ILC logos from linearcollider.org.
Last time we looked at some of the reasoning for why the accelerator community would like a large-scale electron/positron linear collider. This time we'll look at one of the collider proposals in detail. In particular, what will the International Linear Collider (ILC) look like, if it's built?
The biggest question is often, at what energy would we be colliding? The design for the ILC currently states that we will run at 250GeV (giga-electron volts) per beam, or 500GeV total. By comparison, the LHC's target operation is 7000GeV per beam. Again, remember that not all of that 14,000GeV is available for producing particles, whereas at the ILC the full 500GeV would be. As a frame of reference, the famed Higgs boson (one of the main objectives of the LHC) should have a mass of between 115-185GeV. That should put the ILC at exactly the right energy for producing Higgs particles, among other things.
There are two ways we could accelerate to that energy. The first would be to have a reasonably short linear accelerator (or "linac") with a ridiculously high acceleration gradient. This would require a huge R&D effort to figure out how to achieve such a large accelerating field. Alternatively we can have a ridiculously long linac with a reasonable acceleration gradient. The present design for the ILC opts for the latter. The main linacs will accelerate particles using Radio-Frequency, or RF, cavities. (The RF system certainly deserves an entire post of its own, so I won't go into details as to how it works here) Each linac have about 7,000 RF (radio-frequency) cavities, for a total of roughly 14,000. That's a lot of cavities! By comparison, the LHC has a measly 16 RF cavities to handle its acceleration.*
In the current ILC design, each cavity is about a meter long. That puts the total length of each linac at about 7km (4.4 miles), but since you're aiming the two linacs right at each other, that puts the overall end-to-end length of the ILC at a minimum 14km (8.8 miles). There are many other elements in the linacs besides cavities though, so the total end-to-end length is closer to 30km (19 miles)!
Remember why we chose to use a linear accelerator rather than a circular accelerator: linear accelerators don't generate much synchrotron radiation, therefore you don't lose much energy while accelerating the beam. However, this is a double-edged sword-- synchrotron radiation actually works to our advantage with electrons. In circular electron/positron machines, we utilize synchrotron radiation for radiation damping. Here's how it works for a circular accelerator:
Say you have a beam whose particles have a lot of transverse (side-to-side and up-and-down) motion. The beam is somewhat large and sparse, therefore when this beam intersects with the opposing beam, we have relatively few actual collisions between particles. Most of them just pass right by the other beam without interacting.
When the beam is bent in a circle, it radiates a small amount of energy in the form of x-rays. This "cools" the beam in all three dimensions-- horizontally, vertically, and longitudinally (front-to-back). That is to say, we reduce motions in all three dimensions. If we don't replace the lost energy somehow, the beam will just continue to lose more energy every turn until the accelerator can't store the beam anymore. Instead of just letting the beam decay, we can selectively replace the energy in the longitudinal (forward) direction only, via RF cavities. Now we burn off energy in all three dimensions, but only replace it in one. Since the beam only loses a small amount of energy each time it circulates through the ring (~0.001% of its total energy), we need to repeat this several thousand times to have an appreciable effect. The net result is transverse (horizontal and vertical) damping.
Circular electron/positron machines use this to their advantage to increase the beam density by collapsing the beam transversely. For reasons I will elaborate on in another post, damping causes the beam to become a thin "ribbon"-- the vertical dimension is much, much smaller than the horizontal dimension.
In a linear accelerator we have almost no synchrotron radiation, so we can't damp the beam. That means we have to start with a very small beam before traveling down the linacs. Again, we have two options. The first is to have a very good electron (and positron) source, which generates pulses of particles that are of the right characteristics. This turns out to be a very complicated solution, requiring constraints on the source that would prove difficult to attain. The second option is to bolt a circular accelerator on to the start of each linac, allow the beam to settle, and transfer the beam from the circular accelerator to the linear accelerator after the beam has reached the desired state.
As it turns out, the second method is quite a bit easier, and that's how the ILC will operate. The circular accelerators we've bolted on to the start of the linacs are called damping rings. They have no other purpose in life than to improve the beam quality before transferring to the main linacs. In order to achieve the parameters we want, the damping rings will need to be about 7km (about 4.3 miles) in circumference and circulate at an energy of 5GeV (1/50 the final energy). The beam will then travel down a transfer line to the start of the linac.
To put it all together, here's a diagram of the latest proposal of the ILC:
Image from the ILC website, linearcollider.org.
In this cartoon of the ILC, we have a small "starter" linac for the electrons which injects into a storage ring at 5GeV. The beam damps down and is transferred to the main electron linac. About halfway down the electron linac, there is a positron source (using the electron beam to generate x-rays, which hit a metal source and generate positrons). The positrons enter their damping ring and finally their main linac. The two beams then collide at the "IP", or Interaction Point, in the center of the main detector. It sounds like these two beams would be "out of sync" with each other, but the beams will be run more-or-less continuously so the discrepancy won't matter.
There are many details still being worked out for the ILC-- for example, where it will be built-- but much of the initial design work has been completed.
Next time I'll elaborate a little more on the concept of damping rings, how they work, and how my research is directly related to the ILC's damping rings.
---------------------
*From LHC Design Report, Vol.1 Ch. 6.2.1. Please correct me if this is wrong-- I couldn't find a better answer for how many cavities they have!
---------------------
SPECIAL BONUS SURVEY! If you're reading these posts, let me know! I'm curious to see how many people, and who, are interested enough to slug through these posts. Are these articles interesting? Too technical? Too long? Not enough gory details? Anything in particular you liked or disliked, or would like to hear more about? Let me know!
No comments:
Post a Comment