Minutes of the MUCOOL meeting, March 12 1999 A. News: (S. Geer) Due to security concerns, Fermilab management has decided to change the locks for accessing Lab G. Once the new locks are installed, we will be told about new key number etc. Anyway, the other news concerns the removal of the two magnets in Lab G. They should be done the end of the day, literally, but Norbert could confirm this, as he could not access the building.. 1. Report on the design of LH2 vessel, and flow consideration. Ed Black, with support from Dan Kaplan. Ed started his talk by showing the beam profile at the end of the decay channel. The r.m.s. of the X is about 0.54 cm. (same for Y , the beam and the device are assumed to be phi symmetric). Note however that the beam at the entrance of the channel is about twice as big. This is relevant for two reasons (i) optimization ofthe radius of the vessel (ii) distribution of the heat density in the volume. While the former topic was not on the table for today, Ed reported on a plan to understand how serious the heat disssipation problem is. In the current design, the target volume is 3.3 L. (5 cm. radius, L = 42 cm. ). If the heat source (due to ionization) would be uniformly dsitributed in the target, it would imply a temperature elvation of 0.3 degree C per sec. (assuming baseline muon bunch intensity ). However, most of the energy is deposited in volume about 60 times smaller than the vessel volume. How can guide the flow so that it does not go straight trough the vessel? Ed has designed a 3D flow pattern, so that the liquid goes through the entire length of the vessel. We currently are planning to change the target volume once a second. But in addition, we might have to think in terms of cross-flow, or helicoidal path, or initiate tubulence, so that the energy get deposited more uniformly in the vessel. After looking at some analytical expression, Dan and Ed ( and collborators) plan to look at finite element analysis codes to simulate this heat dissipation problem. Candidate codes are ANSYS (it has good reputation, we have expertise at the lab) or ALGOR (cheaper). 2. Emittance exachange section. Rick Fernow worked on the emittance section, based on the use of two double bent solenoids, wedges and r.f. (for bunch rotation, and re-acceleration ). Only the first half of this channel is considered in this report. The initial transverse (longitudinal) emittance is about 650 pi mm mRad. (2.17 mm) respectively. Rick was successful at controling the emittance growth going through the first bend solenoids, via more accurate matching. He is now concentrating on the wedge. This wedge is itself subdivided into 6 pieces. Starting from a Delta E of 8.34 meV, the recored Delta after each of these consecutive sub-wedges are 8.41, 7.94, 7.46, 7.04, 6.71, 6.39, 6.43, MeV, respectively. Clearly, the performance in terms of reducing the momentum spread in these wedges is decreasing. Rick will optimize this number of wedges. In terms of overall ``heating" budget, the 6D emittance gorws by a factor 1.02 through the first 1/2 bend solenoid, 1.08 after the r.f. , 1.17 through the second half bend solenoid, and 1.61 through the first wedge. Rick will patiently keep optimizing!. 3. Status of the high power klystron engineering studies. Presented by Norbert and Al. The use of the Open cell cavity for our cooling channel requires high power klystrons. The current spec call for a 805 MhZ device, capable of delivering 80 MW peak power, 15 micro-sec pulse duration and rep. cylcle of 15 Hz. The plan is to contract out a $50,000 study to one of these comapnies for a rough estimate of the necessary R&D to meet these criteria. We are incontact with two companies : CPI and Litton. Al had a one hour, very positive conversation with a Litton rep. today., they will formally issue their bid for this study very soon. We expect Litton will also do so. In addtion, Norbert reported that the CLIC collaboration (CERN) is also interested in using such high power klystron. They would use a slightly different frequency (907 MhZ), lower peak poer (25 to 50 MW), but much long pulse. They are in contact with Thomson (France). Norbert proposed to share the information that would come out of such studies. Jim Norem asked if there were any commercial application for such device, unfortunately, we don't of any (Modern radar use higher frequencies). 4. Idea for a traget configuration for a "bunched MUCOOL" experiment. Norbert Holtkamp. Norbert presented a few slides, still sketchy at this stage, but quite promising on a first bunched muon channel for early testing of the cooling channels. The idea would be to embed a tungsten (or an other suitable heavy metal) rod in to an existing (antiproton source) transformer, so that a high gradient mangetic field strongly focus (beta = 3 cm.) the pion through the rod. The beam phase space exiting the rod is such that the 2D phase space ellipse is vertical, i.e. , large angular spread but small radial spread. Approx. 22 cm. downstream of the end of this rod, we would place a 8 cm, 1.25 surface field Li-Lens to rotate this phase space. This 8 cm. waist beam can (almost!) fit into an 805 MhZ cavity used for bunching or phase rotate. This pi colletor would be followed a conventional FODO channel, accepting an emittance of 3,000 mm mRad. The maximum beta function of this FODO channel would be roughly 4 m. with an average of 2.5 m. Such quadrupole exists, they more or less match those used in the antiproton source (aperture of 9 cm.) In such a FODO channel, a dipole could be inserted to remove the proton from the beam. 5. pion production yield out of this first MUCOOL target. Nikolai Mokhov started by reminding us of the "baseline" design. He noticed that, once the capture solenoidal field get reduced from 20 T. down to 5 T., and the length of the target is also a bit shorter, the pion yield decreases by a factor of 50. This motivates us to look for an other design, which would not reached the muon collider luminosity, but would be a matter match (i.e., more economical) to the needs of MUCOOL. Nikolay has installed Nrbert target and lens soncfiguration in Mars, and showed preliminary results. He noted that the pion momentum distribution is much flatter coming out of the Li-Lens that in the solenoid-based scheme. Also, the transverse momentum is smaller (~ 20 MeV/c), commensurate to the overal reduced transverse emittance. Since the muon will pick up comparable transverse momentum in the pion decay process, there is not much sens trying to reach lower emittance at this stage. The pi+ (or pi-, but not both!) is about .03 per incident proton. This is about a factor three better than the 5T solenoid (and probably cheaper!) Thus, we are on the right track... The next step is to firm up the design of the FODO channel, and try to understand the longitudinal structure of this beam, and it's use in the cooling channel. Enough horror stories for today.