Within FP6 and FP7 SMI takes part in the development of the internal target system of PANDA as well as in the design of the vacuum system and the interaction zone of the PANDA detector. Our contribution lies in research and development of the (hydrogen) cluster-jet target with the Genova cluster-jet target at GSI. With several setups at SMI we carry out studies on the possibility to use NEG-coated beam pipes near a high-luminosity internal target. This is supplemented by calculations of the vacuum conditions in PANDA.

 

Genova cluster-jet target at GSI

The Genova/Fermilab cluster-jet target, which was set up at GSI, operates with a gas supply system with up to 20 bar inlet pressure for raising the density of the jet.

In several series of measurements between 2005 and 2007 several parameters of the cluster-jet target were measured. For example it was found out that the cluster-jet at the interaction point has a diameter of 6.15 mm with a low density “halo” of 0.1 mm around it. Later, such a “halo” has been also found at the cluster-jet target in Münster.

The maximum target thickness, which could be reached at GSI with this target was 1.4×10¹⁵ atoms/cm² (5×10¹⁴ atoms/cm² was the record at E835, Fermilab). Increasing the target density further was impossible, as we reached the limit of the cooling power of the nozzle head.

In order to overcome these limitations and to optimise the geometry of the nozzle-skimmer arrangement, it was decided to modify this cluster-jet target. In 2008 the concepts for this process were studied, all technical drawings were prepared and all new parts were ordered or manufactured.

 

Nozzle tests

PANDA has to operate at a luminosity of 10³² cm^–2s^–1. Therefore the target thickness of the cluster-jet target has to be high (~5×10¹⁵ protons/cm²). Optimisation of the nozzle shape and the position of the first skimmer have a strong influence on the density of the cluster beam. A measuring device, which measures the density distribution of the cluster-jet right after the nozzle, was constructed at SMI. It turned out in 2008 that the construction of a new cold head for this setup would put too much workload on the mechanical workshop at SMI. Therefore it was considered advantageous to wait and to mount this measuring system inside the Genova/GSI/SMI cluster-jet target, once it is producing a cluster-jet again.

 

PANDA interaction zone with NEG-coated beam pipes
To achieve the desired luminosity in PANDA, not only the target density has to be increased, but also the gas load in the interaction zone has to be minimised. Since it is impossible to install UHV pumps near the interaction zone, we are carrying out investigations on the feasibility of using NEG coated beam pipes at PANDA.

For this purpose we have set up a test system at SMI, which simulates the PANDA interaction zone. It consists of 6 NEG-coated pipes with a 1 µm thin film of Ti(30at%)Zr(30at%)V(40at%) as getter. For (re)acti­vation of the getter we have developed a heating system with backing pumps.

The final aim of several measurements is the value of the duration of the service live till reactivation becomes necessary under conditions similar to the ones in PANDA. For making supplementary measurements we set up a separate UHV-system with a single coated tube, which was NEG-coated at GSI and installed at SMI in 2007.

Unfortunately the coating of this pipe got irreversibly saturated due to a leak during reactivation. Therefore this tube could not be used anymore for further measurements.

After reactivation of the test system with the 6 NEG-coated pipes, there were indications in 2008 that the getter had suffered under the big hydrogen load during one test. A test of the functionality of the NEG-coating on each single pipe revealed that in 4 pipes the coating had gone brittle and in the 2 others the getter had only a very limited pumping speed. Since removing the coating and sputter coating the 6 pipes anew is no option, this system was shut down and by and by dismounted.

The vacuum group at GSI has delivered in 2008 a similar tube, like the one of the supplementary system, so that the remaining measurements can be carried out in the first half of 2009.

Calculations of the PANDA vacuum system

Besides the experimental efforts regarding the internal target and the vacuum system of PANDA it is necessary to also calculate pressure profiles. This helps tremendously in the design of the detector, as the dimensions of the beam pipes are not completely fixed yet.

Calculations for the optimisation of the geometry of the PANDA vacuum system were carried out in 2008, like the pumping system behind the Target Spectrometer, the target beam dump and questions on joining together beam tubes with different diameter. Due to the layout of the interaction cross (only 20 mm inner diameter) and the long distance to the next pumps (~3 m in upstream and downstream directions) the effective pumping speed at the interaction point is extremely low (30 l/s in upstream and 5 l/s in downstream directions). The resulting high density of the residual gas would drastically reduce the current of the antiproton beam and cause a lot of unwanted background events.

In connection with these calculations also the planning and design of the vacuum system has been taken up in 2008. In order to make meaningful simulation studies, CAD drawings of this system have been made by SMI. For the full scale Monte Carlo calculations, which are planned to be carried out in 2009, the 2D CAD drawings have been converted into a 3 dimensional model of the complete target/antiproton beam line.

This 3D-model was presented to the collaboration and integrated into the existing “official” CAD model of the PANDA detector. Changes to the model are constantly being made in close contact with technicians at GSI.

 

Outlook

In the first half of 2009 the Genova/Fermilab cluster-jet target at GSI will be modified according to the designs made in 2008. The cluster production at this Genova/GSI/SMI target will start again after this. Renewed tests to run the target in high-pressure mode will be carried out to produce a cluster-jet with the desired density for PANDA. Furthermore we plan to mount pipes to this target, which will simulate the vacuum system of the PANDA Target Spectrometer in its complete length. This has been seen as a necessary step in order to measure the vacuum situation while running the target, since the gas input of the target cannot be completely simulated by numerical calculations. Additionally a new control and read-out system will be set up by INFN Genova and SMI.

When the Genova/GSI/SMI cluster-jet target will be up again, we will carry out a dedicated beam-time for the measurement of the pressure profiles behind the nozzle with our Pitot tube device and also test alternative nozzle designs.