Proton Driver Meeting

5 - April - 2001
FNAL Black Hole

Presentations:-

• Alberto Marchionni: MINOS as is...
• Larry Wai: What can We do with 1.6MW?
• Erik Zimmerman: High delta-m experiments
• Gerald Jackson: Antiproton Production using an upgraded P.D.

The maximum allowed power for MINOS is 0.4MW. It was designed for 4x10**13 protons on target (POT). One can't just deliver four times more at 120 GeV.

One might need to design a new beam transport system to accommodate the lower energy proposed because the beam will be wider. This means the dipoles will have to have a larger aperture. The replacement of six dipoles would cost on the order of $150k. Weiren has studied this already.

An additional problem arises from the rapid cycle of the Booster: 15Hz as compared to the 2.9s spill. The current horns cannot be ramped at that rate, even for 0.4MW. The current horn is pulsed at 0.5 Hz.

Proposal: put the protons in the Main Injector (MI) using the slip-stacking technique. This already is needed for Run IIB. Then the protons for MINOS could be extracted with the right time structure.

Question: can the MI tolerate a factor four higher intensity? Answer: this is addressed in Appendix D of the design report.

Currently, when simulating the physics to be done at MINOS, one observes a large tail to high neutrino energies. At a later stage of the experiment, when searching for nu_mu to nu_e oscillations, MINOS could reduce this tail using a beam plug. While the benefit is very clear (at least a factor two reduction in the tail), it should be appreciated that the technology for such a beam plug is far from trivial.

What happens if the protons have a much lower energy? Naturally the pion yield per proton is sharply reduced. But it is the power which is a constant rather than the number of protons, so one should look at the yield divided by the proton energy. In this case the optimum is in the 20-30 GeV range -- much lower than the current design of 120 GeV. A relatively simple calculation shows a possible increase of 40% in yield. Caveat: this calculation assumes a thin target which is not really applicable, and the gain in yield wrt 120 GeV is not as great as indicated.

More important is the energy spectrum of the neutrinos, which will be much cleaner than with the current beam. In particular, the high energy tail will be absent, without the need for a beam plug.

Conclusion: If the beam energy is reduced by a factor of four to 30 GeV and the number of protons increased by the same factor, and if the protons are extracted in 1.9s spills from the MI and the total power is limited to 0.4MW, then it is beneficial to MINOS in later stages when this experiment searches for nu_mu/nu_e oscillations. However, MINOS cannot serve as the main justification for an upgrade of the proton driver.

Other physics, however, might provide a good justification -- see the next talk.

Physics studies based on a 300km baseline (FNAL-SLAC).

Assume three families, and that the atmospheric and solar neutrino oscillation parameters are known. These studies are based on off-the-shelf codes from NUMI and MINOS. The scenario is 2006, after MINOS has run. There are five interesting questions:

1. What is sensitivity to nu_mu --> nu_e?
2. Can we utilize matter effects to deduce the sign of delta-m23**2 by comparing nu and nubar interactions?
3. Can we observe CP violation from the shape of the energy spectrum?
4. Can we improve the measurement of nu_mu disappearance already completed by MINOS?
5. Can we observe nu_mu -> nu_tau appearance as in Opera?

There are three Options for consideration.

1. Take the higher proton flux, but spend no new money on beamline or detector (assumed 10 ktyr exposure).
2. Use MINOS, but redo the beamline to handle 1.6MW.
3. Build a new beamline and a new experiment, and put it 3000km away (e.g. at SLAC). The detector would be fine-grained, at least 5 ktyr, something like a liq.Ar TPC.

Results:

1. There is only a small improvement whereby a 1.4sigma effect becomes a 1.8sigma effect.
2. The result is limited by systematics on the background estimate.
3. Limits on nu_mu/nu_e oscillations are improved by an order of magnitude over CHOOZ, and a factor 5 over option A. The limits are statistics limited, as the signal/background is much better than in MINOS: with this E and L, one would be collecting data at the peak of the oscillation probability. Furthermore, the production of background events is reduced, due to the lower solid angle subtended by the experiment. Plots show a huge difference between nu_e and nu_e_bar spectra due to matter effects. Switching the horn currents allows a clear test of C. The exact shape of the nubar spectrum has a demonstrable sensitivity to the CPV phase.

Conclusions:

• There would be a large improvement over CHOOZ limits.
• There could be the discovery of nu_mu/nu_e oscillations.
• There is a possible sensitivity to CP violations.

Future studies:

• nu_mu -> nu_tau appearance
• improvement on mu_mu disappearance

Strawman R&D program:
Soon: build a 10kt prototype based on ICARUS and install it with MINOS.
2006 and later: build the new beamline at Fermilab, and the final detector at SLAC.

Discussion:
This beamline would be very difficult. What would be its depth? How severe would the ground water problem be? This idea is not crazy, however, one has to realize this is a NEW facility, and would cost on the order of $100M.

An entry-level neutrino factory would give a 10x better limit on nu_mu/nu_e oscillations than option-C.

Bottom line: there would be a factor 3.5 more sensitivity, assuming no LSND signal.

Options: 1 BOONE experiment, or 2, with the second one 500m away. Assume 10**20 protons on target.

Justifications: Supernovae element production -- theory would be assisted by a sterile neutrino. A small mixing angle would be most interesting.

Some documentation is in progress and will be made available on the web.

Assume that Run II is over and antiproton stacking is terminated. We take Eproton = 8 GeV and a 15Hz cycle, etc. Assume that less than 20% of the protons are used for producing antiprotons.

Missions:

1. atomic physics (properties of antiHydrogen)
2. nuclear physics (stability at low momentum transfer)
3. biology and medicine (esp. cancer treatment)
4. charmonium production and spectroscopy
5. tests of gravity
6. space fuel
7. production of effective PET isotopes

One would need a means of storage and transport of collections of antiprotons. NASA already has developed a portable Penning trap, and is ready to accept antiprotons for its studies of space propulsion.

The cost of producing antiprotons for non-accelerator uses would be $27.4M/year. Power is only 15% of this.

The demand from NASA and for cancer treatment is way above supply. It would be work taking 1% of the antiprotons produced during Run II (which would not impact collider physics) and devote it to at least some of the missions above.

If the use of this 1% of the antiproton yield during Run II is successful, then private investors are ready to invest on the order of $1B to build a dedicated pbar production facility at Fermilab. The income for the lab could be in the millions of dollars.

In light of this lucrative potential, we would hope that the DoE would see a relatively modest investment in the upgrade of the proton driver as a worthwhile investment.