The Triton and Three-Nucleon Force in Nuclear Lattice ... · Triton and three-nucleon force in...
Transcript of The Triton and Three-Nucleon Force in Nuclear Lattice ... · Triton and three-nucleon force in...
Introduction Lattice χEFT χEFT Simulations Nuclear matter
The Triton and Three-Nucleon Force
in Nuclear Lattice Simulations
B. Borasoy1,2, H. Krebs1, D. Lee3, U.-G. Meißner1,2
1Bonn University
2FZ Julich
3NCSU, Raleigh
INT 06-1 Workshop
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Outline
1 Introduction
2 Chiral effective field theory on the lattice
3 Lattice simulations of chiral effective field theory for the tritonContinuum resultsNuclear lattice results
4 Simulations of nuclear matter
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Introduction
Lattice field theory allows nonperturbative treatment of QCD
Modern lattice QCD simulations work with lattices not muchlarger than size of single nucleon
In forseeable future: lattice QCD simulations not suited toobtain direct results in few- and many-body nuclear physics(beyond lightest nuclei)
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Introduction
Lattice field theory allows nonperturbative treatment of QCD
Modern lattice QCD simulations work with lattices not muchlarger than size of single nucleon
In forseeable future: lattice QCD simulations not suited toobtain direct results in few- and many-body nuclear physics(beyond lightest nuclei)
Employ effective field theory on the lattice:pions and nucleons are point particles on lattice sites,external (axial-) vector fields live on links
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Chiral effective field theory
QCD exhibits (approximate) SU(2)L × SU(2)R chiralsymmetry, broken down spontaneously to SU(2)V
⇒ 3 Goldstone bosons (π+, π−, π0) with small masses
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Chiral effective field theory
QCD exhibits (approximate) SU(2)L × SU(2)R chiralsymmetry, broken down spontaneously to SU(2)V
⇒ 3 Goldstone bosons (π+, π−, π0) with small masses
At low energies: chiral perturbation theory, the effective fieldtheory of QCD, is successful in describing interactions amongmesons and baryons.
Green’s functions are expanded in Goldstone boson massesand small momenta ⇒ chiral counting scheme
ChPT is model independent, effective field theory of QCD
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Chiral perturbation theory
Chiral effective Lagrangian has same symmetries andsymmetry breaking patterns as QCD
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Chiral perturbation theory
Chiral effective Lagrangian has same symmetries andsymmetry breaking patterns as QCD
Consequence of confinement: quarks and gluons do not showup as explicit degrees of freedom in effective Lagrangian
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Chiral perturbation theory
Chiral effective Lagrangian has same symmetries andsymmetry breaking patterns as QCD
Consequence of confinement: quarks and gluons do not showup as explicit degrees of freedom in effective Lagrangian
Pions can be summarized in matrix U(x) ∈ SU(2)
U(x) = exp(
ifπ
π(x))
fπ ≃ 93 MeV,pion decay constant
L = L(U, ∂U, ∂2U, . . . ,M) , M = diag(mu, md)
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Chiral perturbation theory
Chiral effective Lagrangian has same symmetries andsymmetry breaking patterns as QCD
Consequence of confinement: quarks and gluons do not showup as explicit degrees of freedom in effective Lagrangian
Pions can be summarized in matrix U(x) ∈ SU(2)
U(x) = exp(
ifπ
π(x))
fπ ≃ 93 MeV,pion decay constant
L = L(U, ∂U, ∂2U, . . . ,M) , M = diag(mu, md)
Nucleons (p, n) can be included
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Lattice χEFT
Lattice ChPT with mesons has been investigated in
− Myint & Rebbi (1994); Levi, Lubicz & Rebbi (1997)− Shushpanov & Smilga (1999)
Extension to baryonic sector
− Lewis & Ouimet (2001)− Borasoy, Lewis & Ouimet (2002) (2004)
Studies on a finite lattice
− Borasoy & Lewis (2004)
Multi-nucleon effective field theory
− Chandrasekharan, Pepe, Steffen & Mazur (2003)− Lee, Borasoy & Schafer (2004), Lee & Schafer (2005),− Borasoy, Krebs, Lee & Meißner (2005)
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Lattice χEFT
Euclidean SU(2) chiral Lagrangian in mesonic sector
L = L2 + L4 + . . .
L2=f2
π
4
⟨
∇(+)µ U†∇
(+)µ U
⟩
−Bf2
π
2
⟨
M(U + U†)⟩
L4= −14 l1
⟨
∇(±)µ U†∇
(±)µ U
⟩2
− 14 l2
⟨
∇(±)µ U†∇
(±)ν U
⟩⟨
∇(±)µ U†∇
(±)ν U
⟩
+ . . .
with pion fields
U(x) = exp
(
iτaπa(x)
fπ
)
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Lattice χEFT
Use of nearest-neighbor covariant derivative in leading orderLagrangian avoids unphysical states
∇(+)µ U(x) =
1
a
[
Rµ(x)U(x + aµ)L†µ(x) − U(x)
]
with external fields
Lµ(x)=exp [−iaℓµ(x)] = exp [−ia(Vµ(x) − Aµ(x))]
Rµ(x)=exp [−iarµ(x)] = exp [−ia(Vµ(x) + Aµ(x))]
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force
Work in SU(4)-symmetric (Wigner) limit of pionless effectivefield theory: isospin- and spin-symmetric
L = ψ†
(
i∂0 +~∇2
2m
)
ψ −C0
2
(
ψ†ψ)2
−D0
6
(
ψ†ψ)3
ψ includes four nucleon states with mass m
ψ =
p↑p↓n↑
n↓
Renormalized two-body interaction C0 is directly related totwo-body scattering length a2
C0 =4πa2
m
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Continuum regularization
Two-body binding energy B2
B2 =1
ma22
SU(4)-symmetric limit: 1S0 and 3S1 two-body sectors aredegenerate. We choose SU(4)-symmetric two-nucleon bindingenergy B2 = 1 MeV
Three-body interaction D0 not fixed
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Continuum regularization
Two-body binding energy B2
B2 =1
ma22
SU(4)-symmetric limit: 1S0 and 3S1 two-body sectors aredegenerate. We choose SU(4)-symmetric two-nucleon bindingenergy B2 = 1 MeV
Three-body interaction D0 not fixed
Solve homogeneous S-wave bound state equation fordimer-nucleon system with cutoff momentum Λ
⇒ strong dependence of D0 on Λ
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Continuum regularization
Three-body interaction D0 as function of triton binding energy E3
-4e-07
-3e-07
-2e-07
-1e-07
0
1e-07
2e-07
3e-07
4e-07
-14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0
D 0 (
MeV
-5)
E3 (MeV)
Λ/π = 30 MeVΛ/π = 40 MeVΛ/π = 50 MeVΛ/π = 60 MeVΛ/π = 70 MeV
Nontrivial dependence of D0 on Λ is nonperturbative effect,no finite set of diagrams reproduces ultraviolet divergence(Bedaque, Hammer & van Kolck (’99))
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force
D0 scales roughly as Λ−2 for E3 ≪ −1 MeV
Pole in D0(E3) close to continuum threshold fordimer plus nucleon
Pole location decreases for larger Λ
As Λ is increased, new deeper bound states appear which areoutside range of validity of EFT
Triton is identified with shallowest bound state
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force
D0 scales roughly as Λ−2 for E3 ≪ −1 MeV
Pole in D0(E3) close to continuum threshold fordimer plus nucleon
Pole location decreases for larger Λ
As Λ is increased, new deeper bound states appear which areoutside range of validity of EFT
Triton is identified with shallowest bound state
Singular behavior of three-body system could lead to differentcutoff dependence of D0 in different regularization scheme
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force in nuclear lattice simulations
Path integral is evaluated by computing Monte Carlo sampleof world lines
x
t
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force in nuclear lattice simulations
Path integral is evaluated by computing Monte Carlo sampleof world lines
x
t
Strength of two-body coefficient C0 is matched to deuteronbinding energy B2 (we take 1 MeV)
Three-body interaction D0 not fixed
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Two-body interaction
Two-body coupling C0 determined by summingnucleon-nucleon bubble diagrams on the lattice
1 2 nΣn = 0
∞
Tune C0 to reproduce deuteron binding energy B2
a−1(MeV) a−1t (MeV) C0(MeV−2)
40 16 −1.83 × 10−4
50 25 −1.39 × 10−4
60 36 −1.13 × 10−4
70 49 −0.94 × 10−4
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Contributions to path integral
Hopping parameter: h = αt2m
, αt = ata
Single fermion worldline:
hop to a neighboring lattice site during time step: h
no hop during time step: 1 − 6h
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Contributions to path integral
Hopping parameter: h = αt2m
, αt = ata
Single fermion worldline:
hop to a neighboring lattice site during time step: h
no hop during time step: 1 − 6h
Two different fermions at same spatial site and no hop duringtime step: e−C0αt(1 − 6h)2
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Contributions to path integral
Hopping parameter: h = αt2m
, αt = ata
Single fermion worldline:
hop to a neighboring lattice site during time step: h
no hop during time step: 1 − 6h
Two different fermions at same spatial site and no hop duringtime step: e−C0αt(1 − 6h)2
Three different fermions at same spatial site and no hopduring time step: e−3C0αt(1 − 6h)3
Factor e−C0αt for each pairwise interaction
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Contributions to path integral
Hopping parameter: h = αt2m
, αt = ata
Single fermion worldline:
hop to a neighboring lattice site during time step: h
no hop during time step: 1 − 6h
Two different fermions at same spatial site and no hop duringtime step: e−C0αt(1 − 6h)2
Three different fermions at same spatial site and no hopduring time step: e−3C0αt(1 − 6h)3
Factor e−C0αt for each pairwise interaction
Contribution from three-body interaction for three differentfermions: e−D0αt(1 − 6h)3
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force in nuclear lattice simulations
Compute lattice approximation for
〈0, 0, 0| exp [−βH] |0, 0, 0〉
|0, 0, 0〉: state with three nucleons, each of different kind andzero momentum
⇒ measurement of triton ground state energy E3
-1e-08
-5e-09
0
5e-09
1e-08
1.5e-08
2e-08
-14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0
D 0 (
MeV
-5)
E3 (MeV)
1/a = 30 MeV1/a = 40 MeV1/a = 50 MeV1/a = 60 MeV
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force in nuclear lattice simulations
Coupling D0 scales as ∼ a2 for fixed E3 (modulo shift)
Energy region E3 ∼ −1 MeV has not been computed sincethere are many dimer plus nucleon continuum states nearenergy threshold → difficult to extract E3
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force in nuclear lattice simulations
Coupling D0 scales as ∼ a2 for fixed E3 (modulo shift)
Energy region E3 ∼ −1 MeV has not been computed sincethere are many dimer plus nucleon continuum states nearenergy threshold → difficult to extract E3
Results have been confirmed using Hamiltonian lattice withLanczos method
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Hamiltonian lattice
Compare with Hamiltonian lattice using Lanczos method
-1e-08
-5e-09
0
5e-09
1e-08
1.5e-08
2e-08
-14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0
D 0 (
Me
V-5
)
E3 (MeV)
1/a = 30 MeV1/a = 40 MeV1/a = 50 MeV
D0 scales roughly as a2 (modulo shift)
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Hamiltonian lattice
Compare with Hamiltonian lattice using Lanczos method
-1e-08
-5e-09
0
5e-09
1e-08
1.5e-08
2e-08
-14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0
D 0 (
Me
V-5
)
E3 (MeV)
1/a = 30 MeV1/a = 40 MeV1/a = 50 MeV
D0 scales roughly as a2 (modulo shift)
Hamiltonian approach restricted to small volumesand small # of particles
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Hamiltonian lattice
Compare with Hamiltonian lattice using Lanczos method
-1e-08
-5e-09
0
5e-09
1e-08
1.5e-08
2e-08
-14 -13 -12 -11 -10 -9 -8 -7 -6 -5 -4 -3 -2 -1 0
D 0 (
Me
V-5
)
E3 (MeV)
1/a = 30 MeV1/a = 40 MeV1/a = 50 MeV
D0 scales roughly as a2 (modulo shift)
Hamiltonian approach restricted to small volumesand small # of particles
Euclidean lattice method can be generalized to larger nucleonnumbers and more complicated forces amongst nucleons
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force in nuclear lattice simulations
To be done:
More nucleons
Inclusion of four-body force
Effects due to breaking of Wigner symmetry
Higher chiral orders in effective Lagrangian
Inclusion of pions
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Triton and three-nucleon force in nuclear lattice simulations
To be done:
More nucleons
Inclusion of four-body force
Effects due to breaking of Wigner symmetry
Higher chiral orders in effective Lagrangian
Inclusion of pions
Important step towards future many-body simulations witharbitrary number of nucleons
Neutron matter with pions (Lee, Borasoy & Schafer (’04))
→ results for hot neutron matter, T ≈ 20 − 40 MeV, anddensities twice below nuclear matter density
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Lattice action positivity
Cold dilute neutron matter: pionless EFT should provideadequate description of low-energy physics
Implementation on the lattice with a positive semi-definiteaction via Hubbard-Stratonovich transformation→ efficient Monte Carlo simulations
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Lattice action positivity
Cold dilute neutron matter: pionless EFT should provideadequate description of low-energy physics
Implementation on the lattice with a positive semi-definiteaction via Hubbard-Stratonovich transformation→ efficient Monte Carlo simulations
Cold dilute nuclear matter with small proton fraction:3-nucleon force is required for consistent renormalization
3-nucleon force could spoil positivity of lattice action
For triton binding energies of about 8 MeV and assumingfour-nucleon force to be zero (or small) condition for latticeaction positivity [Chen, Lee & Schafer (’04)] is satisfied:no sign problem
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter Continuum results Nuclear lattice results
Concluding remark:
Lattice simulations of chiral effective field theory are a promisingtool to investigate few-body nuclear physics
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Nuclear matter
A central goal of nuclear physics:
understand properties of strongly interacting matterat finite density and temperature
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Nuclear matter
A central goal of nuclear physics:
understand properties of strongly interacting matterat finite density and temperature
Experiments
new data generated by RHIC, Brookhaven
upcoming heavy ion facility planned at GSI, Darmstadt
high energy frontier: ALICE @ LHC
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Nuclear matter
Astrophysical interest
development of early universe
stellar collapse
properties of neutron stars
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Nuclear matter
Astrophysical interest
development of early universe
stellar collapse
properties of neutron stars
Lattice QCD
connects QCD to observed phenomenology
finite temperature ok.
finite density, i.e. chemical potential
⇒ determinant of quark Dirac matrix becomes complex
⇒ highly oscillatory
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Simulations of nuclear matter
Study of neutron matter with nuclear lattice simulations
first step (w/ pions): Lee, Borasoy & Schafer (2004)
pionless theory: Lee & Schafer (2005)
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Simulations of nuclear matter
Study of neutron matter with nuclear lattice simulations
first step (w/ pions): Lee, Borasoy & Schafer (2004)
pionless theory: Lee & Schafer (2005)
w/ pions: sign/phase problem of fermion determinant is muchless severe; well behaved for temperatures above 3 MeV
Better situation than in finite density lattice QCD
Why?
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Simulations of nuclear matter
Study of neutron matter with nuclear lattice simulations
first step (w/ pions): Lee, Borasoy & Schafer (2004)
pionless theory: Lee & Schafer (2005)
w/ pions: sign/phase problem of fermion determinant is muchless severe; well behaved for temperatures above 3 MeV
Better situation than in finite density lattice QCD
Why?
nucleons and pions give a simpler representation of theessential physics in the hadronic phase
nucleons are much heavier than up and down quarks
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Simulations of nuclear matter
Results of numerical simulation at weak coupling agree withresults from perturbation theory for neutron & pionself-energies, shift in average energy
Neutron-neutron contact interaction coupling C determinedby S-wave scattering phase shifts on lattice at zerotemperature and density
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Results
Density versus chemical potential
0
0.5
1
1.5
2
-40 -20 0 20 40 60 80 100
ρ/ρ N
phys
µ-mN
phys (MeV)
ResultsFree neutron
Loop calculation
T phys = 25.0 MeV, a−1 = 150 MeV
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Results
Energy per neutron
20
30
40
50
60
70
80
0 0.5 1 1.5 2
Eph
ys/A
- m
N
phys
(M
eV)
ρ/ρN
phys
ResultsFree neutron
Loop calculation
T phys = 25.0 MeV, a−1 = 150 MeV
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Results
Different lattice volumes
20
30
40
50
60
70
80
0 0.5 1 1.5 2
Ephys
/A -
mN
phys
(M
eV
)
ρ/ρN
phys
L = 3, Tphys = 37.5 MeVL = 4, Tphys = 37.5 MeVL = 5, Tphys = 37.5 MeVL = 3, Tphys = 25.0 MeVL = 4, Tphys = 25.0 MeVL = 5, Tphys = 25.0 MeV
T phys =25 MeV and 37.5 MeV , a−1 = 150 MeV
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Results
Different lattice spacings
25
30
35
40
45
50
55
60
65
0 0.5 1 1.5 2
Ephys
/A -
mN
phys
(M
eV
)
ρ/ρN
phys
a-1 = 150 MeV, L = 3
a-1 = 200 MeV, L = 4
T phys =37.5 MeV, a−1 = 150 MeV and 200 MeV
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Conclusions
Probe larger volumes, lower temperatures, greater
nuclear densities than lattice QCD
Coupling C is determined by fitting to NN scattering data
Cutoff dependence is absorbed into C
Realistic simulation of many-body nuclear phenomena with• no free parameters
• a systematic expansion
• clear theoretical connection to QCD
Include protons, charged pions, higher orders, 3N forces etc.
INT ’06 Triton and 3NF
Introduction Lattice χEFT χEFT Simulations Nuclear matter
Concluding remark:
Lattice simulations of chiral effective field theory are a promisingtool to investigate few- and many-body nuclear physics
INT ’06 Triton and 3NF