Photon-Gluon-Fusion zu D + antiD -Mesonen am COMPASS-Experiment

One of the main goals of the COMPASS experiment at CERN is the
measurement of the gluon contribution DeltaG to the nucleon spin. A
longitudinal polarised muon beam with 160 GeV energy is used. A
6LiD target can be polarised parallel and antiparallel with respect
to the muon spin. In the photon gluon fusion (pgf) a virtual photon
emitted from the muon interacts with a gluon from a target nucleon
via the creation of a quark-antiquark pair. A pgf event can be
identified through the detection of the creation of heavy quarks,
e.g. through the reconstruction of charmed D mesons. The cross
section of the pgf with the creation of D mesons depends on the
polarisation of the photon and the gluon. DeltaG/G can be extracted
from the counting rate asymmetry of the D mesons created with the
different polarisations of the target. In this work D mesons were
reconstructed in different decay channels. The number of
reconstructed D mesons and the signal to background ratio in the
signal region were used to estimate the statistical error of
DeltaG/G. Different kinematic cuts which suppress the background
were optimized with respect to the error of DeltaG/G. The main
contribution to the measurement of DeltaG/G comes from the golden
D* channel (D*+ -> D0 pi+ -> (K- pi+)pi+ and respectively D*-
->D0bar pi- -> (K+ pi-)pi-). The golden D0 channel (D0 ->
K- pi+ and respectively D0bar -> K+ pi-) has a high background
with contributions of some very broad K* resonances. It is
difficult to determine the strengths of these resonances in the Kpi
mass spectrum. The resulting systematic error still has to be
investigated. In total the statistical error of DeltaG/G amounts to
1.45 (using 2002 and 2003 data). The reconstruction probability of
charged tracks as a function of momentum and transverse momentum as
well as the reconstruction efficiency of K0s mesons and the
identification efficiency of charged kaons was determined with
Monte-Carlo simulations. These efficiencies were checked in the
data comparing 2 decay channels of the charged D* resonance and
reconstructing pairs of D mesons. Using these efficiencies the
contribution of other decay channels to the DeltaG/G measurement
could be estimated. Three channels are proposed. Their contribution
corresponds each to 7 - 28 % more data in the golden D* channel.