Efficient Exploitation of Numerical Quadrature with Distance-Dependent Integral Screening in Explicitly Correlated F12 Theory: Linear Scaling Evaluation of the Most Expensive RI-MP2-F12 Term.

Lars Urban, Henryk Laqua, Travis H Thompson, Christian Ochsenfeld
Author Information
  1. Lars Urban: Chair of Theoretical Chemistry, Department of Chemistry, University of Munich (LMU), D-81377 Munich, Germany. ORCID
  2. Henryk Laqua: Chair of Theoretical Chemistry, Department of Chemistry, University of Munich (LMU), D-81377 Munich, Germany.
  3. Travis H Thompson: Chair of Theoretical Chemistry, Department of Chemistry, University of Munich (LMU), D-81377 Munich, Germany.
  4. Christian Ochsenfeld: Chair of Theoretical Chemistry, Department of Chemistry, University of Munich (LMU), D-81377 Munich, Germany. ORCID

Abstract

We present a linear scaling atomic orbital based algorithm for the computation of the most expensive exchange-type RI-MP2-F12 term by employing numerical quadrature in combination with CABS-RI to avoid six-center-three-electron integrals. Furthermore, a robust distance-dependent integral screening scheme, based on integral partition bounds [Thompson, T. H.; Ochsenfeld, C. 044101], is used to drastically reduce the number of the required three-center-one-electron integrals substantially. The accuracy of our numerical quadrature/CABS-RI approach and the corresponding integral screening is thoroughly assessed for interaction and isomerization energies across a variety of numerical integration grids. Our method outperforms the standard density fitting/CABS-RI approach with errors below 1 μE even for small grid sizes and moderate screening thresholds. The choice of the grid size and screening threshold allows us to tailor our ansatz to a desired accuracy and computational efficiency. We showcase the approach's effectiveness for the chemically relevant system valinomycin, employing a triple-ζ F12 basis set combination (CHNO, 5757 AO basis functions, 10,266 CABS basis functions, 735,783 grid points). In this context, our ansatz achieves higher accuracy combined with a 135× speedup compared to the classical density fitting based variant, requiring notably less computation time than the corresponding RI-MP2 calculation. Additionally, we demonstrate near-linear scaling through calculations on linear alkanes. We achieved an 817-fold acceleration for CH and an extrapolated 28,765-fold acceleration for CH, resulting in a substantially reduced computational time for the latter─from 229 days to just 11.5 min. Our ansatz may also be adapted to the remaining MP2-F12 terms, which will be the subject of future work.

References

  1. J Chem Theory Comput. 2022 Jul 12;18(7):4203-4217 [PMID: 35666238]
  2. J Chem Phys. 2016 Aug 7;145(5):054117 [PMID: 27497549]
  3. J Chem Phys. 2010 Oct 21;133(15):154109 [PMID: 20969372]
  4. J Chem Theory Comput. 2015 Nov 10;11(11):5269-76 [PMID: 26574321]
  5. J Chem Phys. 2023 Jun 7;158(21): [PMID: 37259997]
  6. J Chem Phys. 2011 Jan 21;134(3):034113 [PMID: 21261336]
  7. J Chem Phys. 2015 May 21;142(19):194106 [PMID: 26001446]
  8. J Chem Phys. 2008 Feb 28;128(8):084102 [PMID: 18315028]
  9. J Comput Chem. 2012 Mar 15;33(7):810-6 [PMID: 22228538]
  10. Phys Chem Chem Phys. 2008 Jan 7;10(1):106-13 [PMID: 18075688]
  11. J Chem Theory Comput. 2014 Jan 14;10(1):90-101 [PMID: 26579894]
  12. J Chem Theory Comput. 2017 Jun 13;13(6):2571-2580 [PMID: 28448125]
  13. J Chem Phys. 2021 Jan 28;154(4):044101 [PMID: 33514114]
  14. Phys Chem Chem Phys. 2014 Oct 28;16(40):22167-78 [PMID: 25213320]
  15. J Comput Chem. 2011 Aug;32(11):2492-513 [PMID: 21590779]
  16. J Chem Theory Comput. 2022 Oct 11;18(10):5978-5991 [PMID: 36099641]
  17. Phys Chem Chem Phys. 2010 Sep 21;12(35):10460-8 [PMID: 20603665]
  18. J Chem Theory Comput. 2019 Oct 8;15(10):5319-5331 [PMID: 31503475]
  19. J Chem Phys. 2008 Aug 21;129(7):071101 [PMID: 19044752]
  20. J Chem Theory Comput. 2017 Jun 13;13(6):2712-2716 [PMID: 28561575]
  21. J Chem Phys. 2007 Jan 7;126(1):014108 [PMID: 17212491]
  22. J Chem Phys. 2017 Dec 7;147(21):214101 [PMID: 29221401]
  23. J Chem Phys. 2013 Apr 7;138(13):134114 [PMID: 23574215]
  24. J Chem Phys. 2017 Aug 14;147(6):064110 [PMID: 28810785]
  25. J Chem Phys. 2016 May 28;144(20):204112 [PMID: 27250284]
  26. J Chem Phys. 2014 Jan 28;140(4):044118 [PMID: 25669516]
  27. J Chem Theory Comput. 2011 Aug 9;7(8):2427-2438 [PMID: 21836824]
  28. Phys Chem Chem Phys. 2006 May 7;8(17):1985-93 [PMID: 16633685]
  29. J Chem Theory Comput. 2023 Mar 28;19(6):1734-1743 [PMID: 36912635]
  30. J Chem Phys. 2021 Jun 7;154(21):214116 [PMID: 34240990]
  31. J Chem Phys. 2008 Sep 14;129(10):101103 [PMID: 19044900]
  32. J Phys Chem A. 2015 Mar 5;119(9):1642-56 [PMID: 25325889]
  33. J Chem Phys. 2004 Sep 8;121(10):4479-85 [PMID: 15332877]
  34. J Chem Phys. 2017 Jul 14;147(2):024103 [PMID: 28711054]
  35. J Chem Phys. 2018 Mar 21;148(11):114104 [PMID: 29566500]
  36. J Chem Phys. 2022 Sep 14;157(10):104104 [PMID: 36109222]
  37. J Org Chem. 2007 Mar 16;72(6):2118-26 [PMID: 17286442]
  38. J Chem Phys. 2010 Apr 21;132(15):154104 [PMID: 20423165]
  39. J Chem Theory Comput. 2013 Aug 13;9(8):3364-3374 [PMID: 24098094]
  40. J Chem Phys. 2008 Jun 28;128(24):244113 [PMID: 18601323]
  41. J Chem Phys. 2008 May 7;128(17):174103 [PMID: 18465906]
  42. J Chem Theory Comput. 2017 Jul 11;13(7):3153-3159 [PMID: 28636392]
  43. J Chem Theory Comput. 2016 Sep 13;12(9):4254-62 [PMID: 27434098]
  44. Chem Rev. 2012 Jan 11;112(1):75-107 [PMID: 22176553]
  45. J Chem Theory Comput. 2016 Oct 11;12(10):4915-4924 [PMID: 27598837]
  46. J Chem Phys. 2010 Jun 21;132(23):231102 [PMID: 20572681]
  47. J Chem Phys. 2013 May 14;138(18):181104 [PMID: 23676020]
  48. J Chem Phys. 2010 Oct 14;133(14):141103 [PMID: 20949980]
  49. J Chem Phys. 2020 Nov 14;153(18):184115 [PMID: 33187416]
  50. J Chem Theory Comput. 2018 Mar 13;14(3):1403-1411 [PMID: 29431996]
  51. J Chem Phys. 2021 Sep 14;155(10):104109 [PMID: 34525816]
  52. J Chem Theory Comput. 2020 Mar 10;16(3):1456-1468 [PMID: 32053375]
  53. J Chem Phys. 2018 Nov 28;149(20):204111 [PMID: 30501270]
  54. J Chem Theory Comput. 2015 Sep 8;11(9):4226-37 [PMID: 26575918]
  55. J Chem Phys. 2012 May 28;136(20):204110 [PMID: 22667543]
  56. Phys Chem Chem Phys. 2008 Jun 21;10(23):3410-20 [PMID: 18535724]
  57. J Chem Theory Comput. 2015 Nov 10;11(11):5291-304 [PMID: 26574323]
  58. J Chem Phys. 2014 Sep 7;141(9):094106 [PMID: 25194363]
  59. Phys Chem Chem Phys. 2017 Dec 13;19(48):32184-32215 [PMID: 29110012]
  60. J Chem Theory Comput. 2021 Mar 9;17(3):1512-1521 [PMID: 33615784]
  61. J Chem Phys. 2021 Nov 21;155(19):191101 [PMID: 34800963]
  62. J Chem Theory Comput. 2022 Jul 12;18(7):4218-4228 [PMID: 35674337]
  63. J Chem Theory Comput. 2015 Mar 10;11(3):918-22 [PMID: 26579745]
  64. J Chem Phys. 2004 Jul 1;121(1):117-29 [PMID: 15260528]
  65. J Chem Phys. 2007 Dec 14;127(22):221106 [PMID: 18081383]
  66. J Chem Phys. 2019 Jan 28;150(4):044101 [PMID: 30709269]
  67. Chem Rev. 2013 Jan 9;113(1):36-79 [PMID: 23020161]
  68. J Chem Theory Comput. 2015 Apr 14;11(4):1540-8 [PMID: 26574364]
  69. J Chem Phys. 2020 Oct 28;153(16):164115 [PMID: 33138430]
  70. J Phys Chem Lett. 2022 Oct 13;13(40):9332-9338 [PMID: 36178852]
  71. J Chem Theory Comput. 2017 Apr 11;13(4):1691-1698 [PMID: 28245126]
  72. J Chem Phys. 2022 Oct 7;157(13):134107 [PMID: 36209011]
  73. J Chem Theory Comput. 2018 Jul 10;14(7):3451-3458 [PMID: 29905484]
  74. J Chem Phys. 2007 Apr 28;126(16):164102 [PMID: 17477584]

Word Cloud

Created with Highcharts 10.0.0screeningbasednumericalintegralaccuracygridansatzbasislinearscalingcomputationRI-MP2-F12employingcombinationintegralssubstantiallyapproachcorrespondingdensitycomputationalF12functionstimeaccelerationCHpresentatomicorbitalalgorithmexpensiveexchange-typetermquadratureCABS-RIavoidsix-center-three-electronFurthermorerobustdistance-dependentschemepartitionbounds[ThompsonTHOchsenfeldC044101]useddrasticallyreducenumberrequiredthree-center-one-electronquadrature/CABS-RIthoroughlyassessedinteractionisomerizationenergiesacrossvarietyintegrationgridsmethodoutperformsstandardfitting/CABS-RIerrors1μEevensmallsizesmoderatethresholdschoicesizethresholdallowsustailordesiredefficiencyshowcaseapproach'seffectivenesschemicallyrelevantsystemvalinomycintriple-ζsetCHNO5757AO10266CABS735783pointscontextachieveshighercombined135×speedupcomparedclassicalfittingvariantrequiringnotablylessRI-MP2calculationAdditionallydemonstratenear-linearcalculationsalkanesachieved817-foldextrapolated28765-foldresultingreducedlatter─from229daysjust115minmayalsoadaptedremainingMP2-F12termswillsubjectfutureworkEfficientExploitationNumericalQuadratureDistance-DependentIntegralScreeningExplicitlyCorrelatedTheory:LinearScalingEvaluationExpensiveTerm

Similar Articles

Cited By