nuclear shell modelcanhp2015/slide/week2/tabe.pdf · 2015-10-05 · ``computational advances in...
TRANSCRIPT
![Page 1: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/1.jpg)
Infinite basis-space extrapolation of ground-state energies of light nuclei in the no-core Monte Carlo shell model
YIPQS long term workshop``Computational Advances in Nuclear and Hadron Physics”
CANHP2015
Takashi Abe (U of Tokyo)
YITP, Kyoto U
Sep. 29, 2015
(Sep. 28 – Oct. 2, 2015)1
![Page 2: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/2.jpg)
Ab inito approaches
• Major challenge of nuclear physics
- Understand the nuclear structure & reactions from ab-initio calculations w/ realistic nuclear forces (potentials)
- ab-initio approaches in nuclear structure physics (A > 4):
GFMC, NCSM (A ~ 12-16), CC (sub-shell closure +/- 1,2),
Self-consistent Green’s Function theory, IM-SRG, Lattice EFT, …
• demand for extensive computational resources
ab-initio(-like) SM approaches (which attempt to go) beyond standard methods
- IT-NCSM, IT-CI: R. Roth (TU Darmstadt), P. Navratil (TRIUMF), …
- SA-NCSM: T. Dytrych, J.P. Draayer (Louisiana State U), …
- No-Core Monte Carlo Shell Model (MCSM)
2
![Page 3: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/3.jpg)
• Solve the non-relativistic Schroedinger eq.
and obtain the eigenvalues and eigenvectors.
– Ab initio: All nucleons are active, and Hamiltonian consists of realistic NN (+ 3N + …) potentials.
• Two main sources of uncertainties:– Nuclear forces (interactions btw/among nucleons)
In principle, they should be obtained (directly) by QCD.
– Many-body methods
CI: Finite basis space (choice of basis function and truncation)
We have to extrapolate to infinite basis dimensions
3
“Ab initio” in low-energy nuclear structure physics
![Page 4: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/4.jpg)
Shell model (Configuration Interaction, CI)
• Eigenvalue problem of large sparse Hamiltonian matirx
4
Large sparse matrix (in M-scheme)
# non-zero MEs
![Page 5: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/5.jpg)
M-scheme dimension in Nshell truncation
DM
Nshell=1Nshell=2
Nshell=3Nshell=4
Nshell=5
.
.
.
.
.
.
16O (M=0)
12C (M=0)10B (M=3)
4He (M=0)
5
6He (M=0), 6Li(M=1)
7Li (M = 3/2)
8Be (M=0)Current FCI limit
(= emax + 1)
lower p-shell
upper p-shell
No-core calculations
NN only
![Page 6: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/6.jpg)
• Importance truncation
Standard shell model
Monte Carlo shell model
All Slater determinants
Important bases stochastically selected
H =
H ~
Diagonalization
Diagonalization
Review: T. Otsuka , M. Honma, T. Mizusaki, N. Shimizu, Y. Utsuno, Prog. Part. Nucl. Phys. 47, 319 (2001)6
dMCSM ~ O(100)
d > O(1010)
Monte Carlo shell model (MCSM)
![Page 7: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/7.jpg)
• 2nd-quantized non-rel. Hamiltonian (up to 2-body term, so far)
• Eigenvalue problem
• MCSM many-body wave function & basis function
• Deformed SDs
( cα† … spherical HO basis)
7
These coeff. are obtained by the diagonalization.
This coeff. is obtained by a stochastic sampling & CG.
SM Hamiltonian & MCSM many-body w.f.
![Page 8: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/8.jpg)
4He(0+;gs)
12C(0+;gs)
Nshell = 2
Nshell = 2
Nshell = 3
Nshell = 3
Nshell = 4
Nshell = 4
Nshell = 5
Exact result is unknown
JISP16 NN int.
w/ optimum hw
w/o Coulomb force
w/o spurious CoM treatment
8
DM ~ 100
DM ~ 3 x 103
DM ~ 4 x 104
DM ~ 3 x 105
DM ~ 100
DM ~ 8 x 107
DM ~ 6 x 1011
Nshell = 5 DM ~ 6 x 1014
Energies wrt # of basis & energy variance
![Page 9: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/9.jpg)
• Two steps of the extrapolation
1. Extrapolation of our MCSM (approx.) results to the FCI (exact) results in the fixed size of model space -> Energy-variance extrapolation
2. Extrapolation into the infinite model space
- Exponential fit w.r.t. Nmax in the NCFC
- IR- & UV-cutoff extrapolations
9
N. Shimizu, Y. Utsuno, T. Mizusaki, T. Otsuka, T. Abe, & M. Honma, Phys. Rev. C82, 061305(R) (2010)
MCSM
FCI1. Extrp into FCI
2. Extrp into inf.
basis dim.
NCFC
Extrapolations in the no-core MCSM
![Page 10: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/10.jpg)
• Background:
Recent development of computational technologies
(e.g., ~ 10 PFLOPS @ K computer)
& quantum many-body techniques
(e.g., shell model (CI), Lanczos diagonalization ~ 1010 x 1010)
• Purpose:
Full ab initio solutions
• Methods:
Extrapolations to the infinite HO-basis space motivated by an EFT idea
• Goal:
Infinite basis-space extrapolation in the no-core MCSM
10
Extrapolation to the infinite HO-basis space
![Page 11: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/11.jpg)
• Two ways of the extrapolation to the infinite basis space
1. Traditional exponential form (w/ fixed hw)
2. Cutoff extrapolations
- IR-cutoff extrapolation (w/ UV-saturated data)
- IR- & UV-cutoff extrapolations (w/ any data, ideally)
11
S. A. Coon, M. I. Avetian, M. K. G. Kruse, U. van Kolck, P. Maris, J. P. Vary, Phys. Rev. C86, 054002 (2012)
S. A. Coon, arXiv:1303.6358
S. A. Coon, arXiv:1408.0738 (NTSE-2013 proceedings)
P. Maris, A. M. Shirokov, & J. P. Vary, Phys. Rev. C79, 014308 (2009)
R. J. Furnstahl, G, Hagen, T. Papenbrock, Phys. Rev. C86, 031301(R) (2012)
S. N. More, A. Ekstrom, R. J. Furnstahl, G. Hagen, T. Papenbrock, Phys. Rev. C87, 044326 (2013)
R. J. Furnstahl, S. N. More, T. Papenbrock, Phys. Rev. C89, 044301 (2014)
R. J. Furnstahl, G. Hagen, T. Papenbrock, K. A. Wendt, J. Phys. G: Nucl. Part. Phys. 42, 034032 (2015)
Extrapolation to the infinite basis space
E. D. Jurgenson, P. Maris, R. J. Furnstahl, W. E. Ormand & J. P. Vary, Phys. Rev. C87, 054312 (2013)
(N, hω) <-> (λ, Λ)
![Page 12: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/12.jpg)
Λ: UV cutoff -> + infinity
λ: IR cutoff -> 0
Low-energy physics
12
ΛNN: UV cutoff of the NN interaction
~ 600 MeV (JISP16)
λNN: IR cutoff of the NN interaction
~ 60 MeV (JISP16)
Separation of the energy/momentum scale
![Page 13: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/13.jpg)
.
.
.
λ: IR cutoff
Λ: UV cutoff
N
hω
λsc: IR cutoff
R
13
IR & UV cutoffs in HO basis
![Page 14: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/14.jpg)
.
.
.
λ: IR cutoff
Λ: UV cutoff
N
hω
λsc: IR cutoff
R
14
IR & UV cutoffs in HO basis
![Page 15: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/15.jpg)
4He 0+ g.s.
JISP16 NN int.
(w/o Coulomb)
Nshell = 2 - 7
Nsh
ell
Energ
y (
MeV
)
λsc (MeV)
( Λ, λsc )
λsc: IR cutoffΛ: UV cutoff15
hω (MeV)
( Nshell, hω )
Traditional extrapolation IR-cutoff extrapolation
![Page 16: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/16.jpg)
Nsh
ell
Energ
y (
MeV
)
λsc (MeV)
c.f.) NCFC: -29.164(2) MeV
Extrapolated results to infinite Nmax space
MCSM(IR cutoff): ~ -29.14(1) MeV
16
O(100) keV error O(10) keV error
hω (MeV)
MCSM(empirical): -29.4 ~ -29.1 MeV
(Nshell = 3 - 7, hw = 15 - 35 MeV)
4He 0+ g.s.
JISP16 NN int.
(w/o Coulomb)
Nshell = 2 - 7
Traditional extrapolation IR-cutoff extrapolation
![Page 17: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/17.jpg)
ΛUV (MeV)
En
erg
y (
Me
V)
λsc (MeV)
c.f.) NCFC: -29.164(2) MeV
Extrapolated results to infinite Nmax space
MCSM(IR cutoff): ~ -29.14(1) MeV MCSM(UV cutoff): -29.12(1) MeV
17
UV-cutoff extrapolation IR-cutoff extrapolation
O(10) keV error
4He 0+ g.s.
JISP16 NN int.
(w/o Coulomb)
Nshell = 2 - 7
![Page 18: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/18.jpg)
4He 0+ g.s.
JISP16 NN int.
(w/o Coulomb)
Nshell = 2 - 7
Energ
y (
MeV
)
c.f.) NCFC: -29.164(2) MeV
Extrapolated results to infinite Nmax space
MCSM(IR- & UV-cutoff): -29.14(1) MeV
18
IR- & UV-cutoff extrapolation
... Λ: UV cutoff
λsc: IR cutoff
N
hw
O(10) keV error
![Page 19: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/19.jpg)
19
+: Experimental data (AME2012)
x: MCSM w/ JISP16 NN int.
4He (0+;g.s.)
8Be (0+;g.s.)
12C (0+;g.s.)
16O (0+;g.s.)
20Ne (0+;g.s.)
Preliminary
MCSM results show good agreements w/ experimental data up to 12C, slightly overbound for 16O, and clearly overbound for 20Ne.
MCSM results are obtained by traditional extrapolation w/ optimum harmonic oscillator energies.Coulomb interaction is included perturbatively.
Comparison of MCSM results w/ experiments
![Page 20: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/20.jpg)
20
+: Experimental data (AME2012)
x: MCSM w/ JISP16 NN int.
4He (0+;g.s.)
8Be (0+;g.s.)
12C (0+;g.s.)
16O (0+;g.s.)
20Ne (0+;g.s.)
Preliminary
MCSM results show good agreements w/ experimental data up to 12C, slightly overbound for 16O, and clearly overbound for 20Ne.
MCSM results are obtained by traditional extrapolation w/ optimum harmonic oscillator energies.Coulomb interaction is included perturbatively.
Comparison of MCSM results w/ experiments
![Page 21: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/21.jpg)
• MCSM results of g.s. energies for light nuclei can be extrapolated to the infinite basis space.
• JISP16 NN interaction gives good agreement w/ experimental data up to 12C, slightly overbound for 16O, and clearly overbound for 20Ne.
• MCSM algorithm/computation
- Better error estimate for the extrapolations
- Inclusion of the 3-body force
• Physics
- Other observables (rms radius, …)
- Other p- & sd-shell nuclei
21
Summary
Perspective
![Page 22: Nuclear Shell Modelcanhp2015/slide/week2/Tabe.pdf · 2015-10-05 · ``Computational Advances in Nuclear and Hadron Physics” CANHP2015 Takashi Abe (U of Tokyo) YITP, Kyoto U Sep](https://reader034.vdocuments.net/reader034/viewer/2022050308/5f7013a2dd932f4d9b04a46f/html5/thumbnails/22.jpg)
• U of Tokyo– Takaharu Otsuka (Department of Physics & CNS)
– Noritaka Shimizu (CNS)
• JAEA– Yutaka Utsuno
• Iowa State U– James P. Vary
– Pieter Maris
22
Collaborators