1:
2: \documentclass[showpacs,aps,preprint,prc]{revtex4}
3: \usepackage{graphicx}
4:
5: %modified by SAC August 25, 2002
6: %modified by SAC July 29,2002
7:
8:
9: \begin{document}
10: \title{\bf The singular inverse square potential, limit cycles and self-adjoint
11: extensions}
12:
13:
14: \author{M.\ Bawin}
15:
16: \affiliation{Universit\'{e} de Li\`{e}ge, Institut de Physique B5, Sart
17: Tilman, 4000 Li\`{e}ge 1, Belgium }
18:
19: \author{S.\ A.\ Coon}
20: \affiliation{Physics Department, New Mexico State University, Las Cruces, NM
21: 88003, USA and National Science Foundation, 4201 Wilson Blvd., Arlington,
22: Virginia 22230, USA}
23:
24: \date{\today}
25:
26:
27:
28:
29:
30: \begin{abstract}
31:
32: We study the radial Schroedinger equation for a particle of mass m in the
33: field of a singular attractive $ {\alpha}/{r^2}$ potential with $2m\alpha >
34: {1}/{4}$. This potential is relevant to the fabrication of nanoscale atom optical devices, is said to be the potential describing the dipole-bound anions of polar molecules, and is the effective potential underlying the universal behavior of three-body systems in nuclear physics and atomic physics, including aspects of Bose-Einstein condensates, first described by Efimov. New results in three-body physical systems motivate the present investigation. Using the regularization method of Beane et al., we show that the
35: corresponding ``renormalization group flow'' equation can be solved
36: analytically. We find that it exhibits a limit cycle behavior and has
37: infinitely many branches. We show that a physical meaning for self-adjoint
38: extensions of the Hamiltonian arises naturally in this framework.
39: \end{abstract}
40:
41:
42:
43: \pacs{ 34.20.-b, 03.65.Ca, 03.65.Ge, 11.10.Hi,}
44:
45: %\vskip20pt
46:
47:
48: \maketitle
49:
50: In this note, we study the regularization and renormalization of the singular
51: attractive $ {\alpha}/{r^2}$ potential, a problem motivated, in part, by
52: recent results in three-body physics. Specifically, there has been renewed
53: interest in the non-relativistic three-body system with short-range
54: interactions. The current investigations are stimulated by the promise that
55: effective field theories (EFT'S) offer a systematic and model-independent
56: treatment of atomic, nuclear and hadronic physics at low energies. That is, a
57: low-energy system with a clear-cut separation of distance scales can be
58: described by an EFT involving explicitly only the long-wavelength degrees of
59: freedom. The short-range dynamics can be treated as a set of local operators
60: which correspond to delta-function interactions in coordinate space. The
61: details of the short distance physics cannot be of importance to the low energy
62: aspects of the system; if they are, it is an indication of a need for
63: renormalization of the EFT.
64:
65: Such a renormalization of an EFT of a three-body system with delta-function
66: two-body interactions \cite{BHK} has lead to the rediscovery of the
67: one-parameter contact three-body interaction shown to restore the lower bound
68: of this three-body Hamiltonian \cite{AHW}. (Here we mention that the
69: unboundedness of the Hamiltonian from below is interpreted in EFT's as the
70: onset of short-distance physics whose effect must be included in local
71: counterterms \cite{Beane}; hence the three-body counterterm in the EFT
72: equations for 3-body systems, both bound and scattering states \cite{EFTmisc}.)
73: The three-body counterterm exhibits a convergence of the renormalization group
74: flow to one-dimensional limit cycles. This was earlier proved in Ref.
75: \cite{AHW}, a mathematical analysis of the Efimov effect which occurs in bound
76: 3-body systems when more than one of the two-body subsystems has a zero energy
77: resonance \cite{Efi}. The presumed novelty of a renormalization group flow with
78: a limit cycle has inspired additional recent work \cite{Wil}.
79:
80: As the applications of EFT's continue \cite{EFTmisc,Bra}, it will be important
81: to understand how to explicitly renormalize higher orders in an EFT. For
82: example, a renormalized equation for two-nucleon systems with explicit
83: pion-exchange would be of great potential value. Pion exchange gives rise to a
84: singular $ 1/r^3$ potential and the questions arise: Can the resummation of
85: pion graphs be renormalized by a single local operator? Would this operator
86: exhibit a limit cycle as does the three-body contact operator in the pion-less
87: three-nucleon EFT? Could one calculate the evolution of such an operator
88: analytically? A positive answer to the last question would help future
89: numerical work with EFT's. Already, the short distance physics of the $^3S_1$ coupled channels of the single pion exchange potential has been renormalized by a short range four-nucleon counterterm using the method introduced in ref.[3], but the treatment was numerical and these questions were not addressed in that investigation \cite{BeBe}
90:
91:
92: Such questions, coupled with the interesting limit cycle behavior found in the
93: three-body system, have prompted investigations of the renormalization group
94: behavior of the short range counterterms which serve to regularize given long
95: range potentials (including singular potentials) in the two-body Schroedinger
96: equation. Since these long range potentials are often singular at the origin,
97: it has been argued that the short range interaction should not be represented
98: by a 3-dimensional delta-function at the origin. Birse et al.
99: \cite{Birse} choose a delta-shell potential and Beane et al. \cite{Beane}
100: suggest that a simple attractive square well represents a ``smeared out''
101: delta-function potential. In either case, it is said, the details of the short
102: distance regularization should not matter; the low energy aspects of the system
103: should be invariant in the same way under suitable changes of the short range
104: potential.
105:
106: For completeness, we list other regularization and renormalization schemes which do not follow from the separation of scales of an ETF, but also have been applied to the inverse square potential description of physical systems. The problem of a neutral atom interacting with a charged wire \cite{Sch}, relevant to the development of nanoscale atom optical devices, has been treated with the method of self-adjoint extensions \cite{BaCo}. A short distance cutoff scheme which renormalizes the strength of the $1/r^2$ potential, yields a critical dipole moment that has been confronted with the experimental capture of electrons by dipole molecules and formation of anions as an example of quantum mechanical symmetry breaking \cite{cam},\cite{CoH}. These alternative regularizations are not the subject of our present investigation.
107:
108:
109: In this note, we follow the regularization method of Beane et al.~\cite{Beane}
110: to obtain analytically the renormalization group behavior of the coupling
111: constant of the short range attractive square well they use to regularize the
112: long range inverse square potential. The $ 1/r^2$ potential is on the boundary
113: between singular and regular potentials and thus does or does not require a
114: self-adjoint extension, depending on the strength of the interaction. More
115: interesting from the EFT point of view is the Efimov observation \cite{Efim}
116: that the low energy behavior of three-body systems is determined by a long range
117: three-body effective interaction of the form $1/R^2$ where $R$ is built from
118: the relative distances between the particles. Thus, the two-body inverse square
119: potential is the analogue to the interaction in a three-body system in the
120: limit of zero energy resonance (and infinite two-body scattering lengths).
121:
122: A further advantage of our analytic approach to the ``EFT style''
123: renormalization of the inverse square potential is that we can then more
124: readily make contact with the mathematically rigorous and well studied approach
125: to regularization via self-adjoint extensions. The theory of self-adjoint
126: extensions of Hamiltonians underlies the first discussions of limit cycle
127: behavior in three-body systems~\cite{AHW}. The self-adjoint extensions of the
128: inverse square potential are well known \cite{Meetz} and can be compared with
129: the results of our study, which we now begin.
130:
131: The starting point of our study is the
132: $s-$wave reduced radial Schroedinger equation for one particle of mass $m$ in
133: the external potential $V(r)$:
134: \begin{equation}
135: \left( \frac{d^2}{dr^2} - 2mV(r) + k^2 \right) \psi = 0
136: \end{equation}
137: where $V(r)$ is given by \cite{Beane}:
138: \begin{equation}
139: V(r) = -\frac{\alpha_s \theta (R-r)}{R^2} - \frac{\alpha \theta(r-R)}{r^2}
140: \,\,\,\,\,\,
141: (\alpha_s, \alpha > 0).
142: \end{equation}
143: That is, the long range attractive $\alpha/r^2$ second term in eq. (2) is
144: cutoff at a short distance radius $R$ by an attractive square well. As in
145: \cite{Beane}, we first solve eq. (1) for the zero energy solution $(k=0)$ $
146: \psi_o$. It is given by:
147: \begin{eqnarray}
148: \psi_o (r) &=& A \frac{r^{1/2}}{{r_o}^{1/2}}\cos \left( \nu \ln\frac{r}{r_o} +
149: \phi_o \right) \hspace{.75in} \,\,\,\, r>R \\
150: \psi_o (r) &=& A \frac{\cos (\nu \ln\frac{R}{r_o} + \phi_o )}{\sin (K_oR)}
151: \left( \frac{R^{1/2}}{{r_o}^{1/2}}\right) \sin K_or \,\,\,\,\ \ r<R
152: \end{eqnarray}
153: where $ \nu = {(2m\alpha - 1/4)}^{1/2}$, $ \phi_o$ is the zero energy phase
154: \cite{Beane}, ${K_o}^2 = (2m\alpha_s)/{R^2}$ and $r_o$ is an arbitrary scale.
155:
156: The usual matching condition of the wave function and its derivative at
157: $r=R$ then yields:
158: \begin{equation} \label{trans}
159: {(2m\alpha_s)}^{1/2} \cot\{{(2m\alpha_s)}^{1/2}\} = \frac{1}{2} - \nu \tan
160: \left( \nu\ln \left( \frac{R}{r_o}\right) + \phi_o \right)
161: \end{equation}
162: Following \cite{Beane} we now consider eq. (\ref{trans}) to be a transcendental
163: equation {\it defining} the value of the short range coupling constant
164: $\alpha_s$. This equation is of the form $\beta \cot\beta = 1/\omega$ and can
165: be solved exactly in closed form using a method based upon the solution to the Riemann problem in complex variable analysis \cite{Mus}.
166:
167: The solution to eq.(\ref{trans}) then turns out to be \cite{Burn}:
168: \begin{eqnarray}
169: \beta_o &=& \pm \frac{{(\omega - 1)}^{1/2}}{\omega}
170: \exp\left(\frac{1}{\pi}\int_0^1 \arg \Lambda_o(t)\frac {dt}{t} \right) \, ,
171: \,\,\,\, \omega >0 \\
172: \beta_n &=& \pm n\pi \exp \left(\frac{1}{\pi}\int_0^1 \arg
173: \Omega_n(t)\frac {dt}{t} \right)
174: \, , \,\,\, -\infty<\omega < +\infty,\; n= 1,2,...
175: \end{eqnarray}
176: where:
177: \begin{eqnarray}
178: \beta &=& {(2m\alpha_s)}^{1/2}\\
179: \frac{1}{\omega} &=& \frac{1}{2} - \nu \tan \left(\nu \ln
180: \left( \frac{R}{r_o} \right) + \phi_o \right) \\
181: \Lambda_o(t) &=& \lambda (t) + \textstyle \frac{1}{2}\omega t i \pi \\
182: \lambda(t) &=& 1+\textstyle \frac{1}{2}\omega t \ln \frac {1-t}{1+t}\\
183: \Omega_n(t) &=& {\Lambda_o(t)}^2 + n^2{\pi}^2 {\omega}^2t^2
184: \end{eqnarray}
185: Formula (9) does not restrict $\omega$ to be positive, so that all the solutions that we consider follow from the positive branch of (7). If we interpret eq. (\ref{trans}) to be an equation
186: for the running coupling constant $\beta = {(2m\alpha_s)}^{1/2}$, this amounts to requiring that $\beta $ must be defined for all values of ${R}/{r_o}$ except for those points where $\omega$ (or its inverse) vanish. This is potentially important when comparing our analytic solution with {\it numerical} solutions to eq. (5). Indeed, numerical investigations in general will mix solutions that only exist in a limited range of ${R}/{r_o}$-values ($\beta _o$ in eq.(6)) with solutions ($\beta _n$ in eq.(7)) which we consider to be the only physically relevant solutions. From now on, we shall therefore use $\beta $ to mean only a solution of eq. (7), and suppress the subscript $n$ when it is not needed for the discussion. We then find
187: from eq. (7) that eq. (\ref{trans}) has infinitely many roots, in agreement
188: with \cite{Beane}. We have plotted $\beta $ in eq. (\ref{trans}) as a function of
189: $\ln x$ $ (x = {R}/{r_o})$ for fixed $ \phi_o =1.0$, and $n=1$ in figures 1 and
190: 2. The strength of the long range potential $\alpha/r^2$ increases with
191: succeeding figures; $\nu =0.5$ in fig. 1, $\nu = 3.0$ in fig. 2. The running coupling
192: $\beta$ exhibits a limit cycle behavior for all values of $\nu > 0$; the period
193: becomes smaller as the strength of the attractive $\alpha/r^2$ potential
194: increases, according to the argument ($ \nu \ln x + \phi_o $) of the tangent function in the source term of eq. (5). The behavior described by $\beta$, for large enough $\nu$, is of a
195: ``sawtooth" type, with a periodic sharp increase of the value of the coupling
196: constant with decreasing values of $\ln x$. This ``increase'' is actually a genuine discontinuity of $\beta $ at the zeros of $1/\omega $ and must be a multiple of $\pi $. Indeed, a discontinuity can only occur at a zero of $\cot \beta $ in order that $\beta \cot \beta $ be continuous at all points where $1/\omega $ is a continuous function. Once $\beta $ has reached the smallest positive zero ($\beta = \pi /2) $, for some $x-$value, it increases by $\pi $ as $x$ is further decreased. Altering the zero energy phase shift
197: $\phi_o$ from 0 through 2 (and keeping n=1 and $\nu=3$) does not qualitatively
198: change the appearance of the pattern of Fig. 2: the discontinuity moves to lower $x$ , but the magnitude of the discontinuity remains the same multiple of $\pi$. This feature can be
199: traced to the periodicity of the right hand side of eq. (\ref{trans}) with
200: respect to $\phi_o$. The branches which
201: correspond to roots with $n > 1$ are qualitatively similar to Fig 2., but the
202: slow fall-off with decreasing $x$ seen in Fig. 2 increases and the ``saw-tooth" appearance
203: becomes more of a rounded off square wave. This feature of the solutions is illustrated by Fig. 3 which plots $\beta $ for the same strength and initial phase $\nu=3$, $ \phi_o = 1$ as Fig. 2, but n has increased to 16. The magnitude of the discontinuity of this $\beta _{16}$ remains $\pi $, however.
204:
205: Finally we plot as Fig. 4 the analytical solution of
206: eq. (\ref{trans}) for the values $ \nu = 2.0$, $ \phi_o =0.0$ to compare with
207: the numerical solutions of eq. (\ref{trans}), with the same input, displayed in
208: Figure 1 of Ref. \cite{Beane}. The latter numerical solution is in excellent agreement
209: with the analytical solution presented here, if one allows a solution to go from a higher branch ($n=2$) to the next lower branch ($n=1$) as it crosses a zero of $1/\omega$. However, as discussed above and shown on Fig.4, $\beta _1 $ itself must ultimately increase by $\pi $ after reaching its lowest value $\pi/2$, thus exhibiting a limit cycle behavior of the solution.It is evident that the limit cycle behavior of the solutions of equation (5) is a consequence of the requirement that a solution $\beta_n$ be defined for all
210: values of $R/r_o$ for which omega or its inverse is nonzero. This requirement {\it includes solutions for $R \rightarrow 0$} and therefore corresponds to our
211: understanding of the emulation proposed in ref [3] of the contact term which encapsulates the short range dynamics of an EFT. From Figures 1-3, it is clear
212: that the limit cycle behavior of a given branch continues to the left as $R \rightarrow 0$ and lnx becomes arbitrarily small. Consider, however, the behavior
213: for small $R$ of a numerical solution of ref. [3], shown in Figure 4 which segues smoothly between different branches of the analytic solutions of equation
214: (5) as it crosses a zero of $1/\omega$. For some value of negative ln x, as $R$ becomes arbitrarily small, the numerical solution must pass to the $\beta_0$
215: solution. But, as we have noted, this solution exists for only a limited range of $R/r_o$ values. Thus, the two numerical solutions of Fig. 4 would appear to
216: not be defined for arbitrarily small $R/r_o$. One could, however, choose another numerical solution corresponding to a higher value of $n$ which does not
217: pass to $\beta_0$ on the way from large to arbitrarily small $R/r_o$ and avoid this problem. This exercise need not be performed, however, with our
218: requirement of a well defined solution for all values of $R/r_o$ for which omega or its inverse is nonzero.
219:
220:
221: A motivation for both the study of Ref. \cite{Beane} and the present
222: discussion is the expectation that the two-body inverse square potential is the
223: analogue to the interaction in a three-body system in the limit of zero energy
224: resonances. Indeed, the approximate solution of eq. (5) displayed in eq. 8
225: of \cite{Beane} is quite similar to the equation which describes the running of
226: the three-body counterterm of the pion-less three-nucleon EFT's of Refs.
227: \cite{BHK,EFTmisc}. Both equations have poles and the three-body counterterm
228: seems to reach arbitrarily high values. We, to the contrary, find no poles in
229: the analytic solutions of eq. (5). Furthermore, no evidence of multiple
230: branches was found in the renormalized pion-less three-nucleon problem. The
231: renormalization of short distance physics in these two problems needs more
232: understanding in light of the results of Efimov~\cite{Efim}.
233:
234: \begin{figure}
235: \centering
236: \includegraphics{fig1}
237: \caption{The running coupling constant $\beta$ as a function of $\ln x = \ln
238: \frac{R}{r_o}$ for $ \phi_o =1.0$, $\nu = 0.5$, $n=1$}
239: % \label{}
240: \end{figure}
241:
242:
243: \begin{figure}
244: \centering
245: \includegraphics{fig2}
246: \caption{The running coupling constant $\beta$ as a function of $\ln x $ for $
247: \phi_o =1.0$, $\nu = 3.0$, $n=1$}
248: % \label{}
249: \end{figure}
250:
251:
252: \begin{figure}
253: \centering
254: \includegraphics{fig3}
255: \caption{ $\beta$ as a function of $\ln x $ for $ \phi_o =1.0$, $\nu = 3.0, n
256: = 16$}
257: % \label{}
258: \end{figure}
259:
260: \begin{figure}
261: \centering
262: \includegraphics{fig4}
263: \caption{ $\beta$ as a function of $\ln x $ for $ \phi_o =0.0$, $\nu = 2.0$.
264: The three branches, from bottom to top, correspond to n=1 (dash-dotted
265: curve), n=2 (dotted curve) and n=3 (solid curve). As explained in the text, the discontinuity in $\beta$ is always $\pi$ for each curve. The two thicker curves are two
266: numerical solutions for the same parameters taken from ref. [3], as discussed in the text.}
267: \label{}
268: \end{figure}
269:
270:
271: Now we turn to the bound state aspects of the related two problems (three-body
272: system with contact potentials and the $1/r^2$ singular potential) and again
273: find discrepancies. First we show that the regularization method that was used
274: to solve the Schroedinger equation (1) with potential (2) amounts to specifying
275: a particular self-adjoint extension in Case's solution of the bound state
276: (B.S.) spectrum of the attractive singular $ {1}/{r^2}$
277: potential~\cite{Meetz,Case}. In order to do this, we note that the B.S.
278: wavefunction that solves eqs. (1)-(2) is given by:
279:
280:
281: \begin{eqnarray}
282: \psi &=& Cr^{1/2}K_{i\nu}(kr)\,\,\,\, r>R\\
283: \psi &=& C'\sin(Kr)\ \ \ \ \ r<R
284: \end{eqnarray}
285: where:
286: \begin{equation}
287: K^2 = \frac{2m\alpha_s}{R^2} - k^2
288: \end{equation}
289: and $C$ and $C'$ are constants.
290:
291:
292: For $kR <<1$, the matching condition now
293: gives (one still has $KR = {(2m\alpha_s)}^{1/2}$ in that limit):
294: \begin{equation} \label{case}
295: k= \frac{1}{2r_o}\exp \frac{\phi_o + \arg \Gamma (1+i\nu) - (n+1/2)\pi }{\nu}
296: \,\,\,\,\,\,\,\, n=0, \pm 1, \pm 2, ....
297: \end{equation}
298: where we have used eq. (\ref{trans}) together with the small-r behavior
299: of $\psi(r)$ \cite{PP}:
300: \begin{equation}
301: \psi(r) \simeq r^{1/2}\sin \left( \nu \log\frac{kr}{2}-\arg\Gamma
302: (1+i\nu)\right)\, .
303: \end{equation}
304:
305:
306:
307: The spectrum given in eq. (\ref{case}) is essentially the spectrum given by
308: Case \cite{Case}, where Case's arbitrary phase (which fixes the self-adjoint
309: extension) $B$ is now given by:
310:
311: \begin{equation}
312: B= \phi_o + \arg \Gamma (1+i\nu)
313: \end{equation}
314: It is important to note that the binding energy $E_B = (k^2)/{2m}$, after this
315: regularization, no longer
316: depends on the cut-off radius $R$ (for $kR << 1$) but instead on the arbitrary
317: scale $r_o$. Thus, fixing the zero energy phase of the wavefunction $\phi_o$
318: removes the cut-off dependence of the B.S. spectrum for $kR << 1$ . Our result
319: explicitly shows that the physical interpretation of the phase characterizing
320: the self-adjoint extension of the Hamiltonian indeed can be found from a
321: renormalization of the short-range coupling constant in the regularization
322: method described in Ref. \cite{Beane}. By the same token, it illustrates why a
323: cut-off method with a constant strength fails \cite{Meetz} to provide a
324: physical meaning for this arbitrary phase.
325:
326: Note, however, that the ground state of the $1/r^2$ potential remains at
327: negative infinity, and no renormalization of the bound state spectrum has been
328: achieved. Contrast this result of the EFT style renormalization of
329: Ref.~\cite{Beane} with the restoration of the lower bound of the pion-less
330: three-body problem obtained in Refs.~\cite{BHK,AHW}. A clue to this
331: discrepancy may lie in the distinction between the contact interaction used in
332: Ref. 2 and the ``EFT" type \cite{Beane} regularization of the conventional
333: 3-dimensional delta function. That is, the attractive square well in eq. (2)
334: does not provide a unique way of regularizing (``smearing out'') a
335: 3-dimensional delta function, and its limit when $ R \rightarrow 0$ is not
336: the contact interaction discussed in Ref. 2 and
337: 14. In that respect, it would be quite interesting to reexamine the solution of
338: the Schroedinger equation with a singular $\alpha/r^2$ potential in conjunction
339: with local realizations of the contact interaction of Ref. 14 implemented by
340: Kruppa, Varga and Revai ~\cite{KVR}. Such a study might throw additional light
341: on the corresponding renormalization group flow properties in the 3-body
342: problem.
343:
344:
345:
346: \section*{Acknowledgments}
347:
348:
349: The work of M.B. was supported by the National Fund for Scientific Research,
350: Belgium and that of S.A.C. by NSF grant PHY-0070938. We thank Silas Beane for
351: providing us with details of the numerical calculations of Ref. ~\cite{Beane}
352: and Mary Alberg for a communication about our plots of the analytical solutions.
353:
354:
355:
356:
357: \begin{thebibliography}{99}
358:
359: \bibitem{BHK} P.F Bedaque, H.-W. Hammer, and U. van Kolck, Phys. Rev. Lett.
360: {\bf 82}, 463 (1999) [nucl-th/9809025]; Nucl.Phys. {\bf A 646}, 444 (1999)
361: [nucl-th/9811046].
362:
363: \bibitem{AHW} S. Albeverio, R. H\O egh-Krohn, and T. T. Wu, Phys. Letters,
364: {\bf 83A}, 105 (1981).
365:
366: \bibitem{Beane} S. R. Beane, P. F. Bedaque, L. Childress, A. Kryjevski, J.
367: McGuire, and U. van Kolck, Phys. Rev. A {\bf 64} 042103 (2001)
368: [quant-ph/0010073].
369:
370: \bibitem{EFTmisc} P. F. Bedaque, H.-W. Hammer, and U. Van Kolck, Nucl. Phys.
371: {\bf A 676}, 357 (2000) [nucl-th/9906032]; H.-W.Hammer and T. Mehan, Nucl.
372: Phys. {\bf A 698},535 (2001) [nucl-th/0011024]; P. F. Bedaque, E. Braaten, and
373: H.-W.Hammer, Phys. Rev. Lett. {\bf 85}, 908 (2000); P. F. Bedaque, G. Rupak,
374: H. W. Greisshammer, and H.-W. Hammer,``Low energy expansion in the three-body
375: system to all orders and the triton channel'' [nucl-th/0207034].
376:
377: \bibitem{Efi} V. Efimov, Sov. J. Nucl. Phys. {\bf 12}, 589 (1971).
378:
379: \bibitem{Wil} K. G. Wilson, ``A limit cycle for three-body short range forces'',
380: talk presented at the INT program ``Effective Field Theory and Effective
381: Interactions'', Institute for Nuclear Theory, Seattle (2000).
382:
383: \bibitem{Bra} E. Braaten, H.-W. Hammer and M. Kusunoki, ``Universal Equation for
384: Efimov states'' [cond-mat/0201281].
385: \bibitem{BeBe} S.R. Beane,P.F. Bedaque, M.J. Savage and U. van Kolck, Nucl.Phys.{\bf A 700}, 377 (2002) [nucl-th/0104030].
386:
387: \bibitem{Birse} M. C. Birse, J. A. McGovern, and K. G. Richardson, Phys. Lett.
388: {\bf B464}, 169 (1999)[hep-ph/9807302]; T. Barford and M. C. Birse, ``A
389: renormalization group approach to two-body scattering in the presence of
390: long-range forces'' [hep-ph/0206146].
391: \bibitem{Sch} J.Denschlag and J. Schmiedmayer, Europhys.Lett. {\bf 38},405 (1997).
392: \bibitem{BaCo} M. Bawin and S.A. Coon, Phys.Rev. A{\bf 63},034701 (2001).
393: \bibitem{cam} H.E. Camblong, L.N. Epele, H. Fanchiotti and C.A. Garcia Canal, Phys. Rev. Lett. {\bf 87}, 220402 (2001)
394: \bibitem{CoH} S.A. Coon and Barry R. Holstein,Am. J. Phys.{\bf 70}, 513 (2002).
395: \bibitem{Efim} V. E. Efimov, Nucl. Phys. {\bf A 362}, 45 (1981); (E) {\bf 378},
396: 581 (1982); V. Efimov and E. G. Tkachenko, Few-Body Syst. {\bf 4}, 71 (1988).
397: For a review, see V. Efimov, Comments Nucl. Part. Phys. {\bf19}, 271 (1990).
398:
399: \bibitem{Meetz} K. Meetz, Nuovo Cimento {\bf 34}, 690 (1964).
400: \bibitem{Mus} Muskkhelishvili, N.I., Singular Integral Equations (Noordhoff, Groningen, The Netherlands, 1953).
401: \bibitem{Burn} E. E. Burniston and C. E. Siewert, Proc. Camb. Phil. Soc. {\bf
402: 73}, 111 (1973).
403:
404: \bibitem{Case} K. M. Case, Phys. Rev. {\bf80},797 (1950).
405:
406: \bibitem{PP} A. M. Perelemov and V. S. Popov, Teor. Mat. Fiz. {\bf 4},48 (1970)
407: [Theor. Math. Phys. {\bf 4}, 664 (1970].
408:
409: \bibitem{book} S. Albeverio, F. Gesztesy, R. H\O egh-Krohn, and H. Holden, {\it
410: Solvable Models in Quantum Mechanics} (Springer-Verlag,Berlin,1988).
411:
412: \bibitem{KVR} A. T. Kruppa, K. Varga, and J. R\'{e}vai, Phys. Rev C {\bf 63},
413: 064301 (2001).
414:
415: \end{thebibliography}
416: \end{document}
417:
418: