1: \documentstyle[12pt,epsfig,graphics,cite]{article}\pagestyle{myheadings}
2: \textheight=23.5cm\topmargin=-0.6cm\textwidth=16cm
3: \oddsidemargin-0.1cm\evensidemargin-0.1cm\sloppy\frenchspacing\flushbottom
4: \renewcommand{\topfraction}{1}\renewcommand{\bottomfraction}{1}
5: \renewcommand{\textfraction}{0}\renewcommand{\floatpagefraction}{0}
6: \begin{document}\bibliographystyle{plain}\begin{titlepage}
7: \renewcommand{\thefootnote}{\fnsymbol{footnote}}\hfill\begin{tabular}{l}
8: HEPHY-PUB 812/05\\hep-ph/0510372\\October 2005\end{tabular}\\[1cm]
9: \Large\begin{center}{\bf EXACT-PROPAGATOR INSTANTANEOUS BETHE--SALPETER
10: EQUATION FOR QUARK--ANTIQUARK BOUND STATES}\\[1.5cm]\large{\bf LI
11: Zhi-Feng}\\[.3cm]\normalsize Institut f\"ur Theoretische Physik,
12: Universit\"at Wien,\\Boltzmanngasse 5, A-1090 Wien, Austria\\[1cm]\large{\bf
13: Wolfgang LUCHA\footnote[1]{\normalsize\ {\em E-mail address\/}:
14: wolfgang.lucha@oeaw.ac.at}}\\[.3cm]\normalsize Institut f\"ur
15: Hochenergiephysik,\\\"Osterreichische Akademie der
16: Wissenschaften,\\Nikolsdorfergasse 18, A-1050 Wien, Austria\\[1cm]\large{\bf
17: Franz F.~SCH\"OBERL\footnote[2]{\normalsize\ {\em E-mail address\/}:
18: franz.schoeberl@univie.ac.at}}\\[.3cm]\normalsize Institut f\"ur Theoretische
19: Physik, Universit\"at Wien,\\Boltzmanngasse 5, A-1090 Wien,
20: Austria\vfill{\normalsize\bf Abstract}\end{center}\normalsize Recently an
21: instantaneous approximation to the Bethe--Salpeter formalism for the analysis
22: of bound states in quantum field theory has been proposed which retains, in
23: contrast to the Salpeter equation, as far as possible the exact propagators
24: of the bound-state constituents, extracted nonperturbatively from
25: Dyson--Schwinger equations or lattice gauge theory. The implications of this
26: improvement for the solutions of this bound-state equation, that is, the
27: spectrum of the mass eigenvalues of its bound states and the corresponding
28: wave~functions, when considering the quark propagators arising in quantum
29: chromodynamics are explored.\vspace{.5cm} {\em PACS numbers\/}: 11.10.St,
30: 03.65.Ge, 03.65.Pm
31: \renewcommand{\thefootnote}{\arabic{footnote}}\end{titlepage}
32:
33: \section{Introduction}The more than half a century old Bethe--Salpeter
34: formalism\cite{BSE} constitutes a relativistically covariant framework within
35: the realms of quantum field theory for the description of bound states from
36: first principles. The Bethe--Salpeter equation controls a bound-state
37: amplitude encoding, in momentum space, the distribution of the relative
38: momenta of the bound-state constituents. Within elementary particle physics
39: this formalism has been widely applied to quantum electrodynamics (QED) and
40: quantum chromodynamics (QCD). Unfortunately it faces problems of
41: interpretation and of our ignorance of the full interaction kernel~in~QCD.
42:
43: Thus, simplifications of the Bethe--Salpeter equation in form of some
44: three-dimensional reductions are highly desirable. The most famous among all
45: proposals is known as Salpeter equation\cite{SE}. Its formulation, however,
46: is based on assuming all bound-state constituents to interact instantaneously
47: {\em and\/} to propagate as free particles; the latter circumstance~renders
48: hard to implement effects such as spontaneous chiral-symmetry breaking,
49: crucial for QCD.
50:
51: In view of this, an instantaneous Bethe--Salpeter equation which incorporates
52: the exact form of the propagators of the bound-state constituents (to the
53: utmost conceivable extent) has been derived recently\cite{Lucha05:IBSEWEP};
54: this improved bound-state equation reduces, of course, to the Salpeter
55: equation, upon approximation of the exact propagators by their free
56: counterparts. For any description of hadrons as bound states, the exact quark
57: propagators conforming to the QCD Dyson--Schwinger equations are relevant.
58: This work is devoted to the study~of~the consequences of introducing exact
59: quark propagators in this instantaneous~Bethe--Salpeter equation. The most
60: dramatic effect observed is a significant diminution of the level spacing.
61:
62: The paper is organized as follows. Section~\ref{Sec:IBSEWEP} sketches the
63: derivation of our instantaneous Bethe--Salpeter equation with exact
64: propagators of the bound-state constituents presented in
65: Ref.\cite{Lucha05:IBSEWEP}. Approximating all interactions entering in the
66: Bethe--Salpeter equation by their static forms but retaining, as far as
67: possible, exact propagators yields\cite{Lucha05:IBSEWEP} a generalization of
68: Salpeter's equation\cite{SE}, with momentum-dependent masses of the
69: bound-state constituents and with normalization factors of their exact
70: propagators multiplying all interaction terms. Section~\ref{Sec:QP}
71: introduces the exact light-quark propagators obtained within QCD as solution
72: of the Dyson--Schwinger equations. This infinite tower of coupled integral
73: equations calls for a truncation\cite{Maris97a,Maris97b,Roberts98,Ivanov98,
74: Maris99a,Maris99b,Roberts00a,Roberts00b,Alkofer00,Maris00,Maris01,Maris02,
75: Bhagwat02,Tandy03,Maris03,Bhagwat03,Roberts03,Krassnigg03a,Krassnigg03b,
76: Alkofer03a,Krassnigg04} which must not be in conflict with the relevant
77: Ward--Takahashi identity. Section~\ref{Sec:PSFAFBS} summarizes the technique
78: developed for finding the solutions of an instantaneous Bethe--Salpeter
79: equation by first reducing it to a coupled set of radial
80: equations\cite{Lagae92a,Lagae92b,Olsson95,Olsson96}~and then converting it to
81: a matrix eigenvalue problem\cite{Lucha00:IBSEm=0,Lucha00:IBSE-C4,
82: Lucha00:IBSEnzm,Lucha01:IBSEIAS}. Some implications of taking into account
83: exact instead of free propagators of bound-state constituents are
84: analyzed~in~Sec.~\ref{Sec:RD} by application of the entire formalism to a
85: linear confining interaction. Section~\ref{Sec:SCO} scrutinizes our findings.
86: Appendix~\ref{App:GLB} recalls the Hilbert-space basis required for the
87: matrix conversion.
88:
89: \section{Instantaneous Bethe--Salpeter equation with exact propagators}
90: \label{Sec:IBSEWEP}The derivation of the {\em exact-propagator\/}
91: instantaneous Bethe--Salpeter equation\cite{Lucha05:IBSEWEP} parallels the
92: (three-dimensional) reduction of the Bethe--Salpeter equation\cite{BSE} to
93: the {\em free-propagator\/} Salpeter equation\cite{SE}. It may be achieved by
94: several slightly different but equivalent routes.
95:
96: In the framework of the Bethe--Salpeter formalism, a bound state $|{\rm
97: B}(P)\rangle$ of momentum $P$ and mass $M_{\rm B},$ composed of a fermion and
98: an antifermion described by the field operators $\psi_1(x_1),$
99: $\bar\psi_2(x_2),$ resp., is represented, in momentum space, by the
100: Bethe--Salpeter amplitude$$\Psi(p)\equiv\exp({\rm i}\,P\,X)\int{\rm
101: d}^4x\,\exp({\rm i}\,p\,x)\,\langle 0|{\rm T}(\psi_1(x_1)\,\bar\psi_2(x_2))
102: |{\rm B}(P)\rangle\ .$$Here, $X$ and $x$ denote the center-of-momentum and
103: relative coordinates of the two-particle system while $P$ and $p$ label the
104: total and relative momenta of the bound-state constituents. This
105: Bethe--Salpeter amplitude $\Psi$ has to satisfy the homogeneous
106: Bethe--Salpeter equation\begin{equation}\Psi(p)=\frac{{\rm
107: i}}{(2\pi)^4}\,S_1(p_1)\int{\rm d}^4q\,K(p,q)\,\Psi(q)\,S_2(-p_2)\
108: .\label{Eq:BSE}\end{equation}The dynamical ingredients of this equation of
109: motion are the exact propagators $S_i(p)$~of~the two bound fermions $i=1,2$
110: (with individual momenta $p_1,$ $p_2$) and the interaction kernel $K,$ a
111: fully truncated 4-point Green function which encompasses all Bethe--Salpeter
112: irreducible Feynman diagrams for two-particle into two-particle scattering
113: and depends on the relative momenta of initial and final scattering states,
114: $p$ and $q,$ as well as on the total momentum~$P$.
115:
116: The instantaneous approximation to this formalism assumes that the kernel $K$
117: depends just on the spatial components $\mbox{\boldmath{$p$}}$ and
118: $\mbox{\boldmath{$q$}}$ of the relative momenta $p$ and $q$: $K(p,q)=\hat
119: K(\mbox{\boldmath{$p$}},\mbox{\boldmath{$q$}}).$ Its application reduces
120: Eq.~(\ref{Eq:BSE}) to the instantaneous version of the Bethe--Salpeter
121: equation\begin{equation}\Phi(\mbox{\boldmath{$p$}})=\frac{{\rm i}}{2\pi}
122: \int{\rm d}p_0\,S_1(p_1)\,I(\mbox{\boldmath{$p$}})\,S_2(-p_2)\label{Eq:BSE-I}
123: \end{equation}for the Salpeter amplitude (defined by integration of $\Psi(p)$
124: over the time component $p_0$~of~$p$)$$\Phi(\mbox{\boldmath{$p$}})\equiv
125: \frac{1}{2\pi}\int{\rm d}p_0\,\Psi(p)\ ;$$ here the term involving the by
126: assumption now instantaneous interaction is abbreviated~by
127: $$I(\mbox{\boldmath{$p$}})\equiv\frac{1}{(2\pi)^3}\int{\rm d}^3q\,\hat
128: K(\mbox{\boldmath{$p$}},\mbox{\boldmath{$q$}})\, \Phi(\mbox{\boldmath{$q$}})\
129: .$$
130:
131: The fermion propagator $S_i(p)$ is the solution of the fermion
132: Dyson--Schwinger equation. By Lorentz covariance $S_i(p)$ is defined by
133: merely two (Lorentz-scalar) functions $M_i(p^2)$~and $Z_i(p^2),$ in QCD
134: referred to as the quark wave-function renormalization and mass functions:
135: $$S_i(p)=\frac{{\rm i}\,Z_i(p^2)}{\not\!p-M_i(p^2)+{\rm i}\,\varepsilon}\
136: ,\quad\not\!p\equiv p^\mu\,\gamma_\mu\ .$$In the course of the derivation
137: \cite{Lucha05:IBSEWEP} of a generalization of the Salpeter equation
138: towards~exact propagators of all bound-state constituents, two of the present
139: authors (W.~L. and~F.~F.~S.) assumed these functions, $M_i(p^2)$ and
140: $Z_i(p^2),$ to depend only on the spatial components $\mbox{\boldmath{$p$}}$
141: of the momentum $p.$ This allows to substitute $M_i(p^2)$ by
142: $M_i(\mbox{\boldmath{$p$}}^2)$ and $Z_i(p^2)$ by
143: $Z_i(\mbox{\boldmath{$p$}}^2)$~in~$S_i(p).$
144:
145: Then the integral in Eq.~(\ref{Eq:BSE-I}) over the time component $p_0$ can
146: be easily given analytically. Introducing the one-particle energy
147: $E_i(\mbox{\boldmath{$p$}}),$ the (generalized) Dirac Hamiltonian
148: $H_i(\mbox{\boldmath{$p$}}),$ and the energy projection operators
149: $\Lambda_i^\pm(\mbox{\boldmath{$p$}})$ for positive or negative energy of
150: particle $i=1,2$~by\begin{eqnarray*}E_i(\mbox{\boldmath{$p$}})&\equiv&
151: \sqrt{\mbox{\boldmath{$p$}}^2+M_i^2(\mbox{\boldmath{$p$}}^2)}\ ,\quad i=1,2\
152: ,\\[1ex]H_i(\mbox{\boldmath{$p$}})&\equiv&
153: \gamma_0\,[\mbox{\boldmath{$\gamma$}}\cdot\mbox{\boldmath{$p$}}
154: +M_i(\mbox{\boldmath{$p$}}^2)]\ ,\quad i=1,2\ ,\\[1ex]
155: \Lambda_i^\pm(\mbox{\boldmath{$p$}})&\equiv&\frac{E_i(\mbox{\boldmath{$p$}})\pm
156: H_i(\mbox{\boldmath{$p$}})}{2\,E_i(\mbox{\boldmath{$p$}})}\ ,\quad i=1,2\
157: ,\end{eqnarray*}few rather standard manipulations yield\cite{Lucha05:IBSEWEP}
158: our instantaneous Bethe--Salpeter equation for fermion--antifermion bound
159: states, with exact propagators of the bound-state
160: constituents:\begin{equation}\Phi(\mbox{\boldmath{$p$}})
161: =Z_1(\mbox{\boldmath{$p$}}_1^2)\,Z_2(\mbox{\boldmath{$p$}}_2^2)
162: \left(\frac{\Lambda_1^+(\mbox{\boldmath{$p$}}_1)\,\gamma_0\,
163: I(\mbox{\boldmath{$p$}})\,\gamma_0\,\Lambda_2^-(\mbox{\boldmath{$p$}}_2)}
164: {P_0-E_1(\mbox{\boldmath{$p$}}_1)-E_2(\mbox{\boldmath{$p$}}_2)}
165: -\frac{\Lambda_1^-(\mbox{\boldmath{$p$}}_1)\,\gamma_0\,
166: I(\mbox{\boldmath{$p$}})\,\gamma_0\,\Lambda_2^+(\mbox{\boldmath{$p$}}_2)}
167: {P_0+E_1(\mbox{\boldmath{$p$}}_1)+E_2(\mbox{\boldmath{$p$}}_2)}\right).
168: \label{Eq:IBSE}\end{equation}From this, each amplitude
169: $\Phi(\mbox{\boldmath{$p$}})$ has to satisfy the two constraints
170: $\Lambda_1^\pm(\mbox{\boldmath{$p$}}_1)\,\Phi(\mbox{\boldmath{$p$}})\,
171: \Lambda_2^\pm(\mbox{\boldmath{$p$}}_2)=0.$
172:
173: With little effort our bound-state equation (\ref{Eq:IBSE}) may be rephrased
174: as eigenvalue problem,
175: \begin{eqnarray*}&&H_1(\mbox{\boldmath{$p$}}_1)\,\Phi(\mbox{\boldmath{$p$}})
176: -\Phi(\mbox{\boldmath{$p$}})\,H_2(\mbox{\boldmath{$p$}}_2)\\[1ex]
177: &&+\,Z_1(\mbox{\boldmath{$p$}}_1^2)\,Z_2(\mbox{\boldmath{$p$}}_2^2)\,
178: [\Lambda_1^+(\mbox{\boldmath{$p$}}_1)\,\gamma_0\,I(\mbox{\boldmath{$p$}})\,
179: \gamma_0\,\Lambda_2^-(\mbox{\boldmath{$p$}}_2)
180: -\Lambda_1^-(\mbox{\boldmath{$p$}}_1)\,\gamma_0\,I(\mbox{\boldmath{$p$}})\,
181: \gamma_0\,\Lambda_2^+(\mbox{\boldmath{$p$}}_2)]\\[1ex]
182: &&=P_0\,\Phi(\mbox{\boldmath{$p$}})\ ,\end{eqnarray*}with the bound state's
183: energy $P_0$ or, in its rest frame
184: $\mbox{\boldmath{$p$}}_2=-\mbox{\boldmath{$p$}}_1,$ the mass $M_{\rm B}$ as
185: eigenvalue.
186:
187: The Salpeter equation\cite{SE} is obtained by one further step of
188: simplification. Its derivation assumes, in addition to the instantaneous
189: approximation for $K$, that each exact propagator in Eq.~(\ref{Eq:BSE}) can
190: be replaced by the propagator $S_0(p,m_i)$ of a free particle of effective
191: mass~$m_i$:$$S_i(p)\cong S_0(p,m_i)=\frac{{\rm i}}{\not\!p-m_i+{\rm
192: i}\,\varepsilon}\equiv{\rm i}\,\frac{\not\!p+m_i}{p^2-m_i^2+{\rm
193: i}\,\varepsilon}\ ,\quad i=1,2\ .$$Thus the Salpeter equation is recovered
194: from Eq.~(\ref{Eq:IBSE}) in the limit $M_i(p^2)\to m_i,$ $Z_i(p^2)\to 1.$ The
195: exact-propagator instantaneous Bethe--Salpeter equation (\ref{Eq:IBSE})
196: generalizes\cite{Lucha05:IBSEWEP} Salpeter's equation by replacing $m_i$ by
197: $M_i(p^2)$ and introducing factors $Z_i(p^2)$ in the interaction terms.
198:
199: \section{Quark propagator from Dyson--Schwinger equation}\label{Sec:QP}Within
200: QCD the Dyson--Schwinger equation for the quark propagator involves, besides
201: the exact quark propagator, also the exact gluon propagator and the exact
202: quark--gluon vertex. The Dyson--Schwinger equations governing the two latter
203: Green functions couple the quark Dyson--Schwinger equation to the infinite
204: hierarchy of Dyson--Schwinger equations.~Hence, a tractable problem can only
205: be defined by some truncation of this set of integral equations.
206:
207: For the present investigation we employ the so-called
208: ``renormalization-group-improved rainbow--ladder truncation'' scheme
209: \cite{Maris97a,Maris97b,Roberts98,Ivanov98,Maris99a,Maris99b,Roberts00a,
210: Roberts00b,Alkofer00,Maris00,Maris01,Maris02,Bhagwat02,Tandy03,Maris03,
211: Bhagwat03,Roberts03,Krassnigg03a,Krassnigg03b,Alkofer03a,Krassnigg04} applied
212: to the quark Dyson--Schwinger equation and the meson Bethe--Salpeter
213: equation. In this specific scheme the exact gluon propagator and the exact
214: quark--gluon vertex are replaced by their (perturbative) tree-level
215: forms.~The truncation is consistent with the preservation of the axial-vector
216: Ward--Takahashi identity; this is important for all questions related to
217: chiral symmetry and its dynamical breakdown. In this model, all the dynamical
218: information is encoded in some effective coupling strength.
219:
220: Viewed as function of the involved momentum transfer squared this effective
221: coupling is characterized by two main features. In the ultraviolet region it
222: approaches the perturbative behaviour of the strong fine-structure constant,
223: incorporating thereby asymptotic freedom. In the infrared region it exhibits
224: the significant enhancement demanded strongly by studies of the
225: Dyson--Schwinger equation satisfied by the exact gluon propagator. In the
226: particular {\em Ansatz\/} for this effective coupling strength proposed in
227: Ref.\cite{Maris97a} this infrared enhancement is represented partly by the
228: integrable singularity of a momentum-space $\delta$ function, partly by a
229: finite-width approximation to this $\delta$ function. This constitutes the
230: ``model of our~choice.''
231:
232: \newpage
233:
234: In the comprehensive study presented in Ref.\cite{Maris97a}, the exact quark
235: propagators emerging from this truncation model are found as numerical
236: solutions of the quark Dyson--Schwinger equation by fitting main properties
237: of the $\pi$- and {\sl K}-meson system. Like many treatments of
238: Dyson--Schwinger equations the analysis of Ref.\cite{Maris97a} has been
239: performed in Euclidean space, implying that the quark propagators are
240: obtained as Euclidean-space Schwinger functions. Within both QED and QCD, the
241: analytic structure of the exact fermion propagators is still the subject of
242: intense investigations which did not provide a definitive conclusion until
243: now (cf., for instance, Refs.\cite{Maris94,Alkofer03a,Alkofer03b} and
244: references therein). In order to proceed, we must thus assume that the
245: necessary analytic continuation from Euclidean to Minkowski space makes
246: sense, at least for the quark propagators. In this case, the numerically
247: computed functions $M(p^2)$ and $Z(p^2)$ in the propagator of the light {\sl
248: u}- and {\sl d}-quarks may be represented (rather accurately in a range of
249: spacelike momenta) in analytical form by the parametrizations
250: \cite{Roberts02:PC}$$M(p^2)=\frac{a}{1+\displaystyle\frac{p^4}{b}}+m_0\
251: ,\quad Z(p^2)=1-\frac{c}{1-\displaystyle\frac{p^2}{d}}\ ;$$the values of the
252: parameters $a,b,c,d,m_0$ are fixed by interpolation of the numerical
253: results:\begin{eqnarray}&&a=0.745\;{\rm GeV}\ ,\quad b=(0.744\;{\rm GeV})^4\
254: ,\quad m_0=0.0055\;{\rm GeV}\ ,\nonumber\\[1ex]&&c=0.545\ ,\quad
255: d=(1.85508\;{\rm GeV})^2\ .\label{Eq:ParVal}\end{eqnarray}These propagator
256: functions read in their ``$p_0^2=0$'' approximation required for an
257: analytical formulation of the bound-state equation (\ref{Eq:IBSE}) proposed
258: \cite{Lucha05:IBSEWEP} as a generalized Salpeter equation\begin{equation}
259: M(\mbox{\boldmath{$p$}}^2)=
260: \frac{a}{1+\displaystyle\frac{\mbox{\boldmath{$p$}}^4}{b}}+m_0\ ,\quad
261: Z(\mbox{\boldmath{$p$}}^2)=1-
262: \frac{c}{1+\displaystyle\frac{\mbox{\boldmath{$p$}}^2}{d}}\
263: .\label{Eq:ProPar}\end{equation}
264:
265: The behaviour of the quark propagator functions $M(\mbox{\boldmath{$p$}}^2)$
266: and $Z(\mbox{\boldmath{$p$}}^2)$ as functions of $\mbox{\boldmath{$p$}}^2$ is
267: depicted in Fig.~\ref{Fig:PF}. For light quarks the mass function
268: $M(\mbox{\boldmath{$p$}}^2)$ is, of course, dominated by the nonperturbative
269: mechanism responsible for dynamical chiral-symmetry breaking.~Starting at
270: $M(0)=0.7505\;{\rm GeV},$ $M(\mbox{\boldmath{$p$}}^2)$ drops in the vicinity
271: of $\mbox{\boldmath{$p$}}^2=(0.57\;{\rm GeV})^2$ by more than two orders of
272: magnitude, in order to approach in the limit
273: $\mbox{\boldmath{$p$}}^2\to\infty$ the (comparatively~tiny but still
274: nonvanishing and hence explicitly chiral-symmetry breaking) current
275: light-quark mass $m_0=0.0055\;{\rm GeV}.$ In contrast to such drastic
276: variation, the wave-function renormalization function
277: $Z(\mbox{\boldmath{$p$}}^2)$ exhibits an only rather moderate dependence on
278: $\mbox{\boldmath{$p$}}^2.$ With increasing values of
279: $\mbox{\boldmath{$p$}}^2,$ $Z(\mbox{\boldmath{$p$}}^2)$ rises slowly from
280: $Z(0)=0.455$ to its asymptotic value $Z(\mbox{\boldmath{$p$}}^2)\to1$ for
281: $\mbox{\boldmath{$p$}}^2\to\infty.$
282:
283: Within the ``renormalization-group-improved rainbow--ladder truncation''
284: scheme, it is by no means mandatory to implement, as done in the model
285: studied in Refs.\cite{Maris97a,Maris97b,Roberts98,Ivanov98,Roberts00a,
286: Roberts00b,Krassnigg04}, the infrared enhancement in the effective
287: interaction by the sum of an integrable $\delta$ function singularity and its
288: finite-width approximation. The results of the investigations reported in
289: Refs.\cite{Maris99a,Maris99b,Maris00,Maris01,Maris02,Bhagwat02,Tandy03,Maris03,
290: Roberts03,Alkofer03a} demonstrate that propagator functions of very similar
291: shape will be obtained in a model in which the infrared enhancement of the
292: effective-interaction coupling required by hadron phenomenology is provided
293: only by the finite-width representation\cite{Alkofer00}.
294:
295: Moreover, the predictions for the propagator functions $M(p^2)$ and $Z(p^2)$
296: of both models\cite{Maris97a,Maris99a} for the effective coupling in the
297: quark Dyson--Schwinger equation exhibit a remarkable qualitative and
298: quantitative agreement with the results produced by lattice gauge theories. A
299: recent {\em unquenched\/} lattice calculation of the quark propagator in
300: Landau gauge involving two degenerate light ({\sl u\/}/{\sl d\/}) and one
301: heavier ({\sl s\/}) dynamical quarks may be found in Ref.\cite{Bowman05}.
302:
303: \begin{figure}[p]\begin{center}\begin{tabular}{c}
304: \psfig{figure=M-Plot.eps,scale=1}\\[0.05ex](a)\end{tabular}\\[1.5ex]
305: \begin{tabular}{c}\psfig{figure=Z-Plot.eps,scale=1}\\[0.05ex](b)
306: \end{tabular}\\[0.5ex]\caption{Mass function $M(\mbox{\boldmath{$p$}}^2)$ (a)
307: and wave-function renormalization function $Z(\mbox{\boldmath{$p$}}^2)$ (b)
308: of the exact propagator of the (light) {\sl u}- and {\sl d}-quarks obtained
309: by numerical solution of the quark Dyson--Schwinger equation in the
310: ``renormalization-group-improved rainbow--ladder truncation'' model of
311: Ref.\cite{Maris97a} as represented by the simple interpolation of
312: Eqs.~(\ref{Eq:ProPar})~and~(\ref{Eq:ParVal}).}\label{Fig:PF}\end{center}
313: \end{figure}
314:
315: \section{Pseudoscalar fermion--antifermion bound states}\label{Sec:PSFAFBS}
316: Now follow the path, paved in Refs.\cite{Lagae92a,Lagae92b,Olsson95,Olsson96,
317: Lucha00:IBSEm=0,Lucha00:IBSE-C4,Lucha00:IBSEnzm,Lucha01:IBSEIAS}, of the
318: transformation of bound-state equations for Salpeter amplitudes
319: $\Phi(\mbox{\boldmath{$p$}})$ to matrix eigenvalue problems fixing their
320: radial components.
321:
322: Consider, as the perhaps simplest example, fermion--antifermion bound states
323: of spin $J,$ parity $P=(-1)^{J+1}$ and charge-conjugation quantum number
324: $C=(-1)^J.$ In spectroscopic notation such states are labeled by ${}^1J_J.$
325: Because of the constraints $\Lambda_1^\pm(\mbox{\boldmath{$p$}}_1)\,
326: \Phi(\mbox{\boldmath{$p$}})\,\Lambda_2^\pm(\mbox{\boldmath{$p$}}_2)=0$ the
327: general expansion of the Salpeter amplitude $\Phi(\mbox{\boldmath{$p$}})$
328: over a complete set of Dirac matrices involves not the full 16 but only eight
329: independent components. For our ${}^1J_J$ states only two of the latter,
330: $\Phi_1(\mbox{\boldmath{$p$}})$ and $\Phi_2(\mbox{\boldmath{$p$}}),$ are
331: relevant. With our notation for one-particle energy
332: $E(\mbox{\boldmath{$p$}})$ and (generalized) Dirac Hamiltonian
333: $H(\mbox{\boldmath{$p$}})$ introduced in Sec.~\ref{Sec:IBSEWEP} the
334: corresponding Salpeter amplitude $\Phi(\mbox{\boldmath{$p$}})$ reads in the
335: center-of-momentum frame of the particle--antiparticle
336: system$$\Phi(\mbox{\boldmath{$p$}})=\left[\Phi_1(\mbox{\boldmath{$p$}})\,
337: \frac{H(\mbox{\boldmath{$p$}})}{E(\mbox{\boldmath{$p$}})}
338: +\Phi_2(\mbox{\boldmath{$p$}})\right]\gamma_5\ .$$Without loss of generality
339: but for definiteness, focus to exactly the same physical system as studied in
340: Refs.\cite{Lucha00:IBSEm=0,Lucha00:IBSE-C4,Lucha00:IBSEnzm,Lucha01:IBSEIAS}.
341: Aiming at the description of mesons with the quantum numbers of the pion
342: discuss quark--antiquark bound states of spin $J=0,$ that is, pseudoscalar
343: states~of spin-parity-charge conjugation assignment $J^{PC}=0^{-+}.$ Assume
344: the kernel $\hat K(\mbox{\boldmath{$p$}},\mbox{\boldmath{$q$}})$ to be of
345: convolution type with Dirac structure that of time-component Lorentz-vector
346: interactions: $\hat K(\mbox{\boldmath{$p$}},\mbox{\boldmath{$q$}})=\hat
347: K(\mbox{\boldmath{$p$}}-\mbox{\boldmath{$q$}})=
348: V(\mbox{\boldmath{$p$}}-\mbox{\boldmath{$q$}})\,\gamma^0\otimes\gamma^0$ with
349: $V(\mbox{\boldmath{$p$}}-\mbox{\boldmath{$q$}})$ any Lorentz-scalar potential
350: function.
351:
352: Upon factorizing off all dependence of the Salpeter amplitude
353: $\Phi(\mbox{\boldmath{$p$}})$ on angular variables the exact-propagator
354: instantaneous Bethe--Salpeter equation (\ref{Eq:IBSE}) may be reduced
355: \cite{Lagae92a,Olsson95} to a set of coupled equations for the radial factors
356: of all the independent Salpeter components. For bound states composed of
357: particle and corresponding antiparticle we clearly have, with
358: $p\equiv|\mbox{\boldmath{$p$}}|,$
359: $Z_1(\mbox{\boldmath{$p$}}^2)=Z_2(\mbox{\boldmath{$p$}}^2)\equiv Z(p^2)$ and
360: $M_1(\mbox{\boldmath{$p$}}^2)=M_2(\mbox{\boldmath{$p$}}^2)\equiv M(p^2),$
361: which will also enter in $E_1(\mbox{\boldmath{$p$}})=
362: E_2(\mbox{\boldmath{$p$}})=E(p)\equiv\sqrt{p^2+M^2(p^2)}.$ The set of coupled
363: equations governing the radial functions $\Phi_1(p)$ and $\Phi_2(p)$ in our
364: independent Salpeter components $\Phi_1(\mbox{\boldmath{$p$}})$ and
365: $\Phi_2(\mbox{\boldmath{$p$}})$ of ${}^1{\rm S}_0$ states, respectively, can
366: be simply derived by, for instance, ``dressing'' Eq.~(1)~of
367: Ref.\cite{Lucha00:IBSEm=0} or Eq.~(1) of Ref.\cite{Lucha00:IBSEnzm} by
368: insertion of the appropriate factors $Z(p^2)$ in all interaction~terms and by
369: replacement of all constant constituent masses $m$ by the relevant mass
370: functions~$M(p^2)$:\begin{eqnarray}&&2\,E(p)\,\Phi_2(p)+Z^2(p^2)\int
371: \limits_0^\infty\frac{{\rm d}q\,q^2}{(2\pi)^2}\,V_0(p,q)\,\Phi_2(q)=M_{\rm
372: B}\,\Phi_1(p)\ ,\nonumber\\[1ex]&&2\,E(p)\,\Phi_1(p)\label{Eq:RE}\\[1ex]
373: &&+\,Z^2(p^2)\int\limits_0^\infty\frac{{\rm d}q\,q^2}{(2\pi)^2}
374: \left[\frac{M(p^2)}{E(p)}\,V_0(p,q)\,\frac{M(q^2)}{E(q)}
375: +\frac{p}{E(p)}\,V_1(p,q)\,\frac{q}{E(q)}\right]\Phi_1(q)=M_{\rm
376: B}\,\Phi_2(p)\ .\nonumber\end{eqnarray}The configuration- and momentum-space
377: representations of any radial function are related by Fourier--Bessel
378: transformations which involve spherical Bessel functions of the first~kind
379: $j_n(z)$ ($n=0,\pm 1,\pm2,\dots$)\cite{Abramowitz}; as a kind of reminiscence
380: of these, the interaction $V(\mbox{\boldmath{$p$}}-\mbox{\boldmath{$q$}})$~in
381: the kernel $\hat K(\mbox{\boldmath{$p$}}-\mbox{\boldmath{$q$}})$ enters here
382: in form of some static potential $V(r)$ in
383: configuration~space:$$V_L(p,q)\equiv 8\pi\int\limits_0^\infty{\rm
384: d}r\,r^2\,j_L(p\,r)\,j_L(q\,r)\,V(r)\ ,\quad L=0,1,2,\dots\ .$$
385:
386: The particular structure of the set of equations (\ref{Eq:RE}) allows to find
387: its solutions\cite{Lucha00:IBSEm=0,Lucha00:IBSE-C4,Lucha00:IBSEnzm,
388: Lucha01:IBSEIAS}~by inserting one of these relations into the other and
389: obtaining an eigenvalue equation for~$M_{\rm B}^2$:\begin{eqnarray}M_{\rm
390: B}^2\,\Phi_2(p)&=&4\,E^2(p)\,\Phi_2(p)+2\,Z^2(p^2)\,E(p)\int\limits_0^\infty
391: \frac{{\rm d}q\,q^2}{(2\pi)^2}\,V_0(p,q)\,\Phi_2(q)\nonumber\\[1ex]
392: &+&2\,\frac{Z^2(p^2)}{E(p)}\int\limits_0^\infty\frac{{\rm d}q\,q^2}{(2\pi)^2}
393: \left[M(p^2)\,M(q^2)\,V_0(p,q)+p\,q\,V_1(p,q)\right]\Phi_2(q)\nonumber\\[1ex]
394: &+&Z^2(p^2)\int\limits_0^\infty\frac{{\rm d}q\,q^2}{(2\pi)^2}
395: \left[\frac{M(p^2)}{E(p)}\,V_0(p,q)\,\frac{M(q^2)}{E(q)}
396: +\frac{p}{E(p)}\,V_1(p,q)\,\frac{q}{E(q)}\right]\nonumber\\[1ex]
397: &\times&Z^2(q^2)\int\limits_0^\infty\frac{{\rm
398: d}k\,k^2}{(2\pi)^2}\,V_0(q,k)\,\Phi_2(k)\ .\label{Eq:MB2}\end{eqnarray}By
399: expansion over the basis for radial functions summarized in
400: Appendix~\ref{App:GLB} we convert this integral equation to an equivalent
401: matrix eigenvalue problem which can be diagonalized by standard means. As
402: noted in Sec.~\ref{Sec:IBSEWEP}, Salpeter's equation corresponds to the
403: free-propagator approximation $Z(p^2)\cong1$ and $M(p^2)\cong m.$ Thus the
404: studies reported in Refs.\cite{Lucha00:IBSEm=0,Lucha00:IBSE-C4,
405: Lucha00:IBSEnzm,Lucha01:IBSEIAS}~could get the matrix, for a large class of
406: interactions, in algebraic form. Due to the presence~of the true quark
407: propagator functions $Z(p^2)$ and $M(p^2),$ in general this is no longer
408: possible~here. Upon construction of one Salpeter component, $\Phi_2(p),$ as
409: solution of Eq.~(\ref{Eq:MB2}), its companion, $\Phi_1(p),$ follows, for
410: $M_{\rm B}\ne0,$ immediately from the first of the two coupled equations
411: (\ref{Eq:RE}). For vanishing bound-state mass, $M_{\rm B}=0,$
412: Eqs.~(\ref{Eq:RE}) decouple and are thus solved independently.
413:
414: \section{Linear confinement: results and discussion}\label{Sec:RD}Let us
415: eventually apply the formalism developed in Secs.~\ref{Sec:IBSEWEP} through
416: \ref{Sec:PSFAFBS} to a confining~(static) potential of linear shape,
417: $V(r)=\lambda\,r,$ with slope $\lambda=0.2\;\mbox{GeV}^2.$ For confining
418: interactions a time-component Lorentz-vector structure of the kernel appears
419: to be free of all the stability problems encountered by solutions found for a
420: kernel of Lorentz-scalar structure\cite{Parramore95,Parramore96,Olsson95}.
421:
422: Our first goal is to analyze the effect of the dynamical generation of quark
423: masses on the solutions of the instantaneous Bethe--Salpeter equation
424: (\ref{Eq:IBSE}) with exact propagators. To this end, Fig.~\ref{Fig:MBCL}
425: compares, for the three lowest positive-norm $J^{PC}=0^{-+}$ bound states,
426: the mass eigenvalues $M_{\rm B}$ of Eq.~(\ref{Eq:IBSE}) for exact light-quark
427: propagators with $m_0=0$ (corresponding to the chiral limit of QCD) with
428: those of a Salpeter equation for massless constituents
429: \cite{Lucha00:IBSEm=0,Lucha00:IBSE-C4}.
430:
431: \begin{figure}[p]\begin{center}\psfig{figure=MB-Chiral-Limit.eps,scale=1}
432: \caption{Bound-state masses $M_{\rm B}$ (as functions of our variational
433: parameter $\mu$) of the three lowest-lying (positive-norm) $J^{PC}=0^{-+}$
434: eigenstates of our exact-propagator instantaneous Bethe--Salpeter equation
435: (\ref{Eq:IBSE}) for the propagator parametrization (\ref{Eq:ProPar}) but
436: $m_0=0$ ({\sl full~lines\/}) and of the Salpeter equation for massless
437: constituents ({\sl dashed lines\/}), for a time-component Lorentz-vector
438: kernel representing some linear potential $V(r)=\lambda\,r$ of slope
439: $\lambda=0.2\;\mbox{GeV}^2.$ These results arise from diagonalization, for
440: given values of $\mu,$ of $50\times50$ matrices equivalent to our
441: exact-propagator equation but, mimicking Ref.\cite{Lucha00:IBSEm=0}, only
442: $15\times15$ matrices equivalent to Salpeter's equation and in both cases 50
443: terms in all intermediate-step series expansions.}\label{Fig:MBCL}
444: \end{center}\end{figure}
445:
446: The chosen bases for the Hilbert space $L_2(R^+)$ of all with the weight
447: function $w(r)=r^2$ square-integrable (``radial'') functions on the positive
448: real line $R^+$ introduce one additional degree of freedom, by allowing the
449: basis states to depend on a variational parameter $\mu>0.$ As a basis, these
450: vectors constitute a complete orthonormal system for any particular~value of
451: $\mu.$ Therefore, as long as relying exclusively on expansions over the {\em
452: full\/} set of basis vectors our results may be expected to be independent of
453: $\mu.$ Necessary truncations of expansions to a finite number of basis
454: vectors will induce a certain amount of $\mu$-dependence of the~results.
455: However, a reasonable technique involving expansions should exhibit stability
456: with respect to the increase of the number of basis vectors. If taking into
457: account a large enough number of basis vectors, by reducing the dependence on
458: $\mu$ some ``region of stability'' should emerge.
459:
460: With respect to the mass $M_{\rm B}$ of a given bound state, such a region of
461: stability manifests itself in form of a plateau where the numerical value
462: predicted for $M_{\rm B}$ is constant over some nonvanishing range of $\mu.$
463: Beyond doubt, the formation of these plateaus is obvious in
464: Fig.~\ref{Fig:MBCL}. Such extrema of $M_{\rm B}$ disclose the bound states.
465: Compared with the free-propagator~results, the ground state of
466: Eq.~(\ref{Eq:IBSE}) is higher but its radial excitations are lower, to the
467: effect~that~all level spacings are significantly smaller if using exact
468: propagators. In general, it is, of~course, not possible to compensate for
469: these shifts by some change of the parameter values entering in the
470: interaction kernel. For instance, a reduction of the slope $\lambda$ of the
471: linear potential~by~a factor 2 to $\lambda\cong0.1\;\mbox{GeV}^2$ lowers the
472: ground-state energy eigenvalue of Eq.~(\ref{Eq:IBSE}) to the level of its
473: Salpeter-equation counterpart but simultaneously diminishes the level
474: spacings further.
475:
476: Table~\ref{Tab:MB-EPFP} lists the masses $M_{\rm B}$ of the three
477: lowest-lying (positive-norm) $J^{PC}=0^{-+}$ bound states calculated from the
478: exact-propagator instantaneous Bethe--Salpeter equation (\ref{Eq:IBSE}) for
479: the full ($m_0\ne0$) parametrization (\ref{Eq:ProPar}) and (\ref{Eq:ParVal})
480: of the exact light-quark propagators. Because of the relative smallness of
481: the (explicitly chiral-symmetry breaking) current mass $m_0$ these
482: eigenvalues are, of course, less than some 0.5\% larger than the
483: corresponding ``chiral-limit'' values forming the stability plateaus
484: discernible in Fig.~\ref{Fig:MBCL}. These mass values are confronted in
485: Table~\ref{Tab:MB-EPFP} with the corresponding mass eigenvalues $M_{\rm B}$
486: of the Salpeter equation, computed by assuming, for the constituent mass $m$
487: of the light {\sl u}- and {\sl d}-quarks entering their effective
488: propagators, the typical value of $m=0.336\;{\rm GeV}$ frequently adopted by
489: nonrelativistic and relativistic constituent quark models to describe hadrons
490: as bound states of quarks\cite{Lucha91,Lucha92}. Raising in the Salpeter
491: equation the constituent quark mass $m$ from $m=0$ to the canonical value
492: $m=0.336\;{\rm GeV}$ shifts the masses $M_{\rm B}$ of the three lowest states
493: by more than $0.1\;{\rm GeV}$ towards larger values. The net result of this
494: is a further increase of the discrepancy between the level spacings predicted
495: by the exact-propagator equation (\ref{Eq:IBSE}) and the ones arising~from
496: the free-propagator Salpeter equation with some kind of appropriate or
497: reasonable effective mass of the constituent quarks. Therefore, we are forced
498: to conclude that any neglect of the proper behaviour of the
499: momentum-dependent quark mass, $M(p^2),$ by approximating it by a constant
500: constituent mass $m$ is, at least for the light {\sl u}- and {\sl d}-quarks,
501: rather~questionable.
502:
503: \begin{table}[ht]\caption{Bound-state masses $M_{\rm B}$ (in units of GeV)
504: for the three lowest-lying positive-norm $J^{PC}=0^{-+}$ eigenstates (denoted
505: by $1^1{\rm S}_0,$ $2^1{\rm S}_0,$ $3^1{\rm S}_0,$ in usual spectroscopic
506: notation) of our exact-propagator instantaneous Bethe--Salpeter equation
507: (\ref{Eq:IBSE}) with light-quark propagators parametrized by
508: Eqs.~(\ref{Eq:ProPar}), (\ref{Eq:ParVal}) and of the free-propagator Salpeter
509: equation with light-quark constituent mass $m=0.336\;{\rm GeV},$ for
510: time-component Lorentz-vector kernels representing a linear potential
511: $V(r)=\lambda\,r$ of slope $\lambda=0.2\;\mbox{GeV}^2,$ obtained by
512: converting both equations to $50\times50$ matrices and taking into account 50
513: terms in intermediate-step series expansions.}\label{Tab:MB-EPFP}
514: \begin{center}\begin{tabular}{ccc}\hline\hline&&\\[-1.5ex]\multicolumn{1}{c}
515: {State}&\multicolumn{1}{c}{\begin{tabular}{c}Exact-propagator\\bound-state
516: equation\end{tabular}}&\multicolumn{1}{c}{\begin{tabular}{c}(Free-propagator)\\
517: Salpeter equation\end{tabular}}\\[2.5ex]\hline\\[-1.5ex] $1^1{\rm
518: S}_0$&1.750&1.813\\$2^1{\rm S}_0$&1.895&2.410\\$3^1{\rm
519: S}_0$&2.056&2.889\\[1ex]\hline\hline\end{tabular}\end{center}\end{table}
520:
521: \begin{figure}[p]\begin{center}\begin{tabular}{c}
522: \psfig{figure=Phi1-Plot.eps,scale=1}\\[0.05ex](a)\end{tabular}\\[1.5ex]
523: \begin{tabular}{c}\psfig{figure=Phi2-Plot.eps,scale=1}\\[0.05ex](b)
524: \end{tabular}\\[0.5ex]\caption{Configuration-space radial Salpeter component
525: functions $\Phi_1(r)$ (a) and $\Phi_2(r)$ (b) for the lowest positive-norm
526: $J^{PC}=0^{-+}$ bound state of the exact-propagator instantaneous
527: Bethe--Salpeter equation (\ref{Eq:IBSE}) with the quark propagator
528: parametrization (\ref{Eq:ProPar}) ({\sl full lines\/}) and of the Salpeter
529: equation with a light-quark {\em constituent\/} mass $m=0.336\;{\rm GeV}$
530: ({\sl dashed lines\/}) for time-component Lorentz-vector kernels with linear
531: potential $V(r)=\lambda\,r,$ $\lambda=0.2\;\mbox{GeV}^2.$}\label{Fig:Phi12}
532: \end{center}\end{figure}
533:
534: Figure~\ref{Fig:Phi12} illustrates for the $J^{PC}=0^{-+}$ ground state
535: ($1^1{\rm S}_0$ state) of the exact-propagator instantaneous Bethe--Salpeter
536: equation (\ref{Eq:IBSE}) with quark-propagator parametrization
537: (\ref{Eq:ProPar}) the configuration-space behaviour of the radial Salpeter
538: component functions $\Phi_1(r)$ and $\Phi_2(r).$ The norm $\|\Phi\|$ of the
539: Salpeter amplitude $\Phi(\mbox{\boldmath{$p$}})$ for $J^{PC}=0^{-+}$ bound
540: states reads\cite{Lagae92a,Olsson95,Lucha00:IBSEm=0}
541: $$\|\Phi\|^2=4\int\frac{{\rm d}^3p}{(2\pi)^3}\,
542: [\Phi_1^\ast(\mbox{\boldmath{$p$}})\,\Phi_2(\mbox{\boldmath{$p$}})
543: +\Phi_2^\ast(\mbox{\boldmath{$p$}})\,\Phi_1(\mbox{\boldmath{$p$}})]\ ;$$this
544: translates for the radial parts $\Phi_1(p),$ $\Phi_2(p)$ of the Salpeter
545: components $\Phi_1(\mbox{\boldmath{$p$}}),$
546: $\Phi_2(\mbox{\boldmath{$p$}})$~to
547: $$\|\Phi\|^2=\frac{4}{(2\pi)^3}\int\limits_0^\infty{\rm
548: d}p\,p^2\,[\Phi_1^\ast(p)\,\Phi_2(p)+\Phi_2^\ast(p)\,\Phi_1(p)]\ .$$For the
549: plots depicted in Fig.~\ref{Fig:Phi12}, the normalization has been chosen
550: such that $\Phi_2(r)$ satisfies$$\int\limits_0^\infty{\rm
551: d}r\,r^2\,|\Phi_2(r)|^2=\int\limits_0^\infty{\rm d}p\,p^2\,|\Phi_2(p)|^2=1\
552: ;$$the first of Eqs.~(\ref{Eq:RE}) fixes, after a Fourier--Bessel
553: transformation, the normalization of~$\Phi_1(r).$ A normalization factor
554: common to $\Phi_1(p)$ and $\Phi_2(p)$ will then give any desired value
555: to~$\|\Phi\|.$
556:
557: A closer inspection of the radial Salpeter components $\Phi_1(r)$ and
558: $\Phi_2(r)$ reveals a striking similarity to their counterparts found as
559: solutions of the Salpeter equation for a constituent quark mass of
560: $m=0.336\;{\rm GeV}.$ Exact- and free-propagator Salpeter components
561: $\Phi_1(r)$ and $\Phi_2(r)$ show some notable difference only for $r<5\;{\rm
562: GeV}^{-1}$ for $\Phi_1(r)$ and for $r<2\;{\rm GeV}^{-1}$~for $\Phi_2(r).$
563: They are hardly distinguishable from each other for $r>5\;{\rm GeV}^{-1}$ for
564: both $\Phi_1(r)$ and $\Phi_2(r).$ Thus we feel entitled to expect similar
565: predictions for quantities such as decay~rates.
566:
567: \section{Summary, conclusions, and outlook}\label{Sec:SCO}The reduction of
568: the Bethe--Salpeter formalism to the Salpeter equation requires to assume for
569: all bound-state constituents not only an instantaneous interaction but also
570: free-particle propagation. Realizing this fact, a bound-state equation that
571: retains the exact propagators of the bound-state constituents and generalizes
572: Salpeter's equation has been formulated~by consequent application of the
573: instantaneous approximation to all exact propagators
574: too\cite{Lucha05:IBSEWEP}. Of course, this may be extended to Bethe--Salpeter
575: equations for bound states composed of particles that are not, or not all,
576: identical to spin-$\frac{1}{2}$ fermions as well as to three-dimensional
577: reductions\cite{Babutsidze98,Babutsidze99,Kopaleishvili01,Babutsidze03} of
578: the Bethe--Salpeter equation\cite{BSE} different from Salpeter's
579: equation\cite{SE}.
580:
581: The present investigation addressed the question of how realistic
582: descriptions of mesons as quark--antiquark bound states within a general
583: instantaneous Bethe--Salpeter formalism will be modified by such
584: reinstallment of the exact quark propagators, extracted from QCD by analytic
585: continuation from Euclidean to Minkowski space. Interestingly, for the
586: example of a specific interaction used already in earlier studies
587: \cite{Olsson95, Olsson96,Lucha00:IBSEm=0,Lucha00:IBSE-C4,Lucha00:IBSEnzm,
588: Lucha01:IBSEIAS}, we find a drastic shrinking of the level spacings of the
589: bound states while their amplitudes remain practically unchanged.
590:
591: \section*{Acknowledgements}One of us (W.~L.) would like to thank Craig
592: D.~Roberts for a lot of stimulating discussions, as well as for communicating
593: his parametrizations of the (numerical) propagator functions.
594:
595: \appendix\section{The generalized Laguerre basis for radial functions}
596: \label{App:GLB}The Hilbert space $L_2(R^3)$ of all square-integrable
597: functions on the 3-dimensional Euclidean space $R^3$ can be spanned by basis
598: functions each of which is the product of a function of the radial variable
599: and of an angular term, the latter being represented by a spherical harmonic
600: ${\cal Y}_{\ell m}(\Omega)$ for the angular momentum $\ell=0,1,2,\dots$ and
601: its projection $m=-\ell,-\ell+1,\dots,+\ell$ which all depend on the solid
602: angle $\Omega\equiv(\theta,\phi)$ and satisfy the orthonormalization
603: condition$$\int{\rm d}\Omega\,{\cal Y}^\ast_{\ell m}(\Omega)\,{\cal
604: Y}_{\ell'm'}(\Omega)=\delta_{\ell\ell'}\,\delta_{mm'}\ .$$For each value of
605: $\ell$ the radial functions constitute a basis for the Hilbert space
606: $L_2(R^+)$ of all with the weight function $w(r)=r^2$ square-integrable
607: functions on the positive real line $R^+.$ The basis functions of $L_2(R^3)$
608: in configuration and momentum space are related by Fourier transformation.
609: Thus, the configuration-space representation $\phi_i^{(\ell)}(r)$ and
610: momentum-space representation $\phi_i^{(\ell)}(p)$ of the radial factors are
611: related by the Fourier--Bessel transformation
612: \begin{eqnarray*}\phi_i^{(\ell)}(r)&=&{\rm
613: i}^\ell\,\sqrt{\frac{2}{\pi}}\int\limits_0^\infty{\rm
614: d}p\,p^2\,j_\ell(p\,r)\,\phi_i^{(\ell)}(p)\ ,\quad i=0,1,2,\dots\
615: ,\quad\ell=0,1,2,\dots\ ,\\[1ex]\phi_i^{(\ell)}(p)&=&(-{\rm
616: i})^\ell\,\sqrt{\frac{2}{\pi}}\int\limits_0^\infty{\rm
617: d}r\,r^2\,j_\ell(p\,r)\,\phi_i^{(\ell)}(r)\ ,\quad i=0,1,2,\dots\
618: ,\quad\ell=0,1,2,\dots\ .\end{eqnarray*}The spherical Bessel functions of the
619: first kind, $j_n(z)$ ($n=0,\pm 1,\pm 2,\dots$)\cite{Abramowitz}, are remnants
620: of the angular integration. This may be easily deduced with the help of the
621: expansion~of the plane waves over spherical harmonics ${\cal Y}_{\ell m}$ in
622: configuration $(\Omega_r)$ and momentum $(\Omega_p)$~space$$\exp({\rm
623: i}\,\mbox{\boldmath{$p$}}\cdot\mbox{\boldmath{$x$}})=4\pi\,\sum_{\ell=0}^\infty
624: \,\sum_{m=-\ell}^{+\ell}\,{\rm i}^\ell\,j_\ell(p\,r)\,{\cal Y}^\ast_{\ell
625: m}(\Omega_p)\,{\cal Y}_{\ell m}(\Omega_r)\ .$$
626:
627: In terms of orthogonal polynomials of generalized-Laguerre type (for
628: parameter $\gamma$)\cite{Abramowitz},$$L_i^{(\gamma)}(x)=\sum_{t=0}^i\,
629: (-1)^t\left(\begin{array}{c}i+\gamma\\i-t\end{array}\right)\frac{x^t}{t!}\
630: ,\quad i=0,1,2,\dots\ ,$$which, by construction, are orthonormalized [with
631: weight $w(x)=x^\gamma\exp(-x)]$ according~to$$\int\limits_0^\infty{\rm
632: d}x\,x^\gamma\exp(-x)\,L_i^{(\gamma)}(x)\,L_j^{(\gamma)}(x)=
633: \frac{\Gamma(\gamma+i+1)}{i!}\,\delta_{ij}\ ,\quad i,j=0,1,2,\dots\ ,$$our
634: favourite choice of these radial bases is defined in configuration-space
635: representation~by$$\phi_i^{(\ell)}(r)=\sqrt{\frac{(2\,\mu)^{2\,\ell+3}\,i!}
636: {\Gamma(2\,\ell+i+3)}}\,r^\ell\exp(-\mu\,r)\, L_i^{(2\,\ell+2)}(2\,\mu\,r)\
637: ,\quad i=0,1,2,\dots\ .$$These basis functions involve one positive real
638: variational parameter, with the dimension of mass, $\mu.$ The requirement of
639: their normalizability imposes the constraint $\mu>0.$ Then these
640: configuration-space radial basis functions, $\phi_i^{(\ell)}(r),$ satisfy the
641: orthonormalization~condition$$\int\limits_0^\infty{\rm d}r\,r^2\,
642: \phi_i^{(\ell)}(r)\,\phi_j^{(\ell)}(r)=\delta_{ij}\ ,\quad i,j=0,1,2,\dots\
643: .$$Note that the configuration-space representation of our basis functions is
644: chosen to be~real.
645:
646: \newpage
647:
648: The corresponding momentum-space representation $\phi_i^{(\ell)}(p)$ of our
649: basis functions reads\begin{eqnarray*}\phi_i^{(\ell)}(p)
650: &=&\sqrt{\frac{(2\,\mu)^{2\,\ell+3}\,i!}{\Gamma(2\,\ell+i+3)}}\,\frac{(-{\rm
651: i})^\ell\,p^\ell}{2^{\ell+1/2}\,\Gamma\left(\ell+\frac{3}{2}\right)}\,\\[1ex]
652: &\times&\sum_{t=0}^i\,\frac{(-1)^t}{t!}\left(\begin{array}{c}i+2\,\ell+2\\
653: i-t\end{array}\right)\frac{\Gamma(2\,\ell+t+3)\,(2\,\mu)^t}
654: {(p^2+\mu^2)^{(2\,\ell+t+3)/2}}\\[1ex]
655: &\times&F\left(\frac{2\,\ell+t+3}{2},-\frac{1+t}{2};\ell+\frac{3}{2};
656: \frac{p^2}{p^2+\mu^2}\right),\quad i=0,1,2,\dots\ ,\end{eqnarray*} with the
657: hypergeometric series $F(u,v;w;z)$ given in terms of the gamma function
658: $\Gamma$ by\cite{Abramowitz}
659: $$F(u,v;w;z)=\frac{\Gamma(w)}{\Gamma(u)\,\Gamma(v)}\,\sum_{n=0}^\infty\,
660: \frac{\Gamma(u+n)\,\Gamma(v+n)}{\Gamma(w+n)}\,\frac{z^n}{n!}\ .$$The
661: momentum-space radial basis functions $\phi_i^{(\ell)}(p)$ fulfill the
662: orthonormalization condition$$\int\limits_0^\infty{\rm
663: d}p\,p^2\,\phi_i^{\ast(\ell)}(p)\,\phi_j^{(\ell)}(p)=\delta_{ij}\ ,\quad
664: i,j=0,1,2,\dots\ .$$In momentum space, our basis functions are real for
665: $\ell=0,$ as well as for all even values~of~$\ell$:$$\phi_i^{\ast(\ell)}(p)=
666: \phi_i^{(\ell)}(p)\quad\mbox{for\ }\ell=0,2,4,\dots\ ,\quad\forall\
667: i=0,1,2,\dots\ .$$The virtue of our bases is their analytic availability in
668: configuration {\em and\/} momentum space.
669:
670: Mainly for computational convenience, the present investigation makes use of
671: the radial basis functions for two values $\ell=0$ and $\ell=1$ of the
672: angular momentum. Having to deal, in momentum-space representation, with the
673: cumbersome hypergeometric series~$F(u,v;w;z)$ may be avoided by employing
674: simplified expressions equivalent to the above definition\cite{Lucha97}:
675: \begin{eqnarray*}
676: \phi_i^{(0)}(p)&=&\sqrt{\frac{i!}{\mu\,\pi\,\Gamma(i+3)}}\,\frac{4}{p}\,
677: \sum_{t=0}^i\,(-2)^t\,(t+1)\left(\begin{array}{c}i+2\\i-t\end{array}\right)
678: \left(1+\frac{p^2}{\mu^2}\right)^{-(t+2)/2}\\[1ex]
679: &\times&\sin\left((t+2)\arctan\frac{p}{\mu}\right)\\[1ex] &=&\frac{{\rm
680: Im}\{(p+{\rm i}\,\mu)^{2\,i+3}\,[p-{\rm i}\,(3+2\,i)\,\mu]\}}
681: {\sqrt{\mu\,\pi\,(i+1)\,(i+2)}\,p\,(p^2+\mu^2)^{2+i}}\
682: ,\\[1ex]\phi_i^{(1)}(p) &=&-{\rm
683: i}\,\sqrt{\frac{\mu^5}{\pi\,(i+1)\,(i+2)\,(i+3)\,(i+4)}}\,
684: \frac{8}{p^2}\,\sum_{t=0}^i\,\frac{(-2)^t}{t!}
685: \left(\begin{array}{c}i+4\\i-t\end{array}\right)
686: \frac{(t+3)!\,\mu^t}{(p^2+\mu^2)^{(t+3)/2}}\\[1ex]&\times&
687: \left[\frac{\sqrt{p^2+\mu^2}}{t+2}\sin\left((t+2)\arctan\frac{p}{\mu}\right)-
688: \frac{\mu}{t+3}\sin\left((t+3)\arctan\frac{p}{\mu}\right)\right]\\[1ex]
689: &=&\frac{{\rm i}}{2\,\sqrt{\mu^3\,\pi\,(i+1)\,(i+2)\,(i+3)\,(i+4)}\,p^2\,
690: (p^2+\mu^2)^3}\\[1ex]&\times& {\rm Im}\left\{\frac{(p-{\rm
691: i}\,\mu)^{i+5}}{(p+{\rm i}\,\mu)^i}\, [3\,p^3+3\,{\rm
692: i}\,(5+2\,i)\,p^2\,\mu-(5+2\,i)^2\,p\,\mu^2-{\rm i}\,
693: (5+2\,i)\,\mu^3]\right\}.\end{eqnarray*}
694:
695: \small\begin{thebibliography}{30}
696: \bibitem{BSE}E.~E.~Salpeter and H.~A.~Bethe, Phys.~Rev.~{\bf 84} (1951) 1232.
697: \bibitem{SE}E.~E.~Salpeter, Phys.~Rev.~{\bf 87} (1952) 328.
698: \bibitem{Lucha05:IBSEWEP}W.~Lucha and F.~F.~Sch\"oberl, J.~Phys.~G:
699: Nucl.~Part.~Phys.~{\bf 31} (2005) 1133, hep-th/0507281.
700: \bibitem{Maris97a}P.~Maris and C.~D.~Roberts, Phys.~Rev.~C {\bf 56} (1997)
701: 3369, nucl-th/9708029.
702: \bibitem{Maris97b}P.~Maris and C.~D.~Roberts, in: Proc.~of the IV$^{\rm th}$
703: Int.~Workshop on {\em Progress in Heavy~Quark Physics\/}, edited by M.~Beyer,
704: T.~Mannel, and H.~Schr\"oder (University of Rostock, Rostock, 1998), p.~159,
705: nucl-th/9710062.
706: \bibitem{Roberts98}C.~D.~Roberts, in: Proc.~of the 11$^{\rm th}$ Physics
707: Summer School on {\em Frontiers in Nuclear Physics: From Quark--Gluon Plasma
708: to Supernova\/}, edited by S.~Kuyucak (World Scientific, Singapore, 1999),
709: p.~212, nucl-th/9807026.
710: \bibitem{Ivanov98}M.~A.~Ivanov, Yu.~L.~Kalinovsky, and C.~D.~Roberts,
711: Phys.~Rev.~D {\bf 60} (1999) 034018,~nucl-th/9812063.
712: \bibitem{Maris99a}P.~Maris and P.~C.~Tandy, Phys.~Rev.~C {\bf 60} (1999)
713: 055214, nucl-th/9905056.
714: \bibitem{Maris99b}P.~Maris, Nucl.~Phys.~A {\bf 663} (2000) 621,
715: nucl-th/9908044.
716: \bibitem{Roberts00a}C.~D.~Roberts and S.~M.~Schmidt,
717: Prog.~Part.~Nucl.~Phys.~{\bf 45} (2000) S1, nucl-th/0005064.
718: \bibitem{Roberts00b}C.~D.~Roberts, nucl-th/0007054.
719: \bibitem{Alkofer00}R.~Alkofer and L.~von Smekal, Phys.~Rep.~{\bf 353} (2001)
720: 281, hep-ph/0007355.
721: \bibitem{Maris00}P.~Maris, in: Proc.~of the Int.~Conf.~on {\em Quark
722: Confinement and the Hadron Spectrum~IV\/}, edited by W.~Lucha and K.~Maung
723: Maung (World Scientific, New Jersey/London/Singa\-pore/Hong Kong, 2002),
724: p.~163, nucl-th/0009064.
725: \bibitem{Maris01}P.~Maris and P.~C.~Tandy, nucl-th/0109035.
726: \bibitem{Maris02}P.~Maris, A.~Raya, C.~D.~Roberts, and S.~M.~Schmidt,
727: Eur.~Phys.~J.~A {\bf 18} (2003) 231, nucl-th/0208071.
728: \bibitem{Bhagwat02}M.~S.~Bhagwat, M.~A.~Pichowsky, and P.~C.~Tandy,
729: Phys.~Rev.~D {\bf 67} (2003) 054019, hep-ph/0212276.
730: \bibitem{Tandy03}P.~C.~Tandy, Prog.~Part.~Nucl.~Phys.~{\bf 50} (2003) 305,
731: nucl-th/0301040.
732: \bibitem{Maris03}P.~Maris and C.~D.~Roberts, Int.~J.~Mod.~Phys.~E {\bf 12}
733: (2003) 297, nucl-th/0301049.
734: \bibitem{Bhagwat03}M.~S.~Bhagwat, M.~A.~Pichowsky, C.~D.~Roberts, and
735: P.~C.~Tandy, Phys.~Rev.~C {\bf 68} (2003) 015203, nucl-th/0304003.
736: \bibitem{Roberts03}C.~D.~Roberts, Lect.~Notes Phys.~{\bf 647} (2004) 149,
737: nucl-th/0304050.
738: \bibitem{Krassnigg03a}A.~Krassnigg and C.~D.~Roberts, Fizika B {\bf 13}
739: (2004) 143, nucl-th/0308039.
740: \bibitem{Krassnigg03b}A.~Krassnigg and C.~D.~Roberts, Nucl.~Phys.~A {\bf 737}
741: (2004) 7, nucl-th/0309025.
742: \bibitem{Alkofer03a}R.~Alkofer, W.~Detmold, C.~S.~Fischer, and P.~Maris,
743: Phys.~Rev.~D {\bf 70} (2004) 014014, hep-ph/0309077.
744: \bibitem{Krassnigg04}A.~Krassnigg and P.~Maris, J.~Phys.~Conf.~Ser.~{\bf 9}
745: (2005) 153, nucl-th/0412058.
746: \bibitem{Lagae92a}J.-F.~Laga\"e, Phys.~Rev.~D {\bf 45} (1992) 305.
747: \bibitem{Lagae92b}J.-F.~Laga\"e, Phys.~Rev.~D {\bf 45} (1992) 317.
748: \bibitem{Olsson95}M.~G.~Olsson, S.~Veseli, and K.~Williams, Phys.~Rev.~D {\bf
749: 52} (1995) 5141, hep-ph/9503477.
750: \bibitem{Olsson96}M.~G.~Olsson, S.~Veseli, and K.~Williams, Phys.~Rev.~D {\bf
751: 53} (1996) 504, hep-ph/9504221.
752: \bibitem{Lucha00:IBSEm=0}W.~Lucha, K.~Maung Maung, and F.~F.~Sch\"oberl,
753: Phys.~Rev.~D {\bf 63} (2001) 056002, hep-ph/0009185.
754: \bibitem{Lucha00:IBSE-C4}W.~Lucha, K.~Maung Maung, and F.~F.~Sch\"oberl, in:
755: Proc.~of the Int.~Conf.~on {\em Quark Confinement and the Hadron Spectrum
756: IV\/}, edited by W.~Lucha and K.~Maung Maung (World Scientific, New
757: Jersey/London/Singapore/Hong Kong, 2002), p.~340, hep-ph/0010078.
758: \bibitem{Lucha00:IBSEnzm}W.~Lucha, K.~Maung Maung, and F.~F.~Sch\"oberl,
759: Phys.~Rev.~D {\bf 64} (2001) 036007, hep-ph/0011235.
760: \bibitem{Lucha01:IBSEIAS}W.~Lucha and F.~F.~Sch\"oberl, Int.~J.~Mod.~Phys.~A
761: {\bf 17} (2002) 2233, hep-ph/0109165.
762: \bibitem{Maris94}P.~Maris, Phys.~Rev.~D {\bf 50} (1994) 4189.
763: \bibitem{Alkofer03b}R.~Alkofer, W.~Detmold, C.~S.~Fischer, and P.~Maris,
764: Nucl.~Phys.~Proc.~Suppl.~{\bf 141} (2005) 122, hep-ph/0309078.
765: \bibitem{Roberts02:PC}C.~D.~Roberts (private communication).
766: \bibitem{Bowman05}P.~O.~Bowman, U.~M.~Heller, D.~B.~Leinweber,
767: M.~B.~Parappilly, A.~G.~Williams, and J.-B.~Zhang, Phys.~Rev.~D {\bf 71}
768: (2005) 054507, hep-lat/0501019.
769: \bibitem{Abramowitz}{\em Handbook of Mathematical Functions}, edited by
770: M.~Abramowitz and I.~A.~Stegun (Dover, New York, 1964).
771: \bibitem{Parramore95}J.~Parramore and J.~Piekarewicz, Nucl.~Phys.~A {\bf 585}
772: (1995) 705, nucl-th/9402019.
773: \bibitem{Parramore96}J.~Parramore, H.-C.~Jean, and J.~Piekarewicz,
774: Phys.~Rev.~C {\bf 53} (1996) 2449, nucl-th/9510024.
775: \bibitem{Lucha91}W.~Lucha, F.~F.~Sch\"oberl, and D.~Gromes, Phys.~Rep.~{\bf
776: 200} (1991) 127.
777: \bibitem{Lucha92}W.~Lucha and F.~F.~Sch\"oberl, Int.~J.~Mod.~Phys.~A {\bf 7}
778: (1992) 6431.
779: \bibitem{Babutsidze98}T.~Babutsidze, T.~Kopaleishvili, and A.~Rusetsky,
780: Phys.~Lett.~B {\bf 426} (1998) 139, hep-ph/9710278.
781: \bibitem{Babutsidze99}T.~Babutsidze, T.~Kopaleishvili, and A.~Rusetsky,
782: Phys.~Rev.~C {\bf 59} (1999) 976, hep-ph/9807485.
783: \bibitem{Kopaleishvili01}T.~Kopaleishvili, Phys.~Part.~Nucl.~{\bf 32} (2001)
784: 560, hep-ph/0101271.
785: \bibitem{Babutsidze03}T.~Babutsidze, T.~Kopaleishvili, and D.~Kurashvili,
786: Georgian Electronic Scientific J.: Phys.~{\bf 1} (2004) 20, hep-ph/0308072.
787: \bibitem{Lucha97}W.~Lucha and F.~F.~Sch\"oberl, Phys.~Rev.~A {\bf 56} (1997)
788: 139, hep-ph/9609322.
789: \end{thebibliography}\end{document}
790: