# Fileset

[Microscopy_MS_0829final_MDR.pdf](https://mdr.nims.go.jp/filesets/f57fee31-c269-46f5-848a-ad40294af639/download)

## Creator

Toshie Yaguchi, Mia L San Gabriel, [Ayako Hashimoto](https://orcid.org/0000-0002-1985-7667), Jane Y Howe

## Rights

[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[<i>In-situ</i> TEM study from the perspective of&nbsp;holders](https://mdr.nims.go.jp/datasets/5b2237e2-e1c4-4da9-a3f1-83bc28f5845a)

## Fulltext

1 In-situ TEM study from the perspective of holders 1 Toshie Yaguchi1,*, Mia San Gabriel2, Ayako Hashimoto3,4, and Jane Y. Howe2,5 2  3 1Electron Microscope Systems Design Department, Hitachi High-Tech Corporation, 552-53 4 shinko-cho,Hitachinaka-shi, Ibaraki-ken,312-8504, Japan,  5 2Department of Materials Science and Engineering, University of Toronto, 184 College St, 6 Toronto, ON, Canada. 7 3In-situ Electron Microscopy Technique Development Group, National Institute for Materials 8 Science, 1-2-1 Sengen, Tsukuba 305-0047, Japan.  9 4Degree Programs in Pure and Applied Sciences, University of Tsukuba, 1-2-1 Sengen, 10 Tsukuba 305-0047, Japan. 11 5Department of Chemical Engineering and Applied Chemistry, University of Toronto, 200 12 College St, Toronto, ON, Canada. 13  14 *Correspondence should be addressed to 15 Toshie Yaguchi1 , 1Electron Microscope Systems Design Department, Hitachi High-Tech 16 Corporation, 552-53 shinko-cho,Hitachinaka-shi, Ibaraki-ken,312-8504, Japan  17 Phone: +81-80-1150-1135  18 E-mail: toshie.yaguchi.yy@hitachi-hightech.com 19  20 Running title: Holders on in-situ TEM study  21 Keywords: In-situ transmission electron microscopy, Differential pumping, Closed（sealed）22 gas cell, Environmental cell, MEM, ETEMS 23 Total Number of Pages: 72 24 Number of Figures: 13 25   26 mailto:toshie.yaguchi.yy@hitachi-hightech.com2 Abstract 27 During the in-situ TEM observations, the diverse functionalities of different specimen holders 28 play a crucial role. We hereby provide a comprehensive overview of the main types of 29 holders, associated technologies, and case studies pertaining to the widely employed heating 30 and gas heating methods, from their initial developments to the latest advancement. In 31 addition to the conventional approaches, we also discuss the emergence of holders that 32 incorporate a micro-electro-mechanical system (MEMS) chip for in-situ observations. The 33 MEMS technology offers a multitude of functions within a single chip, thereby enhancing the 34 capabilities and versatility of the holders. 35 MEMS chips have been utilized in the environmental-cell designs, enabling customized 36 fabrication of diverse shapes. This innovation has facilitated their application in conducting 37 in-situ observations within gas and liquid environments, particularly in the investigation of 38 catalytic and battery reactions. We summarize recent noteworthy studies conducted using in-39 situ liquid TEM. These studies highlight significant advancements and provide valuable 40 insights into the utilization of MEMS chips in environmental-cells, as well as the expanding 41 capabilities of in-situ liquid TEM in various research domains. 42  43 Abbreviations 44 DP: Differential pumping 45 3 FLG: few-layer graphene 46 hBN: hexagonal boron nitride 47 MEMS: microelectromechanical systems 48 MoS2: molybdenum disulfide 49 SEM: scanning electron microscope(y) 50 SiN: silicon nitride 51 STEM: scanning transmission electron microscope(y) 52 TEM: transmission electron microscope(y) 53 FIB: focused Ion Beam(system)  54 UHV: ultra high voltage   55 E-TEM: Environmental Transmission Electron Microsocpe(y) 56 HVEM: high-voltage electron microscope(y)  57 TMP: turbo molecular pump 58 ED: electron diffraction 59 SOEC: Solid Oxide Electrolysis Cell 60   61 4 Introduction 62 In-situ transmission electron microscopy (TEM) is becoming an increasingly useful approach 63 for characterizing materials and their processing. "In-situ" is a Latin term that translates to "in 64 its position", and it signifies to capture the real-time process of changing phenomena. 65 The history of the in-situ observation began at the time the electron microscope was invented.  66 In 1935, Marton[1] suggested two methods to restrict or control the gaseous or liquid  67 solutions around specimens. One approach is to build an open environmental chamber with a 68 pair of small apertures that can minimize gas leakage into the column. The apertures maintain 69 the chamber at high pressure while keeping the rest of the column at a low enough pressure. 70 This idea of differential pumping was realized by Ruska [2] and Ardenne [3] respectively, in 71 1942.  72 The second approach is to form a closed (sealed) gas cell using electron-transparent windows 73 disposed above and below the sample. The first closed cell with plastic windows was devised 74 by Abrams and McBain [4] from Stanford University in 1944. They observed the movement 75 of liquid and bubbling．The purpose of the initial design was to study the effects of gases on 76 or against contamination by inspecting biological materials in a hydrated state, not in-situ 77 observation. 78 As a means for providing external stimuli to the specimen for in-situ observation, there are 79 modified TEM or specially designed specimen holders.  80 5 The first reaction specimen chamber compatible with TEM based on differential pumping was 81 developed by Hashimoto et al. [5] in 1968. Using a Pt grid heater with heating capabilities of 82 up to 1000 oC, the gas pressure of 300 Torr (~4.0 x104 Pa) was achieved. In 1971, Swann and 83 Tighe [6] designed a side-entry with single tilt hot stage in portion within the cell gap for the 84 AEI EM7 of a high-voltage electron microscope (HVEM).  85 In 1962, the solution of the closed cell was demonstrated by Heide [7] [8]. The specimen 86 chamber was formed by two specimen grids and pieces of thin metal foil to determine a 87 constant spacing. The grid was covered with a low contrast film to separate the vacuum 88 pressure of 10–6 Torr (1.3x10-4 Pa) from the pressure of 760 Torr (1.0x105 Pa) in the closed 89 cell. After that, the closed cell was piled up with various improvements (Escaig and Sella, 90 1966 [9], Fukami, et al. ,1970 [10], Dupouy, 1968 [11]). A closed cell consisted of three films, 91 designed by Fujita et al. [12], held above one atmospheric pressure (1.0 x 105 Pa) in 1976.  92 The idea of in-situ liquid TEM proposed in 1935 [1] was inherited by Abrams and Mcbain, 93 who constructed the first enclosed wet cell in 1994 [4]. The cell consisted of two Pt discs, 94 having 0.1 mm holes covered with thin collodion films. An aqueous suspension of the 95 specimen was placed on the lower window. In 1972, Fullam improved this design that had a 96 triple-layer window with an evaporated film of silicon monoxide and the plastic material [13].  97 Stoyanova et al. (Stoyanova and Mikhailovski, 1959 [14], Stoyanova et al., 1960 [15], 1960 98 [16]) first investigated the possibility of studying wet biological material in a high-pressure 99 6 environmental cell installed in a 75-80 kV microscope. In their design, the environmental cell 100 employed 25 nm-thick collodion/carbon or formvar/carbon composite windows supported 101 across 20–70-µm diameter apertures. The windows were kept apart by a spacer with the 102 thickness ranging from 100 to 2000 µm. For a high-voltage electron microscope (HVEM), a 103 similar cell design was constructed by Dupouy et al. (Dupouy et al.1960 [17]; Dupouy and 104 Perrier 1962 [18]). Allinson designed two side-entry environmental cells in 1970 [19]. The 105 window of single crystal corundum was thinned by ion milling. To avoid the chemical attack 106 of the windows by free water in HVEM, the thick windows of evaporated silicon monoxide 107 deposited on fine mesh copper grids were employed by Nagata and Ishikawa in 1972 [20]. 108 In a side-entry environmental cell reported by Double in 1973 [21], collodion/carbon or 109 collodion/silicon monoxide windows were installed on 200 mesh copper grids. In 1976, 110 Fukami [22] developed a film-sealed environmental cell for a conventional 100 kV TEM. 111 As another observation method for wet specimen, only the lower aperture was served as the 112 film. Using this method, liquid water droplets were observed successfully (e.g., Heide 113 1962[8]), However, various challenges in the development of liquid environment TEM 114 (ETEM) techniques remain.  115 The introduction of microelectromechanical systems (MEMS) technology brought a creative 116 advancement in the window approach. 117 Employing a MEMS-based nano-chip, Creemer et al. [23] miniaturized the gas volume and 118 7 heater into a sealed system. Compared to the furnace-based heater, this technique allows a 119 more rapid thermal response and smaller specimen drift. Since then, MEMS has become 120 widely used not only in window approach, but also in aperture approach. 121 As you can see from the history above, thermal excitation, liquid and/or gaseous environment 122 are examples of typical stimuli used for in-situ TEM. These stimuli can be applied directly to 123 the specimen area in a controllable manner by modifications to the microscope, most 124 commonly via specially designed specimen holders. 125 In this paper. we review the in-situ TEM techniques from the perspective of holders having 126 in-situ capability of heating and gas/liquid environment. 127  128 In-situ heating TEM 129 Here we describe the heating specimen holders for in-situ heating TEM technique. There are 2 130 types of the heating holders. One has an indirect heater and the other is a direct heater.  131 Furnace-type heating specimen holder 132 The furnace-type heating holder has a miniature furnace that heats the specimen indirectly. 133 The advantage of this type is that the temperature of the furnace can be measured with a 134 thermocouple. Also, it is designed to use the conventional 3 mm-diameter specimen. The 135 disadvantage of this holder is that the heating region of furnace is rather huge. Because of the 136 size, it takes tens of minutes to achieve a stable target temperature and to eliminate the 137 8 specimen drift caused by the thermal expansion. In addition, to operate at a high temperature 138 range, it is often necessary to cool the holder by cooling water.  139 In 1956, Takahashi et al. developed a furnace-type specimen holder which could heat the 140 specimen up to 1000 ºC [24]. In the next decades, various improvements have been made to 141 reduce the drift by minimizing the heating volume of the heater. In 1986 [25], Sinclair and 142 Parker. developed commercial heating holders and a video-rate recording system which 143 showed mechanical and thermal stability during the heating process. They observed Si re-144 growth at temperatures between 500 and 800 °C. Heating was implemented with a 145 commercial holder (model number PW 6592), where a Pt pad was heated by an electrical 146 feed-through. The pad held a 3 mm-diameter disk sample, and the temperature was measured 147 using an attached thermocouple. In 2004, M. A. Verheijena et al. [26], modified it for the 148 measurement of the electrical resistance of a TEM sample as a function of temperature. 149 In 1997, Hidaka et al. [27] developed a double tilt specimen heating holder with a heating 150 element of spiral shaped fine metal wire (Hitachi High-Tech). Direct current is supplied by 151 dry battery to obtain the high-resolution TEM image. The heater is a spiral wire, which is 152 linked to two-stage swirl shape, and then a thin-film bulk sample partially milled by focused 153 ion beam (FIB) is inserted. Using this holder, the transformation from -Si3N4 to -Si3N4 via 154 liquid state was observed during heating. The image of  -Si3N4 structure was obtained at 155 1800 °C with resolution of 0.18 nm. After the modification for the observation of 3 156 9 mm−diameter disc specimen, the holder has been applied to the study of the oxidation 157 processes of graphene layers on different facets of Pt nanocrystals in Hitachi H-9500 300 kV 158 TEM by Yuan et al. in 2016 [28] [29].  159 In 2019, Shimada et al. [30] have developed an in-situ annealing system (Figure 1) that consists 160 of double tilt furnace type heating holder and a new thermal control box with a Proportional-161 Integral-Differential (PID) controller to improve the spatial and temporal resolution of the in-162 situ observations in an aberration-corrected TEM (ARM-200F, JEOL). When the sample was 163 being heated at a rate of +1.0 °C/s, the thermal drift occurred in the parallel to axial holder 164 direction with a rate less than 0.1 nm/s. 165 A widely used double tilt heating holder is Gatan 652 heating holder which employs a Ta 166 furnace with a water-cooling system invented by Jones and Swann [31]. The Ta heater is 167 provided on the outer fringe of the specimen fixed position of about 3 mm-diameter. The thin 168 Ta heater is designed so that the heat responsiveness to the heater current is improved, and it 169 is possible to heat at a maximum of 1273 K (1000 oC). Although these types of holders 170 improve their stability, a non-ignorable image shift still exists while increasing in temperature. 171 It is not easy to realize the atomic resolution observation especially at temperatures higher 172 than 500 °C, because the recirculating cooling may also bring vibrations and decrease the 173 imaging stability and resolution.  174 Zheng et al. [32] used Themis transmission electron microscope (TFS, 300 kV) with Gatan 175 10 652 heating holder for their heating experiments. The growth of C-S-H at various 176 temperatures was investigated. The nanoscale morphology, pore structure, element 177 distribution, and phase transformation were correlated with evaluated temperatures.  178 Grid-type heating specimen holder/ Wire-type heating specimen holder 179 In grid-type heating specimen holder or wire-type heating specimen holder, a grid or a fine 180 wire is directly heated by an electric current. Since the specimens are mounted directly onto 181 the heater, they are limited to either powders or flakes. Furthermore, it is hard to directly 182 measure the temperature of the heater. The advantages of this type of holder are the high 183 maximum temperature and high thermal stability.  184 A simple design of a side-entry single tilt heating holder was built for an HVEM by Swann 185 and Tighe in 1971[6]. The specimen is held between two punched Pt-Rh ribbons 25 µm thick 186 by 3.2 mm wide, which are tensioned by spring-loaded current leads and supported over two 187 quarts fibers 0.5 mm in diameter to minimize heat losses. The spring action absorbs the slack 188 induced by heat expansion. Though the stage is not well insulated, it has a low thermal 189 capacity. Power input is 3 A at 12 V to reach a temperature of 500 oC, and the maximum 190 operating temperature is 1000 oC. In 1982, Komatsu et al. [33] developed top-entry stages 191 utilizing localized electron beam heating rather than direct resistance heating of 192 the specimen for ultra-high voltage (UHV) TEM. It consists of a thin Ta tube specimen 193 holder, surrounded by a coaxial Ta tube contains the heater filament and two coaxial stainless 194 11 steel thermal shields. 195 One example of such a direct type of side-entry single tilt heating holder is Kamino holder or 196 Kamino and Saka holder which was developed by Kamino and Saka [34]. An external view of 197 the Kamino holder is shown in Figure 2. A fine W filament of 25 µm in diameter, which is 198 bridged across two electrodes, is heated by direct electric current from a battery. To obtain 199 thermal stability, a dry battery must be used as a power supply. There are two methods that can 200 be used to calibrate the temperature as a function of the electric current. One method is to use 201 an optical pyrometer outside a TEM. The other is to observe the melting of known materials by 202 in-situ TEM. The small thermal mass of the heater leads to a small drift rate. After 15 min., the 203 drift rate becomes as small as 0.1 nm/s. By using the holder, the formation of SiC via solid state 204 reaction between Si and graphite at 1400 °C was successfully observed. On further heating at 205 1500°C, the dynamic observation of a sequence of crystal growth and formation of a grain 206 boundary during sintering at atomic level was achieved [35]. Following this type, two wire 207 types have been developed, which facilitate the heating of more than one material independently 208 [36]. 209 Based on this wire type holder, a variety of versions has been developed. For example, a holder 210 consists of the 3 mm-diameter-grid having a support film and 2 W-filaments for sample 211 evaporation as shown in Figure 3 [37]. It was applied to the generation of various composite 212 nanoparticles and the analysis of structural changes at elevated temperatures by Mori et al. in 213 12 1994 [38]. 214 As another example, the Kamino holder was modified as a compatible heating holder which 215 can be used in the FIB and the scanning TEM (STEM). Tanigaki et al. demonstrated the in-216 situ heating TEM method for an interface reaction using this holder in 2009 [39]. The 217 specimen was prepared by attaching a micro-sample directly to the heater using the FIB 218 micro-sampling technique (Figure 4). It was confirmed that the EDX map and electron 219 diffraction analyses were possible during the reaction. The resolution of this technique was of 220 the order of 0.223 nm at 550 oC. The shape of the tip was modified to secure the path of the 221 focused ion beam to the heater while bonding or thinning the micro-sample on the heater. A 222 rotational function was applied to the holder to observe the sample attached to the filament by 223 tilting it 90o from the FIB position. The other modified holders are described in later sections. 224 In-situ heating TEM – MEMS type specimen holder 225 The integration of MEMS technology into in situ heating TEM specimen holders allows for 226 great customization. MEMS-based heating holders consist of two primary components - a 227 specialized TEM holder and a replaceable chip on which the heating elements are directly 228 patterned. Electrically conductive contact pads on the heating chip directly align with the 229 contact pins on the holder to complete the electrical circuit. MEMS heating chips operate 230 through Joule heating, where a current is driven through the heating element and the electrical 231 energy is converted into heat through resistive losses [40]. The heating element is often 232 13 isolated from the bulk of the silicon chip and embedded in a suspended membrane made of 233 SiN or another insulating material. This not only prevents heat transfer to the rest of the chip, 234 but also minimizes sample drift caused by thermal expansion. In some cases, this large 235 membrane is thin enough to double as the electron-transparent region for TEM imaging [41]. 236 However, most recent designs have opted for a thicker SiN membrane patterned with a series 237 of small electron-transparent windows with reduced thickness [42] [43] [44] [45]. The 238 windows are typically circular with diameters ranging from 5-8 µm but vary in size and 239 shape. As the chips serve as both the heating stage and the specimen support, one issue that 240 arises is the bulging of the SiN membrane due to thermal expansion. Van Omme et al. [44] 241 used finite element analysis to simulate the bulging deformations of their heating chip at 242 various temperatures and optimize the design of the Wildfire Heating Nano-chips 243 manufactured by DENSsolutions. 244 Zhang et al. [46] recently compared the bulging distance, or Z-drift, of the Wildfire chips with 245 the NanoEx heating chips developed by Mele et al. [47]. Both chips exhibited high structural 246 stability at temperatures less than 200° C. However, the NanoEx heating chip was observed to 247 have a significantly larger bulging rate between 200 and 500 °C and a total bulging height of 248 more than 17 µm at 800 °C. Under the same conditions, the Wildfire chip showed minimal 249 bulging behavior until 500 °C and a maximum bulging height of less than 8 µm at 1100 °C. 250 Over the past several years, many have experimented with different patterns and materials for 251 14 optimizing the heating element design. Early heating chips, such as the nanocalorimeter 252 designed by Olson et al. [48] and used by Zhang et al. [41], used a singular, straight heating 253 strip that stretches across a SiN membrane. The Protochips Aduro device used by Allard et al. 254 [49] similarly concentrates the heating region to a small path in a conductive ceramic membrane. 255 In modern heating chips, the heating element is more commonly patterned in a double spiral 256 [42] [43] [44] [45] [50] to optimize Joule heating over the specimen region. Figure 5 shows 257 three examples of the heating element design. Perez Garza et al. [42] designed the double spiral 258 with straight line segments of constant width that follow an octagonal spiral shape, as shown in 259 Figure 5a. In creating the NanoEx heating chips, Mele et al. [47] alternatively patterned a circle 260 in the middle of the heating region surrounded by two heating lines that form hemispheres on 261 either side, shown in Figure 5b. However, these designs introduce corners into the current path, 262 which can result in current crowding [51]. Van Omme et al. [44] designed the DENSsolutions 263 Nano-chips with a double spiral heating element that follows a circular path with no corners to 264 avoid this effect, shown in Figure 5c. While the double spiral design is symmetrical in X and Y 265 directions, it can result in a temperature gradient with the hottest region in the centre of the 266 spiral. Mele et al. [52] proposed that better temperature uniformity can be achieved by 267 increasing the width of the heating element lines closer to the centre of the spiral. This reduced 268 the degree of Joule heating in the central region to counterbalance the nonuniformity of the 269 heating profile. With this configuration, they reported a successful decrease in temperature 270 15 variation between the outer and inner regions of the spiral from 15% to 4%. The same concept 271 of increasing linewidth was adopted by van Omme et al. [44] who reported 98% homogeneity 272 over their heating region.  273 It is common to have at least four contact pads to facilitate four-point probe measurements 274 where current is applied through the two outer contacts, and resistance is measured from the 275 two inner contacts. This allows for resistance measurements without the influence of contact 276 resistance between the contact pins and contact pads [44]. These measurements allow for 277 constant feedback to monitor the temperature of the heating element. Baroncini et al. showed 278 that there is a correlation between the heating power, heating element resistance, and 279 temperature of a microheater that can be quantified using a calibration curve [53]. 280 With MEMS technology, it is possible to increase the functionality of the chip by patterning 281 additional contact pads. Zhao et al. [54] used six contact pads in their heating chip design. 282 Two were used to apply the current to a graphene heating element and four were used to monitor 283 the resistance of a separate platinum thermometer between the heating elements for temperature 284 readings. Bernal et al. [55] reported that the number of contact pads can be increased to at least 285 nine for added functionality in future designs.  286 Selection of the heating element material is critical in dictating the performance of the heating 287 chip. Several materials have been explored in the development of microhotplates and have 288 since been adopted into heating chip designs. Metallic thin films are advantageous for their 289 16 high electrical conductivity, low voltage requirements, and high current density [48]. Platinum 290 is chemically inert at elevated temperatures and has been widely used in MEMS heating 291 devices [47] [50] [56]. Some metals, such as Pt and Ti, require an additional adhesive Cr or Ta 292 layer to be deposited onto the membrane first to avoid delamination [43] [48]. For sensitive 293 applications, this extra layer could influence the electrothermal response and any associated 294 measurements. Mo has been explored as a heating element as it also possesses desirable 295 chemical resistivity, can withstand higher temperatures, and does not require an adhesion 296 layer [42] [43]. However, Mo heating elements require an additional protective layer, such as 297 SiO2, to prevent oxidation at temperatures over 300 oC. 298 Protochips manufactures two types of heating chips – the Fusion Electrothermal E-chips [57] 299 and the Fusion Heating E-chips [58]. The primary difference is that the Fusion Electrothermal 300 E-chips have an additional four tungsten electrodes for electrical characterization of the sample. 301 W has a high melting point of 3422 °C and excellent thermal conductivity, which is compatible 302 with the high temperatures achieved by the heating element [59]. These chips are rated for 303 temperatures up to 900 °C and permit simultaneous heating and electrical measurements. The 304 Protochips designs, unlike the other designs discussed so far, uses a ceramic SiC heating 305 element [58] [60]. Ceramic heating elements have much greater resistance than metal heating 306 elements, and therefore a greater thermal response. Since the resistance of the ceramic heating 307 element is several magnitudes greater than that of the holder-chip contact resistance, the four 308 17 point probe setup is not necessary [61]. The Fusion Heating E-chips have only two contact pads 309 and are rated for temperatures up to 1200 °C with temperature accuracy >95 %. DENSsolutions 310 similarly manufactures two sets of chips – the Wildfire Nano-chips for in situ heating [62] and 311 the Lightning Nano-chips for in situ biasing and heating [63]. The primary difference is that the 312 Lightning Nano-chips have 6-8 contact pads for the additional biasing capability while the 313 Wildfire Nano-chips only have the four contacts required for four-point probe measurements. 314 These chips by DENSsolutions are rated for temperatures up to 1300 °C with temperature 315 accuracy >95 %. In addition to these commercial designs, other materials are still being 316 explored for alternative heating chips. Doped polysilicon has been widely used for heating 317 elements in microheaters outside of ETEM, and shows promising thermal uniformity, stability, 318 and thermal conductivity [53] [64] [65] [66]. However, polysilicon heating elements may 319 encounter issues with long-term stability due to changes in resistivity under thermal and 320 electrical stress [67]. Recently, Zhao et al. [54] used monolayer graphene over suspended SiN 321 as the heating element. They report that the graphene heater has a comparatively low heat 322 capacity, resulting in a fast temperature response, low power consumption, and reduced bulging.  323 On the commercial market, TEM holders for in situ MEMS-based heating such as the 324 Protochips Fusion AX holder, DENSolutions Wildfire series, Mel-Build Corporation double tilt 325 4 electrodes transfer holder, and Hummingbird Scientific MEMS Heating + Biasing holder are 326 available for purchase. Previous studies also list the FEI NanoEx™️-i/v single tilt holder [47], 327 18 Hitachi Blaze heating holder [68] [69], and Protochips Aduro™️ setup [49] as a few additional 328 MEMS type holders. Krisper et al. [70] recently demonstrated the importance of the tilt feature 329 for collecting EDS measurements using the DENSsolutions Wildfire holders. Semiconductor-330 based EDS detectors, such as silicon-drift detectors, are sensitive to the infrared radiation 331 generated by a heated specimen. Tilt capabilities of the holder can be used to align the EDS 332 detector with low takeoff angles from the sample. At these angles, X-ray signals are collected 333 by the detector, but infrared signals are minimal. Since MEMS-based holders include sidewalls 334 to hold the heating chip in place, low takeoff angles are blocked from the detector and careful 335 optimization of the tilt angle is required. 336  337 In-situ gas heating TEM 338 Comparison between opened type and closed type 339 To produce a controlled atmosphere around the specimen, there are two methods. One is 340 “opened type” or “aperture type”. Another is “closed type”. Figure 6 shows a schematic 341 diagram of “opened type” specimen chamber(a) and “closed type” one(b).  342 The “opened type” modifies the specimen chamber by inserting pair of apertures between or 343 within the objective pole pieces to confine the gas leakage, as developed by Boyes and Gai 344 [71]. To avoid a gas leak and maintain the high vacuum in the other critical parts of the TEM, 345 the system would also require the differential pumping system. Therefore, it is also called a 346 19 differential pumping type. As there are no additional membranes on windows, it is possible to 347 maintain a high spatial resolution. Unlike closed type E-cells, this type of ETEM system uses 348 the specimen chamber as the reactor and generally does not require specialized holders with 349 gas flow capabilities. 350 Meanwhile there are some disadvantages of opened type. One major difference is that the 351 opened type system possesses a significantly thicker gas layer than closed type. Currently, the 352 controlled gas pressure in the opened type is lower than in the closed type because the size of 353 the apertures restricts the maximum gas pressure.  354 In closed type, specially designed TEM holders enclose the specimen between two electron 355 transparent membranes, confining the gas or liquid around the specimen. The important feature 356 of closed type is the airtightness of the cell. The length of gas–electron interaction is much 357 shorter than that in the opened type, making it acceptable to increase the gas pressure. Currently, 358 atomic resolution may be obtained at one atmosphere or higher pressure. Yokozawa et al. 359 reported that their system was evaluated with the (de)hydrogenation of Pd at pressures up to 360 4.5 bar (4.5x105 Pa) [72]. A closed type specimen holder is compatible with different TEMs 361 without any further modifications to the microscope itself. Furthermore, the cost of purchasing 362 or modifying a specimen holder is much smaller compared with an opened type ETEM.  363 The fracture of the window’s membrane or leakage of the gas during experiments would 364 deteriorate the vacuum of the column and electron source. Meanwhile, it is important to ensure 365 20 the airtightness and field of view during the assembly of the closed cell. This results in the 366 complexity of the specimen loading operation, which involves challenges such as centering the 367 upper and lower window membranes. Furthermore, the electron-transparent window membrane 368 and gas thickness still interact with the electrons, resulting in the scattering information 369 superimposed on the image obtained. Yaguchi et al. reported that the thicknesses of the 370 membranes and the gas channel would have a significant impact on image quality [73]. In 371 addition, the existence of window membranes hindered the acquisition of EDS signals. 372 Opened type ETEM with heating holders 373 For in situ gas heating experiments, the in-situ heating holders can be used in conjunction with 374 the differential pumping ETEM system, also known as the “opened type” ETEM.  375 Hashimoto et al, (1966, 1959) employed a metallic wire with a high melting point as the 376 specimen support while developing [74, 75] the first reaction specimen chamber with apertured 377 environmental TEM. Although high temperature could be achieved by the directly heating the 378 specimen support element, the discharge of gas into the microscope column established a limit 379 of 10 torr (1.3 x 103 Pa). The requirement for differential pumping was then realized to broaden 380 the pressure range of operation. In 1968, Hashimoto et al. [5] modified the specimen stage 381 design and increased the permissible gas pressure to 300 Torr (4.0 x 104 Pa). An electric current 382 was sent through the ribbon to heat the specimen, which was put on a film covering the hole in 383 the Pt ribbon. The gas piped around the object and spilled into the vacuum of the microscope 384 21 column via the two Pt holes on either side of the ribbon. 385 In 1972, Baker and Harris [76] incorporated a modified gas reaction stage (JEOL Co JEM AGI 386 attachment designed by Hashimoto et al. [5,77]) onto a JEM 7A electron microscope. The basic 387 ideas of the ETEM developed by Hashimoto et al. are still used in the latest equipment. 388 In 1997, Boyes and Gai [71] introduced two pairs of apertures above and below the specimen, 389 which were mounted inside the bores of the objective pole pieces rather than between them as 390 in previous designs so as to keep the high-pressure pathway as small as possible. Additional 391 pumping in the form of turbo molecular and ion getter pumps is used. The gas pressure, 392 usually less than 3x103 Pa, is maintained by a controlled flow directly into the pole piece gap. 393 Thereafter the Titan ETEM G2 was produced by FEI [78] and the more recent Themis ETEM 394 by Thermo Fisher Scientific [79]. The Titan ETEM G2 [78] and Themis ETEM [79] both 395 feature three gas inlets for three different gases, and a mass spectrometer for residual gas 396 analysis of the specimen chamber. All these instruments are additionally configured to support 397 environmental STEM (ESTEM) in addition to ETEM [80]. 398 In 2009, Hitachi H-9500 (LaB6: 300 kV) in-situ TEM featuring with an improved differential 399 pumping capacity for the TEM main unit and mounting a new gas injection mechanism 400 developed by Kishita et al. [81]. There are two vacuum stages developed in the upper portion 401 of the specimen chamber and one vacuum stage in the lower portion, using a turbo molecular 402 pump (TMP) with a pumping speed of 260 L/s. The gaseous atmosphere around the specimen 403 22 is created using a nozzle extending from outside of the column to near the specimen. The 404 system configuration permits in-situ observation with the specimen temperature of 1200 oC or 405 higher and the pressure in the specimen chamber of ∼10 Pa. 406 In this system, a gas injection port is also provided in the pre-evacuation chamber during the 407 introduction of the specimen holder, which permits heating and gas treatment (equivalent to 408 the atmospheric pressure) in the pre-evacuation chamber. Since the pre-evacuation chamber is 409 a part of the TEM, the specimen can be introduced to the specimen chamber right after the 410 reaction without exposing to air atmosphere. These units allow the observation of unaffected 411 phenomena by the electron beam in processing under the atmospheric pressure. Ex-situ 412 observation can be conducted in combination with in-situ observation to provide a reliable 413 analysis of structural change processes undergone by various gas treatments.  414 The Hitachi HF5000 200kV aberration-corrected TEM/STEM employed a combination of 415 small aperture, additional TMP with a pumping speed of 300 L/s, and gas-injection nozzles 416 extending to near the specimen [82] [83] for in-situ observation. The Secondary Electron (SE) 417 detector is a key aspect of this system. Lv et al. applied this system with Hitachi High-Tech 418 Canada, Inc. (HTC) MEMS heating holder to verification on oxidation and reduction 419 reactions of perovskite Solid Oxide Electrolysis Cell (SOEC) catalysts for CO2 electrolysis 420 under gas atmosphere and its CO2 adsorption sites by E-STEM analysis [84].  421 Opened type ETEM/Conventional TEM with gas-injection heating holders 422 23 In 2005, Kamino et al. [85] developed the side-entry specimen holder consisting of a heating 423 element and a gas injection nozzle. Figure 7 shows an external view of a tip of the holder. It 424 could be attached to a conventional TEM without any modification. The heating element is 425 same as the heating holder as described before. A gas injection nozzle with an inner diameter 426 of 0.5 mm was built about 1mm away from the heating element. This holder provides 427 localized gas atmosphere near the specimen. Therefore, the system enables the user to change 428 the pressure of the specimen chamber in a few minutes.  429 The experiment was conducted in a Hitachi H-9500 TEM equipped with a LaB6 cathode. This 430 system allowed for the observation of high-resolution TEM images at temperatures of 1000 431 oC in the ~2 × 10–2 Pa gaseous environment in a conventional TEM without any modifications 432 of the vacuum system [81]. Using this TEM, Yaguchi et al. [86] reported the influence of 433 moisture on the structural changes of heated electrocatalysts of polymer electrolyte membrane 434 fuel cells with the combination of the holder and a moisturized air supply system. The 435 morphological changes of a platinum catalyst dispersed on carbon black (Pt/CB) were 436 dynamically observed in the atmosphere of the highly moisturized air. The experimental 437 results at high moisture content were compared with results obtained with lower moisture 438 content. The dependence of morphological changes of the catalyst on the air humidity was 439 clarified. 440 The holder has been improved to have two built-in heating filaments and an injection nozzle 441 24 by Kamino et al. [87] in 2006. Figure 8 shows the design principle and operational workflow 442 of the holder. While one of the heating filaments works as a “Heater”, the other functions as 443 an “Evaporator”. The two heating filaments can be turned on and off independently. When 444 “Evaporator” is turned on, the precursors are evaporated and deposited onto the substrate on 445 the “Heater”. The substrate temperature can be controlled by the “Heater” during and after the 446 evaporation deposition. Gas injection is controlled from outside the TEM column. The entire 447 process, from the synthesis of the Al2O3 substrate and deposition of AuPd nanoparticles on the 448 Al2O3 support to the observation of nanoparticle behavior on the support surface at elevated 449 temperatures, was done totally in-situ within the TEM pole piece gap.  450 Another modification of the side-entry specimen holder consisting of a heating element is an 451 addition of the air protection transfer mechanism (Figure 9). Using the Hitachi HT7820 TEM 452 (120kV, LaB6) with the holder, the in-situ observation of the degradation process was 453 conducted for a sulfide-based Li4SnS4 glass ceramic under an air-flow environment by 454 Tsukasaki et al. in 2021[88]. A sulfide-based Li4SnS4 glass ceramic is one of the candidates 455 for next-generation all-solid-state batteries materials. Electron diffraction (ED) patterns and 456 hollow cone dark field images could clearly capture morphological changes and the 457 amorphization process caused by air exposure. Moreover, based on the analysis of ED 458 patterns, it is observed that Li4SnS4 is likely to decompose because of the reaction with H2O 459 in air. 460 25 Opened type ETEM with MEMS heating holders 461 For in situ gas heating experiments, the MEMS-based in situ heating holders previously 462 mentioned in the In situ heating MEMS type holder section can be used in conjunction with 463 the differential pumping ETEM system.   464 These ETEMs have been used with commercial MEMS-based heating holders such as the 465 NanoEx-i/v and DENSsolutions Wildfire holders for a wide variety of in situ gas heating 466 applications. While the ETEM/ESTEM systems provide accurate control of gas pressures in the 467 specimen chamber (from 10-3 to 2000 Pa for the Themis ETEM), the addition of MEMS 468 technology brings greater control over the temperature of the heating holder and the imaging 469 stability. In situ experiments can benefit from greater temperature precision provided by the 470 constant four point probe feedback. Innovative designs of the heating chips also result in a 471 reduction of membrane bulging and thermal drift, which allows for high resolution to be 472 achieved with greater stability. In this section, selected studies that highlight the potential of 473 differentially pumped ETEM with MEMS-based holders are discussed. 474 In situ studies involving the gas-solid interactions of catalysts is one application which has 475 largely benefited from the advancements in ETEM with MEMS-based holders. The high 476 resolution and stability of the instruments make it possible to image atomic arrangements and 477 investigate the correlation between structure and catalytic performance [89]. These instruments 478 have been particularly useful in the imaging the single-atom interactions involved in catalysis. 479 26 When Boyes and Gai first developed their modified ETEM with atomic resolution [90], they 480 reported studies of heterogeneous catalyst reactions. In controlled gas environments of up to 50 481 mbar (5 kPa), they observed dynamic particle shape changes, growth of passivating sub-nm 482 overlayers, and the transport of C and Ni atoms in the growth process of carbon nanofibers 483 catalyzed by Ni nanocrystals. When they reconfigured the instrument with ESTEM capabilities, 484 they investigated the dynamics of gas-solid catalyst reactions further and achieved atomic 485 resolution of single platinum atoms in 0.02 mbar (20 Pa) H2 at temperatures greater than 500 °C 486 [80]. The ESTEM addition was highly beneficial for detecting the positions of individual atoms 487 by taking advantage of the HAADF imaging mode with high Z-contrast [91]. However, they 488 reported reduced visibility of single atoms with high mobility at temperatures greater than 489 400 °C. More recently, Boyes et al. [92] and Martin et al. [93] reported similar single atom 490 dynamic studies, now using the modified JEOL 2200 ETEM/ESTEM with advanced MEMS-491 based heating holders made by DENSsolutions. Martin et al. studied the migration of single 492 platinum atoms and the evolution of platinum nanoparticle growth in H2 at temperatures up to 493 600 °C. Although the resolution is still limited by the high mobility of the loose atoms, they 494 attributed both improved temperature control and stage stability to the MEMS-based holders. 495 The potential of advanced MEMS-based holders in the imaging of catalytic interfaces was more 496 recently demonstrated by Yuan et al. [94]. They used the DENSsolutions Wildfire S3 heating 497 holder in the FEI Titan 80-300 ETEM to image the real-time epitaxial rotation of Au 498 27 nanoparticles on a TiO2 (001) surface. At 500 °C, they observed a reversible rotational behavior 499 of the nanoparticles when the gas environment in the ETEM was changed. After replacing the 500 oxygen environment [6.5 mbar (650 Pa) O2] In the chamber with a reactive environment [4.4 501 mbar (440 Pa) CO/O2], they concluded that the Au nanoparticle rotated along the [111] axis, 502 perpendicular to the (001) TiO2 surface. The high stability of the holder was essential to observe 503 the catalyst nanoparticles with atomic resolution under the different environmental conditions. 504 Studies such as these contribute important information about the stability of active catalytic 505 sites and how the catalytic interface can be manipulated through the external environment. 506 Additionally, the phenomenon was found to be temperature dependent as the reversible 507 rotations are not observed at 25 °C. This highlights the benefits of temperature accuracy 508 provided by the MEMS heating chips with four-point probe capabilities. Yuan et al. provide 509 further insights on how these ETEM techniques can be applied in the catalyst design process in 510 their more recent article [95].  511 These instruments allow for high customizability of the sample environment within the 512 specimen chamber. In addition to common gases such as O2 or H2, the ETEM system can 513 accommodate other gaseous species necessary to incite reactions in situ. Diallo et al. [60] 514 introduced digermane gas (Ge2H6) into the Cs-corrected Titan ETEM to observe the nucleation 515 of Ge crystals on a graphene substrate. They used a Protochips Fusion holder and chip with a 516 SiC ceramic heating element to mount the graphene. Growth temperatures ranged from 220 to 517 28 600 °C with and the thermal stability was advertised to be <0.1 °C/min. Studies such as these 518 require slight modifications to the chips themselves. Graphene is electrically conductive and 519 risks short-circuiting the heating elements when mounted onto the heating chip. However, an 520 additional benefit of MEMS-based heating is that the heating chips can easily undergo 521 additional fabrication steps as required. Diallo et al. deposited an extra 10 nm thick Al2O3 layer 522 onto the heating chips prior to mounting the electrically conductive graphene sample to 523 electrically isolate them from the heating elements. These sorts of modifications are made 524 possible as the MEMS heating chip devices maintain the flat structure of a standard substrate 525 that is ideal for microfabrication techniques. 526 Open type E-cell holder with heating 527 The next system also utilizes the differential pumping effect, that is, “open-type”, but uses a 528 specimen holder or chamber for the formation of the gas phase. In early differential pumping 529 systems, a pair of apertures above and below the specimen were installed into the specimen 530 chamber, and the apertures could be taken in/out from the microscope vacuum through the 531 airlock along with the specimen chamber [5,76,77,96].  532 On the other hand, when a differential pumping unit is attached to a side-entry specimen holder, 533 as an open-type cell, a conventional TEM can be used without any modification to the TEM 534 column because the holder is independent of the TEM main body. However, as the tip of the 535 specimen holder is inserted into the gap between the pole piece of objective lens, there are 536 29 significant size restrictions. Since the differential pumping mechanism must be built into the 537 thickness of a few mm, multiple orifice pairs can be hardly arranged unlike an ETEM, and 538 consequently atmospheric pressure cannot be reached. In other words, observations at a lower 539 gas pressure, for example, in the case of a catalytic reaction, will result in fewer gas molecules 540 as the reacting species, thereby reducing the reaction rate. Therefore, it is easier to capture the 541 behavior during the reaction even with a standard CCD camera.  542 Hashimoto et al. developed and used such differential pumping specimen holder system for the 543 in situ observation of catalytic materials [97]. Figure 10 shows a side-entry specimen holder 544 with differential pumping unit and MEMS heating device, which does not use single orifice 545 pair. As shown in Figure 10 (a) the system consists of a differential pumping unit and a MEMS 546 heating unit. The gas reaching near the specimen through a nozzle leaks out through the paired 547 orifices to the vacuum of the microscope column. As a consequence, it is difficult to use 548 corrosive or wetting gases. The inner pressure can be kept up to 20 Pa higher than that of the 549 outer column. For the reduction of microscope vibration, the specimen holder grip (Figure 10 550 (b)) contains a gas tank in order to be used without the connection with an exterior gas source 551 during observation, and the fine control valves driven by a piezo actuator in order to control 552 gas flow electrically. The inner pressure near the specimen can be measured by a tiny pressure 553 gauge. For specimen heating, the specimen is placed on a MEMS heater with holes, which also 554 serves as an orifice as shown in Figure 10 (c) and (d). Shoji el al. observed the Ni#Y2O3 catalyst 555 30 with entangled networks of tens-of-nanometer-thick fibrous phases as catalytic materials for 556 CH4 conversion (Figure11) [98]. Their in-situ TEM/STEM imaging and EELS analysis showed 557 that the Ni center of the Ni#Y2O3 region was stable for its structure and valence state because 558 of its topologically immobilized structure (Figure 11 (b) and (c)), which would contribute to its 559 long-term stable catalytic performance.  560 Next, instead of using a side-entry specimen holder, a differential-pumping-type chamber (cell) 561 that can be inserted and removed in the HVEM. HVEMs have been used for not only 562 observation of thick biological samples, polymers, metals etc. but also improvement of the 563 point-to-point resolution. As a technique to improve such resolution, the use of spherical 564 aberration correctors has been developed and then has been reported for various applications. 565 However, HVEMs are still necessary for observation of thick specimens, and has the feature of 566 the wider gap between the pole pieces of the objective lens, which should be advantageous to 567 in-situ observations. 568 As shown in Figure 12(a), Tanaka et al. [99] developed an environmental HVEM (E-HVEM) 569 with using a differential-pumping-type cell, named ‘Reaction Science HVEM’ (JEOL; 570 JEM1000 K RS), whose high-voltage tank housing as accelerating tube is 6.7 m in height and 571 column is 3.6 m in length. This E-HVEM utilized a symmetrical objective lens with a polepiece 572 gap larger than 15 mm for formation of gas phase, where a new side-entry goniometer to enable 573 inserting and retracting the gas chamber as the environmental open-type cell was constructed 574 31 as shown in Figure 12 (b) [100]. For adopting the open E-cell, three-stage differential pumping 575 system was installed into around the objective lens with additional five turbo molecular pumps 576 and a scrolled pump. This open type E-cell produced a few tenth of atmosphere, for example, 577 13,300 Pa (=100 Torr), which is larger than typical ETEM. Standard various specimen holders 578 including heating ones can be inserted into the gas chamber for various in-situ observation. A 579 point-to-point resolution in TEM and STEM mode was less than 0.15 nm and 1 nm at an 580 accelerating voltage of 1 MV, respectively. Furthermore, a quadrupole mass spectrometer 581 (QMS) was combined to identify the reaction gases in-situ in the specimen chamber [101]. 582 Fujita et al. [102] observed nanoporous NiCo catalyst used also for CH4 conversion through E-583 HVEM in order to investigate the origin of the moderate activity and thermal stability. The 584 grains in the ligaments became finer when the mixture gas (CH4 + CO2) was introduced into 585 the specimen chamber heating at 600 oC, although the pore/ligaments coarsened during heating 586 at 600 oC in vacuum. Furthermore, STEM-EDS analysis of the cooled sample showed the 587 chemical demixing of Ni and Co. This grain refinement and chemical demixing can imply 588 “synergic effects”, which demonstrated higher catalytic activity and durability of bimetallic Ni–589 Co. 590 Closed-type E-cells with membranes, which are generally used in the specimen holder, are 591 necessary for observation in gas atmospheres above atmospheric pressure or in liquids. In 592 addition, any modifications of TEM are not required. 593 32 However, the membranes and dense gas phases sometimes deteriorate image resolution. 594 Furthermore, they become a background and reduce the accuracy when combined with analysis 595 methods such as EELS and holography. In contrast, although the open-type specimen holder or 596 chamber produce the lower ultimate pressure than close-type E-cells, this system has the 597 potential to achieve higher image resolution and less background for EELS analysis and 598 holography than not only the close-type E-cells but also the typical modern ETEM with the 599 larger thickness of the gas phase. 600 Closed type ETEM/E-cell heating holders 601 As mentioned in the previous sections, there are mainly two approaches that enable TEM 602 study of materials in gaseous atmosphere. One is to directly introduce gas to the sample 603 vicinity while maintaining high vacuum elsewhere through the pressure-limiting differential 604 pumping (DP) system. Boyes and Gai developed such system in the 1990’s, modifying a 605 Philips CM30 TEM, achieving an impressive 0.2 nm resolution and 50 mbar (5 kPa) [71]. 606 Later, they built the similar system on aberration-corrected TEM [90]. The advantage of this 607 approach is permitting the use of regular sample holder or windowless heating holder. Spatial 608 resolution is higher because its windowless configuration. However, this approach limits the 609 highest gas or vapor pressure to ~ 100 mbar (10 kPa). Another approach is to introduce gas 610 through closed window cell as part of the sample holder. Such environmental cell, called “E-611 cell” in this paper, is a type of MEMS-based devices with electronically transparent windows 612 typically made of amorphous silicon nitride (SiNx). Closed E-cell has the advantage of 613 33 accommodating high gas or vapor pressure. Gas pressure up to 4 bar (400 Pa) has been 614 achieved [103]. Unlike the open system, which requires customized design and build of an in- 615 situ electron microscope, by having the E-cell sample holder on any matching microscope, it 616 can readily perform TEM analysis in gas. If the E-cell has a hermetic seal, it can also hold 617 liquid or liquid and gas/vapor mixture, which is reviewed in next section.  618 An example of E-cell by Protochips [104] is presented in Figure 13. Dai et al.[105] studied the 619 Rh-doped CaTiO3 powder near atmospheric pressure ranged from 250 to 700 oC using 620 aberration-corrected scanning transmission electron microscopy. It revealed both cyclical 621 precipitation−dissolution of Rh nanoparticles in response to redox cycling of the ambient gas 622 and sintering of the powder [105].  623  624 In-situ liquid-phase electron microscopy 625 Liquid cell electron microscopy is an emerging technique that allows in situ imaging and control 626 of nanoscale phenomena in hermetically sealed liquid cell using TEM, STEM or SEM [106, 627 107]. To be compatible with the vacuum environment required by the TEM, most of the liquid-628 phase microscopy studies have been carried out using sealed cells. The first report of closed-629 cell design was proposed by Abrams and McBain in 1944 [108]. MEMS-based liquid-cell is the 630 enabling technology which drives the liquid-phase microscopy study over the past two decades. 631 In 2003, Ross et al first reported using the silicon nitride E-cell to study nucleation and growth 632 34 of copper clusters in aqueous solutions [109]. This liquid cell has a liquid reservoir and two 633 electrodes. TEM study was carried out using a Hitachi H-9000 TEM operated at 300 kV right 634 after the E cell was connected to a power source for observing the Galvanic growths. Images 635 and videos were recorded during galvanostatic deposition with cathodic current densities of 5 636 and 50 mAcm–2. In 2009, Zheng et al. [110] studied the colloidal nanocrystal growth using a 637 JEOL 3010 TEM. It shows that the platinum nanocrystals can either grow by monomer 638 attachment or by particle coalescence. These two type cells were static cell, which had no liquid 639 flow. Liquid flow cells have become commercially available through various vendors. Liquid 640 flow cell with electrical leads or built-in electrodes permits electrochemical analysis of 641 materials in situ. There are also designs to incorporate a thin-film Joule heater, allowing heating 642 the liquid to its boiling point. Hummingbird has recently developed an optical liquid 643 electrochemistry holder, which makes possible to study photocatalysis, photoelectrochemistry 644 or photochemistry [111].  645 It is also possible to map the electrostatic potentials and magnetic fields in liquids using off-646 axis electron holography [112]. Prozorov et al. demonstrated that the Magnetospirillum 647 magneticum strain AMB-1 and assemblies of magnetic nanoparticles can be studied using off-648 axis electron holography. The electron holograms show sufficient interference fringe contrast 649 to allow the reconstruction of the phase shift of the electron wave and mapping of the magnetic 650 induction from bacterial magnetite nanocrystals.  651 35 Most of commercially available liquid cells have amorphous silicon nitride (SiN) windows for 652 observation. The SiN window can hardly be thinner than 8 nm. SiN is also a dielectric material 653 with a bandgap of 5 eV. It is considered as an electrical insulator, which has charging problem. 654 Ideally, the window membrane should be electronically transparent, mechanically robust, 655 chemically inert, with good electronical and heat conductivity. Low atomic number 2D 656 materials such as hexagonal boron nitride (hBN) and graphene have been experimented for 657 making liquid cells with some success. Initial designs of graphene liquid cell rely on pockets of 658 liquid randomly stored in between two graphene sheets [113]. 659 Clark et al. fabricated liquid cells using lithographically patterned few-layer graphene (FLG), 660 molybdenum disulfide (MoS2), and hBN membranes [114]. Illustrated in Figure 14, it has a 661 monolayer of MoS2 sealed between FLG, with hBN as spacers. They reported that atomically 662 flat hBN sheets form a hermetic seal with graphene and MoS2. This design has a highly 663 controlled total thickness of less than 70 nm which enables single atom tracking. This method 664 could be applicable to a wide variety of applications in not only physical sciences but other 665 fields of research.  666  667 Concluding remarks and Outlook 668 We have reviewed the in-situ TEM techniques, focusing on holders for high temperature, gas 669 reaction, and liquid environments. Those techniques enable the real-time observation of 670 36 materials responses at the atomic scale under specific conditions. 671 In the context of in-situ heating TEM, aside from powder and the 3 mm-diameter disc 672 specimens, it is also feasible to handle specimens prepared using FIB processing. By combing 673 these holders with E-TEM, the in-situ gas heating experiments can be carried out. The E-cell 674 holders utilizing MEMS technology are capable of precise temperature control during the in-675 situ liquid and gas heating experiments. These E-cell holders, along with gas-injection holders, 676 can be utilized in TEM with standard setups. Choosing the appropriate holder aligned with 677 specific experimental goals greatly facilitates in-situ TEM studies. 678 A major challenge in general in-situ research is efficient quantitative data acquisition. This can 679 be addressed through the design of simplified and repeatable workflows, as well as through AI-680 assisted data acquisition and data interpretation.  Additionally, there is a need for a deeper 681 understanding of the interaction between the electron beam and the specimen. Controlling the 682 dose of electron beam is crucial to minimize radiolysis effects. Another approach is to improve 683 the temporal resolution of in-situ TEM by using pulsed electron source [115, 116].  684 In the case of liquid-phase microscopy, advancements in liquid cell design are necessary to 685 address two key aspects: 1) controlling the flow and Brownian motion; and 2) better control 686 over the liquid thickness. By addressing these challenges, in-situ TEM offers unprecedented 687 opportunities to elucidate of material functionality and dynamics with atomic-level precision 688 and high temporal resolution across various fields. 689 37  690 Acknowledgements 691 MSG thanks the support by the Ontario Graduate Scholarship.  JYH ack acknowledges the 692 support by the Natural Sciences and Engineering Research Council (NSERC) of Canada.  693 AH thanks the financial supports by Precursory Research for Embryonic Science and 694 Technology (PRESTO) [No. JPMJPR17S7], Japan Science and Technology Agency (JST), and 695 Grant-in-Aid for Scientific Research (C) (KAKENHI) [No. 25390035], Japan Society for the 696 Promotion of Science (JSPS). 697  698 Data Availability Statements 699 No new data were generated or analysed in support of this research. 700 Conflict of Interest 701 The authors declare that they have no conflict of interest. 702 References 703 [1] Marton L (1935) La microscopie electronique des objets biologiques. Bull. Sci Acad R 704 Belg. 21: 553.  705  706 [2] Ruska E (1942) Beitrag zur übermikroskopischen Abbildung bei höheren Drucken. 707 Kolloid-Zeitschrift. 100(2):212-219.  708  709 38 [3] von Ardenne M (1942) Reaction chamber supermicroscopy by means of the universal 710 electron microscope. Z. physik. Chem. B 52: 61-71.  711  712 [4] Abrams I M and McBain J W (1944) A closed cell for electron microscopy. J Appl Phys. 713 15(8): 607-609.  714  715 [5] Hashimoto H, Naiki T, Eto T, and Fujiwara K (1968) High temperature gas reaction 716 specimen chamber for an electron microscope. Jpn J Appl Phys. 7(8): 946-952.  717  718 [6] Swann P and Tighe N (1971) High voltage microscopy of gas oxide reactions. 719 Jernkontorets Annaler. 155(8): 497-501. 720  721 [7] Heide H G (1960) Elektronenmikroskopie von Objekten unter Atmosphärendruck oder 722 unter Drucken, welche ihre Austrocknung verhindern. Naturwissenschaften. 47(14): 313-317.  723  724 [8] Heide H G (1962) Electron microscopic observation of specimens under controlled gas 725 pressure. J Cell Biol 13(1): 147-152.  726  727 [9] Escaig J and Sella C J. Escaig and C. Sella,(1966) Object cell for observing specimen heated 728 under controlled atmosphere. Proc. 6th Int,Congress for Electron Microscopy. 1: 177. 729  730 [10] Fukami A, Etoh T, Ishihara N, Katoh M, and Fujiwara K (1970) Pressurized specimen 731 chamber for electron microscope. Proc. 7th Int. Congress on Electron Microscopy. 1: 171. 732  733 39 [11] Dupouy G (1968) Electron Microscopy at Very High Voltages, In: Barer R and Cosslett 734 V E (ed.) Adv. Opt. Electron Microsc. 2: pp.168-250 (Academic Press).  735  736 [12] Fujita H, Komatsu M, and Ishikawa I (1976) A universal environmental cell for a 737 3MVclass electron microscope and its applications to metallurgical subjects. Jpn J Appl Phys. 738 15(11): 2221-2228.  739  740 [13] Fullam E F (1972) A Closed wet cell for electron-microscope. Rev. Sci. Instrum. 43: 741 245-247. 742  743 [14] Stoyanova I G and Mikhailovski G A (1959) Method and apparatus for studying wet 744 objects in the electron microscope. Biophysics. 4: 116-121. 745  746 [15] Stoyanova I G, Nekrasova T A, and Zaides A L (1960) A study of collagen in humid 747 condition in a gas microchamber of an electron microscope. The effect of Ionizing Radiation 748 on collagen. Doklady Akad. Nauk  S.S.S.R 130：1366-1369. 749  750 [16] Stoyanova I G, Nekrasova T A, and Biryuzova V I (1960) Study of action of radiation on 751 bacterial cells in moist microchamber of electron microscope. Doklady. Biological Sciences. 752 131(1-6): 185-188. 753  754 [17] Dupouy G, Perrier F, and Durrieu L (1960) Observation of living matter by means of a 755 high voltage electron microscope. C. r. hebd. Séanc. Acad. Sci., Paris. 251: 2836-2841. 756  757 [18] Dupouy G and Perrier F (1962) Microscopie electronique fonctionnant sous une tension 758 40 d’un million de volts. J. de Microscopie. 1: 67-92. 759  760 [19] Allinson D L (1970) Environmental cell for use in a high voltage electron microscope. 7th 761 Int. Congr. Electron Microscopy. Grenoble. I: 169-170. 762  763 [20] Nagata F and Ishikawa I (1972) Observation of wet biological materials in a high voltage 764 electron microscope. J. Appli. Phys. 11: 1239-1244. 765  766 [21] Double D D (1973) Some studies of the hydration of portland cement using high voltage 767 (1MV) electron microscopy. Mat. Sci. Eng. 12: 29-34. 768  769 [22] Fukami A (1974) Merit and Demerit of Observation Method Using Environmental Cell. J 770 Electron Microsc. 23(1): 64. 771  772 [23] Creemer F, Helveg S, Hoveling G H, Ullmann S, Molenbroek A M, Sarro P M, and 773 Zandbergen H W (2008) Atomic-scale electron microscopy at ambient pressure. 774 Ultramicroscopy. 108: 993-998.  775  776 [24] Takahashi N, Takeyama T, Ito K, Ito T, Mihama K, and Watanabe M (1956) High 777 temperature furnace for the electron microscope. J Electron Microsc. 4(1): 16-23.  778  779 [25] Sinclair R and Parker MA (1986) High-resolution transmission electron microscopy of 780 silicon re-growth at controlled elevated temperatures. Nature. 322(6079): 531-533.  781  782 41 [26] Verheijena M A and Donkers J J T M, Thomassen J F P, van den Broek J J , van der Rijt 783 R A F, Dona M J J, and Smit C M (2004) Transmission electron microscopy specimen holder 784 for simultaneous in situ heating and electrical resistance measurements. Rev. Sci. Instrum. 785 75(2): 426-429. 786  787 [27] Hidaka K, Fujiwara T, and Kamino T (1997) Trial Manufacturing of Heating Stage for In-788 situ Observation by TEM. Journal of Surface Analysis. 3(2): 431-434. 789  790 [28] Yuan W, Wang Y, Li H, Wu H, Zhang Z, Selloni A, and Sun C (2016) Real-Time 791 Observation of Reconstruction Dynamics on TiO2(001) Surface under Oxygen via an 792 Environmental Transmission Electron Microscope. Nano Lett. 16: 132-137. 793  794 [29] Yuan W, Jiang Y, Wang Y, Kattel S, Chou L Y, Tsung C K, Wei X, Li J, Zhang X, Wang G, 795 Mao S X, and Zhang Z (2015) In situ observation of facet-dependent oxidation of graphene on 796 platinum in an environmental TEM. Chem. Commun. 51: 350-353. 797  798 [30] Shimada Y, Yoshida K, Inoue K, Shiraishi T, Kiguchi T, Nagai Y, and Konno T J (2019) 799 Evaluation of spatial and temporal resolution on in situ annealing aberration-corrected 800 transmission electron microscopy with proportional–integral–differential controller. 801 Microscopy. 68(3): 271–278.  802  803 [31] Jones J S and Swann P R (1991) Specimen heating holder for electron microscopes. 804 United States Patent, No. 4,996,433. 805  806 [32] Zheng Q, Jiang J, Xu G, Yu J, Tang L, and Li S (2020) New insights into the role of 807 42 Portlandite in the cement system: elastic anisotropy, thermal stability, and structural 808 compatibility with CSH. Cryst, Growth Des. 20: 2477-2488. 809  810 [33] Komatsu M, Fujita H and Matsui H (1982) A new type of very-high temperature heating 811 stage and in situ experiments on α-Al2O3 crystals. Jpn. J. Appl. Phys. 21 (8) : 1233-1237. 812  813 [34] Kamino T and Saka H (1993) Newly developed high resolution hot stage and its 814 applications to materials science. Microsc. Microanal. Microstruct. 4: 127-135. 815  816 [35] Kamino T, Yaguchi T and Saka H (1994) In Situ Study of Chemical Reaction between 817 Silicon and Graphite at l,400°C in a High Resolution/Analytical Electron Microscope. J 818 Electron Microsc.. 43(2): 104-110. 819  820 [36] Kamino T, Sasaki K, and Saka H (1997) High Resolution Electron Microscopy In Situ 821 Observation of Dynamic Behavior of Grain Boundaries and Interfaces at Very High 822 Temperatures. Microscopy and Microanalysis. 3(5): 393-408. 823  824 [37] Yasuda H (1997) Spontaneous alloying in nanometer-sized particles. Denshikenbikyou. 825 32(1): 17-22 (in Japanese). 826  827 [38] Mori H, Yasuda H, and Kamino T (1994) High-resolution electron 828 microscopy study of spontaneous alloying in gold clusters. Philosophical 829 magazine letters. 69(5): 279-283. 830  831 https://www.tandfonline.com/doi/abs/10.1080/09500839408241604https://www.tandfonline.com/doi/abs/10.1080/0950083940824160443 [39] Tanigaki T T, Ito K, Nagakubo Y, Asakawa T, and Kanemura T (2009) An in situ heating 832 TEM analysis method for an interface reaction. J Electron Microsc. 58(5): 281-287.  833  834 [40] Geisberger A A and Sarkar N (2006) Techniques in MEMS Microthermal Actuators and 835 Their Applications. In: Leondes C T (ed.) MEMS/NEMS. pp. 201-261, (Springer, Boston). 836  837 [41] Zhang M, Olson E A, Twesten R D, Wen J G, Allen L H, Robertson I M, and Petrov I 838 (2005) In situ transmission electron microscopy studies enabled by microelectromechanical 839 system technology. J. Mater. Res. 20(7): 1802-1807.  840  841 [42] Perez Garza H H, Morsink D, Xu J, Sholkina M, Pivak Y, Pen M, van Weperen S, and Xu 842 Q (2016) MEMS-based nanoreactor for in situ analysis of solid-gas interactions inside the 843 transmission electron microscope. Micro Nano Lett. 12(2): 69-75.  844  845 [43] Mele L, Santagata F, Mihailovic M, Rossi T, Tran A T, Schellevis H, Creemer J F, and 846 Sarro P M (2012) A molybdenum MEMS microhotplate for high-temperature operation. Sens. 847 Actuator A Phys. 188: 173-180.  848  849 [44] van Omme J T, Zakhozehva M, Spruit R G, Sholkina M, and Perez Garza H H (2018) 850 Advanced microheater for in situ transmission electron microscopy; enabling unexplored 851 analytical studies and extreme spatial stability. Ultramicroscopy 192: 14-20.  852  853 [45] Norcada Inc., "MEMS IN-SITU HEATING," [Online]. Available: 854 https://www.norcada.com/products/in-situ-tem/. [Accessed 15 May 2023]. 855  856 44 [46] Zhang F, Zhang X, Jia Z, and Liu W (2022) Precise drift tracking for in situ transmission 857 electron microscopy via a thon-ring based sample position measurement. Microsc. Microanal. 858 28: 1945-1951.  859  860 [47] Mele L, Konings S, Dona P, Evertz F, Mitterbauer C, Faber P, Schampers R, and Jinschek 861 J R (2016) A MEMS-based heating holder for the direct imaging of simultaneous in-situ 862 heating and biasing experiments in scanning/transmission electron microscopes. Microsc. Res. 863 Tech. 79: 239-250.  864  865 [48] Olson E A, Efremov M Y, Zhang M, Zhang Z, and Allen L H (2003) The design and 866 operation of a MEMS differential scanning nanocalorimeter for high-speed heat capacity 867 measurements of ultrathin films. J. Microelectromechanical Syst. 12(3): 355-364. 868   869 [49] Allard L F, Bigelow W C, Jose-Yacaman M, Nackashi D P, Damiano J, and Mick S E 870 (2009) A new MEMS-based system for ultra-high-resolution imaging at elevated 871 temperatures. Microsc. Res. Tech. 72: 208-215.  872  873 [50] Creemer J F, Helveg S, Kooyman P J, Molenbroek A M, Zandbergen H W, and Sarro P M 874 (2010) A MEMS reactor for atomic-scale microscopy of nanomaterials under industrially 875 relevant conditions. J. Microelectromechanical Syst. 19(2): 254-264.  876  877 [51] Hotovy I, Rehacek V, Mika F, Lalinsky T, Hascik S, Vanko G, and Drzik M (2008) 878 Gallium arsenide suspeded microheater for MEMS sensor arrays. Microsyst. Technol. 14: 879 629-635.  880  881 45 [52] Mele L, Rossi T, Riccio M, Iervolino E, Santagata F, Irace A, Breglio G, Creemer J F, 882 and Sarro P M (2011) Electro-thermal analysis of MEMS microhotplates for the optimization 883 of temperature uniformity. Procedia Eng. 25: 387-390.  884  885 [53] Baroncini M, Placidi P, Cardinali G C, and Scorzoni A (2004) Thermal characterization 886 of a mciroheater for micromachined gas sensors. Sens. Actuator A Phys. 115: 8-14.  887  888 [54] Zhao J, Liang L, Tang S, Zhang G, Su Y, Zhao Y, Li M, Zhang L, Fan S, Li Q, and Wei Y 889 (2023) Graphene microheater chips for in situ TEM. Nano Lett. 23: 726-734.  890  891 [55] Bernal R A, Ramachandramoorthy R, and Espinosa H D (2015) Double-tilt in situ TEM 892 holder with multiple electrical contacts and its application in MEMS-based mechanical testing 893 of nanomaterials. Ultramicroscopy. 156: 23-28.  894  895 [56] Lee K, Park J, Jung S I, Hajra S, and Kim H J (2021) Direct integration of carbon 896 nanotubes on a suspended Pt microheater for hydrogen gas sensing. J. Mater. Sci. Materl. 897 Electron. 32: 19626-19634.  898  899 [57] Protochips, "Electrical E-chips - Finger Configuration - Fusion Select," [Online]. 900 Available: https://protochips.shop/products/fusion-electrical-echips-finger. [Accessed 12 May 901 2023]. 902  903 [58] Protochips, "Heating E-chips - Fusion Select," [Online]. Available: 904 https://protochips.shop/products/fusion-thermal-echips. [Accessed 12 May 2023]. 905  906 46 [59] Castagna M E, Modica R, Cascino S, Moschetti M, Cerantonio V, Messina A, and 907 Santangelo A (2018) A high stability and uniformity W micro hot plate. Sens. Actuator A 908 Phys. 279: 617-623.  909  910 [60] Diallo T M, Aziziyan M R, Arvinte R, Harmand J-C, Patriarche G, Renard C, Fafard S, 911 Ares R, and Boucherif A (2022) In-situ transmission electron microscopy observation of 912 germanium growth on freestanding graphene: unfolding mechanism of 3D crystal growth 913 during van der waals epitaxy. Small. 18: 2101890.  914  915 [61] Protochips, "Technical Note: Closed Loop Temperature Control," [Online]. Available: 916 https://www.protochips.com/wp-content/uploads/2017/11/tech-closed-loop-temperature-917 control.pdf. [Accessed 12 May 2023]. 918  919 [62] DENSsolutions, "The revolutionary Wildfire Nano-Chip," [Online]. Available: 920 https://denssolutions.com/products/wildfire/nano-chip/. [Accessed 12 May 2023]. 921  922 [63] DENSsolutions, "Lightning In Situ TEM Biasing & Heating," [Online]. Available: 923 https://denssolutions.com/products/lightning/. [Accessed 12 May 2023]. 924  925 [64] Khanna V K, Prasad M, Dwivedi V K, Shekhar C, Pankaj A C, and Basu J (2007) Design 926 and electro-thermal simulation of a polysilicon microheater on a suspended membrane for use 927 in gas sensing. Indian J. Pure Appl. Phys. 45: 332-335.  928  929 47 [65] Laconte J, Dupont C, Flandre D, and Raskin J P (2004) SOI CMOS compatible low-930 power microheater optimization for the fabrication of smart gas sensors. IEEE Sens. J4(5): 931 670-680.  932  933 [66] Sberveglieri G, Hellmich W, and Muller G (1997) Silicon hotplates for metal oxide gas 934 sensor elements. Microsyst. Technol. 3: 183-190.  935  936 [67] Rydberg M and Smith U (2000) Long-term stability and electrical properties of 937 compensation doped poly-Si IC-resistors. IEEE Trans. Electron. Devices 47(2): 417-426.  938  939 [68] El-Zoka A A, Howe J Y, Newman R C, and Perovic D D (2019) In situ STEM/SEM 940 study of the coarsening of nanoporous gold. Acta Mater. 182: 67-77.  941  942 [69] Li M, Qiu C, Dogel S, Chen P, Perovic D D, and Howe J Y (2022) Microstructure-943 dependent thermal stability of super-tetragonal nanocomposite films through in situ 944 TEM/EELS study. ACS Appl. Mater. Interfaces. 14: 52316-52323.  945  946 [70] Krisper R, Lammer J, Pivak Y, Fisslthaler E, and Grogger W (2022) The performance of 947 EDXS at elevated sample temperatures using a MEMS-based in situ TEM heating system. 948 Ultramicroscopy. 234: 113461.  949  950 [71] Boyes E D and Gai P L (1997) Environmental high resolution electron microscopy and 951 applications to chemical science. Ultramicroscopy. 67: 219-232.  952  953 48 [72] Yokosawa T, Alan T, Pandraud G, Dam B, and Zandbergen H (2012) In-situ TEM on 954 (de)hydrogenation of Pd at 0.5–4.5 bar hydrogen pressure and 20-400 °C. Ultramicroscopy . 955 112(1): 47-52.  956  957 [73] Yaguchi T, Suzuki M, Watabe A, Nagakubo Y, Ueda K, and Kamino T (2011) 958 Development of a high temperature–atmospheric pressure environmental cell for high-959 resolution TEM. J Electron Microsc. 60(3): 217-225.  960  961 [74] Hashimoto H, Naiki T, Watanabe M, Eto T, Fujiwara K and Nagahama Y (1966) Specimen 962 chamber for observing the reaction process with gas at high temperature. Proc. 6th Int. Congr. 963 for Electron Microscopy. Kyoto１:181. 964  965 [75] Hashimoto H, Naiki T, Mannami M and Fujita K (1959) Electron Microscopic Study by 966 Cine Film, Structure and Property of Thin Film. pp.71-86 (Neugebauer, Newkirk, Vermilyea., 967 John Wiley & Sons Inc.). 968  969 [76] Baker RTK, Harris PS (1972) Controlled atmosphere electron microscopy. J Phys E: Sci 970 Instrum. 5(8): 793-797.  971  972 [77] Hashimoto H, Tanaka K, Yoda E (1958) A specimen treating device at high temperature 973 for the electron microscope. J Electron Microsc. 6(1): 8–11.  974  975 [78] FEI Company, "Titan ETEM G2," 2012. [Online]. Available: 976 https://photos.labwrench.com/equipmentManuals/14127-5457.pdf. [Accessed 12 May 2023]. 977 https://jlc.jst.go.jp/DN/JALC/00300185261?type=list&lang=ja&from=J-STAGE&dispptn=1https://jlc.jst.go.jp/DN/JALC/00300185261?type=list&lang=ja&from=J-STAGE&dispptn=1https://jlc.jst.go.jp/DN/JALC/00300185261?type=list&lang=ja&from=J-STAGE&dispptn=149  978 [79] ThermoFischer Scientific, "Themis ETEM," 2020. [Online]. Available: 979 https://assets.thermofisher.com/TFS-Assets/MSD/Datasheets/themis-etem-datasheet.pdf. 980 [Accessed 12 May 2023]. 981  982 [80] Boyes E D, Ward M R, Lari L, and Gai P L (2013) ESTEM imaging of single atoms under 983 controlled temperature and gas environment conditons in catalyst reaction studies. Ann. Phys. 984 525(6): 423-429. 985  986 [81] Kishita K, Sakai H, Tanaka H, Saka H, Kuroda K, Sakamoto M, Watabe A and Kamino T 987 (2009) Development of an analytical environmental TEM system and its application. J Electron 988 Microsc. 58(6): 331-339.  989  990 [82] Matsumoto H, Konno M, Sato T, Nagaoki I, Yaguchi T and Howe J (2013) Simultanious 991 in situ SEM and STEM analysis of gas/catalyst reaction in a cold field-emission environmental 992 TEM. Microscopy and Analysis. 27(7): 13-18. 993  994 [83] Shirai M, Hanawa A, Kikuchi H, Inada H, Matsumoto H (2019) In-situ observation of 995 catalytic reaction in gas atmosphere using an aberration corrected STEM. Microsc Microanal. 996 25(S2):1526-1527.  997  998 [84] Lv H, Lin L, Zhang X, Li R, Song Y, Matsumoto H, Ta N, Zeng C, Fu Q, Wang G, and 999 Bao X (2021) Promoting exsolution of RuFe alloy nanoparticles on Sr2Fe1.4Ru0.1Mo0.5O6−δ via 1000 repeated redox manipulations for CO2 electrolysis. Nature Communications.12: 5665. 1001  1002 50 [85] Kamino T, Yaguchi T, Konno M, Watabe A, Marukawa T, Mima T, Kuroda K, Saka 1003 H, Arai S, Makino H, Suzuki Y and Kishita K.(2005) Development of a gas 1004 injection/specimen heating holder for use with transmission electron microscope. J Electron 1005 Microsc. 54(6): 497–503.  1006  1007 [86] Yaguchi T, Kanemura T, Shimizu T, Imamura D, Watabe A and Kamino T (2012) 1008 Development of a technique for in situ high temperature TEM observation of catalysts in a 1009 highly moisturized air atmosphere, J Electron Microsc. 61(4): 199–206. 1010  1011 [87] Kamino T, Yaguchi T, Konno M, Watabe A and Nagakubo Y (2006) Development of a 1012 specimen heating holder with an evaporator and gas injector and its application for catalyst. J 1013 Electron Microsc. 55(5): 245–252.  1014  1015 [88] Tsukasaki H, Igarashi K, Wakui A, Yaguchi T, Nakajima H, Kimura T, Sakuda A, 1016 Tatsumisago M, Hayashi A and Mori S (2021) In situ observation of the deterioration process 1017 of sulfide-based solid electrolytes using airtight and air-flow TEM systems. Microscopy. 24: 1018 519-525.  1019  1020 [89] Tao F F and Crozier P A (2016) Atomic-scale observations of catalyst structures under 1021 reaction conditions and during catalysis. Chem. Rev. 116: 3487-3539. 1022  1023 [90] Gai P L, Boyes E D, Helveg S, Hansen P L, Giorgio S, and Henry C R (2007) Atomic-1024 resolution environmental transmission electron microscopy for probing gas-solid reactions in 1025 heterogeneous catalysis. MRS Bull. 32: 1044-1050. 1026 javascript:;javascript:;javascript:;javascript:;javascript:;javascript:;javascript:;javascript:;51  1027 [91] Crewe A V, Wall J, and Langmore J (1970) Visibility of single atoms. Science. 1028 168(3937): 1338-1340.  1029  1030 [92] Boyes E D, LaGrow A P, Ward M R, Martin T E, and Gai P L (2020) Visulizing single 1031 atom dynamics in heterogeneous catalysis using analytical in situ environmental scanning 1032 transmission electron microscopy. Philos. Trans. Royal Soc. A 378(20190605): 1-23. 1033   1034 [93] Martin T E, Mitchell R W, Boyes E D, and Gai P L (2020) Atom-by-atom analysis of 1035 sintering dynamics and stability of Pt nanoparticle catalysts in chemical reactions. Philos. Trans. 1036 Royal Soc. A 378(20190597): 1-21. 1037  1038 [94] Yuan W, Zhu B, Fang K, Li X Y, Hansen T W, Ou Y, Yang H, Wagner J B, Gao Y, Wang 1039 Y, and Zhang Z (2021) In situ manipulation of the active Au-TiO2 interface with atomic 1040 precision during CO oxidation. Science. 371: 517-521. 1041  1042 [95] Yuan W, Fang K, You R, Zhang Z, and Wang Y (2023) Toward in situ atomistic design of 1043 catalytic active sites via controlled atmosphere transmission electron microscopy. Acc. Mater. 1044 Res. 4: 275-286.  1045  1046 [96] Ito T and Hiziya K (1958) A Specimen Reaction Device for the Electron Microscope and 1047 its Applications. Journal of Electron Microscopy. 6: 4-8. 1048  1049 [97] Hashimoto A, Han Y, Akimoto H, Hozumi R and Takeguchi M (2021) Development of a 1050 gas environmental heating specimen holder system using differential pumping. Microscopy. 1051 52 70: 545-549. 1052  1053 [98] Shoji S, Peng X, Imai T, Kumar P S M, Higuchi K, Yamamoto Y, Tokunaga T, Arai S, 1054 Ueda S, Hashimoto A, Tsubaki N, Miyauchi M, Fujita T, and Abe H (2019) Topologically 1055 immobilized catalysis centre for long-term stable carbon dioxide reforming of methane. 1056 Chemical Science. 10: 3701-3705. 1057  1058 [99] Tanaka N, Usukura J, Kusunoki M, Saito Y, Sasaki K, Tanji T, Muto S, and Arai S (2013) 1059 Development of an environmental high-voltage electron microscope for reaction science. 1060 Microscopy. 62: 205-215. 1061  1062 [100] Tanaka N, Fujita T, Takahashi Y, Yamasaki J, Murata K, and Arai S (2020) Progress in 1063 environmental high-voltage transmission electron microscopy for nanomaterials. Phil. Trans. 1064 R. Soc. A. 378: 20190602. 1065  1066 [101] Muto S, Arai S, Higuchi S, Orita K, Ohta S, Tanaka H, Suganuma T, Ibe M, and Hirata 1067 H (2019) Environmental high-voltage S/TEM combined with quadrupole mass spectrometer 1068 for concurrent in-situ structural characterization and detection of product gas molecules 1069 associated with chemical reactions. Microscopy. 68: 185-188. 1070  1071 [102] Fujita T, Higuchi K, Yamamoto Y, Tokunaga T, Arai S, and Abe H (2017) In-Situ TEM 1072 Study of a Nanoporous Ni–Co Catalyst Used for the Dry Reforming of Methane. Metals. 1073 7(10): 406. 1074  1075 https://pubs.rsc.org/en/results?searchtext=Author%3APaskalis%20Sahaya%20Murphin%20Kumarhttps://pubs.rsc.org/en/results?searchtext=Author%3AKimitaka%20Higuchihttps://pubs.rsc.org/en/results?searchtext=Author%3AYuta%20Yamamotohttps://pubs.rsc.org/en/results?searchtext=Author%3ATomoharu%20Tokunagahttps://pubs.rsc.org/en/results?searchtext=Author%3AShigeo%20Araihttps://pubs.rsc.org/en/results?searchtext=Author%3AShigenori%20Uedahttps://pubs.rsc.org/en/results?searchtext=Author%3AAyako%20Hashimotohttps://pubs.rsc.org/en/results?searchtext=Author%3ANoritatsu%20Tsubakihttps://pubs.rsc.org/en/results?searchtext=Author%3AMasahiro%20Miyauchihttps://pubs.rsc.org/en/results?searchtext=Author%3ATakeshi%20Fujitajavascript:;javascript:;javascript:;javascript:;javascript:;javascript:;javascript:;javascript:;https://royalsocietypublishing.org/doi/full/10.1098/rsta.2019.0602https://royalsocietypublishing.org/doi/full/10.1098/rsta.2019.0602https://royalsocietypublishing.org/doi/full/10.1098/rsta.2019.0602https://royalsocietypublishing.org/doi/full/10.1098/rsta.2019.0602https://royalsocietypublishing.org/doi/full/10.1098/rsta.2019.0602https://royalsocietypublishing.org/doi/full/10.1098/rsta.2019.060253 [103] Alan T, Yokosawa T, Gaspar J, Pandraud G, Paul O, Creemer F, Sarro P M, and 1076 Zandbergen H W (2012) Micro-fabricated channel with ultra-thin yet ultra-strong windows 1077 enables electron microscopy under 4-bar pressure. Appl. Phys. Lett. 100:081903. 1078   1079 [104] Protochips Inc., "Atmosphere AX: Machine Vision Environmental Cell for TEM," 1080 [Online]. Available: https://www.protochips.com/products/atmosphere/. [Accessed 10 June 1081 2023]. 1082  1083 [105] Dai S, Zhang S, Katz M B, Graham G W, and Pan X (2017) In situ observation of Rh-1084 CaTiO3 catalysts during reduction and oxidation treatments by transmission electron 1085 microscopy. ACS Catal. 7(3): 1579-1582.  1086  1087 [106] Ross F M (2016) Liquid Cell Electron Microscopy.In Ross F M(ed.) Advances in 1088 Microscopy and Microanalysis (Cambridge University Press, Cambridge).  1089  1090 [107] DiCecco LA, Gao R, Athanasiadou D, Chan RL, Carneiro KMM, Kelly DF, Sone E, 1091 Grandfield K(2022) Exploring Calcium Phosphate Biomineralization Systems Using In Situ 1092 Liquid Phase Electron Microscopy Microsc Microanal. 28(S1):1818-1820. 1093  1094 [108] Abrams I M and McBain J W (1944) A closed cell for electron microscopy. Science. 1095 100(2595): 273-274.  1096  1097 [109] Williamson M J, Tromp R M, Vereecken P M, Hull R, and Ross F M (2003) Dynamic 1098 microscopy of nanoscale cluster growth at the solid-liquid interface. Nat. Mater. 2(8): 532-1099 536.  1100 54  1101 [110] Zheng H, Smith R K, Jun Y W, Kisielowski C, Dahmen U, and Alivisatos A P (2009) 1102 Observation of single colloidal platinum nanocrystal growth trajectories. Science. 324(5932): 1103 1309-1312.  1104  1105 [111] Hummingbird Scientific, "Optical Bulk Liquid Electrochemistry," [Online]. Available: 1106 https://hummingbirdscientific.com/products/optical-bulk-liquid/. [Accessed 16 May 2023]. 1107  1108 [112] Prozorov T, Almeida T P, Kovacs A, and Dunin-Borkowski R E (2017) Off-axis electron 1109 holography of bacterial cells and magnetic nanoparticles in liquid. J. R. Soc. Interface. 1110 14(20170464): 1-10. 1111   1112 [113] Yuk J M, Park J, Ercius P, Kim K, Hellebusch D J, Crommie M F, Lee J Y, Zettl A, and 1113 Alivisatos A P (2012) High-Resolution EM of Colloidal Nanocrystal Growth Using Graphene 1114 Liquid Cells. Science. 336(6077): 61-64.  1115  1116 [114] Clark N, Kelly D J, Zhou M, Zou Y C, Myung C W, Hopkinson D G, Schran C, 1117 Michaelides A, Gorbachev R, and Haigh S J (2022) Tracking single adatoms in liquid in a 1118 transmission electron microscope. Nature. 609(7929): 942-947.  1119  1120 [115] Kuwahara M, Agemura T (2023) Temporal resolution in transmission electron 1121 microscopy using a photoemission electron source. Microscopy, 72(2):97-110. 1122  1123 55 [116] Kim J S, Langrange T, Reed B W, Taheri M L, Armstrong, King W.E, Browning N D, 1124 Campbell G H (2008) Imaging of Transient Structures Using Nanosecond in Situ TEM. Science 1125 321(5895):1472-1475.  1126   1127 56 Figure Legends 1128 Figure 1 In-situ annealing system is constituted of a double tilt furnace type heating holder and 1129 thermal control box with a Proportional-Integral-Differential (PID) controller. Reproduced 1130 from reference [30 ], with the permission of Oxford University Press. 1131  1132 Figure 2 An external view of the direct type of side-entry single tilt heating holder, “Kamino 1133 holder”. A fine W filament of 25 µm in diameter, which is bridged across two electrodes, is 1134 heated by direct electric current from a battery.  1135  1136 Figure 3 Schematic diagram of a holder consists of the 3 mm-diameter-grid having a support 1137 film and 2 W-filaments for sample evaporation. Reproduced from reference [37], with the 1138 permission of The Japanese Society of Microscopy. 1139  1140 Figure 4 Scanning Ion Microscope images showing a typical sample preparation sequence for 1141 the in-situ observation of the interface reaction. (a) The sampling area. (b) After the W-1142 deposition. (c) After the sputtering of surrounding area. (d) After cutting of the bottom. (e) 1143 Picking out from the substrate. (f) Attaching to the filament. (g) The space between the sample 1144 and the filament, after filling with the W-deposition. (h) After thinning. Reproduced from 1145 reference [39], with the permission of Oxford University Press. 1146  1147 Figure 5. Double spiral heating element designs by (a) Perez Garza et al.[42], (b) Mele et al.[47], 1148 and (c) van Omme et al.[44]. (a) shows the heating element that follow an octagonal shape 1149 57 using straight line segments; (b) shows the NanoEx™️ -i/v microheater chip, which concentrates 1150 all the electron-transparent windows in a central large circle. Each window is SiN and consists 1151 of several through holes; (c) shows the Nano-Chip in the Wildfire heating holder. The heating 1152 element of the Nano-Chip has increasing linewidth towards the centre of the double spiral with 1153 no sharp corners and displays excellent temperature uniformity; All images are reproduced with 1154 permission from the respective references. 1155  1156 Figure 6 Schematic diagram of “opened type” specimen chamber(a) and “closed type” one(b). 1157  1158 Figure 7 External view of the gas injection/specimen heating holder. A gas injection nozzle 1159 with an inner diameter of 0.5 mm was built about 1mm away from the heating element. This 1160 holder provides localized gas atmosphere near the specimen. 1161  1162 Figure 8 Procedure for synthesis of metal oxide support and deposition of catalyst nano-1163 particles in a TEM specimen chamber using the specimen heating holder equipped with a gas 1164 injection nozzle and an evaporator. (a) Oxidation of metal to synthesize metal oxide support. 1165 (b) Deposition of catalyst nano-particles on the metal oxide support. (c) Observation of the 1166 nano-catalyst at various environments.  1167  1168 58 Figure 9 External view of the tip of the TEM air protection transfer holder with a W filament 1169 and gas injection nozzle. (a) Shutter off, (b)Shutter on. 1170  1171 Figure 10  “Open-type” specimen holder with MEMS heating device. (a) Schematic diagram 1172 of the holder consisting of a differential pumping unit and a heating unit. (b) Interior of the 1173 specimen holder grip with a gas tank and piezo-driven fine control valve. Cross-sectional 1174 illustration (c) and photograph (d) of the specimen holder head with orifices, membrane-type 1175 heater, and pressure gage. Reproduced from reference [97], with the permission of Oxford 1176 University Press. 1177  1178 Figure 11 In-situ STEM observation on Ni#Y2O3. (a) EDX elemental mapping image of 1179 Ni#Y2O3 sliced specimen. Green: Y; Red: Ni. (b) In situ annular dark-field STEM image 1180 before the catalytic reaction (in vacuum at 450 oC) and (c) during the CH4 conversion 1181 condition (0.3 Pa CH4+CO2 gas at 450 oC). Reproduced from reference [98], with the 1182 permission of Royal Society of Chemistry. 1183  1184 Figure 12 (a) Illustration of a general view of E-HVEM. Reproduced from reference [99], 1185 with the permission of Oxford University Press. (b) Cross-sectional illustrations of a side-1186 entry goniometer to enable insertion of the gas chamber as an environmental cell from the 1187 left-hand side. Reproduced from reference [100], with the permission of the Royal Society 1188 Publishing. 1189  1190 Figure 13 Atomic-scale in situ gas study performed using E-cell. (a) and (b) Protochips 1191 59 Atmospheric AX gas cell [104]. HAADF image of Rh-CaTiO3 powder in 760 Torr of 5%H2/Ar 1192 (labeled by R) after 12 min at 500 °C: (c) large field of view shows diffuse sub-nanometer 1193 bright regions and occasional bright spots, marked by yellow arrows; (d) enlarged image, 1194 containing the small rectangular box in panel (c), where Ca (blue) and Ti (magenta) columns 1195 are identified; and (e) intensity along a line scanned from left to right within the rectangular 1196 box in panel (d); Reproduced with permission from reference[105].  1197  1198 Figure 14 Configuration of the graphene-MoS2-hBN liquid cell. Reproduced with permission 1199 from reference [114].   1200   1201 60 Figure 1           61 Figure 2                           62  Figure 3                            63 Figure 4                         64 Figure 5                 65 Figure 6                            66 Figure 7                          67 Figure 8                             68  Figure 9                69 Figure 10                         70 Figure 11                     71 Figure 12                            72 Figure 13                73 Figure 14