The scanning electron microscope (SEM): how it works and how I use it
Jeremy has spent his career in science and technology, namely acoustics and electronics. Whist reading engineering at Cambridge he was introduced to an early version of the Scanning Electron Microscope (SEM), the Cambridge Stereoscan Mk 1 of 1965 (see Fig. 1), and was blown away by its capabilities, a feeling of which continued throughout his working life. Interestingly, the Cambridge Instrument Company, the makers, foresaw only a need for a handful of these instruments and stopped their production line before demand really took off, a similar story to IBM, who in the 1940’s, foresaw a world demand of only 5 computers.
Jeremy established his SEM in his home in 2016 in part because he was disappointed at results from optical microscopes in pursuing his interest in arachnids. He also wanted to provide an opportunity for school students to experience the capabilities of the instrument and to provide support to business/academia.
In that respect he came to Science Café to allow Café members to benefit from his skills and insights into electron microscopy. Jeremy’s efforts at electron microscopy are those of the extremely skilled practitioner with a readiness to not only publish his results but give his images a highly skilled artistic flourish, making his approach to science similar to that of individuals during the enlightenment who perused a topic of interest for its own sake and were able to interpret results in both a scientific and artistic way. Jeremy’s SEM images are engaging, beautiful and reflect a high degree of scientific and artistic competence, placing him amongst the top echelon of electron microscopists.
The Electron Microscope
Most individuals with a passing association in science will be familiar with the standard light microscope (see Fig. 2) with the eyepiece, objective (various powers), stage (for specimen mounting), condenser and light source.
Fig. 2: Components of an upright light microscope
There is also an inverted optical microscope, which has the same architectural features, but reversed vertically (see Fig.3).
Fig. 3: Inverted and Upright Microscopy at a Glance. Inverted microscopy is popular for live cell imaging- sample access is from the top (for liquid exchange) and no contact between objective and sample, ensuring sterile conditions. The upright microscope is used samples squeezed between slide and coverslip, and fixed samples (e.g. cells and tissue sections)
Fig. 4: Components of an SEM
The component setup of an inverted optical microscope is identical to that of an electron microscope (Fig.4) for which the former provides an excellent analogue. There are major difference in detail, namely illumination is by electron bean, focussing lenses are magnetic coils, not glass optics etc. In an optical microscope, a suitable wavelength for use is 500nm, for electron beams that wavelength is considerably less, at around 5 picometres. Because resolution is a function of wavelength, an electron microscope allows the operator to see considerably smaller detail. In light microscopy, 200nm is the limit of resolution compared to 1nm in the SEM. The SEM also has a sibling, the TEM (Transmission Electron Microscope) which is geared towards internal sample structure rather than with SEM, which looks at surface topography.
In practice too, there are other differences. To avoid scattering of the electron beam, and to preserve the filament, a partial vacuum is required within the instrument, namely at 5x10-4 Pascals in the SEM chamber and 5x10-8 Pa in the electron gun. This level of vacuum is considered low and requires pumps to achieve and maintain that level of pressure.
Among the disadvantages of an SEM over a light microscope are the fact that an SEM is many times more expensive than a conventional light microscope and also, the image does not contain any colour information (see Fig. 8).
The specimen must be conducting to prevent charge build up. If it is not, it can be sputtered with a thin layer of gold or silver, or often, in the case of rocks, carbon. The specimen should also be dry because water molecules will collide with the electron beam disrupting imaging.
Some Technical Points
Scanning- this can take the form of a Raster scan where the specimen is subdivided into a sequence of horizontal strips or scan lines (see Fig.5). Each scan line can be transmitted in the form of an analogue signal, or further divided into discrete pixels for processing. In raster scanning, the beam sweeps horizontally left-to-right at a steady rate, then blanks and rapidly moves back to the left, where it turns back on and sweeps out the next line. During this time, the vertical position is also steadily increasing (downward), but much more slowly – there is one vertical sweep per image frame, but one horizontal sweep per line of resolution.
A more sophisticated method is the so-called Dwell Digital where the scanner dwells on each pixel to increase signal to noise ratio.
Fig. 5: SEM scanning
Imaging- The SEM uses a beam of electrons to create the image of the specimen. The electron beam is scanned across the surface of the object and electrons released at the surface of the specimen by the interaction of the beam and the specimen are detected and used to create an image of the specimen’s surface.
Drying samples- surface tension can distort a sample as it dries. To avoid this, it is dried in a critical point dryer with a particular temperature/pressure combination that prevents distortion. A fluid which works well at room temperature is CO2, which has a critical point at 31.5 Celsius and a pressure of 74 Bars.
Beam-specimen interaction- when the beam hits and enters the specimen, electrons bounce around and the specimen emits secondary electrons and X-rays. The frequency of emitted X-rays depends on the material. The number of secondary electrons emitted is proportional to sine of the angle of that part of the specimen being examined to the horizontal. Steeper areas appear bright and flatter areas darker. A detector is used to count electrons for each point on the sample thereby obtaining an image.
Mounting specimens- specimens are mounted on stumps with double sided sticky tabs, to stabilse the specimen. Form here, they can, if required, be sputtered with gold. The stump can rotate in 3 dimensions (X, Y and Z) for better imaging.
The set up
Having started in 2016, Jeremy is now on his 2nd SEM, plus associated equipment which costs in the region of a small house and houses the kit is a secure outbuilding. The following images perhaps best illustrate equipment required for electron microscopy.
Fig 6: TESCAN MIRA 4 high-resolution field emission (SEM) which uses a Schottky source. Main features are:
(1) Column and specimen chamber, mounted on top of the column console
(2) Housing for turbo pump and some electronics
(3) Housing for further electronics
(4) The MIRA 4 comes with a single 32” screen. There is a second smaller screen to permit easy access to stubs database and image library.
(5) Controlled using TESCAN ESSENCE software installed on a Windows®10 PC, operated using keyboard, mouse & tracker ball.
(6) Uninterruptible power supply unit, preserving electron source if short term power failure.
Fig. 7: Further equipment- Critical Point Drier (LHS), the two gauges register temperature and pressure of the CO2 used for drying. A sputter coater (RHS) shown ready to load stubs on the rotating and tilting holder (‘Rota Coata’). In use, the glass cylinder surrounds the Rota Coata and the gold target assembly rests on top of the cylinder
The main motivation for electron imaging for Jeremy is to make pretty pictures. In its first iteration, an image generally requires the background to be edited, for example, the supporting pad of the specimen can be cracked (and unsightly) and a scale is often automatically included, similarly detracting from the main image. There are easily dealt with using Photoshop, usually within a couple of minutes. Sometimes, it can be longer- considerably longer- an image of a wasp head on one occasion required each isolated hair on a wasp head to be brought out in the correct contrast, a process which took 5 hours. Also, artificial colour can be introduced (electron microscopy does not recognise colour), for example a tardigrade was given an enhancing colour of green (see Fig. 7).
Fig. 8: Image of tardigrade LHS original image, RHS with colour introduced with Photoshop
Fig. 9: Microfossil imaged by light (LHS) and electron (RHS) microscopy.
Fig. 9 illustrates one distinct advantage of electron over light microscopy; a much better depth of field is obtained with the latter. On the LH image, only a small part of the micro-fossil is in focus; in the RH SEM image, the whole specimen is in focus. This is the reason SEM imaging is often used in preference to light microscopy even where the light microscope is well within light resolution limits. Note the light microscopy image is in colour whilst the SEM image is monochrome.
Fig. 10: Some fascinating images (of many) produced by Jeremy
Jeremy displayed a number of his images at the meeting. Two are included above (see Fig 10). As can be seen the quality is excellent, the images sublime and bring home the existence of a sophisticated and complex microworld of which we are largely unaware. Jeremy’s range of work can be better explored on his own website: http://www.jeremypoolesem.org.uk/intro.html
However, science has not been forgotten. Important, ground-breaking, research has been carried out in insect morphology, most notably in identifying the function of ostioles of instars (development stage of insects between each moult), and within the Green Bug (Polomena prasine) Project.