Skip to main content
European Commission logo print header

Ultra-Fast Silicon Detectors: Enabling Discoveries

Periodic Reporting for period 4 - UFSD (Ultra-Fast Silicon Detectors: Enabling Discoveries)

Periodo di rendicontazione: 2020-03-01 al 2021-08-31

The measurement of trajectories of charged particles is ubiquitous in many areas of science, ranging from space science to the determination of the mass of molecules, from beam monitoring in medical treatments to research on fundamental particle interactions in nuclear and particle physics.

State-of-the-art tracking devices work similarly to old pocket cameras: they take a snapshot of a given view. A tracker shows the trajectory of many particles, in the same way as a camera shows a landscape picture with the sky, hills, trees, maybe a lake or the seaside. New digital cameras can tell us much more: they can make a video of the view, showing the same view in subsequent time frames. In this way, we can see, for example, birds flying into view at a certain instant.

The UFSD project proposes to transform present tracking devices by adding the “movie capability”, providing a sequence of frames separated by a very small time difference. In so doing, what has really happened in a given sub-atomic event will be much more understandable

Once space-time tracking is an established technology, it will be used in the next generation of many high-tech instruments. Currently, UFSD sensors have been chosen to be installed in the two most important high-energy physics experiments, ATLAS and CMS at the CERN HL_LHC collider.
1) Production of thin Ultra-fast silicon detectors (UFSD)

The key technological development that has been proposed and produced in the first phase of the project is the production of thin low gain avalanche diodes (LGAD) that we called Ultra-Fast Silicon Detectors (UFSD). UFSDs are a new concept in silicon detector design, merging the best characteristics of standard silicon sensors with the main feature of Avalanche Photo Diodes (APD). The overarching idea is to design silicon detectors with signals that are large enough to assure excellent timing performances but to keep the gain as low as possible. The complicated technological step that needs to be mustered is the capability of keeping the gain value low and be able to segment it.

Charge multiplication in silicon sensors happens when the charge carriers are in electric fields of the order of E ∼ 300 kV/cm. Under this condition, the electrons (and to less extent the holes) acquire sufficient kinetic energy to generate additional e/h pairs. A field value of 300 kV/cm can be obtained by implanting an appropriate charge density that locally generates very high fields. We call this additional doping the "gain layer". The additional doping layer present at the n-p junction in the UFSD design, Figure 1, generates the high field necessary to achieve charge multiplication.


2) Development of the simulation program Weightfield2.
We have developed a full simulation program, Weightfield2 (WF2), with the specific aim of assessing the timing capability of silicon sensors with internal gain. The program (see Figure_2) has been validated by comparing its predictions for minimum ionizing particles and alpha particles with both measured signals and TCAD simulations, finding excellent agreement in both cases. Figure 3 shows on the left the comparison WF2-TCAD for the predicted current produced by a MIP in 300-micron silicon sensors with gain = 14 while on the right the comparison between WF2 and the impulse measured at a beam test with 120 GeV/c pions, using as readout a charge sensitive amplifier. The left side, therefore, shows excellent agreement in the simulation of the mechanisms involved in the current signal. In contrast, the right side shows how the program also correctly simulates the electronic response.

3) Design and production of a custom VLSI chip to readout UFSD sensors.
We designed two full custom amplifier-comparator readout chip for silicon detectors with internal gain designed for precise timing applications. Both ASICs have been developed in UMC 110 nm CMOS technology and aim to achieve a time resolution of ∼ 30 ps. Each channel is independent, and the signal processing chain is composed of (i) trans-impedance amplifier, (ii) single threshold discriminator, (iii) stretcher, and (iv) LVDS driver.
The ASIC is optimized for a sensor capacitance of 3-10 pF and an input charge between 3 and 30 fC. The total power budget per channel is less than 30 mW.


4) Measurements of time resolution of UFSD
We performed several beam tests at the CERN facility in Geneva with π-mesons with a momentum of 180 GeV/c and at FNAL at Fermilab with protons of 120 GeV/c. Many UFSD geometries have been tested successfully, demonstrating an excellent time resolution of about 30 ps.


5) Study of radiation damage to the gain layer
Radiation damage causes three primary effects in UFSD: (i) decrease of charge collection efficiency, (ii) increase of leakage current, and (iii) changes in doping concentration.
Effect (i) and (ii) are not influencing the performances of UFSD.
Changes in doping concentration: UFSD sensors have shown a decrease of gain values for fluences above 1e14neq/cm2, with a complete disappearance of the gain at 1e15neq/cm2. This effect has been understood as a gradual inactivation of acceptors due to radiation defects. Significant progress achieved in the past two years has been the understanding that the addition of impurities, especially that of carbon atoms, reduces radiation damage significantly. Now, carbon-infused UFSDs are the most radiation-hard UFSD sensors ever produced.


Results have been described in many conferences and invited talks. A detailed summary of the grant achievement has been written in a open-source book, "An introduction to Ultra-Fast Silicon Detector"
The project has been very successful, and it is appropriate to state that it has transformed the track-tracking established paradigm introducing the concept of 4-D tracking, the concurrent measurement of space and time with high accuracy.

We proposed to design and manufacture sensors and associated electronics able to measure space and time concurrently with unprecedented precision, and this is what has been done.

Our sensors, the so-called Ultra-Fast Silicon Detector – UFSD, are now planned for installation in the two largest particle experiments, the ATLAS and CMS at the CERN HL-LHC.
A critical step toward this success has been that the largest producer of silicon detectors for research, Hamamatsu photonics, has adopted our design, and it is now offering it in its catalog.

As a demonstration of the call success, almost all major conferences on particle detectors now have extensive sections on 4-D tracking, where, often, members of this project are asked to be chair.


Two aspects of the project will be achieved past the project deadline:

- the production of sensors resistant to radiation with 4D tracking with very small pixels in Q4/2021
- the electronics for small pixels in Q2/2022

Albeit outside the timeline of the project, both achievements are a consequence of the project success.
Difference in design between a normal silicon diode and a UFSD sensor
Comparison between Weightfield2 and TCAD simulation or beam test data
GUI of the Weightfield2 simulation program
Distribution of the signal amplitude at four values of the detector Vbias: (100V,130V , 200V and 230