A Quantum Diaries Survivor

CERN under siege September 10, 2008

This morning CERN appears under siege. The management evidently wanted to arise as much interest as possible, and they got what they bargained for. To give you the flavor of the event, as seen from “behind the curtains”:

• Coming to CERN took half an hour: long queues of cars at the entrances are caused by strict controls of passes by a tripled set of guards on duty. If you are not a user or a registered visitor, abandon the idea of sneaking in.
• Visitor passes can be obtained at Building 33, which is the Visitors center, hosting also a nice exhibit with videos and other interactive displays. However, today there appears to be a queue and a host of reporters crowding the place. I would advise no attempt at getting a pass there either.
• CERN made a honest effort to provide streaming video coverage of the event, but they got sucked in the black hole they organized with their own hands. Never underestimate the power of the web: the traffic is so high that there is little chance to see anything streaming but black photons out of the video frame. As for EVO, it does not even allow people to connect to the interface as of now. If you want to try your luck, however, be my guest.
• The main Auditorium, where a presentation of the events unrolling today is being given, is also completely full, to the point that there are valets diverting the continuous stream of people trying to enter the large theatre to another building (Bldg 40, where the Atlas and CMS experiments have most of their offices), where nothing, however, appears to be going on.
• The CMS control room is full of “experts”. I was originally scheduled to be on a day shift at the CMS tracker, but yesterday evening I was informed with the other shifters scheduled to attend the various parts of the detector that today’s shifts were taken over by the best crew we could put together. This is quite funny in the case of the CMS tracker, for a simple reason: it is going to be kept OFF today! So those experts will have their hands free to reach for glasses of wine and munchies provided as an exception to the rule. They will also be free to smile at the cameras pointed at them: the CMS control room will see a continuous stream of reporters in scheduled tours.

Maybe I need to explain, in this otherwise useless post, why the CMS tracker, which is the heart of the whole detector -it sits at the very core of the giant machine, and it is the first line of observation for the particles produced by the collisions, providing the most precise input on the trajectories of charged particles- why, I was saying, it will be kept OFF during this first historic day of proton running.

The CMS tracker (see picture on the right) is made of thin (300 microns thick) layers of silicon sensors, shaped in wafers with a side lined with narrow electrodes, spaced a few tens of microns from each other. When the silicon gets polarized by a few hundred volts of electric field, it acts as a solid-state ionization detector: the charged particles leave ions along their track, and the freed electrons drift in the electric field across the silicon to the thin strip-like electrodes, yielding a signal there. About 20,000 electrons are collected in two or three strips every time a particle crosses the 300 microns of silicon.

The scale of the tracker is impressive: there are a dozen layers of silicon sensors, for a total area of about a hundred square meters and several million electronic channels. Despite its huge scale (as far as these precision devices go), it is necessary to protect it from high radiation doses, because it deteriorates with time. Be sure to understand: this is a device that will sit for years at the core of CMS, where protons will hit protons forty million times a second, every time producing hundreds of particles flying in every direction. However, these are “normal” operations. Instead, today we will have protons circulating inside CMS for  the first time, and we cannot yet be sure that their orbit is fully under control. If the beam deviates even half an inch from its orbit in the 27 km ring, the  innermost silicon layers of CMS might be sprayed with a lot of unnecessary radation. This is relatively harmless when the detector is off, while it may damage the sensors if they are polarized with the normal electric field. Past experience (for instance, in CDF) has shown that until one gets a well-trained, stable beam, the silicon has to stay off. In CDF we burned a few sensors in a couple of beam incidents a few years ago, and the procedure to closely monitor the stability of the beam before turning on the silicon has become a standard.

So, the tracker will be OFF. A shifter at the tracker (there are two in each shift) usually sits in front of four screens running a graphical user interface which a child of five could easily operate. The shifter just needs to click a few buttons when they become red, and then click a icon showing a cartoon of Albert Einstein to solve most problems. If the problem persists, a phone call solves it. But today, the experts will be even less busy than that…

I will anyway drive to Point 5, where the CMS control room is, later today, to check how things are going and take a few pictures. But I expect no excitement other than that of observing the excitement of those who know less. A 450 GeV beam of protons will make a few turns around the ring: big deal, that is 6.7% of the design energy, and there will be only a few billions of them, which is orders of magnitude less than what LHC will manage next year. I will be more excited to see our detector light up when the first collisions will start. But for that, we will have to wait one full month.