Advertisement

In Situ Transmission Electron Microscopy

  • Frances M. Ross
  • Andrew M. MinorEmail author
Chapter
  • 4.8k Downloads
Part of the Springer Handbooks book series (SHB)

Abstract

In situ transmission electron microscopy provides dynamic observations of the physical behavior of materials in response to external stimuli such as temperature, gas environment, stress, and electric or magnetic fields. The goal of these experiments is to provide insight into atomic or nanoscale phenomena that are not easily accessed by other methods to inform our understanding of materials properties. In this chapter, we aim to demonstrate how in situ microscopy has enhanced our understanding of dynamic phenomena associated with phase transformations, catalysis, crystal growth, liquid-phase processes, electrical and mechanical properties, magnetism, and ferroelectricity. We also address dynamic processes induced by the electron beam, whether intentional or not. In many cases, in situ research is driven by developments in novel instrumentation and testing methodologies. We discuss how advances in aberration correction, sample preparation and manipulation, detector development, and data analysis enable new types of in situ experiments and provide fundamental insight into dynamic materials phenomena.

3.1 A Working Definition of In Situ Transmission Electron Microscopy

Since the earliest days of transmission electron microscopy (), microscopists have understood the potential of microscopy for studying dynamic processes. Images recorded sequentially can be used to track the changes caused by deliberate actions such as heating or straining, or the unavoidable effects of electron irradiation. The class of experiments where a specimen is changed or acted on while it remains under observation (i. e., in situ in the polepiece) is referred to as in situ microscopy. In a sense, every TEM observation is an in situ experiment, since every specimen is affected by the electron beam. But the in situ microscopist aims to act on the specimen in a deliberate way and learn something from the results.

In the most exciting in situ experiments, a controlled change is made to the specimen's environment and is correlated with the resulting change in structure, measured using any of the imaging, analysis or diffraction techniques available, or electronic, optical or mechanical properties, which can also be measured in situ. Preferably both the input, in other words the change in sample environment, and the output, or consequent change in structure or properties, are recorded simultaneously and quantitatively. Given sufficient attention to artifacts, a quantitative understanding of a fundamental physical process can be obtained.

There are numerous advantages to performing experiments in situ. A single experiment gives a continuous view of a process, so may take the place of multiple post mortem measurements. One heating experiment, for example, can provide information that would otherwise have to be extracted by examination of many samples which had been annealed to different temperatures or for different times. Because of microstructural inhomogeneities in comparing before and after states from different samples, in situ observations of before and after from the same region can often provide a clearer picture of a specific phenomenon. Because an in situ experiment is continuously recorded, it is easier to catch a transient phase or observe a nucleation event. In situ experiments can yield specific and detailed kinetic information, measuring for example the motion of individual dislocations under known stress, or the growth rates of individual nanocrystals. Properties can be determined for well-characterized nanostructures, such as the conductivity of single nanotubes or the melting point of precipitates. Finally, growth or catalysis experiments provide a window into the behavior of materials under somewhat realistic processing conditions, since significant changes can occur if we remove a material from the reaction chamber and perform analysis post mortem.

Although in situ experiments provide unique information, this is at the cost of increased experimental complexity. Careful design of the specimen is necessary to minimize thin-foil effects. Tests must be carried out to understand beam effects, and calibration of the applied stimulus is critical.

Given the complexity of many in situ experiments, what is remarkable and inspirational is the variety of materials and phenomena that have been studied in situ. In the majority of experiments, the input to the specimen may be simple beam heating or controlled sample heating. Cooling or straining, application of a voltage or a magnetic field, or even modification with a scanning probe tip are more complex possibilities. For these experiments, a specially designed sample and holder are used which include the required capabilities: heater, electrical contacts, probe tip or mechanical straining. Such experiments can be carried out in most standard microscopes, apart from those with the tiniest polepiece gaps; a side entry design is required to accommodate the sample holder feedthroughs. A second, less common class of experiments is based on changing the environment of the sample by, for example, exposing it to a reactive gas or depositing another material onto it. For such studies, two experimental strategies are possible. The closed-cell approach uses a conventional microscope but achieves environmental control via a modified sample holder in which the sample and reactive environment are enclosed between two electron-transparent windows. The open-cell approach involves modifying the microscope itself, for example by adding gas feedthroughs to the specimen area. The sample is then exposed to the desired environment without the need for windows. A subset of these open-cell experiments involves reactive surfaces for which a clean environment is important. In this case, the entire microscope must then be designed for ultrahigh vacuum (). A UHV sample region allows atomically clean surfaces to be prepared (for example by heating) and then modified controllably in the polepiece. True UHV microscopes are rare as they represent a large investment. They often include side chambers attached to the microscope in which other preparation or deposition treatments can be carried out ex situ.

In terms of collecting the output data, some in situ experiments require atomic-resolution imaging, while others use lower resolution strain or defect imaging, diffraction analysis, or analytical techniques such as elemental mapping. The speed and fidelity of recording data has dramatically increased over the last decades, from analog video tape to charge-coupled device (CCD) recording to modern direct electron detectors capable of recording at more than \({\mathrm{1}}\,{\mathrm{kHz}}\). Measurements of the applied stimulus (say sample temperature, gas pressure, or applied force), and the materials property of interest (say electrical conductivity) need to be collected simultaneously and correlated with the images, diffraction or spectroscopic measurements. This stream of correlated data is extremely powerful but represents a challenge in handling and analysis.

In situ techniques are, in our view, the most exciting frontier for electron microscopy. Instrumental developments have stimulated massive changes in what is possible. The rapid progress is conveyed in several recent general reviews [3.1, 3.2, 3.3] and books [3.4, 3.5]. In this chapter, we describe some of the successes of in situ microscopy in improving our understanding of materials properties and processes. In situ microscopy has a rich history but here we focus on more recent activities, and also consider general experimental requirements for in situ experiments and the recognition and elimination of artifacts. We hope to show how widely in situ microscopy has enhanced our understanding of phenomena associated with phase transformations, crystal growth, electrical and mechanical properties, magnetism and ferroelectricity, implantation and beam effects, and a growing variety of processes that take place in the liquid phase. Improvements such as the larger polepiece gap made possible by aberration correction, more sophisticated data detection and analysis techniques, and enhanced abilities to fabricate holders and specimens of controlled geometry, promise to continue the trajectory of in situ microscopy even further.

3.2 Phase Transformations

The largest body of work accomplished using in situ TEM techniques has been in the area of phase-transformations: melting and crystallization, transformations between crystal structures, and the formation of new phases by solid-state diffusion driven by temperature, voltage, or mechanical strain. An understanding of such transformations is scientifically interesting and technologically essential in, for example, the processing of alloys, the development of new materials having extreme hardness or superplastic, magnetic or shape-memory properties, or development of rechargeable batteries with improved properties. In situ TEM has provided detailed information on the mechanism, kinetics, and structures produced during many phase transformations, both in the bulk and in nanoscale volumes. Microscopy is well suited for such studies because its high resolution allows atomic motions to be visualized; analytical techniques determine the identity of atoms during transformations, and diffraction identifies the phases present under changing conditions. Small precipitates or nuclei can be characterized and their evolution followed, and defects and complex or incommensurate structures analyzed.

The requirements for phase-transformation studies can be as simple as time-resolved imaging and a heating stage. More complex experiments involve cooling, straining, deposition, or application of a voltage. The sample may be in thin-film form or based on a microfabricated platform. It may be an aggregation of nanoparticles forming a thin layer, or a freestanding structure such as a nanowire connected at one or more points. In some cases, the electron beam triggers the phase transformation. More controllably, heating is through a furnace-type sample holder or a resistively heated track on a microfabricated substrate, directly by current flow through the sample, or by using particles of the material of interest attached to a resistively heated wire. Voltage-driven phase transformations involve a variety of sample geometries based around thin films or single nanoparticles. For many phenomena triggered by temperature, it is often possible to choose conditions that give a reaction velocity that is measurable at the time resolution available. Accurate measurement of the sample temperature is a challenge but is essential in obtaining quantitative information, such as activation energy, for reactions carried out in situ. For example, the temperature measured at the furnace by a thermocouple is not the same as the temperature at the region under observation. Methods have therefore been developed to map temperature locally using, for example, the known melting point of In particles [3.6] or calibrated shifts in the plasmon peak through electron energy-loss spectroscopy () [3.7]. However, these methods are limited to particular types of sample and a general solution is not available. An analogous problem for voltage-driven transformations is the difficulty of measuring the voltage at the region under observation. Calibration is a significant problem for electron microscopy, but innovations in microfabricated samples promise better reproducibility and control of the conditions at the region under observation.

In the rest of this section we will discuss transformations in bulk, thin-film materials and nanoparticles triggered by temperature or, less commonly, voltage. We start with solidification, melting, amorphization, and grain growth then discuss reactions between different materials, such as solid-state diffusion reactions. We highlight the examples of silicide formation, battery-relevant reactions, and memory applications. We then examine transformations that take place in nanoscale volumes of material. Comparing these with experiments in larger volumes, a recurring theme will be the finding that small volumes do not transform under the same conditions as larger ones due to the different geometry of phase boundaries, strain states and diffusion pathways and the proximity of the surface. Understanding such effects is critical for materials development.

3.2.1 Crystallization, Melting, and Grain Growth

Amorphization and Crystallization

The crystallization of amorphous materials is an interesting and important process which is uniquely suited to TEM analysis. Initial, elegantly simple experiments involved the recrystallization of silicon, deposited as an amorphous thin film and then heated in cross section in a high-resolution TEM, and showed the power of high-resolution imaging at high temperature [3.10, 3.9]. The nucleation of crystallites was visualized, allowing estimation of the critical nucleus size, and the irregular progress of the reaction front was demonstrated, even though macroscopically the kinetics were consistent with a more continuous ledge mechanism. This pioneering work provided a new view of bulk phase transformations, showing the start–stop motion that is now familiar at the atomic scale. Using multilayer specimens to extend this approach to metal-mediated crystallization of Si, Ge or C demonstrated the mechanism for these reactions as well (Fig. 3.1a-e [3.11, 3.12, 3.13, 3.14, 3.8]). Si crystallization has now been so well studied, both in situ and ex situ, that it can even be used as a calibration tool to measure the temperature in thin specimens [3.15, 3.16]. More recent crystallization studies use plan view rather than cross-sectional geometry allowing many grains to be imaged in materials such as Ta [3.17], the shape-memory alloy NiTi [3.18, 3.19], fluorochlorozirconate glass-ceramics [3.20], and perovskites used for next-generation solar cells [3.21]. Advanced techniques are used to characterize the crystallization process, such as fluctuation electron microscopy, which shows composition fluctuations and short- and medium-range order in a metallic glass as the crystallization temperature is approached [3.22].

Fig. 3.1a-e

Metal-induced crystallization: In situ high-resolution image series recorded during the Ag-mediated crystallization of Ge at \({\mathrm{250}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). The time between frames is \({\mathrm{8}}\,{\mathrm{s}}\). The Ag is the faulted region in the center and the crystalline Ge is in the upper left. The Ag crystal appears to migrate towards the amorphous Ge region but the faults remain fixed (one fault is indicated by a line). The inference is that Ge is supplied by diffusion through the Ag lattice, and the net motion of the Ag is caused by counterdiffusion of Ag atoms. The lack of any amorphous eutectic is clearly demonstrated. From [3.8], reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com/)

A crystallization process relevant to information storage is shown in Fig. 3.2a-g. Phase-change materials such as GeSb and GeSbTe store bits of information as amorphous areas embedded in crystalline regions. A high laser power is used to write amorphous spots, a medium power erases by recrystallizing, and a low power (or other measurement) reads the bits. Crystallization has been measured in films deposited on SiN membranes [3.23, 3.25], freestanding films [3.26], and actual compact disc materials [3.24]. Beam heating shows nucleation and growth kinetics (Fig. 3.2a-ge–g), while more controlled heating experiments measure activation energies (Fig. 3.2a-ga–d). \(\mathrm{SbO_{\mathit{x}}}\) is another potential phase-change material also examined in situ for different stoichiometries \(x\), determining activation energy [3.27]. Although stress effects may change the kinetics in electron-transparent foils, these experiments are useful in allowing transformation parameters to be measured and structural changes to be examined.

Fig. 3.2a-g

Amorphous to crystalline transformation in phase-change materials. (ad) Bright-field TEM images recorded during crystallization of a \({\mathrm{40}}\,{\mathrm{nm}}\) \(\mathrm{Sb_{3.6}Te}\) film on a silicon nitride membrane, at \({\mathrm{85}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) in a heating stage. The growing crystal was prenucleated by heating for \({\mathrm{5}}\,{\mathrm{min}}\) at \({\mathrm{95}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). Reprinted from [3.23], with the permission of AIP Publishing. (eg) Bright-field images displaying stepwise electron irradiation-induced crystallization of an amorphous data mark in \({\mathrm{14}}\,{\mathrm{nm}}\)-thick \(\mathrm{Ga_{15}Sb_{85}}\) after (e\({\mathrm{60}}\,{\mathrm{s}}\), (f\({\mathrm{195}}\,{\mathrm{s}}\), and (g\({\mathrm{226}}\,{\mathrm{s}}\) irradiation at a current density of \({\mathrm{1.5}}\,{\mathrm{nA{\,}mm^{-2}}}\). The specimen was made from a CD-RW/DVD1RW disk consisting of a \(\text{ZnS}{:}\mathrm{SiO_{2}}/\text{GaSb}/\text{ZnS}{:}\mathrm{SiO_{2}}/\text{SiN}/\text{Ag}/\text{SiN}\) layered stack on a polycarbonate substrate, with all layers removed except for the GaSb and surrounding dielectric layers. The phase-change layer was crystallized using a broad laser beam then amorphous data marks were written using a home-built compact disk (CD) or digital versatile disk (DVD) recorder. Reprinted from [3.24], with the permission of AIP

The reverse process of solid-state amorphization can be hard to measure using other experimental techniques. In situ heating of systems such as Ti-Si, Zr-Si, Pt-GaAs [3.28], and Al-Pt [3.29] allow nucleation locations to be determined and diffusion processes to be characterized. In situ observations have led to insights into the role of electrical wind force and defects on the amorphization in phase-change nanowires [3.30]. Mechanically induced crystalline to amorphous transformations can be measured in irradiated Si [3.31]. Amorphization can also be caused by the electron beam, discussed in Sect. 3.7.

The Solid-to-Liquid Transformation and the Structure of the Solid–Liquid Interface

Melting and freezing can be observed in situ by diffraction or imaging. Perhaps the ultimate example is the nanothermometer shown in Fig. 3.3a-e, fabricated by enclosing Ga in a large-diameter carbon nanotube () [3.32, 3.33]. This structure was used to measure the expansion coefficient of liquid Ga and observe different structures on freezing. Several other transformations involving liquids have also been studied in situ. It may at first appear surprising that liquids can be examined at all. However, liquids with low vapor pressure, such as Ga, In, Si or Al, may be imaged in the same way as solids, provided they do not move around too much. If too mobile, they may be coated with a surface layer to maintain shape [3.34, 3.35]. Such an approach yields size-dependent melting points [3.35]. Liquids with high vapor pressure require encapsulation in a closed liquid cell as discussed in Sect. 3.6.

Fig. 3.3a-e

Ga thermometer: Ga contraction and expansion inside a carbon nanotube upon cooling and heating. The background feature is part of the carbon support film. (a) At room temperature, \({\mathrm{21}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\), (b) at \({\mathrm{-40}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\), (c) at \({\mathrm{-80}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\), when solidification occurred. (d) The crystallized Ga was melted at \({\mathrm{-20}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). (e) Reheated to room temperature, \({\mathrm{21}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). Reprinted with permission from [3.32]. Copyright 2004 by the American Physical Society

Theory predicts interfacial ordering in liquid adjacent to a crystal. However, experimental data is difficult to obtain using diffraction techniques, especially for interfaces that are faceted or have small lateral extent. Temperature-controlled TEM can probe both the static structure of the solid–liquid interface and the transformation between solid and liquid. The solid-to-liquid transition in Xe films has been measured by diffraction in an environmental cooling cell [3.36]. Imaging studies, in some cases including energy-filtered or aberration-corrected imaging, show the persistence of order into liquids, for example at the crystal-to-liquid interface in PdSi [3.37], Si [3.38] and AlSi [3.39], and at the interfaces of Xe in Al [3.40] and Al on sapphire [3.41, 3.42]. In this last case, the effect of facet structure on ordering can be distinguished [3.42]. Ledge flow is visible at the AlSi–Si interface [3.43, 3.44], during surface melting in Al and growth of alumina [3.41, 3.45, 3.46, 3.47]. Other examples of in situ imaging of solid–liquid interface dynamics involve the growth of nanowires from eutectic liquid droplets (Sect. 3.3.4) and size-dependent transformations of nanoparticles and inclusions (Sect. 3.2.3).

Grain Growth and Grain-Boundary Motion

Polycrystalline materials change their structure in different ways on annealing or mechanical deformation. Grain-boundary motion during mechanical deformation is discussed in Sect. 3.5.1. In situ heating experiments have examined dewetting [3.48] and grain-boundary motion [3.49, 3.50, 3.51] for metals such as Cu, Au, and Al. Such measurements help to clarify mechanisms, measure activation energies, and even determine the effects of impurities and gas atmosphere. As an example, kinetic parameters measured for nanocrystalline Ag thin films demonstrated that grain growth is dominated by surface diffusion mass transport [3.50]. More complex grain-boundary dynamics can also be studied. An interesting example is the penetration of liquid Ga along grain boundaries in Al, relevant to embrittlement [3.52, 3.53]. The structure and strain field during penetration, the kinetics in different grain orientations, and the effects of dislocations were observed.

Dark-field imaging works well for grain-boundary dynamics in polycrystalline films or at low-symmetry boundaries, but high-resolution heating experiments are useful if there is a symmetrical relation between grains. High-resolution experiments on bicrystals having engineered boundaries with high symmetry, particularly in Au and Cu [3.54, 3.55, 3.56, 3.57, 3.58], enable detailed measurements for the determination of grain-boundary migration mechanisms. Comparison of images at different times can quantify stochastic movements [3.59]. In situ experiments show that collective mechanisms operate during migration, and that unusual structures may form and grow at boundaries (Fig. 3.4a-e). Dislocations may also be emitted, and the details of their structure and relationship with the boundaries can be measured [3.60].

Fig. 3.4a-e

Grain-boundary dynamics: HREM images of a [110] \(\theta=14^{\circ}\) tilt grain boundary in Au at \({\mathrm{893}}\,{\mathrm{K}}\). Individual frames are shown from a video sequence recorded near optimum defocus. (a) GB at \(t={\mathrm{15.37}}\,{\mathrm{s}}\). (b) GB has moved to the right at \(t={\mathrm{44.93}}\,{\mathrm{s}}\) and is near the \((6,-6,1)\) symmetric orientation. (c) Detail of (a) depicted at four different times. A small region composed of eight atomic columns switches orientation between the two grains. Note the stacking disorder and misfit localization at the dislocation cores. Reprinted with permission from [3.54]. Copyright 2002 by the American Physical Society. (d,e) High-resolution image of the \(\text{Cu}\Sigma=3\) interface imaged along \([01\bar{1}]\). The grain boundary is dissociated into a narrow slab of 9R stacked material (fcc stacking but with an intrinsic stacking fault inserted every three planes). (d,e) were recorded \({\mathrm{5}}\,{\mathrm{min}}\) apart after \({\mathrm{400}}\,{\mathrm{kV}}\) electron-beam irradiation and the 9R stacked region has expanded due to changes in the internal stress state induced by the beam. Stacking defects in the 9R structure can be related to the presence of secondary grain-boundary dislocations at the interface. Reprinted from [3.55], with permission from Elsevier

3.2.2 Solid–Solid Transformations

As well as melting, solidification and grain-boundary motion, in situ techniques have been applied to understand transformations between different crystal structures and solid-state reactions involving diffusion. These experiments mostly involve heating of thin films, although transformations have also been initiated by straining, electron-beam heating, electric and magnetic fields and the gas environment, and the materials imaged include nanostructures as well as thin films. High-resolution imaging and analysis, diffraction, or low-resolution weak-beam or bright-field imaging provide information that is complementary to that obtained from other in situ techniques, such as x-ray diffraction, which average over larger volumes.

Diffusionless Transformations

Subtle changes in symmetry can be detected using the sensitive combination of diffraction and high-resolution imaging. These techniques show the transformations between orthogonal, tetragonal and cubic phases in oxides such as \(\mathrm{SrRuO_{3}}\) [3.61]. They also work well for transformations involving charge ordering and incommensurate phases, discussed in Sect. 3.4.4, and changes in structure stimulated by the electron beam, discussed in Sect. 3.7. When studying such phase transformations it is important to consider the experimental artifacts already mentioned for melting and solidification. The examples below illustrate the advantages and some pitfalls of in situ TEM.

Transformations in intermetallic alloys provide an excellent opportunity for in situ microscopy to display its power. For example, for TiNi, the orientation relation between the different phases can be determined, and the dynamics of the emergence of martensite plates during straining can be observed in situ [3.18, 3.63, 3.64]. In situ heating of NiAl alloys [3.62, 3.65] showed how the texture and defect structure in the high-temperature phase are derived from the lower temperature phase, and illustrated the processes occurring during phase decomposition with low-resolution imaging and diffraction (Fig. 3.5). In situ heating has shown phase transformations in Ni-based superalloys induced by oxidation [3.66, 3.67] and changes in ordering at very high temperature [3.68]. Higher resolution imaging showed details of the formation of the gamma phase in TiAl, such as ledge motion at interfaces [3.69]. Other materials examined include the shape-memory alloys CuAlMn [3.70], TiNiHf [3.71], and FeMnSi [3.72] and MgZn alloys [3.73]. Crystallographic relationships, the interaction of dislocations with the transformation front, and the morphology of phases produced on heating or straining were studied. The presence and significance of incommensurate reflections in related materials has been examined using imaging plates and an in-column filter [3.74, 3.75, 3.76, 3.77]. Cooling stages allow an even greater range of transformations to be accessed [3.78].

Fig. 3.5

Phase stability in NiAl: When a martensitic Ni-rich \(\mathrm{Ni_{\mathit{x}}Al_{1-\mathit{x}}}\) sample, with \(x> {\mathrm{63}}\,{\mathrm{at.\%}}\), is annealed at moderate temperatures (\({\mathrm{550}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\)), it transforms into \(\mathrm{Ni_{5}Al_{3}}\). On further heating to \({\mathrm{780}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\), the \(\mathrm{Ni_{5}Al_{3}}\) phase itself decomposes, forming B2 grains in a twinned \(\mathrm{L1_{2}}\) matrix. This image is part of a sequence obtained during heating that shows a B2 grain growing by forming a small extension (marked as X) into the \(\mathrm{L1_{2}}\) matrix, consuming some twins, then rapidly expanding laterally. The \(\mathrm{Ni_{5}Al_{3}}\) phase is undesirable as it degrades the shape-memory properties by inhibiting the transformation back to austenite, and its formation and stability are therefore important. Reprinted from [3.62], with permission from Elsevier

It is also possible to study mechanically induced phase transformations in situ. For example, the stress-induced B2 to B19\({}^{\prime}\) (austenite to martensite) phase transformation was imaged directly [3.79] by recording diffraction patterns during nanocompression testing in real time. This was the first direct evidence that the transformation does exist in NiTi even when the sample size is below \({\mathrm{200}}\,{\mathrm{nm}}\). Correlation of the appearance of the B19\({}^{\prime}\) phase in the diffraction patterns with quantitative data showed that the transformation started at approximately \({\mathrm{1}}\,{\mathrm{GPa}}\) and occurred through a multistep process. Another mechanically induced phase transformation [3.80] involved individual \(\mathrm{VO_{2}}\) nanowires in the TEM strained in tension to drive an M1–M2 structural phase transition. The same sample was strained in situ in a synchrotron x-ray microdiffraction beamline with the same mechanical testing holder, producing a similar result and demonstrating a correlative method to complement the TEM observations.

Thin-foil effects can be important in these transformations. The sample thickness influences the sequence of phases [3.64] and transformation temperatures [3.81]—indeed, transformations do not occur at all at some thicknesses. Electron irradiation can induce transformations (for example in NiMnTi [3.82]) or change kinetics (for example in Ti-Mo alloys on cooling [3.83]). Beam and thin-foil effects are relevant to any in situ transformation experiment. Beam effects should be evaluated by examining unirradiated areas after the transformation. Thin-foil effects can be minimized in some cases by depositing the material of interest onto an electron-transparent membrane. This reduces buckling and provides a more uniform temperature than a conventional specimen of varying thickness, advantageous for quantitative studies [3.18, 3.19, 3.23, 3.50, 3.84, 3.85, 3.86].

Silicide Formation Reactions

Reactions at a planar interface between two solid materials have provided fruitful subjects for in situ TEM. In situ observations allow determination of the diffusing species, the nature of nucleation sites, the sequence of phases, and the relationship between the crystal structures of the initial and final phases. However, we have to be particularly careful to avoid artifacts. For example, if the sample dimensions are comparable to or less than the diffusion lengths of the moving species then surface diffusion may affect the kinetics. Surface nucleation sites may dominate, and beam effects and stress relaxation in thin regions of the foil must be considered. In spite of these issues, a successful body of work has been carried out on these transformations. We illustrate this by discussing silicide formation, a phenomenon of great relevance to the microelectronics industry that has been examined using a range of in situ TEM techniques.

In situ silicidation was initially studied in cross section by heating a metal film such as Ti, Zr, or Cr deposited on Si [3.87, 3.88, 3.89, 3.90] (Fig. 3.6a-da). Plan-view experiments provided the opportunity to examine silicidation by heating metal-implanted Si [3.91] or patterned substrates to study, for example, nucleation-limited transformations in small areas [3.92, 3.93, 3.94] (Fig. 3.6a-db). Structural transformations between silicide phases can also be seen, for example in copper silicides imaged in plan view [3.95]. A third sample geometry is the freestanding nanowire. A mixture of metal nanoparticles and pregrown (oxidized) silicon nanowires is distributed on a grid. At points where metal and silicon make contact, silicide formation takes place on heating. The progression of the silicidation front can be followed along the nanowire to derive nucleation parameters and examine strain effects [3.96, 3.97, 3.98].

Fig. 3.6a-d

Silicide phase transformations: (a) Cross-sectional image series recorded during heating of a \({\mathrm{50}}\,{\mathrm{nm}}\) \(\mathrm{CoSi_{2}}\) film on Si(001), showing interface roughening, protrusions, and eventual pinhole formation. From [3.90], reproduced with permission. (b) Silicide formation in a patterned area. Si wafers were covered with an oxide in which lines were patterned, and then \({\mathrm{12}}\,{\mathrm{nm}}\) Ti was deposited over the wafer. The successive plan-view images show the formation of Ni silicides at different temperatures in an \({\mathrm{800}}\,{\mathrm{nm}}\)-wide line: formation of \(\mathrm{NiSi_{2}}\) (flower-like contrast at \({\mathrm{300}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\)); \(\mathrm{NiSi_{2}}\) pyramids are well formed at \({\mathrm{370}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\); after \({\mathrm{10}}\,{\mathrm{min}}\) at \({\mathrm{400}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) some NiSi has formed in the center of the line but \(\mathrm{NiSi_{2}}\) remains along the edges. Reprinted from [3.92], with the permission of AIP Publishing. (c,d) The C49–C54 phase transformation in \(\mathrm{TiSi_{2}}\) where the Ti was deposited and annealed in situ (c), compared to ex situ deposited \(\mathrm{TiSi_{2}}\) (d). Auger spectroscopy showed the presence of oxygen in the second case. The arrow shows a fixed point on the specimen. In the clean film the phase transition occurs smoothly while in the oxidized film it is strongly pinned at grain boundaries. From [3.99]

Together, these in situ experiments provide a detailed view of the morphology, sequence, and kinetics of silicide formation. Some of the phases are short-lived or hard to see otherwise: as mentioned previously, a single in situ experiment can replace a whole series of ex situ preparations [3.100]. However, choosing the optimal sample geometry is important. In a cross-sectional experiment, for quantitative results any surface diffusion pathways must be suppressed (perhaps by coating the sample) and nucleation sites on the milled surface minimized. The in situ experiment is only an accurate representation of the bulk situation if both the activation energy of the reaction and the final structure produced are comparable with bulk experiments [3.101]. In plan view, surface effects are not as significant but thin-film buckling must be considered. Nanostructured reaction volumes provide an opportunity to control nucleation sites.

Specialized in situ deposition techniques offer an interesting alternative way of looking at silicide formation. Rather than creating the metal/silicon sample ex situ, a silicon substrate is cleaned in situ in a  (ultrahigh vacuum transmission electron microscope) and the metal then deposited in situ. The metal may be deposited onto a cool substrate which is then heated [3.102, 3.99] or deposited at high temperature so that silicide phases form at once as islands [3.103, 3.104, 3.105]. A combined UHV system allowing sequential TEM and STM imaging has been developed to study the structure of such three-dimensional islands in more detail, determining surface reconstruction as well as the sequence of phases [3.106]. In situ deposition has also been used to study more complex silicide reactions, such as oxide- and nitride-mediated epitaxy [3.107, 3.108]. Despite the experimental challenges, the advantages are clear in terms of avoiding contamination or oxidation (or evaluating their effects; see Fig. 3.6a-dc) and revealing kinetic effects such as coarsening that take place during deposition.

Battery-Relevant Transformations

In a battery, energy is stored and released through changes in the chemical environment of atoms as they move within the structure. Charging and discharging the battery may involve insertion or extraction of ions from a host lattice or an exchange between dissolved ions and a solid. Developing new batteries or improving the performance of existing ones relies on deep understanding of these types of reactions. In situ TEM provides a unique approach to following phase transformations such as solid-state diffusion reactions and reactions at the solid–liquid interface.

In lithium-ion batteries the energy is stored by transfer of Li ions in and out of the electrode materials. Figure 3.7a-e presents two in situ TEM examples. Figure 3.7a-ea [3.109] shows changes in the lithiation state of individual nanocrystals in a thin film of \(\mathrm{Fe_{3}O_{4}}\) dispersed on a conductive substrate. Analysis of the symmetry of the lattice image shows the transformation of magnetite (with the spinel structure) to \(\mathrm{Li_{\mathit{x}}Fe_{3}O_{4}}\) with a rocksalt structure. Figure 3.7a-eb [3.111] shows lithiation of a single nanostructure, in this case an Si nanowire. Similar experiments have addressed reactions at anode and cathode materials as well as formation of the solid electrolyte interphase (SEI) layer, a thin solid film composed of organic and inorganic components that forms and changes during the operation of the battery as the electrolyte decomposes.

Fig. 3.7a-e

Phase transformations in battery materials: The transformation from spinel \(\mathrm{Fe_{3}O_{4}}\) to rocksalt structure \(\mathrm{LiFe_{3}O_{4}}\) during in situ lithium intercalation at \(\approx{\mathrm{150}}\,{\mathrm{mA/g}}\) by contact with a W tip coated in Li, with a surface \(\mathrm{Li_{2}O}\) layer acting as the solid electrolyte. (a,b) HRTEM images of an \(\mathrm{Fe_{3}O_{4}}\) single crystal showing the spinel and rocksalt phases. The insets show the FFT of the spinel and rocksalt structures along the \([0\bar{1}1]\) zone axis. (c,d) Filtered images of (a,b) using two sets of spinel (red) and rocksalt (green) diffraction spots to indicate the corresponding phase distribution. From [3.109]. (e) Lithiation of a single-crystal \(\mathrm{SnO_{2}}\) nanowire anode during charging at \({\mathrm{-3.5}}\,{\mathrm{V}}\) against an \(\mathrm{LiCoO_{2}}\) cathode. A schematic of the experimental setup is above, with images at the times indicated showing the initially straight nanowire twisting and bending on charging. Small arrowheads show the position of the reaction front. The red line marks a reference point. After \({\mathrm{1860}}\,{\mathrm{s}}\) the nanowire had elongated by \({\mathrm{60}}\%\) and increased in diameter by \({\mathrm{45}}\%\), a \({\mathrm{240}}\%\) volume expansion. From [3.110]. Reprinted with permission from AAAS

The driving force for these reactions is voltage, rather than temperature as in previous examples. The voltage is supplied by an external circuit connected through the sample holder. Several designs have been developed to apply the electrical bias across the sample. In Fig. 3.7a-ea an STM tip touches a conductive, thin plan-view sample. In Fig. 3.7a-eb, the sample is mounted on a piezo-driven sample holder that is moved to make contact with an electrode composed of an ionic liquid droplet. A third configuration has electrodes patterned on an insulating membrane and covered by a liquid electrolyte in a closed liquid cell, discussed in more detail in Sect. 3.6. Supplying the relevant materials to the reaction zone depends on the experimental configuration. In all-solid-state experiments such as that in Fig. 3.7a-ea, Li is coated on the W tip and a surface lithium oxide layer acts as a solid electrolyte [3.109, 3.112]. In experiments involving ionic liquid droplets, Li ions are present in the ionic liquid [3.110]. In closed liquid cells, the ions are dissolved in an aqueous solution (Sect. 3.6.2). The utility of these experiments comes from correlating the observed phase transformations with the electrical signature of the sample. Thus, it is useful to incorporate multiple electrodes including a reference electrode.

For reactions at anode materials, questions of interest involve the mechanism of the lithiation reaction and the effect of nanoparticle shape and size on stress generation and (anisotropic) volume expansion. In situ TEM has probed the movement of lithium in and out of anode materials such as Si, Ge, graphite, and tin oxide. The amorphization front, the formation of defects, and the changes on reversal are visible. The transformation of crystalline Si to amorphous \(\mathrm{Si_{\mathit{x}}Li_{\mathit{y}}}\) is directly visible through image contrast (Fig. 3.7a-eb), high-resolution imaging [3.113], and elemental mapping [3.114]. Cathode materials are typically oxides such as \(\mathrm{LiFePO_{4}}\). In situ experiments address the mechanism of the phase transformation on lithiation, for example whether it occurs via phase boundary motion or solid solution. These phase transformations are measured through image contrast and diffraction [3.112] or through analytical techniques, including EELS and holography [3.115].

Although these experiments have generally focused on lithium battery reactions, increasingly diverse materials systems are being examined. Several reviews provide information on setting up such experiments and interpreting the results [3.116, 3.117, 3.118, 3.119] and give a sense of the flexibility of such experiments for general study of voltage-driven phase transformations.

Memory-Relevant Transformations

Phase-change memories, which use an amorphous/polycrystalline transformation to encode bits of data, were described in Fig. 3.2a-g. Another type of memory involves resistance switching, where the resistance of a device containing a thin insulating layer is altered by an applied electric field. In situ TEM experiments provide a means to investigate the mechanism of this effect, using strategies related to those in Sect. 3.2.2. In some materials systems, a metal filament is observed to form on biasing, due to diffusion of metallic ions from an electrode [3.120, 3.121, 3.122, 3.123, 3.124, 3.125]. In other materials, oxygen vacancy diffusion is identified as the mechanism that changes the resistivity of the insulating layer [3.126, 3.127, 3.128]. The mechanism may depend on the microstructure [3.129]. The changes in resistivity may be accompanied by structural transformations such as detwinning and superstructure formation [3.127], Wadsley defects and Magnéli phases [3.130] or formation of a thin interfacial reaction layer [3.131].

These experiments have made use of an interesting variety of sample geometries. The layered structure of the device can be viewed edge-on with voltage applied via a movable tip. Alternatively, a device may be attached by FIB onto a microfabricated substrate, or a device can be patterned on a membrane and imaged in plan view. A fascinating variety of TEM tools has been applied, including EELS, (x-ray energy dispersive spectroscopy), and even electron holography to measure the potential distribution changes on switching [3.128]. The phase transformations that take place during switching appear to be materials and condition dependent, and in situ techniques provide a tool of great importance for evaluating new materials and improving device performance.

3.2.3 Size-Dependent Transformations in Nanomaterials

The phase transformations discussed above were generally imaged in bulk materials or thin films, where in situ microscopy visualizes structural changes, nucleation and growth fronts. But as we have seen, particularly with voltage-driven transformations, individual nanostructures and small regions within a larger structure are also accessible. Thus, the techniques applied to transformations in bulk materials can naturally be extended to transformations in small volumes, either embedded in a matrix or freestanding. In the following paragraphs we discuss transformations and stability in small volumes. These experiments confirm the important general conclusion that small particles show different phase diagrams compared to larger volumes of the same material. This is especially important given the many applications of nanostructured materials—for example, in high-strength metal alloys—and individual, freestanding nanoparticles—for example, as catalysts or as components in advanced electronic devices.

Embedded Nanostructures

By focusing on an individual inclusion or precipitate, in situ microscopy provides precise information for nanoscale volumes that can not be obtained otherwise. Excellent quantitative work in several systems shows the opportunities for a wider range of materials.

Pb in Al is a model system where the lack of solubility of Pb in Al means that Pb forms small cuboctahedral inclusions spontaneously with a cube-on-cube orientation relation. Heating experiments allow strain, melting, and diffusion phenomena to be studied with a fascinating range of size-dependent effects (Fig. 3.8a-j). Melting of the Pb is size-dependent with huge supercooling possible, and there is a hysteresis on solidification due to the difficulty of nucleating ledges [3.134]. Strain fields during solidification and melting provide information on the diffusion of point defects [3.135]. In particles at grain boundaries, which have complex structures, each interface melts at a different temperature [3.133, 3.136]. Coimplantation of different materials into Al, such as Cd\(/\)Pb, Sn\(/\)Pb or Tl\(/\)Pb, allows phenomena associated with phase separation, melting, and interface structure to be examined [3.137] and binary phase diagrams determined as a function of size.

Fig. 3.8a-j

Nanoparticle melting phenomena: (a) Size-dependent melting of Pb inclusions in Al. The sample was produced by rapid solidification of an Al-0.5% Pb alloy and the image shows an array of particles at \({\mathrm{423}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\), which is \({\mathrm{96}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) above the bulk melting point. The rounding of most particles indicates their liquid state, while the smallest particles (arrow) are still faceted and solid. (b) By measuring the dependence of inclusion shape on temperature and considering the inclusion shape change kinetics, the step energy as a function of temperature for steps on the inclusion surface can be calculated. The least squares fit indicates a roughening transition at about \({\mathrm{600}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). Reprinted from [3.132], with permission from Elsevier. (cj) Reversible melting of a \({\mathrm{25}}\,{\mathrm{nm}}\) Pb inclusion at a grain boundary in Al. This particle has two different interfaces with two different grains and the two interfaces melt at different temperatures. The thin black line indicates the solid–liquid interface at different temperatures. From [3.133], reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com)

Phase transformations involving precipitate growth have also been measured in situ. In cases where precipitates are pinned on dislocations, diffusion parameters can be measured from their coarsening [3.138] or motion [3.139]. The kinetics of ledge motion and kink nucleation on precipitates can be observed during high-resolution heating experiments [3.140]. For example, for precipitate plates in Al-Cu-Mg-Ag alloys, imaging parallel and perpendicular to the interface demonstrated that precipitates grow by the terrace-ledge-kink mechanism [3.141] and allowed the rate-limiting steps and thermodynamic parameters of kink nucleation to be determined (Fig. 3.9a,b). Other reactions, such as oxidation and reduction, can also be observed in precipitates [3.142, 3.143]. Such precipitate phase-transformation studies have exciting opportunities to yield even more quantitative information via high-resolution imaging and simulations [3.144], and using tomography and XEDS to measure precipitate composition and evolution [3.145].

Fig. 3.9a,b

Precipitate growth mechanism: (a) High-resolution image of a single ledge on a {111} \(\theta\) plate in an Al-Cu-Mg-Ag alloy during growth at about \({\mathrm{220}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\), and (b) graph showing the position of the ledge as a function of time demonstrating growth by irregular motion of ledges. The ledge height is two {111} matrix planes (half a unit cell). The circled area appears blurred in videos due to enhanced atomic motion there. From [3.144], reproduced with permission

Freestanding Nanoparticles

Isolated nanoparticles are interesting for their catalytic or optical properties, driving intense study of the factors determining their shape, phase stability, and sintering. As with precipitates, in situ studies show that freestanding nanostructures have properties that are different from the bulk material.

The earliest in situ studies of freestanding particles demonstrated the dynamic nature of the atomic arrangement [3.146, 3.147]. The large fraction of atoms on or near the surface indeed leads to unusual behavior. TEM has shown that freestanding nanoparticles have structure different from bulk [3.148]. Size-dependent melting can be measured [3.149], phase transformations readily take place by diffusion [3.150, 3.151], and changes are seen in phase stability [3.152, 3.153]. In this context, binary systems such as Au-Sn, Pb-Sn, Bi-Sn, and In-Sn have been extensively studied. Figure 3.10a-f shows mixed composition clusters formed in situ using a two-source evaporator. The binary phase diagram is found to depend strongly on size, with changes in the eutectic temperature [3.154, 3.155]. Melting behavior, phase separation and mixing also depend on the composition and size [3.156, 3.157, 3.158]. These effects reflect a change in solubility or the relatively high cost of forming phase boundaries.

Fig. 3.10a-f

Intermixing in small particles: (ae) Image sequence showing Sn alloying into small Bi particles at \({\mathrm{350}}\,{\mathrm{K}}\). The sample was formed by evaporating Bi onto a grid, followed by Sn evaporation in situ. (a) As-evaporated Bi particle; (be) the same particle during Sn deposition. A crystalline–liquid interface forms (arrows in (b)) and propagates through the crystal until the whole particle becomes liquid. (f) XEDS shows a composition of \({\mathrm{50}}\%\) Sn at this point. Not shown is the asymmetrical behavior of Sn particles during Bi deposition; these become liquid at once without forming a phase boundary, an abrupt crystalline-to-liquid transition that is not expected from the bulk phase diagram and which reflects the energy cost of creating an internal interface. From [3.157], reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com)

Unusual structures may form in freestanding particles when the temperature changes. In Al-Si, a solid Al particle inside a molten Al-Si sphere can form and move with fractional Brownian motion [3.159]. GaSb particles decompose into a crystalline Sb core surrounded by liquid Ga [3.160]. Mixed particles undergo interesting transformations, for example the formation of epitaxial Au–semiconductor interfaces in Au-decorated semiconductor nanowires [3.161]. Stress and anisotropic shape often play an important role in reactions involving freestanding particles. Metals encapsulated within multiwalled carbon onions have changed melting points due to the pressure [3.162, 3.163], and the metal can even migrate through the graphitic covering [3.163]. When there is a solid oxide layer covering a nanoparticle, stress relief can cause cracking [3.164]. Crystallization of metals within carbon nanotubes shows directionality related to the shape of the confined volume [3.165].

Sintering

Sintering of nanoparticles is important in materials processing and catalyst aging. The mechanisms at work, including grain-boundary motion, particle rotation and defect elimination, can be visualized directly in Au or PbSe [3.166, 3.167]. Some materials such as silicon nitride require very high temperature for sintering in situ [3.168]. The gas environment is also important since the sintering of many metals is affected by their surface condition. For example, oxide effects on Fe and Nb sintering have been explored in situ [3.169, 3.170]. More generally, the kinetics of sintering in heterogeneous catalysts, which consist of metal particles on an oxide substrate, depends on the gas environment. Controlled-environment TEM is therefore helpful to ensure the data is relevant to catalyst operation. Such experiments provide information on the effect of support, temperature, and environment [3.171, 3.172].

For sintering of pure metals onto reactive substrates, where UHV conditions are required to avoid oxidation, there are benefits to designing an integrated system where the particles are created and imaged in the same vacuum. This has been achieved by connecting a sputtering chamber to a UHVTEM [3.173], enabling the sintering of metal particles onto oxide-free metal surfaces. Cu sintering onto Cu proceeded by neck growth and grain-boundary motion, whereas Co particles on Cu and Ag foils burrowed beneath the surface to minimize surface energy [3.174].

3.2.4 Summary

In situ microscopy addresses the central physics of phase transformations related to crystallization and melting, diffusion, and the effects of defects. Bulk crystals, embedded nanostructures, and freestanding nanoparticles yield quantitative information on reaction mechanisms and on the relationship of structure to dynamics. This brief survey of results clearly shows that the in situ techniques we have described could be applied to many currently unstudied systems. However, for proper interpretation of results care must be taken with thin-foil effects. Beam effects, strain, surface diffusion, and surface nucleation can render the in situ results less relevant to the real world. The calibration of experimental conditions, particularly temperature, is important, as is choice of accelerating voltage to balance thickness and beam damage. The use of microfabricated samples provides exciting opportunities to control and calibrate temperature and combine heating with other stimuli to obtain quantitative results relevant to materials development.

3.3 Surface Reactions, Catalysis, and Crystal Growth

In situ TEM can provide unique information on reactions that modify surfaces, as well as catalytic reactions and the growth of thin films or nanocrystals. The experiments involve exposing a sample to a reactive environment while imaging under conditions that are sensitive to changes at the sample surface, rather than within its bulk, as was discussed in Sect. 3.2. The reactive environment may include high temperature, flow of reactive gas, a deposition flux, or even exposure to a liquid environment. As we might expect from the discussion in Sect. 3.2, the power of in situ microscopy in these surface studies arises from its ability to observe transient structures and measure reaction kinetics for individual nanostructures. We will show that the experiments indeed contribute to an understanding of surface reactions and growth, leading to improved design of catalysts, control of surface structure, and formation of nanostructures with particular morphology and properties.

Most studies of surface reactions, catalysis, and crystal growth take place in an environmental or controlled-environment TEM (). The column of a standard TEM contains a mildly reducing atmosphere of \(E-6{-}E-7\,{\mathrm{Torr}}\) (\(E-9{-}E-10\,{\mathrm{bar}}\)) and may also be contaminated with hydrocarbons from the sample, holder, or microscope components. By controlling the vacuum environment, the specimen can be exposed to conditions that are, for example, clearly oxidizing or reducing. Gases can be flowed that undergo catalytic reactions, a solvent-rich atmosphere can be set up to control the hydration level, or growth can be studied by supplying appropriate gas- or liquid-phase precursors. In this section, we discuss experiments involving gases. Crystal growth and interfacial reactions that require liquid-phase environments have specific experimental requirements and are discussed separately in Sect. 3.6.

Two main strategies have proven successful for controlled-environment experiments. As mentioned in Sect. 3.1, in open-cell ETEM the gases are leaked directly into the sample region of a TEM. Differential pumping throughout the microscope column ensures that the pressure remains low everywhere apart from near the sample. This minimizes the path length that the electrons must traverse through high-pressure gas and allows the electron gun to be operated normally. Pressures of several \({\mathrm{100}}\,{\mathrm{mTorr}}\) can be achieved at the sample region. However, the base pressure (lowest pressure achievable, which is a measure of the residual contamination expected) is often in the range \({\mathrm{10^{-7}}}\,{\mathrm{Torr}}\). Some open-cell TEMs are therefore designed for ultrahigh vacuum with a base pressure as low as \({\mathrm{10^{-10}}}\,{\mathrm{Torr}}\). UHVTEM offers distinct advantages for samples sensitive to background contamination, but the disadvantage is that the highest pressure achievable at the sample is around \({\mathrm{10^{-5}}}\,{\mathrm{Torr}}\). The sample temperature can be calculated as a function of gas flow and geometry [3.175]. Such microscopes can be complex and expensive but they enable experiments on reactive samples that can not be realized otherwise, especially if adjacent chambers are available for sample preparation. ETEM has a very broad range of applications, summarized in a recent book [3.176], while UHVTEM is a more specialized technique for sensitive materials.

The second approach to ETEM is the closed-cell approach. A pair of electron-transparent windows, typically formed from silicon nitride or graphene and spaced a few micrometers apart, is used to confine a gas within a narrow layer. The gas is supplied through tubes that feed through a specially designed holder. The sample, often composed of nanoparticles, is attached to the interior of one window and is heated by a microfabricated heater. Closed-cell ETEM allows a sample to be exposed to high pressures, even \({\mathrm{1}}\,{\mathrm{atm}}\), without the need for special pumping in the microscope. Functionality such as heating, tilting on two axes, and analytical techniques have in the past been difficult to implement in closed-cell experiments, but recent innovations have expanded the range of capabilities and therefore applications for closed-cell microscopy. Closed cells can even be used for supplying liquids to the sample (Sect. 3.6), provided the window spacing is reduced to a few hundred nanometers for reasonable imaging performance.

Open- and closed-cell ETEM are powerful techniques that have developed rapidly over the last few years and have produced advances in a range of scientifically important areas. Below we summarize some of the science and experimental details. However, the field is so broad that it is difficult to cover all of the interesting ramifications. Aspects of open- and closed-cell ETEM are described in more detail in [3.176, 3.177]. The equipment and science achieved using UHV microscopy are reviewed in [3.178].

3.3.1 Measurement and Modification of Surface Structure

It may appear surprising that TEM is appropriate for surface studies: we might expect the interactions of the electrons with the surface atoms would be swamped by interactions with the atoms within the thin foil. Furthermore, there are significant challenges arising from thin-foil effects, such as temperature nonuniformity, which could be avoided by using techniques such as scanning probe microscopies, (low energy electron microscopy), and controlled-environment SEM. However, TEM has a wide variety of imaging and analytical modes that can be sensitive to the sample surface; it can be highly quantitative in terms of image analysis, and can provide information with good time resolution (unlike typical scanning probe microscopies). Thus, in situ TEM has been used to examine phenomena such as step flow and the development and stability of surface structures such as reconstructions, using controlled heating, beam irradiation, or environmental stimuli such as deposition or exposure to a reactive environment.

In situ TEM initially gained attention as a surface science tool with the successful determination of the Si(111) \(7\times 7\) reconstruction [3.181], an accomplishment not possible with STM and (low energy electron diffraction) at that time. A clean Si surface was prepared by heating in a UHVTEM and diffraction patterns were obtained and analyzed. Since then, many other static and dynamic surface structures have been determined after in situ preparation.

Surface structures may be prepared in a UHV microscope column by heating or deposition onto a thin foil [3.182, 3.183, 3.46], or may be prepared in an adjacent chamber connected to the microscope by UHV [3.184]. Every possible mode of the TEM has been used to analyze these surface structures. In plan view, diffraction techniques have solved reconstructions of metals on Si [3.182, 3.185]. Reflection electron microscopy () has been used extensively to examine surface step dynamics due to electromigration, and the effect of metal adsorption on surface structure [3.179, 3.186, 3.187, 3.188] (Fig. 3.11a-h). REM has also provided useful information on polar surface structures in oxides [3.189] and decomposition of the InP surface on heating [3.190]. Information from REM and plan-view TEM is complementary to that obtained from in situ SEM [3.191]. Profile imaging, in which a surface parallel to the beam is imaged at high resolution, shows directly the periodicity and corrugations associated with surface reconstructions. This was recognized some time ago [3.192, 3.193, 3.194, 3.195], and more recently has helped to determine complex structures like Si(5 5 12) [3.183] as well as faceting, reconstructions, and dynamics of Au-decorated Si surfaces [3.180] (Fig. 3.12a-e) and beam-induced changes in surface structure and stoichiometry [3.196]. A UHVTEM is not necessary for profile imaging of surface reconstructions: an oxygen environment in ETEM was used to form a series of reconstructions on \(\mathrm{TiO_{2}}\) [3.197].

Fig. 3.11a-h

Au deposition at \({\mathrm{800}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) onto a vicinal (\(4^{\circ}\) miscut) Si(001) surface observed using REM. (001) terraces form in (a) and (b); step bunching and formation of step bands are shown in (c) and (d); facet nucleation on the step bands is shown in (f) and (g). Eventually the entire vicinal surface transforms into a hill-and-valley structure of (001) superterraces and (119) and (117) facets. Reprinted with permission from [3.179]. Copyright 1999 by the American Physical Society

Fig. 3.12a-e

The Au-induced reconstruction of a flat Si(001) surface on heating. Au was deposited onto a rough, oxidized Si particle. The irregular sample geometry means that the temperature is not accurately known, but on heating the Au first agglomerated; further heating caused the oxide layer to disappear and the Au to spread over the surface (dark line along the edge in (a)), causing localized faceting into (001) and other terraces. The (001) facets then reconstructed from  (b) to (e), starting from the terrace boundary (arrows AB in (b)). From [3.180], reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com)

Surface diffusion under beam irradiation is evident during profile imaging. This has been used to advantage to study narrow wires in situ. Two adjacent holes are formed in a thin foil and the neck between them observed as it thins and breaks due to inbuilt stresses. Such experiments show a variety of interesting nonbulk-like structures in Au (Fig. 3.13a-d), such as single chains [3.198] and helical multishell wires [3.199, 3.200], with some of the structures related to shear [3.201]. Beam-induced atom migration reduces the dimensions one layer at a time [3.202]. Unusual wire structures also form in Pt [3.202]. The surface diffusion of single atoms, for example W on MgO, can be imaged in TEM [3.203].

Fig. 3.13a-d

Nanostructures formed by surface mobility: Stable Au nanowires imaged during the electron-beam thinning of a specimen. The diameters are \(\mathrm{1.3}\) (a), \(\mathrm{1.1}\) (b), \(\mathrm{0.8}\) (c), and \({\mathrm{0.6}}\,{\mathrm{nm}}\) (d). The wire images are wavy, especially the thinnest, and can be modeled with helical structures. From [3.199]. Reprinted with permission from AAAS

Controlled-environment TEM thus has an excellent track record for creating and observing surface reconstructions and observing step motion and surface diffusion. We now focus on some areas where the ability to characterize and modify surfaces in situ has led to particularly interesting advances.

3.3.2 Oxidation and Reduction

Oxidation and related processes have been intensively studied in situ due to their general importance. Spatially resolved data is powerful when combined with theoretical predictions and with information obtained from post-reaction imaging and spectroscopic techniques. Oxidation, reduction, nitridation, and intercalation reactions can be addressed using open- and closed-cell ETEM, with samples in thin-film or nanoparticle form.

Copper is an excellent example where (open cell) UHVTEM shows that oxidation proceeds via nucleation, growth, and coalescence of oxide islands (Fig. 3.14a-c). Such data was used to develop oxidation theories that go beyond models that assume a continuous oxide film [3.204, 3.206, 3.207, 3.208]. Aberration-corrected ETEM can distinguish the phases present [3.209]. In alloys such as Cu-Ag, morphology can be related to the different reaction rates of each component [3.210]. Closed-cell microscopy provides a path to higher pressure oxidation and has been used, for example, for Zircaloy [3.211]. Oxidation to produce nanostructures was imaged for several metals [3.212, 3.213]. The oxygen may be supplied deliberately or be a result of beam-induced decomposition of a metal oxide. In situ metal oxidation experiments offer possibilities of say improving corrosion resistance by alloying, or optimizing nanostructure formation through processing.

Fig. 3.14a-c

Mechanism of copper oxidation: Dark-field images obtained during oxidation of Cu at \({\mathrm{0.1}}\,{\mathrm{Torr}}\) and \({\mathrm{350}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) in a UHVTEM. Imaging using the \(\mathrm{Cu_{2}O}\) {110} reflection showed that oxidation takes place by \(\mathrm{Cu_{2}O}\) island (a) nucleation (\({\mathrm{5}}\,{\mathrm{min}}\)), (b) growth (\({\mathrm{15}}\,{\mathrm{min}}\)), and (c) coalescence (\({\mathrm{25}}\,{\mathrm{min}}\)). The sample was prepared by floating a \({\mathrm{60}}\,{\mathrm{nm}}\) Cu film onto a support and then removing the surface oxide in situ by annealing at \({\mathrm{350}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) in methanol vapor for \(15{-}30\,{\mathrm{min}}\). The area and number density of the islands were modeled using Johnson–Mehl–Avrami–Kolmogorov theory to give surface diffusion parameters. Reprinted from [3.204], with the permission of AIP Publishing

A defect-free Si–\(\mathrm{SiO_{2}}\) interface is fundamental to transistor operation. Figure 3.15a,ba shows silicon oxidation visualized through forbidden-reflection imaging. This mode is sensitive to the positions of steps on surfaces and at buried interfaces, such as between Si and its oxide, and is complementary to in situ scanning reflection electron microscopy [3.214]. Such experiments show that steps do not move during Si oxidation, meaning that any surface roughness remains during processing [3.205, 3.215].

Fig. 3.15a,b

Mechanism of silicon oxidation: (a) Dark-field image series showing Si(111) in plan view during oxidation in \({\mathrm{2\times 10^{-6}}}\,{\mathrm{Torr}}\) water vapor at \({\mathrm{400}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) at the times indicated (\(1\mathrm{L}={\mathrm{10^{-6}}}\,{\mathrm{Torr{\,}s}}\)). A \(\tfrac{1}{3}(422)\) forbidden reflection was used so that the gray levels correspond to terraces, with intensities repeating every unit cell (3 steps). Steps do not move during oxidation of several layers showing that step sites are no more reactive than terrace sites. Reprinted with permission from [3.205]. Copyright 1992 by the American Physical Society. (b) Si(001) before and after oxidation in \({\mathrm{1}}\,{\mathrm{atm}}\) air at room temperature, recorded using a \(\tfrac{1}{4}(220)\) forbidden reflection. The steps do not move on this surface either

ETEM imaging can provide particularly valuable insights where the material structures are complex. Oxidation and reduction of niobium oxides [3.216, 3.217] forms a series of block structures with changing stoichiometry (Fig. 3.16). A glide shear rearrangement occurs during reduction of vanadium pyrophosphate [3.218], a material important in butane catalysis. ETEM has probed intercalation in layered structures [3.219, 3.220] and de- and rehydroxylation of the lamellar material \(\mathrm{Mg(OH)_{2}}\), which is important in \(\mathrm{CO_{2}}\) sequestration [3.221].

Fig. 3.16

Oxidation of a block structure: High-resolution image of a \(\mathrm{Nb_{22}O_{54}}\) crystal after heating by the electron beam and exposure to \({\mathrm{15}}\,{\mathrm{mbar}}\) oxygen. The partly oxidized structure consists of microdomains of \(\mathrm{Nb_{10}O_{25}}\) (arrowed) in an \(\mathrm{Nb_{22}O_{54}}\) matrix. Such images were used to identify the structure of the \(\mathrm{Nb_{10}O_{25}}\) phase and the complete oxidation sequence from \(\mathrm{Nb_{12}O_{29}}\) to \(\mathrm{Nb_{10}O_{25}}\) in situ. Reprinted from [3.216], with permission from Elsevier

The combination of ETEM imaging with other techniques provides additional information. Examples include the use of EELS to determine oxidation states during reduction of \(\mathrm{CeO_{2}}\) [3.222] and changes in battery electrode materials with oxygen or hydrogen pressure [3.223]. Raman spectroscopy has been used to examine the oxidation products of the two-dimensional () material \(\mathrm{Ti_{3}C_{2}}\) [3.224]. (energy-filtered transmission electron microscopy) measurements can quantify reaction kinetics, as in the reduction of an NiO/ceramic solid oxide fuel cell anode in hydrogen [3.225, 3.226], Fig. 3.17.

Fig. 3.17

Reduction reactions probed using EFTEM: Three image series showing reduction of NiO\(/\) (yttria stabilised zirconia), a standard solid oxide fuel cell anode, during ramping at \({\mathrm{2}}\,{\mathrm{{}^{\circ}\mathrm{C}/min}}\) in \({\mathrm{1.3}}\,{\mathrm{mbar}}\) of \(\mathrm{H_{2}}\). Each column shows a different imaging condition: (1) BF TEM, (2) \(t/\lambda\), which shows volume loss, (3) three-window O K maps, showing O distribution. The inert YSZ phase acts as a reference. A red Ni EFTEM map is superimposed on the first image and also shown in the top-right corner. EEL spectra at the start and end temperature confirm full reduction of the structure. Arrows mark voids indicating the start of the reaction. Black arrowheads show pores at the positions of some initial NiO\(/\)NiO boundaries. White arrowheads show Ni\(/\)Ni boundaries remaining in contact in the final images. The fraction of NiO that converts to Ni yields a three-dimensional picture of the reaction. Reprinted from [3.226], with permission from Elsevier

ETEM is of course applicable to other reactions related to oxidation and reduction. Examples include nitridation of zirconia [3.227] and the reaction of MgO with water vapor [3.228, 3.229]. We anticipate that it will address a greater range of materials in the future.

3.3.3 Catalysis

The central position of catalysis in science and technology has driven researchers to apply every available technique, with electron microscopy playing a strong role. Indeed, catalysis is a very popular subject for ETEM given the outstanding view of structure and dynamics that is possible during catalyst operation. Most ETEM experiments involve heterogeneous catalysts consisting of metal particles on an oxide support (Figs. 3.18a-d and 3.19a-f). The catalyst is examined at elevated temperature and under a controlled gas environment, with atomic resolution possible even at several millibar [3.230]. The movies and data from ETEM experiments help to relate catalyst and substrate structure to reactivity, determine intermediate phases, and measure catalyst stability, for example to sintering. ETEM is used in industrial as well as academic laboratories and we suspect that many results remain proprietary!

The first measurements on catalysts used open-cell ETEM techniques to examine shape changes in oxidizing and reducing environments [3.233], and observe sintering (Sect. 3.2.3), structures and reactions, and the effects of promoter species [3.234, 3.235, 3.236, 3.237, 3.238, 3.239]. Relative surface energies were determined under oxidizing and reducing conditions from particle shapes [3.232] (Fig. 3.19a-f); analytical techniques combined with imaging help for example to study hydrocarbon buildup during catalyst use [3.231] (Fig. 3.18a-d). Over the last ten years, rapid developments in in situ TEM techniques have transformed our view of catalysis. Particularly important advances are in resolution via aberration correction [3.240], quantification (for example, measuring the composition of the gas environment), and the range of experimental conditions accessible (particularly high pressure using closed-cell techniques). An idea of the excitement of the field can be gleaned by comparing reviews of the experimental approaches and results over the last two decades [3.241, 3.242, 3.243, 3.244, 3.245, 3.246].

Fig. 3.18a-d

In situ imaging of \(\text{Pd}/\mathrm{Al_{2}O_{3}}\) catalyst (used for hydrogenation of acetylene) (a) in the as-received condition (room temperature), and (bd) after heating in \({\mathrm{500}}\,{\mathrm{mTorr}}\) steam at \({\mathrm{700}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) for 1, 4 and \({\mathrm{7}}\,{\mathrm{h}}\). Catalysts are regenerated by heating in steam to remove hydrocarbon buildup, but this causes sintering of the metal particles, reducing activity. In situ experiments show that sintering is by conventional Ostwald ripening for fresh catalysts while movement and coalescence occurs for used catalysts. From [3.231], reproduced with permission

Fig. 3.19a-f

In situ imaging of Cu\(/\)ZnO catalyst (the methanol synthesis catalyst) in various gas environments at \({\mathrm{220}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\), together with the corresponding Wulff constructions of the Cu nanocrystals. (a,b) in \({\mathrm{1.5}}\,{\mathrm{mbar}}\) \(\mathrm{H_{2}}\) at \({\mathrm{220}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\); (c,d) in \(3:1\) \(\mathrm{H_{2}}:\mathrm{H_{2}O}\) at a total pressure of \({\mathrm{1.5}}\,{\mathrm{mbar}}\); (e,f) in \({\mathrm{95}}\%\) \(\mathrm{H_{2}}+{\mathrm{5}}\%\) CO at a total pressure of \({\mathrm{5}}\,{\mathrm{mbar}}\). These images allowed the relative surface energies to be determined as a function of environment. From [3.232]. Reprinted with permission from AAAS

Bridging the Pressure Gap

The pressures typically attainable in open-cell ETEM experiments are up to hundreds of mTorr, whereas catalysts in the real world operate at pressures of \({\mathrm{1}}\,{\mathrm{atm}}\) or more. The need to image catalyst behavior at realistic conditions of pressure and temperature has been a driving force for the development of closed cells for ETEM. One version of these closed cells has carbon windows and a heated wire as support [3.248]. Figure 3.20a-d shows a more complex high pressure nanoreactor for catalysis that contains a heater element and silicon nitride windows with thin areas to allow for higher resolution imaging [3.247]. Continued developments allow for improved resolution and rapid heating [3.249]. Closed-cell experiments have imaged a variety of catalysis phenomena. Examples include hydrogenation of Pd [3.250], shape changes in Pt particles during catalytic reactions [3.251], surface layer formation and shape changes [3.252, 3.253], ordering of bimetallic particles [3.254], and oxidation of Co [3.255] and Ag\(/\)AgCl nanocatalysts [3.256].

Fig. 3.20a-d

A high-pressure closed gas cell: (a) Schematic cross section showing gas flow. (b) Optical image of the TEM holder with integrated nanoreactor and four electrical probe contacts. (c) Optical close-up of the nanoreactor membrane. The bright spiral is the Pt heater, the small ovals are the electron-transparent windows and the circles are the \(\mathrm{SiO_{2}}\) spacers that define the minimum height of the gas channel. (d) A low-magnification TEM image of a pair of superimposed \({\mathrm{10}}\,{\mathrm{nm}}\)-thick windows, aligned to create an electron-transparent square for high-resolution imaging. Reprinted from [3.247], with permission from Elsevier

Operando Experiments

Experiments in which the catalysts perform as they would in real life are referred to as operando. Not only are the catalysts exposed to the correct working conditions (atmospheric or higher pressure, the appropriate temperature and mix of feedstock gases) but their performance is measured (gaseous species and heat produced) as well as their structure. Calibration of the conditions is clearly essential if the experiment is to provide useful information. In a closed cell, EELS can be used to measure the local gas temperature via the gas density [3.257]. Closed-cell reactors provide an additional benefit in that all the gas flowing through the reactor can be fed into a mass spectrometer for real-time monitoring, and temperature sensors on the closed-cell chip can provide reaction calorimetry [3.251]. In open-cell experiments, EELS can readily provide the gas composition [3.258], but a relatively small proportion of the gas flows close to the sample. More accurate measurement of catalytic performance can therefore be achieved by adding more material that is not imaged directly, distributed around the sample in the TEM polepiece region [3.259] or downstream [3.260].

Correlative Experiments

In situ catalysis experiments are enhanced by combining the imaging information with other experimental probes. One approach is to design a closed cell that is compatible with multiple instruments. In one operando experiment [3.260], a closed cell was moved between TEM and a synchrotron to allow sequential STEM and x-ray absorption spectroscopy to measure Pt particle rearrangements as well as the ethane produced from ethylene and hydrogen. A second approach is to bring additional probes into the sample area. Photocatalysis can be studied in situ with a light source in this way. High-resolution in situ imaging is then possible, providing a view of reactions such as the decomposition of hydrocarbons deposited on a \(\mathrm{TiO_{2}}\) film [3.261], photodeposition of Pt and photodegradation of \(\mathrm{Cu_{2}O}\) [3.262], and surface modification of anatase nanocrystals [3.263, 3.264]. The exciting recent progress in these areas suggest that in situ studies will continue to have an impact in the future development of new catalysts and in the optimization of existing materials.

3.3.4 Crystal Growth

Crystal growth is well suited to ETEM investigation. As we illustrate below, many materials have been deposited in situ. Carbon and various semiconductors are well studied, and even polymeric reaction products have been observed [3.265, 3.266]. Several deposition techniques have been explored. Evaporation is possible directly onto the substrate, if the polepiece and holder geometry permit. However, it is perhaps easier to flow a reactive precursor gas into the polepiece region and crack (decompose) it at a heated substrate. This is the in situ version of chemical vapor deposition. As with the examples in Sects. 3.3.13.3.3, both open- and closed-cell ETEM are useful for this, with closed-cell experiments generally accessing higher pressures. It is also possible, although less controllable, to supply the growth material by thermal or electron-beam decomposition of a solid source, relying on gas phase or surface diffusion to transport the species to the region under observation. Whichever of these techniques is used to supply the material, the resulting deposit may be a thin film, single crystal or polycrystalline, or an array of nanostructures. Catalytic particles or droplets may also be used to enhance the growth rate locally: the resulting experiments are analogous to the catalysis studies in Sect. 3.3.3 except that the reaction product is a solid crystal rather than a gaseous species. The performance of each catalyst can therefore be imaged directly, an exciting prospect for quantitative understanding of the process.

Carbon Nanostructures

The growth of carbon nanostructures has been extensively studied in situ since the discovery of these interesting materials by TEM. Carbon structures are formed in situ with either a reactive gas environment plus a catalyst or the beam plus a carbon-containing sample (Fig. 3.21a-f).

Fig. 3.21a-f

Carbon nanostructure growth in situ: (a,b) Formation of diamond. Polyhedral graphitic particles produced by arc-discharge were transformed into spherical onions by electron-beam irradiation above \({\mathrm{600}}\,{\mathrm{K}}\). The decreasing distance between shells towards the center showed that the onions are in a state of high self-compression. The nucleation of cubic diamond occurs in the centers of the onions during irradiation above \({\mathrm{900}}\,{\mathrm{K}}\), and the diamond grows until, surprisingly, almost all the onion shells are consumed. (a) After \({\mathrm{2}}\,{\mathrm{h}}\) of irradiation at \({\mathrm{1.25}}\,{\mathrm{MeV}}\) and \({\mathrm{20}}\,{\mathrm{mA{\,}cm^{-2}}}\). (b) After a further hour of irradiation. A typical twin is visible in the diamond. High pressure appears necessary to nucleate the diamond, but further growth is by beam-induced defects. Reprinted from [3.267], with the permission of AIP Publishing. (cf) Growth of multiwalled carbon nanotubes by catalytic chemical vapor deposition. Bright-field images showing an \(\text{Ni}/\mathrm{SiO_{2}}\) catalyst (c) exposed to \(\mathrm{H_{2}}\) at \({\mathrm{450}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\); (d) after exposure to acetylene (\({\mathrm{300}}\,{\mathrm{mTorr}}\) \(\mathrm{H_{2}}+\mathrm{C_{2}H_{2}}\)), (e,f) another part of the sample after \({\mathrm{3}}\,{\mathrm{min}}\). The arrows show an individual Ni particle. At higher temperatures, single-walled tubes formed but with catalysts present at the bases rather than the tips of the tubes. Reprinted from [3.268], with the permission of AIP Publishing

Catalytic growth of carbon nanostructures is imaged using open- or closed-cell ETEM, flowing reactive gases such as methane, propylene, acetylene, or ethylene over the catalyst [3.268, 3.269]. Changes in the shape of individual catalyst particles and the nucleation process can be resolved [3.270, 3.271, 3.272] and the overall process can be followed as a forest of nanotubes forms [3.273]. The relationship between catalyst structure and function can even be measured, suggesting strategies to control key nanotube features such as chirality [3.274]. ETEM experiments have also measured coking, as multilayered carbon shells form around catalyst nanoparticles [3.275] or on metal surfaces [3.276]. An alternative approach to carbon nanotube growth [3.277, 3.278] uses catalysts placed in a multiwall carbon nanotube furnace heated resistively. The source of carbon is irradiation of the carbon walls and the catalyst dynamics and phases can be examined.

The electron beam can drive the synthesis or modification of carbon nanostructures even in the absence of catalysts. The beam may interact with the atmosphere above the specimen producing a plasma that can generate fullerenes [3.279], or irradiation of carbon-containing materials such as C-implanted Cu causes graphitic onions to grow [3.280]. Irradiation of graphitic materials generates point defects that deform existing graphitic sheets to form new structures [3.267, 3.281]. Carbon nanotubes can be modified with the beam to produce more complex structures [3.282, 3.283]. Familiar or new C and BN structures, and even formation of diamond from graphite, can thus be observed in situ [3.162, 3.267, 3.281, 3.284, 3.285, 3.286, 3.287, 3.288, 3.289, 3.290, 3.291].

Epitaxial and Polycrystalline Film Growth

The growth of continuous thin films in situ enables measurements of nucleation, development of surface morphology, grain evolution, and strain effects. As mentioned above, the growth material may be evaporated in situ, as with Au epitaxy on MgO [3.294, 3.295] or silicide growth on Si (Sect. 3.2.2). Alternatively it may be supplied as a reactive gas, as for Al [3.296] or for Ge on Si described below. Growth is viewed edge-on along a vertical edge of the sample [3.276, 3.294, 3.295] or in plan view.

The epitaxy of Ge and SiGe on Si illustrates many of these growth phenomena and also has great relevance to the development of microelectronic devices. UHVTEM is used to provide clean initial surfaces and calibrated conditions of flux and temperature. The initial flat Si surface is obtained by heating the substrate in UHV well above the oxide desorption temperature. Growth is then carried out using chemical vapor deposition gases such as disilane or digermane. STEM [3.297] and REM [3.298, 3.299] provide a view of the developing growth morphology, but plan-view weak-beam imaging provides the greatest sensitivity to the strain fields that develop as Ge is deposited onto Si (Fig. 3.22a-d). The initial surface reconstruction and the nucleation, growth, and coalescence of Ge islands can be followed [3.293, 3.300] on Si(111) and (001). A range of fascinating phenomena now known to be common in other epitaxial systems is observed, such as island shape changes during the introduction of stress-relieving dislocations [3.301, 3.302]. Without the use of in situ TEM, the range of structures observed in these studies would have been tedious to capture ex situ, and dynamic phenomena such as island shape changes and the essential role of Ostwald ripening [3.292] would not have been detected. Changes caused by the presence of surfactants during growth have also been examined in situ [3.303, 3.304]. It is interesting to note that complementary studies using in situ LEEM have been important, as LEEM is more sensitive to surface structure [3.305].

Fig. 3.22a-d

Ge island nucleation, growth, and relaxation: (a) Image series showing spontaneous formation and Ostwald ripening of Ge islands formed during deposition of Ge onto Si at \({\mathrm{5\times 10^{-7}}}\,{\mathrm{Torr}}\) \(\mathrm{Ge_{2}H_{6}}\) and \({\mathrm{650}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). Weak-beam images were acquired in a \((\boldsymbol{g},\,3\boldsymbol{g})\) condition with \(\boldsymbol{g}={\mathrm{220}}\) to show strain contrast. The time of each frame after first appearance of islands is given. (b) The time evolution of every island within the \({\mathrm{0.5}}\,{\mathrm{\upmu{}m}}\times{\mathrm{0.5}}\,{\mathrm{\upmu{}m}}\) area imaged in (a). Large islands grow while smaller ones disappear, but kinetics are not consistent with a simple Ostwald ripening model. (c) A simulation based on a modified Ostwald ripening process in which islands coarsen but also undergo a shape transition at \(V_{\mathrm{c}}\). The fate of islands with different initial sizes is shown. Units are arbitrary but the scale of the plot is chosen to match the data. Reprinted with permission from [3.292]. Copyright 1998 by the American Physical Society. (d) The introduction of dislocations into a larger Ge island. Images were obtained 59, 94, 96, and \({\mathrm{140}}\,{\mathrm{min}}\) after the beginning of growth, with the last three images taken within \({\mathrm{1}}\,{\mathrm{s}}\) of each other. Growth conditions were \({\mathrm{650}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) and \({\mathrm{10^{-6}}}\,{\mathrm{Torr}}\) of \(10:1\) He:\(\mathrm{GeH_{4}}\). Only dislocations in the dark part of the image are visible (D1–D3). Reprinted from [3.293], with permission from Elsevier

When low Ge content (say \({\mathrm{15}}\%\)) SiGe is grown on Si, islands are not seen, but instead a continuous flat layer forms. When sufficiently thick, this film relaxes by introduction of dislocations. The motion of these dislocations can be measured during film deposition and compared with dislocation dynamics during postgrowth annealing, which will be discussed in Sect. 3.5.1. Interestingly, the parameters governing dislocation motion are different during growth, where the surface is H terminated, versus during annealing, where the surface is oxidized [3.306, 3.307]. This is thought to reflect a higher kink nucleation rate under the oxidized surface due to surface stress or increased point defects. Such information is needed for modeling relaxation during device processing, and shows once again the opportunities that arise when materials are observed during growth rather than ex situ.

Catalytic Nanowire Growth

The physics of catalytic carbon nanotube growth and epitaxial semiconductor growth described above combine to create the fascinating catalytic growth process of semiconductor nanowires (Figure. 3.23a-f). These structures grow via the vapor-liquid-solid mechanism, in which the growth species (Si, Ge, \(\text{Ga}+\text{As}\), etc.) are supplied from the gas phase to a catalytic droplet that often contains Au (as in AuSi, AuGe, or AuGa). When the droplet becomes supersaturated, the growth species deposits at the droplet–substrate interface. Growth is in the form of a single-crystal nanowire with the droplet remaining at the tip. Although post growth heating of nanowires helps elucidate the mechanism [3.308], in situ growth provides richer information.

Nanowire growth is examined in situ using UHVTEM to provide a clean initial surface that promotes epitaxy [3.309]. Open-cell ETEM provides higher pressure but less well-controlled vacuum conditions [3.310, 3.311], and closed-cell ETEM [3.312] allows a wider variety of species to be supplied by thermal decomposition. In plan view, growth has been observed qualitatively in Si\(/\)Au [3.313], Si\(/\)Fe [3.314], and GaN [3.315]. However, quantitative measurements are easier in the a side-view geometry, Fig. 3.23a-f. The experiments demonstrate unexpected effects such as surface faceting [3.316] and catalyst diffusion during growth [3.317]. Details of the growth mechanism are resolved by imaging the growth rates of individual nanowires as a function of conditions, as well as the morphology of the growth interface and the pattern of step flow, and the formation of heterojunctions on switching the source gas. Such studies enable rational design of catalysts to achieve a desired synthetic structure [3.318, 3.320]. In situ observations similarly help to explain the factors that control crystal phase so that superlattices of different crystal structure can be created [3.321]. It is even possible to control nanowire growth directions and study nanowire kinking by distorting the catalyst droplet in an electric field [3.322]. The ability to control nanowire structure allows access to new electronic properties. A few other systems have been examined, notably metal oxides [3.323, 3.324], but we anticipate catalytic nanostructure growth will soon be imaged in more materials.

Fig. 3.23a-f

Semiconductor nanowire growth: (a) Single movie frame recorded during Si nanowire growth from a liquid AuSi droplet at \({\mathrm{550}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) and \({\mathrm{2\times 10^{-5}}}\,{\mathrm{Torr}}\) disilane in a Titan ETEM. Reprinted with permission from [3.98]. Copyright 2015 American Chemical Society. (b) Schematic diagram of the pathway of Si from the precursor gas to the growth interface with step geometry also shown. (c) Image series recorded during Si nanowire growth from solid \(\mathrm{AuAl_{2}}\) at \({\mathrm{550}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) and \({\mathrm{2\times 10^{-5}}}\,{\mathrm{Torr}}\) disilane. Step flow is slow (arrows). From [3.318]. Reprinted with permission from AAAS. (d,e) Measured step flow for growth from AuSi and \(\mathrm{Cu_{3}Si}\) catalysts (upper and lower graphs). The incubation time between step flow events is shown as horizontal arrows. A simulation is shown in light brown. (f) Use of the catalyst in (c) to grow Si-Ge heterostructures. The line scan shows the compositional abruptness of the interfaces. Reprinted with permission from [3.319]. Copyright 2015 American Chemical Society

Crystal Growth on Patterned Surfaces and in Nanoscale Systems

An interesting extension of the experiments described above is the study of crystal growth in an environment that is not spatially uniform. By growing on substrates that have been patterned or modified to create areas of different reactivity, growth physics can be probed in interesting ways.

Ge growth on Si is sensitive to substrate patterning. The locations of Ge islands can be controlled during in situ growth onto substrates with lithographically patterned mesas [3.325] or focused ion-beam ()-patterned pits [3.326, 3.327, 3.328] (Fig. 3.24). Measurements of nucleation positions and times help understand the mechanism of island site selection and perhaps aid in fabricating novel electronic devices. In these experiments, FIB patterning was carried out in a chamber connected to the microscope vacuum so that the patterned substrate was not transferred through air. Processing tools that are not in situ (i. e., in the polepiece) but are still within the microscope vacuum system can provide useful control over substrate preparation.

Fig. 3.24

Semiconductor growth on a patterned substrate: On the left is a \(\boldsymbol{g}=220\) bright-field image of Si recorded directly after patterning of dots by FIB using an irradiation time of \({\mathrm{0.1}}\,{\mathrm{ms}}\) per feature. On the right is a (\(\boldsymbol{g},\,3\boldsymbol{g}\)) weak-beam image with \(\boldsymbol{g}=220\) showing the same area after annealing followed by Ge deposition at \({\mathrm{650}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) and \({\mathrm{5\times 10^{-8}}}\,{\mathrm{Torr}}\). The islands nucleate on the FIB-irradiated spots. Reprinted from [3.326], with the permission of AIP Publishing

Crystal growth environments with a spatially limited extent also provide opportunities for examining growth physics. For example, an AuSi droplet on an inert silicon nitride substrate acts as an isolated system for nucleation studies. Exposure of the droplet to Si from the gas phase (the same procedure used to form nanowires) eventually nucleates a silicon nanocrystal. The growth rates and nucleation times of such nanocrystals as a function of droplet size provide detailed information regarding the nanoscale phase diagram [3.329]. Since the composition is varied at constant temperature, these experiments provide a different perspective on the phase diagram, compared to the heating experiments in Sect. 3.2.3.

3.3.5 Summary

We have highlighted insights into surface physics and crystal growth derived from in situ TEM. Experiments in this field, while often requiring dedicated equipment with complex additional preparation chambers, provide information that is difficult or impossible to obtain using other techniques. Continuous and direct observation in situ avoids artifacts arising from ex situ observation, which may be especially significant for growth studies. It is particularly encouraging that realistic growth conditions can be accessed in the TEM, enabling in situ studies of catalysis or CVD to be related to their counterparts in the outside world. Indeed, the relevance of in situ TEM to the catalyst and microelectronics industries is shown by the presence of environmental microscopes in several industrial laboratories.

Many opportunities exist in this area. Growth on patterned surfaces, prepared if necessary with integrated processing tools in the TEM system, seems a particularly exciting way to examine the fundamental processes controlling growth and to fabricate nanoscale objects with particular properties. Another potentially significant area is the correlation of structural or kinetic in situ measurements with macroscopic properties such as conductivity (Sect. 3.4) or stress, as measured by depositing onto a crystalline membrane whose curvature can be measured. This may provide insight into the processes associated with coalescence and grain-boundary motion during polycrystalline film growth, as well as stress relaxation in epitaxial films.

3.4 Functional Materials and Devices

In previous sections the sample was subjected to a temperature or gas environment and a transformation or growth process was observed. We now describe the use of in situ TEM to probe structure–property relationships in materials that are magnetic, superconducting, ferroelectric, or have interesting electrical or optical properties. In these experiments, the relevant stimulus is applied in situ (electric or magnetic field, voltage, current, temperature, light field). If a structural change is induced, such as the motion of a domain boundary or flux line, we obtain insights into how material structure determines functional properties, and how to design materials to achieve a particular performance. Even if there is no observable change, the TEM is still essential since we can apply a stimulus to a nanostructure that is difficult to manipulate and characterize otherwise. The examples we show include correlations between structure and properties both with and without a structural change.

3.4.1 Correlation of Structural and Electrical Properties

TEM can measure electrical properties using electrical biasing holders that apply a voltage to specially designed samples [3.330, 3.331, 3.332]. Individual nanostructures and bulk or patterned TEM samples can be probed to measure the change in resistance as voids form during electromigration, the potential distribution across a p-n junction as a function of bias, or the electrical properties of nanowires or two-dimensional materials. Individual nanostructures are contacted using piezo-driven scanning probe tips. An entire sample or a feature patterned onto a sample is biased via contact pads that make electrical contact to the holder through wire bonds or spring clips. Focused ion-beam processing provides additional capabilities in many in situ biasing experiments, by fabricating devices with a geometry suitable for electrical biasing or by creating electrical contacts to nanostructures.

Electrical Measurements on TEM Samples: Complete Samples as Devices

Electrical phenomena can be examined in bulk materials by applying a voltage across a thinned bulk sample. An interesting example is the measurement of electric fields across p-n junctions, first achieved some decades ago [3.333]. As the bias is applied across an FIB-prepared p-n junction, holography is used to measure the potential distribution [3.334, 3.335, 3.336]. To achieve the eventual aim of understanding the complex fields within real transistors, surface effects from the thin foil must be understood. Similar biasing experiments probe the potential distribution across ferroelectrics: for example, at a bicrystal boundary [3.337, 3.338] holography measurements show breakdown and the presence of trap states associated with dopants.

In situ biasing also allows resistivity to be measured and then correlated with a material’s structure, or with structural changes. Thus, the amorphous to crystalline transformation for phase-change memory materials (Sect. 3.2.1) can be correlated with resistivity, as shown in Fig. 3.25 [3.339]. Similarly, phase transformations in TiNi shape-memory alloys can be related to resistivity [3.64]. Current–voltage characteristics from magnetic tunnel junctions can be correlated with microstructure (Sect. 3.4.2). Another interesting application of in situ resistivity measurement is to determine the cross section and threshold energy for Frenkel pair production by measuring conductivity changes during ion- or electron-beam irradiation [3.340, 3.341].

Fig. 3.25

Correlation of electrical properties with structure: Image series recorded every \({\mathrm{15}}\,{\mathrm{s}}\) during the crystallization of an Al-Ge film at \({\mathrm{450}}\,{\mathrm{K}}\). Reprinted from [3.339], with the permission of AIP Publishing

Fig. 3.26a-f

Correlation of electrical properties with structure: Evolution of a \({\mathrm{300}}\,{\mathrm{nm}}\)-wide Al (\({\mathrm{0.5}}\,{\mathrm{wt\%}}\) Cu) interconnect, \({\mathrm{400}}\,{\mathrm{nm}}\) thick and \({\mathrm{150}}\,{\mathrm{mm}}\) long, on a \({\mathrm{20}}\,{\mathrm{nm}}\) TiN underlayer. \({\mathrm{100}}\,{\mathrm{nm}}\) oxide passivation was deposited on the line, which was then annealed to form a bamboo structure where all grain boundaries run straight across the line. The substrate is a \(\text{SiN}/\mathrm{SiO_{2}}\) bilayer membrane to minimize changes in the stress state. The temperature is \({\mathrm{320}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) and current density is \({\mathrm{2\times 10^{6}}}\,{\mathrm{A{\,}cm^{-2}}}\), and the stress time is (a\(\mathrm{3.5}\), (b\(\mathrm{4.0}\), (c\(\mathrm{4.1}\), (d\(\mathrm{4.5}\), (e\(\mathrm{4.6}\) and (f\({\mathrm{5.5}}\,{\mathrm{h}}\). Reprinted from [3.342], with the permission of AIP Publishing

In situ resistivity measurements can also probe the electrical properties of dislocations [3.343]. In this experiment, dislocations are formed progressively in a p-n junction diode by heating a metastable SiGe layer (Sect. 3.5.1) in situ. Measurement of the diode reverse leakage current as a function of dislocation density yield the leakage current per length of dislocation, useful in device modeling. In this experiment, however, leakage was measured through the whole device, while dislocation density was measured only in the electron-transparent area. It is in fact possible to thin an entire device using FIB [3.344], in this case allowing degradation in a laser structure to be correlated with all dislocations formed during device operation. Such biasing studies can be combined with laser excitation in situ (Sect. 3.4.5) to examine degradation mechanisms and photoplasticity. In situ biasing can be combined with elemental mapping by XEDS to give a full view of chemical redistribution during biasing. An example is the movement of iodine in methylammonium lead iodide perovskite solar cells, measured as a function of current and voltage by biasing thin samples [3.345].

For two-dimensional materials, electrical properties can be examined by careful sample preparation that involves placing the sheet across a pair of microfabricated cantilever electrodes [3.346]. Smaller scale electrical contacts to two-dimensional materials are fabricated by using the electron beam to sculpt a nanostructured geometry from a larger microfabricated structure, for example to cut graphene nanoribbons from graphene sheets attached to electrodes [3.347, 3.348]. Using the electron beam for the final fabrication step allows for a well-defined geometry with a robust electrical connection for two-point [3.347] or three-point [3.348] electrical probing.

Electromigration

Because of its importance and its strong dependence on microstructure, electromigration is one of the most intensively studied phenomena to have been addressed with in situ electrical biasing experiments. Current is passed through a metal line, such as Cu or Al, patterned onto an oxide or silicon nitride substrate, while simultaneously observing void formation and void dynamics at microstructural features such as grain boundaries and precipitates. An example is shown in Fig. 3.26a-f. There are two important concerns in these experiments. Passivation of the metal surface is known to alter failure lifetimes [3.349]. Passivating the lines makes the experimental results more relevant to real life, but it is then harder to discern the void shapes and dynamics. Joule heating in the lines should also be considered [3.350], especially since elevated temperatures and current densities above \({\mathrm{10^{6}}}\,{\mathrm{A{\,}cm^{-2}}}\) are used for accelerated testing. In some experiments, on-chip heat-sink structures are therefore integrated into the samples [3.351].

In situ observations have documented the nucleation of electromigration-induced voids well before failure of the lines: void nucleation sites and subsequent motion depend on the local grain-boundary structure [3.352, 3.353], helping to elucidate the mechanisms at work. The failure mechanism depends on temperature, with interesting void dynamics observed at elevated temperature [3.351, 3.354]. Surface diffusion can be a dominant failure mechanism in electromigration [3.355, 3.356, 3.357]. Thus when barrier or surface layers are present, voids do not migrate, presumably because of changes in surface diffusion [3.342].

Since a thick passivation layer may be essential for meaningful results, an interesting experimental approach is to mount the sample vertically and image in cross section at high voltage [3.358, 3.359]. Such experiments have shown mass transport through Al and TiN layers as well as vertical void and whisker growth. The measurement of local strain by (convergent-beam electron diffraction) during electromigration [3.360] is an exciting development that can help connect the in situ results more closely to models.

Electromigration can also be examined in freestanding, solid conductive nanowires that essentially act as substrates on which droplets of liquid metals are present [3.361, 3.362]. Electromigration mechanisms within the solid or along its surface can be probed by such experiments. Additional in situ measurements on nanowires are discussed in the following section.

Electrical Properties of Individual Nanostructures

The combination of scanning probe microscopy techniques with TEM provides exciting opportunities for measuring the properties of individual nanostructures [3.363]. Nanostructures such as carbon nanotubes or semiconductor nanowires are placed or grown on a scanning probe tip which is then used to approach a flat substrate. Equivalently, the tip is used to approach and contact a single nanostructure that was previously placed or grown on a fixed substrate.

For carbon nanotubes, this type of experiment enables observation of the electric field distribution at the tip on biasing [3.364] or structural changes during field emission [3.365, 3.366, 3.367]. The work function of CNTs can be measured and related to structure [3.368, 3.369, 3.370, 3.371] and CNT growth can be observed in the gap between a biased tip and the substrate [3.372]. After contact is made, the conductivity of a single nanotube can be measured. Figure 3.27a,b shows an interesting example that correlates the chiral indices of double-walled tubes, measured using diffraction, with their transport properties [3.373]. Conductivity measurements show that CNTs are ballistic conductors at room temperature [3.374]. The effects of metal cores within CNTs [3.375] and the changing electrical properties of rheostats made from telescoping multiwalled CNTs [3.376, 3.377] have been correlated with mechanical performance (Sect. 3.5.2). Structural changes taking place during current transport have also been observed [3.378]. The temperature of CNT contacts can be evaluated from the sublimation behavior of material confined within the tube [3.379]. Other materials examined in this way include nanowires formed of GaAs [3.380], ZnO [3.381], GaN [3.382], and BN [3.383].

Fig. 3.27a,b

Simultaneous structure and transport measurement for a double-walled carbon nanotube: (a) Experimental diffraction pattern showing the characteristic layered line structure. The iris-like ring is an artifact. The white arrow indicates the equatorial line and the gray arrows indicate other diffraction lines. (b) Current–voltage characteristic of the same tube. Ohmic behavior is visible up to \({\mathrm{0.5}}\,{\mathrm{V}}\). The insets show the diffraction geometry. The CNT reference axes are (\(x^{\prime}\), \(y^{\prime}\), \(z^{\prime}\)); the beam direction is \(x\) and the diffraction pattern is recorded in the \(yz\) plane. In the lower inset, given \(\Upgamma\), the position of an arbitrary carbon atom (gray dot) is characterized by a translation distance \(z^{\mathrm{f}}_{\mathrm{o}}\) and a rotation angle \(\phi^{\mathrm{f}}_{\mathrm{o}}\) with respect to the tube axis. Reprinted with permission from [3.373]. Copyright 2002 by the American Physical Society

In all these experiments the nature of the contact, for example how the tip is cleaned, is important in ensuring that the measurements relate to the structure rather than the contact. A drop of Hg has been used [3.373, 3.385, 3.386] to measure properties like work function and quantum conductance, but there are still challenges in creating ohmic contacts to nanostructures [3.380]. In an alternative approach, electrical contact is made by growing a nanostructure to make an epitaxial bridge between electrical contacts. For Si nanowires grown between cantilevers, this provides an opportunity to relate the resistivity of the entire bridge to the contact geometry [3.387].

Nanostructures can also be formed by manipulating a substrate with the scanning probe tip. Au filaments form when an Au tip is touched to a substrate and pulled away. Figure 3.28a–c demonstrates conductance–structure correlations in such contacts and their mechanical properties are discussed below in Sect. 3.5.2. Conductance through single and double Au chains is quantized [3.384] and single chains show a metal–insulator transition [3.388]. Conductance and structure change together, showing the dynamic nature of the system [3.389, 3.390].

Fig. 3.28a–c

Electrical properties of Au nanobridges. (a) Quantized conductance of bridges of gold atoms prepared by contacting an Au surface with an STM tip in a UHVTEM. While withdrawing the tip, the conductance changes in units of \(G_{0}=2e^{2}/h=({\mathrm{13}}\,{\mathrm{k\Upomega}})^{-1}\). (b,c) Configuration of the bridge at times A, with two rows of atoms, and B, with one row of atoms, respectively. The distance P–Q, \(\approx{\mathrm{0.89}}\,{\mathrm{nm}}\), is wide enough to have two gold atoms in a bridge if the atoms have the nearest-neighbor spacing of the bulk crystal (\({\mathrm{0.288}}\,{\mathrm{nm}}\)). From [3.384]

Scanning probe tips allow several other opportunities for imaging electrical biasing phenomena. Biasing a single nanostructure to visualize lithiation reactions and biasing a thin film in cross-section geometry to observe switching in ionic conductor memory materials were already described in Sect. 3.2.2. A tip applied to a sample composed of Ni nanoparticles with thin surface oxide layers allows the mechanism of electric field-assisted consolidation to be probed in situ [3.391]. There are many other areas where this experimental approach will provide unique information relating electrical phenomena to structural changes.

3.4.2 Magnetic Materials

It is happily convenient that magnetic microstructure is readily visualized in the TEM via several imaging modes. Lorentz imaging, involving the deflection of electrons by the magnetic fields within the sample, provides information via the Fresnel (easier) or Foucault (higher resolution) modes in TEM or differential phase contrast (DPC; at the highest resolution) in STEM. Holography measures the magnetic fields via their effect on the electron phase. These imaging modes have been used extensively to examine static configurations, but here we discuss experiments where a magnetic structure is deliberately changed [3.392, 3.393, 3.394, 3.395, 3.396]. This is achieved by applying a varying magnetic field to the sample or changing its temperature, possibly while also measuring electrical properties. The resulting correlation of magnetic, electrical, and physical microstructure lead to a better understanding of the dynamics of magnetic domain switching, the mechanisms determining magnetoresistance properties, and phase transformations in magnetic materials.

To apply a controlled magnetic field to the specimen, the microscope is set up so that the sample is in a field-free (or low field) region [3.392]. If a conventional microscope is used, the objective lens can be switched off and other lenses used to provide the magnification. A field is then applied using coils placed in the column or on the specimen holder. Several holder designs incorporating coils have been developed to generate in-plane and out-of-plane fields, preferably without shifting the beam [3.397, 3.398, 3.399, 3.400]. Another design [3.401] uses a sharp needle made of a permanent magnet to produce a strong field near the specimen. A simpler solution that does not involve modification of the sample holder is to leave the objective lens on but tilt the sample, thereby changing the field it experiences. This has been useful in examining superconductors, as we discuss in Sect. 3.4.5.

For further functionality, heating or cooling capabilities allow phenomena to be studied around the Curie temperature and biasing holders allow electrical and magnetic measurements to be combined. Overall, this generates a broad range of experimental opportunities that has been applied to single crystals, polycrystalline thin films and multilayers, and patterned magnetic elements.

Domain-Wall Motion in Magnetic Materials

The motion of domain walls in magnetic materials has been studied for over 30 years to determine how domain walls move in perfect materials, how they interact, and how grain boundaries and other defects alter their motion by pinning. Macroscopic hysteresis loops can be correlated with pinning and domain size in Co\(/\)Pt [3.403], and the mechanism and motion of domain boundaries during magnetization reversals has been correlated with microstructure in materials such as Nd-Fe-B [3.404, 3.405] and CoIr [3.406].

Complex properties can be investigated in multilayered materials where each layer has different magnetic behavior, such as Co\(/\)Fe bilayers [3.408]. Two ferromagnetic materials separated by a conducting layer can show giant magnetoresistance properties which have applications in nonvolatile magnetic random access memory and magnetoresistive read head technologies. When one ferromagnetic layer is pinned by an adjacent antiferromagnetic layer, multistep hysteresis loops make the structures suitable for spin valves or spin tunnel junctions. Magnetic tunnel junctions made of two (or more) magnetic layers separated by an insulating tunnel barrier can also be used for storage or sensing. Details of the magnetization reversal mechanism (for example, whether it occurs by domain nucleation and growth or by spin rotation), and the relationship of the hysteresis loop to the microstructure (interface roughness and grain size) are important questions in these multilayers. For NiFe\(/\)Al-oxide\(/\)Co and Co\(/\)NiFe\(/\)Al-oxide\(/\)NiFe multilayers, Lorentz imaging in Fresnel mode [3.402, 3.409] showed the process of magnetization reversal by wall motion and rotation (Fig. 3.29a-l), and the relationship of grain size and texture to domain-wall motion. Reversal has been examined [3.410, 3.411] in complex structures such as PtMn\(/\)CoFe\(/\)Ru\(/\)CoFe\(/\)Cu\(/\)CoFe\(/\)NiFe used in advanced spin valves, while the effect of heating [3.412] has been studied on Ta\(/\)NiFe\(/\)Cu\(/\)Co\(/\)MnFe\(/\)Ta films to model thermal damage in spin valves. In IrMn\(/\)CoFe films, in situ experiments have shown that magnetization reversal must overcome two energy barriers, explaining features of the hysteresis loop, as well as the dependence of the properties on the layer thicknesses [3.413]. It is clear that in situ experiments provide a unique way of studying the properties of the ever more complex multilayers being developed for modern read heads and other applications.

Fig. 3.29a-l

Magnetization of a multilayer film in situ. (ak) Lorentz–Fresnel images obtained during magnetization of an NiFe\(/\)Al-oxide\(/\)Co junction film. All images show the same area and the direction of the applied field \(H\) is indicated in the last panel. (l) The normalized magnetization versus applied field of the film with the positions of the images indicated. Reprinted from [3.402], with the permission of AIP Publishing

Small Magnetic Elements

Many applications of magnetic materials require the material to be patterned into small elements, making in situ studies of the magnetic properties of such elements valuable. Typically, an electron-transparent substrate such as an SiN layer is patterned with structures of different shapes and sizes using electron-beam lithography. The elements may be far apart so that they can be studied in isolation, or near their neighbors so that crosstalk is observed. It is possible to perform micromagnetic simulations of the complete structures to help understand the microstructural observations and relate them to macroscopic measurements of hysteresis loops.

During switching in situ, the elements can be measured using holography, Fresnel defocus imaging, or DPC. Figures 3.30 and 3.31 compare holography [3.414] and Fresnel imaging [3.407] of Co elements. Note the ripple contrast visible in the Fresnel image from the polycrystalline material. This type of experiment can measure the switching field and show how the different configurations of remanent magnetization and the reversal process depend on the size, aspect ratio, and symmetry of the elements. The temperature and reversal rate dependence can be examined. Permalloy (NiFe) elements have been particularly well studied [3.415, 3.416, 3.417, 3.418, 3.419, 3.420, 3.421, 3.422, 3.423]. The relationship between geometric structure and magnetic properties is directly visible. Examples include the way in which certain end shapes allow easier switching by enabling \(360^{\circ}\) domain walls to form [3.424], and the effects on domain wall pinning of constrictions and protrusions in thin stripes [3.425]. As with the thin-film studies described in the previous section, examining multilayer elements shows how the magnetic properties of each material interact to yield complex reversal behavior [3.426, 3.427, 3.428].

Fig. 3.30

Magnetization of small elements using defocused Fresnel imaging: in situ magnetization of a Co element is shown with the corresponding hysteresis loop. The element is \({\mathrm{25}}\,{\mathrm{nm}}\) thick and \({\mathrm{6}}\,{\mathrm{\upmu{}m}}\) on a side and patterned on an SiN membrane. Images are shown as a function of decreasing applied field, achieved by tilting the sample in a fixed normal field excited by the objective lens (\({\mathrm{160}}\,{\mathrm{Oe}}\)). Large and small arrows show the direction of the applied field and local magnetization, respectively. The nucleation of the reverse domain is indicated with asterisks. Only the upper semiloop of the hysteresis curve is shown; the reverse process is somewhat alike but differs in the reverse switching field. The processes observed in these images are coarsening of ripples, because of coherent spin rotation; nucleation and expansion of reverse domains; wall motion and spin rotation in remaining domains; expulsion of unfavorable boundary domains; and edge annihilation. From [3.407]. Reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com)

Fig. 3.31

Magnetization of small elements using holography: in situ magnetization of two \({\mathrm{30}}\,{\mathrm{nm}}\)-thick Co elements. The magnetic contributions to the phase were obtained in a Philips CM200-FEG TEM equipped with a Lorentz minilens. Phase contours are separated by \(0.21\uppi\,\mathrm{rad}\) and colors represent the magnetization direction. The hysteresis field was applied horizontally in the figure, and should be followed counter-clockwise. The out-of-plane field was \({\mathrm{3600}}\,{\mathrm{Oe}}\). A simulation is also shown where the initial configuration is the S-state in the small cell and the C-state in the large cell, a \({\mathrm{3600}}\,{\mathrm{Oe}}\) field into the page, and square corners on cells. The hysteresis loop is derived from the simulation with the large cell shown in brown and the small cell in black. Reprinted from [3.414], with the permission of AIP Publishing

Analogous phenomena can be measured for continuous films that are etched to form arrays of holes (antidots) [3.429] or structures with quasicrystal symmetry [3.430]. It is also possible to pattern a continuous film other than by etching. In thin Co\(/\)Pt multilayer films, small areas can be created using ion implantation that confine magnetic structures such as skyrmions [3.431].

It is worth remembering that magnetic measurements do not provide the full three-dimensional field. Exciting future directions will include the use of tomographic techniques to extend measurements of small magnetic elements into the third dimension.

Magneto-Electrical Properties

An interesting extension of the in situ magnetizing experiments described above is the linkage of magnetic with electrical properties. This involves the use of a holder with electrical biasing connections. An electrical stimulus is used to change the magnetic configuration, as in the use of current pulses to move domain walls [3.433]. Alternatively, electrical properties such as resistivity are measured while changing the magnetic configuration using an applied field. An example is the study of spin-valve structures made of multilayers such as NiFe\(/\)Cu\(/\)Co\(/\)NiFe\(/\)MnNi [3.434, 3.435, 3.436]. A current is passed through the element as a field is applied to change the magnetization in the upper NiFe (free) layer but not the lower (pinned) layer, as in Sect. 3.4.2. Measuring the change in voltage on changing applied field allows the magnetoresistance to be determined and the effect of current on magnetization reversal to be studied.

A different experimental geometry involves a bridge between two elements of the same material. Rather than using a free and pinned layer as for the spin valve, these elements are patterned with different geometry so that when the field is applied the elements switch at different times. The electrical resistance of the bridge can then be correlated with the domain structure between the larger elements and within the bridge [3.432, 3.437]. An example where a thin permalloy bridge connects two larger elements is shown in Fig. 3.32a-f [3.432]. Current flow changes the switching field, so the experiment can measure the spin-transfer torque efficiency.

Fig. 3.32a-f

Magnetic response imaged with differential phase contrast: (ae) Magnetic induction vector maps of a permalloy bridge constructed from scanning DPC images. The average magnetization direction within the bridge and the applied field are given for each image with magnetization directions shown in the color wheel (f). The sample is \({\mathrm{7}}\,{\mathrm{nm}}\) permalloy on silicon nitride patterned into two large pads connected by a thin bridge. The upper pad is larger and designed to have a lower switching field than the lower pad. Reprinted from [3.432], with the permission of AIP Publishing

Phase Transitions in Magnetic Materials

In situ TEM provides information not just on the process of magnetization reversal but also on the dynamics of the magnetic structure in complex phases and during transitions between phases. Cooling experiments have allowed measurement of magnetic configurations in a wide range of materials and conditions [3.438, 3.439, 3.440]. Figure 3.33a-ka shows magnetic configurations during a transition from a helix to a skyrmion lattice in MnSi [3.441]. Naturally, the transition between ferromagnetic and paramagnetic states has been intensely studied using cooling stages to reach temperatures around the Curie temperature. The types of details accessible include the range over which phases coexist, the formation of magnetic domains on cooling, and associations between microstructural features such as grain boundaries and nucleation points. Some of the interesting materials that have been studied in this way include the magnetic refrigerant \(\mathrm{La(Fe_{0.9},Si_{0.1})_{13}}\) [3.442] and perovskite materials with colossal magnetoresistance such as \(\mathrm{LaSrMnO_{3}}\) [3.443, 3.444, 3.445]. In this case, the mechanism of colossal magnetoresistance was related to behavior observed in the mixed-phase region, where application of a magnetic field creates channels connecting ferromagnetic regions and thereby changes the conductivity. The change in fraction of magnetic domains with temperature can help to determine the critical exponent describing the ferromagnetic phase transition [3.446].

Fig. 3.33a-k

Phase transformations in a magnetic material: (ac) An MnSi thin film showing two different magnetic configurations observed using Lorentz microscopy below \({\mathrm{22.5}}\,{\mathrm{K}}\): stripe domains at zero applied field and a skyrmion lattice with 6-fold symmetry at a field of \({\mathrm{0.18}}\,{\mathrm{T}}\) normal to the film. The magnetic field of a single skyrmion is enlarged. Reprinted with permission from [3.441]. Copyright 2012 American Chemical Society. (dk) Bright-field images of three phases of \(\mathrm{Ni_{2}MnGa}\): the parent phase at \({\mathrm{294}}\,{\mathrm{K}}\), an intermediate phase at \({\mathrm{207}}\,{\mathrm{K}}\), and martensite at \({\mathrm{170}}\,{\mathrm{K}}\) (P, I, and M, respectively). Both the crystal structure (via diffraction) and the magnetic structure (magnetic domains observed by Lorentz microscopy) are visible. BC and DW represent the bend contours and the domain walls, respectively. The magnetic structure does not change during transformation from P to I, but changes significantly during transformation from I to M, giving information on the magnetoelastic interactions within the structure. Reprinted from [3.447], with the permission of AIP Publishing

Materials such as multiferroics, which have phases that show different types of magnetic order (ferromagnetic, ferroelectric, ferroelastic) offer an even more complex playground for interesting structure-property relationships where in situ microscopy can contribute valuable information. An interesting class of such materials is ferromagnetic shape-memory alloys such as \(\mathrm{Ni_{2}MnGa}\) and CoNiAl, where a shape change is induced by a magnetic field. While these show interesting magnetic structures at room temperature [3.448, 3.449], cooling experiments allow magnetic domains to be correlated with grain structure as the materials transform martensitically to a tetragonal phase [3.447, 3.450, 3.451]. An example for \(\mathrm{Ni_{2}MnGa}\) is shown in Fig. 3.33a-kb, and ferroelectric properties are discussed in Sect. 3.4.4. In any such experiments, the effects of unintended variations in sample thickness should be considered for the most quantitative results.

3.4.3 Superconducting Materials

In situ TEM has provided a fascinating glimpse into the physics of superconducting materials. Several groups, most notably that of Tonomura and coworkers, have observed the presence and dynamics of vortices in superconductors using in situ techniques. Single vortices, with their magnetic flux of \(h/2e\), have an observable effect on the phase of the imaging electrons. Thus, Lorentz microscopy or holographic techniques can be used to determine their positions and characteristics. A cooling stage is of course required and the magnetic field is conveniently applied by tilting the sample in the existing field of the objective lens. High-voltage (\({\mathrm{1}}\,{\mathrm{MeV}}\)) TEM is often used to obtain the necessary resolution for these studies [3.452]. However, it is possible to image vortices in a conventional TEM with low-temperature stage [3.453].

These experiments have provided unique insights into superconducting behavior, in particular the formation of vortex lattices and the pinning of vortices, which must be controlled for practical applications of superconductors. Real-time imaging has allowed vortex dynamics to be related to microstructural features for several superconducting materials (Figs. 3.34a,b and 3.35a-d). The motion of vortices was first imaged in Nb foils below \({\mathrm{5}}\,{\mathrm{K}}\) (Fig. 3.34a,b). The effects of grain boundaries on vortex motion were immediately visible [3.456, 3.457, 3.458] and the role of dislocations in pinning vortices and nucleating locally ordered regions of vortices (the Abrikosov lattice) was demonstrated [3.459]. Nb specimens which had been irradiated with an FIB to produce artificial pinning centers showed fascinating vortex dynamics in which local regions of Abrikosov lattice formed with motion mainly at the boundaries of such lattices [3.460]. Regular arrays of vortices could be formed with period matching the pinning point period [3.461]. Vortex annihilation was also observed [3.462] and the interactions between vortices quantified [3.463] by analyzing their motion through the foil. One-way motion of vortices was controlled by FIB-patterning asymmetric channels [3.464].

Fig. 3.34a,b

Vortex motion in superconductors: Pinning of vortices in an Nb thin foil by an array of defects produced by \(\mathrm{Ga^{+}}\) FIB patterning. (a) At \({\mathrm{4.5}}\,{\mathrm{K}}\), the pinning sites create a domain boundary. (b) At a higher temperature of \({\mathrm{8}}\,{\mathrm{K}}\), the pinning sites typically act as cores of edge dislocations in the lattice. Reprinted from [3.454], with permission from Elsevier

Fig. 3.35a-d

Vortex motion in superconductors: A series of Lorentz micrographs of vortices in a field-cooled Bi-2212 film sample obtained at \({\mathrm{50}}\,{\mathrm{K}}\) with a variable magnetic field \(B_{z}\) perpendicular to the foil and a fixed field \(B_{x}={\mathrm{5}}\,{\mathrm{mT}}\) in-plane. \(B_{z}=0\), \(\mathrm{0.02}\), \(\mathrm{0.1}\), and \({\mathrm{0.17}}\,{\mathrm{mT}}\) in (ad), respectively. The arrangement of the vortices in chains is clearly visible. These chains are caused by interactions with horizontal vortices produced by the in-plane field. Reprinted with permission from [3.455]. Copyright 2002 by the American Physical Society

For high-temperature superconductors the pinning of vortices is weak and therefore particularly important to understand and control. The role of defects in vortex dynamics is directly visible in situ [3.465] and pinning by insulating particles has been imaged [3.466]. The effects of artificial pinning centers are highly temperature dependent, giving useful insight into the active mechanisms [3.467]. Furthermore, the vortices adopt unusual chain-like arrangements in these superconductors [3.455, 3.468], as shown in Fig. 3.35a-d. Correlation functions can be used to quantify ordered arrangements of vortices and how these depend on the applied magnetic field [3.453].

3.4.4 Ferroelectric Materials

Ferroelectric domain-boundary motion due to an applied electric field or stress has applications in information storage, and the piezoelectric properties of these materials make them useful as sensors, actuators, and transducers. Two key issues in the development of ferroelectric devices are fatigue, in other words the change in boundary dynamics resulting from repeated cycling, and the effects of film thickness and electrode material on boundary dynamics. Polarized optical microscopy and AFM have been successful in investigating the overall features of boundary motion, but naturally TEM is unparalleled in its ability to relate microstructural features to the motion of the boundaries. In situ experiments are carried out by using a specimen holder with electrical connections to apply an electric field, or to move the domains by heating or straining. A heating biasing holder allows domain dynamics to be studied at different distances from the transition temperature.

As with studies of magnetic materials, two general types of sample are used: mechanically thinned (polycrystalline or single crystal) bulk samples, or films on a thin substrate. For electrical biasing, contacts are patterned on the sample (both on one surface, or one on each surface for bulk samples), or the field may be applied by approaching with a scanning probe tip. A well-controlled specimen and field geometry are of course important for quantitative analysis; nonuniform sample thickness and defects from sample preparation may affect kinetics. For these reasons, experiments on thinned bulk materials tend to provide qualitative information whereas thin films, especially with a controlled electrode geometry and minimal processing, may generate quantitative results. For example, thin-film studies provide the opportunity to investigate the dead layer, in which surface pinning retards domain motion.

Domain motion under in situ biasing and/or heating, including electron-beam heating [3.470, 3.471], has been examined in \(\mathrm{BaTiO_{3}}\) [3.472], \(\mathrm{PbTiO_{3}}\) [3.473] and related materials: \(\mathrm{KNbO_{3}}\) is shown in Fig. 3.36a,b [3.469]. The in situ studies elucidate kinetics and mechanisms. For example, domain growth in \(\mathrm{BaTiO_{3}}\) has been seen to occur by tip motion and then by lateral wall motion [3.472]. In \(\mathrm{BaTiO_{3}}\) and \(\mathrm{KNbO_{3}}\) under heating, biasing and UV irradiation [3.469, 3.474, 3.475], the motion of \(90^{\circ}\) boundaries depends on their curvature and on locking interactions with neighboring domains, and may occur by rippling rather than rigidly. Interestingly, images show the presence of trapped charge at curved or tilted boundaries or at domain tips: charge buildup at boundaries is important in fatigue. The formation of charged domain walls and nonswitchable domains in \(\mathrm{PbZr_{0.2}Ti_{0.8}O_{3}}\) has been followed using holography [3.476]. In multiferroics, similarly complex domain-wall dynamics are also accessible in situ. For example, nucleation and motion of ferroelastic domains in \(\mathrm{BiFeO_{3}}\) take place below the coercive field [3.477]; domain-wall interactions with pre-existing domain walls have been observed by applying a local electric field with a conductive tip [3.478].

Fig. 3.36a,b

Domain-wall motion in a ferroelectric under biasing: Bright-field images of thinned single-crystal \(\mathrm{KNbO_{3}}\) viewed close to the [010] direction. Initially (a), the specimen contains numerous needle-like \(90^{\circ}\) domain walls as well as dislocations (dark wiggly features) and other domain-wall geometries (vertical stripes). After applying an electric field in (b), the needle-like walls move readily, coarsening the domain structure and forming straighter walls which then move less easily. This is because curved and tilted \(90^{\circ}\) domain walls (indicated by 1) must support a geometrically required Maxwellian displacement charge hence experience a direct force from an applied electric field, while charge-neutral domain-wall regions (indicated by 2) do not experience a direct force. This suggests an intrinsic mechanism of fatigue. Reprinted from [3.469], with the permission of AIP Publishing

By analogy with the magnetic phase transitions described in Sect. 3.4.2, ferroelectric phase transitions can be accessed in situ. An example is the electric field-induced transformation of incommensurate modulations in Sn-modified \(\mathrm{Pb(Zr{,}Ti)O_{3}}\) [3.480, 3.481]. Ferroelectric materials such as \(\mathrm{Ba_{2}NaNb_{5}O_{15}}\) show interesting incommensurate phases on heating or cooling. An incommensurate to tetragonal phase transition in \(\mathrm{Ba_{2}NaNb_{5}O_{15}}\) has been examined in diffraction [3.482, 3.483]. The nucleation of the incommensurate phase can be investigated [3.484] as well as the effect of beam-induced defects on the transformation [3.485]. It is possible to combine simulations with in situ studies on patterned ferroelectric elements [3.486].

In piezoelectric applications of relaxor ferroelectrics such as \(\mathrm{Pb(Mg{,}Nb)O_{3}}\)-\(\mathrm{PbTiO_{3}}\), the formation of cracks is important. As the electric field is cycled, the crack propagation pathways are seen to be along domain walls [3.487, 3.488, 3.489]. The structure of domain walls has therefore also been characterized [3.489, 3.490, 3.491].

In straining experiments, domain motion has been observed using high-voltage TEM to image through thick samples, for example in ferroelastic zirconia [3.492]. More recently, in conventional microscopes, the advent of nanomanipulation stages has provided the ability to probe ferroic thin films by applying a voltage or a mechanical strain, or both, normal to the surface. By using FIB cross-sectioned thin films, it is possible to image domain movement and quantitatively measure the necessary potentials to induce switching. Figure 3.37 shows an example in which the ferroelectric switching behavior of a \(\mathrm{BiFeO_{3}}\) film is investigated via an STM probe [3.479]. Ferroelastic domain switching can be induced using either mechanical or electrical probing with the same in situ TEM holder [3.493, 3.494].

Fig. 3.37

Switching in a ferroelectric by a probe: For the sample geometry illustrated schematically, the images show a thin cross-section of a \({\mathrm{100}}\,{\mathrm{nm}}\) \(\mathrm{BiFeO_{3}}\) film before and after switching by a \({\mathrm{4}}\,{\mathrm{V}}\) bias applied to a W tip. Switching occurs by \(71^{\circ}\) rotation of the polarization beneath the tip, with the atomic configurations indicated. From [3.479]. Reprinted with permission from AAAS

3.4.5 Light-Induced Phenomena

In addition to correlating microstructure and electrical or magnetic properties, it is also possible to probe the response of a material to light in situ. Light can be brought to the sample through a fiber optic or mirrors. The presence of light can drive or affect catalysis reactions, as discussed in Sect. 3.3.3. Light can alter the electrical properties of nanostructures, with increasing conductivity measured during simultaneous electrical biasing and illumination [3.496]. Light can also modify the structure, causing melting or crystallization. As an example, the transformation of an amorphous confined Si volume to a single crystal of Si using laser melting is shown in Fig. 3.38. This was achieved [3.495] using a lensed fiber to couple the optical near-field of the laser illumination into the TEM.

Fig. 3.38

Light-induced crystallization with near-field optical probe in a TEM: In the sample geometry illustrated schematically, a fiber-optic probe is manipulated in three dimensions with a piezo motion control system to approach the sample surface in a direction orthogonal to the electron beam for in situ optical near-field probing. The bright-field image series shows an Si cap, as-prepared by FIB, which is melted and recrystallized into a polycrystalline cap structure (pc-Si) by irradiation by several nanosecond laser pulse shots. Further laser pulses transform the cap into a single crystal (sc-Si). Inset are electron diffraction patterns with final zone axis \([\bar{1}12]\). Reprinted with permission from [3.495]. Copyright 2012 American Chemical Society

Light can also be used to probe a material for optical spectroscopy inside the TEM, with light collected simultaneously for measurements such as Raman spectroscopy. One method to perform in situ TEM Raman is to achieve optical access through a viewport installed on the microscope column [3.497, 3.498]. A parabolic mirror is mounted near the specimen to focus the incoming light and collect the outgoing signal to direct it back through the viewport to an optical spectrometer. This method has been used to perform in situ TEM Raman spectroscopy on diamond [3.497]. In situ TEM Raman measurements on silicon and carbon nanotubes inside an environmental TEM have also been performed, employing optical illumination in the far field with a spot of several microns [3.498], and demonstrated the measurement of sample temperature via the downshift of Raman peaks. A modular system for in situ TEM Raman of \(\mathrm{MoS_{2}}\) [3.499] adapts the lensed-fiber approach described above [3.495] with a fiber-based Raman probe for the optical excitation and collection optics, with localized laser illumination applied through a piezo-driven nanomanipulator. In the future it should be possible to integrate near-field nano-optical methods, to combine the high spatial resolution of the TEM with nanoscale optical probing.

3.4.6 Summary

In situ magnetic, electrical, and optical experiments have probed diverse physical phenomena to provide detailed new data. With functional materials appearing in increasingly varied application areas, we anticipate increased demand for the sort of information that only in situ experiments can provide. We anticipate that in situ microscopy will see continued development to address thermoelectric, switching and fatigue phenomena and complex combined stimuli. Advanced imaging will be especially useful in combination with nanoscale sample design that creates well-defined geometries. We envisage interesting results if such studies also include high-speed recording and analytical microscopy. We also note that in real life, materials are usually buried within devices, giving boundary conditions for the electric, magnetic, and strain fields that are different from those found in a TEM foil. Care must be taken that this does not limit the utility of in situ TEM. For modeling practical systems this issue must be addressed, perhaps by using tomographic or other analytical techniques to pick out the material of interest from within a larger structure.

3.5 Mechanical Deformation of Materials

TEM is uniquely suited for studying the mechanical properties of materials, since the fundamental deformation mechanisms are readily observed: the sensitivity of TEM to lattice distortion allows visualization of both elastic and plastic deformation via strain fields. In situ deformation studies aim to impose a known stress on a sample and measure the response quantitatively. Starting in the late 1950s, in situ straining stages were developed that provided dynamic observations of dislocation motion in metals [3.500]. Today, this is a very active area of research. Typically, mechanical deformation is achieved using a straining stage to pull the specimen uniaxially or biaxially with piezo or mechanical linkages. Alternatively, stress can be imposed by heating samples with a thermal expansion mismatch or an inbuilt stress, such as epitaxial films. It is also possible to apply mechanical stimulus locally to an electron-transparent region using nanoindentation or STM probing holders. Straining can even be carried out in a controlled atmosphere to simulate phenomena such as hydrogen embrittlement. The sample itself may be a thin film, a nanostructure or a bulk material, perhaps notched to initiate cracks. Recent advances in sample preparation strategies and microfabricated loading devices have enabled precise alignment of crystalline samples or mechanical testing of individual nanostructures.

In situ experiments play a key role in understanding fundamental deformation phenomena because it is necessary to see the internal structure of a specimen; surface techniques can not give the whole picture, even if, for example, the signature of dislocations can sometimes be seen on the surface. Of course, TEM samples are thin in the beam direction, and the presence of free surfaces nearby must be considered in interpreting results. This can actually be put to good use, though, in studies of tribology or the deformation of individual nanostructures in situ.

3.5.1 Deformation Phenomena

Structural changes during deformation, such as grain-boundary motion, dislocation motion and cracking, have been studied in situ for just about every class of bulk material. These experiments have yielded useful information on dislocation interactions, pinning, the transfer of strain across boundaries, and the effects of temperature. Materials also respond mechanically to irradiation and this is discussed in Sect. 3.7. The study of deformation phenomena has a history going back to the start of high-voltage microscopy, and a review of the pioneering work can be found in [3.501]. Design rules were established later on for extreme heating and straining holders [3.502, 3.503]. More recently, the introduction of quantitative measurement devices has enabled precise knowledge of the stress inside a sample [3.504, 3.505]. Recent reviews illustrate the opportunities this creates [3.506, 3.507].

Deformation Phenomena in Single Crystal and Polycrystalline Materials

For single crystals, it is possible to relate an applied stress to the motion and interactions of specific types of dislocations. In situ observations on single crystals provide detailed measurements of dislocation generation and multiplication mechanisms, as well as geometry, dissociation, interactions, and slip systems (Fig. 3.39). Conversely, observations made on polycrystalline materials allow us to understand the important process of strain transmission across grain boundaries and phenomena such as grain-boundary sliding and grain rotation. Heating, straining, and heating/straining experiments have been the focus of several groups and recent reviews [3.508, 3.509, 3.510, 3.511, 3.512]. We illustrate below the broad range of materials that has been examined in situ.

Fig. 3.39

Dislocation motion on straining: Image series showing single-ended dislocation sources lying on a set of parallel slip planes at \({\mathrm{140}}\%\) strain in single-crystal Al. The schematic diagram shows the dislocation configuration in the first image. The dislocation sources lying on parallel slip planes are numbered 1, 2, and 3 and the previously generated dislocations from source 1 and 2 are labeled 1\({}^{\prime}\) and 2\({}^{\prime}\), respectively. The source size (\(L\)) is measured to be \({\mathrm{7020}}\,{\mathrm{nm}}\). On generation, the dislocations glide and approach the surface with their lines parallel to the surface in edge character and they intersect the surface perpendicularly due to image forces. From [3.513]

In single crystals, dislocation motion has been studied in metals such as Mg, Cu and Al at moderate temperature, while refractory metals require high temperatures (Fig. 3.40) [3.502, 3.509]. A uniaxial stress is most commonly applied, although, for example, cyclic shearing has been described for Al single crystals [3.514]. Plastic deformation does not always require dislocation generation or motion, and in V, Mo and other body-centered cubic metals, deformation may occur by formation of point defects rather than dislocations. This too can be observed in situ with appropriate diffraction and imaging techniques [3.515] (Fig. 3.41). Twinning is an important mode of deformation in many alloys, especially hexagonal alloys of Mg or Ti, and in situ TEM has provided insight into how and when twins form in these materials [3.516, 3.517, 3.518].

Fig. 3.40

Deformation of bulk crystals: Dislocations moving away from a localized dislocation source during in situ deformation of an \(\mathrm{MoSi_{2}}\) single crystal at about \({\mathrm{900}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). The specimen plane is (010) and \(\boldsymbol{g}=[220]\). The loading axis is [201], which is a soft orientation, so dislocations of Burgers vectors \(\tfrac{1}{2}\langle 111\rangle\) are activated on {110} planes. The traces of these slip planes run horizontally. Owing to the generation of these dislocations in localized sources, as shown here, planar slip dominates. From [3.502], reproduced with permission

Fig. 3.41

Deformation of bulk crystals: Deformation analysis using diffraction, showing deformation of a refractory metal without formation of dislocations. The diffraction patterns are obtained near the crack tip during deformation of a notched vanadium foil at the times indicated. As the internal stresses increase, the spots expand perpendicularly to the crack so appear elongated. The central spot is masked to prevent overexposure. Images obtained simultaneously do not show clusters of point defects in V, which is different from the case of Au strained under the same conditions. Reprinted from [3.515], with permission from Elsevier

Bicrystals are an intermediate case between single crystal and polycrystalline metals that provide the opportunity to apply the precision of single-crystal experiments to studies of strain transfer across boundaries. For example, in symmetric \(\Upsigma{}3\) Fe-\({\mathrm{4}}\,{\mathrm{at.\%}}\) Si bicrystals, the primary slip system has been determined as well as the particular slip systems that can transmit strain across the boundary [3.519]. In situ TEM nanocompression was used to test directly the prediction of frictionless glide of an incommensurate grain boundary in Au [3.520].

Polycrystalline metal deformation has been investigated by both straining and heating. The sample geometry must be controlled, for example in terms of thickness uniformity, to avoid artifacts, and surface passivation can affect the results. In polycrystalline Ni, straining experiments have shown that the failure mode depends on grain-boundary structure [3.521], and deformation proceeds by diffusion-controlled grain-boundary-mediated processes rather than dislocation motion within grains [3.522]. Thermal cycling experiments show grain-boundary motion (Sect. 3.2.1) and the interactions of dislocations with grain boundaries and precipitates [3.51]. For Cu and Al, increases in yield stress have been related to particular types of dislocation motion [3.523, 3.524]. Threading dislocation motion and the effect of the passivation layer have also been studied [3.525, 3.526].

Intermetallics have been strained in situ to understand the mechanism of their high-temperature toughness and phase transformations. Dislocation dynamics change as a function of temperature in these materials [3.527]. Dislocation interactions, cross slip, and nucleation of loops control work hardening, ductility, and the critical resolved shear stress [3.528, 3.529, 3.530]. Dislocation motion during creep has also been examined [3.531].

Polysynthetically twinned TiAl crystals have lamellar structures with well-defined boundaries. Similarly to the bicrystal experiments, strain propagation across these boundaries can be quantified in situ [3.532, 3.533]. Transformations in shape-memory alloys have already been mentioned in Sect. 3.2.2, and in situ straining relates the microstructural changes during deformation to the stress-strain response [3.63, 3.70, 3.72]. The cracking of intermetallics is also important in applications. At crack tips, dislocations in NiAl [3.534], and amorphization in NiTi ordered alloys [3.535] can be analyzed.

Quasicrystals have interesting mechanical properties and dislocation geometry. In situ heating and straining has led to the identification of shear systems and models for dislocation motion in these materials [3.536, 3.537, 3.538, 3.539, 3.540].

Ceramics have been important subjects of study since the 1970s, with an interest in comparing dislocation motion over a range of temperatures, for example in zirconia [3.508], alumina [3.503], and \(\mathrm{MoSi_{2}}\) [3.541]. In quartz, strain-induced phase transformations are addressed with combined straining/heating experiments [3.542]. In situ TEM nanocompression of natural quartz revealed both dislocation plasticity and reversible twinning [3.543]. The stress to generate twins was directly measured, leading to the possibility of using twinning in quartz for paleopiezometry.

Finally, in semiconductors, kinetic studies have been important in showing that dislocation motion is consistent with glide governed by the Peierls mechanism [3.511]. From experiments recording dislocation motion and pinning, the kink mean free path and formation and migration energies can be determined [3.544, 3.545, 3.546]. Radiation enhances the glide of dislocations by changing some of these parameters (Sect. 3.7). Low-resolution, dark-field imaging is generally used to record dislocation dynamics. However, it is also possible to observe the thermal motion of kinks in Si using a high-resolution forbidden reflection imaging technique [3.547]. The distribution and pinning of individual kinks is used to infer the kink formation energy and unpinning barrier, and time-resolved imaging shows directly that kink migration is the rate-limiting step in dislocation motion. Similar imaging techniques should be applicable to other materials and could therefore advance our understanding of dislocation motion further [3.548].

Deformation of Multiphase, Composite, Layered, or Amorphous Materials

Deformation experiments in multiphase materials, such as dispersion-strengthened alloys, are particularly important in showing how strain is transmitted between the components. As an example, dispersion-strengthened Al alloys have been examined extensively to characterize the nucleation of dislocations at precipitates and their motion past precipitates [3.549, 3.550]. These phenomena are critical in high-temperature deformation of such alloys. Other phenomena such as grain-boundary migration, coalescence, and elimination of sub-boundaries occur during dynamic continuous crystallization on loading in the TEM [3.551, 3.552], with in situ studies helping to understand the processes. Steel is of course a key multiphase material whose behavior in situ has been studied for some time to understand dislocation motion and slip transmission through interfaces [3.553, 3.554, 3.555, 3.556]. Many other nanocomposite materials provide interesting examples of dislocation–interface interactions. One example is shown in Fig. 3.42 [3.557] and others are reviewed elsewhere [3.558].

Fig. 3.42

Deformation in a Cu-Nb multifilamentary composite: Nanocomposite materials are used in wires requiring high strength and conductivity. This sample was prepared by cold drawing an Nb rod inside a Cu jacket several times until a composite was formed consisting of Cu-\({\mathrm{17}}\,{\mathrm{vol.\%}}\) Nb with \(\mathrm{10^{7}}\) Nb filaments \({\mathrm{40}}\,{\mathrm{nm}}\) across in Cu. The Cu channels have \(\langle 111\rangle\) texture, the Nb filaments have \(\langle 110\rangle\) texture, and the Cu–Nb interfaces are semicoherent. The image series shows the introduction of dislocations under an applied force of \({\mathrm{8}}\,{\mathrm{N}}\). At this point Cu is elastically deformed while Nb is still plastically deforming. The area shown is a Cu channel of width about \({\mathrm{100}}\,{\mathrm{nm}}\) (white) between two Nb filaments (black). In the first image, five dislocation loops are present in parallel planes (a–e). A few seconds later the sixth and seventh loops appear (f, g). By the end of the sequence 13 loops are present, and they have interacted to produce a honeycomb pattern indicated by outlines. This dislocation behavior is used in a plastic flow model which explains the very high ultimate tensile stress of fcc-bcc nanocomposite structures. From [3.557]. Reprinted by permission of the publisher (Taylor & Francis Ltd, http://www.tandfonline.com)

Techniques developed for multiphase materials extend readily to engineered multilayers, to examine the interaction of dislocations or cracks with interfaces. Cross sections of multilayers may be prepared using an FIB to form a thin section at which cracks will initiate [3.559, 3.560]. Alternatively, multilayers (or thin films) may be integrated into a mechanical testing stage. For example, experiments using a piezo stage for cyclic loading have shown that cavitation is the dominant fatigue damage mechanism in solder thin films [3.561]. This stage holds a multilayered structure that is fabricated to include the material of interest as well as a piezoelectric layer. Nanoindentation is also possible for these materials, showing for example that during dislocation–interface interactions in Al\(/\)Nb multilayers, preferential storage of dislocations occurs at interfaces rather than within the layers, and recovery of dislocations is through climb in the interfaces [3.562].

Deformation in amorphous materials may appear hard to access in situ since deformation features are not typically visible. However, in situ TEM does show a range of behavior during deformation of metallic glasses. Examples include the formation and localization of shear bands in metallic glass nanopillars via nanocompression [3.563], formation of a liquid-like layer [3.564], and high ductility (up to \({\mathrm{45}}\%\) strain) and necking [3.565] during in situ tensile testing.

The Effect of a Controlled Environment on Deformation Processes

By combining mechanical testing with controlled-environment TEM, the mechanism of hydrogen embrittlement of steels has been addressed (Fig. 3.43a-d). In situ straining experiments carried out in an environmental cell showed that solute hydrogen increases the speed at which dislocations move through the material, as well as increasing the rate of crack propagation [3.566, 3.567, 3.568]. Quantitative analysis of dislocation dynamics on adding and removing hydrogen were used to distinguish between competing mechanisms for embrittlement. A model was confirmed in which hydrogen shields the dislocation from interactions with other strain centers such as pinning points, other dislocations, or solutes. Very few materials have been strained in a controlled atmosphere, although it has been shown [3.569] that the incorporation of hydrogen from a plasma increases the motion of dislocations in semiconductors. Further studies could provide a sensitive probe of dislocation properties and interactions relevant to real-world applications.

Fig. 3.43a-d

Dislocation motion in a hydrogen atmosphere: Image series showing the effect of H on the mobility of dislocations in 310S stainless steel. (a) A static dislocation configuration in vacuum under constant load; (b\({\mathrm{35}}\,{\mathrm{Torr}}\); (c\({\mathrm{90}}\,{\mathrm{Torr}}\) of hydrogen; (d) composite image formed by subtracting (c) from (a). The static configuration is altered by changes in the interactions between dislocations caused by solute hydrogen. Reprinted with permission from [3.566], John Wiley and Sons

Relaxation of Epitaxially Strained Materials

A different method of straining a material is via epitaxial growth on a lattice mismatched substrate. As more material is deposited the stress builds up. Above a critical thickness the layer may relax by several means, including forming a network of dislocations at the layer–substrate interface (Fig. 3.44a-ca,b). This relaxation process has been examined in situ by growing a layer that is thicker than the critical thickness, but at a temperature too low to allow dislocation nucleation or propagation. When a plan-view specimen is then prepared, relaxation can be triggered by heating in situ. The advantage of this type of experiment is that the stress state is very well defined.

Fig. 3.44a-c

Dislocation dynamics in SiGe heterostructures: (a) Image showing the typical structure of a strain-relaxed SiGe heterostructure imaged in plan view. The structure consists of \({\mathrm{300}}\,{\mathrm{nm}}\) Si\(/\)\({\mathrm{150}}\,{\mathrm{nm}}\) \(\mathrm{Si_{80}Ge_{20}}\)\(/\)Si(001). Dislocation loops consist of straight segments along both upper and lower SiGe–Si interfaces connected by a threading segment that passes through the SiGe layer. Each loop thus appears as a closely spaced pair of straight lines in the bright-field image. (b) Higher magnification view of a threading dislocation segment imaged in a (\(\boldsymbol{g},\,3\boldsymbol{g}\)) weak-beam dark-field condition with \(\boldsymbol{g}=220\). Reprinted from [3.306], with the permission of AIP Publishing. (c) Measured velocities of dislocations as a function of temperature, in \(\mathrm{\AA{}/s}\), in structures similar to that in (a,b). The structures consist of \({\mathrm{300}}\,{\mathrm{nm}}\) Si\(/h\) nm \(\mathrm{Si_{1-\mathit{x}}Ge_{\mathit{x}}}\)\(/\)Si. In the upper graph, full symbols are for \(x=0.14\), \(h={\mathrm{120}}\,{\mathrm{nm}}\) and open symbols are for \(x=0.17\), \(h={\mathrm{145}}\,{\mathrm{nm}}\); in the lower graph, full symbols are for \(x={\mathrm{0.18}}\), \(h={\mathrm{87}}\,{\mathrm{nm}}\) and open symbols are for \(x={\mathrm{0.18}}\), \(h={\mathrm{58}}\,{\mathrm{nm}}\). Predictions of the double-kink theory are shown as solid lines, with the same composition and layer thickness and energy parameter \(Fk={\mathrm{1.0}}\,{\mathrm{eV}}\) at zero stress in Si. After [3.570]

These studies are motivated by their importance to the microelectronics industry, since the positions of individual dislocations, the density of threading arms, and the motion of dislocations during processing are known to affect device performance. The most detailed experiments have therefore been made on \(\mathrm{Si_{1-\mathit{x}}Ge_{\mathit{x}}}\) layers grown on Si, with \(x\) ranging from about \(0.05{-}0.3\) [3.570, 3.571]. Figure 3.44a-cc shows dislocation velocity as a function of layer parameters. Such experiments showed that dislocation propagation is through motion of single kinks, whereas if the \(\mathrm{Si_{1-\mathit{x}}Ge_{\mathit{x}}}\) is capped with a top Si layer, the diffusive kink pair model of dislocation propagation fits the measurements. Activation energies and prefactors were measured for dislocation nucleation and propagation. Note that accurate calibration is necessary if TEM-derived parameters such as activation energy are to be meaningful. In this case, the sample temperature was calibrated using Si regrowth (Sect. 3.2.1), and finite element analysis was used to determine the conditions under which bending of the thin foil could be neglected [3.15]. The parameters determined in situ were used in a processing model for microelectronic device design [3.571].

A similar experimental approach allows other features of the relaxation process to be examined as well. Dislocation nucleation has been studied in SiGe/Si implanted with F [3.16] and with He, which forms platelets and bubbles [3.572, 3.573]. A few other epitaxial systems have been studied, such as \(\mathrm{BaTiO_{3}}\) on \(\mathrm{SrTiO_{3}}\) [3.574], ZnSe on GaAs [3.575], and Al on \(\mathrm{Al_{2}O_{3}}\) [3.576]. Clearly, this technique could yield useful information for many other materials.

We finally note that dislocation configurations in thin films, such as in Fig. 3.44a-ca, are somewhat random. However, in well-controlled geometries such as relaxed epitaxial islands or surfaces with modulated stress fields, dislocation locations can be predicted accurately [3.577, 3.578]. This suggests opportunities for in situ TEM to provide quantitative information on dislocation nucleation and growth in finite nanostructures with built in strain, such as epitaxial islands, nanowires, or ribbons.

3.5.2 Nanomechanical Testing Methodologies

TEM is particularly appropriate for studying the mechanical properties of small volumes of material. In most of the work discussed so far, the same stimulus (strain, heat) is applied to the entire specimen. By applying the stimulus to only a small part of the specimen, the correlation between mechanical input and structural response can be made more precise. In this section we discuss experiments in which small volumes of a specimen are mechanically deformed through mechanical probing (nanoindentation or tension/compression experiments), or individual nanostructures are deformed directly. These experiments make use of MEMS actuation, nanomanipulation techniques and FIB sample preparation to localize the mechanical input to a small area observed by the electron beam.

In Situ Nanoindentation

Nanoindentation involves plastic deformation of a material using a small probe. The probe has the thickness of a typical TEM sample and the experiment is therefore nicely compatible with in situ observation. Integrating a nanoindenter into a specimen holder has enabled localized deformation experiments on single crystals, polycrystalline materials, layered structures, and individual nanostructures. The experimental setup [3.580, 3.581, 3.582, 3.583, 3.584, 3.585] must include some way of making sure the indent occurs in an electron-transparent region of the specimen. For example, to visualize elastic deformation and dislocation formation in Si, a vertically mounted sample can be etched to form a ridge which projects out into the beam [3.581]. Si shows dislocation plasticity on room-temperature nanoindentation [3.586] because the nearby surfaces allow for easy dislocation nucleation in what should otherwise be a brittle material at room temperature. Dislocation and grain-boundary motion in other materials can be studied by depositing the material in a thin film onto an etched Si substrate. In ultrafine-grained Al, for example, grain-boundary motion occurs under the tip [3.587]. This does not occur in Al-Mg, suggesting that solute drag is important [3.588]. Nanoindentation of martensitic steel is shown in Fig. 3.45a-d [3.579], and the formation of dislocations and strain transfer across grain boundaries are visible. Nanoindentation allows the nature and pathway of crack propagation to be determined [3.585, 3.589]. It is also possible to study deformation in multilayers, for example to understand the onset of plasticity in \(\mathrm{In_{\mathit{x}}Ga_{1-\mathit{x}}As}\) multilayers [3.590].

Fig. 3.45a-d

Nanoindentation of steel: Image series showing an Fe-C martensite with a low-angle grain boundary (arrowed), (a) before indentation, (b) at \({\mathrm{21}}\,{\mathrm{nm}}\) penetration depth, showing dislocation emission beneath the indenter, (c) at \({\mathrm{46}}\,{\mathrm{nm}}\) penetration depth, showing dislocation pileup at the grain boundary, and (d) at \({\mathrm{84}}\,{\mathrm{nm}}\) penetration depth, demonstrating dislocation emission at the far side of the grain boundary. From [3.579], reproduced with permission

The more recent development of quantitative force-sensing nanoindentation holders has provided unique insight into the early stages of plastic deformation in metals, contributing to the understanding of ex situ nanoindentation data. This approach showed that the initiation of dislocation plasticity can occur before sustained contact loading, at force levels that are barely discernable during ex situ nanoindentation [3.504]. As shown in Fig. 3.46, correlation of the in situ video and the load versus displacement curve demonstrated that the initiation of defects occurred prior to the first sustained load drop that would normally be interpreted as the yield point in a displacement-controlled experiment. This demonstrates that yielding during a nanoindentation test is a more complicated phenomenon than simply nucleation of defects at the ideal strength.

Fig. 3.46

Quantitative nanoindentation measurements in Al: Structure is correlated with the load versus displacement curve during an in situ TEM nanoindentation test as a Berkovich conductive diamond indenter is used to perform nanoindentation of an initially dislocation-free submicron Al grain. Images show point 1 where the diamond indenter approaches the defect-free Al grain from the bottom of the video frame, and point 2 after the first dislocation burst, occurring before sustained contact that is denoted by the large increases in load that occur around \({\mathrm{70}}\,{\mathrm{nm}}\) in displacement. From [3.504]

Tribology and Nanomanipulation

Nanoindentation is only one of several interesting mechanical experiments that can be carried out in holders incorporating a scanning tip with piezoelectric actuators. A tip can also be used for fundamental studies of tribology and wear by scraping it across a surface. For metals, such observations give insight into sources of friction such as plowing deformation, a common explanation for high friction and wear rates [3.591]. In silicon, observations of wear made with a diamond nanoindentation tip were found to be consistent with atomic attrition and inconsistent with fracture or plastic deformation [3.592].

The tip can also be used to form small necks or grain boundaries by touching it to the surface, deforming both materials (Fig. 3.47a-d). Electrical transport measurements across such necks were discussed in Sect. 3.4.1. Mechanical information is also available, such as the strength of the boundaries formed as two oxidized Si tips come into contact [3.594], and the formation of twins in necks of Si at a W tip [3.595]. In Au [3.596, 3.597], experiments have shown formation of a neck between Au contacts, with compression, shear, deformation, slip, and twinning. These videos give a stunning visual impression of the interaction between tip and surface during scanning, and allow a study of friction at the nanoscale [3.598]. The structures visible in these thin necks can be compared with those seen in narrow regions between holes in plan-view specimens (Sect. 3.3.1).

Fig. 3.47a-d

STM surface manipulation: High-resolution image series showing atomic-scale removal-type mechanical processing of an Au surface. A region six atomic columns wide on the fixed side (B) is removed by the Au tip on the mobile side (A). On both sides the beam is parallel to the [110] axis. The time is (a\({\mathrm{0}}\,{\mathrm{s}}\), (b\({\mathrm{1.2}}\,{\mathrm{s}}\), (c\({\mathrm{3.5}}\,{\mathrm{s}}\), and (d\({\mathrm{5.8}}\,{\mathrm{s}}\). Boxed circles show the unit cell of Au. These images give a visual impression of events which may occur during STM operation as well as phenomena associated with friction. From [3.593], reproduced with permission

Recording STM images as the tip moves across the sample surface yields information that is helpful in interpreting scanning probe microscopy in general. Tip–substrate interactions have been observed in reflection-mode geometry as a graphite specimen was imaged using STM [3.600, 3.601], and changes in the surface were attributed to shearing and abrasion. Using a similar holder in a UHVTEM, the atomic configuration at a tip surface has been correlated with the resolution of STM images it produced [3.595]. Tip motion across a stepped surface provides information on the effects of rastering and surface topography on lateral displacement [3.602, 3.603].

Quantitative Stress Versus Strain Measurements

A long-standing challenge in all in situ TEM mechanical testing experiments is to obtain quantitative data. An interesting technique for measuring local stress is through the curvature of dislocations between pinning points [3.510, 3.513], as was shown in Fig. 3.39. However, this method relies on slowly moving and typically low-density dislocations. Recently, the integration of quantitative force sensing and precise sample micromachining and manipulation has led to more general quantitative methods.

Materials deform in response to stress, and in order to measure the stress applied to the material we need to know the force and the cross-sectional area the force is applied to. For TEM samples, at least one dimension is electron transparent, providing a geometrical constraint. Depending on the loading method (tension, compression, or bending) this initial constraint limits the other two geometrical dimensions to usually within an order of magnitude of the thickness in the beam direction. Therefore, quantitative mechanical testing in situ almost always involves micromachining a bulk sample to the required dimensions (usually with an FIB) or manipulation of a nanostructured sample into a loading platform. MEMS force sensors and displacement measurement systems are integrated with the film of interest to enable quantitative experiments. Stages for uniaxial tensile testing of freestanding thin films have been demonstrated [3.605, 3.606, 3.607, 3.608], but typically these systems require cofabrication of the sample itself with the testing device.

With the integration of high-resolution quantitative force sensors into TEM nanoindentation holders [3.505], the control of sample geometry and/or positioning of the sample becomes a critical factor in performing quantitative testing. Samples for in situ TEM mechanical testing using a probing holder are typically FIB-milled from almost any material, in sizes ranging from nanometers to microns. FIB-milled samples have produced a wealth of knowledge about the effect of size on strength. For example, in situ data [3.609] confirmed a hypothesis related to small-scale deformation known as mechanical annealing, where defects leave the sample due to the nearby surfaces, increasing the stress to deform the sample. Observations such as this would not be possible without quantitative measurements correlated to nanoscale observations of deformation phenomena.

Through in situ nanocompression testing in a TEM it is possible to correlate mechanical data with the microstructural response of a material at length scales that are difficult to approach with other testing methods. For example, using an FIB to produce samples from the matrix of a complex alloy makes it possible to separate dislocation plasticity effects in the matrix from precipitate interactions [3.610].

A further extension of FIB-based sample preparation for in situ TEM nanomechanical tests is the technique of quantitative in situ TEM tensile testing [3.612, 3.613, 3.614], Fig. 3.48a-d. By milling a dog-bone-shaped sample out of a bulk material with an FIB and also milling an inverted diamond gripper, uniaxial quantitative tensile testing is possible with the picoindenter system. Tensile tests have numerous advantages over compression testing, including decreased specimen taper, increased flexibility in sample geometry, and a homogeneously deforming gage section. However, FIB-induced artifacts have clouded these results at the smallest scales, as the machined specimens contain a high density of irradiation defects. Examination of Mo pillars annealed in situ in the TEM to remove FIB-defects has shown [3.615] that FIB-induced defects do indeed result in lower yield stresses than otherwise would be realized in similar non-FIBed specimens. In general, the effect of the FIB-induced defects seems to be an increase in initial defect density that can lead to easier dislocation nucleation than in pure single crystals.

Fig. 3.48a-d

Uniaxial in situ nanomechanical testing techniques: (a) Low-magnification image showing compression geometry for a single-crystal Cu sample. (b) Higher magnification of flat-punch diamond indenter approaching an individual Cu nanopillar. (c) Image showing several tensile samples and a structured conductive diamond tip that acts as a gripper. (d) Higher magnification view of the sample and gripper aligned before testing. Notice the rather high defect density resulting from the fabrication process. From [3.599]

Mechanical Properties of Nanostructures

The mechanical properties of elongated nanostructures are a natural subject for in situ studies. For carbon nanotubes, mechanical parameters such as Young's modulus were first measured by observing the vibration of tubes that extend out as cantilevers [3.616]. The nanostructures vibrate because of coupling to motion in the stage. By using a stage with piezo drives to induce an alternating electrical field between a nanowire and an electrode, controlled frequency vibrations can be set up in the nanowire [3.386, 3.617]. In ZnO wires of rectangular cross section, each vibration direction has its own resonances from which the modulus and time constant or Q factor can be derived [3.604] (Fig. 3.49a-e). These measurements of bending modulus can be related to defects in individual structures [3.618]. In fact, the mechanical resonance depends so sensitively on the structure that modulus measurements have potential use for measuring small masses.

Fig. 3.49a-e

Mechanical properties of a nanobelt: Images showing a cantilevered ZnO nanobelt (a) stationary, (b) at the first harmonic resonance in the \(x\)-direction, frequency \({\mathrm{622}}\,{\mathrm{kHz}}\), and (c) at the first harmonic resonance in the \(y\)-direction, frequency \({\mathrm{691}}\,{\mathrm{kHz}}\). (d) An enlarged image of the nanobelt with the diffraction pattern in the inset. The projected shape of the nanobelt is apparent. (e) The FWHM of the resonance peak measured from a different ZnO nanobelt with resonance at \({\mathrm{230.9}}\,{\mathrm{kHz}}\). Reprinted from [3.604], with the permission of AIP Publishing

The mechanical properties of elongated structures can also be measured by bending or stretching them with an STM tip. Individual Si nanowires [3.621] or CNTs [3.111, 3.622] can be bent, stretched and broken, and stress and strain can be measured during deformation [3.623]. CNTs can be welded to a tip [3.366, 3.624]. A tip can test the mechanical response of telescoping CNTs [3.611] (Fig. 3.50), with simultaneous electrical measurements of conductivity (Sect. 3.4.1). A less quantitative application is a stage developed for stretching chains of nanoparticles, of interest for their use as reinforcing fillers and their presence in diesel exhausts [3.625]. These tip-based experiments can involve analytical measurements as well as imaging; for example, EELS can be used [3.626] to probe changes in electronic structure during bending of multiwalled and bundled single-walled CNTs and correlate the changes with deformation.

Fig. 3.50

Mechanical properties of a carbon nanotube: Image series from a video showing the in situ telescoping of a multiwalled carbon nanotube using an STM tip. In the first five frames the core nanotubes are slowly withdrawn to the right. In the sixth image, which occurred one video frame after the core was released, the core has fully retracted into the outer nanotube housing as a result of the attractive van der Waals force. These experiments demonstrate the reversibility and low friction of the bearing and place limits on interwall frictional forces. From [3.611]. Reprinted with permission from AAAS

Quantitative mechanical testing of individual nanostructures is possible with mechanical probing holders by directly compressing, bending, or pulling a nanostructure oriented on a substrate. For example, Fig. 3.51a-f shows in situ compression of nanocrystalline CdS synthesized into a spherical shell geometry [3.619]. The measured nanoshell strength showed it to be capable of withstanding extreme stresses (approaching the ideal shear strength of CdS). In this example, it is difficult to imagine how the strength of this nanoparticle would have been tested otherwise. In situ TEM mechanical testing not only provides observations of the deformation in a material but also lets one actually find and address the nanoparticle with a testing device. Ex situ, it is difficult to know the degree of alignment with the nanostructure and whether a large drop in load is due to fracture (Fig. 3.51a-ff) or the nanoshell slipping away (which also occurred in some tests).

Fig. 3.51a-f

Compression-to-failure test of an individual nanocrystalline hollow CdS sphere. (ac) Extracted video frames of the in situ compression test, at the times indicated. The estimated contact diameter is shown in red in (b). (d) and (e) are dark-field TEM images of the hollow nanocrystalline CdS ball resting on the Si substrate prior to and after the compression test, respectively. (f) Load and displacement data from the loading portion of the in situ test plotted versus time. The experiment was run in displacement control. From [3.619]

As noted above, tension tests have advantages over compression tests for studying plastic deformation. In tension, it is possible to measure true stress versus true strain and measure mechanical characteristics such as necking and hardening. Recently, a method of testing individual nanowires in tension has been shown by coupling a microfabricated push-to-pull device with a nanoindentation holder. This is shown in Fig. 3.52a-e [3.620] where individual Mo-alloy nanowires are placed across a gap and deformed in tension. In order to study the local strain in a segment of the nanowire under tension, digital image correlation from the in situ TEM video was used [3.620] to show that under tension the exhaustion of a dislocation source led to hardening of the material as the sample became starved of mobile defects to accommodate the imposed deformation.

Fig. 3.52a-e

Quantitative tensile testing of individual nanostructures: A push-to-pull geometry is used to achieve quantitative testing, illustrated for a composite material. (a) Mo-alloy nanofibers, with tailorable dislocation densities, after etching away the NiAl matrix. (b) The fibers are picked up with a micromanipulator and (c) transported to the push-to-pull (PTP) device. The PTP device, micromanipulator, and attached fiber are shown. In the upper right corner the Pt gas injection system is visible. (d) SEM micrograph of the PTP device, with (e) a higher magnification image of the reusable gap across which different fibers can be mounted. Reprinted from [3.620], with permission from Elsevier

3.5.3 Summary

In situ mechanical testing has been successfully performed using several innovative techniques. It is a key tool in the investigation of both strength and ductility in structural materials, and has proven to be advantageous for directly observing and measuring plastic deformation phenomena. Elegant results on nanostructures have been achieved using nanomanipulation with a tip and microfabricated loading devices coupled with force-sensing holders. The challenge in mechanical testing is to make the results even more quantitative (especially as related to local deformation phenomena) and to minimize sample preparation artifacts. Future experiments will involve a greater variety of sample environments. High-temperature straining has long been widely available but it has been more difficult to adapt quantitative techniques such as indentation to work at low or high temperatures. Beyond high-temperature environments, it is critical to study environmentally assisted deformation mechanisms such as stress-corrosion cracking. There is no doubt that these types of experiments will bring our understanding of the mechanical properties of small volumes of materials to a new level of precision.

3.6 Liquid-Phase Processes

A rapidly developing area of in situ TEM is the study of processes that take place in liquids, particularly water. We have seen in Sect. 3.3 that for gas-phase processes—crystal growth, surface reactions, and catalysis—real-time observations greatly improved our understanding of the mechanisms at play. Is it possible to carry out similarly quantitative studies for crystal growth or other processes in liquids? The answer, perhaps surprisingly, is yes. Water can be controlled in the TEM vacuum environment; electrodes can be incorporated, flow and heating are possible, and much of this can be achieved with good spatial and temporal resolution. In this section we present some of the applications and challenges of liquid cell microscopy. For more details of this active field, including static imaging in liquids (with numerous biological applications) as well as in situ experiments, the reader is referred to reviews and a recent book [3.5, 3.627, 3.628].

The concepts of open- and closed-cell ETEM in Sect. 3.3 are directly applicable to water and other high-vapor-pressure liquids. Such liquids must be encapsulated or differentially pumped to keep the microscope pressure low enough for operation. In the closed-cell approach, the liquid is encapsulated between electron-transparent windows, usually made of silicon nitride. These closed liquid cells differ from the high-pressure closed gas cells used in ETEM in that the window spacing must be relatively small since the liquid is of course much denser than even a high-pressure gas. Reducing the liquid thickness improves the image resolution [3.627], but there are practical limits since a liquid layer that is too thin may not be able to provide an accurate representation of a phenomenon that normally occurs in larger volumes of liquid. \({\mathrm{200}}\,{\mathrm{nm}}\) or less is a typical window separation, set by spacer layers between the window chips. However, when loaded in the microscope, the windows are deflected outwards by the pressure difference. Suitable design is therefore needed to control the liquid thickness [3.629]. In the same way that gases are flowed in and out of a closed cell for ETEM, liquids may be pumped into and through a closed liquid cell. Furthermore, heaters and electrodes may be patterned onto the interior surfaces of the windows to enable imaging of electrochemical and temperature-dependent processes. Water may completely fill the closed cell (Sects. 3.6.23.6.4) or there may be a saturated environment with droplets or a thin liquid layer on the interior of the windows (Sect. 3.6.1).

To avoid the background caused by conventional windows, closed cells have been developed that use graphene as the window material [3.630]. Liquid droplets are placed on a supported graphene sample, then a second sample is used to encapsulate the droplets. Graphene liquid cells at present do not permit liquid flow or electrical biasing, but provide improved resolution for objects suspended in liquids.

Open-cell experiments involving water are possible in ETEM, although environmental SEM is a more common choice for such studies. Open-cell ETEM with differential pumping gives a high-resolution view of samples that require a hydrated environment, but electron–gas interactions should be considered [3.631].

Note that not all liquids require closed-cell or true open-cell differentially pumped ETEM techniques. Liquids with low vapor pressure, such as the ionic liquids used in the battery experiments described in Sect. 3.2.2, are directly compatible with conventional microscopes.

3.6.1 Imaging in a Saturated Environment or Under a Thin Liquid Film

Both closed cells and differentially pumped ETEMs can address problems involving thin water layers and environments saturated with water vapor. Early closed-cell experiments (reviewed in [3.501]) employed widely spaced windows, typically made of amorphous carbon supported on metal grids [3.632]. A thin water layer or air saturated with water filled the cell so that it could replicate dynamic processes requiring moisture, such as the hydration of Portland cement, or achieve static imaging of biological and other materials without drying or cryofixation. Static imaging was used to determine the oxidation state of Cr in particles produced by a Cr-reducing bacterium [3.633], and the arrangement and shape of small particles suspended in solvent without drying artifacts [3.634, 3.635]. An in situ use of this technique involves imaging the movement of the myosin head in muscle filaments [3.636, 3.637], by injecting ATP onto biological samples under a thin water layer.

An open cell enables imaging in an environment of varying relative humidity using differential pumping and a cooling stage. This has allowed the deliquescence of salt and aerosol particles, important in the Earth’s atmosphere, to be correlated with their structure and composition [3.638]. Water vapor can be mixed with other species to evaluate humidity effects on catalytic nanoparticle reactions [3.639]. The open cell has also been used to image synthesis of Au nanorods [3.640] and the catalytic reaction sequence used to form nylon [3.266]. The first reaction in this sequence, shown in Fig. 3.53a-d, was observed by injecting the precursor, in a methanol solvent, over a metal/\(\mathrm{TiO_{2}}\) catalyst specimen, while simultaneously heating the catalyst and flowing hydrogen; the second reaction was observed at a higher temperature by introducing adipic acid. Since many commercial polymerization reactions take place from solution, these types of experiments can have impact on the development of industrial processes, in the same way that gas-phase experiments support industrial development of gas reaction catalysts.

Fig. 3.53a-d

The catalytic formation of nylon: One stage in the formation of nylon is the hydrogenation of adiponitrile (ADN) to form hexamethylene diamine (HMD). HMD in turn is reacted with adipic acid and polymerized to produce nylon 6,6. The ADN to HMD reaction is carried out with ADN in a methanol solvent under gaseous hydrogen over a metallic catalyst. (a) High-resolution image of the catalyst, consisting of nanoclusters of Co-Ru (arrowed, c) over rutile titania support (with larger grains, u), at RT. (b) After immersion in adiponitrile in solvent and \(\mathrm{H_{2}}\) gas, at RT. (c) The formation of layers of the product HMD at \({\mathrm{31}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\). The width is indicated by double arrows. (d) Thicker layers form at \({\mathrm{81}}\,{\mathrm{{}^{\circ}\mathrm{C}}}\) (arrowed) after \({\mathrm{3}}\,{\mathrm{min}}\). (S with dots indicated the original catalyst profile.) From [3.266], reproduced with permission

3.6.2 Electrochemical Processes

Electrochemical processes control the performance of batteries during cycling, the deposition of material for coatings, and the corrosion of metals in the environment. Observing microstructural changes as they take place during electrochemical processes and correlating these changes with parameters of current and voltage provides useful insight into many of these important phenomena [3.628, 3.643]. Such observations are made in situ by imaging in a closed cell with electrical connections to the liquid and a stable liquid environment.

The experiments initially addressed the deposition of copper, relevant to the formation of interconnects in integrated circuits [3.641, 3.644]. Three microelectrodes within the cell, connected externally to a potentiostat, allow electrochemical control. The electrolyte fills the entire volume between the windows, and the current passes through the liquid between the working and counter electrodes. The working electrode, at which deposition takes place, is patterned over one of the windows. Figure 3.54a,b shows the nucleation and growth of copper clusters on a gold electrode, with simultaneous measurement of the voltage applied and the resulting current flowing in the cell. When interpreting results from these experiments, care must be taken to verify that the behavior of the small area of electrode under observation is typical of the entire electrode and that the small volume cell behaves like a standard electrochemical cell. Quantitative data on individual clusters can be compared with electrochemical models. Figure 3.54a,b shows nucleation density and growth rate measurements that suggest a need to refine textbook models [3.641].

Fig. 3.54a,b

Electrochemical deposition processes: (a) Electrochemical deposition of copper onto a gold electrode from acidified \(\mathrm{CuSO_{4}}\) solution. A reverse potential was applied for \({\mathrm{2}}\,{\mathrm{s}}\) to clean the electrode, then deposition was carried out at a constant potential of \({\mathrm{-0.07}}\,{\mathrm{V}}\), measured with respect to a Cu reference electrode. The images provide a direct view of the cluster evolution on one part of the electrode. The graph of current versus time has a characteristic shape that can be fitted with a model of cluster nucleation and growth. Reprinted with permission from [3.641]. Copyright 2006 American Chemical Society. (b) Electrochemical deposition and stripping of Li. The current is cycled at \(|J|={\mathrm{10}}\,{\mathrm{mA/cm^{2}}}\) while imaging every \({\mathrm{15}}\,{\mathrm{s}}\) using BF STEM. Images are shown before deposition, near the beginning and end of deposition cycle 1, and near the beginning and end of stripping cycle 1 where disconnected dead Li is evident after stripping finished. The graph shows chronopotentiometry during cycling with the voltage measured versus Ti and inferred versus Li (red graph); arrows show times at which images were taken. Reprinted with permission from [3.642]. Copyright 2015 American Chemical Society

Subsequent experiments imaged processes relevant to batteries and fuel cells. One key area is the structural change on intercalation of lithium into electrode materials [3.645, 3.646]. In this case, results can be compared with intercalation into nanowire electrodes connected only at their ends (Sect. 3.2.2) to explore the effects of different ionic diffusion pathways. Another key area is the formation of the solid electrolyte interphase (Sect. 3.2.2). Nucleation and growth of the SEI has been studied at metal [3.647, 3.648] and carbon [3.649] electrode–electrolyte interfaces, correlating formation with applied voltage. Failure mechanisms can also be addressed, particularly the formation of dendrites during cycling [3.642, 3.650, 3.651]. Finally, observing the behavior of battery materials in the liquid cell helps understand their stability [3.652]. The mechanistic insights from all these experiments will contribute to the development of materials with increased energy density and improved lifetime.

Another distinct application of electrochemical liquid cell microscopy is to corrosion processes. These experiments involve exposing a metal to water with or without an applied potential. Pit formation is visible in thin films and can be related to electrochemical parameters and solution chemistry [3.655]. Bulk materials can be examined by cutting out a lamella and welding it to the liquid cell electrode [3.656]. In the future, these electrochemical liquid cell microscopy applications will be enhanced through liquid cell chip design to control liquid flow and temperature, and by wider use of analytical techniques for chemical analysis.

3.6.3 Liquid-Phase Crystal Growth and Coalescence

Liquid cell microscopy has become a valuable tool for studying nanocrystal growth, assembly, and sintering [3.657]. Crystal growth in a liquid medium aims to create nanocrystals with well-controlled structure and composition under the stimulus of a chemical change, heating, or irradiation. In liquid cell TEM, electron-beam-induced reduction of metal ions in solution is the primary stimulus, although heating and illumination are also used [3.658]. Even though the driving force in situ is thus somewhat different from what is used commercially, the experiments visualize important features of crystal growth.

The liquid cell is filled with a solution containing a metal ion of interest. A wide range of materials have been studied, as described in a recent review [3.657], and some examples are shown in Fig. 3.55a-c. Both conventional liquid cells with silicon nitride windows [3.659] and graphene liquid cells can be used [3.630]. The imaging beam itself triggers the formation of particles. Those that nucleate at the interior surfaces of the windows can be imaged without too much blurring. Those in the liquid close to the window move surprisingly slowly (Sect. 3.6.4), and can also be followed as they grow. The experiments probe the factors controlling nanocrystal morphology, including dose, dose rate, the presence of surfactants and neighboring particles, and the chemical composition of the solution [3.660, 3.661]. The resolution can be high enough to quantify the development of specific facets [3.662]. Measurements of individual particles can distinguish between growth modes such as monomer attachment and coalescence [3.659] and characterize the evolution of the ensemble of all particles in the field of view [3.663]. Nanocrystal etching can also be followed [3.664].

Fig. 3.55a-c

Nanoparticle and bubble dynamics in liquids: (a) Coalescence of two Pt nanoparticles in aqueous solution imaged in a graphene liquid cell at times \(\mathrm{159.18}\), \(\mathrm{159.96}\), \(\mathrm{160.74}\) and \({\mathrm{161.78}}\,{\mathrm{s}}\), respectively. Dimensions of the neck and the particles are indicated. From [3.630]. Reprinted with permission from AAAS. (b) The formation of hollow bismuth oxide particles in aqueous solution, by the Kirkendall effect in which Bi diffuses outwards through the bismuth oxide shell. Reprinted with permission from [3.653]. Copyright 2013 American Chemical Society. (c) The nucleation and coalescence of nanoscale bubbles of hydrogen that form in water due to radiolysis. The images span a \({\mathrm{3}}\,{\mathrm{s}}\) interval. Reprinted with permission from [3.654]. Copyright 2014 American Chemical Society

It is also possible to observe the assembly of particles into aggregates and superlattices [3.665, 3.666] and evaluate nucleation and growth in templates [3.667, 3.668]. During coalescence events, atomistic details of the attachment and sintering processes include shape changes and rotations [3.669]. Such information is relevant to understanding crystal growth by oriented attachment, which controls some key processes in mineral formation.

3.6.4 Fluid Physics

The behavior of fluids at nanoscale dimensions is important in biological processes, catalysis, lubrication, and coatings. Liquid cell microscopy provides opportunities for studying the dynamics and stability of thin films, droplets, and bubbles (Fig. 3.55a-cc), as well as phase-transformation phenomena—condensation, evaporation, boiling, and freezing—with resolution unavailable from other microscopies.

Time-resolved observations help understand the stability of small bubbles formed by heating [3.670] or radiolysis [3.654] and the dewetting of thin water layers by the formation of voids [3.671]. Similarly, time-resolved imaging of condensation and beam-induced motion of droplets provides details of the physics controlling these processes [3.672, 3.673]. The use of liquid cell enclosures with different geometries—cylindrical closed cells within carbon nanotubes or scrolls [3.673, 3.674]—provides complementary information by enabling side views of droplets. An intriguing observation from liquid cell microscopy is the anomalously slow movement of suspended particles in thin liquid layers [3.675, 3.676] and nanodroplets [3.677]. Although this is convenient for imaging as motion-induced blur is reduced, the mechanism of this slow diffusion is still under study. Finally, freezing of water can be imaged by incorporating a Peltier cooler into the liquid cell holder [3.678]. The phase of the growing ice crystal and facets at the ice–water interface can be measured and phenomena such as solute rejection visualized. Overall, liquid cell microscopy provides a promising approach to visualizing many fluid phenomena at small dimensions.

3.6.5 Radiation Chemistry

The electron beam is an intense source of energy that interacts strongly with water. Understanding beam effects is critical, but fortunately we can make use of the knowledge obtained through medical and reactor physics studies. Analysis of energy deposition from the beam into the TEM sample shows that beam-induced heating is negligible [3.659, 3.679] due to the sample's good thermal conductivity. Instead, the deposited energy causes chemical changes via the formation of radical and molecular species by radiolysis. These include hydrated (solvated) electrons \(\mathrm{e}_{\mathrm{h}}^{-}\), hydrogen radicals \(\mathrm{H^{\bullet}}\), hydroxyl radicals \(\mathrm{OH^{\bullet}}\), and \(\mathrm{H_{2}}\). As these species diffuse and react, other species are formed such as \(\mathrm{H_{2}O_{2}}\) and \(\mathrm{H_{3}O^{+}}\), and eventually \(\mathrm{O_{2}}\) and recombined \(\mathrm{H_{2}O}\).

Each radiolytic product affects the outcome of a liquid cell experiment. The most reactive species is the hydrated electron, a free electron surrounded by a cage of water molecules. It reduces metal ions to form metal nanoparticles, whose growth is then directly visible during irradiation (Sect. 3.6.3). \(\mathrm{H_{2}}\) molecules are not chemically reactive but form bubbles if produced at sufficient concentration (Sect. 3.6.4). \(\mathrm{H^{+}}\) changes the pH to lower values on irradiation, which may affect electrochemical processes (Sect. 3.6.2) and alter the stability of nanoparticles. Modeling the expected concentration of each radiolysis species is the first step towards quantifying beam effects. The concentrations are calculated using coupled equations that account for the known formation rate of each species, its destruction rate, and its diffusion out of the irradiated area [3.679]. These calculations are based on parameters for pure water, and there is generally not sufficient knowledge in the literature to include the exact composition of the solution under study. However, we anticipate that radiolysis quantification will become more widespread in liquid cell microscopy research, as well as encouraging us to think harder about beam effects in all in situ experiments.

3.6.6 Outlook

The results described in this section show that dynamic processes in liquids and at liquid–solid interfaces may indeed be observed in real time and with reasonable spatial resolution in situ, providing information that is difficult to obtain using techniques such as scanning probe microscopy or SEM. Future experiments will expand the range of processes and information accessible. In particular, EELS and XEDS are becoming more established, and control of flow, liquid geometry, and temperature improve with every new liquid cell design. We believe that rapid development will continue in this key area of in situ TEM.

3.7 Electron-Beam-Induced Processes

Microscopists often consider electron beam effects to be inconveniences that make it difficult to acquire useful data. However, electron-beam interactions have the useful property that the beam provides a stimulus while also enabling visualization of the effects of that stimulus. Historically, irradiation damage in situ by ions, electrons, or both simultaneously was used to model neutron damage relevant to nuclear energy generation and nuclear waste storage, as well as dating of radioisotope-bearing minerals. High-voltage electron microscopy played an important role in these explorations. However, a recent focus of in situ experiments is the use of electron beams at more conventional energies. Topics that are seeing the greatest activity are beam-assisted deposition and removal of material, structural modification such as amorphization, and even manipulation of single atoms in a sample. Here we describe use of the electron beam as an in situ tool to modify the sample. We also describe complex instruments that allow electron-beam effects to be studied in combination with the effects of ion implantation or irradiation with light. Several reviews [3.680, 3.681] cover aspects of this wide-ranging field.

3.7.1 Interaction with the Column Environment: Beam-Enhanced Deposition

We first discuss a beam-induced phenomenon that is more or less independent of the nature of the specimen. This is the interaction of the electron beam with the atmosphere in the microscope column. In an ETEM experiment, a gas environment is chosen that is a feedstock for a catalyst or the source for a growth process or surface reaction such as oxidation. It has been recognized for some time that the electron beam has a strong effect on such an environment. For example, in the hydrogen embrittlement studies described in Sect. 3.5.1, the electron beam significantly increases the fugacity (an effective partial pressure) of the hydrogen [3.682]. In oxidation and reduction of Pt nanoparticles, the beam activates the gases present [3.683]. Low pressures of reactive species such as water vapor or oxygen—even as residual species during a higher pressure ETEM experiment—can affect growth phenomena [3.274, 3.310]. The cross section for interaction of a high-energy primary beam with the gas is usually low, but when the primary beam hits the sample it generates secondaries that have a higher cross section. The reactions that occur depend on the energy of these secondaries and the nature of the gas environment so it is important to carry out careful comparative experiments.

Beam effects in a reactive atmosphere are used to advantage for deposition. Focused electron-beam-induced deposition () is the process by which an electron beam causes chemical changes in species adsorbed on the sample surface that lead to formation of a solid deposit. The adsorbed species form when a precursor gas such as \(\mathrm{W(CO)_{6}}\) is flowed. Deposition of W from this precursor [3.685] forms structures such as that shown in Fig. 3.56. Similarly, Fe can be deposited from \(\mathrm{Fe(CO)_{5}}\) [3.686, 3.687]. Three-dimensional structures may even be formed using a programmed STEM [3.681, 3.684]. The highest resolution for such nanofabrication is in the low tens of nanometers, if a thin substrate is used to reduce the volume from which secondaries are generated [3.688]. This feature size is competitive with other lithography techniques, and although the throughout and purity of the deposit must be optimized, interesting structures can be formed [3.681].

Fig. 3.56

Fabrication of nanostructures by electron-beam interaction with gas: Nanoscale fabrication was carried out on a carbon grid, using a JEM 2500SE STEM operated at \({\mathrm{200}}\,{\mathrm{kV}}\) with a beam size of \({\mathrm{0.8}}\,{\mathrm{nm}}\) and a beam current of \({\mathrm{0.5}}\,{\mathrm{nA}}\). W(CO)\({}_{6}\) gas was used at a flux of approximately \({\mathrm{2\times 10^{-4}}}\,{\mathrm{Pa{\,}L{\,}s^{-1}}}\). A line was fabricated across a hole in the grid, followed by fabrication of a circle. Reprinted with permission from [3.684], John Wiley and Sons

3.7.2 Beam-Induced Materials Reactions

As the electron beam deposits energy it drives multiple phenomena within the sample such as charging, atomic displacement, sputtering, or heating [3.689]. The outcome depends on the beam parameters, material and sample geometry. Displacement occurs if the beam energy exceeds a displacement threshold energy; charging depends on electrical conductivity; and temperature rise depends on thermal conductivity. These effects are common to all electron microscopy observations, but here we discuss mainly intentional use of the beam to trigger reactions. For quantitative results, we need to separate heating effects from those caused by knock-on damage or electronic excitations. This may require detailed hot-stage measurements or calibration via transformations that occur at known temperature [3.690, 3.691].

Amorphization is the most commonly observed beam-induced transformation, for example in SiC [3.692], Si [3.693], and GaAs [3.694]. In InGaN, beam-induced amorphization can appear like compositional changes [3.695], requiring careful image interpretation. Interestingly, in \(\mathrm{Zr_{2}Ni}\) and \(\mathrm{Zr_{3}Al}\), amorphization kinetics are changed by the presence of hydrogen [3.697], which alters the stability of the defective structure. Thus, the microscope atmosphere should be considered in these transformations as well. The reverse process of beam-induced recrystallization also occurs, for example in semiconductors [3.698]. It may be seen at voltages below those required to create point defects, and may be caused by the formation of dangling bonds at the crystalline–amorphous interface. Recrystallization of strontium titanate can take place from an amorphized sample [3.699] or into an amorphous region from an underlying single-crystal seed layer. Patterning crystallization with a focused beam can produce nanostructured material [3.696, 3.700], as shown in Fig. 3.57.

Fig. 3.57

Beam-induced defect formation: Phase transformation induced by electron-beam exposure in STEM in the indicated area. Layered \(\mathrm{SrNbO_{3.4}}\) transforms to perovskite \(\mathrm{SrNbO_{3}}\) by ejection of O atoms out of the vertex-sharing \(\mathrm{NbO_{6}}\) octahedral slabs. Reprinted with permission from [3.696]. Copyright 2015 American Chemical Society

The electron beam also accelerates oxidation reactions (Sect. 3.3.2) in cases where oxygen is an impurity. For example, structural and chemical evolution occur in CdS nanoribbons due to oxidation under irradiation [3.701].

Other beam-induced reactions include phase separation, formation of nonequilibrium phases, and even decomposition. For example, in borosilicate glasses, which have applications in nuclear waste storage, B-rich phases separate under the beam [3.702]. Beam-induced decomposition can lead to material with nanoscale composition fluctuations. This has been examined particularly in \(\mathrm{SiO_{2}}\), where sputtering and desorption lead to Si-rich regions [3.685, 3.703, 3.704, 3.705]. Beam-induced decomposition occurs in many materials, including \(\mathrm{Al_{2}O_{3}}\), MgO and AlF, and may proceed from either or both surfaces. Several processes are active, including interactions with the atmosphere, changes in surface diffusion [3.706], surface roughening [3.707], and massive atom displacement driven by an electric field induced by the excitations and ionizations of atomic electrons [3.708].

3.7.3 Defect Dynamics

The electron beam can break atomic bonds, creating vacancy-interstitial pairs, or directly eject atoms from the sample, causing mass loss. Imaging these phenomena in situ leads to insights into plasticity, materials stability, and opportunities for creating nanoscale structures with useful electronic properties.

Nanomaterials and Nanofabrication

In two-dimensional materials or in nanostructures such as carbon nanotubes, the formation of point defects, their dynamics after formation, and their reactions to form larger scale defects can be directly visible. The experiments usually involve beam irradiation at elevated temperature to allow defects to form, move, and interact with other point defects or extended defects such as grain boundaries. The beam acts as both the source of energy for creating the defects and the means to image them; it may also be used to control their motion or reactions, yielding data that can be matched with models to understand mechanical properties [3.709]. Materials studied include graphene [3.709, 3.710, 3.711], carbon nanotubes [3.712] (Sect. 3.3.4), two-dimensional silica glass [3.713], boron nitride [3.714], and \(\mathrm{MoS_{2}}\) [3.715]. An example is shown in Fig. 3.58a-c. Defect-induced migration of metal atoms is visible [3.716]. The formation of extended features such as pits caused by vacancy clustering can be followed [3.717]. The relative ratios of different types of edges in graphene holes has been measured as a function of temperature [3.109].

Fig. 3.58a-c

Beam-induced defect formation: A line of vacancies formed in an exfoliated \(\mathrm{MoS_{2}}\) sample after exposure in a Cs-corrected \({\mathrm{80}}\,{\mathrm{keV}}\) TEM at \({\mathrm{10^{6}}}\,{\mathrm{e/(nm^{2}s)}}\) (a). The image is filtered to reduce noise. An atomic model (b) and simulated image (c) are also shown, where brown and orange are S atoms and black is Mo. Calculations show that there is an energy gain from vacancies aligning in this way and the line direction is related to strain in the sample. Reprinted with permission from [3.715]. Copyright 2013 by the American Physical Society

An intense, focused probe in STEM can create enough displacement or diffusion to drill a hole through many types of sample [3.294, 3.295, 3.718, 3.719]. Nanostructures and nanowires between holes can be formed, as was shown in Sects. 3.3.1 and 3.3.4 for Au and C, respectively. Similarly, when the sample is a two-dimensional material, scanning the beam can create structures that are well defined in all three dimensions. Graphene [3.720] and dichalcogenide materials [3.721] have been patterned in this way, and junctions have been formed between different layers in a multiwall CNT [3.722]. This type of beam-induced processing is an excellent approach for building new structures that also enables testing the electronic properties of these materials at nanometer scales. As beam-induced manipulation of nanomaterials develops, it will become possible to build nanodevices based on single dopant atoms and perhaps test physics, say for quantum computers, at these dimensions.

Point Defects in Bulk Materials

In bulk materials individual point defects are difficult to see, but when they cluster into extended defects, measuring the dynamics yields indirect information about the point defects such as their diffusion parameters. The fundamentals of point defect motion have been studied in many materials in this way using high-voltage microscopy [3.724]. For example, by measuring the growth and shrinking of interstitial loops during intermittent irradiation, it is possible to obtain activation energies for vacancy migration and self-diffusion [3.725]. The density of point defects can be measured by examining the formation of jogs in dislocation loops [3.726] while their diffusion parameters can be measured by examining extended defect dynamics far from the irradiated area [3.727]. Loop growth and shrinkage can even probe defects produced by other types of irradiation such as neutrons [3.728, 3.729].

The growth rates of extended defects provide detailed information on interactions between point defects and extended defects. In Cu, irradiation causes growth of stacking fault tetrahedra, and an analysis of size fluctuations showed that growth is by ledge motion after capture of defects [3.730].

Clustering or ordering of point defects is visible in situ. In Si, irradiation forms interstitial clusters whose structure and formation kinetics can be measured [3.723, 3.731], as shown in Fig. 3.59. Strain fields influence point defect clustering [3.731, 3.732], so perhaps clustering could be used to measure strain locally. Imaging defect clusters usually involves weak-beam techniques [3.729]. But EELS can be combined with imaging to characterize ordering, due for example to beam-induced formation and motion of oxygen vacancies [3.733, 3.734].

Fig. 3.59

Electron irradiation damage in Si: HREM image of point defect aggregates created in Si during electron irradiation at room temperature. Irradiation was carried out at \({\mathrm{400}}\,{\mathrm{kV}}\) and \({\mathrm{10^{6}}}\,{\mathrm{e/(nm^{2}{\,}s)}}\) for \({\mathrm{35}}\,{\mathrm{min}}\). The {113} and {111} defects are marked with single and double arrows, respectively. From [3.723]

High-voltage electron irradiation can be used to move atoms from the surface into the bulk of a substrate via elastic collisions of the electrons with the heavy atoms. Systems studied include Au implanted into Si [3.735], Hf into SiC [3.736], and Au into Al [3.737], which forms \(\mathrm{Al_{2}Au}\) phases that move downstream. These experiments are useful in understanding the phenomena taking place during ion-beam processing.

Radiation-Enhanced Defect Motion

In many materials the electron beam influences dislocation motion. Radiation-enhanced dislocation glide during heating or straining has been studied quantitatively by recording the motion of individual dislocations (Fig. 3.60). Such measurements suggest a mechanism based on an enhancement in the creation rate of kink pairs, due to energy released by nonradiative recombination of electron–hole pairs at electronic levels associated with dislocations [3.511, 3.545, 3.569, 3.738, 3.739, 3.740]. The radiation-enhanced motion of individual kinks can actually be observed directly in plan view [3.741]. A radiation-enhanced climb process can also occur, due to absorption of interstitials by dislocations [3.742]. These studies are important in interpreting in situ deformation experiments but they also have relevance to the phenomenon of photoplasticity in optoelectronic device degradation. Quantitative in situ nanomechanical testing has also been used to probe electron irradiation effects. In silica nanoparticles [3.743], consecutive compression tests performed with the electron beam off and on (Fig. 3.61a-c) showed that the material is softer under irradiation, dramatically illustrating the effect of the beam on deformation.

Fig. 3.60

Radiation-induced dislocation glide: The velocity of dislocations in ZnS of different types per unit length \(L\) is shown as a function of the electron-beam intensity. Velocities were measured during in situ straining at \(40\pm{\mathrm{10}}\,{\mathrm{MPa}}\) and \({\mathrm{390}}\,{\mathrm{K}}\). The motion remains slow suggesting that the Peierls mechanism still controls motion, and velocity is proportional to dislocation length showing that kinks form individually and do not collide. The linear behavior at low intensity suggests a change (reduction) in apparent activation energy due to nonradiative recombination of carriers at dislocations, assisting kink formation. At high intensities the recombination rate saturates. After [3.738]

Fig. 3.61a-c

Irradiation effects during mechanical testing: In situ nanomechanical testing is used to probe electron irradiation effects in silica nanoparticles. Two consecutive compression tests are performed with the electron beam off and on, respectively. The silica particle is much softer under electron-beam illumination (c). From [3.743]

3.7.4 Ion-Implantation Phenomena

High-Energy Ion Accelerators

The materials used for nuclear electric power generation or for long-term nuclear waste storage are continuously irradiated with high-energy neutrons from nuclear reactions. Irradiation induces displacement damage and forms helium which causes bubbles, swelling, embrittlement, and radiation-induced segregation. Irradiating a sample with ions and electrons simultaneously in a specialized microscope can simulate the defect types and concentrations created by neutron irradiation. Irradiation-induced defects can be imaged and the effects on phase equilibria, electrical and mechanical properties can be measured. The high dose rates in TEM enable accelerated experiments. Sputtering, ion implantation, and materials processing by impurity doping can also be addressed using such microscopes.

Irradiating a sample with ions while simultaneously imaging or irradiating it with electrons requires a dedicated instrument with tandem accelerators. Several have been built and used for a variety of materials studies [3.745, 3.746]. Unsurprisingly, given the reactor materials in use, many irradiation studies involve steels. Examples include determining the stability of different phases after Xe irradiation [3.747] and measuring bubble motion after He implantation [3.748]. In other materials, defect formation and diffusion processes are examined. In Cu irradiated with Kr, the kinetics of displacement collision cascade formation and the recombination of point defects into clusters were measured [3.749], while in Si, irradiation with Si shrinks the bubbles caused by H implantation [3.750]. Recent studies address materials with complex microstructure—nanotwinned Cu with voids [3.751] and nanoporous Au [3.752]. Both show enhanced radiation tolerance as the voids can absorb radiation-induced defects. Similarly, nanocrystalline Fe shows radiation resistance due to high defect sink density [3.753]. Flexible systems capable of concurrent in situ ion irradiation in the TEM continue to be developed [3.754], coupled to in situ mechanical testing, heating, and ultrafast electron sources.

The fundamentals of bubble motion are studied in a model system consisting of insoluble noble gases such as Ar, Ne, or Xe implanted into Al or a similar matrix. The small precipitates formed are shown in Fig. 3.62a-ja. Implantation can be carried out ex situ [3.744, 3.748, 3.755] or in situ [3.756], so that the coalescence of defects to form the precipitates can be studied as a function of dose. Migration, shape changes, faulting, melting, crystallization, and coalescence occur in these precipitates under \({\mathrm{1}}\,{\mathrm{MeV}}\) electron irradiation [3.744, 3.755] or on heating [3.748]. The shape, nature, and motion of the enclosed phase provide information on interface energies, structure, and diffusion pathways [3.40, 3.757] (Fig. 3.62a-j).

Fig. 3.62a-j

Dynamics of Xe precipitates: The motion and coalescence of two isolated crystalline Xe precipitates in Al, where the Xe was implanted ex situ, during continuous \({\mathrm{1}}\,{\mathrm{MeV}}\) electron irradiation at room temperature. The irradiation causes a damage rate in Al of \(\approx{\mathrm{4\times 10^{-2}}}\) displacements per atom () per second. Each frame shows an area \({\mathrm{16}}\,{\mathrm{nm}}\times{\mathrm{11}}\,{\mathrm{nm}}\). Measured from the first image, the times at which video frames were recorded are (a\({\mathrm{0}}\,{\mathrm{s}}\), (b\({\mathrm{101}}\,{\mathrm{s}}\), (c\({\mathrm{418}}\,{\mathrm{s}}\), (d\({\mathrm{549}}\,{\mathrm{s}}\), (e\({\mathrm{550}}\,{\mathrm{s}}\), (f\({\mathrm{551}}\,{\mathrm{s}}\), (g\({\mathrm{561}}\,{\mathrm{s}}\), (h\({\mathrm{584}}\,{\mathrm{s}}\), and (i\({\mathrm{727}}\,{\mathrm{s}}\). Traces of crystallographic planes are indicated in frame (a). Analysis of such data shows that surface diffusion of Al is responsible for motion and shape changes, while the Xe deforms by shear in response to the reshaping of its cavity. Since the total volume, not the surface area, was conserved during coalescence, cavity pressure depends on the gas structure and not just the interface structure. (j) Mean square displacement of an Xe precipitate in Al containing approximately 38 Xe atoms (occupying a volume equivalent to that of approximately 128 Al atoms), as a function of damage in the Al matrix under \({\mathrm{1}}\,{\mathrm{MeV}}\) electron irradiation. The precipitate moves because of Al jumps on its surface, and this data yields values for the diffusivity and an average jump frequency of about \({\mathrm{5600}}\,{\mathrm{jump/dpa}}\). Reprinted with permission from [3.744]. Copyright 1999 by the American Physical Society

We conclude with a different ion irradiation experiment: the integration of a lower voltage focused ion beam into a TEM [3.758, 3.759]. This tandem microscope allows the damage produced by FIB to be observed directly. This is useful for understanding artifacts and low-voltage damage effects. It also allows sample thinning to be carried out in situ, with the area imaged not exposed to the air. A simpler experiment has an FIB placed in a chamber attached to the TEM vacuum system, allowing FIB patterning of a sample before an in situ growth experiment (Sect. 3.3.4). Direct imaging of the effects of FIB is increasingly important since FIB is used for materials processing, surface patterning, growth, and fabrication of nanostructures.

3.8 Outlook

In this chapter we have attempted to summarize some of the frenetic activity in the field of in situ electron microscopy and to illustrate the huge variety of experiments possible. In spite of experimental complexity, many successful and informative experiments have mimicked real-world processes inside the microscope column.

It seems clear that there is no one in situ microscope. Some in situ experiments take place in expensive microscopes which are modified to achieve a controlled specimen environment or are designed for UHV. But other experiments are conducted in conventional instruments, using standard holders: even irradiating the sample with the electron beam produces in situ data. Broader fields for experimentation are enabled by purchasing commercial in situ holders (heating, cooling, straining, nanoindenters, electrical biasing holders, liquid cells) or developing customized holders, and by adding capabilities to measure a property of the sample, such as electrical transport, simultaneously with imaging its structure.

The rapid developments in instrumentation make this an exciting time for in situ microscopy. Sample holder design becomes more complex yet more precise and reproducible. The design of the sample itself increasingly incorporates innovative use of FIB preparation or integration of thin-film materials into micromachined substrates. Furthermore, as is clear throughout this chapter, the increased interest in nanostructures and low-dimensional materials fits perfectly with the ability of TEM to analyze small volumes with minimal sample preparation.

Apart from thin-foil effects and beam-induced artifacts, the main limitation of in situ microscopy has been the small space available in the polepiece. In the future, we anticipate that this problem will be mitigated by ongoing developments in aberration correction. Corrected microscope designs can increase the polepiece gap without loss of resolution. Extra space enables experiments requiring combined stimuli (heating and irradiation, straining in a gas environment, simultaneous \(E\) and \(B\) fields) or new stimuli (lasers, micromanipulators, microfluidics). Extra space also improves calibration of the sample environment. It is hard to overemphasize the importance of this for quantitative results. For example, one could include a fiber optic coupled to a radiation thermometer [3.760], a thickness monitor for growth experiments, a pressure gauge, or MEMS-based sensors. Another very significant benefit resulting from aberration correction is the ability to go to low voltages without extreme loss of resolution. The ability to choose voltage to optimize beam damage will prove especially useful for in situ experiments involving two-dimensional materials. For other materials, sample thickness, or more generally sample design, will remain a constraint to decreasing the voltage.

As well as better integration with aberration-corrected microscope designs, there are several other areas where we anticipate in situ experiments will expand. Even today, relatively few in situ experiments use analytical techniques such as energy-filtered imaging, EELS, or XEDS to examine time-resolved chemical changes. The need for rapid acquisition and the dose sensitivity of the sample limit the signal-to-noise ratio possible. Thus, the ongoing improvements in detectors are particularly promising for in situ analytical studies. Similarly, monochromators and chromatic aberration correction will improve in situ chemical analysis, allowing more efficient use of the dose to the sample. Correlative microscopy, where TEM is combined with data from light microscopy or x-ray techniques (moving the sample in its holder between instruments), will expand the range of length scales and signals that are used for understanding materials processes.

A distinctive feature of electron microscopy is its vast range of imaging modes, but many of these have not yet seen much use in situ. Convergent-beam electron diffraction is an example that is beginning to see use with high-speed detectors to map local strain in situ [3.761]. Techniques such as tomography, holography, or approaches for imaging amorphous materials will also become more frequently used in situ, especially where the phenomenon of interest is slow enough and the material beam tolerant enough to make image acquisition feasible. Custom electron-beam manipulation with the precision seen in lithography tools will help sample patterning and allow more dose-efficient imaging.

High-speed imaging opens new phenomena for study. However, it exacerbates the problems of dose tolerance, data collection, and data analysis that in situ microscopists already experience. Although computational improvements help with acquisition, searching and analysis of data, the vast data rates from modern detectors continue to make this a challenge. Intelligent interrogation of video data, from object detection and tracking to more complex measurements, is currently a limitation for in situ analysis, as custom software often has to be developed for each experiment. Improved modeling of beam effects and the physical processes taking place in the sample in situ are also key areas for development. Modeling is essential for extracting the most quantitative information from an experiment, and can in principle guide the microscopist’s choice of experimental conditions to perform observations that are the most critical tests of a theory.

A final comment arises from the complexity of controlled-environment in situ microscopes, which has increased since the first development of such instruments [3.762]. Some processes, say sample heating, must take place in the polepiece as images are recorded. But other aspects of the experiment can be carried out outside the polepiece while still in the vacuum system. Examples include ion sputtering or high-temperature annealing to clean the sample, some deposition tools such as evaporators, and temperature calibration. Moving some capabilities ex situ but within the same vacuum system improves the reliability of complex microscope systems and provides a more flexible approach to experiments, especially in a multiuser environment where time is a constraint.

The field of in situ transmission electron microscopy has advanced tremendously since its beginning in the 1950s. We anticipate even more exciting results over the next decades.

Notes

Acknowledgements

Frances Ross would like to acknowledge the inspirational scientists with whom she has had the pleasure of working: her PhD advisor, the late Mike Stobbs, who emphasized quantitative information from TEM; Murray Gibson, Robert Hull, Uli Dahmen, and Ruud Tromp; and Mike McDonald, Mark Reuter and Arthur Ellis, whose outstanding technical help made the in situ experiments possible; also the IBM T.J. Watson Research Center with its far-sighted view of basic research, her many collaborators, and her family for their patience and support. Andrew Minor would like to thank his scientific mentors who over the years have guided and supported him, including J.W. Morris, Jr., Eric Stach, and Uli Dahmen. In addition, he appreciates all of the students, postdocs, collaborators, and funding sources who have enabled him to work in the exciting field of in situ electron microscopy. Lastly, he is grateful for the love and support of his family.

References

  1. R. Ramachandramoorthy, R. Bernal, H.D. Espinosa: ACS Nano 9, 4675–4685 (2015)Google Scholar
  2. T. Xu, L. Sun: Small 11, 3247–3262 (2015)Google Scholar
  3. M.L. Taheri, E.A. Stach, I. Arslan, P.A. Crozier, B.C. Kabius, T. LaGrange, A.M. Minor, S. Takeda, M. Tanase, J.B. Wagner, R. Sharma: Ultramicroscopy 170, 86–95 (2016)Google Scholar
  4. G. Dehm, J.M. Howe, J. Zweck: In-Situ Electron Microscopy: Applications in Physics, Chemistry and Materials Science (Wiley, Hoboken 2012)Google Scholar
  5. F.M. Ross (Ed.): Liquid Cell Electron Microscopy (Cambridge Univ. Press, Cambridge 2016)Google Scholar
  6. T. Brintlinger, Y. Qi, K.H. Baloch, D. Goldhaber-Gordon, J. Cumings: Nano Lett. 8, 582 (2008)Google Scholar
  7. M. Mecklenburg, W.A. Hubbard, E.R. White, R. Dhall, S.B. Cronin, S. Aloni, B.C. Regan: Science 347, 629–632 (2015)Google Scholar
  8. T.J. Konno, R. Sinclair: Philos. Mag. B 71, 179–199 (1995)Google Scholar
  9. M.A. Parker, R. Sinclair: Nature 322, 531–533 (1986)Google Scholar
  10. R. Sinclair, M.A. Parker, K.B. Kim: Ultramicroscopy 23, 383–395 (1987)Google Scholar
  11. T.J. Konno, R. Sinclair: Philos. Mag. B 66, 749–765 (1992)Google Scholar
  12. T.J. Konno, R. Sinclair: Philos. Mag. B 71, 163–178 (1995)Google Scholar
  13. T.J. Konno, R. Sinclair: Acta Metall. Mater. 43, 471–484 (1995)Google Scholar
  14. R. Sinclair, T. Itoh, R. Chin: Microsc. Microanal. 8, 288–304 (2002)Google Scholar
  15. R. Hull, J. Bean: Mater. Res. Soc. Bull. 19(6), 32–37 (1994)Google Scholar
  16. E.A. Stach, R. Hull, J.C. Bean, K.S. Jones, A. Nejim: Microsc. Microanal. 4, 294–307 (1998)Google Scholar
  17. M.T. Janish, W.M. Mook, C.B. Carter: Scr. Mater. 96, 21–24 (2014)Google Scholar
  18. H.-J. Lee, A.G. Ramirez: Appl. Phys. Lett. 85, 1146 (2004)Google Scholar
  19. H.-J. Lee, H. Ni, D.T. Wu, A.G. Ramirez: Appl. Phys. Lett. 87, 124102 (2005)Google Scholar
  20. C.J. Alvarez, Y. Liu, R.L. Leonard, J.A. Johnson, A.K. Petford-Long: J. Am. Ceram. Soc. 96, 3617–3621 (2013)Google Scholar
  21. J.A. Aguiar, S. Wozny, T.G. Holesinger, T. Aoki, M.K. Patel, M. Yang, J.J. Berry, M. Al-Jassim, W. Zhou, K. Zhu: Energy Environ. Sci. 9, 2372–2382 (2016)Google Scholar
  22. L. He, J.P. Chu, C.L. Li, C.M. Lee, Y.C. Chen, P.K. Liaw, P.M. Voyles: Thin Solid Films 561(87), 92 (2014)Google Scholar
  23. B.J. Kooi, J.T.M. De Hosson: J. Appl. Phys. 95, 4714–4721 (2004)Google Scholar
  24. M. Kaiser, L. van Pieterson, M.A. Verheijen: J. Appl. Phys. 96, 3193–3198 (2004)Google Scholar
  25. B.J. Kooi, W.M.G. Groot, J.T.M. De Hosson: J. Appl. Phys. 95, 924–932 (2004)Google Scholar
  26. A.K. Petford-Long, R.C. Doole, C.N. Afonso, J. Solis: J. Appl. Phys. 77, 607–613 (1995)Google Scholar
  27. T. Missana, C.N. Afonso, A.K. Petford-Long, R.C. Doole: Philos. Mag. A 79, 2577–2590 (1999)Google Scholar
  28. R. Sinclair: Mater. Res. Soc. Bull. 19(6), 26–31 (1994)Google Scholar
  29. B. Blanpain, J.-M. Legresy, J.W. Mayer: J. Phys. Colloq. 51(C-4), 131–137 (1990)Google Scholar
  30. S.-W. Nam, H.-S. Chung, Y.C. Lo, L. Qi, J. Li, Y. Lu, A.T.C. Johnson, Y. Jung, P. Nukala, R. Agarwal: Science 336, 1561–1566 (2012)Google Scholar
  31. Y.-C. Wang, W. Zhang, L.-Y. Wang, Z. Zhuang, E. Ma, J. Li, Z.-W. Shan: NPG Asia Mater. 8, e291 (2016)Google Scholar
  32. Z. Liu, Y. Bando, M. Mitome, J. Zhan: Phys. Rev. Lett. 93, 095504 (2004)Google Scholar
  33. Y. Gao, Y. Bando: Nature 415, 599–599 (2002)Google Scholar
  34. T. Kato, K. Nunome, Y. Morimoto, K. Nishimura, H. Saka: Philos. Mag. Lett. 80, 187–192 (2000)Google Scholar
  35. Y. Senda, K. Sasaki, H. Saka: Philos. Mag. 84, 2635–2649 (2004)Google Scholar
  36. T.E.A. Zerrouk, M. Hamichi, J.D.H. Pilkington, J.A. Venables: Phys. Rev. B 50, 8946–8949 (1994)Google Scholar
  37. J.M. Howe: Philos. Mag. A 74, 761–775 (1996)Google Scholar
  38. H. Nishizawa, F. Hori, R. Oshima: J. Cryst. Growth 236, 51–58 (2002)Google Scholar
  39. G.A. Storaska, K.T. Moore, J.M. Howe: Philos. Mag. 84, 2619–2634 (2004)Google Scholar
  40. S.E. Donnelly, R.C. Birtcher, C.W. Allen, I. Morrison, K. Furuya, M. Song, K. Mitsuishi, U. Dahmen: Science 296, 507–510 (2002)Google Scholar
  41. S.H. Oh, Y. Kauffmann, C. Scheu, W.D. Kaplan, M. Rühle: Science 310, 661–663 (2005)Google Scholar
  42. M. Gandman, Y. Kauffmann, C.T. Koch, W.D. Kaplan: Phys. Rev. Lett. 110, 086106 (2013)Google Scholar
  43. S. Arai, S. Tsukimoto, S. Muto, H. Saka: Microsc. Microanal. 6, 358–361 (2000)Google Scholar
  44. S. Arai, S. Tsukimoto, H. Saka: J. Electron Microsc. 52, 79–84 (2003)Google Scholar
  45. K. Sasaki, H. Saka: Mater. Res. Soc. Symp. Proc. 466, 185–190 (1997)Google Scholar
  46. T. Kamino, K. Sasaki, H. Saka: Microsc. Microanal. 3, 393–408 (1997)Google Scholar
  47. H. Saka, S. Arai, S. Tsukimoto, K. Sasaki: Mater. Sci. Forum 294–296, 617–620 (1999)Google Scholar
  48. F. Niekiel, P. Schweizer, S.M. Kraschewski, B. Butz, E. Spiecker: Acta Mater. 90, 118–132 (2015)Google Scholar
  49. R.-M. Keller, W. Sigle, S.P. Baker, O. Kraft, E. Arzt: Mater. Res. Soc. Symp. Proc. 436, 221–226 (1997)Google Scholar
  50. R. Dannenberg, E. Stach, J.R. Grozac, B.J. Dresser: Thin Solid Films 379, 133–138 (2000)Google Scholar
  51. B. Kaouache, P. Gergaud, O. Thomas, O. Bostrom, M. Legros: Microelectron. Eng. 70, 447–454 (2003)Google Scholar
  52. R.C. Hugo, R.G. Hoagland: Scr. Mater. 38, 523–529 (1998)Google Scholar
  53. R.C. Hugo, R.G. Hoagland: Scr. Mater. 41, 1341–1346 (1999)Google Scholar
  54. K.L. Merkle, L.J. Thompson, F. Phillipp: Phys. Rev. Lett. 88, 225501 (2002)Google Scholar
  55. D.L. Medlin, G.H. Campbell, C.R. Carter: Acta Mater. 46, 5135–5142 (1998)Google Scholar
  56. K.L. Merkle, L.J. Thompson: Mater. Lett. 48, 188–193 (2001)Google Scholar
  57. K.L. Merkle, L.J. Thompson, F. Phillipp: Philos. Mag. Lett. 82, 589–597 (2002)Google Scholar
  58. M.L. Bowers, C. Ophus, A. Gautam, F. Lançon, U. Dahmen: Phys. Rev. Lett. 116, 106102 (2016)Google Scholar
  59. A. Gautam, C. Ophus, F. Lançon, P. Denes, U. Dahmen: Ultramicroscopy 151, 78–84 (2015)Google Scholar
  60. G. Lucadamo, D.L. Medlin: Acta Mater. 50, 3045–3055 (2002)Google Scholar
  61. J.C. Jiang, X.Q. Pan: Philos. Mag. Lett. 80, 271–279 (2000)Google Scholar
  62. D. Schryvers, Y. Ma: Mater. Lett. 23, 105–111 (1995)Google Scholar
  63. Y. Gao, Y.Q. Liu, K.H. Wu: Mater. Res. Soc. Symp. Proc. 404, 171–176 (1996)Google Scholar
  64. X.-G. Ma, K. Komvopoulos: J. Mater. Res. 20, 1808–1813 (2005)Google Scholar
  65. D. Schryvers, M. Yandouzi, L. Toth: Thin Solid Films 326, 126–133 (1998)Google Scholar
  66. Q. Ding, Z. Shen, S. Xiang, H. Tian, J. Li, Z. Zhang: J. Alloys Compd. 651, 255–258 (2015)Google Scholar
  67. S. Gao, Y. Zhou, C.-F. Li, Z.-Q. Liu, T. Jin: J. Alloys Compd. 671, 458–464 (2016)Google Scholar
  68. Q. Zhang, Y. Chang, L. Cu, Y. Luo, B. Ge: Scr. Mater. 126, 55–57 (2017)Google Scholar
  69. J.M. Howe, W.T. Reynolds Jr., V.K. Vasudevan: Z. Metallkd. 95, 275–278 (2004)Google Scholar
  70. J. Dutkiewicz, M. Bartsch, S. Miura, U. Messerschmidt, H. Kato: J. Phys. IV Colloq. 5(C2), 275–280 (1995)Google Scholar
  71. X.D. Han, R. Wang, Z. Zhang, D.Z. Yang: Mater. Lett. 30, 23–28 (1997)Google Scholar
  72. B. Jiang, T. Tadakit, H. Mori, T.Y. Hsu: Mater. Trans. JIM 38, 1072-77–1078–1082 (1997)Google Scholar
  73. X. Gao, J.F. Nie: Scr. Mater. 57, 655–658 (2007)Google Scholar
  74. T. Tamiya, D. Shindo, Y. Murakami, Y. Bando, K. Otsuka: Mater. Trans. JIM 39, 714–723 (1998)Google Scholar
  75. W. Cai, Y. Murakami, K. Otsuka: Mater. Sci. Eng. A 273–275, 186–189 (1999)Google Scholar
  76. Y. Murakami, D. Shindo: Philos. Mag. Lett. 81, 631–638 (2001)Google Scholar
  77. S. Ii, M. Nishida, Y. Murakami, D. Shindo: J. Phys. IV Proc. 112, 1035–1038 (2003)Google Scholar
  78. L.E. Tanner, A.R. Pelton, G. Van Tendeloo, D. Schryvers, M.E. Wall: Scr. Metall. Mater. 24, 1731–1736 (1990)Google Scholar
  79. J. Ye, R.K. Mishra, A. Pelton, A.M. Minor: Acta Mater. 58, 490–498 (2010)Google Scholar
  80. H. Guo, K. Chen, Y. Oh, K. Wang, C. Dejoie, S.A. Syed Asif, O.L. Warren, Z.W. Shan, J. Wu, A.M. Minor: Nano Lett. 11, 3207–3213 (2011)Google Scholar
  81. T. Kuninori, E. Sukedai, H. Hashimoto: Mater. Trans. JIM 37, 1404–1407 (1996)Google Scholar
  82. D. Schryvers, D.E. Lahjouji, B. Slootmaekers, P.L. Potapov: Scr. Mater. 35, 1235–1241 (1996)Google Scholar
  83. H. Matsumoto, E. Sukedai, H. Hashimoto: Mater. Trans. JIM 40, 1436–1443 (1999)Google Scholar
  84. F.M. Ross, R.R. Kola, R. Hull, J.C. Bean: Mater. Res. Soc. Symp. Proc. 318, 697–702 (1993)Google Scholar
  85. T.L. Morkved, W.A. Lopes, J. Hahm, S.J. Sibener, H.M. Jaeger: Polymer 39, 3871–3875 (1998)Google Scholar
  86. A.W. Grant, Q.-H. Hu, B. Kasemo: Nanotechnology 15, 1175–1181 (2004)Google Scholar
  87. H. Tanaka, T.J. Konno, R. Sinclair, N. Hirashita: J. Appl. Phys. 78, 4982–4987 (1995)Google Scholar
  88. H. Tanaka, N. Hirashita, R. Sinclair: Jpn. J. Appl. Phys. 35(2), L479 (1996)Google Scholar
  89. H. Tanaka, N. Hirashita, R. Sinclair: Jpn. J. Appl. Phys. 37(1), 4284–4287 (1998)Google Scholar
  90. M.V. Sidorov, B. Kardynal, D.J. Smith: Microsc. Microanal. 4, 317–324 (1998)Google Scholar
  91. C. Hayzelden, J.L. Batstone: J. Appl. Phys. 73, 8279–8231 (1993)Google Scholar
  92. V. Teodorescu, L. Nistor, H. Bender, A. Steegen, A. Lauwers, K. Maex, J. Van Landuyt: J. Appl. Phys. 90, 167–174 (2001)Google Scholar
  93. C. Ghica, L. Nistor, H. Bender, A. Steegen, A. Lauwers, K. Maex, J. Van Landuyt: J. Mater. Res. 16, 701–708 (2001)Google Scholar
  94. L.M. Gignac, V. Svilan, L.A. Clevenger, C. Cabral, C. Lavoie: Mater. Res. Soc. Symp. Proc. 441, 225–230 (1997)Google Scholar
  95. C.-Y. Wen, F. Spaepen: Philos. Mag. 87(35), 5581–5599 (2007)Google Scholar
  96. Y.-C. Chou, W.-W. Wu, S.-L. Cheng, B.-Y. Yoo, N. Myung, L.J. Chen, K.N. Tu: Nano Lett. 8, 2194–2199 (2008)Google Scholar
  97. Y.-C. Chou, W.-W. Wu, L.J. Chen, K.N. Tu: Nano Lett. 9, 2337–2342 (2009)Google Scholar
  98. Y.-C. Chou, W. Tang, C.-J. Chiou, K. Chen, A.M. Minor, K.N. Tu: Nano Lett. 15, 4121–4128 (2015)Google Scholar
  99. F.M. Ross: IBM J. Res. Dev. 44, 489–501 (2000)Google Scholar
  100. M.H. Wang, L.J. Chen: J. Appl. Phys. 71, 5918–5925 (1992)Google Scholar
  101. R. Sinclair, T. Yamashita, M.A. Parker, K.B. Kim, K. Holloway, A.F. Schwartzman: Acta Cryst. A 44, 965–975 (1988)Google Scholar
  102. J.M. Gibson, J.L. Batstone, R.T. Tung: Appl. Phys. Lett. 51, 45–47 (1987)Google Scholar
  103. F.M. Ross, P.A. Bennett, R.M. Tromp, J. Tersoff, M. Reuter: Micron 30, 21–32 (1999)Google Scholar
  104. H.P. Sun, Y.B. Chen, X.Q. Pan, D.Z. Chi, R. Nath, Y.L. Foo: Appl. Phys. Lett. 86, 71904 (2005)Google Scholar
  105. R. Nath, C.W. Soo, C.B. Boothroyd, M. Yeadon, D.Z. Chi, H.P. Sun, Y.B. Chen, X.Q. Pan, Y.L. Foo: Appl. Phys. Lett. 86, 201908 (2005)Google Scholar
  106. M. Tanaka, M. Han, M. Takeguchi, K. Furuya: Philos. Mag. 84, 2699–2709 (2004)Google Scholar
  107. M.W. Kleinschmit, M. Yeadon, J.M. Gibson: Appl. Phys. Lett. 75, 3288–3290 (1999)Google Scholar
  108. R.K.K. Chong, M. Yeadon, W.K. Choi, E.A. Stach, C.B. Boothroyd: Appl. Phys. Lett. 82, 1833–1835 (2003)Google Scholar
  109. K. He, S. Zhang, J. Li, X. Yu, Q. Meng, Y. Zhu, E. Hu, K. Sun, H. Yun, X.-Q. Yang, Y. Zhu, H. Gan, Y. Mo, E.A. Stach, C.B. Murray, D. Su: Nat. Commun. 7, 11441 (2016)Google Scholar
  110. J.Y. Huang, L. Zhong, C.M. Wang, J.P. Sullivan, W. Xu, L.Q. Zhang, S.X. Mao, N.S. Hudak, X.H. Liu, A. Subramanian, H. Fan, L. Qi, A. Kushima, J. Li: Science 330, 1515–1520 (2010)Google Scholar
  111. M. Wang, D. Golberg, Y. Bando: Adv. Mater. 22, 4071–4075 (2010)Google Scholar
  112. F. Wang, H.-C. Yu, M.-H. Chen, L. Wu, N. Pereira, K. Thornton, A. Van der Ven, Y. Zhu, G.G. Amatucci, J. Graetz: Nat. Commun. 3, 1201 (2012)Google Scholar
  113. X.H. Liu, J.W. Wang, S. Huang, F. Fan, X. Huang, Y. Liu, S. Krylyuk, J. Yoo, S.A. Dayeh, A.V. Davydov, S.X. Mao, S.T. Picraux, S. Zhang, J. Li, T. Zhu, J.Y. Huang: Nat. Nanotechnol. 7, 749–756 (2012)Google Scholar
  114. M. Gu, Z.G. Wang, J.G. Connell, D.E. Perea, L.J. Lauhon, F. Gao, C.M. Wang: ACS Nano 7, 6303 (2013)Google Scholar
  115. K. Yamamoto, Y. Iriyama, T. Hirayama: Microscopy 66, 50–61 (2017)Google Scholar
  116. X.H. Liu, Y. Liu, A. Kushima, S. Zhang, T. Zhu, J. Li, J.Y. Huang: Adv. Energy Mater. 2, 722–741 (2012)Google Scholar
  117. C.M. Wang: J. Mater. Res. 30, 326–339 (2014)Google Scholar
  118. M.T. Janish, D.T. Mackay, Y. Liu, K.L. Jungjohann, C.B. Carter, M.G. Norton: J. Mater. Sci. 51, 589–602 (2016)Google Scholar
  119. C.-M. Wang, H.-G. Liao, F.M. Ross: MRS Bulletin 40, 46–52 (2015)Google Scholar
  120. Z. Xu, Y. Bando, W. Wang, X. Bai, D. Golberg: ACS Nano 4, 2515–2522 (2010)Google Scholar
  121. T. Fujii, M. Arita, Y. Takahashi, I. Fujiwara: Appl. Phys. Lett. 98, 212104 (2011)Google Scholar
  122. Y. Yang, W. Lu, Y. Yao, J. Sun, C. Gu, L. Gu, Y. Wang, X. Duan, R. Yu: Sci. Rep. 4, 3890 (2014)Google Scholar
  123. X. Tian, S. Yang, M. Zeng, L. Wang, J. Wei, Z. Xu, W. Wang, X. Bai: Adv. Mater. 26, 3649–3654 (2014)Google Scholar
  124. W.A. Hubbard, A. Kerelsky, G. Jasmin, E.R. White, J. Lodico, M. Mecklenburg, B.C. Regan: Nano Lett. 15, 3983–3987 (2015)Google Scholar
  125. Y. Yang, P. Gao, S. Gaba, T. Chang, X. Pan, W. Lu: Nat. Commun. 3, 732 (2012)Google Scholar
  126. Y.-T. Huang, S.-Y. Yu, C.-L. Hsin, C.-W. Huang, C.-F. Kang, F.-H. Chu, J.-Y. Chen, J.-C. Hu, L.-T. Chen, J.-H. He, W.-W. Wu: Anal. Chem. 85, 3955–3960 (2013)Google Scholar
  127. J. Norpoth, S. Mildner, M. Scherff, J. Hoffmann, C. Jooss: Nanoscale 6, 9852–9862 (2014)Google Scholar
  128. M. Duchamp, V. Migunov, A.H. Tavabi, A. Mehonic, M. Buckwell, M. Munde, A.J. Kenyon, R.E. Dunin-Borkowski: Resolut. Discov. 1, 27–33 (2016)Google Scholar
  129. K. D’Aquila, Y. Liu, H. Iddir, A.K. Petford-Long: Phys. Status Solidi Rapid Res. Lett. 9, 301–306 (2015)Google Scholar
  130. R.J. Kamaladasa, A.A. Sharma, Y.-T. Lai, W. Chen, P.A. Salvador, J.A. Bain, M. Skowronski, Y.N. Picard: Microsc. Microanal. 21, 140–153 (2015)Google Scholar
  131. K. Baek, S. Park, J. Park, Y.-M. Kim, H. Hwang, S.H. Oh: Nanoscale 9, 582–593 (2017)Google Scholar
  132. H. Gabrisch, L. Kjeldgaard, E. Johnson, U. Dahmen: Acta Mater. 49, 4259 (2001)Google Scholar
  133. U. Dahmen, S. Hagege, F. Faudot, T. Radetic, E. Johnson: Philos. Mag. 84, 2651–2662 (2004)Google Scholar
  134. H. Gabrisch, L. Kjeldgaard, A. Vaernholt Olesen, E. Johnson, U. Dahmen: Mater. Res. Soc. Symp. Proc. 580, 257–264 (2000)Google Scholar
  135. L.H. Zhang, E. Johnson, U. Dahmen: Mater. Res. Soc. Symp. Proc. 821, 181–186 (2004)Google Scholar
  136. P. Bhattacharya, V. Bhattacharya, K. Chattopadhyay: J. Electron Microsc. 48(Suppl.), S1047–S1054 (1999)Google Scholar
  137. E. Johnson, A. Johansen, C. Nelson, U. Dahmen: J. Electron Microsc. 51(Suppl.), S201–S209 (2002)Google Scholar
  138. M. Legros, B. Kaouache, P. Gergaud, O. Thomas, G. Dehm, T.J. Balk, E. Arzt: Philos. Mag. 85, 3541–3552 (2005)Google Scholar
  139. E. Johnson, M.T. Levinsen, S. Steenstrupt, S. Prokofjev, V. Zhilin, U. Dahmen, T. Radetic: Philos. Mag. 84, 2663–2673 (2004)Google Scholar
  140. J.M. Howe: Mater. Trans. JIM 39, 3–23 (1998)Google Scholar
  141. W.E. Benson, J.M. Howe: Philos. Mag. A 75, 1641–1663 (1997)Google Scholar
  142. T. Isshiki, M. Ohta, K. Nishio, M. Shiojiri: Thin Solid Films 271, 84–87 (1995)Google Scholar
  143. B.J. Kooi, J.T.M. De Hosson: Mater. Res. Soc. Symp. Proc. 654, AA4.1.1 (2001)Google Scholar
  144. J.M. Howe, T.M. Murray, K.T. Moore, A.A. Csontos, M.M. Tsai, A. Garg, W.E. Benson: Microsc. Microanal. 4, 235–247 (1998)Google Scholar
  145. S.K. Malladi, Q. Xu, M.A. van Huis, F.D. Tichelaar, K.J. Batenburg, E. Yücelen, B. Dubiel, A. Czyrska-Filemonowicz, H.W. Zandbergen: Nano Lett. 14, 384–389 (2014)Google Scholar
  146. D.J. Smith, A.K. Petford-Long, L.R. Wallenberg, J.-O. Bovin: Science 233, 872–875 (1986)Google Scholar
  147. S. Iijima, T. Ichihashi: Phys. Rev. Lett. 56, 616–619 (1986)Google Scholar
  148. A.S. Barnard, N.P. Young, A.I. Kirkland, M.A. van Huis, H. Xu: ACS Nano 3, 1431–1436 (2009)Google Scholar
  149. J.M. Howe: Interfaces in Materials (Wiley, New York 1997)Google Scholar
  150. A.O. Yalcin, Z. Fan, B. Goris, W.-F. Li, R.S. Koster, C.-M. Fang, A. van Blaaderen, M. Casavola, F.D. Tichelaar, S. Bals, G. Van Tendeloo, T.J.H. Vlugt, D. Vanmaekelbergh, H.W. Zandbergen, M.A. van Huis: Nano Lett. 14(6), 3661–3667 (2014)Google Scholar
  151. A.O. Yalcin, B. Goris, R.J.A. van Dijk-Moes, Z. Fan, A.K. Erdamar, F.D. Tichelaar, T.J.H. Vlugt, G. Van Tendeloo, S. Bals, D. Vanmaekelbergh, H.W. Zandbergen, M.A. van Huis: Chem. Commun. 51, 3320–3323 (2015)Google Scholar
  152. K. Chatterjee, J.M. Howe, W.C. Johnson, M. Murayama: Acta Mater. 52, 2923–2935 (2004)Google Scholar
  153. Y. Li, Z. Lia, F. Fang, X. Wang, L. Li, J. Zhu: Appl. Phys. Lett. 105, 182901 (2014)Google Scholar
  154. H. Yasuda, K. Mitsuishi, H. Mori: Phys. Rev. B 64, 094101 (2001)Google Scholar
  155. J.G. Lee, H. Mori, H. Yasuda: Phys. Rev. B 65, 132106 (2002)Google Scholar
  156. H. Yasuda, H. Mori, K. Furuya: Philos. Mag. Lett. 80, 181–186 (2000)Google Scholar
  157. J.G. Lee, H. Mori: Philos. Mag. 84, 2675–2686 (2004)Google Scholar
  158. J.G. Lee, H. Mori: Phys. Rev. Lett. 93, 235501 (2004)Google Scholar
  159. T. Yokota, J.M. Howe, W.A. Jesser, M. Murayama: J. Appl. Phys. 95, 5756–5761 (2004)Google Scholar
  160. H. Yasuda, H. Mori, J.G. Lee: Phys. Rev. B 70, 214105 (2004)Google Scholar
  161. A. Figuerola, M. van Huis, M. Zanella, A. Genovese, S. Marras, A. Falqui, H.W. Zandbergen, R. Cingolani, L. Manna: Nano Lett. 10(8), 3028–3036 (2010)Google Scholar
  162. F. Banhart, E. Hernandez, M. Terrones: Phys. Rev. Lett. 90, 185502 (2003)Google Scholar
  163. A.K. Schaper, F. Phillipp, H. Hou: J. Mater. Res. 20, 1844–1850 (2005)Google Scholar
  164. G.A. Storaska, J.M. Howe: Mater. Sci. Eng. A 368, 183–190 (2004)Google Scholar
  165. T. Daiming, C.L. Ren, R. Lv, W.J. Yu, P.X. Hou, M. Wang, X. Wei, Z. Xu, N. Kawamoto, Y. Bando, M. Mitome, C. Liu, H.M. Cheng, D. Golberg: Nano Lett. 15(8), 4922–4927 (2015)Google Scholar
  166. M.A. van Huis, L.T. Kunneman, K. Overgaag, Q. Xu, G. Pandraud, H.W. Zandbergen, D. Vanmaekelbergh: Nano Lett. 8, 3959–3963 (2008)Google Scholar
  167. A.P. Lange, A. Samanta, H. Majidi, S. Mahajan, J. Ging, T.Y. Olson, K. van Benthem, S. Elhadj: Acta Mater. 120, 364 (2016)Google Scholar
  168. T. Kamino, T. Yaguchi, M. Ukiana, Y. Yasutomi, H. Saka: Mater. Trans. JIM 36, 73–75 (1995)Google Scholar
  169. T. Vystavel, G. Palasantzas, S.A. Koch, J.T.M. De Hosson: Appl. Phys. Lett. 83, 3909–3911 (2003)Google Scholar
  170. T. Vystavel, G. Palasantzas, S.A. Koch, J.T.M. De Hosson: Appl. Phys. Lett. 82, 197–199 (2003)Google Scholar
  171. S.B. Simonsen, I. Chorkendorff, S. Dahl, M. Skoglundh, J. Sehested, S. Helveg: J. Am. Chem. Soc. 132, 7968–7975 (2010)Google Scholar
  172. A.T. DeLaRiva, T.W. Hansen, S.R. Challa, A.K. Datye: J. Catal. 308, 291–305 (2013)Google Scholar
  173. M. Yeadon, J.C. Yang, R.S. Averback, J.W. Bullard, D.L. Olynick, J.M. Gibson: Appl. Phys. Lett. 71, 1631–1633 (1997)Google Scholar
  174. C.G. Zimmermann, M. Yeadon, K. Nordlund, J.M. Gibson, R.S. Averback, U. Herr, K. Samwer: Phys. Rev. Lett. 83, 1163–1166 (1999)Google Scholar
  175. P.M. Mortensen, T.W. Hansen, J.B. Wagner, A.D. Jensen: Ultramicroscopy 152, 1–9 (2015)Google Scholar
  176. T.W. Hansen, J.B. Wagner (Eds.): Controlled Environment Transmission Electron Microscopy: Principles and Practice (Springer, Cham 2016)Google Scholar
  177. J.R. Jinschek: Chem. Commun. 50, 2696 (2014)Google Scholar
  178. H. Poppa: J. Vac. Sci. Technol. A 22, 1931–1947 (2004)Google Scholar
  179. H. Minoda, K. Yagi, F.-J. Meyer zu Heringdorf, A. Meier, D. Kahler, M. Horn von Hoegen: Phys. Rev. B 59, 2363–2375 (1999)Google Scholar
  180. T. Kamino, T. Yaguchi, M. Tomita, H. Saka: Philos. Mag. A 75, 105–114 (1997)Google Scholar
  181. K. Takayanagi, Y. Tanishiro, S. Takahashi, M. Takahashi: Surf. Sci. 164, 367–392 (1985)Google Scholar
  182. Y. Oshima, T. Hirata, T. Yokoyama, H. Hirayama, K. Takayanagi: Surf. Sci. 465, 81–89 (2000)Google Scholar
  183. J. Liu, M. Takeguchi, M. Tanaka, H. Yasuda, K. Furuya: J. Electron Microsc. 50, 541–544 (2001)Google Scholar
  184. L.D. Marks, E. Bengu, C. Collazo-Davila, D. Grozea, E. Landree, C. Leslie, W. Sinkler: Surf. Rev. Lett. 5, 1087–1106 (1998)Google Scholar
  185. C. Collazo-Davila, D. Grozea, L.D. Marks: Phys. Rev. Lett. 80, 1678–1681 (1998)Google Scholar
  186. K. Aoki, H. Minoda, Y. Tanishiro, K. Yagi: Surf. Rev. Lett. 5, 653–663 (1998)Google Scholar
  187. H. Minoda, K. Yagi: Surf. Sci. 437, L761–6 (1999)Google Scholar
  188. A.V. Latyshev, D.A. Nasimov, V.N. Savenko, A.L. Aseev: Thin Solid Films 367, 142–148 (2000)Google Scholar
  189. M. Gajdardziska-Josifovska, R. Plass, M.A. Schofield, D.R. Giese, R. Sharma: J. Electron Microsc. 51(Suppl.), S13–S26 (2002)Google Scholar
  190. M. Gajdardziska-Josifovska, M.H. Malay, D.J. Smith: Surf. Rev. Lett. 4, 655–669 (1997)Google Scholar
  191. Y. Homma, P. Finnie: J. Cryst. Growth 237-239, 28–34 (2002)Google Scholar
  192. L.D. Marks: Phys. Rev. Lett. 51, 1000 (1983)Google Scholar
  193. J.-O. Bovin, R. Wallenberg, D.J. Smith: Nature 317, 47–49 (1985)Google Scholar
  194. M. Mitome, K. Takayanagi, Y. Tanishiro: Phys. Rev. B 42, 7238–7241 (1990)Google Scholar
  195. D.J. Smith, M. Gajdardziska-Josifovska, P. Lu, M.R. McCartney, J. Podbrdsky, P.R. Swann, J.S. Jones: Ultramicroscopy 49, 26–36 (1993)Google Scholar
  196. X.G. Ning, M. Avalos-Borja, H.Q. Ye: J. Appl. Phys. 79, 125–128 (1996)Google Scholar
  197. W. Yuan, Y. Wang, H. Li, H. Wu, Z. Zhang, A. Selloni, C. Sun: Nano Lett. 16, 132–137 (2016)Google Scholar
  198. H. Koizumi, Y. Oshima, Y. Kondo, K. Takayanagi: Ultramicroscopy 88, 17–24 (2001)Google Scholar
  199. Y. Kondo, K. Takayanagi: Science 289, 606–608 (2000)Google Scholar
  200. Y. Oshima, Y. Kondo, K. Takayanagi: J. Electron Microsc. 52, 49–55 (2003)Google Scholar
  201. Y. Oshima, A. Onga, K. Takayanagi: Phys. Rev. Lett. 91, 205503 (2003)Google Scholar
  202. Y. Oshima, H. Koizumi, K. Mouri, H. Hirayama, K. Takayanagi, Y. Kondo: Phys. Rev. B 65(121401), 1–4 (2002)Google Scholar
  203. N. Tanaka, H. Kimata, T. Takashima, T. Kizuka: Surf. Rev. Lett. 5, 723–729 (1998)Google Scholar
  204. J.C. Yang, D. Evan, L. Tropia: Appl. Phys. Lett. 81, 241–243 (2002)Google Scholar
  205. F.M. Ross, J.M. Gibson: Phys. Rev. Lett. 68, 1782–1785 (1992)Google Scholar
  206. J.C. Yang, A.B. Kolasa, J.M. Gibson, M. Yeadon: Appl. Phys. Lett. 73, 2841–2843 (1998)Google Scholar
  207. G. Zhou, J.C. Yang: Phys. Rev. Lett. 89, 106101 (2002)Google Scholar
  208. G. Zhou, J.C. Yang: Surf. Sci. 531, 359–367 (2003)Google Scholar
  209. H. Liu, H. Zheng, L. Li, H. Sheng, S. Jia, F. Cao, X. Liu, B. Chen, R. Xing, D. Zhao, J. Wang: Nano Res. (2017),  https://doi.org/10.1007/s12274-017-1429-2CrossRefGoogle Scholar
  210. L. Wang, J.C. Yang: J. Mater. Res. 20, 1902–1909 (2005)Google Scholar
  211. W. Harlow, H. Ghassemi, M.L. Taheri: J. Nucl. Mater. 474, 126–133 (2016)Google Scholar
  212. X. Li, S. Cheng, S. Deng, X. Wei, J. Zhu, Q. Chen: Sci. Rep. 7, 40911 (2017)Google Scholar
  213. S. Rackauskas, H. Jiang, J.B. Wagner, S.D. Shandakov, T.W. Hansen, E.I. Kauppinen, A.G. Nasibulin: Nano Lett. 14, 5810–5813 (2014)Google Scholar
  214. M. Ichikawa: J. Phys. Condens. Matter 11, 9861–9870 (1999)Google Scholar
  215. F.M. Ross, J.M. Gibson, R.D. Twesten: Surf. Sci. 310, 243–266 (1994)Google Scholar
  216. M.J. Sayagues, J.L. Hutchison: J. Solid State Chem. 163, 137–143 (2002)Google Scholar
  217. M.J. Sayagues, J.L. Hutchison: J. Solid State Chem. 124, 116–122 (1996)Google Scholar
  218. P.L. Gai, K. Kourtakis: Science 267, 661 (1995)Google Scholar
  219. L. Diebolt, R. Sharma, M. McKelvy, W.S. Glaunsinger: Mat. Res. Soc. Proc. 404, 183–188 (1995)Google Scholar
  220. M.V. Sidorov, M.J. McKelvy, R. Sharma, W.S. Glaunsinger: J. Solid State Chem. 141, 330–337 (1998)Google Scholar
  221. M.J. McKelvy, R. Sharma, A.V.G. Chizmeshya, R.W. Carpenter, K. Streib: Chem. Mater. 13, 921–926 (2001)Google Scholar
  222. R. Sharma, P.A. Crozier, Z.C. Kang, L. Eyring: Philos. Mag. 84, 2731–2747 (2004)Google Scholar
  223. K. Karki, Y. Huang, S. Hwang, A.D. Gamalski, M.S. Whittingham, G. Zhou, E.A. Stach: ACS Appl. Mater. Interfaces 8, 27762–27771 (2014)Google Scholar
  224. H. Ghassemi, W. Harlow, O. Mashtalir, M. Beidaghi, M.R. Lukatskaya, Y. Gogotsi, M.L. Taheri: J. Mater. Chem. A 2, 14339–14343 (2014)Google Scholar
  225. Q. Jeangros, T.W. Hansen, J.B. Wagner, R.E. Dunin-Borkowski, C. Hébert, J. Van herle, A. Hessler-Wyser: Chem. Commun. 50, 1808–1810 (2014)Google Scholar
  226. Q. Jeangros, T.W. Hansen, J.B. Wagner, R.E. Dunin-Borkowski, C. Hébert, J. Van Herle, A. Hessler-Wyser: Ultramicroscopy 169, 11–21 (2016)Google Scholar
  227. R. Sharma, D. Naedele, E. Schweda: Chem. Mater. 13, 4014–4018 (2001)Google Scholar
  228. R. Sharma, M.J. McKelvy, H. Bearat, A.V.G. Chizmeshya, R.W. Carpenter: Philos. Mag. 84, 2711–2729 (2004)Google Scholar
  229. M. Gajdardziska-Josifovska, R. Sharma: Microsc. Microanal. 11, 524–533 (2005)Google Scholar
  230. J.R. Jinschek, S. Helveg: Micron 11, 1156–1168 (2012)Google Scholar
  231. R.-J. Liu, P.A. Crozier, C.M. Smith, D.A. Hucul, J. Blackson, G. Salaita: Microsc. Microanal. 10, 77–85 (2004)Google Scholar
  232. P.L. Hansen, J.B. Wagner, S. Helveg, J.R. Rostrup-Nielsen, B.S. Clausen, H. Topsoe: Science 295, 2053–2055 (2002)Google Scholar
  233. P.L. Gai, B.C. Smith, G. Owen: Nature 348, 430–432 (1990)Google Scholar
  234. P.L. Gai: Adv. Mater. 10, 1259–1263 (1998)Google Scholar
  235. P.L. Gai, K. Kourtakis, S. Ziemecki: Microsc. Microanal. 6, 335–342 (2000)Google Scholar
  236. P.L. Gai: Top. Catal. 21, 161–173 (2002)Google Scholar
  237. T.W. Hansen, J.B. Wagner, P.L. Hansen, S. Dahl, H. Topsoe, C.J.H. Jacobsen: Science 294, 1508–1510 (2001)Google Scholar
  238. R. Sharma: Microsc. Microanal. 7, 494–506 (2001)Google Scholar
  239. R. Sharma: J. Mater. Res. 20, 1695–1707 (2005)Google Scholar
  240. S. Takeda, Y. Kuwauchi, H. Yoshida: Ultramicroscopy 151, 178–190 (2015)Google Scholar
  241. E.D. Boyes, P.L. Gai: Ultramicroscopy 67, 219–232 (1997)Google Scholar
  242. H. Topsøe: J. Catal. 216, 155–164 (2003)Google Scholar
  243. J.C. Yang, M.W. Small, R.V. Grieshaber, R.G. Nuzzo: Chem. Soc. Rev. 41, 8179–8194 (2012)Google Scholar
  244. T.W. Hansen, J.B. Wagner: ACS Catalysis 4, 1673–1685 (2014)Google Scholar
  245. P.A. Crozier, T.W. Hansen: MRS Bulletin 40, 38–45 (2015)Google Scholar
  246. F. Tao, P.A. Crozier: Chem. Rev. 116, 3487–3539 (2016)Google Scholar
  247. J.F. Creemer, S. Helveg, G.H. Hoveling, S. Ullmann, A.M. Molenbroek, P.M. Sarro, H.W. Zandbergen: Ultramicroscopy 108, 993–998 (2008)Google Scholar
  248. S. Giorgio, S. Sao Joao, S. Nitsche, D. Chaudanson, G. Sitja, C.R. Henry: Ultramicroscopy 106, 503–507 (2006)Google Scholar
  249. L.F. Allard, S.H. Overbury, W.C. Bigelow, M.B. Katz, D.P. Nackashi, J. Damiano: Microsc. Microanal. 18, 656–666 (2012)Google Scholar
  250. T. Yokosawa, T. Alan, G. Pandraud, B. Dam, H. Zandbergen: Ultramicroscopy 112, 47–52 (2012)Google Scholar
  251. S.B. Vendelbo, C.F. Elkjær, H. Falsig, I. Puspitasari, P. Dona, L. Mele, B. Morana, B.J. Nelissen, R. van Rijn, J.F. Creemer, P.J. Kooyman, S. Helveg: Nat. Mater. 13, 884–890 (2014)Google Scholar
  252. S. Zhang, C. Chen, M. Cargnello, P. Fornasiero, R.J. Gorte, G.W. Graham, X. Pan: Nat. Commun. 6, 7778 (2015)Google Scholar
  253. S. Zhang, P.N. Plessow, J.J. Willis, S. Dai, M. Xu, G.W. Graham, M. Cargnello, F. Abild-Pedersen, X. Pan: Nano Lett. 16, 4528–4534 (2016)Google Scholar
  254. L. Dubau, J. Nelayah, S. Moldovan, O. Ersen, P. Bordet, J. Drnec, T. Asset, R. Chattot, F. Maillard: ACS Catalysis 6, 4673–4684 (2016)Google Scholar
  255. H.L. Xin, K. Niu, D.H. Alsem, H. Zheng: Microsc. Microanal. 19, 1558–1568 (2014)Google Scholar
  256. Y.A. Wu, L. Li, Z. Li, A. Kinaci, M.K.Y. Chan, Y. Sun, J.R. Guest, I. McNulty, T. Rajh, Y. Liu: ACS Nano 10, 3738–3746 (2016)Google Scholar
  257. S.B. Vendelbo, P.J. Kooyman, J.F. Creemer, B. Morana, L. Mele, P. Dona, B.J. Nelissen, S. Helveg: Ultramicroscopy 133, 72–79 (2013)Google Scholar
  258. S. Chenna, P.A. Crozier: ACS Catalysis 2, 2395–2402 (2012)Google Scholar
  259. B.K. Miller, T.M. Barker, P.A. Crozier: Ultramicroscopy 156, 18–22 (2015)Google Scholar
  260. Y. Li, D. Zakharov, S. Zhao, R. Tappero, U. Jung, A. Elsen, PhBaumann, R.G. Nuzzo, E.A. Stach, A.I. Frenkel: Nat. Commun. 6, 7583 (2015)Google Scholar
  261. K. Yoshida, J. Yamasaki, N. Tanaka: Appl. Phys. Lett. 84, 2542–2544 (2004)Google Scholar
  262. F. Cavalca, A.B. Laursen, B.E. Kardynal, R.E. Dunin-Borkowski, S. Dahl, J.B. Wagner, T.W. Hansen: Nanotechnology 23, 075705 (2012)Google Scholar
  263. B.K. Miller, P.A. Crozier: Microsc. Microanal. 19, 461–469 (2013)Google Scholar
  264. L. Zhang, B.K. Miller, P.A. Crozier: Nano Lett. 13, 679–684 (2013)Google Scholar
  265. P. Crozier, V. Oleshko, A. Westwood, R. Cantrell: In situ environmental transmission electron microscopy of gas phase ziegler-natta catalytic polymerization of propylene. In: Proc. Electron Microsc. Anal. Inst. Phys. Conf. Ser. (2001) pp. 393–396Google Scholar
  266. P.L. Gai: Microsc. Microanal. 8, 21–28 (2002)Google Scholar
  267. F. Banhart: J. Appl. Phys. 81, 3440–3445 (1997)Google Scholar
  268. R. Sharma, Z. Iqbal: Appl. Phys. Lett. 84, 990–992 (2004)Google Scholar
  269. R. Sharma, P. Rez, M.M.J. Treacy, J. Stuart: J. Electron Microsc. 54, 231–237 (2005)Google Scholar
  270. S. Helveg, C. Lopez-Cartes, J. Sehested, P.L. Hansen, B.S. Clausen, J.R. Rostrup-Nielsen, F. Abild-Pedersen, J.K. Norskov: Nature 427, 426–429 (2004)Google Scholar
  271. H. Yoshida, S. Takeda, T. Uchiyama, H. Kohno, Y. Homma: Nano Lett. 8, 2082–2086 (2008)Google Scholar
  272. X. Feng, S.W. Chee, R. Sharma, K. Liu, X. Xie, Q. Li, S. Fan, K. Jiang: Nano Res. 4, 767 (2011)Google Scholar
  273. V. Balakrishnan, M. Bedewy, E.R. Meshot, S.W. Pattinson, E.S. Polsen, F. Laye, D.N. Zakharov, E.A. Stach, A.J. Hart: ACS Nano 10, 11496–11504 (2016)Google Scholar
  274. A.R. Harutyunyan, G. Chen, T.M. Paronyan, E.M. Pigos, O.A. Kuznetsov, K. Hewaparakrama, S.M. Kim, D. Zakharov, E.A. Stach, G.U. Sumanasekera: Science 326, 116–120 (2009)Google Scholar
  275. J. Wu, S. Helveg, S. Ullmann, Z. Peng, A.T. Bell: J. Catal. 338, 295–304 (2016)Google Scholar
  276. J. Kling, T.W. Hansen, J.B. Wagner: Carbon 99, 261–266 (2016)Google Scholar
  277. J.A. Rodríguez-Manzo, M. Terrones, H. Terrones, H.W. Kroto, L. Sun, F. Banhart: Nat. Nanotechnol. 2, 307 (2007)Google Scholar
  278. D.-M. Tang, C. Liu, W.-J. Yu, L.-L. Zhang, P.-X. Hou, J.-C. Li, F. Li, Y. Bando, D. Golberg, H.-M. Cheng: ACS Nano 8, 292–301 (2014)Google Scholar
  279. A.P. Burden, J.L. Hutchison: Carbon 36, 1167–1173 (1998)Google Scholar
  280. H. Abe, S. Yamamoto, A. Miyashita: J. Electron Microsc. 51, 183–187 (2002)Google Scholar
  281. D. Ugarte: Nature 359, 707–709 (1992)Google Scholar
  282. M. Terrones, F. Banhart, N. Grobert, J.-C. Charlier, H. Terrones, P.M. Ajayan: Phys. Rev. Lett. 89(075505), 1–4 (2002)Google Scholar
  283. M. Terrones, H. Terrones, F. Banhart, J.-C. Charlier, P.M. Ajayan: Science 288, 1226–1229 (2000)Google Scholar
  284. Q. Ru, M. Okamoto, Y. Kondo, K. Takayanagi: Chem. Phys. Lett. 259, 425–431 (1996)Google Scholar
  285. K. Niihara, T. Sekino, A. Nakahira, T. Kamino, T. Yaguchi: Mater. Lett. 27, 121–124 (1996)Google Scholar
  286. F. Banhart: Rep. Prog. Phys. 62, 1181–1221 (1999)Google Scholar
  287. E. Bengu, L.D. Marks: Phys. Rev. Lett. 86, 2385–2387 (2001)Google Scholar
  288. V.V. Roddatis, V.L. Kuznetsov, Y.V. Butenko, D.S. Su, R. Schlogl: Phys. Chem. Chem. Phys. 4, 1964–1967 (2002)Google Scholar
  289. H.E. Troiani, A. Camacho-Bragado, V. Armendariz, J.L. Gardea Torresday, M.J. Yacaman: Chem. Mater. 15, 1029–1031 (2003)Google Scholar
  290. A. Gloter, K. Suenaga, H. Kataura, R. Fujii, T. Kodama, H. Nishikawa, I. Ikemoto, K. Kikuchi, S. Suzuki, Y. Achiba, S. Iijima: Chem. Phys. Lett. 390, 462–466 (2004)Google Scholar
  291. M.S. Wang, J.Y. Wang, Q. Chen, L.M. Peng: Adv. Funct. Mater. 15, 1825–1831 (2005)Google Scholar
  292. F.M. Ross, J. Tersoff, R.M. Tromp: Phys. Rev. Lett. 80, 984–987 (1998)Google Scholar
  293. M. Hammar, F.K. LeGoues, J. Tersoff, M.C. Reuter, R.M. Tromp: Surf. Sci. 349, 129–144 (1996)Google Scholar
  294. T. Kizuka, N. Tanaka: Phys. Rev. B 56, R10079–82 (1997)Google Scholar
  295. T. Kizuka, N. Tanaka: Philos. Mag. Lett. 76, 289–297 (1997)Google Scholar
  296. J. Drucker, R. Sharma, K. Weiss, J. Kouvetakis: J. Appl. Phys. 77, 2846–2848 (1995)Google Scholar
  297. M. Krishnamurthy, J.S. Drucker, J.A. Venables: Mater. Res. Soc. Symp. Proc. 202, 77–82 (1991)Google Scholar
  298. H. Minoda, K. Yagi: J. Cryst. Growth 163, 48–53 (1996)Google Scholar
  299. S. Maruno, H. Nakahara, S. Fujita, H. Watanabe, Y. Kusumi, M. Ichikawa: Rev. Sci. Instrum. 68, 116–119 (1997)Google Scholar
  300. F.K. LeGoues, M. Hammar, M.C. Reuter, R.M. Tromp: Surf. Sci. 349, 249–266 (1996)Google Scholar
  301. F.K. LeGoues, M.C. Reuter, J. Tersoff, M. Hammar, R.M. Tromp: Phys. Rev. Lett. 73, 300–303 (1994)Google Scholar
  302. F.K. LeGoues, J. Tersoff, M.C. Reuter, M. Hammar, R. Tromp: Appl. Phys. Lett. 67, 2317–2319 (1995)Google Scholar
  303. S. Maruno, S. Fujita, H. Watanabe, Y. Kusumi, M. Ichikawa: Appl. Phys. Lett. 68, 2213–2215 (1996)Google Scholar
  304. A. Portavoce, M. Kammler, R. Hull, M.C. Reuter, M. Copel, F.M. Ross: Phys. Rev. B 70(9), 195306–195301 (2004)Google Scholar
  305. F.M. Ross, R.M. Tromp, M.C. Reuter: Science 286, 1931–1934 (1999)Google Scholar
  306. E.A. Stach, R. Hull, R.M. Tromp, M.C. Reuter, M. Copel, F.K. LeGoues, J.C. Bean: J. Appl. Phys. 83, 1931–1937 (1998)Google Scholar
  307. E.A. Stach, R. Hull, R.M. Tromp, F.M. Ross, M.C. Reuter, J.C. Bean: Philos. Mag. A 80, 2159–2200 (2000)Google Scholar
  308. A.I. Persson, M.W. Larsson, S. Stenstrom, B.J. Ohlsson, L. Samuelson, L.R. Wallenberg: Nat. Mater. 3, 677–681 (2004)Google Scholar
  309. F.M. Ross: Rep. Prog. Phys. 73, 114501–114522 (2010)Google Scholar
  310. Y.-C. Chou, F. Panciera, M.C. Reuter, E.A. Stach, F.M. Ross: Chem. Commun. 52, 5686–5689 (2016)Google Scholar
  311. A.D. Gamalski, J. Tersoff, E.A. Stach: Nano Lett. 16, 2283 (2016)Google Scholar
  312. F. Lenrick, M. Ek, K. Deppert, L. Samuelson, L.R. Wallenberg: Nano Res. 7, 1188–1194 (2014)Google Scholar
  313. Y. Wu, P. Yang: J. Am. Chem. Soc. 123, 3165 (2001)Google Scholar
  314. G. Zhou, J.C. Yang, F. Xu, J.A. Barnard, Z. Zhang: Mater. Res. Soc. Symp. Proc. 737, F6.3 (2002)Google Scholar
  315. E.A. Stach, P.J. Pauzauskie, T. Kuykendall, J. Goldberger, R. He, P. Yang: Nano Lett. 3, 867 (2003)Google Scholar
  316. F.M. Ross, J. Tersoff, M.C. Reuter: Phys. Rev. Lett. 95, 146104 (2005)Google Scholar
  317. J.B. Hannon, S. Kodambaka, F.M. Ross, R.M. Tromp: Nature 440, 69–71 (2006)Google Scholar
  318. C.-Y. Wen, M.C. Reuter, J. Bruley, J. Tersoff, S. Kodambaka, E.A. Stach, F.M. Ross: Science 326, 1247–1250 (2009)Google Scholar
  319. C.-Y. Wen, M.C. Reuter, D. Su, E.A. Stach, F.M. Ross: Nano Lett. 15, 1654–1659 (2015)Google Scholar
  320. F. Panciera, Y.-C. Chou, M.C. Reuter, D. Zakharov, E.A. Stach, S. Hofmann, F.M. Ross: Nat. Mater. 14, 820–825 (2015)Google Scholar
  321. D. Jacobsson, F. Panciera, J. Tersoff, M.C. Reuter, S. Lehmann, S. Hofmann, K.A. Dick, F.M. Ross: Nature 531, 317–322 (2016)Google Scholar
  322. F. Panciera, M.M. Norton, S.B. Alam, S. Hofmann, K. Mølhave, F.M. Ross: Nat. Commun. 7, 12271 (2016)Google Scholar
  323. A. Klamchuen, M. Suzuki, K. Nagashima, H. Yoshida, M. Kanai, F. Zhuge, Y. He, G. Meng, S. Kai, S. Takeda, T. Kawai, T. Yanagida: Nano Lett. 15, 6406–6412 (2015)Google Scholar
  324. G. Meng, T. Yanagida, H. Yoshida, K. Nagashima, M. Kanai, F. Zhuge, Y. He, A. Klamchuen, S. Rahong, X. Fang, S. Takeda, T. Kawai: Nanoscale 6, 7033–7038 (2014)Google Scholar
  325. F.M. Ross, M. Kammler, M.E. Walsh, M.C. Reuter: Microsc. Microanal. 10, 105–111 (2004)Google Scholar
  326. M. Kammler, R. Hull, M.C. Reuter, F.M. Ross: Appl. Phys. Lett. 82, 1093–1095 (2003)Google Scholar
  327. A. Portavoce, R. Hull, F.M. Ross: Nanoscale lateral control of Ge quantum dot nucleation sites on Si(001) using focused ion beam implantation. In: Lateral Alignment of Epitaxial Quantum Dots, Nanoscience and Technology, ed. by O.G. Schmidt (Springer, Berlin 2005) pp. 397–428Google Scholar
  328. M. Gherasimova, R. Hull, M.C. Reuter, F.M. Ross: Appl. Phys. Lett. 93, 023106 (2008)Google Scholar
  329. B.J. Kim, J. Tersoff, S. Kodambaka, M.C. Reuter, E.A. Stach, F.M. Ross: Science 322, 1070–1073 (2008)Google Scholar
  330. J. Cumings, E. Olsson, A.K. Petford-Long, Y. Zhu: MRS Bulletin 33(2), 101–106 (2008)Google Scholar
  331. Q. Chen, L.M. Peng: Ultramicroscopy 111, 948–954 (2011)Google Scholar
  332. S. Canepa, S.B. Alam, D.-T. Ngo, F.M. Ross, K. Mølhave: In situ TEM electrical measurements. In: Controlled Atmosphere Transmission Electron Microscopy, ed. by T.W. Hansen, J.B. Wagner (Springer, Cham 2016) pp. 281–300Google Scholar
  333. E.H. Darlington, U. Valdre: J. Phys. E 8, 321–324 (1975)Google Scholar
  334. A.C. Twitchett, R.E. Dunin-Borkowski, R.F. Broom, P.A. Midgley: J. Phys. Condens. Matter 16, 181–192 (2004)Google Scholar
  335. A.C. Twitchett, R.E. Dunin-Borkowski, P.A. Midgley: Phys. Rev. Lett. 88, 238302 (2002)Google Scholar
  336. A.C. Twitchett, R.E. Dunin-Borkowski, R.J. Hallifax, R.F. Broom, P.A. Midgley: Microsc. Microanal. 11, 66–78 (2009)Google Scholar
  337. K.D. Johnson, V.P. Dravid: Microsc. Microanal. 5, 428–436 (1999)Google Scholar
  338. K.D. Johnson, V.P. Dravid: Interface Sci. 8, 189–198 (2000)Google Scholar
  339. M.A. Verheijen, J.J.T.M. Donkers, J.F.P. Thomassen, J.J. van den Broek, R.A.F. van der Rijt, M.J.J. Dona, C.M. Smit: Rev. Sci. Instrum. 75, 426–429 (2004)Google Scholar
  340. W.E. King, K.L. Merkle, M. Meshii: Phys. Rev. B 23, 6319–6334 (1981)Google Scholar
  341. K. Haga, W.E. King, K.L. Merkle, M. Meshii: Nucl. Instrum. Methods Phys. Res. B 16, 134–142 (1986)Google Scholar
  342. J.T. Lau, J.A. Prybyla, S.K. Theiss: Appl. Phys. Lett. 76, 164–166 (2000)Google Scholar
  343. F.M. Ross, R. Hull, D. Bahnck, J.C. Bean, L.J. Peticolas, C.A. King: Appl. Phys. Lett. 62, 1426–1428 (1993)Google Scholar
  344. D.T. Mathes, R. Hull, K.D. Choquette, K.M. Geib, A.A. Allerman, J.K. Guenter, B. Hawkins, R.A. Hawthorne: Proceedings SPIE 4994, 67–82 (2003)Google Scholar
  345. Q. Jeangros, M. Duchamp, J. Werner, M. Kruth, R.E. Dunin-Borkowski, B. Niesen, C. Ballif, A. Hessler-Wyser: Nano Lett. 16, 7013–7018 (2016)Google Scholar
  346. Z. Xu, Y. Bando, L. Liu, W. Wang, X. Bai, D. Golberg: ACS Nano 5, 4401–4406 (2011)Google Scholar
  347. Y. Lu, C.A. Merchant, M. Drndic, A.T.C. Johnson: Nano Lett. 11, 5184–5188 (2011)Google Scholar
  348. J.A. Rodríguez-Manzo, Z.J. Qi, A. Crook, J.-H. Ahn, A.T.C. Johnson, M. Drndić: ACS Nano 10, 4004–4010 (2016)Google Scholar
  349. J.C. Doan, S. Lee, S.-H. Lee, P.A. Flinn, J.C. Bravman, T.N. Marieb: J. Appl. Phys. 89, 7797–7808 (2001)Google Scholar
  350. W.C. Shih, A.L. Green: Mater. Res. Soc. Symp. Proc. 356, 417 (1994)Google Scholar
  351. J.A. Prybyla, S.P. Riege, S.P. Grabowski, A.W. Hunt: Appl. Phys. Lett. 73, 1083–1085 (1998)Google Scholar
  352. S.P. Riege, J.A. Prybyla, A.W. Hunt: Appl. Phys. Lett. 69, 2367–2369 (1996)Google Scholar
  353. H.B. Heersche, G. Lientschnig, K. O'Neill, H.S.J. van der Zant, H.W. Zandbergen: Appl. Phys. Lett. 91, 072107 (2007)Google Scholar
  354. T. Kozlova, H.W. Zandbergen: Ultramicroscopy 158, 74–80 (2015)Google Scholar
  355. R.W. Vook: Mater. Chem. Phys. 36, 199–216 (1994)Google Scholar
  356. C.Y. Chang, R.W. Vook: Thin Solid Films 223, 23–30 (1993)Google Scholar
  357. T. Kozlova, M. Rudneva, H.W. Zandbergen: Nanotechnology 24, 505708 (2013)Google Scholar
  358. H. Okabayashi, H. Kitamura, M. Komatsu, H. Mori: Appl. Phys. Lett. 68, 1066–1068 (1996)Google Scholar
  359. H. Mori, H. Okabayashi, M. Komatsu: Thin Solid Films 300, 25–29 (1997)Google Scholar
  360. J. Nucci, S. Krämer, E. Arzt, C.A. Volkert: J. Mater. Res. 20, 1851–1859 (2005)Google Scholar
  361. Q. Liu, J.H. Wu, K. Xu, A. Lu, Y. Bando, D. Golberg, J.Q. Hu: Nano Lett. 15, 2809–2816 (2015)Google Scholar
  362. B.C. Regan, S. Aloni, R.O. Ritchie, U. Dahmen, A. Zettl: Nature 428, 924–927 (2004)Google Scholar
  363. D. Erts, A. Løhmus, R. Lohmus, H. Olin: Appl. Phys. A 72, S71–S74 (2001)Google Scholar
  364. J. Cumings, A. Zettl, M.R. McCartney, J.C.H. Spence: Phys. Rev. Lett. 88(056804), 1–4 (2002)Google Scholar
  365. Z.L. Wang, R.P. Gao, W.A. de Heer, P. Poncharal: Appl. Phys. Lett. 80, 856–858 (2002)Google Scholar
  366. T. Kuzumaki, Y. Horiike, T. Kizuka, T. Kona, C. Oshima, Y. Mitsuda: Diam. Relat. Mater. 13, 1907–1913 (2004)Google Scholar
  367. C.H. Jin, J.Y. Wang, M.S. Wang, J. Su, L.M. Peng: Carbon 43, 1026–1031 (2005)Google Scholar
  368. R. Gao, Z. Pan, Z.L. Wang: Appl. Phys. Lett. 78, 1757–1759 (2001)Google Scholar
  369. X.D. Bai, E.G. Wang, P.X. Gao, Z.L. Wang: Nano Lett. 3, 1147–1150 (2003)Google Scholar
  370. Z. Xu, X.D. Bai, E.G. Wang, Z.L. Wang: Appl. Phys. Lett. 87, 163106 (2005)Google Scholar
  371. Z. Xu, X.D. Bai, E.G. Wang, Z.L. Wang: J. Phys. Condens. Matter 17, L507–L512 (2005)Google Scholar
  372. J. Yamashita, H. Hirayama, Y. Ohshima, K. Takayanagi: Appl. Phys. Lett. 74, 2450–2452 (1999)Google Scholar
  373. M. Kociak, K. Suenaga, K. Hirahara, Y. Saito, T. Nakahira, S. Iijima: Phys. Rev. Lett. 89, 155501 (2002)Google Scholar
  374. P. Poncharal, C. Berger, Y. Yan, Z.L. Wang, W.A. de Heer: J. Phys. Chem. B 106, 12104–12118 (2002)Google Scholar
  375. Q. Liu, R. Zou, Y. Bando, D. Golberg, J.Q. Hu: Prog. Mater. Sci. 70, 1–49 (2015)Google Scholar
  376. J. Cumings, A. Zettl: Phys. Rev. Lett. 93(086801), 1–4 (2004)Google Scholar
  377. K.E. Moore, O. Cretu, M. Mitome, D. Golberg: Carbon 107, 225–232 (2016)Google Scholar
  378. J.Y. Huang, S. Chen, S.H. Jo, Z. Wang, D.X. Han, G. Chen, M.S. Dresselhaus, Z.F. Ren: Phys. Rev. Lett. 94, 236802 (2005)Google Scholar
  379. P.M.F.J. Costa, U.K. Gautam, Y. Bando, D. Golberg: Nat. Commun. 2, 421–426 (2011)Google Scholar
  380. M.W. Larsson, L.R. Wallenberg, A.I. Persson, L. Samuelson: Microsc. Microanal. 10, 41–46 (2004)Google Scholar
  381. P.M.F.J. Costa, D. Golberg, G. Shen, M. Mitome, Y. Bando: J. Mater. Sci. 43, 1460–1470 (2008)Google Scholar
  382. B. Liu, Y. Bando, M. Wang, D. Golberg: J. Nanosci. Nanotechnol. 10, 3945–3951 (2010)Google Scholar
  383. Z. Xu, D. Golberg, Y. Bando: Nano Lett. 9, 2251–2254 (2009)Google Scholar
  384. H. Ohnishi, Y. Kondo, K. Takayanagi: Nature 395, 780–783 (1998)Google Scholar
  385. P. Poncharal, S. Frank, Z.L. Wang, W.A. de Heer: Eur. Phys. J. D 9, 77–79 (1999)Google Scholar
  386. Z.L. Wang, R.P. Gao, Z.W. Pan, Z.R. Dai: Adv. Eng. Mater. 3, 657–661 (2001)Google Scholar
  387. C. Kallesøe, C.-Y. Wen, T.J. Booth, O. Hansen, P. Bøggild, F.M. Ross, K. Mølhave: Nano Lett. 12, 2965–2970 (2012)Google Scholar
  388. T. Kizuka, S. Uhmehara, S. Fujisawa: Jpn. J. Appl. Phys. 40(2), L71–L74 (2001)Google Scholar
  389. T. Kizuka, H. Ohmi, T. Sumi, K. Kumazawa, S. Deguchi, M. Naruse, S. Fujisawa, S. Sasaki, A. Yabe, Y. Enomoto: Jpn. J. Appl. Phys. 40, L170–L173 (2001)Google Scholar
  390. Y. Oshima, K. Mouri, H. Hirayama, K. Takayanagi: Surf. Sci. 531, 209–216 (2003)Google Scholar
  391. C.S. Bonifacio, T.B. Holland, K. van Benthem: Acta Mater. 63, 140–149 (2014)Google Scholar
  392. J.N. Chapman: J. Phys. D 17, 623–647 (1984)Google Scholar
  393. R.E. Dunin-Borkowski, M.R. McCartney, D.J. Smith, S.S.P. Parkin: Ultramicroscopy 74, 61–73 (1998)Google Scholar
  394. Y. Zhu, M. De Graef (Eds.): Magnetic Imaging and Its Applications to Materials (Academic Press, Cambridge 2001)Google Scholar
  395. L.A. Rodríguez, C. Magén, E. Snoeck, C. Gatel, L. Marín, L. Serrano-Ramón, J.L. Prieto, M. Muñozi, P.A. Algarabel, L. Morellon, J.M. De Teresa, M.R. Ibarra: Ultramicroscopy 134, 144–154 (2013)Google Scholar
  396. D.-T. Ngo, L.T. Kuhn: Nanosci. Nanotechnol. 7, 045001 (2016)Google Scholar
  397. T. Uhlig, M. Heumann, J. Zweck: Ultramicroscopy 94, 193–196 (2003)Google Scholar
  398. G. Yi, W.A.P. Nicholson, C.K. Lim, J.N. Chapman, S. McVitie, C.D.W. Wilkinson: Ultramicroscopy 99, 65–72 (2004)Google Scholar
  399. J.W. Lau, M.A. Schofield, Y. Zhu: Ultramicroscopy 107, 396–400 (2007)Google Scholar
  400. M. Arita, R. Tokuda, K. Hamada, Y. Takahashi: Mater. Trans. 55, 403–409 (2014)Google Scholar
  401. H.S. Park, Y.-G. Park, Y. Gao, D. Shindo, M. Inoue: J. Appl. Phys. 97, 33908 (2005)Google Scholar
  402. A.C.C. Yu, C.C.H. Lo, A.K. Petford-Long, D.C. Jiles, T. Miyazaki: J. Appl. Phys. 91, 780–784 (2002)Google Scholar
  403. D.M. Donnet, V.G. Lewis, J.N. Chapman, K. O'Grady, H.W. van Kesteren: J. Phys. D 26, 1741–1745 (1993)Google Scholar
  404. P. Thompson, O. Gutfleisch, J.N. Chapman, I.R. Harris: J. Phys. D 30, 1854–1860 (1997)Google Scholar
  405. V.V. Volkov, Y. Zhu: J. Magn. Magn. Mater. 214, 204–216 (2000)Google Scholar
  406. D.-T. Ngo, A. Hashimoto, S. McVitie, S. Saito, J.N. Chapman, M. Takahashi: J. Phys. D 44, 095001 (2011)Google Scholar
  407. V.V. Volkov, Y. Zhu, M. Malac: Philos. Mag. 84, 2607–2617 (2004)Google Scholar
  408. A. Kohn, C. Wang, A.K. Petford-Long, S.G. Wang, R.C.C. Ward: J. Appl. Phys. 103, 063918 (2008)Google Scholar
  409. A.C.C. Yu, A.K. Petford-Long, K. O'Grady, T. Miyazaki: J. Appl. Phys. 91, 5234–5239 (2002)Google Scholar
  410. C.K. Lim, J.N. Chapman, M. Rahman, A.B. Johnston, D.O. O'Donnell: J. Phys. D 35, 2344–2351 (2002)Google Scholar
  411. C.K. Lim, J.N. Chapman, M. Rahman, A.B. Johnston, D.O. O'Donnell: J. Appl. Phys. 95, 1510–1515 (2004)Google Scholar
  412. X. Portier, A.K. Petford-Long, T.C. Anthony, J.A. Brug: Appl. Phys. Lett. 75, 1290–1292 (1999)Google Scholar
  413. Y.G. Wang, A.K. Petford-Long: J. Appl. Phys. 92, 6699–6707 (2002)Google Scholar
  414. R.E. Dunin-Borkowski, M.R. McCartney, B. Kardynal, D.J. Smith: Appl. Phys. Lett. 75, 2641–2643 (1999)Google Scholar
  415. A.B. Johnston, J.N. Chapman, B. Khamsehpour, C.D.W. Wilkinson: J. Phys. D 29, 1419–1427 (1996)Google Scholar
  416. S. McVitie, J.N. Chapman: Microsc. Microanal. 3, 146–153 (1997)Google Scholar
  417. K.J. Kirk, J.N. Chapman, S. McVitie, P.R. Aitchison, C.D.W. Wilkinson: Appl. Phys. Lett. 75, 3683–3685 (1999)Google Scholar
  418. A. Schneider, H. Hoffmann, J. Zweck: J. Magn. Magn. Mater. 257, 1–10 (2003)Google Scholar
  419. M. Schneider, H. Hoffmann, S. Otto, T. Haug, J. Zweck: J. Appl. Phys. 92, 1466–1472 (2002)Google Scholar
  420. M. Schneider, H. Hoffmann, J. Zweck: Appl. Phys. Lett. 79, 3113–3115 (2001)Google Scholar
  421. X.X. Liu, J.N. Chapman, S. McVitie, C.D.W. Wilkinson: Appl. Phys. Lett. 84, 4406–4408 (2004)Google Scholar
  422. J.W. Lau, M. Beleggia, M.A. Schofield, G.F. Neumark, Y. Zhu: J. Appl. Phys. 97, 10E702 (2005)Google Scholar
  423. C. Phatak, A.K. Petford-Long, O. Heinonen: Phys. Rev. Lett. 108, 067205 (2012)Google Scholar
  424. X. Portier, A.K. Petford-Long: Appl. Phys. Lett. 76, 754–756 (2000)Google Scholar
  425. D. Petit, H.T. Zeng, J. Sampaio, E. Lewis, L. O'Brien, A.-V. Jausovec, D. Read, R.P. Cowburn, K.J. O’Shea, S. McVitie, J.N. Chapman: Appl. Phys. Lett. 97, 233102 (2010)Google Scholar
  426. M. Tanase, A.K. Petford-Long, O. Heinonen, K.S. Buchanan, J. Sort, J. Nogués: Phys. Rev. B 79, 014436 (2009)Google Scholar
  427. M. Tanase, A.K. Petford-Long: Microsc. Res. Tech. 72, 187–196 (2009)Google Scholar
  428. S. Zhang, A.K. Petford-Long, O. Heinonen, C. Phatak: Appl. Phys. Lett. 105, 212409 (2014)Google Scholar
  429. A.Y. Toporov, R.M. Langford, A.K. Petford-Long: Appl. Phys. Lett. 77, 3063–3065 (2000)Google Scholar
  430. V. Brajuskovic, F. Barrows, C. Phatak, A.K. Petford-Long: Sci. Rep. 6, 34384 (2016)Google Scholar
  431. S. Zhang, A.K. Petford-Long, C. Phatak: Sci. Rep. 6, 31248 (2016)Google Scholar
  432. M.C. Hickey, D.-T. Ngo, S. Lepadatu, D. Atkinson, D. McGrouther, S. McVitie, C.H. Marrows: Appl. Phys. Lett. 97, 202505 (2010)Google Scholar
  433. E.M. Hempe, M. Kläui, T. Kasama, D. Backes, F. Junginger, S. Krzyk, L.J. Heyderman, R. Dunin-Borkowski, U. Rudiger: Phys. Status Solidi A 204, 3922–3928 (2007)Google Scholar
  434. X. Portier, A.K. Petford-Long, R.C. Doole, T.C. Anthony, J.A. Brug: Appl. Phys. Lett. 71, 2032–2034 (1997)Google Scholar
  435. X. Portier, E.Y. Tsymbal, A.K. Petford-Long, T.C. Anthony, J.A. Brug: Phys. Rev. B 58, R591–R594 (1998)Google Scholar
  436. X. Portier, A.K. Petford-Long, T.C. Anthony, J.A. Brug: J. Appl. Phys. 85, 4120–4126 (1999)Google Scholar
  437. T. Haug, K. Perzlmaier, C.H. Back: Phys. Rev. B 79, 024414 (2009)Google Scholar
  438. T. Asaka, Y. Anan, T. Nagai, S. Tsutsumi, H. Kuwahara, K. Kimoto, Y. Tokura, Y. Matsui: Phys. Rev. Lett. 89, 207203 (2002)Google Scholar
  439. J.Q. He, V.V. Volkov, M. Beleggia, T. Asaka, J. Tao, M.A. Schofield, Y. Zhu: Phys. Rev. B 81, 094427 (2010)Google Scholar
  440. J. Rajeswari, P. Huang, G.F. Mancini, Y. Murooka, T. Latychevskaia, D. McGrouther, M. Cantoni, E. Baldini, J.S. White, A. Magrez, T. Giamarchi, H.M. Rønnow, F. Carbone: Proc. Natl. Acad. Sci. 112, 14212–14217 (2015)Google Scholar
  441. A. Tonomura, X. Yu, K. Yanagisawa, T. Matsuda, Y. Onose, N. Kanazawa, H.S. Park, Y. Tokura: Nano Lett. 12, 1673–1677 (2012)Google Scholar
  442. N. Kawamoto, Y. Murakami, D. Shindo, K. Fukamichi: J. Appl. Phys. 100, 043913 (2006)Google Scholar
  443. Y. Murakami, J.H. Yoo, D. Shindo, T. Atou, M. Kikuchi: Nature 423, 965–968 (2003)Google Scholar
  444. J.H. Yoo, Y. Murakami, D. Shindo, T. Ohsuna: Philos. Mag. 84, 2597–2606 (2004)Google Scholar
  445. J.H. Yoo, Y. Murakami, D. Shindo, T. Atou, M. Kikuchi: Phys. Rev. Lett. 93, 047204 (2004)Google Scholar
  446. C. Phatak, A.K. Petford-Long, H. Zheng, J.F. Mitchell, S. Rosenkranz, M.R. Norman: Phys. Rev. B 92, 224418 (2015)Google Scholar
  447. H.S. Park, Y. Murakami, D. Shindo, V.A. Chernenko, T. Kanomata: Appl. Phys. Lett. 83, 3752–3754 (2003)Google Scholar
  448. C. Phatak, O. Heinonen, M. De Graef, A. Petford-Long: Nano Lett. 16, 4141–4148 (2016)Google Scholar
  449. A. Budruk, C. Phatak, A.K. Petford-Long, M. De Graef: Acta Mater. 59, 4895–4906 (2011)Google Scholar
  450. M. De Graef, M.A. Willard, M.E. McHenry, Y. Zhu: IEEE Trans. Magn. 37, 2663–2665 (2001)Google Scholar
  451. Y. Murakami, D. Shindo, K. Oikawa, R. Kainuma, K. Ishida: Acta Mater. 50, 2173–2184 (2002)Google Scholar
  452. A. Tonomura: J. Electron Micrsoc. 52, 11–19 (2003)Google Scholar
  453. J.C. Loudon, P.A. Midgley: Ultramicroscopy 109, 700–729 (2009)Google Scholar
  454. K. Harada, T. Matsuda, H. Kasai, O. Kamimura, A. Tonomura: Ultramicroscopy 67, 133–141 (1997)Google Scholar
  455. A. Tonomura, H. Kasai, O. Kamimura, T. Matsuda, K. Harada, T. Yoshida, T. Akashi, J. Shimoyama, K. Kishio, T. Hanaguri, K. Kitazawa, T. Masui, S. Tajima, N. Koshizuka, P.L. Gammel, D. Bishop, M. Sasase, S. Okayasu: Phys. Rev. Lett. 88, 237001 (2002)Google Scholar
  456. K. Harada, T. Matsuda, J. Bonevich, M. Igarashi, S. Kondo, G. Pozzi, U. Kawabe, A. Tonomura: Nature 360, 51–53 (1992)Google Scholar
  457. J.E. Bonevich, K. Harada, T. Matsuda, H. Kasai, T. Yoshida, G. Pozzi, A. Tonomura: Phys. Rev. Lett. 70, 2952–2955 (1993)Google Scholar
  458. A. Tonomura: J. Electron Microsc. 51(Suppl.), S3–S11 (2002)Google Scholar
  459. S. Horiuchi, M. Cantoni, M. Uchida, T. Tsuruta, Y. Matsui: Appl. Phys. Lett. 73, 1293–1294 (1998)Google Scholar
  460. T. Matsuda, K. Harada, H. Kasai, O. Kamimura, A. Tonomura: Science 271, 1393–1395 (1996)Google Scholar
  461. K. Harada, O. Kamimura, H. Kasai, F. Matsuda, A. Tonomura, V.V. Moshchalkov: Science 274, 1167–1170 (1996)Google Scholar
  462. K. Harada, H. Kasai, T. Matsuda, M. Yamasaki, A. Tonomura: J. Electron Microsc. 46, 227–232 (1997)Google Scholar
  463. C.-H. Sow, K. Harada, A. Tonomura, G. Crabtree, D.G. Grier: Phys. Rev. Lett. 80, 2693–2696 (1998)Google Scholar
  464. Y. Togawa, K. Harada, T. Akashi, H. Kasai, T. Matsuda, F. Nori, A. Maeda, A. Tonomura: Phys. Rev. Lett. 95, 87002 (2005)Google Scholar
  465. K. Harada, H. Kasai, O. Kamimura, T. Matsuda, A. Tonomura, S. Okayasu, Y. Kazumata: Phys. Rev. B 53, 9400–9405 (1996)Google Scholar
  466. Z. Akase, H. Kasai, S. Mamishin, D. Shindo, M. Morita, A. Tonomura: J. Appl. Phys. 111, 033912 (2012)Google Scholar
  467. A. Tonomura, H. Kasai, O. Kamimura, T. Matsuda, K. Harada, Y. Nakayama, J. Shimoyama, K. Kishio, T. Hanaguri, K. Kitazawa, M. Sasase, S. Okayasu: Nature 412, 620–622 (2001)Google Scholar
  468. T. Matsuda, O. Kamimura, H. Kasai, K. Harada, T. Yoshida, T. Akashi, A. Tonomura, Y. Nakayama, J. Shimoyama, K. Kishio, T. Hanaguri, K. Kitazawa: Science 294, 2136–2138 (2001)Google Scholar
  469. A. Krishnan, M.M.J. Treacy, M.E. Bisher, P. Chandra, P.B. Littlewood: Am. Inst. Phys. Conf. Proc. 535, 191–200 (2000)Google Scholar
  470. H.P. Xu, Q. Li, D. Feng, G.J. Shen, Q.C. Guan, J.Y. Wang: Ferroelectrics 146, 137–144 (1993)Google Scholar
  471. F.-T. Huang, F. Xue, B. Gao, L.H. Wang, X. Luo, W. Cai, X.-Z. Lu, J.M. Rondinelli, L.Q. Chen, S.-W. Cheong: Nat. Commun. 7, 11602 (2016)Google Scholar
  472. E. Snoeck, L. Normand, A. Thorel, C. Roucau: Phase Trans. A46, 77–88 (1994)Google Scholar
  473. S.B. Ren, C.J. Lu, H.M. Shen, Y.N. Wang: Mater. Res. Soc. Symp. Proc. 404, 95–100 (1995)Google Scholar
  474. A. Krishnan, M.E. Bisher, M.M.J. Treacy: Mater. Res. Soc. Symp. Proc. 541, 475–480 (1999)Google Scholar
  475. A. Krishnan, M.M.J. Treacy, M.E. Bisher, P. Chandra, P.B. Littlewood: Integr. Ferroelectr. 43, 31–49 (2002)Google Scholar
  476. M.-G. Han, M.S.J. Marshall, L. Wu, M.S. Schofield, T. Aoki, R. Twesten, J. Hoffman, F.J. Walter, C.H. Ahn, Y. Zhu: Nat. Commun. 5, 4693 (2014)Google Scholar
  477. C.R. Winkler, A.R. Damodaran, J. Karthik, L.W. Martin, M.L. Taheri: Micron 43, 1121–1126 (2012)Google Scholar
  478. H. Chang, S.V. Kalinin, S. Yang, P. Yu, S. Bhattacharya, P.P. Wu, N. Balke, S. Jesse, L.Q. Chen, R. Ramesh, S.J. Pennycook, A.Y. Borisevich: J. Appl. Phys. 110, 052014 (2011)Google Scholar
  479. C.T. Nelson, P. Gao, J.R. Jokisaari, C. Heikes, C. Adamo, A. Melville, S.H. Baek, C.M. Folkman, B. Winchester, Y. Gu, Y. Liu, K. Zhang, E. Wang, J. Li, L.Q. Chen, C.B. Eom, D.G. Schlom, X. Pan: Science 334(6058), 968–971 (2011)Google Scholar
  480. H. He, X. Tan: Appl. Phys. Lett. 85, 3187–3189 (2004)Google Scholar
  481. H. He, X. Tan: Phys. Rev. B 72, 24102 (2005)Google Scholar
  482. X.-Q. Pan, M.-S. Hu, M.-H. Yao, D. Feng: Phys. Status Solidi A 92, 57–68 (1985)Google Scholar
  483. S. Mori, N. Yamamoto, Y. Koyama, Y. Uesu: Phys. Rev. B 55, 11212–11217 (1997)Google Scholar
  484. M. Verwerft, G. Van Tendeloo, J. Van Landuyt, S. Amelinckx: Ferroelectrics 88, 27–36 (1988)Google Scholar
  485. S. Barre, H. Mutka, C. Roucau, A. Litzler, J. Schneck, J.C. Toledano, S. Bouffard, F. Rullier-Albenque: Phys. Rev. B 43, 11154–11161 (1991)Google Scholar
  486. X. Lin, C. Murray, V.P. Dravid: Statics and dynamics of charged interfaces in electroceramics. In: Microscopy and Microanalysis, ed. by G.W. Bailey, et al. (Springer, New York 1998) pp. 552–553Google Scholar
  487. Z. Xu, X. Tan, P. Han, J.K. Shang: Appl. Phys. Lett. 76, 3732–3734 (2000)Google Scholar
  488. X. Tan, Z. Xu, J.K. Shang, P. Han: Appl. Phys. Lett. 77, 1529–1531 (2000)Google Scholar
  489. X.L. Tan, H. He, J.K. Shang: J. Mater. Res. 20, 1641–1653 (2005)Google Scholar
  490. X. Tan, J.K. Shang: J. Appl. Phys. 96, 2805–2810 (2004)Google Scholar
  491. X. Tan, J.K. Shang: J. Appl. Phys. 95, 635–639 (2004)Google Scholar
  492. B. Baufeld, D. Baither, U. Messerschmidt, M. Bartsch, A.H. Foitzik, M. Ruhle: J. Am. Ceram. Soc. 80, 1699–1705 (1997)Google Scholar
  493. J.X. Zhang, B. Xiang, Q. He, J. Seidel, R.J. Zeches, P. Yu, S.Y. Yang, C.H. Wang, Y.H. Chu, L.W. Martin, A.M. Minor, R. Ramesh: Nat. Nanotechnol. 6, 98–102 (2011)Google Scholar
  494. P. Gao, C. Nelson, J. Jokisaari, J. Britson, C. Duan, M. Trassin, S.H. Baek, H. Guo, L. Li, Y. Wang, A.M. Minor, C.B. Eom, R. Ramesh, L.Q. Chen, X. Pan: Nat. Commun. 5, 4801 (2014)Google Scholar
  495. B. Xiang, D.J. Hwang, J.B. In, S.B. Ryu, J.H. Yoo, O. Dubon, A.M. Minor, C.P. Grigoropoulos: Nano Lett. 12, 2524–2529 (2012)Google Scholar
  496. C. Zhang, T. Wei, Z. Xu, X. Wang, J. Liu, S. Li, T. Daiming, D. Liu, L. Meiyong, Y. Bando, D. Golberg: Nanoscale 6, 8084–8090 (2014)Google Scholar
  497. Y. Ohno: Appl. Phys. Express 5, 125204 (2012)Google Scholar
  498. M. Picher, S. Mazzucco, S. Blankenship, R. Sharma: Ultramicroscopy 150, 10–15 (2015)Google Scholar
  499. F.I. Allen, E. Kim, N.C. Andresen, G.P. Grigoropoulos, A.M. Minor: Ultramicroscopy 178, 33–37 (2017)Google Scholar
  500. H.G.F. Wilsdorf: Rev. Sci. Instrum. 29, 323 (1958)Google Scholar
  501. E.P. Butler, K.F. Hale: Dynamic Experiments in the Electron Microscope, Practical Methods of Electron Microscopy, Vol. 9 (North-Holland, Amsterdam 1981)Google Scholar
  502. U. Messerschmidt, D. Baither, M. Bartsch, B. Baufeld, B. Geyer, S. Guder, A. Wasilkowska, A. Czyrska-Filemonowicz, M. Yamaguchi, M. Feuerbacher, K. Urban: Microsc. Microanal. 4, 226–234 (1998)Google Scholar
  503. M. Komatsu, H. Mori, K. Iwasaki: J. Am. Ceram. Soc. 77, 839–842 (1994)Google Scholar
  504. A.M. Minor, S.A. Syed Asif, Z.W. Shan, E.A. Stach, E. Cyrankowski, T.J. Wyrobek, O.L. Warren: Nat. Mater. 5, 697–702 (2006)Google Scholar
  505. O.L. Warren, Z. Shan, S.A. Syed Asif, E.A. Stach, J.W. Morris Jr., A.M. Minor: Mater. Today 10(4), 59–60 (2007)Google Scholar
  506. M. Legros: Small-scale plasticity. In: Mechanics of Nano-Objects, ed. by O. Thomas, A. Ponchet, S. Forest (Presse des Mines, Paris 2011)Google Scholar
  507. Q. Yu, M. Legros, A.M. Minor: MRS Bulletin 40, 62–68 (2015)Google Scholar
  508. U. Messerschmidt, D. Baither, B. Baufeld, M. Bartsch: Mater. Sci. Eng. A 233, 61–74 (1997)Google Scholar
  509. H. Mori: Mater. Trans. JIM 39, 959–966 (1998)Google Scholar
  510. F. Pettinari, A. Couret, D. Caillard, G. Molenat, N. Clement, A. Coujou: J. Microsc. 203, 47–56 (2001)Google Scholar
  511. G. Vanderschaeve, C. Levade, D. Caillard: J. Microsc. 203, 72–83 (2001)Google Scholar
  512. U. Messerschmidt, M. Bartsch: Mater. Chem. Phys. 81, 518–523 (2003)Google Scholar
  513. S.H. Oh, M. Legros, D. Kiener, G. Dehm: Nat. Mater. 8, 95 (2009)Google Scholar
  514. M.E. Kassner, M.A. Wall, M.A. Delos-Reyes: Metall. Mater. Trans. A 28, 595–609 (1997)Google Scholar
  515. M. Komatsu, Y. Matsukawa, K. Yasunaga, M. Kiritani: Mater. Sci. Eng. A 350, 25–29 (2003)Google Scholar
  516. Q. Yu, Z.-W. Shan, J. Li, X. Huang, L. Xiao, J. Sun, E. Ma: Nature 463(18), 335–338 (2010)Google Scholar
  517. Q. Yu, L. Qi, K. Chen, R.K. Mishra, J. Li, A.M. Minor: Nano Lett. 12, 887–892 (2012)Google Scholar
  518. Q. Yu, L. Qi, R.K. Mishra, J. Li, A.M. Minor: Proc. Natl. Acad. Sci. 110(33), 13289–13293 (2013)Google Scholar
  519. J. Gemperlova, A. Jacques, A. Gemperle, T. Vystavel, N. Zarubova, M. Janecek: Mater. Sci. Eng. A 324, 183–189 (2002)Google Scholar
  520. F. Lancon, J. Ye, D. Caliste, T. Radetic, A.M. Minor, U. Dahmen: Nano Lett. 10, 695–700 (2010)Google Scholar
  521. R.C. Hugo, H. Kung, J.R. Weertman, R. Mitra, J.A. Knapp, D.M. Follstaedt: Acta Mater. 51, 1937–1943 (2003)Google Scholar
  522. Z. Shan, E.A. Stach, J.M.K. Wiezorek, J.A. Knapp, D.M. Forstaedt, S.X. Mao: Science 305, 654–657 (2004)Google Scholar
  523. G. Dehm, E. Arzt: Appl. Phys. Lett. 77, 1126–1128 (2000)Google Scholar
  524. T.J. Balk, G. Dehm, E. Arzt: Acta Mater. 51, 4471–4485 (2003)Google Scholar
  525. R.-M. Keller-Flaig, M. Legros, W. Sigle, A. Gouldstone, K.J. Hemker, S. Suresh, E. Arzt: J. Mater. Res. 14, 4673–4676 (1999)Google Scholar
  526. M. Legros, K.J. Hemker, A. Gouldstone, S. Suresh, R.M. Keller-Flaig, E. Arzt: Acta Mater. 50, 3435–3452 (2002)Google Scholar
  527. D. Häussler, M. Bartsch, M. Aindow, I.P. Jones, U. Messerschmidt: Philos. Mag. A 79, 1045–1071 (1999)Google Scholar
  528. M. Legros, A. Couret, D. Caillard: Philos. Mag. 73, 61–99 (1996)Google Scholar
  529. M. Legros, Y. Minonishi, D. Caillard: Philos. Mag. A 76, 995–1032 (1997)Google Scholar
  530. M. Legros, D. Caillard: J. Microsc. 203, 90–98 (2001)Google Scholar
  531. J. Malaplate, D. Caillard, A. Couret: Philos. Mag. 84, 3671–3687 (2004)Google Scholar
  532. S. Zghal, A. Coujou, A. Couret: Philos. Mag. A 81, 345 (2001)Google Scholar
  533. S.G. Pyo, N.J. Kim: J. Mater. Res. 20, 1888–1901 (2005)Google Scholar
  534. D. Caillard, C. Vailhe, D. Farkas: Philos. Mag. A 79, 723–739 (1999)Google Scholar
  535. S. Watanabe, Y. Haishi, T. Suda, S. Ohnuki, H. Takahashi, M. Kiritani: Radiat. Eff. Defects Solids 157, 101–108 (2002)Google Scholar
  536. U. Messerschmidt, M. Bartsch, M. Feuerbacher, B. Geyer, K. Urban: Philos. Mag. A 79, 2123–2135 (1999)Google Scholar
  537. U. Messerschmidt: J. Microsc. 203, 68–71 (2001)Google Scholar
  538. D. Caillard, C. Roucau, L. Bresson, D. Gratias: Acta Mater. 50, 4499–4509 (2002)Google Scholar
  539. F. Mompiou, D. Caillard, M. Feuerbacher: Philos. Mag. 84, 2777 (2004)Google Scholar
  540. M. Bartsch, P. Schall, M. Feuerbacher, U. Messerschmidt: J. Mater. Res. 20, 1814–1824 (2005)Google Scholar
  541. S. Guder, M. Bartsch, U. Messerschmidt: Philos. Mag. A 82, 2737–2754 (2002)Google Scholar
  542. E. Snoeck, C. Roucau: Phys. Rev. B 45, 12720–12724 (1992)Google Scholar
  543. E. Tochigi, E. Zepeda-alarcon, H.-R. Wenk, A.M. Minor: Phys. Chem. Miner. 41, 757–765 (2014)Google Scholar
  544. H. Gottschalk, N. Hiller, S. Sauerland, P. Specht, H. Alexander: Phys. Status Solidi A 138, 547–555 (1993)Google Scholar
  545. G. Vanderschaeve, C. Levade, D. Caillard: J. Phys. Condens. Matter 12, 10093–10103 (2000)Google Scholar
  546. T. Kruml, D. Caillard, C. Dupas, J.-L. Martin: J. Phys. Condens. Matter 14, 12897–12902 (2002)Google Scholar
  547. H.R. Kolar, J.C.H. Spence, H. Alexander: Phys. Rev. Lett. 77, 4031–4034 (1996)Google Scholar
  548. J.C.H. Spence, H.R. Kolar, H.R. Hembree, C.J. Humphreys, J. Barnard, R. Datta, C. Koch, F.M. Ross, J.F. Justo: Philos. Mag. 86, 4781–4796 (2006)Google Scholar
  549. J.S. Vetrano, S.M. Bruemmer, L.M. Pawlowski, I.M. Robertson: Mater. Sci. Eng. A 238, 101–107 (1997)Google Scholar
  550. R. Haitenhauer, M. Bartsch, U. Messerschmidt, P. Haasen, P.-J. Wilbrandt: Philos. Mag. A 70, 447–461 (1994)Google Scholar
  551. J.S. Vetrano, S.M. Bruemmer, I.M. Robertson: Mater. Res. Soc. Symp. Proc. 404, 177–182 (1995)Google Scholar
  552. L.M. Dougherty, I.M. Robertson, J.S. Vetrano: Acta Mater. 51, 4367–4378 (2003)Google Scholar
  553. D. Caillard, J.L. Martin, B. Jouffrey: Acta Metall. 28, 1059–1067 (1980)Google Scholar
  554. B. Verhaeghe, F. Louchet, B. Doisneau-Cottignies, Y. Brechet, J.-P. Massoud: Philos. Mag. A 76, 1079–1091 (1997)Google Scholar
  555. M. Janecek, F. Louchet, B. Doisneau-Cottignies, Y. Brechet: Philos. Mag. A 80, 1605–1619 (2000)Google Scholar
  556. W. Zielinski, W. Swiatnicki, M. Barstch, U. Messerschmidt: Mater. Chem. Phys. 81, 476–479 (2003)Google Scholar
  557. L. Thilly, M. Veron, O. Ludwig, F. Lecouturier, J.P. Peyrade, S. Askenazy: Philos. Mag. A 82, 925–942 (2002)Google Scholar
  558. F. Louchet, J.-J. Blandin, M. Veron: Adv. Eng. Mater. 3, 608–612 (2001)Google Scholar
  559. M.A. Wall, T.W. Barbee, T.P. Weihs: Mater. Res. Soc. Symp. Proc. 404, 43–48 (1995)Google Scholar
  560. M.A. Wall, T.W. Barbee Jr.: Mater. Res. Soc. Symp. Proc. 480, 245–250 (1997)Google Scholar
  561. X. Tan, T. Du, J.K. Shang: Appl. Phys. Lett. 80, 3946–3948 (2002)Google Scholar
  562. N. Li, J. Wang, J.Y. Huang, A. Misra, X. Zhang: Scr. Mater. 63(4), 363–366 (2010)Google Scholar
  563. Z.W. Shan, J. Li, Y.Q. Cheng, A.M. Minor, S.A. Syed Asif, O.L. Warren, E. Ma: Phys. Rev. B 77, 155419–155416 (2008)Google Scholar
  564. J.T.M. De Hosson: Microsc. Res. Tech. 72, 250–260 (2009)Google Scholar
  565. H. Guo, P.F. Yan, Y.B. Wang, J. Tan, Z.F. Zhang, M.L. Sui, E. Ma: Nat. Mater. 6, 735–739 (2007)Google Scholar
  566. I.M. Robertson, D. Teter: Microsc. Res. Tech. 42, 260–269 (1998)Google Scholar
  567. D.F. Teter, I.M. Robertson, H.K. Birnbaum: Acta Mater. 49, 4313–4323 (2001)Google Scholar
  568. P. Sofronis, I.M. Robertson: Philos. Mag. A 82, 3405–3413 (2002)Google Scholar
  569. K. Maeda, K. Suzuki, Y. Yamashita, Y. Mera: J. Phys. Condens. Matter 12, 10079–10091 (2000)Google Scholar
  570. R. Hull, J.C. Bean, D. Bahnck, L.J. Peticolas Jr., K.T. Short, F.C. Unterwald: J. Appl. Phys. 70, 2052–2065 (1991)Google Scholar
  571. R. Hull, J.C. Bean, C. Buescher: J. Appl. Phys. 66, 5837–5843 (1989)Google Scholar
  572. D.M. Follstaedt, S.M. Myers, J.A. Floro, S.R. Lee: Nucl. Instrum. Methods Phys. Res. B 127/128, 375–378 (1997)Google Scholar
  573. N. Hueging, M. Luysberg, K. Urban, D. Buca, S. Mantl: Appl. Phys. Lett. 86(3), 42112–42111 (2005)Google Scholar
  574. H.P. Sun, X.Q. Pan, J.H. Haeni, D.G. Schlom: Appl. Phys. Lett. 85, 1967–1969 (2004)Google Scholar
  575. S. Lavagne, C. Levade, G. Vanderschaeve: Proc. Microsc. Semicond. Mater. 12, 219–224 (2001)Google Scholar
  576. G. Dehm, T. Wagner, T.J. Balk, E. Arzt, B.J. Inkson: J. Mater. Sci. Technol. 18, 113–117 (2002)Google Scholar
  577. X.H. Liu, F.M. Ross, K.W. Schwarz: Phys. Rev. Lett. 85, 4088–4091 (2000)Google Scholar
  578. M. Kammler, D. Chidambarrao, K.W. Schwarz, C.T. Black, F.M. Ross: Appl. Phys. Lett. 87, 133116 (2005)Google Scholar
  579. T. Ohmura, A.M. Minor, E.A. Stach, J.W. Morris Jr.: J. Mater. Res. 19, 3626–3632 (2004)Google Scholar
  580. N. Gane: Proc. Royal Soc. A 317, 367–391 (1970)Google Scholar
  581. M.A. Wall, U. Dahmen: Microsc. Res. Tech. 42, 248–254 (1998)Google Scholar
  582. E.A. Stach, T. Freeman, A.M. Minor, D.K. Owen, J. Cumings, M.A. Wall, T. Chraska, R. Hull, J.W. Morris Jr., A. Zettl, U. Dahmen: Microsc. Microanal. 7, 507–517 (2001)Google Scholar
  583. A.M. Minor, J.W. Morris, E.A. Stach: Appl. Phys. Lett. 79, 1625–1627 (2001)Google Scholar
  584. M.S. Bobji, C.S. Ramanujan, R.C. Doole, J.B. Pethica, B.J. Inkson: Mater. Res. Soc. Symp. Proc. 778, 105–110 (2003)Google Scholar
  585. S. Ii, C. Iwamoto, K. Matsunaga, T. Yamamoto, M. Yoshiya, Y. Ikuhara: Philos. Mag. 84, 2767–2775 (2004)Google Scholar
  586. A.M. Minor, E.T. Lilleodden, M. Jin, E.A. Stach, D.C. Chrzan, J.W. Morris: Philos. Mag. 85, S323–S330 (2005)Google Scholar
  587. M. Jin, A.M. Minor, E.A. Stach, J.W. Morris Jr.: Acta Mater. 52, 5381–5387 (2004)Google Scholar
  588. W.A. Soer, J.T.M. De Hosson, A.M. Minor, J.W. Morris Jr., E.A. Stach: Acta Mater. 52, 5783–5790 (2004)Google Scholar
  589. K. Matsunaga, S. Ii, C. Iwamoto, T. Yamamoto, Y. Ikuhara: Nanotechnology 15, S376–S381 (2004)Google Scholar
  590. S. Korte, W.J. Clegg: Acta Mater. 58(1), 59–66 (2010)Google Scholar
  591. A.P. Merkle, L.D. Marks: Appl. Phys. Lett. 90, 064101 (2007)Google Scholar
  592. T.D.B. Jacobs, R.W. Carpick: Nat. Nanotechnol. 8, 108–112 (2013)Google Scholar
  593. T. Kizuka, N. Tanaka, S. Deguchi, M. Naruse: Microsc. Microanal. 4, 218–225 (1998)Google Scholar
  594. T. Kizuka, K. Hosoki: Appl. Phys. Lett. 75, 2743–2745 (1999)Google Scholar
  595. Y. Naitoh, K. Takayanagi, Y. Oshima, H. Hirayama: J. Electron Microsc. 49, 211–216 (2000)Google Scholar
  596. T. Kizuka: Phys. Rev. Lett. 81(16), 4448–4451 (1998)Google Scholar
  597. T. Kizuka: Phys. Rev. B 57, 11158–11163 (1998)Google Scholar
  598. S. Fujisawa, T. Kizuka: Tribol. Lett. 15, 163 (2003)Google Scholar
  599. D. Kiener, A.M. Minor: Nano Lett. 11, 3816–3820 (2011)Google Scholar
  600. J.C.H. Spence, W. Lo, M. Kuwabara: Ultramicroscopy 33, 69–82 (1990)Google Scholar
  601. W. Lo, J. Spence: Ultramicroscopy 48, 433 (1992)Google Scholar
  602. S. Fujisawa, T. Kizuka: Jpn. J. Appl. Phys. 42(2), L1182 (2003)Google Scholar
  603. H. Ohnishi, Y. Kondo, K. Takayanagi: Surf. Sci. 415, L1061–4 (1998)Google Scholar
  604. X.D. Bai, P.X. Gao, Z.L. Wang, E.G. Wang: Appl. Phys. Lett. 82, 4806–4808 (2003)Google Scholar
  605. M.A. Haque, M.T.A. Saif: Exp. Mech. 43, 248–255 (2003)Google Scholar
  606. M.A. Haque, M.T.A. Saif: Proc. Natl. Acad. Sci. 101, 6335–6340 (2004)Google Scholar
  607. M.A. Haque, M.T.A. Saif: J. Mater. Res. 20, 1769–1777 (2005)Google Scholar
  608. K. Hattar, J. Han, M.T.A. Saif, I.M. Robertson: J. Mater. Res. 20, 1869–1877 (2005)Google Scholar
  609. Z.W. Shan, R.K. Mishra, S.A.S. Asif, O.L. Warren, A.M. Minor: Nat. Mater. 7(2), 115–119 (2008)Google Scholar
  610. J. Ye, R.K. Mishra, A.M. Minor: Scr. Mater. 59(9), 951–954 (2008)Google Scholar
  611. J. Cumings, A. Zettl: Science 289, 602–604 (2000)Google Scholar
  612. D. Kiener, P. Hosemann, S.A. Maloy, A.M. Minor: Nat. Mater. 10, 608–613 (2011)Google Scholar
  613. D. Kiener, A.M. Minor: Acta Mater. 59, 1328–1337 (2011)Google Scholar
  614. D. Kiener, P. Kaufmann, A.M. Minor: Adv. Eng. Mater. 14(11), 960–967 (2012)Google Scholar
  615. M.B. Lowry, D. Kiener, M.M. LeBlanc, C. Chisholm, J.N. Florando, J.W. Morris, A.M. Minor: Acta Mater. 58(15), 5160–5167 (2010)Google Scholar
  616. A. Krishnan, E. Dujardin, T.W. Ebbesen, P.N. Yianilos, M.M.J. Treacy: Phys. Rev. B 58, 14013–14019 (1998)Google Scholar
  617. Z.L. Wang, Z.R. Dai, R. Gao, J.L. Gole: J. Electron Microsc. 51(Suppl.), S79–S85 (2002)Google Scholar
  618. R. Gao, Z.L. Wang, Z. Bai, W.A. de Heer, L. Dai, M. Gao: Phys. Rev. Lett. 85, 622–625 (2000)Google Scholar
  619. Z.W. Shan, G. Adesso, A. Cabot, M.P. Sherburne, S.A. Syed Asif, O.L. Warren, D.C. Chrzan, A.M. Minor, A.P. Alivisatos: Nat. Mater. 7(12), 947–952 (2008)Google Scholar
  620. C. Chisholm, H. Bei, M. Lowry, J. Oh, S.A. Syed Asif, O.L. Warren, Z.W. Shan, E.P. George, A.M. Minor: Acta Mater. 60, 2258–2264 (2012)Google Scholar
  621. D. Tang, C. Ren, M. Wang, X. Wei, N. Kawamoto, C. Liu, Y. Bando, M. Mitome, N. Fukata, D. Golberg: Nano Lett. 12, 1898–1904 (2012)Google Scholar
  622. T. Kuzumaki, H. Sawada, H. Ichinose, Y. Horiike, T. Kizuka: Appl. Phys. Lett. 79, 4580–4582 (2001)Google Scholar
  623. K. Asaka, T. Kizuka: Phys. Rev. B 72, 115431 (2005)Google Scholar
  624. H. Hirayama, Y. Kawamoto, Y. Ohshima, K. Takayanagi: Appl. Phys. Lett. 79, 1169–1171 (2001)Google Scholar
  625. Y.J. Suh, S.V. Prikhodko, A.K. Friedlander: Microsc. Microanal. 8, 497 (2004)Google Scholar
  626. K. Suenaga, C. Colliex, S. Iijima: Appl. Phys. Lett. 78, 70–72 (2001)Google Scholar
  627. N. De Jonge, F.M. Ross: Nat. Nanotechnol. 6, 695–704 (2011)Google Scholar
  628. F.M. Ross: Science 350, aaa9886 (2015)Google Scholar
  629. E. Jensen, K.R. Molhave: Encapsulated liquid cells for transmission electron microscopy. In: Liquid Cell Electron Microscopy, ed. by F.M. Ross (Cambridge Univ. Press, Cambridge 2016) pp. 35–55Google Scholar
  630. J.M. Yuk, J. Park, P. Ercius, K. Kim, D.J. Hellebusch, M.F. Crommie, J.Y. Lee, A. Zettl, A.P. Alivisatos: Science 336, 61–64 (2012)Google Scholar
  631. J. Shah: Philos. Mag. 84, 2749–2765 (2004)Google Scholar
  632. A. Fukami, K. Fukushima, N. Kohyama: Observation technique for wet clay minerals using film-sealed environmental cell equipment attached to high-resolution electron microscope. In: Microstructure of Fine Grained Sediments from Mud to Shale, ed. by R.H. Bennett, W.R. Bryant, M.H. Hulbert (Springer, New York 1991) pp. 321–331Google Scholar
  633. T.L. Daulton, B.J. Little, K. Lowe, J. Jones-Meehan: Microsc. Microanal. 7, 470–485 (2001)Google Scholar
  634. K.H. Foo, W.-A. Chiou, A. Ishikawa, K. Fukushima, M. Meshii: Microsc. Microanal. 7(Suppl. 2), 1216–1217 (2001)Google Scholar
  635. W.-A. Chiou, A. Ishikawa, K.H. Foo, H. Miyata, H. Tsukada, K. Fukushima: In-situ study of small particles in liquid environment. In: Proc. ICEM15, Vol. 1 (2002) pp. 937–938Google Scholar
  636. H. Sugi, T. Akimoto, K. Sutoh, S. Chaen, N. Oishi, S. Suzuki: Proc. Natl. Acad. Sci. USA 94, 4378–4382 (1997)Google Scholar
  637. H. Sugi, S. Chaen, T. Akimoto, H. Minoda, T. Miyakawa, Y. Miyauchi, M. Tanokura, S. Sugiura: Sci. Rep. 5, 15700 (2015)Google Scholar
  638. M.E. Wise, G. Biskos, S.T. Martin, L.M. Russell, P.R. Buseck: Aerosol Sci. Technol. 39, 849–856 (2005)Google Scholar
  639. K. Yoshida, A.N. Bright, M.R. Ward, L. Lari, X. Zhang, T. Hiroyama, E.D. Boyes, P.L. Gai: Nanotechnology 25, 425702 (2014)Google Scholar
  640. P.L. Gai, M.A. Harmer: Nano Lett. 2, 771–774 (2002)Google Scholar
  641. A. Radisic, P.M. Vereecken, J.B. Hannon, P.C. Searson, F.M. Ross: Nano Lett. 6, 238–242 (2006)Google Scholar
  642. A.J. Leenheer, K.L. Jungjohann, K.R. Zavadil, J.P. Sullivan, C.T. Harris: ACS Nano 9, 4379–4389 (2015)Google Scholar
  643. N. Hodnik, G. Dehm, K.J. Mayrhofer: Acc. Chem. Res. 49, 2015–2022 (2016)Google Scholar
  644. M.J. Williamson, R.M. Tromp, P.M. Vereecken, R. Hull, F.M. Ross: Nat. Mater. 2, 532–536 (2003)Google Scholar
  645. M.E. Holtz, Y. Yu, D. Gunceler, J. Gao, R. Sundararaman, K.A. Schwarz, T.A. Arias, H.D. Abruña, D.A. Muller: Nano Lett. 14, 1453–1459 (2014)Google Scholar
  646. M. Gu, L.R. Parent, B.L. Mehdi, R.R. Unocic, M.T. McDowell, R.L. Sacci, W. Xu, J.G. Connell, P. Xu, P. Abellan, X. Chen, Y. Zhang, D.E. Perea, J.E. Evans, L.J. Lauhon, J.-G. Zhang, J. Liu, N.D. Browning, Y. Cui, I. Arsland, C.-M. Wang: Nano Lett. 13, 6106–6112 (2013)Google Scholar
  647. R.L. Sacci, N.J. Dudney, K.L. More, L.R. Parent, I. Arslan, N.D. Browning, R.R. Unocic: Chem. Commun. 50, 2104–2107 (2014)Google Scholar
  648. Z. Zeng, W.-I. Liang, H.-G. Liao, H.L. Xin, Y.-H. Chu, H. Zheng: Nano Lett. 14, 1745–1750 (2014)Google Scholar
  649. R.L. Sacci, J.M. Black, N. Balke, N.J. Dudney, K.L. More, R.R. Unocic: Nano Lett. 15, 2011–2018 (2015)Google Scholar
  650. E.R. White, S.B. Singer, V. Augustyn, W.A. Hubbard, M. Mecklenburg, B. Dunn, B.C. Regan: ACS Nano 6, 6308–6317 (2012)Google Scholar
  651. B.L. Mehdi, J. Qian, E. Nasybulin, C. Park, D.A. Welch, R. Faller, H. Mehta, W.A. Henderson, W. Xu, C.M. Wang, J.E. Evans, J. Liu, J.-G. Zhang, K.T. Mueller, N.D. Browning: Nano Lett. 15, 2168–2173 (2015)Google Scholar
  652. P. Abellan, B.L. Mehdi, L.R. Parent, M. Gu, C. Park, W. Xu, Y. Zhang, I. Arslan, J.-G. Zhang, C.-M. Wang, J.E. Evans, N.D. Browning: Nano Lett. 14, 1293–1299 (2014)Google Scholar
  653. K.-Y. Niu, J. Park, H. Zheng, A.P. Alivisatos: Nano Lett. 13, 5715–5719 (2013)Google Scholar
  654. J.M. Grogan, N.M. Schneider, F.M. Ross, H.H. Bau: Nano Lett. 14, 359–364 (2013)Google Scholar
  655. S.W. Chee, S.H. Pratt, K. Hattar, D. Duquette, F.M. Ross, R. Hull: Chem. Commun. 51, 168–171 (2015)Google Scholar
  656. X. Zhong, M.G. Burke, S. Schilling, S.J. Haigh, N.J. Zaluzec: Microsc. Microanal. 20(S3), 1514–1515 (2014)Google Scholar
  657. H.-G. Liao, K. Niu, H. Zheng: Chem. Commun. 49, 11720–11727 (2013)Google Scholar
  658. J.E. Evans, K.L. Jungjohann, N.D. Browning, I. Arslan: Nano Lett. 11(7), 2809–2813 (2011)Google Scholar
  659. H. Zheng, R.K. Smith, Y.W. Jun, C. Kisielowski, U. Dahmen, A.P. Alivisatos: Science 324, 1309–1312 (2009)Google Scholar
  660. T.J. Woehl, C. Park, J.E. Evans, I. Arslan, W.D. Ristenpart, N.D. Browning: Nano Lett. 14, 373–378 (2013)Google Scholar
  661. D. Alloyeau, W. Dachraoui, Y. Javed, H. Belkahla, G. Wang, H. Lecoq, S. Ammar, O. Ersen, A. Wisnet, F. Gazeau, C. Ricolleau: Nano Lett. 15(4), 2574 (2015)Google Scholar
  662. H.-G. Liao, D. Zherebetskyy, H. Xin, C. Czarnik, P. Ercius, H. Elmlund, M. Pan, L.-W. Wang, H. Zheng: Science 345, 916–919 (2014)Google Scholar
  663. T.J. Woehl, J.E. Evans, I. Arslan, W.D. Ristenpart, N.D. Browning: ACS Nano 6, 8599–8610 (2012)Google Scholar
  664. Y. Jiang, G. Zhu, F. Lin, H. Zhang, C. Jin, J. Yuan, D. Yang, Z. Zhang: Nano Lett. 14, 3761–3765 (2014)Google Scholar
  665. J.M. Grogan, L. Rotkina, H.H. Bau: Phys. Rev. E 83, 061405 (2011)Google Scholar
  666. J. Park, H. Zheng, W.C. Lee, P.L. Geissler, E. Rabani, A.P. Alivisatos: ACS Nano 6, 2078–2085 (2012)Google Scholar
  667. M.H. Nielsen, J.R.I. Lee, Q.N. Hu, T.Y.J. Han, J.J. De Yoreo: Faraday Discuss. 159, 105–121 (2012)Google Scholar
  668. L.R. Parent, D.B. Robinson, T.J. Woehl, W.D. Ristenpart, J.E. Evans, N.D. Browning, I. Arslan: ACS Nano 6, 3589–3596 (2012)Google Scholar
  669. D.S. Li, M.H. Nielsen, J.R.I. Lee, C. Frandsen, J.F. Banfield, J.J. De Yoreo: Science 336, 1014–1018 (2012)Google Scholar
  670. E.R. White, M. Mecklenburg, S.B. Singer, S. Aloni, B.C. Regan: Appl. Phys. Express 4, 055201 (2011)Google Scholar
  671. U. Mirsaidov, C.-D. Ohl, P. Matsudaira: Soft Matter 8, 7108–7111 (2012)Google Scholar
  672. U.M. Mirsaidov, H. Zheng, D. Bhattacharya, Y. Casana, P. Matsudaira: Proc. Natl. Acad. Sci. 109, 7187–7190 (2012)Google Scholar
  673. D. Bhattacharya, M. Bosman, V.R.S.S. Mokkapati, F.Y. Leong, U. Mirsaidov: Microsc. Microanal. 20, 407–415 (2014)Google Scholar
  674. D. Mattia, Y. Gogotsi: Microfluid. Nanofluid. 5, 289–305 (2008)Google Scholar
  675. H. Zheng, S.A. Claridge, A.M. Minor, A.P. Alivisatos, U. Dahmen: Nano Lett. 9, 2460–2465 (2009)Google Scholar
  676. A. Verch, M. Pfaff, N. de Jonge: Langmuir 31, 6956–6964 (2015)Google Scholar
  677. J. Lu, Z. Aabdin, N.D. Loh, D. Bhattacharya, U. Mirsaidov: Nano Lett. 14, 2111–2115 (2014)Google Scholar
  678. K. Tai, Y. Liu, S.J. Dillon: Microsc. Microanal. 20, 330–337 (2014)Google Scholar
  679. N.M. Schneider, M.M. Norton, B.J. Mendel, J.M. Grogan, F.M. Ross, H.H. Bau: J. Phys. Chem. C 118, 22373–22382 (2014)Google Scholar
  680. R.C. Birtcher, M.A. Kirk, K. Furuya, G.R. Lumpkin, M.O. Ruault: J. Mater. Res. 20, 1654–1683 (2005)Google Scholar
  681. S. Jesse, A.Y. Borisevich, J.D. Fowlkes, A.R. Lupini, P.D. Rack, R.R. Unocic, B.G. Sumpter, S.V. Kalinin, A. Belianinov, O.S. Ovchinnikova: ACS Nano 10, 5600–5618 (2016)Google Scholar
  682. G.M. Bond, I.M. Robertson, H.K. Birnbaum: Scr. Metall. 20, 653–658 (1986)Google Scholar
  683. H. Yoshida, H. Omote, S. Takeda: Nanoscale 21, 13113–13118 (2014)Google Scholar
  684. M. Shimojo, K. Mitsuishi, M. Tanaka, M. Han, K. Furuya: J. Microsc. 214, 76–79 (2004)Google Scholar
  685. K. Furuya, K. Mitsuishi, M. Shimojo, M. Takeguchi: Rev. Adv. Mater. Sci. 5, 381–386 (2003)Google Scholar
  686. M. Tanaka, F. Chu, M. Shimojo, M. Takeguchi, K. Mitsuishi, K. Furuya: Appl. Phys. Lett. 86, 183104 (2005)Google Scholar
  687. M. Takeguchi, M. Shimojo, K. Furuya: Nanotechnology 16, 1321–1325 (2005)Google Scholar
  688. W.F. van Dorp: Appl. Phys. A 117, 1615–1622 (2014)Google Scholar
  689. R.F. Egerton, P. Li, M. Malac: Micron 35, 399–409 (2004)Google Scholar
  690. H. Guo, K.V. Wang, Y. Deng, Y. Oh, S.A. Syed Asif, O.L. Warren, Z.W. Shan, J. Wu, A.M. Minor: Appl. Phys. Lett. 102(23), 231909–231014 (2013)Google Scholar
  691. H. Guo, M.I. Khan, C. Cheng, W. Fan, C. Dames, J. Wu, A.M. Minor: Nat. Commun. 5, 4986 (2014)Google Scholar
  692. H. Inui, H. Mori, A. Suzuki, H. Fujita: Philos. Mag. B 65, 1–14 (1992)Google Scholar
  693. S. Takeda, J. Yamasaki: Phys. Rev. Lett. 83, 320–323 (1999)Google Scholar
  694. H. Yasuda, H. Mori: J. Electron Microsc. 48, 581–584 (1999)Google Scholar
  695. T.M. Smeeton, M.J. Kappers, J.S. Barnard, M.E. Vickers, C.J. Humphreys: Appl. Phys. Lett. 83, 5419–5421 (2003)Google Scholar
  696. C. Chen, Z. Wang, F. Lichtenberg, Y. Ikuhara, J.G. Bednorz: Nano Lett. 15, 6469–6474 (2015)Google Scholar
  697. D.K. Tappin, I.M. Robertson, H.K. Birnbaum: Phys. Rev. B 51, 14854–14860 (1995)Google Scholar
  698. I. Jencic, M.W. Bench, I.M. Robertson, M.A. Kirk: J. Appl. Phys. 78, 974–982 (1995)Google Scholar
  699. Y. Zhang, J. Lian, C.M. Wang, W. Jiang, R.C. Ewing, W.J. Weber: Phys. Rev. B 72, 094112 (2005)Google Scholar
  700. S. Jesse, Q. He, A.R. Lupini, D.N. Leonard, M.P. Oxley, O. Ovchinnikov, R.R. Unocic, A. Tselev, M. Fuentes-Cabrera, B.G. Sumpter, S.J. Pennycook, S.V. Kalinin, A.Y. Borisevich: Small 11, 5895–5900 (2015)Google Scholar
  701. X. Huang, T. Jones, H. Fan, M.-G. Willinger: Adv. Mater. Interfaces 3, 1600751 (2016)Google Scholar
  702. K. Sun, L.M. Wang, R.C. Ewing, W.J. Weber: Philos. Mag. 85, 597–608 (2005)Google Scholar
  703. S. Fujita, S. Maruno, H. Watanabe, M. Ichikawa: Appl. Phys. Lett. 69, 638–640 (1996)Google Scholar
  704. G.S. Chen, C.B. Boothroyd, C.J. Humphreys: Philos. Mag. A 78, 491–506 (1998)Google Scholar
  705. X.W. Du, M. Takeguchi, M. Tanaka, K. Furuya: Appl. Phys. Lett. 82, 1108–1110 (2003)Google Scholar
  706. Y. Mera, K. Suzuki, K. Maeda: Physica B 340–342, 488–491 (2003)Google Scholar
  707. D. Grozea, E. Landree, L.D. Marks: Appl. Phys. Lett. 71, 2301–2303 (1997)Google Scholar
  708. N. Jiang: J. Phys. D 46, 305502 (2013)Google Scholar
  709. A.W. Robertson, G.-D. Lee, K. He, Y. Fan, C.S. Allen, S. Lee, H. Kim, E. Yoon, H. Zheng, A.I. Kirkland, J.H. Warner: Nano Lett. 15, 5950–5955 (2015)Google Scholar
  710. A. Hashimoto, K. Suenaga, A. Gloter, K. Urita, S. Iijima: Nature 430, 870–873 (2004)Google Scholar
  711. J. Kotakoski, C. Mangler, J.C. Meyer: Nat. Commun. 5, 3991 (2014)Google Scholar
  712. K. Urita, K. Suenaga, T. Sugai, H. Shinohara, S. Iijima: Phys. Rev. Lett. 94, 155502 (2005)Google Scholar
  713. P.Y. Huang, S. Kurasch, J.S. Alden, A. Shekhawat, A.A. Alemi, P.L. McEuen, J.P. Sethna, U. Kaiser, D.A. Muller: Science 342, 2249227 (2013)Google Scholar
  714. A.L. Gibb, N. Alem, J.-H. Chen, K.J. Erickson, J. Ciston, A. Gautam, M. Linck, A. Zettl: J. Am. Chem. Soc. 135, 6758–6761 (2013)Google Scholar
  715. H.-P. Komsa, S. Kurasch, O. Lehtinen, U. Kaiser, A.V. Krasheninnikov: Phys. Rev. B 88, 035301 (2013)Google Scholar
  716. K. Urita, Y. Sato, K. Suenaga, A. Gloter, A. Hashimoto, M. Ishida, T. Shimada, H. Shinohara, S. Iijima: Nano Lett. 4, 2451–2454 (2004)Google Scholar
  717. T. Pham, A.L. Gibb, Z. Li, S.M. Gilbert, C. Song, S.G. Louie, A. Zettl: Nano Lett. 16, 7142–7147 (2016)Google Scholar
  718. S.D. Berger, I.G. Salisbury, R.H. Milne, D. Imeson, C.J. Humphreys: Philos. Mag. B 55, 341–358 (1987)Google Scholar
  719. C.A. Walsh: Philos. Mag. A 59, 227–246 (1989)Google Scholar
  720. Q. Xu, M.-Y. Wu, G.F. Schneider, L. Houben, S.K. Malladi, C. Dekker, E. Yucelen, R.E. Dunin-Borkowski, H.W. Zandbergen: ACS Nano 7, 1566–1572 (2013)Google Scholar
  721. X. Liu, T. Xu, X. Wu, Z. Zhang, J. Yu, H. Qiu, J.-H. Hong, C.-H. Jin, J.-X. Li, X.-R. Wang: Nat. Commun. 4, 1776 (2013)Google Scholar
  722. J.A. Rodríguez-Manzo, A.V. Krasheninnikov, F. Banhart: Chem. Phys. Chem. 13, 2596 (2012)Google Scholar
  723. L. Fedina, A. Gutakovskii, A. Aseev, J. van Landuyt, J. Vanhellemont: Philos. Mag. A 77, 423–435 (1998)Google Scholar
  724. M. Kiritani: Radiat. Eff. Defects Solids 148, 233–267 (1999)Google Scholar
  725. S. Arai, C. Morita, K. Arakawa, M. Kiritani: J. Electron Microsc. 44, 1–7 (1995)Google Scholar
  726. K. Arakawa, Y. Satoh, S. Arai, M. Kiritani: Philos. Mag. A 80, 2041–2055 (2000)Google Scholar
  727. S. Arai, Y. Satoh, K. Arakawa, C. Morita, M. Kiritani: J. Electron Microsc. 53, 21–27 (2004)Google Scholar
  728. M. Horiki, S. Arai, Y. Satoh, M. Kiritani: J. Nucl. Mater. 255, 165–173 (1998)Google Scholar
  729. M.A. Kirk, M.L. Jenkins, H. Fukushima: Mater. Res. Soc. Symp. Proc. 540, 521–526 (1999)Google Scholar
  730. A. Arakawa, S. Arai, H. Orihara, K. Ono, M. Kiritani: J. Electron Microsc. 51(Suppl.), S225–S229 (2002)Google Scholar
  731. J. Vanhellemont, A. Romano-Rodriguez, L. Fedina, J. Van Landuyt, A. Aseev: Mater. Sci. Techn. 11, 1194–1202 (1995)Google Scholar
  732. L. Fedina, O. Lebedev, G. Van Tendeloo, J. Van Landuyt: Proc. Microsc. Semicond. Mater. 10, 55–58 (1997)Google Scholar
  733. R.F. Klie, N. Browning: J. Electron Microsc. 50(Suppl.), 59 (2001)Google Scholar
  734. Y. Ding, Y. Chen, K.C. Pradel, M. Liu, Z.L. Wang: J. Appl. Phys. 120, 214302 (2016)Google Scholar
  735. H. Mori, H. Yasuda, H. Koyama, N. Sumida, H. Fujita: Nucl. Instrum. Methods Phys. Res. B 63, 291–296 (1992)Google Scholar
  736. H. Yasuda, H. Mori, T. Sakata, M. Naka, H. Fujita: Philos. Mag. B 65, 1001–1009 (1992)Google Scholar
  737. J.G. Lee, E. Taguchi, H. Mori: J. Electron Microsc. 51(Suppl.), S195–S200 (2002)Google Scholar
  738. C. Levade, A. Faress, G. Vanderschaeve: Philos. Mag. A 69, 855–870 (1994)Google Scholar
  739. M. Werner, E.R. Weber, M. Bartsch, U. Messerschmidt: Phys. Status Solidi A 150, 337–341 (1995)Google Scholar
  740. I. Yonenaga, M. Werner, M. Bartsch, U. Messerschmidt, E.R. Weber: Phys. Status Solidi A 171, 35–40 (1999)Google Scholar
  741. M. Inoue, K. Suzuki, H. Amasuga, Y. Mera, K. Maeda: J. Appl. Phys. 83, 1953–1957 (1998)Google Scholar
  742. I. Yonenaga, P.D. Brown, C.J. Humphreys: Mater. Sci. Eng. A 253, 148–150 (1998)Google Scholar
  743. K. Zheng, C. Wang, Y.-Q. Cheng, Y. Yue, X. Han, Z. Zhang, Z. Shan, S.X. Mao, M. Ye, Y. Yin, E. Ma: Nat. Commun. 1, 24 (2010)Google Scholar
  744. R.C. Birtcher, S.E. Donnelly, M. Song, K. Furuya, K. Mitsuishi, C.W. Allen: Phys. Rev. Lett. 83, 1617–1620 (1999)Google Scholar
  745. C.W. Allen, E.A. Ryan: Microsc. Res. Tech. 42, 255–259 (1998)Google Scholar
  746. M.O. Ruault, F. Fortuna, H. Bernas, J. Chaumont, O. Kaitasov, V.A. Borodin: J. Mater. Res. 20, 1758–1768 (2005)Google Scholar
  747. F. Chu, M. Song, K. Mitsuishi, H. Yasuda, K. Furuya: J. Electron Microsc. 51(Suppl.), S231–S234 (2002)Google Scholar
  748. K. Ono, K. Arakawa, K. Hojou, M. Oohasi, R.C. Birtcher, S.E. Donnelly: J. Electron Microsc. 51(Suppl.), S245–S251 (2002)Google Scholar
  749. T.L. Daulton, M.A. Kirk, L.E. Rehn: Philos. Mag. A 80, 809–842 (2000)Google Scholar
  750. M.O. Ruault, F. Fortuna, H. Bernas, M.C. Ridgway, J.S. Williams: Appl. Phys. Lett. 81, 2617–2619 (2002)Google Scholar
  751. Y. Chen, J. Li, K.Y. Yu, H. Wang, M.A. Kirk, M. Li, X. Zhang: Acta Mater. 111, 148–156 (2016)Google Scholar
  752. J. Li, C. Fan, J. Ding, S. Xue, Y. Chen, Q. Li, H. Wang, X. Zhang: Sci. Rep. 7, 39484 (2017)Google Scholar
  753. O. El-Atwani, J.E. Nathaniel II, A.C. Leff, B.R. Muntifering, J.K. Baldwin, K. Hattar, M.L. Taheri: J. Nucl. Mater. 484, 236–244 (2017)Google Scholar
  754. K. Hattar, D.C. Bufford, D.L. Buller: Nucl. Instrum. Methods Phys. Res. 338, 56–65 (2014)Google Scholar
  755. C.W. Allen, R.C. Birtcher, S.E. Donnelly, K. Furuya, N. Ishikawa, M. Song: Appl. Phys. Lett. 74, 2611–2613 (1999)Google Scholar
  756. M. Song, K. Mitsuishi, H. Yasuda, K. Furuya: J. Electron Microsc. 51(Suppl.), S211–S214 (2002)Google Scholar
  757. C.W. Allen, R.C. Birtcher, S.E. Donnelly, M. Song, K. Mitsuishi, K. Furuya, U. Dahmen: Philos. Mag. Lett. 83, 57–64 (2003)Google Scholar
  758. M. Tanaka, K. Furuya, T. Saito: Microsc. Microanal. 4, 207–217 (1998)Google Scholar
  759. P. Gnauck, W. Nisch, E. Plies: Cross-sectional investigation of the internal interface between biological cell tissue and implant materials in the transmission electron microscope (TEM). In: Proc. Int. Congr. Electron Microsc. 14, Vol. 3, ed. by H.A. Calderon Benavides, M.J. Yacaman (IEEE, Piscataway 1998) pp. 515–516Google Scholar
  760. T. Isshiki, K. Nishio, Y. Deguchi, H. Yamamoto, M. Shojiri: HRTEM observation of ceramics with a new heating stage coupled to radiation thermometer. In: Proc. Int. Congr. Electron Microsc. 14, Vol. 3, ed. by H.A. Calderon Benavides, M.J. Yacaman (1998) pp. 521–522Google Scholar
  761. C. Gammer, J. Kacher, C. Czarnik, O.L. Warren, J. Ciston, A.M. Minor: Appl. Phys. Lett. 109(8), 081906 (2016)Google Scholar
  762. H. Poppa: J. Vac. Sci. Technol. 2, 42–48 (1965)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Dept. of Materials Science and EngineeringMassachusetts Institute of TechnologyCambridge, MAUSA
  2. 2.University of California, Berkeley, and Lawrence Berkeley National LaboratoryBerkeley, CAUSA

Personalised recommendations