“…Despite several decades of intensive investigation there is still no complete consensus as to the origin of exchange bias [1,2], or the large body of phenomena that accompany the exchange-induced anisotropy, such as coercivity enhancement [1,2], training [1,2], and magnetization reversal asymmetry [4,5]. One issue that has been the subject of numerous investigations is that of the microstructure dependence of the exchange bias H E , particularly with regard to the AF layer [1,2,[6][7][8][9][10][11][12][13][14][15][16]. The interest in this issue arises because many exchange bias models predict a strong sensitivity to the existence of defects and disorder [6,[17][18][19][20][21], either at the AF/F interface or in the bulk of the AF layer.…”