Application of X-ray fluorescence analyzer in cement production Abstract: As a routine analysis method, X-ray fluorescence spectrum began in 1950s and has developed for 50 years, and now it has become one of the necessary methods for material composition analysis. X-ray fluorescence spectrum analyzer is an advanced instrument analysis method. In recent twenty years, with the development of science and computer technology, this analysis method has been widely used. Advanced foreign countries have widely used X-ray fluorescence spectrometer to analyze the chemical composition of cement products and their raw materials and control the composition of raw materials in the fields of cement scientific research and cement industrial production, and achieved good results. The new dry cement production has the characteristics of high output, stable quality and high degree of automation, which also puts forward higher requirements for product quality. X-ray fluorescence analyzer has the advantages of fast analysis speed, wide detection elements, high accuracy and simple operation. It is an important analytical instrument for quality control of modern cement enterprises. It can simplify laboratory personnel, greatly reduce the labor intensity of analysts, provide timely analysis data for production control and guide production. As on-line quality control of cement production, it has been widely used. In March 2004, Shandong and Wang Chao Cement Co., Ltd. invested in the construction of a new dry production line with a daily output of 2,500 tons of clinker. Completed in February 2005, Venus200X-ray fluorescence analyzer produced by Panaco (formerly Philips Instruments of the Netherlands) was introduced in March to guide the production. After several months of debugging and maintenance, the company has developed and established six working curves of raw meal, clinker, sandstone, clay, red stone and limestone, which can stably analyze the contents of SiO2 _ 2, Al _ 2O _ 3, Fe _ 2O _ 3, CaO, MgO, K2O, Na2O and SO3 in various materials, providing accurate and reliable basis for raw material batching and raw material incoming control in cement enterprises. 1. Instrument configuration and measurement conditions: PW41/kloc-0 Venus 200X-ray fluorescence analyzer, sequential low-power scanning, allowable analysis range Be-U, PW2550 X-ray tube, Sc as rake end window, 300μ m thick beryllium window, equipped with DELL Pentium IV computer, circulating water cooling system, Shanghai Nanjing Cement. See attached table 1 for the measurement conditions of each element. Second, the source factors of errors in material analysis and the solutions to some factors that affect the accuracy of analysis data and the stability of instruments mainly include the following aspects: 1. Sample preparation: The X-ray fluorescence analyzer of PANalytical Company allows the sample to be powder tablet, molten tablet, solid, powder and liquid without being limited by its state and shape. In the application of cement industry, samples can be prepared by powder pressing and melting. Although the melting sheet can eliminate the particle size effect and mineral effect, it can obtain better accuracy. However, it has not been widely used because of its time-consuming operation and high cost. Powder tablet method is simple and fast, and it is the first choice for XRF sample preparation in domestic cement plants at present. Our company adopts grinding and tabletting method. First of all, the sample to be analyzed should be as dry as possible to improve the accuracy of sample preparation. It is required that the sample to be tested be made into powder with fineness not more than 80μ m, and the diameter of the sample should be as large as possible, preferably not less than 32mm. The material must be representative, which can be sampled continuously or at multiple points, fully stirred and reduced to the required grams for inspection. 2. Grinding and tabletting: 2. 1 Determination of grinding time: The sample to be tested must be ground by a special grinder, which can be divided into manual grinding and mechanical vibration grinding. Mechanical vibration grinding has high efficiency, easy control and good sample reproducibility. It is very important to choose a suitable grinding tool for grinding, especially when analyzing trace elements. Our company chooses tungsten carbide abrasive tools, which can minimize the analysis error. Grinding fineness should not be greater than 200 mesh (76um). After repeated experiments and comparisons, we determined that the grinding time of raw clinker and various raw materials was 150s. If the grinding time was too long, it would cause sticky grinding, and if the fineness was too short, it could not meet the analysis requirements. 2.2 Addition of grinding AIDS and binders: Adding appropriate grinding AIDS when grinding samples is helpful to improve grinding efficiency and clean the bowl. The purpose of adding binder is to make the sample better shaped and pressed into a sample with smooth surface and unexpected fracture. Grinding AIDS and adhesives must be organic substances that do not contain the components to be tested. After analysis and test, the suitable additives are triethylamine, alcohol and paraffin. See Table 2-2.3 for details. Sample tabletting: carefully put the prepared powder into the mold and press it into tablets with an automatic press under a certain pressure. X-ray fluorescence intensity is closely related to the pressure and particle size of the pressed sample. Steel ring can be used, the pressure is set at 24t, and the holding time is 30s, which can meet the requirements. 2.4. Verification of sample preparation error After selecting the conditions of sample preparation according to 2. 1 and 2.2, verify the reproducibility of sample preparation. Taking the raw meal as an example, after fully stirring the samples of the same raw material, press 10 samples into the fluorescence analyzer in turn for measurement. Statistics of analysis results are shown in Table 3. It can be seen that the analysis results of the same sample preparation 10 times are extremely high. 2.5 Before the analyzer is used, the validity and accuracy of the instrument stability must be tested. With 50 grams of samples with accurate analysis results, put them into a material bowl (without two steel rings) and run for 3 minutes, then divide them into four parts and make them into tablets for measurement. The measurement results are shown in Table 4. As can be seen from Table 4, the appearance errors of several groups are particularly small, indicating that the stability of the instrument is very good. 2.6. Establishment of application program (standard curve) X-ray fluorescence analyzer is a relative measuring instrument, which calculates the working curve by measuring the standard sample whose chemical analysis results have been accurately known, and processing the characteristic X-ray intensity data obtained by the computer through a series of mathematics. Establishing application program (standard curve) is the basis of accurate analysis of fluorometer, and the following problems should be paid attention to: (1) Selecting representative standard samples. Each set of curves should have at least 8 samples (our company should select at least 10 standard samples when making each set of curves), and take samples in the mining areas used in actual production, so that the physical properties of the analysis samples are the same, and the content range of each element should cover the range that can be achieved in actual production, and be in the middle of production control indicators as far as possible to ensure the accuracy of analysis. (2) Accurate analysis of standard samples. In order to eliminate human error to the greatest extent, standard samples generally need three analysts with more than three years of analytical experience to do parallel chemical analysis, and finally take the average value. Classical wet chemical analysis is the basis for the calibration of almost all rapid spectroscopic techniques, and the chemical analysis value must be accurate, otherwise nothing can be discussed, and the accuracy of the analysis results must be absolutely guaranteed, providing an accurate basis for the calibration of fluorescence analysis. (3) When establishing a new application program (standard curve), operating procedures, various parameters selected by the instrument, sample preparation links of tabletting, environmental conditions, etc. It must be strictly consistent with analysis and application to reduce system error. 2.7. After the application program of calibration curve (standard curve) is established, in order to make the working curve put into normal use, it is necessary to correct the curve, and in order to ensure the accuracy of instrument analysis after use, it is necessary to compare the samples (standard samples) with the fluorescence analysis on a regular basis. If the linear regression is not very good, it is necessary to adjust the working curve slightly by increasing or decreasing the empirical coefficient or standard items. If the working curve is too large, a new working curve must be created. Since our company was put into production in March 2005, in order to ensure the accuracy of instrumental analysis, we insist on making standard samples once a day and comparing them with chemical analysis once every three days. Taking raw meal as an example, the comparison is shown in Table 5. As can be seen from the data in Table 4 and Table 5, the stability and accuracy of the X-ray fluorescence analyzer are good. X-ray fluorescence analyzer has gradually replaced the traditional wet chemical analysis method in the inspection of raw materials and raw materials in our company, which provides accurate and timely analysis results for production and lays a solid foundation for stabilizing raw material batching and improving clinker quality. 2.8 Precautions for sample analysis: 2.8. 1 When preparing samples, the grinding time, additives, set pressure and holding time must be strictly consistent with the state when calibrating the working curve, otherwise the uneven distribution of grinding particle size will bring systematic errors and affect the analysis results. 2.8.2 When grinding different kinds of materials, if the same bowl is used, it must be washed with the sample to be tested or with clean water, otherwise, there will be a big error due to the pollution of other materials. 2.8.3 Before tabletting, the samples must be evenly distributed in the steel ring, and there can be no stacking distribution, otherwise the local difference of tabletting density will be caused, which will affect the analysis results. 2.8.4 The prepared samples should not be polished and cleaned, nor should they be left in the air for too long before measurement, otherwise the measurement results will be affected. 2.8.5 In order to ensure the normal operation of the instrument, the environment must be kept dry and clean. After the sample is pressed, the dust around the pressing ring should be removed as far as possible without affecting the smoothness of the measured surface, and it should be sucked up with a vacuum cleaner to avoid bringing it into the machine, polluting the measurement environment, affecting the analysis results and damaging the service life of the instrument. 2.8.6 Operators must be familiar with the operating procedures first, otherwise the selected analysis program is wrong, the working curve is wrong, and the result will be completely wrong. The operation of sample preparation and tabletting will also bring different degrees of errors, so the operators are required to standardize the operation and make random checks and comparisons from time to time to reduce human errors.