MODULE XCLSG2P CONTAINS * * ..File: xclsg2p.f FL_S * * * ..Calculation of the 2-loop x-space MS(bar) coefficient functions * for F2 via compact parametrizations involving only logarithms. * Singlet, mu_r = mu_f = Q. Expansion parameter: alpha_s/(4 pi). * * ..The relative accuracy of the coefficient functions, as well as of * the convolution results, amounts to a few thousandth. * * ..Reference: W.L. van Neerven and A. Vogt, * hep-ph/0006154 = Nucl. Phys. B588 (2000) 345 * ..The user should also cite the original calculations, * J. Sanchez Guillen et al, Nucl. Phys. B353 (1991) 337 and * E.B. Zijlstra and W.L. van Neerven, Phys. Lett. B273 (1991) 476 * * * ===================================================================== * * * ..This is the pure singlet piece, denoted by C2S in WvN's program. * Six numerical coefficients (all but the one of 1/y, which is * exact up to truncation) are fitted to his results, using x values * between 10^-6 and 1-10^-6. * FUNCTION CLS2A (Y, NF) IMPLICIT REAL*8 (A-Z) INTEGER NF * DL = LOG (Y) DL1 = LOG (1.-Y) * CLS2A = NF * ( (15.94 - 5.212 * Y) * (1.-Y)**2 * DL1 1 + (0.421 + 1.520 * Y) * DL**2 + 28.09 * (1.-Y) * DL 2 - (2.370/Y - 19.27) * (1.-Y)**3 ) * RETURN END FUNCTION * * --------------------------------------------------------------------- * * * ..This is the gluon contribution, denoted by C2G2 in WvN's program. * Six numerical coefficients are fitted as above, the one of 1/y is * exact up to truncation. * FUNCTION CLG2A (Y, NF) IMPLICIT REAL*8 (A-Z) INTEGER NF * DL = LOG (Y) DL1 = LOG (1.-Y) * CLG2A = NF * ( (94.74 - 49.20 * Y) * (1.-Y) * DL1**2 1 + 864.8 * (1.-Y) * DL1 + 1161.* Y * DL * DL1 2 + 60.06 * Y * DL**2 + 39.66 * (1.-Y) * DL 3 - 5.333 * (1./Y - 1.) ) * RETURN END FUNCTION * * =================================================================av== END MODULE XCLSG2P