Since its invention, transmission electron microscopy (TEM) has been an invaluable addition to the materials science toolbox. It has been routinely used in fields such as catalysis, semiconductors and photonics as illustrated in a number of review papers [1–7]. No other tool provides the versatility of spatial information at high resolution along with the simultaneous acquisition of spectroscopic information. Since the early days of TEM, non-vacuum imaging has been pursued. For more than half a century the capability of imaging features at elevated temperatures in non-vacuum conditions using electrons has been available [8–11]. However, it is only within the last decade that commercially available TEMs have offered this option [12–16]. Scientists are now routinely exposing samples to heat, gas, liquid, stress, and light while performing electron microscopical investigations.
In order to expose samples to a gaseous environment, certain modifications have to be made. Gas can be introduced either via the microscope column or via the sample holder. In the former case, the microscope column itself is fitted with a gas inlet through the objective lens and differential pumping apertures in the upper and lower pole pieces. This setup is known as the differentially pumped column and is described by Boyes et al. . The other option is to inject gases via the sample holder. This has been described by Kishita et al. . In either case, microscopes capable of exposing samples to a gaseous 5 alpha reductase are now known as environmental transmission electron microscopes or ETEMs.
In order to simulate the working environment of e.g. an industrial catalyst, a gas atmosphere along with an ability to heat the sample is necessary. Heating is typically done using a heating holder. However, this also represents a challenge to conventional knowledge of TEM experimentation. Conventionally, heating experiments in the TEM have been carried out in high vacuum. This means that very little power is needed to heat the thermally isolated sample region (traditionally a metal grid 3mm in diameter) as there is virtually no gas atmosphere to remove heat from the sample. In an ETEM however, there is a flow of gas and thus a heat sink around the sample. Hence additional heat needs to be supplied in order to maintain a certain set temperature. Keeping the power constant when introducing gas can result in a significant temperature drop (typically hundred degrees under the conditions evaluated in the present work). Furthermore, as heat (in traditional heating holders) is supplied via the periphery of the 3mm metal grid, there may be a temperature gradient across the sample, rendering the center of the 3mm grid at a lower temperature than what is measured (typically by thermocouples) at the furnace supplying the heat. However, to the best knowledge of the authors, the level of non-uniformity of the sample holder temperature field in a gaseous atmosphere has not been investigated previously.
Using a simplified model of the inside of an objective lens, the gas flow, temperature and pressure field inside the ETEM chamber in steady state is calculated. The simulation was made using the Weakly Compressible Navier–Stokes and General Heat Transfer packages of the commercial software program .
The CFD model was defined and solved using ; a commercial multiphysics modeling and simulation software. A Gatan 652 double tilt heating holder in an FEI Titan 80-300 differentially pumped ETEM was used as basis for the model. Fig. 1(a), (b), and (c) shows a two dimensional cross-section of the modeled geometry of respectively the x–y-plane, y–z-plane, and x–z-plane.
The model is divided into thirteen regions, denoted as CX, where X denotes a specific region. Table 1 describes the function and material of each of these regions.
A sample holder is inserted from the left of Fig. 1(b). This consists of three sections: the holder barrel (C4), the holder tip for the furnace and sample (C5), and the furnace (C6). When using the ETEM the sample is placed in the center of the furnace on a grid (C13); the center of this grid is used as origin in the drawings of Fig. 1. A cold trap (C8) is placed around the holder and beneath the holder is the objective aperture, which consist of an objective aperture barrel (C9) and the objective diaphragm (C10). A sniffer (C7) connected to a quadropole mass spectrometer is placed parallel to the gas flow with the orifice facing the sample. The upper and lower pole pieces (truncated cones, C2 and C3) are located above and below the sample. Each pole piece has a Pt diaphragm (C11 and C12) at the cone center. C2 and C3 were assumed to be equivalent to 304 steel in the lack of better material knowledge and also considering that relevant material properties do not change much between different types of steel.