It was demonstrated that, during the formation of Pd2Si, point defects were injected into the Si substrate and led to an appreciable degree of dopant diffusion at temperatures as low as 200C. This was the lowest temperature at which dopant diffusion had ever been observed. Buried marker layers exhibited asymmetrical diffusion, with migration occurring preferentially towards the silicide-forming interface. The results showed that a large gradient in point defect concentration did not develop across the width of a marker layer. This showed that it was not possible for a large point defect gradient to be created by trapping as they diffused through the doped layers.

M.Wittmer, P.Fahey, G.J.Scilla, S.S.Iyer, M.Tejwani: Physical Review Letters, 1991, 66[5], 632-5