SciPost Submission Page

Integrating micromagnets and hybrid nanowires for topological quantum computing

by Malcolm J. A. Jardine, John P. T. Stenger, Yifan Jiang, Eline J. de Jong, Wenbo Wang, Ania C. Bleszynski Jayich, Sergey M. Frolov

Submission summary

As Contributors: Sergey Frolov · Malcolm Jardine
Arxiv Link: (pdf)
Code repository:
Date submitted: 2021-04-14 20:08
Submitted by: Frolov, Sergey
Submitted to: SciPost Physics
Academic field: Physics
  • Condensed Matter Physics - Experiment
  • Condensed Matter Physics - Computational
Approaches: Experimental, Computational


Majorana zero modes are expected to arise in semiconductor-superconductor hybrid systems, with potential topological quantum computing applications. One limitation of this approach is the need for a relatively high external magnetic field that should also change direction at nanoscale. This proposal considers devices that incorporate micromagnets to address this challenge. We perform numerical simulations of stray magnetic fields from different micromagnet configurations, which are then used to solve for Majorana wavefunctions. Several devices are proposed, starting with the basic four-magnet design to align magnetic field with the nanowire and scaling up to nanowire T-junctions. The feasibility of the approach is assessed by performing magnetic imaging of prototype patterns.

Current status:
Editor-in-charge assigned

Submission & Refereeing History

Reports on this Submission

Anonymous Report 2 on 2021-5-18 (Invited Report)


The authors propose and study a new way to induce locally a magnetic field using micromagnets in order to allow the emergence of Majorana zero modes in a single nanowire as well as in more complex devices such as T-junctions. Numerical stray field simulations are presented along with numerical simulations of Majorana wavefunctions and energy spectrum at one dimension. First experimental realization of an inhomogeneous stray field required for a T-junction and produced by micromagnets is shown, pointing both the feasibility and the limitations of the proposal.

The original idea presented in this work constitutes an interesting and new experimental technique that should be considered in the race for performing topological quantum computing based on Majorana qubits. Overall, the results and analysis support the main conclusions and claims and the novelty of the approach presented here is of interest for the field of Majorana nanowires. I recommend it therefore for publication. Nevertheless, the paper suffers from unclarities. More precisely, the following issues should be addressed in a revised version prior to any publication.

Requested changes

1) The advantage of the method is not clearly pointed out in the introduction. It is only shortly mentioned in the text and the reader has to make some effort to gather together the different advantages scattered all over the text. According to the present work, this method allows to avoid any external magnetic fields to induce MZM and to locally induce stray fields that can be oriented differently in the samples. Therefore, it opens the way to the measurement of complicated devices like T-junctions braiding devices. This would be also the case of nanowire being covered by an insulating ferromagnetic layer. Nevertheless, the authors implied in the text that the stray field induced in the nanowire with the present method should much more homogeneous. Another advantage might the greater freedom offered to address and/or manipulate individually the MZM.

Some issues are not mentioned or are insufficiently discussed in the present version of the manuscript:
2) In order to generate a MZM, the nanowire needs to be covered by a superconducting thin film. Due to the Meissner effect, such a superconducting layer will locally influence the configuration of the stray field.
3) The effect of the presence of a disorder that is definitively present in any nanowire should be discussed and how the MZM are sensitive to such a static disorder.
4) The authors assume for their numerical simulations that the stray field is constant over the all cross section of the nanowire. Some arguments to support this claim are missing in the text. How homogeneous is the field in the y-direction? How would be the emergence of MZM sensitive to any inhomogeneous stray field in the y direction?
5) It looks like the experimental realization of the micromagnets have a different aspect ratio than the one considered in numerical simulations. How will the aspect ratio impact the configuration of the stray field?
6) One can notice some tiny asymmetry between right and left (red and green) in the double Dragonfly as well as in the T-junction when the problem is supposed to be mirror symmetric. This can be seen in the wavefunction as well as in the energy levels of the system. Is there any explanation for this asymmetry?
7) In the double Dragonfly configuration, not only the total length is responsible for the reduction of the overlap of the MZM but also the decay of the wavefunction that appears to be stronger by comparing figures 3.a and 4.a.
Some sentences are not clear and should be rephrased or developed:
8) “Nevertheless, left and right MZM form a pair due to field rotation provided by the central magnet.” A non-expert reader might appreciate to have more details there.
9) The term “activated” is confusing. You mean that the wire is “pinched off”.

Some less important remarks:
10) There is no reference to S4, S6, S9 in the main text and S7 is only mentioned in the supplementary informations.
11) You mention “Twenty-four T-junctions with three Dragonflies each are imaged, and six of 76 total Dragonflies”: it corresponds to 72 configurations (24 x 3), doesn’t it?
12) There is some color mistake in the figure 4 as well as in figures S1, S2 and S3: the small magnet below the NW has the wrong color (blue <-> red) whereas the arrow indicates the right magnetization.

  • validity: good
  • significance: good
  • originality: high
  • clarity: ok
  • formatting: reasonable
  • grammar: -

Anonymous Report 1 on 2021-5-18 (Invited Report)


The authors propose an original architecture that attempts to meet both basic requirements (for realizing Majorana zero modes) and practical requirements (for setting up the micromagnets).
The article contains both numerical simulations and an experimental proof-of-concept.
The source code and scripts of the simulations and plots are publicly available in a git repository.


While being an original proposal, I am not convinced that the proposal is particularly promising for realizing a scabable Majorana architecture. The initialization of the micromagnets in the proof-of-concept is not yet completely successful and the nanowires still have to be incorporated, as well as additional hardware for initialization, manipulation, and readout (e.g., gates).
The presentation of the results could be improved.


The authors propose an original micromagnet architecture (with a so-called Dragonfly as building block) to realize scalable networks for Majorana zero modes without requirement of a (strong) external magnetic field. While the proposed architecture and experimental proof-of-concept is not fully convincing, I think that this work is worth publishing in SciPost Physics and could inspire further research in this direction of combining Majorana nanowire in network structures with micromagnets or, more generally, nonuniform magnetization profiles. I do find that the presentation of the results are hard to follow in some places and the manuscript could definitely benefit from some improvements, so my verdict is minor revision. The problems with the presentation are listed in detail below, in the section for requested changes.

Requested changes

I would like to see a statement on the validity of the 1D model with magnetization profiled averaged in the transverse direction. Based on Fig. 1, it seems that there could be a strong variation in the transverse direction, especially where the micromagnet ends are close to the nanowire, and this might invalidate the results obtained with the 1D model to some extent. In this context, it would also be interesting to see the vector field with a bit more detail inside and near the nanowire region (a zoom-in of Fig. 1 with nanowire region visible).

The method section on the micromagnetic simulations in the main text is very brief. Some basic information on the simulation approach and the parametrization of the micromagents could be added there. Or, at the very least, a reference to the relevant supplementary section must be added.

The position of the nanowire and extent of the topological region should be clearly indicated on Figs. 2-4, S1, S3-S5, S7. And another suggestion for improvement: replace Fig. 2 or 3b by a vector field (now they are exact copies). In general, I think that a vector field is much easier to read than plotting the profile $\theta(x)$. Also, the introduction of $\theta$ in Fig. 2 is not the most intuitive when considering the reference frame depicted in Fig. 1.

3) Is there a particular reason why the T-junction setup does not contain a double Dragonfly in the bottom leg? I would expect that it would further improve the Majorana polarization. I would like to see some statement of this design choice.

It would be much easier to parse the list of energies in Fig. 6 when they are presented in a table and when the energies of the red, green, and blue colored states in the subfigures are indicated on an axis. (minor comment: it is not clearly stated that these values are for pairs of states)

On Eq. (4) in the Supplementary Information: $\epsilon_0$ is not defined and the momentum operator $\mathbf{p}$ should not appear anymore.

The color coding of one of the micromagnets on the bottom right of the double Dragonfly (on Figs. 2, S1, S2, S3) is incorrect, and also of one of the wave functions on Fig. 6d (purple instead of red).

7) Some figures in the Supplementary are not referred to in the text and some don't appear in the right order.

-'Schroedinger' on p. 1
- 'micromagentic' on p. 2
- 'mircomagnets' on p. 4
- 'cite i' (-> site $i$) on p. 7
- 'disretized' on p. 7
- 'Shrödinger' on p. 4,7
- 'exited' (-> excited) on p. 9
- 'x position' (-> y position) on Fig. S5b

  • validity: good
  • significance: ok
  • originality: high
  • clarity: ok
  • formatting: good
  • grammar: -

Login to report or comment