1: %%UNIX --- UPDATED ON 13/8/97
2: %====================================================================%
3: % sprocl.tex 27-Feb-1995 %
4: % This latex file rewritten from various sources for use in the %
5: % preparation of the standard proceedings Volume, latest version %
6: % by Susan Hezlet with acknowledgments to Lukas Nellen. %
7: % Some changes are due to David Cassel. %
8: %====================================================================%
9: %\input{psfig}
10: %\input psfig
11: %for BibTeX - sorted numerical labels by
12: %order of first citation.
13: % A useful Journal macro
14: % Some useful journal names
15: % Some other macros used in the sample text
16: %temp replacemt due to no font
17: %%BEGINNING OF TEXT
18:
19: \documentstyle[sprocl,psfig]{article}
20: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
21: %TCIDATA{OutputFilter=Latex.dll}
22: %TCIDATA{LastRevised=Wed Sep 20 03:23:27 2000}
23: %TCIDATA{<META NAME="GraphicsSave" CONTENT="32">}
24: %TCIDATA{CSTFile=article.cst}
25:
26: \font\eightrm=cmr8
27: \bibliographystyle{unsrt}
28: \arraycolsep1.5pt
29: \def\Journal#1#2#3#4{{#1} {\bf #2}, #3 (#4)}
30: \def\NCA{\em Nuovo Cimento}
31: \def\NIM{\em Nucl. Instrum. Methods}
32: \def\NIMA{{\em Nucl. Instrum. Methods} A}
33: \def\NPB{{\em Nucl. Phys.} B}
34: \def\PLB{{\em Phys. Lett.} B}
35: \def\PRL{\em Phys. Rev. Lett.}
36: \def\PRD{{\em Phys. Rev.} D}
37: \def\ZPC{{\em Z. Phys.} C}
38: \def\st{\scriptstyle}
39: \def\sst{\scriptscriptstyle}
40: \def\mco{\multicolumn}
41: \def\epp{\epsilon^{\prime}}
42: \def\vep{\varepsilon}
43: \def\ra{\rightarrow}
44: \def\ppg{\pi^+\pi^-\gamma}
45: \def\vp{{\bf p}}
46: \def\ko{K^0}
47: \def\kb{\bar{K^0}}
48: \def\al{\alpha}
49: \def\ab{\bar{\alpha}}
50: \def\be{\begin{equation}}
51: \def\ee{\end{equation}}
52: \def\bea{\begin{eqnarray}}
53: \def\eea{\end{eqnarray}}
54: \def\CPbar{\hbox{{\rm CP}\hskip-1.80em{/}}}
55: %\input{tcilatex}
56:
57: \begin{document}
58:
59: \title{INSTANTONS AND NUCLEON MAGNETISM }
60: \author{H. FORKEL}
61: \address{Institut f{\"u}r Theoretische Physik, Universit{\"a}t Heidelberg, \\
62: D-69120 Heidelberg, Germany}
63: \maketitle
64: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
65:
66: \abstracts{ We construct improved QCD sum rules for the nucleon magnetic
67: moments by implementing direct-instanton contributions to the operator
68: product expansion of the nucleon correlator in a magnetic background field.
69: The instanton contributions are found to affect only those sum
70: rules which had previously been considered unstable. The new sum rules
71: show a high degree of stability and reproduce the experimental values
72: of the nucleon magnetic
73: moments for values of the magnetic quark condensate susceptibility which
74: are consistent with other estimates. (Invited talk given at ``Hadron
75: Physics 2000'', Caraguatatuba, S\~ao Paulo, Brazil (April 10-15,
76: 2000). }
77:
78:
79: Over the last years, an increasingly multifaceted picture of the role of QCD
80: instantons \cite{bel75} in hadron physics has begun to emerge from hadron
81: models with more or less instanton-induced structure \cite{hor78},
82: instanton-liquid vacuum models\cite{sch98}, a generalized operator product
83: expansion including instanton contributions (IOPE) \cite{for00} and, most
84: recently, lattice calculations \cite{lat}.
85:
86: One of the major benefits of sharpening this picture by identifying specific
87: hadronic instanton effects will be to end the puzzling elusiveness of
88: explicit glue in hadrons at low energies. Indeed, in most successful models
89: of the classical mesons and baryons, gluons play at best a marginal role.
90: The situation is further obfuscated by the fact that gluons do not directly
91: couple to electroweak probes. Since much of what we know about the classical
92: hadrons derives from their response to such probes, it is particularly
93: desirable to pin down traces of nonperturbative glue in electromagnetic
94: hadron observables.
95:
96: In this talk, I will report on an investigation \cite{aw99} which takes a
97: step towards uncovering the hidden role of glue in the response to
98: electromagnetic fields by studying the impact of instantons on the nucleon
99: magnetic moments in the recently developed IOPE sum-rule approach \cite
100: {for00,FB}. While in general the range of applicability of the IOPE (as
101: a short-distance expansion) is more limited than that of instanton vacuum
102: models or lattice calculations, it profits from the transparency of an
103: analytical (yet largely model-independent) method and takes, in contrast to
104: instanton models, all long-wavelength vacuum fields and also perturbative
105: fluctuations into account.
106:
107: In order to study the nucleon's magnetic moments in the framework of the
108: IOPE, we start from a correlation function which contains the response of
109: the nucleon to the electromagnetic current at short, spacelike distances.
110: (The information on the magnetic moments is subsequently extracted by
111: dispersion techniques \cite{for00,svz79}.) At first sight, the natural
112: candidate for such a correlation function would be
113: \begin{equation}
114: \Pi _{\mu }\left( p,q\right) =-\int d^{4}xe^{ipx}\int
115: d^{4}ye^{iqy}\left\langle 0\left| T\eta \left( x\right) J_{\mu }
116: \left( y\right) \bar{\eta}\left( 0\right) \right| 0\right\rangle
117: \label{corr0}
118: \end{equation}
119: where the nucleon interpolating fields $\eta (x)$ (with either proton or
120: neutron quantum numbers) are composite operators of massless up and down
121: quark fields \cite{I},
122: \begin{equation}
123: \eta _{p}(x)=\epsilon ^{abc} [u^{a^{T}}(x)C\gamma _{\alpha }u^{b}(x)]
124: \gamma_{5}\gamma ^{\alpha }d^{c}(x),\qquad \eta_{n}
125: =\eta_{p}(u\leftrightarrow d), \label{interpol} \end{equation}
126: and $J_{\mu }$ is the electromagnetic current. However,
127: obtaining the nucleon magnetic moments requires taking the $q\rightarrow 0$
128: limit in which the matrix element of (\ref{corr0}) is probed mainly
129: in the large-distance region $\left| y\right| ,\left|
130: x-y\right| \gg \Lambda ^{-1}$ where the operator product expansion breaks
131: down\footnote{
132: The OPE\ of this matrix element can be used, however, to obtain the nucleon
133: form factors at intermediate momentum transfers $Q\sim 1$ GeV \cite{bel93}$.$
134: }.
135:
136: Fortunately, there exists an alternative approach, due to Ioffe and Smilga
137: \cite {IS} and Balitsky and Yung \cite{BY}, which can be directly applied to
138: the calculation of the magnetic moments. The underlying idea is to treat the
139: long-wavelength photons at the same level as the soft QCD vacuum fields
140: (which make up the condensates), i.e. to account for them in the operators
141: of the (I)OPE (see below). This amounts to considering the electromagnetic
142: field as a constant, classical background field (which, incidentally, is
143: reminiscent of the experimental setup used in actual measurements of the
144: moments). In other words, the electromagnetic probe does not
145: appear anymore as an external field but rather as part of the
146: ``magnetized'' vacuum state.
147:
148: We are thus led to consider the nucleon correlation function
149: \begin{eqnarray}
150: \Pi (p) &=&i\int d^{4}x\,e^{ipx}{\langle 0|T\eta (x)\bar{\eta}(0)|0\rangle }
151: _{F} \label{corr1} \\
152: &=&\Pi _{0}(p)+\sqrt{4\pi \alpha }\Pi _{\mu \nu }(p)F^{\mu \nu }+O\left(
153: F^{2}\right) \label{corr4}
154: \end{eqnarray}
155: in the background of a constant electromagnetic field $F_{\mu \nu }$. We
156: have not written down higher orders of $F$ since the information on the
157: magnetic moments is part of the linear response
158: \begin{equation}
159: \Pi _{\mu \nu }(p)=(\rlap/p\sigma _{\mu \nu }+\sigma _{\mu \nu }\rlap/
160: p)\,\Pi _{1}(p^{2})+i(\gamma _{\mu }p_{\nu }-\gamma _{\nu }p_{\mu })\rlap/
161: p\,\,\Pi _{2}(p^{2})+\sigma _{\mu \nu }\,\Pi _{3}(p^{2}) \label{lresp}
162: \end{equation}
163: to the (arbitrarily weak) external field. Above we have given the general
164: decomposition of $\Pi _{\mu \nu }(p)$ in terms of three independent Lorentz
165: and spinor structures which are associated with one chirally-even ($\Pi _{1}$)
166: and two chirally-odd invariant amplitudes ($\Pi _{2}$ and $\Pi _{3}$). The
167: relation to the magnetic moments can be made explicit by writing
168: \begin{equation}
169: \Pi (p)=i\int d^{4}x\,e^{ipx}{\langle 0|T\eta (x)\bar{\eta}(0)e}^{-i\int
170: d^{4}yA^{\mu }J_{\mu }}{|0\rangle } \label{corr3}
171: \end{equation}
172: where $J_{\mu }={\bar{q}\gamma }_{\mu }{Qq}$ is the
173: electromagnetic current with quark charge matrix $Q$ and $A_{\mu }$
174: is the external vector potential. In
175: fixed-point gauge, and specializing to a constant magnetic background field
176: \begin{equation}
177: B_{i}=-\frac{1}{2}\varepsilon_{ijk} F_{jk}, \hspace{1.2cm}
178: \left(E_{i}=F_{i0}=0\right)
179: \end{equation}
180: with the potential
181: \begin{equation}
182: A_{\mu }\left( y\right) =-\frac{1}{2}F_{\mu \nu }y^{\nu },
183: \end{equation}
184: the above exponential can be rewritten as
185: \begin{equation}
186: \int d^{4}yA^{\mu }J_{\mu }=\frac{1}{2}F_{ij}\int
187: d^{4}yy_{j}J_{i}=\int dt\vec{\mu}\cdot \vec{B}
188: \end{equation}
189: where $\vec{\mu}$ is the magnetic-moment operator
190: \begin{equation}
191: \vec{\mu}=\frac{1}{2}\int d^{3}x\left( \vec{x}\times \vec{J}
192: \right) .
193: \end{equation}
194: Expanding (\ref{corr3}) to first order in $F$ and inserting nucleon
195: intermediate states, we thus obtain
196: \begin{equation}
197: \Pi _{\mu \nu }(p)F^{\mu \nu }=-\varepsilon _{ijk}\Pi _{ij}(p)B_{k}\sim {
198: \langle N}\left( p\right) {|}\vec{\mu}{|N}\left( p\right) {\rangle }\cdot
199: \vec{B} \label{polecontrib}
200: \end{equation}
201: in terms of the nucleon magnetic moments, as anticipated.
202:
203: In order to write down sum rules for ${\langle N}\left( p\right) {|}\vec{\mu}
204: {|N}\left( p\right) {\rangle }$, we need a QCD description of the correlator
205: (\ref{corr1}) at momenta $s=-p^{2}\simeq 1{\rm GeV}^{2}$, i.e. at distances
206: $x\sim 0.2\,{\rm fm}$. Such a description is provided by the nonperturbative
207: IOPE which factorizes (\ref{corr1}) into contributions from soft ($
208: k_{i}<\nu $) and hard ($k_{i}\geq \nu $) field modes (with momenta $k_i$),
209: where $\nu \sim 0.5\,{\rm GeV}$ is the operator renomalization scale. The
210: IOPE is generated by splitting each diagram contributing to (\ref{lresp}) in
211: all possible ways into a hard and a soft subgraph. The hard subgraphs, with
212: the integration range of each internal momentum restricted\footnote{
213: In practice, this restriction is often unnecessary.} to be larger than $\nu$
214: , give rise to the Wilson coefficients and receive, beyond the standard
215: perturbative contributions \cite{IS}, direct instanton
216: contributions which we will evaluate below. The soft subgraphs yield
217: hadron-channel independent condensates, i.e. vacuum expectation values of
218: local, composite QCD operators renormalized at $\nu$.
219:
220: The new feature brought in by the magnetic background field is that the
221: ``magnetized'' vacuum state $\left| 0\right\rangle _{F}$ singles out a
222: preferred direction (that of the field strength $\vec{B}$) and therefore
223: ceases to be a Lorentz scalar. Hence, several Lorentz-covariant operators
224: of the OPE (considered here up to dimension eight) aquire finite expectation
225: values, the $F$-induced condensates
226: \begin{eqnarray}
227: \langle 0|\bar{q}\sigma _{\mu \nu }q|0\rangle _{F} &=&\sqrt{4\pi \alpha }
228: \chi F_{\mu \nu }\langle 0|\bar{q}q|0\rangle , \label{magncond1} \\
229: g\langle 0|\bar{q}G_{\mu \nu }q|0\rangle _{F} &=&\sqrt{4\pi \alpha }\kappa
230: F_{\mu \nu }\langle 0|\bar{q}q|0\rangle , \\
231: g\langle 0|\bar{q}\gamma _{5}\tilde{G}_{\mu \nu }q|0\rangle _{F} &=&
232: \frac{i}{2}\sqrt{4\pi \alpha }\xi F_{\mu \nu }\langle 0|\bar{q}q|0\rangle .
233: \end{eqnarray}
234: ($G_{\mu \nu }=\frac{1}{2}\lambda _{a}G_{\mu \nu }^{a}$, $\tilde{G}_{\mu \nu
235: }=\frac{1}{2}\varepsilon _{\mu \nu \rho \sigma }G_{\rho \sigma }$ with
236: $\varepsilon _{0123}=-1$.) The parameters $\chi ,\kappa $, and $\xi$
237: quantify the vacuum response to weak electromagnetic fields and thus play
238: the role of generalized susceptibilities. The magnetic susceptibility of the
239: quark condensate, $\chi $, for example, originates from the induced spin
240: alignment of quark-antiquark pairs in the vacuum. The corresponding vacuum
241: expectation value is the lowest-dimensional induced condensate and
242: will therefore play a leading role in the magnetic moment sum rules.
243:
244: Altogether, the IOPE of the invariant amplitudes $\Pi _{i}(p)$ in Eq. (\ref
245: {lresp}) can thus be written as
246: \begin{equation}
247: \Pi _{i}(p^{2})=\sum_{n}C_{i,n}\left( p^{2};\nu \right) \langle 0|
248: {\cal O}_{n}\left[ \nu \right] |0\rangle _{F} \label{iope}
249: \end{equation}
250: where the ${\cal O}_{n}$ are local, composite QCD operators with the
251: appropriate quantum numbers (the multi-index $n$ can contain Lorentz
252: indices) and dimension $d\leq 8$\footnote{
253: Contributions from higher-dimensional operators are strongly suppressed, see
254: below.}. The $C_{n}$ are the Wilson coefficients. It might be worth
255: reiterating that the instanton contributions to (\ref{iope}) factorize into
256: those from soft modes (contained in the condensates $\langle 0|
257: {\cal O}_{n}|0\rangle $) and those due to hard modes which add nonperturbative
258: structure to the $C_{n}$. The latter can be of substantial size since
259: $\bar{\rho}<\nu ^{-1}\simeq 0.4\,{\rm fm}$ and have to be calculated explicitly.
260: (Nevertheless, they were neglected in the usual, perturbative treatment.)
261:
262: The calculation of the IOPE coefficients is described in Ref. \cite
263: {for00,aw99} to which we refer for details. Similar to the
264: phenomenologically known values of the lowest-dimensional condensates,
265: the bulk properties of the instanton size distribution are generated by
266: long-distance vacuum dynamics and thus taken as input. We will use the
267: standard ``instanton liquid'' values \cite{sch98} $\bar{\rho}\simeq
268: \frac{1}{3}\,{\rm fm}$ for the average instanton size and $\bar{R}\simeq 1{\rm fm}$
269: for the average separation between neighboring (anti)instantons. Since the
270: instanton size distribution $n(\rho )$ is rather sharply peaked around
271: $\bar{\rho}$, we can thus approximately set $n(\rho )=\bar{n}\delta (\rho
272: -\bar{\rho})$ with $\bar{n}\simeq $ 0.5 fm$^{-4}$. Moreover,
273: $\bar{\rho}^{-1}\gg \Lambda _{QCD} $ implies that the hard instanton
274: contributions can be calculated semiclassically. Since, finally,
275: multi-instanton correlations are negligible at the relevant distances $x\sim
276: 0.2\,{\rm fm}\ll \bar{R}$, this amounts to evaluating the hard subgraphs in
277: the background of an (anti-) instanton field and subsequently averaging over
278: the instanton-parameter distribution. The main contributions typically arise
279: from the quark zero modes in the instanton field (which aquire an effective
280: mass \cite{shi80} $\bar{m}(\rho )=-\frac{2}{3}\pi^{2}\rho^{2}\langle
281: \bar{q}q\rangle $ due to the interactions with ambient vacuum fields)
282: and can be treated exactly, while
283: the continuum modes are approximated by plane waves.
284:
285: More specifically, the leading instanton contributions to the correlator
286: (\ref{corr1}) can be traced to the graph in which two of the quarks propagate
287: in zero modes while the third interacts with the background field through the
288: magnetized quark condensate \footnote{Background-field induced transitions
289: between zero and continuum modes (which can be dominant in other correlators
290: \cite{for95}) are absent here.}. This contribution is purely nonperturbative
291: and difficult to account for, e.g., in quark-based hadron models \cite{for97}.
292: After averaging over the instanton size distribution, it reads
293: \begin{eqnarray}
294: {\langle 0|T\eta _{p}(x)\bar{\eta}_{p}(0)|0\rangle }_{F,inst} &=&\Pi
295: _{0}^{inst}(x)-\frac{2^{3}e_{u}}{3\pi ^{4}}\frac{\bar{\rho}^{4}}{\bar{m}^{2}}
296: \times \nonumber \\
297: &&\langle \bar{q}\sigma _{\mu \nu }q\rangle _{F}\,\sigma _{\mu \nu }\int
298: d^{4}x_{0}\frac{1}{(r^{2}+\bar{\rho}^{2})^{3}(x_{0}^{2}+\bar{\rho}^{2})^{3}}
299: \label{instinx}
300: \end{eqnarray}
301: for the proton. $\Pi_{0}^{inst}$ contains the instanton contributions for
302: $F=0$, calculated in \cite{FB}, and $r=x-x_{0}$, where $x_{0}$ specifies the
303: center of the instanton. The corresponding neutron correlator is
304: obtained by replacing $e_{u}$ with $e_{d}$. Above, we have used the
305: self-consistency condition \cite{CDG}
306: \begin{equation}
307: \langle \bar{q}q\rangle =-2\int
308: d\rho \,\frac{n(\rho )}{\bar{m}(\rho )}=-2\,
309: \frac{\bar{n}}{\bar{m}(\bar{\rho})}
310: \end{equation}
311: to eliminate the $\bar{n}$ dependence
312: from (\ref{instinx}). For later use in the sum rules, we now calculate the
313: Fourier and Borel transform\footnote{ For the Borel transform we follow the
314: convention of Ref. \cite{I}.} of (\ref {instinx}), from which we obtain
315: \begin{equation} \widehat{\Pi }_{3}(M^{2})=\frac{e_{u}}{128\pi ^{4}}a\chi
316: \bar{\rho}^{2}M^{6}I(z^{2})
317: \label{incontr2}
318: \end{equation}
319: where $M$ denotes the Borel mass parameter and $z= M\bar{\rho}$, $%
320: a= -(2\pi )^{2}\langle \bar{q}q\rangle $. The zero-mode loop gives rise
321: to the integral
322: \begin{equation}
323: I(z^{2})=\int_{0}^{1}\frac{d\alpha }{\alpha ^{2}(1-\alpha )^{2}}e^{-
324: \frac{z^{2}}{4\alpha (1-\alpha )}}=4e^{-\frac{z^{2}}{2}}\left[ K_{0}
325: \left( \frac{z^{2}}{2}\right) +K_{1}\left( \frac{z^{2}}{2}\right) \right]
326: \label{instint}
327: \end{equation}
328: and exhibits the unique exponential Borel-mass dependence which is
329: characteristic for instanton contributions \cite{FB}.
330:
331: A first important lesson of Eq. (\ref{incontr2}) is that, to leading order,
332: direct instantons contribute almost exclusively\footnote{
333: The small direct--instanton contributions to $\Pi _{2}$ have no appreciable
334: impact on the sum rules and are neglected.} to the chirally-odd amplitude
335: $\Pi_{3}$. This is crucial because in the analysis of Ref. \cite{IS}, were
336: this contribution was not accounted for, exactly the $\Pi_{3}$ sum-rule
337: failed to show a fiducial stability region \cite{ISP}. The previous neglect
338: of the instanton contributions thus offers a potential explanation for
339: this instability. An additional hint in this direction comes from two other
340: chirally-odd nucleon sum rules \cite{FB,for97} where stabilization due
341: to instantons was indeed found to take place\footnote{ A previously
342: contemplated connection between the instability of the original $\Pi_{3}$
343: sum rule and the perturbative infrared singularities of the ``pragmatic'' OPE
344: seems, in view of newer results \cite{WPC}, unlikely \cite{aw99}.}.
345:
346: We are now ready to proceed to the quantitative analysis of the IOPE sum
347: rule for $\Pi_{3}$. The latter results from equating the Borel transform of
348: the standard OPE of Ref. \cite{IS} and the instanton contributions (\ref
349: {incontr2}) to the Borel-transformed double dispersion relation for the
350: correlator (\ref{corr1}), with a spectral function parametrized in terms of
351: the nucleon pole contribution (cf. Eq. (\ref{polecontrib})) and a continuum
352: based on local duality. Including the infrared-divergent term encountered in
353: Ref. \cite{IS}, duly truncated at the OPE renormalization point, the IOPE
354: sum rule (for the proton) reads
355: \begin{eqnarray}
356: &&aM^{2}\left\{ \left[ e_{u}-\frac{1}{6}e_{d}(1+4\kappa +2\xi )\right]
357: E_{1}(M)\right. \nonumber \\
358: &&+\frac{1}{6}e_{u}\frac{m_{0}^{2}}{M^{2}}\left[ \ln {\frac{M^{2}}{\nu^{2}}}
359: -\gamma _{EM}\right] L^{-\frac{4}{9}}+\frac{1}{6}e_{d}M^{2}\chi
360: E_{2}(M) L^{-\frac{16}{27}} \nonumber \\
361: &&\left. -\frac{1}{8}e_{u}\chi \rho _{c}^{2}M^{4}I(z^{2})L^{
362: -\frac{16}{27}}\right\} \nonumber \\
363: &=&\frac{1}{4}\tilde{\lambda}_{N}^{2}me^{-\frac{m^{2}}{M^{2}}}\left[
364: \frac{\mu _{p}}{M^{2}}-\frac{\mu _{p}^{a}}{2m^{2}}+A_{p}\right],
365: \label{sumrule}
366: \end{eqnarray}
367: where $m$ is the nucleon mass, $\mu _{p}$ ($\mu _{p}^{a}$) the proton's
368: (anomalous) magnetic moment, $W$ the continuum threshold, and $\lambda _{N}$
369: the coupling of the current (\ref{interpol}) to the nucleon state, $\langle
370: 0|\eta |N\rangle =\lambda _{N}u$. For the mixed quark condensate we use the
371: standard parametrization $\langle \bar{q}\sigma _{\mu \nu }G^{\mu \nu
372: }q\rangle =-m_{0}^{2}\langle \bar{q}q\rangle $ with $m_{0}^{2}=0.8\,
373: {\rm GeV}^{2}$, and $\gamma _{EM}\simeq 0.577$ is the Euler-Mascheroni
374: constant. The additional parameters $A_{p,n}$ determine the strength of
375: electromagnetically induced transitions between the nucleon and its excited
376: states. The sum rule for the neutron is obtained from (\ref{sumrule}) by
377: interchanging $e_{u}$ and $e_{d}$ and by replacing $\mu _{p},\mu
378: _{p}^{a}\rightarrow \mu _{n}$ and $A_{p}\rightarrow A_{n}$. We have also
379: defined $\tilde{\lambda}_{N}^{2}=32\pi ^{4}\lambda _{N}^{2}$, $L=\ln
380: (M/\Lambda )/\ln (\mu /\Lambda )$ ($\Lambda =0.1\,{\rm GeV}$), and
381: transferred, using the standard expressions
382: \begin{equation}
383: E_{n}(M)=1-e^{-\frac{W^{2}}{M^{2}}}\left[ 1+\sum_{1}^{n}\frac{1}{j}\left(
384: \frac{W^{2}}{M^{2}}\right) ^{j}\right] ,
385: \end{equation}
386: the continuum contributions to the IOPE-side of the sum rules. The
387: appropriate form of these contributions has recently been clarified in Ref.
388: \cite{iof95}.
389:
390: The quantitative analysis of the IOPE\ sum rule (\ref{sumrule}) is based on
391: the minimization of the relative deviation between its two sides. Since the
392: fiducial domain, i.e. the Borel-mass region in which the IOPE truncation is
393: justified while the nucleon pole still dominates over the continuum, is (as
394: usual) not large enough to determine all unknown parameters from a combined
395: fit, we follow the procedure of Ref. \cite{IS} and obtain the coupling
396: $\tilde{\lambda}^{2}=2.93\,{\rm GeV}^{6}$ and the continuum threshold
397: $W=1.66\,{\rm GeV}$ by fitting the instanton-improved nucleon mass sum rule
398: of Ref. \cite{FB} to the experimental nucleon mass. The values of the two
399: susceptibilities $\kappa =-0.34\pm 0.1$ and $\xi =-0.74\pm 0.2$ were
400: estimated in independent work by Kogan and Wyler \cite{KW}. This enables us
401: to fit both sides of the sum rules (\ref{sumrule}) by varying $\chi $
402: and $A_{p}$ (or $A_{n}$, respectively) while keeping the magnetic
403: moments fixed at their experimental values.
404:
405: The fits are performed in the fiducial Borel mass domain $0.8\,{\rm GeV}\leq
406: M\leq 1.15\,{\rm GeV,}$ where the highest-dimensional operators contribute
407: at most 10 \% to the IOPE and the continuum contribution does not
408: exceed 50 \%. This fiducial domain is larger, incidentally, than
409: that of the sum rules based on $\Pi_{1}$ and $\Pi_{2}$.
410:
411: \begin{figure}[tbp]
412: \psfig{figure=mountmusr.ps,height=2.7in,width=4.5in,bbllx=72bp,bblly=230bp,%
413: bburx=539pt,bbury=570pt}
414: \caption{The OPE (dashed line) and direct instanton (dotted line)
415: contributions to the new $\protect\sigma_{\protect\mu\protect\nu}$ sum rules
416: for the proton (positive range) and neutron. Their sum (dot-dashed line) is
417: compared to the RHS (solid line).}
418: \label{fig1}
419: \end{figure}
420:
421: Figure \ref{fig1} shows, both for the proton and the neutron sum rules, the
422: direct-instanton contributions, the remaining OPE including the continuum
423: contributions, their sum (which makes up the left-hand side of Eq. (\ref
424: {sumrule})), and the right-hand sides. The fit quality of both the proton
425: and neutron sum rules is quite impressive. Figure \ref{fig1} shows,
426: furthermore, that the direct-instanton contributions can reach
427: the magnitude of the remaining terms in the OPE, which explains why their
428: previous neglect had a detrimental impact on the sum-rule stability. In Fig.
429: \ref{fig2} the optimized sum rules are solved for $\mu_{N}$ and plotted as a
430: function of the Borel mass. The resulting functions $\mu_{p,n}(M)$ therefore
431: specify the value of the magnetic moment which is required to make both sides
432: of the sum rule (\ref{sumrule}) match exactly at a given value of $M$. As a
433: consequence of the high fit quality, $\mu (M)$ is practically
434: $M$-independent. The numerical results are $\chi \simeq -4.96\,{\rm GeV}^{-2}$,
435: $A_{p}\simeq 0.28{\rm GeV}^{2}$ for the proton and $\chi \simeq -4.73\,{\rm
436: GeV}^{-2}$, $A_{n}\simeq -0.27{\rm GeV}^{-2}$ for the neutron sum rule. The
437: values of the quark condensate susceptibility $\chi $ (at $\nu =0.5\,{\rm
438: GeV}$) lie within the range obtained from other estimates \cite{IS,WPC,BK}
439: and differ somewhat from the value $\chi \simeq -5.7\,{\rm GeV}^{-2}$ found
440: in the two- and three-pole models of Ref. \cite{BK}.
441:
442: In conclusion, we have recovered a third reliable sum rule for the nucleon
443: magnetic moments. In contrast to the other two, it receives previously
444: neglected direct-instanton contributions which arise from the interplay with
445: long-wavelength vacuum fields. Our new sum rule is built on the chirally-odd
446: amplitude $\Pi _{3}$ of the nucleon correlator in an electromagnetic
447: background field and found to be at least as stable as the other two
448: (although it had previously been regarded as flawed). The new sum rule adds
449: to the predictive power of the background-field sum rules and strengthens
450: their mutual consistency.
451:
452: Moreover, our results reinforce a systematic pattern which emerged from the
453: study of direct-instanton effects in the pion \cite{for95}, nucleon \cite
454: {FB,for97} and glueball \cite{for200} channels: those sum rules which worked
455: satisfactorily without instanton corrections receive little or no direct
456: instanton contributions while previously less reliable or completely
457: unstable sum rules are stabilized by large instanton contributions. This
458: pattern points not only towards the importance of direct instantons in
459: particular sum rules, but also supports the adequacy of their semiclassical
460: implementation into the OPE. Our results show that these conclusions
461: continue to hold in the presence of a ``magnetized'' vacuum.
462:
463: \begin{figure}[tbp]
464: \psfig{figure=mountmu2.ps,height=2.7in,width=4.5in,angle=-90}
465: \caption{The Borel mass dependence of the magnetic moments of the proton
466: (upper) and neutron calculated from the optimal fit of LHS and RHS.}
467: \label{fig2}
468: \end{figure}
469:
470: \section*{Acknowledgments}
471:
472: I would like to thank the organizers for this interesting and
473: stimulating meeting, and for choosing such a beautiful location. I would
474: also like to thank Mountaga Aw and Manoj Banerjee for their collaboration on
475: the work presented here, which was supported by the Deutsche
476: Forschungsgemeinschaft under habilitation grant Fo 156/2-1 and by the U.S.
477: Dept. of Energy under grant number DE-FG02-93ER-40762.
478:
479: \section*{References}
480:
481: \begin{thebibliography}{99}
482: \bibitem{bel75} A.A. Belavin, A.M. Polyakov, A.S. Schwartz and
483: Yu.S.Tyupkin, Phys. Lett. B {\bf 59}, 85 (1975); R. Jackiw and C. Rebbi,
484: Phys. Rev. Lett. {\bf 37}, 172 (1976); C.G. Callan Jr., R.F. Dashen
485: and D.J. Gross, Phys. Lett. B {\bf 63}, 334 (1976); G. 't Hooft, Phys.
486: Rev. Lett. {\bf 37}, 8 (1976).
487:
488: \bibitem{hor78} D. Horn and S. Yankielowicz, Phys. Lett. B {\bf 76}, 343
489: (1978); W.U. Blask, U. Bohn, M. Huber, B. Metsch, and H. Petry, Z. Phys. A
490: {\bf 337}, 327 (1990); A.E. Dorokhov, Y.A. Zubov, and N.I. Kochelev, Sov. J.
491: Part. Nucl. {\bf 23}, 522 (1992); A. Christov et al., Prog. Part. Nucl.
492: Phys. {\bf 37}, 91 (1996).
493:
494: \bibitem{sch98} T. Sch{\"{a}}fer and E. Shuryak, Rev. Mod. Phys.
495: {\bf 70}, 323 (1998);
496: D.I. Diakonov and V.Yu. Petrov, Nucl. Phys. {\bf B245}, 259 (1984);
497: Phys. Lett. B {\bf 147}, 351 (1984); Nucl. Phys. {\bf B272}, 457 (1986).
498:
499: \bibitem{for00} H. Forkel, {\it Instantons, OPE and Hadron Structure}, to
500: be published.
501:
502: \bibitem{lat} M.C. Chu and S. Huang, Phys. Rev. D {\bf 45}, 2446 (1992);
503: M.-C. Chu, J.M. Grandy, S. Huang and J. Negele, Phys. Rev. D {\bf 49}, 6039
504: (1993); P. van Baal, Nucl. Phys. Proc. Suppl. {\bf 63}, 126 (1998).
505:
506: \bibitem{aw99} M. Aw, M.K. Banerjee and H. Forkel, Phys. Lett. B {\bf 454,}
507: 147 (1999).
508:
509: \bibitem{FB} H. Forkel and M.K. Banerjee, Phys. Rev. Lett. {\bf 71}, 484
510: (1993).
511:
512: \bibitem{svz79} M.A. Shifman, A.I. Vainshtein and V.I. Zakharov, Nucl.
513: Phys. {\bf B147}, 385, 448 (1979).
514:
515: \bibitem{I} B.L. Ioffe, Nucl. Phys. {\bf B188}, 317 (1981), Nucl. Phys.
516: {\bf B191}, 591(1981); V.M. Belyaev and B.L. Ioffe, Sov. Phys. JETP
517: {\bf 56}, 493 (1982).
518:
519: \bibitem{bel93} V.M. Belyaev and I.I. Kogan, Int. J. Mod. Phys. A {\bf 8},
520: 153 (1993).
521:
522: \bibitem{IS} B.L. Ioffe and A.V. Smilga, Nucl. Phys. {\bf B232}, 109
523: (1984).
524:
525: \bibitem{BY} I.I. Balitsky, Phys. Lett. B {\bf 114}, 53 (1982); I.I.
526: Balitsky and A.V. Yung, Phys. Lett. B {\bf 129}, 328 (1983).
527:
528: \bibitem{shi80} M.A. Shifman, A.I. Vainshtein and V.I. Zakharov, Nucl.
529: Phys. {\bf B163}, 46 (1980).
530:
531: \bibitem{for95} H. Forkel and M. Nielsen, Phys. Lett. B {\bf 345}, 55
532: (1995).
533:
534: \bibitem{for97} H. Forkel and M. Nielsen, Phys. Rev. D {\bf 55}, 1471
535: (1997).
536:
537: \bibitem{CDG} C.G. Callen Jr., R. Dashen and D.J. Gross, Phys. Rev. D
538: {\bf 17}, 2717, 2763, (1978); D.G. Caldi, Phys. Rev. Lett. {\bf 39}, 121 (1977).
539:
540: \bibitem{ISP} B. Ioffe and A. Smilga, private communication.
541:
542: \bibitem{WPC} S.L. Wilson, J. Pasupathy and C.B. Chiu, Phys. Rev. D {\bf 36}
543: , 1451 (1987); C.B. Chiu, S.L. Wilson, J. Pasupathy and J.P. Singh, Phys.
544: Rev. D {\bf 36}, 1553 (1987).
545:
546: \bibitem{iof95} B. Ioffe, Phys. Atom. Nucl. {\bf 58}, 1408 (1995).
547:
548: \bibitem{KW} I.I. Kogan and D. Wyler, Phys. Lett. B {\bf 274}, 100 (1992).
549:
550: \bibitem{BK} V.M. Belyaev and Ya.I. Kogan, Yad. Fiz. {\bf 40}, 1035 (1984)
551: [Sov. J. Nucl. Phys. {\bf 40}, 659 (1984)]; Ya. Balitsky, A.V.
552: Kolesnichenko, and A.V. Yung, Yad. Fiz. {\bf 41}, 282 (1985) [Sov. J. Nucl.
553: Phys. {\bf 41}, 178 (1985)].
554:
555: \bibitem{for200} H. Forkel, Heidelberg preprint HD-TVP-00-1 (2000),
556: hep-ph/0005004.
557: \end{thebibliography}
558:
559: \end{document}
560: