RU2314516C2 - Method of determining mass fractions of components in materials and alloys - Google Patents

Abstract

FIELD: spectral analysis.

SUBSTANCE: method comprises exciting specimen radiation in the low-temperature plasma, recording spectrum of specimen radiation, measuring intensity of analytic line of the element and comparison line, calculation of the concentration of the element in the sample from the physical model that includes terms for parameters characterizing the stable state of the low-temperature plasma in the standard specimen with respect to the sample and capability of the specimen to emit the low-temperature plasma with respect to the standard specimen for each element.

EFFECT: enhanced precision.

1 dwg

Description

The invention relates to atomic emission spectral analysis of materials and alloys, namely, to determine the content of mass fractions of elements in materials and alloys.

In classical atomic emission analysis, a number of methods are known for determining the content of mass fractions of elements in materials and alloys.

For example, there is a known method for determining mass fractions in emission spectral analysis of materials and products, based on the empirical Lomakin-Sheybe formula — representing the dependence of the intensity I of low-temperature plasma radiation (NTP) on the mass fractions of Ci elements in the material under study [1].

According to this method, the calculation of the parameters of the mass fractions of the elements Ci is carried out according to the calibration graphs I = f (logCi) in small intervals of variation of Ci, and the boundaries of these intervals are determined experimentally by the appropriate selection of standard reference samples (CO).

The disadvantage of this method is the need to have a large number of standard samples, which leads to additional time spent on their analysis and creates certain difficulties in express analyzes.

Technical solutions are known in which, using a single set of COs, about 50 grades of aluminum alloys are analyzed on multichannel atomic emission spectrometers. A method has been developed for constructing empirical calibration characteristics on large ranges of changes in the content of such elements as copper, zinc, magnesium [2].

Due to the fact that such a construction using one analytical line for some elements is impossible due to reabsorption and self-absorption, the problem is solved using two analytical lines of the element being determined. Non-linearity was taken into account using polynomial models of calibration characteristics (GC), the coefficients of which were calculated by the method of least squares.

To increase the accuracy of determination of impurities in aluminum alloys using a single set of state SO (GSO), instrumental weighing was introduced, the essence of which is reduced to a set of such GC, by which, when determining the composition of the sample, the maximum probability of obtaining actually observed element contents is ensured. "Instrumental" weighing allows you to reduce the number of used sets of CO, calibration graphs that are constantly in the computer memory, or special tables.

The optical spectral analysis method described above is based on indirect measurements of the chemical composition, and the relative intensity of the spectral line of this element when there is a fundamental relationship between them is taken as a measure of the percentage of the element being determined in the alloy sample.

The connection is established according to the calibration of the spectrometer using a GSO kit of known composition.

However, the implementation of this method requires:

a large amount of work related to the recording of computer calibration characteristics;

the presence of a set of several WITH;

when analyzing the spectrum excitation of not only the sample, but also the GSO set, as well as measuring the intensity of the spectral lines of the sample and GSO.

The closest solution, taken as a prototype, is a method for determining the content of mass fractions of elements in materials and alloys, in which the amount of CO is reduced to one [3].

The known method includes a sample radiation excitation, measuring the blackening of the main line element and the comparison line, calculating the content of the test cell in the sample, recording the emission spectrum SB for determining the parameter M _e characterizing ability to radiation NTP sample relative to CO, for each element, and a parameter (AH) _ep , characterizing the steady state of NTP in CO in relation to the sample, related by the ratio

where M = (2 / π) arctan {[S _i (S _i + S _icp - ΔS _e ) / S _i (S _i + S _iav + ΔS _e )] C _i };

b _uh = 1- (1 / π) arctg [(AH) _uh C _ij ];

C _IE , ΔS _e - element content and the difference in blackening of the main line and the comparison line in CO;

S _i , S _icp - blackening of the main line and the comparison line for the sample;

B, D - proportionality coefficients, determined by the table of correspondence according to the value of M, and calculate the approximate content of the desired element in the sample C _x according to the system of equations:

S ₀ = 250 - the maximum blackening determined by a microphotometer (nameplate value).

An analysis of the known methods for determining the mass fraction of elements in materials and alloys shows that the use of the currently existing semi-empirical theory of calculating the quantitative content of elements does not provide for the totality of existing volume interactions of particles in a cloud of scientific and technological progress, accompanied by energy transformations of atoms and ions. That is why the range of reliable analysis by the control JI method is limited, which significantly reduces the accuracy and reliability of the analysis when using industry-specific JI and JI enterprises.

The limitation of their application is also determined by internal energy transformations when the quantitative ratios of the components of the samples change. All this leads to the fact that even in a limited range of concentration changes, errors may occur in the determination of C _x .

The objective of the invention is to further increase the accuracy and reliability of determining the quantitative content of elements in materials and alloys.

The problem is achieved by introducing into consideration isolated calculation systems and the method of successive approximations, taking into account different matrix compositions and possible differences in the experimental conditions for the standard and the sample. To calculate a wide range of changes in the concentrations of the analyzed elements, one control standard is used.

To determine the content of mass fractions of elements in materials and alloys, a method is used that includes excitation of the radiation of the sample in a low-temperature plasma, registration of the emission spectrum of the sample, measurement of the intensity of the analytical line of the element and the comparison line, calculation of the content of the desired element in the sample, according to a physical model containing the expression of the parameter ( αχ) _eh , characterizing the stable state of low-temperature plasma in a standard sample with respect to the sample, and the parameter M _ie , characterizing the method the ability to emit low-temperature plasma relative to the standard sample for each element, determined by the relations

Where

;

;

С _ei , _{ΔР ei} - element content and intensity difference

the analyzed line and the line of comparison of the element i in the standard sample;

P _xi , P _xicp — intensities of the analyzed line and the line of comparison of element i in the sample;

B, D - proportionality coefficients determined by the parameter M _ei , calculate the energy parameters of radiation of the sample relative to the standard L ′ _he and L ′ _eh

make adjustments and the calculation of the new values of the intensities of the spectral lines in the sample P _x and P _HSR so that

wherein? P _x, ΣP _x, - the difference and the sum of the intensities of the analyzed line, and comparing the line element in the sample;

P ′ _x , P ′ _xcp , P _x , P _xsr - measured and adjusted sample parameters;

L _he = L ′ _he + ΔL ′;

ΔL ′ = 1- (L ′ _he + L ′ _eh );

calculate the content of the desired element in the sample C _x according to the system of equations

Where

moreover, at intermediate stages, the calculated concentration C _st = C _{x is} taken as a reference, the calculated value of the spectral line radiation intensity for the reference is determined by the expression

where v = (f + fd ² + 1 + d ² ) / [2d (f-1)];

d = tg [(π / 2) · (P _e + P _esr ) / 2P ₀ ];

h = [tg (π / 2) · (P _e / P ₀ )] / [tg (π / 2) · (P _esr / P ₀ )];

f = U _eh h = (1 / U _he ) h,

and the concentration calculation and the adjustment of the radiation parameters are performed until the specified accuracy of calculations

where Δ _Rst is the calculated value of the radiation parameter in the internal standard;

Δ _Рх - measured value of the radiation parameter in the sample;

- относительная погрешность вычислений.

- relative error of calculations.

According to the proposed method, a number of virtual standards are calculated with control by radiation intensity, the content of which is successively approaching the desired value.

The method is an integral part of an automated system for determining the quantitative composition of elements in materials and alloys with photoelectric recording of spectra. As radiation receivers, solid-state multichannel charge-coupled linear devices are used that convert the radiation intensities of a continuous spectrum into a discrete signal, after processing of which the parameters of the analyzed spectral lines are obtained. As the parameters involved in the further calculation of the quantitative content, the intensities or areas maxima are selected, limited by the contours of the spectral lines. Regardless of the registration method, you can take the intensity display P.

The initial data for the calculation are the measured parameters of the radiation intensity of the spectral lines of the studied element R _x , R _e and the comparison lines R _xsr , R _esr for the standard and the test sample, as well as the concentration of C _e element in the standard. For a non-insulated system, sample parameters are denoted by P ′ _x ; P ′ _xcp :

where ΣP _e = P _e + P _esr ; ΣP _x = P _x + P _xcp ;

? P = P _{e e} - P _ESD; ΔP _x = P _x - P _hsr .

The initial data of system (1) are transferred to an isolated state, detaching themselves from differences in the matrix composition and possible differences in the experimental conditions for the standard and the test sample.

For this, the relative energy parameters of the radiation of the sample are calculated relative to the standard L _he , L _eh

and parameters of amplification of spectral radiation of a sample element relative to a standard U _he , U _eh

where P ₀ - the maximum value of the measured parameter (photoelectrically analyzing V ₀ = 10B).

The condition for compatible isolated systems are the equations

If, in this case, radiation absorption does not occur in the reference sample system, then this sum is 1. If a part of the energy is absorbed (ΣL ′ <1) or a part of the energy is radiated into the environment (ΣL ′> 1), then to bring the system into isolated state should take this difference into account by amending

In accordance with the received L _he corrected and obtained new values of P _x and R _xsr , it should be _borne in mind that the parameters ΔP _x = const and ΔP _e = const, since they are a measure of the quantitative content of the element in the sample and standard:

tg (π / 2) L _ho = (? P _x + P _HSR) / P _HSR (ΣP _'x -? P _e) / (ΣR' _x +? P _e)

after conversions:

P _xsr = (ΔP _x ΣP ′ _x - ΔP _x ΔP _e ) / [tg (π / 2) L _he (ΣP ′ _x + ΔP _e ) + ΔP _e - ΣP _x ];

and the calculation formula is converted to:

tg (π / 2) L _{ho = [(ΔP x + P xcp} ) / P xcp] [( _x + 2P? P _HSR - _e? P) / _(x? P +? P + 2P _{HSR e)].}

After replacing the desired value of P _xsr , the reduced quadratic equation of the following form is solved:

The solution to this equation is the parameter of the line of comparison of the sample element P _xsr and then from the condition P _x = ΔP _x + P _{xsr the} intensity of the sample element is determined.

As a result of the transformations, the initial equilibrium system with the parameters arises for the calculations:

In subsequent transformations, we will find the parameters of the internal standard and compare the obtained values with the initial measured parameters of the sample in an isolated system (2).

A necessary condition for the relationship of the internal standard and the sample express equation parameters ΣR _e = ΣR _x = const.

This equilibrium system for the desired P _x and P _xsr will be symmetric with respect to the data ΣP _e / 2 and ΔP _x / 2, i.e.

To find the intermediate value of the concentration of the sample element C ′ _x = C _st1 at the first stage of calculations, the identity with respect to C _{st1 is solved} [1]:

Where

- radiation parameter of the reference standard;

- sets the energy parameter of the sample element C _x , or reference C _e, respectively (susceptibility to spectral radiation);

- coefficient of self-absorption of atoms and ions in a low-temperature plasma;

Next, based on equations (2) to L _Oh, calculated difference .DELTA.P _CT1 sample cell according to claim 5 corresponding to the found value _CT1 C:

Where

The data of the control standard and samples are recorded at the beginning of the second stage of calculations:

Points 4-7 are repeated at the second stage of calculations. At the end of these calculations, the initial data of the equilibrium system are determined similar to (18).

The steps are repeated until, with a given degree of accuracy, the obtained value of the difference of the input parameters of the internal standard ΔP _sti at the last stage of calculations i does not coincide with the measured value in the sample ΔP _x , i.e.:

for a given level of relative error δ _P.

The experimental data for determining the percentage of Si, Cu, Mg, and Mn in AK5M cast aluminum alloy obtained on the upgraded MFS-8M photovoltaic installation are shown in the table. As receivers of spectral radiation, SKCCD-type recording units based on diode arrays were used. As a control standard S _ke and the test sample, standard samples from the GSO kit No. 11 were used. The value of the test sample was compared with the pre-known content of C _GSO .

Table Experimental data of the method of successive approximations Element C _ke With _GSO C _xi data by stages δ _C ,% C ₁ C ₂ C ₃ C ₄ = C _x Si 4.09 5.29 5.124 - - 5.331 0.78 6.28 5.660 6.065 - 6.160 1.91 7.31 6.032 6.950 7.541 7.506 2.68 Cu 0.98 2.06 1.784 2.222 2.187 2.170 5.33 2.84 2.378 2.755 2.821 2.726 4.01 6.75 5.630 6.344 6.952 6.935 2.70 Mg 0.188 0.308 0.2757 0.321 - 0.315 2.27 0.482 0.394 0.467 - 0.501 3.94 0.497 0.406 0.486 - 0.506 1.81 Mn 0.156 0.219 0.203 - - 0.226 3.20 0.402 0.355 0.414 - 0.391 2.74 0.740 0.580 0.663 0.800 0.768 3.81

Example calculations for photoelectric analysis.

Below are the intermediate calculation results for MP in the Ak5M alloy according to the table.

The initial data of the measured voltages are proportional to the intensities of the spectral lines (V _o = 10B).

We find L _ho, this compute L _{'ho, oh} L and ΔL:

L ′ _he = 0.7137; L _eh = 0.1397; ΔL = 0.1466; then L _he = 0.8603.

We determine the parameter V _xcp from expression (17) and then V _x for an equilibrium isolated system for which condition (19) is satisfied for symmetry reasons. Then we get V _xcp = 1.5450; V _x = 0.4300.

Thus, the initial equilibrium system at the initial stage of calculations has the following form:

First step. We find the content C _x1 , from the solution of identity (10) using (11) - (13), (2), (3):

C _x1 = 0.58; a _x = 0.5501; b _x = 0.9016; Q _x = 0.4142; a _st1 = 1.6360; b _st1 = 0.9204;

Q _st1 = 0.1160; To _st1 = 1.3443; K _x = 1.4002; U _he = 3.573; Q _st1 U _he = 0.4586.

Then C _x1 = C ₁ = 0.58 is the internal standard of the first stage.

We determine the parameters V _x and V _xcp . For this, in accordance with equations (15) - (18), we find the values of d, h, K, f and v, and then the parameter ΔV _x = ΔV _st1 from formula (14) with C ₁ = 0.58. Then we get:

K (C _x , C _e ) = 1.333; a _st1 = 1.6360; b _st1 = 0.9204; K _st1 = 1.3443; Q _st1 = 0.1160; Q _x = 0.4140;

d = 0.1563; h = 0.0765; f = 4.8825; v = 5.0484.

Then ΔV _x = ΔV ₁ = -1.2700, or for symmetry reasons for equilibrium systems we have:

V ₁ = 0.9875 + (ΔV ₁ / ₂₎ =0.9875+0.6350=0.3525;

V _1cp = 0.9875- (ΔV ₁ / ₂₎ = _{0.9875-0.6350} =1.6225.

Second phase. Source system:

We find C _x2 :

C _x2 = 0.663; and _x = 0.4444; b _x = 0.8975; Q _x = 0.5321; _Thu2 = 0.5501; _a2 b = 0.9016;

Q _a2 = 0.4142; to _a2 = 1.4002; _x = 1.4096; U _he = 1.2843; Q _st2 U _he = 0.532.

Then C _x2 = C ₂ = 0.663 is the internal standard of the second stage.

We define the parameters V _x and V _xcp for C ₃ = 0.75 by analogy with the previous:

K (C _x, C _a2) -1.0660; Q _x = 0.5321; Q _x = 0.4142;

d = 0.1563; h = 0.2128; f = 3.9; v = 5.5372.

Then ΔV _x = ΔV ₃ = -1.1566, and from the symmetry condition for equilibrium systems we have:

V ₂ = 0.9875 + (ΔV ₂ /2)=9875+0.5783=0.4092;

V _2cp = 0.9875- (ΔV ₂ /2)=0.9875-0.5783=1.5658.

The third stage. Source system:

Find C _x3 .

C _x3 = 0.8; and _x = 0.4212; b _x = 0.8965; Q _x = 0.5667; and _st3 = 0.4444; b _st3 = 0.8975;

Q _st3 = 0.5321; to _st3 = 1.4002; _x = 1.4118; U _x3 = 1.066; Q _st3 U _he = 0.5671.

Then C _x3 = C ₃ = 0.80 is the internal standard of the third stage.

We define the parameters V _x and V _xcp at C ₃ = 0.8. Similar to the previous one:

K (C _x , C _St3 ) = 1.0660; Q _x = 0.5667; Q _st3 = 0.5321;

d = 0.1563; h = 0.2564; f = 3.7215; v = 5.5915.

Then ΔV _x = ΔV ₃ = -1.082, and from the symmetry condition for equilibrium systems we have:

V ₃ = 0.9875 + (ΔV ₃ /2)=0.9875+0.5410=0.4465;

V _3av = 0.9875- (ΔV ₃ / ₂₎ = _{0.9875-0.5410} =1.5285.

The fourth stage. Source system:

Find C _x4 :

C _x3 = 0.77; and _x = 0.4348; b _x = 0.8971; Q _x = 0.5459; a _st4 = 0.4212; b _st4 = 0.8965;

Q _st4 = 0.5667; to _st4 = 1.4118; _x = 1.4105; U _xst4 = 0.9524; Q _st4 U _xe = 0.5397.

Then C _x4 = C _x = 0.768 is the internal standard of the fourth stage.

We define the parameters V _x and V _xcp at C ₄ = 0.768. Similarly:

ΔV _x = (4V _o / π) · arctg [-v ± [(v ² -1) ^0.5 ]];

K (C _x , C _st4 ) = 1.01; Q _x = 0.5459; About _st4 = 0.5665;

d = 0.1563; h = 0.2869; f = 3.6532;

v = 5.7474. Then ΔV _x = ΔV ₄ = -1.114;

and from the symmetry condition for equilibrium systems we have:

V ₄ = 0.9875 + (ΔV ₁ / ₂₎ =0.9875+0.5570=0.4305;

V _4cp = 0.9875- (ΔV ₁ / ₂₎ = _{0.9875-0.5570} =1.5455.

Thus, at the fourth and final stage, the parameters of the analytical pair under study have the following numerical values:

Thus, the relative error in determining the voltage difference in the standard compared to the element voltage difference in the sample at the fourth stage of calculations was

δ _p = (ΔV _x -ΔV _st4 ) / ΔV _cf = (1.115-1.114) /1.1145=0.1%.

The error in the determination of C _x = C _GSO = 0.740 was

δ _C = (C _x - C _GSO) / C = _GSO (0.768-0.74) /0.74=3.8% at the P = 0.156 _Oe.

The error of analytical calculations can be determined from the joint solution of equations (3), (4), (6) and (7).

Diagrams of changes in the percentage of elements and voltages at the output of the diode arrays are shown in the drawing. The relative errors obtained are also indicated there.

The obtained experimental data and calculations allow us to talk about a wider range of use of one control standard. As experimental studies of various grades of materials show, the content of elements in samples relative to control standards may differ by 10 or more times. Moreover, the error of the final result is practically unchanged and can range from tenths of percent to (5-7)%.

With traditional methods for determining the content using calibration graphs with a ratio of 10> (C _x / C _e )> 5, the relative error can vary from 5 to 25%. At smaller ranges of concentration changes, the measurement error decreases. The limiting case of calculations with traditional methods of processing measurement results is the difference in the contents of elements in samples and samples no more than 10 times.

The technical and economic efficiency of the proposed method for determining the content of mass fractions of elements in materials and alloys is to increase the accuracy and reliability of determining the quantitative content of elements in materials and alloys. As a result, the cost-effectiveness, expressness and quality of the analyzes are improved by reducing the cost of acquiring expensive GSO sets and using them in current analyzes, and reducing the time for analysis. The method provides the necessary accuracy and reliability of the results in accordance with the requirements of GOST.

Literature

1. Nagibina I.M., Mikhailovsky Yu.K. Photographic and photoelectric spectral instruments and emission spectroscopy technique. - L .; Engineering, 1981. - S.92-103.

2. Morozov N.A. Methods of optical spectral analysis of aluminum alloys using computers. - Factory Laboratory, 1986, No. 9. - S.21-28.

3. Pat. 2035718 Russia, MKI G01N 21/67. A method for determining the mass fractions of elements in materials and alloys. // Nikitenko B.F., Odinets A.I., Kazakov N.S., Kuznetsov V.P., Kuznetsov A.A. 1995. Bulletin No. 14. - The prototype.

Claims (1)

A method for determining the mass fraction of elements in materials and alloys, including excitation of sample radiation in a low-temperature plasma, recording the emission spectrum of the sample, measuring the intensity of the analytical line of the element and the comparison line, calculating the content of the desired element in the sample, characterized in that the radiation intensity of the analytical line of the element in the standard with a sequential approximation of the element content to the desired value and adjust the radiation parameters of the sample for To measure the analytic line of the element, a physical model is used that contains the expressions of the parameter (αχ) _eh , which characterizes the stable state of low-temperature plasma in the standard sample with respect to the sample and the parameter M _ie , which characterizes the ability to emit low-temperature plasma relative to the standard sample for each element, defined by the relations

Where

With _ei ΔP _ei is the content of the element and the intensity difference of the analyzed line and the line of comparison of element i in the standard sample; P _xi , P _xicp — intensities of the analyzed line and the line of comparison of element i in the sample; B, D - proportionality coefficients determined by the parameter M _ei , calculate the energy parameters of radiation of the sample relative to the standard L ′ _he and L ′ _eh

make adjustments and the calculation of the new values of the intensities of the spectral lines in the sample P _x and P _HSR so that

where ΔР _x , ΣР _x , is the difference and the sum of the intensities of the analyzed line and the line of comparison of the element in the sample; P ′ _x , P ′ _xcp , P _x , P _xsr - measured and adjusted sample parameters; L _he = L ′ _he + ΔL ′; ΔL ′ = 1- (L ′ _he + L ′ _eh ); calculate the content of the desired element in the sample C _x according to the system of equations

Where

moreover, at intermediate stages, the calculated concentration C _st = C _{x is} taken as a reference, the calculated value of the spectral line radiation intensity for the reference is determined by the expression

where v = (f + fd ² + 1 + d ² ) / [2d (f-1)]; d = tg [(π / 2) · (P _e + P _esr ) / 2P ₀ ]; h = [tg (π / 2) · (P _e / P ₀ )] / [tg (π / 2) - (P _esr / P ₀ )]; f = U _eh h = (1 / U _eh ) h, and the concentration calculation and the adjustment of the radiation parameters are performed until the specified accuracy of calculations

where Δ _Rst is the calculated value of the radiation parameter in the internal standard; Δ _Рх - measured value of the radiation parameter in the sample;

- relative error of calculations.

Images