$B - \bar B$ Mixing in the HQET

Abstract

We present a high statistics, quenched lattice calculation of the B-parameters $B_{B_d}$ and $B_{B_s}$, computed at lowest order in the HQET. The results were obtained using a sample of 600 quenched gauge field configurations, generated by Monte Carlo simulation at $β=6.0$ on a $24^{3}\times 40$ lattice. For the light quarks the SW-Clover action was used; the propagator of the lattice HQET was also tree-level improved. Our best estimate of the renormalization scale independent B-parameter is $\hat{B}_{B_d} = 1.03 \pm 0.06 \pm 0.18$. $\hat{B}_{B_d}$ has been obtained by using ``boosted'' perturbation theory to calculate the renormalization constants which relate the matrix elements of the lattice operators to the corresponding amplitudes in the continuum. Due to the large statistics, the errors in the extraction of the matrix elements of the relevant bare operators are rather small. The main systematic error, corresponding to $\pm 0.18$ in the above result, comes from the uncertainty in the evaluation of the renormalization constants, for which the one-loop corrections are rather large. The non-perturbative evaluation of these constants will help to reduce the final error. We also obtain $\hat{B}_{B_s}/\hat{B}_{B_d} = 1.01 \pm 0.01$ and $f^2_{B_s}\hat{B}_{B_s}/f^2_{B_d}\hat{B}_{B_d} = 1.38 \pm 0.07$.