7 views

Uploaded by nagy0peter

- 10.1.1.155
- Molpro Slides
- 2003-Jul-9
- LAA_10748
- Merzbacher 3rd Problems 1 and 2
- (a) 1.759 × 1011 C/Kg (b) 17.59 × 10−11 C/Kg
- Linear Response Theory
- Homework 1 particle physcs
- PS2 Solutions
- 2018Spring Phd
- k 0336268
- Electron Configuration
- Crystals
- merz3
- Apéndices Piezoelectricidad
- The Macroscopic Electric Field Inside a Dielectric
- Nano Optics and Atomics_ Transport of Light and Matter Waves.pdf
- 51377
- Tegus Thesis
- The Unwinding: An Inner History of the New America

You are on page 1of 1

**with separately scaled energy terms
**

Péter Nagy and Ágnes Szabados

Laboratory of Theoretical Chemistry, Eötvös University,Budapest, Hungary

nagy0peter@gmail.com

Short content Introduction

• Feenberg-scaling in perturbation theory (PT) A widely applied approach in quantum chemistry to obtain wavefunctions and energies is perturbation theory (PT). There exist numerous

• Grimme-scaling in Møller-Plesset PT multireference function based formulations which provide perturbative corrections. Multiconfigurational PT (MCPT) is one of these[1].

To improve the second order Møller-Plesset PT a special scaling philosophy was introduced by Grimme [4] in the single-reference frame-

• Multiconfigurational PT (MCPT)

work. This scaling is based on the systematic separation of the different contributions of doubly excited states. It was shown[2] that a similar

• Grimme-scaling as a two parameter Feenberg-scaling scaling can be obtained by a system-dependent optimization procedure, as suggested by Feenberg[3].

• Extension of the theories to MCPT In this work scaling methods are investigated within MCPT, following the ideas of Feenberg and Grimme. We introduce various novel scaling

• Results and conclusion strategies suited for covalent bond dissociation. The basic theory of the applied methods and pilot applications on small model systems are

presented.

**Feenberg-scaling in PT Multiconfigurational PT Generalized Feenberg scaling
**

• Initial partition: Ĥ = Ĥ (0) + Ŵ • Reference function: Ψ0

(0)

= c0|0i +

model space

P

ck |ki • Partitioning: Ĥ = Ĥ (0)′(p) + Ŵ ′(p)

1 µ

k6=0 ∂

E (2)(p) + E (3)(p) = 0

• Repartition: Ĥ = Ĥ (0) + Ŵ − Ĥ (0) µ ∈ ℜ • Determine (p) from (1): ∂p (2)

|1 − µ

{z } | 1 {z

−µ } |0i: principal determinant, |ki: excited determinant

• Separate the energy into terms BG and AG,H: B and A:

Ĥ (0)′ Ŵ ′ • Bi-orthogonal basis set in the CI space:

3 (0)

groups

P in G

P groups

P

∂ P

• Determine µ from: ∂µ E (i)′ = 0 (1) |Ψ0 i and |ki ; k = 0, . . . , NF CI E (2)(p) = pG bk = pGBG

i=0 (0) G k G

**(2) 2 hΨ̃0 | = c10 h0| and hk̃| = hk| − cck0 h0| groups
**

P P in H

in G P

(E ) E (3)(p) =

• The new energy: E (2)′ = E (2)−E (3) and E (3)′ = 0 pGpH akl =

• Bi-orthogonal MCPT formulae (definition of b and a): G,H k6=0 l6=0

Grimme’s spin-component scaling (SCS) (2)

1,2×

P (0)

1 h0|Ŵ |kihk̃|Ŵ |Ψ0 i P

groups

P

(pGpH(AGH + AHG) − δGHpGBG)

EMCPT = − c0 ∆k = bk

• Separate doubles: parallel (’triplet’, |Tk i = b+ + − −

σ aσ iσ jσ |0i) k6=0 k6=0

G,H

**+ − − • Substitution to (2): A p = B provide the paramerters
**

and anti-parallel (’singlet’, e.g. |Sk i = b+σ σ iσ jσ |0i)

a (3)

1,2×

P 1−4×

P h0|Ŵ |kihk̃|Ŵ |lihl̃|Ŵ |Ψ0 i

(0) P

EMCPT = c0∆l ∆k = akl

k6=0 l6=0 k,l6=0

• Singles (L) , triples and quadruples (J) also appear, Specification of the partitioning

at difference with ordinary MP2 and MP3

• (Møller–Plesset: nondiagonal H (0) in MRPT)

• Grouping variants:

• (Epstein-Nesbet: degeneracy leads often to divergence)

◦ S(TLJ): two groups, S and the others, two optimized param.

• Davidson-Kapuy:

|0i |Sk i |Tki ◦ Opt-Fee: one group, no separation, one optimized param.

• System independent scaling parameters: pS = 1.2; pT = 13 ◦ Grimme: scaling the S and T groups with Grimme‘s param. ∆DK DK DK

k = Ek − E0 = hk|F̂ |ki − h0|F̂ |0i

**2× 2× ◦ STLJ: four groups, four optimized param. • Denominator as a difference of IPs and EAs [5]:
**

(2) P |h0|H|Tk i|2 P |h0|H|Sk i|2

ESCS-MP2 = −pT − pS occ. virt.

∆k ∆k

ǫ⊕

P P

Tk Sk ∆IP/EA

k = i − ǫ⊖

a

i i

Parameters pS and pT can be determined following Feenberg, (0) (0) (0) (0)

hΨ0 |Ĥ|Ψ0 i hΨ |a+Ĥa |Ψ i

ǫ⊕

i = − 0 (0) i + i (0)0 (ǫ⊖

a analogously)

but the concept of two parameters requires generalization. Averaging method (0) (0)

hΨ0 |Ψ0 i hΨ0 |ai ai|Ψ0 i

• Averaging before determining the scaling factors,

• sum for certain principals, |ki in the model space

• weight them with their importance:

model

Pspace

c2k BGk

kP

BGAV = c2k

and analogusly for AAV

GH

Scaling in SR case Scaling in DK partitioning

Bond fission of HF molecule, 3−21G basis, IP/EA partitioning Bond fission of HF molecule, 3−21G basis

0.1

CAS 2x2 ref. 0.005

0.08 SRPT2

SRPT3 0.012

Consequences

∆E = EPT−EFCI / a. u.

∆E = EPT−EFCI / a. u.

**S(TLJ)−SRPT3 0.009 0
**

0.06 Opt−Fee−SRPT3

Grimme−SRPT2 0.006

MCPT3 • Essential, non-obvious factors: the partitioning,

0.04 0.003 the grouping strategy and the choice of the princi- −0.005

0 pal determinant MCPT2

MCPT3

0.02 −0.003 S(TLJ)−MCPT3

0.8 1 1.2 1.4 Summary −0.01 Opt−Fee−MCPT3

Grimme−MCPT2

0 STLJ−MCPT3

• Feenberg-scaling is generalizable for the MR case

1 2 3 4 5 6 • We explored several partitionings, grouping vari- 1 2 3 4 5 6

Bond length / Å Bond length / Å

ants and we averaged for principal determinants

• Scaled SRPT does not compete with MCPT upon dissociation • Some scaling variants ameliorate MCPT, e.g. Opt–Fee, Grimme

• There are cases where scaling may provide better

results • Grouping matters a lot (especially scaling of non-doubles)

Further plans

Scaling in IP/EA partitioning • Find a generally applicable scaling strategy Scaling, averaging and damping

Bond fission of HF molecule, 3−21G basis • Reduce the calculation cost by finding Grimme- Bond fission of HF molecule, 3−21G basis

type system independent scaling parameters 0.01

AV−MCPT3

0.006 AV−S(TLJ)−MCPT3

AV−Opt−Fee−MC

0.008 AV−STLJ−MCPT3

0.004 References NONAV−MCPT3

∆E = EPT−EFCI / a. u.

∆E = EPT−EFCI / a. u.

**NONAV−S(TLJ)−MCPT3
**

NONAV−STLJ−MCPT3

0.002 0.006

[1] Á. Szabados, Z. Rolik, G. Tóth, and P. R.

0 Surján. J. Chem. Phys., 122, 114104, (2005).

MCPT2 0.004

MCPT3 [2] Á. Szabados. J. Chem. Phys., 125, 214105,

−0.002 S(TLJ)−MCPT3

Opt−Fee−MCPT3

(2006).

−0.004 Grimme−MCPT2 [3] E. Feenberg. Phys. Rev., 103, 1116, (1956). 0.002

STLJ−MCPT3

for comparison: DK−MCPT3

−0.006

[4] S. Grimme. J. Chem. Phys., 118, 9095, (2003).

0

1 2 3 4 5 6 [5] A. Zaitsevskii, JP. Malrieu. Chem. Phys. Lett., 1 2 3 4 5 6

Bond length / Å 223, 597, (1995). Bond length / Å

• IP/EA denominators remove irregular behaviour at large distance • Averaging and damping improves parallelity

• Scaling may improve MCPT, but the grouping is still an issue

- 10.1.1.155Uploaded byAdam Jermyn
- Molpro SlidesUploaded byshadraws
- 2003-Jul-9Uploaded bySAN JUAN BAUTISTA
- LAA_10748Uploaded bybaconyummy4
- Merzbacher 3rd Problems 1 and 2Uploaded byLuke Burgess
- (a) 1.759 × 1011 C/Kg (b) 17.59 × 10−11 C/KgUploaded bytavva111
- Linear Response TheoryUploaded bynaveedulhaq07
- Homework 1 particle physcsUploaded byAndres Monroy
- PS2 SolutionsUploaded byYenhao Philip Chen
- 2018Spring PhdUploaded byYan-Tung Lin
- k 0336268Uploaded byInternational Organization of Scientific Research (IOSR)
- Electron ConfigurationUploaded byJulia Dhia Azhar
- CrystalsUploaded byShyamol Bose
- merz3Uploaded byjose492432
- Apéndices PiezoelectricidadUploaded byAnonymous TlGnQZv5d7
- The Macroscopic Electric Field Inside a DielectricUploaded by馮維祥
- Nano Optics and Atomics_ Transport of Light and Matter Waves.pdfUploaded byjuridoc
- 51377Uploaded byrpodila
- Tegus ThesisUploaded byvarunganesh29

- ductiity and seismic behaviourUploaded bysakolkong
- Ultrasound tailoring of industrially important nano particlesUploaded byijsret
- c 030102011023Uploaded bytheijes
- Thesis Fluid Solid InteractionUploaded byphucphungvan
- THERMODYNAMICS-2.pptxUploaded byGetachew Tikue
- Leaf Spring2Uploaded byYogi Bhimani
- zumdahl_ch11Uploaded bymelbaz1
- FuelCellLabReport.docxUploaded byKyle Miller
- Cold Reheat EPRIUploaded byIgorToporsky
- Gauss’s LawUploaded bydwalz
- Shape Memory AlloyUploaded byRajesh Govekar
- Membranes Under PrestressUploaded bySparrow Jack
- API Effects of Oil and Chemically DispersedUploaded byHiram Flores
- Ies 2017Uploaded bytvineeth
- Jet DyeingUploaded bySoumen Kundu
- Dynamic Magnification Factor for Impact LoadUploaded byAbukalam Mulla
- Fka Report MassReduction PotentialUploaded byKinezAndrei
- FloTron Configuration Guide, 2015Uploaded byNova Fabrica Ltd
- DMLS OverviewUploaded byRishikesh Patil
- A Study on Supersonic Coherent JetUploaded bychuankit92
- BP1 Bolt Connection CapacityUploaded byballisnothing
- Bridge Design ManualUploaded byKaratunga Kenneth
- 2012 Crodamide Customer PresentationUploaded byVictor Castrejon
- Design Sheet Pile Walls Using Truline Update 1 8-8-13Uploaded byVenkataLakshmiKorrapati
- Dehuminification Without Re-heatUploaded bymaximgh
- Composite Design and TheoryUploaded byapi-3700351
- Curing Concrete _ Concrete Technology _ Portland Cement Association (PCA)Uploaded byJay Patel
- DesalinationUploaded byIsmael Khalil
- JMBC2010Uploaded byXinDa Zhu
- Water Resource EngineeringUploaded byDaeng Sikki