CERN Accelerating science

Developing radiation hardness electronics for future detectors

Particle colliders generate an extremely large amount of high-energy interactions in order to produce and detect very rare events that can be used to improve our understanding of the underlying physical laws. In doing so, a significant background is generated, which in turn degrades the performance of the detector through radiation damage. This damage can be subdivided into two categories: displacement damage, mainly affecting the silicon sensors via the alteration of its solid state structure; and total ionizing dose (TID), impacting the performance of transistors present in the front-end readout electronics through the trapping of charge in their insulating oxides.

Moreover, Monte Carlo codes such as FLUKA can be used to simulate the radiation environment in high-energy accelerators. For future machines such as the FCC, such calculations are the only means of estimating the radiation levels. These simulations show that the sensors and readout electronics present in the first layer of the so-called inner barrel detector will be exposed to 1 MeV neutron equivalent fluence (representative of displacement damage) and TID levels of 6×1017 neq/cm2 and 400 MGy respectively. Such values are roughly a factor 50 larger than those expected for the High-Luminosity LHC (HL-LHC), which already pose extremely challenging constraints on the radiation hardness of the detectors.

In order to illustrate the impact of radiation on electronics, we consider a state-of-the-art transistor technology, typically described through its technical node or minimal distance between the transistor drain and source. A cross section of a MOSFET transistor is shown below. Whereas the very aggressive scaling of the transistor size has in general improved their resistance to TID through the shrinking of the gate oxide and therefore reduction of the sensitive region, for highly scaled technologies such as 65 and 28 nm, the impact of TID is significant already at 10 MGy, as shown below. When considering the target levels for FCC introduced above, it can be easily understood that encountering transistor technologies resistant enough is a huge challenge.

In addition to the impact of radiation on the detector performance, the radiation levels in the accelerator itself, though orders of magnitude lower for locations hosting electronic systems, can also negatively impact the performance of the machine through the failure of critical equipment such as powering, controls, cryogenics, vacuum or machine protection. Such systems are typically based on Commercial-Off-The-Shelf (COTS) components, which are not designed to operate in harsh radiation environments. Though the cumulative degradation mechanisms mentioned above also apply, for accelerator applications it is typically Single Event Effects (SEE) induced stochastically by nuclear reactions from the hadrons present in the radiation field which pose a strongest threat to a successful operation. 

In order to ensure the reliability and availability of the machine, such systems need to be qualified against radiation. Due to the very large number of semiconductor devices involved, characterizing them individually is not feasible cost and timewise. Therefore, full systems are tested in the CHARM facility, which mimics the radiation field in the accelerator through the interaction of a 24 GeV proton beam with a 50 cm copper target. Images of the facility can be seen below. The same beam is used further upstream to qualify detectors against displacement damage for the experiment environment in the IRRAD facility.

Testing at a system level can significantly reduce the cost and time with respect to discrete component testing, however in order to obtain relevant and reliable results, the observability of the various failure modes and identification of the faulty sub-systems or components need to be carefully defined. These challenges are common to other applications such as space, avionics and ground-level. For this reason, the RADSAGA Innovative Training Network (radsaga.web.cern.ch) coordinated by CERN and funded by the European Commission unites a broad variety of radiation effects experts in the industrial, research facility and university domains, with the objective of addressing the challenges related to facility calibration and emerging effects at component and system level. The goal of the project is to deliver a handbook on radiation qualification at system level which could serve as a guideline to the radiation effects community. Challenges such as the radiation hardness assurance for HL-LHC and FCC will be at the heart of the project objectives.

Finally, the RADECS 2017 international conference, to be held in Geneva in October and in which challenges such as those introduced above will be treated, will also provide an ideal opportunity to interact with world-wide experts on radiation effects in electronics.