Computer calculation of probability for binary collisions of electrons with ions and molecules

Computer calculation of rate coefficient for binary collision i < σᵢν > as a function of temperature is presented, and the Maxwell electron velocity distribution function is chosen. The finite elements of the fifth order made it possible to significantly speed up the process of calculation...

Повний опис

Збережено в:
Бібліографічні деталі
Опубліковано в: :Вопросы атомной науки и техники
Дата:2021
Автори: Marchuk, Yu.M., Kulyk, Yu.S., Moiseenko, V.Е.
Формат: Стаття
Мова:Англійська
Опубліковано: Національний науковий центр «Харківський фізико-технічний інститут» НАН України 2021
Теми:
Онлайн доступ:https://nasplib.isofts.kiev.ua/handle/123456789/195421
Теги: Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Цитувати:Computer calculation of probability for binary collisions of electrons with ions and molecules / Yu.M. Marchuk, Yu.S. Kulyk, V.Е. Moiseenko // Problems of Atomic Science and Technology. — 2021. — № 4. — С. 154-156. — Бібліогр.: 5 назв. — англ.

Репозитарії

Digital Library of Periodicals of National Academy of Sciences of Ukraine
_version_ 1860247892111917056
author Marchuk, Yu.M.
Kulyk, Yu.S.
Moiseenko, V.Е.
author_facet Marchuk, Yu.M.
Kulyk, Yu.S.
Moiseenko, V.Е.
citation_txt Computer calculation of probability for binary collisions of electrons with ions and molecules / Yu.M. Marchuk, Yu.S. Kulyk, V.Е. Moiseenko // Problems of Atomic Science and Technology. — 2021. — № 4. — С. 154-156. — Бібліогр.: 5 назв. — англ.
collection DSpace DC
container_title Вопросы атомной науки и техники
description Computer calculation of rate coefficient for binary collision i < σᵢν > as a function of temperature is presented, and the Maxwell electron velocity distribution function is chosen. The finite elements of the fifth order made it possible to significantly speed up the process of calculation < σᵢν >. The result of the approximation is a smooth function and the values of this function, its first and second derivatives, have no jumps at the mesh nodes and the accuracy of calculation is within the limits of statistical errors for the source data. These advantages and the results will be used in future tasks. Запропоновано комп’ютерний розрахунок вірогідності парних зіткнень < σᵢν > як функції від температури. Вибрана максвелiвська функція розподілу електронів за швидкостями. Скінченні елементи п'ятого порядку дозволили значно прискорити процес обчислення < σᵢν >. Результатом апроксимації є гладка функція. Значення цієї функції, її першої та другої похідних не мають стрибків у вузлах сітки, а точність знаходиться в межах статистичних помилок вхідних даних. Переваги даного методу будуть використані в майбутніх задачах. Предложен компьютерный расчет вероятности парных столкновений < σᵢν > как функции от температуры. Выбрана максвелловская функция распределения электронов по скоростям. Конечные элементы пятого порядка позволили значительно ускорить процесс расчета < σᵢν >. Результатом аппроксимации является гладкая функция. Значения этой функции, ее первой и второй производных не имеют скачков в узлах сетки, а точность расчета находится в пределах статистических ошибок исходных данных. Эти преимущества и результаты будут использованы в будущих задачах.
first_indexed 2025-12-07T18:39:07Z
format Article
fulltext ISSN 1562-6016. ВАНТ. 2021. № 4(134) 154 https://doi.org/10.46813/2021-134-154 COMPUTER CALCULATION OF PROBABILITY FOR BINARY COLLISIONS OF ELECTRONS WITH IONS AND MOLECULES Yu.M. Marchuk, Yu.S. Kulyk, V.Е. Moiseenko National Science Center “Kharkov Institute of Physics and Technology”, Kharkiv, Ukraine Computer calculation of rate coefficient for binary collision i   as a function of temperature is presented, and the Maxwell electron velocity distribution function is chosen. The finite elements of the fifth order made it pos- sible to significantly speed up the process of calculation i   . The result of the approximation is a smooth func- tion and the values of this function, its first and second derivatives, have no jumps at the mesh nodes and the accu- racy of calculation is within the limits of statistical errors for the source data. These advantages and the results will be used in future tasks. PACS: 29.17.+w; 41.75.Lx INTRODUCTION In the non-equilibrium plasma, elastic and inelastic collisions determine the rates of many collision proc- esses. Collisions of electrons with heavy particles (ion or with atom) are of the certain interest for plasma pro- duction and sustain. If you consider the case of elastic collisions, scattering is not accompanied by an appre- ciable loss of energy; in the case of inelastic interac- tions, the energy losses are higher. The probability of electron scattering through a cer- tain angle  is determined by the value of the interac- tion potential of the scattering particle with the incident electron. It is proportional to the differential scattering cross section    /d d d d     and the angular dependence of the scattering cross section  /d d  can take as a simple form, which simplifies calculations and can be quite complex. Total rate coefficient for collisions with the relative velocity  is defined as    /i d d d     . The frequency of collisions for monoenergetic elec- trons is determined by the expression    e i n    where n is the density of particles which electron col- lides. FINITE ELEMENT APPROXIMATION Many problems in applied mathematics are reduced to solving different equations of various types (linear and non-linear). One of the powerful numerical meth- ods is the finite element method, which is a universal tool for solving problems of mathematical physics. The finite element method is based on the fact that any arbitrary and continuous function  x in the computational domain, for example, the temperature, can be approximated by a set of piecewise continuous functions defined on a finite number of subdomains (finite elements) [1]. The region of integration is divided into elements in such a way that on each of them the unknown function could be approximated, for example, by polynomials. Inside each element, an unknown function is repre- sented by a linear combination of basis functions, un- known coefficients, while the values of the function are at the nodal points of the approximation. As coordinate functions, functions are taken that are identically zero everywhere, except for one finite element, inside which they coincide with the basis functions. Thus, the prob- lem of approximating a function is reduced to a system of linear algebraic equations, the number of equations of which coincides with the number of nodal points. These elements have common anchor points and to- gether they approximate the shape of the region. The essence of the problem is that we need to ap- proximate the known experimental data for electron impact ionization average cross section. This method meets our requirements, since excessive accuracy is not needed and the time spent on the calculation is rather small. Also, the advantages of the method are that the first and second derivatives at the mesh nodes are matching. For this reason, this method is suitable for calculations and can be used in work [2]. FORMULATION OF THE PROBLEM There are Lagrangian and Hermitian finite elements and each of them has its own advantages. The main disadvantage of Lagrangian finite elements is that they include additional points on the segment and deriva- tives at the main nodes are not matching. We use fifth degree polynomials to approximate a function for which the first and the second derivative are matching at the mesh nodes. This is essential advantage of Her- mitian finite elements. In accordance with the above written we will be use the fifth-order Hermitian finite element approximation for the rate coefficient. A pow- erful tool for simulating processes in plasma is de- scribed in the work [3]. Using the general formula for the rate coefficient for probability of binary collisions:   3 i e iS f d         , (1) we assume that the velocity distribution function is the normalized Maxwellian: 2 3/ 2 2 2 1 expe eM Te Te f f             , (2) where  is the velocity of electrons. Thermal velocity 2 /Te ekT m  depends on temperature, where k is the Boltzmann constant, T is the temperature of elec- trons, em is the mass of the electrons. ISSN 1562-6016. ВАНТ. 2021. № 4(134) 155 Next, we take the  values for the ionization cross- section of the hydrogen atom from review [4] and ap- proximate with columns (piecewise constant elements) which defined on finite number of subdomains (Fig. 1). Fig. 1. Function approximation using piecewise constant elements for electron impact ionization cross-section [4] The required function S depends on the electron temperature. Next, we introduce a temperature data grid with equidistant nodes i. We make the following change of variable: 2 2/ Teu   . At each grid interval, we represent the  as a con- stant value and analytically calculate the integral. The number of such integrals is equal to the number of in- tervals; therefore, we can represent equation (1) as a sum by the number of nodal points i .          1 1 1 1 1 exp 1 exp N i i i i i S A u u u u           , (3) where N is the number of points, 2 /Te iA    , iu and 1iu  is the lower and upper integration. It is also of interest to calculate the first and second derivatives of the function S with respect to tempera- ture:     1 2 1 / exp 2 1 N i i i i Ad S dT u u u T             2 1 1 1exp 2 1i i i A u u u T         , (4)      3 21 2 2 1 2 2 1 / exp 2 N i i i i i u u uAd S dT u T T                   3 2 1 1 1 1 2 2 1exp 2 i i i i u u uA u T T              . (5) A program has been written that makes it possible using equations (3) - (5) to calculate the S for any value of temperature of electrons. S is the continuous in temperature and all derivatives exist and this is the advantage of this program. But the main disadvantage is that the program calculates slowly, since (3) - (5) contains summation over the number of intervals and it takes time. The task is to simplify the S calculation, we will approximate the results using Hermitian finite elements with a polynomial of the fifth degree. Next, we intro- duce a new grid of temperature data. M is the number of intervals, k is the discrete variable which enumerates of mesh nodes and varies from 1 to M .       1 2 1 2 3 2| k kT T T i i i dS d SS S dT dT                              2 4 1 5 6 2 1 1 i i i dS d SS dT dT                        . (6) Below are the finite elements for the fifth degree polynomial [5]:   3 4 5 1 1 10 15 6        ,    3 4 5 2 6 8 3k          ,     2 2 3 4 5 3 1 3 2 k          ,        3 4 5 4 1 10 1 15 1 6 1           ,          3 4 5 5 1 6 1 8 1 3 1k                ,           2 3 4 52 6 1 1 3 1 1 1 2 ,k                 where 1k k kT T   ,    1 k k k T T T T      . (7) NUMERICAL RESULTS OF MODELLING The next step is to write a Fortran program. In order to calculate S using formula (6) in Fortran, we need to know the constants, namely iS ,  / idS dT , and   2 2/ i d S dT which are calculated at each nodal point. Since these constants are required, at the prepara- tory stage, our program launches a program that calcu- lates the equations (3) - (5) and fills with these values at the nodes of its own mesh. When this stage is passed our program is ready to work. Fig. 2. S values which are calculated by direct and approximation method at each temperature value S is the deviation values of the direct method from the approximation method For numerical experiments, we chose a temperature range from 0 to 1000 eV and chose 100000 temperature ISSN 1562-6016. ВАНТ. 2021. № 4(134) 156 values in which we will calculate S . Next, we made the calculation using the new program (we will call the approximation method) and using the old program (direct method) and compared the time of their calcula- tion, the results are shown below: 0.109201apprt s is the approximation method, 3.08882cyclet s is the direct method. Numerical simulation results are presented in Fig. 2. The figure shows that the result for one function calcu- lated in two different method is the same. CONCLUSIONS At the first stage, a method was developed for cal- culating S for electronic collisions using tabular val- ues of  . The calculation of S as a function of tem- perature is proposed. A distinctive feature is that S function is quickly calculated, and also the result of the approximation of S is a smooth function and the values of this function have no jumps. The application of polynomials in finite elements of the fifth order made it possible to speed up the process of calculating S by about 30 times 28.27cycle appr t t  . These advantages will be used in future tasks. REFERENCES 1. O. Zenkevich, K. Morgan. Finite elements and ap- proximation. New York: "John Wiley & Sons", 1983, 328 p. 2. Y.S. Kulyk, V.Е. Moiseenko, T. Wauters, A.I. Lyssoivan. Modelling of radio-frequency wall conditioning in short pulses in a stellarator. Prob- lems of Atomic Science and Technology. Series “Plasma Physics”. 2021, № 1(131), p. 9-14. 3. J.D. Lore et al. Implementation of the 3D edge plasma code EMC3-EIRENE on NSTX // Nuclear Fusion. 2012, v. 52, № 5, p. 054012. 4. R.K. Janev, D. Reiter, U. Samm. Collision proc- esses in low-temperature hydrogen plasmas. Jülich: Forschungszentrum, Zentralbibliothek, 2003, t. 4105. 5. C.C. Ike, E.U. Ikwueze. Fifth degree Hermittian polynomial shape functions for the finite element analysis of clamped simply supported Euler- Bernoulli beam //American Journal of Engineering Research (AJER). 2018, v. 7, № 4, p. 97-105. Article received 04.06.2021 КОМПЬЮТЕРНЫЙ РАСЧЕТ ВЕРОЯТНОСТИ ПАРНЫХ СТОЛКНОВЕНИЙ ЭЛЕКТРОНОВ С ИОНАМИ И МОЛЕКУЛАМИ Ю.М. Марчук, Ю.С. Кулик, В.Е. Моисеенко Предложен компьютерный расчет вероятности парных столкновений i   как функции от темпера- туры. Выбрана максвелловская функция распределения электронов по скоростям. Конечные элементы пято- го порядка позволили значительно ускорить процесс расчета i   . Результатом аппроксимации является гладкая функция. Значения этой функции, ее первой и второй производных не имеют скачков в узлах сетки, а точность расчета находится в пределах статистических ошибок исходных данных. Эти преимущества и результаты будут использованы в будущих задачах. КОМП'ЮТЕРНИЙ РОЗРАХУНОК ЙМОВІРНОСТІ ПАРНИХ ЗІТКНЕНЬ ЕЛЕКТРОНІВ З ІОНАМИ ТА МОЛЕКУЛАМИ Ю.М. Марчук, Ю.С. Кулик, В.Є. Моісеєнко Запропоновано комп’ютерний розрахунок вірогідності парних зіткнень i   як функції від темпера- тури. Вибрана максвелiвська функція розподілу електронів за швидкостями. Скінченні елементи п'ятого порядку дозволили значно прискорити процес обчислення i   . Результатом апроксимації є гладка фун- кція. Значення цієї функції, її першої та другої похідних не мають стрибків у вузлах сітки, а точність знахо- диться в межах статистичних помилок вхідних даних. Переваги даного методу будуть використані в майбу- тніх задачах.
id nasplib_isofts_kiev_ua-123456789-195421
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
issn 1562-6016
language English
last_indexed 2025-12-07T18:39:07Z
publishDate 2021
publisher Національний науковий центр «Харківський фізико-технічний інститут» НАН України
record_format dspace
spelling Marchuk, Yu.M.
Kulyk, Yu.S.
Moiseenko, V.Е.
2023-12-05T10:06:25Z
2023-12-05T10:06:25Z
2021
Computer calculation of probability for binary collisions of electrons with ions and molecules / Yu.M. Marchuk, Yu.S. Kulyk, V.Е. Moiseenko // Problems of Atomic Science and Technology. — 2021. — № 4. — С. 154-156. — Бібліогр.: 5 назв. — англ.
1562-6016
PACS: 29.17.+w; 41.75.Lx
DOI: https://doi.org/10.46813/2021-134-154
https://nasplib.isofts.kiev.ua/handle/123456789/195421
Computer calculation of rate coefficient for binary collision i < σᵢν > as a function of temperature is presented, and the Maxwell electron velocity distribution function is chosen. The finite elements of the fifth order made it possible to significantly speed up the process of calculation < σᵢν >. The result of the approximation is a smooth function and the values of this function, its first and second derivatives, have no jumps at the mesh nodes and the accuracy of calculation is within the limits of statistical errors for the source data. These advantages and the results will be used in future tasks.
Запропоновано комп’ютерний розрахунок вірогідності парних зіткнень < σᵢν > як функції від температури. Вибрана максвелiвська функція розподілу електронів за швидкостями. Скінченні елементи п'ятого порядку дозволили значно прискорити процес обчислення < σᵢν >. Результатом апроксимації є гладка функція. Значення цієї функції, її першої та другої похідних не мають стрибків у вузлах сітки, а точність знаходиться в межах статистичних помилок вхідних даних. Переваги даного методу будуть використані в майбутніх задачах.
Предложен компьютерный расчет вероятности парных столкновений < σᵢν > как функции от температуры. Выбрана максвелловская функция распределения электронов по скоростям. Конечные элементы пятого порядка позволили значительно ускорить процесс расчета < σᵢν >. Результатом аппроксимации является гладкая функция. Значения этой функции, ее первой и второй производных не имеют скачков в узлах сетки, а точность расчета находится в пределах статистических ошибок исходных данных. Эти преимущества и результаты будут использованы в будущих задачах.
en
Національний науковий центр «Харківський фізико-технічний інститут» НАН України
Вопросы атомной науки и техники
Gas discharge, plasma-beam discharge, and their applications
Computer calculation of probability for binary collisions of electrons with ions and molecules
Комп'ютерний розрахунок ймовірності парних зіткнень електронів з іонами та молекулами
Компьютерный расчёт вероятности парных столкновений электронов с ионами и молекулами
Article
published earlier
spellingShingle Computer calculation of probability for binary collisions of electrons with ions and molecules
Marchuk, Yu.M.
Kulyk, Yu.S.
Moiseenko, V.Е.
Gas discharge, plasma-beam discharge, and their applications
title Computer calculation of probability for binary collisions of electrons with ions and molecules
title_alt Комп'ютерний розрахунок ймовірності парних зіткнень електронів з іонами та молекулами
Компьютерный расчёт вероятности парных столкновений электронов с ионами и молекулами
title_full Computer calculation of probability for binary collisions of electrons with ions and molecules
title_fullStr Computer calculation of probability for binary collisions of electrons with ions and molecules
title_full_unstemmed Computer calculation of probability for binary collisions of electrons with ions and molecules
title_short Computer calculation of probability for binary collisions of electrons with ions and molecules
title_sort computer calculation of probability for binary collisions of electrons with ions and molecules
topic Gas discharge, plasma-beam discharge, and their applications
topic_facet Gas discharge, plasma-beam discharge, and their applications
url https://nasplib.isofts.kiev.ua/handle/123456789/195421
work_keys_str_mv AT marchukyum computercalculationofprobabilityforbinarycollisionsofelectronswithionsandmolecules
AT kulykyus computercalculationofprobabilityforbinarycollisionsofelectronswithionsandmolecules
AT moiseenkove computercalculationofprobabilityforbinarycollisionsofelectronswithionsandmolecules
AT marchukyum kompûterniirozrahunokimovírnostíparnihzítknenʹelektronívzíonamitamolekulami
AT kulykyus kompûterniirozrahunokimovírnostíparnihzítknenʹelektronívzíonamitamolekulami
AT moiseenkove kompûterniirozrahunokimovírnostíparnihzítknenʹelektronívzíonamitamolekulami
AT marchukyum kompʹûternyirasčetveroâtnostiparnyhstolknoveniiélektronovsionamiimolekulami
AT kulykyus kompʹûternyirasčetveroâtnostiparnyhstolknoveniiélektronovsionamiimolekulami
AT moiseenkove kompʹûternyirasčetveroâtnostiparnyhstolknoveniiélektronovsionamiimolekulami