arXiv subjects

Andrei Alexandru

Publications and source records attributed to Andrei Alexandru.

At least 109 records · Page 6Linked to original sources

arXiv2012

Chiral Polarization Scale at Finite Temperature

We study the chiral polarization properties of low-lying Dirac eigenmodes at finite temperature using the overlap operator. Results for pure gauge theory on both sides of deconfinement phase transition are presented. We find that the polarization scale decreases as we increase the temperature, but it remains non-zero as we cross in the deconfined phase and vanishes only when $T\approx 1.4 T_c$. This is caused by the presence of near-zero modes which, we find, are chirally polarized.

hep-lat↗

arXiv2012

Resonance parameters of the rho-meson from asymmetrical lattices

We present a lattice QCD calculation of the parameters of the $ρ$ meson decay. The study is carried out on spatially asymmetric boxes using nHYP-smeared clover fermions with two mass-degenerate quark flavors. Our calculations are carried out at a pion mass $m_π=304(2)$ MeV on the set of lattices $V=24^2\times η24\times48$ with $η=1.0,1.25$, and 2.0 with lattice spacing $a=0.1255(7)$ fm. The resonance mass $m_ρ=827(3)(5)$ MeV and coupling constant $g_{ρππ}=6.67(42)$ are calculate using the P-wave scattering phase shifts. We construct a 2$\times$2 correlation matrix to extract the energy of the scattering states and compute the phase shifts using the finite volume formula. By varying the degree of asymmetry, we are able to compute a set of phase shifts that are evenly distributed throughout the spectral region where the $ρ$ decays.

hep-lat↗

arXiv2011

Critical point of $N_f = 3$ QCD from lattice simulations in the canonical ensemble

A canonical ensemble algorithm is employed to study the phase diagram of $N_f = 3$ QCD using lattice simulations. We lock in the desired quark number sector using an exact Fourier transform of the fermion determinant. We scan the phase space below $T_c$ and look for an S-shape structure in the chemical potential, which signals the coexistence phase of a first order phase transition in finite volume. Applying Maxwell construction, we determine the boundaries of the coexistence phase at three temperatures and extrapolate them to locate the critical point. Using an improved gauge action and improved Wilson fermions on lattices with a spatial extent of $1.8 \fm$ and quark masses close to that of the strange, we find the critical point at $T_E = 0.925(5) T_c$ and baryon chemical potential $μ_B^E = 2.60(8) T_c$.

hep-ph↗

arXiv2011

Electric polarizability of hadrons with overlap fermions on multi-GPUs

Electric polarizability is an important parameter for the internal structure of hadrons. Previous studies of polarizabilities have been done at relatively heavy pion masses, leaving the chiral region largely unexplored. In this report, we use overlap fermions which are known to be computationally demanding to properly capture the chiral dynamics. We present an implementation strategy to construct overlap on multi-GPUs. We find that our GPU code has an equivalent of \sim30 CPU cores to 1 GPU. We also present preliminary results for the polarizability of the neutral pion.

hep-lat↗

arXiv2011

Spin Polarizabilities on the Lattice

Spin polarizabilities provide information on the internal structure of hadrons in the presence of weak external electromagnetic fields, and are actively studied by Compton scattering experiments. They provide finer detail than the regular polarizabilities since they require space and time-varying fields. Using an effective action in the weak field limit, we have identified methods to isolate each of the physical quantities ($μ, α, β, γ_{E1}, γ_{M1}, γ_{E2}, γ_{M2}$) for spin-1/2 hadrons, both neutral and charged. We also perform a lattice QCD simulation to investigate the feasibility of the effective action approach.

hep-lat↗

arXiv2011

Absolute X-distribution and self-duality

Various models of QCD vacuum predict that it is dominated by excitations that are predominantly self-dual or anti-self-dual. In this work we look at the tendency for self-duality in the case of pure-glue SU(3) gauge theory using the overlap-based definition of the field-strength tensor. To gauge this property, we use the absolute X-distribution method which is designed to quantify the dynamical tendency for polarization for arbitrary random variables that can be decomposed in a pair of orthogonal subspaces.

hep-lat↗

arXiv2011

$ρ$ meson decay on asymmetrical lattices

We present a lattice QCD calculation of the characteristics of the $ρ$ meson decay. The study is carried out on spatially asymmetric boxes using nHYP-smeared clover fermions in the quenched approximation. The resonance mass and coupling constant are calculate using the P-wave scattering phaseshifts, of the isospin I=1 two-pion system. We use pion masses m_π= 418 MeV and m_π=312 MeV. In both cases, the $ρ$ decay is kinematically feasible. We work on lattice sizes N_z X 24^2 X 48 with lattice spacing a=0.1 fm and N_z=24,30,34,48.

hep-lat↗

arXiv2011

How Self-Dual is QCD?

Vacuum characteristics quantifying dynamical tendency toward self-duality in gauge theories could be used to judge the relevance of classical solutions or the viability of classically motivated vacuum models. Here we decompose the field strength of equilibrium gauge configurations into self-dual and anti-self-dual parts, and apply absolute X-distribution method to the resulting polarization dynamics in order to construct such characteristics. Using lattice regularization and focusing on pure-glue SU(3) gauge theory at zero temperature, we find evidence for positive but very small dynamical tendency for self-duality of vacuum in the continuum limit.

hep-lat↗

arXiv2011

The Analysis of Space-Time Structure in QCD Vacuum II: Dynamics of Polarization and Absolute X-Distribution

We propose a framework for quantitative evaluation of dynamical tendency for polarization in arbitrary random variable that can be decomposed into a pair of orthogonal subspaces. The method uses measures based on comparisons of given dynamics to its counterpart with statistically independent components. The formalism of previously considered X-distributions is used to express the aforementioned comparisons, in effect putting the former approach on solid footing. Our analysis leads to definition of a suitable correlation coefficient with clear statistical meaning. We apply the method to the dynamics induced by pure-glue lattice QCD in local left-right components of overlap Dirac eigenmodes. It is found that, in finite physical volume, there exists a non-zero physical scale in the spectrum of eigenvalues such that eigenmodes at smaller (fixed) eigenvalues exhibit convex X-distribution (positive correlation), while at larger eigenvalues the distribution is concave (negative correlation). This chiral polarization scale thus separates a regime where dynamics enhances chirality relative to statistical independence from a regime where it suppresses it, and gives an objective definition to the notion of "low" and "high" Dirac eigenmode. We propose to investigate whether the polarization scale remains non-zero in the infinite volume limit, in which case it would represent a new kind of low energy scale in QCD.

hep-lat↗

arXiv2011

Efficient implementation of the overlap operator on multi-GPUs

Lattice QCD calculations were one of the first applications to show the potential of GPUs in the area of high performance computing. Our interest is to find ways to effectively use GPUs for lattice calculations using the overlap operator. The large memory footprint of these codes requires the use of multiple GPUs in parallel. In this paper we show the methods we used to implement this operator efficiently. We run our codes both on a GPU cluster and a CPU cluster with similar interconnects. We find that to match performance the CPU cluster requires 20-30 times more CPU cores than GPUs.

hep-lat↗

arXiv2010

Hadron electric polarizability -- finite volume corrections

We use the background field method to extract the polarizability for the neutral "pion". In our previous study we found that the polarizability for this system is negative which is believed to be a finite volume artifact. To address this issue, we carry out simulations for different lattice sizes and we also look at the influence of the boundary conditions on these results. We find that for pion masses lower than 400 MeV the polarizability remains negative even on larger lattices. An infinite volume extrapolation is attempted, but the results are not conclusive due mainly to a lack of an analytical form for the finite volume corrections for this system.

hep-lat↗

arXiv2010

Magnetic Moments of Negative-Parity Baryons from Lattice QCD

We report preliminary results for the magnetic moments of negative-parity baryons extracted from mass shifts in the presence of static external magnetic fields. The calculations are done on $24^3\times 48$ quenched lattices using standard Wilson actions, with $β$=6.0 and pion mass down to about 520 MeV, and 1000 configurations. Reasonable signals for the negative-parity states are observed and the sign of their magnetic moments is established. The results are compared to model calculations.

hep-lat↗

arXiv2010

Absolute Measure of Local Chirality and the Chiral Polarization Scale of the QCD Vacuum

The use of the absolute measure of local chirality is championed since it has a uniform distribution for randomly reshuffled chiral components so that any deviations from uniformity in the associated "X-distribution" are directly attributable to QCD-induced dynamics. We observe a transition in the qualitative behavior of this absolute X-distribution of low-lying eigenmodes which, we propose, defines a chiral polarization scale of the QCD vacuum.

hep-lat↗

arXiv2010

Finite density phase transition of QCD with $N_f=4$ and $N_f=2$ using canonical ensemble method

In a progress toward searching for the QCD critical point, we study the finite density phase transition of $N_f = 4$ and 2 lattice QCD at finite temperature with the canonical ensemble approach. We develop a winding number expansion method to accurately project out the particle number from the fermion determinant which greatly extends the applicable range of baryon number sectors to make the study feasible. Our lattice simulation was carried out with the clover fermions and improved gauge action. For a given temperature, we calculate the baryon chemical potential from the canonical approach to look for the mixed phase as a signal for the first order phase transition. In the case of $N_f=4$, we observe an "S-shape" structure in the chemical potential-density plane due to the surface tension of the mixed phase in a finite volume which is a signal for the first order phase transition. We use the Maxwell construction to determine the phase boundaries for three temperatures below $T_c$. The intersecting point of the two extrapolated boundaries turns out to be at the expected first order transition point at $T_c$ with $μ= 0$. This serves as a check for our method of identifying the critical point. We also studied the $N_f =2$ case, but do not see a signal of the mixed phase for temperature as low as 0.83 $T_c$.

hep-lat↗

arXiv2010

QCD at non-zero density and canonical partition functions with Wilson fermions

We present a reduction method for Wilson Dirac fermions with non-zero chemical potential which generates a dimensionally reduced fermion matrix. The size of the reduced fermion matrix is independent of the temporal lattice extent and the dependence on the chemical potential is factored out. As a consequence the reduced matrix allows a simple evaluation of the Wilson fermion determinant for any value of the chemical potential and hence the exact projection to the canonical partition functions.

hep-lat↗

arXiv2009

Neutron electric polarizability

We use the background field method to extract the "connected" piece of the neutron electric polarizability. We present results for quenched simulations using both clover and Wilson fermions and discuss our experience in extracting the mass shifts and the challenges we encountered when we lowered the quark mass. For the neutron we find that as the pion mass is lowered below $500\MeV$, the polarizability starts rising in agreement with predictions from chiral perturbation theory. For our lowest pion mass, $m_π=320\MeV$, we find that $α_n = 3.8(1.3)\times 10^{-4}\fm^3$, which is still only one third of the experimental value. We also present results for the neutral pion; we find that its polarizability turns negative for pion masses smaller than $500\MeV$ which is puzzling.

hep-lat↗

arXiv2009

Study of QCD critical point using canonical ensemble method

The existence of the QCD critical point at non-zero baryon density is not only of great interest for experimental physics but also a challenge for the theory. We use lattice simulations based on the canonical ensemble method to explore the finite baryon density region and look for the critical point. We scan the phase diagram of QCD with three degenerate quark flavors using clover fermions with $m_π\approx 700{MeV}$ on $6^3\times4$ lattices. We measure the baryon chemical potential as we increase the density and we see the characteristic "S-shape" that signals the first order phase transition. We determine the phase boundary by Maxwell construction and report our preliminary results for the location of critical point.

hep-lat↗

arXiv2008

Winding number expansion for the canonical approach to finite density simulations

The canonical partition function approach was designed to avoid the overlap problem that affects the lattice simulations of nuclear matter at high density. The method employs the projections of the quark determinant on a fix quark number sector. When the quark number is large, the evaluation of the projected determinant becomes numerically unstable. In this paper a different evaluation method based on expanding the determinant in terms of loops winding around the lattice is studied. We show that this method is stable and significantly faster than our original algorithm. This greatly expands the range of quark numbers that we can simulate effectively.

hep-lat↗