Voltage control of nanoscale magnetism and magnetic skyrmions : application to energy efficient computing devices

We have recently demonstrated strain and acoustic wave induced switching of magnetization via the Villari effect in magnetostrictive Cobalt and Nickel nanomagnets [1-4] and magnetostrictive soft layer of a magnetic tunnel junction [5]. However, one common issue observed in all experiments involving magnetostrictive elements with lateral dimensions ~100-300 nm is the lack of reliability with which the magnetization is switched.This is due to acombination of many factors: low magneto mechanical coupling in Cobalt and Nickel, thermal noise at room temperature and defects that pin the magnetization. This problem can potentially get worse when one scales to lateral dimensions less than 50 nm to compete with Spin Transfer Torque Random Access Memory (STT-RAM).This talk will discuss the above experimental work with complementary modeling that accounts for defects and thermal noise and recent work on polarized neutron reflectometry to understand depth dependent magnetization rotation due to differential strain transfer [6] to access the potential extent to which such strain switched nanomagnetic devices could scale to < 50nm lateral dimensions.

Furthermore, recent work shows thatdirect voltage control of magnetic anisotropy (VCMA) in conjunction with magnetic skyrmion states [7-9] may offer a robust mechanism for switching nanomagnets. For example, inclusion of Dzyaloshinskii-Moriya Interaction (DMI) can lead to an intermediate skyrmion state during the reversal of a perpendicular-MTJ from the ferromagnetic “up” to “down” state. Recent simulations show that forcing such reversal through a specific (skyrmion) state could be more robust to both thermal noise and defects than precessional VCMA switching schemes will also be discussed [9]. Such a skyrmion’s core also oscillates resonantly with the applied voltage induced change in anisotropy, which can be exploited for implementing unique neuromorphic functionalities such as a “resonate and fire neuron” in an energy efficient manner [10]. Preliminary experiments in this direction will also be presented.

Acknowledgements:
* The strain control of magnetism research performed in J. Atulasimha’s group at VCU (N. D'Souza, M. Salehi Fashami, M. M. Al-Rashid) in collaboration with S. Bandyopadhyay group at VCU (A. Biswas, H. Ahmad) G. P. Carman group at UCLA, J.P. Wang group at U. of Minnesota and B. Kirby and D. B. Gopman at NIST.
Simulations on VCMA switching of skyrmions/skyrmion mediated switchingwas performed in Atulasimha’s group (D. Bhattacharya, M. M. Al-Rashid, M. A. Azam) and experiments performed in collaboration with K. L. Wang group at UCLA. 
Grants that supported work at VCU: NSF grants NEB2020: ECCS-1124714, CAREER: CCF-1253370, SHF: CCF-1216614, ECCS 1609303; SRC under NRI task 2203.001; VCU Quest Commercialization Grant and Virginia Microelectronics Seed Grant.

REFERENCES: 
[1] N. D'Souza, M. Salehi Fashami, S. Bandyopadhyay and J. Atulasimha, Nano Letters, 16, 1069, 2016.
[2] V. Sampath, N. D’Souza, D. Bhattacharya, G. M. Atkinson, S. Bandyopadhyay, and J. Atulasimha,
Nano Letters, 16, 5681, 2016.
[3] M. Salehi-Fashami, M. Al-Rashid, W.Y. Sun, P. Nordeen, S. Bandyopadhyay, A.C. Chavez, G.P. Carman, J. 
Atulasimha,Nanotechnology (letter), 27, 43LT01, 2016.
[4] A. Biswas, H. Ahmad, J. Atulasimha, S. Bandyopadhyay, Nano Letters, 17 (6), 3478, 2017.
[5] Z. Zhao, et al., Appl. Phys. Lett., 109, 102403, 2016.
[6] M. M. Al-Rashid, A. Grutter, B. Kirby, J. Atulasimha, https://arxiv.org/abs/1801.08224
[7] D. Bhattacharya, M. M. Al-Rashid, J. Atulasimha, Scientific Reports, 6, 31272, 2016.
[8] D. Bhattacharya, M. M. Al-Rashid, J. Atulasimha,Nanotechnology, 28, 425201 2017
[9] D. Bhattacharya, J. Atulasimha,ACS Appl. Mater. Interfaces, 10 (20), 17455, 2018.
[10] M. A. Azam, D. Bhattacharya, D. Querlioz, J. Atulasimha https://arxiv.org/abs/1806.00076