1 Introduction

In-situ electron microscopy is a fast-growing and leading field in the materials research. It is an indispensable tool in understanding the material behavior in real-time. Additionally, the rapid miniaturization of the components for technological devices, that approaches micron and nanoscale dimensions, requires accurate size-specific properties (Gianolaet 2011) as it has been shown that the properties at a small length scale can be significantly different from the bulk counterpart (Dimiduk et al. 2005; Uchic et al. 2004; Volkert et al. 2008). Besides, in the multi-phase materials, the properties measured at a small length scale can be used as input to carry out microstructure-based modeling, which further assists in understanding the bulk deformation behavior (Guo et al. 2014).

To understand the material behavior in real-time, and at small length scales, a number of improvements in the microscopy techniques have been made over the years. Several two-dimensional (2D) microscopy techniques [e.g., optical, Raman, scanning electron microscopy (SEM), electron backscattered diffraction (EBSD), electron channeling contrast imaging (ECCI)] and three-dimensional (3D) techniques [e.g., transmission electron microscopy (TEM), X-ray tomography, neutron imagining, atom probe tomography (APT), focused ion beam (FIB) tomography] cover a range of length scales (Singh et al. 2014a,2015,2017; Gianolaet 2011; Volkert et al. 2008; Guo et al. 2014; Legros et al. 2010). Among these, the electron microscopy technique is well suited to elucidate the microstructural changes, with time, for micron- and submicron-sized specimens.

In-situ electron microscopy can be used to understand different phenomena depending upon the attachments equipped in SEM. For instance, a customized or commercial mechanical testing jig assists in unraveling the complex deformation mechanisms (Singh et al. 2014a). Further, EBSD and digital image correlation (DIC) techniques can assist in understanding the strain partitioning between different phases or grains in the materials. Novel microscopy techniques, such as electron channeling contrast imaging (ECCI) (Slama et al. 2019), can further assist in the quantification of the deformation systems generated under the application of load. FIB machining and deposition enable the sample preparation, which aids in evaluating the site-specific properties at small length scales using specialized attachments, such as nanoindentation system in a SEM (Sarvesha et al. 2020; Guo et al. 2014; Singh et al. 2014b). In recent years, irradiation properties of the nuclear materials have been studied by irradiating different ions, such as Fe+ , W+, He+, on the microscale test structures in assistance with FIB (Armstrong et al. 2015). Furthermore, a couple of in-situ SEM studies have also been carried out to evaluate the corrosion properties in environmental SEM (Proff et al. 2010), where a special in-situ jig was used to simulate the corrosion environment (Kim et al. 2020).

A powerful way of capturing real-time deformation microstructure is acoustic emission (AE) along with the conventional in-situ measurements. The AE method is based on sound emitted due to the activation of deformation mechanisms or fracture during deformation and thus provides highly time-resolved information (Wisner et al. 2015).

Figure 1 summarizes the various in-situ tests performed in SEM. In-situ tests in combination with SE/BSE imaging and EBSD in SEM are the most attractive methods used to understand the deformation behavior. This article discusses a few case studies showing the use of in-situ electron microscopy in evaluating the deformation behavior of metallic systems (Sn whisker, 7075 aluminum alloys, and AZ80 magnesium alloy).

Fig. 1
figure 1

Summary of the in-situ test performed in a scanning electron microscope

Full size image

2 Case Studies

2.1 Case Study I: In-Situ Tensile Behavior of Sn Whiskers

Environment-friendly Pb-free solders are predominantly based on Sn-rich alloys, where whiskers are known to grow on the Sn plates (Mathew et al. 2009). An understanding of the mechanical response of Sn whisker is vital for the soldering community. Whiskers are single crystals and contain minimal defects, which assists in the free flow of electron, thus making them an excellent conductor. The phenomenon of Sn whisker formation has been studied for decades (Galyon 2005); however, only a few studies have been performed to evaluate the mechanical properties of Sn whiskers (Dunn 1987; Powell and Skove 2004). This is due to adversity involved in handling whiskers. This case study deals with the in-situ evaluation of tensile properties of Sn whisker using a micro–electro–mechanical system (MEMS) tensile stage in assistance with a focused ion beam (FIB) (Singh et al. 2014a).

Sn whiskers on a Sn plate, having a very high aspect ratio, are shown in Fig. 2a. The steps of in-situ sample preparation using FIB are shown in Fig. 2. Manipulator needle was brought close to the base of the Sn whisker (Fig. 2b) and carefully welded using Pt (Fig. 2c). Sn whisker was then cut using Ga ions, as shown in Fig. 2d. The whisker was then brought to the MEMS device and welded in the trenches using Pt, as shown in Fig. 2e–g. In the end, the needle was released by cutting the whisker using Ga ions (Fig. 2h).

Fig. 2
figure 2

a SEM image of Sn whiskers. Steps of whisker lift off and placement in MEMS inside FIB/SEM, b manipulator needle in contact with whisker, c welding of whisker using Pt filler, d FIB cutting of the whisker, e–g placement and welding of whisker in the trenches, and h FIB cutting to release the needle (Singh et al. 2014a)

Full size image

The MEMS device is shown in Fig. 3a (details of construction of the MEMS stage are reported elsewhere (Han and Saif 2006; Han et al. 2007)), and welded whisker on it is shown in Fig. 3b. A quasistatic loading was applied during the tensile test until fracture. The fractured Sn whisker is shown in Fig. 3c. The magnified image of the whisker shows that its surface is irregular. Therefore, to evaluate the stress, the surface area was calculated using image processing software (ImageJ), as shown in Fig. 3d. Two successful tests were performed, and in both the case linear stress–strain behavior was observed for the Sn whisker (Fig. 3e). The values of Young’s modulus of both the whiskers were measured to be 42 and 45 GPa, which are close to the theoretical value of the modulus for pure Sn. The fracture strength and strain to failure for the two tests were 720 MPa, 880 MPa, and 2%, 3%, respectively. The observed high fracture strength and limited ductility may be attributed to a lack of defects in the single crystal Sn whiskers.

Fig. 3
figure 3

a Optical micrograph of MEMS stage for tensile testing of whisker, b zoomed image of Sn whisker before loading, c fracture Sn whisker, d SE micrograph of fractured surface and inset is a binary image of the fractured surface, e nominal stress–strain in the behavior of Sn whisker (Singh et al. 2014a)

Full size image

Limited ex-situ experiments have been performed to evaluate the mechanical properties of Sn whiskers (Dunn 1987; Powell and Skove 2004). Powell and Skove (2004) observed the linear load–displacement behavior. However, the properties of the whiskers were not reported. Dunn (1987) obtained Young’s modulus using the cantilever method. However, a large variability (8–85 GPa) was observed. One of the crucial shortcomings in Dunn’s study was the measurement of the cross-sectional area in which the whisker was assumed to have a circular cross section. However, it should be noted that in the bending theory, Young’s modulus varies with the fourth power of the radius, and thus, a small change in the radius would result in a significant deviation in the measurement of modulus. In comparison with the ex-situ testing, which had inherent limitations in handling the whiskers, the in-situ tensile testing in SEM assisted in evaluating the reliable mechanical properties of the Sn whisker.

2.2 Case Study II: Evaluation of Mechanical Properties of the Constituent Particles in Al 7075 by Micro-pillar Compression

Aluminum alloys are predominantly used as structural materials due to their high strength-to-weight ratio (Bucci et al. 2000; Starke and Staley 2011). These alloys contain second-phase particles (Fe-containing: Al7Cu2Fe and Si-containing: Mg2Si), also called as constituent particles, which form during casting and do not dissolve in the subsequent thermomechanical treatment of the alloy (Payne et al. 2010; Xue et al. 2007). These particles have been observed to affect the overall deformation behavior of the alloy as they are the source for crack nucleation and propagation during static or dynamic loading (Pearson 1975; Weiland et al. 2009; Maire et al. 2011; Lugo et al. 2011). Hence, the overall mechanical response of the Al7075 alloy depends on the characteristic of constituent particles. This present case study deals with the evaluations of the stress–strain behavior of the constituent particles under uniaxial compression of micro-pillars (Singh et al. 2015).

The SEM micrograph (Fig. 4a) shows two types of constituent particles (one black and another white). The corresponding EDS analysis (Fig. 4b) indicates that the bright particle is Fe-bearing inclusions and the dark particle is Si-bearing inclusion. The approximate atomic ratio of Al, Cu, and Fe in Fe-bearing inclusion is 7:2:1, indicating them to be Al7Cu2Fe. Similarly, for Si-bearing inclusion, the atomic ratio of Mg and Si is 2:1, indicating them to be Mg2Si. The micro-pillar fabrication sequence for Mg2Si using dual-beam focused ion beam (FIB) is shown in Fig. 4c–f. The same procedure was followed for Al7Cu2Fe and Al7075 matrix. Initially, a circular trench of ~25 µm diameter was milled out (shown in Fig. 4c), which provides clearance for the flat punch during compression. A course pillar with ~7 µm was made at the inclusion center (refer to Fig. 4d). Further, the pillar diameter was reduced to ~3.5 µm (Fig. 4e), and milling was continued until an aspect ratio of 2.7–3.4 was maintained. During the final process, a low current was utilized, so the taper is less than 3° (shown in Fig. 4f).

Fig. 4
figure 4

a SEM image showing Fe- and Si-bearing inclusions in Al7075, b EDS analysis of the second-phase particles, and c–f micro-pillar fabrication using dual-beam FIB

Full size image

The SEM micrographs before loading and after fracture of the pillars are shown in Fig. 5a1, b1, c1 and a2, b2, c2, respectively. The stress–strain behavior of the constituent particles and the matrix of the repeat test is shown in Fig. 5d. Linear stress–strain behavior is observed till fracture for the Al7Cu2Fe particles. The fracture stress is ~2.5 ± 0.2 GPa, and strain is approximately 2%. SEM micrograph of the fractured pillar (Fig. 5a2) indicates catastrophic failure. In the case of the Mg2Si particle, the stress value increased linearly with strain in the initial region, followed by strain bursts. The significant strain hardening is due to the dislocation activity, which is also visible in the SEM micrograph (steps indicated by the arrow in Fig. 5b2). The compressive yield strength of Mg2Si was calculated to be ~1.8 ± 0.1 GPa. The Al7075 matrix has similar stress–strain behavior as Mg2Si with a lower yield strength of ~0.5 ± 0.1 GPa. The strength values are in-line with the hardness value obtained by nanoindentation (Kim et al. 2020), i.e., the highest hardness for Al7Cu2Fe particle followed by Mg2Si particle and Al7075 matrix.

Fig. 5
figure 5

a1, b1, c1 and a2, b2, c2 show Al7Cu2Fe, Mg2Si, Al7075 micro-pillars before loading and after fracture, respectively, and (d) stress–strain behavior of the second-phase particles and the matrix (Singh et al. 2015)

Full size image

Previously, studies have been conducted on the evaluation of mechanical properties of Al7Cu2Fe and Mg2Si by fabricating bulk samples through the powder metallurgy route. While both Al7Cu2Fe and Mg2Si were found to be brittle (Laplanche et al. 2014; Wang et al. 2007; Muñoz-Palos et al. 1996; Takeuchi et al. 1996), limited plasticity for Mg2Si was observed in this study. This difference in behavior could be attributed to the difference in microstructure like grain size, defect concentration, and type (such as residual porosity), etc. It has been recently shown in several micro-pillar compression studies that a brittle material (in bulk) can show plastic deformation at a small length scale (Östlundet 2009; Michler et al. 2007; Korte and Clegg 2011; Östlund 2011). Hence, reducing the size of the sample suppresses cracking, resulting in plastic deformation when the test sample is made significantly small (Östlundet 2009). Oswald (2003) evaluated the hardness and Young’s modulus using nanoindentation and predicated stress–strain behavior. Even though the highest strength and modulus was reported for Al7Cu2Fe, a similar yield strength was predicated for Al-matrix and Mg2Si due to their similar hardness values. However, the present study shows that Mg2Si exhibits higher yield strength as compared to the matrix, which corroborates with the hardness values reported in other studies (Singh et al. 2014c).

In summary, the mechanical properties of the constituent particles were evaluated by micro-pillar compression testing, the knowledge of which will allow a better understanding of the bulk deformation behavior of Al7075 alloys.

2.3 Case Study III: Role of Second-Phase Particles on the Deformation Behavior of AZ80 Magnesium Alloys

The role of second-phase particles on the deformation behavior of AZ80 alloy has been divided into two separate studies. One study (Sarvesha et al. 2021) deals with understanding the role of these particles on the fracture process of the alloy using in-situ tensile testing, and another study (Sarvesha et al. 2020) is on the measurement of mechanical properties of the particles using micro-pillar compression.

2.3.1 Effect of Second-Phase Particle on Crack Propagation in Magnesium alloy via in-situ Tensile Testing

Magnesium and its alloys exhibit the highest strength-to-weight ratio, which assists in a weight reduction, thus have potential in the automobile, aerospace, and electronics industries (Kulekci 2008; Blawert et al. 2004). Mg-Al-Zn alloys, designated as AZ series alloys, which are a prominent class of magnesium alloys, contain Mg-Al precipitate and Al-Mn inclusions, (Zindal et al. 2018,2017). These second-phase particles have been found to affect the mechanical properties of magnesium alloys significantly (Lugo et al. 2011; Yakubtsov et al. 2008).

A typical as-cast optical micrograph (OM) of Mg-9.3Al-0.45Zn-0.15Mn is shown in Fig. 6a, which shows two types of micron-sized second-phase particles. The corresponding backscattered electron (BSE) image (Fig. 6b) shows gray and white particles due to difference in composition. The energy dispersive spectroscopy (EDS) map (Fig. 6c) shows that the gray particle is enriched with Mg and Al with Zn segregation. In contrast, the white particle is enriched with Mn and Al with the segregation of Fe and Si. Furthermore, the optical microscopic image (dark field mode) shows that Mg-Al precipitate has porosities, which is called a discontinuous eutectic precipitate. In addition, additional dark regions (indicated by arrows in Fig. 6a) correspond to continuous precipitates, which are formed adjacent to eutectic precipitate (refer SE micrograph in Fig. 6b). Electron backscatter diffraction (EBSD) maps suggest that the micron-sized Mg-Al precipitate corresponds to Mg17Al12 (space group \({\text{I}}\overline{4}3{\text{m}}\)), which are polycrystalline and forms along the grain boundaries of the α-Mg matrix, whereas the Al-Mn inclusion corresponds to Al8Mn5 particle (space group \({\text{ R}}3{\text{m}}\)) and has a rhombohedral crystal structure.

Fig. 6
figure 6

a Optical micrograph (OM)-bright filed (BF) image, b SE micrograph, c EDS map of Mg, Al, Zn, Mn and Fe. d, e EBSD maps of Mg17Al12/α-Mg matrix and Al8Mn5 particle, respectively

Full size image

Gatan in-situ tensile testing stage, along with the sample dimension, are shown in Fig. 7a, b, respectively. The sequence of secondary electrons (SE) micrographs at different strain (ɛ) levels (refer Fig. 7c) shows that crack propagation occurred predominantly through Mg17Al12 precipitates. The visible crack appeared at strain (ɛ) 0.0376. To further ascertain the strain at which the crack initiates in Mg17Al12, in-situ testing was performed with a focus on a single Mg17Al12 particle, as shown in Fig. 8a. It can be observed that at ɛ = 0.017, the crack initiates in the second-phase particle, which is in the linear portion of the stress–strain curve (shown in inset). With an increase in strain, more cracks appear perpendicular to the loading direction, as indicated by arrows in Fig. 8a.

Fig. 7
figure 7

a In-situ tensile testing stage, b sample dimension, c SEM images showing crack propagation through Mg17Al12 precipitates (arrow indicate reference region) in an as-cast sample

Full size image
Fig. 8
figure 8

a SEM micrographs of in-situ tensile testing of Mg17Al12 at different strain levels along with the stress–strain diagram. SE micrograph and corresponding area IPF-X and KAM map b before loading and c after fracture. (Loading is along X direction). (For legends, refer Fig. 6)

Full size image

Further, to understand the strain partitioning between the matrix and the precipitate, in-situ tensile testing, assisted with EBSD, was performed. A 70° pre-tilt in-situ stage (shown in Fig. 7a) was employed for EBSD scanning. The secondary electron (SE) micrographs (at 70° tilt) along with the EBSD maps, before loading and after fracture, are shown in Fig. 8b, c, respectively. Kernel average misorientation (KAM) maps in Fig. 8b, c are generated from EBSD scanning. KAM calculates the average misorientation between each pixel and its nearest neighbors and indicates strain in different phases during external loading. KAM map in Fig. 8b shows no strain in the matrix as well as in the particle before loading. However, after fracture, the strain is predominately distributed in the matrix, as the matrix shows higher KAM values (Fig. 8c). The KAM value of the precipitate is approximately zero, indicating the Mg17Al12 fractured in a brittle manner.

Previously, Yakubtsov et al. (2008) observed an increase in the ductility of AZ80 (Mg-8.2Al-0.51Zn) alloy with a decrease in the volume fraction of Mg17Al12precipitate. Lü et al. (2000) observed the crack formation near Mg17Al12 precipitate during tensile deformation, and it was postulated that the crack initiated at the interface between Mg/Mg17Al12. In the present study, it is evident from in-situ tensile testing, fracture initiates at Mg17Al12 at low strain and propagates through the Mg17Al12 phase.

2.3.2 Mechanical Property Evaluation of Second-Phase Particle in Magnesium Alloy by Micro-pillar Compression

It is clear from the previous study (Sect. 2.3.1) that the second-phase particles significantly affect the deformation characteristics of magnesium alloys. Therefore, it is essential to understand the mechanical properties of the individual particle. However, due to the small size of these particles, their mechanical properties cannot be evaluated by conventional techniques. Thus, this study deals with the evaluation of mechanical properties (stress–strain behavior) of these second-phase particles using in-situ micro-pillar compression (Sarvesha et al. 2020).

The micro-pillar fabrication process has already been discussed in Case II. The undeformed and deformed pillars of the matrix and second-phase particles are shown in Fig. 9, and the stress–strain behavior is shown in Fig. 10a. During compression, slip bands form on the α-Mg matrix surface (shown in Fig. 8a1 inset), which manifests as strain bursts in stress–strain curves (Fig. 10a)—a general characteristic of ductile metals/alloy (Schneider et al. 2013). In contrast, the stress–strain behavior of Mg17Al12 showed linear behavior until a massive strain burst (Fig. 10a). The corresponding SE micrograph shows a noticeable change in the shape of the Mg17Al12 pillar (Fig. 9b2), which could be due to strong strain localization without dislocation movement (Song et al. 2018). Further, it is evident from the stress–strain behavior of Al8Mn5 (Fig. 10a) particle that it exhibits the highest strength and ductility. SEM micrograph of the deformed Al8Mn5 pillar (Fig. 9c2) shows slip lines, which manifest as load drops in stress–strain curves (Fig. 10a). Al8Mn5 particle exhibits the highest yield strength of ~3084 ± 630 MPa, followed by Mg17Al12 with ~814 ± 141 MPa, and α-Mg matrix with ~134 ± 22 MPa.

Fig. 9
figure 9

a1, b1, c1 and a2, b2, c2 before and after deformation of α-Mg matrix, Mg17Al12, and Al8Mn5 micro-pillar, respectively (Sarvesha et al. 2020)

Full size image
Fig. 10
figure 10

a Stress–strain behavior of the matrix and particles, b TEM-BF, and c SADP of deformed Al8Mn5 pillar. d–f are the SE micrographs, AFM images, and line profile of the indents made using nanoindentation on Al8Mn5, Mg17Al12, and α-Mg matrix, respectively (Sarvesha et al. 2020)

Full size image

The site-specific transmission electron microscopy (TEM) of Al8Mn5 deformed micro-pillar shows the dislocation structure (shown by the arrow in Fig. 10b). It can be noticed that one type of dislocation structure is queued on the slip plane. The selective area diffraction pattern (SADP) along the \([100]\) zone axis is shown in Fig. 10c. The angle between \(\left( {01\overline{1}} \right)\) and \(\left( {021} \right)\) is around 88.7°, which corresponds to the rhombohedral crystal structure (Zeng et al. 2018).

Additionally, to verify the deformation behavior of each phase, qualitative analysis was also performed using nanoindentation. The indentation was performed at 60 mN load for the Al8Mn5 particle and 35 mN load for the Mg17Al12 particle and the matrix. SEM micrographs, atomic force microscopy (AFM) images, and corresponding line profiles of the indents made on the Al8Mn5 particle, the Mg17Al12 particle, and the matrix are shown in Fig. 10d–f, respectively. The line profiles along the corner (C) to edge (E) were used to observe the pile-up and sink-in phenomenon. Line profiles for the Al8Mn5 particle clearly show pile-up along all the edges, whereas the Mg17Al12 particle does not show pile-up or sink-in along the edges. The α-Mg matrix shows both pile-up and sink-in (indicated in AFM in Fig. 10f). In addition, both pile-up and sink-in were visible along the same edge of the indent (shown in SE micrograph in Fig. 10f). It is already known that the plastically deformable materials show pile-up characteristics (Fischer-Cripps 2004). The observed pile-up’s in the case of Al8Mn5 particle corroborate with the observation made previously, i.e., the Al8Mn5 particles exhibit ductility. In contrast, the absence of pile-up and sink-in in the case of Mg17Al12 particles indicates the absence of plastic deformation, which is also evident from micro-pillar compression (Fig. 9).

A couple of studies are available for the bulk deformation of Mg17Al12 (Ragani et al. 2011; Fukuchi and Watanabe 1980). However, no study has been carried out to evaluate the stress–strain behavior of Al8Mn5 particles. Ragani et al. (2011) fabricated bulk Mg17Al12 through levitation casting and measured the room temperature compressive fracture strength and strain to be ~325 MPa and ~2%, respectively. Fukuchi and Watanabe (1980) evaluated the bulk properties of Mg17Al12 at elevated temperature (between 360 and 435 °C), where particles were found to exhibit ductility. However, it should be noted that both the studies were carried out in the bulk form, which can show different properties than that at a small length scale. In fact, it has been shown that the mechanical response at large and small length scales might be different (Singh et al. 2015; Michler et al. 2007; Korte and Clegg 2011). In the present case, interestingly, the Mg17Al12 particle exhibits distinct behavior at room temperature, i.e., sudden failure after a significant strain burst, unlike the brittle nature reported in the bulk samples (Ragani et al. 2011).

3 Summary

The mechanical properties of materials change significantly with a reduction in size. Due to the inherent problems associated the conventional mechanical testing in understanding the deformation behavior of materials at a small length scale, in-situ electron microscopy has gained interest in recent years. The advancement in instrumentation in the electron microscopes, such as FIB and nanoindentation in SEM, has made it possible to understand the deformation behavior in real time. Also, techniques such as EBSD/ECCI/DIC assist in the quantification of the deformation characteristics of the materials. The present article shows the importance of electron microscopy in understanding the deformation behavior of metallic systems by using three case studies.