CERN Accelerating science

Setting a roadmap for future experimental technologies

by Panos Charitos

In elementary-particle physics as in most fields of science, advances in understanding are closely coupled to advances in technology. Machines of higher luminosity open opportunities for the observation of rare and unexpected processes while higher-energy accelerators are needed to cross thresholds for suspected new phenomena.  Advances in accelerator technology must be accompanied by advances in detector technology as more complicated particle collisions are produced. 

The technological sophistication of the LHC detectors is almost incomprehensible as they include several subdetector systems, contain millions of detecting elements and support a research program for the international particle physics community. The volume of data that will be produced during the high-luminosity upgrade of the LHC (HL-LHC) and in future colliders call for even more sophisticated technologies.  Further advances are necessary to enable the processing of larger and more complex data samples that boost and eventually boost the performance beyond today’s state-of-the-art.

The first workshop on future experimental R&D took place at CERN's main auditorium.

To respond to this challenge, CERN’s Experimental Physics department has launched a process to define its R&D programme on new Experimental Technologies from 2020 onwards. Experts from all over the world have signed up to the eight working groups that were created to define a solid R&D programme in technological key areas and more than 450 physicists and engineers have participated to this workshop.

This initiative initially spans a 5-year period with a foreseen extension beyond this period. It covers detector upgrades beyond LHC phase II and includes concepts developed for CLIC and Future Circular Colliders (FCC). Christian Joram, coordinator of the R&D steering committee and organizer of this meeting, explains: “beyond the high luminosity LHC upgrade the landscape of experiments is only vaguely defined and may evolve in different directions. Therefore we want to launch an R&D programme that concentrates on advancing key technologies rather than developing specialised applications.”

As physics experiments get more ambitious, the detectors they use have to keep pace. Improvements include better electronic readout, better modelling and simulation tools, and better computational techniques for reconstructing the recorded information by the detector. There are certain improvements that will be mutually beneficial for different types of detectors. One example is the increased accuracy in timing. According to Joram: “ultrafast (sub-ns) timing is a promising way to mitigate pile-up in very high luminosity environments and will almost certainly impact the development of all classes of detectors, be it silicon, gas or photodetectors”.

Manfred Krammer, Head of CERN’s EP department and co-organizer of the workshop, emphasized in his opening speech the collaborative spirit of this effort: “We look for synergies, complementarity and cooperation. CERN’s R&D programme is not an isolated path but is carried out thanks to strong partnerships with other laboratories and research institutes” and he added “it is timely to think how the industry can be involved in joint R&D efforts and what we can learn from industrial innovation in fields related to detector technologies”.

Every working group had typically two meetings before the workshop, with up to 70 persons attending, many from external institutes. In their talks the convenors reported about the material and opinions collected during these meetings. The talks covered a variety of topics reflecting the EP department’s diversity and the strong collaboration with partners from all over the world. 

In the domain of silicon detectors (WG 1), the depleted CMOS technology attracts a lot of interest, as it allows building low-mass and high-resolution sensors, which have the potential to cover large areas at affordable cost. In the last years, substantial progress was achieved in terms of radiation hardness, obviously a key requirement for almost all future experiments. Another trend are Low Gain Avalanche Detectors (LGAD), which have achieved timing precision in the range of a few tens of picoseconds. Finally, there is also a strong need to continue the development of classic silicon detectors (pixels, strips and pads) and the associated electronics.     

Gas detectors (WG 2) will continue to play an important role in future experiments as they can cover very large surfaces at moderate cost. Micro pattern technologies have boosted their performance and a wealth of new ideas exists in terms of materials, production techniques and readout modalities, how this success story can be pursued. An important question is the optimum choices of the gases and their compatibility with environmental standards.

Working group 3 is dealing with calorimetry and light based detectors, two topics often tightly linked. Highly granular calorimeter concepts are in the focus, allowing decomposing jets in their individual particles (‘particle flow calorimetry’). The requirements are manifold and so are the options in terms of active media (e.g. scintillators, noble liquids, silicon), absorbers and readout concepts. Again, fast timing will play a key role. In the last years, we could witness a revolution in the world of photosensors. The so-called Silicon Photomultiplier (SiPM) has matured and is becoming a standard tool for many applications. Still, its high dark count rate and moderate radiation hardness are constraining its use when it comes to the detection of low light levels, e.g. in Cherenkov detectors. This is still the domain of vacuum tubes, where modern finely segmented multi-anode devices, micro-channel plate tubes (for ultra-fast timing) and completely new concepts go far beyond the classic PMT.

The working group on detector mechanics showed an impressive potpourri of advanced materials, design tools and production technologies, which have the potential to change the way we build detectors, but also boost their performance. Ultra-light and precise or also very large-size carbon-fibre components become accessible and lead to solutions which were inconceivable ten years ago. Environment-friendly cooling technologies, combined with advanced 3D printing and micro-fabrication technologies reduce the material budgets of vertex and tracking detectors. Tight cooperation with high-tech industries is the way to go.

Working groups 5 and 6 gave us insights in the rapidly evolving worlds of electronics and fast data transmission. The numbers of custom-specific chips in the LHC experiments is counted in millions, however the HEP community is seen as an expert customer with very special requirement. The ever-decreasing feature sizes in the CMOS manufacturing processes lead to benefits (higher functionality), but also to unknowns (radiation hardness) and risks (much higher cost). The main challenges are the strongly increasing ionising doses and the necessity to read and transfer huge amounts of data. Even though the parameter space is complex, the two working groups could already identify promising routes.

At the other end of the data link sit the processing farms with their advanced software for pattern recognition, track reconstruction and other computations. This is the topic of WG 7. The speakers discussed approaches on how to deal with the track reconstruction challenge at an event pile-up of 1000 (FCC-hh scenario) and sketched the potential of machine learning. For a population of more than ten thousand physicist, the EP department is a key provider of software frameworks and toolkits. For the Exabyte era already at the horizon, simple extrapolation of the existing will not be sufficient and fresh concepts for data management will be needed.

Last but not least, WG 8 presented partly already ongoing studies of experimental magnets for LHeC, CLIC and various FCC detector flavours. For FCC-ee a ultra-thin 2T concept is being studied with a free bore of 4.4 m and 6 m length, whose material budget should stay below 1 radiation length. The FCC-hh baseline design foresees a very large main solenoid (a free bore of 10 m and a length of 20 m) and forward solenoids at both ends. All developments require progress in the superconducting cable to meet the increased strength requirements and a multitude of more generic studies.     

A second workshop will take place later this autumn, and will start the preparation of a report summarizing the proposed R&D programme. In the meantime, all working groups will filter the proposals discussed during the workshop and agree on roadmaps for future R&D lines. This will culminate in a final report on time for the upcoming European Strategy Update.

Finally, it should be noted that developments of detectors for high-energy particle physics benefit many other sectors from healthcare and medical imaging to industrial manufacturing and quality monitoring. The R&D programme on detector technologies shows that there are some big challenges and innovative ideas, and the benefits can be realised by a series of step-by-step improvements that will be documented in the final report due by the end of this year.

You can find more information and a full list of the presentations on the indico page:


The author would like to thank the members of the steering committee for their fruitful comments.