Speaker
Dr
Matteo Martini
(LNF -INFN)
Description
A precise measurement of the $K_S \rightarrow \gamma \gamma$ rate
is an important
test of Chiral Perturbation Theory predictions. The decay
amplitude can be evaluated
at the leading $p^4$ order providing an estimate of the $BR$ for
this decay of
$(2.1\pm0.1)\times10^{-6}$. The latest experimental determination
of $BR(K_S
\rightarrow \gamma \gamma)$ is a precise measurement from NA48,
$(2.78\pm0.07)\times 10^{-6}$, which differs from $\chi$PT $p^4$
prediction of about
30\%. This seems to indicate the presence of important
contributions from higher
order corrections.
KLOE analysis on 1.6 fb$^{-1}$ of data acquired during years
2001-2002 and
2004-2005 benefits from the tagging technique, which allows for
the first time
this decay to be identified with a pure $K_S$ ``beam'', without
the background from
$K_L \rightarrow \gamma \gamma$ decay, and with completely different
systematics respect to fixed target experiments. Event counting
is performed from a
fit to the bidimensional distribution of the two-photon invariant
mass versus the
angle between photon momenta in the $K_S$ rest frame. In this
plane the best
separation is achieved between the signal and the main source of
background, which is
represented by $K_S \rightarrow \pi^0 \pi^0$ events with two missing
photons. The result of this analysis is presented, which is
competitive with present
measurements.
$K_S \rightarrow e^+e^-$ decay is a $\Delta S = 1$ weak neutral
current process. The
Standard Model expectation for its $BR$ is $1.6\times10^{-15}$,
which has been
evaluated by Chiral Perturbation Theory with 10\% error. The best
experimental limit
on this decay, achieved by CPLEAR experiment, is $BR <
1.4\times10^{-7}$ at 90\% CL.
We performed a direct search of $K_S\rightarrow e^+e^-$ decay by
analysing 1.3
fb$^{-1}$ of data. The analysis exploits the excellent KLOE
drift chamber momentum
resolution to identify the signal through $e^+e^-$ invariant mass
reconstruction.
Further background rejection comes from calorimeter particle
identification, which is
based on time of flight, shower
longitudinal profile and $E/p$. The result of this search is
presented, which
improves on the previous experimental limit by a factor of 10.
Primary author
Collaboration KLOE
(INFN/LNF)
