Photocurrent measurements in chlorine‐doped amorphous selenium

Abstract

Transient photocurrent (TPC) measurements have been carried out on evaporated a‐Se films doped with either 12.5 or 67 ppm of Cl and compared with analogous measurements on undoped a‐Se samples. The two doping levels lead to strikingly different resulting current traces. While the 12.5 ppm doping level weakens, but does not eliminate the signature of the two hole traps that are observed in the TPC signals of undoped a‐Se, no indication for those levels remains in the 67 ppm Cl‐doped samples. It is argued that suppression of the trap generated by the negative‐U D− centre agrees with the notion that they are being supplanted by doping‐induced Cl− centres. The enhanced density of D+ centres and accompanying cross‐linking of Se chains by those three‐fold coordinated sites leads in turn to the weakening and disappearance of the shallower hole trap ascribed to dihydral angle variations. Results from constant‐photocurrent and time‐of‐flight (TOF) experiments support this interpretation of the TPC observations.

Effect of As alloying on the valence band tail states in a‐Se

“…The absence of that type of defect in the As-doped samples fits in with the fact that the As cross-links the Se chains, thereby making them shorter and making dihedral angle variations less probable. The same effect was in fact observed with the more highly Cldoped samples where the increased density of cross-linking Se 3+ centers similarly shortened the Se chains and suppressed the dihedral-angle defect [13]. At the same time, the cross-linking makes the material more rigid, which in turn may be the reason for the steeper band tail energy distribution.…”

“…The above numerical values of E* will, therefore, underestimate the actual depth of lar dip in the TPC curves at earlier times also deserves to be clarified. Indeed, in pure a-Se [11], and in 12.5 ppm Cldoped a-Se [13], there is a dip around 10 -6 s, that is the signature of a shallow neutral defect linked to the dihedral angle variations in Se chains. The absence of that type of defect in the As-doped samples fits in with the fact that the As cross-links the Se chains, thereby making them shorter and making dihedral angle variations less probable.…”

“…While these latter ones have been shown to correspond to the thermal transitions involving the a-Se negative-U centers [8], the former ones can be linked to local distortions of the dihedral bonding angle along a Se chain [12]. In a recent study [13] it was seen that a-Se doping with 12.5 ppm of Cl did not disturb that defect structure, but that 67 ppm of Cl did reduce the density of the shallow centers below observability. In this work we will explore the changes that As doping introduces in the a-Se DOS and try to arrive at a reasoned explanation for the observed hole lifetime decrease.…”

Frequency- and time-resolved photocurrents in vacuum-deposited stabilised a-Se films: the role of valence alternation defects

“…In this work, we suggest that VAP defects, and particularly, the interconversion between the two stable VAP configurations can explain our experimental data without invoking any stable populations of the various neutral overcoordinated and hypervalent defects that have been proposed to exist [31]. Previous VAP-based interpretations of various electrical and optical properties of a-Se such as photoconductivity, post-transit time-of-flight (TOF) signals, photodarkening and interrupted field TOF have demonstrated the usefulness of VAP defect models [27,29,30,[32][33][34][35][36][37][38][39]. In fact a VAP defect-based model has been successfully used to explain the "ghosting" and the drop in sensitivity of a-Se photoconductors [40].…”