The capabilities of the Atom software used in the spectral analytical equipment manufactured by the VMK-Optoelektronika company are presented. Spectral measurements using various sources — inductively coupled plasma, microwave plasma, arc and spark discharge in air or argon atmosphere, atomic absorption, etc. — are discussed. Procedures allowing the traceability of measurement results are described.
The calculation steps starting with calculation of the spectral line intensity to determination of the element content in the sample are briefly considered, as well as the estimation of statistical indicators and the metrological control of the measurement procedure according to GOST R ISO 5725. The advanced data export tools allow the transfer of results to external information systems, and, moreover, expand the functionality of the program through interaction with other applications. Various options for printing out using compact and extended presentations of results in the form of reports are available. The Atom software covers several methods of spectral analysis, including time-resolved spectrometry (scintillation, absorption, and analysis of non-metallic inclusions). Methods of qualitative and semi-quantitative analysis are offered in the form of separate additional tools. The Atom software distribution kit contains auxiliary information systems: a database of spectral lines, an alloy grade guide, a catalog of standard samples with certified elemental composition, and a database of the standard metrological characteristics of measurement methods. A distinctive feature of the Atom software is a well-thought-out user interface for the operation of any spectral-analytical equipment with high performance. The modern modular architecture of the program provides effective management of the project complexity and enables independent development of individual components using suitable tools.

The aim of this work was to develop a tool for generating and sending diagnostic reports and backup copies of the software for spectral systems based on multichannel analyzers of emission spectra (MAES) controlled by the Atom software. To match the goal, we compiled a list of parameters and characteristics of the spectral system available to the software that can be used for troubleshooting. The information about the characteristics of the installed Atom software and operating system, the settings of the Atom software and all its modules, the settings for recording spectra by the MAES analyzer, the failure reports and auxiliary data that are available in the files with spectral data appeared to be the most useful parameters. In accordance with this list, we developed a new software tool AtomReport for generating reports on the settings and operation of the device, as well as for making backup copies of the software. Diagnostics based on generated reports provided identification of incorrect settings for analysis, recording or mathematical processing of spectra, reproduction and elimination of malfunctions in the software. Practical examples of improving the analytical results using the remote diagnostics of the equipment are presented. The ability of obtaining backup copies of the Atom software increases the stability of the spectral system by offering the possibility of prompt restoring of the operational status of the software after any changes made by the user of the spectral system and allows the diagnostics of malfunctions using a backup copy in a test debugging environment. Moreover, the developed tool can be used to identify malfunctions in the equipment itself using the data stored during recording of the spectra.

Photodetector arrays are widely used in atomic emission spectrometry as a part of spectrum analyzers. When the width of a spectral line is comparable with the array structure pitch, the recorded signal becomes dependent on the position of the line relative to the cells of the photodetector array. This dependence is usually explained by the loss of high spatial frequencies according to the Kotelnikov theorem, and the pixel response function of the cells (the dependence of the cell output signal on the position of a point light spot on the cell surface) is considered rectangular and determined by the cell size. However, this approximation can sometimes lead to significant errors. The aim of this study is to determine experimentally the pixel response function of BLPP-2000 and BLPP-4000 photodetector array cells used in high-speed MAES analyzers by precise moving of a light spot with a diameter of 0.7 μm and a wavelength of 405 nm. The array cell width is 14 and 7 μm, respectively. It is shown that when a light spot enters between the cells of BLPP-2000 and BLPP-4000 photodetectors, no information is lost. As for the cells of back-illuminated CCD BLPP-2000 arrays, the coefficient of mutual influence is 30% and the integral signal does not depend on the position of the light spot. The presence of isolating areas (loci) between the cells in CMOS BLPP-4000 arrays led to insignificant oscillations of the integral signal depending on the position of the light spot, but, however, provided reduction of the coefficient of mutual influence to 5%.

The design of a Grand-2000 atomic emission spectrometer is considered. The optical efficiency and spectral resolution of a Grand-2000 spectrometer are compared with the corresponding characteristics of a Grand spectrometer with multi-element emission spectrum analyzers based on BLPP-2000 and BLPP-4000 photodetector arrays. The optical efficiencies are compared using the ratio of spectral line intensities recorded on the two spectrometers using the same types of photodetectors. The resolution of the spectrometers (FWHM of the spectral line) is measured using different types of photodetectors. It is shown that, when using BLPP-4000 photodetectors, the resolution of a Grand-2000 spectrometer is twice as large (4 pm), compared to a Grand spectrometer. The optical efficiency of a Grand-2000 spectrometer is 3 – 8 times lower, depending on the wavelength. It is also shown that a Grand-2000 spectrometer with BLPP-2000 photodetector arrays has better optical efficiency in the short-wavelength region compared to a Grand spectrometer with BLPP-4000 photodetector arrays at the same values of the spectral resolution.

The method of scintillation atomic emission spectrometry (SAES) allows determination of the form of the element present in the sample and the size of inclusions in addition to the total content of the element in the direct analysis of powder samples. The aim of the work is to verify the expedience of using the direct method of analysis of individual gold particles in studying the possible forms of gold appearance in pyrites and sulfides, including the so-called «invisible» finely divided gold. The measurements were carried out both on individual pyrite single crystals, and on the samples of sulfide gold and brown coal deposits of Kazakhstan. The studies were carried out on a Grand-Potok complex (VMK Optoelektronika, Ltd.) equipped with a laser ablation system and a system for injection of aqueous solutions and aerosols samples. The samples of pyrite crystals of various sizes were analyzed to determine the content and speciation of gold in the samples. No dependence of the gold content on the crystal size was noted for all the studied pyrites. It is shown that gold and other precious metals are concentrated in the surface layers of pyrite crystals and in crystal lattice defects in the bulk of the crystal (according to SAES data during laser ablation of the sample in an arc discharge). The total content of gold in the bulk of pyrite crystals is about 2 g/ton, platinum and silver is less than 0.02 g/ton, while the content of Au, Ag, Pt on the surface ranges from 2 to 5 g/ton. The detection limit of gold determination by the SAES method is 0.01 g/ton. The SAES method can be used for determination of the gold speciation in pyrites and sulfides (finely dispersed, individual particles up to 1 μm or even less).

Two main methods of approximate quantitative atomic emission spectral analysis have traditionally been developed in the Central Laboratory of the Karpinsky All-Russian Research Geological Institute (VSEGEI): evaporation from the channel of carbon electrodes and a more efficient spill method. When comparing these two methods for introducing a sample into the arc discharge, two factors should be taken into account: the amount of material entering the discharge and the completeness of evaporation of chemical elements. A sample weighing about 40 mg is almost completely evaporated from the channel of a carbon electrode. The spill-injection method provides a uniform supply of a large powder sample (400 – 500 mg) into the arc discharge. The first method of analysis is well suited for determination of volatile and nonvolatile elements, whereas the second method provides stable conditions for evaporation and excitation during all the time when the sample is supplied into the discharge. The spill-injection method is more sensitive to changes in the bulk composition of the sample and the particle size of the sample material compared to evaporation from the electrode channel. These are the main factors affecting the magnitude of the systematic error in determining the concentrations of semi-volatile elements. The errors that make up the total error of approximate quantitative atomic emission spectral analysis are systematically analyzed in spectral laboratories. Comparison of the results of analysis of the same samples (state standard samples) with the results of chemical methods over a period of several years revealed that the maximum contribution to the total error is attributed to the visual assessment of the content and visual interpolation, as well as to a discrepancy in interpretation of results after a long time, etc. The impact of various errors has been reduced through the use of the MAES analyzer and the wide capabilities of the Atom software, which offer a correct consideration of the background with the option of individual settings, the use of coefficients accounting for the interference effect, and the possibility of analysis by several analytical lines of each chemical element thus allowing the use of constant calibration curves for a wide range of rock compositions.

A method of quantitative simultaneous determination of 23 trace elements (Ba, Be, Cu, Co, Cr, Ga, La, Mo, B, Ni, Pb, Sc, Sn, Sr, V, Y, Yb, Zn, Zr, Fe, Mn, P, Ti) and 6 macronutrients (Mg, Al, Si, Na, K, Ca) in soils, rocks and ash of plant material using atomic emission spectrometry is described. A two-jet argon arc plasmatron is used as the source of spectra excitation. The samples to be analyzed are introduced into the plasma flow jet by injection of an air suspension of a finely dispersed powder under gas (argon) pressure. A multichannel analyzer of emission spectra (MAES) is used for spectra recording. State standard samples of soils, rocks, and plant ash of various compositions and genesis are used as reference samples. The optimal conditions for excitation, registration of spectra and processing of the useful signal were determined to minimize the random and systematic errors of element determinations. Evaluation of the metrological characteristics, carried out using natural standard samples, revealed no significant systematic error. The relative standard deviation of the method ranges from 0.07 to 0.15, and the calculated value of the Student’s criterion for all the elements considered was shown to be less than the tabular value.

An atomic emission technique for the analysis of a fuel salt containing uranium and a coolant based on lithium and beryllium fluorides has been developed and tested. When developing the technique, the conditions accepted at Rosatom enterprises for the analysis of beryllium oxide and lithium carbonate were taken into account. The complexity of the structure of the arc spectrum of the matrix is noted. X-ray phase analysis of the sample residues in the electrodes revealed that the source of molecular bands in the spectra is lithium fluoroberillate. The analytical lines of Al, B, Ca, Cd, Cr, Cu, Fe, Mg, Ni, Pb, Si and Zn, free from overlap with the lines of the molecular spectrum and uranium were selected. A method of extreme experiment design is used to select optimal conditions for the arc excitation of the sample: the type and strength of the discharge current (alternating current 12 A), exposure time (20 sec), the shape of the electrode («glass») and the mass of the material (30 mg). Recommendations are given and implemented for the preparing samples for calibration using pure lithium fluoroberillate as a matrix material when introducing controlled elements in the form of certified reference materials of graphite (graphite collector of trace impurities). Calibration graphs in logarithmic coordinates are linear with angular coefficients close to unity.
The metrological characteristics of the technique are evaluated in analysis of real samples: the repeatability, intermediate precision of the results and the limits of the element detection.

Conditions for the analysis of liquid samples on a MFS-8 spectrometer upgraded with a MAES photodiode array by arc atomic emission spectrometry with dry residue technique are specified. It is shown that optimization of the electrode shape along with the method of calculating the intensity of the spectral line and time of the base exposure leads to the expansion of the operating range of calibration curves. The arc plasma parameters (temperature and electron density) are similar when the spectra of dry residues of aqueous solutions, saliva, blood serum, and dry wine are excited. This provides direct (without digestion) determination of trace elements in these samples with the detection limits at a level of μg/liter. When a sample of vegetable oil is deposited on the electrode, it is likely to be absorbed in depth and does not enter the arc plasma in full, which makes the direct analysis impossible. Nevertheless, the determination of trace elements in oils by the dry residue technique is possible after acid digestion of the sample.

A Polyvac E980 (Hilger Analytical) photoelectric vacuum optical emission spectrometer has been modified with a multichannel analyzer of emission spectra (MAES). The line of argon supply to the spark stand has been completely renewed. A precision mass flow controller with a microprocessor control unit has been installed to adjust the argon supply for different operation modes and fix the current values with the possibility of saving data for each spectrum. The microprocessor control unit coordinates the operation of the control lines and interrupts the operation of the spark generator when the stand is open or argon is absent. The modified vacuum spark spectrometer, originally designed for the analysis of iron-based alloys, has become suitable for determining the composition of non-ferrous alloys. Analytical programs have been developed for the rapid spectral analysis of AK12M2 and AK9ch aluminum alloys. The main methodological difficulty of the analysis is attributed to the necessity of monitoring high concentrations of silicon, copper, and iron along with the content of calcium impurity in the same sample. Calcium has a negative effect on the quality of castings, thus making the possibility of Ca determination an important motivation for the modification. After the preparatory work and the construction of calibration curves, the short-term and long-term repeatability of the results of spectral analysis of standard samples of aluminum alloys has been studied on the modified spectrometer. The results completely meet the standard requirements for determining the composition of aluminum alloys.

The method of atomic emission spectrometry with arc spectrum excitation provides determination of rare earth elements (REE) in geological samples by direct analysis of solid samples without a long-term preliminary sample preparation. Using the analysis of a rare-earth ore sample from the Tomtor niobium-rare-earth deposit as an example, we compared the analytical capabilities of a Grand Potok installation and a two-jet arc plasmatron (TJP) in combination with a DFS-458S spectrograph equipped with a multichannel emission spectrum analyzer (MAES). It is shown that a higher temperature of the TJP plasma (~7500 K) compared to that of the arc discharge (~5500 K) and the shift in the recording range of a DFS-458S spectrum to longer wavelengths (280 – 245 nm) provided a more reliable determination of REE using a larger number of analytical lines with lower detection limits. The correctness of the results of REE determination was confirmed by comparison with the data obtained by the ICP-MS method.

The current system of the temperature control for the electrothermal atomizer (ETA) used in atomic absorption spectrometers for simultaneous multielement analysis, is unable to provide high characteristics of the analysis when the calibration of the built-in optical pyrometer becomes irrelevant due to the natural wear of graphite cells upon operation. As the control of the ETA efficiency using an external calibrated pyrometer is laborious, it is advisable to use the dependence of temperature on the time of appearance of atomic vapors of elements. We have studied the possibility of controlling the temperature of graphite cells in the electrothermal atomizer of a multielement atomic absorption spectrometer with a continuous spectrum source by the time dependence of absorption signals of chemical elements. The correctness of the calibration was checked by recording the absorption signals of a sample containing chemical elements of different volatility with subsequent evaluation of the time and the corresponding temperature of the appearance of atomic vapors of the elements. The obtained temperatures of the appearance of atomic vapors of Al, Cd, In, Mn, Ni, Pb and V ranged within 640 – 1940°C. When the heating rate is changed by a factor of more than 3, the vapor appearance temperature for the selected elements differs by less than 5%. Using a deliberate change in the calibration of the built-in optical pyrometer, we have simulated a situation in which the relevance of the calibration was lost, e.g., due to the wear of a graphite cell. The experiment revealed a shift of the correlation graph between the actual and measured temperatures of the appearance of vapors of elements in the case of incorrect calibration of the feedback pyrometer in the coordinates «Real temperatures» — «Measured temperatures». The method presented in the study can be used to check the correctness of the calibration of the built-in pyrometer and to determine the necessity of replacing a worn graphite cell.