As the first long shutdown since the start-up of the LHC continues, many teams at CERN are already preparing for future improvements in performance that were foreseen when the machine restarts after the second long shutdown, in 2019. The LHCb collaboration, for one, has recently approved the choice of technology for the upgrade of its Vertex Locator (VELO), giving the go-ahead for a new pixel detector to replace the current microstrip device.

The collaboration is working towards a major upgrade of the LHCb experiment for the restart of data-taking in 2019. Most of the subdetectors and electronics will be replaced so that the experiment can read out collision events at the full rate of 40 MHz. The upgrade will also allow LHCb to run at higher luminosity and eventually accumulate an order of magnitude more data than was foreseen with the current set-up.

The job of the VELO is to peer closely at the collision region and reconstruct precisely the primary and secondary interaction vertices. The aim of the upgrade of this detector is to reconstruct events with high speed and precision, allowing LHCb to extend its investigations of CP violation and rare phenomena in the world of beauty and charm mesons.

The new detector will contain 40 million pixels, each measuring 55 μm square. The pixels will form 26 planes arranged perpendicularly to the LHC beams over a length of 1 m (see figure). The sensors will come so close to the interaction region that the LHC beams will have to thread their way through an aperture of only 3.5 mm radius.

Operating this close to the beams will expose the VELO to a high flux of particles, requiring new front-end electronics capable of spitting out data at rates of around 2.5 Tbits/s from the whole VELO. To develop suitable electronics, LHCb has been collaborating closely with the Medipix3 collaboration. The groups involved have recently celebrated the successful submission and delivery of the Timepix3 chip. The VeloPix chip planned for the read-out of LHCb’s new pixel detector will use numerous Timepix3 features. The design should be finalized about a year from now.

An additional consequence of the enormous track rate is that the VELO will have to withstand a considerable radiation dose. This means that it requires highly efficient cooling, which must also be extremely lightweight. LHCb has therefore been collaborating with CERN’s PH-DT group and the NA62 collaboration to develop the concept of microchannel cooling for the new pixel detector. Liquid CO2 will circulate in miniature channels etched into thin silicon plates, evaporating under the sensors and read-out chips to carry the heat away efficiently. The CO2 will be delivered via novel lightweight connectors that are capable of withstanding the high pressures involved. LHCb will be the first experiment to use evaporative CO2 cooling in this way, following on from the successful experience with CO2 cooling delivered via stainless steel pipes in the current VELO (CERN Courier June 2012 p29).

All of these novel concepts combine to make a "cool" pixel detector, well equipped to do the job for the LHCb upgrade.

• For more information, see the LHCb Upgrade Letter of Intent and Framework TDR at the LHCb webpage https://twiki.cern.ch/twiki/bin/view/LHCb/LHCbUpgrade.