1: \documentclass[12pt,preprint]{article}
2: %\usepackage{feynmf}
3: \usepackage{amsmath}
4: %\usepackage{amssymbol}
5: \usepackage{graphics}
6: \usepackage{epsfig}
7: %\usepackage{psfig}
8: \renewcommand{\vec}[1]{{\bf #1}}
9: %\renewcommand{\baselinestretch}{2}\normalsize
10: \setlength{\baselineskip}{20mm}
11: \textwidth 15.0 true cm
12: \textheight 22.0 true cm
13: \headheight 0 cm
14: \headsep 0 cm
15: \topmargin 0.4 true in
16: \oddsidemargin 0.25 true in
17: \newcommand{\eqb}{\begin{equation}}
18: \newcommand{\eqe}{\end{equation}}
19: \newcommand{\dmb}{\begin{displaymath}}
20: \newcommand{\dme}{\end{displaymath}}
21: \newcommand{\pd}{\partial}
22: \newcommand{\ep}{\varepsilon}
23: \newcommand{\eab}{\begin{eqnarray}}
24: \newcommand{\eae}{\end{eqnarray}}
25: \newcommand{\ra}{\right\rangle}
26: \newcommand{\la}{\left\langle}
27: \newcommand{\e}{\mbox{e}}
28: \newcommand{\be}{\begin{equation}}
29: \newcommand{\ee}{\end{equation}}
30: \newcommand{\sgn}{\text{sgn}\,}
31: \newcommand{\munu}{{\mu\nu}}
32: \newcommand{\ad}{{\dot{\alpha}}}
33: \newcommand{\bd}{{\dot{\beta}}}
34: \newcommand{\La}{\Lambda}
35: \setlength{\unitlength}{1mm}
36:
37: \begin{document}
38: \begin{titlepage}
39: \begin{flushright}
40: \end{flushright}
41: \vspace{0.6cm}
42:
43: \begin{center}
44: \Large{Yang-Mills thermodynamics: The confining phase}
45:
46: \vspace{1.5cm}
47:
48: \large{Ralf Hofmann}
49:
50: \end{center}
51: \vspace{1.5cm}
52:
53: \begin{center}
54: {\em Institut f\"ur Theoretische Physik\\
55: Universit\"at Frankfurt\\
56: Johann Wolfgang Goethe - Universit\"at\\
57: Robert-Mayer-Str. 10\\
58: 60054 Frankfurt, Germany}
59: \end{center}
60: \vspace{1.5cm}
61:
62: \begin{abstract}
63: We summarize recent nonperturbative results obtained for the
64: thermodynamics of an SU(2) and an SU(3)
65: Yang-Mills theory being in its
66: confining (center) phase. This phase is associated with a
67: dynamical breaking of the local magnetic
68: center symmetry. Emphasis is
69: put on an explanation of the involved
70: concepts.
71:
72: \end{abstract}
73:
74: \end{titlepage}
75:
76: \noindent{\sl Introduction.} This is the last one in a series of three papers giving an
77: abbreviated presentation of nonperturbative concepts and results for the
78: thermodynamics of an SU(2) or an SU(3) Yang-Mills theory as obtained in
79: \cite{Hofmann2005,HerbstHofmannRohrer2004,HerbstHofmann2004}. Here we discuss the
80: confining or center phase.
81:
82: The three unexpected results for the confining phase are the spin-1/2
83: nature of the massless (neutral, Majorana) and massive (charged) excitations, the Hagedorn
84: nature of the transition from the confining to the preconfining phase,
85: and the exact vanishing of the pressure and energy density of the
86: ground state in the confining phase of an SU(2) or SU(3)
87: Yang-Mills theory.
88:
89: The first result clashes
90: with the perception about
91: bosonic glueballs being the observable excitations of
92: pure SU(3) Yang-Mills theory at
93: zero temperature. This statement seems to be supported by
94: lattice simulations \cite{MorningstarPeardon1999} and
95: by analysis based on the QCD-sum-rule method \cite{Narison1984}.
96: We have discussed in \cite{Hofmann2005} why lattice
97: simulations of pure SU(2) and SU(3) Yang-Mills theory
98: run into a severe finite-size problem
99: at low temperatures and thus are unreliable. QCD spectral
100: sum rules \cite{ShifmanVainshteinZakharov1978}, on the other hand, {\sl assume}
101: the existence of a lowest resonance with finite coupling
102: to the currents of a given production
103: channel (these currents are formulated as local functionals
104: of the fundamental fields in the QCD Lagrangian). The resonance's properties are
105: determined subsequently by assuming the analyticity of the associated correlation
106: function in the external momentum and by appealing to an operator product expansion in the
107: deep euclidean region. Analyticity, however, must break
108: down across two phase boundaries (deconfining-preconfining, preconfining-center) provided
109: that the effects arising due to the deviation from the thermodynamical limit can,
110: on a qualitative level, be neglected in the production process. As a consequence,
111: the QCD sum rule method is probably unreliable for the
112: investigation of the spectrum of a {\sl pure} SU(2) and SU(3)
113: Yang-Mills theory. (We hasten to add that
114: the situation is different for real-world hadronic resonances because the dynamical mixing
115: of pure SU(3) and pure dual SU(3) gauge theories
116: may restore a quasi-analytical behavior of the relevant
117: correlation functions \cite{Hofmann2005}. After all the overwhelming
118: phenomenological successes of QCD sum rules are a lot more more than
119: coincidence. The term `dynamical mixing' includes the occurrence of
120: the fractional Quantum Hall effect \cite{Laughlin1983,TsuiStoermerGossard1982}
121: which renders quarks to be emerging phenomena,
122: for a discussion see \cite{Hofmann2005}.)
123:
124: The second result -- the Hagedorn nature of the
125: transition to the truly confining phase -- was suspected
126: to occur for the real-world strong interactions
127: a long time ago \cite{Hagedorn1965,Veneziano1968}. Subsequently performed lattice
128: simulations seemed to exclude a Hagedorn transition
129: (diverging partition function above the critical point) even in the case of a
130: pure SU(2) or SU(3) Yang-Mills theory \cite{LuciniTeperWenger}. Again, this a
131: consequence of the lattice's failure to properly
132: capture the infrared physics in
133: thermodynamical simulations at low temperature, for an
134: extended discussion see \cite{Hofmann2005}.
135:
136: The third result, namely the excact vanishing of the ground-state pressure
137: and energy density of the Yang-Mills theory at zero temperature, commonly is
138: used as a normalization assumption in lattice computations \cite{Bielfeld19961999}
139: and not obtained as a dynamical result. In \cite{Hofmann2005} we have shown the absoluteness,
140: that is, the {\sl gravitational} measurablility of the finite and exactly
141: computable energy-momentum tensor associated with the
142: ground-state in the deconfining and preconfining phases: An immediate consequence of
143: the fact that these ground states are determined by
144: radiatively protected BPS equations. (Since these equations are first-order as
145: opposed to second-order Euler-Lagrange equations the usual shift
146: ambiguity in the corresponding potentials is absent.) Recall, that the
147: finiteness of the ground-state energy density and
148: pressure in the deconfining and preconfining phases arises from averaged-over interactions
149: between and radiative corrections within solitonic field configurations.
150: Being (euclidean) BPS saturated, classical configurations in the deconfining phase the
151: latter are free of pressure and energy density in isolation. The same applies
152: to the massless, interacting magnetic monopoles which, by their condensation, form the ground state in
153: the preconfining phase. In the confining phase configurations that are
154: free of pressure and energy density do also exist (single center-vortex loops). In contrast to the
155: other phases propagating gauge field fluctuations are, however, absent in the confining phase.
156: Only contact interactions occur between the center solitons, which, however, do not elevate the vanishing
157: energy density of the isolated soliton to a finite value for the
158: ensemble. The proof for this relies on computing the curvature
159: of the potential for the spatial coarse-grainined center-vortex
160: condensate at its zeros and by comparing this curvature with the square of the maximal
161: resolution that is allowed for in the
162: effective theory \cite{Hofmann2005}, see below.
163:
164: The outline of this paper is as follows: First, we
165: discuss the occurence of isolated, instable, that is, contracting and
166: collapsing center-vortex loops in the preconfining phase.
167: From the evolution of the magnetic coupling constant in this
168: phase we conclude that center-vortex loops become stable, particle-like
169: excitations at the deconfining-confining phase boundary.
170: Second, we point out the spin-1/2 nature of these particles, and we derive a dimensionless
171: parameter with discrete values describing the condensate
172: of pairs of single center-vortex loops
173: after spatial coarse-graining. A discussion of
174: the creation of center-fluxes (local phase jumps of the vortex condensate)
175: by the decay of the monopole
176: condensate in the preconfining phase is given. Third, we
177: construct potentials for the vortex condensates which, in their physical effects,
178: are uniquely determined by the remaining local symmetry and by the positive
179: semi-definiteness of the energy density: Particle creation
180: by local phase jumps of the order parameter
181: may only go on so long as the energy density feeding into their
182: creation is nonvanishing. Fourth, we discuss in detail
183: the remarkable result that for SU(2) and SU(3) Yang-Mills dynamics
184: the confining phase's ground-state energy density is exactly nil.
185: In particular, we stress the fact that radiative
186: corrections to the tree-level result are
187: entirely absent. Fifth, we give an estimate for the density of
188: static fermion states and thus establish the Hagedorn nature of
189: the transition from the confining to the preconfining phase. Finally,
190: we summarize our results in view of its implications
191: for particle physics and for cosmology.\vspace{0.3cm}\\
192: \noindent{\sl Instable center-vortex loops in the preconfining phase.}
193: Here we discuss the SU(2) case only, results for SU(3) follow by
194: simple doubling. The ground-state of the preconfining
195: phase is a condensate of magnetic monopoles peppered with
196: instable defects: closed magnetic flux lines whose core
197: regions dissolve the condensate locally and thus restore the
198: dual gauge symmetry U(1)$_D$ (for SU(3): U(1)$^2_D$).
199: It was shown in \cite{Hofmann2005} that the magnetic
200: flux carried by a given vortex-loop solely depends on
201: the {\sl charge} of the monopoles and antimonopoles contributing
202: to the explicit magnetic current inside the vortex core. Thus the various species
203: of vortex-loops, indeed, are mapped one-to-one
204: onto the nontrivial center elements of SU(2) or SU(3): They deserve the
205: name center-vortex loops. In the magnetic phase, center-vortex
206: loops are, however, instable as we show now. To derive the classical
207: field configuration associated with an infinitely
208: long vortex line one considers an Abelian Higgs model
209: with no potential and a magnetic coupling $g$.
210: (We need to discuss the energy-momentum tensor
211: of the solitonic configuration {\sl relative} to the ground state
212: obtained by spatially averaging over instable vortex loops. Thus
213: we need to substract the temperature dependent
214: ground-state contribution which is reached far away from the considered vortex core
215: as a result of the applicable spatial coarse-graining,
216: see \cite{Hofmann2005} for details.) The following
217: ansatz is made for the static dual gauge field $a^D_\mu$ \cite{NielsenOlesen1973}:
218: %*******
219: \eqb
220: \label{ANOansatz}
221: a_4^D=0\,,\ \ \ \ \ \ \ \ \ a_i^D=\epsilon_{ijk}\hat{r}_j e_k\, A(r)
222: \eqe
223: %********
224: where $\hat{r}$ is a radial unit vector in
225: the $x_1x_2$ plane, $r$ is the distance from the vortex core,
226: and $\vec{e}$ denotes a unit vector along the vortex' symmetry axis which we choose to coincide with
227: the $x_3$ coordinate axis. No analytical
228: solution with a finite energy per vortex length
229: is known to the system of the two coupled equations of motion
230: honouring the ansatz (\ref{ANOansatz}) and the Higgs-field
231: decomposition $\varphi=|\varphi|(r)\exp[i\theta]$. An approximate
232: solution, which assumes the constancy of $|\varphi|$, is given as
233: %***********
234: \eqb
235: \label{A(r)}
236: A(r)=\frac{1}{gr}-|\varphi|K_1(g|\varphi|r)\longrightarrow \frac{1}{gr}-
237: |\varphi|\sqrt{\frac{\pi}{2g|\varphi|r}}\exp[-g|\varphi|r]\,,\ \ \ \ \ (r\to\infty)\,.
238: \eqe
239: %************
240: In Eq.\,(\ref{A(r)}) $K_1$ is a modified Bessel function. Outside the core region
241: the isotropic pressure $P_v(r)$ in the $x_1x_2$ plane is, up to an exponentially
242: small correction, given as
243: %**********
244: \eqb
245: \label{vortexpres}
246: P_v(r)=-\frac{1}{2}\frac{\Lambda_M^3\beta}{2\pi}\,\frac{1}{g^2 r^2}\,.
247: \eqe
248: %***********
249: Notice that we have substituted the asymptotic value
250: $|\varphi|=\sqrt{\frac{\Lambda_M^3\beta}{2\pi}}\,,\ (\beta\equiv\frac{1}{T})\,,$
251: as it follows from the spatially coarse-grained
252: action in the preconfining (or magnetic) phase \cite{Hofmann2005,Hofmann20052}. Notice also the minus
253: sign on the right-hand side of Eq.\,(\ref{vortexpres}): The
254: configuration in Eq.\,(\ref{A(r)}) is static due to its cylindrical symmetry
255: but highly instable w.r.t. bending of the vortex axis. In particular,
256: the pressure inside a center-vortex {\sl loop} is more negative than outside causing the soliton to
257: contract, and, eventually, to dissolve. Bending of the vortex axis occurs because
258: there are no isolated magnetic charges in the preconfining phase which could serve as sources
259: for the magnetic flux. An equilibrium between vortex-loop
260: creation by the spatially and temporally {\sl correlated} dissociation of large-holonomy calorons
261: and vortex-loop collapse is responsible for the
262: negative pressure of the ground state in the preconfining phase.
263: The typical core-size $R$ of a center-vortex loop evaluates as
264: $R\sim\frac{1}{m_{a^{\tiny D}_\mu}}=\frac{1}{g}\sqrt{\frac{\Lambda_M^3}{\beta}}$ and
265: its energy as $E_v\sim\frac{\pi}{g}\sqrt{\frac{\Lambda_M^3\beta}{2\pi}}$. (This takes
266: into account an estimate for $\varphi$'s gradient contribution to the total
267: energy of the soliton.)
268:
269: Notice that core-size $R$,
270: energy $E_v$, and pressure $P_v(r)$ of a center-vortex vanish
271: in the limit $g\to\infty$. This situation
272: is reached at the critical temperature $T_{c,M}$ where the magnetic
273: coupling diverges in a logarithmic fashion: $g\sim-\log(T-T_{c,M})$ \cite{Hofmann2005}.
274: At $T_{c,M}$ the creation of single center-vortex loops at rest with respect to
275: the heat bath (i) does not cost any energy
276: and (ii) entails the existence of stable and massless particles. The latter do, in turn,
277: condense pairwise into a new ground state.\vspace{0.3cm}\\
278: \noindent{\sl Pairwise condensation of
279: single center-vortex loops: Ground-state decay and
280: change of statistics.} We consider a static,
281: circular contour $C(\vec{x})$ of infinite radius -- an $S_1$ --
282: which is centered at the point $\vec{x}$. In addition, at finite coupling $g$
283: we consider a system $S$ of two single center-vortex loops, 1 and 2, which both
284: are pierced by $C(\vec{x})$ and which contribute opposite units of
285: center flux $F_{v_1}=\frac{2\pi}{g}=\oint_{C(\vec{x})} dz_\mu\, a^D_{1,\mu}=-F_{v_2}$
286: through the minimal surface spanned by $C(\vec{x})$. Depending on
287: whether 1 collapses before or after 2 or whether 1 moves away
288: from $C$ before or after 2 the total center flux $F$ through $C'$s
289: minimal surface reads
290: %***********
291: \eqb
292: \label{fluxiso}
293: F=\left\{\begin{array}{c}
294: \hspace{-3cm}\pm\frac{2\pi}{g}\ \ \ \ \ \ (\mbox{either 1 or 2 is pierced by $C(\vec{x})$})\\
295: 0\ \ \ \ \ \ \ \ (\mbox{1 and 2 or neither 1 nor 2 are pierced by $C(\vec{x})$})\,.\end{array}\right.
296: \eqe
297: %************
298: The limit $g\to\infty$, which dynamically takes place at $T_{c,M}$, causes
299: the center flux of the isolated system $S$ to vanish and renders single center-vortex
300: loops massless and stable particles. The center flux of the isolated system $S$
301: does no longer vanish if we couple $S$ to the heat bath. Although 1 and
302: 2 individually are spin-1/2 fermions the system $S$ obeys
303: bosonic statistics. (Both, 1 and 2, come in two
304: polarizations: the projection of the dipole moment, generated by
305: the monopole current inside the core of the center-vortex loop, onto a
306: given direction in space either is parallel or
307: antiparallel to this direction.) Thus, assuming the spatial momentum
308: of 1 and 2 to vanish, the quantum statistical average flux reads
309: %*****
310: \eab
311: \label{avfluxsysV}
312: \lim_{g\to\infty}\,F_{\tiny\mbox{th}}&=&4\pi F\,\int d^3p\,\,\delta^{(3)}(\vec{p})\, n_B(\beta |2\,E_v(\vec{p})|)\nonumber\\
313: &=&0,\pm\frac{8\pi}{\beta|\varphi|}=0,\pm 4\,\lambda^{3/2}_{c,M}\,.
314: \eae
315: %*********
316: According to Eq.\,(\ref{avfluxsysV}) there are finite, discrete, and dimensionless
317: parameter values for the description of the
318: macroscopic phase
319: %********
320: \eqb
321: \label{phasePhi}
322: \Gamma\frac{\Phi}{|\Phi|}(\vec{x})\equiv \lim_{g\to\infty}\exp[i\la \oint_{C(\vec{x})}dz_\mu\,
323: a^D_\mu\ra]
324: \eqe
325: %**********
326: associated with the Bose condensate of the system $S$.
327: In Eq.\,(\ref{phasePhi}) $\Gamma$ is an undetermined and dimensionless
328: complex constant. Notice that taking the limit of vanishing spatial momentum for
329: each single center-vortex loop {\sl is} the implementation of
330: spatial coarse-graining. This coarse-graining is performed
331: down to a resolution $|\Phi|$ which is determined by the (existing) stable solution to the
332: equation of motion in the effective theory, see below.
333:
334: For convenience we normalize the parameter
335: values given by $\lim_{g\to\infty}\,F_{\tiny\mbox{th}}$
336: as $\hat{\tau}\equiv \pm 1,0$.\vspace{0.3cm}\\
337: \noindent{\sl Coarse-grained action and center jumps.}
338: To investigate the decay of the monopole condensate at $T_{c,M}$ (pre- and reheating)
339: and the subsequently emerging equilibrium situation, we need to find conditions to constrain
340: the potential $V_C$ for the macroscopic field $\Phi$ in such a way
341: that the dynamics arising from it is unique. Recall that
342: at $T_{c,M}$ the dual gauge modes of the preconfining phase
343: decouple. Thus the entire process of fermionic pre- and reheating in the confining phase
344: is described by spatially and temporally discontinuous changes of the
345: modulus (energy loss) and phase (flux creation) of
346: the field $\Phi$. Since the condensation of
347: the system $S$ renders the expectation of the 't Hooft loop finite (proportional
348: to $\Phi$) the magnetic center symmetry $Z_2$ (SU(2)) and $Z_3$ (SU(3))
349: is dynamically broken as a discrete gauge symmetry. Thus,
350: after return to equilibrium, the ground state of the confining phase
351: must exhibit $Z_2$ (SU(2)) and $Z_3$ (SU(3)) degeneracy. This implies that for SU(2)
352: the two parameter values $\hat{\tau}=\pm 1$ need to be identified while each of the
353: three values $\hat{\tau}=\pm 1,0$ describe a distinct
354: ground state for SU(3). Let us now discuss how either one of
355: these degenerate ground states is reached. Spin-1/2 particle creation
356: proceeds by single center vortex loops being sucked-in from
357: infinity. (The overall pressure is still negative during the
358: decay of the monopole condensate thus facilitating the in-flow
359: of spin-1/2 particles from spatial infinity.) At a given point $\vec{x}$ an observer
360: detects the in-flow of a massless fermion in terms of the field
361: $\Phi(\vec{x})$ rapidly changing its phase by a forward center jump
362: (center-vortex loop gets pierced by $C(\vec{x})$) which is followed by the associated
363: backward center jump (center-vortex loop lies inside $C(\vec{x})$).
364: Each phase change corresponds to a tunneling
365: transition inbetween regions of positive curvature in
366: $V_C$. If a phase jump has
367: taken place such that the subsequent potential
368: energy for the field $\Phi$ is still positive then
369: $\Phi$'s phase needs to perform additional jumps in order to
370: shake off $\Phi$'s energy completely. This can only
371: happen if no local minimum exists at a finite value of $V_C$.
372: If the created single center-vortex loop moves sufficiently fast it can subsequently
373: convert some of its kinetic energy into mass by twisting: massive, self-intersecting
374: center-vortex loops arise. These particles are also spin-1/2
375: fermions: A $Z_2$ or $Z_3$ monopole, constituting the
376: intersection point, reverses the center
377: flux \cite{Reinhardt2001}, see Fig.\,\ref{Fig-0}.
378: %***********************
379: \begin{figure}
380: \begin{center}
381: \leavevmode
382: %\epsfxsize=9.cm
383: \leavevmode
384: %\epsffile[80 25 534 344]{}
385: \vspace{4.5cm}
386: \special{psfile=Fig-0.ps angle=0 voffset=-120
387: hoffset=-170 hscale=60 vscale=50}
388: \end{center}
389: \caption{The creation of an isolated $Z_2$
390: monopole by self-intersection of a
391: center-vortex loop. \label{Fig-0}}
392: \end{figure}
393: %************************
394:
395: If the SU(2) (or SU(3)) pure
396: gauge theory does not mix with any other preconfining or deconfining
397: gauge theory, whose propagating gauge modes would couple
398: to the $Z_2$ (or $Z_3$) charges, a soliton generated
399: by $n$-fold twisting is stable in isolation and possesses a
400: mass $n\,\Lambda_C$. Here $\Lambda_C$ is the mass of the
401: charge-one state (one self-intersection). After a sufficiently large and even number
402: of center jumps has occurred the
403: field $\Phi(\vec{x})$ settles in one of
404: its minima of zero energy density. Forward - and backward tunneling inbetween
405: these minima corresponds to the spontaneous on-shell generation of a massless,
406: single center-vortex loop of zero momentum. In a WKB-like approximation one
407: expects the associated euclidean trajectory to have a
408: large action which, in turn, predicts large suppression. We conclude that tunneling
409: between the minima of zero energy density
410: is forbidden.
411:
412: Let us summarize the results of our above discussion: (i) the potential
413: $V_C$ must be invariant under magnetic
414: center jumps $\Phi\rightarrow\exp[\pi i]\Phi$\,\,(SU(2)) and
415: $\Phi\rightarrow\exp[\pm \frac{2\pi}{3}]\Phi$\,\,(SU(3))
416: only. (An invariance under a larger continuous or
417: discontinuous symmetry is excluded.) (ii)
418: Fermions are created by a forward - and backward
419: tunneling corresponding to local center jumps in $\Phi$'s
420: phase. (iii) The minima of
421: $V_C$ need to be at zero-energy density and are
422: all related to each other by center transformations,
423: no additional minima exist. (iv)
424: Moreover, we insist on the occurrence of one mass
425: scale $\Lambda_C$ only to parameterize the potential $V_C$. (As it was
426: the case for the ground-state physics in the de -
427: and preconfining phases.) (v) In addition, it is clear that the potential $V_C$ needs to
428: be real.\vspace{0.1cm}\\
429: \noindent\underline{SU(2) case:}\vspace{0.1cm}\\
430: A generic potential $V_C$ satisfying (i),(ii), (iii), (iv), and (v) is given by
431: %*********
432: \eqb
433: \label{2potC}
434: V_C=\overline{v_C}\,v_C\equiv\overline{\left(\frac{\Lambda_C^3}{\Phi}-\La_C\,\Phi\right)}\,
435: \left(\frac{\Lambda_C^3}{\Phi}-\La_C\,\Phi\right)\,.
436: \eqe
437: %*********
438: The zero-energy minima of $V_C$
439: are at $\Phi=\pm \Lambda_C$. It is clear that
440: adding or subtracting powers $(\Phi^{-1})^{2l+1}$ or
441: $\Phi^{2k+1}$ in $v_C$, where $k,l=1,2,3,\cdots$,
442: generates additional zero-energy
443: minima, some of which are {\sl not} related by center transformations
444: (violation of requirement (iii)). Adding $\Delta V_C$, defined as
445: an {\sl even} power of a Laurent expansion
446: in $\bar{\Phi}\Phi$, to $V_C$ (requirements (iii) and (v)), does in general
447: destroy property (iii). A possible exception is
448: %*******
449: \eqb
450: \label{DeltaVC}
451: \Delta V_C=\lambda\left(\Lambda_C^2-
452: \Lambda_C^{-2(n-1)}\left(\bar{\Phi}\Phi\right)^n\right)^{2k}
453: \eqe
454: %*********
455: where $\lambda>0; k=1,2,3,\cdots; n\in {\bf Z}$. Such a term, however,
456: is irrelevant for the description of the tunneling
457: processes (requirement (ii)) since the associated euclidean
458: trajectories are essentially along U(1) Goldstone
459: directions for $\Delta V_C$ due to the
460: pole in Eq.\,(\ref{2potC}). Thus adding $\Delta V_C$ does not cost
461: much additional euclidean action and therefore does not affect
462: the tunneling amplitude in a significant way. As for the
463: curvature of the potential at its minima, adding $\Delta V_C$ does not lower
464: the value as obtained for $V_C$ alone. One
465: may think of multiplying $V_C$ with a positive, dimensionless
466: polynomial in $\Lambda_C^{-2}\bar{\Phi}\Phi$ with
467: coefficients of order unity. This, however,
468: does not alter the physics of the pre - and reheating
469: process. It increases the curvature of the
470: potential at its zeros and therefore does not alter the result that quantum
471: fluctuations of $\Phi$ are absent after relaxation.
472: \vspace{0.1cm}\\
473: \noindent\underline{SU(3) case:}\vspace{0.1cm}\\
474: A generic potential $V_C$ satisfying (i),(ii), (iii), (iv), and (v) is given by
475: %*********
476: \eqb
477: \label{3potC}
478: V_C=\overline{v_C}\,v_C\equiv\overline{\left(\frac{\Lambda_C^3}{\Phi}-\Phi^2\right)}\,
479: \left(\frac{\Lambda_C^3}{\Phi}-\Phi^2\right)\,.
480: \eqe
481: %*********
482: The zero-energy minima of $V_C$
483: are at $\Phi=\Lambda_C\exp\left[\pm\frac{2\pi i}{3}\right]$ and $\Phi=\Lambda_C$.
484: Again, adding or subtracting powers $({\Phi}^{-1})^{3l+1}$ or $(\Phi)^{3k-1}$
485: in $v_C$, where $l=1,2,3,\cdots$ and $k=2,3,4,\cdots$,
486: violates requirement (iii). The same discussion for adding
487: $\Delta V_C$ to $V_C$ and for
488: multiplicatively modifying $V_C$ applies as in the SU(2) case.
489: In Fig.\,\ref{Fig-1} plots of the potentials in Eq.\,(\ref{2potC}) and Eq.\,(\ref{3potC}) are shown.
490: %***********************
491: \begin{figure}
492: \begin{center}
493: \leavevmode
494: %\epsfxsize=9.cm
495: \leavevmode
496: %\epsffile[80 25 534 344]{}
497: \vspace{5.5cm}
498: \special{psfile=Fig-1.ps angle=0 voffset=-150
499: hoffset=-190 hscale=50 vscale=50}
500: \end{center}
501: \caption{The potential $V_C=\overline{v_C(\Phi)}v_C(\Phi)$ for the center-vortex
502: condensate $\Phi$. Notice the regions of negative tangential curvature
503: inbetween the minima.\label{Fig-1}}
504: \end{figure}
505: %************************
506: The ridges of negative tangential curvature are classically forbidden:
507: The field $\Phi$ tunnels through these ridges, and a
508: phase change, which is determined by an element of the center $Z_2$ (SU(2)) or $Z_3$ (SU(3)),
509: occurs locally in space. This is the afore-mentioned generation of
510: one unit of center flux.\vspace{0.3cm}\\
511: \noindent{\sl No vacuum energy after relaxation.}
512: The action describing the process of relaxation of $\Phi$
513: to one of $V_C$'s minima is
514: %**********
515: \eqb
516: \label{actionPhi}
517: S = \int d^4x
518: \left(\frac{1}{2}\,\overline{\partial_\mu \Phi}
519: \partial^\mu \Phi - \frac12\, V_C \right) \,.
520: \eqe
521: %***********
522: Once $\Phi$ has settled into $V_C$'s
523: minima $\Phi_{\tiny\mbox{min}}$ there are no
524: quantum fluctuations $\delta\Phi$ to
525: be integrated out anymore. Let us show this:
526: Writing $\Phi=|\Phi|\exp\left[i\frac{\theta}{\La_c}\right]$, we have
527: %*******
528: \eqb
529: \label{minimacur}
530: \left.\frac{\pd^2_{\theta} V_C(\Phi)}{|\Phi|^2}\right|_{\Phi_{\tiny\mbox{min}}}=
531: \left.\frac{\pd^2_{|\Phi|} V_C(\Phi)}
532: {|\Phi|^2}\right|_{\Phi_{\tiny\mbox{min}}}
533: =\left\{\begin{array}{c}8\,\ \ \ \ \ (\mbox{SU(2)})\\
534: 18\,\ \ \ \ \ (\mbox{SU(3)})\end{array}\right.\,.
535: \eqe
536: %*******
537: Thus a potential fluctuation $\delta\Phi$ would
538: be harder than the maximal resolution $|\Phi_{\tiny\mbox{min}}|$ corresponding to
539: the effective action Eq.\,(\ref{actionPhi}) that arises after
540: spatial coarse-graining. Thus quantum fluctuations are already
541: contained in the classical configuration $\Phi_{\tiny\mbox{min}}$: The cosmological constant
542: in the confining phase of an SU(2) or
543: SU(3) Yang-Mills theory vanishes exactly. Again, adding the term
544: $\Delta V_C$ of Eq.\,(\ref{DeltaVC}) to the potentials in Eqs.\,(\ref{2potC}) and
545: (\ref{3potC}) or performing the above multiplicative modification
546: does not lower the value for the curvature as
547: obtained in Eq.\,(\ref{minimacur}) and therefore
548: does not change this result.\vspace{0.3cm}\\
549: \noindent{\sl Estimate for density of states, Hagedorn nature of the transition.}
550: That the transition from the confining to the preconfining phase
551: is of the Hagedorn nature is shown by an estimate for
552: the density of massive spin-1/2 states.
553: The multiplicity of massive fermion states, associated with center-vortex loops possessing
554: $n$ self-intersections, is given by twice the number $L_n$
555: of bubble diagrams with $n$ vertices in a scalar $\lambda \phi^4$ theory.
556: In \cite{BenderWu1969} the minimal number of such diagrams $L_{n,min}$ was
557: estimated to be
558: %******
559: \eqb
560: \label{NOD}
561: L_{n,min}=n!3^{-n}\,.
562: \eqe
563: %******
564: The mass spectrum is equidistant. That is,
565: the mass $m_n$ of a state with $n$ self-intersections of
566: the center-vortex loop is $m_n\sim n\,\La_C$. If we only
567: ask for an estimate of the density of {\sl static}
568: fermion states $\rho_{n,0}=\tilde{\rho}(E=n\La_C)$ of energy
569: $E$ then, by appealing to Eq.\,(\ref{NOD}) and Stirling's formula,
570: we obtain \cite{Hofmann2005}
571: %******
572: \eab
573: \label{statdes}
574: \rho_{n,0}&>&\frac{\sqrt{8\pi}}{3\La_C}\,\exp[n\log n]\Big(\log n+1\Big)\,\ \ \ \ \ \mbox{or}\nonumber\\
575: \tilde{\rho}(E)&>&\frac{\sqrt{8\pi}}{3\La_C}\exp[\frac{E}{\La_C}\log\frac{E}{\La_C}]
576: \Big(\log\frac{E}{\La_C}+1\Big)\,.
577: \eae
578: %*******
579: Eq.\,(\ref{statdes}) tells us that the density of static fermion
580: states is more than exponentially increasing with energy $E$. The partition function
581: $Z_{\Phi}$ for the system of static fermions thus is estimated as
582: %******
583: \eab
584: \label{partfunctionphi}
585: Z_{\Phi}&>&\int_{E^*}^\infty dE\,\tilde{\rho}(E)\,n_F(\beta E)\nonumber\\
586: &>&\frac{\sqrt{8\pi}}{3\La_C}\,\int_{E^*}^\infty dE\,\exp\left[\frac{E}{\La_C}\right]\,
587: \exp[-\beta E]\,,
588: \eae
589: %*******
590: where $E^*\gg \Lambda_C$ is the energy where we start to trust
591: our approximations. Thus $Z_{\Phi}$ diverges at some
592: temperature $T_H<\Lambda_C$. Due to the logarithmic factor in the exponent
593: arising in estimate Eq.\,(\ref{statdes}) for $\tilde{\rho}(E)$ we
594: would naively conclude that $T_H=0$. This, however, is an artefact
595: of our assumption that all states with $n$ self-intersections
596: are infinitely narrow. Due to the existence of contact interactions between
597: vortex lines and intersection points this assumption is
598: the less reliable the higher the total energy of a
599: given fluctuation. (A fluctuation of large energy has a higher
600: density of intersection points and vortex lines and
601: thus a larger likelihood for the occurrence of contact interactions which mediate the
602: decay or the recombination of a given state with $n$ self-intersections.)
603: At the temperature $T_H$ the entropy
604: wins over the Boltzmann suppression in energy, and the partition
605: function diverges. To reach the point $T_H$ one would, in a spatially homogeneous way,
606: need to invest an infinite amount of energy into the system
607: which is impossible. By an (externally induced) violation of
608: spatial homogeneity and thus by a sacrifice of thermal
609: equilibrium the system may, however, condense densly packed (massless)
610: vortex intersection points into a new ground state. The latter's excitations exhibit a
611: power-like density of states and thus are described by a
612: finite partition function. This is the celebrated Hagedorn
613: transition from below.\vspace{0.3cm}\\
614: \noindent{\sl Summary in view of particle physics and cosmology.} The confining
615: phase of an SU(2) and SU(3) pure Yang-Mills theory is characterized
616: by a condensate of single center-vortex loops and a dynamically broken, local magnetic
617: $Z_2$ (SU(2)) and $Z_3$ (SU(3)) symmetry: No massless or finite-mass gauge bosons exist.
618: Single center-vortex loops
619: emerge as massless spin-1/2 particles due to
620: the decay of a monopole condensate. A fraction of zero-momentum,
621: single center-vortex loops subsequently condenses by the formation
622: of Cooper-like pairs. Protected from radiative corrections, the
623: energy density and the pressure of this condensate is
624: precisely zero in a thermally equilibrated situation. The spectrum of particle
625: excitations is a tower of spin-1/2 states with
626: equidistant mass levels. A massive state emerges by twisting
627: a single center-vortex loop hence generating
628: self-intersection point(s). This takes place when
629: single center-vortex loops collide. The process of mass generation thus
630: is facilitated by converting (some of) the kinetic energy of a single center-vortex loop
631: into the (unresolvable) dynamics of a flux-eddy
632: marking the self-intersection point, see Fig.\,(\ref{Fig-0}).
633: Due to their over-exponentially increasing multiplicity
634: heavy states become instable by the contact interactions facilitated by
635: dense packing. In a spatially extended
636: system (such as the overlap region for two colliding, heavy, and
637: ultrarelativistic nuclei) there is a finite value
638: in temperature, comparable to the Yang-Mills scale $\Lambda_C$,
639: where a given, spatially nonhomogeneous
640: perturbation induces the condensation of vortex intersections.
641: This is the celebrated (nonthermal) Hagedorn transition.
642:
643: The existence of a Hagedorn-like density of states explains
644: why in an isolated system, governed by a single SU(2) Yang-Mills theory,
645: the center-flux eddy in a spin-1/2 state with a single self-intersection appears to be
646: structureless for external probes of all momenta
647: with one exception: If the externally supplied resolution is
648: comparable to the Yang-Mills scale $\Lambda_C$, that is, close
649: to the first radial excitation level of a BPS monopole
650: \cite{ForgasVolkov2003} then the possibility of converting the
651: invested energy into the entropy associated with
652: the excitation of a large number of instable and
653: heavy resonances does not yet exist. As a consequence, the center
654: of the flux eddy -- a BPS monopole -- is excited itself and
655: therefore reveals part of its structure. For an externally supplied
656: resolution, which is considerable below $\Lambda_C$, there is nothing to
657: be excited in a BPS monopole \cite{ForgasVolkov2003} and thus the object
658: appears to be structureless as well.
659:
660: There is experminental evidence \cite{Alvensleben1968,Ashkin1953,Scott1951} that this situation
661: applies to charged leptons being the spin-1/2 states with a single self-intersection of
662: SU(2) Yang-Mills theories with scales comparable with the associated
663: lepton masses \cite{Hofmann2005}. The corresponding neutrinos are
664: Majorana particles (single center-vortex loops) which is also supported
665: by experiment \cite{Klapdor2004}. The weak symmetry SU(2)$_W$ of the Standard Model (SM) is
666: identified with SU(2)$_e$ where the subscript $e$ refers to
667: the electron. The important difference compared with the SM is
668: that the pure SU(2)$_e$ gauge theory {\sl by itself provides} for a nonperturbative
669: breakdown of its continuous gauge symmetry in
670: two stages (deconfining and preconfining phase)
671: and, in addition, generates the electron neutrino and the electron
672: as the only stable and apparently structureless solitons in its confining
673: phase: No additional, fundamentally charged, and fluctuating Higgs
674: field is needed to break the weak gauge symmetry. The confining phase is associated with a
675: discrete gauge symmetry -- the magnetic center symmetry -- being
676: dynamically broken.
677:
678: As far as the cosmological-constant problem is concerned the state of affairs
679: is not as clear-cut as it may seem. Even though each pure SU(2) or
680: SU(3) gauge theory does not generate a contribution to the vacuum
681: energy in its confining phase one needs to include
682: gravity, the dynamical mixing of various
683: gauge-symmetry factors, and the anomalies of
684: emerging global symmetries in the analysis
685: to obtain the complete picture on
686: the Universe's present ground state. We hope to be able to pursue
687: this program in the near future. Notice that today's ground-state
688: contribution due to an SU(2) Yang-Mills
689: theory of scale comparable to the present temperature of the
690: cosmic microwave background is small as
691: compared to the measured value \cite{Hofmann20054}. This SU(2) Yang-Mills
692: theory masquerades as the U(1)$_Y$ factor of the Standard Model
693: within the present cosmological epoch.
694:
695:
696: \section*{Acknowledgements}
697: The author would like to thank Mark Wise for useful conversations
698: and for the warm hospitality extended to him during his
699: visit to Caltech in May 2005. Financial support by
700: Kavli Institute at Santa Barbara and by
701: the physics department of UCLA is thankfully acknowledged.
702:
703:
704: \begin{thebibliography}{10}
705:
706: \bibitem{Hofmann2005}
707: R. Hofmann, Int. J. Mod. Phys. A {\bf 20}, 4123 (2005).
708:
709: \bibitem{HerbstHofmannRohrer2004}
710: U. Herbst, R. Hofmann, and J. Rohrer, Acta Phys. Pol. B {\bf 36}, 881 (2005).
711:
712: \bibitem{HerbstHofmann2004}
713: U. Herbst and R. Hofmann, hep-th/0411214.
714:
715: \bibitem{MorningstarPeardon1999}
716: C. J. Morningstar and M. J. Peardon, Phys. Rev. D {\bf 60}, 034509 (1999).
717:
718: \bibitem{Narison1984}
719: S. Narison, Z. Phys. C {\bf 26}, 209 (1984).
720:
721: \bibitem{ShifmanVainshteinZakharov1978}
722: M. A. Shifman, A. I. Vainshtein, V. I. Zakharov, Nucl. Phys. B {\bf 147}, 385 (1979).\\
723: M. A. Shifman, A. I. Vainshtein, V. I. Zakharov, Nucl.Phys. B {\bf 147}, 448 (1979).\\
724: M. A. Shifman, A. I. Vainshtein, V. I. Zakharov, Nucl. Phys. B {\bf 147}, 519 (1979).
725:
726: \bibitem{Laughlin1983}
727: R. B. Laughlin, Phys. Rev. Lett. {\bf 50}, 1395 (1983).
728:
729: \bibitem{TsuiStoermerGossard1982}
730: D. C. Tsui, H. L. St\"ormer, and A. C. Gossard, Phys. Rev. Lett. {\bf 48}, 1559 (1982).
731:
732: \bibitem{Hagedorn1965}
733: R. Hagedorn, Nuovo Cim. Suppl. {\bf 3}, 147 (1965).\\
734: R. Hagedorn and J. Ranft, Nuovo Cim. Suppl. {\bf 6}, 169 (1968).
735:
736: \bibitem{Veneziano1968}
737: G. Veneziano, Nuovo Cim. A {\bf 57}, 190 (1968).
738:
739: \bibitem{Bielfeld19961999}
740: G. Boyd {\sl et} al., Phys. Rev. Lett. {\bf 75}, 4169 (1995).\\
741: G. Boyd {\sl et} al., Nucl. Phys. {\bf B469}, 419 (1996).\\
742: J. Engels, F. Karsch, T. Scheideler, Nucl. Phys. B {\bf 564}, 302 (1999).
743:
744: \bibitem{LuciniTeperWenger}
745: B. Lucini, M. Teper, and U. Wenger, JHEP {\bf 0401}, 061 (2004).\\
746: B. Lucini, M. Teper, and U. Wenger, JHEP {\bf 0502}, 033 (2005).
747:
748: \bibitem{NielsenOlesen1973}
749: H. B. Nielsen and P. Olesen, Nucl. Phys. B {\bf 61}, 45 (1973).
750:
751: \bibitem{Hofmann20052}
752: R. Hofmann, hep-th/0507122.
753:
754: \bibitem{Reinhardt2001}
755: H. Reinhardt, Nucl. Phys. B {\bf 628}, 133 (2002).
756:
757: \bibitem{BenderWu1969}
758: C. M. Bender and T. T. Wu, Phys. Rev. {\bf 184}, 1231 (1969).
759:
760: \bibitem{ForgasVolkov2003}
761: P. Forgacs and M. Volkov, Phys. Rev. Lett. {\bf 92}, 151802.
762:
763: \bibitem{Alvensleben1968}
764: H. Alvensleben {\sl et al.}, Phys. Rev. Lett. {\bf 21}, 1501 (1968).
765:
766: \bibitem{Ashkin1953}
767: A. Ashkin, L. A. Page, and W. M. Woodard, Phys. Rev. {\bf 94}, 357 (1953).
768:
769: \bibitem{Scott1951}
770: M. B. Scott, A. O. Hanson, and E. M. Lyman, Phys. Rev. {\bf 84}, 638 (1951).
771:
772: \bibitem{Klapdor2004}
773: H. V. Klapdor-Kleingrothaus, I. V. Krivosheina, A. Dietz, and
774: O. Chkvorets, Phys. Lett. B {\bf 586}, 198 (2004).
775:
776: \bibitem{Hofmann20054}
777: R. Hofmann, hep-ph/0508176.
778:
779: \end{thebibliography}
780:
781: \baselineskip25pt
782: \end{document}
783: