Abstract

With the aim of developing a terabit integrated non-volatile memory cell, many researchers have intensively studied perpendicular spin-transfer torque magnetic random access memory (p-STT MRAM) cells because of their fast write time (~10 ns), non-volatile memory operation, extremely low power consumption, and no scaling-down limit of less than 20 nm. A p-STT MRAM cell consists of a selective device and a perpendicular magnetic tunnel junction (p-MTJ) spin valve. In particular, CFB (Co2Fe6B2 or Co4Fe4B2)/MgO based p-MTJ spin valves have been widely studied since they have demonstrated excellent device performance characteristics. In particular, Key issues for obtaining a higher TMR ratio in CFB/MgO-based p-MTJ spin-valves are (1) the interfacial-perpendicular magnetic anisotropy (i-PMA) between the free or pinned Co2Fe6B2 layer and the MgO tunneling barrier and (2) the tunneling barrier’s body centered-tetragonal (bct) crystallinity. In addition, since CFB/MgO-based p-MTJ spin valves should achieve an excellent device performance at the back-end-of-line temperature of ~350°C the tunneling barrier bct crystallinity becomes more and more critical at a higher ex-situ magnetic annealing temperature. Thus, in our study, in order to obtain the best crystallinity in an MgO tunneling barrier at an ex-situ magnetic annealing temperature greater than 350°C, we investigated how the radio-frequency (RF) sputtering power of a 0.65-1.15 nm thick MgO tunneling barrier affects the TMR ratio and the resistance-area product (RA) for p-MTJ spin valves by using a current-in-plane tunneling (CIPT) technique, and how it affects the tunneling barrier bct crystallinity by using cross-sectional transmission electron microscopy (x-TEM). Figure 1 shows the dependencies of the TMR ratio on the tunnelling barrier RF power for Co2Fe6B2-MgO based p-MTJ spin valves with [Co/Pt]n-SyAF layers. For RF power of 400 to 500 W, the TMR ratio was almost zero (≈5%) when the tunneling barrier thickness (tMgO ) was less than 0.7 nm. This indicates that the tunneling barrier’s crystalline structure would be amorphous. The TMR ratio rapidly increased and then increased approximately linearly as the tMgO was increased from 0.8 to 1.15 nm. It is particularly noteworthy that, lower tunneling barrier RF power led to higher TMR ratios; i.e., the highest TMR ratio sequence was obtained at 500, 400, and 300 W. This implies that lower RF power would lead to better tunneling bct crystallinity. In particular, At 300 W, the TMR ratio was 168% for tMgO = 1.15 nm. However, the TMR ratio for all tMgO at 250 W abruptly dropped to almost zero (≈1.2%). we investigated the dependency of the tunneling barrier bct crystallinity on the RF power by x-TEM. For the valve that was RF sputtered at 300 W, the tunneling barrier layer was well bct crystallized and its thickness was very uniformly distributed; in Figure 2(a), it is 1.05 nm. This indicates there is maximum Δ1 coherent tunneling of perpendicular spin-electrons between the free and pinned Co2Fe6B2 layers. However, for the valve that was RF sputtered at 250 W, the tunneling barrier layer had an almost completely amorphous structure with a locally crystallized poly grain. This indicates there is almost no Δ1 coherent tunneling of perpendicular spin-electrons between the free and pinned Co2Fe6B2 layers. A comparison of Figure 2(a) with Figure 2(b) indicates that the tunneling barrier bct crystallinity was abruptly degraded when the tunneling barrier RF power was decreased from 300 to 250 W. As a result, the TMR ratio of the p-MTJ spin valves abruptly decreased from 168 to 2%. In our presentation, we will review in detail the mechanism by which the RF sputtering power performs a good bct MgO crystallinity. In addition, we will discuss correlation of MgO crystallinity between x-TEM images and optical properties via spectroscopic ellipsometry to understand how the RF power affects the crystallinity of the MgO tunneling barrier. *This work was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) grant funded by the Korea government (No. 2014R1A2A1A01006474 & No. 1004608) and Brain Korea 21 PLUS Program in 2014. Figure 1

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.