Silicon detector dark current evolution with fluence in a pion beam

Peter Booth, Craig Buttar, Ian Dawson, Paul, Dervan, Chris Grigson,

Ben Kitchener, Richard Nicholson, University of Sheffield

Note: The results on this page are preliminary !


Previous work has shown that the silicon detector macroscopic damage parameters: α-the current damage parameter and β-the acceptor introduction parameter, can be extracted from a measurement of the dark current evolution with fluence during irradiation with a proton beam. In this work, a similar study has been undertaken using the pion irradiation facility at PSI..

The experiment

A portable version of ATLAS PS irradiation facility was developed for setting up at PSI (and in other facilities). The system maintains the detectors at ~-10C and biases them, and continuously monitors the detector currents and temperatures within the system. The fluence is monitored using a rate signal from a SEC, that is calibrated using the activation of Al foils and measuring the production of 22Na and 24Na.

The setup is shown above. The detectors are mounted on Al-phenolic-Al sandwich (figure 1), two detectors are then attached to an Al-frame. The frames are mounted on an Al cold plate (figure 2)which is kept cold by flowing ethanol-water mix through a channel in the cold plate. Temperatures in the box are monitored using 3 Pt100s, two are located on the detector plates and one is on the cold plate. The temperature of the cold plate is controlled using a heating mat underneath it which is in turn controlled by a PID controller. This allows the temperature of cold plate to maintained within 1C.

The bias and current monitoring are made using a Keithley 6517A with a 10 channel scanner card. During irradiation the detectors are maintained at 100V, and at the end of a run and I-V is taken. The temperatures are monitored using an HP 34970A data acquisition and switch unit. The K6517A and the HP34790A are controlled the data recorded via GPIB using a Labview control program.

In the CERN PS the magnet supercycle is such that the detectors are irradiated with a 0.5s every few seconds, depending on the number of spill in the 14s supercycle. The currents are monitored during the periods between spills. The PSI pion beam has an essentially continuous structure, 40ns bunches separated by 1ns. Unlike the PS, the dark current cannot be measured when the beam is on. Therefore, to measure the dark current the beam was switched off and an I-V curve measured for a range of fluences.

Preliminary results

The plot above shows the current evolving with fluence for a bias of 100V, and the temperature when the current has been measured. The temperature varies over only ~1.5C but this must be corrected for before determining α and β.

The plot above shows the dark current evolution with fluence after the currents have been corrected to -10C.

A summary plot of the data for two of the detectors is shown above. (a) Shows the current evolution with fluence in the low fluence region, where the current has been corrected to-10C. (b) shows the current evolution with fluence where the currents have been corrected to +20C. (c) is a log-log plot of the low fluence region, and demonstrates that the current increases linearly with fluence (power law=0.99). (d) is a log-log plot of the high fluence region and gives a power law 0.40. This is somewhat less than the expected square root behaviour. The current has been fitted to a square root fluence dependence to derive β. The results are summarised below:

  Detector 1 Detector 2
Alpha (-10C) 7.0x10-18Acm-1  7.0x10-18Acm-1
Alpha (+20) 1.1x10-18Acm-1  1.2x10-18Acm-1
low fluence power law 0.99 0.99
High fluence power law 0.39 0.39
Beta (-10C) 0.05cm-1  

The values are compatible with the results in the NIM paper, giving a relative damage factor for pions at 191MeV and protons with T=23GeV of 2.50.4, compared to a prediction of 1.8 from NIEL curves (note the NEIL curves for protons go to T=10GeV). The error arises largely due to the uncertainty in the fluence.


We are very grateful to K. Gabathuler and the staff of Paul Scherrer Institute (PSI), Viilligen, Switzerland for providing the pion irradiation facility and support during our stay. We also  wish to acknowledge Maurice Glaser (CERN) for providing infrastructure at PSI, helping us to setup and for determining the fluence. Tim Jones, Dave Muskat and Mike Wormald from Liverpool University, and Shaun Roe at CERN helped to prepare the samples. This work  has been funded by the UK Particle Physics and Astronomy Research Council.


Back to top of page

Back To Home Page

Page maintained by Craig Buttar
Last modified 06/01/2003