<<

Nature Vol. 295 11 February 1982 457

Estimates of the energies of the Making muonium in vacuum indicate that 50 per cent have energies of less than 5 keV and 95 per cent from C.J. Batty and G. Marshall of less than 20 ke V. There are two measurements in MuoNIUM is the name given to the simple general the will capture and lose an particular which would benefit from atomic state consisting of a positively many times before it reaches copious quantities of muonium in vacuum. charged muon and an electron. For atomic thermal energies, and if the likelihood for One is a sensitive search for the conversion and chemical purposes it is aptly described the capture process is large compared with of muonium to antimuonium (µ-e+) which as a light isotope of with only that for electron loss as thermalization can be accommodated in some versions of about one-ninth the hydrogen . becomes complete, muonium will be the electroweak theory. The interaction of However the properties of its 'nucleus' formed. In most chemical and solid-state muonium with drastically reduces make muonium both interesting and useful applications the interaction of muonium the expected conversion rate, so that in physics and chemistry. Two recent with the target material is studied, but this current limits on the process are not papers describe a novel way of forming interaction can be a hindrance when particularly restrictive. The other is the muonium which will extend its use to studying certain other phenomena. It is in muonium, the energy measurements in vacuum. therefore of interest to develop an difference between the 2s ½ and 2p½ states The muon, identified in 1936 in cosmic alternative method of muonium pro­ of the n = 2 atomic level. Since both the ray studies, was one of the earliest of the duction so that measurements can be made muon and the electron are and thus elementary to be discovered. But on the in vacuum. have no structure, quantum elec­ despite an enormous amount of study and A group working at the LAMPF ac­ trodynamics can predict the Lamb shift a considerable knowledge of its properties, celerator facility in New Mexico has very accurately, in contrast to the case for its existence remains an enigma. Like the recently reported the observation of fast atomic hydrogen. The fast muonium electron and the , negative and ("'10 keV) muonium in vacuum (P.. emerging from the foil in the LAMPF positive are weakly interacting Bolton et al. Phys. Rev. Lett. 47, 1441; experiment should contain a 2s component point-like leptons and are not influenced 1981). The authors allowed an intense low­ of about one-eighth the total muonium by the felt by . energy (4 MeV) muon beam to slow down intensity. The muon differs from the electron in a polyethylene degrader before it passed This component was also shown to exist essentially only in its mass, which is 207 through a thin foil target in vacuum. It is in another recent experiment performed by times that of the electron. predicted, on the basis of a group from the TRIUMF facility It is relatively simple to produce intense, measurements, that a significant fraction in Vancouver, Canada(C.J. Oram et al. J. high-quality muon beams from the decay of the muons emerging with energies less Phys. B14, L789; 1981) using a similar of obtained at some than 20 keV will be in the form of muon beam and thin foil target. Muonium accelerators. So it has been possible to muonium. produced in the skin of the foil emerged to make a thorough study of the properties of In order to separate the small neutral pass through two successive sets of parallel the muon, and then with this knowledge, to muonium component from the charged plates, used to produce high electric fields use the muon as a tool for measurements in remainder of the beam, a 5 kG magnetic at right angles to the direction of motion of other fields. Thus the and magnetic sweeping field is used immediately after the the atoms. The electric fields will introduce moment of the muon, together with the target foil. The neutral particles travel 160 through the Stark effect a 2p component to fact that its weak decay (with a mean cm from the target through the magnetic any 2s state which may be present. Decay lifetime of 2.2 µs) into an electron plus two field to a beam stop. The can then proceed from the 2p to the 1 sstate is asymmetric or violating, expected from the decay of the muon are by emission of UV light with a lifetime of have been used to give information on the detected in an array of scintillation about 3 ns. The light can be detected by a magnetic fields inside a variety of counters and a NaI detector. If the e+ are wavelength shifter and photomultiplier materials. This technique of muon spin from the decay ofµ + then they should have system mounted close to the set of rotation, or µSR in analogy with NMR and a characteristic energy spectrum with an plates. A clock is started when a muon is ESR, is a subject of intense interest at endpoint energy of 53 MeV. incident on the thin foil and is stopped several laboratories throughout the world. This characteristic energy spectrum, when an event is recorded by the UV Another application involves the superimposed on a flat background of detection system. formation of muonic atoms, in which a events due to cosmic rays and dominated at If voltage is applied only to the second negative muon is captured in an atomic energies below 30 Me V by counts due to the set of plates the time spectrum should show orbital, effectively replacing one of the electron contamination in the incident three muon components: an exponential . Measurements of the beam, has been observed in the experiment decay due to muon-decay positrons emitted in the subsequent atomic cascade reported. The introduction of a few Torr of incident on the photomultiplier system; a can determine the distribution of charge in gas into the normally evacuated flat background due to uncorrelated the nucleus. (< 5 x 10-5 Torr) apparatus causes the events; and a foreground peak, about 150 The formation of muonium is fun­ muonium signal to disappear due to the ns wide, due to muonium in the 2s state damental to experiments with muons in ionizing collisions µ+ e- + He->-µ+ + e- + decaying in the region of high electric field. fields as diverse as quantum electro­ He; the µ+ are then swept away by the The foreground peak should largely dynamics, , solid-state and magnet, and the background spectrum can disappear when an electric field is applied surface physics, and chemistry. For most be measured. The results show that, for the to the first set of plates, as this would cause applications muonium can be formed by particular muon beam used, the pro­ the 2s state muonium atoms emerging from arresting a beam of energetic positive duction rate for muonium is about 3 lo-4 the foil to decay rapidly to the ls ground muons in a suitable solid, liquid, or gas per incident µ+ regardless of which of state before they reach the second set of target, in the same way that hydrogen several target foils is used. The conclusions plates and detector system. atoms can be formed by stopping proton are consistent with proton beam data and Further experiments are planned to beams. Near the end of its range, the with models of the neutralization process. improve the production of muonium in positive particle loses energy by ionization, vacuum. In this way, the goals of an and in the low-velocity regime (cor­ accurate determination of the Lamb shift responding to less than about 20 keV for C.J. Batty and G. Marshall are at the and a sensitive search for the conversion of muons) there is a high probability that it Rutherford Appleton Laboratory, Chilton, muonium to antimuonium should be will form a neutral muonium atom. In Didcot, Oxfordshire OX11 OQX. achieved. D

0028-0836/ 82/0li0457• 01$01 .00 © 1981 Macmillan Journals Ltd