پديد آورنده :
امامي، زينب
عنوان :
محاسبه ي انرژي و طول پيوند دو تايي آهن به دو روش مونت كارلوي كوانتومي و نظريه ي تابع چگالي
مقطع تحصيلي :
كارشناسي ارشد
محل تحصيل :
اصفهان: دانشگاه صنعتي اصفهان، دانشكده فيزيك
صفحه شمار :
هفت، 84ص.: مصور،نمودار
يادداشت :
ص.ع. به فارسي و انگليسي
استاد راهنما :
مجتبي اعلائي
استاد مشاور :
هادي اكبر زاده
توصيفگر ها :
VMC , ِDMC , كدGAUSSIAN , كدCASINO
تاريخ نمايه سازي :
16/12/90
تاريخ ورود اطلاعات :
1396/10/12
چكيده فارسي :
به فارسي و انگليسي: قابل رويت در نسخه ديجيتالي
چكيده انگليسي :
Calculation of length and binding energy of iron dimer by the two methods of quantum Monte Carlo and density functional theory Zeinab Emami z emami@ph iut ac ir October 10 2011 Department of Physics Isfahan University of Technology Isfahan 84156 8311 IranDegree M Sc Language FarsiM Alaei m alaei@cc iut ac irH Akbarzadeh akbarzad@cc iut ac irAbstractThe main aim of this thesis is studying of iron dimer by the Quantum Monte Carlo QMC Method The QMC method is astochastic method which is used to solve the Schr dinger equation The character of being stochastic is due to using randomnumbers As a matter of fact QMC is a Monte Carlo method sor solving the Srodinger equation The main advantage of thismethod is its accuracy is independent of the integral dimension Also this method due to its independence of the integraldimension is the best accessible method to deal with complicated systems such as strongly correlated systems as well as largesystems in which modern quantum chemistry methods such as configuration interaction fail to apply Variational Monte Carlomethod VMC and diffusion Monte Carlo method DMC are the most popular QMC method in electronic structure calculationswhich are used in the present project The CASINO code was used for the QMC calculations This code is capable of dealingwith limited systems such as various kinds of atoms molecules crystals systems with periodic boundary conditions 2 dimensional and 3 dimensional homogenous and inhomogeneous gases in crystalline or fluid phases Beginning the QMCcalculations requires an input trial wave function There are several ways to produce the trial wave function One of the ways is touse the output file of the density functional theory DFT method To perform the DFT calculations the GAUSSINA03 code wasused The more similar is the input trial wave function to the ground wave function of the system the more accurate is thevariational QMC method For the same reason a great deal of attention was assigned to choosing the basis set used in the DFTcalculations in order to find the optimum basis set for DFT calculations Before finding the optimum basis set the propermultiplicity of the system was obtained and found to be 7 Ultimately the output file of the DFT calculations by GAUSSIAN03code was used to produce the trial wave function for the QMC method At the beginning of the calculations the variance of theVMC calculations was high Studying the texts clarified that the intense fluctuations of the core electrons causes the enhancementin the variance of the calculations Therefore an effective core potential with 12 core electrons was utilized in performing thecalculations and the variance of the calculations fell off to an acceptable value The wave function produced by the VMCcalculations was used as the guide function in order to impose the importance sampling in the DMC method The DMCcalculations were performed with the same effective core potential used in the VMC method To enhance the accuracy of thecalculations some different ways such as changing the number of steps and walkers and optimizing the backflow factor wereexamined Applying the backflow factor enhanced the calculation time to 6 times while did not change the variance so much Therefore finally the backflow factor was omitted For different via atoms length chosen around the bond length obtained by theDFT method the DMC calculations were performed and the diagram of energy versus via atoms length was fitted to the 4 th orderpolynomial Ultimately the iron bond length was found to be 2 054 and its binding energy turned out to be 1 143 eV which isin a good agreement with the experimental reports In addition to the main goal of the project other results were obtained Forexample it was observed that for walker numbers larger than a plausible value the variance of the calculations does not changesignificantly Also the prominence of using effective core potential in the large and strongly correlated system was clarified Key words
استاد راهنما :
مجتبي اعلائي
استاد مشاور :
هادي اكبر زاده