NA48 measures direct CP violation

30 August 1999


With a first result announced from a major experiment at CERN, 1999 looks like being a vintage year for the measurement of CP violation ­the effect that enables nature to differentiate between matter and antimatter.

Earlier this year, the KTeV experiment at Fermilab (CERN Courier April) confirmed that CP is violated “directly” in the way quarks decay and transform into each other. The new result from the NA48 experiment at CERN underlines this direct CP violation and provides a valuable new benchmark.

CP violation, which is vital to our understanding of particle interactions and of the evolution of the universe in the wake of the Big Bang, is difficult to understand, hard to study and awkward to measure.

In 1956, physicists were shocked to discover that the weak force can differentiate between right and left. To reimpose order on their theories, physicists introduced charge/parity (CP) symmetry, in which physics should remain the same if a left-handed particle changes into a right-handed antiparticle, and so on.

For CP, the neutral kaon plays a special role. For CP purposes, this particle has to come in two forms: long-lived, decaying into three pions; and short-lived, decaying into two pions. In 1964, physicists received another shock when Christenson, Cronin, Fitch and Turlay discovered at Brookhaven that long-lived kaons can also decay into two pions. CP symmetry is not exact.

There are two ways in which this can happen. In the first, the long-lived kaon is a mixture of quantum states, mainly CP-odd with just a small amount of CP-even. In addition, CP could also be violated in the actual quark reactions underlying the particle transformations. In the decay of the neutral kaon, a strange quark slips off the map, producing pions composed of only up and down quarks. CP violation via this route is called “direct”.

To establish whether such direct CP violation happens, physicists have to compare two ratios: that of long-lived kaons decaying into two neutral pions with those going into two charged pions and the equivalent ratio for short-lived kaons. If these two ratios do not tally exactly, direct CP violation happens. Quark effects contribute to CP violation.

For several years the results from two major experiments ­ NA31 at CERN and E731 at Fermilab ­ could not be reconciled. The former gave the difference of the ratio of ratios from unity (divided by a conventional numerical factor) of 2.3 ± 0.65 x 10-3. The latter gave a much smaller figure, compatible with zero.

Does direct CP violation happen or not? To resolve the dilemma, new studies were begun. The early CP violation experiment at CERN had made separate runs with long-lived and short-lived kaons, leaving the door open to possible changes creeping in from one run to the next. Both recent experiments overcame this by using simultaneous beams of long- and short-lived kaons and taking data on charged and neutral pion production by both kaon beams at the same time. Earlier this year, KTeV at Fermilab reported a value of 2.8 ± 0.41 x 10-3 in tune with the earlier CERN measurement, but a bit on the high side, which surprised some people. The new result from NA48 is 1.85 ± 0.73 x 10-3.

The NA48 flagship experiment providing the new measurement is a major research investment. Installed in CERN’s highest-intensity proton beamline it uses a large and sophisticated detector. Some of the CERN protons go to make long-lived kaons, and the remaining particles are bent by a crystal and used to make a parallel beam of short-lived kaons. In this way the protons giving birth to short-lived kaons are “tagged” and the two varieties of kaons are clearly differentiated, even though they eventually decay in the same way.

Decays into charged pions are measured by a magnetic spectrometer, while the pions from the neutral pions are pinpointed by the pride of the experiment ­ a special liquid krypton calorimeter with better than 1% energy resolution and subnanosecond time resolution.

Handling NA48 data requires a major effort in dataprocessing power, with data from all detector modules being pipelined to CERN’s main computer centre via a dedicated fast link. The NA48 result comes from data taken in 1997 and will be about 10% of the total amount of data that the experiment expects to accumulate in three years of running. However, even this initial sample amounts to almost 5 million kaon decays into pairs of pions, more than1 million of them being decays of long-lived kaons into pairs of pions, the direct descendant of the handful of events seen in 1964 that established CP violation as a physics phenomenon.

CP violation has still only been seen in the decays of neutral kaons. However new experiments are setting out to measure CP violation via new routes.

bright-rec iop pub iop-science physcis connect