B. Panzer

First Evidence for Direct CP Violation

Since CP violation was first observed in the decay of the long-lived neutral kaon into two pions [1], it remains one of the enigmas in particle physics. Whilst CP violation is manifest in neutral kaon decays, the search for CP-violating effects has elsewhere been unsuccessful. In the phenomenology of CP violation in the neutral kaon system [2], the short- and long-lived mass eigenstates are usually defined in terms of the CP eigenstates K1 (CP = + 1) and K2 (CP = - 1) as Ks ≈ K1 + ?K2 and KL ≈ K2 + ?K1. The parameter ? describes CP violation induced by kaon state-mixing. Direct CP violation may also occur in the decay of K2 into two pions with a relative amplitude ?′, which is non-zero in the case of a phase difference between the amplitudes A0 and A2 for the decay into isospin O and 2 states of two pions. Before the present measurement, all experimental results were compatible with ? = 2.27 × 10-3 exp (i 43.70) and with the Superweak Model [3], in which state-mixing is the only source of CP violation and ?′ = 0. In the Standard Model with six weakly interacting quarks [4], direct CP violation as well as state-mixing is introduced by transitions via heavy-quark intermediate states. Based on this, a small, but non-zero, value of ?′ is predicted [5]. To a good approximation, ?′ /? is related to the double ratio R of the relative decay rates of the long- and short-lived neutral kaons into two neutral and two charged pions as Re (?′/?) = 1/6 × (1 - R).

A measurement of the phases of the CP-violating amplitudes in K0→2π decays and a test of CPT invariance

Abstract The phases of the CP-violating amplitudes in K0→π+π− and K0→2π0 decays, φ+−=46.9°±2.2° and φ00=47.1°±2.8°, have been measured in the same experiment, and a direct comparison gives the phase difference φ00−φ+−=0.2°±2.9°. This result leads to an upper limit on possible CPT violation in the K0 mass matrix, of |(m K 0 −m K 0 )/m K 0 | −18 at the 95% confidence level and is the most stringent test of the equality of particle and antiparticle masses.

The Beam and Detector for a High Precision Measurement of {CP} Violation in Neutral Kaon Decays

The K0 beam and detector used for a high-precision measurement of the CP-violation parameter ϵ′ at the CERN Super Proton Synchrotron (SPS) are described. The beam provides KL and KS alternately through a common decay region. The detection of the decays is based on wire chambers and calorimeters without employing a magnet. The trigger and readout system achieve a high selectively for the suppressed, CP-violating, two-pion decays of the KL by incorporation of hard-wired processors. The readout is based on Fastbus for maximum data rates.

Observation of the decay KL→π0γγ

Abstract The decay mode K L → π 0 γγ has been observed with a signal of 21 events and an expected background of 1.5±0.9 events. A branching ratio for decays with invariant γγ masses above 280 MeV of (2.1±0.6)×10 −6 is calculated. This result is compared with the values estimated from theoretical models and has implications for the CP conserving contribution to K L → π 0 e + e − decay.

Observation of the decay Ks → 2γ and measurement of the decay rates KL → 2γ and KS → 2γ

Abstract The decay rates of K L → 2γ and K S → 2γ have been measured at the CERN SPS. The results are Γ (K L →2 γ )/ χ (K L →2 π 0 )=0.632±0.004±0.008 and Γ (K S →2 γ )/ Γ (K L →2 γ )=2.3 ±1.0±0.4. This is the first observation of K S →2γ decays.

Observation of the decay K(L) ---> pi0 gamma gamma

Abstract The decay mode K L → π 0 γγ has been observed with a signal of 21 events and an expected background of 1.5±0.9 events. A branching ratio for decays with invariant γγ masses above 280 MeV of (2.1±0.6)×10 −6 is calculated. This result is compared with the values estimated from theoretical models and has implications for the CP conserving contribution to K L → π 0 e + e − decay.

Measurement of the rate of the decay KL→e+e−γ and observation of a form factor in this decay

Abstract A large sample of Dalitz decays, K L → e + e − γ , has been observed. The branching ratio is Γ(K L →e + e − γ) Γ(K L → all )=(9.2±0.5±0.5)×10 −6 in good agreement with theoretical predictions. We observe an enhancement at high masses in the distribution of the invariant electron-positron pair mass, compared to the distribution expected from QED. This excess is interpreted as being due to virtual mesons contributing to the photon propagator.

A large-area transition radiation detector

Abstract The construction and the operation of a large-area transition radiation detector (TRD) for the NA31 experiment at CERN are described. The TRD incorporates several novel features for stabilizing the detector response. The density of the gas mixture (xenon+helium+methane) in the detection chambers is matched to the carbon dioxide gas in the surrounding radiators by tuning the helium concentration to avoid a hydrostatic pressure difference, which would deform the chamber walls. The chamber pressure is continuously regulated by computer control to maintain it to within 1 μbar of the radiator pressure. The gas gain of each of the four chambers is regulated to better than 0.2% by changing the high voltage under computer control, using the pulse-height spectra of 16 109 Ca sources mounted on the chambers. The results of performance studies are described. The detector has a pion efficiency of 98.7% with an electron rejection of a factor of 10.

A measurement of the decay KL→π0γγ

Abstract The full data set of the experiment NA31 at CERN has been used to analyse the decay mode K L → π 0 γγ . A signal of 63 events has been observed with an estimated background of 6.0±1.7 events, corresponding to a branching ratio of (1.7±0.3)×10 −6 consistent with our previous result based on partial statistics. The invariant mass spectrum of the two photons is found to be consistent with chiral perturbation theory, and the decay is dominated by the J = 0 two photon state.

Search for the decay KL→π0e+e−

A search for the decay KL→π0e+e− has been performed at the CERN SPS. One candidate event has been found, compatible with an expected background of 1.5 events. The corresponding upper limit for the branching ration is Γ(KL→π0e+e−)Γ(KL → all)<4×10−8 with 90% confidence.