Over 60?years ago, Buckminster Fuller, a visionary inventor and architect, introduced a revolutionary concept in structural engineering: the geodesic dome. It was the result of years of experimentation and search for a general shape that would exhibit high structural integrity and, at the same time, be light and affordable to build. The domes gained international visibility when they were used in large-scale projects, such as the American pavilion at Expo 67 in Montreal. These structures defined principles for structural integrity that can be applied on a large range of scales. Remarkably, with a radius below one nanometer, the structure of the C60 molecule is reminiscent of that of a geodesic dome and it was therefore named ‘buckminsterfullerene’.

Today, the advent of modern nanofabrication tools has opened exciting new possibilities for defining structures on scales ranging from hundreds of micrometers down to tens of nanometers. However, only recently could such small structures be realized in three dimensions – and Buckminster Fuller’s principles have again proven essential. Indeed, his ideas find direct applications in structures such as mechanical metamaterials, which combine macroscopic mechanical properties with bio-inspired architectures to create materials with unprecedented strength and properties [1].

Now, we add an additional degree of freedom to such materials: magnetism. In ferromagnetic nanostructures, the relationship between geometry and magnetic configuration is well understood. The main driver is the magnetostatic interaction, which can lead to the formation of complex domain patterns. These patterns are characterized by dynamics generated by their constitutive elements, such as uniformly magnetized domains walls, or vortex structures.

More recently, the study of three-dimensional nanostructures, mostly cylinders and nanotubes, has demonstrated that curvature could give rise to novel magnetochiral effects in the presence of external fields [2]. Such effects have so far been predicted by theory and by simulations and are promising for applications, in particular for data storage concepts, which rely on magnetic domain walls to store bits of information. Experimentally, such structures are only starting to be studied [3] and novel measurement techniques are currently being developed [4], [5], [6], [7].

Using patterned ferromagnets, it is also possible to exploit the magnetostatic interaction to create so-called artificial spin systems. These are a novel class of magnetic materials, which consist in lithographically defined nanomagnets displaying emergent behavior. Such materials have been studied in two dimensions, and it has recently been shown that their properties could be tailored to support and manipulate topological defects [8], [9] as well as to create metamaterials for spin waves [10], and even thermal ratchets [11].

A logical extension of this work is the realization of such materials in three dimensions: the use of the third dimension indeed allows to fully achieve the potential of artificial spin structures, by offering maximal configurability and the possibility of optimizing the magnetostatic interaction between the different elements of the system. It also allows tailoring novel dynamics characterized by the propagation of domain walls along curved structures in connected artificial spin systems, i.e. systems that are not only magnetostatically but also exchange coupled. Such systems require specific fabrication as well as measurement techniques, which are still being actively developed.

As a proof of concept for the fabrication of a three-dimensional artificial spin system, we have chosen a mesoscopic buckyball. In this structure, the vertices correspond to the location of the carbon atoms in the C60 molecule and the solid bars connecting the vertices correspond to the molecular bonds. The structure allows investigating the emergent magnetic properties of such systems, while the connected network of bars can also define paths for domain wall propagation that could conceivably be used to perform logical operations [12], [13]. However, the fabrication of such magnetic architectures with nanoscale features remains challenging.

Over the past decade, multi-photon laser polymerization has emerged as a high-resolution free-form three-dimensional structuring technique. The technique provides great flexibility in envisioning and designing structures for various applications and so far has been widely used to produce metamaterials, photonic crystals, and scaffolds for tissue engineering [14]. Commercially available three-dimensional lithography systems reach sub-micrometer resolution in the plane perpendicular to the optical axis of the writing beam. However, the resolution along the axis is typically two to three times lower due to a weaker confinement of light along this direction and, as a result, elongated voxels are produced. In three-dimensional laser lithography, high-resolution buckyball structures are typically produced by properly combining single-voxel lines. However, when writing small structures, oval rather than round bar cross-sections are obtained, resulting in difficulties to disentangle various magnetic phenomena occurring in the structure. One solution to mitigate the initial voxel elongation is composing the bars of a few overlapping lines, but this increases the overall dimensions of the structure. A way around it is to scale down polymerized 3D structures by pyrolysis and plasma etching [15].

We implement three-dimensional laser lithography and resist post-processing to develop three-dimensional mesoscopic magnetic structures. Polymerized scaffolds, where voxel asymmetry is compensated by writing larger structures, are pyrolyzed and treated in oxygen plasma to scale them down to sizes relevant for magnetic investigations. However, the used resists are typically nonmagnetic, and post-processed scaffolds need to be further altered to endow them with magnetic properties. We have achieved this by depositing a uniform conductive film of iridium using atomic layer deposition followed by electroplating of a magnetic material – in this case, nickel. The result is a mesoscale 3D structure composed of magnetic nanotubes. One of these magnetic buckyballs is featured on the cover of this issue of Materials Today. The image shows a scanning electron micrograph of a 5-µm diameter Buckyball composed of nickel tubes, around 300?nm in diameter, and having sidewall thickness of ca. 30?nm. Further studies of these architectures are expected to give insights into three-dimensional nanomagnetism and provide a basis for the design of novel magnetic microdevices.

See full text for more information.

Architectural structures open new dimensions in magnetism Magnetic buckyballs


GS and CD acknowledge funding from the EU-H2020 Research and Innovation program under Grant Agreement No. 654360 NFFA-Europe. SG acknowledges funding by the European Union's Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement No. 708674. SG is grateful to Riccardo Hertel, Claire Donnelly and Laura Heyderman for fruitful discussions and collaborations.

Recommended articlesCiting articles (0)


[1]L.R. Meza, et al.

Proc. Natl. Acad Sci. U.S.A., 112 (2015), pp. 11502-11507

CrossRefView Record in Scopus

[2]R. Hertel

SPIN, 3 (2013), p. 1340009

CrossRefView Record in Scopus

[3]A. Fernández-Pacheco

Nat. Commun. (2017), p. 15756

View Record in Scopus

[4]C. Phatak, et al.

Nano Lett., 14 (2014), pp. 759-764

CrossRefView Record in Scopus

[5]C. Donnelly, et al.

Phys. Rev. Lett., 114 (2015), Article 115501


[6]R. Streubel, et al.

Nat. Commun., 6 (2015), p. 7612

[7]C. Donnelly, et al.

Nature, 547 (2017), pp. 328-331

CrossRefView Record in Scopus

[8]E. Mengotti, et al.

Nature Physics, 7 (2011), pp. 68-74

CrossRefView Record in Scopus

[9]C. NisoliThe Role of Topology in Materials

Springer (2018), pp. 85-112

CrossRefView Record in ScopusGoogle Scholar

[10]E. Iacocca, et al.

Phys. Rev. B, 93 (2016), Article 134420


[11]S. Gliga, et al.

Nat. Mater., 16 (2017), p. 1106

CrossRefView Record in Scopus

[12]K. Zeissler, et al.

Sci. Rep., 3 (2013), p. 1252

[13]A. Pushp, et al.

Nat. Phys., 9 (2013), pp. 505-511

CrossRefView Record in Scopus

[14]M. Malinauskas, et al.

Light Sci. Appl., 5 (2016), p. e16133


[15]G. Seniutinas, et al.

Microelectron. Eng., 191 (2018), pp. 25-31

ArticleDownload PDFView Record in Scopus

Read full text on ScienceDirect

DOI: 10.1016/j.mattod.2019.05.001