arXiv subjects
Publications and source records attributed to K. F. Liu.
We study the pion-baryon scattering process $π+B \to (n-1)π+B$ in a QCD theory with a large number (N_c) of colors. It is known that this scattering amplitudes decreases with N_c like N_c^{1-n/2}, and that its individual tree diagrams grow like N_c^{n/2}. The only way these two can be consistent is for n-1 powers of N_c to be cancelled when the Feynman diagrams are summed. We prove this to be true in tree order for any n.
We introduce a new method for estimating determinants or determinant ratios of large matrices, which combines the techniques of Padè approximation with rational functions and $Z_{2}$ noise estimation of traces of large matrices. The method requires simultaneously solving several matrix equations, which can be conveniently accomplished using the MR method. We include some preliinary results, and indicate potential applications to non-Hermitian matrices, and Hybrid Monte Carlo without pseudofermions.
We introduce a dynamical fermion algorithm which is based on the hybrid Monte Carlo (HMC) algorithm, but without pseudofermions. The molecular dynamics steps in HMC are retained except the derivatives with respect to the gauge fields are calculated with the $Z_2$ noise. The determinant ratios are estimated with the Pade - $Z_2$ method. Finally, we use the Kennedy-Kuti linear accept/reject method for the Monte Carlo step which is shown to respect detailed balance. We comment on the comparison of this algorithm with the pseudofermion algorithm.
In the presence of a large parameter, such as mass or energy, leading behavior of individual Feynman diagrams often get cancelled in the sum. This is known to happen in large-$N_c$ QCD in the presence of a baryon, and also in the case of high-energy electron-electron as well as quark-quark scatterings. We present an exact combinatorial formula, involving multiple commutators of the vertices, which can be used to compute such cancellations. It is a non-abelian generalization of the eikonal formula, and will be applied in subsequent publications to study the consistency of large-$N_c$ QCD involving baryons, as well as high-energy quark-quark scattering in ordinary QCD.
We report on a lattice QCD calculation of the $πN σ$ term, the scalar form factor, and $\langle N|\bar{s}s|N\rangle$. The disconnected insertion part of $\sg$ is found to be $1.8 \pm 0.1$ times larger than the connected insertion contribution. The $q^2$ dependence of $\sg(q^2)$ is about the same as $G_E(q^2)$ of the proton so that $\sg(2m_π^2) - \sg(0) = 6.6 \pm 0.6$ MeV. The ratio $y= ß/\ud = 0.36 \pm 0.03$. Both results favor a $\sg \sim 53$ MeV, slightly larger than our direct calculation of $\sg = 49.7 \pm 2.6$ MeV. We also compute $F_s$ and $D_s$ and find that the agreement with those from the octect baryon mass splittings crucially depends on the inclusion of the large disconnected insertion. Finally, we give our result for the $K N σ$ term.
We report on a lattice QCD calculation of the flavor-singlet axial coupling constant $g_A$ of the proton from the axial-vector current. The simulation is carried out at $β= 6$ on a quenched $16^3 \times 24$ lattice. An extrapolation to the chiral limit shows that the connected insertion (valence and cloud parts) is $0.62 \pm 0.09$, which is close to $g_A^8$. The disconnected insertion (vacuum polarization from the sea quark) for the $u$ or $d$ quark is $- 0.12\pm 0.01 $ and the $s$ quark also contributes $-0.12\pm 0.01$. The total $g_A^1$ is thus $0.25 \pm 0.12$ which is in agreement with experiments. In addition, we find $g_A^3 = 1.20 \pm 0.11$, $g_A^8 = 0.61 \pm 0.13$ and $F_A/D_A = 0.60 \pm 0.02$, also in agreement with experiments.
The $πNN$ form factor $g_{πNN}(q^2)$ is obtained from a quenched lattice QCD calculation of the pseudoscalar form factor $g_P(q^2)$ of the proton with pion pole dominance. We find that $g_{πNN}(q^2)$ fitted with the monopole form agrees well with the Goldberger-Treiman relation and is much preferred over the dipole form. The monopole mass is determined to be $0.75 \pm 0.14 {\rm GeV}$ which shows that $g_{πNN}(q^2)$ is rather soft. The extrapolated $πN$ coupling constant $g_{πNN} = 12.7 \pm 2.4$ is quite consistent with the phenomenological values. We also compare $g_{πNN}(q^2)$ with the axial form factor $g_A(q^2)$ to check the pion dominance in the induced pseudoscalar form factor $h_A(q^2)$ vis à vis chiral Ward identity.
We introduce a $Z_2$ noise for the stochastic estimation of matrix inversion and discuss its superiority over other noises including the Gaussian noise. This algorithm is applied to the calculation of quark loops in lattice quantum chromodynamics that involves diagonal and off-diagonal traces of the inverse matrix. We will point out its usefulness in its applications to estimating determinants, eigenvalues, and eigenvectors, as well as its limitations based on the structure of the inverse matrix.
Results for the isovector axial form factors of the proton from a lattice QCD calculation are presented for both point-split and local currents. They are obtained on a quenched $16^{3} \times 24$ lattice at $β= 6.0$ with Wilson fermions for a range of quark masses from strange to charm. We determine the finite lattice renormalization for both the local and point-split currents of heavy quarks. Results extrapolated to the chiral limit show that the $q^2$ dependence of the axial form factor agrees reasonably well with experiment. The axial coupling constant $g_A$ calculated for the local and the point-split currents is about 6\% and 12\% smaller than the experimental value respectively.
Matrix elements of the form $ <0| Tr \; g^{2}GG |G> $ are calculated using the lattice QCD Monte Carlo method. Here, $|G>$ is a glueball state with quantum numbers $ 0^{++}$, $ 2^{++}$, $ 0^{-+}$ and $G$ is the gluon field strength operator. The matrix elements are obtained from the hybrid correlation functions of the fuzzy and plaquette operators performed on the $12^{4}$ and $14^{4}$ lattices at $β= 5.9 $ and $5.96$ respectively. These matrix elements are compared with those from the QCD sum rules and the tensor meson dominance model. They are the non-perturbative matrix elements needed in the calculation of the partial widths of $J/Ψ$ radiative decays into glueballs.