Effects of thermal neutrons on MSGCs
Introduction
MSGCs in the CMS central detector will be subjected to very high fluxes of neutrons and photons mainly due to albedo radiation from the surrounding calorimeters 1, 2.
Although neutrons themselves do not cause any ionization and are not directly seen in particle detectors, they can indirectly lead to an increase of the signal rate of gaseous detectors. An important mechanism, by which neutrons generate signals, is scattering from hydrogen, in which case the recoiling proton is recorded. Usually recoil protons will not exit solid materials, so this mechanism can give a signal only when it occurs in a gas. A more serious problem is the material activation. Neutron capture is usually followed by a gamma de-excitation. Although the neutron energy may have been less than 1 eV the resulting nuclear excitation can amount to several MeV. The emitted photon can convert to an e+e− pair, Compton scatter or produce a photoelectron in the detector.
Studies have been performed to estimate the neutron background 1, 3in the CMS detector. On the MSGC layers in the tracker, between r=50 cm and r=100 cm, the total neutron flux will be about and the low-energy neutron flux (E<100 keV) will be between 0.6 and . Fig. 1 shows the neutron kinetic energy spectrum on the r=80 cm tracker layer [3].
Section snippets
Experimental set-up
To investigate the effects of slow neutrons on MSGCs, 4 small detectors with aluminium strips ( active area) were exposed to a flux of neutrons in the CEA (Saclay) ISIS reactor facility.
Fig. 2 shows a schematical cross-section of the four tested chambers manufactured on different substrates:
- •
chamber 1: on a standard D263 glass
- •
chamber 2: on a diamond-like coating made by ICMC1
- •
chamber 3: on a coating made by SURMET2
Gain measurement
The gain of the MSGCs 1–3 was measured in the laboratory before their exposure to the neutron flux.
Immediately after the reactor was stopped, an Fe55 source was placed in front of the chamber 1 on a D263 glass. With cathode voltage equal to −600 V, the gain measured in the range of around 650 is compatible with the measurements shown in Fig. 4.
Comparative gain measurements have been made in the laboratory with the same experimental conditions as those before the irradiation. In order to measure
Induced activity by irradiation
For chamber 4 the γ spectrum (Fig. 5) was measured just after the ISIS reactor was stopped. Table 2 presents the main elements identified in the present analysis. We can compare this Table with Table 3 which shows the D263 glass composition measured by neutron activation analysis (NAA) and by Rutherford backscattering spectroscopy (RBS) [6].
Antimony, sodium and potassium are present in the glass. Detectors were mounted on a ceramic card. Gold is the main component of the conductive ink used for
Conclusion
In order to detect potential effects of induced activity, MSGCs made on different substrates were exposed to a neutron instantaneous flux close to the low-energy neutron instantaneous flux expected in CMS tracker.
The gain was measured before and after the exposures. These measurements show that MSGC performance will probably not be affected by the slow neutrons at LHC.
Measurements made on MSGC 4 show induced activity of chamber element components, especially substrate components, and of
Acknowledgements
We wish to express our gratitude to S. Barthe, A.M. Bergdolt, J. Coffin, H. Eberlé, J.M. Helleboid, M. Hoffer, M.R. Kapp, G. Schuster, M.H. Sigward and R. Wortmann for their technical support.
References (6)
- M. Huhtinen, P. Aarnio, HU-SEFT R 94-01,...
- M.J. Esten et al., Proc. of the Int. Workshop on Micro-Strip Gas Chambers, Lyon, 30 November–2 December...
- M. Huhtinen, CERN CMS TN/95-198,...