1: \documentstyle[aps,psfig,eqsecnum,twocolumn]{revtex}
2: \setlength{\evensidemargin}{0.5cm}
3: \setlength{\oddsidemargin}{-1.cm}
4: \renewcommand{\theequation}{\arabic{equation}}
5: %\renewcommand{\baselinestretch}{1.021}
6: \renewcommand{\baselinestretch}{1.04}
7:
8:
9: %%%%%%%%%%%%%% Some Macros %%%%%%%%%%%%%%%%%%%%%%%%
10: \def\di{\displaystyle}
11: \def\tab{&\di}
12: \def\step{}
13:
14: \def\eq#1{(\ref{#1})}
15: \def\Eq#1{Eq.~(\ref{#1})}
16: \def\Es#1{Eqs.~(\ref{#1})}
17: \def\tr{{\rm tr}}
18: \def\Tr{{\rm Tr}}
19: \def\STr{{\rm STr}}
20: \def\ov{\over}
21: \def\s0#1#2{\mbox{\small{$ \frac{#1}{#2} $}}}
22: \def\0#1#2{\frac{#1}{#2}}
23: \def\fig#1{fig.~(\ref{#1})}
24: \def\eq#1{(\ref{#1})}
25: \def\Eq#1{Eq.~(\ref{#1})}
26: \def\llangle{\left\langle}
27: \def\rrangle{\right\rangle}
28: \def\ad{{\rm ad}}
29:
30:
31: %%%%%%%%%% Set-up %%%%%%%%%%%%
32:
33: %\date{\today}
34:
35:
36:
37:
38: \makeatletter
39:
40: \renewenvironment{thebibliography}[1]
41: {\frenchspacing\small
42: \begin{list}{[\arabic{enumi}]}
43: {\usecounter{enumi}\parsep=2pt\topsep 0pt
44: \settowidth{\labelwidth}{[#1]}
45: \leftmargin=\labelwidth\advance\leftmargin\labelsep
46: \rightmargin=0pt\itemsep=0pt\sloppy}}{\end{list}}
47:
48: \makeatother
49:
50:
51:
52:
53:
54:
55:
56:
57: \begin{document}
58: \twocolumn[\hsize\textwidth\columnwidth\hsize\csname
59: @twocolumnfalse\endcsname
60: \title{Perturbation theory and renormalisation group equations}
61:
62: \author{
63: \hfill %%%% required for preprint number
64: Daniel F. Litim\,${}^*$
65: and
66: Jan M. Pawlowski\,${}^\dagger$
67: %%%% the following two lines for the preprint number
68: \hfill
69: \raisebox{23mm}[0mm][0mm]{\makebox[0mm][r]{
70: CERN-TH-2001-327\ \ FAU-TP3-01-10\ \ {\tt hep-th/0111191}}}%
71: }
72: \address{${}^*${\it
73: Theory Division,
74: CERN,
75: CH-1211 Geneva 23.
76: }\\ ${}^\dagger${\it
77: Institut f\"ur Theoretische Physik III,
78: Universit\"at Erlangen,
79: D-91054 Erlangen.
80: }}
81:
82: \maketitle
83:
84:
85: \begin{abstract}\noindent
86: {We discuss the perturbative expansion of several one loop improved
87: renormalisation group equations. It is shown that in general the
88: integrated renormalisation group flows fail to reproduce
89: perturbation theory beyond one loop.
90: \\
91:
92: {PACS: 05.10.Cc,\ 11.10.Gh\ \ \
93: %05.10.Cc Renormalization group methods
94: %11.10.Gh Renormalization
95: E-mail: Daniel.Litim@cern.ch\,, \ jmp@theorie3.physik.uni-erlangen.de}
96:
97:
98: }
99: \end{abstract}
100: \vskip2.pc]
101:
102:
103: %********|*********|*********|*********|*********|*********|*********|****
104: \noindent
105: {\bf 1. Introduction}\\[-1ex]
106: %********|*********|*********|*********|*********|*********|*********|****
107:
108: \noindent
109: Renormalisation group (RG) methods are an essential ingredient in the
110: study of non-perturbative problems in continuum and lattice
111: formulations of quantum field theory. A number of RG equations have
112: been proposed, where the starting point is the (infrared) regularised
113: one loop effective action. Taking the derivative w.r.t.\ the infrared
114: scale together with a subsequent one loop improvement leads to a flow
115: for the effective action. The merit of such an equation is its
116: flexibility, as it allows for non-perturbative approximations not
117: bound to the weak coupling regime. Thus, these flows are particularly
118: interesting for theories where one has to resort to truncations
119: because the full problem is too hard to attack. Indeed, surprisingly
120: good results concerning critical exponents in scalar theories have
121: been obtained within simple approximations to a particular version of
122: a one loop improved RG \cite{Litim:2001hk}, based on a proper-time
123: representation of the one loop effective action \cite{Liao:1996fp}. It
124: has also been suggested that the proper-time RG may be an interesting
125: tool for gauge theories, since the regularisation respects a local
126: non-Abelian gauge symmetry \cite{Liao:1997nm}.\step
127:
128: However, results obtained within a truncated system are only as good
129: as the accompanying quality checks. Apart from the inherent problems
130: of these checks, the present situation requires additional care, since
131: most of the one loop improved RG lack a first principle derivation.
132: Such flows suffer from a severe conceptual problem. It is unclear,
133: whether they are only approximations to flows for the full effective
134: action or whether they represent an exact flow. The latter is indeed
135: known to hold true for Exact RG (ERG) flows \cite{Polchinski,CW}
136: (for reviews see \cite{Bagnuls:2000ae}). They can be obtained within a
137: one loop improvement, but also from a first principle derivation,
138: mostly done within a path integral representation. The strength of
139: exact RG flows is that systematic approximations of the integrated
140: flow correspond to systematic approximations to the full quantum
141: theory. This property, in combination with the convergence behaviour
142: of the flow, is at the root of the predictive power of the
143: formalism. The similarity of the different one loop improved flows,
144: including ERG flows, has fuelled hopes that the scenario just
145: described for exact flows may be valid in general. \step
146:
147: Based on this picture, and prior to an application of a general
148: one loop improved flow to any physical problem, it is mandatory to
149: either prove that a given flow is exact, or to unravel its inherent
150: approximations. A way to settle these questions consist in a detailed
151: comparison of one loop improved flows with known exact flows. Within
152: the derivative expansion, this has been studied in
153: \cite{Litim:2001hk}. In this note, we take a different route and
154: study one loop improved RG equations within perturbation theory. It
155: is shown that they only represent, in general, approximations to flows
156: in the full theory. This result is achieved by a structural analysis
157: of the flows, and by calculating the diagrammatic representation of
158: the two loop contributions to the effective action generated by the
159: flow through an iterative formal integration. In general neither the
160: graphs nor the combinatorial factors of the two loop diagrams that
161: originate from one loop improved flows, are the correct ones. A full
162: account of the present calculation together with a discussion of
163: related issues will be presented in \cite{consistent}.
164: \\[2ex]
165:
166:
167: %********|*********|*********|*********|*********|*********|*********|****
168: \noindent
169: {\bf 2. One loop improved renormalisation group}\\[-1ex]
170: %********|*********|*********|*********|*********|*********|*********|****
171:
172: \noindent
173: We briefly review the philosophy of a one loop improved
174: renormalisation group. The starting point is the formal equation for
175: the one loop effective action:
176: \begin{eqnarray}\label{1loop}
177: \Gamma^{\rm 1-loop}=S_{\rm cl}
178: +\s012 \Tr\ln S^{(2)}\,.
179: \end{eqnarray}
180: The trace in \eq{1loop} is ill-defined and requires -at least- an UV
181: regularisation. A one loop improved RG is derived from \eq{1loop} by
182: first employing an explicit regularisation, taking the derivative
183: w.r.t.\ the cut-off scale $k$ and then substituting $S^{(2)}$ by
184: $\Gamma^{(2)}$. Here, we
185: concentrate on infrared regularisations; this does not make a
186: difference for the flow itself, which in either case
187: should be local in momentum space, e.g.~only a small momentum
188: range about $q^2\approx k^2$ contributes to the flow at fixed $k$.
189: \step
190:
191: Let us start with the derivation of the ERG flow
192: \cite{Polchinski,CW,Bagnuls:2000ae}. Adding an infrared regulator $R$
193: (a momentum dependent mass term) to $S^{(2)}$ in \eq{1loop} and
194: proceeding according to the one loop improvement philosophy, we arrive
195: at
196: \begin{eqnarray}\label{exact}
197: \partial_t \Gamma_k=\s012 \Tr \left(\Gamma_k^{(2)}+R\right)^{-1}
198: \partial_t R\, ,
199: \end{eqnarray}
200: where $t=\ln k$ is the logarithmic infrared scale introduced via $R$.
201: The regulator $R$ has to meet some requirements as a function of
202: momentum and the cut-off scale, which are discussed at length in the
203: literature. For our purpose these consistency requirements are
204: irrelevant, since we only want to perform iterative formal
205: integrations.
206: \step
207:
208: We emphasise that a general exact flow is the flow of some operator
209: insertion within the theory. A first principle derivation of the ERG,
210: for example, is based on the insertion $\s012 \int \phi R\phi$.
211: Insisting on the one loop nature of the flow, one is {\it bound} to an
212: insertion which is at most quadratic in the fields. Otherwise, the
213: corresponding exact flow would also contain higher loop contributions.
214: We conclude that an exact flow with a
215: one loop structure must depend {\it linearly} on the full propagator.
216: This is indeed the case for the ERG flow \eq{exact}.\step
217:
218: Another possibility for regularising the expression in \eq{1loop} is
219: to modify the trace itself by inserting an operator
220: $\rho$ multiplicatively \cite{Liao:2000yu}. This amounts to the
221: replacement $\Tr \ln S^{(2)}\to \Tr \rho \ln S^{(2)}$ in \eq{1loop}
222: and leads to the one loop improved RG flow
223: \begin{eqnarray}\label{operator}
224: \partial_t \Gamma_k=\s012 \Tr\ \partial_t\rho\ \ln \Gamma_k^{(2)}\,.
225: \end{eqnarray}
226: The multiplicative structure of this flow is particularly convenient,
227: when used in numerical applications. Note, that opposed to \eq{exact},
228: the flow \eq{operator} depends on the logarithm of $\Gamma_k^{(2)}$.
229: Based on this structure, we can already conclude that \eq{operator}
230: cannot be exact. \step
231:
232: Finally we consider a regularisation based on a proper-time
233: representation of \eq{1loop},
234: \begin{eqnarray}\label{RegDef}
235: \Gamma^{\rm 1-loop}=S_{\rm cl}
236: -\s012 \int\0{ds}{s} \ \Tr\ \exp \left(-s\,S_{\rm cl}^{(2)}\right)\,.
237: \end{eqnarray}
238: Now we multiply the integrand in \eq{RegDef} by a regularising
239: function $f(s\,\Lambda^2)-f(s\,k^2)$ \cite{Oleszczuk:1994st}.
240: Proceeding along the lines of the one loop improvement we arrive at
241: \cite{Liao:1996fp}
242: \begin{eqnarray}\label{PTRG}
243: \partial_t \Gamma_k= \s012 \int_0^\infty \0{ds}{s}\
244: \partial_t f \ \Tr\exp\left(-s\Gamma^{(2)}_k\right)\,.
245: \end{eqnarray}
246: In order to facilitate the perturbative calculations below, we cast
247: the flow equation \eq{PTRG} in a form which is more convenient for
248: this purpose. This alternative representation also reveals more clearly
249: the structure of the proper-time flows. To that end, we expand a
250: general proper-time flow in the following basis set of regulator
251: functions $f$:
252: \begin{eqnarray}
253: \label{fm}
254: \partial_t f(x;m)& = & \0{2}{\Gamma(m)} x^m \exp\left({-x}\right)\,.
255: \end{eqnarray}
256: Here, $x=k^2 s$. Note that the IR behaviour is controlled by the term
257: $e^{-x}$, where $x$ serves as a mass. These flows cover all
258: proper-time flows that have been studied in the literature
259: \cite{Litim:2001hk,Liao:1996fp,Liao:1997nm,Floreanini:1995aj,Schaefer:1999em,Meyer:2000bz,Papp:2000he,Bohr:2001gp,Meyer:2001zp,Bonanno:2001yp,Mazza:2001bp,Zappala:2001nv}.
260: Moreover, linear combinations $\sum_m d_m\, f(x;m)$ of \eq{fm} with
261: $\sum_m d_m=1$ cover all flows with mass-like IR behaviour.
262: The trace in \eq{PTRG} can be written in terms of the normalised
263: eigenfunctions $\Psi_n$ of $\Gamma^{(2)}_k$ with $ \Gamma^{(2)}_k
264: \Psi_n=\lambda_n \Psi_n$. Within this representation we deal with
265: simple $s$-integrals. By performing the $s$-integration we arrive at
266: \cite{consistent}
267: \begin{eqnarray}\label{PTRG-CS}
268: \partial_t \Gamma_k=\Tr \left({k^2\over \Gamma_k^{(2)}+k^2}\right)^m.
269: \end{eqnarray}
270: The operator kernel inside the trace is the $m$th power of a
271: Callan-Symanzik kernel. We note that the {\it functional dependence}
272: of \eq{PTRG-CS} on $\Gamma^{(2)}$ depends on the regularisation.
273: Above, we have argued that an exact one loop flow has to depend
274: linearly on the full propagator. Hence, \eq{PTRG-CS} is not exact for
275: $m\neq 1$ due to the non-linear dependence of \eq{PTRG-CS} on the full
276: propagator.\step
277:
278: In addition, \eq{PTRG-CS} also signals that, at least in perturbation
279: theory, the deviation of a general proper-time flow from an exact flow
280: is regularisation-dependent. In contrast, for both \eq{exact} and
281: \eq{operator}, the functional dependence on $\Gamma^{(2)}$ and,
282: thus, the result of their formal integration is independent of
283: the regularisation. For \eq{PTRG}, however, linear combinations of
284: \eq{PTRG-CS} span the space of all kernels which decay at least as
285: $(\Gamma^{(2)}+k^2)^{-1}$ and reproduce the one loop effective action.
286: A general kernel trivially leads to a non-unique endpoint of
287: the flow. This result also implies that \eq{PTRG}, in general, is not
288: an exact flow.
289: \\[2ex]
290:
291: %********|*********|*********|*********|*********|*********|*********|****
292: \noindent
293: {\bf 3. Effective action at one loop}\\[-1ex]
294: %********|*********|*********|*********|*********|*********|*********|****
295:
296: \noindent
297: Thus, prior to any use of the flows \eq{operator} and \eq{PTRG}, it is
298: mandatory to collect more information on their inherit deviation from
299: exact flows. Here, this is done by explicitly calculating one loop and
300: two loop effective actions following from the flows.
301: This also serves as an independent proof of our general statements.
302: We restrict ourselves to a scalar
303: theory with one species of fields, but with general interaction. The
304: results are easily generalised to arbitrary field content. As the
305: flows \eq{exact}, \eq{operator} and \eq{PTRG} are derived as one loop
306: improved flows from the one loop effective action \eq{1loop}, their
307: integrals reproduce the one loop effective action in the limit, where
308: the infrared cut-off tends to zero. It is instructive to see how this
309: comes about. The one loop contribution $\Delta\Gamma_1$ is given by
310: \begin{eqnarray}\label{formal1loop}
311: \Delta\Gamma_1= \int_\Lambda^k {d k'\over k'}
312: \left(\partial_{t'}\Gamma_{k'}\right)_{\rm 1-loop}\, .
313: \end{eqnarray}
314: Here, $\left(\partial_{t'}\Gamma_{k'}\right)_{\rm 1-loop}$
315: stands for the right-hand sides in either of the flow equations \eq{exact},
316: \eq{operator} or \eq{PTRG}, with $\Gamma_k^{(2)}$
317: substituted by $S^{(2)}$. This is sufficient to obtain the effective
318: action at one loop. \step
319:
320: Consequently, integrating the ERG flow \eq{exact} leads to
321: \begin{eqnarray}\label{exact1loop}
322: \Delta\Gamma_1=
323: \s012 \Tr \, \left[\ln (S^{(2)}+R)\right]_\Lambda^k\, .
324: \end{eqnarray}
325: Note that even for $k\neq 0$ the expression functionally resembles the
326: one loop contribution to the effective action. Indeed, it is the UV
327: regularised one loop contribution for a theory with propagator
328: $S^{(2)}+R$.\step
329:
330: Integrating the one loop improved flow \eq{operator} leads to
331: \begin{eqnarray}\label{operator1loop}
332: \Delta\Gamma_1=
333: \s012\Tr \left[\rho\,\ln S^{(2)}\right]_\Lambda^k\,.
334: \end{eqnarray}
335: Again this resembles the one loop effective action for any $k$. In
336: contrast to an ERG flow, however, it is impossible to
337: interpret \eq{operator1loop} as the one loop contribution of
338: an UV-regularised modified theory.\step
339:
340: Integrating the proper-time flow \eq{PTRG-CS} at one loop, we get
341: after a straightforward algebra
342: \begin{eqnarray}
343: \Delta\Gamma_1=\s0{1}{2 m}\Tr \left[ \left(\s0{k'^2}{
344: S^{(2)}}\right)^m
345: {}_2 F_1\left(m,m;m+1;-
346: \s0{k'^2}{S^{(2)}}\right)\right]_\Lambda^k,\!\!\!
347: \label{PTRG-1loop}\end{eqnarray}
348: where ${}_p F_q(x,y;z;w)$ is the generalised hyper-geometric series.
349: For integer $m$, the series in ${}_2 F_1$ in \eq{PTRG-1loop} can be
350: summed up and there is a simpler representation for the one loop
351: contribution:
352: \begin{eqnarray}
353: \Delta\Gamma_1=\s012\Tr \left[ \ln \left(\mbox{\small $S^{(2)}+{k'}^2$}\right)
354: -\sum_{n=1}^{m-1} \s01n
355: \left(\s0{{k'}^2}{ S^{(2)}+{k'}^2}\right)^n\right]_\Lambda^k.
356: \label{PTRG-m}\end{eqnarray}
357: For $k\neq 0$ \eq{PTRG-1loop} does not resemble the one loop
358: contribution to the effective action. Of course, for $k\to 0$,
359: \eq{PTRG-1loop} reproduces the one loop effective action
360: $\s012[\Tr\ln(S^{(2)}+{k}^2)]_{\rm ren}$ where the
361: renormalisation at $\Lambda$ is included. \\[2ex]
362:
363:
364: %********|*********|*********|*********|*********|*********|*********|****
365: \noindent
366: {\bf 4. Effective action at two loop}\\[-1ex]
367: %********|*********|*********|*********|*********|*********|*********|****
368:
369: \noindent
370: As the ERG flow \eq{exact} has a first principle derivation,
371: obviously it has to reproduce the correct two loop result.
372: Structurally it belongs to the same class as the usual Callan-Symanzik
373: flow, and the calculation of diagrams and combinatorial prefactors of
374: either flow goes along the same lines. Here, we only present the result
375: of such a calculation. The two loop contribution $\Delta\Gamma_2$ to
376: the effective action obeying \eq{exact} is given by
377: \begin{eqnarray}
378: \nonumber
379: \lefteqn{\Delta\Gamma_2 =
380: \int_{pp'qq'}
381: \Bigl[ {1\over 8}\, G_{pp'}\ S_{p'pqq'}^{(4)}\ G_{q'q}}
382: \hspace{1.4cm}\\
383: & &\di -{1\over {12}}\,\int_{ll'}
384: G_{pp'}\ S_{p'lq}^{(3)} \ G_{ll'}\
385: S_{l'pq'}^{(3)}\ G_{q'q}
386: \Bigr]_{\rm ren.}\, ,
387: \label{profield}
388: \end{eqnarray}
389: where the subscript ${}_{\rm ren.}$ indicates that these are
390: renormalised diagrams due to the subtractions at $\Lambda$. We have
391: introduced the abbreviations $G_{pp'}\equiv(S^{(2)}+R)^{-1}(p,p')$,
392: the vertices $S_{p_1\cdots p_n}^{(n)} \equiv \delta^{(n)}S /
393: \delta\phi(p_1) \cdots \delta\phi (p_n)$, and a convenient short-hand
394: notation for the momentum integrals $\int_{p_1\cdots p_n}\equiv\int
395: \s0{d^dp_1}{(2\pi)^d}\cdots \s0{d^dp_n}{(2\pi)^d}$. The combinatorial
396: factors in \eq{profield} are in agreement with perturbation
397: theory. Again, even for $k\neq 0$ the result \eq{profield} functionally
398: resembles the perturbative structure. This analysis can be
399: easily extended to any loop order. Note that one can always rewrite
400: the integrands as total $t'$-derivatives. Thus, the precise form of the
401: regulator $R$ is irrelevant for the result, as it should.\step
402:
403: Expanding the one loop improved flow equation \eq{operator} at two
404: loop leads to the following expression:
405: \begin{eqnarray}\label{sb2loop}
406: \Delta\Gamma_2=\s012 \int_\Lambda^k \0{dk'}{k'}
407: \int_{pp'qq'}\ \Delta{\Gamma^{(2)}_{1,pp'}}\ G_{p'q}\
408: \partial_{t'} \rho_{qq'}
409: \end{eqnarray}
410: and $G=1/S^{(2)}$. It is easy to rewrite the expression
411: on the right hand side of \eq{sb2loop} as a total derivative, since the
412: only $k$-dependence of $\Delta \Gamma_1^{(2)}$ is given by $\rho$.
413: We finally get
414: \begin{eqnarray}\nonumber
415: \lefteqn{\Delta\Gamma_2=
416: \int_{pp'qq'}
417: \Bigl[ {1\over 8}\,
418: (G\, \rho)_{pp'}\ S_{p'pqq'}^{(4)}\ (G\, \rho)_{q'q}}\hspace{.75cm}\\
419: & &\di -{1\over 8}\,\int_{ll'}
420: (G\, \rho)_{pp'}\ S_{p'lq}^{(3)} \ G_{ll'}\
421: S_{l'pq'}^{(3)}\ (G\, \rho)_{q'q}
422: \Bigr]_{\rm ren.}.
423: \label{profieldsc}
424: \end{eqnarray}
425: Again, as for \eq{profield}, the result does not depend on the
426: regulator for $k=0$, where $\rho= 1$. Differentiating
427: \eq{profieldsc} w.r.t.~$k$ leads to the integrand of \eq{sb2loop}, as
428: it should. The combinatorial factors of the diagrams in
429: \eq{profieldsc} do not match those in \eq{profield}. Thus the flow
430: \eq{operator} fails to reproduce perturbation theory beyond one loop.
431: \step
432:
433: Finally we discuss the proper-time flow \eq{PTRG}. Below \eq{PTRG-CS},
434: we have already argued that the flow \eq{PTRG} is not an exact flow
435: for a general regulator. Here, as an explicit example, we calculate
436: the two loop effective action for $m=2$. Expanding the flow \eq{PTRG}
437: at two loop we get
438: \begin{eqnarray}\label{2-loop}
439: \Delta\Gamma_2 =\! -2 \int_\Lambda^k \0{dk'}{k'} \int_{pp'}
440: \ \Delta\Gamma_{1,pp'}^{(2)}\ (G\ k'^2\ G\ k'^2\ G)_{p'p} ,
441: \end{eqnarray}
442: where $G_{pp'}\equiv (S^{(2)}+k'^2)^{-1}(p,p')$. Note, that it is
443: impossible to rewrite the integrand in \eq{2-loop} as a total
444: derivative w.r.t.\ the scale parameter $t'$. This already is a strong
445: hint at the fact that one cannot get the correct two loop result. Let
446: us cast \eq{2-loop} in a form which shows explicitly how it deviates
447: from perturbation theory. Using partial $t'$-integrations we obtain
448: from \eq{2-loop}, after some lengthy but straightforward algebra,
449: \begin{eqnarray}\nonumber
450: \Delta\Gamma_2 &=&
451: \int_{pp'qq'} \Bigl[ \018\, G_{pp'}\ S_{p'pqq'}^{(4)}\ G_{q'q}
452: \\ && \nonumber \di
453: \qquad\quad
454: -\0{1}{12}\int_{ll'} G_{pp'}\ S_{p'lq}^{(3)} \ G_{ll'}\
455: S_{l'pq'}^{(3)}\ G_{q'q}\Bigr]_{\rm ren.}
456: \\ \di \nonumber
457: &-&
458: \012 \, \int_\Lambda^k \0{dk'}{k'} \int_{pp'qq'll'}
459: \Bigl[(G\ k'^2\, G)_{pp'} \ S_{p'ql}^{(3)}
460: \\ &&\di
461: \qquad \qquad \times(G\ k'^2\, G)_{qq'} \ S_{q'pl'}^{(3)}\
462: (G\ k'^2\, G)_{l'l}\Bigr] \,.
463: \label{deviation}
464: \end{eqnarray}
465: Differentiating \eq{deviation} w.r.t.~$k$ leads to the integrand of
466: \eq{2-loop}, as it should. The first two terms in \eq{deviation}
467: correspond to the correct two loop result as presented in
468: \eq{profield}. The last term denotes the deviation from standard
469: perturbation theory. The $d\ln k'$-integrand of the last term in
470: \eq{deviation} is the non-standard diagram depicted in Fig.~1. The
471: last term on the right-hand side of \eq{deviation} cannot be absorbed
472: in renormalisation constants. It contains arbitrary powers in fields
473: and momenta and does not integrate to zero in the limit $k\to0$ and
474: $\Lambda\to\infty$. For massive theories both limits are safe.
475: Consequently this term displays a non-trivial deviation of the
476: proper-time flow from perturbation theory. The form of the integrand
477: is that of the sunset graph where all propagators have been
478: substituted by their squares. This is clearly related to the fact
479: that the form of the proper-time flow is that of a Callan-Symanzik
480: flow with all propagators substituted by their squares.
481:
482: %********|*********|*********|*********|*********|*********|*********|****
483: %******************| Fig1: pPTRG3.ps |*********|*******************|****
484: %********|*********|*********|*********|*********|*********|*********|****
485: \begin{figure}
486: \begin{center}
487: \unitlength0.001\hsize
488: \begin{picture}(600,430)
489: \psfig{file=pPTRG3.ps,width=.55\hsize}
490: \end{picture}
491: \begin{minipage}{\hsize}{
492: { \small {\bf Figure 1:} The integrand of the non-standard term in
493: \eq{deviation}. The two vertices $S^{(3)}$ are denoted
494: by $\bullet$, the six internal lines are the propagators
495: $G=(S^{(2)}+k'^2)^{-1}$, and the three insertions correspond to
496: $k'^2$.}}
497: \end{minipage}
498: \end{center}
499: \vskip-3ex
500: \end{figure}
501: %************************************************************************|
502:
503: To be more explicit, consider the example of a massive $\phi^4$-theory
504: with mass $M$ and quartic interaction $ \s0{\lambda}{4!}\, \int d^d x
505: \,\phi^4$. The contribution of the non-standard diagram to the
506: propagator is obtained after taking the second derivative with respect
507: to the fields in \eq{deviation} at $\phi=0$. We find
508: \begin{eqnarray}\nonumber
509: \lefteqn{\lambda^2 \int_\infty^0 \0{d k}{ k} \int \frac{d^d q}{(2\pi)^d}
510: \frac{d^d l}{(2\pi)^d}\,\Bigl[ {k^2\over (k^2+M^2+q^2)^2}}
511: \hspace{.6cm}
512: \\ & &\di
513: \times {k^2\over (k^2+M^2+l^2)^2}
514: {k^2\over (k^2+M^2+(l+q-p)^2)^2}\Bigr]\, .
515: \label{propphi4} \end{eqnarray}
516: The integrand it strictly positive. Hence the integral is
517: non-vanishing. Moreover it has a non-trivial momentum dependence.
518: This can be seen by evaluating the limits $p\to 0$ and $p\to\infty$.
519: For $p\to 0$ we are left with a non-vanishing constant. In turn, for
520: $p\to\infty$ the expression in \eq{propphi4} vanishes.\\[2ex]
521:
522: %********|*********|*********|*********|*********|*********|*********|****
523: \noindent
524: {\bf 5. Discussion}\\[-1ex]
525: %********|*********|*********|*********|*********|*********|*********|****
526:
527: \noindent
528: Having established that neither \eq{operator} nor, in general,
529: \eq{PTRG} provide exact flows, we want to understand what precisely
530: causes the deviation from perturbation theory. First we recall the
531: argument made prior to \eq{operator}: A general exact flow is related
532: to the flow of an operator insertion in the theory. Demanding,
533: additionally, that the flow has a one loop structure restricts
534: possible insertions to operators quadratic in the fields. Consequently
535: such a flow has to depend linearly on the full propagator. \step
536:
537: For a general flow it might be hard to decide, whether one has such a
538: situation. Already for general proper-time flows we had to take the
539: detour of expanding general flows in the basis \eq{fm} in order to
540: reach to a conclusion. Thus, we would like to provide an additional
541: criterion, which also reflects the necessity of a linear dependence on
542: the full propagator. Indeed, a sufficient condition for a RG equation
543: to reproduce perturbation theory can be deduced from the iterative
544: structure of the perturbation series: It suffices that the solution of
545: a RG equation has the same iterative structure even at non-vanishing
546: cut-off. Without this property, the corresponding RG equation has to
547: satisfy an infinite tower of iterative constraints in order to
548: reproduce perturbation theory in the limit, where the infrared cut-off
549: tends to zero. Consequently, one can assess from the structure of the
550: one loop effective action at $k\neq 0$, whether a flow is likely to
551: reproduce perturbation theory. \step
552:
553: The iterative structure discussed above is absent in the one loop
554: effective action given in \eq{PTRG-1loop} for $k\neq 0$. Moreover, it
555: cannot be regained by considering linear combinations of regulators
556: \eq{fm}. Despite this discouraging fact, let us shed some more light
557: on the structure of proper-time flows. It is not possible to integrate
558: a general proper-time flow beyond one loop without knowing the precise
559: form of the regulator. Still, there are recursive relations between
560: different proper-time flows at a given loop order. These relations
561: tell us how the flows differ from each other for {\it arbitrary} $m$,
562: integer or not. At two loop, and with $G=(S^{(2)}+k^2)^{-1}$, the most
563: general recursion relation is given by \cite{consistent}
564: \begin{eqnarray}\nonumber
565: \Delta\Gamma_{2,m}&-&\Delta\Gamma_{2,m-1}=
566: \012\int_\infty^0
567: \0{dk}{k}\,\Tr \big[\left({G\,k^2}\right)^{m-1}\, G
568: \\ \di
569: && \times\, \displaystyle
570: (\s0{m}{m-1}\,{k}^2 G-1)
571: \,{\delta^2\over(\delta\phi)^2}\Tr\, (G \,{k}^2)^{m-1}
572: \big]\,,
573: \label{recursive} \end{eqnarray}
574: apart from irrelevant terms from the different renormalisation
575: procedures for the two flows. The difference \eq{recursive} (or, more
576: generally, $\Delta\Gamma_{2,m}-\Delta\Gamma_{2,m-n}$ with integer $n$)
577: depends on arbitrarily high powers of the fields and does not
578: integrate to zero. \step
579:
580: \Eq{recursive} can be used to give an independent explicit proof of
581: the non-exactness of general proper-time flows. To that end, let us
582: assume for a moment that the proper-time flow for a particular $m_0$
583: is exact. Then it follows from \eq{recursive} that all flows with
584: $m=m_0+n$ for integer $n$ are {\it not} exact, because the
585: corresponding terms \eq{recursive} do not vanish identically in the
586: fields. Hence, of all proper-time flows of the form \eq{PTRG} with
587: regulators \eq{fm} or finite linear combination thereof, the set of
588: exact flows is of measure zero. This has an immediate consequence for
589: flows with integer $m$. The Callan-Symanzik flow ($m=1$) is exact, but
590: any flow with integer $m>1$, or any linear combinations thereof, are
591: not exact. Hence, the structure of the findings for $m=2$ is present
592: for arbitrary $m$, and \eq{recursive} provides an independent explicit
593: proof for the general statement derived after \eq{PTRG-CS}.
594:
595: Thus, for proper-time flows, we arrive at the following picture. The
596: only known exact proper-time flow is the Callan-Symanzik flow. Other
597: exact proper-time flows -if they exist- would require a linear
598: dependence on the full propagator, possibly in some disguise. Based on
599: our findings, no further exact flows can be found within the set of
600: regulators \eq{fm}, which covers all flows previously studied in the
601: literature. Of course, it is not excluded, that a regulator, which is
602: represented by an infinite series of regulators \eq{fm}, is exact.
603: However, there is no {\it a priori} criterion upon which one could
604: embark on and construct such a regulator. \step
605:
606:
607:
608: To summarise, we have shown that the one loop improved flows
609: \eq{operator} and, in general, \eq{PTRG} are not exact flows. We have
610: shown explicitly, that they fail at the first non-trivial order,
611: at two loop. These results imply
612: that hopes expressed in the literature -- suggesting that the
613: RG flows \eq{operator} and \eq{PTRG} correspond to exact flows only
614: with a different implementation of the regularisation -- cannot be
615: maintained. In fact, these flows are {\it substantially} different
616: from exact flows, and describe at best approximations to the latter.
617: Justification of their use requires a deep understanding of the
618: inherent approximation in order to furnish these methods with
619: predictive power. This question has only been addressed within the
620: derivative expansion \cite{Litim:2001hk}. However, the potential
621: benefits of general one loop improved RG flows within numerical
622: implementations justify further investigations. An extensive study of
623: this problem, including a more detailed account of the present
624: calculations, will be given elsewhere \cite{consistent}.
625: \\
626:
627: {\it Acknowledgements:} JMP thanks CERN for hospitality. DFL has been
628: supported by a Marie-Curie fellowship under EC contract
629: no.~HPMF-CT-1999-00404.
630: \\[2ex]
631: \newpage
632: %********|*********|*********|*********|*********|*********|*********|****
633: \noindent
634: {\bf References}\\[-1ex]
635: %********|*********|*********|*********|*********|*********|*********|****
636:
637: %********|*********|*********|*********|*********|*********|*********|****
638: \begin{thebibliography}{99}
639: %********|*********|*********|*********|*********|*********|*********|****
640:
641: %\cite{Litim:2001hk}
642: \bibitem{Litim:2001hk}
643: D.~F.~Litim and J.~M.~Pawlowski,
644: %``Predictive power of renormalisation group flows: A comparison,''
645: Phys.\ Lett.\ B {\bf 516} (2001) 197 [hep-th/0107020].
646: %%CITATION = HEP-TH 0107020;%%
647:
648:
649:
650: %\cite{Liao:1996fp}
651: \bibitem{Liao:1996fp}
652: S.~B.~Liao,
653: %``On connection between momentum cutoff and the proper time regularizations,''
654: Phys.\ Rev.\ {\bf D53} (1996) 2020
655: [hep-th/9501124].
656: %%CITATION = HEP-TH 9501124;%%
657:
658:
659:
660: %\cite{Liao:1997nm}
661: \bibitem{Liao:1997nm}
662: S.~B.~Liao,
663: %``Operator Cutoff Regularization and Renormalization Group in Yang-Mills Theory,''
664: Phys.\ Rev.\ {\bf D56} (1997) 5008
665: [hep-th/9511046].
666: %%CITATION = HEP-TH 9511046;%%
667:
668: \bibitem{Polchinski}
669: J.\,Polchinski, Nucl.~Phys.~{\bf B231} (1984) 269.
670:
671: \bibitem{CW}
672: C.\,Wetterich, Phys.~Lett.~{\bf B301} (1993) 90;\\
673: %\cite{Ellwanger:1994mw}
674: %\bibitem{Ellwanger:1994mw}
675: U.~Ellwanger,
676: %``Flow equations for N point functions and bound states,''
677: Z.\ Phys.\ C {\bf 62} (1994) 503;\\
678: %[hep-ph/9308260].
679: %%CITATION = HEP-PH 9308260;%%
680: %\cite{Morris:1994qb}
681: %\bibitem{Morris:1994qb}
682: T.~R.~Morris,
683: %``The Exact renormalization group and approximate solutions,''
684: Int.\ J.\ Mod.\ Phys.\ {\bf A9} (1994) 2411.
685: %[hep-ph/9308265].
686: %%CITATION = HEP-PH 9308265;%%
687:
688:
689: %\cite{Bagnuls:2000ae}
690: \bibitem{Bagnuls:2000ae}
691: %\bibitem{Reviews}
692: C.~Bagnuls and C.~Bervillier,
693: %``Exact renormalization group equations: An introductory review,''
694: Phys.\ Rept.\ {\bf 348} (2001) 91
695: [hep-th/0002034];\\
696: %%CITATION = HEP-PH 0002034;%%
697: %\cite{Litim:1998nf}
698: D.~F.~Litim and J.~M.~Pawlowski,
699: %{\it On gauge invariant Wilsonian flows},
700: %in: The Exact Renormalisation Group, Faro, Portugal, p.~168-185.
701: hep-th/9901063;\\
702: %%CITATION = HEP-TH 9901063;%%
703: J.~Berges, N.~Tetradis and C.~Wetterich,
704: %``Non-perturbative renormalisation flow in quantum field theory
705: %and statistical physics,''
706: hep-ph/0005122.
707: %%CITATION = HEP-PH 0005122;%%
708:
709:
710: \bibitem{consistent}
711: D.~F.~Litim and J.~M.~Pawlowski, hep-th/0202188.
712:
713:
714:
715: %\cite{Liao:2000yu}
716: \bibitem{Liao:2000yu}
717: S.~B.~Liao, C.~Y.~Lin and M.~Strickland,
718: %``Self-consistent renormalization group flow,''
719: hep-th/0010100.
720: %%CITATION = HEP-TH 0010100;%%
721:
722:
723:
724:
725: %\cite{Oleszczuk:1994st}
726: \bibitem{Oleszczuk:1994st}
727: M.~Oleszczuk,
728: %``A Symmetry preserving cutoff regularization,''
729: Z.\ Phys.\ {\bf C64} (1994) 533.
730: %%CITATION = ZEPYA,C64,533;%%
731:
732: %\cite{Floreanini:1995aj}
733: \bibitem{Floreanini:1995aj}
734: R.~Floreanini and R.~Percacci,
735: %``The Heat kernel and the average effective potential,''
736: Phys.\ Lett.\ B {\bf 356} (1995) 205
737: [hep-th/9505172].
738: %%CITATION = HEP-TH 9505172;%%
739:
740: %\cite{Schaefer:1999em}
741: \bibitem{Schaefer:1999em}
742: B.~J.~Sch\"afer and H.~Pirner,
743: %``The equation of state of quarks and mesons in a renormalization group
744: %flow picture,''
745: Nucl.\ Phys.\ {\bf A660} (1999) 439
746: [nucl-th/9903003].
747: %%CITATION = NUCL-TH 9903003;%%
748:
749: %\cite{Meyer:2000bz}
750: \bibitem{Meyer:2000bz}
751: J.~Meyer, G.~Papp, H.~J.~Pirner and T.~Kunihiro,
752: %``Renormalization group flow equation at finite density,''
753: Phys.\ Rev.\ C {\bf 61} (2000) 035202
754: [nucl-th/9908019].
755: %%CITATION = NUCL-TH 9908019;%%
756:
757: %\cite{Papp:2000he}
758: \bibitem{Papp:2000he}
759: G.~Papp, B.~J.~Sch\"afer, H.~J.~Pirner and J.~Wambach,
760: %``On the convergence of the expansion of renormalization group flow equation,''
761: Phys.\ Rev.\ D {\bf 61} (2000) 096002
762: [hep-ph/9909246].
763: %%CITATION = HEP-PH 9909246;%%
764:
765: %\cite{Bohr:2001gp}
766: \bibitem{Bohr:2001gp}
767: O.~Bohr, B.~J.~Sch\"afer and J.~Wambach,
768: %``Renormalization group flow equations and the phase transition in O(N)
769: %models,''
770: Int.\ J.\ Mod.\ Phys.\ A {\bf 16} (2001) 3823
771: [hep-ph/0007098].
772: %%CITATION = HEP-PH 0007098;%%
773:
774: %\cite{Meyer:2001zp}
775: \bibitem{Meyer:2001zp}
776: J.~Meyer, K.~Schwenzer, H.~J.~Pirner and A.~Deandrea,
777: %``Renormalization group flow in large N(c),''
778: hep-ph/0110279.
779: %%CITATION = HEP-PH 0110279;%%
780:
781:
782:
783: %\cite{Bonanno:2001yp}
784: \bibitem{Bonanno:2001yp}
785: A.\,Bonanno and D.\,Zappal\`a,
786: %``Towards an accurate determination of the critical exponents with the
787: %renormalization group flow equations,''
788: Phys.~Lett.~{\bf B504} (2001) 181
789: [hep-th/0010095].
790: %%CITATION = HEP-TH 0010095;%%
791:
792: %\cite{Mazza:2001bp}
793: \bibitem{Mazza:2001bp}
794: M.~Mazza and D.~Zappal\`a,
795: %``Proper time regulator and renormalization group flow,''
796: Phys.\ Rev.\ {\bf D64} (2001) 105013
797: [hep-th/0106230].
798: %%CITATION = HEP-TH 0106230;%%
799:
800: %\cite{Zappala:2001nv}
801: \bibitem{Zappala:2001nv}
802: D.~Zappal\`a,
803: %``Improving the Renormalization Group approach to the quantum-mechanical
804: %double well potential,''
805: Phys.\ Lett.\ {\bf A290} (2001) 35
806: [quant-ph/0108019].
807: %%CITATION = QUANT-PH 0108019;%%
808:
809: \end{thebibliography}
810:
811: \end{document}
812:
813:
814: