Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
motivation simulation results outlook
Strange nucleon form factors and isoscalarcharges with Nf = 2 + 1 O(a)-improved Wilson
fermions
Dalibor Djukanovic, Konstantin Ottnad, Harvey Meyer,Georg von Hippel, Jonas Wilhelm, Hartmut Wittig
21 June, 2019
1 / 32
motivation simulation results outlook
table of contents
1 motivation
2 simulationform factorsensemblessetuprenormalizationz-expansion
3 resultsstrange vector form factorsstrange axial vector form factorsstrange tensor chargequark flavor contributions to gA and gT
4 outlook
2 / 32
motivation simulation results outlook
Standard Model observables
weak mixing angle ΘW1
APV =GµQ
2
4πα√
2
{QW (p)− F (Ei ,Q
2)}
QW (p) = 1− 4 sin2 Θw
→ F (Ei ,Q2) contains G s
E , G sM and G s
A
1Becker et al., 1802.047593 / 32
motivation simulation results outlook
Standard Model observables
contribution of intrinsic spin of quark flavor f
G fA(0) = g f
A = ∆f
1
2=∑f
(1
2∆f + Lf
)+ Jg
1
neutron electric dipole moment
dn =∑f
dγf gfT
⇒ appearance of quark-disconnected contributions
1X.-D. Ji, Phys. Rev. Lett. 78, 610 (1997), arXiv:hep-ph/96032494 / 32
motivation simulation results outlook
Previous Work
Strange nucleon form factors with Nf = 2 + 1O(a)-improved Wilson fermions
D. Djukanovic, H. Meyer, K. Ottnad, G. von Hippel,J. Wilhelm, H. Wittig, arXiv:1810.10810
Strange electromagnetic form factors of the nucleon withNf = 2 + 1 O(a)-improved Wilson fermions
D. Djukanovic, K. Ottnad, J. Wilhelm, H. Wittig,arXiv:1903.12566
Nucleon isovector charges and twist-2 matrix elements withNf = 2 + 1 dynamical Wilson quarks
T. Harris, G. v. Hippel, P. Junnarkar, H. Meyer, K. Ottnad,J. Wilhelm, H. Wittig, L. Wrang, arXiv:1905.01291
5 / 32
motivation simulation results outlook
form factors
construct a ratio
RJµ(~q;~p′; Γν) =CN
3,Jµ(~q, z0;~p′, y0; Γν)
CN2 (~p′, y0)
√CN
2 (~p′, y0)CN2 (~p′, z0)CN
2 (~p′-~q, y0-z0)
CN2 (~p′-~q, y0)CN
2 (~p′-~q, z0)CN2 (~p′, y0-z0)
s.d.= M1
νµ(~q, ~p′)G1(Q2) + M2νµ(~q, ~p′)G2(Q2)
(overdetermined) system of equations at each Q2
M ~G = ~R ; M =
M11
...M1
N
M21
...M2
N
, ~G =
(G1
G2
), ~R =
R1...
RN
minimizing the least-squares function
χ2 = (~R −M~G )T C−1 (~R −M~G )
6 / 32
motivation simulation results outlook
form factor parametrization
⟨N, ~p, s |Jµ(x)|N, ~p′, s ′
⟩= us(~p) Jµ us
′(~p′)e iq·x
axial vector current
Aµ = γµγ5GA(Q2) + γ5qµ2m
GP(Q2)
vector current
Vµ = γµF1(Q2) + iσµνqν
2mF2(Q2)
F1,F2 ↔ GE,GM
tensor current at Q2 = 0
Tµν = iσµνgT
7 / 32
motivation simulation results outlook
ensembles
Coordinated Lattice Simulations (CLS)1,2
Nf = 2 + 1 O(a)-improved Wilson fermionstree-level improved Luscher-Weisz gauge actionopen boundary conditions in timetr Mq = const
β a [fm] N3s × Nt mπ[MeV] mK [MeV] Ndis
cfg Nconcfg
H102 3.40 0.08636 323 × 96 352 436 997 997
H105 3.40 0.08636 323 × 96 278 461 1019 1019
C101 3.40 0.08636 483 × 96 223 473 489 1956
N401 3.46 0.07634 483 × 128 289 462 701 701
N203 3.55 0.06426 483 × 128 345 441 768 1536
N200 3.55 0.06426 483 × 128 282 462 852 852
D200 3.55 0.06426 643 × 128 203 482 234 936
N302 3.70 0.04981 483 × 128 359 458 1177 1177
1CLS, 1411.39822CLS, 1608.08900
8 / 32
motivation simulation results outlook
setup
nucleon interpolator
Nα(x) = εabc
(uaβ(x) (Cγ5)βγ db
γ (x))
ucα(x)
smeared-smeared quark propagators
truncated solver method for two- and three-point functions1,2
CN2/3 =
1
NLPsrc
NLPsrc∑
n=1
CN2/3(xn)LP︸ ︷︷ ︸
biased estimate
+
1
NHPsrc
NHPsrc∑
n=1
(CN
2/3(xn)HP − CN2/3(xn)LP
)︸ ︷︷ ︸
bias correction
1G. Bali et al., PoS (LATTICE 2015) 3502E. Shintani et al., Phys. Rev. D91 114511 (2015)
9 / 32
motivation simulation results outlook
lattice currents
improved local axial current
Aµ(~z , z0)Imp. = Aµ(~z , z0) + acA ∂µP(~z , z0)
improved conserved vector current
Vµ(~z , z0)Imp. = Vµ(~z , z0) + acV ∂νTνµ(~z , z0)
non-perturbative determination of cA and cV1,2
local tensor current
1Alpha Collaboration, Nucl. Phys. B 896 (2015) 5552Gerardin et al., Phys. Rev. D 99, 014519 (2019)
10 / 32
motivation simulation results outlook
contributions
disconnected: G l(Q2) and G s(Q2)
ux y
d
u
zl/s
connected: Gu+dcon. (Q2)
ux y
d
u
z
+ · · ·
11 / 32
motivation simulation results outlook
contributions
disconnected:
three-point function factorizes
CN,l/s3,Jµ
(~q, z0;~p′, y0; Γν) =⟨Ll/sJµ
(~q, z0) · CN2 (~p′, y0; Γν)⟩G
parity and spin projectors
Γ0 =1
2(1 + γ0) , Γk = Γ0 iγ5γk
connected:
sequential propagators for three-point functiondrop A0 due to large excited-state contaminationparity and spin projector
Γ =1
2(1 + γ0)(1 + iγ5γ3)
12 / 32
motivation simulation results outlook
source positions
disconnected:
NHPsrc = 1 or 4, NLP
src = 32on 7 timeslices evenly distributed around Nt/2separated by seven timeslices without sources
connected:
NHPsrc = 1 or 4, NLP
src = 16− 48on 1 timeslicessame sources for two- and three-point functionsalready used in Mainz isovector charges paper1
1K. Ottnad et al., 1905.0129113 / 32
motivation simulation results outlook
quark loopsLl/sΓµ
(~q, z0) = −∑~z∈Λ
e i~q·~z⟨tr[S l/s(z ; z) Γµ
]⟩G
= −∑~z∈Λ
e i~q·~z⟨η†(z) Γµ s l/s(z)
⟩G ,η
hierarchical probing1
ηn → hn � η
4D noise vectors, 2×512 Hadamard vectors/configuration
1Stathopoulos et al., 1302.401814 / 32
motivation simulation results outlook
excited-state contamination
G eff(Q2, z0, y0) = G (Q2) + A1(Q2) e−∆z0 + B1(Q2) e−∆′(y0−z0)
+ C1(Q2) e−∆z0−∆′(y0−z0)
disconnected:plateau method y0 ∼ 1 fmsummation method y0 ∈ [0.5, 1.3] fm
S(Q2, y0) =
y0−1∑z0=1
G eff(Q2, z0, y0)
= K (Q2) + y0G (Q2) + O(e−∆y0 , e−∆′y0
)
connected + disconnected:summation method with 3 or 4 y0 ∈ [1.0, 1.5] fmtwo-state fits
15 / 32
motivation simulation results outlook
renormalization
Nf = 3 periodic boundary condition ensembles
β ∈ {3.40, 3.46, 3.55}RI′-MOM scheme in Landau gauge
ZO 〈p |OΓ| p〉|p2=µ2 = 〈p |OΓ| p〉tree|p2=µ2 .
extrapolation to the chiral limitperturbative subtraction of leading-order lattice artifacts1
conversion to RGI scheme
MS as intermediate schemeMS β- and γ-functionsfit residual µ-dependence
convert to MS at µ = 2.0 GeV
1G. von Hippel et al., PoS LATTICE2016 (2016) 19416 / 32
motivation simulation results outlook
renormalization
start with flavor-diagonal basis
OaΓ(x) = ψ(x)Γλaψ(x) , ψ = (u, d , s)T , a ∈ {3, 8, 0} ,
ZΓ =
Z 33Γ 0 00 Z 88
Γ 00 0 Z 00
Γ
, Z 33Γ = Z 88
Γ , Z iq = Zq .
basis transformation Ou−dΓ (x)R
Ou+dΓ (x)ROs
Γ(x)R
=
Zu−d ,u−dΓ 0 0
0 Zu+d ,u+dΓ Zu+d ,s
Γ
0 Z s,u+dΓ Z s,s
Γ
Ou−d
Γ (x)
Ou+dΓ (x)Os
Γ(x)
Zu−d ,u−d
Γ already used in Mainz isovector charges paper1
1K. Ottnad et al., 1905.0129117 / 32
motivation simulation results outlook
renormalization
linear extrapolation to β = 3.7
error multiplied with factor 10checks performed for gu−d
A in Mainz isovector charges paper1
⇒ consistent results compared to using ZSFA
0.75
0.76
0.77
0.78
0.79
0.8
0.81
0.82
0.83
3.4 3.45 3.5 3.55 3.6 3.65 3.7
(Z33 A)R
GI
β
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
3.4 3.45 3.5 3.55 3.6 3.65 3.7
(Z00 A)R
GI
β
1K. Ottnad et al., 1905.0129118 / 32
motivation simulation results outlook
z-expansion
fit the Q2-dependence
G (Q2) =5∑
k=0
akz(Q2)k
z(Q2) =
√tcut + Q2 −
√tcut√
tcut + Q2 +√tcut
, tcut = (2mK )2
Gaussian priors
ak = 0± 5 max{|a0|, |a1|} ∀ k > 1 (1)
for electric form factor: a0 = 0
charge radius
r2 ≡ −6dG
dQ2
∣∣∣∣Q2=0
= − 3
2tcuta1 (2)
19 / 32
motivation simulation results outlook
strange vector form factors
mK ≈ 460 MeV
−0.004
−0.002
0
0.002
0.004
0.006
0.008
0.01
0.012
0 0.2 0.4 0.6 0.8 1 1.2
Gs E(Q
2)
Q2 [GeV2]
a = 0.08636 fma = 0.07634 fma = 0.06426 fm
−0.035
−0.03
−0.025
−0.02
−0.015
−0.01
−0.005
0
0.005
0.01
0.015
0 0.2 0.4 0.6 0.8 1 1.2
Gs M(Q
2)
Q2 [GeV2]
a = 0.08636 fma = 0.07634 fma = 0.06426 fm
β = 3.55
−0.006
−0.004
−0.002
0
0.002
0.004
0.006
0.008
0.01
0.012
0 0.2 0.4 0.6 0.8 1
Gs E(Q
2)
Q2 [GeV2]
mK = 482MeVmK = 462MeVmK = 441MeV
−0.07
−0.06
−0.05
−0.04
−0.03
−0.02
−0.01
0
0.01
0 0.2 0.4 0.6 0.8 1
Gs M(Q
2)
Q2 [GeV2]
mK = 482MeVmK = 462MeVmK = 441MeV
20 / 32
motivation simulation results outlook
extrapolation
SU(3) HBChPT1
−0.4
−0.35
−0.3
−0.25
−0.2
−0.15
−0.1
−0.05
−1.14 −1.12 −1.1 −1.08 −1.06 −1.04 −1.02 −1 −0.98
χ2/dof = 0.84
( r2) s E/t
0
log(√t0mK)
β = 3.40
H102
H105
C101
β = 3.46
N401
β = 3.55
N203 N200
D200
β = 3.70
N302
−0.9
−0.8
−0.7
−0.6
−0.5
−0.4
−0.3
−0.2
−0.1
0
0.1
2.7 2.75 2.8 2.85 2.9 2.95 3 3.05 3.1
χ2/dof = 1.13
( r2) s M/t
0
1/(√t0mK)
β = 3.40
H102
H105C101
β = 3.46
N401
β = 3.55
N203
N200
D200
β = 3.70
N302
−0.03
−0.025
−0.02
−0.015
−0.01
−0.005
0.32 0.325 0.33 0.335 0.34 0.345 0.35 0.355 0.36 0.365 0.37
χ2/dof = 2.67
µs
√t0mK
β = 3.40
H102H105
C101
β = 3.46
N401
β = 3.55
N203
N200
D200
β = 3.70N302
(r2)sE (mK ) = a0 + a1 log(mK )
µs(mK ) = a2 + a3mK
(r2)sM(mK ) = a4 + a5/mK
1T. R. Hemmert et al., nucl-th/990407621 / 32
motivation simulation results outlook
error budget
(r2)sE [fm2] µs (r2)sM [fm2] χ2/DOFMain result -0.0039(13) -0.020(5) -0.010(5) 0.84, 2.67, 1.13Variations:
Doubling prior width -0.0049(18) -0.022(6) -0.015(9) 0.68, 1.83, 0.69Plateau method ∼ 1 fm -0.0040(9) -0.009(6) -0.0026(53) 1.12, 2.54, 1.58Including O(a2) -0.0030(16) -0.016(6) -0.007(6) 0.75, 2.98, 2.05No cut in Q2 -0.0034(9) -0.016(4) -0.007(4) 2.36, 2.77, 1.69
(r2)sE = −0.0039(13)(18) fm2
µs = −0.020(5)(12)
(r2)sM = −0.010(5)(11) fm2
22 / 32
motivation simulation results outlook
comparison
−0.01 −0.005 0 −0.1 −0.05 0 −0.04 −0.02 0
this work
LHPC
ETMC
χQCD
1505.01803
1801.09581
1705.05849
(r2)sE [fm2] µs (r2)sM [fm2]
(r2)sE = −0.0039(13)(18) fm2
µs = −0.020(5)(12)
(r2)sM = −0.010(5)(11) fm2
23 / 32
motivation simulation results outlook
strange axial vector form factors
mK ≈ 460 MeV
−0.055
−0.05
−0.045
−0.04
−0.035
−0.03
−0.025
−0.02
−0.015
−0.01
0 0.2 0.4 0.6 0.8 1 1.2
Gs A(Q
2)
Q2 [GeV2]
a = 0.08636 fma = 0.07634 fma = 0.06426 fm
−3.5
−3
−2.5
−2
−1.5
−1
−0.5
0
0 0.2 0.4 0.6 0.8 1 1.2
Gs P(Q
2)
Q2 [GeV2]
a = 0.08636 fma = 0.07634 fma = 0.06426 fm
β = 3.55
−0.07
−0.06
−0.05
−0.04
−0.03
−0.02
−0.01
0 0.2 0.4 0.6 0.8 1
Gs A(Q
2)
Q2 [GeV2]
mK = 482MeVmK = 462MeVmK = 441MeV
−3
−2.5
−2
−1.5
−1
−0.5
0
0.5
1
1.5
2
2.5
0 0.2 0.4 0.6 0.8 1
Gs P(Q
2)
Q2 [GeV2]
mK = 482MeVmK = 462MeVmK = 441MeV
24 / 32
motivation simulation results outlook
extrapolation
linear in m2π
−0.05
−0.045
−0.04
−0.035
−0.03
−0.025
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
χ2/dof = 0.43
gs A
t0m2π
β = 3.40
H102
H105
C101
β = 3.46
N401
β = 3.55
N203
N200
D200
β = 3.70
N302
−1
−0.8
−0.6
−0.4
−0.2
0
0.2
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
χ2/dof = 0.23( r2) s A
/t0
t0m2π
β = 3.40
H102H105
C101
β = 3.46
N401
β = 3.55
N203
N200
D200
β = 3.70
N302
25 / 32
motivation simulation results outlook
comparison
−0.075 −0.05 −0.025 0
this work
ETMC
PNDME
χQCD
JLQCD
1703.08788
1806.10604
1806.08366
1805.10507
gsA
g sA = −0.035(4)(8)
(r2)sA = −0.005(5)(10) fm2
variations: same as for strange vector form factors,but O(a) lattice artifacts
26 / 32
motivation simulation results outlook
strange tensor charge
−0.03
−0.02
−0.01
0
0.01
0.02
0.03
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
χ2/dof = 1.44
gs T
t0m2π
β = 3.40
H102
H105
C101
β = 3.46
N401
β = 3.55
N203
N200
D200 β = 3.70
N302
−0.05 −0.025 0 0.025 0.05
this work
ETMC
PNDME
JLQCD
1703.08788
1812.03573
1805.10507
gsT
g sT = −0.0037(79)(441)
variations: plateau method ∼ 1 fm, including O(a) latticeartifacts
27 / 32
motivation simulation results outlook
quark flavor contributions to gA and gT
guA/T =
1
2
(gu+d−2sA/T + 2g s
A/T + gu−dA/T
)gdA/T =
1
2
(gu+d−2sA/T + 2g s
A/T − gu−dA/T
)u + d − 2s renormalizes as u − d
additional noise cancellation for disconnected l − s1
gu−dA/T determined in Mainz isovector charges paper2
g sA/T already determined in this work
determine gu+d−2sA/T with two-state fit
gO(z0, y0) = gO + AO
(e−∆z0 + e−∆(y0−z0)
)+ BOe
−∆y0
gap ∆ determined as in Mainz isovector charges paper1
1V. Gulpers et al., PoS (LATTICE2014) 1282K. Ottnad et al., 1905.01291
28 / 32
motivation simulation results outlook
two-state fit
N302
0.6
0.65
0.7
0.75
0.8
0 0.5
y0/a = 20
0 0.5
y0/a = 22
0 0.5
y0/a = 24
0 0.5
y0/a = 26
(gu
+d−
2sA
)eff bar
e(z 0,y
0)
z0/y0 z0/y0 z0/y0 z0/y0
29 / 32
motivation simulation results outlook
extrapolation
linear in m2π
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
χ2/dof = 0.93
gu+d−2s
A
t0m2π
β = 3.40
H102
H105
C101
β = 3.46
N401
β = 3.55
N203
N200D200
β = 3.70 N302
0.45
0.5
0.55
0.6
0.65
0.7
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08
χ2/dof = 2.35
gu+d−2s
Tt0m
2π
β = 3.40
H102
H105C101
β = 3.46N401β = 3.55
N203
N200
D200
β = 3.70
N302
variations: summation method, including O(a) lattice artifacts
30 / 32
motivation simulation results outlook
comparison
0.6 0.8 1 1.2 −0.6 −0.35 −0.1
this work
ETMC
χQCD
PNDME
1706.02973
1806.08366
1806.10604
guA gdA
0.6 0.8 1 1.2 −0.3 −0.1 0.1
this work
JLQCD
ETMC
PNDME
1805.10507
1703.08788
1812.03573
guT gdT
guA = 0.84(3)(5)
gdA = −0.40(3)(5)
guT = 0.77(4)(5)
gdT = −0.19(4)(5)
31 / 32
motivation simulation results outlook
outlook
improve renormalization at β = 3.7
more configurations and ensembles
electric and magnetic charge radius and magnetic moment ofproton and neutron
(r2)p/n =1
2
[1
3(r2)u+d−2s ± (r2)u−d
]
µp/n =1
2
[1
3µu+d−2s ± µu−d
]
32 / 32