GGA Level Research Articles

Halogen Thermochemistry Assessed with Density Functional Theory: Systematic Errors, Swift Corrections and Effects on Electrochemistry.

Despite its sizable errors, density functional theory (DFT) is extensively used to evaluate thermochemical properties of gases, liquids and their interfaces with solids. As numerous halogen-containing compounds appear as reactants, products and/or electrolytes in electrochemical reactions, and ionic effects are currently an active area of research, it is important to evaluate the accuracy of DFT for halogen thermochemistry. Herein, we assess the formation energies of interhalogens, hydrogen halides, diatomic and atomic halogens and their ions using six widespread functionals at the GGA, meta-GGA and hybrid levels. We observe that DFT errors with respect to experiments are correlated with the electronegativity of the species and there are systematic trends across functionals, such that swift corrections were devised. Specifically, the average of the mean absolute errors for the six functionals decreased from 0.19 eV before the corrections to 0.08 eV after them. Besides, the overall maximum absolute error (MAX) decreased from 0.76 to 0.44 eV and the average of the MAXs decreased from 0.51 to 0.24 eV. Finally, we illustrate the qualitative and quantitative impact of gas-phase errors on the predictions of surface Pourbaix diagrams.

Reducing training data needs with minimal multilevel machine learning (M3L)

For many machine learning applications in science, data acquisition, not training, is the bottleneck even when avoiding experiments and relying on computation and simulation. Correspondingly, and in order to reduce cost and carbon footprint, training data efficiency is key. We introduce minimal multilevel machine learning (M3L) which optimizes training data set sizes using a loss function at multiple levels of reference data in order to minimize a combination of prediction error with overall training data acquisition costs (as measured by computational wall-times). Numerical evidence has been obtained for calculated atomization energies and electron affinities of thousands of organic molecules at various levels of theory including HF, MP2, DLPNO-CCSD(T), DFHFCABS, PNOMP2F12, and PNOCCSD(T)F12, and treating them with basis sets TZ, cc-pVTZ, and AVTZ-F12. Our M3L benchmarks for reaching chemical accuracy in distinct chemical compound sub-spaces indicate substantial computational cost reductions by factors of ∼1.01, 1.1, 3.8, 13.8, and 25.8 when compared to heuristic sub-optimal multilevel machine learning (M2L) for the data sets QM7b, QM9 LCCSD(T) , Electrolyte Genome Project, QM9 AECCSD(T) , and QM9 EACCSD(T) , respectively. Furthermore, we use M2L to investigate the performance for 76 density functionals when used within multilevel learning and building on the following levels drawn from the hierarchy of Jacobs Ladder: LDA, GGA, mGGA, and hybrid functionals. Within M2L and the molecules considered, mGGAs do not provide any noticeable advantage over GGAs. Among the functionals considered and in combination with LDA, the three on average top performing GGA and Hybrid levels for atomization energies on QM9 using M3L correspond respectively to PW91, KT2, B97D, and τ-HCTH, B3LYP ∗ (VWN5), and TPSSH.

Metal-ligand bond in group-11 complexes and nanoclusters.

Density functional theory is used to study geometric, energetic, and electronic properties of metal-ligand bonds in a series of group-11 metal complexes and ligand-protected metal clusters. We study complexes as the forms of M-L (L = SCH3, SC8H9, PPh3, NHCMe, NHCEt, NHCiPr, NHCBn, CCMe, CCPh) and L1-M-L2 (L1 = NHCBn, PPh3, and L2 = CCPh). Furthermore, we study clusters denoted as [M13L6Br6]- (L = PPh3, NHCMe, NHCEt, NHCiPr, NHCBn). The systems were studied at the standard GGA level using the PBE functional and including vdW corrections via BEEF-vdW. Generally, Au has the highest binding energies, followed by Cu and Ag. PBE and BEEF-vdW functionals show the order Ag-L > Au-L > Cu-L for bond lengths in both M-L complexes and metal clusters. In clusters, the smallest side group (CH3) in NHCs leads to the largest binding energy whereas no significant variations are seen concerning different side groups of NHC in M-L complexes. By analyzing the projected density of states and molecular orbitals in complexes and clusters, the M-thiolate bonds were shown to have σ and π bond characteristics whereas phosphines and carbenes were creating σ bonds to the transition metals. Interestingly, this analysis revealed divergent behavior for M-alkynyl complexes: while the CCMe group displayed both σ and π bonding features, the CCPh ligand was found to possess only σ bond properties in direct head-to-head binding configuration. Moreover, synergetic effects increase the average binding strength to the metal atom significantly in complexes of two different ligands and underline the potential of adding Cu to synthesize structurally richer cluster systems. This study helps in understanding the effects of different ligands on the stability of M-L complexes and clusters and suggests that PPh3 and NHCs-protected Cu clusters are most stable after Au clusters.

Linear-response time-dependent density functional theory approach to warm dense matter with adiabatic exchange-correlation kernels

We present a new methodology for the linear-response time-dependent density functional theory (LR-TDDFT) calculation of the dynamic density response function of warm dense matter in an adiabatic approximation that can be used with any available exchange-correlation (XC) functional across Jacob's Ladder and across temperature regimes. The main novelty of the presented approach is that it can go beyond the adiabatic local density approximation (ALDA) and generalized LDA (AGGA) while preserving the self-consistence between the Kohn-Sham (KS) response function and adiabatic XC kernel for extended systems. The key ingredient for the presented method is the combination of the adiabatic XC kernel from the direct perturbation approach with the macroscopic dynamic KS response from the standard LR-TDDFT method using KS orbitals. We demonstrate the application of the method for the example of warm dense hydrogen, for which we perform a detailed analysis of the KS density response function, the RPA result, the total density response function and of the adiabatic XC kernel. The analysis is performed using LDA, GGA, and meta-GGA level approximations for the XC effects. The presented method is directly applicable to disordered systems such as liquid metals, warm dense matter, and dense plasmas.

A computational study of adsorption of H2S and SO2 on the activated carbon surfaces

An effort has been made to explore the adsorption potential of activated carbon (AC) structures for adsorption of hydrogen sulfide (H2S) and sulfur dioxide (SO2) employing density functional theory (DFT) at GGA level. 2-ring, 3-ring, 6-ring, and 9-ring carbon structures are used as adsorbent surfaces. The above mentioned rings are examined by creation of defect and inclusion of hydrogen as well to get the clear view close to experimental observations of adsorption properties of activated carbon. The adsorption properties depend upon many factors including whether adsorbate adsorbs in planer or non-planer mode, defect creation in the adsorbent substrate, atomic hydrogen insertion in the system and size of the adsorbent system. Our calculations show that side by side (planer) interaction binds the molecules much more strongly in comparison with molecules adsorbed upon the surface in non-planer mode. If vacancy is created at the central position of the surface, the molecules bind with substantial binding energy. However, overall Eads of the molecules varies randomly and no consistency could be achieved. Additionally, smaller sized structures are favorable relative to the bigger surfaces. The highest Eads for both the molecules is −2.97 eV, though not on the same substrate system. Finally, it can be argued that activated carbon is very useful material for adsorbing the noxious gases.

A forgotten participant in pore deblocking of zeolites: dicarbonate in NaMeA zeolites, Me = Na, K, Rb, Cs.

The dynamics of carbon dioxide, carbonate anion (CO32-), and dicarbonate anion (C2O52-) in NaKA zeolite is studied at the DFT GGA level using ab initio molecular dynamics (AIMD). We show the easy formation of C2O52- dicarbonate from the reaction between CO32- and CO2 at high CO2 loading and their equilibrium at low CO2 loading. We have found that the dicarbonate anion can contact up to six cations (Me+ and Na+, Me = Na, K, Rb, Cs), which could reduce the separation properties of NaMeA zeolites relative to CO2 mixtures. The K+ interaction with dicarbonate C2O52- species pushes the cation from 8R site in full analogy with the carbonate's deblocking studied earlier. The easy C2O52- formation in NaMeA is confirmed by modeling reaction of C2O52- formation at the DFT GGA (PBE-D3) and hybrid levels (B3LYP, HISS, HSE06) with cNEB. The calculated intensities for high and low frequency branches of valence vibrations in C2O52- are compared with calculated ones for Me2C2O5 molecules and known IR spectroscopic data in the NaMeA zeolites. This new mechanism of deblocking could be important for a wide family of narrow pore zeolites (CHA, RHO, KFI, etc.) at room temperature where the carbonates are observed in the IR spectra. The possibility of tricarbonate formation is discussed.

Testing the r2SCAN Density Functional for the Thermodynamic Stability of Solids with and without a van der Waals Correction.

A central aim of materials discovery is an accurate and numerically reliable description of thermodynamic properties, such as the enthalpies of formation and decomposition. The r2SCAN revision of the strongly constrained and appropriately normed (SCAN) meta-generalized gradient approximation (meta-GGA) balances numerical stability with high general accuracy. To assess the r2SCAN description of solid-state thermodynamics, we evaluate the formation and decomposition enthalpies, equilibrium volumes, and fundamental band gaps of more than 1000 solids using r2SCAN, SCAN, and PBE, as well as two dispersion-corrected variants, SCAN+rVV10 and r2SCAN+rVV10. We show that r2SCAN achieves accuracy comparable to SCAN and often improves upon SCAN's already excellent accuracy. Although SCAN+rVV10 is often observed to worsen the formation enthalpies of SCAN and makes no substantial correction to SCAN's cell volume predictions, r2SCAN+rVV10 predicts marginally less accurate formation enthalpies than r2SCAN, and slightly more accurate cell volumes than r2SCAN. The average absolute errors in predicted formation enthalpies are found to decrease by a factor of 1.5 to 2.5 from the GGA level to the meta-GGA level. Smaller decreases in error are observed for decomposition enthalpies. For formation enthalpies r2SCAN improves over SCAN for intermetallic systems. For a few classes of systems-transition metals, intermetallics, weakly bound solids, and enthalpies of decomposition into compounds-GGAs are comparable to meta-GGAs. In total, r2SCAN and r2SCAN+rVV10 can be recommended as stable, general-purpose meta-GGAs for materials discovery.

Hole-mediated ferromagnetic coupling in two-dimensional CrI3/VSe2 van der Waals heterostructures

Two-dimensional (2D) ferromagnetic (FM) semiconductors play imperative roles in the development of spintronic devices. However, the Curie temperature (Tc) of most 2D FM semiconductors are fairly low for practical applications. Herein, using the first-principles calculations, it revealed that the magnetic coupling strength within CrI3 can be modulated via the construction of CrI3/VSe2 van der Waals heterostructure. The system remains semiconducting with the band gap of 0.22 eV (GGA level). Its Tc can be boosted to ∼210 K via p-type doping, well above the liquid nitrogen temperature, which makes it a step closer to practical application. The greatly promoted FM coupling is originated from the additional super-exchange paths as well as the enhanced in-plane Cr-I-Cr super-exchange interaction. It is found that the interlayer magnetic coupling is tunable upon the stacking alignment, which is feasible to manipulate the magnetic orders using external mechanical motion.

Heat transport in liquid water from first-principles and deep neural network simulations

We compute the thermal conductivity of water within linear response theory from equilibrium molecular dynamics simulations, by adopting two different approaches. In one, the potential energy surface (PES) is derived on the fly from the electronic ground state of density functional theory (DFT) and the corresponding analytical expression is used for the energy flux. In the other, the PES is represented by a deep neural network (DNN) trained on DFT data, whereby the PES has an explicit local decomposition and the energy flux takes a particularly simple expression. By virtue of a gauge invariance principle, established by Marcolongo, Umari, and Baroni, the two approaches should be equivalent if the PES were reproduced accurately by the DNN model. We test this hypothesis by calculating the thermal conductivity, at the GGA (PBE) level of theory, using the direct formulation and its DNN proxy, finding that both approaches yield the same conductivity, in excess of the experimental value by approximately 60%. Besides being numerically much more efficient than its direct DFT counterpart, the DNN scheme has the advantage of being easily applicable to more sophisticated DFT approximations, such as meta-GGA and hybrid functionals, for which it would be hard to derive analytically the expression of the energy flux. We find in this way, that a DNN model, trained on meta-GGA (SCAN) data, reduce the deviation from experiment of the predicted thermal conductivity by about 50%, leaving the question open as to whether the residual error is due to deficiencies of the functional, to a neglect of nuclear quantum effects in the atomic dynamics, or, likely, to a combination of the two.

A comparative density functional theory study of oxygen doping versus adsorption on graphene to tune its band gap

Graphene, having a perfect two-dimensional crystal structure, has many excellent features such as a high specific surface area, and extraordinary electrical, thermal and mechanical properties. However, its usage in electronic devices is possible only if band gap of desired value is induced in this gapless semi-metal. Therefore, first principle calculations have been carried out to investigate the role of oxygen (O) doping versus adsorption and, the impurity concentration and coverage to induce band gap in graphene employing PBE at GGA level. The band gap is induced owing to production of vacancies, dissociative adsorption of oxygen, subsituational doping and pre-dissociated oxygen adsorption. It is interesting to note that band gap is introduced by both the processes of doping and adsorption of O. The oxygen doping leads to induction of two energy gaps, smaller in value above and larger below the Fermi level; while adsorption irrespective of adsorption configuration produces single direct gap. Increase both in concentration and coverage leads to enhance band gap value maximum being 1.85 eV in case of hexagonal doping of 12.5% concentration with an exception in adsorption case. The results allow us to conclude that adsorption is as useful as doping to tune the band gap in graphene enabling its applications in designing high performance electronic devices.

Electronic band structure, phase stability, magnetic and thermoelectric characteristics of the quaternary Heusler alloys CoCuZrAs and CoRhMoAl: Insights from DFT computations

The search for new materials with stable structures and unique functional properties is important for the development of future technologies. Herein, by performing first-principles calculations based on density functional theory (DFT), we have explored two new quaternary Heusler alloys, CoCuZrAs and CoRhMoAl. The total energy calculations reveal that CoCuZrAs (CoRhMoAl) is stable in the Y2 (Y1)-type structure among the three possible atomic configurations and that both alloys are ferromagnetic in the ground state. Electronic properties computed at the GGA level show that the present alloys have metallic behavior, whereas CoRhMoAl is a half-metal at GGA + U (U ≥ 1 eV) due to the opening of a band gap at the Fermi level in the spin-down channel. Furthermore, CoRhMoAl retains its half-metallicity when pressure is applied (in the range of 0–25 GPa). The magnetic data reveal that CoCuZrAs has a total spin moment of 2.34 µB, whereas CoRhMoAl has an integral spin moment of 3.0 µB, which is in line with the Slater-Pauling formula Mt = Zt-24 (where Mt = total spin moment/cell, Zt = total valence electrons/cell). In CoCuZrAs, the Co, Cu, Zr and As atoms contribute individual spin moments of 2.16 µB, −0.03 µB, 0.18 µB and 0.04, respectively, whereas in CoRhMoAl, the Co, Rh, Mo, and Al atoms carry spin moments of 1.81, 0.17, 1.14 and −0.09 µB, respectively. These results are also supported by the corresponding spin density plots, where the spin density is mainly located at Co atoms. The thermoelectric properties calculated up to 1000 K exhibit interesting features that might be useful in energy production.

Hunting the elusive shallow n-type donor – An ab initio study of Li and N co-doped diamond

We report calculated energetics (at the GGA, and at the B3LYP, HSE06 hybrid density functional levels of theory) and electronic properties (B3LYP, HSE06) of Li and N co-doped diamond containing LiNx clusters (x = 1–4). Defect energies and structural properties at the GGA and hybrid (B3LYP, HSE06) levels of theory are very similar. Our results for isolated Li and N dopants agree with previous work in that they are not suitable candidates for shallow n-type donors. Here, we investigate the effects of addition of N atoms as a co-dopant with Li, which vary with the Li:N ratio and whether Li is interstitial or substitutional. Reaction energies, geometries, bonding and densities of states (DOS) are examined. Nitrogen stabilises substitutional rather than interstitial lithium. Atom-projected DOS show that the chief contributions to the defect states are not from orbitals on the dopant atoms themselves but from states associated with neighbouring carbon atoms. Only the tetrahedral 1:4 cluster, LiCN4, with substitutional Li is likely to behave as a shallow donor. We propose a non-molecular synthetic route for preparation of the LiCN4-doped material via prior generation of the N4V centre at temperatures at which Li is mobile but LiCN4 clusters are not.

Antimonene/bismuthene vertical Van-der Waals heterostructure: A computational study

In this paper, the structural, electronic, mechanical and optical properties of antimonene-bismuthene Van-der Waals heterostructure (Sb-Bi HS) were calculated based on the first principle density functional theory. We explored different stacks of Sb-Bi HS to find the most and the least stable staking for this heterostructure. At the GGA level of theory, the most stable model is a semiconductor with an indirect band gap of 159 meV. However, when the spin-orbit (SO) interaction is considered, the VBM and CBM touch the Fermi level and the HS becomes a semimetal. Our results also show that the electronic properties of the HS are robust against the external electric field and biaxial strain. Young modulus was calculated as 64.3N/M which predicts this HS as a resistant material against being stretched or compressed. The calculated optical properties, similar to monolayer antimonene, are completely dependent on the polarization of incident light and differ when parallel or perpendicular polarization is considered. Moreover, the absorption coefficient for perpendicular polarization in the visible region is significantly increased in comparison with the monolayer antimonene. High structural stability, electronic and mechanical robustness against electric field and strain, along with polarization-dependent optical properties of this HS, promise for its applications in beam splitters and nano-scale mirrors.

Theoretical study of the antioxidant capacity of the flavonoids present in the Annona muricata (Soursop) leaves.

A theoretical approach was used to evaluate the antioxidant capacity of 20 flavonoids reported in Annona muricata leaves. The theoretical study was at the GGA level using the wB97XD functional and the cc-pvtz basis set. The calculations were performed in gas phase and implicit solvent phase. The flavonol robinetin (03c) and the flavanol gallocatechin (01c) are species that exhibited the best antioxidant capacity in the HAT, SEPT, and SPLET mechanisms. On the other hand, in the SET I mechanism, flavonol quercetin (03b) was the best, and in the SET II mechanism, the most favored species is the flavanol catechin (01a). However, these species do not achieve to overcome the antioxidant capacity presented by the Trolox.

First-principles study of localized and delocalized electronic states in crystallographic shear phases of niobium oxide

Crystallographic shear phases of niobium oxide form an interesting family of compounds that have received attention both for their unusual electronic and magnetic properties, as well as their performance as intercalation electrode materials for lithium-ion batteries. Here, we present a first-principles density-functional theory study of the electronic structure and magnetism of H-Nb$_2$O$_5$, Nb$_{25}$O$_{62}$, Nb$_{47}$O$_{116}$, Nb$_{22}$O$_{54}$, and Nb$_{12}$O$_{29}$. These compounds feature blocks of niobium-oxygen octahedra as structural units, and we show that this block structure leads to a coexistence of flat and dispersive energy bands, corresponding to localised and delocalised electronic states. Electrons localise in orbitals spanning multiple niobium sites in the plane of the blocks. Localised and delocalised electronic states are both effectively one-dimensional and are partitioned between different types of niobium sites. Flat bands associated with localised electrons are present even at the GGA level, but a correct description of the localisation requires the use of GGA+U or hybrid functionals. We discuss the experimentally observed electrical and magnetic properties of niobium suboxides in light of our results, and argue that their behaviour is similar to that of $n$-doped semiconductors, but with a limited capacity for localised electrons. When a threshold of one electron per block is exceeded, metallic electrons are added to existing localised electrons. We propose that this behaviour of shear phases is general for any type of $n$-doping, and should transfer to doping by alkali metal (lithium) ions during operation of niobium oxide-based battery electrodes. Future directions for theory and experiment on mixed-metal shear phases are suggested.

Properties of Novel Non-Silicon Materials for Photovoltaic Applications: A First-Principle Insight.

Due to the low absorption coefficients of crystalline silicon-based solar cells, researchers have focused on non-silicon semiconductors with direct band gaps for the development of novel photovoltaic devices. In this study, we use density functional theory to model the electronic structure of a large database of candidates to identify materials with ideal properties for photovoltaic applications. The first screening is operated at the GGA level to select only materials with a sufficiently small direct band gap. We extracted twenty-seven candidates from an initial population of thousands, exhibiting GGA band gap in the range 0.5–1 eV. More accurate calculations using a hybrid functional were performed on this subset. Based on this, we present a detailed first-principle investigation of the four optimal compounds, namely, TlBiS2, Ba3BiN, Ag2BaS2, and ZrSO. The direct band gap of these materials is between 1.1 and 2.26 eV. In the visible region, the absorption peaks that appear in the optical spectra for these compounds indicate high absorption intensity. Furthermore, we have investigated the structural and mechanical stability of these compounds and calculated electron effective masses. Based on in-depth analysis, we have identified TlBiS2, Ba3BiN, Ag2BaS2, and ZrSO as very promising candidates for photovoltaic applications.

An Expansive Exploration for Novel Liquid Electrolytes through Data-Driven Physics-Based Modeling

Electrolyte development plays a key role in the design of current and future battery chemistries. Almost all of the advanced battery technologies such as lithium metal, high voltage cathodes, dual ion batteries, metal air batteries and other beyond lithium ion battery technologies are bottlenecked by the performance of the electrolytes.[i] ,[ii],[iii] There is an increasing need for electrolytes with large electrochemical windows, wide temperature operating windows, very good ionic conductivity for operation at high rates of charge and discharge, low viscosity, good solubility and low flammability and toxicity from a safety standpoint. Thus, there are a large number of constraints on the electrolytes and they are becoming increasingly stringent. To add to the complexity of electrolyte design, there exist multiple trade-offs between all these properties as shown in previous literature.[iv] Although various solid electrolytes, ionic liquids and polymer electrolytes are being researched, organic liquid electrolytes still outperform the other electrolytes when we consider all the properties required. The current approach of new electrolyte design is largely intuition based using certain structural features in the organic molecules. Currently only a few hundred solvents and salts have been demonstrated in all the battery technologies while there are over 20 million organic molecules discovered. The richness of liquid organic compounds is multiplied even further if we start considering mixtures as most of the properties vary non-linearly as you mix two organic compounds. In this work, we suggest a much better data-driven systematic exploration of new electrolytes. The methodology employs determination of various electronic structure features calculated from simple and scalable density functional theory (DFT) calculations and topological features of the molecules. We will identify unique parameters which are non-linear functions of the above features for each of the properties above along with features that govern mixtures. We currently have structures for over 1 million molecules with over 100,000 data points for each property. This gives a unique capability to identify parameters accurately, explore multiple trade-offs and different methods of overcoming these trade-offs. The cathodic stability of the electrolyte can be described in terms of the highest occupied molecular orbital (HOMO) energy level of each of the components. The HOMO level will be estimated using GGA level DFT, quantitative structure-property relationships (QSPR) and higher order theories and comparisons will be shown. The solubility of a salt is governed by the solvation of the individual ions and the dielectric constant of the solvent as shown by the extended Debye- Hückel and other models. We have developed a modified Ising model to describe solvation shell around the Li+ ions, and the model has been solved under mean field approximations to derive the fraction of all species in the Li+ solvation shell.[v] The model can also be extended to describe other cations and anions and is applicable to mixtures. The Ising model coefficients used for the model are the ion-solvent interactions, ion-ion interactions and solvent-solvent interactions can be calculated from DFT. The dielectric constant, important for solubility is correlated to the dipole moment of molecule. This relationship is quite qualitative as in a liquid, the molecules exist as multiple conformers. To provide more quantitative data, we have identified the volume of molecule calculated using DFT along with number of rotatable bonds as additional geometric features to determine the dielectric constant. The dielectric constant along with the solvation energies will be used to give a feature for solubility. The viscosity of an electrolyte is dependent on the solvation shell structure of ions and the viscosity of the solvent which is related to intermolecular interactions and will be explored. We will also show the relationship between the Li+ solvation energy and charge transfer resistance. We believe this electrolyte exploration method is much more systematic, quantitative, scalable and also able to identify outliers to certain trends. This method and database will create a new frontier for electrolyte discovery for most electrochemical systems and also other areas of science. In the future, we will add features for other physical and chemical properties and also extend these to inorganic compounds.

КОМПЬЮТЕРНОЕ МОДЕЛИРОВАНИЕ ЭЛЕКТРОННОЙ СТРУКТУРЫ ГЕМАТИТА

Представлены результаты компьютерного моделирования из первых принципов электронной структуры объемного гематита α-Fe2O3 при нормальных условиях. Расчеты проводились методом линеаризованных присоединенных плоских волн (ЛППВ) в рамках теории функционала плотности (density functional theory, DFT) и с помощью DFT, модифицированной постоянной Хаббарда U. Были получены спектры полной и локальных парциальных плотностей электронных состояний, зонные структуры. Путем сопоставления рассчитанных и известных экспериментальных данных найден метод адекватного описания электронной структуры α-Fe2O3.
 
 Расчеты были проведены на вычислительных мощностях Центра Обработки Данных (ЦОД) Воронежского государственного университета

Quantifying confidence in density functional theory predictions of magnetic ground states

The success of descriptor-based material design relies on eliminating bad candidates and keeping good candidates for further investigation. While DFT has been widely successfully for the former, often times good candidates are lost due to the uncertainty associated with the DFT-predicted material properties. Uncertainty associated with DFT predictions has gained prominence and has led to the development of exchange correlation functionals that have built-in error estimation capability. In this work, we demonstrate the use of built-in error estimation capabilities within the BEEF-vdW exchange correlation functional for quantifying the uncertainty associated with the magnetic ground state of solids. We demonstrate this approach by calculating the uncertainty estimate for the energy difference between the different magnetic states of solids and compare them against a range of GGA exchange correlation functionals as is done in many first principles calculations of materials. We show that this estimate reasonably bounds the range of values obtained with the different GGA functionals. The estimate is determined as a post-processing step and thus provides a computationally robust and systematic approach to estimating uncertainty associated with predictions of magnetic ground states. We define a confidence value (c-value) that incorporates all calculated magnetic states in order to quantify the concurrence of the prediction at the GGA level and argue that predictions of magnetic ground states from GGA level DFT is incomplete without an accompanying c-value. We demonstrate the utility of this method using a case study of Li and Na-ion cathode materials and the c-value metric correctly identifies that GGA level DFT will have low predictability for NaFePO$_4$F.

3R and 2H polytypes of MoS2: DFT and DFPT calculations of structural, optoelectronic, vibrational and thermodynamic properties

We report the results of a theoretical study on the behavior of the structural, optoelectronic, vibrational, including infrared and Raman theoretical spectra, phonon spectrum, and thermodynamic properties of 3R- and 2H- polytypes of molybdenum disulfide (MoS2) using density functional theory (DFT) considering both the local density and generalized gradient approximation, LDA and GGA, respectively. Calculated lattice parameters are close to the experimental measurements, and an indirect band gap E(A→KΓ)=1.33 eV (0.68 eV) was obtained within the GGA (LDA) level of calculation, considering the 3R-polytype, and for the 2H- polytype an indirect band gap E(Γ→KΓ)=1.30 eV (0.70 eV) was obtained within the GGA (LDA) approximation. The complex dielectric function and absorption of 3R-MoS2 and 2H-MoS2 polytypes were shown to be sensitive to the plane of polarization of the incident light. The phonon dispersion relation together with density of states (DOS) as well as theoretical peaks of the infrared (IR) and Raman spectra in the frequency range of 0–800 cm−1 was analyzed and assigned, considering the norm-conserved pseudopotentials. The thermodynamic potentials, the specific heat at constant volume and Debye temperature of the 3R-MoS2 and 2H-MoS2 polytypes are also calculated, whose dependence on the temperature are discussed.