Method for analyzing liquids containing particles, method for managing a hot-dip galvanizing bath, method for manufacturing hot-dip galvanized steel sheets, and analytical apparatus for liquids containing particles.

Abstract

This method of analyzing a particle-containing liquid is for analyzing a particle-containing liquid by irradiating the same with a laser and running spectrometry on the light emission that occurs at every laser irradiation, the method comprising: a step for focusing the laser; a step for obtaining the light emission intensity of a first element selected from elements included in the liquid and a light emission intensity of a second element selected from elements contained in the liquid and the particles; a step for normalizing the light emission intensity of the second element with the light emission intensity of the first element to obtain a normalized intensity of the second element; a step for creating a frequency distribution graph of the normalized intensity; a step for identifying liquid-induced peaks and particle-induced peaks in the frequency distribution graph; and a step for quantifying the abundance of the particles on the basis of the ratio of a frequency derived from the particle-induced peaks to the total frequency of the frequency distribution graph.

Description

Technical Field

[0001] The present invention relates to a method for analyzing a liquid containing particles, a method for managing a hot-dip galvanizing bath, a method for manufacturing a hot-dip galvanized steel sheet, and an apparatus for analyzing a liquid containing particles. This application claims priority based on Japanese Patent Application No. 2022-202524 filed in Japan on December 19, 2022, the content of which is incorporated herein by reference.

Background Art

[0002] In a molten metal bath such as a hot-dip galvanizing bath in the production of plated steel sheets, monitoring and management of the state of the molten metal is required for quality control of hot-dip galvanized steel sheets, which is one of the products obtained using the bath, and management of operating conditions.

[0003] Patent Document 1 proposes a laser emission spectroscopic analysis method for molten metal as a method for analyzing molten metal during refining. Further, Patent Document 2 discloses a method and an apparatus in which Laser-Induced Breakdown Spectroscopy (LIBS) is applied to the analysis of molten metal materials.

[0004] Further, Patent Document 3 discloses an immersion sensor for analysis of a liquid or a melt to which LIBS is applied, wherein a measurement point is located at a position where the liquid or the melt is newly supplied into a sample chamber from an inlet opening.

[0005] Further, Non-Patent Document 1 discloses a method of obtaining information on only molten Al by taking the Al / Zn ratio from the spectrum of each LIBS laser shot on molten zinc, creating a frequency distribution diagram, and excluding the dross portion.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

[0007] [Non-Patent Document 1] Eric Baril et al, "Novel Method for On-Line Chemical Analysis of Continuous Galvanizing Baths", Galvatech2004 Conference Proceedings, 1095-1104 [Overview of the Initiative] [Problems that the invention aims to solve]

[0008] Molten metals often contain particles on the micrometer scale, and these particles can affect the quality of industrial products manufactured from them. For example, alumina in molten steel and molten aluminum, and oxides and nitrides in molten titanium, can act as inclusions, causing cracks and defects in products. In the case of galvanized steel sheets, which are manufactured by continuously immersing steel sheets in molten zinc, solid matter containing Zn-Al-Fe alloys called dross can adhere to the steel sheets during plating, resulting in defects and poor appearance. Therefore, controlling particles in molten metals is extremely important for improving product quality and yield.

[0009] However, Patent Documents 1-3 do not disclose the analysis of particles in liquids such as molten metal, and it is unclear whether they can analyze the amount of particles present. Furthermore, the technology disclosed in Non-Patent Document 1 is aimed at improving the quantitative accuracy of molten Al, and does not reach the point of separating and quantifying particles and molten Al.

[0010] Conventionally, to quantify particles in molten metal, a portion of the molten metal is first taken as an analytical sample using a ladle or similar tool. Next, this molten metal is solidified using a mold or similar tool. However, during this process, the state of the particles in the sampled analytical material may change due to alloying or other factors. To avoid this change in particle state, rapid solidification is performed by water cooling, or by using a mold made of a material with good thermal conductivity, such as copper. Afterward, the solidified analytical sample is removed from the mold, cut, and polished to obtain an observation surface. This observation surface is then observed using an optical microscope or electron microscope to measure the number (concentration), size, and elemental composition of particles per unit volume. However, this method requires time and effort for cutting, polishing, and observing the sample, and because it involves observing the cross-section of the analytical sample, there are issues with the representativeness of the obtained quantitative values.

[0011] Therefore, the present invention has been made in view of the above problems, and aims to provide a method for analyzing a liquid containing particles, a method for managing a hot-dip galvanizing bath, and a method for manufacturing a hot-dip galvanized steel sheet, which can easily quantify the amount of particles present in the liquid. [Means for solving the problem]

[0012] The inventors have found that by controlling the laser focusing conditions in LIBS, it is possible to distinguish between the emission spectrum of the liquid and the emission spectrum of particles in the liquid. Furthermore, in a frequency distribution graph of the normalized intensity obtained by normalizing the emission intensity of the emission spectrum, the amount of particles in the liquid can be quantified based on the ratio of the frequency originating from the particle-induced peak to the total frequency in the frequency distribution graph.

[0013] Based on the above findings, the gist of the present invention is as follows: [1] A method for analyzing a liquid containing particles according to one aspect of the present invention, comprising: irradiating the liquid containing particles with a laser and analyzing the light emitted each time the laser is irradiated, the method comprising: a laser focusing step of focusing the laser; an emission intensity measurement step of obtaining the emission intensity of a first element selected from elements contained in the liquid and a second element selected from elements contained in the liquid and the particles, and whose concentration in the liquid and the concentration in the particles are different; and the emission intensity of the second element The method includes: a normalization step to obtain a normalized intensity of the second element by normalizing it by the luminescence intensity of the first element; a frequency distribution graph creation step to create a frequency distribution graph representing the frequency distribution of the normalized intensity; a peak identification step to identify a liquid-induced peak, which is a peak caused by the liquid, and a particle-induced peak, which is a peak caused by the particles, in the frequency distribution graph; and a particle quantification step to calculate the sum of the frequencies originating from the particle-induced peaks and to quantify the amount of particles based on the ratio of the frequencies originating from the particle-induced peaks to the total frequency in the frequency distribution graph. [2] The method for analyzing a liquid containing particles as described in [1] above includes a calibration curve preparation step in which a calibration curve is prepared in advance, which is made by preparing liquids containing particles in different amounts, measuring the amount of particles in the liquid by analyzing a sample taken from the liquid, and determining the ratio of the sum of frequencies derived from the particle-induced peak to the total frequency of the frequency distribution graph, thereby creating a calibration curve in advance that shows the relationship between the ratio of the sum of frequencies derived from the particle-induced peak to the total frequency of the frequency distribution graph and the amount of particles, and in the particle quantification step, the amount of particles may be quantified using the calibration curve. [3] In the method for analyzing a liquid containing particles as described in [1] above, in the peak identification step, the liquid-induced peak and the particle-induced peak may be identified by calculating an approximation curve in the frequency distribution graph, and in the particle quantification step, the frequency originating from the particle-induced peak may be quantified from the calculated approximation curve. [4] The method for analyzing a liquid containing particles as described in any of [1] to [3] above may include an average particle diameter measurement step of measuring the average particle diameter of the particles present in the liquid before the laser focusing step. [5] In the method for analyzing a liquid containing particles described in any of [1] to [4] above, the diameter of the laser focusing spot may be set to a value smaller than 10 times the average particle diameter of the particles in the laser focusing step. [6] In the method for analyzing a liquid containing particles described in any of [1] to [5] above, the laser focusing step may be performed by changing the diameter of the laser focusing spot, and the emission intensity measurement step may be performed for each diameter of the focusing spot to determine the laser focusing conditions. [7] In the analytical method for a liquid containing particles described in any of [1] to [6] above, the liquid may be a molten metal. [8] In the analytical method for a liquid containing particles described in any of [1] to [7] above, the liquid may be a molten zinc alloy containing at least aluminum in molten zinc. [9] In the method for analyzing a liquid containing particles described in any of [1] to [8] above, the diameter of the laser focusing spot may be set to less than 300 μm in the laser focusing step.

[10] In the method for analyzing a liquid containing particles described in any of [1] to [9] above, the diameter of the laser focusing spot may be controlled to be 5 μm or more and less than 100 μm in the laser focusing step.

[0014]

[11] A method for managing a molten zinc plating bath according to another aspect of the present invention is a method for managing a molten zinc plating bath, which involves irradiating a molten zinc plating bath containing particles with a laser and analyzing the light emitted each time the laser is irradiated, thereby analyzing the molten zinc plating bath and managing the molten zinc plating bath, comprising: a laser focusing step of focusing the laser; a light emission intensity measuring step of obtaining the light emission intensity of a first element selected from elements contained in the molten zinc plating bath and the light emission intensity of a second element selected from elements contained in the liquid and the particles, the concentration in the liquid and the concentration in the particles being different, and The method includes: a normalization step of normalizing the luminescence intensity of a second element by the luminescence intensity of the first element to obtain a normalized intensity of the second element; a frequency distribution graph creation step of creating a frequency distribution graph representing the frequency distribution of the normalized intensity; a peak identification step of identifying a liquid-induced peak, which is a peak caused by the liquid, and a particle-induced peak, which is a peak caused by the particles, in the frequency distribution graph; and a particle quantification step of calculating the sum of the frequencies originating from the particle-induced peaks and quantifying the amount of particles based on the ratio of the frequencies originating from the particle-induced peaks to the total frequency in the frequency distribution graph.

[0015]

[12] A method for manufacturing a hot-dip galvanized steel sheet according to yet another aspect of the present invention includes a step of managing a hot-dip galvanized bath managed by the hot-dip galvanized bath management method described in

[11] above.

[0016]

[13] An apparatus for analyzing a liquid containing particles according to still another aspect of the present invention is an analysis apparatus for analyzing the liquid containing particles by irradiating the liquid containing particles with laser light and spectroscopically analyzing luminescence generated each time the laser light is irradiated, the analysis apparatus comprising: a laser focusing unit that focuses the laser light; a luminescence intensity measurement unit that obtains a luminescence intensity of a first element selected from elements contained in the liquid, and a luminescence intensity of a second element that is contained in the liquid and the particles and is selected from elements having a concentration in the liquid different from a concentration in the particles; a normalization unit that normalizes the luminescence intensity of the second element with the luminescence intensity of the first element to obtain a normalized intensity of the second element; a frequency distribution graph creation unit that creates a frequency distribution graph representing a frequency distribution of the normalized intensities; a peak identification unit that identifies, in the frequency distribution graph, a liquid-originated peak that is a peak originating from the liquid and a particle-originated peak that is a peak originating from the particles; and a particle quantification unit that obtains a total sum of frequencies derived from the particle-originated peak, and quantifies the abundance of the particles based on a ratio of the frequencies derived from the particle-originated peak to the total frequency of the frequency distribution graph. Effects of the Invention

[0017] As described above, according to the above aspect of the present invention, it is possible to easily quantify the abundance of particles present in a liquid. Brief Description of the Drawings

[0018] [Figure 1] It is a diagram illustrating an example of a configuration of a component analysis system applicable to the method for analyzing particles in a liquid according to one embodiment of the present invention. [Figure 2] It is a functional block diagram illustrating a functional configuration of a signal processing unit. [Figure 3] It is a flowchart illustrating a flow of the particle analysis method according to the same embodiment. [Figure 4] This is an example of an emission spectrum of an Al-containing molten zinc alloy measured by LIBS. [Figure 5] This is another example of an emission spectrum of an Al-containing molten zinc alloy measured by LIBS. [Figure 6] This is an example of a frequency distribution graph in the same embodiment, and is an example of a frequency distribution graph of an Al-containing molten zinc alloy including dross. [Figure 7] The following are examples of peaks identified in the frequency distribution graph of the same embodiment, including peaks caused by dross and peaks caused by liquid. [Figure 8] This is an example of a graph showing a calibration curve in the same embodiment, where the horizontal axis represents the number of dross particles and the vertical axis represents the ratio of the frequency derived from the particle-induced peak to the total frequency in the frequency distribution graph. [Figure 9] This is an example of a graph showing a calibration curve in the same embodiment, where the horizontal axis represents the Al concentration in the Al-containing molten zinc alloy, and the vertical axis represents the normalized intensity at the peak top position of the liquid-induced peak in the frequency distribution graph. [Figure 10] This figure shows an example of a schematic configuration of a hot-dip galvanizing apparatus to which the method for analyzing particles in a liquid according to one embodiment of the present invention can be applied. [Figure 11] This is a plan view showing an example of a schematic configuration of a hot-dip galvanizing apparatus to which the method for analyzing particles in a liquid according to the same embodiment can be applied. [Figure 12] This is a frequency distribution graph for level 1 in the example. [Figure 13] This is a frequency distribution graph for level 2 in the example. [Figure 14] This is a frequency distribution graph for level 3 in the example. [Figure 15] This is a frequency distribution graph for level 4 in the example. [Figure 16] This is a frequency distribution graph of the comparative examples. [Modes for carrying out the invention]

[0019] Preferred embodiments of the present invention will be described in detail below with reference to the attached drawings.

[0020] <Particle analysis system> First, before describing the method for analyzing a liquid containing particles according to one embodiment of the present invention, an example of a particle component analysis system capable of performing the method for analyzing a liquid containing particles according to this embodiment will be described. Figure 1 is a diagram showing an example of the configuration of the component analysis system 11 according to this embodiment. As shown in Figure 1, the component analysis system 11 according to this embodiment has a configuration for performing LIBS and includes, for example, a laser device 12, a cylindrical probe 13, a transmission cable 14, a support member 17, a fixing member 18, a drive unit 19, and a processing device 20. In the description of the particle analysis device, the liquid will be described as molten metal, which is one embodiment thereof.

[0021] The laser device 12 is a device having a laser oscillator (not shown) that irradiates pulsed laser light (hereinafter also simply referred to as "laser"), and an optical system (not shown) that guides the pulsed laser light to a cylindrical probe 13 whose tip is immersed in molten metal 3, and guides the plasma-induced light received via the cylindrical probe 13 to a transmission cable 14. The laser oscillator has the function of generating plasma at the gas-liquid interface between the gas supplied to the molten metal 3 by the cylindrical probe 13 (described later) and the molten metal 3 using the laser it oscillates. Preferably, the laser oscillator has the function of irradiating pulsed laser light that non-selectively evaporates the molten metal 3 and the components of the particles contained in the molten metal 3. The laser oscillator has an output of 10 to 200 mJ / pulse with pulse irradiation such as a Q-switched method, and emits a laser with a wavelength of approximately the fundamental wave to the third harmonic. For example, the laser oscillator may be an Nd:YAG laser or diode laser, which are widely used as high-power pulsed lasers. The pulse frequency should be, for example, around 1 to 50 Hz. The laser pulse width should preferably be several nanoseconds to several psec.

[0022] Furthermore, the laser device 12 may be provided with an adjustment mechanism for adjusting the laser spot diameter and other parameters.

[0023] The optical system includes an optical system for guiding the laser emitted from the laser oscillator to the cylindrical probe 13, and an optical system for guiding the plasma-induced light received through the cylindrical probe 13 to the transmission cable 14. These optical systems are composed of optical components such as lenses, mirrors or dichroic mirrors, or collimators. The optical system may include an adjustment mechanism for adjusting the laser spot diameter, etc.

[0024] The cylindrical probe 13 is a cylindrical member that constitutes the optical path between the molten metal 3 and the laser device 12. One end is connected to the laser device 12, and the other end, which is the open end 131, is immersed in the molten metal 3. The pulsed laser light emitted from the laser device 12 is guided through the cylindrical probe 13 and focused by dichroic mirrors and lenses to focus on the molten metal 3 containing the particles to be analyzed on the open end 131 side. In addition, an inert gas is supplied to the cylindrical probe 13 from an arbitrary position toward the open end 131. As shown in Figure 1, the inert gas is released from the open end 131 into the molten metal 3 and foams up, creating bubbles 15. When the laser guided through the cylindrical probe 13 is irradiated onto the molten metal 3, plasma 16 is generated at the gas-liquid interface between the generated bubbles 15 and the molten metal 3. The elements constituting the particles and molten metal are decomposed to the atomic level by the plasma 16, excited, and emit light when they relax. The inert gas is preferably an inert gas commonly used in plasma emission spectrometry, such as Ar or He. Furthermore, the open end 131 of the cylindrical probe 13 may be perpendicular to the longitudinal direction or angled.

[0025] The transmission cable 14 is a cable for transmitting light, which is emitted as plasma in the molten metal 3 and guided to the laser device 12 via the cylindrical probe 13, to the processing device 20. The transmission cable 14 is made of a general-purpose light guide cable, such as an optical fiber cable.

[0026] The support member 17 supports the laser device 12 and the cylindrical probe 13. The support member 17 can be, for example, a plate-shaped base. The support member 17 is driven by a drive means, such as a built-in motor, of the drive unit 19 (described later), so as to have an angle with respect to the horizontal plane. This makes it possible to change the angle of the cylindrical probe 13 with respect to the vertical direction. The support member is not limited to the illustrated configuration and may be, for example, a member capable of gripping or suspending the laser device 12 and the cylindrical probe 13.

[0027] The fixing member 18 is a rod-shaped member positioned outside the plating tank 4 and fixed in position relative to the plating tank 4. The drive unit 19 is attached to the fixing member 18. The drive unit 19 fixes the support member 17 and is configured to be movable along the fixed member 18. The drive unit 19 moves to any position along the fixed member 18 by a drive means such as a built-in motor. This makes it possible to change the immersion depth of the open end 131 of the cylindrical probe 13 in the plating bath 3. In other words, by making the support member 17 that supports the laser device 12 and the cylindrical probe 13 movable relative to the fixed member 18 using the drive unit 19, the position of the open end 131 of the cylindrical probe 13 becomes controllable.

[0028] The processing unit 20 has the function of spectrally analyzing and detecting the light transmitted through the transmission cable 14, and processing the obtained signal. The processing unit 20 comprises a detection unit 21 and a signal processing unit 22.

[0029] The detection unit 21 spectrally analyzes the light transmitted via the transmission cable 14 and records the spectrally analyzed light as an emission spectrum. The emission spectrum is collected and recorded for each laser shot (pulse). For example, the detection unit 21 can be realized by a spectrometer and a photoelectric converter. The spectrometer has sufficient wavelength resolution and sensitivity to the emission intensity at the wavelength of the particles to be measured and the elements in the molten metal, and is preferably of the Echelle or Czerny-Turner type. The photoelectric converter has an operating speed capable of processing the laser signal for all pulses at a predetermined frequency and has sufficient sensitivity to the wavelengths of the particles to be measured and the constituent elements of the molten metal. The photoelectric converter may be, for example, an optical sensor such as a CCD (Charge Coupled Device) or CMOS (Complementary Metal Oxide Semiconductor) equipped with an image intensifier, or a PMT (Photomultiplier Tube). The detection unit 21 in this embodiment measures signals in wavelength bands that include wavelengths corresponding to the elements contained in the molten metal 3 and the elements contained in the particles, and outputs these signals as measurement data. The measurement data relating to the signals corresponding to each element detected by the detection unit 21 is output to the signal processing unit 22.

[0030] The signal processing unit 22 has the function of processing the acquired measurement data (signals). The signal processing unit 22 calculates the peak intensity of the wavelengths of the particles to be measured and the elements in the molten metal from the collected spectrum, normalizes them, calculates their frequency of occurrence, creates a frequency distribution graph, and separates the peaks. Figure 2 is an example of a functional block diagram showing the functional configuration of the signal processing unit 22. The signal processing unit 22 includes, as shown in Figure 2, a laser focusing unit 221 that focuses a laser, an emission intensity measuring unit 222 that obtains the emission intensity of a first element selected from the elements contained in the molten metal 3 and the emission intensity of a second element selected from elements contained in the molten metal 3 and particles, and whose concentration in the molten metal 3 and the concentration in the particles are different, a normalization unit 223 that normalizes the emission intensity of the second element by the emission intensity of the first element to obtain the normalized intensity of the second element, a frequency distribution graph creation unit 224 that creates a frequency distribution graph representing the frequency distribution of the normalized intensity, a peak identification unit 225 that identifies liquid-induced peaks, which are peaks caused by the molten metal 3, and particle-induced peaks, which are peaks caused by particles, in the frequency distribution graph, and a particle quantification unit 226 that calculates the total frequency originating from the particle-induced peak and quantifies the amount of particles based on the ratio of the frequency originating from the particle-induced peak to the total frequency in the frequency distribution graph. In the laser focusing unit 221, for example, the optical system is controlled to achieve a predetermined diameter (focusing diameter) of the laser's focused spot. Specifically, the focusing device (for example, an optical system or lens system such as a collimator and focusing lens) is controlled to achieve the target focusing diameter. If appropriate focusing conditions are determined in advance through preliminary tests and the optical system is set to those conditions, control of the focusing conditions as a signal processing step is unnecessary.

[0031] Furthermore, the signal processing unit 22 is implemented by hardware including, for example, an arithmetic unit such as a CPU or processor, main memory such as ROM (Read-only memory) or RAM (Random Access Memory), and auxiliary memory such as a hard disk or flash memory. The signal processing unit 22 may be composed of one piece of hardware or multiple pieces of hardware. In addition, the signal processing unit 22 may be implemented by an embedded system.

[0032] <Methods for analyzing liquids> Next, with reference to Figure 3, a method for analyzing a liquid containing particles according to one embodiment of the present invention will be described. Figure 3 is a flowchart showing the flow of the method for analyzing a liquid containing particles according to this embodiment. Below, a method for analyzing a liquid containing particles using the component analysis system 11 described above will be described. However, the present invention is not limited to such examples.

[0033] The method for analyzing a liquid containing particles according to this embodiment involves irradiating the liquid containing particles with a laser and analyzing the light emitted with each laser irradiation by spectral analysis. The method for analyzing a liquid containing particles according to this embodiment includes: a laser focusing step (step S1) for focusing a laser; an emission intensity measurement step (step S2) for obtaining the emission intensity of a first element selected from elements contained in the liquid and the emission intensity of a second element selected from elements contained in the liquid and the particles, where the concentration in the liquid and the concentration in the particles are different; a normalization step (step S3) for obtaining a normalized intensity of the second element by normalizing the emission intensity of the second element with the emission intensity of the first element; a frequency distribution graph creation step (step S4) for creating a frequency distribution graph representing the frequency distribution of the normalized intensity; a particle-induced peak identification step (step S5) for identifying a liquid-induced peak, which is a peak caused by the liquid, and a particle-induced peak, which is a peak caused by the particles, in the frequency distribution graph; and a particle quantification step (step S6) for calculating the sum of the frequencies originating from the particle-induced peaks and quantifying the amount of particles based on the ratio of the frequencies originating from the particle-induced peaks to the total frequency in the frequency distribution graph. The liquid being measured is not particularly limited and may be, for example, water containing particles or molten metal. The molten metal may be a molten zinc alloy (Al-containing molten zinc alloy) in which at least aluminum is contained in molten zinc. Note that the first element is a different element from the second element.

[0034] [Laser focusing process (Step S1)] In the laser focusing process, the diameter of the laser spot (focusing diameter) is focused to an appropriate diameter that allows for the identification of liquid-induced peaks and particle-induced peaks in the peak identification process described later, and the laser is irradiated onto the liquid containing the particles. In this process, it is preferable to adjust the focusing diameter to be smaller than, for example, 10 times the average particle diameter of the particles. This is to prevent the signal caused by the particles from being buried by the signal from the liquid matrix. In this case, it becomes easier to identify liquid-induced peaks and particle-induced peaks in the peak identification process described later, that is, it becomes easier to separate the liquid-induced peaks and particle-induced peaks. For example, if a particle present in the laser focusing spot has a single dimension, and that dimension, i.e., the diameter of the particle, is 1 / 10th the diameter of the laser focusing spot, then the area occupied by the particle in the laser focusing spot will be 1 / 100th. In other words, it is assumed that the intensity of the emission spectrum from the particle will be 1% of the emission spectrum from the non-particle areas, i.e., from the molten metal. On the other hand, in alloyed hot-dip galvanizing baths, the particles being measured are dross, and their composition has been reported to be 44 to 45 mass% Al, 37 to 38 mass% Fe, and 18 to 19 mass% Zn (ISIJ International, Vol. 61 (2021), No. 3, pp. 937-944). Generally, the composition of alloyed hot-dip galvanizing baths is about 0.12 to 0.14 mass% Al and about 0.03 to 0.05 mass% Fe. Therefore, the particles contain Al at a concentration of 314 to 375 times and Fe at a concentration of 740 to 1266 times relative to the molten metal. Thus, if one particle with a diameter of 1 / 10th the laser focusing spot diameter (average particle diameter) is present, the emission intensity of Al will be approximately 3 times higher, and the emission intensity of Fe will be approximately 7 to 12 times higher. On the other hand, for lasers commonly used in LIBS, with an output of several tens of mJ / pulse and a pulse width of several nanoseconds, the stability of emission intensity is generally between a few percent and over ten percent. In the case of molten metal, considering subtle fluctuations in the position of the analysis surface, it is expected that the stability will be 20% or more. Here, the lower limit of quantification in analysis is the minimum amount or concentration at which the analyte can be quantified by a given analytical method, and is generally considered to be 10 times the background variability (standard deviation). If the stability of the emission intensity when no particles are present is 20%, then 10 times that is 200%, meaning that an increase in signal intensity of about twice the original emission intensity does not yield a signal significant enough for quantitative analysis. Here, if the diameter of the particles is greater than 1 / 10 of the diameter of the laser focused spot, it can be calculated that a signal intensity of about 3 times (300%) or more can be obtained for Al, and about 7 to 12 times (700 to 1200%) or more can be obtained for Fe. Therefore, a signal significantly exceeding 200%, which is 10 times the variability of emission intensity, can be obtained, making quantitative analysis possible. If the diameter of the particles is smaller than 1 / 10 of the diameter of the laser irradiation spot, the proportion of the emission signal from the particles in the obtained emission signal decreases exponentially with the square of the particle's diameter, and it becomes insignificant relative to the stability of the emission intensity, so a signal intensity sufficient to distinguish and analyze the particles cannot be obtained. Therefore, although the emission intensity and variability of a laser are greatly affected by the wavelength, power output, pulse width, focusing conditions, and state of the object being analyzed, for quantitative analysis, it is preferable that the size of the laser's focused spot be less than 10 times the average diameter (average particle size) of the particles. For example, if the average particle size is 10 to 20 μm, the diameter of the laser focused spot is preferably around 50 μm. On the other hand, techniques such as cutting with an aperture can be used to focus the laser, but with such techniques, it can be difficult to focus while maintaining the photon density. Therefore, in order to perform laser-induced emission analysis while maintaining analytical sensitivity, the diameter of the focused spot is preferably 5 μm or more. In conventional LIBS, the laser focused diameter is about 300 μm to 1 mm, which is significantly larger than the diameter of particles contained in the liquid. Therefore, with conventional methods, it is difficult to distinguish between liquid and particles and analyze them accurately. In this embodiment, from the viewpoint of high-precision analysis, it is preferable to set the laser focused diameter to less than 300 μm, and more preferably to less than 200 μm, less than 100 μm, less than 80 μm, or less than 60 μm. When the diameter of the laser focused spot is reduced, the frequency of particles being contained within the spot decreases, but it is possible to improve the analytical accuracy by increasing the laser oscillation frequency or appropriately setting the measurement time. Regarding the diameter of the laser spot when irradiating a liquid containing particles, it is preferable to first determine the diameter of the focused beam at the position of the object to be analyzed (the surface position of the liquid containing the particles to be analyzed) using, for example, thermal paper or a laser beam profiler, depending on the laser and focusing device (for example, an optical system or lens system such as a collimator and focusing lens) to be used, and then select or adjust an appropriate optical system according to the average particle size of the particles to be measured. Here, examples of selecting or adjusting the optical system and lens system include selecting or adjusting the aperture and focal length of the lens, and adjusting the position of the lens in the optical axis direction. If the diameter of the laser's focused spot is much larger than the particle size (average particle size), the probability of particles being present directly beneath the laser increases, but the probability of particles not being present directly beneath the laser decreases significantly. Therefore, in the frequency distribution graph creation process described later, it becomes impossible to obtain a frequency distribution graph with two peaks. Accordingly, it is necessary to reduce the diameter of the focused spot to increase the probability that particles are not present directly beneath the laser.On the other hand, if the focused spot diameter is made very small, the probability of a particle being directly beneath the laser also decreases significantly. As a result, in the frequency distribution graph creation process described later, it may not be possible to obtain a frequency distribution graph with two peaks. Therefore, there is an appropriate range for the focused spot diameter of a laser. The average particle size of particles in a liquid can be determined by measuring the particle size distribution, for example, by laser scattering measurement, if the liquid containing the particles is an aqueous solution. In the case of molten metal, the molten metal can be sampled, solidified, cut, and polished. The cut and polished surface can then be measured under a microscope, and the average particle size can be determined using image processing. In the peak identification process, the appropriate spot diameter (focusing diameter) that can distinguish between liquid-induced peaks and particle-induced peaks depends not only on the particle diameter (average particle diameter) but also on the difference in chemical composition between the liquid and particles (difference in the content of the second element described later) and the variation in particle diameter. For example, if the difference in chemical composition is small, a smaller spot diameter is required. Therefore, even if the spot diameter is mechanically set by multiplying the particle diameter by a certain coefficient, it may not be possible to distinguish (separate) between liquid-induced peaks and particle-induced peaks. It is essential to focus the light to a spot diameter that can distinguish between liquid-induced peaks and particle-induced peaks in the peak identification process. If liquid-induced peaks and particle-induced peaks can be distinguished (separated) in the peak identification process, there is no need to adjust the ratio between the average particle diameter and the focused spot diameter. For example, if, by chance, liquid-induced peaks and particle-induced peaks can be distinguished (separated) in the peak identification process when focusing with a certain optical system and lens system through preliminary tests, the focusing conditions obtained by chance with that optical system and lens system may be adopted (without the need to adjust the spot diameter itself).

[0035] In the laser focusing process, it is preferable to focus the laser under multiple focusing conditions in advance, and then perform the emission intensity measurement process, normalization process, frequency distribution graph creation process, and particle-induced peak identification process described later for each focusing condition, in order to determine the optimal focusing condition that can identify (separate) liquid-induced peaks and particle-induced peaks. This allows for optimization of the focusing conditions, i.e., a more appropriate focusing diameter, making it possible to distinguish between liquids and particles with higher precision. For example, it is preferable to control the diameter of the laser focusing spot to 5 μm or more and less than 100 μm.

[0036] [Luminous intensity measurement process (Step S2)] In the emission intensity measurement process, the emission intensity of a first element selected from the elements contained in the liquid and the emission intensity of a second element selected from elements contained in the liquid and particles, and whose concentration in the liquid is different from that in the particles, are obtained. This process is performed using LIBS. Specifically, the emission spectrum is collected for each laser pulse, and from the collected spectrum, the emission intensity of a first element selected from the elements contained in the liquid and the emission intensity of a second element selected from elements contained in the liquid and particles, and whose concentration in the liquid is different from that in the particles, are calculated. Since the emission spectrum includes continuous light generated from the plasma in addition to the emission from each element, the peak intensity is calculated by subtracting the baseline (background due to continuous light from the plasma) and then using the peak height method or the peak area method. The supply of inert gas, laser irradiation, and detection and spectroscopic analysis of plasma emission can be performed using conventionally known methods.

[0037] The first element can be any element contained in the liquid, but it is preferable that it be the element with the highest content among the elements constituting the liquid, as this reduces the change in concentration and improves measurement accuracy. The second element, as described above, is an element contained in both the liquid and the particles, and whose concentration in the liquid is different from that in the particles.

[0038] For example, if the liquid is an aluminum-containing molten zinc alloy, then the aluminum-containing molten zinc alloy is an alloy called dross (Fe2Al5Zn), which has a higher aluminum content than the aluminum-containing molten zinc alloy. x ) Contains many particles. In Al-containing molten zinc alloys containing dross, for example, the emission intensity of these elements is calculated by setting the first element to Zn and the second element to Al. Figure 4 is an example of an emission spectrum estimated when a laser is irradiated onto the molten metal portion of an Al-containing molten zinc alloy measured by LIBS. In Figure 5, the emission intensity of Zn is relatively small and the emission intensity of Al is relatively large compared to the emission spectrum in Figure 4. From this, Figure 5 is an example of an emission spectrum when a laser is irradiated onto an Al-containing molten zinc alloy containing a lot of dross. Since Al-containing molten zinc alloy is a liquid, it easily vaporizes due to the thermal energy added by the laser, and many atoms are excited and emit light in the plasma generated by the laser. Therefore, as shown in Figure 4, when irradiated onto molten metal, the emission intensity of Zn is high. On the other hand, when a laser is irradiated onto dross, since dross is a solid and has a high boiling point, the number of atoms that vaporize and are introduced into the plasma is smaller than that of molten metal. As a result, as shown in Figure 5, when irradiated onto dross, the emission intensity of Zn is low. Furthermore, the ratio of the luminescence intensity of Al to that of Zn shows that the luminescence intensity of Al is relatively higher in dross compared to molten metal. This is because dross, which is an alloy of Al, Fe, and Zn, has a relatively larger proportion of Al compared to molten metal, where Al is dissolved in a relatively dilute solution.

[0039] The emission spectrum includes emission associated with relaxation from an electronically excited state (atomic beam) and emission associated with relaxation from an ionized state (ionic beam). Preferably, the first and second elements are selected from the same type of ionic or atomic beam, and more preferably, elements with similar energy levels leading to excitation are selected. However, the selection is not limited to this, and elements with similar emission intensity or wavelength may also be chosen.

[0040] (Standardization process (Step S3)) In the normalization process, the normalized intensity of the second element is obtained by normalizing the emission intensity of the second element by the emission intensity of the first element. In other words, in the normalization process, the ratio of the emission intensity of the second element to the emission intensity of the first element (emission intensity of the second element / emission intensity of the first element) is calculated.

[0041] (Frequency distribution graph creation process (Step S4)) In the frequency distribution graph creation process, a frequency distribution graph (histogram) is created that represents the distribution of the frequency of occurrence of the normalized intensity of the second element normalized in the normalization process. The emission wavelength and emission intensity of the elements reflect the elemental composition of the laser irradiation site. In this process, in order to classify emission caused solely by the liquid and emission caused by particles in the next process, the particle-induced peak identification process, a frequency distribution graph is created for the elements contained in the particles (the second element) from the aforementioned normalized intensities, with the normalized intensity on the horizontal axis and the frequency of occurrence of each normalized intensity for each pulse on the vertical axis. When quantifying particles, it is preferable to obtain the sum of frequencies originating from particle-induced peaks from the approximation curve, so the width of each bin in the histogram is kept the same. The bins on the horizontal axis of the frequency distribution graph and the number of pulses n per analysis are preferably determined by the size and number of particles in the liquid to be measured, and the degree of liquid flow, so that the frequency distribution graph draws a smooth curve. For example, if the liquid contains particles at a concentration of several tens of ppm, it is preferable to obtain data from several thousand to 10,000 pulses. To obtain higher analytical accuracy, it is preferable to increase the integration time to collect more data or to increase the number of bottles when creating the histogram.

[0042] Figure 6 shows an example of a frequency distribution graph created for an Al-containing molten zinc alloy containing dross, with the Al / Zn intensity on the horizontal axis and its frequency of occurrence on the vertical axis. As shown in Figure 6, a bimodal frequency distribution graph (with two peaks) is obtained for Al-containing molten zinc alloys containing dross.

[0043] (Peak Identification Process (Step S5)) In the peak identification step, liquid-induced peaks, which are peaks caused by the liquid, and particle-induced peaks, which are peaks caused by the particles, are identified in the frequency distribution graph. If there is an overlap in the distribution in the frequency distribution graph, the peaks are separated using the inverse convolution method. Specifically, it is preferable to separate the respective peaks by fitting an approximation curve using known methods such as the Gauss-Newton method, or more specifically, the Levenberg-Marquardt method or Powell's Dog Leg method. In other words, if there is an overlap in the distribution in the frequency distribution graph, it is preferable to separate the liquid-induced peaks and particle-induced peaks by calculating an approximation curve in the frequency distribution graph. In the frequency distribution graph, if the liquid-induced peak and the particle-induced peak do not overlap, it is not necessary to calculate an approximation curve. In such cases, the frequencies originating from the liquid-induced peak and the frequencies originating from the particle-induced peak can be easily determined without separating them, and the particle quantification process described later may be performed. In this embodiment, identifying liquid-induced peaks and particle-induced peaks means separating them using methods such as fitting an approximation curve, as necessary. However, if the liquid-induced peaks and particle-induced peaks do not overlap, identifying liquid-induced peaks and particle-induced peaks means confirming that the liquid-induced peaks and particle-induced peaks do not overlap.

[0044] When a laser is shone only on a liquid, the emission signal contains information about the elements that make up the liquid. When a laser is shone on a portion of the liquid that contains particles, the emission signal contains information about the elements that make up the particles in addition to the liquid. Since particles in a liquid are solids with a different composition than the liquid, when a laser is shone on a portion of the liquid that contains particles, the emission signal will contain different elemental information or have a different emission intensity than when only the liquid is shone. Therefore, the frequency distribution graph created in the frequency distribution graph creation process contains information about the particles in the liquid. For example, if the target is an element common to both the particles and the liquid, and the composition of the particles contains more of that element than the composition of the liquid, the frequency distribution graph will have two peaks in its intensity distribution. In this case, if the content of the second element in the particles is greater than the content of the second element in the liquid, the lower-intensity peak will reflect the liquid, and the higher-intensity peak will reflect the particles. Both peaks exhibit some degree of spread, but while the lower-intensity peaks, originating from the liquid, tend to exhibit a distribution close to a normal distribution based on analytical variability, the higher-intensity peaks, originating from the particles, are often broader, incorporating variability based on particle size in addition to analytical variability.

[0045] Figure 7 shows an example of peaks identified in the frequency distribution graph in this embodiment, and is an example of a peak for dross and a peak for Al-containing molten zinc alloy. When the liquid is an Al-containing molten zinc alloy, as described above, the ratio of the emission intensity of Al to the emission intensity of Zn in the dross (Al / Zn) is relatively larger than the Al / Zn ratio of the molten metal. Therefore, it can be determined that the peak located on the low Al / Zn side is due to molten zinc, and the peak located on the high Al / Zn side is due to dross.

[0046] (Particle quantification process (Step S6)) In the particle quantification step, the sum of frequencies originating from particle-induced peaks is determined, and the amount of particles is quantified based on the ratio of the frequencies originating from particle-induced peaks to the total frequencies in the frequency distribution graph. When calculating an approximation curve of the frequency distribution graph in the peak identification step, it is preferable to quantify the frequencies originating from particle-induced peaks from the calculated approximation curve. In this case, the area of ​​the approximation curve originating from particle-induced peaks may be considered as the frequencies originating from particle-induced peaks, and the sum of the areas of the approximation curves originating from liquid-induced peaks and the approximation curves originating from particle-induced peaks may be considered as the total frequencies in the frequency distribution graph. In other words, "area of ​​approximation curve originating from particle-induced peaks" / ("area of ​​approximation curve originating from liquid-induced peaks" + "area of ​​approximation curve originating from particle-induced peaks") may be considered as the ratio of the frequencies originating from particle-induced peaks to the total frequencies in the frequency distribution graph. If necessary, the abundance of particles may be quantified using parameters other than the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph, provided that these parameters are uniquely determined from the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph. For example, instead of the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph, the abundance of particles may be quantified using the ratio of the frequency originating from liquid-induced peaks to the frequency originating from particle-induced peaks. Furthermore, if the total frequency can be limited to a fixed frequency by fixing the laser pulse frequency, laser pulse width, and LIBS measurement time, including the calibration curve creation process, the abundance of particles may be quantified simply using the frequency of particle-induced peaks (without calculating the total frequency). In this embodiment, these variations are not excluded.

[0047] (Calibration curve creation process (Step S0)) To determine the number density of particles, it is preferable to include a calibration curve creation step, for example, before the particle quantification step. In the calibration curve creation step, it is preferable to prepare liquids containing particles of different abundances in advance, measure the abundance of particles in the liquid (e.g., number density) by analyzing samples taken from the liquids, and create a calibration curve in advance that shows the relationship between the ratio of the frequency derived from particle-induced peaks to the total frequency of the frequency distribution graph obtained by applying the above-described analytical method to the liquid when taking the samples from the liquids, and the abundance of the particles. Multiple standard samples are used that differ in the amount of particles and at least one of the average particle size or particle size distribution. If the liquid is molten metal, the quantification of particles in the standard sample is performed, for example, by observing the cut and polished surface as described above. The components of the molten metal are determined by cutting a part of the analysis sample with a drill or the like, dissolving the cutting powder with acid, and then analyzing it with a high-frequency inductively coupled plasma analyzer or atomic absorption spectrometer. The composition and temperature of the molten metal may also be calculated using a phase diagram or the like. Since the analytical value is the sum of the molten metal and particles, the composition of the molten metal can be calculated by subtracting the quantitative result of the particles obtained by observing the cut and polished surface as described above, or by calculation from the phase diagram. Figure 8 shows an example of a calibration curve graph in this embodiment, where the horizontal axis represents the number of dross particles and the vertical axis represents the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph. Here, the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph was calculated based on the area of ​​the region enclosed by the curve obtained by curve fitting the frequency distribution graph and the x-axis of the frequency distribution graph. In other words, "area of ​​the approximation curve originating from particle-induced peaks" / ("area of ​​the approximation curve originating from liquid-induced peaks" + "area of ​​the approximation curve originating from particle-induced peaks") was considered as the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph. Note that even in the calibration curve creation process, if the liquid-induced peaks and particle-induced peaks do not overlap in the frequency distribution graph, it is not necessary to calculate the approximation curve. If they overlap, it is preferable to calculate the approximation curve. In order to create a highly accurate calibration curve, it is preferable to create a calibration curve in which the laser transmitter, the optical system for light guidance and focusing (e.g., focusing diameter), and the optical system for guiding the plasma-induced light to the transmission cable 14 are different.

[0048] Next, if a calibration curve has been created, in the particle quantification step, the amount of particles present in the liquid containing the particles to be measured is quantified based on the calibration curve, using the ratio of the frequency originating from the particle-induced peak to the total frequency in the frequency distribution graph as a result of the measurement. For example, the number of dross particles in an Al-containing molten zinc alloy can be quantified by applying the ratio of the frequency originating from the particle-induced peak to the total frequency in the measured frequency distribution graph to a calibration curve like the one shown in Figure 8.

[0049] The calibration curve creation step may be omitted. For example, when using this analytical method to reduce defects and appearance defects in plated steel sheets, the calibration curve creation step may be omitted, and the "ratio of frequencies originating from particle-induced peaks to the total frequency in the frequency distribution graph" itself may be used as the control target. The plating bath may be managed or controlled without calculating the amount of particles (dross) using the calibration curve created in the calibration curve creation step. More specifically, if the relationship between the defect rate (or appearance defect rate) of plated steel sheets and the "ratio of frequencies originating from particle-induced peaks to the total frequency in the frequency distribution graph" can be clarified, an appropriate range for the "ratio of frequencies originating from particle-induced peaks to the total frequency in the frequency distribution graph" may be determined in advance according to the clarified content, and the plating bath operating conditions (such as plating bath temperature and chemical composition of the plating bath) may be managed or controlled so that the "ratio of frequencies originating from particle-induced peaks to the total frequency in the frequency distribution graph" falls within the predetermined appropriate range.

[0050] Since there is some variation in the area ratio of particles (dross) to the laser irradiation area in each laser pulse, the width of the particle-induced peak in the frequency distribution graph is often larger than the width of the liquid-induced peak. As described above, the inventors have found that the ratio of the frequency derived from the particle-induced peak to the total frequency in the frequency distribution graph shows a good correlation with the number density of particles in the liquid.

[0051] (Method for determining elemental concentration in liquid (Step S7)) Furthermore, in this method, if the liquid contains a second element, it is also possible to quantify the second element in the liquid from the frequency distribution graph. Figure 9 is an example of a calibration curve graph in this embodiment, where the horizontal axis is the Al concentration in the liquid (Al-containing molten zinc alloy), and the vertical axis is the normalized intensity (luminescence intensity of the second element / luminescence intensity of the first element) at the peak top position caused by the liquid in the frequency distribution graph (the center position of the liquid-induced peak on the horizontal axis in Figure 7). The vertical axis in Figure 9 is the normalized intensity, and the unit of the vertical axis is dimensionless. When quantifying Al in an Al-containing molten zinc alloy, for example, as shown in Figure 9, a correlation (calibration curve) is created in advance between the normalized intensity at the peak top position of the liquid-induced peak and the Al in the liquid (Al-containing molten zinc alloy). Then, by measuring the normalized intensity at the peak top position of the liquid-induced peak using LIBS on the liquid (Al-containing molten zinc alloy) to be measured, the Al concentration in that liquid (Al-containing molten zinc alloy) can be quantified using the pre-created calibration curve. Therefore, by using not only the correlation between the number of particles and the ratio of the frequency originating from the particle-induced peak to the total frequency in the frequency distribution graph, but also the correlation between the concentration of the second element in the liquid and the normalized intensity at the peak top position of the liquid-induced peak in the frequency distribution graph, it is possible to quantify not only the amount of particles in the liquid containing the particles being measured, but also the concentration of the second element in the liquid. Therefore, the method for analyzing a liquid containing particles according to this embodiment preferably further includes a step of quantifying the concentration of an element in the liquid, which quantifies the concentration of the second element in the liquid based on the correlation between the concentration of the second element in the liquid and the normalized intensity at the peak top position of the liquid-induced peak in the frequency distribution graph.

[0052] According to the method for analyzing liquids containing particles in this embodiment, by creating a calibration curve, the cutting, polishing, and observation of samples for each liquid for particle quantification are omitted, saving time and effort. Therefore, particles can be quantified easily.

[0053] <Hot-dip galvanizing equipment> The liquid particle analysis system according to this embodiment can be applied to a hot-dip galvanizing apparatus. Here, an example of a hot-dip galvanizing apparatus will be described. Figure 10 is a side view showing the schematic configuration of a hot-dip galvanizing apparatus 1 to which the liquid particle analysis method according to this embodiment can be applied, and Figure 11 is a plan view showing the schematic configuration of the hot-dip galvanizing apparatus 1 shown in Figure 10. As an example of a molten metal bath, the hot-dip galvanizing bath 3 (hereinafter simply referred to as "plating bath") in the hot-dip galvanizing apparatus 1 will be described as a representative example, but the present invention is not limited thereto and can be applied to any other molten metal bath.

[0054] The hot-dip galvanizing apparatus 1 is a device for continuously depositing molten zinc onto the surface of a steel strip 2 by immersing the steel strip 2 in a plating bath 3 filled with molten zinc. The hot-dip galvanizing apparatus 1 comprises a plating tank 4, a snout 5, a pair of upper and lower support rolls 7, 7, an inductor 8, a gas wiping device 9, an alloying furnace 10, and a component analysis system 11.

[0055] The plating tank 4 stores a plating bath 3 made of molten zinc. In addition to Zn, the plating bath 3 may contain, for example, about 0.12 to 0.15 mass% of Al and about 0.02 to 0.1 mass% of Fe. The temperature of the plating bath 3 is about 440 to 480°C. The snout 5 is positioned at an angle so that one end is immersed in the plating bath 3. The in-bath roll 6 is positioned at the bottom of the inside of the plating tank 4. The in-bath roll 6 rotates along the arrow shown in the figure due to contact with and shearing the steel strip 2.

[0056] The support roll 7 is positioned inside the plating bath 4, downstream of the bath roll 6 in the direction of transporting the steel strip 2, and is positioned to sandwich the steel strip 2 from both the left and right sides as it is fed out from the bath roll 6. The support roll 7 is rotatably supported by bearings (not shown) (e.g., sliding bearings, rolling bearings, etc.). Note that there may be only one support roll, three or more support rolls, or no support rolls at all.

[0057] The inductor 8 is an example of a heating device for heating the plating bath 3 filled in the plating tank 4. As shown in Figure 10, multiple inductors 8 are provided on the side wall of the plating tank 4 to adjust the temperature of the plating bath 3 to a predetermined temperature. Note that the means of heating the plating bath 3 is not limited to the inductor 8, and known technologies can be used.

[0058] The gas wiping device 9 is positioned above the plating tank 4 and has the function of controlling the amount of molten metal adhering to the surface of the steel strip 2 by blowing gas (e.g., nitrogen, air) onto both sides of the steel strip 2 to scrape off the molten metal adhering to the surface of the steel strip 2.

[0059] The alloying furnace 10 is an example of a heating device that heats the steel strip 2 to a predetermined temperature after gas wiping. The alloying furnace 10 raises the temperature of the steel strip 2 by heating, promoting the alloying of the molten metal plating layer attached to the surface of the steel strip 2. Known technologies, such as induction heating heaters, can be used as the alloying furnace 10.

[0060] The steel strip 2, which has been annealed in the upstream annealing furnace, is immersed in the plating tank 4 filled with the plating bath 3 via a snout 5, and is pulled up vertically through the bath rolls 6 and support rolls 7, 7, and transported out of the plating bath 3. After the steel strip 2 is transported out of the plating bath 3, the basis of the molten metal adhering to the surface is adjusted by a gas wiping device 9, and then it passes through the alloying furnace 10.

[0061] As shown in Figure 3, the component analysis system 11 is positioned near the corner of the plating bath 3, but it is not limited to the illustrated configuration and can be positioned at any location on the plating bath 3 and used for measurements at any location on the plating bath 3. The component analysis system 11 is as described above.

[0062] As described above, the method for analyzing a liquid containing particles, which is one embodiment of the present invention, can be applied to a hot-dip galvanizing apparatus and is suitable for managing a hot-dip galvanizing bath. Therefore, in another aspect, this embodiment is a method for managing a hot-dip galvanizing bath, which involves irradiating a hot-dip galvanizing bath containing particles with a laser and performing spectroscopic analysis of the light emitted each time the laser is irradiated, and comprising: a laser focusing step of focusing a laser; a light emission intensity measurement step of obtaining the light emission intensity of a first element selected from elements contained in the liquid and a second element selected from elements contained in the liquid and particles, and whose concentration in the liquid and the concentration in the particles are different; and the light emission intensity of the second element being measured against the light emission intensity of the first element. The present invention also relates to a method for managing a molten zinc plating bath, which includes: a normalization step of obtaining a normalized intensity of a second element by normalizing it; a frequency distribution graph creation step of creating a frequency distribution graph representing the frequency distribution of the normalized intensity; a particle-induced peak identification step of identifying a liquid-induced peak, which is a peak caused by the liquid, and a particle-induced peak, which is a peak caused by the particles, in the frequency distribution graph; and a particle quantification step of calculating the sum of the frequencies originating from the particle-induced peaks and quantifying the amount of particles based on the ratio of the frequencies originating from the particle-induced peaks to the total frequency in the frequency distribution graph. When managing a molten zinc plating bath, if necessary, a calibration curve creation step may be performed to create a calibration curve by preparing liquids containing particles in different amounts in advance, and then the amount of particles may be quantified using the calibration curve in the particle quantification step. Alternatively, when managing a molten zinc plating bath, if necessary, the peak identification step may be performed to identify the liquid-induced peak and the particle-induced peak by calculating an approximation curve in the frequency distribution graph, and then the particle quantification step may be performed to quantify the frequency originating from the particle-induced peak from the calculated approximation curve.

[0063] Furthermore, hot-dip galvanized steel sheets can be suitably manufactured using a hot-dip galvanizing bath controlled in this manner. Therefore, in another aspect, the present invention also relates to a method for manufacturing hot-dip galvanized steel sheets, which includes a step of controlling a hot-dip galvanizing bath controlled by a method for controlling a hot-dip galvanizing bath.

[0064] The present invention has been described above based on preferred embodiments, but the present invention is not limited thereto. The above is merely illustrative, and any device that has a configuration substantially identical to the technical idea described in the claims of the present invention and produces similar effects is included within the technical scope of the present invention.

[0065] For example, in the laser focusing process, it is preferable to adjust the focusing diameter to be less than 10 times the average particle diameter of the particles. To adjust the focusing diameter in this way, an average particle diameter measurement step may be included before the laser focusing process to measure the average particle diameter of the particles present in the liquid. By measuring the particle size distribution of the particles present in the liquid and measuring the average particle diameter of the particles, it is possible to easily control the laser focusing conditions. The average particle diameter can be measured by known methods. For example, methods for measuring fine particles in a liquid include light scattering, light diffraction, microscopy such as electron microscopes and X-ray microscopes, electrostatic methods, sedimentation methods, inertial methods, and diffusion methods. In the case of liquid metal, the size of the fine particles contained can be determined by taking a sample of the liquid metal, rapidly cooling and solidifying it, cutting and polishing it, and observing it under a microscope. If the lower limit of the average particle diameter of the particles present in the liquid is clear, the average particle diameter measurement step is unnecessary, and the focusing diameter may be set to 10 times or less of that lower limit of the average particle diameter.

[0066] Furthermore, it is preferable that the liquid itself is flowing, or that there is a mechanism to induce flow, so that the laser does not repeatedly irradiate the same part of the liquid, but rather irradiates a new surface of the liquid. Mechanisms for inducing flow include utilizing thermal convection, or, if the liquid is molten metal, forcibly agitating it with a metal pump. Alternatively, an efficient method involves immersing the laser and the probe for transmitting the light into the liquid and then blowing in an inert gas to agitate the liquid inserted into the probe, thereby creating a new surface for analysis. This method ensures that the laser-irradiated area is always a new surface of liquid, thus guaranteeing the representativeness of the analysis.

[0067] Furthermore, while the liquid analysis method described above used a liquid containing one type of particle, specifically an Al-containing molten zinc alloy containing dross, as an example, the liquid analysis method described above can also be applied when the liquid contains multiple particles with different compositions. For example, when the liquid contains multiple types of particles with different compositions, a histogram is obtained showing the liquid's peak and multiple peaks representing each of the particles with different compositions. Furthermore, the analytical methods for liquids containing particles described above are not limited to molten metals, but can be broadly applied to suspended particles in liquids. For example, dissolved carbon dioxide in water and suspended organic matter such as microorganisms and microplastics can be separated and quantified by focusing on carbon, which is a common element. Alternatively, suspended matter that forms as a solid in liquid oil due to deterioration of the oil, such as oxidation or saponification, can be separated and analyzed. [Examples]

[0068] The present invention will be described in more detail below with reference to examples. The examples described below are merely examples of the present invention and do not limit it.

[0069] [Example 1] Using a laser emission spectrometer, we analyzed the plating bath used in continuous hot-dip galvanizing using the Zenzimer method. A small amount of Al is added to the molten zinc plating bath, and a small amount of Fe dissolves into the bath to allow the iron plate to pass through. As a result, an alloy called top dross (Fe2Al5Zn) is found near the surface of the plating bath. x The particles were suspended with an average particle size of 10 μm to 20 μm and a particle size distribution width of several μm to approximately 100 μm. Furthermore, the amount of dross in the plating bath was varied by adjusting the components, temperature, and holding time.

[0070] A laser was generated using a Quantel Ultra100 manufactured by Lumibird. The laser oscillation conditions were: wavelength: 1064 nm, power output: 100 mJ / pulse, frequency: 20 Hz. The emitted laser light passed through a 1000 mm long cylindrical probe made of Sialon and was focused by a lens with an anti-reflective coating so that it was focused on the surface of the plating bath. The focal length f of the lens was set to 900 mm, and the laser beam diameter was set to 50 μm.

[0071] The light emitted from the surface of the plating bath upon laser irradiation was transmitted via a cylindrical probe, then shifted 90° from the laser's optical axis using a dichroic mirror, focused with a lens, and finally transmitted to the spectrometer via a 15m long, 32-core optical fiber (Mitsubishi Electric Wire & Cable, Type I standard fiber). A double-grating spectrometer (NP250-2) manufactured by SOL Instrument was used. An ICCD camera (istar) manufactured by Andor was used as the detector.

[0072] For each laser irradiation pulse, the emission intensity of Al (the second element) at a wavelength of 309.27 nm was divided by the emission intensity of Zn (the first element) at a wavelength of 303.58 nm, and this value was denoted as Al / Zn. Spectra were collected 1000 times, and a frequency distribution graph was created with Al / Zn on the x-axis and its frequency on the y-axis. In the frequency distribution graph, 25 bins were used for 1000 data points. Next, an approximation based on the Levenberg-Marquardt method was performed, and it was possible to separate two peaks in the frequency distribution graph, namely the liquid-induced peak and the particle-induced peak.

[0073] For each molten zinc alloy with the Al concentration and dross count shown in Table 1 (Levels 1-4), the liquid-induced peaks and particle-induced peaks (dross peaks) in the frequency distribution graph were identified using the method described above. Since the Al content in the particles (dross) is greater than the Al content in the liquid, and therefore relatively larger than the Al / Zn ratio in the molten metal, the peaks located on the high Al / Zn side were determined to be peaks caused by dross. In addition, the Al concentration and dross abundance of the molten zinc alloy were measured using conventional methods. Specifically, approximately 300g of molten metal for Levels 1-4 was collected using a ladle. The collected analytical samples were rapidly solidified using a copper water-cooled mold. The analytical samples were removed from the mold and cut in half. One half was drilled, 1g of the resulting chips was weighed, completely decomposed with nitric acid and hydrochloric acid, and then subjected to ICP emission spectrometry for quantitative analysis of the Al concentration. The remaining half was then prepared by obtaining a smooth surface using a milling machine and creating a cross-section using a mirror polishing machine. After performing Nital etching on this cross-section, it is observed using an optical microscope, and the number of dross particles (10 μm or larger) per unit area of ​​particles (particles / cm²) is determined. 2 The measurement was taken. Cross-sectional observation was performed on 20 new cross-sections by repeatedly milling and mirror polishing an area of ​​approximately 2 cm x 2 cm. That is, approximately 80 cm 2 The following observations were made. These observations were performed on analytical samples collected under conditions of levels 1 to 4. The average diameter of the dross at levels 1 to 4 was less than 5 μm.

[0074] The upper row of the Reference Value column in Table 1 shows the Al concentration in molten zinc (Al concentration in Table 1) determined by ICP emission spectroscopy, and the lower row of the Reference Value column shows the number of dross particles determined by cross-sectional observation. The upper row of the Examples column in Table 1 shows the normalized intensity at the peak top position of molten zinc in the frequency distribution graph, and the lower row of the Examples column shows the ratio of the frequency originating from particle (dross) peaks to the total frequency in the frequency distribution graph. Figure 12 shows the frequency distribution graph with peak separation for level 1, Figure 13 shows the frequency distribution graph with peak separation for level 2, Figure 14 shows the frequency distribution graph with peak separation for level 3, and Figure 15 shows the frequency distribution graph with peak separation for level 4. Here, as shown in Figures 12 to 15, in levels 1 to 4, the liquid-induced peaks and particle-induced peaks overlapped with each other, so as described above, an approximation curve fitting based on the Levenberg-Marckert method was performed. Furthermore, the area of ​​the approximation curve derived from the particle-induced peak (Area of ​​the approximation curve derived from the liquid-induced peak) plus the area of ​​the approximation curve derived from the particle-induced peak (Area of ​​the approximation curve derived from the particle-induced peak) was considered as the ratio of the frequency derived from the particle-induced peak to the total frequency in the frequency distribution graph.

[0075] [Table 1]

[0076] Here, Figure 8 shows the number of dross particles larger than 10 μm per unit area (particles / cm²) observed using an optical microscope at levels 1 to 4. 2 This shows the relationship between the number of dross particles and the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph. As shown in Figure 8, the number of dross particles is proportional to the ratio of the frequency originating from particle-induced peaks to the total frequency in the frequency distribution graph, so it was found that quantitative analysis of dross can be performed using the calibration curve shown in Figure 8. For this Zenzimer-type continuous molten zinc plating bath, the ratio of the frequency originating from particle (dross) peaks to the total frequency in the frequency distribution graph was measured at multiple different timings using the same laser optical system (i.e., the same focusing conditions) and the same LIBS method as described above. The number of dross particles determined by LIBS was then calculated using the calibration curve in Figure 8. Simultaneously, analytical samples were taken using the same method as described above, and the number of dross particles was measured by cross-sectional observation. The ratio of the number of dross particles determined by LIBS to the number of dross particles determined by cross-sectional observation was within 1.00 ± 0.10, indicating that the number of dross particles determined by LIBS was in good agreement with the number of dross particles determined by cross-sectional observation.

[0077] Furthermore, Figure 9 shows the relationship between the Al concentration in the liquid (Al-containing molten zinc alloy) at levels 1 to 4 and the normalized intensity at the peak top position of the liquid-induced peak. As shown in Figure 9, the Al concentration in the liquid (Al-containing molten zinc alloy) is proportional to the normalized intensity at the peak top position of the liquid-induced peak. Therefore, it was found that quantitative analysis of the Al concentration in molten zinc can be performed using the calibration curve shown in Figure 9.

[0078] On the other hand, for levels 1 to 4, the conventional measurement methods, namely the measurement of Al concentration in molten zinc by ICP emission spectroscopy and the measurement of the number of dross particles by cross-sectional observation, required one operator to work for approximately 10 days. Therefore, when measuring the number of dross particles or Al concentration of molten zinc containing dross, the method according to the present invention allows measurement using a calibration curve with LIBS, making it possible to measure these parameters quickly and easily compared to conventional methods.

[0079] [Example 2] Under Level 1 conditions, the focusing lens was changed to a focusing diameter of 500 μm, and the analysis was performed. A histogram was created by applying the same signal processing as in the inventive example. The results are shown in Figure 15. Although several spectra with a high Al / Zn ratio, which are thought to be due to dross, were detected, the histogram was broad. This is thought to be because the laser focusing diameter was very large compared to the dross, so even if dross was present directly beneath the laser, the emission from the molten metal became dominant and was averaged out in the spectrum. Therefore, it was found that by setting the focusing diameter within an appropriate range, it is possible to distinguish between molten zinc and dross with higher accuracy.

[0080] Although preferred embodiments of the present invention have been described in detail above with reference to the attached drawings, the present invention is not limited to these examples. It is clear to any person with ordinary skill in the art to which the present invention belongs that various modifications or alterations can be conceived within the scope of the technical idea described in the claims, and these are also understood to fall within the technical scope of the present invention. [Explanation of Symbols]

[0081] 1. Hot-dip galvanizing apparatus 2 Steel strip 3. Plating bath (molten metal) 4 Plating tank 5 Snout 6. Roll during bath 7 Support Roles 8 Inductors 9. Gas wiping device 10 Alloying furnace 11-Component Analysis System 12 Laser devices 13. Cylindrical probe 131 Open end 14 Transmission Cables 15 bubbles 16 Plasma 17 Support Member 18 Fixing member 19 Drive unit 20 Processing Units 21 Detection unit 22 Signal Processing Unit 221 Laser focusing section 222 Luminous intensity measurement unit 223 Standardization Section 224 Frequency Distribution Graph Creation Section 225 Peak Identification Section 226 Particle quantification section

Claims

1. A method for analyzing a liquid containing particles, comprising irradiating the liquid containing the particles with a laser and analyzing the light emitted with each laser irradiation, wherein the liquid containing the particles is analyzed by spectral analysis, A laser focusing step for focusing the aforementioned laser, A light emission intensity measurement step to obtain the light emission intensity of a first element selected from the elements contained in the liquid, and the light emission intensity of a second element selected from elements contained in the liquid and the particles, where the concentration in the liquid and the concentration in the particles are different. A normalization step of normalizing the luminescence intensity of the second element by the luminescence intensity of the first element to obtain a normalized intensity of the second element, A frequency distribution graph creation step is to create a frequency distribution graph that represents the distribution of the frequency of the normalized intensity, A peak identification step in the frequency distribution graph to identify liquid-induced peaks, which are peaks caused by the liquid, and particle-induced peaks, which are peaks caused by the particles, A particle quantification step in which the sum of the frequencies originating from the particle-induced peak is calculated, and the amount of the particle is quantified based on the ratio of the frequencies originating from the particle-induced peak to the total frequency of the frequency distribution graph, A method for analyzing liquids containing particles, including [specific particles].

2. A liquid containing particles of different amounts is prepared in advance. By analyzing a sample taken from the liquid, the amount of the particles present in the liquid is measured, When collecting the sample from the liquid, by applying the analytical method described in claim 1 to the liquid, the ratio of the sum of the frequencies derived from the particle-induced peaks to the total frequency of the frequency distribution graph is determined. This includes a calibration curve creation step of creating a calibration curve in advance that shows the relationship between the ratio of the total frequency derived from the particle-induced peak to the total frequency of the frequency distribution graph and the abundance of the particles. The method for analyzing a liquid containing particles according to claim 1, wherein the particle quantification step involves quantifying the amount of particles using the calibration curve.

3. In the peak identification step, the liquid-induced peak and the particle-induced peak are identified by calculating an approximation curve in the frequency distribution graph. The method for analyzing a liquid containing particles according to claim 1 or 2, wherein the particle quantification step involves quantifying the frequency originating from the particle-induced peak from the calculated approximation curve.

4. Before the laser focusing step, A method for analyzing a liquid containing particles according to claim 1 or 2, comprising an average particle diameter measurement step for measuring the average particle diameter of the particles present in the liquid.

5. The method for analyzing a liquid containing particles according to claim 4, wherein in the laser focusing step, the diameter of the laser focusing spot is set to a value smaller than 10 times the average particle diameter of the particles.

6. A method for analyzing a liquid containing particles according to claim 1 or 2, wherein in the laser focusing step, the diameter of the laser focusing spot is changed, the emission intensity measurement step is performed, and for each diameter of the focusing spot, the normalization step, the frequency distribution graph creation step, and the peak identification step are performed to determine the laser focusing conditions.

7. The method for analyzing a liquid containing particles according to claim 1 or 2, wherein the liquid is molten metal.

8. The method for analyzing a liquid containing particles according to claim 1 or 2, wherein the liquid is a molten zinc alloy containing at least aluminum in molten zinc.

9. The method for analyzing a liquid containing particles according to claim 1 or 2, wherein in the laser focusing step, the diameter of the laser focusing spot is set to less than 300 μm.

10. The method for analyzing a liquid containing particles according to claim 1 or 2, wherein in the laser focusing step, the diameter of the laser focusing spot is controlled to be 5 μm or more and less than 100 μm.

11. A method for managing a molten zinc plating bath, comprising irradiating a molten zinc plating bath containing particles with a laser and analyzing the light emitted each time the laser is irradiated, thereby analyzing the molten zinc plating bath and managing the molten zinc plating bath, A laser focusing step for focusing the aforementioned laser, A light emission intensity measurement step to obtain the light emission intensity of a first element selected from the elements contained in the molten zinc plating bath, and the light emission intensity of a second element selected from elements contained in the molten metal of the molten zinc plating bath and the particles, wherein the concentration in the molten metal of the molten zinc plating bath and the concentration in the particles are different; A normalization step of normalizing the luminescence intensity of the second element by the luminescence intensity of the first element to obtain a normalized intensity of the second element, A frequency distribution graph creation step is to create a frequency distribution graph that represents the distribution of the frequency of the normalized intensity, A peak identification step is performed to identify, in the frequency distribution graph, a liquid-induced peak which is a peak caused by the molten metal of the molten zinc plating bath, and a particle-induced peak which is a peak caused by the particles. A particle quantification step, which involves calculating the sum of the frequencies originating from the particle-induced peak and quantifying the amount of the particles based on the ratio of the frequencies originating from the particle-induced peak to the total frequency in the frequency distribution graph, A method for managing a molten zinc plating bath, including the following.

12. A method for manufacturing a hot-dip galvanized steel sheet, comprising the step of managing a hot-dip galvanized bath managed by the hot-dip galvanizing bath management method described in claim 11.

13. An analytical apparatus for analyzing a liquid containing particles by irradiating the liquid with a laser and performing spectroscopic analysis of the light emitted with each laser irradiation, A laser focusing unit for focusing the aforementioned laser, A light emission intensity measuring unit that obtains the light emission intensity of a first element selected from the elements contained in the liquid, and the light emission intensity of a second element selected from elements contained in the liquid and the particles, where the concentration in the liquid and the concentration in the particles are different. A normalization unit that normalizes the luminescence intensity of the second element by the luminescence intensity of the first element to obtain a normalized intensity of the second element, A frequency distribution graph creation unit creates a frequency distribution graph representing the frequency distribution of the normalized intensity, In the frequency distribution graph, a peak identification unit identifies liquid-induced peaks, which are peaks caused by the liquid, and particle-induced peaks, which are peaks caused by the particles. A particle quantification unit calculates the sum of the frequencies originating from the particle-induced peak and quantifies the amount of the particles based on the ratio of the frequencies originating from the particle-induced peak to the total frequency in the frequency distribution graph, An analytical device for liquids containing particles, including those that include particles.

Images