Properties of HPK UFSD after neutron irradiation up to 6e15 n/cm2

Abstract

In this paper we report results from a neutron irradiation campaign of Ultra-Fast Silicon Detectors (UFSD) with fluences of 1e14, 3e14, 6e14, 1e15, 3e15 and 6e15 neq/cm 2 . The UFSD used in this study are circular 50 µm thick Low-Gain Avalanche Detectors (LGAD), with a 1.0 mm diameter active area. They have been produced by Hamamatsu Photonics (HPK), Japan, with pre-irradiation internal gain in the range 5-70 depending on the bias voltage. The sensors were tested pre-irradiation and post-irradiation with minimum ionizing particles (MIPs) from a 90 Sr β-source. The leakage current, internal gain and the timing resolution were measured as a function of bias voltage at -20 o C and -30 o C. The timing resolution of each device under test was extracted from the time difference with a second calibrated UFSD in coincidence, using the constant fraction discriminator (CFD) method for both. The dependence of the gain upon the irradiation fluence is consistent with the acceptor removal mechanism; the highest gain decreases from 70 before radiation to 2.6 after a fluence of 6e15 n/cm 2 . Consequently, the timing resolution was found to deteriorate from 20 ps to 50 ps. The results indicate that the most accurate time resolution is obtained varying with fluence the CFD value used to determine the time of arrival, from 0.1 for pre-irradiated sensors to 0.6 at the highest fluence. Key changes to the pulse shape induced by irradiation, i.e. (i) the contribution of charge multiplication not limited to the gain layer zone, (ii) the shortening of the rise time and (iii) the reduced pulse height, were compared with the WF2 simulation program and found to be in agreement.

Timing performance of small cell 3D silicon detectors

“…The central columns (n + )of two such test-structures 400 µm apart were connected together to the input of the amplifier, while the neighboring n + electrodes were grounded. The first stage of amplification uses fast transimpedance amplifier designed by UCSC [15] followed by a second amplifier which gives signals large enough to be relatively easily recorded by a 40 GS/s digitizing oscilloscope with 2.5 GHz bandwidth. The reference time required for measurement of the time resolution was provided from a non-irradiated LGAD detector produced by HPK [7,15].…”

“…The first stage of amplification uses fast transimpedance amplifier designed by UCSC [15] followed by a second amplifier which gives signals large enough to be relatively easily recorded by a 40 GS/s digitizing oscilloscope with 2.5 GHz bandwidth. The reference time required for measurement of the time resolution was provided from a non-irradiated LGAD detector produced by HPK [7,15]. It is 50 µm thick, has a diameter of 0.8 mm and high gain of ∼ 60 at 330 V and room temperature.…”

“…The time resolution of HPK-50D sensors was measured (in the same setup with a method described below) by using two such detectors and was determined to be 26 ps. More details on timing and charge collection measurements with these LGADs can be found in [7,15]. The electrons were collimated to an angle of < 1 • by a combination of collimator, small cell size and the circular opening in the PCB boards hosting the sensors.…”

Comparison of 35 and 50μm thin HPK UFSD after neutron irradiation up to 6 ⋅ 1015 neq/cm2

“…Results on irradiated CNM LGAD 300 µm, 75 µm and 45 µm thick sensors were presented in [12] [13]. Recently a radiation campaign with neutrons with 50 µm thick LGAD produced by Hamamatsu Photonics, Japan (HPK) has been reported [14]. In all cases the timing resolution has been shown to deteriorate with fluence due to the decreasing value of the gain.…”

“…In Section 2 we will describe the characteristics of 35 µm thick UFSD, followed in Section 3 by a short description of the irradiation facility. In Section 4, a short description of the experimental set-up is presented, details were previously reported in [6] [14]. In Section 5, we will describe the data analysis including the extraction of the gain and the time resolution, and in Section 6 the results on bias dependence of charge collection and gain, pulse characteristics and timing resolution for a range of fluences will be presented.…”