TY - JOUR

T1 - Wilson's operator expansion

T2 - Can it fail?

AU - Novikov, V. A.

AU - Shifman, M. A.

AU - Vainshtein, A. I.

AU - Zakharov, V. I.

PY - 1985/1/28

Y1 - 1985/1/28

N2 - Within the framework of some simple models we discuss the status of the operator product expansion (OPE) in the presence of nonperturbative effects. We consider, in particular, the 4d Higgs model, 2d sigma model and the Schwinger model. The general formulation of OPE is presented and it is demonstrated that there exists a consistent procedure allowing one to define unambiguously both coefficient functions and matrix elements of composite operators. One of the key elements of the procedure is the introduction of an auxiliary parameter, the normalization point μ. For the simplest T-products discussed in the literature earlier we construct the corresponding OPE explicitly. Then we check its validity by comparing the results for the two-point functions with independent direct calculations of the same correlators. Although the general procedure is standard and does not vary from one theory to another, numerically the relative role of perturbative and nonperturbative contributions in vacuum condensates is different in different theories. The two extremes considered are the λφ{symbol}4 theory with no spontaneous breaking of the symmetry and the O(N) sigma model in the limit N → ∞. In the former case there is only perturbative contribution to (φ{symbol}2), while in the latter case the perturbative pieces are suppressed by 1/N factors. Numerically QCD is much closer to the O(N) sigma model in the large-N limit. Comments on specific features of QCD are presented.

AB - Within the framework of some simple models we discuss the status of the operator product expansion (OPE) in the presence of nonperturbative effects. We consider, in particular, the 4d Higgs model, 2d sigma model and the Schwinger model. The general formulation of OPE is presented and it is demonstrated that there exists a consistent procedure allowing one to define unambiguously both coefficient functions and matrix elements of composite operators. One of the key elements of the procedure is the introduction of an auxiliary parameter, the normalization point μ. For the simplest T-products discussed in the literature earlier we construct the corresponding OPE explicitly. Then we check its validity by comparing the results for the two-point functions with independent direct calculations of the same correlators. Although the general procedure is standard and does not vary from one theory to another, numerically the relative role of perturbative and nonperturbative contributions in vacuum condensates is different in different theories. The two extremes considered are the λφ{symbol}4 theory with no spontaneous breaking of the symmetry and the O(N) sigma model in the limit N → ∞. In the former case there is only perturbative contribution to (φ{symbol}2), while in the latter case the perturbative pieces are suppressed by 1/N factors. Numerically QCD is much closer to the O(N) sigma model in the large-N limit. Comments on specific features of QCD are presented.

UR - http://www.scopus.com/inward/record.url?scp=25044460747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=25044460747&partnerID=8YFLogxK

U2 - 10.1016/0550-3213(85)90087-2

DO - 10.1016/0550-3213(85)90087-2

M3 - Article

AN - SCOPUS:25044460747

SN - 0550-3213

VL - 249

SP - 445

EP - 471

JO - Nuclear Physics, Section B

JF - Nuclear Physics, Section B

IS - 3

ER -