Introduction

A properly aligned, bent single crystal is capable of deflecting a high energy beam of positively charged particles through a relatively large angle. In this process channeled particles traveling near a plane or axial channeling direction follow the gradually bending crystal. Crystal septa have several limitations in acceptance. Only particles within the critical angle for channeling are accepted. The planer critical angle is five microradians at 800 GeV for the (110) plane in silicon and is inversely proportional to the square root of the momentum at relativistic energies. On the other hand particle beams in experimental areas at accelerators tend to have angular divergences on the orderof several hundred microradians. Thus the angular acceptance of a crystal in the direction of bend is typically 0.025. Likewise, the spatial acceptance in the direction of the bend is limited by the thickness of the crystal which is in turn set by the requirement that the crystal bend without breaking. Crystals up to several mm thick have been used. This should be contrasted to external beam spots which range from 3 to10 mm. Perpendicular to the bend the acceptance is good since the crystal can be quite wide and a substantial fraction of the particles captured in a plane will be bent. I explored some possible application of bent crystals in 1980 [Carrigan, Fermilab Pub 80/45 (1985)].

Anomalous losses at points of high curvature

In 1983  we discovered that anomalous losses of particles occurred halfway trhough the deflection for crystals bent with so-called three point jigs [Baker et al., Phys. Lett. 137B, 129 (1984)]. These losses were traced to distortions of the crystals at the bending point. They provided some of the first experimental information on bending dechanneling and led to the introduction of four point benders and other more sophisticated bending arrangements.