Neural network force fields for simple metals and semiconductors: construction and application to the calculation of phonons and melting temperatures
We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids. Training and test sets are efficiently generated from global structural prediction runs, at the same time assuring the structural variety and importance of sampling the relevant regions of...
Gespeichert in:
| Veröffentlicht in: | Physical chemistry chemical physics : PCCP 2019-03, Vol.21 (12), p.656-6516 |
|---|---|
| Hauptverfasser: | , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 6516 |
|---|---|
| container_issue | 12 |
| container_start_page | 656 |
| container_title | Physical chemistry chemical physics : PCCP |
| container_volume | 21 |
| creator | Marques, Mário R. G Wolff, Jakob Steigemann, Conrad Marques, Miguel A. L |
| description | We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids. Training and test sets are efficiently generated from global structural prediction runs, at the same time assuring the structural variety and importance of sampling the relevant regions of phase space. The neural networks are trained to yield not only good formation energies, but also accurate forces and stresses, which are the quantities of interest for molecular dynamics simulations. Finally, we construct, as an example, several force fields for both semiconducting and metallic elements, and prove their accuracy for a variety of structural and dynamical properties. These are then used to study the melting of bulk copper and gold.
We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids. |
| doi_str_mv | 10.1039/c8cp05771k |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1039_C8CP05771K</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2188985485</sourcerecordid><originalsourceid>FETCH-LOGICAL-c440t-d437a9ebe940071608eec4b65c79d58a047f91ee93dccdcd2681b6af08aa2eb73</originalsourceid><addsrcrecordid>eNpdkUFP3DAQhS1UBJRy4d7KUi-o0lJ77cR2b9WKlgpEe2jPkTOelIATB9tR1R_C_8W7C1upJ8_zfH4ezSPklLNzzoT5CBomVinF7_fIEZe1WBim5atdrepD8jqlO8YYr7g4IIei9EUl9RF5vME5Wk9HzH9CvKddiIC069G7tBY09cPkkQ6YrU_Ujo4mHHoIo5shh5g-0VKnHIvqw7gB7DT5HuxG50DzLVKwHma_vQodnW7DWF5t6AF97sffNOMwYbR5jpjekP2ufIcnz-cx-fXl4ufqcnH9_eu31efrBUjJ8sJJoazBFo1kTPGaaUSQbV2BMq7SlknVGY5ohANw4Ja15m1tO6atXWKrxDE52_pOMTzMmHIz9AnQeztimFOz5FobXRZVFfT9f-hdmONYpiuUkVwqtVxTH7YUxJBSxK6ZYj_Y-LfhrFmH1az06scmrKsCv3u2nNsB3Q59SacAb7dATLDr_ktbPAH6J50h</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2194147725</pqid></control><display><type>article</type><title>Neural network force fields for simple metals and semiconductors: construction and application to the calculation of phonons and melting temperatures</title><source>Royal Society Of Chemistry Journals 2008-</source><source>Alma/SFX Local Collection</source><creator>Marques, Mário R. G ; Wolff, Jakob ; Steigemann, Conrad ; Marques, Miguel A. L</creator><creatorcontrib>Marques, Mário R. G ; Wolff, Jakob ; Steigemann, Conrad ; Marques, Miguel A. L</creatorcontrib><description>We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids. Training and test sets are efficiently generated from global structural prediction runs, at the same time assuring the structural variety and importance of sampling the relevant regions of phase space. The neural networks are trained to yield not only good formation energies, but also accurate forces and stresses, which are the quantities of interest for molecular dynamics simulations. Finally, we construct, as an example, several force fields for both semiconducting and metallic elements, and prove their accuracy for a variety of structural and dynamical properties. These are then used to study the melting of bulk copper and gold.
We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids.</description><identifier>ISSN: 1463-9076</identifier><identifier>EISSN: 1463-9084</identifier><identifier>DOI: 10.1039/c8cp05771k</identifier><identifier>PMID: 30843548</identifier><language>eng</language><publisher>England: Royal Society of Chemistry</publisher><subject>Computer simulation ; Construction ; Free energy ; Gold ; Heat of formation ; Molecular dynamics ; Neural networks ; Test sets</subject><ispartof>Physical chemistry chemical physics : PCCP, 2019-03, Vol.21 (12), p.656-6516</ispartof><rights>Copyright Royal Society of Chemistry 2019</rights><lds50>peer_reviewed</lds50><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c440t-d437a9ebe940071608eec4b65c79d58a047f91ee93dccdcd2681b6af08aa2eb73</citedby><cites>FETCH-LOGICAL-c440t-d437a9ebe940071608eec4b65c79d58a047f91ee93dccdcd2681b6af08aa2eb73</cites><orcidid>0000-0002-7420-5098 ; 0000-0003-1420-4840 ; 0000-0002-3923-4475 ; 0000-0003-0170-8222</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>314,780,784,27924,27925</link.rule.ids><backlink>$$Uhttps://www.ncbi.nlm.nih.gov/pubmed/30843548$$D View this record in MEDLINE/PubMed$$Hfree_for_read</backlink></links><search><creatorcontrib>Marques, Mário R. G</creatorcontrib><creatorcontrib>Wolff, Jakob</creatorcontrib><creatorcontrib>Steigemann, Conrad</creatorcontrib><creatorcontrib>Marques, Miguel A. L</creatorcontrib><title>Neural network force fields for simple metals and semiconductors: construction and application to the calculation of phonons and melting temperatures</title><title>Physical chemistry chemical physics : PCCP</title><addtitle>Phys Chem Chem Phys</addtitle><description>We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids. Training and test sets are efficiently generated from global structural prediction runs, at the same time assuring the structural variety and importance of sampling the relevant regions of phase space. The neural networks are trained to yield not only good formation energies, but also accurate forces and stresses, which are the quantities of interest for molecular dynamics simulations. Finally, we construct, as an example, several force fields for both semiconducting and metallic elements, and prove their accuracy for a variety of structural and dynamical properties. These are then used to study the melting of bulk copper and gold.
We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids.</description><subject>Computer simulation</subject><subject>Construction</subject><subject>Free energy</subject><subject>Gold</subject><subject>Heat of formation</subject><subject>Molecular dynamics</subject><subject>Neural networks</subject><subject>Test sets</subject><issn>1463-9076</issn><issn>1463-9084</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2019</creationdate><recordtype>article</recordtype><recordid>eNpdkUFP3DAQhS1UBJRy4d7KUi-o0lJ77cR2b9WKlgpEe2jPkTOelIATB9tR1R_C_8W7C1upJ8_zfH4ezSPklLNzzoT5CBomVinF7_fIEZe1WBim5atdrepD8jqlO8YYr7g4IIei9EUl9RF5vME5Wk9HzH9CvKddiIC069G7tBY09cPkkQ6YrU_Ujo4mHHoIo5shh5g-0VKnHIvqw7gB7DT5HuxG50DzLVKwHma_vQodnW7DWF5t6AF97sffNOMwYbR5jpjekP2ufIcnz-cx-fXl4ufqcnH9_eu31efrBUjJ8sJJoazBFo1kTPGaaUSQbV2BMq7SlknVGY5ohANw4Ja15m1tO6atXWKrxDE52_pOMTzMmHIz9AnQeztimFOz5FobXRZVFfT9f-hdmONYpiuUkVwqtVxTH7YUxJBSxK6ZYj_Y-LfhrFmH1az06scmrKsCv3u2nNsB3Q59SacAb7dATLDr_ktbPAH6J50h</recordid><startdate>20190320</startdate><enddate>20190320</enddate><creator>Marques, Mário R. G</creator><creator>Wolff, Jakob</creator><creator>Steigemann, Conrad</creator><creator>Marques, Miguel A. L</creator><general>Royal Society of Chemistry</general><scope>NPM</scope><scope>AAYXX</scope><scope>CITATION</scope><scope>7SR</scope><scope>7U5</scope><scope>8BQ</scope><scope>8FD</scope><scope>JG9</scope><scope>L7M</scope><scope>7X8</scope><orcidid>https://orcid.org/0000-0002-7420-5098</orcidid><orcidid>https://orcid.org/0000-0003-1420-4840</orcidid><orcidid>https://orcid.org/0000-0002-3923-4475</orcidid><orcidid>https://orcid.org/0000-0003-0170-8222</orcidid></search><sort><creationdate>20190320</creationdate><title>Neural network force fields for simple metals and semiconductors: construction and application to the calculation of phonons and melting temperatures</title><author>Marques, Mário R. G ; Wolff, Jakob ; Steigemann, Conrad ; Marques, Miguel A. L</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c440t-d437a9ebe940071608eec4b65c79d58a047f91ee93dccdcd2681b6af08aa2eb73</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2019</creationdate><topic>Computer simulation</topic><topic>Construction</topic><topic>Free energy</topic><topic>Gold</topic><topic>Heat of formation</topic><topic>Molecular dynamics</topic><topic>Neural networks</topic><topic>Test sets</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Marques, Mário R. G</creatorcontrib><creatorcontrib>Wolff, Jakob</creatorcontrib><creatorcontrib>Steigemann, Conrad</creatorcontrib><creatorcontrib>Marques, Miguel A. L</creatorcontrib><collection>PubMed</collection><collection>CrossRef</collection><collection>Engineered Materials Abstracts</collection><collection>Solid State and Superconductivity Abstracts</collection><collection>METADEX</collection><collection>Technology Research Database</collection><collection>Materials Research Database</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>MEDLINE - Academic</collection><jtitle>Physical chemistry chemical physics : PCCP</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Marques, Mário R. G</au><au>Wolff, Jakob</au><au>Steigemann, Conrad</au><au>Marques, Miguel A. L</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Neural network force fields for simple metals and semiconductors: construction and application to the calculation of phonons and melting temperatures</atitle><jtitle>Physical chemistry chemical physics : PCCP</jtitle><addtitle>Phys Chem Chem Phys</addtitle><date>2019-03-20</date><risdate>2019</risdate><volume>21</volume><issue>12</issue><spage>656</spage><epage>6516</epage><pages>656-6516</pages><issn>1463-9076</issn><eissn>1463-9084</eissn><abstract>We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids. Training and test sets are efficiently generated from global structural prediction runs, at the same time assuring the structural variety and importance of sampling the relevant regions of phase space. The neural networks are trained to yield not only good formation energies, but also accurate forces and stresses, which are the quantities of interest for molecular dynamics simulations. Finally, we construct, as an example, several force fields for both semiconducting and metallic elements, and prove their accuracy for a variety of structural and dynamical properties. These are then used to study the melting of bulk copper and gold.
We present a practical procedure to obtain reliable and unbiased neural network based force fields for solids.</abstract><cop>England</cop><pub>Royal Society of Chemistry</pub><pmid>30843548</pmid><doi>10.1039/c8cp05771k</doi><tpages>11</tpages><orcidid>https://orcid.org/0000-0002-7420-5098</orcidid><orcidid>https://orcid.org/0000-0003-1420-4840</orcidid><orcidid>https://orcid.org/0000-0002-3923-4475</orcidid><orcidid>https://orcid.org/0000-0003-0170-8222</orcidid></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1463-9076 |
| ispartof | Physical chemistry chemical physics : PCCP, 2019-03, Vol.21 (12), p.656-6516 |
| issn | 1463-9076 1463-9084 |
| language | eng |
| recordid | cdi_crossref_primary_10_1039_C8CP05771K |
| source | Royal Society Of Chemistry Journals 2008-; Alma/SFX Local Collection |
| subjects | Computer simulation Construction Free energy Gold Heat of formation Molecular dynamics Neural networks Test sets |
| title | Neural network force fields for simple metals and semiconductors: construction and application to the calculation of phonons and melting temperatures |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-04T07%3A01%3A06IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Neural%20network%20force%20fields%20for%20simple%20metals%20and%20semiconductors:%20construction%20and%20application%20to%20the%20calculation%20of%20phonons%20and%20melting%20temperatures&rft.jtitle=Physical%20chemistry%20chemical%20physics%20:%20PCCP&rft.au=Marques,%20M%C3%A1rio%20R.%20G&rft.date=2019-03-20&rft.volume=21&rft.issue=12&rft.spage=656&rft.epage=6516&rft.pages=656-6516&rft.issn=1463-9076&rft.eissn=1463-9084&rft_id=info:doi/10.1039/c8cp05771k&rft_dat=%3Cproquest_cross%3E2188985485%3C/proquest_cross%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2194147725&rft_id=info:pmid/30843548&rfr_iscdi=true |