Equal-double junctions in 24 GeV/ c proton-irradiated MCZ n- and p-type Si detectors: A systematic transient current technique investigation

Abstract

We undertook systematic transient current technique (TCT) studies, measuring the shapes of electron- and hole-transient currents in three sets of samples irradiated by 24 GeV/ c protons at fluences 1.6×10 14–2.4×10 15 p/cm 2. We carried out these measurements after leaving the samples to anneal for 22–23 days at room temperature. The three sets comprised (1) magnetic Czochralski (MCZ) n-type Si detectors; (2) MCZ p-type Si detectors; and (3) float-zone (FZ) n-type Si detectors (control set). The control set showed no surprises. The space charge sign inversion (SCSI) had already occurred at the lowest fluence (1.6×10 14 p/cm 2), and the double junction/double peak effect was readily apparent, with the first junction, the minor one, near the p + contact, which changes very little with bias voltages. It is superseded by the second junction near the n + contact (negative space charge) at biases higher than the full-depletion voltage. For both MCZ n-type and p-type detectors, the double junction/double peak effect also was initiated at the lowest fluence, but the standard SCSI evident in FZ n-type detectors (wherein the negative space charge dominates the entire detector) was not seen in that fluence range. However, in these two groups, the double junction/peak effect persisted into subsequent higher fluences with almost equal junctions near the p + and n + contacts, regardless of bias voltages, which may be much larger than the full-depletion voltages. This new effect, termed the equal-double-junction effect, is unique for the 24 GeV/ c proton-irradiated MCZ (n and p) Si detectors. It is evident by the almost identical shapes in TCT currents, before trapping corrections, for both electrons (red laser on the p + contact) and holes (on the n + contact), with the first peak always dominating a small second peak at any bias voltages. After trapping corrections, the heights of the two peaks are about the same, suggesting the existence of nearly equal-double junctions in the detector.