NF Note 10

CERN Neutrino Factory Working Group Technical Note
R. Horne, C.D. Johnson
(from previously unpublished material)

Post-irradiation Examination of a High-Intensity Proton Beam Target

This Technical Note was distributed at CERN in manuscript form on 10/10/85. It contained several colour photographs.

There has recently been a renewed interest in high-intensity and high-power proton targetry for neutrino production and nuclear spallation sources. We have taken advantage of the newly available techniques of Web publishing and computer enhancement of photo images to re-issue the original document in an updated form.

At the time of this study the CERN Proton Synchrotron, PS, could deliver a beam onto the antiproton target of up to 5 bunches, spaced by 105 ns, per machine cycle (pulse) with the following properties:


Energy  26 GeV
Intensity (maximum)  2.5 x1012 p/bunch
Repetition rate (max.)   0.4 Hz
Bunch length (4sigmaz) 25 ns   
Beam spot size at target of (4sigmar) 1.6 mm 
Energy deposited in target (50 mm)  1.5 kJ per pulse (5 bunches)
Peak energy density in beam:          50 kJ mm-2 per pulse
Mean power absorbed in target at maximum repetition rate   600 W

The antiprotons (and other secondary particles - notably negative pions) produced in the target were collected (focused) by a lithium lens placed very close to the target as illustrated in Figure 1: Target + Li Lens. The high-density target was a 3 mm diameter x 50 mm length rod of rhenium (later iridium was used) pressed into a graphite cylinder, itself a press fit inside a finned titanium container that was bolted on to a finned aluminium heat sink with a black anodised finish (seen to the left of the target assembly in Figure 1). The whole assembly - target plus heat sink was air cooled. Some photographs of the target and target/lens assembly installed in the antiproton production zone are shown below.

Click on the photos to see enlarged images

a) A target assembly, b) a similar target mounted upstream of the lithium lens, c) the target/lens installed in the beamline with illumination for TV monitoring

The component pieces of a typical target before assembly are shown in Figure 2. In this version a copper heat sink was installed between the graphite and the titanium container. The rhenium core was made up from pieces 3 mm diameter and 10 mm long. The alumina disk at the downstream end of the rhenium core was introduced to improve the fatigue lifetime of the downstream exit window of the titanium container. Previously two steel containers had ruptured on axis at the exit window after 105 beam pulses. This was attributed to fatigue failure.

The target chosen for post-irradiation dissection had the simpler form shown in Figure 1. By removing the upstream plug and then cutting through the target snout immediately in front of the alumina disc, the target and its surrounding graphite cylinder could be pressed out from the titanium container The graphite cylinder was then sheared longitudinally with a special tool exposing the rhenium rods for visual inspection. The target had been irradiated for six days and, taking account of beam intensity fluctuations and duty cycle, the estimated total proton flux was between 1 and 1.5 1017 protons. Previous experience with tungsten and rhenium targets had led us to expect that the rods would be shattered by beam-induced thermal shock soon after the start of irradiation at full intensity. Our target was designed to contain the fragments within the graphite sheath. The time evolution of normalised yield of secondary particles (antiprotons, electrons, negative pions) measured seperately after startup, indicated that breakup reduced the effective target density by less than 10%. Earlier results from long-term irradiation of copper targets in graphite had indicated a more important loss of up to 30% in effective density due to fracture and void formation.

Our purpose in carrying out this post-irradiation target inspection was to study the fracture pattern and to confirm our calculation that most of the damage would occur in the downstream end of the rhenium rod where the heat load was greatest, see Figure 3.

In the following set of four pictures, Photos 1 and 2 show the target before dissection. The target snout was subsequently sliced through at the interface between the graphite and the alumina disc and a further slice of 10 mm length was cut off the end of the snout. Photo 3 is a view of this slice from the downstream end exposing the rhenium core, the graphite plug and the outer sleeve of titanium. Photo 4 shows the upstream face of the alumina disc - it exhibits radial cracks. Click for an enlarged view.

Photo 1                                 Photo2

Photo 3                                 Photo4

The graphite slug containing the 10 mm downstream end of the target was then pushed out from the titanium sleeve and split. The rhenium was in fragments and these are shown in Photo 5.

The remaining upstream 40 mm length of rhenium was still inside the main body of the target (Photo 6). It was pressed out and the surrounding graphite cylinder was split over its entire length. The results are shown in Photos 7 and 8. The greatly enlarged views (click on the photo) show that the rhenium in the upstream 30 mm has broken into many short lengths and that the region from 30 mm to 40 mm had disintegrated in a similar way to the 10 mm end section (Photo 5). This qualitatively confirms the prediction (Figure 3) that the maximum thermal shock is expected in the downstream 40% of the target.

Photo 5

Photo 6

Photo 7

Photo 8

Summary   These photographs confirmed our understanding that within the antiproton targets the high-density target material (rhenium or iridium) was shattered by beam-induced thermal shock, but that the fragments could be contained to provide a target with an effective density better than 90% of the solid material. A subsequent improvement in target design - intended for higher intensity proton beams - was water cooled and in some assemblies the 3mm diameter iridium core material was plated to a diameter of 6 mm with nickel to improve the containment of the fragments. A target of this design is shown below. Its lifetime was in excess of 107 beam pulses with no significant depletion of effective density.

Schematic sectional drawing and component parts of the Antiproton Collector (ACOL) target.

This technical note was originally written in manuscript form, dated 10/10/85.