GOST 7728-79
GOST 7728−79 magnesium Alloys. Methods of spectral analysis (with Change No. 1)
GOST 7728−79
Group B59
INTERSTATE STANDARD
MAGNESIUM ALLOYS
Methods of spectral analysis
Magnesium alloys.
Methods for spectral analysis
ISS 77.120.20
AXTU 1709
Date of introduction 1981−07−01
INFORMATION DATA
1. DEVELOPED by the Ministry of aviation industry of the USSR
2. APPROVED AND put INTO EFFECT by Decision of the USSR State Committee on standards
3. REPLACE GOST 7728−68
4. REFERENCE NORMATIVE AND TECHNICAL DOCUMENTS
| The designation of the reference document referenced |
Item number |
| GOST 83−79 |
2.2 |
| GOST 84−76 |
2.2 |
| GOST 195−77 |
2.2 |
| GOST 244−76 |
2.2 |
| GOST 804−93 |
2.2, 3.2 |
| GOST 1535−91 |
2.2 |
| GOST 2856−79 |
Introduction |
| GOST 2789−73 |
2.3 |
| GOST 3240.0−76−3240.21−76 |
1.4, 2.5 |
| GOST 3773−72 |
2.2 |
| GOST 4160−74 |
2.2 |
| GOST 6709−72 |
2.2 |
| GOST 14957−76 |
Chapeau |
| GOST 19627−74 |
2.2 |
| GOST 23208−83 |
2.4.2 |
| TU 16. K71−087−90 |
2.2 |
5. Resolution of Gosstandart No. 665 from
6. EDITION with Change No. 1, approved in may 1991 (IUS 8−91)
This standard specifies the methods of spectral analysis for determination of the main alloying components and impurities (aluminum, manganese, zinc, beryllium, copper, silicon, iron, Nickel, zirconium, cadmium, calcium, cerium, lanthanum, lithium, yttrium, neodymium, praseodymium and India) in the magnesium alloys (wrought and casting) according to GOST and GOST 2856 14957.
1. GENERAL REQUIREMENTS
1.1. The content of alloying components and impurities in alloys is determined by the calibration graphs. Will use two methods of calibration instruments:
the method of «three standards»;
the method of «controlling standard».
The spectrum — photographic and photoelectric.
The analysis of the photographic method of calibration graphs constructed in the coordinates:
where is the difference of pochernenija lines of the designated element and an element of comparison;
— mass fraction of the element in the standard samples (CO);
The analysis of the PV method, the calibration graphs constructed in the coordinates:
,
where — mass fraction of the element in the standard samples;
— the readings of the output of the measuring device, is proportional to the logarithm of the relative intensity of the lines defined by item and line comparison.
Note. For quantometer in which the readings of the output device «a» is proportional to the relative intensity of spectral lines, calibration curve constructed in the coordinates:
.
(Changed edition, Rev. N 1).
1.2. For evaporation of the sample and the excitation spectrum using spark and arc sources of light.
1.3. For the calibration of instruments used state standard samples (GSO) N 423−73−429−73; 820−76−823−76; 1797−80−1803−80; 2772−83−2776−83; 740−75−747−75; 740−84 P-744−84 P; 2329−82−2535−82; 2336−82−2343−82.
It is permitted to use industry standard sample (CCA) N 1−81−4-81; 9−81−12−81; 5−81−8-81; 62−82−65−82; 74−83−83−83, standard samples of enterprises (SOP), as well as newly-issued standard samples of composition of magnesium alloys in all categories.
(Changed edition, Rev. N 1).
1.4. Check of correctness of definition of a mass fraction of elements is carried out by comparing the results of spectral analysis with the results of the analysis performed by chemical methods according to GOST 3240.0-GOST 3240.21.
The permissible value of the absolute differences in percent must be calculated by the formula
where the result of analysis of sample made by chemical method, %;
— the result of a sample analysis performed by the spectral method, %;
— the relative standard deviation characterizing the reproducibility of spectral analysis;
— the relative standard deviation characterizing the reproducibility of chemical analysis results.
Such a test must be conducted no less than once per quarter for each of the analyzed grades of the alloy.
1.5. Sampling is carried out according to normative-technical documentation.
1.4, 1.5. (Added, Rev. N 1).
2. PHOTOGRAPHIC METHOD OF SPECTRAL ANALYSIS
2.1. The essence of the method
The method is based on the excitation spectrum of the arc or spark discharge, with subsequent registration on the photographic plate by a spectrograph.
2.2. Apparatus, materials and reagents
Spectrograph with quartz optics medium variance type of ISP-30 spectrograph of DFS-13.
Light sources: spark generator type IG-3 or IVS-23 and arc — type DG-2 or IVS-28.
Microphotometer type MF-2 or IPV-460.
The damping of the three — and nine -.
The spectral coals in the form of bars brands high purity-7−3, p-2, p-3, with a diameter of 6 mm.
Bars of magnesium brand MG GOST 804 with a diameter of 6−8 mm.
Rods of copper of grades M0, M1 according to GOST 1535 or TU 16. K71−087−90 with a diameter of 6−8 mm.
Photographic plates of the spectral types 1, 2, 3, ES, UPS sensitivity from 3 to 20 units, the SFC-01, the SFC-02, the SFC-04, SFC-05.
The lathe table.
A device for grinding coal.
Developer.
Solution I:
Distilled water according to GOST 6709 — 1000 cm,
metol (para-methylaminophenol) — 2 g,
sanitarily sodium (sodium sulfite) crystalline — 104 g or anhydrous sodium sanitarily according to GOST 195 — 52 g,
hydroquinone (paradoxians) according to GOST 19627 — 10 g,
potassium bromide according to GOST 4160 — 2 g.
Solution II:
Distilled water according to GOST 6709 — 1000 cm,
sodium carbonate crystal according to GOST 84 — 108 g or
the anhydrous sodium carbonate according to GOST 83 — 54 g.
Before the manifestation of the solutions I and II mixed (3:1).
Fixer.
Distilled water according to GOST 6709 — 1000 cm,
the crystalline sodium thiosulfate (sodium hyposulphite) according to GOST 244 — 300 g
ammonium chloride according to GOST 3773 — 60
Allowed the use of other instruments, equipment and materials subject to receipt of the metrological characteristics meet the requirements of this standard.
2.3. Preparation of samples for analysis
For analysis use samples of the following shapes and sizes (after sharpening):
bars with a diameter from 5 to 50 mm, length 30−100 mm;
in the form of «fungus», of a disk or plate with a minimum thickness of 3.5 mm, with a diameter of 30−50 mm.
Molded samples are prepared by pouring liquid metal into a mold dvuhromovo split (finger pattern) or metallic (or water-cooled metal) mold, providing the samples in the form of «fungus». The material of the mold — cast iron or steel grade of St.3.
In the first case, the diameter of the specimen should be 5−9 mm, length 30−100 mm, the second diameter of the cap or disk is 30−50 mm, thickness — not less than 5 mm.
In the analysis sheets and bars size less than that provided by the standard, adjust the calibration graphs for the DIS, having the same dimensions, shape of test sample (AO).
Ubeskrivelig the surface of the samples sharpen on a plane, a parameter of surface roughness should be not more than 20 µm according to GOST 2789.
With molded samples in the form of «fungus» remove a layer not less than 1.5 mm, with rod shaped samples of 5−10 mm. If the sample has a flat end face, allowed the removal of a layer at a depth of 0.2−0.5 mm.
Sample preparation and analysis must be identical for this series of measurements. On the treated surface of the specimen are not allowed shell scratches, cracks, slag inclusions. Protivoelektrodom sharpen: on a spherical surface with a radius of 3−6 mm, cone angle 120° or truncated cone with ground diameter of 1.0−1.7 mm with a sharpening angle of 40−60°.
2.4. Analysis
The conditions of analysis of the photographic method are given in table.1.
Table 1
| The conditions of analysis | |||||
| Equipment, materials and parameters | In all alloys: Al, Be, In, Y, Cd, Ca, Si, La, Li, Mn, Cu, Nd, Ce, Zn, Zr, Fe |
In grades МА8, MA18-CE in the alloy brand ВМЛ5-Nd |
The alloy grade MA11-Nd, Pr |
In all alloys — impurities Fe, Si, Cu, Ni |
The alloy grade Ml5pch-Zr |
| Spectrograph |
ISP-30 |
ISP-30 |
DFS-13 |
ISP-30 |
ISP-30 |
| Generator |
Types of IG-3, IVS-23 (the scheme is simple or complex) |
Types of IG-3, IVS-23 (the scheme is simple or complex) |
Types of IG-3, IVS-23 (scheme complex) |
Types DG-2, IVS-28 |
Types of IG-3, IVS-23 (scheme complex) |
| Gap width, mm |
0,015−0,020 | ||||
| Capacitance, µf |
0,005−0,01 |
0,01 |
0,02 |
- |
0,01−0,02 |
| A self-induction, mH |
0−0,05 |
0,05 |
0,01 |
- |
0,05 |
| Current, A |
1,6−3,0 |
2.0 to 3.0 |
4,0 |
2,5−4,5 |
2,0−4,0 |
| The analytical gap, mm |
2,0 |
2.0 to 2.5 |
2.0 to 2.5 |
1,5−1,8 |
2.0 to 2.5 |
| Defining the gap gap, mm |
3,0 |
0,5−0,8 |
3,0 | ||
| The roasting |
20−40 |
30 |
30 |
5−10 |
Without firing |
| The counter |
Coal or magnesium |
Coal, magnesium, copper |
Coal or magnesium | ||
| Photographic plates |
Types 1, 2, ES |
Types 1, 2, ES |
Types 2, 3, S, 1 |
Types 2, 3, ES, OFS |
Types 2, 3 |
| Coordinates of calibration curve |
|||||
Notes:
1. Parameters are within the specified values.
2. The exposure time is selected depending on the sensitivity of the used photographic plates should be not less than 15 s.
3. Carrying out analysis with the use of pairs of electrodes, with sharpening one of the electrodes on the plane.
4. In the analysis sheets and bars with dimensions smaller than required by standard allowed the choice of other modes of operation of the light sources.
5. When determining the mass fraction of calcium below 0.1%, it is recommended to pre-firing carbon electrodes at a current of 16−18 A.
Wavelength of analytical spectral lines and ranges of mass fractions given in table.2.
Table 2
| The designated element |
The wavelength of the element, nm |
Wavelength comparison lines, nm |
The range of detectable concentrations, % | ||
| Aluminium |
I 396,15 I 394,40 II 358,69 I 308,22 |
I 332,99 307,40 I 291,55 |
0,01−0,5 0,5−1,5 3,0−12,0 3,0−12,0 | ||
| Beryllium |
II 313,04 |
I 332,99 307,40 Background |
0,0005−0,01 | ||
| Iron |
I 358,12 I 302,06 II 259,94 II 238,20 |
I 332,99 307,40 Background |
0,002−0,1 0,002−0,1 0,002−0,1 0,002−0,1 | ||
| Indium |
I 410,17 I 325,60 I 303,94 |
I 332,99 I 332,99 307,40 |
0,2−1,0 0,2−1,0 0,2−1,0 | ||
| Yttrium |
II 319,56 II 320,03 |
I 332,99 307,40 |
1,0−3,0 1,0−3,0 | ||
| Cadmium |
I 361,05 I 346,62 I 326,11 |
I 332,99 |
0,1−1,0 0,3−2,0 0,5−2,0 | ||
| Calcium |
II 396,85 II 393,37 II 315,89 |
I 333,21 I 332,99 |
Of 0.01−0.2 Of 0.01−0.2 0,2−0,5 | ||
| Silicon |
I 288,16 I 251,61 I 251,61 I 288,16 |
307,40 I291,55 Background |
0,05−0,5 0,05−0,5 0,001−0,05 0,001−0,05 | ||
| Lantan |
II 394,91 II 375,91 II 338,09 II 317,17 |
I 332,99 307,40 |
0,2−2,0 0,2−2,0 0,2−2,0 0,2−2,0 | ||
| Li |
I 323,26 |
I 332,99 |
8,0−12,0 | ||
| Manganese |
II 347,41 II 346,03 II 344,20 II 294,92 II 259,37 |
I 332,99 I 291,55 307,40 |
0,7−2,5 0,7−2,5 0,7−2,5 0,01−0,7 0,01−0,7 | ||
| Copper |
I 327,40 I 324,75 |
I 332,99 307,40 |
0,02−0,5 0,003−0,5 | ||
| Neodymium |
II 430,36 II 410,95 II 401,22 II 406,10 II 401,22 II 406,10 II 380,54 |
I 332,99 Background I I 332,99 |
1,6−3,0 1,6−3,0 1,0−5,0 1,0−5,0 0,03−0,6 0,03−0,6 0,01−0,6 | ||
| Nickel |
I 352,45 I 314,48 II 239,45 |
I 332,99 Background |
0.01 0.01 0.01 | ||
| Praseodymium |
II 410,07 |
Background |
0,08−0,5 | ||
| Cerium |
II 418,66 II 413,76 II 401,24 II 320,17 |
I 416,73 Background I 322,99 Background 307,40 |
0,07−0,5 0,07−0,5 0,5−3,0 1,0−3,0 | ||
| Zinc |
I 334,50 I 330,29 I328,53 II 255,80 II 250,20 |
I 332,99 307,40 I 291,55 307,40 I 291,55 |
0,05−1,5 0,05−4,0 2,0−10,0 2,0−10,0 2,0−10,0 | ||
| Cubic Zirconia |
II 343,82 II 339,20 II 327,93 II 327,30 II 339,20 |
I 332,99 307,40 Background |
0,04−0,8 0,04−0,8 0,2−1,0 0,2−1,0 0,002−0,06 | ||
Notes:
1. Spectral lines, limited in the table paranteza may be respectively combined in any analytical pair.
2. If the comparison lines using the background, the latter is measured near the line of the element.
3. The Roman numeral I to the values of wavelengths is from a line to a neutral atom, figure II — once ionized atom.
2.4.1. When working according to the method of «three standards» perform the following operations:
choose FROM the analyzed alloy in an amount of not less than three;
spectra of co and AO are photographed on the same photographic plate at the selected conditions of analysis with randomization of the order of shooting. For each co and AO photographed three spectrum;
measure the blackening of analytical lines selected, count the difference of pochernenija
for the analytical line pairs, and the arithmetic average of the
three spectra;
build a calibration curve in the coordinates: 
This schedule is suitable for the analysis of those samples, the spectra of which are filmed along with on a single photographic plate;
the contents of the element in AO find for the calibration schedule.
The exposure time is selected such to provide normal pochernenija for all analytical lines.
If the analysis of the small mass fraction of the blackening of analytical lines of the element lies in the region of neoteric, you must use a response curve is carefully constructed in neoteric. The calibration graph constructed in this case in the coordinates:
where is the intensity of the line of the designated element;
— the intensity of the comparison lines or background area of the element.
It allowed the construction of calibration curve in the coordinates 
the difference pucherani line of the element and the background near the line.
2.4.2. When working according to the method of «reference standard» except WITH, you need to build a calibration chart, choose the SOP, which must satisfy the following requirements:
a) the chemical composition should be as close as possible to the middle of the range mass fraction, indicated in the GOST 2856, GOST 23208*;
_______________
* On the territory of the Russian Federation GOST 23208−2003. — Note the CODE.
b) the shape, size and physico-mechanical properties (the method of casting, handling) it needs to meet AO.
The work begins with the construction of the basic calibration curve on a single photographic plate spectra photographed WITH this alloy and SOP five times.
On average the photometric estimates in the normal pochernenija analytical lines build permanent calibration schedule of the main photographic plates in the coordinates 
In the analysis of industrial samples: for the working plate photographed spectra AO 3 times each and the spectra of the DIS 4 times. Determine the difference pucherani analytical line pairs for SOP and AO, i.e., 
. The conversion factor is
introduced to account for the properties of the emulsion working plates, calculated by the formula
where is the difference of pochernenija auxiliary line pair of magnesium or a difference of pochernenija line of magnesium to two steps of the attenuator, calculated for several spectra WITH and SOPS for the main plates;
— the difference between pochernenija the same lines of magnesium and the same steps of the attenuator, calculated for a working photographic plates SOP and AO.
For calculation of the coefficient of use of a pair of lines of magnesium Mg 333,21 nm — nm Mg 332,99 etc.
Through the point with coordinates 
Allowed to build a calibration curve for working glass plates in the coordinates: 
,
— the angle of the calibration chart for the main plates, constructed in the coordinates
To simplify the calculations calibration curve can be artificially reduced to 45°. In this case, instead of the coefficient factor should be used
2.5. Processing of the results
The final result of the analysis be the arithmetic mean of three parallel measurements obtained on the three spectrograms, if the condition:
where is the greatest result of the parallel measurements;
— the lowest result of the parallel measurements;
— the relative standard deviation characterizing the precision of measurements;
— the arithmetic mean calculated from
parallel measurements (
3).
When conducting Express-analysis allowed the analytical result to calculate the two parallel measurements and
provided that
Values and
for photographic analysis method are given in table.3.
Table 3
| The designated element |
The range of detectable concentrations, % | Photographic method |
Photovoltaic method | ||
| no more | |||||
| Beryllium |
0,0005−0,001 |
0,17 |
0,14 |
- |
- |
| Beryllium, iron, silicon, copper, Nickel, zirconium |
0,001−0,01 |
0,15 |
0,12 |
0,15 |
0,12 |
| Aluminum, iron, cadmium, calcium, silicon, manganese, copper, neodymium, cerium, zinc, zirconium, praseodymium |
0,01−0,1 |
0,10 |
0,08 |
0,08 |
0,06 |
| Aluminum, indium, calcium, cadmium, lanthanum, manganese, copper, neodymium, praseodymium, cerium, zinc, zirconium, silicon |
0,1−0,5 |
0,06 |
0,05 |
0,05 |
0,04 |
| Aluminium, indium, yttrium, calcium, cadmium, Lanta, manganese, neodymium, cerium, zirconium, copper, zinc |
0,5−2,0 |
0,05 |
0,04 |
0,04 |
0,03 |
| Aluminium, calcium, neodymium, cerium, zinc, yttrium, manganese |
2,0−5,0 |
0,05 |
0,04 |
0,04 |
0,03 |
| Aluminum, zinc, lithium |
5,0−15,0 |
0,04 |
0,03 |
0,03 |
0,025 |
Calculation methodology and
is given in the Appendix.
If the difference between the analysis result and one of the limit values of the content item (for this alloy for the standard for the brand) in absolute value less than or equal to 
is the number of samples (
1 or 2);
— the average result of the analysis of one or two definitions.
2.4−2.5. (Changed edition, Rev. N 1).
3. THE METHOD OF PHOTOELECTRIC SPECTRAL ANALYSIS
3.1. The essence of the method
The method is based on the excitation spectrum of the alloy arc or spark discharge with the registration of the intensity of the lines using the photovoltaic installation.
3.2. Equipment and material
Installation of photovoltaic (quantometer) type DFS-10m, DFS-36, MFS-4, MFS-8.
The spectral coals in the form of bars brands high purity-73, C2, C3 with 6 mm diameter.
Bars of magnesium brand MG GOST 804 with a diameter of 6−8 mm.
The lathe table.
A device for grinding coal.
Allowed the use of other spectral instruments, equipment and materials subject to receipt of the metrological characteristics meet the requirements of this standard.
(Changed edition, Rev. N 1).
3.3. Sample preparation
The samples prepared as described in step 2.3.
3.4. Analysis
By the photoelectric method of analysis for the calibration of the instrument using the method of «three standards» and «reference standard».
The conditions of analysis of the photovoltaic method are given in table.4.
Table 4
| Option |
The conditions of analysis | ||||
| Quantometer DFS-10M, generator powerplant-1 |
Quantometer DFS-36, generator UGE-4 | ||||
| Arc mode | Low-voltage spark mode | Arc mode | Low-voltage spark mode | High-voltage spark mode (scheme complex) | |
| The voltage, V |
220±5 | ||||
| The current in the circuit, And |
2,5−3,0 |
- | |||
| Capacitance, µf |
- |
40−60 |
- |
40−60 |
0,005; 0,01 |
| Inductance, µh |
0 |
0; 150 | |||
| The phase of ignition, deg |
90 |
- |
- | ||
| The number of digits in the half life current |
- |
1 |
- |
1; 2 (VP the method Mgr.) |
1; 2; 3 (VP control method) |
| The analytical gap, mm |
1,5 |
1,5; 2,0 | |||
| The roasting |
5−7 |
10−20 |
5−7 |
10−20 |
20−30 |
| The exhibition, with |
20−40 |
20−60 | |||
| The counter |
Coal or magnesium |
||||
| Coordinate system |
| ||||
or
Note. Options set within the specified values.
The wavelength of analytical spectral lines and ranges of mass fractions given in table.5.
Table 5
| The designated element |
Wavelength of lines of the designated element, nm |
Range mass fraction, % |
| Aluminium |
I 396,15 I 309,27 I 308,21 I 256,80 |
1,0−12,0 1,0−5,0 4,0−12,0 0,05−12,0* |
| Beryllium |
II 313,04 |
0,001−0,01 |
| Iron |
II 271,44 II 275,57 II 259,94 II 238,20 |
0,01−0,1 0,01−0,1 0,01−0,1 0,01−0,1 |
| Cadmium |
II 226,50 |
0,10−2,5* |
| Silicon |
I 288,16 I 251,61 |
0,05−0,5 0,05−0,5 |
| Indium |
I 303,94 |
0,2−1,0 |
| Yttrium |
II 360,07 |
1,0−3,0 |
| Lantan |
II 398,85 II 392,92 |
0,4−1,5 0,4−1,5 |
| Manganese |
II 258,37 II 257,61 II 294,92 |
0,05−2,5 0,05−2,5 0,1−0,6 |
| Copper |
I 510,55 I 327,40 |
0,1−1,0 0,01−0,5 |
| Nickel |
I 341,48 |
0,001−0,01* |
| Zinc |
I 334,50 |
0,2−4,0 |
| Cubic Zirconia |
II 343,82 II 339,20 |
0,01−1,0 0,01−1,5 |
| Neodymium |
II 430,36 II 406,11 II 401,22 |
1,0−2,5 1,0−5,0 |
| Cerium |
II 418,6 II 401,24 |
0,05−0,4* 0,05−0,4* |
| Line comparison of magnesium |
307,40 I 518,36 I 552,84 I 291,54 I 277,98 II 279,08 II 280,77 |
________________
* Data from quantometer with multimode light source.
The analytical lines chosen depending on the mass fraction of the element in the sample, the possibility of placing the output of cracks on the slide quantometer etc.
The use of other analytical lines, provided that they provide accuracy and sensitivity to meet the requirements of this standard.
The width of the entrance slit of quantometer (0,02−0,06 mm) and the width of the output slit (0,05; 0,10; 0,15; 0,20) is selected depending on the mass fraction of an element and the degree of doping of the alloy.
The contents of the element in AO determined using a calibration curve constructed in the coordinates: 
.
3.5. Processing of the results
Processing of results conducted as specified in claim 2.5.
Value or
for PV analysis method are given in table.3.
3.4, 3.5. (Changed edition, Rev. N 1).
APPLICATION (reference). TO EVALUATE THE ACCURACY OF SPECTRAL ANALYSIS
APP
Reference
1. The accuracy of the spectral method of analysis is determined by the magnitude of systematic and random errors provided that the failures are excluded from the calculations. It is assumed that the random errors obey the normal distribution law.
2. When properly configured, the spectral instrument and the implementation of the recommendations of the standard procedure of analysis the main sources of systematic errors are errors associated with the influence of structure and chemical composition of samples on the results of the analysis.
These errors should be identified by comparison of the results of the analysis of samples, performed chemical and spectral methods on a large sample (minimum 20 samples).
If the existence of errors installed, they eliminate the adjustment of the position calibration curve SOP.
(Changed edition, Rev. N 1).
3. The result of the analysis of samples, obtained for example as the arithmetic mean of the two (three) of parallel measurements, i.e. in two (three) spectra should be regarded as one definition.
4. Reproducibility of spectral analysis can be characterized by the relative standard deviation of a single determination.
5. For calculation choose at least five samples of one brand of alloy having approximately the same chemical composition, and within 5 days conduct their analysis series (one series per day). Spectra in each series is carried out in a different sequence, i.e. randomization.
One series of spectra recorded on the same photographic plate.
On each photographic plate to receive three of the spectrum of each sample and three FROM each spectrum. The latter is necessary for the construction or adjustment of the calibration graphs.
By the photoelectric registration before making a measurement, adjust the position of the calibration graphs, and then spectra of the samples.
Just from each sample get for 5 days for 15 measurements (five definitions).
For each sample calculate the standard deviation () by the formula
where is the average mass fraction of the element in
the sample is calculated from five determinations;
— mass fraction of the element at the
-th definition in
the sample is calculated from three measurements;
— the number of definitions (
5).
Then compute the standard deviation by the formula
where ,
, …,
— standard deviation, calculated respectively for first, second, etc. samples according to the formula (1);
— the number of samples (
5).
The relative standard deviation characterizing the reproducibility of the analysis calculated by the formula
where is the average mass fraction of element in the samples, calculated according to the formula
where ,
, …,
— the average mass fraction of element respectively in the first, second, etc. samples, calculated from 5 determinations.
4, 5. (Changed edition, Rev. N 1).
6. Convergence measurements are characterized by relative mean square error of a single measurement .
7. The value of the find in series of 20 parallel measurements for one sample with correctly configured equipment. First calculate the standard deviation
by the formula
where is the average mass fraction of element in the sample is calculated from 20 parallel measurements;
— mass fraction of element in the sample computed from the
-th measurement;
— the number of measurements in the series (
20).
Next calculate the relative standard deviation characterizing the repeatability of measurements, according to the formula
8. When conducting analyses, it is often necessary to estimate the error of the result of the analysis and confidence limits. At p = 0.95 and excluded systematic error
is calculated by the formula
where is the number of definitions by which the computed result of the analysis of the sample (usually in the spectral analysis of
1 or 2);
the result of analysis of the sample, calculated according to the
definitions.
Calculated means that with a reliability of 95% the true value of the designated value lies in the interval values:
the most likely result is .
9. A comprehensive assessment of the operation of the generator excitation spectrum, the spectral device and electronic measuring devices are recommended periodically by determining the relative standard deviation for a series of 20 parallel measurements according to the formula (5).
The value found is compared
, i.e. the standard deviation, which was calculated previously, when you configured the instrument.
The comparison is carried out according to the criteria.
If 
, it indicates that the instrument requires tuning.
At confidence probability 0.95 and the number of measurements in the series (20)
This check is recommended 1−2 times per month.
TS.
10. The offset of the calibration curve relative to the base (drift level SOP) is considered significant if it exceeds the standard deviation of the results of 4 measurements calculated according to the SOP, i.e. when
where is the number of parallel measurements for SOP, which is controlled by the position of the chart (
4);
The position of the calibration chart, it is recommended to monitor 1−2 samples 2−3 times per shift.
7−10. (Changed edition, Rev. N 1).
