the direction of its movement during the implementation of said additive manufacturing or welding process.

[0058] In one variant, this assembly is separated from the heating and material supply device of the additive manufacturing or welding installation.

[0059] Thanks to this analysis system, an online measurement system is available, using energy-dispersive X-ray spectroscopy, interrogating the melt pool area, or, where applicable, the solidified bead, to enable elemental analysis with the best compromise between the measurement area and the signal-to-noise ratio. It is understood that the smaller the measurement area, the greater the noise relative to the useful signal, with quantification, for example, of the elemental proportion, via a calibration set. The analysis system can establish a map, particularly a quantitative one, of the surface and depth distribution of the element(s) of interest through a cross-measurement approach performed by said at least one multi-pixel spectral detector or said at least two single-pixel spectral detectors.

[0060] The analysis system can make it possible to produce one or more maps of the distribution of chemical species of the melt pool or the solidified bead in situ during the additive manufacturing or welding process.

[0061] Said spectral detector(s) are preferably provided with a collimation system.

[0062] The analysis system may include two single-pixel spectral detectors arranged to form a predetermined angle between them.

[0063] The ionizing source is preferably an X-ray tube. The ionizing source may be equipped with a collimation system.

[0064] The data processing system may include a computer connected to the detector(s).

[0065] The data processing system can be configured to determine at least one physico-chemical quantity, in particular the quantity and / or concentration, of said at least one chemical species in said at least one given measurement zone, or even in all or part of the melt pool or solidified bead.

[0066] The data processing system can be configured to establish at least a partial, in particular quantitative, map representative of the distribution of said at least one chemical species in the melt pool or the solidified bead.

[0067] The invention also relates, according to another of its aspects, in combination with the above, to an additive manufacturing or welding installation comprising an additive manufacturing or welding system and an analysis system as defined above.

[0068] The additive manufacturing process and the welding process share the common feature of using the shaping of material in a liquid state at very high temperatures, thus creating a molten pool and then a weld bead that solidifies upon cooling. It is this molten pool and / or this solidified weld bead that is analyzed by the analytical method according to the invention and / or using the analytical system according to the invention.

[0069] The additive manufacturing process of a part from the metal alloy can be chosen, without limitation, from the group consisting of an additive manufacturing process by deposition of material under concentrated energy with supply of material in the form of wire or powder (in English “Direct Energy Deposition” (DED)) and an additive manufacturing process on powder bed, in particular selective laser melting (in English “Selective Laser Melting” (SLM)) or electron beam melting (in English “Electron Beam Melting” (EBM)), preferably deposition of material under concentrated energy (DED).

[0070] The welding process of at least two metal parts with possibly the addition of a metal alloy rod close to the two parts considered can be chosen, in a non-limiting way, from the group consisting of TIG welding (for "Tungsten Inert Gas" in English, i.e. Tungsten Inert Gas), MIG welding and MAG welding (respectively for "Metal Inert Gas" in English, i.e. Metal Inert Gas and for "Metal Active Gas" in English, i.e. Metal Active Gas).

[0071] The additive manufacturing or welding system will be adapted for the implementation of the chosen additive manufacturing or welding process. Brief description of the drawings

[0072] The invention will be better understood upon reading the detailed description which will to follow, with non-exhaustive examples of its implementation, and to the examination of the attached drawing, on which

[0073] [Fig.1] [Fig.1] represents, in block diagram, an example of an analysis method according to the invention,

[0074] [Fig.2] Fig.2 schematically represents, in cross-sectional view, an example of an installation for implementing the analysis method according to the invention,

[0075] [Fig. 3] [Fig. 3] represents a graph illustrating the spectrum of an example source ionizing radiation that can be used for the implementation of the invention, with the number of photons Np on the ordinate and the energy e in keV on the abscissa,

[0076] [Fig.4] [Fig.4] represents a graph illustrating the transmission of photons from the K electronic layer of the elements Ni, Fe and Mo as a function of the thickness of an Invar® alloy,

[0077] [Fig. 5] [Fig. 5] is a schematic cross-sectional view of an example of analytical system for implementing the analytical method according to the invention,

[0078] [Fig. 6] [Fig. 6] schematically represents, in top view, a material deposited by additive manufacturing or welding including a melt pool, and an example of mapping of certain measurement areas of said melt pool that can be established using the analysis method according to the invention implemented using the analysis system of [Fig.5],

[0079] [Fig.7] [Fig.7] represents a graph illustrating the ratio of the densities of pro Abilities of the spectral detectors of the analysis system [Fig. 5] as a function of alloy thickness for the Ka lines of iron and nickel,

[0080] [Fig.8] [Fig.8] schematically represents, in cross-sectional view, another example of an analysis system according to the invention for implementing the analysis method according to the invention, and

[0081] [Fig.9] [Fig.9] is a view similar to [Fig.6] illustrating the implementation of the analysis method according to the invention using the analysis system of [Fig.8]. Detailed description

[0082] In the following description, identical elements or elements with identical functions bear the same reference numeral. For the sake of brevity, they are not described opposite each figure; only the differences between the embodiments are described.

[0083] In the figures, the actual proportions have not been respected, for the sake of clarity.

[0084] Figure [1] shows a block diagram illustrating different steps of the analysis method according to the invention.

[0085] The analytical method according to the invention aims to analyze at least one chemical species or even several chemical species, in particular its / their distribution, in a melt bath or a solidified bead produced by an additive manufacturing process or a welding process of a metallic alloy.

[0086] In a step a, ionizing radiation is emitted using at least one ionizing source into at least a portion of the melt pool or solidified bead. The ionizing source acts as an excitation source. The ionizing source, preferably an X-ray tube, preferably emits in an energy range between 1 keV and 300 keV, preferably between 2 keV and 50 keV, and even more preferably between 5 keV and 30 keV.

[0087] In a step b, fluorescent photons thus emitted by said at least a part of the irradiated melt bath or solidified bead are detected using at least one multi-pixel spectral detector or at least two, in particular exactly two, single-pixel spectral detectors.

[0088] Also during step b, a signal representing the fluorescence photon count rate, measured by a spectral detector in at least one given measurement area of ​​the melt pool or solidified bead and for at least one given fluorescence line, is generated by said or each of said spectral detectors.

[0089] The counting rate is measured in counts per second (cps), a measurement related to the number of X-ray photons interacting with the detector. X-ray photons produced by fluorescence or scattering cause ionization in the detector. These charges migrate under the influence of the detector's polarizing electric field and generate current pulses whose amplitude is proportional to the photon energy. A discriminator can separate the pulses according to their amplitude, and thus the incident photons can be counted proportionally according to their energy. They are then arranged from lowest to highest energy in channels, producing a histogram or spectrum.

[0090] It should be noted that said at least a portion of the irradiated molten pool or solidified bead is preferably larger than the measurement area. Measurements can be carried out in several given measurement areas of said at least a portion of the irradiated molten pool or solidified bead.

[0091] In step c, an analysis is performed on the photon counting rate, for example by estimating the net count using a Gaussian fit, which provides information on the quantity of a chemical species originating from a given measurement area. By processing the signal from step b, it is possible to indicate, for example, for each given measurement area, a representative value of said chemical species, for example its concentration.

[0092] It is thus possible to carry out at least a partial mapping of said at least one chemical species in the melt pool or solidified bead established from the signal generated in step b. This allows us to study, for example, the distribution of said at least one chemical species in all or part of the melt pool or solidified bead.

[0093] Step c may include the determination, in at least one given measurement area of ​​the melt pool or solidified bead, of at least one physicochemical quantity, in particular the concentration, of said at least one chemical species and / or a post-processing including a correction step relating to escape photons, stacking phenomena and / or detection efficiency, to then apply a Gaussian product fit to the fluorescence lines of interest allowing an estimation of the net count, i.e. a count measurement without the contribution of noise and finally, via calibration sets, link the net count measurement to a physicochemical quantity such as, for example, the concentration of the chemical species.

[0094] In a particular embodiment of the invention, at least steps a and b are carried out during the additive manufacturing or welding process of the metal alloy, for example, as soon as the metal alloy melts in a given area. This makes it possible to analyze the weld pool with the metal alloy still in its molten state.

[0095] The analysis method may include a step 0, as in the example illustrated in [Fig.1], prior to step a, consisting of determining theoretically, for example by analytical method or by numerical simulation, for a given metallic alloy and a given additive manufacturing or welding process, the depth sensitivity for at least one given chemical species of that alloy.

[0096] The aim is to evaluate the maximum subsurface depth, below the surface of the melt pool, to which a given chemical species of the metal alloy can be detected using the spectral detector(s). This depth sensitivity could correspond, for example, to a transmission rate equal to 5% of the transmitted photon flux reaching the detector. Such a step 0 can make it possible, for each chemical species of interest in the metal alloy under consideration, to estimate the depth to which signal processing can be carried out in step c.

[0097] Figure 2 shows an example of an additive manufacturing or welding installation 5 according to the invention comprising an additive manufacturing or welding system 10 and an analysis system 15 according to the invention.

[0098] The additive manufacturing or welding system 10 includes, as illustrated in this example, at least one device 11 for heating and supplying material in the form of wire or powders of metal alloy A to allow the use of molten material, in liquid form at very high temperatures, so as to create a melt pool B which, upon cooling, gradually forms a solidified bead C, as seen in this figure.

[0099] In the illustrated example, the process implemented by the additive manufacturing or welding system 10 is a laser DED type additive manufacturing process using powder or metal wire as raw material.

[0100] The analysis method according to the invention allows the analysis of the outer surface S of the melt pool B, the analysis in the melt pool B under the outer surface S as well as, where appropriate, in the solidified bead C, for example which has just been solidified, provided that the depth sensitivity of the chemical species concerned allows it and / or that the chemical species have an atomic number between 12 and 92.

[0101] In the illustrated example, the metal alloy is an Invar® alloy with the reference Alloy 36 / K93603 / ASTM F 1684. The so-called chemical species of interest present in this alloy are molybdenum, iron and nickel.

[0102] The heating and material supply device 11, which deposits and heats the material, moves in the direction of movement 12 indicated by the arrow to implement the additive manufacturing process.

[0103] The additive manufacturing installation 5, in this example DED, further comprises an analysis system 15 according to the invention comprising an ionizing source 16 equipped with a collimation system 27 and capable of emitting a beam of ionizing rays, in the illustrated example a divergent beam F of X-rays.

[0104] The analysis system 15 in this example further comprises two single-pixel spectral detectors 17, namely a first spectral detector 17a and a second spectral detector 17b. Each of these spectral detectors 17a and 17b is provided with a pinhole collimation system 18. They are arranged so as to be able to detect fluorescence photons emitted from a measurement area Zm located at the apex of the angle α formed by their respective orientations, as visible.

[0105] The analysis system 15 is attached to the additive manufacturing system 10 by a mechanical system 20 arranged to attach on the one hand the ionizing source 16 and the spectral detectors 17 to each other and on the other hand this assembly of the ionizing source 16 and the spectral detectors 17 to the additive manufacturing system 10. Thus, the additive manufacturing system 10 and the assembly of the ionizing source 16 and the spectral detectors 17 move together as a unit during the implementation of the additive manufacturing process.

[0106] The invention thus makes it possible, in real time and during the implementation of additive manufacturing or welding, to analyze the melt pool B (including the surface S of the melt pool by adjusting the angular orientation of the detectors) and / or the solidified bead C in order to deduce information, in particular the distribution, relating to at least one chemical species of the metal alloy within the melt pool or the solidified cord.

[0107] The analysis system 15 also includes a data processing system 19, illustrated in dotted lines on [Fig.2], which may consist of a computer, which is remote and not necessarily attached to the additive manufacturing system 10, but connected at least to the detectors 17 in order to be able to process the signals generated by them, in particular.

[0108] The main parameters of the additive manufacturing process, in the example considered, are the power of the heating source, which can vary between 0.2 kW and 5 kW, and the scanning speed, which can range from 0.5 cm / s to 10 cm / s. These parameters result in a molten bath with a thickness that can vary between 500 µm and 5 mm. For the sake of clarity, a bath thickness of 1 mm will be considered hereafter.

[0109] To generate a fluorescence signal, the ionizing source 16 is required, consisting in the illustrated example of an X-ray tube with a tungsten anode of maximum energy of 100 keV, the spectrum of the number of photons Np as a function of energy e (expressed in keV) of which is as illustrated in [Fig. 3] in the example considered. The ionizing source 16 is equipped with a collimation system 27, in particular of the single-hole type.

[0110] In this configuration, 60% of the incident photons are attenuated by 1 mm of alloy A, and there is a 1% to 2% probability that their interactions with the material constituting the melt pool will generate fluorescence lines from the K, L, M, etc. electronic layers. The parameterization of the incident beam F thus makes it possible to produce X-ray fluorescence lines throughout the entire thickness of the intended melt. In order for the fluorescence lines produced in an Invar®-type alloy, associated with molybdenum, iron, and nickel, to be collected by the spectral detectors 17, the attenuation that these lines will undergo due to alloy A can be studied, as indicated above in preliminary step 0. Somewhat arbitrarily, it is considered that, to perform the measurement, a minimum of 5% of the fluorescent photon flux must reach the spectral detectors 17.

[0111] For example, if we consider the Ka and K[3] lines, a photon flux of energy of 7.48 keV corresponding to the nickel (Ni Ka) Ka line will be attenuated to 95% by 15 pm of Invar®.

[0112] Similarly, for the higher energy molybdenum Ka fluorescence line (Mo Ka), 17.48 keV, 5% of the photon flux will be transmitted after passing through 92 pm of Invar® thickness.

[0113] Figure 4 shows the transmission rate T of photons from the electronic layer K of molybdenum, nickel and iron (in %) as a function of the thickness Ep, expressed in pm, of the Invar® traversed.

[0114] We can thus find, using this graph of [Fig.4], the depth sensitivity to which can be expected with the elements molybdenum, iron and nickel for a respective transmission of 5%, calculated from NIST data which can be accessed on the website www.nist.gov.

[0115] The table below reproduces these depth sensitivity values:

[0116] [Tables 1] 5% transmission for the fluorescence lines considered: Fluorescence line Ni Ka Ni K[3 Fe Ka Fe K[3 Mo Ka Mo K[3 Invar® thickness (in µm) 15 19 46 59 92 128

[0117] Thus, for the Invar® alloy considered, molybdenum will allow a depth sensitivity of about one hundred micrometers compared to about fifty micrometers for the iron element and about twenty micrometers for the nickel element.

[0118] In order to be able to extract this sensitivity in depth by measurement, one can refer to the analysis system 15 according to the invention shown in isolation on [Fig.5].

[0119] In this example, the angle a between the two detectors 17 is 45°. Moreover, detector 17a in the illustrated example is disposed at 90° to the surface S of the melt bath which it overlooks while detector 17b is disposed at 45° to the surface S of the bath which it overlooks, as seen in [Fig.5].

[0120] The energy-centered count rate of a fluorescence line of interest is defined, with No a reference count rate and Ni, N2 the count rates measured respectively by the two detectors 17b and 17a, such that, according to Beer-Lambert's law:

[0121] [Math.2] = N g e

[0122] [Math.3] N2 —

[0123] with Xi and x2> the thicknesses traversed from the emission point and p the linear attenuation coefficient of the alloy considered at the energy of the fluorescence line of interest, which can be calculated using NIST data.

[0124] The goal being to determine the value xb, we can write:

[0125] [Math.4] N 1 _ a - n2 - A - e

[0126] Referring to [Fig.5], we can write the thickness x2 as follows:

[0127] [Math.5] x 2~ COS^Z

[0128] We can thus take report A and write:

[0129] [Math.6] -In(A)=*](1-cosa)

[0130] Since the value of A is determined by the measurement, we can calculate xp

[0131] [Math.7] ln( A x î ~--7——i PV ' cos^a

[0132] Figure 5 shows the illumination zone O, lit by the ionizing beam at least at some point. Collimating the spectral detectors 17 results in an inspection or measurement zone Zm that is smaller than the illumination zone O, which itself may be smaller than the volume of the molten bath B. The measurement follows the movement of the heat source formed by the heating and material supply device 11, as described above. The measurement is therefore averaged over a measurement area that can be adjusted by collimation. Typically, a measurement area of ​​around 0.03 mm² is targeted.

[0133] As seen previously, to have depth sensitivity of the measurement, the ratio of the counting rates between the two detectors must

[0134] [Math. 8] HAS A = w; is significantly different from unity, and in this example, a variation of at least 5% is estimated to be necessary, that is

[0135] [Math.9] |1 - A | > 5%

[0136] Figure 7 shows the ratio A between two spectral detectors 17 with the same configuration as in Figure 5, again for an Invar® metallic alloy. These curves are from data simulated by a Monte Carlo code for particle transport called PENELOPE (see, for example, the publication on this subject by Sempau, J., et al. entitled "An algorithm for Monte Carlo simulation of coupled electron-photon transport." Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 132.3 (1997): 377-390 and that by Sempau, J., et al. entitled "Experimental benchmarks of the Monte Carlo code PENELOPE." Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms 207.2 (2003): 107-123).

[0137] It can be seen in [Fig. 7] that the values ​​of A are significant for the Ka lines Iron and nickel exhibit a plateau effect at approximately 50 pm and 25 pm alloy thicknesses, respectively; beyond these thicknesses, the bath can be considered infinite. This is due to the self-absorption of spectral lines by the alloy.

[0138] Once the measurements have been taken, they can be reprocessed as indicated above in step c, so as to represent them for example on a map illustrated in [Fig.6] in grey levels.

[0139] For fluorescence lines of interest, in this case Ka of iron, Ka of nickel and Ka of molybdenum, and for each position illustrated on [Fig.6] by three measured zones Zmb Zm2 and Zm3, it is possible to display, as in the map 30 illustrated on the right part of [Fig.6], a grey level representative of the concentration of a given element for each of the measured zones ZmB Zm2 and Zm3.

[0140] The first zone measured in the direction of movement 12 is the Zmi zone for which the concentration of iron, nickel and molybdenum could be quantified.

[0141] The second measured area is the Zm2 area for which it was again possible to carry out measurements of the concentration of iron, nickel and molybdenum.

[0142] The same applies to the zone Zm3 and to all Zm; with k equal to 1,2, [Math. io] n py, even though only the zones Zmb Zm2, Zm3 have been represented here.

[0143] Figures 8 and 9 show another embodiment of the analysis system 15 according to the invention comprising an ionizing source 16 with collimation system 27 and a single multi-pixel spectral detector 35 comprising a plurality of pixels 36 numbered Pb P2, P3, P4, P5. The spectral detector 35 can of course comprise more than five pixels, for example sets ranging from 16x16 pixels to 512 x 512 pixels, arranged in this example in two dimensions, in a matrix.

[0144] The pixel matrix P(i,j) with i,j = 1, 2, ..., n ; [Math. 11] ng 1^1, spatially encodes the measured surface, as illustrated in the (Y, Z) direction, as seen in Figure 8, where Z is the direction of advance of the heating system and Y is the depth. A collimation mask 37 allows, for each pixel P(i,j), a depth sensitivity through the selection of solid angles of the inspected bath zone Zm, thus measuring a count rate N(i,j) varying according to the concentration of the elements of interest in the melt bath B for traversed thicknesses x(i,j). The display of the measurement can then be like the image in Figure 6, with the difference that the surface will not be encoded by a pixel giving, for elements of interest, a count value (a grey level) but by a matrix of pixels giving respectively count values ​​(i.e. a grey level matrix), at a given measurement position Zmt, with k = 1, 2, ..., n; [Math.12] we FJ, visible on [Fig.9]. .

[0145] Figure 9 thus illustrates, for the embodiment of Figure 8, an XRF mapping along a direction, in this example along the Z-axis of the melt pool B with tracking of the movement of the heat source. The display of the measurements is visualized in function of the measured area Zm; with k equal to 1.2, [Math. 13] ne

[0146] Of course, the invention is not limited to the examples just described.

[0147] In particular, the radiation to which the melt pool or solidified bead is subjected via the ionizing source can consist of gamma rays or a sufficiently accelerated electron or ion beam, or a neutron beam.

[0148] The additive manufacturing process may differ from a DED manufacturing process as described above. It may, for example, but not limited to, consist of a powder bed additive manufacturing process, in particular selective laser melting (SLM) or electron beam melting (EBM).

[0149] The process implemented may alternatively be not an additive manufacturing process but a welding process. The welding process may be chosen, without limitation, from the group consisting of TIG welding (for "Tungsten Inert Gas" in English, i.e. Tungsten Inert Gas), MIG welding and MAG welding (respectively for "Metal Inert Gas" in English, i.e. Metal Inert Gas and for "Metal Active Gas" in English, i.e. Metal Active Gas).

[0150] The metallic alloy may be different from Invar®, for example being chosen from the group consisting of iron alloys, copper alloys, in particular bronze or brass alloys, aluminum alloys, titanium alloys, nickel-based superalloys, in particular Inconel 625, Inconel 718, 316L, tungsten carbide alloys, other low coefficient of expansion alloys, in particular those based on iron and nickel, and any other metallic alloy.

[0151] The chemical species analyzed in the melt bath or solidified bead will be adapted according to the metal alloy considered.

Claims

Demands

1. Method for analyzing, by energy-dispersive X-ray spectroscopy, at least one chemical species in a melt pool (B) or a solidified bead (C), said melt pool (B) or said solidified bead (C) being produced by a metal additive manufacturing process or a metal alloy welding process, the analytical method comprising the following steps: a. Step a: emission of ionizing radiation into at least a part of the molten pool or solidified bead using at least one ionizing source (16), b. Step b: - detection, using at least one multi-pixel spectral detector (35) or using at least two single-pixel spectral detectors (17; 17a, 17b), of fluorescence photons emitted by said at least a portion of the fused bath (B) or the solidified bead (C) irradiated, - cross-measurements taken by said at least one multi-pixel spectral detector (35), respectively said at least two single-pixel spectral detectors (17; 17a, 17b) and - generation of at least one signal representing the fluorescence photon counting rate by said multi-pixel spectral detector (35), respectively each of said single-pixel spectral detectors (17; 17a, 17b), in at least one given measurement zone (Zmb Zm2, Zm3, Zmt) of the melt pool (B) or solidified bead (C) and for at least one given fluorescence line, a. Step c: processing of said at least one signal in order to provide at least one piece of information on said at least one chemical species in the melt pool (B) or the solidified bead (C).

2. An analytical method according to claim 1, wherein at least steps a and b are carried out during the additive manufacturing or welding process, in particular from the melting of the metal alloy in at least one measurement zone (Zmi, Zm2, Zm3, Zmt) given of the melt pool.

3. Analysis method according to claim 1 or 2, wherein said spectral detector(s) (35; 17; 17a, 17b) comprise a collimation system (36; 18), the collimation system (36; 18) being selected from the group consisting of a hole collimation system and a mask collimation system.

4. Analytical method according to any one of the preceding claims, wherein, during step a, the ionizing source (16) emits in an energy range between 1keV and 300keV, preferably between 2 keV and 50 keV and more preferably between 5 keV and 30 keV.

5. Analytical method according to any one of the preceding claims, wherein the ionizing source (16) implementing step a is an X-ray tube capable of emitting an X-ray beam and said spectral detector(s) (35; 17; 17a, 17b) implementing step b are configured to detect fluorescence photons emitted by one or more chemical species present in the metal alloy (A) within the melt pool (B) or the solidified bead (C).

6. Analytical method according to any one of the preceding claims, comprising a theoretical determination step prior to step a, relating to an evaluation of measurement sensitivity or depth sensitivity, this preliminary step being carried out by analytical or numerical modelling.

7. Analytical method according to the preceding claim, wherein the preliminary step comprises an evaluation of the depth sensitivity carried out by numerical modeling of the radiation-matter interaction, in particular with regard to the absorption, by the melt pool or the bead, of fluorescence X-ray photons emitted at depth, for a given metallic alloy and a given additive manufacturing or welding process, for at least one given chemical species of said alloy.

8. An analytical method according to any one of the preceding claims, wherein step b and / or c comprises determining the depth sensitivity of the chemical species associated with said fluorescence line, from the energy-centered count rate of a fluorescence line of a given chemical species, measured by the spectral detector(s) (35; 17; 17a, 17b), and taking into account of the arrangement of the multi-pixel spectral detector (35) or of the relative arrangement of the single-pixel spectral detectors (17; 17a, 17b) during measurement.

9. Analysis method according to the preceding claim, employing two single-pixel spectral detectors (17; 17a, 17b) with collimation system (18), method wherein the ratio A between the count rates of the two spectral detectors (17; 17a, 17b) is such that [Math. 14] [1 - A | is greater than or equal to 5%.

10. Analysis method according to any one of claims 1 to 8, implementing at least one multi-pixel spectral detector (35) with collimation system (37).

11. Analytical method according to any one of the preceding claims, wherein step c comprises processing said at least one signal in order to determine at least one physico-chemical quantity, in particular the quantity and / or concentration, of said at least one chemical species in said at least one given measurement zone (Zmi, Zm2, Zm3, Zmt), or even in all or part of the melt pool (B) or the solidified bead (C).

12. Analytical method according to any one of the preceding claims, wherein step c comprises establishing at least a partial, in particular quantitative, mapping representative of the distribution of said at least one chemical species in the melt pool (B) or the solidified bead (C).

13. Analytical method according to any one of the preceding claims, wherein step c comprises a Gaussian fit produced on the fluorescence lines of interest allowing an estimation of the net count and optionally the implementation of calibration sets so as to link the measurement of the net count to a physico-chemical quantity of the chemical species.

14. Product computer program comprising program code instructions which, when executed by a computer, implement at least step c of the analysis method according to any one of claims 1 to 13.

15. Analytical system (15) for the analysis by energy-dispersive X-ray spectroscopy of at least one chemical species in a molten pool or a solidified bead produced by an additive manufacturing process or a welding process of a metallic alloy, in particular an analytical system for implementing the analytical method according to any one of claims 1 to 13, comprising: i. an ionizing source (16) configured to emit ionizing rays into at least a part of the molten pool (B) or the solidified bead (C), ii. at least one multi-pixel spectral detector (35) or at least two single-pixel spectral detectors (17; 17a, 17b) configured to detect fluorescence photons emitted by said at least a portion of the irradiated melt pool (B) or solidified bead (C), and to generate at least one signal representing the fluorescence photon count rate, in at least one given measurement zone (ZmB, Zm2, Zm3, Zmt) of the melt pool (B) or solidified bead (C) and for at least one given fluorescence line and iii. a data processing system configured to provide, from said at least one signal from said or said spectral detectors (35; 17; 17a, 17b), at least one piece of information on said at least one chemical species in the melt pool or solidified bead, including information on a representative value of said at least one chemical species in the melt pool (B) or solidified bead (C).

16. Analysis system (15) according to claim 15, wherein the entire ionizing source (16) and said spectral detector(s) (35; 17; 17a, 17b) is attached to a heating and material supply device (11) of an additive manufacturing or welding installation (5) for the implementation of said additive manufacturing or welding process, or is mechanically fixed following said heating and material supply device (11) in the direction of its movement during the implementation of said additive manufacturing or welding process.

17. Analysis system (15) according to claim 15 or 16, wherein said spectral detector(s) (35; 17; 17a, 17b) are provided with a collimation system (36; 18).

18. Analysis system (15) according to any one of claims 15 to 17, comprising two single-pixel spectral detectors (17; 17a, 17b) arranged to form a predetermined angle (a) between them.

19. Analysis system (15) according to any one of claims 15 to 18, wherein the ionizing source (16) is an X-ray tube and the ionizing source (16) is provided with a collimation system (27).

20. Additive manufacturing or welding installation (5) comprising an additive manufacturing or welding system (10) and an analysis system (15) according to any one of claims 15 to 19.

Images