Search

Topological insulators are substances which are bulk insulators but which carry current via special "topologically protected" edge states. The understanding of long range topological order in these systems is built around the idea of a Berry connection, which is a gauge connection obtained from the phase of the electron wave function transported over momentum space rather than coordinate space. The phase of a closed Wilson loop of the Berry connection around the Brillouin zone defines a topological order parameter which labels discrete flux vacua. The conducting states are surface modes on the domain walls between discrete vacua. Evidence from large-$N_c$ chiral dynamics, holographic QCD, and Monte Carlo observations has pointed to a picture of the QCD vacuum that is very similar to that of a topological insulator, with discrete quasivacua labelled by $\theta$ angles that differ by mod $2\pi$. In this picture, the domain walls are membranes of Chern-Simons charge, and the quark condensate consists of surface modes on these membranes, which are delocalized and thus support the long range propagation of Goldstone pions. The Berry phase in QED2 describes charge polarization of fermion-antifermion pairs, while in 4D QCD it describes the polarization of Chern-Simons membranes., Comment: 7 pages, no figures, talk presented at Lattice 2014

The dynamical effects of topological charge in two-dimensional QED can be expressed in terms of a topological order parameter via a Berry phase construction. The Berry phase describes the electric charge polarization of the vacuum in a manner similar to the theory of polarization in topological insulators. The topological order parameter labels discrete vacua which differ by units of electric flux. Here the associated Berry connection is explicitly constructed from the Dirac Hamiltonian eigenstates by introducing a small attractive Thirring coupling, so that there is still a stable boson in the limit of zero EM coupling. The Berry connection arises from the analytic structure of the Bethe ansatz states in complex rapidity near the free fermion point., Comment: 16 pages, 3 figures

Both theoretical arguments and Monte Carlo observations indicate that the topological structure of the QCD vacuum consists of a laminated array of extended, coherent codimension-one membranes of alternating sign. Large-$N_c$ arguments, supported by gauge/string holography, indicate that these membranes are domain walls which separate discrete "flux vacua" with values of the topological $\theta$ parameter which differ by $\pm 2\pi$. This exposes a close analogy with 2D U(1) gauge theory, where $\theta$ can be interpreted as electric polarization, and the domain walls are pointlike charged particles. In 4D QCD, the $\theta$ parameter represents background Ramond-Ramond flux, which can be interpreted as a polarization of the charged membranes in the vacuum. In this framework, the chiral condensate is formed from the quark surface modes on the membranes. Massless Goldstone boson propagation takes place due to a coordination between bulk oscillations of the polarization field $\theta$ and the surface currents represented by the Chern-Simons 3-form on the brane surface. This coordination is enforced by overall gauge invariance which imposes an anomaly inflow constraint between bulk and surface currents., Comment: 6 pages, Talk presented at Chiral Dynamics Workshop, Jefferson Lab, Aug. 2012

Large $N_c$ and holographic arguments, as well as Monte Carlo results, suggest that the topological structure of the QCD vacuum is dominated by codimension-one membranes which appear as thin dipole layers of topological charge. Such membranes arise naturally as $D6$ branes in the holographic formulation of QCD based on IIA string theory. The polarizability of these membranes leads to a vacuum energy $\propto \theta^2$, providing the origin of nonzero topological susceptibility. Here we show that the axial U(1) anomaly can be formulated as anomaly inflow on the brane surfaces. A 4D gauge transformation at the brane surface separates into a 3D gauge transformation of components within the brane and the transformation of the transverse component. The in-brane gauge transformation induces currents of an effective Chern-Simons theory on the brane surface, while the transformation of the transverse component describes the transverse motion of the brane and is related to the Ramond-Ramond closed string field in the holographic formulation of QCD. The relation between the surface currents and the transverse motion of the brane is dictated by the descent equations of Yang-Mills theory., Comment: 22 pages, 3 figures

Topological excitations in the QCD vacuum take the form of coherent codimension one sheets with positive and negative sheets juxtaposed in a dipole layer. These sheets may be interpreted as Luscher's "Wilson bags," which are domain walls between discrete quasivacua labelled by a local value of the $\theta$ parameter equal to $2\pi k$, with $k$ given by the number of units of background Ramond-Ramond flux. This picture of the vacuum is closely analogous to Coleman's description of the 2D massive Schwinger model, where $\theta$ is interpreted as a background electric field, and a pointlike charged particle is a domain wall between vacua which differ by one unit of background electric flux. The main effect of the Ramond-Ramond U(1) field in low energy QCD dynamics is the generation of phenomenologically important contact terms: (1) The dominant contact term in the topological charge correlator which leads to positive topological susceptibility, (2) The $\eta'$ mass insertion, and (3) An $SU(N_f)\times SU(N_f)$ invariant Nambu-Jona Lasinio 4-quark interaction., Comment: 7 pages, 1 figure, talk presented at Lattice 2011, Lake Tahoe, CA

The Witten-Sakai-Sugimoto construction of holographic QCD in terms of D4 color branes and D8 flavor branes in type IIA string theory is used to investigate the role of topological charge in the chiral dynamics of quarks in QCD. The QCD theta term arises from a compactified 5-dimensional Chern-Simons term on the D4 branes. This term couples the QCD topological charge to the Ramond-Ramond $U(1)$ gauge field of IIA string theory. The nonzero topological susceptibility of pure-glue QCD can be attributed to the presence of D6 branes, which constitute magnetic sources of the RR gauge field. The topological charge of QCD is required, by an anomaly inflow argument, to coincide in space-time with the intersection of the D6 branes and the D4 color branes. This clarifies the relation between D6 branes and the coherent, codimension-one topological charge membranes observed in QCD Monte Carlo calculations. Using open-string/closed-string duality, we interpret a quark loop (represented by a D4-D8 open string loop) in terms of closed-string exchange between color and flavor branes. The role of the RR gauge field in quark-antiquark annihilation processes is discussed. RR exchange in the s-channel generates a 4-quark contact term which produces an $\eta'$ mass insertion and provides an explanation for the observed spin-parity structure of the OZI rule. The $(\log {\rm Det\;U})^2$ form of the $U(1)$ anomaly emerges naturally. RR exchange in the t-channel of the $q\overline{q}$ scattering amplitude produces a Nambu-Jona Lasinio interaction which may provide a mechanism for spontaneous breaking of $SU(N_f)\times SU(N_f)$., Comment: 20 pages, 7 figures

Monte Carlo studies of $CP^{N-1}$ sigma models have shown that the structure of topological charge in these models undergoes a sharp transition at $N=N_c\approx 4$. For $NN_c$ it is dominated by extended, thin, 1-dimensionally coherent membranes of topological charge, which can be interpreted as domain walls between discrete quasi-stable vacua. These vacua differ by a unit of background electric flux. The transition can be identified as the delocalization of topological charge, or "instanton melting," a phenomenon first suggested by Witten to resolve the conflict between instantons and large $N$ behavior. Implications for $QCD$ are discussed., Comment: 7 pages, 3 figures. Talk presented at Lattice 2008, Williamsburg, VA

The behavior of Wilson loops with fractional charge is used to study the $\theta$-dependence of the free energy density $\epsilon(\theta)$ for the $CP^1$, $CP^5$, and $CP^9$ sigma models in two spacetime dimensions. The function $\epsilon(\theta)$ is extracted from the area law for a Wilson loop of charge $q=\theta/2\pi$. For $CP^1$, $\epsilon(\theta)$ is smooth in the region $\theta\approx\pi$ and well-described by a dilute instanton gas throughout the range $0<\theta<2\pi$. For $CP^5$ and $CP^9$ the energy exhibits a clear cusp and evidence for discrete, degenerate vacua at $\theta = \pi$, as expected from large $N$ arguments. For $CP^9$ the $\theta$-dependence is in good quantitative agreement with the leading order large $N$ prediction $\epsilon(\theta)={1/2}\chi_t\theta^2$ throughout the range $0<\theta<\pi$., Comment: 16 pages, 10 figures

Monte Carlo studies of pure glue SU(3) gauge theory using the overlap-based topological charge operator have revealed a laminar structure in the QCD vacuum consisting of extended, thin, coherent, locally 3-dimensional sheets of topological charge embedded in 4D space, with opposite sign sheets interleaved. Studies of localization properties of Dirac eigenmodes have also shown evidence for the delocalization of low-lying modes on effectively 3-dimensional surfaces. In this talk, I review some theoretical ideas which suggest the possibility of 3-dimensionally coherent topological charge structure in 4-dimensional gauge theory and provide a possible interpretation of the observed structure. I begin with Luscher's ``Wilson bag'' integral over the 3-index Chern-Simons tensor. The analogy with a Wilson loop as a charged world line in 2-dimensional $CP^{N-1}$ sigma models suggests that the Wilson bag surface represents the world volume of a physical membrane. The large-N chiral Lagrangian arguments of Witten also indicate the existence of multiple ``k-vacuum'' states with discontinuous transitions between k-vacua at $\theta=$ odd multiples of $\pi$. The domain walls between these vacua have the properties of a Wilson bag surface. Finally, I review the AdS/CFT duality view of $\theta$ dependence in QCD. The dual realtionship between topological charge in gauge theory and Ramond-Ramond charge in type IIA string theory suggests that the coherent topological charge sheets observed on the lattice are the holographic image of wrapped D6 branes., Comment: 15 pages, 3 figures. Talk presented at Lattice 2006, Tucson, AZ, July 23-27, 2006

Recent work by Armoni, Shifman, and Veneziano suggests a large-N equivalence between supersymmetric Yang-Mills Theory and one-flavor QCD. One consequence of this "orientifold projection" is that scalar and pseudoscalar mesons in one-flavor QCD should have degenerate mass since they lie within the same Wess-Zumino supermultiplet. We use lattice calculations to investigate the mass shifts caused by "double-hairpin" annihilation diagrams in quenched QCD to test for this degeneracy. Similar quark-antiquark annihilation processes are studied in the 2-dimensional CP(N-1) model with quenched fermions., Comment: 7 pages, 2 figures (2 figure files), PoS style, Lattice 2006 Poster (Beyond QCD)

The anomalous scaling behavior of the topological susceptibility $\chi_t$ in two-dimensional $CP^{N-1}$ sigma models for $N\leq 3$ is studied using the overlap Dirac operator construction of the lattice topological charge density. The divergence of $\chi_t$ in these models is traced to the presence of small instantons with a radius of order $a$ (= lattice spacing), which are directly observed on the lattice. The observation of these small instantons provides detailed confirmation of L\"{u}scher's argument that such short-distance excitations, with quantized topological charge, should be the dominant topological fluctuations in $CP^1$ and $CP^2$, leading to a divergent topological susceptibility in the continuum limit. For the $\CP$ models with $N>3$ the topological susceptibility is observed to scale properly with the mass gap. These larger $N$ models are not dominated by instantons, but rather by coherent, one-dimensional regions of topological charge which can be interpreted as domain wall or Wilson line excitations and are analogous to D-brane or ``Wilson bag'' excitations in QCD. In Lorentz gauge, the small instantons and Wilson line excitations can be described, respectively, in terms of poles and cuts of an analytic gauge potential., Comment: 33 pages, 12 figures

In an effort to clarify the significance of the recent observation of long-range topological charge coherence in QCD gauge configurations, we study the local topological charge distributions in two-dimensional $CP^{N-1}$ sigma models, using the overlap Dirac operator to construct the lattice topological charge. We find long-range sign coherence of topological charge along extended one-dimensional structures in two-dimensional spacetime. We discuss the connection between the long range topological structure found in $CP^{N-1}$ and the observed sign coherence along three-dimensional sheets in four-dimensional QCD gauge configurations. In both cases, coherent regions of topological charge form along membrane-like surfaces of codimension one. We show that the Monte Carlo results, for both two-dimensional and four-dimensional gauge theory, support a view of topological charge fluctuations suggested by Luscher and Witten. In this framework, the observed membranes are associated with boundaries between ``k-vacua,'' characterized by an effective local value of $\theta$ which jumps by $\pm 2\pi$ across the boundary., Comment: 26 pages

The recently observed long-range coherent structure of topological charge fluctuations in QCD is compared with theoretical expectations based on the AdS/CFT brane construction of nonsupersymmetric gauge theory by Witten. Similar observations of coherent topologicalcharge structure in 2D $CP^{N-1}$ sigma models are interpreted in terms of Wilson lines representing world lines of screened electric charges. The analogy between 2D U(1) and 4D Yang-Mills theory leads to the interpretation of the observed coherent sheets of topological charge in QCD as screened ``Wilson bags'' first suggested by Luscher. The duality between the Wilson bag surface and a wrapped 6-brane in IIA string theory is discussed. The complete screening of the force between bag surfaces for integer values of the bag charge $\theta/2\pi$ corresponds to the observation by Polchinski that the net force between d-branes from closed string exchange vanishes for quantized values of Ramond-Ramond charge., Comment: Talk presented at Lattice 2005

We calculate the lattice two-point function of topological charge density in pure-glue QCD using the discretization of the operator based on the overlap Dirac matrix. Utilizing data at three lattice spacings it is shown that the continuum limit of the correlator complies with the requirement of non-positivity at non-zero distances. For our choice of the overlap operator and the Iwasaki gauge action we find that the size of the positive core is ~2a (with a being the lattice spacing) sufficiently close to the continuum limit. This result confirms that the overlap-based topological charge density is a valid local operator over realistic backgrounds contributing to the QCD path integral, and is important for the consistency of recent results indicating the existence of a low-dimensional global brane-like topological structure in the QCD vacuum. We also confirm the divergent short-distance behavior of the correlator, and the non-integrable nature of the associated contact part., Comment: 13 pages, 5 figures

We have recently presented evidence that in configurations dominating the regularized pure-glue QCD path integral, the topological charge density constructed from overlap Dirac operator organizes into an ordered space-time structure. It was pointed out that, among other properties, this structure exhibits two important features: it is low-dimensional and geometrically global, i.e. consisting of connected sign-coherent regions with local dimensions 1<= d < 4, and spreading over arbitrarily large space--time distances. Here we show that the space-time structure that is responsible for the origin of topological susceptibility indeed exhibits global behavior. In particular, we show numerically that topological fluctuations are not saturated by localized concentrations of most intense topological charge density. To the contrary, the susceptibility saturates only after the space-time regions with most intense fields are included, such that geometrically global structure is already formed. We demonstrate this result both at the fundamental level (full topological density) and at low energy (effective density). The drastic mismatch between the point of fluctuation saturation (~ 50% of space-time at low energy) and that of global structure formation (<4% of space-time at low energy) indicates that the ordered space-time structure in topological charge is inherently global and that topological charge fluctuations in QCD cannot be understood in terms of individual localized pieces. Description in terms of global brane-like objects should be sought instead., Comment: 10 pages, 3 figures; v2: typos corrected, minor modifications; v3: misprint in Eqs. (2,3) fixed

Local topological charge structure in the 2D CP(N-1) sigma models is studied using the overlap Dirac operator. Long-range coherence of topological charge along locally 1D regions in 2D space-time is observed. We discuss the connection between these results and the recent discovery of coherent 3D sheets of topological charge in 4D QCD. In both cases, coherent regions of topological charge form along surfaces of approximmate codimension 1., Comment: Lattice2004(topology)

In a recent investigation of chiral behavior in quenched lattice QCD, the flavor-singlet pseudoscalar ``hairpin'' vertex associated with the eta prime mass was studied for pion masses ranging from approximately 275 to 675 MeV. Throughout this mass range, the quark-disconnected pseudoscalar correlator is well-described by a pure double-pion-pole diagram with a p^2 independent mass insertion. The residue of the double pole was found to exhibit a significant quark mass dependence, evidenced by a negative slope of the effective mass insertion (m_0^{eff})^2 as a function of m_{\pi}^2. It has been observed by Sharpe that, with a consistent NLO calculation in quenched chiral perturbation theory, this mass dependence is uniquely predicted in terms of the single-pole coefficient \alpha_{\Phi} and the Leutwyler parameter L_5. Since \alpha_{\Phi} is found to be approximately zero, the chiral slope of the double-pole residue determines a value for L_5. This provides a consistency check between the chiral slope of the hairpin mass insertion and that of the pion decay constant. We investigate the consistency of these mass dependences in our Monte Carlo results at two values of lattice spacing. Within statistics, the slopes are found to be consistent with the Q\chiPT prediction, confirming that the observed negative slope of m_0^{eff} arises as an effect of the L_5 Leutwyler term., Comment: 4 pages, 1 figure, LaTeX

Lattice topological charge associated with Ginsparg-Wilson fermions exhibits generic topological stability over quantum ensemble of configurations contributing to the QCD path integral. Moreover, the underlying chiral symmetry leads to the suppression of ultraviolet noise in the associated topological charge densities ("chiral smoothing"). This provides a solid foundation for the direct study of the role of topological charge fluctuations in the physics of QCD vacuum. Using these tools it was recently demonstrated that: (a) there is a well-defined space-time structure (order) in topological charge density (defined through overlap fermions) for typical configurations contributing to QCD path integral; (b) this fundamental structure is low-dimensional, exhibiting sign-coherent behavior on subsets of dimension less than four and not less than one; (c) the structure has a long-range global character (spreading over maximal space-time distances) and is built around the locally one-dimensional network of strong fields (skeleton). In this talk we elaborate on certain aspects and implications of these results., Comment: 3 pages, 1 figure; Lattice2003(topology)

The results of a high-statistics study of scalar and pseudoscalar meson propagators in quenched lattice QCD are presented. For two values of lattice spacing, $\beta=5.7$ ($a \approx .18$ fm) and 5.9 ($a \approx .12$ fm), we probe the light quark mass region using clover improved Wilson fermions with the MQA pole-shifting ansatz to treat the exceptional configuration problem. The quenched chiral loop parameters $m_0$ and $\alpha_{\Phi}$ are determined from a study of the pseudoscalar hairpin correlator. From a global fit to the meson correlators, estimates are obtained for the relevant chiral Lagrangian parameters, including the Leutwyler parameters $L_5$ and $L_8$. Using the parameters obtained from the singlet and nonsinglet pseudoscalar correlators, the quenched chiral loop effect in the nonsinglet scalar meson correlator is studied. By removing this QCL effect from the lattice correlator, we obtain the mass and decay constant of the ground state scalar, isovector meson $a_0$., Comment: 36 pages, 12 figures, LaTeX

While sign-coherent 4-dimensional structures cannot dominate topological charge fluctuations in the QCD vacuum at all scales due to reflection positivity, it is possible that enhanced coherence exists over extended space-time regions of lower dimension. Using the overlap Dirac operator to calculate topological charge density, we present evidence for such structure in pure-glue SU(3) lattice gauge theory. It is found that a typical equilibrium configuration is dominated by two oppositely-charged sign-coherent connected structures (``sheets'') covering about 80% of space-time. Each sheet is built from elementary 3-d cubes connected through 2-d faces, and approximates a low-dimensional curved manifold (or possibly a fractal structure) embedded in the 4-d space. At the heart of the sheet is a ``skeleton'' formed by about 18% of the most intense space-time points organized into a global long-range structure, involving connected parts spreading over maximal possible distances. We find that the skeleton is locally 1-dimensional and propose that its geometrical properties might be relevant for understanding the possible role of topological charge fluctuations in the physics of chiral symmetry breaking., Comment: 4 pages RevTeX, 4 figures; v2: 6 pages, 5 figures, more explanations provided, figure and references added, published version

Recently, we have pointed out that sign-coherent 4-dimensional structures can not dominate topological charge fluctuations in QCD vacuum at all scales. Here we show that an enhanced lower-dimensional coherence is possible. In pure SU(3) lattice gauge theory we find that in a typical equilibrium configuration about 80% of space-time points are covered by two oppositely-charged connected structures built of elementary 3-dimensional coherent hypercubes. The hypercubes within the structure are connected through 2-dimensional common faces. We suggest that this coherence is a manifestation of a low-dimensional order present in the QCD vacuum. The use of a topological charge density associated with Ginsparg-Wilson fermions ("chiral smoothing") is crucial for observing this structure., Comment: 3 pages, 1 figure; Proceedings of the "Confinement V" Conference, Gargnano, Italy, Sep 10-14, 2002

We present results of a high-statistics study of scalar and pseudoscalar meson propagators in quenched QCD at two values of lattice spacing, beta = 5.7 and 5.9, with clover-improved Wilson fermions. The study of the chiral limit is facilitated by the pole-shifting ansatz of the modified quenched approximation. Pseudoscalar masses and decay constants are determined as a function of quark mass and quenched chiral log effects are estimated. A study of the flavor singlet eta prime hairpin diagram yields a precise determination of the eta prime mass insertion. The corresponding value of the quenched chiral log parameter delta is compared with the observed QCL effects. Removal of QCL effects from the scalar propagator allows a determination of the mass of the lowest lying isovector scalar $q\bar{q}$ meson., Comment: Lattice2002(spectrum); 3pgs(Latex), 8figs.(ps)

The structure of topological charge fluctuations in the QCD vacuum is strongly restricted by the spectral negativity of the Euclidean 2-point correlator for $x\neq 0$ and the presence of a positive contact term. Some examples are considered which illustrate the physical origin of these properties., Comment: Lattice 2002 Conference Proceedings

We introduce the Dirac eigenmode filtering of topological charge density associated with Ginsparg-Wilson fermions as a tool to investigate the local structure of topological charge fluctuations in QCD. The resulting framework is used to demonstrate that the bulk of topological charge in QCD does not appear in the form of unit quantized lumps. This means that the mixing of "would-be" zeromodes associated with such lumps is probably not the prevalent microscopic mechanism for spontaneous chiral symmetry breaking in QCD. To characterize the coherent local behavior in topological charge density at low energy, we compute the charges contained in maximal coherent spheres enclosing non-overlapping peaks. We find a continuous distribution essentially ending at ~0.5. Finally, we study, for the first time, the overlap-operator topological-charge-density correlators and find consistency with non-positivity at nonzero physical distance. This represents a non-trivial check on the locality (in gauge paths) of the overlap Dirac operator for realistic gauge backgrounds., Comment: 3 pages, 4 figures, talk, Lattice2002(topology)

The local structure of low-lying eigenmodes of the overlap Dirac operator is studied. It is found that these modes cannot be described as linear combinations of 't Hooft "would-be" zeromodes associated with instanton excitations that underly the Instanton Liquid Model. This implies that the instanton liquid scenario for spontaneous chiral symmetry breaking in QCD is not accurate. More generally, our data suggests that the vacuum fluctuations of topological charge are not effectively dominated by localized lumps of unit charge with which the topological "would-be" zeromodes could be associated., Comment: Presented by I. Horvath at the NATO Advanced Research Workshop "Confinement, Topology, and other Non-Perturbative Aspects of QCD", January 21-27, 2002, Stara Lesna, Slovakia. 12 pages, 6 figures, uses crckapb.sty

We consider the lattice topological charge density introduced by Hasenfratz, Laliena and Niedermayer and propose its eigenmode expansion as a tool to investigate the structure of topological charge fluctuations in QCD. The resulting effective density is built from local chiralities studied previously. At every order of the expansion the density exactly sums up to the global topological charge, and the leading term describes the maximally smooth space-time distribution of charge relevant for propagating light fermions. We use this framework to demonstrate our previous suggestion that the bulk of topological charge in QCD does not effectively appear in the form of quantized unit lumps. Our conclusion implies that it is unlikely that the mixing of "would-be" zeromodes associated with such lumps is the prevalent microscopic mechanism for spontaneous chiral symmetry breaking in QCD. We also present first results quantitatively characterizing the space-time behavior of effective densities. For coherent fluctuations contained in spherical regions we find a continuous distribution of associated charges essentially ending at ~0.5., Comment: 5 pages, 4 figures, RevTex; v2: typo in eq. (3) fixed; v3: References added, minor cosmetic changes, unchanged conclusions, published version

The reasons for using low-lying Dirac eigenmodes to probe the local structure of topological charge fluctuations in QCD are discussed, and it is pointed out that the qualitative double-peaked behavior of the local chiral orientation probability distribution in these modes is necessary, but not sufficient for dominance of instanton-like fluctuations. The results with overlap Dirac operator in Wilson gauge backgrounds at lattice spacings ranging from a~0.04 fm to a~0.12 fm are reported, and it is found that the size and density of local structures responsible for double-peaking of the distribution are in disagreement with the assumptions of the Instanton Liquid Model. More generally, our results suggest that vacuum fluctuations of topological charge are not effectively dominated by locally quantized (integer-valued) lumps in QCD., Comment: 29 pages, 13 figures; v2: minor improvements in presentation, results and conclusions unchanged, version to appear in PRD

A truncated determinant algorithm is used to study the physical effects of the quark eigenmodes associated with eigenvalues below 420 MeV. This initial high statistics study focuses on coarse ($6^4$) lattices (with O($a^2$) improved gauge action), light internal quark masses and large physical volumes. Three features of full QCD are examined: topological charge distributions, string breaking as observed in the static energy and the eta prime mass., Comment: Lattice2001(confinement); 3pgs(Latex), 4figs.(ps)

The scalar isovector meson propagator is studied in quenched QCD. For the lightest quark masses used, this propagator is dominated by a quenched chiral loop effect associated with the $\eta'$-$\pi$ two-meson intermediate state. Both the time dependence and the pion mass dependence of the effect are well-described by quenched chiral perturbation theory., Comment: 3 pages, 3 figures, Lat2001 Proceedings

We discuss the utility of low-lying Dirac eigenmodes for studying the nature of topological charge fluctuations in QCD. The implications of previous results using the local chirality histogram method are discussed, and the new results using the overlap Dirac operator in Wilson gauge backgrounds at lattice spacings ranging from a~0.04 fm to a~0.12 fm are reported. While the degree of local chirality does not change appreciably closer to the continuum limit, we find that the size and density of local structures responsible for chiral peaking do change significantly. The resulting values are in disagreement with the assumptions of the Instanton Liquid Model. We conclude that the fluctuations of topological charge in the QCD vacuum are not locally quantized., Comment: 3 pages, 4 figures, Lattice2001(confinement)

The scalar, isovector meson propagator is analyzed in quenched QCD, using the MQA pole-shifting ansatz to study the chiral limit. In addition to the expected short-range exponential falloff characteristic of a heavy scalar meson, the propagator also exhibits a longer-range, negative metric contribution which becomes pronounced for smaller quark masses. We show that this is a quenched chiral loop effect associated with the anomalous structure of the $\eta '$ propagator in quenched QCD. Both the time dependence and the quark mass dependence of this effect are well-described by a chiral loop diagram corresponding to an $\eta '- \pi$ intermediate state, which is light and effectively of negative norm in the quenched approximation. The relevant parameters of the effective Lagrangian describing the scalar sector of the quenched theory are determined., Comment: 29 pages, 10 figures, Latex

The low-lying eigenmodes of the Dirac operator associated with typical gauge field configurations in QCD encode, among other low-energy properties, the physics behind the solution to the $U_A(1)$ problem (i.e. the origin of the $\eta'$ mass), the nature of spontaneous chiral symmetry breaking, and the physics of string-breaking, quark-antiquark pair production, and the OZI rule. Moreover, the space-time chiral structure of these eigenmodes reflects the space-time topological structure of the underlying gauge field. We present evidence from lattice QCD on the local chiral structure of low Dirac eigenmodes leading to the conclusion that topological charge fluctuations of the QCD vacuum are not instanton-dominated. The result supports Witten's arguments that topological charge is produced by confinement-related gauge fluctuations rather than instantons., Comment: 35 pages, 11 figures

Virtual quark pair screening leads to breaking of the string between fundamental representation quarks in QCD. For unquenched four dimensional lattice QCD, this (so far elusive) phenomenon is studied using the recently developed truncated determinant algorithm (TDA). The dynamical configurations were generated on an Athlon 650 MHz PC. Quark eigenmodes up to 420 MeV are included exactly in these TDA studies performed at low quark mass on large coarse (but O($a^2$) improved) lattices. A study of Wilson line correlators in Coulomb gauge extracted from an ensemble of 1000 two-flavor dynamical configurations reveals evidence for flattening of the string tension at distances R $\geq$ approximately 1 fm., Comment: 16 pages, 5 figures, Latex (deleted extraneous eps figure file)

Several lattice calculations which probe the chiral and topological structure of QCD are discussed. The results focus attention on the low-lying eigenmodes of the Dirac operator in typical gauge field configurations., Comment: Talk presented at the DPF2000 Conference

A study of the chiral behavior of pseudoscalar masses and decay constants is carried out in quenched lattice QCD with Wilson fermions. Using the modified quenched approximation (MQA) to cure the exceptional configuration problem, accurate results are obtained for pion masses as low as $\approx$ 200 MeV. The anomalous chiral log effect associated with quenched $\eta'$ loops is studied in both the relation between $m_{\pi}^2$ vs. $m_q$ and in the light-mass behavior of the pseudoscalar and axial vector matrix elements. The size of these effects agrees quantitatively with a direct measurement of the $\eta'$ hairpin graph, as well as with a measurement of the topological susceptibility, thus providing several independent and quantitatively consistent determinations of the quenched chiral log parameter $\delta$. For $\beta=5.7$ with clover-improved fermions $(C_{sw} =1.57)$ all results are consistent with $\delta=0.065\pm 0.013$ ., Comment: 51 pages, 20 figures, Latex

The OZI rule is prominent in hadronic phenomena only because OZI violation is typically an order of magnitude smaller than expected from large N_c arguments. With its standard ^3P_0 pair creation operator for hadronic decays by flux tube breaking, the quark model respects the OZI rule at tree level and exhibits the cancellations between OZI-violating meson loop diagrams required for this dramatic suppression. However, if the quark model explanation for these cancellations is correct, then OZI violation would be expected to be large in the nonet with the same quantum numbers as the pair creation operator: the 0^{++} mesons. Experiment is currently unable to identify these mesons, but we report here on a lattice QCD calculation which confirms that the OZI rule arises from QCD in the vector and axial vector mesons as observed, and finds a large violation of the rule in the scalar mesons as anticipated by the quark model. In view of this result, we make some remarks on possible connections between the ^3P_0 pair creation model, scalar mesons, and the U_A(1) anomaly responsible for the large OZI violation which drives the \eta' mass. In particular, we note that our result favors the large N_c and not the instanton interpretation of the solution to the \eta' mass problem., Comment: 29 pages, 9 figures; the original version of this paper had a sign error which affects the interpretation of some of the results; some new references added and typos corrected

Preliminary results of a study of quenched chiral logarithms at beta=5.7 are presented. Four independent determinations of the quenched chiral log parameter delta are obtained. Two of these are from estimates of the eta-prime mass, one from the residue of the hairpin diagram and the other from the topological susceptibility combined with the Witten-Veneziano formula. The other two determinations of delta are from measurement of virtual eta-prime loop effects in mpi^2 vs. quark mass and in the chiral behavior of the pseudoscalar decay constant. All results are consistent with delta=.080(15). The expected absence of quenched chiral logs in the axial-vector decay constant is also observed., Comment: Talk presented at Lattice '99 by H. Thacker

A truncated determinant algorithm is used to study the physical effects of the quark eigenmodes associated with eigenvalues below 400 MeV. This initial study focuses on coarse lattices (with O(a^2) improved gauge action), light internal quark masses and large physical volumes. Four bellwether full QCD processes are discussed: topological charge distributions, the eta prime propagator, string breaking as observed in the static energy and the rho decay into two pions., Comment: LATTICE99(Confinement); 3pgs(Latex), 4figs.(eps)

The computational requirements and dynamics of Monte Carlo simulations of unquenched QCD incorporating the infrared quark eigenmodes (up to $\approx \Lambda_{QCD}$) exactly and UV modes via a loop representation are discussed. The accuracy of such a loop representation is studied for a variety of lattice volumes and quark masses. The method has been successfully applied for lattices up to 10$^3$x20 at $a \simeq$0.17F with improved (clover) action, and allows simulations at or near kappa critical., Comment: LATTICE99(algorithms), 3 pages, espcrc2.sty

The modelling of the ultraviolet contributions to the quark determinant in lattice QCD in terms of a small number of Wilson loops is examined. Complete Dirac spectra are obtained for sizeable ensembles of SU(3) gauge fields at $\beta$=5.7 on 6$^4$, 8$^4$, and 10$^4$ lattices allowing for the first time a detailed study of the volume dependence of the effective loop action generating the quark determinant. The connection to the hopping parameter expansion is examined in the heavy quark limit. We compare the efficiency and accuracy of various methods- specifically, Lanczos versus stochastic approaches- for extracting the quark determinant on an ensemble of configurations., Comment: 20 pages, LaTeX

For purposes of regularization as well as numerical simulation, the discretization of Lorentz invariant continuum field theories on a space-time lattice is often convenient. In general, this discretization destroys the rotational or Lorentz-frame independence of the theory, which is only recovered in the continuum limit. The Baxter 8-vertex model may be interpreted as a particular discretization of a self-interacting massive Dirac fermion theory in two dimensions (the massive Thirring model). Here it is shown that, in the 8-vertex/massive Thirring model, the Lorentz frame independence of the theory remains undisturbed on the lattice. The only effect of the discretization is to compactify the manifold of Lorentz frames. The relationship between this lattice Lorentz symmetry and the Yang-Baxter relations is discussed., Comment: 15 pages, no figures

Using the pole shifting procedure of the modified quenched approximation (MQA) to cure the exceptional configuration problem, accurate hadron hadron spectrum calculations can be obtained at very light quark mass. Here we use the MQA to extend and improve our previous investigation of chiral logs in the pion mass. At beta=5.7 for Wilson fermion, we see clear evidence for quenched chiral logarithms in the pion mass as a function of quark mass. The size of the observed chiral log exponent delta is in good agreement with the value obtained from a direct calculation of the eta' hairpin diagram., Comment: 3 pages, 4 figures, Lattice 98 talk

Virtual quark effects in lattice QCD are studied in an approach where the infrared modes are included in a precise and gauge-invariant manner. At fixed physical volume the required number of modes does not increase in the continuum limit, and the acceptance is tolerable even for very light quark masses. Topological charge distributions at varying quark mass, and the static quark potential, showing the onset of string-breaking, are presented in both QED2 and QCD4., Comment: LATTICE98 (algorithms)

Within a truncated determinant algorithm, two alternatives are discussed for including systematically the remaining ultraviolet modes. Evidence is presented that these modes are accurately described by an effective action involving only small Wilson loops., Comment: LATTICE98(algorithms) ; typos corrected

A new approach to the inclusion of virtual quark effects in lattice QCD simulations is presented. Infrared modes which build in the chiral physics in the light quark mass limit are included exactly and in a gauge invariant way. At fixed physical volume the number of relevant infrared modes does not increase as the continuum limit is approached. The acceptance of our procedure does not decrease substantially in the limit of small quark masses. Two alternative approaches are discussed for including systematically the remaining ultraviolet modes. In particular, we present evidence that these modes are accurately described by an effective action involving only small Wilson loops., Comment: 37 pages (Latex) including 18 figures (EPS)

We examine artifacts associated with the chiral symmetry breaking induced through the use of Wilson-Dirac fermions in lattice Monte Carlo computations. For light quark masses, the conventional quenched theory can not be defined using direct Monte Carlo methods due to the existence of nonintegrable poles in physical quantities. These poles are associated with the real eigenvalue spectrum of the Wilson-Dirac operator. We show how this singularity structure can be observed in the analysis of both QED in two dimensions and QCD in four dimensions., Comment: 32 pages (Latex) including 13 figures (EPS)

The calculation of the eta-prime hairpin diagram is carried out in the modified quenched approximation (MQA) in which the lattice artifact which causes exceptional configurations is removed by shifting observed poles at kappa

In lattice QCD with Wilson fermions, exceptional configurations arise in the quenched approximation at small quark mass. The origin of these large previously uncontrolled lattice artifacts is identified. A simple well-defined procedure (MQA) is presented which removes the artifacts while preserving the correct continuum limit., Comment: Talk presented by E. Eichten at Lattice 97, Edinburgh(UK), July97. 6 pages, LaTeX, 1 table, 5 figures

Lattice QED2 with the Wilson formulation of fermions is used as a convenient model system to study artifacts of the quenched approximation on a finite lattice. The quenched functional integral is shown to be ill-defined in this system as a consequence of the appearance of exactly real modes for physical values of the fermion mass. The location and frequency of such modes is studied as a function of lattice spacing, lattice volume, topological charge and improved action parameters. The efficacy of the recently proposed modified quenched approximation is examined, as well as a new approach to the interpolation from the quenched to full dynamical theory employing a truncated form of the fermion determinant., Comment: Talk presented by A. Duncan at LATTICE97 (theoretical developments)

In continuum QCD, nontrivial gauge topologies give rise to zero eigenvalues of the massless Dirac operator. In lattice QCD with Wilson fermions, these zero modes appear as exactly real eigenvalues of the Wilson-Dirac operator and hence as poles in the quark propagator in the vicinity of critical hopping parameter. It is shown that "exceptional configurations" which arise in the quenched approximation at small quark mass are the result of the fluctuation of the position of zero mode poles to sub-critical values of the hopping parameter on particular gauge configurations. We describe a procedure for correcting these lattice artifacts by first isolating the contribution of zero mode poles to the quark propagator and then shifting the sub-critical poles to the critical point. This procedure defines a modified quenched approximation in which accurate calculations can be carried out for arbitrarily small quark masses., Comment: 36 pages, Latex, with 11 embedded figures