“…Thus, exposure of metal surfaces to ambient conditions before ALD TB growth implies that a native oxide IL forms before nucleation of the ALD TB. ,,,,− The IL has a significant impact on the quality of the ALD TB growth on top of it because it is typically defective with oxygen vacancies and pinholes that leads to the defective growth continue to ALD TBs, which explains the significant increase in leakage with ex situ ALD TBs at thickness below 2–5 nm. − However, the control over the formation of a native oxide IL, which can even form in an in situ ALD process, has been optimized in our previous work. , The defective ALD TBs obtained in ex situ ALD processes are illustrated in the low E b together with soft dielectric breakdown. In the in situ ALD processes, the IL can be reduced to a negligible level ∼ 0.1–0.2 Å, which not only enables sub-nm think pinhole-free ALD TBs to be achieved but also reduces the defect concentration in ALD-Al 2 O 3 TBs, as illustrated in the E b of 1.31–1.47 eV that is significantly higher than the previously reported E b ∼ 0.3–0.6 eV for AlO x TBs. ,, This work therefore illustrates the critical importance in controlling the FM/ALD TB interface and the potential of the in situ ALD process for the fabrication of ultrathin, high-quality TBs for MTJs. − , Finally, it is worth mentioning that the ALD-Al 2 O 3 TBs are highly uniform, as reflected in the negligible TB thickness dependence of E b values. − This is in contrast to the strong TB thickness dependence of the TB height in the AlO x TBs obtained via oxygen diffusion into Al or Al alloys …”