Advertisement

Ion Microscopy

  • Gregor HlawacekEmail author
Chapter
Part of the Springer Handbooks book series (SHB)

Abstract

Helium ion microscopy ( ) is a relatively young imaging and nanofabrication technique, which is based on a gas field ion source ( ). It rasters a narrow beam of helium ions across the surface of the specimen, to obtain high-resolution surface-sensitive images. Usually, secondary particles such as electrons are collected for image formation but also photons, backscattered atoms or sputtered sample atoms can be used for image formation. Thanks to the very high brightness of the source, a lateral resolution of \({\mathrm{0.5}}\,{\mathrm{nm}}\) can be achieved. The method is in particular suitable for obtaining high-resolution images of insulating samples (such as ceramic materials and biological samples) as the built-in charge compensation allows us to observe such specimens without any additional conductive coatings. In this chapter, I will introduce the method and briefly sketch the underlying physics. In the remainder of the chapter, a number of imaging modes will be discussed and selected examples will be presented. Finally, an outlook is presented on the ongoing efforts to add analytical capabilities to the method.

During the last century, several of the newly invented microscopy techniques were closely related to milestones in technology, and nanotechnology in particular. With regard to the related instruments based on charged particles, this list comprises transmission electron microscopy ( ), scanning electron microscopy ( ) and liquid metal ion source ( )-based focused ion beam ( ) instruments. In the recent past, helium ion microscopy (HIM) , based on a gas field ion source (GFIS) , was added to the already impressive list of enabling charged-particle microscopy and nanofabrication techniques.

Already in his seminal lecture [14.1], Richard P. Feynman speculated about the use of ions to reveal—and, more importantly, to modify—the smallest details in the atomic structure of matter. However, he was not the first one to play with this idea. Already in 1932, Knoll and Ruska put forward the idea of an ion microscope [14.2]. Although the earliest attempts actually reached back to the 1940s [14.3, 14.4, 14.5, 14.6], it was not until the 1970s that Levi-Seti [14.7, 14.8, 14.9] and Orloff [14.10] started to develop GFIS-based imaging techniques. However, they were quickly outperformed by other groups working with LMISs, notably gallium-based LMISs [14.11]. The latter became the quasi-standard in charged particle nanomachining and rapid prototyping at sub-micrometer-length scales during the past few decades.

However, at the beginning of the 21st century, a new and successful attempt was undertaken to realize a GFIS-based ion microscope. The technique was introduced in 2006 [14.12, 14.13] and quickly gained popularity. Although marketed as the helium ion microscope, it soon became clear that the technique also provides powerful nanomachining capabilities. This use case—in the spirit of Richard P. Feynman—received an additional boost with the introduction of neon as an alternative gas [14.14, 14.15]. Its higher mass made it a more powerful high-resolution alternative to helium when it comes to materials modification and removal.

In the following text, I would like to first discuss the basics of beam formation and the technical challenges associated with the formation and controlled manipulation of an ion beam with a diameter of less than \({\mathrm{0.5}}\,{\mathrm{nm}}\). This section will be concluded by a brief introduction of alternative-source technologies that go beyond Ga LMIS. After discussing the signals available in the HIM, I will present a number of use cases where the HIM excels.

In the past, a couple of review papers [14.16, 14.17] and monographs [14.18, 14.19] have been published discussing a wide range of use cases. The reader is referred to them, in particular for detailed information on nanoengineering applications.

14.1 Fundamentals of Beam Formation

14.1.1 The Gas Field Ion Source

While helium ion microscope [14.19] and atom probe tomography ( ) [14.20] share field ion microscopy ( ) [14.21] as a common ancestor, they utilize the work of Müller et al in very different ways. In the 1950s, Müller et al were the first to actually see atoms in real space. They placed a sharp metal needle in a vacuum vessel and applied a high voltage between the needle and a phosphor screen. The field strength at the apex of the metal tip is enhanced due to the geometry and therefore the resultant field becomes strong enough to ionize gas atoms and eventually atoms of the metallic needle. Subsequently, these ions are accelerated perpendicular to the local surface normal and towards the phosphor screen. The tip can be treated as a point source as long as the distance to the much larger screen is sufficiently long. When a homogeneous field distribution is achieved between the point-like source and the flat screen, a distortion-free magnification of several orders of magnitude can be achieved. Individual atoms that stick out of an otherwise atomically flat crystal terrace will yield a higher ion current because the field enhancement is stronger at such a protrusion. The same is true for atoms at the corners of a terrace, or to a lesser extent, for the edge atoms. Surface science quickly utilized the method to obtain detailed information on atomic processes at surfaces. See Fig. 14.1a,b for two examples, one historic and one recent.

Fig. 14.1a,b

FIM images of tungsten tips. (a) FIM image obtained by Müller et al [14.22] in the 1950s from a (110) tungsten tip with a radius of approximately \({\mathrm{94}}\,{\mathrm{nm}}\). (b) FIM image of a (111) tungsten tip with \({\mathrm{12}}\,{\mathrm{nm}}\) tip radius obtained in a modern system by Pitters et al reprinted from [14.23], with permission from Elsevier

The gas field ion source makes use of this technique to create narrow beamlets of ionized noble-gas atoms which originate from only a few atoms at the apex to the tip. Gas field ion sources have been the subject of scientific interest for a long time [14.24, 14.25, 14.26, 14.27], and many of these investigations have focused on the use in microscopy [14.10, 14.9] or nanomachining applications [14.28]. It is remarkable that some of the engineering solutions found today with modern helium ion microscopes could already be found in some of the early works [14.29].

Prerequisite to a successful implementation of the modern GFIS system is the reliable and easy fabrication of an atomically sharp metal tip. A multitude of methods have been proposed for creating nanotips that terminate in only a few atoms. These methods include the deposition of atoms on the apex [14.30, 14.31], supertip formation [14.32], annealing in the presence of an electric field [14.33, 14.34], gas atom facet-induced reactions involving gas adsorption [14.35, 14.36, 14.37], faceting with thin metallic overlayers [14.38, 14.39] and field-controlled chemical etching [14.40, 14.41] (an exhaustive list of literature on this topic can be found in [14.42]). With all these nanotips , adsorbed helium can be ionized at only a few atoms at the apex. The result is the high current density and source brightness which is required for a working microscope.

The operational properties of such tips and the influence of various parameters is the subject of ongoing studies. A few of the parameters that have been studied include the tip shape [14.43], the influence of pressure and gas type [14.26] on the achievable currents, and also the amount and creation of double-charged ions [14.44, 14.45, 14.46]. Also, more advanced source concepts such as single-atom tips ( ) are under investigation [14.47].

A problem that is often forgotten is the fact that the final beam also has a significant amount of neutral atoms in it. These neutral atoms are formed immediately after the ionization of the gas atoms when they travel through the high-pressure region around the tip apex. Neutral atoms are formed either through a charge exchange between the newly generated ion and a still neutral gas atom or a scattering event without charge exchange. In either case, a neutral particle with a high kinetic energy might be created with a trajectory aimed at the sample. A good source design needs to take this into account and minimize the travel distance of the beam through this high-pressure region.

For this to be useful in an academic or industrial environment, the source formation and maintenance should not be the main task of the operator. This goal was reached only in 2006 [14.13]. In the currently available GFIS-based tools, this is achieved by a combination of closed source and well-known concepts related to FIM.

In today's commercially available GFIS-based tools, the source-formation process begins with an annealing step that flashes the pre-shaped tungsten single crystal to high temperatures in the presence of a mild electric field. By changing annealing parameters, such as duration and temperature profile, as well as time and strength of the applied electric field, the shape of the tip can be influenced. This is needed to (a) compensate for age-induced changes in the tip shape and (b) within limits, optimize the resulting best imaging voltage ( ) at which the source will be operated.

Fig. 14.2

Sketch of the most important parts of the ion optical system used for helium ion microscopy and a more detailed representation of the active ion optical parts. After [14.12, 14.48]

Once a proper tip shape has been achieved, scanning field ion microscopy ( ) is used to remove unwanted adatoms and form the desired atomic structure at the apex. SFIM is performed by scanning the image of the tip, which is magnified by the first lens over an aperture further down the column (Fig. 14.2). Whenever intensity is present in the image, ions will reach a sample and secondary electrons ( s) can be detected. The goal of this process is to form an arrangement of three tungsten atoms at the apex of the tip. Ideally, this trimer is positioned in the middle of a hexagon formed by the next (and subsequent) atomic layers. An example is presented in Fig. 14.3. However, the selectivity of the ionization process at the trimer is high, so that the corner and edge atoms of the next terrace which are visible in Fig. 14.3 can only be seen when the detector is saturated by the high current emitted from the trimer.

Fig. 14.3

SFIM image of a GFIS trimer. The trimer and the corner and edge atoms of the supporting lattice plane are visible

At this point it is appropriate to discuss the choice of imaging gas and the consequences for the source stability. The ideal gas for GFIS-based ion microscopy must fulfill a number of requirements; namely, it is or has:
  • Not reactive

  • Light

  • High ionization potential .

Obviously, the used imaging gas should not react with the source or the sample. This should also be true for its ionized state in which it should not form compounds with other gas atoms, surface material or the sample. For microscopy applications, the gas should also be as light as possible. Only for a low atomic number \(Z\) element can surface damage be minimized and can a large penetration depth be guaranteed. The latter is important because it ensures that the unavoidable damage that is created by a high-energy particle beam occurs as far away from the surface as possible.

The last point seems counterintuitive and requires a little more attention. In Table 14.1, a number of gases typically found inside the GFIS source chamber are listed. While the first two are currently used intentionally, the others will be present in small quantities inside the source vacuum. From the gases presented in the list, He and Ne have the highest ionization potential and the lowest polarizability . All other listed gases have a lower ionization potential and therefore require a lower field strength to become ionized. As a consequence, these gas atoms cannot reach the high field apex of the tip because they will be ionized earlier, before reaching the sensitive trimer at the tip. This is important as the polarizability of most of the other gases is significantly higher. As a result, they will be accelerated stronger and potentially damage the surface of the tip by sputtering. The more reactive ones like oxygen, nitrogen and so on can potentially also react with surface atoms and hence change the shape of the tip. Shape changes, on the other hand, will in turn change the field distribution around the apex and consequently influence the operation of the source. Although this process might be utilized intentionally to modify the tip shape during tip fabrication, it is unwanted during continuous operation with any source gas, but in particular with He or Ne.

Table 14.1

Ionization potentials and polarizability of common gases intentionally or unintentionally present in the GFIS vacuum [14.42, 14.49, 14.50]

Gas

Ionization potential (eV)

Polarizability (\({\mathrm{10^{-24}}}\,{\mathrm{cm^{3}}}\))

He

\(\mathrm{24.6}\)

\(\mathrm{0.20}\)

Ne

\(\mathrm{21.6}\)

\(\mathrm{0.29}\)

Ar

\(\mathrm{15.8}\)

\(\mathrm{1.63}\)

\(\mathrm{H_{2}}\)

\(\mathrm{15.6}\)

\(\mathrm{0.80}\)

\(\mathrm{N_{2}}\)

\(\mathrm{14.5}\)

\(\mathrm{1.74}\)

CO

\(\mathrm{14.0}\)

\(\mathrm{1.97}\)

\(\mathrm{O_{2}}\)

\(\mathrm{13.6}\)

\(\mathrm{1.57}\)

\(\mathrm{H_{2}O}\)

\(\mathrm{12.6}\)

\(\mathrm{1.43}\)

Once the desired tip shape has been reached, the ionization conditions will be optimized for maximum intensity (find the BIV) at one of the trimer atoms. The trimer atom of choice is then pre-aligned with the optical axis of the first lens and guided down the column.

In the very recent past, additional gases have become available. The most widely used is neon [14.14], which can be utilized in commercially available microscopes. However, other gases such as nitrogen (\(\mathrm{N_{2}}\))  [14.51] and hydrogen (\(\mathrm{H_{2}}\))  [14.52] are used in imaging as well as nanofabrication applications.

However, so far, most of the results have been achieved with helium and neon, followed by nitrogen. Besides the fact that many more machines are available that can utilize the former gases, the latter currently seems to be more demanding in terms of tip preparation and stability. While the heavier gases are mostly used for nanofabrication applications, they also provide slightly different contrast when used for imaging [14.53]. Other gases currently under investigation but not matured enough for microscopy or focused ion beam (FIB) applications include argon  [14.54] and krypton  [14.55].

Neon operation is very similar to helium operation with the exception that a lower extraction field is used. This is related to the lower ionization potential of neon compared to helium. Unfortunately, this also allows other gases to approach the vicinity of the tip apex. Therefore, extra care has to be taken with respect to the contamination level of the source gas .

14.1.2 Optics

The helium ion microscopy (HIM) optical column uses two electrostatic lenses and two sets of multipoles to guide the ionized atoms to the desired location on the sample (Fig. 14.2 for a sketch of the column). Electrostatic lenses are the technology of choice here due to the weaker and mass-dependent focusing action of magnetic lenses for ions. The first or condenser lens follows directly after the beam extraction and focuses the beam inside the column. After the crossover , a user-selectable aperture limits the beam size and current . The reaming part of the beam is focused on the sample by the second or objective lens . The user can control the size of the aperture and the position of the crossover independently and hence has fine-grained control over both the beam opening angle and the beam current.

The achievable beam diameter at the sample can be estimated using a formalism introduced by Barth and Kruit [14.56]. The various contributions to the final spot size are:
  • \(d_{\mathrm{I}}\) size of the source image

  • \(d_{\mathrm{A}}\) contribution due to diffraction

  • \(d_{\mathrm{S}}\) contribution due to spherical aberration

  • \(d_{\mathrm{c}}\) contribution due to chromatic aberration.

They can be summed to give the overall \(d_{50}\) probe size
$$d_{\mathrm{p}}=\sqrt{\left[\left(d^{4}_{\mathrm{A}}+d^{2}_{\mathrm{S}}\right)^{\frac{1.3}{4}}+d^{1.3}_{\mathrm{I}}\right]^{\frac{2}{1.3}}+d^{2}_{\mathrm{C}}}\;.$$
(14.1)
The size of the source image
$$d_{\mathrm{I}}=\sqrt{\frac{4I_{\mathrm{p}}}{B_{\mathrm{r}}V_{\mathrm{p}}\uppi^{2}\alpha_{\mathrm{i}}^{2}}}$$
(14.2)
can be calculated using the specified probe current \(I_{\mathrm{p}}\), the reduced brightness of the source \(B_{\mathrm{r}}\), the acceleration voltage \(V_{\mathrm{p}}\) used and the half-angle of the beam on the image side \(\alpha_{\mathrm{i}}\). The contributions due to diffraction
$$d_{\mathrm{A}}=\frac{{\mathrm{7.78\times 10^{-12}}}}{\sqrt{V_{\mathrm{p}}}\alpha_{\mathrm{i}}}$$
(14.3)
and spherical aberration
$$d_{\mathrm{S}}={\mathrm{0.18}}C_{\mathrm{s}}\alpha_{\mathrm{i}}^{3}$$
(14.4)
further depend on the spherical aberration coefficient \(C_{\mathrm{s}}\) of the column for a given column magnification. Finally, the influence of the chromatic aberration
$$d_{\mathrm{C}}={\mathrm{0.34}}C_{\mathrm{C}}\frac{\Updelta V}{V_{\mathrm{p}}}\alpha_{\mathrm{i}}$$
(14.5)
can be calculated with the chromatic aberration coefficient \(C_{\mathrm{C}}\) at the given column magnification and the energy spread of the ions emitted from the GFIS \(\Updelta V\) .

The influence of the source properties is particularly prominent in (14.2), (14.3) and (14.5). The GFIS is ideally suited as it has very high brightness, an atom-sized virtual source, low energy spread and an extremely short de Broglie wavelength. A number of key parameters for the GFIS are tabulated in Table 14.2.

Table 14.2

Key properties of the GFIS under He operation [14.17, 14.57, 14.58, 14.59]

Quantity

Value

Unit

Extraction voltage

\(25{-}35\)

kV

Acceleration voltage

\(5{-}35\)

kV

Brightness

\(\mathrm{5\times 10^{9}}\)

\(\mathrm{A{\,}cm^{-2}{\,}sr^{-1}}\)

Red brightness

\(\mathrm{1\times 10^{9}}\)

\(\mathrm{A{\,}cm^{-2}{\,}sr^{-1}{\,}V^{-1}}\)

Energy spread

\(<{\mathrm{1}}\)

eV

De Broglie wavelength

\(\approx{\mathrm{80}}\)

fm

Virtual source size

\(<{\mathrm{0.25}}\)

nm

Typical maximum total current

\(\approx{\mathrm{150}}\)

pA

Using (14.1), the \(d_{50}\) probe size can be calculated. In Fig. 14.4a,b, the minimum probe size is plotted versus the image-side half-angle \(\alpha_{\mathrm{i}}\) for a \({\mathrm{30}}\,{\mathrm{keV}}\) helium beam with a beam current of \({\mathrm{0.5}}\,{\mathrm{pA}}\) (Fig. 14.4a,ba) and \({\mathrm{20}}\,{\mathrm{pA}}\) (Fig. 14.4a,bb) [14.48]. While the latter is rarely used for imaging, the first one is a realistic value suitable for high-resolution imaging.

Fig. 14.4a,b

Probe size versus image-side half-angle \(\alpha_{\mathrm{i}}\). The two plots show the results for a (a\({\mathrm{0.5}}\,{\mathrm{pA}}\) and (b\({\mathrm{20}}\,{\mathrm{pA}}\) helium beam with an energy of \({\mathrm{30}}\,{\mathrm{keV}}\). After [14.48]

From the plots, it becomes clear that the probe size in the HIM is limited by the already very high brightness and the chromatic aberration. Concepts have been discussed in the literature to further reduce the aberration contribution [14.60, 14.61]. Although, not realized yet, a lateral resolution of \({\mathrm{0.3}}\,{\mathrm{\AA{}}}\) is predicted.

The magnification of the ion optical column [14.48]
$$M=\frac{\alpha_{0}}{\alpha_{\mathrm{i}}}\sqrt{\frac{V_{\text{ext}}}{V_{\mathrm{p}}}}$$
(14.6)
(not of the microscope) can be calculated using the source-side half-angle \(\alpha_{0}\) and the extraction voltage \(V_{\text{ext}}\). With the optimal values also used in Fig. 14.4a,ba, a value of \(M={\mathrm{0.72}}\) is calculated. This is much higher than for other comparable instruments. As a result, the HIM is very sensitive to vibrations, as this will be amplified by the large column magnification.

14.1.3 Auxiliary Instrumentation

An imaging method that is utilizing a \({\mathrm{0.5}}\,{\mathrm{nm}}\) wide beam to achieve imaging resolutions of the same order has special needs in terms of isolation from the environment. This is particularly true for floor or instrument vibration and acoustic coupling to the surroundings. These circumstances and a few other requirements need further attention.

As mentioned above, the microscope uses a set of apertures which are installed close to the crossover point roughly halfway along the column. These apertures, together with the adjustable crossover position, define the maximum ray angle as well as the probe size and probe current. For helium, thin gold foils are typically used for the apertures, while the higher sputter yield of neon necessitates the use of molybdenum.

The SFIM mode used during source formation requires a second set of deflectors not used during normal imaging. These deflectors are positioned above the aperture and sweep the source image over an aperture or other restriction inside the column. As mentioned before, an image is than created using a conductive sample in the main chamber. The role of the sample is to convert the transmitted ions into detectable ion-induced secondary electrons ( s).

An electrostatic blanker is used to control beam exposure. The most important parameter requiring attention is blanker speed. While it is obvious that a fast blanker is a good choice, the position of the blanker must also be considered to reduce the unwanted blanking tail. This is a result of ions passing through the not fully closed blanker. Such an ion might still hit the sample in an unwanted location. While the existing blanker is reasonably fast for most microscopy applications, specific nanofabrication or analysis techniques require a faster blanker [14.62].

First-time users are often surprised by the fact that the top of the microscope seems to be bent in some direction. This, however, is normal and achieved by a flexible connection between the column and the source. This allows aiming the beam along the optical axis of the column.

Additionally, the source needs to be cooled to liquid-nitrogen temperature . While older versions of the instrument used a pulse-tube cooler, today this is achieved by means of a solid nitrogen dewar. This system provides stable temperatures with a stability better than \({\mathrm{0.5}}\,{\mathrm{K}}\) over the cooling cycle. It also is vibration-free during the cooling cycle. However, twice a day the sublimated nitrogen has to be replenished. During this approximately two-hour period, the system experiences vibrations and temperature-induced drift that makes stable operation difficult.

Many other small and large details ensure that the system experiences minimal vibrations . An obvious one is the heavy granite support on which the experimental chamber is mounted by means of passive air dampers. It helps to lower the eigenfrequency of the system and thus makes it less susceptible to floor vibrations. A system of Helmholtz coils is normally used to reduce the negative influence of electromagnetic AC fields present in the vicinity of the tool.

A key feature of the HIM is its capability to image insulating samples without the need of an additional coating. This is made possible by the fact that, different to a scanning electron microscope (SEM), samples in the HIM will always charge positive. This is easy to understand because positive primary ions are impinging on the sample, and electrons are emitted by the sample in the form of iSE. An installed electron flood gun can be used in an interleaved mode with the primary beam to compensate for the accumulated charge. Important parameters include the flood time, the energy of the electrons, and the delay between the end of the flood time and the restart of imaging. The user must set these values based on experience and by judging the achieved image result.

One important component is the vacuum system. Despite the fact that the GFIS operates in the \({\mathrm{10^{-6}}}\,{\mathrm{mbar}}\) range He or Ne pressure, it requires an ultrahigh vacuum ( ) base pressure. This is important to minimize contamination that could adsorb or create an unwanted chemical attack on the tip. This is achieved by the combination of a turbo pump—with a high noble gas pumping speed—and an ion-getter pump for the mid-column region. Further, this part of the tool is fabricated from stainless steel to enable the use of UHV-rated CF-type flanges.

The sample chamber, on the other hand, is made from nickel-plated low-carbon steel and has vacuum levels only in the mid-\({\mathrm{10^{-7}}}\,{\mathrm{mbar}}\) range. This compromise between optimal imaging conditions and efficient sample handling is supported by a downstream plasma cleaner which can be used to clean the chamber overnight by a series of plasma cleaning/nitrogen purge cycles [14.63]. The device creates oxygen radicals that transform hard-to-pump hydrocarbons into more volatile compounds that can be removed from the vacuum. However, better vacuum levels and therefore better image quality can be achieved with a redesign of the sample chamber and an optimized pumping strategy at only a small additional penalty in sample handling [14.64, 14.65].

14.1.4 Alternative Source Technologies

Over the past few decades, other ion source technologies have also been investigated for their potential use in microscopy applications. However, the development of alternative sources is not limited to the well-known gallium liquid metal ion source (LMIS) . In the past, liquid metal alloy ion sources ( ) have gained increasing importance [14.24, 14.66]. After identifying a suitable alloy that has a low vapor pressure and a low melting point, many (isotopically) single-element focus beams can be generated. In addition to single-charged monomers, small clusters which might be multiple charged can also be extracted. However, given the high atomic mass of most elements used in such sources, they lend themselves more to micro- and nanomachining than high-resolution imaging.

Other researchers have developed ion sources based on a laser-cooled atom magneto-optical trap ion source ( ). The obvious advantage of such a system is the very small energy spread. In addition, many elements (i. e., Li, Na, K, Rb, Cs, Fr, Mg, Ca, Sr, He, Ne, Ar, Kr, Xe, Al, Ag, Cr, Er, Cd, Hg and Yb) lend themselves to laser cooling and could potentially be used in such a device. Different strategies for trapping the cooled atoms are possible, and to date, both magnetic [14.67] and electric-field trapping using a Paul trap  [14.68] have been realized. The latter has been used as a deterministic single-ion source. While ion extraction is typically a stochastic process, this realization of an ion source makes it possible to extract and position individual ions with nanometer precision [14.68].

Finally, although strictly speaking not an ion microscope, neutral-beam helium microscopy has been developed in recent years [14.69, 14.70, 14.71, 14.72]. This method is based on helium-atom scattering [14.73] but spatially resolved. The advantage is the unique surface-sensitive interaction of the thermal atoms with the substrate that can reveal—in a damage-free way—information about the surface structure [14.74] and dynamic processes on it [14.75]. However, the method is limited by the fact that He, with its low polarizability , has no spin or charge and is consequently very difficult to control in its neutral state. The aforementioned realizations have overcome this issue by using Fresnel lenses [14.70] or a pinhole [14.71]. While both methods have their drawbacks, impressive progress has been made in the field, and interesting contrast mechanisms can be exploited [14.76].

14.2 Signals

The ion microscope can make use of multiple signals, which stem from the primary or various secondary particles. It is therefore ideally suited for a number of multimodal imaging techniques. Such correlative imaging techniques become increasingly important as additional information that goes beyond the original data sets can be obtained in this way.

Before we go into the details of the various signals, some general remarks about the interaction volume and the achievable resolution are appropriate. It is crucial to realize that although the tool is usually referred to as an ion microscope, most of the images created are based on SEs emitted from the sample. In Fig. 14.5, trajectories of electrons and Ga and He atoms are presented. These trajectories have been calculated using Monte Carlo methods and give an idea of the interaction volume of these various beams with a sample. The selected energies are typical for a low-voltage scanning electron microscope ( ), gallium FIB and HIM. Also marked is the characteristic escape depth of SEs. For both electron-induced secondary electrons ( s) and iSEs , the characteristic length scale is around \({\mathrm{1}}\,{\mathrm{nm}}\), with the iSE escape depth being slightly smaller [14.79]. The gallium beam, with its large beam diameter and strong scattering, has the largest volume from which SEs can be emitted. LVSEM using a low acceleration voltage and a good beam diameter still samples a large volume because the electrons experience strong scattering close to the surface. The helium ion beam has a minimal beam diameter, and strong lateral scattering only occurs deep inside the sample. Consequently, the volume in which SEs can be excited is minimal.

Fig. 14.5

Comparison of the scattering cascade of various charged particles. For Ga and He, results have been obtained using SRIM [14.77]. CASINO [14.78] was used for \(\mathrm{e^{-}}\). Reprinted from [14.59], with the permission of AIP Publishing

An energy of \({\mathrm{1}}\,{\mathrm{keV}}\) has been chosen for the \(\mathrm{e^{-}}\)-beam to simulate a condition where surface sensitivity is required. However, even under these conditions, HIM has a higher surface sensitivity due to the lower iSE escape depth, which is related to the lower and narrower secondary electron energy distribution ( ) of the iSE as compared to eSE energy distribution [14.80]. Further, advanced filtering routines must be used in SEM to suppress energetic backscattered electrons ( s). These BSEs can originate from deep inside the sample. Although they carry important information on the sample composition, they reduce the surface sensitivity, particularly for high primary beam energies.

In the following sections, the various signals are discussed. The sections are roughly sorted by the relative importance of the signals, based on how frequently they are currently used in imaging and analytics.

14.2.1 Electrons

For a secondary electron (SE) to be emitted from the sample surface, a number of different processes have to occur. First, a sufficient amount of energy has to be transferred from the impinging particle to the target electron. Second, the excited electron has to be transported to the surface of the sample, where, third, it has to overcome the work function to be able to leave the sample.

For the generation of an iSE, there are two possible routes for the energy transfer from the impinging atom to the target electron. At low velocities of the impinging particle, potential emission ( ) [14.81] can occur. Although this process can happen at high velocities, kinetic emission ( ) soon becomes the dominant mechanism through which iSEs are generated. The transition to KE occurs around \({\mathrm{5}}\,{\mathrm{keV}}\) for He atoms [14.79, 14.82]. Consequently, we will have closer look at KE, because it is the most relevant process for iSE generation in the HIM.

According to Bethe [14.83], the rate of iSE generation can be described by
$$\delta_{\text{SE}}=-\frac{1}{\epsilon}\frac{\mathrm{d}E}{\mathrm{d}s}\;.$$
(14.7)
Here, \(\epsilon\) is a scaling constant and \(\mathrm{d}E/\mathrm{d}s\) the stopping power of the incident particle in \(\mathrm{eV/\AA{}}\). As discussed above for the case of He, no significant nuclear collisions are expected in the first few nanometers below the surface. Consequently, \(\mathrm{d}E/\mathrm{d}s\) depends mostly on the electronic stopping power.
At least two types of electrons generated in the sample have to be distinguished. iSEs liberated by kinetic emission induced by the primary ion are called iSE1, whereas iSEs generated by a secondary process are called iSE2. Also, at least two different types of secondary processes must be distinguished here. iSE2 can be the result of KE induced from recoiling atoms or collisions of iSE1 with other target electrons. Although the number of iSE1 is large, the latter process has a very small probability, because the maximum energy transferred to a target electron during kinetic emission is rather limited. The maximum transferred to a target electron by a head-on collision can be estimated using [14.84, 14.85, 14.86]
$$\Updelta E=2m_{\mathrm{e}}\left[v+\left(\frac{v_{\mathrm{F}}}{2}\right)\right]^{2}.$$
(14.8)
Here, \(m_{\mathrm{e}}\) is the electron mass, \(v\) the velocity of the incoming ion, and \(v_{\mathrm{F}}\) the Fermi velocity of the target electrons. According to (14.8), the maximum energy transferred to the target electrons of a gold sample is \({\mathrm{45}}\,{\mathrm{eV}}\) for the case of a \({\mathrm{35}}\,{\mathrm{keV}}\) He beam. This energy is below or only equal to the energy necessary to generate an eSE for most materials [14.87].

iSE2 generated by recoiling atoms are also of less importance, as the number of recoils close to the sample surface is small, and thus only a small number of such iSE2 are generated. Ramachandra et al [14.79] have calculated the \(\text{iSE2}/\text{iSE1}\) ratio. Their conclusion is that for higher primary energies, a smaller ratio can be achieved and thus a higher resolution should be possible, as the electrons will originate from a smaller volume.

Once the electrons have been liberated in the bulk, they still need to escape from the solid. This process is usually described as a diffusion process of the iSE with a characteristic length scale that is on the order of \({\mathrm{1}}\,{\mathrm{nm}}\) for nearly all materials. Consequently, HIM is extremely surface-sensitive, because electrons can only escape if they are generated within the first few nanometers below the sample surface. The ion-induced secondary electron yield ( ) has been measured and changes in a material-dependent way from one to eight [14.59].

Finally, to leave the sample, the electron has to overcome the work function. This is a material property that also depends on the crystal structure of the surface, which can also be changed by adsorbates . Due to the already small energy of the iSE, HIM is very sensitive to changes in the work function, which can result in a rich materials contrast [14.88, 14.89]. Unfortunately, the local iSEY will always be a combination of the production and the escape processes (assuming that the detection system is equally efficient for all positions and energies—in particular, the former is not completely correct). It is therefore often difficult to attribute a contrast change to a change in work function, as the iSE production may also be altered.

In Fig. 14.6a,b, the ion-induced secondary electron energy distribution ( ) of the iSE created in an HIM is presented. While in Fig. 14.6a,ba calculated iSEY for various techniques are presented, actual measured iSE yields are shown in Fig. 14.6a,bb. The maximum of the distribution is usually smaller and at lower energies compared to the distribution found in an SEM [14.80, 14.90]. However, these calculations still overestimate the energy of the peak of the distribution. This is possibly related to bulk plasmon excitations which are not considered in the calculations [14.91, 14.92].

Fig. 14.6a,b

Ion-induced secondary electron energy distribution (iSEED) in the HIM. (a) Calculation of the iSEED for various techniques. (b) Measured iSEED for various metals. The inset shows the material (work function)-dependent position of the maximum energy in the iSEED. After [14.80, 14.91]

In the end, all these processes increases the surface sensitivity of the method. A comparative study of SEM and HIM using carbon-coated gold nanorods underline the magnitude of this effect. While the approximately \({\mathrm{1}}\,{\mathrm{nm}}\)-thick carbon shell is visible in the HIM under standard imaging conditions, very low acceleration voltages are necessary in SEM to achieve a comparable contrast (Fig. 14.14) [14.93]. Another example of the surface sensitivity and, in particular, the sensitivity to changes in the work function is presented in a study that used HIM to visualize the different half unit cell terminations found in \(\mathrm{Ti_{3}SiC_{2}}\) [14.88]. Here, only the topmost surface layer differs between adjacent half unit cell high steps. While titanium terminates the full unit cell terraces, the terraces at half the unit cell are terminated by silicon. However, the difference in work function between these two surface terminations is sufficient to be visualized by the ion microscope. It is also possible to obtain quantitative information on the work function using the iSEED. With measurements such as the one presented in Fig. 14.6a,b, Petrov and Vyvenko were able to calculate the work function of the target material [14.91, 14.94].

To obtain the best imaging results, extra care must be taken with respect to sample cleanliness . As discussed previously, the HIM will readily detect already very thin carbon films. A well-known and usually unwanted carbon film present in many HIM and SEM images manifests as the dark squares found after obtaining high-resolution images in both techniques. These dark squares are the result of beam-induced cracking of residual hydrocarbons originating from either the sample or the vacuum chamber. An optimized sample handling strategy and regular cleaning of the specimen chamber will improve the situation for cases in which imaging depends on very clean surfaces.

With the exception of the special cases presented above, so far no approach has been developed that allows one to obtain quantitative information from the SE data. With a software package developed by P. Rack et al, the calculation of iSE yields for a large number of materials is possible. However, the calculation requires tabulated data for the materials of interest and is therefore only of limited use for analytical applications where the materials are not known beforehand [14.79, 14.95, 14.96].

The most commonly used contrast mechanism in HIM is topography contrast, which is similar to that obtained in SEM. The dependence of the iSE yield on the surface tilt relative to the incoming beam can be described by the well-known secant law
$$\delta_{\text{SE}}(\theta)=\delta_{\text{SE}}(0)\sec\theta\;.$$
(14.9)

However, some measurements indicate that the response curve is actually flatter, and a reduced edge effect is therefore to be expected [14.16]. A comparison of experimental iSEY versus the tilt angle for various materials [14.16], together with results from simulations [14.97] and the secant law (14.9), is presented in Fig. 14.7. In contrast, several more recent articles report a very strong edge effect for thin films and two-dimensional () materials [14.100, 14.99]. This is encouraging, as imaging of 2-D materials is notoriously difficult. For selected examples such as 2-D membranes, HIM clearly outperforms SEM [14.101, 14.102] .

Fig. 14.7

Comparison of the angular dependence of the iSEY for different materials (Cu Exp., Au Exp., W Exp.) [14.16] with simulation results (Cu Sim. \({\mathrm{5}}\,{\mathrm{kV}}\), Cu Sim. \({\mathrm{25}}\,{\mathrm{kV}}\)) [14.97] and the secant law (14.9). Reproduced from [14.98]

14.2.2 Ions

Although they seem to be the obvious choice in an ion microscope, the backscattered particles are rarely used in current HIM. An annular multichannel plate can be used to obtain, in a simple way, images that provide some qualitative material contrast based on backscattered helium (BSHe ). The mechanism behind this contrast can be understood by examining the Rutherford scattering cross section
$$\mathrm{d}\sigma=\left(\frac{q^{2}Z_{1}Z_{2}}{4E_{0}}\right)^{2}\frac{\mathrm{d}\Omega}{\sin^{4}\frac{\theta}{2}}\;.$$
(14.10)
Here, \(q\) is the elementary charge, and \(Z_{1}\) and \(Z_{2}\) denote the atomic number of the incoming ion and the target material, respectively. If the target atom is at rest, \(E_{0}\) is the energy of the impinging particle, \(\mathrm{d}\Omega\) is an arbitrary part of solid angle, and \(\theta\) is the scattering angle. The cross section has a quadratic dependence on the atomic number of the target atoms. As a result, a much higher number of backscattered particles is expected for elements further down the periodic system of the elements. Unfortunately, the actual dependence of the BSHe yield also depends on the screening of the nucleus by the electrons. Measured backscatter yields for a large number of elements are presented in Fig. 14.8. While the overall trend shows a clear increase in yield with increasing atomic number, an oscillation following the rows of the periodic table is clearly visible. It is important to realize that the amount of BSHe is small, and a detector with a high efficiency is required. Even for a very high BSHe yield of \({\mathrm{20}}\%\)—as is the case for a gold target—the BSHe will be distributed over a large solid angle. Under standard imaging conditions, usually only about \(\mathrm{500}\) ions are used per pixel; therefore, the detector must have high sensitivity.
Fig. 14.8

Measured BSHe yields plotted against the atomic number \(Z_{2}\). While an overall increase in BSHe yield can be observed, the periodic structure of the table of elements is clearly visible. After [14.17]

A second possibility in addition to counting the number of backscattered primary particles is to measure their kinetic energy. The energy
$$\begin{aligned}\displaystyle E_{1}&\displaystyle=E_{0}\left(\frac{M_{1}}{M_{1}+M_{2}}\right)^{2}\\ \displaystyle&\displaystyle\quad\,\times\left(\cos\theta\pm\sqrt{\left(\frac{M_{1}}{M_{2}}\right)^{2}-\sin^{2}\theta}\right)^{2}\end{aligned}$$
(14.11)
of a particle with mass \(M_{1}\) that gets scattered into the backscatter angle \(\theta\) depends on its primary energy \(E_{0}\) and the mass \(M_{2}\) of the collision partner. As both the primary energy and the mass of the impinging particle are known, the mass of the collision partner can be determined, once the energy of the backscattered particle is measured under the angle \(\theta\). This methodology is similar to the well-known Rutherford backscattering spectrometry ( ), which is usually performed at much higher energies of around \({\mathrm{1}}\,{\mathrm{MeV}}\).

Utilizing these signals is complicated by the fact that most of the backscattered helium is neutral [14.103, 14.104, 14.105], and the collision process inside the sample is characterized by a large number of multiple and dual collisions [14.106, 14.107, 14.108, 14.109, 14.110]. The difference between the various scattering mechanisms is depicted in Fig. 14.9a-d. While single scattering (Fig. 14.9a-da) is desired, dual, plural and multiple scattering will result in the same scattering angle, but different particle energies. This complicates the analysis and limits the achievable resolution. Nevertheless, various attempts have been made to utilize those signals [14.111, 14.112, 14.113, 14.99].

Fig. 14.9a-d

Different trajectories for backscattered particles, for the different cases of single scattering (a), dual scattering (b), plural scattering (c) and multiple scattering (d). The exit angle is identical for all four cases, while the energy of the backscattered particles will differ. After [14.114]

14.2.3 Photons

In its most general form, ionoluminescence ( ) can be defined as the emission of photons due to the optical transition of an electronic system that has been excited by an energetic ion. Light emission can stem from:
  1. 1.

    Excited backscattered neutral helium [14.115, 14.116]

     
  2. 2.

    Excited sputtered particles and molecular complexes [14.117, 14.118, 14.119]

     
  3. 3.

    The sample material directly [14.118, 14.120].

     
For the case of HIM, the latter is the most relevant in terms of signal strength and materials information. Two main cases can be distinguished here. IL can originate from delocalized states or, in other words, the recombination of free electrons from the conduction band with holes from the valence band. This process is called intrinsic ionoluminescence (IL)  [14.121]. If the light emission is related to the presence of impurity atoms or ions, it is termed extrinsic emission  [14.121]. Here, several cases have to be considered because these impurity atoms can either act as activators , which might need a sensitizer to actually emit light (\(\mathrm{Tb^{3+}}\) can be activated by \(\mathrm{Ce^{3+}}\)), or finally, they can also act as quenchers and suppress luminescence if they are present. For example, \(\mathrm{Fe^{2+}}\) can act as a quencher for \(\mathrm{Mn^{2+}}\) in apatite [14.122]. Activators can work in various ways:
  1. 1.

    Transition metal ions with \(\mathrm{d}^{\mathrm{n}}\) electronic configuration (e. g., \(\mathrm{Ti^{3+}}\), \(\mathrm{Cr^{3+}}\), \(\mathrm{Mn^{2+}}\))

     
  2. 2.

    Ions with s\({}^{2}\)-configuration (e. g., \(\mathrm{Tl^{+}}\), \(\mathrm{Pb^{+}}\), \(\mathrm{Sb^{3+}}\))

     
  3. 3.

    Rare-earth elements (REE\({}^{2+/3+}\))

     
  4. 4.

    Actinides (e. g., \(\mathrm{UO_{2}^{2+}}\), \(\mathrm{Cm^{3+}}\)).

     
Since IL is, in many respects, similar to cathodoluminescence , databases available for the interpretation of cathodoluminescence ( ) results can also be utilized to interpret IL measurements. However, IL studies are complicated by the fact that the ion beam not only induces the desired light emission from the sample but also influences the sample in a more drastic way than an electron does in the case of CL. The ion-induced defect generation often results in target coloration and enhanced emission for ionic materials [14.123, 14.124], but usually quenches the intrinsic band structure emission from semiconductors [14.125, 14.126]. On the other hand, with its broad range of accessible fluences and fluxes and its high resolution, HIM enables in situ studies of exactly these degradation processes [14.127].

14.3 Imaging

14.3.1 High-Resolution Imaging

High-resolution surface imaging is one of the key features of HIM. The following points have to be considered.

Low Beam Current (\(\approx\) 200 fA)

The amount of damage created in the sample is proportional to the fluence. When the irradiated area becomes smaller and the pixel density becomes higher, damage will occur rapidly.

Small Aperture (5 \(\upmu\)m or 10 \(\upmu\)m)

This will ensure a low beam current and a high-quality beam because only the part close to the optical axis can reach the sample and minimizes the signal created by neutral He atoms hitting the sample unintentionally. (The number of neutral atoms depends on the gas pressure in the source chamber. In current devices, once created and traveling down the optical column, neutral particles can only be filtered mechanically with the use of an aperture.)

Large Spot Control (\(> 6\))

This value corresponds to the distance between the central beam crossover and the aperture position. Larger distances will provide a cleaner beam with lower currents.

High Acceleration Voltage (Ideally \(> \) 30 keV)

This ensures that, close to the surface, nuclear collisions are minimized and the electronic stopping is maximized. This results in a minimum beam straggle close to the surface and a maximum iSEY.

This list is far from complete, and the importance of the various contributions varies from sample to sample. In particular, the reduction of the primary current will lower the signal-to-noise ratio. To compensate, longer dwell times and line or frame averaging must be used. However, an increased recording time can lead to image distortion as drift becomes more noticeable (including drift of the column alignment as well as sample drift). An additional problem arises from the fact that cracked hydrocarbons will quickly accumulate in the imaged area effectively obscuring small features on the surface [14.128]. This problem can only be overcome by ensuring that both the sample and vacuum chamber are as clean as possible. The downstream plasma cleaner typically used in HIM mitigates this problem [14.63]. Although the use of the plasma cleaner is safe for many samples, care has to be taken with delicate (e. g., biological) samples. For certain use cases, sample modification by the plasma cleaner can also be utilized to reveal the bulk structure of samples, which is otherwise not or only difficult to access [14.129]. A comparative study between a standard vacuum chamber and a custom-built UHV chamber demonstrated the possible benefits that a better vacuum would bring [14.65].

Of the many possible examples that demonstrate the high-resolution capabilities, one is presented in Figs. 14.10 and 14.11. The figures give an overview of the morphologies that are formed by transition metal oxides grown from mixtures of \(\mathrm{Fe(acac)_{3}}\) and \(\mathrm{Co(acac)_{2}}\) by pulsed-spray evaporation chemical vapor deposition ( ) [14.130, 14.131, 14.132]. The presented magnification series shows details of the morphology starting from a field of view ( ) of \({\mathrm{10}}\,{\mathrm{\upmu{}m}}\) in the top row to \({\mathrm{300}}\,{\mathrm{nm}}\) in the bottom row. At all magnification steps, the fine detail and the rich morphology can be admired.

Fig. 14.10

High-resolution images of a mixture of TMO layers grown by PSE-CVD (part 1). The mixing ratio of the precursors is given at the top. The FOV decreases from top to bottom: 10, 2, and \({\mathrm{1}}\,{\mathrm{\upmu{}m}}\), 500 and \({\mathrm{300}}\,{\mathrm{nm}}\). Reproduced from [14.133]. The images shown in the figures have been reproduced in part from [14.131] with permission of The Royal Society of Chemistry

Fig. 14.11

High-resolution images of a mixture of TMO layers grown by PSE-CVD (part 2). The mixing ratio of the precursors is given at the top. The FOV decreases from top to bottom: 10, 2, and \({\mathrm{1}}\,{\mathrm{\upmu{}m}}\), 500 and \({\mathrm{300}}\,{\mathrm{nm}}\). Reproduced from [14.133]. The images shown with in the figures have been reproduced in part from [14.131] with permission of The Royal Society of Chemistry

14.3.2 Imaging Insulators and Biological Samples

A key feature of the HIM is its ability to image insulating and charging samples without the need for a coating and without compromising the lateral resolution. This ability of the HIM is naturally of interest to biological sciences. A direct comparison between variable-pressure SEM (VPSEM) and HIM is presented in Fig. 14.12a-d. The difference in image fidelity is clearly visible by comparing the low-magnification images in Fig. 14.12a-da,b. The difference becomes even clearer when increasing the magnification. The VPSEM image presented in Fig. 14.12a-dc is blurry and lacks details that are easy to resolve in the corresponding HIM image in Fig. 14.12a-dd. In numbers, this corresponds to an edge resolution of \({\mathrm{55}}\,{\mathrm{nm}}\) for the VPSEM data and \({\mathrm{3.4}}\,{\mathrm{nm}}\) for the HIM data (measured at the indicated positions). This large difference in imaging performance is most likely the result of the challenging combination of sample charging and the required large depth of focus [14.135]. HIM clearly outperforms SEM in these two areas. In another example, the low-voltage field emission scanning electron microscope ( ) shows a somewhat better performance, but for this case as well, HIM outperforms the SEM-based imaging at the highest magnification levels [14.136].

Fig. 14.12a-d

Comparison of images obtained by VPSEM (a,c) with images obtained by HIM (b,d) imaging performance on uncoated butterfly wings (Papilio ulysses black dorsal scales). Reprinted with permission from [14.134], published by John Wiley and Sons

The benefit of having a combined helium/neon GFIS microscope is demonstrated by a study of predator worms [14.136]. Here, the resolving power of the helium ion beam has been used to image Pristionchus pacificus Fig. 14.13a-ga. To reveal the tooth hidden behind the mouth structure, the focused neon beam available in the HIM has been employed to remove the outer sheath of the mouth. An example of the resolving power of HIM, even with delicate samples with small contrast, is presented in Fig. 14.13a-gb. The image shows the surface of crystal of flat bovine liver catalase. The weak rectangular texture present in the image stems from the protein crystal unit cell. The inset shows the fast Fourier transform (FFT) of the data, which allows one to obtain the correct unit cell dimension of \({\mathrm{8.8}}\,{\mathrm{nm}}\times{\mathrm{6.7}}\,{\mathrm{nm}}\) [14.17]. It is remarkable that the contrast is the result of only small changes in the electron density between the position of the protein and the somewhat lower electron density between the proteins. Other examples include uncoated leaf surfaces at low, medium and high magnification in Fig. 14.13a-gc–e. The samples have been collected, stored in a vacuum chamber for several hours to remove excess water and then directly inserted into the HIM for imaging [14.19].

Finally, Fig. 14.13a-gf,g shows low- and high-magnification images of a diatom. Also in this case, no additional sample preparation aside from removing the organic material was performed.

Fig. 14.13a-g

Various examples of biological samples imaged with the HIM. All samples are uncoated, and the images have been acquired with the use of the electron flood gun in the HIM. (a) Predator roundworm where a part of the mouth has been removed to reveal the underlying tooth. Reproduced from [14.136]. (b) Protein crystal. The spots in the FFT reveal the crystal structure of the sample. Reprinted from [14.17], with the permission of AIP Publishing. (c) Low, (d) medium and (e) high magnification of the uncoated surface of Euphorbia myrsinites leaves. From [14.19]. (f) Low and (g) high magnification of a diatom. Sample courtesy of A. Jantschke, Bioanalytical Chemistry, TU Dresden

14.3.3 Surface Sensitivity

The energy distribution of the iSE generated in the HIM, and the lack of medium- and high-energy backscattered electrons, makes the technique extremely surface-sensitive. This allows the undisturbed imaging of the true sample surface. Here, undisturbed specifically refers to interference from signals related to bulk properties and not surface properties. The surface sensitivity can be further increased by making use of channeling into a crystalline substrate (Sect. 14.4.4 for further details). The reduced electron production in the bulk optimizes the signal-to-noise ratio for the surface layer.

Figure 14.14 makes possible a direct comparison between an HIM image and a conventional SEM image obtained from exactly the same sample area. A blanket-like carbon layer is covering the gold nanorods in the HIM image. In the SEM image—obtained with an acceleration voltage of \({\mathrm{2}}\,{\mathrm{kV}}\)—this blanket is invisible. It even seems that the gold nanorods are separated by a \({\mathrm{2}}\,{\mathrm{nm}}\) vacuum gap. While the in-lens detector used is very efficient, in this case it does not discriminate between the high-energy eSEs and BSEs created in the Au nanorod and the much smaller number of eSEs generated in the carbon layer. Only after lowering the acceleration voltage to \({\mathrm{220}}\,{\mathrm{V}}\) and bringing the sample within approximately \({\mathrm{1.5}}\,{\mathrm{mm}}\) in front of the pole piece of the SEM can an image comparable to the HIM be recorded. Although this low-energy capability is becoming more widespread, it is by no means a standard option in conventional SEM. A similar comparison is presented in Fig. 14.15a-c [14.98]. While clearly the high-voltage field emission scanning electron microscope ( ) image presented in Fig. 14.15a-ca shows no surface detail, the one obtained with only \({\mathrm{1}}\,{\mathrm{keV}}\) acceleration voltage (Fig. 14.15a-cb) has more details at small length scales. However, the HIM image (Fig. 14.15a-cc) obtained at \({\mathrm{30}}\,{\mathrm{keV}}\)—and therefore not compromising the performance or versatility of the tool—is the richest in detail.

Fig. 14.14

(a) HIM and (b) conventional SEM image obtained from the same sample area. While the blanket-like carbon layer is clearly visible in the HIM image, the gold nanorods seem to be separated by a \({\mathrm{2}}\,{\mathrm{nm}}\) gap in the SEM image. Reprinted from [14.93], with permission from Elsevier

Fig. 14.15a-c

Highly structured silicon surface imaged with (a) FESEM at \({\mathrm{15}}\,{\mathrm{keV}}\), (b) FESEM at \({\mathrm{1}}\,{\mathrm{keV}}\) and (c) HIM at \({\mathrm{30}}\,{\mathrm{keV}}\). From [14.98]

Other demonstrations of the surface sensitivity include the visualization of various surface terminations found in single crystals of \(\mathrm{Ti_{3}SiC_{2}}\) [14.88]. Depending on the position of the cleavage plane with respect to the unit cell, the atomically flat terraces are terminated by either titanium (full-unit cell) or silicon (half-unit cell). The related minute change in work function leads to a change in the iSEY, which appear as darker and brighter areas.

This contrast mechanism is not limited to bulk samples, but can also be applied to the important class of 2-D materials. Recently, it has been demonstrated that HIM can be used to count the number of layers in few-layer graphene [14.137] as well as hexagonal boron nitride [14.138].

Fig. 14.16

iSE intensity of hexagonal boron nitride as a function of the layer number and primary-beam energy. Reprinted from [14.138], with the permission of AIP Publishing

From the plot of the iSE intensity versus the layer number presented in Fig. 14.16, it becomes clear that the number of layers can easily be counted independent of the primary energy used. A model has been developed by the authors to explain the observed contrast and actually extract numbers from it [14.138].

However, when imaging sensitive 2-D materials, one needs to keep in mind the damage that can be caused by the energetic ion beam. In particular for unprotected graphene, it has been shown that it is very difficult to obtain clear images without damaging the sample [14.100]. However, several experimental [14.139, 14.140, 14.141] and theoretical [14.142] reports show that encapsulating 2-D materials in other 2-D materials can mitigate the problem.

14.3.4 Voltage Contrast and Subsurface Imaging

Because electrons are most commonly used for image generation in HIM, it is also sensitive to variations in the local electronic properties of the materials. Conductive structures buried below high-resistance materials can be visualized using the static capacitive contrast. The subsurface features reduce the surface field build-up due to charging and allow electrons to escape more easily. This and many more likely contrast mechanisms based on changes in the surface electrical field are collectively referred to as voltage contrast mechanisms.

An example is presented in Fig. 14.17a,b. From the comparison of SEM (Fig. 14.17a,ba) with HIM (Fig. 14.17a,bb), it is obvious that the effect is stronger in the latter case. This is explained by the difference between the SEED in the HIM and from the one found in the SEM. The lower-energy iSE is more strongly influenced by changes in the surface potential than the somewhat more energetic eSEs formed in the SEM.

Fig. 14.17a,b

Comparison of voltage contrast in SEM (a) and HIM (b). Reprinted from [14.143], with the permission of AIP Publishing

Buried structures can also be visualized by using BSHe. In Fig. 14.18, simultaneously recorded iSE and BSHe images of a semiconductor/metal structure are shown. The bright areas at the top and the bottom are Pd electrodes. They are separated by a buried Si layer which is sandwiched between two \(\mathrm{SiO_{2}}\) layers. The thickness of the top thermal oxide layer is \({\mathrm{114}}\,{\mathrm{nm}}\). See the inset in Fig. 14.18 for a schematic representation of the sample structure. While the iSE image on the left is rich in surface contrast on the Pd, as well as the oxide surface, the Pd distribution is revealed in the BSHe image (right). However, only the center area (dark: no Pd) between the electrodes (bright: high Pd content) is free of Pd. During the thermal treatment of the sample, Pd diffused (gray: intermediate Pd concentration) into the buried Si layer and formed stoichiometric \(\mathrm{Pd_{2}Si}\) [14.113]. The \(\mathrm{Pd_{2}Si}\) formed below the top oxide layer increases the backscatter probability and gives rise to the stronger signal as compared to the pure \(\mathrm{Si/SiO_{2}}\) stack in the center of the image. The shape of the diffusion front is clearly visible. Although other approaches are possible, BSHe imaging through the top oxide layer is the simplest method to map out the shape of the diffusion front with high resolution in a very short time.

Fig. 14.18

Visualization of a buried Pd diffusion front. Left: surface-sensitive iSE image. Right: Mass-sensitive bulk image obtained using BSHe. The buried \(\mathrm{Pd_{2}Si}\) diffusion front is visible as a light gray area between the bright Pt deposits. FOV: \({\mathrm{5}}\,{\mathrm{\upmu{}m}}\). In the inset, a schematic representation of the sample structure is presented. Reprinted from [14.113], with permission from Elsevier

14.3.5 Scanning Transmission Ion Microscopy

Scanning transmission ion microscopy ( ) had already been investigated before atomically sharp GFIS sources became available [14.9]. However, since HIM has become available, renewed interest in this method has arisen. In its simplest form, STIM can be used for endpoint detection when milling pores into a membrane [14.144]. This has been exploited by several groups, but in particular has been used for the controlled fabrication of sub-\({\mathrm{5}}\,{\mathrm{nm}}\) pores for biomolecule detection [14.145, 14.146, 14.147].

Even more advanced are applications that make use of the transmitted ions to generate signals that allow us to investigate bulk sample properties of nanostructures. Although advanced techniques such as backside secondary ion mass microscopy ( ) are possible, so far only transmitted ions and iSEs emitted from the backside of the sample have been used.

Figure 14.19a-d presents stopping and range of ions in matter ( )-based calculations to help in understanding the principle and limitations of STIM bright-field ( ) and dark-field ( ) imaging. The top row in Fig. 14.19a-d presents the calculated exit angles of the transmitted helium after it has passed through the support (left) or a gold nanoparticle present on this support (right). The increase in scattering angle for the latter case is related to the increased cross section \(\mathrm{d}\sigma\) (14.10) for the heavy element nanoparticle. For a given scattering angle \(\theta\), the cross section, and therefore the chance for a scattering event, increases with increasing \(Z_{2}\). If such angle distributions are collected for each pixel in a STIM image, post-processing will enable us to mass-filter the acquired information by selecting the appropriate cutoff angle between the BF (small angles) and the DF (large angles). However, the scattering will also broaden the finely focused helium beam, which will affect the achievable resolution for thick and/or heavy specimens. In the lower row of Fig. 14.19a-d, the distance of the exit position of the helium atom from the beam center is plotted for the same configuration. While the beam certainly broadens, the loss in resolution for this rather extreme case is at an acceptable level.

Fig. 14.19a-d

Scattering-induced angular distribution in STIM: (a) Histogram of the exit angle of helium after traversing \({\mathrm{10}}\,{\mathrm{nm}}\) of \(\mathrm{SiO_{2}}\), often used as a sample support. (b) Exit angle after traversing the sample support presented in (a) plus \({\mathrm{10}}\,{\mathrm{nm}}\) of gold to simulate a gold nanoparticle. Expected lateral resolution in STIM: (c) Exit position of helium after traversing the sample support. (d) Exit positions of helium after traversing the \(\mathrm{SiO_{2}}\) support and an additional \({\mathrm{10}}\,{\mathrm{nm}}\) of Au. Please note the 10\(\times\) larger \(y\)-axis

In Fig. 14.20a-d, various STIM-related imaging modes are compared. In Fig. 14.20a-da, the topside iSE image of a thinned sample of the gate region of a semiconductor device is given as a reference. Here, contrast is related to the iSEY and metals, and other conductors appear bright, whereas insulators such as gate oxides appear dark. Figure 14.20a-db is a bright-field image of the same area. Although the image has a somewhat lower resolution, the metal lines appear dark, as they are formed from heavier materials and result in more scattering, which reduces the intensity of the primary beam. A higher-quality image can be achieved by utilizing the backside iSE signal (Fig. 14.20a-dc). These iSEs are generated by the transmitted primary and recoil atoms. As the achievable resolution in this mode is defined by the width of the ion beam exiting the sample at the bottom, it will be inferior to topside iSE imaging. However, the contrast is a combination of the contrast mechanism of the bright-field image with the high signal-to-noise ratio of an iSE image. Finally, a high-resolution dark-field image of carbon black nodules is presented in Fig. 14.20a-dd. The resolution of this image is \({\mathrm{0.45}}\,{\mathrm{nm}}\) and is very rich in gray levels. By changing the cutoff between dark- and bright-field images, the contrast can be further optimized. By carefully optimizing the cutoff angle for dark-field imaging, the internal composition of core shell particles also can be revealed. For this purpose, an annular microchannel plate (MCP ) detector can be used at various distances from the sample [14.148].

Fig. 14.20a-d

Comparison of four different STIM imaging modes. (a) Topside iSE image, (b) bright-field STIM and (c) backside iSE image. (d) Is a dark-field STIM image of a carbon black sample. Reprinted from [14.143], with the permission of AIP Publishing

Due to the similarity of STIM to scanning transmission electron microscopy ( ), it is intriguing to think about diffraction-based contrast. A number of results have been presented in the past indicating that a more complex contrast mechanism than the ones described earlier should be possible. Examples are presented in Fig. 14.21a-c. The authors interpret the contrast visible in Fig. 14.21a-ca as the result of the interference of the Bragg-diffracted beam with the incident beam as they travel through the sample. Knowing the wavelength of the incident helium beam and the structure factor, one can predict the periodicity of the observed intensity variations. For the MgO cube presented in Fig. 14.21a-ca, the observation matches nicely with the expected values. In addition, STIM provides a way to reveal information about the internal structure of a crystalline sample. In Fig. 14.21a-cb, a BF image of defects present in MgO cubes is shown. However, care must be taken in such investigations, as the impinging ion beam will also damage the crystal structure and create additional defects. Finally, it should be mentioned that also in STIM, channeling—to reveal the crystal structure of the specimen (Sect. 14.4.4)—should be possible, and first results look promising [14.150].

Fig. 14.21a-c

Examples of diffraction-based contrast in STIM. BF (a) and DF (b) image of MgO crystals obtained with a \({\mathrm{40}}\,{\mathrm{keV}}\) \(\mathrm{He^{+}}\) beam. (c) Line defects in MgO revealed in a \({\mathrm{40}}\,{\mathrm{keV}}\) \(\mathrm{He^{+}}\) BF image. The horizontal FOV is \({\mathrm{200}}\,{\mathrm{nm}}\) in all cases. Reprinted from [14.149], reproduced with permission

An interesting variant of HIM where the detector is also placed below the sample is scanning reflection ion microscopy ( ) [14.151]. This method lets the beam strike the sample at glancing angles and enables imaging of the surface free of elemental contrast, with good resolution, perpendicular to the surface. As the imaging is based on the collection of atoms and not electrons, surface charging has only a weak effect, and the method also works for insulators .

14.4 Analytical Approaches to Reveal Composition and Other Material Parameters

Before the different analytical approaches realized to date are discussed, a few comments on the connection between lateral resolution and sensitivity are appropriate. It is obvious, but often overlooked for nanosized objects, that the concentration of the target element must be sufficiently large such that at least one atom is present in the voxel to be analyzed. However, in HIM, a second complication arises that is related to sputtering and more generally the damage induced by the energetic particles. Even the helium beam—used for most imaging tasks—will sputter the sample surface. While for SIMS (Sect. 14.4.2) sputtering is a required process, this is not the case for other analytical methods that rely on backscattering of primary particles (Sect. 14.4.1) or the production of secondary particles like photons (Sect. 14.4.3) or electrons .

Taking this into account, one can calculate the achievable element sensitivity for a given analyte volume. In Figs. 14.22 and 14.23, results of such calculations are presented for backscatter spectrometry ( ) and SIMS, respectively. For the chosen examples (Fe and Si), it turns out that the achievable lateral resolution is around \({\mathrm{100}}\,{\mathrm{nm}}\) for BS [14.111, 14.114] and between 2 and \({\mathrm{20}}\,{\mathrm{nm}}\) [14.152] for SIMS. In this comparison, one has to keep in mind that SIMS is destructive by definition, while BS in principle can be performed with minimal damage to the sample. On the other hand, on also has to remember that for SIMS, the detection of a single particle is sufficient to determine the presence of an element, while, for BS, a statistically relevant number of particles needs to be obtained from the analyte volume.

Fig. 14.22

Minimum feature size for which a given concentration of iron can be detected using backscatter spectrometry. Results for a \({\mathrm{5}}\,{\mathrm{nm}}\)-thick layer (dashed) and a \({\mathrm{20}}\,{\mathrm{nm}}\)-thick layer (solid) analyzed by \({\mathrm{2}}\,{\mathrm{MeV}}\) He (red), \({\mathrm{30}}\,{\mathrm{keV}}\) He (black) and \({\mathrm{30}}\,{\mathrm{keV}}\) Ne (blue) are shown. The analysis assumes that at least \(\mathrm{1000}\) particles are detected. After [14.114]

Fig. 14.23

Detection limit for Si ions using Ne ions as a function of the feature size. The usefull yields are \(\mathrm{2\times 10^{-6}}\) without oxygen flooding and \(\mathrm{2\times 10^{-2}}\) with oxygen flooding. After [14.152]

In general—as for other microscopy methods—analytics in HIM benefits substantially from combining multiple data sets. This method of combining data sets from the same sample area—ideally acquired simultaneously—is often referred to as correlative imaging. To maximize the usefulness of the approach, it is important to combine data sets with orthogonal properties. In HIM these could be, for example, a relatively low-resolution BSHe image (Sect. 14.4.1) with a high-resolution iSE image. The enhanced resolution of the latter generally enables the precise attribution of the low-resolution elemental signal to structures smaller than the resolution of the low-resolution composition data.

14.4.1 Backscattered Particles

The use of backscattered He was one of the first attempts to add analytical capabilities to HIM [14.112, 14.153, 14.154]. However, the silicon drift detector used had an energy resolution of only between \(\mathrm{4}\) and \({\mathrm{5}}\,{\mathrm{keV}}\). This, and the small solid angle, led to only a small number of successful applications. Nevertheless, these early attempts gave hope that, with a better detector, good results could be achieved. In Fig. 14.24, results obtained with the discussed silicon drift detector are presented. While the actual recorded data does not resemble the well-known box profiles typically obtained in RBS, the simulated spectra for detectors with better energy resolution are encouraging [14.155].

Fig. 14.24

Energy distribution of \({\mathrm{30}}\,{\mathrm{keV}}\) He backscattered from a thin HfO layer on Si. The experimental data are presented together with a fit obtained through SIMNRA. In addition, simulated spectra based on detectors with better energy resolution are presented. Solid: \({\mathrm{4.2}}\,{\mathrm{keV}}\), dashed: \({\mathrm{1.0}}\,{\mathrm{keV}}\), dotted: \({\mathrm{0.1}}\,{\mathrm{keV}}\). After [14.155]

However, it turns out that at these very low energies, it is not only the detector resolution that limits the achievable energy resolution, but processes inside the sample also limit the achievable fidelity of the results. In classical RBS, which is usually performed at energies of several hundred keV, one assumes—correctly for the majority of experiments—that the impinging particle undergoes only a single collision on the way in and on the way out. The measured energy loss, therefore, is related only to the electronic losses during the passage of the particle thorough the sample and the energy transfer during the single small impact parameter collision at the turning point. As a result, the mass of the target atom can be calculated with a very high precision. At only a few keV, this assumption does not hold, and the impinging particle undergoes a series of small-angle collisions (large impact parameter). The sum of the small momentum changes, along with the related uncertainty in the exit angle, leads to a larger straggling of the exit energies of the detectable particles already at the surface. This effect is taken into account by most of the modern BS analysis software packages. However, while dual scattering is only considered by SIMNRA [14.156] and WINDF [14.157], plural scattering is not considered by any of the existing codes. These two terms refer to two (dual) or more (plural) large-angle collisions. A careful analysis of ion scattering trajectories obtained through Monte Carlo simulations by using SRIM [14.77] reveals that for \({\mathrm{30}}\,{\mathrm{keV}}\) He in Fe, it is more likely to have two collisions with an energy loss larger than \({\mathrm{1}}\%\) than only one collision with such an energy loss [14.114]. See Fig. 14.25 for a statistical analysis and sample trajectories depicting the situation. The fact that such an analysis can be performed also shows that these effects can be taken into account for the analysis of experimental results that suffer from these effects. Please note that, under the given circumstances, an energy loss of \({\mathrm{1}}\%\) or \({\mathrm{3}}\,{\mathrm{keV}}\) is close to the energy resolution of the silicon drift detector used. A more detailed discussion on plural and multiple scattering can be found in [14.106, 14.107, 14.108, 14.109, 14.110].

Fig. 14.25

(a) Trajectories and collision events with an energy loss of more than \({\mathrm{1}}\%\) for two different energies obtained with SRIM. (b) Statistical analysis of SRIM simulations. Histograms for collisions with an energy loss of more that \({\mathrm{1}}\%\) (left) and \({\mathrm{0.1}}\%\) are shown. At low primary energies, multiple collisions are more likely than the desired single scattering events. From [14.114]

Nevertheless, better detector schemes enable pushing both the lateral, as well as the energy resolution, close to the theoretical limit. A time-of-flight ( )-based realization has recently been introduced that made possible a time resolution of \({\mathrm{2.7}}\%\) and a lateral resolution of \({\mathrm{50}}\,{\mathrm{nm}}\) [14.62]. The TOF setup has been implemented by utilizing the already present beam blanker to create ion beam pulses with a duration of only \({\mathrm{20}}\,{\mathrm{ns}}\). The stop signal is obtained from an MCP , which records the impinging backscattered particles. In Fig. 14.26a-ca, an image of a multi-element sample is presented. Here, higher intensity values correspond to a shorter TOF, or in other words, a heavier element in the sample. The time resolution achieved corresponds to an energy resolution of \({\mathrm{1.6}}\,{\mathrm{keV}}\) for He at the surface of HfO. In terms of lateral resolution , the achieved value is close to the theoretical limit of a few tens of nanometers.

Fig. 14.26a-c

Spatially resolved element analysis in HIM. (a,b) Image of the multi-element standard used. The pixel intensity is given by the shortest measured TOF for this pixel. Higher brightness corresponds to the shorter TOF of the higher target mass. (c) Energy spectra obtained from the marked areas. The energy of the surface edge reveals the element present in the particular area. Reprinted from [14.62], with permission from Elsevier

14.4.2 Sputtered Particles

The addition of neon as an imaging gas to HIM makes it possible to efficiently use sputtered particles for elemental analysis. Helium is used in the HIM for imaging purposes because of its low mass and the fact that it is inert to both the source, as well as the sample. The low mass results in minimal damage to the sample under normal imaging conditions. Using neon —which is five times heavier—instead of helium for scanning the sample surface leads to significant damage. Furthermore, the range of neon is shorter and the kinetic energy of the ion is deposited closer to the surface. As a result, the sputter yield for neon on most materials is at least one. Typical values for He are between \(\mathrm{0.01}\) and \(\mathrm{0.1}\). Sputter yields for He and Ne obtained using different analytical models are presented in Fig. 14.27a,b. According to Sigmund [14.158], the sputter yield for an ion with energy \(E\)
$$Y(E)=0.42\frac{\alpha S_{\mathrm{n}}}{U_{\mathrm{s}}}$$
(14.12)
depends on the nuclear stopping force \(S_{\mathrm{n}}\), a parameter \(\alpha\) which depends on the mass ratio of projectile and target as well as the surface binding energy \(U_{\mathrm{s}}\). Sputter yields for He, calculated using (14.12), deviate strongly from the other theoretical values. This is related to the parameter \(\alpha\) in Sigmund's original equation that has been approximated only for mass ratios in the range of \({\mathrm{0.2}}\lesssim M_{2}/M_{1}\lesssim{\mathrm{0.5}}\). Helium exceeds this mass range for practically all target elements. The two newer analytical models (M84 [14.159], Y96 [14.160]) use a different polynomial fit for \(\alpha\) and also consider electronic stopping. They further add extra empirical parameters that help to fit the calculated values to a large number of experimental data. Nevertheless, the two later models retain the principal dependencies obtained by Sigmund much earlier. The graphs presented in Fig. 14.27a,b show that when one wants to estimate sputter yields for He, it is advisable to use this newer, slightly more complex equations.
Fig. 14.27a,b

Calculated sputter yield for \({\mathrm{30}}\,{\mathrm{keV}}\) He (a) and Ne (b) for most elements. The labels refer to different models taken from the following references: Sigmund (circles) [14.158], M84 (triangles) [14.159], Y96 (squares) [14.160]. The solid line refers to sputter yields obtain by using SRIM [14.77]. The SRIM data is present in both graphs to highlight the difference

A major problem of SIMS in HIM is associated with the fact that practically all sputtered particles are sputtered as neutral atoms. In SIMS, this problem is usually circumvented by using reactive elements as the impinging ion. Typically, this is either oxygen or cesium. In HIM, this problem could be solved by using Cs flooding of the sample surface. The enhancement factors between no reactive gas flooding and with reactive gas flooding are as high as \(\mathrm{10^{5}}\), but mostly around \(\mathrm{10^{2}}\) to \(\mathrm{10^{3}}\) [14.161, 14.162]. However, these values are high enough for efficient use of the technique.

A SIMS add-on has been developed which utilizes a magnetic sector analyzer to obtain a maximum \(\Updelta M/M\approx{\mathrm{500}}\) [14.152, 14.163]. First results have been achieved as can be seen in Fig. 14.28. The signals of \(\mathrm{{}^{48}Ti}\) and \(\mathrm{{}^{39}K}\) obtained from a \(\mathrm{TiO_{2}/Au/TiO_{2}}\) photo-catalytic multilayer structure have been combined. A lateral resolution in mass-filtered mode between \(\mathrm{10}\) and \({\mathrm{15}}\,{\mathrm{nm}}\) has been demonstrated with the described setup [14.152, 14.163].

Fig. 14.28

Overlay of \(\mathrm{{}^{48}Ti}\) and \(\mathrm{{}^{39}K}\) obtained from a \(\mathrm{TiO_{2}/Au/TiO_{2}}\) photocatalytic multilayer structure. While \(\mathrm{{}^{48}Ti}\) provides an overview of the structure, the \(\mathrm{{}^{39}K}\) signal shows impurities. From [14.152]

Klingner et al demonstrated that TOF-SIMS can also be implemented in the HIM [14.62]. However, currently, this implementation is inferior to the previous one in terms of lateral, as well as mass, resolution. However, the advantage over the above-discussed magnetic sector design is its lower volume and weight. Consequently, less disturbance of the high-resolution imaging capabilities is expected .

14.4.3 Photons

A limited number of studies have been performed with ionoluminescence inside the helium ion microscope. However, despite the fact that the method is scientifically interesting and enables one to analyze a large number of scientific questions, to date no practical application has been developed. In particular for semiconducting materials, the beam-induced damage to the crystal structure quickly quenches the intrinsic emission from the sample. This has been investigated by Veligura et al [14.126], as well as Boden et al [14.125]. Both studies concluded that the signal that can be obtained is mostly limited to emission from defects . In any case, the signal quenches rapidly, and the intensities are rather small. An interesting exception seems to be semiconducting nanostructures. Particularly if they are smaller than the size of the collision cascade and also freestanding, IL emission can be observed even for very large fluences. Two snapshots from a series of images of nanowires recorded with increasing fluence are presented in Fig. 14.29a-d. Due to the relatively high fluence and poor thermal contact, the shape of the nanowires starts to change after doubling the fluence (compare the top horizontal nanowire in Fig. 14.29a-da,b). Nevertheless, an IL signal with unchanged intensity can be observed. It should be noted that, due to the extremely weak signal in this case, no spectral analysis could be performed. It is likely that the observed emission is not dominated by emission from defects created by the ion beam because most of the defects are formed deep inside the substrate and not in the nanowire itself [14.164]. This result is supported by another study that was able to obtain IL signals from quantum dots but not from bulk materials [14.125].

Fig. 14.29a-d

HIM iSE images and panchromatic IL images obtained from GaP nanowires. The applied fluences are (a,c\({\mathrm{1.4\times 10^{17}}}\,{\mathrm{cm^{-2}}}\) and (b,d\({\mathrm{2.8\times 10^{17}}}\,{\mathrm{cm^{-2}}}\). Reprinted from [14.126], with permission from Elsevier

However, the sensitivity of the method to defects can be utilized for fundamental studies of damage formation and to study the behavior of the created emission centers [14.127, 14.165].

Ionoluminescence of minerals and of samples that contain rare-earth elements has turned out to be the most promising application for IL in HIM. Sapphire, ruby, yttrium aluminum garnet and calcium fluorite have been investigated. In all cases, the emission was found to be related to various defects in the crystal. Ruby showed an intensely sharp emission from R-lines around \({\mathrm{694}}\,{\mathrm{nm}}\) [14.166]. Others such as cerium-activated yttrium aluminum garnet showed a broad emission from \(\mathrm{Ce^{3+}}\) centers that were quenched with increasing fluence [14.125]. Sapphire was exceptionally stable, and luminescence was able to be measured up to fluences close to the amorphization limit. In this case, emission from impurities (\(\mathrm{Cr^{3+}}\)) could also be detected [14.167].

Finally, immunofluorescence has also been tested with both organic [14.168] and inorganic rare-earth-doped particles. Figure 14.30 shows an example obtained on a mouse tooth that has been labeled with an organic fluorophore .

Fig. 14.30

iSE and IL images of Alexa Fluor® 488-tagged mouse tooth. Rows of prisms can be identified (marked by circles). From [14.168]

14.4.4 Structural Information

Channeling is well understood and routinely exploited in ion beam analysis ( ) to obtain information on crystal structure and defects [14.169, 14.170]. The method is very powerful and is capable of identifying the position of interstitial atoms with unprecedented accuracy. It was originally described by Lindhard [14.171] and is usually performed at high energies in the MeV regime. In HIM, with a typical primary ion beam energy of only \({\mathrm{30}}\,{\mathrm{keV}}\), channeling is useful to obtain information on the structure of the near-surface region with high lateral resolution . For helium with energies between 1 and \({\mathrm{100}}\,{\mathrm{keV}}\) (depending on the target material), the acceptance angle for channeling [14.171, 14.172, 14.173, 14.19]
$$\Psi_{2}=\left(\frac{Ca}{d\sqrt{2}}\Psi_{1}\right)^{\frac{1}{2}}$$
(14.13)
depends on the Thomas-Fermi screening length \(a\), a constant \(C=\sqrt{3}\), the spacing between the atoms in the string \(d\), and the critical angle for channeling
$$\Psi_{1}=\sqrt{\frac{U_{\mathrm{a}}(r_{\text{min}})}{E}}\;.$$
(14.14)
Here, \(U_{\mathrm{a}}(r_{\text{min}})\) is the average potential at the distance of the closest approach of an ion with energy \(E\) to a string of atoms [14.173]. An image obtained of a polycrystalline gold film showing channeling contrast is presented in Fig. 14.31. The individual grains can be distinguished from each other as their different orientations result in different channeling conditions. Here, dark means that the crystal lattice is in a channeling condition and has a low index orientation parallel to the beam. Bright areas correspond to grains that are not aligned with the beam, and small impact parameter collisions will occur more often and closer to the surface. This results in an enhanced backscatter yield compared to the channeling case. Please also note that twins are clearly visible in the image.
Fig. 14.31

Image of a polycrystalline gold film obtained using BSHe. Due to channeling the various grains have different backscatter probabilities and can be distinguished from each other. The straight segments observed in some of the grains are twins. The image has been recorded using \({\mathrm{1.11\times 10^{15}}}\,{\mathrm{cm^{-2}}}\) He ions at \({\mathrm{20}}\,{\mathrm{keV}}\) under normal incidence. The FOV is \({\mathrm{15}}\,{\mathrm{\upmu{}m}}\). Reproduced from [14.65]

Fig. 14.32

Surface of a polycrystalline gold film for different azimuthal orientations and a fixed polar angle of \(35^{\circ}\). The image has been recorded using \(\approx{\mathrm{5\times 10^{14}}}\,{\mathrm{cm^{-2}}}\) \({\mathrm{15}}\,{\mathrm{keV}}\) He ions. The FOV is \({\mathrm{10}}\,{\mathrm{\upmu{}m}}\) and the images have been aligned to each other. Reproduced from [14.19]

However, in HIM, most images are not recorded using BSHe but SEs. As is evident by comparing Figs. 14.31 and 14.32, the signal-to-noise ratio for the latter is significantly better despite a lower applied fluence. This is a consequence of the high iSEY in HIM.

Recording channeling-intensity pictures under various polar and azimuth angles, the crystal orientation of the sample surface can be mapped. This makes use of the transparency model originally developed to understand sputtering [14.174, 14.175, 14.176]. This model is based on the following assumptions [14.174]:
  1. 1.

    iSEs are generated due to kinetic electron emission

     
  2. 2.

    Only collisions that occur close to the surface play a role

     
  3. 3.

    The escape probability is independent of the orientation

     
  4. 4.

    The ion–electron emission coefficient is proportional to the probability of a collision.

     
With the exception of (3), these conditions are fulfilled in HIM. The escape probability (3) depends on the work function of the sample, for which HIM is very sensitive  [14.88, 14.89], so care must be taken here. A more detailed discussion of the part relevant for HIM can be found in [14.177] and more generally in [14.178, 14.179]. Using this model, channeling maps like the one presented in Fig. 14.33 can be created. Subsequently, such maps can be used to index the grains in a polycrystalline sample. The method is not limited to HIM, and the obtained results are similar to what can be achieved by using electron backscatter diffraction ( ) in SEM [14.180].
Fig. 14.33

Indexing of crystal orientation. The fcc-channeling map used for indexing the polycrystalline gold film shown in Figs. 14.31 and 14.32. The colors correspond to the in-plane rotation angle of the gold grains. After [14.65]

Using the iSE signal instead of the BSHe signal has the advantage that the method becomes very surface-sensitive. This not only reduces the negative effects of beam-induced damage on the result but also makes possible the analysis of epitaxial relationships between thin surface layers and the underlying bulk material. An example is presented in Fig. 14.34. Using characteristic features on the surface, the sample has been aligned with the \([\overline{11}0]\) direction parallel to the beam and the \([1\overline{1}0]\) direction parallel to the fast-scan direction. In this channeling condition, the He atoms can travel along the fcc lattice with only a small chance for a small impact parameter collision. However, as the sample is covered by two monolayers (ML ) of Ag, parts of the Ag surface have an hcp stacking with the underlying bulk Pt crystal [14.181] (Fig. 14.34c for a sketch of the situation). This results in an enhanced probability for small impact parameter collisions and hence a larger iSEY. The threefold symmetric nature of this fcc/hcp pattern shows up in the HIM image and the corresponding FFT is presented in Fig. 14.34a and b, respectively [14.89].

Fig. 14.34

(a) High-resolution HIM image of the Ag\(/\)Pt(111) surface. Three sets of densely packed parallel lines rotated by \(120^{\circ}\) can be seen. The \([1\overline{1}0]\) direction runs roughly from left to right. (b) FFT showing the regular arrangement of the alternating fcc/bcc areas leading to the line pattern visible in (a). (c) Model showing the samples atoms (Yellow: Ag; Purple: Pt) and the incoming He beam (green). (a,b) reprinted from [14.89], with permission from Elsevier; (c) reproduced from [14.177]

The discussion on the ability of HIM to detect defects at the interface of an only two-atom-layer-thick adlayer underlines the importance of good vacuum conditions. As in the example of the silver layer, HIM will also detect very thin films of hydrocarbons [14.182, 14.93]. While the former is interesting, the latter is mostly unintentional and negatively affects the imaging performance. See also the discussion in Sect. 14.3.1 .

14.5 Conclusion

In the last decade, helium ion microscopy (HIM) has matured into a versatile imaging technology for a number of technological, environmental, and medical or biological questions. While initially limited to pure imaging tasks, using helium, based on the exceptional high brightness of the gas field ion source (GFIS), the method is under active development and has quickly evolved. Today, several alternative gases can be used to enable new contrast mechanisms and to expand the possibilities beyond imaging. In this context, it is important to realize that to date, HIM provides several analytical methods which are complementary to existing analytical methods available in scanning electron microscopy (SEM) or transmission electron microscopy (TEM). The highly focused \(\mathrm{He^{+}}\) or \(\mathrm{Ne^{+}}\) beam with a diameter of only \(\mathrm{0.5}\) and \({\mathrm{1.8}}\,{\mathrm{nm}}\), respectively, has enabled new record resolutions in secondary ion mass spectrometry (SIMS) as well as in Rutherford backscattering spectrometry (RBS). These two methods are special, as they provide unprecedented sensitivity in the case of SIMS and the possibility for standard free quantification on the nanometer scale in the case of RBS. In addition, several analytical approaches closer to what exists in other nanoscale techniques are under active development and show promising results. This includes ionoluminescence (IL) in the case of photons and several transmission-based approaches (scanning transmission ion microscopy (STIM)) which can be compared to TEM or scanning transmission electron microscopy (STEM).

An important asset of HIM is the capacity to image insulating and in particular biological samples. The use of an electron flood gun for charge compensation requires no compromise in terms of resolution or chamber vacuum, which could negatively affect the imaging performance. Several examples have been discussed that highlight this special feature of the technique.

Finally, it should be mentioned that HIM has also redefined the limits for charged particle-based lithography [14.183]. With the introduction of neon as an imaging gas and the integration of a gas injection system ( ), countless applications are possible that go beyond the possibilities of state-of-the-art Ga-based focused ion beam (FIB) systems.

Notes

Acknowledgements

This work is based on the input and hard work of many other people, in particular Vasilisa Veligura (University of Twente, Enschede, The Netherlands) and Nico Klingner (Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany) who worked on IL and TOF, respectively; but also Raoul van Gastel and Bene Poelsema—from the University of Twente in Enschede, the Netherlands—who brought me into the field of HIM. Further, I want acknowledge help from my colleagues Rene Heller, Lothar Bischoff and Stefan Facsko.

References

  1. R.P. Feynman: There's plenty of room at the bottom, Eng. Sci. 23(5), 22–36 (1960)Google Scholar
  2. M. Knoll, E. Ruska: Das Elektronenmikroskop, Z. Phys. 78(5/6), 318–339 (1932)CrossRefGoogle Scholar
  3. H. Boersch: Über hochauflösende Abbildung mittels Ionenstrahlen. (Ionen-Übermikroskopie.), Naturwissenschaften 30(46/47), 711–712 (1942)CrossRefGoogle Scholar
  4. C. Magnan: The proton microscope, Nucleonics 4(4), 52–66 (1949)Google Scholar
  5. P. Chanson, C. Magnan: Sur les premieres images obtenues avec un microscope protonique, C.R. Hebd. Seances Acad. Sci. 233(23), 1436–1438 (1951)Google Scholar
  6. P. Chanson, C. Magnan: Sur le contraste des images en microscopie protonique du a la diffusion elastique et inelastique des protons dans les elements legers, C.R. Hebd. Seances Acad. Sci. 238(17), 1701–1703 (1954)Google Scholar
  7. W.H. Escovitz, T.R. Fox, R. Levi-Setti: Hydrogen ion neutralization: A new contrast mechanism in the scanning transmission ion microscope, Ultramicroscopy 1(3/4), 271–274 (1976)CrossRefGoogle Scholar
  8. R. Levi-Setti: Proton scanning microscopy: Feasibility and promise, Scanning Electron Microsc. 7, 125–134 (1974)Google Scholar
  9. W.H. Escovitz, T.R. Fox, R. Levi-Setti: Scanning transmission ion microscope with a field ion source, Proc. Natl. Acad. Sci. U.S.A. 72(5), 1826–1828 (1975)CrossRefGoogle Scholar
  10. J.H. Orloff, L.W. Swanson: Study of a field-ionization source for microprobe applications, J. Vac. Sci. Technol. 12(6), 1209 (1975)CrossRefGoogle Scholar
  11. R.L. Seliger, J.W. Ward, V. Wang, R.L. Kubena: A high-intensity scanning ion probe with submicrometer spot size, Appl. Phys. Lett. 34(5), 310 (1979)CrossRefGoogle Scholar
  12. J. Morgan, J.A. Notte, R. Hill, B.W. Ward: An introduction to the helium ion microscope, Microsc. Today 14(4), 24–31 (2006)CrossRefGoogle Scholar
  13. B.W. Ward, J.A. Notte, N.P. Economou: Helium ion microscope: A new tool for nanoscale microscopy and metrology, J. Vac. Sci. Technol. B 24(6), 2871 (2006)CrossRefGoogle Scholar
  14. J.A. Notte, F. Rahman, S.M. McVey, S. Tan, R.H. Livengood: The neon gas field ion source-stability and lifetime, Microsc. Microanal. 16(S2), 28–29 (2010)CrossRefGoogle Scholar
  15. R.H. Livengood, S. Tan, R. Hallstein, J.A. Notte, S. McVey, F.H.M. Rahman: The neon gas field ion source—A first characterization of neon nanomachining properties, Nucl. Instrum. Methods Phys. Res. A 645(1), 136–140 (2011)CrossRefGoogle Scholar
  16. D.C. Bell: Contrast mechanisms and image formation in helium ion microscopy, Microsc. Microanal. 15(02), 147–153 (2009)CrossRefGoogle Scholar
  17. G. Hlawacek, V. Veligura, R. van Gastel, B. Poelsema: Helium ion microscopy, J. Vac. Sci. Technol. B 32(2), 020801 (2014)CrossRefGoogle Scholar
  18. D.C. Joy: Helium Ion Microscopy, Springer Briefs in Materials (Springer, New York 2013) p. 64CrossRefGoogle Scholar
  19. G. Hlawacek, A. Gölzhäuser (Eds.): Helium Ion Microscopy, Nano Science and Technology (Springer, Basel 2016) p. 633Google Scholar
  20. M. Miller, A. Cerezo, M. Hetherington, G. Smith: Atom Probe Field Ion Microscopy (Oxford Univ. Press, Oxford 2006)Google Scholar
  21. J.J. Hren, S. Ranganathan (Eds.): Field-Ion Microscopy (Springer, New York 1968) p. 258Google Scholar
  22. E.W. Müller: Das Feldionenmikroskop, Z. Phys. 131(1), 136–142 (1951)CrossRefGoogle Scholar
  23. J.L. Pitters, R. Urban, C. Vesa, R.A. Wolkow: Tip apex shaping of gas field ion sources, Ultramicroscopy 131, 56–60 (2013)CrossRefGoogle Scholar
  24. V.N. Tondare: Quest for high brightness, monochromatic noble gas ion sources, J. Vac. Sci. Technol. A 23(6), 1498 (2005)CrossRefGoogle Scholar
  25. J.H. Orloff, L.W. Swanson: Angular intensity of a gas-phase field ionization source, J. Appl. Phys. 50(9), 6026 (1979)CrossRefGoogle Scholar
  26. K. Horiuchi, T. Itakura, H. Ishikawa: Emission characteristics and stability of a helium field ion source, J. Vac. Sci. Technol. B 6(3), 937 (1988)CrossRefGoogle Scholar
  27. M. Sato: Current–Pressure characteristics of Ar field ion source at high pressures, Jpn. J. Appl. Phys. 31(Part 2, No. 3A), L291–L292 (1992)CrossRefGoogle Scholar
  28. K. Horiuchi, T. Itakura, H. Ishikawa: Fine pattern lithography using a helium field ion source, J. Vac. Sci. Technol. B 6(1), 241 (1988)CrossRefGoogle Scholar
  29. T. Itakura, K. Horiuchi, S. Yamamoto: Focusing column for helium field ion source, Microelectron. Eng. 3(1–4), 153–160 (1985)CrossRefGoogle Scholar
  30. H.-W. Fink: Mono-atomic tips for scanning tunneling microscopy, IBM J. Res. Dev. 30(5), 460–465 (1986)CrossRefGoogle Scholar
  31. T.T. Tsong: Atom-Probe Field Ion Microscopy (Cambridge Univ. Press, Cambridge 1990) p. 400CrossRefGoogle Scholar
  32. R. Börret, K. Jousten, K. Böhringer, S. Kalbitzer: Long time current stability of a gas field ion source with a supertip, J. Phys. D 21(12), 1835–1837 (1988)CrossRefGoogle Scholar
  33. V.T. Binh, J. Marien: Characterization of microtips for scanning tunneling microscopy, Surf. Sci. 202(1/2), L539–L549 (1988)CrossRefGoogle Scholar
  34. V.G. Butenko, Y.A. Vlasov, O.L. Golubev, V.N. Shrednik: Point sources of electrons and ions using microprotrusion on the top of a tip, Surf. Sci. 266(1–3), 165–169 (1992)CrossRefGoogle Scholar
  35. A. Szczepkowicz: Oxygen-covered tungsten crystal shape: Time effects, equilibrium, surface energy and the edge-rounding temperature, Surf. Sci. 605(17/18), 1719–1725 (2011)CrossRefGoogle Scholar
  36. H. Wengelnik, H. Neddermeyer: Oxygen-induced sharpening process of W(111) tips for scanning tunneling microscope use, J. Vac. Sci. Technol. A 8(1), 438–440 (1990)CrossRefGoogle Scholar
  37. C. Vesa, R. Urban, J.L. Pitters, R.A. Wolkow: Robustness of tungsten single atom tips to thermal treatment and air exposure, Appl. Surf. Sci. 300, 16–21 (2014)CrossRefGoogle Scholar
  38. C.-C. Chang, H.-S. Kuo, I.-S. Hwang, T.T. Tsong: A fully coherent electron beam from a noble-metal covered W(111) single-atom emitter, Nanotechnology 20(11), 115401 (2009)CrossRefGoogle Scholar
  39. T.-Y. Fu, L.-C. Cheng, C.-H. Nien, T.T. Tsong: Method of creating a Pd-covered single-atom sharp W pyramidal tip: Mechanism and energetics of its formation, Phys. Rev. B 64(11), 113401 (2001)CrossRefGoogle Scholar
  40. J.L. Pitters, R. Urban, R.A. Wolkow: Creation and recovery of a W(111) single atom gas field ion source, J. Chem. Phys. 136(15), 154704 (2012)CrossRefGoogle Scholar
  41. R. Urban, J.L. Pitters, R.A. Wolkow: Gas field ion source current stability for trimer and single atom terminated W(111) tips, Appl. Phys. Lett. 100(26), 263105 (2012)CrossRefGoogle Scholar
  42. R. Urban, R.A. Wolkow, J.L. Pitters: Single atom gas field ion sources for scanning ion microscopy. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 31–61CrossRefGoogle Scholar
  43. S. Kalbitzer: Optimised operation of gas field ion source, Appl. Phys. A 79(8), 1901–1905 (2004)CrossRefGoogle Scholar
  44. S. Nagai, T. Iwata, R. Okawa, K. Kajiwara, K. Hata: Investigation of stability and charge state of Ne and Ar gas field ion source by time-of-flight mass spectrometry, Jpn. J. Appl. Phys. 53(5), 058004 (2014)CrossRefGoogle Scholar
  45. E.W. Müller, S.B. McLane, J.A. Panitz: Field adsorption and desorption of helium and neon, Surf. Sci. 17(2), 430–438 (1969)CrossRefGoogle Scholar
  46. E.W. Müller, S.V. Krishnaswamy, S.B. Mclane: Atom-probe FIM analysis of the interaction of the imaging gas with the surface, Surf. Sci. 23, 112 (1970)CrossRefGoogle Scholar
  47. H.-S. Kuo, I.S. Hwang, T.-Y. Fu, Y.C. Lin, C.C. Chang, T.T. Tsong: Noble metal/W(111) single-atom tips and their field electron and ion emission characteristics, Jpn. J. Appl. Phys. 45(11), 8972–8983 (2006)CrossRefGoogle Scholar
  48. J.A. Notte, J. Huang: The helium ion microscope. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 3–30CrossRefGoogle Scholar
  49. E.W. Müller, T.T. Tsong: Field Ion Microscopy, Principles and Applications (Elsevier, New York 1969) p. 380CrossRefGoogle Scholar
  50. E.A. Mason, E.W. McDaniel: Appendix III: Tables of properties useful in the estimation of ionneutral interaction energies. In: Transport Properties of Ions in Gases (Wiley-VCH, Weinheim 2005) pp. 531–539CrossRefGoogle Scholar
  51. F. Aramaki, T. Kozakai, O. Matsuda, O. Takaoka, Y. Sugiyama, H. Oba, K. Aita, A. Yasaka: Photomask repair technology by using gas field ion source, Proc. SPIE 8441, 84410D (2012)CrossRefGoogle Scholar
  52. F. Aramaki, T. Ogawa, O. Matsuda, T. Kozakai, Y. Sugiyama: Development of new FIB technology for EUVL mask repair, Proc. SPIE 7969, 79691C (2011)CrossRefGoogle Scholar
  53. M.E. Schmidt, A. Yasaka, M. Akabori, H. Mizuta: Nitrogen gas field ion source (GFIS) focused ion beam (FIB) secondary electron imaging: A first look, Microsc. Microanal. 23(4), 758–768 (2017)CrossRefGoogle Scholar
  54. H.-S. Kuo, I.-S. Hwang, T.-Y. Fu, Y.-H. Lu, C.-Y. Lin, T.T. Tsong: Gas field ion source from an Ir/W\(\langle\)111\(\rangle\) single-atom tip, Appl. Phys. Lett. 92(6), 063106 (2008)CrossRefGoogle Scholar
  55. H. Shichi, S. Matsubara, T. Hashizume: Characteristics of krypton ion emission from a gas field ionization source with a single atom tip, Jpn. J. Appl. Phys. 56(6S1), 06GC01 (2017)CrossRefGoogle Scholar
  56. J.E. Barth, P. Kruit: Addition of different contributions to the charged particle probe size, Optik 101, 101–109 (1996)Google Scholar
  57. R. Hill, F.H.M. Rahman: Advances in helium ion microscopy, Nucl. Instrum. Methods Phys. Res. A 645(1), 96–101 (2011)CrossRefGoogle Scholar
  58. N. Ernst, G. Bozdech, H. Schmidt, W.A. Schmidt, G.L. Larkins: On the full-width-at-half-maximum of field ion energy distributions, Appl. Surf. Sci. 67(1–4), 111–117 (1993)CrossRefGoogle Scholar
  59. J.A. Notte, B.W. Ward, N.P. Economou, R. Hill, R. Percival, L. Farkas, S. McVey: An introduction to the helium ion microscope, AIP Conf. Proc. 931, 489–496 (2007)CrossRefGoogle Scholar
  60. F.W. Martin: A proposal for improved helium microscopy, Microsc. Microanal. 20(5), 1619–1622 (2014)CrossRefGoogle Scholar
  61. F.W. Martin: Particle-beam column corrected for both chromatic and spherical aberration, US Patent 9275817B2 (2013)Google Scholar
  62. N. Klingner, R. Heller, G. Hlawacek, J. von Borany, J.A. Notte, J. Huang, S. Facsko: Nanometer scale elemental analysis in the helium ion microscope using time of flight spectrometry, Ultramicroscopy 162, 91–97 (2016)CrossRefGoogle Scholar
  63. D.C. Bell, M.C. Lemme, L.A. Stern, C.M. Marcus: Precision material modification and patterning with He ions, J. Vac. Sci. Technol. B 27(6), 2755 (2009)CrossRefGoogle Scholar
  64. R. van Gastel, L. Barriss, C.A. Sanford, G. Hlawacek, L. Scipioni, A. Merkle, D. Voci, C. Fenner, H.J.W. Zandvliet, B. Poelsema: Design and performance of a ear ultra high vacuum helium ion Microscope, Microsc. Microanal. 17(S2), 928–929 (2011)CrossRefGoogle Scholar
  65. V. Veligura, G. Hlawacek, R. van Gastel, H.J.W. Zandvliet, B. Poelsema: Channeling in helium ion microscopy: Mapping of crystal orientation, Beilstein J. Nanotechnol. 3, 501–506 (2012)CrossRefGoogle Scholar
  66. L. Bischoff, P. Mazarov, L. Bruchhaus, J. Gierak: Liquid metal alloy ion sources—An alternative for focussed ion beam technology, Appl. Phys. Rev. 3(2), 021101 (2016)CrossRefGoogle Scholar
  67. B. Knuffman, A.V. Steele, J.J. McClelland: Cold atomic beam ion source for focused ion beam applications, J. Appl. Phys. 114(4), 044303 (2013)CrossRefGoogle Scholar
  68. G. Jacob, K. Groot-Berning, S. Wolf, S. Ulm, L. Couturier, S.T. Dawkins, U.G. Poschinger, F. Schmidt-Kaler, K. Singer: Transmission microscopy with nanometer resolution using a deterministic single ion source, Phys. Rev. Lett. 117(4), 1–6 (2016)CrossRefGoogle Scholar
  69. D.A. MacLaren, B. Holst, D.J. Riley, W. Allison: Focusing elements and design considerations for a scanning helium microscope (SHeM), Surf. Rev. Lett. 10(02n03), 249–255 (2003)CrossRefGoogle Scholar
  70. M. Koch, S. Rehbein, G. Schmahl, T. Reisinger, G. Bracco, W.E. Ernst, B. Holst: Imaging with neutral atoms—A new matter-wave microscope, J. Microsc. 229(1), 1–5 (2008)CrossRefGoogle Scholar
  71. M. Barr, A. Fahy, A. Jardine, J. Ellis, D. Ward, D.A. MacLaren, W. Allison, P.C. Dastoor: A design for a pinhole scanning helium microscope, Nucl. Instrum. Methods Phys. Res. B 340, 76–80 (2014)CrossRefGoogle Scholar
  72. A. Fahy, M. Barr, J. Martens, P.C. Dastoor: A highly contrasting scanning helium microscope, Rev. Sci. Instrum. 86(2), 023704 (2015)CrossRefGoogle Scholar
  73. B. Poelsema, G. Comsa: Scattering of Thermal Energy Atoms, Springer Tracts in Modern Physics, Vol. 115 (Springer, Berlin, Heidelberg 1989)CrossRefGoogle Scholar
  74. F. Traeger: Helium atom scattering from oxide surfaces, ChemPhysChem 7(5), 1006–1013 (2006)CrossRefGoogle Scholar
  75. A.P. Jardine, J. Ellis, W. Allison: Quasi-elastic helium-atom scattering from surfaces: Experiment and interpretation, J. Phys. Condens. Matter 14(24), 315 (2002)CrossRefGoogle Scholar
  76. M. Barr, A. Fahy, J. Martens, A.P. Jardine, D.J. Ward, J. Ellis, W. Allison, P.C. Dastoor: Unlocking new contrast in a scanning helium microscope, Nat. Commun. 7, 10189 (2016)CrossRefGoogle Scholar
  77. J.F. Ziegler, J.P. Biersack, M.D. Ziegler: SRIM, The Stopping and Range of Ions in Matter (SRIM, Chester 2008)Google Scholar
  78. H. Demers, N. Poirier-Demers, A.R. Couture, D. Joly, M. Guilmain, N. de Jonge, D. Drouin: Three-dimensional electron microscopy simulation with the CASINO Monte Carlo software, Scanning 33(3), 135–146 (2011)CrossRefGoogle Scholar
  79. R. Ramachandra, B.J. Griffin, D.C. Joy: A model of secondary electron imaging in the helium ion scanning microscope, Ultramicroscopy 109(6), 748–757 (2009)CrossRefGoogle Scholar
  80. K. Ohya, T. Yamanaka, K. Inai, T. Ishitani: Comparison of secondary electron emission in helium ion microscope with gallium ion and electron microscopes, Nucl. Instrum. Methods Phys. Res. B 267(4), 584–589 (2009)CrossRefGoogle Scholar
  81. H.D. Hagstrum: Theory of auger ejection of electrons from metals by ions, Phys. Rev. 96(2), 336–365 (1954)CrossRefGoogle Scholar
  82. J. Ferrón, E. Alonso, R.A. Baragiola, A. Oliva-Florio: Dependence of ion-electron emission from clean metals on the incidence angle of the projectile, Phys. Rev. B 24(8), 4412–4419 (1981)CrossRefGoogle Scholar
  83. H.A. Bethe: Minutes of the Washington, D.C., Meeting May 1–3, 1941, Phys. Rev. 59(11), 913–945 (1941)CrossRefGoogle Scholar
  84. K. Ohya, K. Inai, A. Nisawa, A. Itoh: Emission statistics of x-ray induced photoelectrons and its comparison with electron- and ion-induced electron emissions, Nucl. Instrum. Methods Phys. Res. B 266(4), 541–548 (2008)CrossRefGoogle Scholar
  85. M. Rösler: Theory of ion-induced kinetic electron emission from solids. In: Ionization of Solids by Heavy Particles, ed. by R.A. Baragiora (Plenum, Boston 1993) pp. 27–58CrossRefGoogle Scholar
  86. H. Kudo: Ion-Induced Electron Emission from Crystalline Solids, Springer Tracts in Modern Physics, Vol. 175 (Springer, Berlin, Heidelberg 2002)Google Scholar
  87. Y. Lin, D.C. Joy: A new examination of secondary electron yield data, Surf. Interface Anal. 37(11), 895–900 (2005)CrossRefGoogle Scholar
  88. K. Buchholt, P. Eklund, J. Jensen, J. Lu, A.L. Spetz, L. Hultman: Step-flow growth of nanolaminate Ti3SiC2 epitaxial layers on 4H-SiC(0001), Scr. Mater. 64(12), 1141–1144 (2011)CrossRefGoogle Scholar
  89. G. Hlawacek, M. Jankowski, H. Wormeester, R. van Gastel, H.J.W. Zandvliet, B. Poelsema: Visualization of steps and surface reconstructions in helium ion microscopy with atomic precision, Ultramicroscopy 162, 17–24 (2015)CrossRefGoogle Scholar
  90. Y.V. Petrov, O.F. Vyvenko, A.S. Bondarenko: Scanning helium ion microscope: Distribution of secondary electrons and ion channeling, J. Surf. Investig. 4(5), 792–795 (2010)CrossRefGoogle Scholar
  91. Y. Petrov, O. Vyvenko: Secondary electron emission spectra and energy selective imaging in helium ion microscope, Proc. SPIE 8036, 80360O (2011)CrossRefGoogle Scholar
  92. P. Riccardi, P. Barone, A. Bonanno, A. Oliva, R.A. Baragiola: Angular studies of potential electron emission in the interaction of slow ions with Al surfaces, Phys. Rev. Lett. 84(2), 378–381 (2000)CrossRefGoogle Scholar
  93. G. Hlawacek, I. Ahmad, M.A. Smithers, E.S. Kooij: To see or not to see: Imaging surfactant coated nano-particles using HIM and SEM, Ultramicroscopy 135C, 89–94 (2013)CrossRefGoogle Scholar
  94. Y.V. Petrov, O.F. Vyvenko: Secondary electron generation in the helium ion microscope: Basics and imaging. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 119–146CrossRefGoogle Scholar
  95. R. Ramachandra: A Study of Helium Ion Induced Secondary Electron Production, Ph.D. Thesis (Univ. Tennessee, Knoxville 2009)Google Scholar
  96. R. Timilsina, P.D. Rack: Monte Carlo simulations of focused ion beam induced processing. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 89–118CrossRefGoogle Scholar
  97. R. Ramachandra, B.J. Griffin, D.C. Joy: Modeling for metrology with a helium beam, Proc. SPIE 6922, 69221W–69221W–8 (2008)CrossRefGoogle Scholar
  98. S.A. Boden: Introduction to imaging techniques in the HIM. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 149–172CrossRefGoogle Scholar
  99. G. Behan, D. Zhou, M. Boese, R.M. Wang, H. Zhang: An investigation of nickel cobalt oxide nanorings using transmission electron, scanning electron and helium ion microscopy, J. Nanosci. Nanotechnol. 12(2), 1094–1098 (2012)CrossRefGoogle Scholar
  100. D.S. Fox, Y. Zhou, A. O'Neill, S. Kumar, J.J. Wang, J.N. Coleman, G.S. Duesberg, J.F. Donegan, H. Zhang, A. O'Neill: Helium ion microscopy of graphene: Beam damage, image quality and edge contrast, Nanotechnology 24(33), 335702 (2013)CrossRefGoogle Scholar
  101. A. Gölzhäuser: Helium ion microscopy of carbon nanomembranes. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 225–244CrossRefGoogle Scholar
  102. A. Beyer, H. Vieker, R. Klett, H.M. zu Theenhausen, P. Angelova, A. Gölzhäuser: Imaging of carbon nanomembranes with helium ion microscopy, Beilstein J. Nanotechnol. 6(1), 1712–1720 (2015)CrossRefGoogle Scholar
  103. G. Behan, J.F. Feng, H. Zhang, P.N. Nirmalraj, J.J. Boland: Effect of sample bias on backscattered ion spectroscopy in the helium ion microscope, J. Vac. Sci. Technol. A 28(6), 1377 (2010)CrossRefGoogle Scholar
  104. D.-E. Arafah, J. Meyer, H. Sharabati, A. Mahmoud: Charge-state measurements of backscattered ions from Au films, Phys. Rev. A 39(8), 3836–3841 (1989)CrossRefGoogle Scholar
  105. T.M. Buck, G.H. Wheatley, L.C. Feldman: Charge states of 25–150 keV H and 4He backscattered from solid surfaces, Surf. Sci. 35, 345–361 (1973)CrossRefGoogle Scholar
  106. W. Eckstein, M. Mayer: Rutherford backscattering from layered structures beyond the single scattering model, Nucl. Instrum. Methods Phys. Res. B 153(1–4), 337–344 (1999)CrossRefGoogle Scholar
  107. P. Bauer, E. Steinbauer, J.P. Biersack: Rutherford backscattering beyond the single scattering model, Nucl. Instrum. Methods Phys. Res. B 79(1–4), 443–445 (1993)CrossRefGoogle Scholar
  108. E. Steinbauer, P. Bauer, J. Biersack: Monte Carlo simulation of RBS spectra: Comparison to experimental and empirical results, Nucl. Instrum. Methods Phys. Res. B 45(1–4), 171–175 (1990)CrossRefGoogle Scholar
  109. A. Weber, H. Mommsen: Background in Rutherford backscattering spectra: A simple formula, Nucl. Instrum. Methods Phys. Res. 204, 559–563 (1983)CrossRefGoogle Scholar
  110. A. Weber, H. Mommsen, W. Sarter, A. Weller: Double scattering in Rutherford backscattering spectra, Nucl. Instrum. Methods Phys. Res. 198, 527–533 (1982)CrossRefGoogle Scholar
  111. N. Klingner: Ionenstrahlanalytik im Helium-Ionen-Mikroskop, Ph.D. Thesis (TU Dresden, Dresden 2016)Google Scholar
  112. S. Sijbrandij, J.A. Notte, L. Scipioni, C. Huynh, C.A. Sanford: Analysis and metrology with a focused helium ion beam, J. Vac. Sci. Technol. B 28(1), 73 (2010)CrossRefGoogle Scholar
  113. R. van Gastel, G. Hlawacek, H.J.W. Zandvliet, B. Poelsema: Subsurface analysis of semiconductor structures with helium ion microscopy, Microelectron. Reliab. 52(9/10), 2104–2109 (2012)CrossRefGoogle Scholar
  114. R. Heller, N. Klingner, G. Hlawacek: Backscattering spectrometry in the helium ion microscope: Imaging elemental compositions on the nm scale. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 265–295CrossRefGoogle Scholar
  115. C. Kerkdijk, E.W. Thomas: Light emission induced by H+ and He+ impact on a clean copper surface, Physica 63(3), 577–598 (1973)CrossRefGoogle Scholar
  116. W. Baird, M. Zivitz, E. Thomas: Excited–state formation as H+ and He+ ions scatter from metal surfaces, Phys. Rev. A 12(3), 876–884 (1975)CrossRefGoogle Scholar
  117. W.F. van der Weg, P.K. Rol: On the excited state of sputtered particles, Nucl. Instrum. Methods Phys. Res. 38, 274–276 (1965)CrossRefGoogle Scholar
  118. C.W.W. White: Ion induced optical emission for surface and depth profile analysis, Nucl. Instrum. Methods Phys. Res. 149(1–3), 497–506 (1978)CrossRefGoogle Scholar
  119. N.H. Tolk, I.S.T. Tsong, C.W. White: In situ spectrochemical analysis of solid surfaces by ion beam sputtering, Anal. Chem. 49(1), 16A–30A (1977)CrossRefGoogle Scholar
  120. D. Ghose, R. Hippier: Ionoluminescence. In: Luminescence of Solids, ed. by D.R. Vij (Springer, New York 1998) pp. 189–220CrossRefGoogle Scholar
  121. U. Scherz: Grundlagen der Festkörperphysik. In: Lehrbuch der Experimentalphysik, 2nd edn., ed. by R. Kassing (De Gruyter, Berlin 1992)Google Scholar
  122. G.M. Filippelli, M.L. Delaney: The effects of manganese(II) and iron(II) on the cathodoluminescence signal in synthetic apatite, SEPM J. Sediment. Res. 63, 167–173 (1993)Google Scholar
  123. K. Schwartz: Excitons and radiation damage in alkali halides. In: Atomic Physics Methods in Modern Research, Lecture Notes in Physics, Vol. 1, ed. by K.P. Jungmann, J. Kowalski, I. Reinhard, F. Träer (Springer, Berlin, Heidelberg 1997) pp. 351–366CrossRefGoogle Scholar
  124. V. Veligura, G. Hlawacek, U. Jahn, R. van Gastel, H.J.W. Zandvliet, B. Poelsema: Creation and physical aspects of luminescent patterns using helium ion microscopy, J. Appl. Phys. 115(18), 183502 (2014)CrossRefGoogle Scholar
  125. S.A. Boden, T.M.W. Franklin, L. Scipioni, D.M. Bagnall, H.N. Rutt: Ionoluminescence in the helium ion microscope, Microsc. Microanal. 18(6), 1253–1262 (2012)CrossRefGoogle Scholar
  126. V. Veligura, G. Hlawacek, R. van Gastel, H.J.W. Zandvliet, B. Poelsema: Investigation of ionoluminescence of semiconductor materials using helium ion microscopy, J. Lumin. 157, 321–326 (2015)CrossRefGoogle Scholar
  127. V. Veligura, G. Hlawacek, R. van Gastel, H.J.W. Zandvliet, B. Poelsema: A high resolution ionoluminescence study of defect creation and interaction, J. Phys. Condens. Matter 26(16), 165401 (2014)CrossRefGoogle Scholar
  128. J.I. Goldstein, D.E. Newbury, D.C. Joy, C.E. Lyman, P. Echlin, E. Lifshin, L. Sawyer, J.R. Michael: Scanning Electron Microscopy and X-Ray Microanalysis, 3rd edn. (Springer, Berlin 2003)CrossRefGoogle Scholar
  129. A.J. Pearson, S.A. Boden, D.M. Bagnall, D.G. Lidzey, C. Rodenburg: Imaging the bulk nanoscale morphology of organic solar cell blends using helium ion microscopy, Nano Lett. 11(10), 4275–4281 (2011)CrossRefGoogle Scholar
  130. Z.-Y. Tian, P. Mountapmbeme, P.M. Kouotou, A. El Kasmi, P.H. Tchoua Ngamou, K. Kohse-Höinghaus, H. Vieker, A. Beyer, A. Gölzhäuser: Low-temperature deep oxidation of olefins and DME over cobalt fernite, Proc. Combust. Inst. 35, 2207 (2015)CrossRefGoogle Scholar
  131. P.M. Kouotou, H. Vieker, Z.-Y. Tian, P.T. Ngamou, A.E. Kasmi, A. Beyer, A. Gölzhäuser, K. Kohse-Höinghaus: Structure-activity relation of spinal-type Co-Fe oxides for low-temperature CO oxidation, Catal. Sci. Technol. 4, 3359 (2014)CrossRefGoogle Scholar
  132. P.M. Kouotou, Z.-Y. Tian, H. Vieker, A. Beyer, A. Gölzhäuser, K. Kohse-Höinghaus: Selective synthesis of \(\alpha\)-Fe2O3 thin films and effect of the deposition temperature and lattice oxygen on the catalytic combustion of propene, J. Mater. Chem. A 1, 10495 (2013)CrossRefGoogle Scholar
  133. H. Vieker, A. Beyer: HIM applications in combustion science: Imaging of catalyst surfaces and nascent soot. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 187–203CrossRefGoogle Scholar
  134. S.A. Boden, A. Asadollahbaik, H.N. Rutt, D.M. Bagnall: Helium ion microscopy of Lepidoptera scales, Scanning 34(2), 107–120 (2012)CrossRefGoogle Scholar
  135. H. Guo, H. Itoh, C. Wang, H. Zhang, D. Fujita: Focal depth measurement of scanning helium ion microscope, Appl. Phys. Lett. 023105(2014), 1–5 (2014)Google Scholar
  136. M.S. Joens, C. Huynh, J.M. Kasuboski, D.C. Ferranti, Y.J. Sigal, F. Zeitvogel, M. Obst, C.J. Burkhardt, K.P. Curran, S.H. Chalasani, L.A. Stern, B. Goetze, J.A.J. Fitzpatrick: Helium ion microscopy (HIM) for the imaging of biological samples at sub-nanometer resolution, Sci. Rep. 3, 3514 (2013)CrossRefGoogle Scholar
  137. Y. Zhou, D.S. Fox, H. Zhang: Helium ion microscopy for two-dimensional materials. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 245–262CrossRefGoogle Scholar
  138. H. Guo, J. Gao, N. Ishida, M. Xu, D. Fujita: Characterization of two-dimensional hexagonal boron nitride using scanning electron and scanning helium ion microscopy, Appl. Phys. Lett. 104(3), 031607 (2014)CrossRefGoogle Scholar
  139. G. Nanda, S. Goswami, K. Watanabe, T. Taniguchi, P.F.A. Alkemade: Defect control and n-doping of encapsulated graphene by helium-ion-beam irradiation, Nano Lett. 15(6), 4006–4012 (2015)CrossRefGoogle Scholar
  140. R. Zan, Q.M. Ramasse, R. Jalil, T. Georgiou, U. Bangert, K.S. Novoselov: Control of radiation damage in MoS 2 by graphene encapsulation, ACS Nano 7(11), 10167–10174 (2013)CrossRefGoogle Scholar
  141. G. Algara-Siller, S. Kurasch, M. Sedighi, O. Lehtinen, U. Kaiser: The pristine atomic structure of MoS 2 monolayer protected from electron radiation damage by graphene, Appl. Phys. Lett. 103(20), 203107 (2013)CrossRefGoogle Scholar
  142. M. Kalbac, O. Lehtinen, A.V. Krasheninnikov, J. Keinonen: Ion-irradiation-induced defects in isotopically-labeled two layered graphene: Enhanced in-situ annealing of the damage, Adv. Mater. 25(7), 1004–1009 (2013)CrossRefGoogle Scholar
  143. L. Scipioni, C.A. Sanford, J.A. Notte, B. Thompson, S. McVey: Understanding imaging modes in the helium ion microscope, J. Vac. Sci. Technol. B 27(6), 3250 (2009)CrossRefGoogle Scholar
  144. L. Scipioni, D.C. Ferranti, V.S. Smentkowski, R.A. Potyrailo: Fabrication and initial characterization of ultrahigh aspect ratio vias in gold using the helium ion microscope, J. Vac. Sci. Technol. B 28(6), C6P18 (2010)CrossRefGoogle Scholar
  145. J. Yang, D.C. Ferranti, L.A. Stern, C.A. Sanford, J. Huang, Z. Ren, L.-C. Qin, A.R. Hall: Rapid and precise scanning helium ion microscope milling of solid-state nanopores for biomolecule detection, Nanotechnology 22(28), 285310 (2011)CrossRefGoogle Scholar
  146. A.R. Hall: Solid-state nanopores: From fabrication to application, Microsc. Today 20(05), 24–29 (2012)CrossRefGoogle Scholar
  147. O.K. Zahid, A.R. Hall: Helium ion microscope fabrication of solid-state nanopore devices for biomolecule analysis. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 447–470CrossRefGoogle Scholar
  148. T.J. Woehl, R.M. White, R.R. Keller: Dark-field scanning transmission ion microscopy via detection of forward-scattered helium ions with a microchannel plate, Microsc. Microanal. 22(3), 544–550 (2016)CrossRefGoogle Scholar
  149. J.A. Notte, R. Hill, S.M. McVey, R. Ramachandra, B.J. Griffin, D.C. Joy: Diffraction imaging in a He+ on beam scanning transmission microscope, Microsc. Microanal. 16(05), 599–603 (2010)CrossRefGoogle Scholar
  150. J. Wang, S.H.Y. Huang, C. Herrmann, S.A. Scott, F. Schiettekatte, K.L. Kavanagh: Focussed helium ion channeling through Si nanomembranes, J. Vac. Sci. Technol. B 36(2), 021203 (2018)CrossRefGoogle Scholar
  151. Y.V. Petrov, O.F. Vyvenko: Scanning reflection ion microscopy in a helium ion microscope, Beilstein J. Nanotechnol. 6, 1125–1137 (2015)CrossRefGoogle Scholar
  152. T. Wirtz, D. Dowsett, P. Philipp: SIMS on the helium ion microscope: A powerful tool for high-resolution high-sensitivity nano-analytics. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 297–323CrossRefGoogle Scholar
  153. S. Sijbrandij, J.A. Notte, B. Thompson, C. Huynh, C.A. Sanford, L. Scipioni: A new detector for backscattered helium ions in the 30 keV energy range, Microsc. Microanal. 15(S2), 220–221 (2009)CrossRefGoogle Scholar
  154. S. Sijbrandij, B. Thompson, J.A. Notte, B.W. Ward, N.P. Economou: Elemental analysis with the helium ion microscope, J. Vac. Sci. Technol. B 26(6), 2103 (2008)CrossRefGoogle Scholar
  155. R. van Gastel, G. Hlawacek, S. Dutta, B. Poelsema: Backscattered helium spectroscopy in the helium ion microscope: Principles, resolution and applications, Nucl. Instrum. Methods Phys. Res. B 344, 44–49 (2015)CrossRefGoogle Scholar
  156. M. Mayer: SIMNRA, a simulation program for the analysis of NRA, RBS and ERDA, AIP Conf. Proc. 475, 541–544 (1999)CrossRefGoogle Scholar
  157. N.P. Barradas, C. Jeynes, R.P. Webb: Simulated annealing analysis of Rutherford backscattering data, Appl. Phys. Lett. 71(2), 291–293 (1997)CrossRefGoogle Scholar
  158. P. Sigmund: Theory of sputtering. I. Sputtering yield of amorphous and polycrystalline targets, Phys. Rev. 184(2), 383–416 (1969)CrossRefGoogle Scholar
  159. N. Matsunami, Y. Yamamura, Y. Itikawa, N. Itoh, Y. Kazumata, S. Miyagawa, K. Morita, R. Shimizu, H. Tawara: Energy dependence of the ion-induced sputtering yields of monatomic solids, At. Data Nucl. Data Tables 31(1), 1–80 (1984)CrossRefGoogle Scholar
  160. Y. Yamamura, H. Tawara: Energy dependence of ion-induced sputtering yields from monatomic solids at normal incidence, At. Data Nucl. Data Tables 62(2), 149–253 (1996)CrossRefGoogle Scholar
  161. L. Pillatsch, N. Vanhove, D. Dowsett, S. Sijbrandij, J.A. Notte, T. Wirtz: Study and optimisation of SIMS performed with He+ and Ne+ bombardment, Appl. Surf. Sci. 282, 908–913 (2013)CrossRefGoogle Scholar
  162. D. Dowsett, T. Wirtz, N. Vanhove, L. Pillatsch, S. Sijbrandij, J.A. Notte: Secondary ion mass spectrometry on the helium ion microscope: A feasibility study of ion extraction, J. Vac. Sci. Technol. B 30(6), 06F602 (2012)CrossRefGoogle Scholar
  163. T. Wirtz, P. Philipp, J.-N. Audinot, D. Dowsett, S. Eswara: High-resolution high-sensitivity elemental imaging by secondary ion mass spectrometry: from traditional 2-D and 3-D imaging to correlative microscopy, Nanotechnology 26(43), 434001 (2015)CrossRefGoogle Scholar
  164. V. Veligura, G. Hlawacek, R.P. Berkelaar, R. van Gastel, H.J.W. Zandvliet, B. Poelsema: Digging gold: keV He+ ion interaction with Au, Beilstein J. Nanotechnol. 4, 453–460 (2013)CrossRefGoogle Scholar
  165. S. Ogawa, T. Iijima, S. Awata, R. Sugie, N. Kawasaki, Y. Otsuka: Characterization of damage in SiO2 during helium ion microscope observation by luminescence and TEM-EELS, Microsc. Microanal. 18(S2), 814–815 (2012)CrossRefGoogle Scholar
  166. V. Veligura, G. Hlawacek: Ionoluminescence. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 325–351CrossRefGoogle Scholar
  167. V. Veligura: Material Characterization and Modification Using Helium Ion Microscopy: Various Examples, Ph.D. Thesis (Univ. Twente, Enschede 2014)Google Scholar
  168. T.M.W. Franklin: Scanning Ionoluminescence Microscopy with a Helium Ion Microscope, Ph.D. Thesis (Univ. Southampton, Southampton 2012), p. 198Google Scholar
  169. I. Stensgaard: Surface studies with high-energy ion beams, Rep. Prog. Phys. 55(7), 989–1033 (1992)CrossRefGoogle Scholar
  170. L.C. Feldman, J.W. Mayer, S.T.A. Picraux: Materials Analysis by Ion Channeling: Submicron Crystallography (Academic Press, New York 1982) p. 300Google Scholar
  171. J. Lindhard: Influence of crystal lattice on motion of energetic charged particles, Mat. Fys. Medd. Dan. Vidensk. Selsk. 34(14), 1–64 (1965)Google Scholar
  172. A.A. van Gorkum: Channeling computations for low energy (25–2500 eV) helium in tungsten, Phys. Lett. A 75(1/2), 134–136 (1979)CrossRefGoogle Scholar
  173. M. Nastasi, J. Mayer, Y. Wang: Ion Beam Analysis (CRC, Boca Raton 2014) p. 472CrossRefGoogle Scholar
  174. G.D. Magnuson, C.E. Carlston: Electron ejection from metals due to 1- to 10-keV noble gas ion bombardment. I. Polycrystalline Materials, Phys. Rev. 129(6), 2403–2408 (1963)CrossRefGoogle Scholar
  175. M.T. Robinson: Theoretical aspects of monocrystal sputtering. In: Sputtering by Particle Bombardment, ed. by R. Behrisch (Springer, Berlin 1981) pp. 73–144CrossRefGoogle Scholar
  176. U. Von Gemmingen: Ion induced secondary electron emission from single crystal surfaces, Surf. Sci. 120(2), 334–345 (1982)CrossRefGoogle Scholar
  177. G. Hlawacek, V. Veligura, R. van Gastei, B. Poelsema: Channeling and backscatter imaging. In: Helium Ion Microscopy, ed. by G. Hlawacek, A. Gölzhäuser (Springer, Basel 2016) pp. 205–224CrossRefGoogle Scholar
  178. M.A. Karolewski, R.G. Cavell: Modelling the directional and energy dependence of 5-10 keV Ar+ ion-induced secondary electron yields from Cu crystals, Surf. Sci. 605(19/20), 1842–1851 (2011)CrossRefGoogle Scholar
  179. M. Nègre, J. Mischler, N. Bénazeth, C. Noguera, D. Spanjaard: Diffraction effects in the angular distribution of secondary electrons emitted under ionic bombardment of a (111) surface of silver: Experimental results and theoretical model, Surf. Sci. 78(1), 174–180 (1978)CrossRefGoogle Scholar
  180. C. Langlois, T. Douillard, H. Yuan, N.P. Blanchard, A. Descamps-Mandine, B. Van de Moortèle, C. Rigotti, T. Epicier: Crystal orientation mapping via ion channeling: An alternative to EBSD, Ultramicroscopy 157, 65–72 (2015)CrossRefGoogle Scholar
  181. M. Jankowski, H. Wormeester, H.J.W. Zandvliet, B. Poelsema: Temperature-dependent formation and evolution of the interfacial dislocation network of Ag/Pt(111), Phys. Rev. B 89(23), 235402 (2014)CrossRefGoogle Scholar
  182. G. Hlawacek, V. Veligura, S. Lorbek, T.F. Mocking, A. George, R. van Gastel, H.J.W. Zandvliet, B. Poelsema: Imaging ultra thin layers with helium ion microscopy: Utilizing the channeling contrast mechanism, Beilstein J. Nanotechnol. 3, 507–512 (2012)CrossRefGoogle Scholar
  183. M.G. Stanford, B.B. Lewis, K. Mahady, J.D. Fowlkes, P.D. Rack: Review article: Advanced nanoscale patterning and material synthesis with gas field helium and neon ion beams, J. Vac. Sci. Technol. B 35(3), 030802 (2017)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.Institute for Ion Beam Physics & Materials ResearchHelmholtz Zentrum Dresden RossendorfDresdenGermany

Personalised recommendations