Superconducting Linac Booster Project

 

The desirability of having higher energy heavy ion beams for experiments in the areas of nuclear and atomic physics is by now well established. A few heavy ion boosters are already in operation at some laboratories. Most of these boosters are based on the technique of radio-frequency acceleration using cavity resonators made of both normal and superconducting materials. For the Pelletron at Bombay, a superconducting linear accelerator is proposed as the booster; the accelerating elements being lead coated quarter wave resonators (QWR). A conscious decision was taken at the very beginning to develop indigenously as many sub-systems of the Linac as possible such as the cavity resonators, liquid helium cryostats, RF control electronics, magnetic quadrupole doublets etc. The Linac has a modular design, each module accommodating four QWRs operating at 150 MHz. The optimum velocity acceptance for the cavities is $\beta$=0.1. A total of eight to seven modules are planned. The injection path of the Linac has a beam sweeper, a phase pick up cavity, a superconducting beam buncher to compress the beam from the Pelletron to less than 200 psec, a foil stripper and a magnetic quadrupole triplet apart from the usual beam diagnostic and pumping elements. The beam through the Linac is periodically focused using magnetic quadrupole doublets located in between cryostat modules.

The quarter wave resonators operating at 150 MHz are made of OFHC copper. A schematic drawing of a QWR is shown in Fig. The tapered inner conductor and the cylindrical outer wall are connected together by the base plate which is e-beam welded to these parts. The base plate to the outer wall has a deep penetration weld from the outer side and a cosmetic one from the inside. This cosmetic weld has been the most difficult operation in the fabrication of the cavities. The machining and the e-beam welding of the cavities are done at the central workshop, BARC. The two drift tubes on the outer can are hydrogen brazed into position after adjusting the cavities to the designed frequency.  Since Pb is the superconducting element in the design of our cavities, the resonators made of OFHC cooper have to be coated with a thin layer of Pb ~ 2 mm, on the inner surface. The performance of the cavity largely depends on the surface qualities of this Pb layer. A lot of care is thus taken in the preparation of the cavity surface prior to the lead plating.

 First, the resonator is subjected to a sequence of mechanical polishing with fine abrasives of various grades using a tumbling machine. The final tumbling is done using walnut shells. A submicron surface finish is achieved by the mechanical polishing. The next stage of surface preparation involves electropolishing in a chemical bath of orthophosphoric acid, alcohol and water maintained at a temperature less than 180 C. After electropolishing, the resonator is washed, wetted and etched mildly with dilute citric acid. The resonator is then washed with de-ionized water and the lead fluoborate plating solution is poured into it as quickly as possible. All these operations are done in a specially designed covered sink with air flow controls. The lead deposition is done using pulsed electroplating at a current density of about 0.7 mA/cm2 The plating is done for a specified time to obtain 2 mm thick Pb coating. The plated Pb layer is thoroughly washed with high quality DI water, passivated with EDTA and dried with flowing Ar gas. The dried resonator is stored in vacuum.

The performance of the resonator is evaluated by measuring, in the superconducting state, its quality factor Q as a function of the electric field developed in the cavity. These measurements are done in a liquid helium cryostat developed in-house. This cryostat can hold one resonator and has all the connections and controls needed to measure the cavity Q as a function of the electric field. RF power to the cavity is coupled through an inductive loop which can be driven in and out of the cavity remotely by a stepper motor. After proper bake out of the resonator, it is cooled to liquid helium temperature. In the superconducting state, to measure Q at different field levels, the cavity is pulse powered under critical coupling at various field values as measured from the cavity pick up probe. The decay of the stored energy $U$ in the cavity is then observed to measure the characteristic decay time $\tau$ of the cavity given by

U=Uo /exp (-t/t)

                        The quality factor Q=wt, where w is the resonant frequency. This procedure to measure Q is good at relatively low field levels. A typical decay time for the QWR measured in our laboratory is about 180 msec. At high field levels, since the loss mechanisms are not purely resistive, Q is calculated from the measured input RF power in CW mode using the relation

Q= wU/P

where P is the measured input power and U is the stored energy for the given Eacc. The Q vs Eacc curve for a QWR measured in our laboratory is shown in Fig.4 before and after helium conditioning[4]. At a power dissipation of about 7 watts, it has been possible to run the cavity in CW mode at a field of 2.8 MV/m without any serious problems. Two resonators built indigenously have been tested so far and three more are under different stages of testing. As mentioned earlier the most difficult problem for us so far has been the cosmetic e-beam weld between the end plate and the outer can; even a slight porosity in this weld causes improper lead film deposition resulting in poor Q values.

      The modular liquid helium cryostat is designed to hold four b=0.1 QWRs. The various resonator and cryogenic parameters are controlled and monitored from outside using UHV compatible vacuum feed throughs. The liquid helium storage tank to which the four resonators are attached has a liquid helium holding capacity of 45 litres. The resonators are cooled by liquid helium through gravity flow. Both the helium tank and the resonators are surrounded by liquid nitrogen shields maintained at 77K. A few layers of high quality aluminized mylar wrappings on the nitrogen shields provide partial thermal insulation at 77K. The static heat load at 4.2K of this cryostat with all attachments is about 2 watts. 

References

1.      M.B. Kurup, R.G. Pillay and H.G. Devare, Indian J. Phys. 62A, 706 (1988).

2.      R.G. Pillay, M.B. Kurup, A.K. Jain, D. Biswas, S.A. Kori and B. Srinivasan, Indian J. Pure and Appl. Phy. 27, 671 (1989).

3.      B. Srinivasan, Animesh Jain, S.A. Kori and R.G. Pillay, DAE Symp. Nucl. Phys. 32B, (1989).

4.      B. Srinivasan, DAE Symp. Nucl. Phys. 36A, 157 (1993).

   back to pelletron home page



back to pelletron page