RU2616747C1 - Method for cryolite ratio determination for electrolyte with calcium, magnesium and potassium fluorides additives using xrf method - Google Patents

Abstract

FIELD: chemistry.

SUBSTANCE: invention relates to a method for determination of the cryolite ratio (CO) of molar ratio (NaF+KF)/AlF₃) with added potassium, magnesium and calcium fluorides. The method includes plotting of calibration curves for Na, F, Ca, Mg, using industry standard samples (ISS) of electrolyte composition for aluminum production electrolysers, that have passed metrological certification, and for K - with synthetic electrolyte samples with established procedure of preparation with potassium content error values in the form of regression. Calibration characteristics for fluorine, sodium, calcium, and magnesium are plotted as a regression dependence: C_i=(b, I, +c_i)(1+M)i-Σ-p_iC_i, j≠k≠I, where i is the defined element, j is the element involved in absorption (imposition) of the defined element, k is the element involved in excitation of the defined element, b_i, c_i are coefficients of the regression equation for the i-th element, determined by the least square method, s/imp., wt %, respectively, I_i is the measured intensity of i-th element fluorescent emission, imp./s, C_i is concentration of the i-th potassium, wt %, C_j is concentration of the j-th element of the matrix, wt %, P_j is the intensity overlap correction coefficient, p_j=0÷1; (1+M)_i is the matrix effect coefficient, the value of which is determined by the formula:

where i is the defined element, α_j is the absorption coefficient, which is calculated by the least square method (wt %)^-1; β_k is the excitation coefficient, which is calculated by the least square method; _Ck is the concentration of the k-th element of the matrix, wt %; N is the number of elements, and potassium calibration characteristic is plotted as a regression dependence: C=aI²+bI+c, where: a, b, c are regression equation coefficients for potassium, determined by the least square method (s/imp.)², s/imp., wt %, respectively; I is the measured intensity of analytical potassium line fluorescence, imp./s; C is potassium concentration, wt %. Calibration characteristic for Na determination is plotted with α and β correction by Al, and calibration characteristic for F determination is plotted considering Na imposing and β correction by Al.

EFFECT: achievement of accuracy in СO determination in the electrolyte with addition of K.

2 cl, 5 dwg, 1 tbl

Description

The invention relates to the field of metallurgy, in particular to the production of aluminum by electrolysis of a cryolite melt, and can be used to determine the cryolite ratio (KO), (molar ratio (NaF + KF) / AlF ₃ ) with the addition of calcium and magnesium fluorides of magnesium and potassium.

For the modern development of aluminum production technology, electrolytes with the addition of calcium and potassium fluorides are of great interest. Moreover, the publications consider the value of potassium supplements from 3% mass and higher, since the background value of less than 3% does not give a noticeable change in the physicochemical and thermodynamic properties of the electrolyte. Magnesium enters the electrolyzer with alumina and accumulates in it in a significant amount (up to 1 wt.% In terms of MgF ₂ ), even if the technology does not provide for direct additives of magnesium fluoride. The proposed method works both in electrolytes with the addition of magnesium fluoride, and without it.

The WAMI methodological recommendations (“Methods for determining the cryolite ratio of the electrolyte of aluminum electrolysis cells”, Leningrad. 1974, p. 36) provide three methods for the analysis of KO, which are still used in domestic aluminum plants to determine KO. The accuracy of determining KO by two chemical methods: “hot titration” of molten electrolyte with solid sodium fluoride and “titration with thorium nitrate”, is estimated at 0.05 units. KO (P = 0.95) and serves as a guide for the accuracy of technological control. Moreover, the convergence of the methods (the divergence of the results with parallel determinations of the CO) is estimated at 0.02 units. KO. For the third, diffractometric, method of determining KO, the magnitude of the absolute error is given, determined, as follows from table 1, page 34, for a simplified electrolyte composition (without technologically necessary additives of calcium and magnesium fluorides), which, depending on the range of KO, 0, 03-0.05 units KO. The YOU’s report on the topic No. 5-79-773, stage 20, dated 03/27/1981 indicates the reproducibility of the diffractometric method for determining CO in real samples of industrial electrolyte, characterized by a standard deviation of 0.063 units. KO.

The diffractometric method for the determination of CO in an electrolyte with potassium addition gives insufficient accuracy, since the NaF-KF-AlF system is not sufficiently studied from the point of view of phase formation. When the potassium content is less than 8 wt. % in a chilled sample of electrolyte, the elpasolite KNa ₂ AlF ₆ mainly crystallizes. A further increase in potassium concentration leads to the formation of phases that are absent in the powder diffraction databases.

The X-ray fluorescence method for determining KO by elemental composition in the technological control of the electrolyte composition is not used due to insufficient accuracy and is used only to control the addition of calcium and magnesium fluorides.

A known method for determining the cryolite ratio [RF Patent No. 2424379, IPC С25С 3/06, publ. from July 20, 2011], including the construction of calibration characteristics for Na, F, Ca, Mg, using industry standard electrolyte samples, sampling the electrolyte and preparing the sample for analysis, measuring the fluorescence intensity from the K _α lines of Na, F, Ca, Mg determination of the concentrations of Na, F, Ca, Mg and determination of the cryolite ratio by the concentrations of Na, F, Ca, Mg.

The disadvantages of the method include the fact that it does not take into account potassium additives and, therefore, does not provide the necessary accuracy for the determination of KO in calcium - potassium electrolytes.

Closest to the proposed method in technical essence and the achieved result is a method for determining the cryolite ratio [Feret FR Characterization of bath electrolyte by X-Ray fluorescence. // Light Metals. 1988. The Minerals. Metals & Materials Society. 697-702], including the manufacture of synthetic samples with a known content of Al, F, Na, K, Ca, Mg, simulating the composition of chilled electrolyte samples, measuring the fluorescence intensity of the analytical K _α lines of Al, F, Na, K, Ca on these samples, Mg, construction of calibration characteristics using the De Jonch bond equation, measurement of the fluorescence intensity of analytical lines on electrolyte samples, calculation by the calibration characteristics of the concentrations of the determined elements, calculation of the cryolite ratio from the calculated concentrations lementov.

Since the article does not provide data on the accuracy of calculating the KO from the measured concentrations of electrolyte elements, our estimate of the random component of the error in calculating the KO based on the standard deviations of the analysis of elemental composition given in the article is characterized by a value of ± 0.10 units. abs. KO.

The disadvantages of the method include the following: a) for constructing calibration characteristics, synthetic samples are used that simulate the composition of chilled electrolyte samples that have not passed metrological certification and do not have established errors in the certified value of element concentration; b) the content of potassium fluoride in calibration samples less than 1% does not allow to use the method for electrolytes with the addition of potassium fluoride; c) the method has insufficient accuracy for calculating the cryolite ratio for technological control of the chemical composition.

The technical result of the proposed method is to achieve the necessary accuracy of determining KO in an electrolyte with the addition of calcium fluorides of magnesium and potassium to a value of ± 0.03 units. abs. KO.

The specified technical result is achieved by the fact that in the method for determining the cryolite ratio of the electrolyte with the addition of calcium, magnesium and potassium fluorides by the X-ray fluorescence method, which includes measuring the intensity of analytical lines from Na, F, Ca, Mg, K in calibration samples, constructing calibration characteristics from Na, F , Ca, Mg, K, sampling the electrolyte and preparation of a sample for analysis, measurement of the intensity of fluorescence of K _α radiation lines Na, F, Ca, Mg, K, Na concentration determination, F, Ca, Mg, K and determining the ratio of cryolite according to the concentrations of Na, F, Ca, Mg, K, according to the invention, calibration characteristics are constructed for Na, F, Ca, Mg, using industry standard samples (TOC) of the electrolyte composition of aluminum electrolyzers that have passed metrological certification, and for K - using synthetic electrolyte samples with an error determined by the preparation procedure for the content value in the form of a regression dependence, while the calibration characteristics for fluorine, sodium, calcium, and magnesium are constructed in the form of a regression dependence :

where i is the determined element, j is the element involved in the absorption (overlap) of the determined element, k is the element involved in the excitation of the determined element;

b _i , c _i - coefficients of the regression equation for the i-th element, determined by the least squares method, s / imp., mass. % respectively;

I _i - the measured intensity of the fluorescent radiation of the i-th element, imp./s;

C _i - the concentration of the i-th determined element, mass. %;

C _j is the concentration of the jth element of the matrix, mass. %;

p _j - correction coefficient of overlapping intensity, p _j = 0 ÷ 1;

(1 + M) _i is the coefficient of influence of the matrix, the value of which is determined by the formula:

where i is the determined element, α _j is the absorption coefficient, the value of which is calculated by the least squares method, (% mass.) ^-1 ;

β _k - excitation coefficient, the value of which is calculated by the method of least squares, dimensionless quantity;

C _k is the concentration of the kth element of the matrix, mass. %;

N is the number of elements

and the calibration characteristic for potassium is built in the form of a regression dependence:

where: a, b, c are the coefficients of the regression equation for potassium, determined by the least squares method, (s / imp.) ² , s / imp., mass. % respectively;

I - the measured intensity of the fluorescent radiation of the analytical line of potassium, imp./s;

C is the concentration of potassium, mass. %

The proposed method is complemented by private distinctive features that contribute to the achievement of the specified technical result.

The calibration characteristic for determining Na can be built using α and β correction for Al, and the calibration characteristic for determining F can be built taking into account the superposition of Na and using β correction in A1.

Since the coefficients α _j and β _k are empirical and are introduced to take into account the absorption and excitation of the fluorescent radiation of the element being determined in the sample and are calculated by the least squares method with the help of TOC and synthetic samples of potassium-containing electrolyte when constructing calibration characteristics, their dimensions are chosen so as to comply with the dimension concentration using equation (1).

To calculate the KO by the proposed method, the concentration of Na, F, Ca, Mg, and K is determined in a sample of chilled electrolyte. Next, the amount of fluorine bound in Na, Ca, Mg, and K fluorides is calculated by stoichiometry. Aluminum fluoride (AlF ₃ ) is calculated from the total concentration fluorine minus fluorine bound in fluorides of Na, Ca, Mg and K.

Samples for constructing a calibration characteristic for K are prepared as follows. Take the pre-calculated amount of reagents sodium fluoride (NaF), aluminum fluoride (AlF ₃ ), calcium fluoride (CaF ₂ ), potassium fluoride (KF) and aluminum dioxide (Al ₂ O ₃ ). All reagents of the hch category (chemical purity). Mix thoroughly and alloy in a platinum crucible at a temperature of 950 ° C. The exposure time at 950 ° C is 20 minutes. After that, the crucible is cooled without removing from the muffle furnace to a temperature of 500 ° C, and then in air. To control weight loss, the crucible is weighed before and after melting.

The essence of the method lies in the fact that in the samples of electrolyte selected and prepared for analysis, the fluorescence radiation intensity is measured from the K _α lines of Na, F, Ca, Mg, K, and the concentrations C _{i of the} listed elements are determined from the previously constructed calibration characteristics.

The cryolite ratio, according to certain concentrations of C _i elements, is calculated by the formula:

- концентрации фтористого натрия и фтористого алюминия и фтористого калия соответственно. Криолитовое отношение в данном случае есть мольное отношение суммы фтористого натрия и фтористого калия к фтористому алюминию. Сомножители 2 и 1,45 учитывают, во сколько раз молекулярная масса фтористого натрия и фтористого калия соответственно меньше молекулярной массы фтористого алюминия. Концентрация фтористого натрия вычисляется по формуле:Where

- concentrations of sodium fluoride and aluminum fluoride and potassium fluoride, respectively. The cryolite ratio in this case is the molar ratio of the sum of sodium fluoride and potassium fluoride to aluminum fluoride. Co-factors 2 and 1.45 take into account how many times the molecular weight of sodium fluoride and potassium fluoride is respectively less than the molecular weight of aluminum fluoride. The concentration of sodium fluoride is calculated by the formula:

,

,

where μ _NaF , μ _Na are the molecular weights of sodium fluoride and sodium, respectively, C _Na is the concentration of sodium in the electrolyte sample (determined by the calibration characteristic).

The concentration of potassium fluoride is calculated by the formula:

,

,

where μ _KF , μ _K are the molecular weights of potassium fluoride and potassium, respectively, C _K is the concentration of potassium in the electrolyte sample (determined by the calibration characteristic).

The concentration of aluminum fluoride is calculated by the formula:

,

,

- молекулярные массы алюминия фтористого и фтора соответственно,

- концентрация фтора в алюминии фтористом.Where

- molecular weights of aluminum fluoride and fluorine, respectively,

- the concentration of fluorine in aluminum by fluoride.

The fluorine concentration in aluminum by fluoride is calculated by the formula:

,

,

концентрации фтора во фтористом натрии, фтористом кальции, фтористом магнии и фтористом калии соответственно вычисляются по формулам:where C _F is the fluorine concentration in the electrolyte sample (determined by the calibration characteristic),

fluorine concentrations in sodium fluoride, calcium fluoride, magnesium fluoride and potassium fluoride are respectively calculated by the formulas:

,

,

,

,

,

,

where C _Ca , C _Mg , are the concentrations of calcium fluoride and magnesium fluoride in the electrolyte sample, respectively (determined by calibration characteristics).

Finally, the formula for calculating the cryolite ratio will take the form:

In the proposed method, due to the use of OCO electrolyte that has undergone metrological certification, synthetic electrolyte samples certified according to the preparation procedure, and taking into account the influence of the matrix using expression (3) when constructing calibration characteristics, the accuracy of determination of sodium, fluorine, calcium, magnesium, potassium, allowing to determine KO with an accuracy of ± 0.03 units. abs. KO, that is, sufficient for technological control of the chemical composition of the electrolyte in the production of aluminum.

The method is illustrated by examples and drawings, where in FIG. 1 shows a calibration characteristic for F; in FIG. 2 - calibration characteristic for Na; in FIG. 3 - calibration characteristic for Mg; in FIG. 4 is a calibration curve for Ca, in FIG. 5 - calibration characteristic according to K.

Example 1 (prototype). Since it was not possible to reproduce the conditions by the prototype method, a theoretical estimate was made of the random component of the measurement error of the cryolite ratio using standard deviations obtained when constructing calibration characteristics in the work of Feret FR Characterization of bath electrolyte by X-Ray fluorescence. For this purpose, samples of the composition of the electrolyte C3, C12, made on the basis of the CCA material, were added by adding potassium fluoride. The elemental composition of the samples is presented in table 1. Certified KO values for the preparation procedure are 1.39 and 2.21 units. abs. KO, respectively, with errors Δ _C3 = 0.015 and Δ _C12 = 0.017 units. abs. KO, respectively.

The measurement of these elements by the method of the prototype, taking into account the error will give the following results. C3: F 54.015 ± 0.270; Na - 17.104 ± 0.171; Ca - 2.38 ± 0.047; K - 4.83 ± 0.097% mass. C12: F - 50.779 ± 0.253; Na - 17.046 ± 0.170; Ca - 1.840 ± 0.037; K - 13.897 ± 0.278% mass.

The magnitude of the random component of the error in determining the QO was estimated by the formula (6):

In numerical terms, this will be 0.1 units. abs. KO. Therefore, the KO value for sample C3 can be determined as 1.39 ± 0.1 units. abs. KO, and for sample C12 as 2.21 ± 0.1 units. abs. KO. That is, the absolute discrepancy between the certified and found KO values can be more than 0.1 units. abs. KO.

Example 2 (the proposed method). The fluorescence intensity K _α of the F, Na, Ca, K lines was measured in the same samples C3 and C12 prepared for measurement in the form of an extruded tablet with a diameter of 40 mm and a thickness of 4 mm. Note that samples C3 and C12 were not used to build calibration characteristics. The obtained values of net intensities (intensities of analytical signals minus background intensities) for sample C3: F - 29.859 kpp./s; Na - 64,250 kpp./s; K - 380.083 kimp./s; Ca - 204.218 kpp. / S, for sample C12: F - 23.431 kpp./s; Na - 63.771 kimp / s; K - 1064.642 kimp./s; Ca - 125.5159 kyp./s. According to the calibration characteristics presented in drawings 1-5, find the concentration of elements for sample C3: F - 53.880 ± 0.130; Na - 17,100 ± 0,070; Ca - 2.35 ± 0.015, K - 4.91 ± 0.11 wt. %, for sample C12: F - 51.960 ± 0.130; Na - 17.580 ± 0.070; Ca - 1.980 ± 0.015, K - 13.690 ± 0.11 wt. % The regression equation for fluorine:

C _F = (0.1989I + 44.2904) (1 + 0.0036C _Al ),

b _F = 0.1989; c _F = 44.2904; α _A1 = 0.0036;

regression equation for sodium:

C _Na = (0.2378I + 1.4442) (1 + 0.0063C _Al ) -∑0.0085C _F ,

b _Na = 0.2378; c _Na = 1,4442; α _Al = 0.0063; p _F = 0.0085;

regression equation for potassium:

C _k = 9.45 × 10 ^-7 I ² + 0.0115I + 0.4184,

a _K = 9.45 × 10 ^-7 ; b _K = 0.0115; c _K = 0.4184;

regression equation for calcium:

C _Ca = (0.0090I + 0.0495) (1 + 0.051C _K )

b _Ca = 0.0090; with _Ca = 0.0495; α _K = 0.051;

The value of KO is calculated according to equation (5). The value of the random component of the error in determining the QO is estimated by the formula (6). Calculated for sample C3 KO = 1.41 ± 0.030 units. abs. KO, for sample C12 KO = 2,200 ± 0,030 units. abs. KO. Certified KO values are 1.39 and 2.20 units. abs. KO, respectively. The absolute discrepancy between the certified and found values is 0.02 and - 0.01 units. abs. KO, respectively.

Claims (22)

1. A method for determining the cryolite ratio of an electrolyte with the addition of calcium, magnesium and potassium fluorides by the X-ray fluorescence method, including measuring the intensity of analytical lines from Na, F, Ca, Mg, K in calibration samples, constructing calibration characteristics from Na, F, Ca, Mg, K, , taking an electrolyte sample and preparing a sample for analysis, measuring the fluorescence intensity from the K _α lines of Na, F, Ca, Mg, K, determining the concentrations of Na, F, Ca, Mg, K and determining the cryolite ratio from the concentrations of Na, F, Ca , Mg, K, characterized in that g The calibration characteristics for Na, F, Ca, Mg are constructed using industry standard samples (TOC) of the electrolyte composition of aluminum electrolyzers that have passed metrological certification, and according to K - using synthetic electrolyte samples with an error in the potassium content in the form of regression established by the preparation procedure dependencies, while the calibration characteristics for fluorine, sodium, calcium, and magnesium are built in the form of a regression dependence:

j ≠ k ≠ i where i is the determined element, j is the element involved in the absorption - superposition of the determined element, k is the element involved in the excitation of the determined element; b _i , c _i - coefficients of the regression equation for the i-th element, determined by the least squares method, s / imp., wt. % respectively; I _i - the measured intensity of the fluorescent radiation of the i-th element, imp./s; C _i - the concentration of the i-th determined element, wt. %; C _j is the concentration of the jth element in the matrix, wt. %; p _j - correction coefficient of overlapping intensity, p _j = 0 ÷ 1; (1 + M) _i is the coefficient of influence of the matrix of the sample, the value of which is determined by the formula:

where i is the determined element, α _j is the absorption coefficient, the value of which is calculated by the least squares method, (wt.%) ^-1 ; β _k - excitation coefficient, the value of which is calculated by the method of least squares, dimensionless quantity; With _k is the concentration of the k-th element of the sample matrix, wt. %; N is the number of elements and the calibration characteristic for potassium is built in the form of a regression dependence: C = aI ² + bI + c where: a, b, c are the coefficients of the regression equation for potassium, determined by the least squares method, (s / imp.) ² , s / imp., wt. % respectively; I - the measured intensity of the fluorescent radiation of the analytical line of potassium, imp./s; C is the concentration of potassium, wt. % 2. The method according to p. 1, characterized in that the calibration characteristic for determining Na is built using α and β correction for Al, and the calibration characteristic for determining F is built taking into account the superposition of Na and using β correction for Al.

Images