Symposium MA03 : Directed Matter—Atom-by-Atom Assembly with Electron Beams and Scanning Probes

Recent advancements in transmission electron microscopy (TEM) have made it possible to directly visualise and determine sites of individual atoms in low-dimensional materials. Here we present recent results on in-situ dynamical, voltage-dependent behaviour after the interaction with 80kV-20kV electrons, using our recently developed high-resolution low-voltage SALVE instrument [1]. We discuss examples of interaction mechanisms in single-layer transition metal dichalcogenides, the study of their atomic and electronic structure, including charge density wave (CDW) materials at room temperature as well as at cryogenic temperatures in both reciprocal and real spaces. 1T-TaS₂ in particular is investigated in its pristine form as well as intercalated with pyridine (C₅H₅N), and triethylenediamine (C₆H₁₂N₂). Using momentum-resolved valence electron energy loss spectroscopy (MR-VEELS), we have determined the nature and dispersion of the plasmon and interband excitations.
As a second class of materials we report on in-situ studies about the atomic structure of, in-between graphene, encapsulated calcium sulfate solution. We found that only a few-layer thick crystalline anhydrite (CaSO₄) in the so-called AII phase has been formed and we addressed the liquid-solid crystallization process to internal pressure inside the graphene pockets [2], similarly to the crystallization in [3,4].
We present another class of in-situ studies between two graphene layers, regarded as a fundamental unit of a graphite anode, and relate our results towards novel electrochemical energy storage solutions. In particular, a miniaturized electrochemical cell is presented. We reversibly lithiate single-crystalline bilayer graphene devices in controlled manner using an electrochemical gate confined to a device protrusion. Electron-beam stimulated chemical reaction inside carbon nanotubes are also presented for a wide range of transition metals and discussed in the light of nanocatalysis. [5]
[1] M. Linck, P. Hartel, S. Uhlemann, F. Kahl, H. Müller, J. Zach, M Haider, M. Niestadt M. Bischoff, J. Biskupek, Z. Lee, T. Lehnert, F. Börrnert, H. Rose, and U. Kaiser 117 (2016) 076101.
[2] T. Lehnert, M. Kinyanjui, A. Ladenburger, D. Rommel, K. Wörle, F. Börrnert, K. Leopold and U. Kaiser, ACS nano 11 (2017) 7967-7973.
[3] E. Khestanova, F. Guinea, L. Fumagalli, A.K. Geim and I.V. Grigorieva; Nature Communications, 7:12587, 2016
[4]K.S. Vasu, E. Prestat, J. Abraham, J. Di, R.J. Kashtiban, J. Beheshtian, J. Sloan, P. Carbone, M. Neek-Amal, S.J. Haigh, A.K. Geim and R.R. Nair; Nature Communications, 7:12168, 2016
[5] The authors greatly acknowledge funding from the German Research Foundation (DFG) and the Ministry of Science, Research and the Arts (MWK) of the federal state Baden-Württemberg, Germany in the frame of the SALVE project (www.salve-project.de).

Electron-beam induced transformations provide unprecedented opportunities for shaping materials atom-by-atom, and for studying the underlying processes in real time using in-situ microscopy. High-energy electrons can drive materials transformations via different mechanisms. Knock-on displacement of atoms has been used extensively to induce defects, such as vacancies in 2D materials. Here, we discuss this process in the context of structural transformations between different stable phases of few-layer metal chalcogenide semiconductors, which are initiated by the electron-induced generation of chalcogen vacancies and provide access to layered crystals with unique optoelectronic properties. While direct atomic displacements have long been explored, other interactions of relativistic electrons with matter are only now being recognized. For example, a focused electron beam represents a localized evanescent source of supercontinuum light. We can use this effect to stimulate and image the plasmon-mediated growth of anisotropic metal nanoprisms in solution. This novel approach can support the development of non-thermal chemical processes that depend on plasmonic light harvesting and the transfer of non-equilibrium charge carriers.

Transmission Electron Microscopy (TEM) is a widely used materials characterization method, where electron beam alternating the sample needs to be avoided generally. While in some cases, electron beam can be used to introduce reactions that are useful for the understanding of materials properties. For example, for the study of nucleation and growth of inorganic nanoparticles from a liquid precursor solution, electron beam can trigger the reactions while serving as a light source for imaging. Although the electron beam induced chemical reactions can be complex and often hard to elucidate, the clear trend of inorganic nanoparticle growth may provide useful information for understanding of the reactions. We explore a wide range of systems undergoing chemical reactions induced by electron beam. Control experiments that lead to better understanding of the reactions are also introduced. Phenomena that are system specific and those that are more general will be highlighted. Perspectives on the future development in this area will be provided.

Extremely long electron and nuclear spin coherence times have recently been demonstrated in isotopically pure Si-28 [1,2] making silicon one of the most promising semiconductor materials for spin based quantum information. The two level spin state of single electrons bound to shallow phosphorus donors in silicon in particular provide well defined, reproducible qubits [3] and represent a promising system for a scalable quantum computer in silicon. An important challenge in these systems is the realisation of an architecture, where we can position donors within a crystalline environment with approx. 20-50nm separation, individually address each donor, manipulate the electron spins using ESR techniques and read-out their spin states.

We have developed a unique fabrication strategy for a scalable quantum computer in silicon using scanning tunnelling microscope lithography to precisely position individual P donors in Si [4] aligned with nanoscale precision to local control gates [5] necessary to initialize, manipulate, and read-out the spin states [6]. During this talk I will focus on demonstrating single-shot spin read-out [8] and ESR control of precisely-positioned P donors in Si. I will also describe our approaches to scale up using rf reflectometry [9] and the investigation of 3D architectures for implementation of the surface code [10] and highlight that the device produced have the lowest noise characteristics of any silicon device to date [11].

References
[1] K. Saeedi et al., Science 342, 130 (2013).
[2] J. T. Muhonen et al., Nature Nanotechnology 9, 986 (2014).
[3] B.E. Kane, Nature 393, 133 (1998).
[4] M. Fuechsle et al., Nature Nanotechnology 7, 242 (2012).
[5] B. Weber et al., Science 335, 6064 (2012).
[6] H. Buch et al., Nature Communications 4, 2017 (2013).
[7] B. Weber et al., Nature Nanotechnology 9, 430 (2014).
[8] T. F. Watson et al., Science Advances 3, e1602811 (2017).
[9] M.G. House et al., Nature Communications 6, 8848 (2015)
[10] C. Hill et al., Science Advances 1, e1500707 (2015).
[11] S. Shamim et al., Nano Letters 16, 5779 (2016).

The novel properties of graphene-like honeycomb structure have spurred tremendous interest in investigating other two-dimensional (2D) layered structures beyond graphene. In this lecture, I will present construction of graphene, silicene, germanene, hafnium honeycomb lattice, monolayer PtSe₂ as well as HfTe₃/HfTe₅, a superconductor-topological insulator layered heterostructure, on transition metal surfaces (TMS) (for example, Ru(0001), Pt(111), Hf(0001) and Ir(111)). Molecular beam epitaxial growth technique is used to form the large scale 2D atomic crystals on TMS. Low electron energy diffraction (LEED) and scanning tunneling microscopy/spectroscopy (STM/S) together with density functional theory (DFT) calculations are employed to confirm the formed structures on the TMS. In addition, we have successfully intercalated Si-layer at the interface between the formed graphene and the Ru(0001). The intercalation mechanism has been clarified with STM observations at an atomic level and the DFT calculations. We expect that these new 2D crystals materials will show very interesting physical property and its promising potential applications in nanoscale devices.

In collaboration with Y.L. Wang, S.X. Du, H.M. Guo, L. Huang, H.T. Yang, J.T. Sun, Y. Pan, L. Meng, L.F. Li, G. Li, Y.Q. Wang, X. Wu, L.Z. Zhang, S.R. Song, J.B. Pan et al. from Institute of Physics, CAS; Z.H. Qin from Wuhan Institute of Physics and Mathematics, CAS; S.Y. Zhou from Tsinghua University; S. Pantelides from Vanderbilt University, US; A. Ferrari from University of Cambridge, UK; M. Ouyang from Maryland University, US; W.A. Hofer from the University of Liverpool, F. Liu from University of Utah, US.

References

[1] Y. Pan et al., Adv. Mater. 21 (2009) 2777.
[2] L. Meng et al., Nano Lett. 13 (2013) 685.
[3] L.F. Li et al., Nano Lett. 13 (2013). 4671.
[4] L.F. Li et al., Adv. Mater. 26 (2014) 4820.
[5] Y.L. Wang et al., Nano Lett. 15 (2015) 4013.
[6] G. Li et al., J. Am. Chem. Soc. 137 (2015) 7099.
[7] Y.Q. Wang et al., Adv. Mater. DOI: 10.1002/adma.201600575 (2016).

Symmetry-protected Dirac nodal line semimetals (DNLSs) receive much attention because of their exotic physical properties and potential applications in dissipationless spin devices. Several compounds (Ca₃P₂, Cu₃PdN, etc.) have been found to be three-dimensional DNLSs, in which the topological Dirac nodal line fermions (DNLFs) can be driven to form a Weyl fermion, Dirac fermion, or other topological phase if the protected symmetry is broken. However, DNLFs in 2D atomic crystal have rarely been explored because of their high vulnerability to symmetry breaking, even in challenging cryogenic experiments. Here we propose that free-standing monolayer CuSe with honeycomb structure is endowed with the exotic 2D DNLF protected by mirror reflection symmetry, based on first-principles calculation. The DNLF state is evidenced by nontrivial edge states that arise as the crossing bands open the gaps with spin-orbit coupling. Motivated by the promising 2D DNLF feature of CuSe, we constructed monolayer CuSe on a Cu(111) surface by molecular beam epitaxy and confirmed success with scanning tunneling microscopy. The good agreement of angle resolved photoemission spectroscopy with the calculated band structures of CuSe/Cu(111) demonstrates that it is a monolayer CuSe with a distorted honeycomb lattice. Our theoretical and experimental results, considered together, establish planar transition-metal honeycomb structure as a new platform to study 2D DNLFs.

In collaboration with Lei Gao, Jia-Tao Sun, Jian-Chen Lu, Hang Li, Kai Qian, Shuai Zhang, Tian Qian, Hong Ding and Hong-Jun Gao from the Institute of Physics, CAS; X. Lin and Y. Y. Zhang from the University of CAS.

Scanning probe microscopies, with their ability for atomic scale resolution imaging and patterning, are expected to enable fabrication of quantum electronic devices at scales never achievable before. Atomically precise patterning of a hydrogen passivated Si (100) surfaces by a scanning tunneling microscope (STM) allows selective doping to fabricate planar nanowires and quantum dots. These devices are then encapsulation by an epitaxial layer of Si. A key challenge in this method is locating the encapsulated devices and aligning electrical contacts to them with a high degree of accuracy. Kelvin Probe Force Microscopy (KPFM), which enables nanometer-scale imaging of the surface potential, is capable of imaging encapsulated Phosphorous devices. Based on topography and surface potential images acquired simultaneously, we can locate devices with respect to the etched fiducial marks, and then align and fabricate electrical contacts using an e-beam lithography tool. The presentation will focus mainly on subsurface imaging of buried devices and a strategy for aligning electrical contacts to STM patterned devices using KPFM.

Scanning transmission electron microscopy (STEM) is emerging as fundamentally new kind of tool for the direct assembly of nanostructures. Atomically precise manipulation with STEM relies on recent advances in instrumentation that have enabled non-destructive atomic-resolution imaging at lower electron energies. Graphene, the atomically thin layer of hexagonally bonded carbon, is the ideal sample for such experiments. In an effort to control its properties, heteroatom dopants have been introduced into graphene both during growth and using post-growth methods, with ion implantation being a particularly promising example of the latter technique.

While momentum transfer from highly energetic electrons often leads to atom ejection, interesting dynamics can be induced when the transferable kinetic energies are comparable to bond strengths in the material [1]. Operating in this regime, very recent experiments have revealed the potential for single-atom manipulation of Si heteroatoms in the graphene lattice using the Ångström-sized electron beam [2]. In our latest experiments, we have achieved 36 controlled single-site jumps with a manipulation rate already comparable to state-of-the-art in fully automated scanning tunneling microscopy. Sample quality thus appears to be the principal challenge in creating 2D nanostructures from multiple Si atoms in the near future [3].

To enable such successes, it has been vital to understand the relevant atomic-scale phenomena through accurate dynamical simulations. Although excellent agreement between experiment and theory for the specific case of atomic displacements from graphene has been recently achieved using density functional theory molecular dynamics [4], in many other cases quantitative accuracy remains a challenge. I will discuss our recent reanalysis of available experimental data on beam-driven dynamics of N, B, and Si heteroatoms in light of such simulations [5], and present our latest manipulation trials with implanted P and Ge.

[1] Susi, T., Kotakoski, J., Kepaptsoglou, D., Mangler, C., Lovejoy, T.C., Krivanek, O.L., Zan, R., Bangert, U., Ayala, P., Meyer, J.C., Ramasse, Q., Phys. Rev. Lett. 113, 115501 (2014). doi:10.1103/PhysRevLett.113.115501
[2] Susi, T., Meyer, J.C., Kotakoski, J., Ultramicroscopy 180, 163-172 (2017). doi:10.1016/j.ultramic.2017.03.005
[3] Nosraty Alamdary, D., Kotakoski, J., Susi, T., Physica Status Solidi B, 1700188 (2017).
[4] Susi, T., Hofer, C., Argentero, G., Leuthner, G.T., Pennycook, T.J., Mangler, C., Meyer, J.C., Kotakoski, J., Nat. Commun. 7:13040 (2016), doi: 10.1038/ncomms13040
[5] Susi, T., Kapaptsoglou, D., Lin, Y.-C., Ramasse, Q., Meyer, J.C., Suenaga, K., Kotakoski, J., 2D Materials 4, 042004 (2017). doi:10.1088/2053-1583/aa878f

Two-dimensional (2D) materials have fascinating properties due to their monolayer nature and are promising candidates for flexible nanoelectronic and optoelectronic applications. Defects are well known to have profound influence on the performance of these materials. In order to fully develop their potential, it is essential to understand the atomic structures and dynamical behaviors of the intrinsic defects, and their related electronic properties.

Excitation of dynamical evolution of defects and simultaneous atomic resolution imaging can be realized with an aberration corrected electron beam inside the scanning transmission electron microscope (STEM). This method offers time-resolved direct tracking of the atomic motion during the structural changes induced by the high energy electrons. By controlling the scanning pattern of the electron beam, we can even manipulate the evolution of the defects.

In this talk, I will first show the atomic scale characterizations of complex defect structures in common 2D materials, such as graphene and MoS₂, and elaborate how they affect the physical properties of the materials by combing density functional theory (DFT) calculations. I will then demonstrate the atom-by-atom structural evolutions as monitored by sequential Z-contrast (STEM) imaging and its underlying physics, such as Se vacancy-induced inversion domain nucleation in MoSe₂, the origin of novel 2D Pd₂Se₃ phase driven by interlayer fusion in layered PdSe₂, and the electron beam induced synthesis of hexagonal MoSe₂ from square FeSe [1-3]. At the end of the talk, I will discuss the in-situ fabrication of highly stable metallic nanowires with MX stoichiometry within the transition-metal dichalcogenide (TMD) monolayers by steering the electron beam with atomic precision [4]. These nanowires can be made into different morphology and can be also alloyed either in cation or anion, effectively tuning their electronic structures [5]. These tunable nanowires could, therefore, serve as ultrasmall interconnects in future flexible nanocircuits fabricated entirely from the same monolayer.

Reference:
[1] Junhao Lin, et al., ACS Nano, 9, 5189 (2015)
[2] Junhao Lin, et al., Physical Review Letters, 119, 016101 (2017)
[3] John A. Brehm#, Junhao Lin#, et al., under review
[4] Junhao Lin, et al., Nature Nanotechnology, 9, 436 (2014)
[5] Junhao Lin, et al., ACS Nano, 10, 2782 (2016)

The ultimate control of materials will be to construct, from the ground up, the structure and functionality desired by placing single atoms where we want them. Atomic-scale control of individual atoms would profoundly impact efforts in areas such as quantum computing and semiconductor manufacturing, for example. Not only would this capability be revolutionary with regard to manufacturing exotic materials and devices it would be brought to bear fruitfully on atomic-scale chemistry and our understanding of atomic processes. Being able to stick a few atoms together, controllably, and examine how they bond, what structures they form, and measure the mechanical, optical, and electrical properties is the pinnacle of understanding materials and perfecting material design.
As a starting platform, graphene has emerged as an “ideal” test bed for atomic manipulation in the scanning transmission electron microscope (STEM). While there are practical engineering challenges to overcome, such as how to reliably clean graphene^{1, 2} and how to introduce the desired materials in a controllable way,^{3, 4} graphene is robust against a 60 kV electron beam, the position (in x and y) of every atom in the lattice can be directly interpreted from the image (as opposed to a 3D crystal), the atomic species can be determined through image intensity analysis,⁵ and it is to date the only host material where controllable single atom dopant motion has been demonstrated.^{6, 7} Here, we practically examine current developments toward atomic scale control of defects in graphene with a STEM beam as a manipulation tool. In particular, we will discuss methodology and mechanisms employed to construct, in situ, atomic-scale structures in graphene with single atoms. Such experiments involve three or four material control paradigms which will also be discussed: Macroscopic sample fabrication and treatment, mesoscopic material transfer and/or in situ material treatments involving the full sample (such as ion deposition or in suit heating), nanoscopic material manipulation in situ (i.e. sputtering with the electron beam in STEM), and, finally, manipulation of single atoms and defects to fabricate a desired structure. While the greatest interest will naturally occur around the manipulation of single atoms, each step in the preparation of a sample conducive to such control is equally important and will be addressed.
1. M. Tripathi et al, physica status solidi (RRL) – Rapid Research Letters 11 (8), 1700124-n/a (2017).
2. O. Dyck et al, arXiv preprint arXiv:1709.00470 (2017).
3. Q. M. Ramasse et al, ACS Nano 6 (5), 4063-4071 (2012).
4. S. Toma et al, 2D Materials 4 (2), 021013 (2017).
5. O. L. Krivanek et al, Nature 464 (7288), 571-574 (2010).
6. T. Susi et al, Ultramicroscopy 180, 163-172 (2017).
7. O. Dyck et al, Applied Physics Letters 111 (11), 113104 (2017).

The edge structure of a two-dimensional (2D) material plays an important role in its growth and greatly affects electronic properties. To develop controlled edge engineering methods, it is critical to understand the mechanisms and kinetics of how edge structures form and evolve when subjected to different chemical environments. Here, using atomic-scale in situ scanning transmission electron microscopy (STEM), we observe the dynamics of edge structure formation and by combined electron beam irradiation and thermal effects. The most commonly observed edges are Se- or Mo- terminated zigzag edges, and MoSe nanowire (NW) terminated edges. Density functional theory (DFT) calculation on 59 different hypothesized edges confirms that the experimentally observed edges have the lowest formation energy under metal-rich environments. Unique functional properties of NW terminated edges are revealed using DFT calculation. Edge reconstruction from zigzag edges to NW terminated edges was directly observed experimentally and understood using ab initio molecular dynamics (AIMD). The combined atomic scale engineering, theory and simulation presented in this work helps to pave the way to engineering the edge of 2D materials for targeted functional applications via controlled electron beam irradiation.

In situ aberration-corrected STEM imaging was conducted at Oak Ridge National Laboratory’s Center for Nanophase Materials Sciences (CNMS), a U.S. Department of Energy Office of Science User Facility. Synthesis science (XL, DBG, CMR, KX) was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences (BES), Materials Sciences and Engineering Division. This research used resources of the National Energy Research Scientific Computing Center, a DOE Office of Science User Facility supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. W.Z., J.D. and F.D. acknowledge the support from the Institute for Basic Science (IBS-R019-D1) of Korea.

Atomically precise manipulation of graphene and other 2D materials holds promise for fields as diverse as quantum electronics and nanopore sensing. However, manipulation of these materials often assumes a ‘pristine’ system free of contaminants and other detrimental constituents. Recently, several studies have begun to describe and characterize the prevalence and implications of adsorbed hydrocarbon-based contaminants on graphene and other emerging 2D materials. These studies indicate that contamination often occludes the majority of exposed surfaces of the materials of interest and that it has high surface mobility and low volatility – a combination that causes it to be challenging to remove or control in scanning probe and beam-based modification. Although present on virtually all graphene devices with an exposed surface, only recently have the implications of the contamination begun to be acknowledged in the literature. With a growing body of research characterizing the contamination with Raman, XPS, AFM, SIMS, and especially STEM, an understanding of the challenge is emerging. Here we provide an overview of sources of contamination, characterization of the properties and composition, an assessment of the implications for utilizing these materials, along with strategies for mitigation.

Manipulation and control of the matter at the atomic level is one of the ultimate goals in nanoscience. As device elements continue to shrink, and new device concepts such as oxide electronics are being proposed based on unique materials properties, it is vitally important to be able to manipulate a wide range of materials at atomic scale. We have recently demonstrated atomic-level sculpting of 3d crystalline oxide nanostructures from metastable amorphous precursor in a scanning transmission electron microscope (STEM) [1]. SrTiO₃ nanowires were fabricated epitaxially from the crystalline substrate following the beam path, producing crystalline structures as small as 1-2 nm and the process can be observed in situ with atomic resolution. The details of the process led us to conclude that the relevant energy transfer is knock-on in nature, rather than associated with local heating. High localization associated with this process allowed us to fabricate arbitrary shaped structures via control of the position and scan speed of the electron beam.
We have explored the utility of this approach beyond STO/STO structures. Similar processes can be indiced ar Si/amorphous Si interfaces [2]. The growth behavior and the resulting feature shape are well described by modeling using a two-temperature approach. The growth is strongly affected by the presence of SiO2 at the crystalline-amorphous interface. For linear trajectories of the electron beam, the growth rate appears to depend on crystallographic direction.
We have further investigated using beam-induced growth to create heterostructures, where BaTiO3 features were grown on SrTiO3 substrate. Due to substantial lattice mismatch, the beam-induced features exhibit island-type growth with 2-unit-cell-thick pseudomorphic layer. The STEM probe has nm-scale size in beam direction, resulting in localization of the electron beam impact on a similar scale. The prospects for creating functional heterostructures with this approach, as well as its unique utility for uncovering atomic-scale structure property relationships, will also be discussed.
* Research supported by the U.S. Department of Energy (DOE), Basic Energy Sciences (BES), Division of Materials Sciences and Engineering, and by ORNL's Laboratory Directed Research and Development Fund. A portion of this research was conducted at ORNL’s Center for Nanophase Materials Sciences (CNMS), which is sponsored by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. DOE.

[1] Stephen Jesse et al., Small, 11 5895 (2015)
[2] Nan Jiang et al., MRS Bulletin, 42 , 653 (2017).

The properties of semiconductors and insulators are very sensitive to the presence and location of impurities and defects, meaning that it is therefore important to image these defects down to the level of single atoms. Scanning transmission electron microscopy (STEM) is one of the leading techniques to directly identify stable atomic structures and has solved a wide variety of materials problems. Recent improvements in the optics and mechanical stability of the electron microscope now allow us to observe dynamics at the atomic-scale, although the time resolution is still in the order of seconds. In this presentation, we show the dynamic observation of single atom diffusion in a luminescent aluminum nitride (AlN) and the phase transition from spinel to rocksalt in lithium manganese oxide (LiMn₂O₄) by atomic-resolution dynamic STEM imaging.

When a specimen is illuminated by a high-energy electron beam, the incident electron transfers energy to the atoms, promoting dopant diffusion in bulk materials and, moreover, enabling the stimulation of local phase transitions without heating or cooling the sample. One good example is the case of single dopant diffusion in AlN, which is important because the dopant atoms contribute to the luminescence color. The doped Ce atom has unusually large ionic radius in AlN and this size mismatch enhances the mobility of Ce dopants in AlN, enabling us to track the movement of single Ce dopants via dynamic Z-contrast STEM imaging. The second example is the phase transition of spinel LiMn₂O₄. Following some charge-discharge electrochemical processes, this material will locally form rock-salt type structures through cation mixing between lithium and transition metals. We have observed this gradual transition at the atomic-scale by using time-sequential imaging and spectroscopy. The presented dynamic STEM observations could open the way to in-depth understanding of the atomistic phenomena in this and other energy-relevant materials. A part of this work was supported by JSPS KAKENHI Grant Number JP17H06094.

The dopants in graphene (like Si, P, and N) is found to be dynamically mobile, where the atom can travel through the lattice by exchanging sites with the neighbor C atom. Here, we investigate on the machism of lattice hopping through low voltage (60keV) aberration corrected scanning transmission electron microscope (AC-STEM), and we find that the basic machanism is categorized into two types: out-of-plane exchange, and in-plane exchange. Details of the machanism is further studied by density functional theory (DFT), comparing the energy barrier of transition state for different atom species through nudged elastic band (NEB) theory. This fundamental study of lattice dynamics of dopant atoms will pave the way for further investigation on engineering the configuration of dopant atoms in graphene.

Hydrocarbon contamination and deposition in a scanning transmission electron microscope (STEM) is generally considered a detriment. Techniques such as plasma cleaning and thermal annealing are designed specifically to reduce or prevent hydrocarbon contamination. While contamination is clearly a detriment to the vast majority of microscopy studies, here we seek to address how such contamination may be useful, as it relates to nanometer-scale in situ material manipulation. There have been exciting recent developments in atomic scale material control in the STEM where single Si atoms have been controllably moved through a graphene lattice via e-beam manipulation.^1-3 This is exciting for several reasons, 1) the STEM has the capability of focusing the probe onto single atoms in graphene so that addressing individual atoms is possible, 2) individual atomic species may be determined via image intensity or single atom spectroscopy, and 3) structures fabricated can be directly imaged during fabrication. Given this progress toward atomic manipulation in graphene, it is worth exploring additional material control mechanisms with this material system. Given, also, that mechanisms for material manipulation in STEM are limited mostly to e-beam exposure, we approach this topic from the perspective of exploration: if some process is observed (i.e. hydrocarbon deposition), can we control when it occurs, can we precisely control the process (i.e. deposit a single layer of graphene), can we use this phenomenon in a larger process chain? Answering such questions will establish reliable capabilities which will expand the toolbox of material control techniques available in STEM. Here we will highlight several interesting examples where we make use of the ubiquitously observed hydrocarbon deposition in STEM to controllably pattern structures, heal holes, remove dopant atoms from the graphene lattice, and move nanoparticles. Exploring such capabilities is necessary in the pursuit of developing general atomic scale material control in a STEM.

1. S. Toma, K. Demie, L. Yung-Chang, M. R. Quentin, C. M. Jannik, S. Kazu and K. Jani, 2D Materials 4 (4), 042004 (2017).
2. O. Dyck, S. Kim, S. V. Kalinin and S. Jesse, Applied Physics Letters 111 (11), 113104 (2017).
3. O. Dyck, S. Kim, E. Jimenez-Izal, A. N. Alexandrova, S. V. Kalinin and S. Jesse, ArXiv e-prints,
arXiv:1710.09416, (2017).

Transmission electron microscopy (TEM) is traditionally used as a tool to characterise materials, providing atomic resolution imaging of low dimensional nanostructures such as graphene and carbon nanotubes. In this context, damage to materials imaged by TEM (caused by collisions with the highly energetic electrons) is generally considered as something to be avoided or limited. However, with detailed understanding of the effects of the electron beam (e-beam), the energy transmitted from it can be used to drive chemical reactions that would be otherwise unfeasible.

A mechanistic understanding of beam-driven chemical reactions can be achieved with the comparison of experimental TEM images to the results of modelling. The dynamic response of nanotube-encapsulated organic species to the stimulus of the e-beam has been simulated using density functional theory (DFT) molecular dynamics. By combining these results with an accurate analytical model of the interaction of relativistic electrons and atomic nuclei, the experimentally observed behaviour of these systems under the e-beam has been quantitatively characterised.

The elemental dependence of the transfer of energy from the e-beam was shown to play a key role in determining reaction products, and is responsible for the very high susceptibility of carbon-hydrogen bonds to irradiation damage.¹ Deuteration is an effective remedy for overcoming this limitation, increasing lifetimes of organic molecules under electron irradiation and therefore enhancing the accuracy of structural analysis by TEM.² A close iterative collaboration between theory and microscopy was used to establish TEM as an effective tool for chemical reaction discovery and the characterisation of previously unknown reaction mechanisms.³ This has initially been demonstrated with two example reactions, in which organic molecule precursors are activated by the e-beam, eventually forming novel one-dimensional materials.⁴

As another test case, we study transformations of point defects in graphene.⁵ The cross-sections and threshold energies of irreversible (atom emission) and reversible (bond rotation) processes are measured. Observation of statistically significant number of events at variable experimental conditions allows us to decouple beam induced and thermal reaction pathways and obtain independent estimations of the cross-sections and activation energies for direct and back bond rotations. The back rotation is characterized by a very high value of the cross-section. Comparisons to theoretical estimations indicate that the assumed mechanism of direct knock-on damage cannot be the main cause of SW defect healing under electron beam.

1. Nanoscale, 2013, 5, 6677–6692
2. Small, 2015, 11, 622–629
3. Acc. Chem. Res., 2017, 50, 1797-1807
4. ACS Nano, 2017, 11, 2509–2520.
5. Carbon, 2016, 105, 176-182

Molecular electronics where single molecules perform basic functionalities of digital circuits is a fascinating concept that one day may augment or even replace nowadays semiconductor technologies. The tautomerization of molecules, i.e., the bistable functional position of hydrogen protons within an organic frame, has recently been intensively discussed as a potential avenue towards nano-scale switches. It has been shown that tautomerization can be triggered locally or non-locally, for example by a scanning tunneling microscope (STM) tip positioned directly above or in close vicinity to the molecule. Whereas consensus exists that local switching is caused by inelastic electrons which excite vibrational molecular modes, the detailed processes responsible for non-local tautomerization switching and —even more important in the context of this work— methods to control, engineer, and potentially utilize this process are largely unknown.
Here, we demonstrate for H₂Pc molecules on Ag(111) that the tautomerization processes are mediated by surface state electrons with a well-defined dispersion relation. We are able to controllably decrease or increase the probability of non-local, hot electron-induced tautomerization by atom-by-atom–designed Ag nanostructures. We show that Ag atom walls act as potential barriers which exponentially damp the hot electron current between the injection point and the molecule, reducing the switching probability by up to 83% for a four-atom wide wall. By placing the molecule in one and the STM tip in the other focal point of an elliptical nanostructure we could coherently focus hot electrons onto the molecule which led to an almost tripled switching probability. Finally, we will discuss to what extent interference phenomena may be used to control tautomerization. Our results demonstrate that the tautomerization switching of single molecules can remotely be controlled by the utilizing suitable nanostructures and may pave the way for designing new tautomerization-based switches.

A scanning tunneling microscope (STM) can do more than atomic imaging and manipulation, its tunneling current can also be used for the excitation of light, converting electron energy to photon energy. In this talk, I shall first demonstrate the realization of STM based single-molecule electroluminescence by adopting a strategy of both efficient electronic decoupling and nanocavity plasmonic enhancement. The emission intensity, achieved through optimized material combination for molecule, spacer, tip, and substrate, is strong and stable enough for us to carry out second-order photon correlation measurements. The observation of an evident photon antibunching effect demonstrates clearly the nature of single-photon emission from an isolated single molecule that is electrically excited by tunneling electrons. Strikingly, the spectral peak in an isolated monomer is found to split when a molecular dimer is artificially constructed through STM manipulation, which suggests that the excitation energy from tunneling electrons is likely to rapidly delocalize over the whole molecular dimer and form a delocalized exciton. The spatial distribution of the excitonic coupling for different energy states in a dimer can be visualized in real space through sub-nanometer resolved electroluminescence imaging technique, which correlates very well with the local optical responses predicted in terms of coherent intermolecular dipole-dipole coupling. Finally, I shall also demonstrate electrically driven single-photon superradiance for constructed molecular oligomers, which strongly suggests the formation of multi-molecule quantum entangled states. These findings open up new avenues to fabricate electrically driven quantum light sources and to study intermolecular energy transfer at the single-molecule level.

Nanostructure fabrication based on charged-particle beams has for many years been assumed to be at or near its limit. However, the fundamental origins of those limits were not well understood. Secondary electron generation and scattering in materials certainly played a role, but the magnitude and resolution limit introduced by that role was not fully quantified. In recent years, however, the advent of improved microscopy methods, resists with excellent image contrast, and analytic tools such as electron-energy-loss spectroscopy have allowed deeper into the single-nanometer-lithography domain. It has, for instance, pointed to volume plasmons as playing a significant role in the exposure process. Helium ions have also been used to realize sub-10-nm-length-scale lithography. The advent of new bright sources for helium ions with sub-nm-dimension probes suggests novel methods of patterning and perhaps direct milling could permit access to the sub-10-nm or even perhaps sub-1-nm length scales in material modification. In this talk, I will discuss some recent results in this field, but also point to open questions that should be addressed if the ultimate capabilities of beam-based fabrication are to be realized.

The magnetic and electronic properties of materials often find their origin in basic atomic-scale interactions. Yet, due to the large number of atoms involved, many phenomena can be very difficult to predict: we call these ‘emergent’. The ability to build structures atom-by-atom by means of scanning tunneling microscopy (STM) may provide an excellent platform to explore emergence as a function of system size. For example, by properly tuning the anisotropy of magnetic atoms a thin insulator, we have been able to engineer finite spin chains hosting spin waves [1] as well as the beginnings of a quantum phase transition at a critical magnetic field [2]. Unfortunately, the maximum size of such assembled structures is often limited due to e.g. crystal impurity, crystal strain, and general uncontrollability of the STM tip shape, hampering the reliability with which atoms can be manipulated. In this talk, I will demonstrate how atomic assembly can be enhanced dramatically by switching to manipulation of atomic vacancies, rather than adatoms, on a chlorine-terminated copper surface [3]. The resulting structures, comprising thousands of vacancies positioned on an exactly defined grid, are found to be stable up to 77 K. We use this new technique to construct two-dimensional artificial crystals of various size and atomic spacing, and investigate their collective electronic properties through local tunneling spectroscopy [4].


[1] A. Spinelli et al., Nature Matererials 13, 782 (2014)
[2] R. Toskovic et al., Nature Physics 12, 656 (2016)
[3] F. E. Kalff et al., Nature Nanotechnology 11, 926 (2016)
[4] J. Girovsky et al., SciPost Physics 2, 020 (2017)

Hydrogen resist lithography (HRL), based on STM patterning of an atomic hydrogen resist layer on Si(100) is an enabling methodology for atomic precision fabrication. The genesis of HRL¹, and its underlying physical mechanisms², will be discussed along with examples of it use. These include atomic scale patterning, creating templates for covalent molecular attachment to silicon, and modifying substrate nanostructure interactions by desorbing hydrogen from their interfaces. For these latter experiments, we have developed a dry contact transfer (DCT) method of integrating 0D, 1D and 2D nanostructures with clean silicon and III-V semiconductor surfaces, as an interesting route towards fabricating hybrid nanoelectronic systems³. DCT has enabled us to integrate carbon nanotubes, graphene and atomically precise graphene nanoribbons (GNRs) with silicon, GaAs and InAs substrates. STM imaging and spectroscopy, coupled with our atomic resolution STM-based hydrogen resist process have been used to study the interactions of carbon nanotubes, graphene and atomically precise graphene nanoribbons with silicon, GaAs and InAs substrates. By STM and STS, we have observed the metallic zigzag edge state in graphene⁴, carbon nanotube-substrate lattice alignment effects⁵, and the electronic structure of GNRs^{6, 7}. This talk will also show a method for creating sub-5nm metal wires for contacting nanostructures⁸, a SPM probe sharpening technique for producing 1 nm radii probes⁹, and a spin-off of the HRL research in which deuterium has been found to harden CMOS transistor technology against the effects of hot-carrier degradation¹⁰.

References:

1. Lyding, J.; Shen, T.; Hubacek, J.; Tucker, J.; Abeln, G., Applied Physics Letters 1994, 64 (15), 2010-2012.
2. Shen, T.; Wang, C.; Abeln, G.; Tucker, J.; Lyding, J.; Avouris, P.; Walkup, R., Science 1995, 268 (5217), 1590-1592.
3. Albrecht, P.; Lyding, J., Applied Physics Letters 2003, 83 (24), 5029-5031.
4. Ritter, K.; Lyding, J., Nature Materials 2009, 8 (3), 235-242.
5. Ruppalt, L.; Lyding, J., Nanotechnology 2007, 18 (21).
6. Radocea, A.; Sun, T.; Vo, T.; Sinitskii, A.; Aluru, N.; Lyding, J., Nano Letters 2017, 17 (1), 170-178.
7. Pour, M.; Lashkov, A.; Radocea, A.; Liu, X.; Sun, T.; Lipatov, A.; Korlacki, R.; Shekhirev, M.; Aluru, N.; Lyding, J.; Sysoev, V.; Sinitskii, A., Nature Communications 2017, 8.
8. Ye, W.; Martin, P. A. P.; Kumar, N.; Daly, S. R.; Rockett, A. A.; Abelson, J. R.; Girolami, G. S.; Lyding, J. W., Acs Nano 2010, 4 (11), 6818-6824.
9. Schmucker, S.; Kumar, N.; Abelson, J.; Daly, S.; Girolami, G.; Bischof, M.; Jaeger, D.; Reidy, R.; Gorman, B.; Alexander, J.; Ballard, J.; Randall, J.; Lyding, J., Nature Communications 2012, 3.
10. Lyding, J.; Hess, K.; Kizilyalli, I., Applied Physics Letters 1996, 68 (18), 2526-2528.

Recent, many lithography methods using scanning probe microscopy (SPM) have been reported. In particular, experiments were conducted using atomic force microscopy (AFM) software designed for lithographic purpose. These methods have led to the achievement of high resolution and low cost. However, there is still a throughput challenge due to limitation of cantilever-based scanning probe system. Additionally, the recent significant issues are miniaturization and high integration of various devices. The nano-scale materials are excellent solutions for these issues. However, the controlled assembly of nanowire is a key challenge in the development of a range of bottom-up device. Nano-scale combing assembly technique can be a solution. However, nano-scale combing technique needs electron-beam (e-beam) lithography accordingly it also causes high cost and throughput problem. Here, we present new lithographic method overcome these two issues. This method deals with the problems by combining the nano-scale combing technique with the hard-tip, soft-spring lithography method. We proceeded with lithography by attaching the hard tip arrays of a centimeter-scale to a scanning probe microscopy instead of a cantilever. High-throughput large-area patterns of nano-scale were obtained using this method. Additionally, assembly of nanowire can be achieved by combining this patterning process with nano-scale combing technique without e-beam lithography.

The ability to precisely control and combine the composition, size, shape and conformation of nanoscale building blocks represents a major challenge in current fabrication processes. Solution-based fabrication methods are particularly attractive due to their scalability, low cost, and mild operating conditions. While a high degree of size and shape control has been achieved during the last decades for the synthesis of nanoscale colloidal particles, the next challenge is to establish the knowledge and tools required to build arbitrarily designed hybrid 3D architectures with a high degree of placement control and material combination. In this respect, a promising strategy is to control solution-based growth at the nanoscale using scanning probes in liquid.

Our research focuses on the direct writing of functional metallic nanostructures using an electrochemical Atomic Force Microscope (AFM) as a nano-electrode in aqueous solutions containing metal salts, localizing electrochemical deposition in a 3D printing fashion. Our approach is to survey and control the phase formation and growth at the liquid/solid interface in electrodeposition with the AFM tip.

We investigate favourable parameters for local nucleation and growth, both numerically and experimentally. From the numerical point of view, we consider the physics of electrochemical and mass transport processes under DC and AC conditions. We match the deposition dynamics with the in-situ topography profiles of Cu-based structures as they grow (from 2D lines to 3D rods). As a first conclusion, we highlight the main parameters that define the writing resolution in 3 dimensions.

As an example of a functional metal structure, we fabricate a Cu plasmonic nanoantenna. The performance of the antenna is linked to the metal quality through plasmon damping effects. Reciprocally, we exploit the antenna performance as to understand how do the local electrochemical conditions affect the crystalline quality of the growth antenna.

Ultimately, this technique can be extended to different materials and morphologies, such as semiconducting and chiral structures. Moreover, scaling up the process is possible by means of mass parallelization. As such we envision this 3D-printing technique to be both useful for the manufacturing and rapid prototyping of novel light-matter nanodevices, as well as consider this a viable technique for large scale fabrication.

Unlike atomically thin graphene, transition metal dichalcogenide (TMD) monolayers are three-atom thick and can exist in many different polymorphs where metal atom coordination changes from trigonal prismatic (1H phase) to octahedral and distorted octahedral (1T and 1T’ phases).¹ This diversity of crystal types provides an additional tool to control the band structure of TMDs nanostructures so that their electronic properties can span in a wide range from semiconducting and metallic to topological insulators, quantum spin Hall insulators and two-dimensional superconductors.² Thermodynamically favourable phase of WSe₂ monolayers is the trigonal prismatic semiconducting 1H phase which can be converted into the metastable octahedral phase as a result of either an electron transfer (i.e. during lithiation) or applied mechanical strain. Indeed, direct formation of 1H structure type of WSe₂ via wet chemical approaches or physical deposition techniques has only been reported so far.³ In this work, we report on colloidal synthesis of 1T’ WSe₂ nanostructures from tungsten carbonyl precursor in a coordinating solvent oleic acid at 300^oC. WSe₂ nanostructures demonstrate a well-defined flower-like morphology with atomically thin individual petals reaching 50 nm in lateral size. Furthermore, we found that 1T’ phase of WSe₂ can be converted into the semiconducting 1H phase upon annealing at 400^oC in argon atmosphere preserving the starting flower-like morphology. The phase conversion was confirmed by high-resolution transmission electron microscopy, Raman and optical spectroscopy. We suggest that formation of the metastable 1T’ phase of WSe₂ in a low temperature synthesis from tungsten carbonyl can be caused by an excess of electrons on the metal centres during the growth.

References:
¹ Acerce et al, Nature, 2017, 549, 370–373
² Yang et al, Nat. Phys., 2017, 13, 931–937
³ Barrera et al, J. Mater. Chem. C, 2017, 5, 2859–2864.

Inducing and controlling electric dipoles is hindered in the ultrathin limit by the finite screening length of surface charges at metal-insulator junctions, though this effect can be circumvented by specially designed interfaces. We demonstrate that non-zero electric polarization can be induced and reversed in a hysteretic manner in bilayers made of ultrathin insulators whose electric polarization cannot be switched individually [1]. Using scanning tunneling microscopy and atomic force microscopy, we explore the interface between ionic rock salt alkali halides such as NaCl or KBr and polar insulating Cu₂N terminating bulk copper. The strong compositional asymmetry between the polar Cu₂N and the vacuum gap breaks inversion symmetry in the alkali halide layer, inducing out of plane dipoles that are stabilized in one orientation (self-poling). The dipole orientation can be reversed by a critical electric field applied from the tip of a scanning tunneling microscope, producing sharp switching of the tunnel current passing through the junction. Furthermore, the switching is stabilized in the presence of defects that can be manipulated with atomic-scale precision. These results provide a new way to induce, probe, and manipulate electric dipoles at the atomic scale.

[1] J. Martinez-Castro et al., Nature Nanotechnology; DOI: 10.1038/s41565-017-0001-2

We probed the polar CaF2(111) surface by non-contact atomic force microscopy using
a bare metal tip, a CuO tip and a CO-terminated tip. The tips are engineered by poking, CO or CuO pickup and subsequent COFI
imaging (see Welker et al., Science 336, 444 (2012)). In the non-contact regime far from the surface,
the contrast is entirely determined by electrostatics and can be calculated at a precision of about 10 percent using
Coulomb's law and assuming a point charge at the tip apex (positive for the metal tip, negative for CO and CuO tips).
While metal tips turn unstable when reaching the repulsive regime, using CO or CuO tips allow to image the surface at very high resolution
as well. Interestingly, the CO tips provide less van-der-Waals background force, while CuO tips show less artifacts due to tip bending.

Liquid cell Transmission Electron Microscopy (TEM) has attracted a lot of research attention owing to its capacity to access materials properties and dynamics in liquids at a sub-nanometer resolution. In a standard configuration, it utilizes a sealed electron transparent silicon nitride membrane window to separate high-vapour liquid pressure from the vacuum environment. Recently, it has been shown fabrication of the nanofluidic channels directly onto the liquid cell window enabling to monitor the self-assembly dynamics in a dimensionally confined system [1]. Because of the manufacturing difficulty, the channel profile is of high sidewalls surface roughness and exceeds 100 nm which significantly limits its application. An improved technique of SiNx liquid cells nanostructuration is required to advance this field.
Here, we demonstrate the fabrication of dense high aspect ratio silicon nanopatterns of high resolution onto the liquid cell Si₃N₄ membrane window with the use of new electron beam lithography (EBL) resist, ‘SML’ (EM Resist Ltd.) and optimized deep reactive ion etch (DRIE) process parameters. Using SF₆/C₄F₈ etch chemistry [2,3], we found an etch selectivity of SML to Si in the order of 1:4 and etch rate of 138 nm/min. Such etch chemistry is extremely useful for direct SML pattern transfer into silicon, without the hard-mask layer, thereby reducing the surface roughness of the substrate material. We achieved a dense silicon nanolines of a sub 20 nm half-pitch to a depth of 200 nm at 1.2 nm line edge roughness from SEM images. This new manufacturing process allows liquid cell TEM membranes fabrication of target dimension specific to their research at high accuracy and precision. We explore this technique for the prototyping of the active elements of the silicon electronics onto liquid cell membranes, for in-situ TEM imaging of wet chemical etching of the nanofins and elastocapillary interaction of the nanopillars.
References
[1] E. Mielle, S. Raj, Z. Baraissov, P.Král, and U. Mirsaidov. Dynamics of Templated Assembly of Nanoparticle Filaments within Nanochannels, Advanced Materials 29, 1702682 (2017)
[2] Y.Lisunova, M. Spieser, R.D.D. Juttin, F. Holzner, J. Brugger, High-aspect ratio nanopatterning via combined thermal scanning probe lithography and dry etching. Microelectronic Engineering 180, 20 (2017)
[3] C. Rawlings, M. Zientek, M. Spieser, Y. Lisunova, J.Brugger, U. Duerig, and A. W. Knoll, Fabrication of nanometer accurate 3D profiles using closed loop thermal Scanning Probe Lithography, in submission

Hydrogen depassivation lithography has enabled unprecedented sub-nanometer precision in the positioning of dopant atoms in silicon[1], advancing the field of silicon quantum electronics. However, as donor-based Quantum Information Processing devices scale from single-qubit devices towards multi-qubit devices such as crossbar architectures[2], and other applications such as artificial 2D materials made from large arrays of dopant patches are proposed[3], atomically precise lithography is required over increasingly large areas with improved reproducibility.
After developing the ZyVector™ automated STM lithography system with real-time piezo creep correction, we have previously demonstrated open-loop atomic precision patterning (i.e. lithography errors of less than one dimer row or pixel) over length scales up to 100 nm. On scales up to 500 nm, position errors of up to 2.5% were observed. These errors on the larger scale are caused by hysteresis, both directly and through the effect of the hysteresis on the creep errors.
In this work, we address these errors by continuing to optimize the correction of piezo creep, and also implementing real-time correction of hysteresis. Data comparing the effect of the real time positioning corrections on creep in xy and z will be offered. For movements within small areas, creep correction reduces positioning errors by more than 90%. Our current hysteresis model, which we continue to improve, corrects hysteresis errors by about 80%. Real-time corrections in the z direction greatly reduce the settling time of the tip after landing, from about 90 minutes to about 30s. The reduced settling time and minimized z drift after landing allows processes which require the feedback loop to be switched off, such as spectroscopy or atom manipulation, to be performed more efficiently.
In parallel with real-time position corrections, we have developed automatic fiducial alignment routines, allowing any remaining errors to be corrected. The tip position can either be aligned to previously-drawn patterns or to deliberate fiducial marks. A large pattern can therefore be stitched together from an array of write fields, i.e. areas within which atomic precision can be obtained. Thus, the precise patterning achieved for small areas can be scaled to large areas.
Taking all these techniques together, we present a set of design rules, which will allow for successful atomic-precision patterning from the single-atom to the micron scale.
1 M. Fuechsle, et al. Nat Nano 7 242-246 (2012) DOI:: 10.1038/nnano.2012.21
2 C. D. Hill, et al. Science Advances 1 (2015) DOI: 10.1126/sciadv.1500707
3. J. Salfi, et al. Nat. Commun., vol. 7, p. 11342, 2016. DOI: 10.1038/ncomms11342

The removal or addition of individual atoms, rows of atoms, or pre-determined blocks of atoms can be extremely useful for device applications or for defining custom platforms for exploring fundamental science. I will describe the application of electron beams and helium ion beams to atomic-scale geometry manipulations in low dimensional materials, primarily graphene and hexagonal boron nitride. Experimental results will be presented along with relevant theoretical considerations.

Phosphorous delta layer devices in silicon can be fabricated with atomic precision by performing hydrogen depassivation lithography using a scanning tunneling microscope (STM). To date, this process has not been thermally compatible with CMOS processing. We present a low temperature STM sample preparation that enables significant processing of devices prior to STM patterning. This preparation enables a CMOS compatible fabrication path that scales from the nanoscale STM patterned device to macroscopic bond pads using only optical lithography. Using low-temperature electrical transport, we demonstrate a high yield of delta-layer based, nanoscale electrical devices across numerous fabrication runs. This all-optical fabrication pathway enables much faster cycles of learning, opening the door to deeper understanding of fabricated devices and more ambitious process development.

This work was supported by the Laboratory Directed Research and Development Program at Sandia National Laboratories, and was performed, in part, at the Center for Integrated Nanotechnologies, a U.S. DOE, Office of Basic Energy Sciences user facility. Sandia National Laboratories is managed and operated by National Technology and Engineering Solutions of Sandia, LLC., a wholly owned subsidiary of Honeywell International, Inc., for the U.S. Department of Energy under contract DE-NA-0003525.