arXiv subjects
Publications and source records attributed to M. Göckeler.
Using established relations between the scattering matrix in infinite volume and the two-particle spectrum in a periodic box, we compute the mass and width of the rho meson from simulations of N_f=2 flavors of dynamical clover fermions at small pion masses 2 m_pi < m_rho.
High luminosity accelerators have greatly increased the interest in semi-exclusive and exclusive reactions involving nucleons. The relevant theoretical information is contained in the nucleon wavefunction and can be parametrized by moments of the nucleon distribution amplitudes, which in turn are linked to matrix elements of local three-quark operators. These can be calculated from first principles in lattice QCD. Defining an RI-MOM renormalization scheme, we renormalize three-quark operators corresponding to low moments non-perturbatively and take special care of the operator mixing. After performing a scheme matching and a conversion of the renormalization scale we quote our final results in the MSbar scheme at mu=2 GeV.
We consider the renormalisation of lattice QCD operators with one and two covariant derivatives related to the first and second moments of generalised parton distributions and meson distribution amplitudes. Employing the clover fermion action we calculate their non-forward quark matrix elements in one-loop lattice perturbation theory. For some representations of the hypercubic group commonly used in simulations we determine the sets of all possible mixing operators and compute the matrices of renormalisation factors in one-loop approximation. We describe how tadpole improvement is applied to the results.
We investigate the Operator Product Expansion (OPE) on the lattice by directly measuring the product (where J is the vector current) and comparing it with the expectation values of bilinear operators. This will determine the Wilson coefficients in the OPE from lattice data, and so give an alternative to the conventional methods of renormalising lattice structure function calculations. It could also give us access to higher twist quantities such as the longitudinal structure function F_L = F_2 - 2 x F_1. We use overlap fermions because of their improved chiral properties, which reduces the number of possible operator mixing coefficients.
Algorithmic progress in recent years made it possible to simulate QCD with Nf=2 flavours of O(a)-improved Wilson fermions at very light quark masses. We present the current results for baryon spectrum states, the nucleon axial coupling and the lowest moment of unpolarised nucleon structure functions. Special emphasis is given to a comparison of our calculations with results from chiral effective theories.
Results are presented for the lowest moment of the distribution amplitude for the K-star vector meson. Both longitudinal and transverse moments are investigated. We use two flavours of O(a) improved Wilson fermions, together with a non-perturbative renormalisation of the matrix element.
Semi-exclusive and exclusive processes are becoming more and more important in high energy physics since they are excellently suited to study the internal hadronic structure. To analyze such processes the knowledge of the hadron distribution amplitudes, which are universal for different reactions, is essential. Only rather indirect information on these nonperturbative functions can be obtained from measurements. In this work we report on a lattice QCD computation of moments of nucleon distribution amplitudes using suitable three-quark operators. However, these operators have to be renormalized and the mixing is even more complicated than in the continuum. Using the symmetry group of the hypercubic lattice we therefore derive and implement irreducibly transforming three-quark operators, which allow us to control the mixing pattern and will finally lead to quantitative predictions in the MSbar scheme. We present preliminary results for leading-twist and next-to-leading twist nucleon distribution amplitudes based on the QCDSF/UKQCD simulations with 2 flavors of dynamical clover fermions.
We investigate mesons containing charm quarks on fine lattices with a^{-1} \sim 5 GeV. The quenched approximation is employed using the Wilson gauge action at β= 6.6 and nonperturbatively O(a) improved Wilson quarks. We present results for decay constants using various interpolating fields and give preliminary results for form factors of semileptonic decays of D_s mesons to light pseudoscalar mesons.
We compute the electromagnetic form factor of the pion using non-perturbatively O(a) improved Wilson fermions. The calculations are done for a wide range of pion masses and lattice spacings. We check for finite size effects by repeating some of the measurements on smaller lattices. The large number of lattice parameters we use allows us to extrapolate to the physical point. For the square of the charge radius we find =0.441(20) fm^2, in good agreement with experiment.
We present the first calculation in lattice QCD of the lowest two moments of transverse spin densities of quarks in the nucleon. They encode correlations between quark spin and orbital angular momentum. Our dynamical simulations are based on two flavors of clover-improved Wilson fermions and Wilson gluons. We find significant contributions from certain quark helicity flip generalized parton distributions, leading to strongly distorted densities of transversely polarized quarks in the nucleon. In particular, based on our results and recent arguments by Burkardt [Phys. Rev. D 72 (2005) 094020], we predict that the Boer-Mulders-function $h_1^\perp$, describing correlations of transverse quark spin and intrinsic transverse momentum of quarks, is large and negative for both up and down quarks.
We present data for the axial coupling constant g_A of the nucleon obtained in lattice QCD with two degenerate flavours of dynamical non-perturbatively improved Wilson quarks. The renormalisation is also performed non-perturbatively. For the analysis we give a chiral extrapolation formula for g_A based on the small scale expansion scheme of chiral effective field theory for two degenerate quark flavours. Applying this formalism in a finite volume we derive a formula that allows us to extrapolate our data simultaneously to the infinite volume and to the chiral limit. Using the additional lattice data in finite volume we are able to determine the axial coupling of the nucleon in the chiral limit without imposing the known value at the physical point.
We present new results for light quark masses. The calculations are performed using two flavours of $O(a)$ improved Wilson fermions. We have reached lattice spacings as small as $a \sim 0.07$fm and pion masses down to $m_π \sim 340$MeV in our simulations. This gives us significantly better control on the chiral and continuum extrapolations.
We present first results for Wilson coefficients of operators up to first order in the covariant derivatives for the case of Wilson fermions. They are derived from the off-shell Compton scattering amplitude $\mathcal{W}_{μν}(a,p,q)$ of massless quarks with momentum $p$. The Wilson coefficients are classified according to the transformation of the corresponding operators under the hypercubic group H(4). We give selected examples for a special choice of the momentum transfer $q$. All Wilson coefficients are given in closed analytic form and in an expansion in powers of $a$ up to first corrections.
Due to improvements in computer performance and algorithms, the rapidly increasing cost for unquenched Wilson-type fermions with lighter quarks has been ameliorated and new simulations are now possible. Here we present results using two flavours of O(a)-improved Wilson fermions for meson decay constants at pseudoscalar masses down to 320MeV. Results are at several lattice spacings down to about 0.07fm and include a non-perturbative determination of the renormalisation constant. This enables us to attempt contact with (partially quenched) chiral perturbation theory.
We consider the renormalisation of composite quark-antiquark operators with one and two lattice covariant derivatives related to the lowest moments of generalised parton distributions (GPDs) and meson distribution amplitudes (DAs). Their matrix elements are calculated in one-loop lattice perturbation theory for non-zero momentum transfer. Using clover and overlap fermions we present the resulting matrices of mixing and renormalisation factors. For overlap fermions we explicitly check the absence of mixing with lower-dimensional operators of different chirality in particular representations of the hypercubic group. This feature favours the use of chiral fermions.
Based on lattice simulations with two flavours of dynamical, O(a)-improved Wilson fermions we present results for the first two moments of the distribution amplitudes of pseudoscalar mesons at several values of the valence quark masses. By extrapolating our results to the physical masses of up/down and strange quarks, we find the first two moments of the K^+ distribution amplitude and the second moment of the pi^+ distribution amplitude. We use nonperturbatively determined renormalisation coefficients to obtain results in the MSbar scheme. At a scale of 4 GeV^2 we find a_2^pi=0.201(114) for the second Gegenbauer moment of the pion's distribution amplitude, while for the kaon, a_1^K=0.0453(9)(29) and a_2^K=0.175(18)(47).
The QCDSF/UKQCD collaboration has an ongoing program to calculate nucleon matrix elements with two flavours of dynamical O(a) improved Wilson fermions. Here we present recent results on the electromagnetic form factors, the quark momentum fraction and the first three moments of the nucleon's spin-averaged and spin-dependent generalised parton distributions, including preliminary results with pion masses as low as 320 MeV.
Using the O(a) Symanzik improved action an estimate is given for the strange quark mass for unquenched (n_f=2) QCD. The determination is via the axial Ward identity (AWI) and includes a non-perturbative evaluation of the renormalisation constant. Numerical results have been obtained at several lattice spacings, enabling the continuum limit to be taken. Our results indicate a value for the strange quark mass (in the MS-bar scheme at a scale of 2 GeV) in the range 100 - 130 MeV. A comparison is also made with other recent lattice determinations of the strange quark mass using dynamical sea quarks.