Spintronics Device for Stand-by Power Free Nonvolatile CMOS VLSI

Recent progress in perpendicular magnetic-easy axis magnetic tunnel junctions (MTJs), a spintronics device, offers a high potential building block for constructing not only stand-alone fast and nonvolatile RAMs in the 30 nm feature size and beyond but also nonvolatile CMOS VLSI employing logic-in-memory architecture [1]. The shift from in-plane to perpendicular is prompted by the need for a high crystalline anisotropy that is available in perpendicular materials for reducing the device size. In addition, current-induced switching is inherently more efficient with perpendicular easy axis. However, satisfying both high tunnel magnetoresistance (TMR) ratio over 100{\%} and low switching current was a challenge, because of the mismatch between MgO (100) - CoFe(B) bcc (100) structure needed to obtain high TMR and the crystal structure of perpendicular materials. It was shown that a strong perpendicular interface anisotropy exists at the MgO-CoFeB interface [2, 3], strong enough ($K_{i}$ = 1.3 mJ/m$^{2})$ to overcome demagnetization energy and make the easy axis perpendicular when the ferromagnetic electrode thickness is thin enough. First principle calculation by Nakamura \textit{et al.} showed that the perpendicular anisotropy is due to the oxygen-iron bond that reduces contribution of in-plane crystalline anisotropy [4]. By the use of this perpendicular easy axis, a 40 nm$\phi $ MgO-CoFeB MTJ with high TMR ($>$100 {\%}) and low switching current of 49 $\mu $A was realized [2]. It was also pointed out that activation volume for reversal plays an important role in determining the thermal stability of the MTJs [5]. I will discuss how the MTJs are incorporated in CMOS VLSIs to make them nonvolatile and stand-by power free. \\[4pt] [1] S. Ikeda, \textit{et al.} IEEE Trans. Electron Devices, \textbf{54}, 991, 2007. \\[0pt] [2] S. Ikeda, \textit{et al.} Nature Mat., \textbf{9}, 721, 2010. \\[0pt] [3] M. Endo, \textit{et al.} Appl. Phys. Lett., \textbf{96}, 212503, 2010. \\[0pt] [4] K. Nakamura \textit{et al}., Phys. Rev. B, 81, 220409(R), 2010 \\[0pt] [5] H. Sato, \textit{et al. }Appl. Phys. Lett. \textbf{99}, 042501, 2011.