shows a beam spot pattern recorded on an imaging plate
when we used a capillary with an outlet diameter of 20 lm. Since
this image was taken far from the capillary outlet (65 cm downstream),
the pattern expresses the divergence angle distribution
of protons extracted from the capillary outlet. The figure shows
that the micro beam obviously consists of two different components.
The central intense spot (called ‘‘core’’ in this paper) is
attributed to protons passing through the capillary without collision;
the surrounding doughnut-shape pattern (called ‘‘halo’’) is
due to protons scattered by the capillary inner wall. Note that
the signal at the core was saturated because the beam intensity
of the core was much higher than that of the halo. The divergence
angle of the core was around 1 mrad in this case, but it can change
depending on the incident beam optics. On the other hand, the
averaged divergence angle of the halo component was 7–8 mrad.
The inner edge, i.e. the minimum divergence angle of the halo component
is determined by the taper angle of the capillary inner wall
near the outlet. Since this halo component contributes X-ray production
in the target as well as the core component, it affects the
spatial resolution of micro-PIXE measurement, which will be discussed
later based on results from Monte-Carlo simulations