Over its rather long history, focused electron beam induced deposition (FEBID) has mostly been used as an auxiliary process in passivating surfaces in sample preparation for transmission electron microscopy. This has changed over the last one and a half decades. On the one hand, FEBID has been established as the leading technical approach to lithography mask repair on the industrial scale. On the other hand, FEBID-related technical and methodological developments, FEBID-derived materials, and FEBID-based device fabrication have had a significant impact in various areas of basic and applied research, such as nanomagnetism and superconductivity, plasmonics, and sensing. Despite this dynamic development, the FEBID user base does still form a rather exclusive club of enthusiasts. In this Perspective, our aim is to provide sufficient insight into the basics of FEBID, its potential, as well as its challenges, to scientists working in the broader fields of materials science, nanotechnology, and device development. It is our hope to spark growing interest and even excitement into FEBID which, as we believe, still has to live up to its full potential.

Focused electron beam induced deposition, FEBID, is a unique and highly flexible direct-write nanofabrication approach. It is based on the electron-induced dissociation of a previously adsorbed precursor gas in the focus of an electron beam typically provided by a scanning electron microscope (SEM). Considering the world-wide installation base of SEMs, many thousands of these instruments could, in principle, be upgraded to become FEBID machines at a relatively small cost by adding a gas injection system (GIS). Why are so few groups are using FEBID and even fewer groups are actively working on advancing the field? There are certainly many reasons for this and we venture to collect some of them here: (i) too few scientists know about FEBID in sufficient depth to appreciate its potential; (ii) those scientists who do know about FEBID are not aware of recent developments and dismiss it as a technology that can only provide impurity-rich material and its use is thus limited to fabricate protection layers for sample preparation in cross section transmission electron microscopy (TEM); (iii) FEBID materials cannot compete property-wise with nanostructured materials obtained from conventional thin film routes and lithography; (iv) FEBID is too complicated; just consider the multitude of different possible dissociation channels at play during electron irradiation of an adsorbed precursor layer; (v) although some FEBID materials may be interesting for applications, the process itself is too slow and thus not scalable toward higher throughput fabrication; (vi) although FEBID might be useful in some application fields, it cannot be considered as enabling technology for new directions in research; (vii) FEBID literature is too specialized and thus the start in using the technique is hindered by the lack of introductory material.

In this Perspective, we aim to give an overview of, what we believe, are the most relevant developments in FEBID over the last couple of years which may help to provide the reader with a more informed perspective of the potential of this fascinating technology. We also go into some issues that still need to be addressed to make FEBID more accessible to non-specialists and widen its materials base for different application fields. We then proceed to describing our perspective on promising developments in FEBID before concluding this Perspective. In order to have a solid ground to start from, we begin with a very condensed description of the FEBID process itself.

FEBID is accomplished by the electron-induced dissociation of an adsorbed precursor that leads to fragments, part of which remain as a permanent deposit on the provided sample surface whereas the other fragments are sufficiently volatile to leave the surface to be pumped away by the vacuum system. To be specific, in most instances, metalorganic precursors are used and the targeted precursor fragment to be remaining on the surface is the metal.1,2 Several different dissociation channels are active in parallel, such as dissociative electron attachment (DEA) and dissociative ionization (DI) (see also below), all of which have characteristic and energy-dependent dissociation cross sections σi(E).3 Depending on their bond strengths and the interaction of the different ligands with the surface, the formed deposits most often do also contain ligands or smaller fragments thereof in conjunction with the metal.4 The electronic properties can, therefore, vary from fully metallic toward insulating depending on the overall deposit composition and its microstructure.5 It should be noted that simple structures such as lines, areas, and nanopillars can be easily deposited without a complete understanding of the process and their writing can be controlled by choosing the deposition conditions using the SEM software and GIS control [Fig. 1(a)]. The actual resolution of FEBID depends on the experimental parameters related to the electron beam settings, precursor fragmentation characteristics, and substrate used for deposition. FEBID in a SEM is regularly reported to give access to feature sizes with lateral resolution below 10 nm.6 Even sub-nm control has been demonstrated experimentally using a transmission microscope for FEBID, showing point deposits of an average width at half maximum of 1.0 nm on a thin membrane.7 For further details related to feature sizes1,8 and deposit compositions,4 we refer to papers specifically focussing on these topics.

FIG. 1.

(a) The first step in FEBID is defining the target structure, e.g., by using a CAD program. (b) FEBID is done within a SEM with beam control done via a pattern file generated from the previously defined target structure. The pattern file consists of beam position commands for which writing layer, as indicated in (a) with the letter “n.” The numbering in the schematics refers to the different elementary processes associated with FEBID, such as 2: diffusion and 3′: desorption. (c) Monte Carlo simulation result shown for ten exemplary electron trajectories caused by scattering within a bow-like 3D target structure. The energy loss experienced by the primary electron is indicated by the color bar. Inelastic scattering events near the surface of the structure can result in secondary electron emission, which is most important for precursor dissociation. (d) Interrelationship between reactive force field calculations and combining solutions of the diffusion reaction equation with Monte Carlo simulations to simulate FEBID growth; see the text for details.

FIG. 1.

(a) The first step in FEBID is defining the target structure, e.g., by using a CAD program. (b) FEBID is done within a SEM with beam control done via a pattern file generated from the previously defined target structure. The pattern file consists of beam position commands for which writing layer, as indicated in (a) with the letter “n.” The numbering in the schematics refers to the different elementary processes associated with FEBID, such as 2: diffusion and 3′: desorption. (c) Monte Carlo simulation result shown for ten exemplary electron trajectories caused by scattering within a bow-like 3D target structure. The energy loss experienced by the primary electron is indicated by the color bar. Inelastic scattering events near the surface of the structure can result in secondary electron emission, which is most important for precursor dissociation. (d) Interrelationship between reactive force field calculations and combining solutions of the diffusion reaction equation with Monte Carlo simulations to simulate FEBID growth; see the text for details.

Close modal

More details regarding the fragmentation processes and the range of electronic properties exhibited by FEBID structures are discussed below. Here, we will cast a brief glance at the dynamics of the growth process, which will help to elucidate the multiscale nature in space and time associated with FEBID growth.

In Fig. 1(b), the essential components of a FEBID system are schematically depicted that include the focused electron beam, the precursor (gas) injector, and the sample surface. Into the focus of the electron beam represented by the electron flux density ϕe the gas injection system provides a precursor gas flux density. While this flux of gaseous precursor is typically provided by a capillary with a narrow aperture (diameter ∼ 0.5 mm) and has a directional component, additional diffuse precursor flux density stems from the continuous desorption and adsorption processes of precursor molecules going on in the surrounding area of the deposition zone.9 We subsume both these precursor flux density components into one flux density ϕg. For simplicity and given the fact that for the vast majority of precursors used in FEBID, the energy-dependent dissociation cross section is not accurately known, we use an effective and energy-integrated cross section σ in the following. We note that it is mainly the slow secondary electrons (energy < 50 eV) that are most important for the growth process. If the local area density of adsorbed precursor n(x,y;t) is known at any given time t, where (x,y) denotes the surface position, the local thickness change Δℎ can be calculated from the deposit volume Vd associated with the sticking dissociation fragments as follows (see, e.g., Ref. 10):

Δh=VdσϕenΔt,
(1)

during a short time interval Δt over which n(x,y;t) can be assumed to remain constant. By rastering the electron beam over the surface, keeping the beam stationary at selected dwell points (x,y) for a pre-defined dwell time tD, a deposit is formed whose shape (laterally and into the vertical dimension) is solely determined by the chosen beam pattern and the precursor dynamics. Assuming now a Langmuir adsorption model (but see Ref. 11 for going beyond Langmuir) for the precursor supplemented by diffusion and dissociation one is led to a surface reaction-diffusion equation of the following form (see, e.g., Ref. 10):

nt=D(2nx2+2ny2)-nτ-ϕeσn+sϕg(1-nn0).
(2)

Here, n0 denotes the area density associated with a fully filled monolayer of precursor and s is the precursor sticking coefficient [0,1]. The surface diffusion coefficient D and average precursor residence time τ both dependent on the temperature following an Arrhenius law.12,13 If all the model parameters governing this partial differential equation are known, it is possible to solve for n(x,y;t) with (x,y) specifying coordinates on the growing surface. Using this result, the local height Δh increases at time step Δt can be calculated from which the evolved surface is obtained and used again in determining the precursor density for the next time step solving again the reaction-diffusion equation.9 An additional complication arises from the fact that it is not simply the primary electron beam that determines ϕe but, in fact, the electron emission flux density caused by the secondary electrons excited by the primary beam at its time-dependent position in addition to those secondaries which are generated off the beam along the various scattering paths the primary and backscattered electrons can take. This emission profile depends itself on the growing shape of the deposit and its material composition, as well as on the shape and material composition of the substrate, as is exemplarily shown in Fig. 1(c). Accompanying Monte Carlo simulations can be used to calculate the emission profile and update it as the shape of the deposit's changes over time.9 Thus, it is possible to simulate FEBID growth quite accurately by numerically solving the discretized diffusion-reaction equation in conjunction with Monte Carlo simulations for determining the secondary electron flux distribution over the growing deposits. Also, the local energy dissipation distribution as calculated by the Monte Carlo simulation can be used to determine the time-dependent temperature distribution within the growing deposit by solving the heat conduction equation for the growing structure.14 This then impacts the local diffusion coefficients and average precursor residence times, both of which are thermally activated quantities.

It is instructive to have a closer look at the time scales associated with the different processes entering the diffusion-reaction equation and the heat conduction inside a growing structure. Figure 2 gives a schematic overview using model parameters found in the FEBID literature determined from several experiments, such as Refs. 8 and 15. The time multiscale nature of the FEBID simulation problem is quite apparent. FEBID growth simulation using the combined reaction-diffusion and Monte Carlo approach is suitable to simulate growth shape evolution up to sizes of several 100 nm to a few μm. Ideally, one would start from first principle calculations of the precursor dynamics including the energetically preferred geometrical arrangements of the precursor molecules on a given surface microstructure. This has been done in selected cases.16 Next, one would be interested in describing the precursor-specific electron-dissociation processes on the microscopic level. So far, this is only realistically possible on the basis of classical force field approaches complemented by a reactive component, i.e., reactive force-field calculations.17 These combined with Monte Carlo simulations are approaching the time scales associated with early stages of growth as described in the diffusion-reaction equation.18 It is, therefore, realistic to assume that the reaction-diffusion model parameters will become quantifiable by the reactive force-field approach applied to selected precursors. By this, a reliable and shape-realistic simulation of, in particular, three-dimensional (3D) FEBID growth is envisioned [see schematic in Fig. 1(d)]. Of course, all these model simulation approaches crucially depend on accompanying experimental studies for reference and calibration. Taken together, this will have a big impact on the development of software codes for the semi-automated generation of beam steering commands that result in shape-true 3D FEBID structures for various application fields in 3D nanomagnetism,19,20 3D nano-superconductivity, 3D plasmonics,21,22 3D nano-sensors,23 and other areas of research. The current state of the art of semi-automated beam steering command generation, given a desired 3D target structured, is briefly described in Sec. III A 1.

FIG. 2.

Schematic overview of time scale for model parameters used in FEBID simulations.

FIG. 2.

Schematic overview of time scale for model parameters used in FEBID simulations.

Close modal

As a direct-write nanofabrication approach, FEBID has several unique advantages, a selection of which is indicated in the center of the schematic shown in Fig. 3. It relies on technology and methodology from several fields (inner ring) and can be an enabling technology in various areas (outer ring). In this section, we will briefly review some recent promising developments but we will also point out some research areas that need particular attention for bringing FEBID up to its full potential.

FIG. 3.

Overview summarizing different topics of further progress based on unique features of the FEBID process enabled by technology and methodology from several fields.

FIG. 3.

Overview summarizing different topics of further progress based on unique features of the FEBID process enabled by technology and methodology from several fields.

Close modal

1. Development of 3D deposition optimization and process automatization

The transfer process from 3D shape definition toward actual beam control in a multi-parameter FEBID experiment [see Figs. 1(a) and 1(b)] is crucial from the operator point of view. Ideally, the complexity of the actual processes going on in FEBID can be hidden from the operator by software tools taking into account both growth calibration data and simulation-derived insights into the growth process. The long-term goals of FEBID process development, therefore, are to provide such software tools, which can be used to a large degree like automatic slicers known from, e.g., 3D printing of polymers. However, it needs to be stressed that 3D FEBID is intrinsically capable to yield multi-functional materials. In this regard, FEBID goes far beyond current 3D printing technologies, be it on the macro-, micro-, or nano-scale. On the downside, one has to appreciate that deposit composition and purity are still an issue and that the FEBID process itself is non-local in nature, which has to be taken into account by suitably setting the growth parameters and by how going about to do the slicing process. Slicing is the process of deriving beam position control settings from the locations where the actual target 3D shape is crossing a virtual plane perpendicular to the beam direction as it is moving from the bottom to the top of the target structure with a speed dictated by the local growth rate [see Fig. 1(a), slicing plane indicated with letter n]. In this slicing process, several aspects have to considered, such as, the reduced diffusive precursor replenishment with increasing height, the precursor consumption at the location of a beam dwell event and in its surroundings (proximity effects), and possible beam-induced heating.14 Presently, two semi-automated slicer software solutions are available in the public domain, which are mostly suitable for wireframe-like 3D FEBID structures.24,25 More recently, also curved 3D sheet-like structures have been addressed with explicitly taking beam-induced heating effects into account.14 Ongoing work is focusing on combining these slicing approaches into a next-generation software solution that is hoped to foster the broader distribution of 3D FEBID to non-specialists in a broad range of application fields. The complexity of nanoscale structures that can be written by the FEBID approach is illustrated in Fig. 4.

FIG. 4.

(a) Self-sensing 3D thermal nanoprobe, showing the electrode splitting together with the 3D PtCx tetrapod, which electrically bridges the electrodes; the partly colored close-ups illustrate the geometry of the tetrapods from side and top view. Adapted with permission from Sattelkow et al., ACS Appl. Mater. Interfaces 11, 22655 (2019). Copyright 2019 American Chemical Society. (b) and (c) Ferromagnetic cobalt Möbius strip with arrows to help visualize the morphology. Scale bars are 1 μm. Adapted with permission from Skoric et al., Nano Lett. 20, 184 (2020). Copyright 2020 American Chemical Society. (d) Complex 3D PtCx vase prepared by automated pattern file generation and the predefined geometry as model for the deposition. Adapted with permission from Keller and Huth, Beilstein J. Nanotechnol. 9, 2581 (2018). Copyright 2020 John Wiley and Sons.

FIG. 4.

(a) Self-sensing 3D thermal nanoprobe, showing the electrode splitting together with the 3D PtCx tetrapod, which electrically bridges the electrodes; the partly colored close-ups illustrate the geometry of the tetrapods from side and top view. Adapted with permission from Sattelkow et al., ACS Appl. Mater. Interfaces 11, 22655 (2019). Copyright 2019 American Chemical Society. (b) and (c) Ferromagnetic cobalt Möbius strip with arrows to help visualize the morphology. Scale bars are 1 μm. Adapted with permission from Skoric et al., Nano Lett. 20, 184 (2020). Copyright 2020 American Chemical Society. (d) Complex 3D PtCx vase prepared by automated pattern file generation and the predefined geometry as model for the deposition. Adapted with permission from Keller and Huth, Beilstein J. Nanotechnol. 9, 2581 (2018). Copyright 2020 John Wiley and Sons.

Close modal

2. Hybrid methods

Combination of gas-phase approaches can provide benefits when compared to the individual processes. For instance, site-selective deposition of specific nanomaterials or 3D scaffold preparation can be facilitated by FEBID could be combined with atomic layer deposition (ALD) of metal, which will give access to a high purity coating of the FEBID deposits.28 However, non-selective deposition during the ALD cycles has to be prevented. Currently, the control of nucleation during area-selective deposition is a general concern in the ALD process and enabled by (a) competitive adsorption of precursors on different surfaces or (b) atomic layer etching (ALE) cycles removing additional undesired nuclei on non-growth surfaces.29 Therefore, a potential necessity of combining the suggested FEBID/ALD hybrid approach with ALE or preferential co-adsorption of additional nucleation inhibitors must be considered.30 

The FEBID/ALD hybrid approach as suggested will enable the preparation of nanostructures by a direct-write application and subsequent coating with a conformal, highly pure, inorganic layer. The resulting structures will enable studies of geometry-dependent effects in plasmonics, magnetism, etc. Further benefits of this combined approach include (i) a faster feature size evolution due to the simultaneous growth of a multitude of FEBID seeds when compared to the pure direct-writing of a final feature size as well as (ii) shape retention of a written complex nanostructure, which is not fully preserved when a post-growth purification is applied to gain access to pure noble metal deposits.

3. Flexibility of FEBID regarding substrate material and shape

Typically, FEBID nanostructures are grown on bulk substrates such as Si, SiO2, TiO2, or Al2O3 and carbon or Si3N4 membranes are often used, e.g., for TEM structural investigations. However, FEBID techniques can be used to deposit nanostructures on almost any substrate material of any shape with surfaces accessible to the electron beam. In addition, limitations to be considered include charging effects and associated drift as well as deposit resolution. In the recent past, FEBID nanostructures have been prepared on various substrate types, such as flexible, transparent polycarbonates,31 as well as thin insulating layers of either self-assembled monolayers32 or metalorganic frameworks.33 Workarounds in terms of minimizing charging effects can be realized by working in proximity of metallic pads allowing the deposition of material with similar structural and physical properties on the insulator surface as is obtained on standard substrates.34 

Moreover, in situ fabrication of a hybrid structure consisting of a carbon membrane/FEBID deposit has been demonstrated recently.32 The carbon membrane is prepared by the electron-induced cross-linking of pre-made self-assembled monolayers. As the membrane can be separated from its support, this allows the transfer of the hybrid structure containing the FEBID nanostructure to literally any other surface. This strategy might pave the way to approaches using the transfer of functional nanoscale hybrid nanostructures to literally any other substrate. In addition, this example illustrates that different electron-induced reactions can be combined giving access to in situ prepared heterostructures. Similarly, but more applicable to large area preparation on non-charging surfaces, electrically conducting polymer films containing carbon nanotubes35 can be prepared or a transfer of large area graphene to any other substrate can be envisioned in order to prevent local charging effects and thus enabling high quality FEBID on any substrate material.

To date, the actual electron-induced precursor decomposition/fragmentation channels and parameter-dependent effects are still not fully understood since so far in situ monitoring of the dissociation processes in FEBID has not yet been accomplished.

The interplay of the different fragmentation paths/products and dependencies of deposit quality/composition on the experimental conditions illustrates the complex processes involved in the FEBID.4 Some understanding of the electron–molecule interactions leading to electron-induced fragmentation channels of precursors and potential intermediates is gained by combining information from single molecule dissociation in the gas phase,3 as well as surface-bound precursor condensates being exposed to electron flooding.15 In addition to these nowadays established methods, very interesting results have been gained employing nm-sized noble gas condensates with adsorbed precursors. This approach can be considered to be a hybrid of the aforementioned approaches allowing to identify specific polynuclear intermediates formed by intermolecular interactions.36 The observed bi-/polymetallic intermediates in these experiments are formed from the monometallic precursor by the simultaneous interplay of neutral dissociation of one precursor moiety and DEA to another aggregate constituent. Such effects are not observed in single collision experiments for the ligand cleavage in monomeric precursor species. Apparently, the cluster approach complements and bridges between surface science and single collision experiments.

Importantly, thermal contributions to an effective ligand stripping from metal-containing precursors can be generally considered being an integral part of the fragmentation processes during FEBID. The writing process is typically carried out on substrates at room temperature with potentially higher temperatures in a small volume of the substrate/deposit where the electron beam impacts the surface. In this respect, the impact of thermal effects for complete ligand stripping in the electron-induced decomposition of a heteronuclear Fe-Co-carbonyl37 and in the electron-supported carbon removal from FEBID-PtCx38 have been observed. The HFeCo3(CO)12 molecule liberates 4–5 carbonyl ligands by initial electron–molecule interaction and subsequent release of all remaining carbonyl ligands by temperature increase to 25 °C, while alternative treatment by further electron bombardment results in a C–O fragmentation associated with composite formation (Fig. 5).37 However, the process of thermal decomposition of intermediates is not universal and structural similarities and identical ligands are no guarantee for obtaining the same purity in the deposits. The structurally similar precursor H2FeRu3(CO)12 leads to metal contents of only ∼26 at. % in FEBID, which can be related to more thermally stable intermediates formed in an initial step of the decomposition cascade.39 These examples illustrate that it is of utmost importance to study the potentially dominating decomposition channels to explain the FEBID products composition. Simultaneously, a larger body of data on the understanding of electron–molecule interactions depending on the chemical nature of the precursor derivatives will support the quest to establish a more generalized precursor design strategy dedicated to this direct-write technique.

FIG. 5.

Schematic illustrating different stages of ligand cleavage in the fragmentation of HFeCo3(CO)12 and two subsequent effects leading to (i) a composite formation by further electron bombardment or (ii) complete ligand stripping by a thermally activated process step resulting in a pure FeCo3 alloy. Reprinted with permission from Ragesh Kumar et al., J. Phys. Chem. C 122, 2648 (2018). Copyright 2018 American Chemical Society.

FIG. 5.

Schematic illustrating different stages of ligand cleavage in the fragmentation of HFeCo3(CO)12 and two subsequent effects leading to (i) a composite formation by further electron bombardment or (ii) complete ligand stripping by a thermally activated process step resulting in a pure FeCo3 alloy. Reprinted with permission from Ragesh Kumar et al., J. Phys. Chem. C 122, 2648 (2018). Copyright 2018 American Chemical Society.

Close modal

However, the aforementioned experimental conditions in these studies are typically distinctively different from the actual FEBID process. It would be highly desirable to be able to gain more insight in situ and understand the chemistry on the surface during the deposition process. In situ investigations could include (i) ultrafast optical spectroscopy including FEBID-synchronized pump–probe techniques, i.e., laser-pulse assisted photoemission40 at the electron source of the microscope, (ii) femtosecond infrared spectroscopy41 that could reveal some processes on the surface, such as the evolution of carbonyl signals. In particular, signatures of surface-bound intermediates during the FEBID process would provide valuable insight into the actual fragmentation process/cascade, which at the same time would yield important clues for FEBID-specific precursor design. In addition, deeper insight into electron-induced reactions could be provided by data related to dissociation studies on nanoparticles of a specific size, such as highly defined clusters of inorganic materials being in situ stabilized without the use of surfactants. Materials that come to mind are fullerenes, which go along the work that has been done with noble gas particles but would allow to do these studies at higher temperatures closer to room temperature.

Another aspect requiring advancement is the FEBID-specific precursor design and associated testing of new derivatives and ligand systems. Post-growth purification techniques to remove carbon from composite deposits are not generally applicable and mostly limited to Pt and Au deposits.42,43 Progress in this area of research requires close collaboration between FEBID practitioners and synthetic chemists with an open mind of the involved scientists. We suggested some guidelines for the precursor development in a recent review covering the current state-of-the art on the chemical understanding of specific precursors and ligand systems used/tested for FEBID.4 These methodologies under discussion also include the single-source precursor approach targeting specific compounds in a single gas source deposition process, which has been successfully applied in materials synthesis by thermal conversion under various conditions.

1. Superconductors

FEBID of superconductors is a very recent achievement with the first report describing the deposition of granular PbxCyOδ microstrips using a (CH3CH2)4Pb precursor.44 The microstrips had a Tc of 7.2 K, similar to the value obtained in bulk lead. Despite the good superconducting properties, the observed non-specific deposition associated with prominent co-deposition disqualifies this process for the growth of nanoscale superconductors.

Subsequently, the deposition of W-based FEBID material from W(CO)6 has been described with the highest Tc values being between 4 and 5 K.45 The demonstration of direct-write Josephson junctions using this process illustrates the principle potential for applications in metrology and quantum computing. However, the conditions to drive the metal content high enough in order to increase the critical temperature are not suitable for high resolution and site-selective deposition by FEBID due to the high currents required. Moreover, the achieved Tc is still low and inhomogeneities in the deposits that still contain a large amount of carbon and oxygen will have a significant impact on the device performance being so close to the boiling point of liquid helium at 4.15 K.

Thus, FEBID-derived superconducting materials and some potential applications for nanoscale superconductivity have been shown, but the resolution of the nanostructure writing process still needs to be improved and, ideally, also Tc should be increased. At present, the main bottleneck for the development of FEBID-derived superconductors is the lack of suitable precursors for writing deposits with nanoscale resolution.

On the other hand, the FIBID technique, which uses focused ion beams instead of electrons, has been already successfully employed to grow samples for studies in the field of nano-superconductivity. This has been possible despite the unavoidable Ga implantation as well as sputtering processes associated with Ga-FIBID. For example, WC-based vertical46 and NbC-based free-standing47 3D nanowires were demonstrated with Tc well above the boiling point of liquid 4He. Such superconducting nanowires can be envisioned as building blocks for future 3D nanodevices and, more in general, for applications in quantum information processing and nanoelectronics. Very recent studies demonstrated combinations of FIBID- and FEBID-derived hybrid ferromagnetic/superconductor nano-deposits allowing to study different aspects in the field of fluxonics.48 Remarkably, the FIBID-based superconductor NbC stands out as a material with exceptionally high vortex velocities up to 15 km/s and close-to-perfect edge barriers for studying vortex dynamics.47 Moreover, in combination with FEBID-based ferromagnetic FeCo3, these superconductor nanostructures have been used to demonstrate magnon Cherenkov radiation in a hybrid device.49 

On another note, the co-deposit associated with FEBID, i.e., the penumbra-like deposition of low-metal content material caused by secondary electrons stemming from backscattered electrons around the focal area of the beam, can serve as an electronic coupling medium between closely spaced superconducting islands. This combination does allow the direct writing of Josephson networks consisting of a two-dimensional array of superconducting islands with defined island spacing and the possibility of tuning the Josephson-coupling strength.50 

2. Magnetism

FEBID-derived magnetic material until 2015 has been reviewed51 while in the following years there has been a focus on scanning probe microscopy, such as magnetic force microscopy (MFM),52 as well as on integrating FEBID materials into new device architectures53 and investigations related to topography and formation of heterostructures.

In the last five years, dramatic development has taken place due to a breakthrough in the preparation of 3D magnetic nanostructures19 with new FEBID pattern-generation approaches allowing for the fabrication of complex 3D nano-architectures.54,55 From the theoretical side, novel magneto-chiral effects and topologically induced magnetization configurations were predicted in curved magnetic geometries. Therefore, engineering the topography of 3D nanostructures by tailoring the shape allows to introduce a curvature-induced effective magnetic interaction. Very recently, the potential impact of direct-write 3D FEBID in this curvilinear nanomagnetism was demonstrated by the growth of double-helix geometries with controlled magnetic chirality and incorporation of a change in chirality (Fig. 6).56,57 The change in the chiral vector implemented in this nanostructure as indicated by the star in Fig. 6(a) takes place via the formation of an asymmetric vortex. Thus, chiral spin states have been imprinted in the structure via geometrical chirality only. Moreover, the FEBID approach also enables the formation of localized complex 3D spin textures and intentional topological defects at regions mediating the transition between geometrical chiralities overcoming limitations using standard approaches based on bulk and thin film magnetic systems.

FIG. 6.

(a) Colored SEM image of the Co nanostructure prepared by FEBID using Co2(CO)8, consisting of two double-helices of opposite chirality joined at the tendril perversion marked with *. Scale bar 250 nm, image tilt 45°. (b) Micromagnetic simulations of the linkage between two double-helices of opposite chirality with antiparallel magnetic alignment of their strands (blue and red arrows). (c) Subset of spins after untwisting the magnetization state. Line cross sections are taken between the centers of the two helix strands at different heights and rotated into the xy plane revealing that the change in magnetic chirality (transition from a RH to a LH Bloch wall) takes place via the formation of an asymmetric vortex (gray area). Adapted with permission from Sanz-Hernández et al., ACS Nano 14, 8084 (2020). Copyright 2020 American Chemical Society.

FIG. 6.

(a) Colored SEM image of the Co nanostructure prepared by FEBID using Co2(CO)8, consisting of two double-helices of opposite chirality joined at the tendril perversion marked with *. Scale bar 250 nm, image tilt 45°. (b) Micromagnetic simulations of the linkage between two double-helices of opposite chirality with antiparallel magnetic alignment of their strands (blue and red arrows). (c) Subset of spins after untwisting the magnetization state. Line cross sections are taken between the centers of the two helix strands at different heights and rotated into the xy plane revealing that the change in magnetic chirality (transition from a RH to a LH Bloch wall) takes place via the formation of an asymmetric vortex (gray area). Adapted with permission from Sanz-Hernández et al., ACS Nano 14, 8084 (2020). Copyright 2020 American Chemical Society.

Close modal

The study of spin waves and their use in information processing systems, which is also termed magnonics, is another rapidly growing field in magnetism where a significant impact of 3D FEBID magnetic nanostructures can be expected.19,58 Downscaling phase-setting elements to the nanometer range while keeping them simple, programmable, and integrable with other circuit elements are among the current challenges in nanomagnonics. The ability to fine-tune the spin-wave phase front at the lateral nanoscale will enable the fabrication of spin-wave lenses acting by analogy with phased-array antennas for electromagnetic waves and thus complementing the existing approaches to generate and steer spin-wave beams by tailoring the geometry of electromagnetic-to-spin wave transducers. In this respect, a spinwave phase shifter by FIB milling a nanogroove in the middle of a FEBID-derived FeCo3 magnonic waveguide has been demonstrated.59 Finally, the characterization of magnetic FEBID nano-objects can be carried out by spin-wave spectroscopy by using an Au-coplanar waveguide for spin-wave excitation, as shown for Co3Fe nanodiscs in the 100 nm size range.60 A variation and control of eigenmodes has been achieved by the deposition of an additional Co3Fe ring structure with similar dimensions on top of the nanodiscs.61 The prepared Co3Fe structures have been named nanovolcanoes and can be viewed as multi-mode resonators with the potential to act as 3D building blocks for nanomagnonics.

The investigation of the magnetic configuration of 3D FEBID nanostructures is not trivial. The first measurements were carried out by magneto-optical Kerr effect (MOKE) on tilted Co nanowires complemented by micromagnetic simulations.62 More recently, measurements of the stray field of complex FEBID FeCo3 nano-trees and nano-cubes were carried out by micro-Hall magnetometry, which was also supported by simulation in order to analyze the magnetic configuration in zero and applied field to study the magnetization reversal.20,54 Moreover, nano-SQUID magnetometry63 and imaging of 3D complex nanostructures by x-ray magnetic circular dichroism (XMCD) have been performed, which enabled a reconstruction of the complete 3D nanoscale magnetic configuration of double-helix nanostructures.57 

Beyond the analysis of the magnetic configuration or the measurement of their magnetic stray field, FEBID magnetic nanostructures can be employed in various applications of scanning probe microscopy (SPM). Most recent advancements show significant improvements in sensitivity and resolution by tuning the geometry and the composition of FEBID Co, Fe,64 and Co3Fe52 tips and by employing Co nanowires as MFM transducers64 taking advantage of improvements in deposition techniques.

In general, magnetic properties are dependent on topological features and specific geometries can be prepared by FEBID without limitations known for top-down approaches. Therefore, a large impact on different aspects in modern applied and fundamental magnetism research such as fluxonics and magnonics can be expected by the continuous development of deposition strategies and also extending the available material compositions.

3. Magnetostrictive materials—A new challenge

FEBID of pure metallic material without the need of post-processing as well as potentially compound materials depends on the availability of suitable precursor derivatives. This is intertwined with an in-depth understanding of the processes related to known precursors/ligands. In this respect, also well-known but neglected precursors, such as Ni(CO)4, could come into play when FEBID of magnetostrictive materials is considered. Even though this precursor is highly toxic, by suitable precautions, including encasing of the precursor feed section in an additional ventilated box and appropriate adsorbent, it could nevertheless be used. For instance, the preparation of Ni-containing alloys of Fe or Co could be envisioned using a dual-GIS deposition approach using monometallic carbonyls as precursors. FEBID deposition of such magnetostrictive nanoscaled alloys could pave the way for further development of sensors or combined with actuators. New materials and applications being developed will increase the visibility of the FEBID approach and reveal the vast potential for different fields of application.

1. FEBID of polymers

Another application could be related to a surface-controlled, site-selective polymer formation from the gas phase. This will require to limit/prevent an electron-induced C-H cleavage, while at the same time electron-induced initiation of polymerization events from molecules transiently adsorbing to the surface should be initiated. In contrast to the typical solution processing of polymers, this FEBID polymerization process will certainly be limited by a restricted structural arrangement of polymer chains in such a gas phase approach. Radiation-induced polymerization is a well-investigated method for polymer formation in liquids or gels65 and is also related to e-beam lithography using thin resist layers but distinct differences will have to be considered. To the best of our knowledge, no reports on the intentional formation of surface-anchored functional polymers have been described in the literature but work in this direction could open up new opportunities. For this, the monomers must be volatile enough for a steady gaseous supply and at the same time not receptive to side reactions. On a related topic, very recently, electron-induced cross-linking of structures from liquid prepolymer solutions has been demonstrated allowing the formation of patterned hydrogels as well as inorganic–organic hybrid materials containing gold nanoparticles.66 This example illustrates the versatility of FEBID extending the materials classes being printed on a surface to hydrogels as shown in Fig. 7.

FIG. 7.

SEM images of 3D poly(ethylene glycol) diacrylate hydrogel feature examples patterned using the focused electron beam. (a) Donut-like gel feature made by writing a 1 μm wide ring with a radius of 10 μm. (b) Dome structures formed out of four overlapping coaxial rings by varying electron beam energy and dose (c) for every ring. Adapted with permission from Gupta et al., ACS Nano 14, 12982 (2020). Copyright 2020 American Chemical Society.

FIG. 7.

SEM images of 3D poly(ethylene glycol) diacrylate hydrogel feature examples patterned using the focused electron beam. (a) Donut-like gel feature made by writing a 1 μm wide ring with a radius of 10 μm. (b) Dome structures formed out of four overlapping coaxial rings by varying electron beam energy and dose (c) for every ring. Adapted with permission from Gupta et al., ACS Nano 14, 12982 (2020). Copyright 2020 American Chemical Society.

Close modal

Similarly, a new field for FEBID could be related to nano-medicine and biology. In this respect, the well-known effects of beam damage of cellular tissue by a focused electron beam have to be considered and therefore a direct growth in solution containing living cells is deemed not to be very fruitful. Hence, the formation of structures in living organic matter might not be the best strategy for FEBID. We rather envision much more impact of FEBID in the preparation of patterned surfaces allowing us to study morphological effects on cells and bacteria interacting with these structures. Surface nano-topographies with specific dimensions have been shown to kill various types of bacterial strains through a mechanical mechanism,67 while regulating stem cell differentiation and tissue regeneration.68,69 In this respect, the size and shape effects as well as material composition effects could be studied using the direct-write capacities of FEBID. Recently, the effect of different pattern geometries/sizes has been shown to affect the direct penetration of nanopatterns into the bacterial cell wall leading to the disruption of the cell wall and cell death of bacteria.70 In addition, the demonstration of Si nanowires penetrating living cells71 does also illustrate that it might be possible to record information on processes related to cell migration and the effects of the artificial surface structuring on the cell's speed and direction during their movement. For this purpose, FEBID writing of 3D PtCx nanoarches, similar to other reported sensor applications,23 should allow to collect electrical readout related to the nanostructure bending caused by the cell migration. Therefore, FEBID could be well suited to provide valuable contributions in the understanding of nano-bio-interactions.

2. FEBID of alloys: Direct deposition and layered heterostructures as intermediates

The growth of FEBID alloys and heterostructures as intermediates to obtain alloys, such as bilayer or multilayer systems, allows to obtain materials with tunable structural, electrical, and magnetic properties not present in the single component materials. The growth of FEBID alloys can be achieved by three approaches including (i) the co-deposition of two or three different precursors using individual GISs, (ii) the intermixing of multilayers by post-growth processing, and (iii) the use of heteronuclear single-source precursors. Examples for the simultaneous injection of individual precursors using multiple GISs include the preparation of binary PtSi,72 CoPt,73 CoSi,74 and FeSi75 or ternary compounds Co2FeSi by a hybrid approach.76 However, one of the major problems that can be encountered by this approach is competition for adsorption sites on the substrate of the different precursors requiring tuning of the deposition parameters such as individual precursor injection to obtain a specific ratio.75 

The multi-GIS approach can also be used for the individual deposition of bilayer or multilayer structures of two or more components by alternating deposition with different precursor gases (Fig. 8). This approach allows to prepare intermediate layers containing only single metal deposits that can be converted into alloys by post-growth processing including thermal annealing77 or electron-induced diffusion at room temperature.76 However, this approach is limited to planar deposits and a 3D nanostructure deposition is not possible by FEBID alone and would require other growth strategies such as combinations with other gas-phase deposition techniques. Here, we refer to the hybrid approaches discussed below.

FIG. 8.

TEM lamella of a [Co2FeSi]12 multilayer grown by the use of HFeCo3 (CO)12, Fe(CO)5, and Si5H12. Right: Zoom of the multilayer structure. The dark and bright layers correspond to Co2Fe and Si layers, respectively. After growth, the multilayer was treated by electron beam irradiation to induce intermixing the formation of the Co2FeSi compound, see darker regions in the upper part of the multilayer. Reprinted with permission from Porrati et al., Nanotechnology 28, 415302 (2017). Copyright 2017 IOP Publishing.

FIG. 8.

TEM lamella of a [Co2FeSi]12 multilayer grown by the use of HFeCo3 (CO)12, Fe(CO)5, and Si5H12. Right: Zoom of the multilayer structure. The dark and bright layers correspond to Co2Fe and Si layers, respectively. After growth, the multilayer was treated by electron beam irradiation to induce intermixing the formation of the Co2FeSi compound, see darker regions in the upper part of the multilayer. Reprinted with permission from Porrati et al., Nanotechnology 28, 415302 (2017). Copyright 2017 IOP Publishing.

Close modal

In order to tackle the issue of different adsorption characteristics of individual precursors and the limitations of the layer-based approach, single-source precursors containing all the phase forming elements in one molecule have also been tested as sources for FEBID allowing to deposit MnSix,78 CoSi, and Co2Si,79 as well as FeCo3.80 Particularly relevant is the HFeCo3(CO)12 precursor leading to FeCo3 deposits with a high metal content of about 80 at.% in planar deposits80 and essentially pure FeCo3 in 3D nanowire-based deposits54 due to the very favorable decomposition characteristics as discussed in Sec. III B.

For future studies to expand the accessible alloy material combinations, especially the layer-base growth technique as well as the synthesis and application of suitable single-source precursors are the most promising paths. Therefore, precursor development is crucial to also access binary compounds that contain non-metal components.

FEBID-PtCx deposits have been used to prepare sensor devices. The nanogranular microstructure of this PtCx material with nm-sized Pt particles in a C-matrix results in tunability of the electrical properties dominated by tunneling. The combination of this FEBID material based on nanogranular metals with well-established micro-electro-mechanical systems (MEMS) enables the fabrication of nanogranular tunneling resistor strain sensors, as was demonstrated as an alternative for the typically optical readout in atomic force microscopy (AFM) imaging.81 Similarly, the electrical readout mechanism can be used in 3D nanoscale resonators allowing to distinguish between different gaseous environments that cause a frequency shift in the FEBID-derived nanoarches (Fig. 9).23 

FIG. 9.

(a) Tilted SEM image of an arch shaped 3D Pt–C nanoresonator allowing an electrical readout. (b) The sequential shift of resonance frequencies during exposure to different gas species, namely, ambient air (purple), nitrogen (blue), oxygen (red), and SF6 (green). After each exposure step, high vacuum is subsequently resumed, which reveals the reversible (shaded blue) and irreversible (shaded in red) characteristics. Adapted with permission from Arnold et al., Adv. Funct. Mater. 28, 1707387 (2018). Copyright 2018 John Wiley and Sons.

FIG. 9.

(a) Tilted SEM image of an arch shaped 3D Pt–C nanoresonator allowing an electrical readout. (b) The sequential shift of resonance frequencies during exposure to different gas species, namely, ambient air (purple), nitrogen (blue), oxygen (red), and SF6 (green). After each exposure step, high vacuum is subsequently resumed, which reveals the reversible (shaded blue) and irreversible (shaded in red) characteristics. Adapted with permission from Arnold et al., Adv. Funct. Mater. 28, 1707387 (2018). Copyright 2018 John Wiley and Sons.

Close modal

This sensing mechanism based on mechanical deformation and a simple electrical readout could be extended to a multitude of sensing applications in different media/environments. The PtCx deposits are rather inert and, therefore, encapsulation is a minor issue, while the strain-sensitive resistor can be easily placed on a pre-constructed MEMS platform. The portfolio for sensing applications does include the health sector and cantilever-based sensors for SARS-CoV-2 virus detection are in development. A large benefit of such FEBID resistor-based readout is the simplicity that does not rely on color detection or other detection mechanisms and is operated at room temperatures. Moreover, the sensitivity of the readout does allow the detection of rather small quantities of the analytes. Hence, the fundamental properties of FEBID materials can be harnessed and applied to the fabrication of future devices in sensing.

In this context, the tip editing of commercially available cantilevers or the entire fabrication of custom-made probes for scanning probe microscopy (SPM) by FEBID has been reviewed in a recent article.52 Another recent development in SPM demonstrated the fabrication of a FeCo3-Pt granular ferromagnetic Hall sensor grown by co-deposition using two precursors employing a dual-GIS geometry.82 The nanogranular ferromagnetic Hall devices fabricated can be tailor-made for any given probe geometry, while a magnetic stray field sensitivity has been demonstrated.

Yet another potential sensor application based on FEBID-derived materials has been opened up by demonstrating the sensitivity of the electrical conductance of nanogranular Pt to changes in the dielectric environment nearby. Such changes modulate the charging energy associated with the grain-to-grain tunneling process. This sensing principle, which can be theoretically understood on the mean-field level,83 has been demonstrated to detect sub-monolayers of adsorbed water84 and the paraelectric-to-ferroelectric transition of an organic ferroelectric.85 With the extension from DC to AC conductance, as has very recently been done for nanogranular Pt FEBID structures,86 a new approach to dielectric spectroscopy on the sub-micrometer scale can be envisioned.

In recent years, direct-write nanofabrication by focused electron beam induced deposition (FEBID) has seen tremendous progress in several areas. This comprises (1) a significant increase in the number and purity of FEBID materials for magnetism and superconductivity, (2) the demonstration of a broad range of sensor-related applications, in particular, in scanning probe microscopy, (3) reliable 3D fabrication approaches for wire-frame and sheet-like nanostructures, and (4) several advancements in FEBID-specific precursor synthesis, to name the most important. This list can be extended to include promising results in the fabrication of 3D plasmonic structures by FEBID.9,21,22 Certainly, there are challenges that have to be addressed by the specialists in this field, but it is hoped that this perspective will help to foster a broader interest in this fascinating direct-write approach to functional nanostructures from 0 to 3D.

M.H. thanks the Deutsche Forschungsgemeinschaft (DFG) for financial support through Project Nos. HU 752/15-1 and HU 752/16-1. S.B. acknowledges funding by the Deutsche Forschungsgemeinschaft in the Heisenberg Programme (BA 6595/1-1; Grant No. 413940754). This work was conducted within the Frankfurt Center for Electron Microscopy (FCEM).

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

The authors have no conflicts to disclose.

1.
I.
Utke
,
P.
Hoffmann
, and
J.
Melngailis
,
J. Vac. Sci. Technol. B
26
,
1197
(
2008
).
2.
M.
Huth
,
F.
Porrati
,
C.
Schwalb
,
M.
Winhold
,
R.
Sachser
,
M.
Dukic
,
J.
Adams
, and
G.
Fantner
,
Beilstein J. Nanotechnol.
3
,
597
(
2012
).
3.
O.
Ingolfsson
, Low-Energy Electrons Fundamentals and Applications (
Jenny Stanford Publishing
,
New York
,
2019
), p.
47
.
4.
S.
Barth
,
M.
Huth
, and
F.
Jungwirth
,
J. Mater. Chem. C
8
,
15884
(
2020
).
5.
A.
Botman
,
J. J. L.
Mulders
, and
C. W.
Hagen
,
Nanotechnology
20
,
372001
(
2009
).
6.
W. F.
van Dorp
,
Appl. Phys. A
117
,
1615
(
2014
).
7.
W. F.
van Dorp
,
B.
van Someren
,
C. W.
Hagen
,
P.
Kruit
, and
P. A.
Crozier
,
Nano Lett.
5
,
1303
(
2005
).
8.
I.
Utke
,
V.
Friedli
,
M.
Purrucker
, and
J.
Michler
,
J. Vac. Sci. Technol. B
25
,
2219
(
2007
).
9.
J. D.
Fowlkes
,
R.
Winkler
,
B. B.
Lewis
,
M. G.
Stanford
,
H.
Plank
, and
P. D.
Rack
,
ACS Nano
10
,
6163
(
2016
).
10.
M.
Toth
,
C.
Lobo
,
V.
Friedli
,
A.
Szkudlarek
, and
I.
Utke
,
Beilstein J. Nanotechnol.
6
,
1518
(
2015
).
11.
D.
Sanz-Hernández
and
A.
Fernández-Pacheco
,
Beilstein J. Nanotechnol.
8
,
2151
(
2017
).
12.
A.
Szkudlarek
,
M.
Gabureac
, and
I.
Utke
,
J. Nanosci. Nanotechnol.
11
,
8074
(
2011
).
13.
J.
Cullen
,
C. J.
Lobo
,
M. J.
Ford
, and
M.
Toth
,
ACS Appl. Mater. Interfaces
7
,
21408
(
2015
).
14.
E.
Mutunga
,
R.
Winkler
,
J.
Sattelkow
,
P. D.
Rack
,
H.
Plank
, and
J. D.
Fowlkes
,
ACS Nano
13
,
5198
(
2019
).
15.
R. M.
Thorman
,
T. P.
Ragesh Kumar
,
D. H.
Fairbrother
, and
O.
Ingólfsson
,
Beilstein J. Nanotechnol.
6
,
1904
(
2015
).
16.
K.
Muthukumar
,
H. O.
Jeschke
,
R.
Valentí
,
E.
Begun
,
J.
Schwenk
,
F.
Porrati
, and
M.
Huth
,
Beilstein J. Nanotechnol.
3
,
546
(
2012
).
17.
P.
de Vera
,
A.
Verkhovtsev
,
G.
Sushko
, and
A. V.
Solov’yov
,
Eur. Phys. J. D
73
,
215
(
2019
).
18.
P.
de Vera
,
M.
Azzolini
,
G.
Sushko
,
I.
Abril
,
R.
Garcia-Molina
,
M.
Dapor
,
I. A.
Solov’yov
, and
A. V.
Solov’yov
,
Sci. Rep.
10
,
20827
(
2020
).
19.
A.
Fernández-Pacheco
,
L.
Skoric
,
J. M.
De Teresa
,
J.
Pablo-Navarro
,
M.
Huth
, and
O. V.
Dobrovolskiy
,
Materials
13
,
3774
(
2020
).
20.
M.
Al Mamoori
,
L.
Keller
,
J.
Pieper
,
S.
Barth
,
R.
Winkler
,
H.
Plank
,
J.
Müller
, and
M.
Huth
,
Materials
11
,
289
(
2018
).
21.
R.
Winkler
,
F.-P.
Schmidt
,
U.
Haselmann
,
J. D.
Fowlkes
,
B. B.
Lewis
,
G.
Kothleitner
,
P. D.
Rack
, and
H.
Plank
,
ACS Appl. Mater. Interfaces
9
,
8233
(
2017
).
22.
D.
Kuhness
,
A.
Gruber
,
R.
Winkler
,
J.
Sattelkow
,
H.
Fitzek
,
I.
Letofsky-Papst
,
G.
Kothleitner
, and
H.
Plank
,
ACS Appl. Mater. Interfaces
13
,
1178
(
2021
).
23.
G.
Arnold
,
R.
Winkler
,
M.
Stermitz
,
A.
Orthacker
,
J.-H.
Noh
,
J. D.
Fowlkes
,
G.
Kothleitner
,
M.
Huth
,
P. D.
Rack
, and
H.
Plank
,
Adv. Funct. Mater.
28
,
1707387
(
2018
).
24.
L.
Keller
and
M.
Huth
,
Beilstein J. Nanotechnol.
9
,
2581
(
2018
).
25.
J. D.
Fowlkes
,
R.
Winkler
,
B. B.
Lewis
,
A.
Fernández-Pacheco
,
L.
Skoric
,
D.
Sanz-Hernández
,
M. G.
Stanford
,
E.
Mutunga
,
P. D.
Rack
, and
H.
Plank
,
ACS Appl. Nano Mater.
1
,
1028
(
2018
).
26.
J.
Sattelkow
,
J. E.
Fröch
,
R.
Winkler
,
S.
Hummel
,
C.
Schwalb
, and
H.
Plank
,
ACS Appl. Mater. Interfaces
11
,
22655
(
2019
).
27.
L.
Skoric
,
D.
Sanz-Hernández
,
F.
Meng
,
C.
Donnelly
,
S.
Merino-Aceituno
, and
A.
Fernández-Pacheco
,
Nano Lett.
20
,
184
(
2020
).
28.
A. J. M.
Mackus
,
J. J. L.
Mulders
,
M. C. M.
van de Sanden
, and
W. M. M.
Kessels
,
J. Appl. Phys.
107
,
116102
(
2010
).
29.
H.-B.-R.
Lee
and
S. F.
Bent
,
Atomic Layer Deposition of Nanostructured Materials (
Wiley, 2012
)
, p.
193
.
30.
A. J. M.
Mackus
,
M. J. M.
Merkx
, and
W. M. M.
Kessels
,
Chem. Mater.
31
,
2
(
2019
).
31.
P.
Peinado
,
S.
Sangiao
, and
J. M.
De Teresa
,
ACS Nano
9
,
6139
(
2015
).
32.
C.
Preischl
,
L. H.
Le
,
E.
Bilgilisoy
,
F.
Vollnhals
,
A.
Gölzhäuser
, and
H.
Marbach
,
Small
16
,
2003947
(
2020
).
33.
M.
Drost
,
F.
Tu
,
L.
Berger
,
C.
Preischl
,
W.
Zhou
,
H.
Gliemann
,
C.
Wöll
, and
H.
Marbach
,
ACS Nano
12
,
3825
(
2018
).
34.
S.
Sangiao
,
L.
Morellón
,
M. R.
Ibarra
, and
J. M.
De Teresa
,
Solid State Commun.
151
,
37
(
2011
).
35.
B.
Zhou
,
Y.
Li
,
G.
Zheng
,
K.
Dai
,
C.
Liu
,
Y.
Ma
,
J.
Zhang
,
N.
Wang
,
C.
Shen
, and
Z.
Guo
,
J. Mater. Chem. C
6
,
8360
(
2018
).
36.
J.
Lengyel
,
J.
Kočišek
,
M.
Fárník
, and
J.
Fedor
,
J. Phys. Chem. C
120
,
7397
(
2016
).
37.
T. P.
Ragesh Kumar
,
I.
Unlu
,
S.
Barth
,
O.
Ingólfsson
, and
D. H.
Fairbrother
,
J. Phys. Chem. C
122
,
2648
(
2018
).
38.
Z.
Warneke
,
M.
Rohdenburg
,
J.
Warneke
,
J.
Kopyra
, and
P.
Swiderek
,
Beilstein J. Nanotechnol.
9
,
77
(
2018
).
39.
T. P.
Ragesh Kumar
,
P.
Weirich
,
L.
Hrachowina
,
M.
Hanefeld
,
R.
Bjornsson
,
H. R.
Hrodmarsson
,
S.
Barth
,
D. H.
Fairbrother
,
M.
Huth
, and
O.
Ingólfsson
,
Beilstein J. Nanotechnol.
9
,
555
(
2018
).
40.
G.
Saathoff
,
L.
Miaja-Avila
,
M.
Aeschlimann
,
M. M.
Murnane
, and
H. C.
Kapteyn
,
Phys. Rev. A
77
,
022903
(
2008
).
41.
C. R.
Baiz
,
P. L.
McRobbie
,
J. M.
Anna
,
E.
Geva
, and
K. J.
Kubarych
,
Acc. Chem. Res.
42
,
1395
(
2009
).
42.
B.
Geier
,
C.
Gspan
,
R.
Winkler
,
R.
Schmied
,
J. D.
Fowlkes
,
H.
Fitzek
,
S.
Rauch
,
J.
Rattenberger
,
P. D.
Rack
, and
H.
Plank
,
J. Phys. Chem. C
118
,
14009
(
2014
).
43.
D.
Belić
,
M. M.
Shawrav
,
E.
Bertagnolli
, and
H. D.
Wanzenboeck
,
Beilstein J. Nanotechnol.
8
,
2530
(
2017
).
44.
M.
Winhold
,
P. M.
Weirich
,
C. H.
Schwalb
, and
M.
Huth
,
Appl. Phys. Lett.
105
,
162603
(
2014
).
45.
T. J.
Blom
,
T. W.
Mechielsen
,
R.
Fermin
,
M. B. S.
Hesselberth
,
J.
Aarts
, and
K.
Lahabi
,
ACS Nano
15
,
322
(
2021
).
46.
R.
Córdoba
,
A.
Ibarra
,
D.
Mailly
, and
J. M.
De Teresa
,
Nano Lett.
18
,
1379
(
2018
).
47.
O. V.
Dobrovolskiy
,
D. Y.
Vodolazov
,
F.
Porrati
,
R.
Sachser
,
V. M.
Bevz
,
M. Y.
Mikhailov
,
A. V.
Chumak
, and
M.
Huth
,
Nat. Commun.
11
,
3291
(
2020
).
48.
O. V.
Dobrovolskiy
,
V. M.
Bevz
,
E.
Begun
,
R.
Sachser
,
R. V.
Vovk
, and
M.
Huth
,
Phys. Rev. Appl.
11
,
054064
(
2019
).
49.
O. V.
Dobrovolskiy
,
Q.
Wang
,
D. Y.
Vodolazov
,
B.
Budinska
,
R.
Sachser
,
A. V.
Chumak
,
M.
Huth
, and
A. I.
Buzdin
, arXiv:2103.10156 (
2021
).
50.
F.
Porrati
,
F.
Jungwirth
,
S.
Barth
,
O. V.
Dobrovolskiy
, and
M.
Huth
,
Adv. Funct. Mater.
(to be published).
51.
J. M.
De Teresa
,
A.
Fernández-Pacheco
,
R.
Córdoba
,
L.
Serrano-Ramón
,
S.
Sangiao
, and
M. R.
Ibarra
,
J. Phys. D: Appl. Phys.
49
,
243003
(
2016
).
52.
H.
Plank
,
R.
Winkler
,
C. H.
Schwalb
,
J.
Hütner
,
J. D.
Fowlkes
,
P. D.
Rack
,
I.
Utke
, and
M.
Huth
,
Micromachines
11
,
48
(
2020
).
53.
M.
Gavagnin
,
H. D.
Wanzenboeck
,
D.
Belić
, and
E.
Bertagnolli
,
ACS Nano
7
,
777
(
2013
).
54.
L.
Keller
,
M. K. I.
Al Mamoori
,
J.
Pieper
,
C.
Gspan
,
I.
Stockem
,
C.
Schröder
,
S.
Barth
,
R.
Winkler
,
H.
Plank
,
M.
Pohlit
,
J.
Müller
, and
M.
Huth
,
Sci. Rep.
8
,
6160
(
2018
).
55.
R.
Winkler
,
J. D.
Fowlkes
,
P. D.
Rack
, and
H.
Plank
,
J. Appl. Phys.
125
,
210901
(
2019
).
56.
R.
Streubel
,
P.
Fischer
,
F.
Kronast
,
V. P.
Kravchuk
,
D. D.
Sheka
,
Y.
Gaididei
,
O. G.
Schmidt
, and
D.
Makarov
,
J. Phys. D: Appl. Phys.
49
,
363001
(
2016
).
57.
D.
Sanz-Hernández
,
A.
Hierro-Rodriguez
,
C.
Donnelly
,
J.
Pablo-Navarro
,
A.
Sorrentino
,
E.
Pereiro
,
C.
Magén
,
S.
McVitie
,
J. M.
de Teresa
,
S.
Ferrer
,
P.
Fischer
, and
A.
Fernández-Pacheco
,
ACS Nano
14
,
8084
(
2020
).
58.
M.
Gavagnin
,
H. D.
Wanzenboeck
,
S.
Wachter
,
M. M.
Shawrav
,
A.
Persson
,
K.
Gunnarsson
,
P.
Svedlindh
,
M.
Stöger-Pollach
, and
E.
Bertagnolli
,
ACS Appl. Mater. Interfaces
6
,
20254
(
2014
).
59.
O. V.
Dobrovolskiy
,
R.
Sachser
,
S. A.
Bunyaev
,
D.
Navas
,
V. M.
Bevz
,
M.
Zelent
,
W.
Śmigaj
,
J.
Rychły
,
M.
Krawczyk
,
R. V.
Vovk
,
M.
Huth
, and
G. N.
Kakazei
,
ACS Appl. Mater. Interfaces
11
,
17654
(
2019
).
60.
O. V.
Dobrovolskiy
,
S. A.
Bunyaev
,
N. R.
Vovk
,
D.
Navas
,
P.
Gruszecki
,
M.
Krawczyk
,
R.
Sachser
,
M.
Huth
,
A. V.
Chumak
,
K. Y.
Guslienko
, and
G. N.
Kakazei
,
Nanoscale
12
,
21207
(
2020
).
61.
O. V.
Dobrovolskiy
,
N. R.
Vovk
,
A. V.
Bondarenko
,
S. A.
Bunyaev
,
S.
Lamb-Camarena
,
N.
Zenbaa
,
R.
Sachser
,
S.
Barth
,
K. Y.
Guslienko
,
A. V.
Chumak
,
M.
Huth
, and
G. N.
Kakazei
,
Appl. Phys. Lett.
118
,
132405
(
2021
).
62.
A.
Fernández-Pacheco
,
L.
Serrano-Ramón
,
J. M.
Michalik
,
M. R.
Ibarra
,
J. M.
De Teresa
,
L.
O'Brien
,
D.
Petit
,
J.
Lee
, and
R. P.
Cowburn
,
Sci. Rep.
3
,
1492
(
2013
).
63.
M. J.
Martínez-Pérez
,
J.
Pablo-Navarro
,
B.
Müller
,
R.
Kleiner
,
C.
Magén
,
D.
Koelle
,
J. M.
de Teresa
, and
J.
Sesé
,
Nano Lett.
18
,
7674
(
2018
).
64.
M.
Jaafar
,
J.
Pablo-Navarro
,
E.
Berganza
,
P.
Ares
,
C.
Magén
,
A.
Masseboeuf
,
C.
Gatel
,
E.
Snoeck
,
J.
Gómez-Herrero
,
J. M.
de Teresa
, and
A.
Asenjo
,
Nanoscale
12
,
10090
(
2020
).
65.
A.
Ashfaq
,
M.-C.
Clochard
,
X.
Coqueret
,
C.
Dispenza
,
M. S.
Driscoll
,
P.
Ulański
, and
M.
Al-Sheikhly
,
Polymers
12
,
2877
(
2020
).
66.
T.
Gupta
,
E.
Strelcov
,
G.
Holland
,
J.
Schumacher
,
Y.
Yang
,
M. B.
Esch
,
V.
Aksyuk
,
P.
Zeller
,
M.
Amati
,
L.
Gregoratti
, and
A.
Kolmakov
,
ACS Nano
14
,
12982
(
2020
).
67.
I.
Izquierdo-Barba
,
J. M.
García-Martín
,
R.
Álvarez
,
A.
Palmero
,
J.
Esteban
,
C.
Pérez-Jorge
,
D.
Arcos
, and
M.
Vallet-Regí
,
Acta Biomater.
15
,
20
(
2015
).
68.
L. E.
McNamara
,
R. J.
McMurray
,
M. J. P.
Biggs
,
F.
Kantawong
,
R. O. C.
Oreffo
, and
M. J.
Dalby
,
J. Tissue Eng.
2010
,
120623
.
69.
K.
Modaresifar
,
S.
Azizian
,
M.
Ganjian
,
L. E.
Fratila-Apachitei
, and
A. A.
Zadpoor
,
Acta Biomater.
83
,
29
(
2019
).
70.
M.
Ganjian
,
K.
Modaresifar
,
M. R. O.
Ligeon
,
L. B.
Kunkels
,
N.
Tümer
,
L.
Angeloni
,
C. W.
Hagen
,
L. G.
Otten
,
P.-L.
Hagedoorn
,
I.
Apachitei
,
L. E.
Fratila-Apachitei
, and
A. A.
Zadpoor
,
Adv. Mater. Interfaces
6
,
1900640
(
2019
).
71.
W.
Kim
,
J. K.
Ng
,
M. E.
Kunitake
,
B. R.
Conklin
, and
P.
Yang
,
J. Am. Chem. Soc.
129
,
7228
(
2007
).
72.
M.
Winhold
,
C. H.
Schwalb
,
F.
Porrati
,
R.
Sachser
,
A. S.
Frangakis
,
B.
Kämpken
,
A.
Terfort
,
N.
Auner
, and
M.
Huth
,
ACS Nano
5
,
9675
(
2011
).
73.
F.
Porrati
,
E.
Begun
,
M.
Winhold
,
C. H.
Schwalb
,
R.
Sachser
,
A. S.
Frangakis
, and
M.
Huth
,
Nanotechnology
23
,
185702
(
2012
).
74.
F.
Porrati
,
B.
Kämpken
,
A.
Terfort
, and
M.
Huth
,
J. Appl. Phys.
113
,
053707
(
2013
).
75.
F.
Porrati
,
R.
Sachser
,
G. C.
Gazzadi
,
S.
Frabboni
, and
M.
Huth
,
J. Appl. Phys.
119
,
234306
(
2016
).
76.
F.
Porrati
,
R.
Sachser
,
G. C.
Gazzadi
,
S.
Frabboni
,
A.
Terfort
, and
M.
Huth
,
Nanotechnology
28
,
415302
(
2017
).
77.
O. V.
Dobrovolskiy
,
M.
Kompaniiets
,
R.
Sachser
,
F.
Porrati
,
C.
Gspan
,
H.
Plank
, and
M.
Huth
,
Beilstein J. Nanotechnol.
6
,
1082
(
2015
).
78.
F.
Porrati
,
S.
Barth
,
R.
Sachser
,
F.
Jungwirth
,
M.
Eltsov
,
A. S.
Frangakis
, and
M.
Huth
,
J. Phys. D: Appl. Phys.
51
,
455301
(
2018
).
79.
F.
Jungwirth
,
F.
Porrati
,
A.
Schuck
,
M.
Huth
, and
S.
Barth
,
ACS Appl. Mater. Interfaces
13
,
48252
(
2021
).
80.
F.
Porrati
,
M.
Pohlit
,
J.
Müller
,
S.
Barth
,
F.
Biegger
,
C.
Gspan
,
H.
Plank
, and
M.
Huth
,
Nanotechnology
26
,
475701
(
2015
).
81.
M.
Dukic
,
M.
Winhold
,
C. H.
Schwalb
,
J. D.
Adams
,
V.
Stavrov
,
M.
Huth
, and
G. E.
Fantner
,
Nat. Commun.
7
,
12487
(
2016
).
82.
R.
Sachser
,
J.
Hütner
,
C. H.
Schwalb
, and
M.
Huth
,
Nanomaterials
11
,
348
(
2021
).
83.
M.
Huth
,
F.
Kolb
, and
H.
Plank
,
Appl. Phys. A
117
,
1689
(
2014
).
84.
F.
Kolb
,
K.
Schmoltner
,
M.
Huth
,
A.
Hohenau
,
J.
Krenn
,
A.
Klug
,
E. J. W.
List
, and
H.
Plank
,
Nanotechnology
24
,
305501
(
2013
).
85.
M.
Huth
,
A.
Rippert
,
R.
Sachser
, and
L.
Keller
,
Mater. Res. Express
1
,
046303
(
2014
).
86.
M.
Hanefeld
,
P.
Gruszka
, and
M.
Huth
,
Sci. Rep.
11
,
15163
(
2021
).