[rvt]Efracio M.

Flores

bulk EuHfO3 from first - principles

Maycol Avalos Sarmientoa , David Olivera Sierrab , Quedin Ovalle Alarconb , Alexandher Chipa Cardenasb ,
Mark Anthony Mosqueira Arandob , guino adriel ayala chaucab
a Universidad Nacional Micaela Bastidas de Apurı́ma, Perú
b São Paulo State University, Brasil

Abstract
ZnSe were performed to identify reasons for the appearance of secondary phases,
Keywords: Multiferroic, polarization, bandgap.

1. Introduction

EuHfO3 on conducting, grounded bases and performing the experiments in situ on the freshly prepared
samples. The measurements were performed at a sample temperature of 18–20 K. We compare the data
with nonrelativistic density functional theory (DFT) calculations based on the local density approximation
(LDA) using the lowtemperature bulk space group P21=n. In this work we use the generalized gradi-
ent approximation (GGA) to the density functional theory (DFT) within a plane-wave pseudopotential
scheme24–26 as implemented in Quantum ESPRESSO. To further explore the electronic structure of the
EuHfO3 cubic, we calculated the projected density of states (PDOS). model system was achieved by replac-
ing ions. The calculated electronic structures reported herein werecarried out using spin-polarized density
functional theory (DFT+U) with periodic model as implemented in Vienna Ab Initio Simulation Package
(VASP 5.3). We employed Dudarev’s approach for DFT+U calculations to correct for the self-interaction er-
ror inherent in current exchange-correlation functionals DFT when applied to transition metals with tightly
localized d electrons. A key advance in the understanding of the bismuthates came in the work of Franc-
cini and co-workers, who showed that DFT calculations using the HSE hybrid density functional could
account quantitatively for the CD-CDW found in EuHfO3. //The accurate and careful measurement of
heat flow is the basis of thermodynamics. In particular, calorimetry, the process of measuring heat into
and out of systems, has lead to remarkae scientific dynamic concept of Gibbs free energy is expressed wit
the enthalpy, T is the temperature, and S is the entropy of the system. The absolute values of these state
variables are impossible to obtain for practical applications. Instead, scientific and engineering fields ob-
tain the critical thermodynamic information by determining the relative change in Gibbs energy. This is
accomplished by measuring changes Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s40192-018-0108-9) contains supplementary material, which is available to autho-
rized users. Taylor D. Sparks sparks@eng.utah.edu 1 Department of Materials Science and Engineering,
University of Utah, Salt Lake City, USA in entropy and enthalpy which are fundamentally tied to heat
transfer and heat capacity. Under most conditions, the change in a system is expressed by the equation

✩ This document is a collaborative effort.

∗ Corresponding author
∗∗ Principal corresponding author

Email address: efracio01@gmail.com (guino adriel ayala chauca)

1 This is the specimen author footnote.

Preprint submitted to Elsevier February 1, 2022

Determining the heat capacity is critical to find changes in entropy and enthalpy, as calorimetry is based on
the relationsh Ti Cp(T )dT for the enthalpy = Tf Ti Cp(T ) T dT for entropy, where Cp is the temperature-
dependent heat capacity at constant pressure. For convenience, heat capacity values are typically fit by
constants: a, b, c, d, e to a power series such as Cp =ynamic uses, measuring heat flow into and out of a
system as a function of temperature is an attractive way to probe phase transitions. First-order transitions
are eerized by latent heating. Higher-order phenomena can also be monitored [14]; examples include the
characterization of glassy transition temperatures in polymers and applications for determining magnetic
and ferroelectric ordering. In addition to studying fundamental science, knowing the value of heat capacity
as a function of temperature is valuable from a design perspective. Considerations of heat capacity as a
physical property are especially useful when Integr Mater Manuf Innov considering design parameters for
thermal engineering applications. The foundational importance of heatr emphasized by the inclusion of heat
capacity tables in almost all introductory engineering textbooks. Moreover, heat capacity also directly influ-
ences other psuch as thermal conductivity. Thermal conductivity measurements are rare due to the difficulty
associated with measuring actual heat flux and properly handling loss from phenomena such as radiation.
Thermal conductivity is instead calculated using thermal diffusivity ( . Despite the utility of heat capacity
for engineers and scientists, careful measurements as a function ofe are not published for all materials of in-
terest. Even in well established repositories of thermochemical data, such as the NIST: JANAF tables, there
are only several hundred crystalline compounds reported. Considering the scope of other crystal structure
databases, such as the inorganic crystal structures database (ICSD) hosting over 188,000 i there is a lack of
corresponding heat capacity data. Heat capacity measurements require equipment and resources that many
labs do not have, which contributes to this information imbalance. A common method for determining heat
capacity is differential scanning calorimetry. Proper use of this technique requires multiple runs (basline,
standard, and sample) which can take hours or days. The measurements are also sensitive to error from
oxidation, poor thermal contact, sintering, volatilization, and others [7]. To overcome the challenges associ-
ated with measuring heat capacity, some researchers have turned to calculating heat capacity for compounds
along side their measurements [17]. A common approach for predicting heat capacity uses Neumann-Kopp
(N-K) estimations based on summing the heat capacity of the constituent elements. While this approach
works surprisingly well at room temperature, it does poorly at high temperatures. Therefore, in an attempt
to extend the utility of the Neumann-Kopp technique, Leitner et al. [9] used the concept of working with
mixed-oxide constituents rather than individual elements. This method works better than the standard
Neumann-Kopp approach at elevated temperatures; however, it relies on a case-b usage as there is no guar-
antee that such oxides and their thermodynamic data exist. Another approach uses contributions from
cation and anion groups [10] to calculate heat capacity—referred to as cation/anion contribution (CAC).
Implementing calculations using the CAC method is both easier and faster than the mixed-oxide method as
they use identifiable cation/anion pairs and a temperature dependent power series. Finally, a more modern
approach to predicting heat capacity relies on the computational determination of phonon frequencies [11]
or other methods based around density functional theory (DFT). The modern approach of simulation is
limited in its use due to the speed of first-principle calculations. In this work, we address a new approach
for determining heat capacity based in the emerging field of materials informatics. This technique leverages
machine learning (ML) algorithms and big data approaches to make statistically calculations. For this work,
we show that a model can be built by training with publicly available data in the NIST: JANAF tables.
The results of this model allow us to compare the performance of machine learning to the performance of
Neumann-Kopp calculations and group contribution methods such as CAC. In this context, we discuss the
potential value of using this approach instead of, or in concert with, the traditional experimental approaches.

In the present work provides a study on the EuHfO3[1] by means of [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
[12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36]
[37] [38] [39] [40]

2
 Fig. 1. Ilustracion de la imagen cubica del trabajo hablado.

2. Computational methodology

For the calculation of the total energy of the EuHfO3, the LINUX operating system is used, for which the
Virtual Box is initially downloaded, which is a virtualization software, also called type 2 hypervisor. It is used
to generate virtualizations of the operating system within a existing physical computer, generating what are
known as virtual machines. Installed within this software is Ubuntu, which is a community-developed operat-
ing system that is perfect for laptops, desktops, and servers. With this software, you can download Quantum
ESPRESSO, which is an integrated set of open-source computer codes for electronic structure calculation and
nanoscale material modeling. It is based on the theory of the density function, plane waves and pseudopo-
tentials. It is divided into a core core of components and a set of plugins that perform more advanced tasks,
as well as a number of third-party packages designed to be interoperable with the core components, such
as: gfortran, libblas-dev, liblapack-dev, libffw3- dev, libopenmpi-dev, etc. The installation is done within
the UBUNTU terminal. Then we proceed to download the pseudopotentials Eu, Hf and O on the following
page: http://www.quantum-espresso.org/ we download these files, we proceed to download the primitive cif
file for the EuHfO3 cell with 5 atoms on the next page https://materialsproject.org/materials/mp-22942/.
Subsequently, the downloaded upf files are installed in the UBUNTU terminal, in order to modify the struc-
ture viewer and the Quantum ESPRESSO input generator, the downloaded primitive cell is used, then the
generated file is copied to the terminal to be able to find the total energy and the other calculations. Our to-
tal energy calculations are based on density functional theory approximating local density, pseudopotentials
that preserve the norm, and the Ceperley-Alder expression for exchange and connections. Wave functions
are expanded in the wave plane using a shear energy E Pw ¡40 Ry. The Brillouin area show is performed
with special Ne points of the Monkhorst-Pack scheme where Np varies between 40 and 100. A Gaussian
broadening method has been used for partially filled bands. To find the position of static equilibrium, they
have calculated the total energy for various values of the cubic lattice constant. The calculated values can
fit well with the Murnaghan equation of state that yields the lattice constant of equilibrium and the volume
modulus. The control calculations produce results very similar to the total energies as for the first choice,
within a constant general change in the energies. To verify the method, we first determine the total energy
of cubic BaBiO3 for six different volumes of the primitive cell between 40 and 100 times the experimental
volume. The calculations of the values are insensitive to different base choices, used to interpolate the charge
density between the spheres and to different k-point tests using 4 and 9 k-points. The control calculations
produce results very similar to the total, the resistivity data confirm our prediction that EuHfO3 remains
insulating up to at least 80 GPa. The authors also report a theoretical phase diagram in excellent agree-
ment with ours, confirming our main conclusion that BaBiO3 remains a charge-ordered insulator up to high
pressures, forming more and more distorted structures.
EuHfO3 on conducting, grounded bases and performing the experiments in situ on the freshly prepared
samples. The measurements were performed at a sample temperature of 18–20 K. We compare the data with
nonrelativistic density functional theory (DFT) calculations based on the local density approximation (LDA)
using the lowtemperature bulk space group P21=n. In this work we use the generalized gradient approxi-
mation (GGA) to the density functional theory (DFT) within a plane-wave pseudopotential scheme24–26 as

3
 Fig. 2. EuHfo3.pdosa tm1(Eu)detalledeldesarrolloenBU N T U terminal.

implemented in Quantum ESPRESSO. To further explore the electronic structure of the EuHfO3 cubic, we
calculated the projected density of states (PDOS). model system was achieved by replacing ions. The cal-
culated electronic structures reported herein werecarried out using spin-polarized density functional theory
(DFT+U) with periodic model as implemented in Vienna Ab Initio Simulation Package (VASP 5.3). We
employed Dudarev’s approach for DFT+U calculations to correct for the self-interaction error inherent in
current exchange-correlation functionals DFT when applied to transition metals with tightly localized d elec-
trons. A key advance in the understanding of the bismuthates came in the work of Franccini and co-workers,
who showed that DFT calculations using the HSE hybrid density functional could account quantitatively for
the CD-CDW found in EuHfO3. This computationally more intensive HSE functional corrects the tendency
of LDA calculations to overestimate screening. The HSE functional (and equivalent approaches) accounts
for the non-local dynamic correlations present in the bismuthates.
All electronic structure calculations reported in this work were performed with the Quantum-ESPRESSO
package [? ] []

sin(x)
1

0
y

-1

-2π -π 0 π 2π
x

Fig. 3. Illustration of Relaxed ZnO surfaces of 6 layers with 3 layers of ZnSe and ZnTe used in the current
work.

3. Results and discussion

Machine learning, in itself, requires no knowledge of the underlying physical principles when predicting
heat capacity. The process of converting the element-derived features into a prediction are handled by the
algorithm using purely statistical methods. In the case of heat capacity, the origin and mechanism has been
well known for many years.
We expect the most important features to align with the physical mechanisms of heat capacity. In fact,
the three most important features when predicting heat capacity at elevated temperatures, not including
Neumann-Kopp calculations, are: 1. valence-shell s electrons 2. Unfilled f valence electrons 3. Outer

4
 Fig. 4. ” EuHfo3.pdosa tm1(Eu)w tc1(s”

Fig. 5. ”EuHfo3.pdost ot”

Fig. 6. ”EuHfo3.pdost op(eu)

shell electrons These features are all properties typically associated with the bonding of the molecule. The
traditional understanding of heat capacity, which utilizes the vibrational frequencies in the solid, relates well
to this observation since bond strength and type will influence vibrational frequencies. Thus, the observed
features reinforce the currently understood physical mechanism of heat capacity. At room temperature,
machine learning performs on par with Neumann-Kopp and has comparable performance to the CAC method
up to 1000 K, as seen in Fig. 5. Beyond 1000, CAC error begins to rise due to the limitations with using a
power series. All models have similar temperature behavior, with random forest exhibiting the lowest error.
(A comparison of support vector regression and linear regression to CAC error with respect to temperature
5
 Fig. 7. ”EuHfo3.pdosa pt”

can be seen in Figures S6 and S7 in the supplementary information.) The Neumann-Kopp method of
predicting heat capacity has traditionally been abandoned at elevated temperature values, forcing some to
rely on the cumbersome Cation/Anion contribution method, or DFT calculations. These processes are time
intensive, and in some instances, oxide and Cation/Anion values themselves are not well results in this
work suggest that machine learning predictions can not only replace as a high-throughput method at room
temperature, but can also be used at elevated temperatures with acceptable prediction accuracy. Machi

Fig. 8. ”EuHfo3.pdosa pt”

3.1. Crystal Structure

In ?? B), the ZnO/ZnSe interface

4. Conclusions

In summary, in this work were implemented DFT+U calculations to investigate the electronic structure,
optical properties of three bulk

6
 Table 1. Lattice parameters determined by DFT (GGA) for the structures studied
Structure Calculated Lattice parameters Lattice parameters experimental
ZnO a= 3.290Å c=5.31Å a= 3.249Å c/a=1,60Å[? ]
ZnSe a= 5.3123Å a= 5.67Å[? ]
ZnTe a= 5.802740Å a= 6.11Å[? ]

Table 2. Lattice parameters determined by DFT + U for the structures studied

Structure Calculated Lattice parameters Energy of band gap (Eg )
ZnO a= 3.250Å c=5.21Å Eg = 3.3eV
ZnSe a= 5.7095Å Eg = 2.6725eV
ZnTe a= 6.13685Å Eg = 2.257eV

Acknowledgements

The authors are grateful by financial support

References
[1] P. V. Balachandran, A. A. Emery, J. E. Gubernatis, T. Lookman, C. Wolverton, A. Zunger, Predictions of new ab o 3
perovskite compounds by combining machine learning and density functional theory, Physical Review Materials 2 (2018)
043802.
[2] Detección de perovskitas abo3 estables y metaestables mediante el aprendizaje automático y el proyecto de materiales,
Computational Materials Science (????).
[3] S. A. Dar, V. Srivastava, U. K. Sakalle, V. Parey, G. Pagare, Dft investigation on electronic, magnetic, mechanical and
thermodynamic properties under pressure of some eumo3 (m= ga, in) perovskites, Materials Research Express 4 (2017)
106104.
[4] Y. Liu, V. R. Cooper, B. Wang, H. Xiang, Q. Li, Y. Gao, J. Yang, Y. Zhou, B. Liu, Discovery of ab o3 perovskites
as thermal barrier coatings through high-throughput first principles calculations, Materials Research Letters 7 (2019)
145–151.
[5] T. Birol, C. J. Fennie, Origin of giant spin-lattice coupling and the suppression of ferroelectricity in eutio 3 from first
principles, Physical Review B 88 (2013) 094103.
[6] D. Yuan, S. Zhang, Z. Xiang, Y. He, Y. Wang, Y. Liu, X. Zhao, X. Zhou, Q. Zhang, Highly efficient removal of thorium
in strong hno3 media using a novel polymer adsorbent bearing a phosphonic acid ligand: A combined experimental and
density functional theory study, ACS applied materials & interfaces 11 (2019) 24512–24522.
[7] Aplicación de cálculos de primeros principios de alto rendimiento en innovación cerámica, Journal of Materials Science
Technology (????).
[8] T. Mazaheri, B. Sun, J. Scher-Zagier, A. Thind, D. Magee, P. Ronhovde, T. Lookman, R. Mishra, Z. Nussinov, Stochastic
replica voting machine prediction of stable cubic and double perovskite materials and binary alloys, Physical Review
Materials 3 (2019) 063802.
[9] J. Cheng, K. Yang, Design of two-dimensional electron gas systems via polarization discontinuity from large-scale first-
principles calculations, Journal of Materials Chemistry C 6 (2018) 6680–6690.
[10] ¿cómo podemos controlar los ”bloques de elementos” en los cristales de óxido de metal en transición?, in: Springer (Ed.),
Nuevos materiales poliméricos basados en bloques de elementos, ????
[11] B. Sun, T. Mazaheri, J. Scher-Zagier, D. Magee, P. Ronhovde, T. Lookman, Z. Nussinov, Stochastic replica voting machine
prediction of stable perovskite and binary alloys (????).
[12] M. Boubchir, H. Aourag, Materials genome project: Mining the ionic conductivity in oxide perovskites, Materials Science
and Engineering: B 267 (2021) 114984.
[13] S. Vorotylo, K. Bardasova, A. Saenko, et al., Phase formation in mechanochemical synthesis of europium and lanthanum
hafnates: Ab initio modeling and experimental study, - (2020) 63–68.
[14] G. Pilania, P. V. Balachandran, C. Kim, T. Lookman, Finding new perovskite halides via machine learning, Frontiers in
Materials 3 (2016) 19.
[15] G. Pilania, P. Balachandran, J. E. Gubernatis, T. Lookman, Classification of abo3 perovskite solids: a machine learning
study, Acta Crystallographica Section B: Structural Science, Crystal Engineering and Materials 71 (2015) 507–513.
[16] J. Schmidt, J. Shi, P. Borlido, L. Chen, S. Botti, M. A. Marques, Predicting the thermodynamic stability of solids
combining density functional theory and machine learning, Chemistry of Materials 29 (2017) 5090–5103.

7
[17] A. A. Emery, C. Wolverton, High-throughput dft calculations of formation energy, stability and oxygen vacancy formation
energy of abo 3 perovskites, Scientific data 4 (2017) 1–10.
[18] Q. Xu, Z. Li, M. Liu, W.-J. Yin, Rationalizing perovskite data for machine learning and materials design, The journal of
physical chemistry letters 9 (2018) 6948–6954.
[19] Z. Li, Q. Xu, Q. Sun, Z. Hou, W.-J. Yin, Thermodynamic stability landscape of halide double perovskites via high-
throughput computing and machine learning, Advanced Functional Materials 29 (2019) 1807280.
[20] P. V. Balachandran, B. Kowalski, A. Sehirlioglu, T. Lookman, Experimental search for high-temperature ferroelectric
perovskites guided by two-step machine learning, Nature communications 9 (2018) 1–9.
[21] C. J. Bartel, C. Sutton, B. R. Goldsmith, R. Ouyang, C. B. Musgrave, L. M. Ghiringhelli, M. Scheffler, New tolerance
factor to predict the stability of perovskite oxides and halides, Science advances 5 (2019) eaav0693.
[22] A. Ali, H. Park, R. Mall, B. Aı̈ssa, S. Sanvito, H. Bensmail, A. Belaidi, F. El-Mellouhi, Machine learning accelerated
recovery of the cubic structure in mixed-cation perovskite thin films, Chemistry of Materials 32 (2020) 2998–3006.
[23] G. Song, J. Su, S. Fang, J. Tong, X. Xu, F. Yang, N. Zhang, F. Chang, Crystal structural, dielectrical properties and
high temperature magnetic phase transition of sm1-xcaxfeo3 (x= 0–0.3), Journal of Alloys and Compounds 757 (2018)
279–287.
[24] N. V. Orupattur, S. H. Mushrif, V. Prasad, Catalytic materials and chemistry development using a synergistic combination
of machine learning and ab initio methods, Computational Materials Science 174 (2020) 109474.
[25] Z. Li, L. E. Achenie, H. Xin, An adaptive machine learning strategy for accelerating discovery of perovskite electrocatalysts,
ACS Catalysis 10 (2020) 4377–4384.
[26] K. Min, E. Cho, Accelerated discovery of potential ferroelectric perovskite via active learning, Journal of Materials
Chemistry C 8 (2020) 7866–7872.
[27] B. R. Sutherland, Solar materials find their band gap, Joule 4 (2020) 984–985.
[28] P. Ilanchezhiyan, G. M. Kumar, C. Siva, H. Cho, S. Tamilselvan, S. Seal, T. Kang, D. Kim, Enhancing defect densities
in smerxfe1-xo3 nanostructures and tuning their electrical characteristics for photocatalytic and photoresponse functions,
Journal of Materials Research and Technology 9 (2020) 12585–12594.
[29] C. Yang, Y. Gan, M. Lee, C. Ren, K. S. Brinkman, R. D. Green, X. Xue, Structural evolution, electrochemical kinetic
properties, and stability of a-site doped perovskite sr 1- x yb x coo 3- δ, Journal of Materials Chemistry A 8 (2020)
10450–10461.
[30] Q.-Y. Zhang, J.-J. Han, Y. Huang, Y. Chen, X. Yan, W.-Z. Lang, Effect of ba non-stoichiometry in ba1-xzr0. 1ce0. 7y0.
2o3-δ on its structure defect, sinterability and hydrogen permeability, Ceramics International 46 (2020) 19564–19573.
[31] H. Kwon, J. Park, B.-K. Kim, J. W. Han, Effect of b-cation doping on oxygen vacancy formation and migration in labo
3: A density functional theory study, Journal of the Korean Ceramic Society 52 (2015) 331–337.
[32] E. D. Wachsman, Development of a lower temperature sofc, ECS Transactions 25 (2009) 783.
[33] H. Aourag, A. Guittom, Mining the bulk positron lifetime, physica status solidi (b) 246 (2009) 392–396.
[34] H. Shi, C. Su, R. Ran, J. Cao, Z. Shao, Electrolyte materials for intermediate-temperature solid oxide fuel cells, Progress
in Natural Science: Materials International (2020).
[35] X. Yang, L. Jia, B. Pan, B. Chi, J. Pu, J. Li, Mechanism of proton conduction in doped barium cerates: a first-principles
study, The Journal of Physical Chemistry C 124 (2020) 8024–8033.
[36] Y. Zhuo, A. M. Tehrani, A. O. Oliynyk, A. C. Duke, J. Brgoch, Identifying an efficient, thermally robust inorganic
phosphor host via machine learning, Nature communications 9 (2018) 1–10.
[37] S. K. Kauwe, J. Graser, A. Vazquez, T. D. Sparks, Machine learning prediction of heat capacity for solid inorganics,
Integrating Materials and Manufacturing Innovation 7 (2018) 43–51.
[38] T. Lookman, P. V. Balachandran, D. Xue, R. Yuan, Active learning in materials science with emphasis on adaptive
sampling using uncertainties for targeted design, npj Computational Materials 5 (2019) 1–17.
[39] A. O. Oliynyk, L. A. Adutwum, J. J. Harynuk, A. Mar, Classifying crystal structures of binary compounds ab through
cluster resolution feature selection and support vector machine analysis, Chemistry of Materials 28 (2016) 6672–6681.
[40] C. Kim, G. Pilania, R. Ramprasad, Machine learning assisted predictions of intrinsic dielectric breakdown strength of
abx3 perovskites, The Journal of Physical Chemistry C 120 (2016) 14575–14580.