Accurate Prediction Of Heats Of Formation By A Combined Method Of DFT And Neural Network Correction

In this disertation,we propose the X1 method which combines the density functional theory method with a neural network(NN) correction for an accurate yet efficient prediction of heats of formation.It calculates the final energy by using B3LYP/6-311+G(3df,2p) at the B3LYP/6-311+G(d,p) optimized geometry to obtain the B3LYP standard heats of formation at 298 K with the unscaled zero-point energy and thermal corrections at the latter basis set.The NN parameters cover 15 elements of H,Li,Be,B,C,N,O,F,Na,Mg,Al,Si,P,S,and Cl.The performance of X1 is close to the Gn theories,giving a mean absolute deviation of 1.43 kcal/mol for the G3/99 set of 223 molecules up to 10 nonhydrogen atoms and 1.48 kcal/mol for the X1/07 set of 393 molecules up to 32 nonhydrogen atoms(J.Chem.Phys. 127(2007)214105).Ninety-five additional tests from reference(J.Chem.Phys. 119(2003)11501) have been performed.X1 reduces its MAD from 11.10 to 1.69 kcal/mol,validating the effectiveness of the X1 method for the accurate prediction of△_∫H~θ.The overall mean absolute deviation of the X1 method from experiment for the 488 heats of formation is 1.52 kcal/mol compared to 9.44 kcal/mol for the original B3LYP results.We use 31 radicals and 115 molecules to set up 141 bond dissociation reactions.For the total of 144 heats of formations and 141 bond energies,B3LYP leads to mean absolute deviations of 4.57 and 6.31,respectively,while X1 reduces the corresponding errors to 1.41 and 2.46 kcal/mol.In this disertation,we discuss in detail how to set up the X1 neural network.We give examples,showing how to apply the X1 method and how the applicability of X1 can be extended.The X1 method offers an eXpress way to calculate accurate heats of formation of larger molecules,inaccessible by accurate wavefunction-based methods such as G2 or G3.The X1 method greatly eXtends the reliability and applicability of the B3LYP method,providing a valuable complement to the experimental determination of thermo-chemistry.