Method for evaluating water content of metallurgical slag auxiliary material by using LIBS (Laser-induced Breakdown Spectroscopy) technology

Abstract

The invention mainly relates to the technical field of metallurgy, and provides a method for evaluating the water content of a metallurgical slag auxiliary material by utilizing an LIBS (Laser-induced Breakdown Spectroscopy) technology, aiming at efficiently and quickly obtaining the water content of the metallurgical slag auxiliary material, the method is characterized in that oxides in the slag auxiliary material are replaced by adopting hydroxide compounds to prepare standard samples with different H contents; laser pulses are excited by the LIBS equipment to burn out test point positions on the surfaces of different samples, spectral information of the test point positions is collected, and a plurality of calibration curves of H atom characteristic spectral line intensity and H content are obtained; fitting on the basis of the plurality of H atom characteristic spectral line intensities and the calibration curve of the H content to obtain a fitting function of the H content and the H atom spectral intensity; the H atomic spectrum intensity of the to-be-detected slag auxiliary material is obtained, the H content in the to-be-detected slag auxiliary material is obtained based on the fitting function, the moisture content of the slag auxiliary material is obtained based on the H content, operation is easy, repeatability is high, sample preparation is not needed, the test analysis period can be greatly shortened, and the method is suitable for the fast-paced production process.

Description

TECHNICAL FIELD

[0001] The present application mainly relates to the field of metallurgical technology, and particularly relates to a method for evaluating water content of slag auxiliary material for metallurgy by using LIBS technology. BACKGROUND

[0002] H2O content is a key index for judging the quality of slag auxiliary material involved in electroslag remelting. If the water content of slag auxiliary material involved in ESR process is too high, various metallurgical defects of electroslag ingot are prone to occur, such as white spots or hydrogen embrittlement caused by H increase. Non-protective atmosphere electroslag furnace can pre-eliminate water vapor in the slag through graphite electrode, so hydrogen-induced defects are not prone to occur. However, protective atmosphere electroslag furnace cannot exchange electrodes during smelting, and usually can only directly melt slag through metal electrodes, so the quality of slag auxiliary material directly affects the quality of electroslag ingot, and metallurgical defects caused by slag auxiliary material are more prone to occur.

[0003] At present, the protective atmosphere electroslag furnace mostly uses pre-melted slag with relatively low melting point, relatively less impurities and water content, to avoid the above problems. However, it is found in production practice that CaO involved in metallurgical slag auxiliary material is prone to react with water vapor in the air to generate Ca(OH)2. Especially in the relatively humid southern region, high-alkalinity powder slag (or porous structure block slag) is prone to adsorb a large amount of water vapor due to the large specific surface area, so that the water content of the slag material exceeds the standard. In theory, block pre-melted slag has relatively weak water absorption compared with powder slag auxiliary material. However, there is a gap between domestic pre-melted slag and imported pre-melted slag in aspects of material selection, sintering process, post-processing and the like. It is found that the domestic pre-melted slag with the same main component ratio is more prone to moisture absorption.

[0004] The slag auxiliary material (including some pre-melted slag with high CaO%) is generally subjected to a long-time drying treatment before electroslag remelting production, which can reduce the water content to a certain extent, but cannot guarantee complete removal. Unlike the dehydration reaction of oxide crystal water, the dehydration reaction of Ca(OH)2 involves chemical bond breaking, and it is difficult to ensure sufficient reaction by conventional drying treatment process, so it is necessary to detect the water content of the dried pre-melted slag to determine whether the pre-melted slag can be used for electroslag remelting. In order to avoid affecting the production rhythm, the detection of H2O content should be completed in a short time.

[0005] The loss on ignition method is a conventional method for detecting the H2O content of a compound, that is, by completely drying the sample at high temperature, the mass difference between the sample before and after drying is obtained, that is, the mass of H2O in the sample. This method is effective for compounds with stable structure and no chemical reaction (or phase change) at high temperature, while once the drying process involves reaction or phase change, it is easy to cause test error. The pre-melted slag involved in the electroslag remelting process often contains a large proportion of CaF2, which will react with water vapor in the air at high temperature to generate HF and CaO, and the rapid volatilization of HF will cause the total mass of the pre-melted slag to change, so it is impossible to determine whether the mass change of the molten slag after high-temperature drying is due to dehydration or chemical reaction.

[0006] Laser-induced breakdown spectroscopy (LIBS) is a real-time in-situ elemental composition qualitative / quantitative analysis technology that has gradually developed in the past two decades, which has the characteristics of no sample pretreatment (or only the minimum treatment); no special requirements for the test environment; and can directly analyze gaseous, liquid and solid samples. The application of LIBS in the metallurgical field has attracted attention in recent years, such as rapid quantitative analysis of steel element composition, trace element detection, element distribution and segregation characterization, and quantitative detection of molten slag (mixed slag) composition. SUMMARY

[0007] The technical problem to be solved by the present application is to more accurately evaluate the water content of slag auxiliary materials for metallurgy based on LIBS technology.

[0008] The technical scheme adopted by the present application to solve the above technical problem is: A method for evaluating the water content of slag auxiliary materials for metallurgy using LIBS technology, the method comprising: Step S1: replacing oxides in the slag auxiliary materials with hydroxides to prepare standard samples with different H contents; Step S2: the LIBS device excites laser pulses to burn test points on the surface of different samples, collects spectral information of the test points, and obtains a calibration curve of the intensity of multiple H atomic characteristic spectral lines and H content; Step S3: obtaining a fitting function of H content and H atomic spectral intensity based on the calibration curve of the intensity of multiple H atomic characteristic spectral lines and H content; Step S4: obtaining the H atomic spectral intensity of the slag auxiliary materials to be tested, obtaining the H content in the slag auxiliary materials to be tested based on the fitting function, and obtaining the water content of the slag auxiliary materials based on the H content.

[0009] Further, the preparation of the standard sample in step S1 comprises: Step S11: replacing the oxides with hydroxides according to the oxide proportion of the raw materials in the slag auxiliary materials, and drying; Step S12: Use a planetary ball mill with Al2O3 grinding balls to thoroughly mix the dried slag and auxiliary materials; Step S13: Compress the mixed residue into tablets as standard samples.

[0010] Furthermore, step S2 includes: Step S21: Place the standard sample in an airtight chamber, fill the chamber with Ar gas, and control the gas pressure to ≤1 atm; Step S22: Turn on the laser and spectrometer in the LIBS device respectively, adjust the lens focal length so that the laser pulse spot is focused on the surface of the standard sample, the laser triggers the spectrometer, and sets the signal acquisition delay time, laser pulse threshold width, energy and frequency; Step S23: The laser pulse burns off the test points on the standard sample to form plasma, and the spectral signal at the test points is collected; Step S24: Obtain calibration curves of the intensity of multiple H atom characteristic spectral lines and H content based on the acquired spectral signals.

[0011] Furthermore, step S23 also includes: performing noise reduction and normalization processing on the acquired spectral signal.

[0012] Furthermore, in step S23, multiple test points are set on the surface of a sample, and a calibration curve is established based on the average spectral signal intensity of multiple test points to determine the intensity of the H atomic characteristic spectral line of the standard sample versus the H content.

[0013] Furthermore, the characteristic spectral lines of the H atom are HA or Hβ spectral lines.

[0014] The beneficial effects of this invention are: This invention uses LIBS (Laser-Induced Plasma Emission Spectroscopy) as a detection method. By replacing the oxides in the slag additives with hydroxides to prepare standard samples, the characteristic spectral lines of H atoms are extracted from the LIBS spectrum of the standard samples. A calibration curve is established, and a fitting function is obtained to obtain the H content and spectral line intensity. This allows for the quantitative analysis of H2O content in slag additives. The method provides rapid, accurate, and low-cost H2O content analysis. The testing method is simple to operate, highly repeatable, and requires no sample preparation, significantly shortening the testing and analysis cycle. It is suitable for fast-paced production processes.

[0015] Instruction manual illustrations Figure 1 The LIBS spectrum of electroslag premelted slag; Figure 2 The un-noise-reduced Hα spectral lines were acquired by a low-sensitivity, high-resolution echelle grating spectrometer. Figure 3 The HA spectrum after noise reduction; Figure 4 This is the calibration curve for the water content of the pre-melted slag. Detailed Implementation

[0016] The present invention provides a method for evaluating the water content of metallurgical slag additives using LIBS technology. The core of solving the above-mentioned technical problem is: replacing the oxides in the slag additive samples with hydroxides to prepare standard samples with different H contents, obtaining standard H atom spectra of different samples, fitting a fitting function between H content and H atom spectral intensity based on the standard spectra, obtaining the H atom spectral intensity of the slag additive to be tested, obtaining the H content in the slag additive to be tested based on the fitting function, and obtaining the water content of the slag additive based on the H content, thereby achieving quantitative detection of water content in slag additives.

[0017] Specifically, the method for evaluating the water content of metallurgical slag additives using LIBS technology according to the present invention includes the following steps: 1) Raw material preparation: Prepare the corresponding oxides and fluorides according to the composition of the pre-melted slag. Select powders with an average size >1000 mesh. If the material is granular or coarse powder, appropriate mechanical or manual grinding is required.

[0018] 2) Sample Preparation: Weigh the raw materials according to the composition ratio of the pre-melted slag. Adjust the H content in the standard sample by replacing some of the oxides in the slag with hydroxides. The range of hydroxide percentage (i.e., H content) in the sample should be set to encompass the H content of the moisture in the test sample as much as possible. To avoid the reaction of water vapor in the air with various oxides, sample preparation and testing should be completed as quickly as possible. If possible, sample weighing, mixing, and subsequent tableting can be carried out in a glove box.

[0019] 3) The sample was thoroughly mixed using a planetary ball mill with Al2O3 grinding balls. Since the H content in the slag was low, the uniformity could be improved by increasing the total slag amount (with the same component ratio) and appropriately extending the grinding time.

[0020] 4) The prepared mixed powder is compressed into tablets using a tablet press and a mold to obtain standard samples. Since no binder is added during this process and the samples are not sintered afterward, the compressed tablets are often relatively loose. The tableting process conditions are adjusted according to the material characteristics, but the overall tableting process should be kept as consistent as possible, and the thickness of the compressed standard sample tablets should be as consistent as possible.

[0021] 5) Place the standard sample prepared in 4) in a small, airtight chamber. The chamber is connected to a vacuum pump, a pressure gauge, and an Ar (or high-purity N2) gas cylinder via pipelines, both of which are equipped with valves. Before testing, turn on the vacuum pump and its corresponding valve to purge the air, then close the vacuum pump's corresponding valve and open the Ar gas cylinder and its corresponding valve to fill the chamber with gas. Repeat this operation three times until the chamber is filled with Ar gas, with the pressure controlled at ≤1 atm. The airtight chamber and Ar atmosphere can minimize the interference of water vapor in the air on the test results.

[0022] 6) The LIBS involved in the test uses a nanometer-scale pulsed laser with a pulse width of 6-8 ns (except for constructing a double-pulse LIBS gain signal, lasers with excessively narrow pulse widths, such as picosecond (ps) or femtosecond (fs) lasers, should not be used); the pulse wavelength is 1064 nm or 532 nm (due to the involvement of an airtight cavity, laser pulses ≤266 nm are not used); the maximum energy output range is 100 mJ; and the pulse frequency is 1-10 Hz. A high-sensitivity signal is used in the spectrometer; if the spectrometer configuration prioritizes resolution, a proportional ICCD should be considered to precisely control the test window. During the test, the signal-to-noise ratio is improved through signal accumulation, and the acquired spectra are denoised. The spectrometer's test range should include HI 656.279 nm.

[0023] 7) Turn on the laser (preheat the laser) and spectrometer in the LIBS equipment (calibrate the spectrometer with an Hg-Ar lamp; if an echelle grating spectrometer is used, perform radiation calibration with a tungsten lamp beforehand); adjust the lens focal length so that the laser pulse spot is focused on the surface of the standard sample; set the trigger mode to laser triggering the spectrometer, set the signal acquisition delay time to 0.8~1.0 ms, and the gate width to 2.5 ms (if the spectrometer is not equipped with an ICCD, do not set the gate width); set the laser pulse energy to 80~100 mJ and the pulse frequency to 1~2 Hz.

[0024] 8) Excite the laser pulse to burn the sample surface to form plasma and collect the spectral signal; Since it is difficult to make the composition distribution of the prepared standard sample completely uniform, it is necessary to collect signals from different positions and then accumulate and average them. Therefore, multiple test points are set in each standard sample, and only 1 to 2 sets of data are collected at each test point.

[0025] 9) Noise reduction processing is performed on the collected raw data. Many algorithms exist for spectral data noise reduction; a suitable algorithm can be selected based on the spectrometer's performance (sensitivity / resolution). In this embodiment, the Riley algorithm is chosen as the data preprocessing method. The processed data is then fitted using Lorentz Fitting or Voigt Fitting to obtain the characteristic spectral lines of H atoms in the standard sample. The characteristic spectral lines are then normalized. A calibration curve is established between the intensity of the characteristic spectral lines of H atoms in the standard sample and the H content. A polynomial fitting is then performed on the calibration curve to obtain the fitting function y=f(i), where y is the H content (wt% or mol%), and i is the intensity of the H atom spectral line corresponding to the test sample. The H₂O content in the test sample is calculated based on the H atom content.

[0026] Example: 1) Preparation of standard samples: In this embodiment, the main raw materials of the pre-melted slag are CaF2, Al2O3, and CaO. A reagent with a mesh size >1000 is selected and dried in a drying oven at 110℃ for 5 hours. A powder is prepared according to a mass ratio (wt%) of CaF2:Al2O3:CaO = 40:30:30. Ca(OH)2 is used to replace CaO in the slag at different molar ratios (mol%) to form mixed reagents with different H2O contents. The different reagents are thoroughly mixed using a planetary ball mill to ensure uniform component distribution. The mixed reagents are then pressed into tablets using a tablet press and a mold to prepare standard samples. The mass of reagents added to the mold is approximately the same, and the tableting conditions are identical to ensure that the thickness of the standard samples is as similar as possible.

[0027] 2) Sample loading: Load the standard sample / test sample into the small sealed cavity and seal the cavity with bolts and rubber gaskets; the top of the sealed cavity is equipped with a plane quartz lens and a three-way valve is connected to one side, which is connected to the vacuum pump and Ar gas cylinder respectively; first turn on the vacuum pump to purge the air in the cavity, and after turning off the vacuum pump, fill the cavity with about 0.1MPa of Ar gas through the Ar gas cylinder. Repeat the above operation three times to fully replace the gas in the cavity.

[0028] 3) LIBS Equipment Debugging: ① Before testing, the laser needs to be preheated (Nd:YAG pulsed laser, pulse width 8ns, wavelength 1064nm, pulse energy output range 1~100mJ, pulse frequency adjustable range 1~10Hz), and the ICCD camera equipped with the spectrometer needs to be cooled (ETB Aryelle200 echelle grating spectrometer, detection wavelength range 380~900nm, resolution λ / Δλ≥8000, the spectrometer has additional signal gain for the 600~700nm wavelength range); ② An in-situ excitation and in-situ signal acquisition optical path design is adopted, and the lens position is adjusted so that the focused laser spot falls on the sample surface in the sealed cavity (bi-convex quartz lens); ③ Spectrometer calibration: Before testing, the spectrometer is calibrated with an Hg-Ar lamp to avoid large shifts in the position of characteristic spectral lines; the spectrometer is calibrated with radiation using a W lamp, and the test file is saved.

[0029] 4) LIBS parameter settings: ① Use laser to trigger the spectrometer's triggering mechanism; ② Laser: Pulse energy output set to 80mJ, pulse frequency set to 1Hz; ③ Spectrometer: Delay time set to 0.8ms, test window set to 2.5ms, signal gain set to 1200.

[0030] 5) Plot the calibration curve: ① Test the standard samples separately, such as... Figure 1 As shown, only one set of spectral signals was collected for each pulse ablation point, and five sets of data were collected for each standard sample surface, obtaining results as follows: Figure 2 The image shows the mean spectral signal; ② Extract the spectral range where α is located, and use the Riley algorithm to denoise the spectral data to improve the signal-to-noise ratio of the α spectral line, obtaining the image shown. Figure 3 The noise-reduced HA spectrum is shown below; ③ The HA spectrum is fitted using the Lorentz function to obtain the spectral intensity value; ④ The H content calibration curve in the pre-melted slag is plotted: as shown below. Figure 4 As shown, a curve is plotted with the intensity of the α spectral line as the Y-axis and different H atom contents as the X-axis; then, a polynomial fitting is performed on the curve to obtain the fitting function.

[0031] 6) Perform the above operations on the test sample, and substitute the obtained HA spectral line intensity value into the fitting function to obtain the H2O content based on the H atom content.

Claims

1. A method for evaluating the water content of metallurgical slag additives using LIBS technology, characterized in that, The method includes: Step S1: Replace the oxides in the slag additives with hydroxides to prepare standard samples with different H contents; Step S2: The LIBS device excites laser pulses to burn test points on the surface of different samples, collects spectral information of the test points, and obtains calibration curves of multiple H atom characteristic spectral line intensities and H content; Step S3: Obtain the fitting function of H content and H atom spectral intensity based on the calibration curve fitting of multiple H atom characteristic spectral line intensities and H content; Step S4: Obtain the H atom spectral intensity of the slag additive to be tested, obtain the H content in the slag additive based on the fitting function, and obtain the moisture content of the slag additive based on the H content.

2. The method for evaluating the water content of metallurgical slag additives using LIBS technology according to claim 1, characterized in that, The preparation of standard tests in step S1 includes: Step S11: Replace the oxides with hydroxides according to the proportion of oxides in the raw materials in the slag auxiliary materials, and then dry them; Step S12: Use a planetary ball mill with Al2O3 grinding balls to thoroughly mix the dried slag and auxiliary materials; Step S13: Compress the mixed residue into tablets as standard samples.

3. The method for evaluating the water content of metallurgical slag additives using LIBS technology according to claim 1, characterized in that, Step S2 includes: Step S21: Place the standard sample in an airtight chamber, fill the chamber with Ar gas, and control the gas pressure to ≤1 atm; Step S22: Turn on the laser and spectrometer in the LIBS device respectively, adjust the lens focal length so that the laser pulse spot is focused on the surface of the standard sample, the laser triggers the spectrometer, and sets the signal acquisition delay time, laser pulse threshold width, energy and frequency; Step S23: The laser pulse burns off the test points on the standard sample to form plasma, and the spectral signal at the test points is collected; Step S24: Obtain calibration curves of the intensity of multiple H atom characteristic spectral lines and H content based on the acquired spectral signals.

4. The method for evaluating the water content of metallurgical slag additives using LIBS technology according to claim 3, characterized in that, Step S23 also includes: performing noise reduction and normalization processing on the acquired spectral signal.

5. A method for evaluating the water content of metallurgical slag additives using LIBS technology according to claim 4, characterized in that, In step S23, multiple test points are set on the surface of a sample, and a calibration curve of the intensity of the characteristic spectral line of H atoms in the standard sample versus the H content is established based on the average spectral signal intensity of the multiple test points.

6. A method for evaluating the water content of metallurgical slag additives using LIBS technology according to any one of claims 1-5, characterized in that, The characteristic spectral lines of the H atom are Hα or Hβ lines.

Images