Comments, with reply, on "Shallow-junction diode formation by implantation of arsenic and boron through TiSi/sub 2/ film and rapid thermal annealing" by L. Rubin et al

The commenters clarify several points that may lead readers to the conclusion that the proposal in the above-titled paper by L. Rubin et al. (ibid., vol.37, p.183-90, Jan. 1990) is a viable approach to making shallow junctions for VLSI application. It is argued that (1) the junctions leakage data in the paper are unacceptably high for VLSI application; (2) the junction depth cited is only that below the silicide and thus is misleading; (3) the implantation-through-silicide process leaves damage in the Si which cannot be annealed out without high temperature (the high-temperature anneal subsequently compromises the silicide integrity, and thus is not a viable process); and (4) the amount of dopant in the Si is a small percentage of the total dose and is thus very sensitive to small variations of the TiSl/sub 2/ thickness, and therefore imposes a severe manufacturability question. In reply, N. Herbots and L. Rubin state that they never claimed to present a process that was ready for use in circuit manufacturing. Rather, they presented an investigation of a technique, and stated that it had the promise of being useful for manufacturing if it is further refined.<>