arXiv subjects
Publications and source records attributed to P. M. Stevenson.
In their defence of "maximum conformality" methods, Brodsky et al make the astonishing claim that any RG transformation a'=a(1+V_1 a + ...) in QCD must have V_1 proportional to b=(33-2n_f)/6. It is well known that this is not true. I emphasize again the correctness and central importance of the Celmaster-Gonsalves relation for the prescription dependence of the Lambda parameter.
Physical quantities in QCD do not depend upon $α_s(Q)$. There is no way to measure $α_s(Q)$ experimentally. If those statements sound shocking, please read on. They are actually well-known facts, though ones that are constantly being ignored in the QCD literature. Renormalized perturbation theory is not an ordinary power-series expansion; its renormalization-scheme ambiguity is not merely a minor nuisance. Rather, it is a structure in which invariance under redefinitions of the coupling is a fundamental symmetry -- a symmetry that, like any other, deserves respect. I speak bluntly because without a change in mindset perturbative QCD can never become a proper, scientific enterprise.
The so-called "principle of maximal conformality" is ineffective and does nothing to resolve the renormalization-scheme-dependence problem. Some essential facts about that problem are summarized. It is stressed that RG invariance is a symmetry and that any viable method for resolving the scheme-dependence problem should be formulatable in terms of the invariants of that symmetry.
Perturbative calculations of factorized physical quantities, such as moments of structure functions, suffer from renormalization- and factorization-scheme dependence. The application of the principle of minimal sensitivity to "optimize" the scheme choices is reconsidered, correcting deficiencies in the earlier literature. The proper scheme variables, RG equations, and invariants are identified. Earlier results of Nakkagawa and Niegawa are recovered, even though their starting point is, at best, unnecessarily complicated. In particular, the optimized coefficients of the coefficient function C are shown to vanish, so that C^opt=1. The resulting simplifications mean that the optimization procedure is as simple as that for purely-perturbative physical quantities.
The slope of the beta function at a fixed point is commonly thought to be RG invariant and to be the critical exponent gamma* that governs the approach of any physical quantity R to its fixed-point limit: R*-R proportional to Q^gamma*. Chyla has shown that this is not quite true. Here we define a proper RG invariant, the "effective exponent" gamma(Q), whose fixed-point limit is the true gamma*.
The recent QCD calculations of the five-loop beta function and of R(e+e-) to O(alpha_s^4) provide one more term in the Banks-Zaks expansion in (16.5-nf). There is no longer any hope that the expansion could extend, even crudely, to low nf. Above nf=9, however, the results appear to be reasonably consistent from order to order.
Perturbative QCD with nf flavours of massless quarks becomes simple in the hypothetical limit nf -> 16.5, where the leading beta-function coefficient vanishes. The Banks-Zaks (BZ) expansion in a0=(8/321)(16.5-nf) is straightforward to obtain from perturbative results in MSbar or any renormalization scheme (RS) whose nf dependence is `regular.' However, `irregular' RS's are perfectly permissible and should ultimately lead to the same BZ results. We show here that the `optimal' RS determined by the Principle of Minimal Sensitivity does yield the same BZ-expansion results when all orders of perturbation theory are taken into account. The BZ limit provides an arena for exploring optimized perturbation theory at arbitrarily high orders. These explorations are facilitated by a `master equation' expressing the optimization conditions in the fixed-point limit. We find an intriguing strong/weak coupling duality a -> a*^2/a about the fixed point a*.
Perturbative QCD, when optimized by the principle of minimal sensitivity at fourth order, yields finite results for R(e+e-)(Q) down to Q=0. For two massless flavours (n_f=2) this occurs because the couplant "freezes" at a fixed point of the optimized beta function. However, for larger n_f's, between 6.7 and 15.2, the infrared limit arises by a novel mechanism in which the evolution of the optimized beta function with energy Q is crucial. The evolving beta function develops a minimum that, as Q -> 0, just touches the axis at a_p (the "pinch point"), while the infrared limit of the optimized couplant is at a larger value, a^star (the "unfixed point"). This phenomenon results in R approaching its infrared limit not as a power law, but as R -> R^star-const./|ln Q|^2. Implications for the phase structure of QCD as a function of n_f are briefly considered.
Physical quantities in QCD are independent of renormalization scheme (RS), but that exact invariance is spoiled by truncations of the perturbation series. "Optimization" corresponds to making the perturbative approximant, at any given order, locally invariant under small RS changes. A solution of the resulting optimization equations is presented. It allows an efficient algorithm for finding the optimized result. Example results for R(e+e-)=3(Sum q_i^2)(1+R) to fourth order (NNNLO) are given that show nice convergence, even down to arbitrarily low energies. The Q=0 "freezing" behaviour, R=0.3+/-0.3, found at third order is confirmed and made more precise; R=0.2+/-0.1. Low-energy results in the MS-bar scheme, by contrast, show the typical pathologies of a non-convergent asymptotic series.
Does massless (lambda Phi^4)_4 theory exhibit spontaneous symmetry breaking (SSB)? The raw 1-loop result implies that it does, but the "RG-improved" result implies the opposite. I argue that the appropriate "low-energy effective theory" is a nonlocal field theory involving an attractive, long-range interaction Phi^2(x) Phi^2(y)/z^4, where z=|x-y|. RG improvement then requires running couplings for both this interaction and the original pointlike interaction. A crude calculation in this framework yields SSB even after "RG improvement" and closely agrees with the raw 1-loop result.
Predictions for (phi^4)_4 theory from renormalization-group-improved perturbation theory, as formulated by Luescher and Weisz, are compared to published (and some unpublished) data from lattice Monte-Carlo simulations of the 4-dimensional Ising model. Good agreement is found in all but one respect:-- the change in the wavefunction-renormalization constant Z_R across the phase transition is significantly greater than predicted. A related observation is that propagator data in the broken phase show deviations from free-propagator form -- deviations that become larger, not smaller, closer to the continuum limit. More data closer to the critical point are needed to clarify the situation.
Hydrodynamics is the appropriate "effective theory" for describing any fluid medium at sufficiently long length scales. This paper treats the vacuum as such a medium and derives the corresponding hydrodynamic equations. Unlike a normal medium the vacuum has no linear sound-wave regime; disturbances always "propagate" nonlinearly. For an "empty vacuum" the hydrodynamic equations are familiar ones (shallow water-wave equations) and they describe an experimentally observed phenomenon -- the spreading of a clump of zero-temperature atoms into empty space. The "Higgs vacuum" case is much stranger; pressure and energy density, and hence time and space, exchange roles. The speed of sound is formally infinite, rather than zero as in the empty vacuum. Higher-derivative corrections to the vacuum hydrodynamic equations are also considered. In the empty-vacuum case the corrections are of quantum origin and the post-hydrodynamic description corresponds to the Gross-Pitaevskii equation. I conjecture the form of the post-hydrodynamic corrections in the Higgs case. In the 1+1-dimensional case the equations possess remarkable `soliton' solutions and appear to constitute a new exactly integrable system.
Low-momentum excitations of a dilute Bose-Einstein condensate behave as phonons and move at a finite velocity v_s. Yet the atoms making up the phonon excitation each move very slowly; v_a = p/m --> 0. A simple "cartoon picture" is suggested to understand this phenomenon intuitively. It implies a relation v_s/v_a = N_ex, where N_ex is the number of excited atoms making up the phonon. This relation does indeed follow from the standard Bogoliubov theory.
The Higgs vacuum -- with its constant background field -- is not `empty' but is a kind of medium. Any ordinary medium, viewed at sufficiently long length scales, has a hydrodynamic description and can propagate sound waves. The vacuum medium is unusual; the speed of sound is formally infinite and there is no linear sound-wave regime. Instead, long-wavelength disturbances are described by some intrinsically nonlinear hydrodynamic equations.
The Higgs vacuum is a kind of medium. In any medium one generally expects sound waves for sufficiently long wavelengths (>> mean free path). I briefly describe how the broken-symmetry vacuum can be viewed as a Bose-Einstein condensate of `phion' particles. This picture yields a natural notion of the `mean free path'. I speculate that this is at the millimeter-centimeter scale.
Using a high-statistics lattice simulation of the Ising limit of $(λΦ^4)_4$ theory, we have measured the susceptibility and propagator in the broken phase. We confirm our earlier finding of a discrepancy between the field re-scaling implied by the propagator data and that implied by the susceptibility. The discrepancy becomes {\it worse} as one goes closer to the continuum limit; thus, it cannot be explained by residual perturbative effects. The data are consistent with an unconventional description of symmetry breaking and ``triviality'' in which the re-scaling factor for the finite-momentum fluctuations tends to unity, but the re-scaling factor for the condensate becomes larger and larger as one approaches the continuum limit. In the Standard Model this changes the interpretation of the Fermi-constant scale and its relation to the Higgs mass.
We discuss the phase transition in 3+1 dimensional lambda Phi^4 theory from a very physical perspective. The particles of the symmetric phase (`phions') interact via a hard-core repulsion and an induced, long-range -1/r^3 attraction. If the phion mass is sufficiently small, the lowest-energy state is not the `empty' state with no phions, but is a state with a non-zero density of phions Bose-Einstein condensed in the zero-momentum mode. The condensate corresponds to the spontaneous-symmetry-breaking vacuum with neq 0 and its excitations ("phonons" in atomic-physics language) correspond to Higgs particles. The phase transition happens when the phion's physical mass m is still positive; it does not wait until m^2 passes through zero and becomes negative. However, at and near the phase transition, m is much, much less than the Higgs mass M_h. This interesting physics coexists with `triviality;' all scattering amplitudes vanish in the continuum limit, but the vacuum condensate becomes infinitely dense. The ratio m/M_h, which goes to zero in the continuum limit, can be viewed as a measure of non-locality in the regularized theory. An intricate hierarchy of length scales naturally arises. We speculate about the possible implications of these ideas for gravity and inflation.
We discuss the physics of the 3+1 dimensional lambda Phi^4 quantum field theory in terms of the statistical mechanics of a gas of particles (`atoms') that interact via a -1/r^3-plus-hard-core potential. The hard-core potential, delta^(3)(r), arises from the bare vertex diagram, while the attractive, long-range -1/r^3 potential is due to exchange of a particle pair via the t,u-channel "fish" diagram. (Higher-order diagrams preserve this form of the interparticle potential.) For sufficiently small atom mass, the lowest-energy state is not the `empty' state with no atoms, but a state with a non-zero density of spontaneously created atoms, Bose-condensed in the zero-momentum mode. This corresponds to the spontaneous-symmetry-breaking phase transition, and the `phonon' excitations of the Bose condensate correspond to Higgs particles. The important point is that the phase transition happens while the atom's physical mass m is still positive: it does not wait until m^2 passes through zero and becomes negative, contrary to the assumption of a second-order transition, on which renormalization-group-improved perturbation theory is based.