hep-ex0106011/m09.tex
1: \documentclass[acus]{JAC2000}
2: %
3: % M09 
4: % Sven Menke
5: % alpha_s at low Q^2 from tau and e+e- data
6: % Invited talk given at 'e+e- Physics at Intermediate Energies',
7: % SLAC, 30 April - 2 May 2001
8: % 
9: \def\bfseries{\fontseries\bfdefault\selectfont\boldmath}
10: %%  
11: %%  Use \documentclass[boxit]{JAC2000}                      
12: %%  To draw a frame with the correct margins on the output.
13: %%  
14: %%  Use \documentclass[acus]{JAC2000}                      
15: %%  For US letter paper layout
16: %% 
17: 
18: \usepackage[final]{graphics}
19: \usepackage{citesort}
20: 
21: \newcommand{\CIPTGGVL} {\ensuremath{-0.003}}
22: \newcommand{\CIPTGGEM} {\ensuremath{ 0.011}}
23: \newcommand{\CIPTSSVL} {\ensuremath{ 0.0012}}
24: \newcommand{\CIPTSSEM} {\ensuremath{ 0.0047}}
25: \newcommand{\CIPTSEVL} {\ensuremath{-0.0010}}
26: \newcommand{\CIPTSEEM} {\ensuremath{ 0.0029}}
27: \newcommand{\CIPTCSMT} {\ensuremath{0.16}}
28: \newcommand{\CIPTcCHSQ}{\ensuremath{0.63}}
29: \newcommand{\MEANcGGVL}{\ensuremath{ 0.001}}
30: \newcommand{\MEANASVLMT}{\ensuremath{ 0.348}}
31: \newcommand{\MEANASXMMT}{\ensuremath{ 0.009}}
32: \newcommand{\MEANcVSVL}{\ensuremath{ 0.0256}}
33: \newcommand{\MEANcVEVL}{\ensuremath{-0.0080}}
34: \newcommand{\MEANcASVL}{\ensuremath{-0.0197}}
35: \newcommand{\MEANcAEVL}{\ensuremath{ 0.0041}}
36: 
37: %%
38: %%   VARIABLE HEIGHT FOR THE TITLE BOX (default 35mm)
39: %%
40: 
41: \setlength{\titleblockheight}{26mm}
42: \newcommand{\alphas}    {\ensuremath{\alpha_{\rm s}}}
43: \newcommand{\Pgne}      {\ensuremath{\mathrm{\nu_e}}}
44: \newcommand{\Pgngt}     {\ensuremath{\mathrm{\nu_\tau}}}
45: \newcommand{\Pgtm}      {\ensuremath{\mathrm{\tau^-}}}
46: \newcommand{\Pgtp}      {\ensuremath{\mathrm{\tau^+}}}
47: \newcommand{\Ph}        {\ensuremath{\mathrm{h}}}
48: \newcommand{\Pe}        {\ensuremath{\mathrm{e}}}
49: \newcommand{\Pep}       {\ensuremath{\mathrm{e^+}}}
50: \newcommand{\Pem}       {\ensuremath{\mathrm{e^-}}}
51: 
52: % add words to TeX's hyphenation exception list
53: \hyphenation{author another created financial paper re-commend-ed}
54: \hyphenation{si-mul-ta-neous si-mul-ta-neous-ly}
55: 
56: \title{\flushright{M09}\\[-2mm] \centering $\alpha_{\rm\bf s}$ AT LOW
57:   $Q^2$ FROM $\Pep\Pem$ AND $\tau$ DATA\thanks{Work supported by
58:     Department of Energy contract DE--AC03--76SF00515.}}
59: 
60: \author{S. Menke, Stanford Linear Accelerator Center,
61:     2575 Sand Hill Road, Menlo Park, CA 94205, USA}
62: %
63: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
64: %
65: \begin{document}
66: 
67: \maketitle
68: 
69: \begin{abstract} It has been shown in recent analyses by 
70:   ALEPH~\cite{art:Andreas98} and OPAL~\cite{art:Sven} that precision
71:   QCD tests are possible with hadronic $\tau$ decays by comparing
72:   spectral moments of the hadronic decay ratio of the $\tau$ with QCD
73:   calculations.  In principle $\Pep\Pem$ data can be used in a similar
74:   manner by evaluating spectral moments of $R$.  The current
75:   $\Pep\Pem$ data is compared with the OPAL $\tau$ data and a
76:   prediction is made on the achievable accuracy of QCD tests with the
77:   projected precision of PEP-N~\cite{art:PEP-N}.
78: \end{abstract}
79: 
80: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
81: 
82: \section{INTRODUCTION}
83: %
84: The $\tau$ lepton is the only lepton heavy enough to decay into
85: hadrons. The observed spectra of the non-strange hadrons in $s$, where
86: $\sqrt{s}$ is the mass of the final state hadronic system, give the
87: non-strange spectral functions.  Decays with an even number of pions
88: in the final state belong to the vector current while decays with an
89: odd number of pions belong to the axial-vector current. A comparison
90: of weighted integrals over the spectral functions of the vector and
91: axial-vector current with QCD predictions can give fundamental
92: parameters of the
93: theory\cite{art:Braaten88,art:Braaten89,art:Narison88,art:Braaten,art:Pich92},
94: including the strong coupling constant \alphas. Different integrals
95: (moments) are used to measure power corrections of non-perturbative
96: origin and the strong coupling simultaneously, thus reducing the
97: theoretical uncertainties on \alphas{\ }connected with the
98: non-perturbative terms.
99: 
100: The QCD tests can also be performed for energy scales smaller than
101: $m_\tau$, if the integrals over the spectral functions are performed
102: with an upper integration limit $m_{\tau'}^2 = s_0\le m_\tau^2$ and
103: replacing $m_\tau^2$ by $m_{\tau'}^2$ in the
104: integrals~\cite{art:Andreas98} thus creating the spectral moments of a
105: hypothetical $\tau'$ lepton with a mass below $m_\tau$.
106: 
107: In the same manner $\Pep\Pem$ annihilation into hadrons can be used to
108: extract the spectral function of the vector-current and compare with
109: theory by means of its spectral moments.
110: 
111: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
112: 
113: \section{THEORY}
114: %
115: 
116: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
117: 
118: \subsection{Hadronic $\tau$ decays}
119: %
120: The vector $v(s)$ and axial-vector $a(s)$ spectral functions (the
121: absorptive parts of the vacuum polarization correlators) are given by
122: the spectra in $s$ of the final state hadrons, normalized to the
123: branching ratios $B$ and corrected for the phase-space:
124: \begin{eqnarray}
125:   v/a(s)\! & = & \!\!\frac{1}{2\pi} {\rm Im}\Pi_{\rm V/A}(s) \nonumber\\
126:            & = & \! \!\frac{\displaystyle
127:     m_\tau^2 \!\sum_{\Ph_{\rm V/A}} 
128:     \frac{\displaystyle B_{\tau\rightarrow \Ph_{\rm V/A} \Pgngt}}{\displaystyle
129:       B_{\tau\rightarrow \Pe \Pgne \Pgngt}}
130:     \frac{\displaystyle w_{\rm V/A}}{\displaystyle N_{\rm V/A}}
131:     \frac{\displaystyle {\rm d} N_{\rm V/A}}{\displaystyle {\rm d} s}}
132:     {\displaystyle 6 S_{\rm EW} |V_{\rm ud}|^2 
133:     \left(1\! -\! \frac{s}{m_\tau^2}\right)^2
134:     \left(1\! +\! 2\frac{s}{m_\tau^2}\right)
135:     }, 
136:   \label{eq:spectral} 
137: \end{eqnarray}
138: where the sum is performed over non-strange hadronic final states
139: $h_{\rm V/A}$ with angular momentum $J\!=\!1$. $N_{\rm V/A}$ is the
140: number of taus that decay into the hadron $\Ph_{\rm V/A}$ plus
141: neutrino, $w_{\rm V/A}$ denotes the appropriate weight of the hadronic
142: mode to the vector or axial-vector current, $S_{\rm EW} = 1.0194$ is
143: an electroweak correction term~\cite{art:Marciano} and $|V_{\rm ud}|^2
144: = 0.9477 \pm 0.0016$ is the squared CKM weak mixing matrix
145: element~\cite{art:PDG00}.
146: 
147: Within the framework of QCD weighted integrals over the spectral
148: functions (so called moments) have been
149: calculated~\cite{art:Diberder}:
150: \begin{eqnarray}
151:   \lefteqn{R_{\tau,{\rm V/A}}^{kl}\! = 6 S_{\rm EW} |V_{\rm ud}|^2\!\!
152:     \int\limits_0^{m_\tau^2}\! 
153:     \frac{{\rm d} s}{m_\tau^2}
154:     \left( 1\! -\! \frac{s}{m_\tau^2}\right)^{2+k}\!
155:     \left(\frac{s}{m_\tau^2}\right)^l} \nonumber\\  
156:   & & \phantom{R_\tau =} \times
157:   \left[ \left(1\!+\!2\frac{s}{m_\tau^2}\right) v/a(s) + v^0/a^0(s)
158:   \right]\label{eq:rkl}, 
159: \end{eqnarray}
160: where the scalar spectral function $v^0(s)$ vanishes, since no scalar
161: particle has been observed in $\tau$ decays, while the pseudo-scalar
162: spectral function $a^0(s)$ is given by the pion pole, assuming that
163: the pion is the only pseudo-scalar final-state in non-strange $\tau$
164: decays:
165: \begin{equation}
166:   a^0(s) = 
167:   \frac{\displaystyle m_\tau^2 
168:     \frac{\displaystyle B_{\tau \rightarrow \pi \Pgngt}}{
169:       \displaystyle B_{\tau \rightarrow \Pe \Pgne \Pgngt}}
170:     \delta(s-m_\pi^2)}
171:   {\displaystyle 6 S_{\rm EW} |V_{\rm ud}|^2
172:     \left(1\!-\!\frac{s}{m_\tau^2}\right)^{2}}.
173:   \label{eq:pionpole}
174: \end{equation}
175: 
176: The moments are used to compare the experiment with theory. In what
177: follows, ten moments for $kl = 00,10,11,12,13$ for $\rm V$ and $\rm A$
178: are used.  The first moments~$R_{\tau,{\rm V/A}}^{00}$ are the total
179: normalized decay rates of the $\tau$ into vector and axial-vector
180: mesons. In the na{\"\i}ve parton model these two rates are identical and
181: add up to the number of colors. Since only non-strange currents are
182: considered in this work the na{\"\i}ve expectation has to be multiplied by
183: $|V_{\rm ud}|^2$.  Including the perturbative and non-perturbative
184: contributions, equation (\ref{eq:rkl}) is usually written as
185: \cite{art:Diberder}:
186: \begin{equation}
187:   R_{\tau,{\rm V/A}}^{kl}\! = 
188:   \frac{3}{2} S_{\rm EW} |V_{\rm ud}|^2 
189:   \biggl( 1 +
190:   \delta_{\rm pert}^{kl} + \!\!\!\!\!\!\sum_{D = 2,4,6,8}\!\!\!\!\!\!
191:   \delta_{\rm V/A}^{D,kl} \biggr),        
192:   \label{eq:rklQCD}      
193: \end{equation}
194: where $\delta_{\rm pert}^{kl}$ are perturbative QCD corrections
195: ($\approx 20\,\%$ for $kl\!=\!00$) and the $\delta_{\rm V/A}^{D,kl}$
196: terms are the so-called power corrections ($\approx 1\,\%$ for
197: $kl\!=\!00$).
198: 
199: The perturbative term $\delta_{\rm pert}^{kl}$ is known to third order
200: in \alphas{\ }and partly known to fourth order in \alphas. For $kl=00$
201: the Contour Improved Perturbation Theory (CIPT) result
202: is~\cite{art:Diberder}:
203: \begin{eqnarray}
204:   \delta_{\rm pert}^{00} = \sum_{n=1}^4 \frac{K_n}{2\pi i} \!\!\!
205:   \oint\limits_{|s|=m_\tau^2} \!\!\!\frac{{\rm d} s}{s}
206:   \left(1-2\frac{s}{m_\tau^2}+2\frac{s^3}{m_\tau^6}\right.\nonumber\\
207:   \left.-\frac{s^4}{m_\tau^8}\right)
208:   \left(\frac{\alphas(-s)}{\pi}\right)^n,\label{eq:CIPT}
209: \end{eqnarray}
210: where the $K_1=1, K_2=1.63982, K_3=6.37101$ are
211: known~\cite{art:Chetyrkin,art:Dine,art:Celmaster,art:Gorishnii,art:Surguladze}
212: and $K_4 = 25\pm50$ has been
213: estimated~\cite{art:Diberder,art:Pich97,art:Kataev}.  The
214: Taylor-expansion in $\alphas(m_\tau^2)$ of the CIPT result
215: is~\cite{art:Braaten,art:Diberder}:
216: \begin{equation}
217:   \label{eq:FOPT}
218:   \delta_{\rm pert}^{00} =  a_{\rm s}
219:   +  5.2023      a^2_{\rm s}
220:   + 26.366      a^3_{\rm s}   
221:   + (78.003 + K_4) a^4_{\rm s},
222: \end{equation}
223: with $a_{\rm s} = \alphas(m_\tau^2)/\pi$. It is referred to as Fixed
224: Order Perturbation Theory (FOPT).  The third method considered in this
225: paper resums the leading term of the $\beta$-function to all orders in
226: \alphas{\ }by inserting so-called Renormalon Chains
227: (RCPT)~\cite{art:Ball,art:Neubert,art:Maxwell95,art:Maxwell96}.  The
228: fixed-order corrected version (up to the third order in \alphas)
229: quoted in the lower portion of table 6 in reference~\cite{art:Neubert}
230: is used in the fit.
231: 
232: The power corrections $\delta_{\rm V/A}^{D,kl}$ in the framework of
233: the Operator Product Expansion (OPE) \cite{art:Wilson} are
234: proportional to $m_\tau^{-D}$.  The dimension $D=2$ terms are
235: mass-corrections of perturbative
236: origin~\cite{art:Braaten,art:Diberder} and are small for non-strange
237: $\tau$ decays.  Corrections of higher dimension are of
238: non-perturbative origin, absorbing the long-distance dynamics into
239: vacuum matrix elements~\cite{art:SVZ1,art:SVZ2,art:SVZ3,art:Braaten}.
240: In contrast to the perturbative part the power corrections differ for
241: the vector and the axial-vector currents.
242: 
243: If one neglects the small $s$-dependence of the power corrections, the
244: $\delta_{\rm V/A}^{D,kl}$ terms can be expressed for all $kl$ values
245: by a product of the power correction for $kl = 00$ and a simple
246: integral over the $kl$-dependent weight-functions~\cite{art:Diberder}.
247: This approach is used for the dimension $D=6$ and $D=8$ terms, taking
248: $\delta_{\rm V/A}^{6/8,00}$ as free parameters.  For the dimension
249: $D=2$ and $D=4$ terms the full $s$-dependence is taken into account
250: for the theoretical description of the moments \cite{art:Diberder}.
251: The least precisely known $D=4$ parameter, the gluon condensate which
252: is known only to $50\,\%$ \cite{art:Braaten} is also taken as a free
253: parameter in the fit, while the $D=2$ term is calculated from the
254: quark masses and the strong coupling. The quark-masses and
255: -condensates needed to complete the $D=2,4$ terms are taken
256: from~\cite{art:Braaten}.
257: 
258: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
259: 
260: \subsection{$\Pep\Pem$ annihilation into hadrons}
261: %
262: The ratio $R_{\Pep\Pem}$ is defined as:
263: \begin{eqnarray}
264:   R_{\Pep\Pem}(s) \!\!& = &\!\! \frac{3s}{4\pi\alpha^2}
265:   \sigma_{\Pep\Pem\rightarrow\textrm{hadrons}}(s)\nonumber\\
266:   \!\!&=&\!\! 12\pi\,{\rm Im}\Pi^\gamma(s) 
267:   = 6 v^\gamma(s),
268:   \label{eq:R}
269: \end{eqnarray}
270: with $v^\gamma(s) \approx\sum_{f={\rm u,d,s}}Q_f^2 v(s)$ being the
271: vector spectral function with (except for the isoscalar contributions
272: and the charge dependent factor) similar properties as $v(s)$
273: in~(\ref{eq:spectral}) and~(\ref{eq:rkl}) and $Q_f$ denoting the
274: charge of the quark flavor $f$.
275: 
276: In massless perturbative QCD $\sum_f Q_f^2 v(s)$ and $v^\gamma(s)$ are
277: identical and most conveniently expressed in form of the
278: Adler-function~\cite{art:Adler}:
279: \begin{eqnarray}
280:   D_{\rm P}(-s) &=& -4\pi^2s\frac{{\rm d}\Pi^\gamma(-s)}{{\rm d}s}\nonumber\\ 
281:   &=& \sum_{f={\rm u,d,s}}\!\!\!Q_f^2\left(1+\sum_{n=1}^4 
282:     K_n\frac{\alphas^n(s)}{\pi^n}\right)  \label{eq:Adler},
283: \end{eqnarray}
284: with (for three flavors) the same $K_n$ as in~(\ref{eq:CIPT}).
285: 
286: The mass corrections to the Adler-function are given
287: by~\cite{art:Chetyrkin90}:
288: \begin{eqnarray}
289:   D_{\rm mass}(-s) & = & \!\!-\!\!\sum_{f={\rm u,d,s}}\!\!\!Q_f^2
290:   \frac{m_f^2(s)}{s}\Biggl[6+28\frac{\alphas(s)}{\pi}\nonumber\\
291:   & & \hspace{-1.5cm}+\!\Biggl(\!259.666
292:   - 2.25\!\!\!\!\!\sum_{f'={\rm u,d,s}}\!\!\!\!\!
293:   \frac{m_{f'}^2(s)}{m_f^2(s)}\Biggr)
294:   \frac{\alphas^2(s)}{\pi^2}\Biggr]  \label{eq:Dmass},
295: \end{eqnarray}
296: where the running quark masses $m_f(s)$ are calculated from
297: scale-invariant mass parameters and an evolution equation which is
298: know to four loops~\cite{art:Vermaseren97}.
299: 
300: Finally the non-perturbative part of the Adler function
301: reads~\cite{art:Braaten}:
302: \begin{eqnarray}
303:   D_{\rm NP}(-s)\!\!\! & = & \!\!\!8\pi^2\!\!\!\!\sum_{f={\rm u,d,s}}\!\!\!Q_f^2
304:   \left[\frac{1}{12}
305:     \left(1-\frac{11}{18}\frac{\alphas(s)}{\pi}\right)\!
306:     \frac{\langle\frac{\alphas}{\pi}GG\rangle}{s^2}\right.\nonumber\\
307:   & & + \left(2+\frac{2}{3}\frac{\alphas(s)}{\pi}\right)
308:   \frac{\langle m_f \bar\psi_f\psi_f\rangle}{s^2}\nonumber\\
309:   & & + \frac{4}{27}\frac{\alphas(s)}{\pi}\!\!\sum_{f'={\rm u,d,s}}\!\!\!
310:   \frac{\langle m_{f'} \bar\psi_{f'}\psi_{f'}\rangle}{s^2}\nonumber\\
311:   & & + \left.\frac{3}{2}\frac{\langle{\cal O}_6\rangle}{s^3} 
312:     + 2\frac{\langle{\cal O}_8\rangle}{s^4}\right]\label{eq:DNP},
313: \end{eqnarray}
314: with the dimension 4 contributions from the gluon condensate
315: $\langle\alphas/\pi GG\rangle$ and the quark condensates $\langle m_f
316: \bar\psi_f\psi_f\rangle$ and the dimension 6 and 8 operators
317: $\langle{\cal O}_6\rangle$ and $\langle{\cal O}_8\rangle$.
318: 
319: The total Adler function is simply the sum of the perturbative
320: part~(\ref{eq:Adler}), the mass corrections~(\ref{eq:Dmass}) and the
321: non-perturbative part~(\ref{eq:DNP}):
322: \begin{equation}
323:   \label{eq:Dtot}
324:   D(-s) = D_{\rm P}(-s) + D_{\rm mass}(-s) + D_{\rm NP}(-s).
325: \end{equation}
326: 
327: Moments of similar form to~(\ref{eq:rkl}) can be defined for
328: $\Pep\Pem$ data~\cite{art:Davier}:
329: \begin{equation}
330:   \label{eq:repemkl}
331:   R_{\Pep\Pem}^{kl}(s_0) = \!\!\!\int\limits_{4m_\pi^2}^{s_0}\!\!
332:   \frac{{\rm d}s}{s_0}
333:   \left(1-\frac{s}{s_0}\right)^k\left(\frac{s}{s_0}\right)^l
334:   R_{\Pep\Pem}(s).
335: \end{equation}
336: In order to guarantee the validity of the OPE the endpoint in the
337: integrals should be suppressed by at least two powers of $s$ and the
338: moments chosen here are therefore $kl=20,30,31,32,33$. As in $\tau$
339: decays the moments can be rewritten as integrals around the circle
340: $|s|=s_0$ in the complex $s$-plane:
341: \begin{equation}
342:   \label{eq:rklcirc}
343:   R_{\Pep\Pem}^{kl}(s_0) = 6\pi i\!\!\!\oint\limits_{|s|=s_0}\!\!
344:   \frac{{\rm d}s}{s_0}\!
345:   \left(1-\frac{s}{s_0}\right)^k\!\!\left(\frac{s}{s_0}\right)^l\!\!
346:   \Pi^\gamma(s),
347: \end{equation}
348: and further written as integrals over the Adler function after
349: integrating by parts:
350: \begin{eqnarray}
351:   R_{\Pep\Pem}^{kl}(s_0) & = &\frac{3}{2\pi i}\!\!\!\oint\limits_{|s|=s_0}\!\!\!
352:   \frac{{\rm d}s}{s}\!
353:   \sum_{m=0}^k\!\frac{(-1)^m{k\choose m}}{m+l+1}\nonumber\\
354:   & & \times\left[1-\left(\frac{s}{s_0}\right)^{m+l+1}\right]D(s).  
355:   \label{eq:rkld}
356: \end{eqnarray}
357: 
358: In the QCD fits the dimension $D=6$ and $D=8$ terms are fitted using
359: $\langle{\cal O}_{6,8}\rangle$ as free parameters.  Note that this
360: differs from the $\tau$ data fits where the contributions to the
361: moment $kl=00$ are used as free parameters.  For the dimension $D=2$
362: and $D=4$ terms the full $s$-dependence is taken into account for the
363: theoretical description of the moments as described above.  In
364: contrast to the non-strange $\tau$ data where the least precisely
365: known $D=4$ parameter is the gluon condensate the largest uncertainty
366: for the $\Pep\Pem$ data comes from the strange quark mass $m_{\rm
367:   s}(1\,{\rm GeV})$ which is used as a free parameter for the $D=2$
368: and $D=4$ terms. The other quark-masses and -condensates needed to
369: complete the $D=2,4$ terms are taken from~\cite{art:Braaten}.
370: 
371: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
372: 
373: \section{RESULTS OF QCD FITS TO $\tau$ DATA} 
374: %
375: \begin{table*}[htb]
376:   \begin{center}
377:     \caption{\em Comparison of the QCD fit results (CIPT) to the
378:       moments of vector (axial-vector) current, the simultaneous fit
379:       of all moments for both currents, and the moments for the sum of
380:       both currents~\cite{art:SvenPhD}.  The quoted errors contain
381:       statistical errors only.}
382:     \begin{tabular}[t]{cllllllll}
383:       \hline
384:          & \multicolumn{2}{c}{V} & \multicolumn{2}{c}{A} & 
385:          \multicolumn{2}{c}{V and A } & \multicolumn{2}{c}{V + A }\\
386:          observable & \phantom{+}value & \phantom{+}error & 
387:          \phantom{+}value & \phantom{+}error & \phantom{+}value & 
388:          \phantom{+}error & \phantom{+}value & \phantom{+}error \\
389:       \hline               
390:    $\alphas(m_\tau^2)$   & 
391:    $\phantom{+}0.341$          & 
392:    $\pm 0.017$           &
393:    $\phantom{+}0.357$          & 
394:    $\pm 0.019$           &
395:    $\phantom{+}0.347$          & 
396:    $\pm 0.012$           &
397:    $\phantom{+}\MEANASVLMT$     &
398:    $\pm\MEANASXMMT$ \\
399:    $\langle\frac{\alphas}{\pi}\,GG\rangle/{\rm GeV^4}$ & 
400:    $\phantom{+}0.002$          & 
401:    $\pm 0.010$           &
402:    $-0.011$              & 
403:    $\pm 0.020$           &
404:    $\phantom{+}\MEANcGGVL$     & 
405:    $\pm 0.008$           &
406:    $\CIPTGGVL$           &
407:    $\pm\CIPTGGEM$\\
408:    $\delta_{\rm V}^6$    &  
409:    $\phantom{+}0.0259$         & 
410:    $\pm 0.0041$          &
411:      \phantom{+}---            & 
412:      \phantom{+}---            &
413:    $\phantom{+}\MEANcVSVL$     & 
414:    $\pm 0.0034$          & 
415:      \phantom{+}---            & 
416:      \phantom{+}---            \\
417:    $\delta_{\rm V}^8$& 
418:    $-0.0078$             & 
419:    $\pm 0.0018$          &
420:      \phantom{+}---            & 
421:      \phantom{+}---            &
422:    $\MEANcVEVL$          & 
423:    $\pm 0.0013$          &
424:      \phantom{+}---            & 
425:      \phantom{+}---            \\
426:    $\delta_{\rm A}^6$    &  
427:      \phantom{+}---            & 
428:      \phantom{+}---            &
429:    $-0.0246$             & 
430:    $\pm 0.0086$          &
431:    $\MEANcASVL$          & 
432:    $\pm 0.0033$          &
433:      \phantom{+}---            & 
434:      \phantom{+}---            \\
435:    $\delta_{\rm A}^8$    & 
436:      \phantom{+}---            & 
437:      \phantom{+}---            &
438:    $\phantom{+}0.0067$         & 
439:    $\pm 0.0050$          &
440:    $\phantom{+}\MEANcAEVL$     & 
441:    $\pm 0.0019$          &
442:      \phantom{+}---            & 
443:      \phantom{+}---            \\
444:    $\delta_{\rm V+A}^6$  &  
445:      \phantom{+}---            & 
446:      \phantom{+}---            &
447:      \phantom{+}---            & 
448:      \phantom{+}---            &
449:      \phantom{+}---            & 
450:      \phantom{+}---            &
451:      $\phantom{+}\CIPTSSVL$    &
452:      $\pm\CIPTSSEM$      \\
453:    $\delta_{\rm V+A}^8$  &  
454:      \phantom{+}---            & 
455:      \phantom{+}---            &
456:      \phantom{+}---            & 
457:      \phantom{+}---            &
458:      \phantom{+}---            & 
459:      \phantom{+}---            &
460:      $\CIPTSEVL$         &
461:      $\pm\CIPTSEEM$      \\
462:       \hline                                                
463:   $\chi^2/{\rm d.o.f.}$ &
464:   $\phantom{+}0.07/1$         & 
465:                         &
466:   $\phantom{+}0.06/1$         & 
467:                         &
468:   $\phantom{+}\CIPTcCHSQ/4$   &
469:                         &
470:   $\phantom{+} \CIPTCSMT/1$   &
471:                         \\
472:       \hline                       
473:     \end{tabular}
474:     \label{tab:V,A,VaA,VpA} \end{center}
475: \end{table*}                                
476: The fit results of QCD parameters to the ten moments $R_{\tau,{\rm
477:     V/A}}^{kl}$ for $kl=00,10,11,12,13$ as reported
478: in~\cite{art:Sven,art:SvenPhD} are shown in
479: Tab.~\ref{tab:V,A,VaA,VpA}.  Only the CIPT fits are presented here,
480: since the focus is the stability of the QCD fits and the two other
481: approaches lead to similar results within the theoretical
482: uncertainties. The various fits demonstrate the stability of the
483: method and that the perturbative parameter $\alpha_{\rm s}$ can be
484: measured together with the non-perturbative parts even if most of the
485: non-perturbative parts do not cancel as it is the case in the first
486: three fits.
487: 
488: Note that the statistical error on $\alpha_{\rm s}$ in the last two
489: fits is different only because the last fit uses additional
490: information from the $\tau$ lifetime and the leptonic branching
491: ratios~\cite{art:Sven}.
492: 
493: \begin{figure}[htb]
494: \centering
495: \resizebox{.5\textwidth}{!}{
496: \includegraphics{fig01.eps}}
497: \caption{\em $R_{\tau'}^{V+A}(s_0)$ for a hypothetical $\tau'$ lepton
498:   versus the upper integration limit $s_0 = m_{\tau'}^2$.}
499: \label{fig:rtau}
500: \end{figure}
501: The extension to energy scales $s<m_\tau^2$ is demonstrated in
502: Fig.~\ref{fig:rtau}. Here the moment
503: $R_{\tau',V+A}^{00}(s_0=m_{\tau'}^2)$ is shown together with the fit
504: results to the three different theories FOPT, CIPT and RCPT for the
505: perturbative part plus non-perturbative parts.
506: 
507: The error on the theoretical curves (or alternatively their spread)
508: indicates that the OPE is applicable to values as low as $1.5\,{\rm
509:   GeV}^2$ which can be used to extract the running of the strong
510: coupling constant from $\tau$ data alone.
511: 
512: The three considered theories differ in the treatment of unknown
513: higher order terms which is partially taken into account as
514: theoretical uncertainty by means of the error on $K_4$. As a
515: conservative approach one could take the average of all three results
516: and their spread as additional theoretical uncertainty. From the
517: results in~\cite{art:Sven} one obtains therefore:
518: \begin{equation}\label{eq:asmtau}
519:   \alpha_{\rm s}(m_\tau^2) = 0.326 \pm 0.007_{\rm exp} \pm 0.022_{\rm theo},
520: \end{equation}
521: which translates to
522: \begin{equation}\label{eq:asmz}
523:   \alpha_{\rm s}(m_{\rm Z}^2) = 0.1194 \pm 0.0008_{\rm exp} \pm 0.0027_{\rm theo}
524: \end{equation}
525: at $m_{\rm Z}=91.188\,{\rm GeV}$.
526: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
527: 
528: \section{AVERAGING OF $\Pep\Pem$ DATA}
529: %
530: Exclusive $\Pep\Pem$ channels up to a CMS energy of $2.2\,{\rm
531:   GeV}$~\cite{Barkov:1985ac,Vasserman:1979hw,Vasserman:1981xq,Amendolia:1984di,Quenzer:1978qt,Bisello:1989hq,art:CMD2:2pi,Dolinsky:1991vq,Cosme:1976tf,Parrour:1976rt,Cosme:1979qe,Cordier:1980qg,Antonelli:1992jx,art:SND:2pipi0,Kurdadze:1986tc,Bisello:1991du,Esposito:1980by,Barkov:1988gp,Cordier:1982zn,art:CMD2:4pi,Antonelli:1988fw,Bisello:1981sh,Ivanov:1981wf,Ivanov:1982cr,Delcourt:1981eq,Mane:1981ep,Bisello:1988ez,Cordier:1982en,Mane:1982si}
532: are combined in this analysis to calculate the total hadronic cross
533: section.  The narrow $\omega$ and $\phi$ resonances are excluded from
534: the exclusive channels and added to the total cross section as
535: relativistic Breit-Wigner curves with $s$-dependent
536: widths~\cite{Eidelman:1995ny,art:PDG00}.  The weights of individual
537: channels and the treatment of systematic errors is taken
538: from~\cite{art:Davier} but the statistical errors and the final result
539: are obtained in a slightly different manner:
540: \begin{Enumerate}
541: \item a common equidistant binning in energy is chosen for all
542:   channels ($N=300$ bins from $E_{\rm min} = 0\,{\rm GeV}$ to $E_{\rm
543:     max} = \sqrt{5}\,{\rm GeV}$) and for every experiment and channel
544:   a histogram is filled with weighted averages of cross section
545:   measurements falling in the same bin:
546:   \begin{equation}
547:     \label{eq:trap}
548:     d_i = \frac{\displaystyle\sum_{\{j|N E_j/E_{\rm max}\in[i-1,i]\}}{
549:         \!\!\!\!\!\!\!\!\!\!\!\!\!\!\sigma_j/\Delta^2\sigma_j}}
550:     {\displaystyle\sum_{\{j|N E_j/E_{\rm max}\in[i-1,i]\}}{
551:         \!\!\!\!\!\!\!\!\!\!\!\!\!\!1/\Delta^2\sigma_j}},\quad i=1\dots N,
552:   \end{equation}
553:   where the squared errors $\Delta_i^2$ are given by the inverse of
554:   the denominator in~(\ref{eq:trap}).
555: \item gaps at $k=i\dots j$ between bins $i$ and $j$ of individual
556:   measurements are interpolated with the trapezoidal rule:
557:   \begin{equation}
558:     \label{eq:trap1}
559:     d_k = c_k d_i + (1-c_k) d_j,\quad c_k = 
560:     \frac{k-i}{j-i}.
561:   \end{equation}
562: \item the statistical errors are interpolated with the same procedure:
563:   \begin{equation}
564:     \label{eq:trap2}
565:     \Delta_k = c_k \Delta_i + (1-c_k) \Delta_j,
566:   \end{equation}
567:   and are therefore larger compared to Gaussian propagated errors by
568:   the factor:
569:   \begin{equation}
570:     \label{eq:trap3}
571:     r_k = \frac{c_k \Delta_i + 
572:       (1-c_k) \Delta_j}{c_k \Delta_i 
573:       \oplus (1-c_k) \Delta_j}.
574:   \end{equation}
575: \item the correlation matrix for the interpolated parts is given by:
576:   \begin{equation}
577:     \label{eq:trap4}
578:     \!\!\!\!V^{\rm stat}_{kl} = r_k r_l \left( c_k c_l \Delta_i^2 
579:       + (1-c_k)(1-c_l)\Delta_j^2\right).
580:   \end{equation}
581: \item the systematic errors are also interpolated according to the
582:   trapezoidal rule~(\ref{eq:trap2}) and all systematic errors are
583:   assumed to be $100\,\%$ correlated.
584: \item the final error matrix is given by the sum of both:
585:   \begin{equation}
586:     \label{eq:trap5}
587:     V_{kl} = V^{\rm stat}_{kl} + V^{\rm sys}_{kl}.
588:   \end{equation}
589: \item the averaged cross section over all experiments in bin $i$ for
590:   one exclusive channel is obtained from the weighted mean of all
591:   measurements and its error by usual error propagation. In case of
592:   inconsistent sets of data points for a particular bin its final
593:   error is scaled by the ratio $S = \sqrt{\chi^2/\chi^2_{68\,\%}}$, if
594:   $S>1$, where $\chi^2$ is the calculated $\chi^2$ for the individual
595:   experiments being consistent with the average and $\chi^2_{68\,\%}$
596:   is the $\chi^2$ value corresponding to a confidence level of
597:   $68\,\%$.
598: \end{Enumerate}
599: After this interpolation and averaging procedure the total hadronic
600: cross section is given by the sum of the exclusive channels as
601: in~\cite{art:Davier}.  The resulting values for $R_{\Pep\Pem}$ are
602: shown in Fig.~\ref{fig:repem}.
603: \begin{figure}[htb]
604: \centering
605: \resizebox{.5\textwidth}{!}{
606: \includegraphics{fig02.eps}}
607: \caption{\em $R_{\Pep\Pem}(s)$ from exclusive channels. The dark gray
608:   band denotes statistical errors; the light gray band shows the sum
609:   of systematic and statistical errors. The dark solid line shows the
610:   relative uncertainty; the small dark error band beyond $s=2\,GeV^2$
611:   shows the projected uncertainty after 5 years of
612:   PEP-N~\cite{art:PEP-N}.  The narrow $\omega$ and $\phi$ resonances
613:   and the massless QCD prediction are also indicated by solid lines.}
614: \label{fig:repem}
615: \end{figure}
616: 
617: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
618: 
619: \section{RESULTS OF QCD FITS TO $\Pep\Pem$ DATA}
620: %
621: The five moments $R_{\Pep\Pem}^{kl}(4\,{\rm GeV}^2)$ for
622: $kl=20,30,31,32,33$ are given in Tab.~\ref{tab:repemkl}.  Their
623: correlations are given in Tab.~\ref{tab:repemklcorr}. The result of
624: the CIPT fit to these moments is given in Tab.~\ref{tab:repemfit}.
625: The correlations of the fit parameters are shown in
626: Tab.~\ref{tab:repemfitcorr}.
627: \begin{table}[htb]
628:   \begin{center}
629:     \caption{\em Moments $R_{\Pep\Pem}^{kl}(4\,GeV^2)$ for
630:       $kl=20,30,31,32,33$ from exclusive $\Pep\Pem$ data.  The errors
631:       include statistical and systematic uncertainties; projected
632:       errors after 5 years running of PEP-N and theoretical
633:       uncertainties are also given.}
634:     \begin{tabular}[t]{cllll}
635:       \hline
636:          $kl$ & $R_{\Pep\Pem}^{kl}$ & stat.+sys. & PEP-N  & theo.\\
637:       \hline               
638:         $20$  & $0.760$  & $0.014$  & $0.009$  & $0.008$ \\    
639:         $30$  & $0.569$  & $0.008$  & $0.007$  & $0.012$ \\
640:         $31$  & $0.1206$ & $0.0027$ & $0.0016$ & $0.0040$\\
641:         $32$  & $0.0355$ & $0.0015$ & $0.0006$ & $0.0000$\\
642:         $33$  & $0.0143$ & $0.0009$ & $0.0003$ & $0.0000$\\
643:       \hline
644:     \end{tabular}\label{tab:repemkl} 
645:   \end{center}
646: \end{table}
647: 
648: \begin{table}[htb]
649:   \begin{center}
650:     \caption{\em Correlations of the moments $R_{\Pep\Pem}^{kl}(4\,GeV^2)$ for
651:       $kl=20,30,31,32,33$ in percent. The values correspond to the
652:       quadratic sum of experimental and theoretical errors.}
653:     \begin{tabular}[t]{crrrr}
654:       \hline
655:          $kl$ & $30$     & $31$     & $32$     & $33$    \\
656:       \hline               
657:         $20$  & $+86.6$  & $ +3.8$  & $+74.5$  & $+70.4$ \\    
658:         $30$  &          & $-44.2$  & $+33.6$  & $+30.0$ \\
659:         $31$  &          &          & $+54.9$  & $+50.9$ \\
660:         $32$  &          &          &          & $+98.8$ \\
661:       \hline
662:     \end{tabular}\label{tab:repemklcorr} 
663:   \end{center}
664: \end{table}
665: 
666: \begin{table}[htb]
667:   \begin{center}
668:     \caption{\em Results from the QCD fits to the moments 
669:       $R_{\Pep\Pem}^{kl}(4\,GeV^2)$ for $kl=20,30,31,32,33$ from
670:       exclusive $\Pep\Pem$ data.  The errors include statistical and
671:       systematic uncertainties; projected errors after 5 years running
672:       of PEP-N and theoretical uncertainties are also given.}
673:     \begin{tabular}[t]{cllll}
674:       \hline
675:          obs. & \phantom{+}val. & err. & PEP-N  & theo.\\
676:       \hline               
677:         $\alpha_{\rm s}(4\,{\rm GeV}^2)$ & 
678:         $\phantom{+}0.286$  & $0.031$  & $0.027$  & $0.015$ \\ 
679:         $m_{\rm s}/{\rm GeV}$ & 
680:         $\phantom{+}0.220$  & $0.036$  & $0.026$  & $0.059$ \\
681:         $\langle{\cal O}_6\rangle/{\rm GeV}^6$ & 
682:         $-0.0041$ & $0.0007$ & $0.0005$ & $0.0002$\\
683:         $\langle{\cal O}_8\rangle/{\rm GeV}^8$ & 
684:         $\phantom{+}0.0043$ & $0.0004$ & $0.0002$ & $0.0002$\\
685:       \hline
686:         $\chi^2/{\rm d.o.f.}$ & $\phantom{+}0.04/1$ & & \\
687:       \hline
688:     \end{tabular}\label{tab:repemfit} 
689:   \end{center}
690: \end{table}
691: 
692: \begin{table}[htb]
693:   \begin{center}
694:     \caption{\em Correlations of the QCD fit parameters to the moments 
695:       $R_{\Pep\Pem}^{kl}(4\,GeV^2)$ for $kl=20,30,31,32,33$ in
696:       percent. The values correspond to the quadratic sum of
697:       experimental and theoretical errors.}
698:     \begin{tabular}[t]{crrr}
699:       \hline
700:          obs. & $m_{\rm s}$ & $\langle{\cal O}_6\rangle$ & 
701:         $\langle{\cal O}_8\rangle$ \\
702:       \hline               
703:         $\alpha_{\rm s}$            & $+77.2$  & $+71.4$  & $-42.0$ \\    
704:         $m_{\rm s}$                 &          & $+35.9$  & $-8.2$  \\
705:         $\langle{\cal O}_6\rangle$  &          &          & $-93.2$ \\
706:       \hline
707:     \end{tabular}\label{tab:repemfitcorr} 
708:   \end{center}
709: \end{table}
710: The value of the strong coupling at $4\,{\rm GeV}^2$ corresponds to
711: \begin{equation}\label{eq:asepemmz}
712:   \alpha_{\rm s}(m_{\rm Z}^2) = 0.117 \pm 0.005_{\rm exp} \pm 0.002_{\rm theo}
713: \end{equation}
714: at $m_{\rm Z}=91.188\,{\rm GeV}$ in good agreement with the $\tau$
715: result~\ref{eq:asmz} but with a larger experimental uncertainty.
716: \begin{figure}[htb]
717: \centering
718: \resizebox{.5\textwidth}{!}{
719: \includegraphics{fig03.eps}}
720: \caption{\em The moments $R_{\Pep\Pem}^{kl=20,30,31,32,33}(s_0)$
721:   versus the integration limit $s_0$. The shaded bands show the
722:   experimental moments with statistical and systematic uncertainties
723:   from top to bottom in the order $kl=20,30,31,32,33$. The dotted
724:   curves denote central values and $\pm 1\sigma$ ranges for the
725:   theoretical expectations (CIPT) using the fit values to the moments
726:   at $s_0=4\,GeV^2$ as input.}
727: \label{fig:epemfit}
728: \end{figure}
729: Figure~\ref{fig:epemfit} shows the moments of $R_{\Pep\Pem}$ as a
730: function of the upper integration limit together with theoretical
731: predictions using the fitted values at $s_0=4\,{\rm GeV}^2$ as input.
732: Reasonable agreement between the extrapolated and the measured moments
733: is observed for all five moments down to $s_0\approx2.5\,{\rm GeV}^2$.
734: The first three moments agree within errors with the extrapolated
735: results down to $s_0\approx1.5\,{\rm GeV}^2$.
736: 
737: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
738: 
739: 
740: \section{CONCLUSIONS}
741: %
742: The comparison of QCD fits to different currents (V, A, V and A, V+A)
743: in non-strange hadronic $\tau$ decays probes the stability of the OPE
744: at low energy scales. The good agreement of perturbative and
745: non-perturbative parameters among these fits demonstrates that QCD
746: fits can reliably be performed at these low energy scales.  The use of
747: spectral moments is therefore extended to $\Pep\Pem\rightarrow{\rm
748:   hadrons}$ data in order to fit QCD parameters at $4\,{\rm GeV}^2$.
749: While the extraction of the strong coupling $\alpha_{\rm s}$ is not
750: competitive in terms of experimental uncertainties with the
751: measurement from $\tau$ data it provides an important cross check and
752: probes an energy region not accessible with $\tau$ decays. Furthermore
753: the sensitivity to the mass of the strange quark can be used to
754: constrain this parameter. A value of $m_{\rm s}(1\,{\rm GeV}) =
755: (220\pm36\pm59)\,{\rm MeV}$ has been observed.  The experimental
756: uncertainty could be reduced by $30\,\%$ after 5 years of running with
757: PEP-N.
758: 
759: \section{ACKNOWLEDGMENTS}
760: I would like to thank the organizers Stan Brodsky and Rinaldo Baldini
761: for giving me the opportunity to present this work. I am also indebted
762: to Andreas H{\"o}cker who provided me with FORTRAN files of the $\Pep\Pem$
763: data.
764: 
765: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
766: \begin{thebibliography}{10}
767: 
768: \bibitem{art:Andreas98}
769: ALEPH Collaboration, R.~Barate et~al.,
770: \newblock Eur.~Phys.~J. {\bf C4} (1998) 409.
771: 
772: \bibitem{art:Sven}
773: OPAL Collaboration, {K. Ackerstaff} et~al.,
774: \newblock Eur.~Phys.~J. {\bf C7} (1999) 571.
775: 
776: \bibitem{art:PEP-N}
777: R.~Baldini et~al.,
778: \newblock
779:   http://www.slac.stanford.edu/grp/rd/epac/\\LOI/LOI-2000.3-Experiment.pdf
780:   (2000).
781: 
782: \bibitem{art:Braaten88}
783: E.~Braaten,
784: \newblock Phys.~Rev.~Lett. {\bf 60} (1988) 1606.
785: 
786: \bibitem{art:Braaten89}
787: E.~Braaten,
788: \newblock Phys.~Rev. {\bf D39} (1989) 1458.
789: 
790: \bibitem{art:Narison88}
791: S.~Narison and A.~Pich,
792: \newblock Phys.~Lett. {\bf B211} (1988) 183.
793: 
794: \bibitem{art:Braaten}
795: E.~Braaten, S.~Narison, and A.~Pich,
796: \newblock Nucl.~Phys. {\bf B373} (1992) 581.
797: 
798: \bibitem{art:Pich92}
799: F.~{Le~Diberder} and A.~Pich,
800: \newblock Phys.~Lett. {\bf B286} (1992) 147.
801: 
802: \bibitem{art:Marciano}
803: {W.J. Marciano} and {A. Sirlin},
804: \newblock Phys.~Rev.~Lett. {\bf 61} (1988) 1815.
805: 
806: \bibitem{art:PDG00}
807: {D.E Groom} et~al.,
808: \newblock Eur.~Phys.~J. {\bf C15} (2000) 1.
809: 
810: \bibitem{art:Diberder}
811: F.~{Le~Diberder} and A.~Pich,
812: \newblock Phys.~Lett. {\bf B289} (1992) 165.
813: 
814: \bibitem{art:Chetyrkin}
815: {K.G. Chetyrkin}, {A.L. Kataev}, and {F.V. Tkachev},
816: \newblock Phys.~Lett. {\bf B85} (1979) 277.
817: 
818: \bibitem{art:Dine}
819: M.~Dine and J.~Sapirstein,
820: \newblock Phys.~Rev.~Lett. {\bf 43} (1979) 668.
821: 
822: \bibitem{art:Celmaster}
823: W.~Celmaster and {R.J. Gonsalves},
824: \newblock Phys.~Rev.~Lett. {\bf 44} (1980) 560.
825: 
826: \bibitem{art:Gorishnii}
827: {S.G. Gorishnii}, {A.L. Kataev}, and {S.A. Larin},
828: \newblock Phys.~Lett. {\bf B259} (1991) 144.
829: 
830: \bibitem{art:Surguladze}
831: {L.R. Surguladze} and {M.A. Samuel},
832: \newblock Phys.~Rev.~Lett. {\bf 66} (1991) 560,
833: \newblock Erratum: ibid. {\bf 66} (1991) 2416.
834: 
835: \bibitem{art:Pich97}
836: A.~Pich,
837: \newblock  {\bf FTUV-97-03} (1997).
838: 
839: \bibitem{art:Kataev}
840: {A.L. Kataev} and {V.V. Starshenko},
841: \newblock Mod.~Phys.~Lett. {\bf A10} (1995) 235.
842: 
843: \bibitem{art:Ball}
844: P.~Ball, M.~Beneke, and {V.M. Braun},
845: \newblock Nucl.~Phys. {\bf B452} (1995) 563.
846: 
847: \bibitem{art:Neubert}
848: M.~Neubert,
849: \newblock Nucl.~Phys. {\bf B463} (1996) 511.
850: 
851: \bibitem{art:Maxwell95}
852: {C.N. Lovett-Turner} and {C.J. Maxwell},
853: \newblock Nucl.~Phys. {\bf B452} (1995) 188.
854: 
855: \bibitem{art:Maxwell96}
856: {C.J. Maxwell} and {D.G. Tonge},
857: \newblock Nucl.~Phys. {\bf B481} (1996) 681.
858: 
859: \bibitem{art:Wilson}
860: {K.G. Wilson},
861: \newblock Phys.~Rev. {\bf 179} (1969) 1499.
862: 
863: \bibitem{art:SVZ1}
864: {M.A. Shifman}, {A.I. Vainshtein}, and {V.I. Zakharov},
865: \newblock Nucl.~Phys. {\bf B147} (1979) 385.
866: 
867: \bibitem{art:SVZ2}
868: {M.A. Shifman}, {A.I. Vainshtein}, and {V.I. Zakharov},
869: \newblock Nucl.~Phys. {\bf B147} (1979) 448.
870: 
871: \bibitem{art:SVZ3}
872: {M.A. Shifman}, {A.I. Vainshtein}, and {V.I. Zakharov},
873: \newblock Nucl.~Phys. {\bf B147} (1979) 519.
874: 
875: \bibitem{art:Adler}
876: {S.L. Adler},
877: \newblock Phys.~Rev. {\bf D10} (1974) 3714.
878: 
879: \bibitem{art:Chetyrkin90}
880: K.~Chetyrkin and J.~K{{\"u}}hn,
881: \newblock Phys. Lett. {\bf B248} (1990) 359.
882: 
883: \bibitem{art:Vermaseren97}
884: J.~Vermaseren, S.~Larin, and T.~van Ritbergen,
885: \newblock Phys. Lett. {\bf B405} (1997) 327.
886: 
887: \bibitem{art:Davier}
888: M.~Davier and A.~H{\"o}cker,
889: \newblock Phys. Lett. {\bf B419} (1998) 419.
890: 
891: \bibitem{art:SvenPhD}
892: S.~Menke,
893: \newblock BONN-IR-98-13.
894: 
895: \bibitem{Barkov:1985ac}
896: L.~M. Barkov et~al.,
897: \newblock Nucl. Phys. {\bf B256} (1985) 365.
898: 
899: \bibitem{Vasserman:1979hw}
900: I.~B. Vasserman et~al.,
901: \newblock Yad. Fiz. {\bf 30} (1979) 999.
902: 
903: \bibitem{Vasserman:1981xq}
904: I.~B. Vasserman et~al.,
905: \newblock Yad. Fiz. {\bf 33} (1981) 709.
906: 
907: \bibitem{Amendolia:1984di}
908: S.~R. Amendolia et~al.,
909: \newblock Phys. Lett. {\bf B138} (1984) 454.
910: 
911: \bibitem{Quenzer:1978qt}
912: A.~Quenzer et~al.,
913: \newblock Phys. Lett. {\bf B76} (1978) 512.
914: 
915: \bibitem{Bisello:1989hq}
916: DM2, D.~Bisello et~al.,
917: \newblock Phys. Lett. {\bf B220} (1989) 321.
918: 
919: \bibitem{art:CMD2:2pi}
920: CMD-2, A.~Bondar et~al.,
921: \newblock talk at ICHEP 2000, Osaka, Japan  (2000).
922: 
923: \bibitem{Dolinsky:1991vq}
924: S.~I. Dolinsky et~al.,
925: \newblock Phys. Rept. {\bf 202} (1991) 99.
926: 
927: \bibitem{Cosme:1976tf}
928: G.~Cosme et~al.,
929: \newblock Phys. Lett. {\bf B63} (1976) 349.
930: 
931: \bibitem{Parrour:1976rt}
932: G.~Parrour et~al.,
933: \newblock Phys. Lett. {\bf B63} (1976) 357.
934: 
935: \bibitem{Cosme:1979qe}
936: G.~Cosme et~al.,
937: \newblock Nucl. Phys. {\bf B152} (1979) 215.
938: 
939: \bibitem{Cordier:1980qg}
940: A.~Cordier et~al.,
941: \newblock Nucl. Phys. {\bf B172} (1980) 13.
942: 
943: \bibitem{Antonelli:1992jx}
944: DM2, A.~Antonelli et~al.,
945: \newblock Z. Phys. {\bf C56} (1992) 15.
946: 
947: \bibitem{art:SND:2pipi0}
948: M.~Achasov et~al.,
949: \newblock Phys. Lett. {\bf B462} (1999) 365.
950: 
951: \bibitem{Kurdadze:1986tc}
952: L.~M. Kurdadze et~al.,
953: \newblock JETP Lett. {\bf 43} (1986) 643.
954: 
955: \bibitem{Bisello:1991du}
956: DM2, D.~Bisello et~al.,
957: \newblock Nucl. Phys. Proc. Suppl. {\bf 21} (1991) 111.
958: 
959: \bibitem{Esposito:1980by}
960: B.~Esposito et~al.,
961: \newblock Lett. Nuovo Cim. {\bf 28} (1980) 195.
962: 
963: \bibitem{Barkov:1988gp}
964: L.~M. Barkov et~al.,
965: \newblock Sov. J. Nucl. Phys. {\bf 47} (1988) 248.
966: 
967: \bibitem{Cordier:1982zn}
968: A.~Cordier et~al.,
969: \newblock Phys. Lett. {\bf B109} (1982) 129.
970: 
971: \bibitem{art:CMD2:4pi}
972: CMD-2, R.~R. Akhmetshin et~al.,
973: \newblock Phys. Lett. {\bf B475} (2000) 190.
974: 
975: \bibitem{Antonelli:1988fw}
976: DM2, A.~Antonelli et~al.,
977: \newblock Phys. Lett. {\bf B212} (1988) 133.
978: 
979: \bibitem{Bisello:1981sh}
980: D.~Bisello et~al.,
981: \newblock Phys. Lett. {\bf B107} (1981) 145.
982: 
983: \bibitem{Ivanov:1981wf}
984: P.~M. Ivanov et~al.,
985: \newblock Phys. Lett. {\bf B107} (1981) 297.
986: 
987: \bibitem{Ivanov:1982cr}
988: P.~m.~Ivanov et~al.,
989: \newblock JETP Lett. {\bf 36} (1982) 112.
990: 
991: \bibitem{Delcourt:1981eq}
992: B.~Delcourt et~al.,
993: \newblock Phys. Lett. {\bf B99} (1981) 257.
994: 
995: \bibitem{Mane:1981ep}
996: F.~Mane et~al.,
997: \newblock Phys. Lett. {\bf B99} (1981) 261.
998: 
999: \bibitem{Bisello:1988ez}
1000: DM2, D.~Bisello et~al.,
1001: \newblock Z. Phys. {\bf C39} (1988) 13.
1002: 
1003: \bibitem{Cordier:1982en}
1004: A.~Cordier et~al.,
1005: \newblock Phys. Lett. {\bf B110} (1982) 335.
1006: 
1007: \bibitem{Mane:1982si}
1008: F.~Mane et~al.,
1009: \newblock Phys. Lett. {\bf B112} (1982) 178.
1010: 
1011: \bibitem{Eidelman:1995ny}
1012: S.~Eidelman and F.~Jegerlehner,
1013: \newblock Z. Phys. {\bf C67} (1995) 585.
1014: 
1015: \end{thebibliography}
1016: 
1017: %
1018: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
1019: %
1020: 
1021: \end{document}
1022: 
1023: