Observation of Interference between Čerenkov and Synchrotron Radiation

Associated with the experimental discoveries of Čerenkov radiation and synchrotron radiation, separate theoretical analyses were based on the different physical circumstances of the two phenomena. In recent years, Erber and collaborators have noted in a series of papers5−13 that when a relativistic charged particle traverses a dielectric medium subject to an applied magnetic field the resulting radiation is best considered as a single process which can be interpreted as Čerenkov radiation or synchrotron radiation only in limiting cases. Striking interference effects are predicted, which have been observed for the first time in the experiment described in this Letter. Approximate conditions for interference can be deduced from the usual understanding of Čerenkov and synchrotron radiation. In both cases the energy for the radiation is derived from the charged particle, and not the surrounding medium. The small momentum transfers to the field and the medium are essentially unobservable. An interference effect might then be expected when the frequency and angular distributions of Čerenkov and synchrotron radiation are similar. As synchrotron radiation is emitted with angular spread ∼ 1/γ, where E = γmc is the energy of the charged particle, interference can occur when the Čerenkov angle is also ∼ 1/γ, which occurs close to the Čerenkov threshold for large γ. Then if the Čerenkov angle is varied the angular distribution of Čerenkov radiation sweeps through that of synchrotron radiation and the total radiation rate can be observed to oscillate. Čerenkov radiation is unpolarized on average, while synchrotron radiation is largely polarized perpendicular to the applied magnetic field, so the interference effect will be more pronounced for the latter polarization. To observe this effect we have performed an experiment with 344and 378-MeV electrons at the Bates Linear Accelerator Laboratory. The detector, sketched in Fig. 1, consisted of a 5.1-m-long Čerenkov counter which was filled with helium gas at various pressures so as to vary the index of refraction and hence the Čerenkov angle. A weak magnetic field was applied transverse to the axis of the counter by a pair of coils whose aspect ratio was chosen to provide maximum field uniformity. The resulting spectrum of synchrotron radiation peaked in the infrared, with a useful tail into optical frequencies. The electron beam was deflected 7.5 cm horizontally by the magnetic field as it traversed the counter. To keep the electrons within the good-field region the downstream end of the counter was displaced by 3.75 cm from the field-off beam axis. The magnetic field was measured with a Rawson-Lush Model 784 digital gaussmeter to vary by no more than 1% within ±4 cm of the detector axis. The ends of the magnet coils were outside the active length of the counter to avoid the (interesting) complication of synchrotron radiation in a nonuniform magnetic field.