Parallel beams from nuclear microprobe lens systems

A nuclear microprobe lens system is normally used to focus a diverging beam into a fine probe. It does this by making the astigmatism terms in the first-order transfer matrix: (x/θ) and (y/phi) zero. Thus the position of a ray vector in the image plane (x,y) in the linear approximation will not depend on the divergence of the ray vector in the object plane (θ,phi). This is accomplished at the expense of magnifying the object divergence by factors (θ/θ) and (phi/phi) in the xoz and yoz planes, respectively, leading to relatively steep convergence angles at the Gaussian image plane where the specimen is located. An alternative operation mode of the nuclear microprobe lens system is possible where the convergence angle of the beam after the lens system in the linear approximation is made independent of the divergence angle of the beam prior to the lens system. In this case the terms (θ/θ) and (phi/phi) in the first-order transfer matrix are zero. This mode of the lens system does not produce a focused beam, so is most suitable for crystals without lateral structure. As an example, we present measurements from the <0 0 0 1> α-Al2O3 axis, which show significant improvement in channeling angular widths for planar channeling mode even when compared to the unfocused mode