1: \def\be{\begin{equation}}
2: \def\ee{\end{equation}}
3: \def\bea{\begin{eqnarray}}
4: \def\eea{\end{eqnarray}}
5: \def\ie{\hbox{\it i.e.}{ }}
6: \def\etc{\hbox{\it etc.}{ }}
7: \def\eg{\hbox{\it e.g.}{ }}
8: \def\cf{\hbox{\it cf.}{ }}
9: \def\viz{\hbox{\it viz.}{ }}
10:
11: \documentclass[12pt]{article}
12: \textwidth=14.5cm
13: \textheight=21cm
14: \usepackage{epsfig,latexsym}
15: \begin{document}
16: \thispagestyle{empty}
17: \vspace*{-0.5 cm}
18: \vspace*{-1.2in}
19: \vspace*{0.7 in}
20: \begin{center}
21: {\large \bf Improving the Renormalization Group approach to}
22: \\
23: \vspace*{0.3 cm}
24: {\large \bf the quantum-mechanical double well potential}\\
25: \vspace*{2cm}
26: {\bf D. Zappal\`a\footnote{E-mail address: dario.zappala@ct.infn.it}
27: }\\ \vspace*{.3cm}
28: {\it INFN, Sezione di Catania}\\
29: {\it Dipartimento di Fisica, Universit\`a di Catania}\\
30: {\it Corso Italia 57, I-95129, Catania, Italy} \\
31:
32:
33:
34: \vspace*{1 cm}
35: {\bf ABSTRACT} \\
36: \end{center}
37: The gap between ground and first excited state of
38: the quantum-mechanical double well is calculated using the Renormalization
39: Group equations to the second order in the derivative expansion,
40: obtained within a class of proper time regulators.
41: Agreement with the exact results
42: is obtained both in the strong and weak coupling regime.
43: \\
44: \vskip 0.5 cm
45: \noindent
46: Pacs 11.10.Hi ; 03.65.Ca
47:
48: \parskip 0.3cm
49: \vspace*{3cm}
50: \vfill\eject
51: \setcounter{page}{1}
52: \voffset -1in
53: \vskip2.0cm
54:
55: \newcommand{\fa}{\phi^a}
56: \newcommand{\fb}{\phi^b}
57: \newcommand{\p}{\partial_{\mu}}
58: \newcommand{\dd}{\delta^{ab}}
59: \newcommand{\nn}{\nonumber}
60:
61: One of the most representative problems in quantum mechanics
62: is the tunnelling of a particle through a potential barrier.
63: It does not have a classical counterpart and cannot
64: be handled in the usual perturbative approach.
65: The double well potential provides a typical example
66: of quantum tunnelling and, when employed in field theory,
67: it also represents the most elementary toy model for
68: the spontaneous symmetry breaking phenomenon and the related
69: phase transitions and therefore it is extremely relevant in particle
70: physics and in cosmology.
71: The spectrum of the quantum mechanical double well can of course
72: be evaluated numerically and it is known that the energy gap $\Delta E$
73: between the first excited state and the ground state, which appears
74: as a consequence of the tunnelling between the two classical vacua,
75: has a singular behavior in the quartic coupling $\lambda$:
76: $\Delta E\propto {\rm exp}(-1/\lambda)$ \cite{cole}. The exponential
77: behavior is well reproduced by the dilute istanton gas calculation,
78: $\Delta E =2\sqrt{2 \sqrt{2}/(\pi\lambda)}{\rm exp}(-1/(3 \sqrt{2}\lambda))$,
79: which however is reliable only for very small values of $\lambda$
80: and becomes soon very different from the exact value of $\Delta E$ as
81: $\lambda$ grows.
82:
83: Some time ago it has been suggested \cite{aoki,tetradis} that $\Delta E$
84: could be evaluated by making use of the Renormalization Group (RG).
85: The original formulation of the non-perturbative RG obtained through
86: a recursion procedure is due to Wilson \cite{wil} and since then
87: a differential formulation of the RG flow equations has been developed
88: \cite{a1}-\cite{a6}. The RG equations involve the full action
89: of the problem considered and determine its flow
90: as a function of a momentum scale $k$, starting from the bare action
91: defined at a high momentum scale $k=\Lambda$ down to the infrared region
92: $k\sim 0$. In the $k\to 0$ limit the running action contains the effects
93: of all the fluctuations below the scale $\Lambda$ and can be identified with
94: the effective action.
95:
96: In \cite{aoki,tetradis} the equation for the full
97: action was approximated by an equation for the potential, which corresponds
98: to the lowest order approximation in a systematic derivative expansion
99: of the action. In \cite{aoki} a further approximation was considered
100: by expanding the potential in a polynomial series and then truncating this
101: series. The flow equation for the potential was then reduced to a set of
102: coupled ordinary differential equations for the various coefficients of
103: the series. In \cite{tetradis} the full partial differential equation
104: for the potential was instead studied with no further approximation.
105: In both cases the results are practically the same: when $k$ is lowered
106: the potential evolves from the bare double well shape,
107: reducing the barrier until it totally disappears.
108: When $k$ is close to zero the potential has become convex with the central
109: region between the original minima almost flat. This is already an interesting
110: feature because the convexity of the effective potential is an exact
111: property \cite{c1} which can be observed by making use of non-perturbative approaches
112: (see e.g. \cite{c2})
113: and cannot be recovered in perturbation theory.
114: On a more quantitative level, however the flow equation for the potential
115: is not fully satisfactory. In fact when it is used to compute the quantity
116: $\Delta E$, which as explained in \cite{aoki} is related to the second
117: derivative of the effective potential at the origin, it provides
118: accurate results for large values of the coupling $\lambda$,
119: but when approaching $\lambda=0$ the correct exponential behavior
120: is not recovered. Therefore the regime
121: in which the quantum tunnelling effects are relevant ($\lambda\sim 0$)
122: is not well described neither by the
123: approximation considered in \cite{aoki} nor by the one in \cite{tetradis}.
124:
125: In this Letter we examine the natural improvement on the lowest
126: order approximation of the flow equations in the derivative expansion,
127: and specifically we include the next term in this expansion, i.e. the
128: coefficient of the kinetic term of the action, $Z(k,x)$, which
129: depends both on the running scale $k$ and on the spatial coordinate $x$.
130: In field theory this is nothing else than a wave function renormalization
131: and in quantum mechanics it has the role of a position dependent effective
132: mass. We expect that the inclusion of this term,
133: which represents the first correction to the fully local information
134: carried by the potential could be helpful in describing the
135: complex dynamics of the tunnelling.
136:
137: Our aim is not only the analysis of a possible improvement in determining
138: $\Delta E$, due to the inclusion of $Z(k,x)$. In fact we are particularly
139: interested in checking the reliability of a particular version of the RG
140: flow obtained by means of a regulator introduced in the Schwinger Proper
141: Time (PT) formalism \cite{b1,b2,io1,io2}. As discussed in \cite{io2} and more
142: in detail in \cite{liti} this particular flow, unlike the so called
143: Exact Renormalization Group (ERG) flow,
144: does not have a first principle derivation
145: and there is no proof of convergence to the effective action
146: in the $k\to 0$ limit. However, as shown in \cite{io1,io2}, the PT RG
147: equations are particularly interesting because they provide excellent
148: determinations of the critical exponents at the non-gaussian fixed point
149: in three dimensions, certainly comparable to the ones obtained
150: with the ERG equations. The problem of determining $\Delta E$
151: in the quantum mechanical double well is therefore another good check
152: for the PTRG because at least in this case the exact results are available.
153:
154: In \cite{io1,io2} the flow equations of the PTRG are derived
155: in the required approximation of the derivative expansion of the action,
156: i.e. the coupled partial differential flow equations for the potential
157: and for the wave function renormalization are displayed for the particular
158: PT regulator
159: $f_k(s Z k^2)={\rm exp}(-s Z k^2)\sum^m_{i=0}{{(sZ k^2)^i}/ {i!}}$ where $s$
160: is the proper time and $m$ is an integer
161: index. Actually the best values for the critical exponents in \cite{io1,io2}
162: are obtained in the limit $m\to\infty$ and therefore here we are particularly
163: interested in this limit. In \cite{io2} the limit $m\to\infty$
164: is formally taken in the fixed point equations by redefining the field
165: variables. It is also noticed that the second derivative of
166: the potential determined by the $m$-dependent equations behaves
167: like $1/m$ for large $m$. This behavior would imply in the present case
168: that $\Delta E=0$ in the large $m$ limit. However this problem can be
169: avoided if one, instead of considering the field redefinition as in \cite{io2},
170: would simply start with a slightly different cut-off function
171: $\widetilde
172: f_k(s Z k^2)={\rm exp}(-s Z m k^2)\sum^m_{i=0}{{(sZ m k^2)^i}/ {i!}}$
173: where $k^2$ has been replaced by $m k^2$.
174: This is practically equivalent to what has been done in \cite{liti} where
175: the same rescaling of $k$ has been performed directly in the flow equations.
176: However the latter approach mixes the two limits $m\to \infty$ and $k\to 0$
177: and this could generate some confusion.
178:
179: If the new cut-off function $\widetilde f_k$ is used instead of $f_k$,
180: a new set of $m$-dependent flow equations are obtained whose large $m$ limit
181: (this time with no redefinition of the field or of the scale $k$)
182: yields again the equations which have been determined in \cite{io2}
183: and \cite{liti}. The $m$-dependent flow equations obtained with the cut-off
184: $\widetilde f_k$ as we shall see below provide for the double well problem,
185: in the limit $m\to \infty$
186: a non-vanishing second derivative of the potential and therefore
187: a finite value of $\Delta E$. The procedure outlined in
188: \cite{io1} to derive the flow equations, with the cut-off $f_k$ replaced by $\widetilde f_k$,
189: yields the two coupled partial differential equations for the
190: running potential $V(k,x)$ and the wave function renormalization $Z(k,x)$
191: \be
192: \label{eq:uflow}
193: k{{\partial V}\over{\partial k}} =\alpha (k^2 m)^{D/2} \Biggl({{Z k^2}\over{Z k^2+V''/m}}
194: \Biggr)^{m+1-D/2}
195: \ee
196: \bea
197: &&k{{\partial Z}\over{\partial k}} =\alpha (k^2 m)^{D/2}
198: \Biggl({{Z k^2}\over{Z k^2+V''/m}}\Biggr)^{m+1-D/2}\Biggl\lbrack{{(m+1-D/2)}\over{m(Z k^2+V''/m)}}
199: \biggl(-Z''
200: \nonumber\\
201: &&+{{(4+18D-D^2) (Z')^2}\over{24Z}}\biggr)
202: +{{(10-D)(m+1-D/2)(m+2-D/2)}\over{6 m^2(Z k^2+V''/m)^2}}Z'V'''
203: \nonumber\\
204: &&-{{(m+1-D/2)(m+2-D/2)(m+3-D/2)}\over{6m^3(Z k^2+V''/m)^3}}Z(V''')^2\Biggr \rbrack
205: \label{eq:zflow}\eea
206: where each prime indicates a derivative w.r.t. the spatial coordinate $x$ and
207: the constant $\alpha$ is expressed in terms of gamma functions
208: \be
209: \label{eq:alfa}
210: \alpha={{\Gamma(m+1-D/2)}\over{(4 \pi)^{D/2} \Gamma(m+1)}}
211: \ee
212: Eqs. (\ref{eq:uflow},\ref{eq:zflow}) in the limit $m\to \infty$ become
213: \be
214: \label{eq:ufminf}
215: k{{\partial V}\over{\partial k}} = \left({k^2\over{4\pi}}\right)^{D/2} e^{-V''/(Z k^2)}
216: \ee
217: \bea
218: &&k{{\partial Z}\over{\partial k}} =
219: \left({k^2\over{4\pi}}\right)^{D/2} e^{-V''/(Z k^2)}\nonumber\\
220: &&\times\Biggl(
221: -{{Z''}\over{Z k^2}}+{{(4+18D-D^2) (Z')^2}\over{24Z^2k^2}}
222: +{{(10-D)Z'V'''}\over{6 (Z k^2)^2}}-
223: {{Z(V''')^2}\over{6(Z k^2)^3}}\Biggr )
224: \label{eq:zfminf}
225: \eea
226: Note that all equations are expressed in terms of dimensionful
227: quantities (with $\hbar=1$).
228: The determination of $\Delta E$ requires the numerical integration
229: of the coupled equations for $V$ and $Z$ with $D=1$.
230: We have to fix the initial conditions of the differential equations
231: which we take at a value $k=\Lambda$, with $\Lambda$ much larger than
232: any other scale in the problem. The double well potential
233: \be
234: \label{eq:dw}
235: V_{dw}(x)={1\over 2}M^2 x^2+\lambda x^4
236: \ee
237: corresponds to the initial condition of the flow of the potential $V(k,x)$
238: whereas, according to the normalization of the kinetic term in the bare action,
239: we take $Z=1$ as initial condition for $Z(k,x)$.
240: The scale of all the dimensionful quantities is fixed by the choice
241: $M^2=-1$. We shall also consider, for comparison, the convex
242: anharmonic oscillator, and in this case we choose $M^2=1$.
243: Correspondingly the scale at which the initial conditions
244: of the parameters are fixed is chosen $\Lambda=1500$.
245: The numerical resolution of the two coupled partial differential
246: equations is performed with the help of the NAG routines.
247:
248: In Fig. 1 $V''(0,0)$ and $Z(0,0)$ obtained from Eqs. (\ref{eq:uflow},\ref{eq:zflow})
249: are reported versus $1/m$. Namely
250: black circles and diamonds correspond respectively
251: to $Z(0,0)$ and $V(0,0)$ for $M^2=1$ and $\lambda=0.4$.
252: White circles and diamonds correspond to
253: $Z(0,0)$ and $V(0,0)$ for $M^2=-1$ and $\lambda=0.05$.
254: In all cases we have inserted at $1/m=0$ the corresponding
255: values obtained from Eqs. (\ref{eq:ufminf},\ref{eq:zfminf}).
256: The convergence to these latter values
257: is clear. Therefore, as we had anticipated, in the limit
258: $m\to \infty$, Eqs.(\ref{eq:uflow},\ref{eq:zflow}) provide non-vanishing values of the second
259: derivative of the potential which converge to those obtained
260: from Eqs.(\ref{eq:ufminf},\ref{eq:zfminf}).
261:
262: Since we are particularly interested in the $1/m=0$ case,
263: from now on we consider Eqs.(\ref{eq:ufminf},\ref{eq:zfminf}).
264: The qualitative behavior of the running potential along the flow is
265: explained in detail in \cite{tetradis} and we observe a similar trend.
266: In Fig. 2 the second derivative of the running potential (continuous lines)
267: and the wave function renormalization (dashed lines) obtained at various
268: values of $k$ (namely $k=1500,10,1,0.5,0.2,0.1,0$) for $M^2=-1$ and
269: $\lambda=0.06$ are plotted. Below
270: $k=0.2$ the second derivative of the potential
271: at the origin becomes positive and the potential convex.
272: At the same time $Z$, which for most of the running stays close to one,
273: rapidly increases in the region between the classical minima with a sharp
274: fall to one outside that region. The highest peak corresponds to the final
275: output at $k=0$.
276:
277: Finally we turn to the evaluation of the energy gap $\Delta E$.
278: As explained in \cite{aoki}, as long as the potential is analysed keeping
279: $Z=1$ fixed during the flow, $\Delta E$ is to be identified with
280: $\sqrt{V''(0,0)}$,
281: but in the approximation here considered with a running wave function
282: renormalization, the energy gap $ \Delta E$ corresponds,
283: in field theory language, to the renormalized mass, i.e.
284: \be
285: \label{eq:de}
286: \Delta E= \sqrt{V''(0,0)/Z(0,0)}
287: \ee
288: Therefore we determined the values of $\Delta E$ from
289: Eqs.(\ref{eq:ufminf},\ref{eq:zfminf}) according to Eq. (\ref{eq:de})
290: (these results are indicated as $\Delta E_{PTnlo}$) and,
291: for comparison, we also evaluated $\Delta E$ to the lowest
292: order in the derivative expansion (indicated as $\Delta E_{PTlo}$),
293: i.e. solving the flow in
294: Eqs.(\ref{eq:ufminf},\ref{eq:zfminf})
295: keeping the wave function renormalization constant $Z=1$.
296: The energy gap to the lowest order
297: in the derivative expansion with the flow equation
298: used in \cite{tetradis}, which is identical to the Wegner-Houghton local
299: potential equation in $D=1$ (see \cite{a1})
300: \be
301: \label{eq:whe}
302: k{{\partial V}\over{\partial k}} = {-k \over {2\pi}} {\rm log}\left(1+ {V''\over {k^2}}\right)
303: \ee
304: has also been calculated ($\Delta E_{wh}$) and
305: all these results are eventually compared
306: with the exact estimate $\Delta E_{exact}$, derived by numerically
307: solving the quantum mechanical eigenvalue problem.
308: In Table 1 these estimates of $\Delta E$ are collected
309: for various values of the coupling $\lambda$ both for $M^2=1$ and
310: $M^2=-1$. The values found for $Z(0,0)$ are also included.
311: From the accuracy of the numerical code the estimated
312: errors in Table 1 are at most two or three units on the last digit.
313:
314: In the $M^2=1$ case, as expected from \cite{aoki,tetradis}, everything is
315: rather smooth. In fact we observe that
316: the differences among the various
317: determinations of $\Delta E$ are extremely small and $Z(0,0)$ is practically equal
318: to its bare value $Z(\Lambda,x)=1$. However little discrepancies appear
319: at large values of $\lambda$ (strong coupling regime) and $\Delta E_{wh}$
320: turns out to be a slightly better approximation
321: than $\Delta E_{PTlo}$ and to be comparable to
322: $\Delta E_{PTnlo}$.
323:
324: The double well problem with $M^2=-1$ is rather different. Here,
325: from \cite{aoki,tetradis} we know that the RG equations work better
326: in the strong coupling regime. Actually this feature is true for
327: $\Delta E_{wh}$ which, in the region where the tunnelling plays an
328: important role ($\lambda<0.1$), does not reproduce the exponential
329: behavior of $\Delta E_{exact}$. On the other side, $\Delta E_{PTlo}$
330: which is less accurate at $\lambda=0.4$, turns out to be more reliable
331: at small values of $\lambda$. But the remarkable result is that
332: the higher order approximation is very accurate. In fact the correction
333: due to the wave function renormalization increases when $\lambda$ is
334: decreased and the agreement between $\Delta E_{PTnlo}$ and
335: $\Delta E_{exact}$ is always excellent.
336: Unfortunately below $\lambda=0.05$ the peak in $Z(0,x)$ shown
337: in Fig. 2 becomes very narrow, generating problems in the numerical
338: resolution of the flow equations and we were not able to determine
339: the corresponding values of $Z(0,0)$ and of the energy gap.
340: In any case for such small values of $\lambda$ the numerical errors
341: are no longer negligible and a more refined analysis should be performed.
342: Even an approximated analytical investigation of this regime,
343: similar to the one discussed in \cite{tetradis}, would be useful,
344: although the problem of two coupled equations is much more difficult
345: to handle.
346:
347:
348: In conclusion we have determined the energy gap between the ground state
349: and the first excited state of the quantum mechanical potential $V_{dw}$ in
350: Eq. (\ref{eq:dw}) both for positive (quartic anharmonic potential)
351: and for negative (double well potential) curvature $M^2$, by making use
352: of the RG techniques. In particular we used
353: a Proper Time regulated version of the RG
354: flow equations in the derivative expansion
355: approximation, truncated to the second order.
356: The results show the reliability of the PT RG flow
357: which, even to the lowest order provides good estimates
358: of the energy gap and, above all, they show the
359: sensible improvement coming from the inclusion
360: of the wave function renormalization in the
361: quantitative analysis of a tunnelling process.
362:
363:
364: \flushleft{\bf Acknowledgement}
365:
366: The author is grateful to Alfio Bonanno for many fruitful discussions.
367:
368: \begin{thebibliography}{12}
369:
370: \bibitem{cole} S. Coleman, {\it Aspects of Symmetry},
371: (Cambridge University Press), Cambridge, 1985.
372: \bibitem{aoki} K.-I. Aoki, A. Horikoshi, M. Taniguchi and H. Terao, in the
373: proceedings of the Workshop on {\it The Exact Renormalization Group},
374: Faro, Portugal, September 1998, (World Scientific, Singapore 1999),
375: arXiv:hep-th/9812050.
376: \bibitem{tetradis} A. S. Kapoyannis and N. Tetradis, Phys. Lett. {\bf A 276}, 225, (2000).
377: \bibitem{wil} L.P. Kadanoff, Physica {\bf 2}, 263, (1966);
378: K.G. Wilson, Phys. Rev. {\bf B4}, 3174 and 3184, (1971);
379: K.G. Wilson and M.E. Fisher, Phys. Rev. Lett. {\bf 28}, 240, (1972);
380: K.G. Wilson and J. Kogut, Phys. Rep. {\bf 12}, 75, (1974).
381: \bibitem{a1} F. J. Wegner and A. Houghton, Phys. Rev {\bf A8}, 401 (1973).
382: \bibitem{a2}
383: J.F. Nicoll, T.S. Chang and H.E. Stanley, Phys. Rev. Lett. {\bf 33}, 540 (1974);
384: Phys. Rev. {\bf A 13}, 1251 , (1976);
385: J.F. Nicoll and T.S. Chang, Phys. Lett. {\bf A62}, 287, (1977);
386: T.S. Chang, D. D. Vvedensky and J.F. Nicoll, Phys. Rep. {\bf 217}, 280, (1992).
387: \bibitem{a3} J. Polchinski, Nucl. Phys. {\bf B 231}, 269 (1984).
388: \bibitem{a4} A. Hasenfratz and P. Hasenfratz, Nucl Phys. {\bf B270},
389: 685 (1986).
390: \bibitem{a5}
391: C. Wetterich, Nucl. Phys. {\bf B352}, 529, (1991);
392: Z. Phys. {\bf C57}, 451, (1993); {\bf C60} 461, (1993);
393: Phys Lett. {\bf B301}, 90, (1993).
394: \bibitem{a6} T. Morris, Int. J. Mod. Phys. {\bf A 9} 2411, (1994);
395: Phys. Lett. {\bf B329}, 241, (1994).
396: \bibitem{c1} K. Symanzik, Commun. Math. Phys. {\bf 16}, 48, (1970);
397: J. Iliopulos, C. Itzykson and A. Martin, Rev. Mod. Phys. {\bf 47}, 165, (1975);
398: T.L. Curtright and C.B. Thorn, Jour. Math. Phys. {\bf 25}, 541, (1984).
399: \bibitem{c2} D.J.E. Callaway and D.J. Maloof, Phys. Rev. {\bf D27}, 1948, (1983);
400: D.J.E. Callaway, Phys. Rev. {\bf D27}, 2974, (1983);
401: V. Branchina, P. Castorina and D. Zappal\`a, Phys. Rev. {\bf D41}, 1948, (1990).
402: \bibitem{b1} M. Oleszczuk, Z. Phys. {\bf C64}, 533, (1994).
403: \bibitem{b2} S.-B. Liao, Phys. Rev. {\bf D53}, 2020, (1996).
404: \bibitem{io1} A. Bonanno and D. Zappal\`a, Phys. Lett. {\bf B504}, 181, (2001).
405: \bibitem{io2} M.Mazza and D. Zappal\`a, arXiv:hep-th/0106230,to appear in
406: Phys. Rev. {\bf D}.
407: \bibitem{liti} D.F. Litim and J.M. Pawlowski, Preprint: CERN-TH-2001-169, FAU-TP3-01-05,
408: arXiv:hep-th/0107020.
409: \end{thebibliography}
410: \vfill\eject
411: \begin{figure}
412: \psfig{figure=sol1.eps,height=13.5 true cm,width=14.5 true cm,angle=-90}
413: \caption{
414: $V''(0,0)$ ( black and white diamonds) and $Z(0,0)$ (black and white circles) obtained
415: from Eqs.(\ref{eq:uflow},\ref{eq:zflow}) respectively for $M^2=1$, $\lambda=0.4$
416: and for $M^2=-1$, $\lambda=0.05$, {\it vs} $1/m$.
417: The limiting values at $1/m=0$ are obtained from Eqs. (\ref{eq:ufminf},\ref{eq:zfminf}).
418: }
419: \end{figure}
420: \vfill\eject
421:
422: \begin{figure}
423: \psfig{figure=sol4.eps,height=13.5 true cm,width=14.5 true cm,angle=-90}
424: \caption{
425: $V''(k,x)$ (continuous lines) and $Z(k,x)$ (dashed lines) plotted {\it vs} $x$
426: at $k=1500,10,1,0.5,0.2,0.1,0$, as obtained from Eqs. (\ref{eq:ufminf},\ref{eq:zfminf})
427: for $M^2=-1$ and $\lambda=0.06$.
428: }
429: \end{figure}
430: \vfill\eject
431:
432: \begin{center}
433: TABLE 1
434: \end{center}
435: \vspace*{.8 cm}
436: \begin{table} [h] \centering{
437: \begin{tabular}{|c||c|c|c|c||c|} \hline
438: \hline
439: $\lambda$&$\Delta E_{wh}$&$\Delta E_{PTlo}$ & $\Delta E_{exact}$
440: & $\Delta E_{PTnlo}$ & $Z(0,0)$\\
441: \hline
442: \hline
443: \multicolumn{6}{|c|}{$M^2=1$}\\
444: \hline
445: $1.0 $&$ 1.9291 $&$ 1.9464 $&$ 1.9341 $&$ 1.9380 $&$1.0052 $\\
446: $0.4 $&$ 1.5450 $&$ 1.5556 $&$ 1.5482 $&$ 1.5498 $&$1.0037 $\\
447: $0.1 $&$ 1.2091 $&$ 1.2127 $&$ 1.2104 $&$ 1.2109 $&$1.0013 $\\
448: $0.05 $&$ 1.1201 $&$ 1.1218 $&$ 1.1208 $&$ 1.1210 $&$1.0006 $\\
449: $0.03 $&$ 1.0774 $&$ 1.0784 $&$ 1.0779 $&$ 1.0780 $&$1.0003 $\\
450: $0.02 $&$ 1.0538 $&$ 1.0544 $&$ 1.0540 $&$ 1.0542 $&$1.0002 $\\
451: \hline
452: \multicolumn{6}{|c|}{$M^2=-1$}\\
453: \hline
454: $0.4 $&$ 0.9654 $&$ 0.9897 $&$ 0.9667 $&$0.9730 $&$1.0217 $\\
455: $0.3 $&$ 0.8173 $&$ 0.8404 $&$ 0.8166 $&$0.8233 $&$1.0273 $\\
456: $0.2 $&$ 0.6212 $&$ 0.6416 $&$ 0.6159 $&$0.6227 $&$1.0416 $\\
457: $0.1 $&$ 0.3297 $&$ 0.3280 $&$ 0.2969 $&$0.3027 $&$1.1321 $\\
458: $0.07 $&$ 0.2238 $&$ 0.1848 $&$ 0.1539 $&$0.1562 $&$1.3343 $\\
459: $0.06 $&$ 0.1902 $&$ 0.1311 $&$ 0.1031 $&$0.1028 $&$1.5548 $\\
460: $0.05 $&$ 0.1576 $&$ 0.0806 $&$ 0.0562 $&$0.0532 $&$2.1270 $\\
461: $0.04 $&$ 0.1259 $&$ 0.0496 $&$ 0.0210 $&$ ---- $&$ ---- $\\
462: $0.03 $&$ 0.0947 $&$ 0.0329 $&$ 0.0036 $&$ ---- $&$ ---- $\\
463: $0.02 $&$ 0.0637 $&$ 0.0204 $&$ 0.0003 $&$ ---- $&$ ---- $\\
464: \hline
465: \end{tabular}
466: }
467: \caption{
468: Various determinations of $\Delta E$, including its exact estimate, and of
469: $Z(0,0)$ for some values of the coupling $\lambda$, both in the $M^2=1$
470: and in the $M^2=-1$ case.
471: }
472: \end{table}
473: \end{document}
474:
475: