Today I would like to go all the way back to my first column, in which I discussed stringing batteries in a row to create a large enough electric field to be interesting for particle physics. We quickly decided that batteries were not a practical solution. However, it is also not practical to inject beam from a very low-energy particle source directly into even a modest-energy synchrotron. The problem arises because the beam size is too large, and it increases rapidly at low energies since the individual particles in a bunch repel one another. At high energies this effect is not so noticeable since the beam’s longitudinal velocities are much greater than the transverse velocities. One solution to get from our low-energy particle source to a synchrotron is to go through a linear accelerator, or linac, first.
At Fermilab we initially inject beam from a negative ion source into something called a radio-frequency quadrupole. The RFQ accelerates the ions to 750 thousand electronvolts, or eV, while keeping them focused. Next, the ions pass through Fermilab’s linear accelerator, or Linac. Traveling down the Linac, the ions’ energy increases by a factor of more than 500 — to 400 million eV, which is the energy they need to be injected into the next part of the accelerator chain, the 8 billion-eV Booster, a circular accelerator or synchrotron.
As the negative ions enter the Booster, they pass through an injection magnet followed immediately by a thin stripping foil that removes the electrons from the negative ions (see illustration above), leaving only protons, which then go on to circulate through the Booster. (Protons are sent to targets to make beams for experiments or are sent directly to the experiments.) Using negative ions at the point of injection to the Booster allows us to inject 10 or more turns into the Booster for each acceleration cycle, thus enabling higher intensity per Booster cycle.
From the Booster, beam is sent into the Recycler and then on to the Main Injector, which accelerates the beam and before sending it to Fermilab’s various experiments.
The upstream half of the Fermilab Linac consists of large RF cavities, or tanks, with copper drift tubes inside along the center line of the cavities (see illustration below). The cavity resonates at 201 megahertz (MHz). The beam passes through a hole in the center of the drift tubes and experiences the accelerating field when it is in the gaps between drift tubes. In order to maintain synchronization between the beam and the oscillating field, each successive drift tube is longer than the previous one. This ensures that the beam is in the gap only when there is an accelerating field present. The process gets the beam up to 116.5 million eV.
The downstream end of the Linac has side-coupled RF cavities that oscillate at a frequency of 805 MHz. They are much more efficient in their use of linear space than the drift tube arrangement, resulting in a 400 million-eV beam energy for injection into the Booster.
Efficient Booster operation requires a gap in the beam to allow the extraction kickers to turn on without spraying beam into the Booster magnets. Removing ions to create this gap inevitably leads to residual radioactivity. However, the residual activity can be minimized by by implementing this procedure as early in the acceleration process as possible: the lower the beam energy, the lower the residual activity. Currently this process is being implemented just downstream of the RFQ. This will facilitate making more intense beam for the neutrino program and other experiments.
This column along with the previous columns I have written completes an overview of the Fermilab accelerator complex. When the proposed PIP-II accelerator is complete, the entire linac will be replaced by a linac that uses superconducting RF cavities.