Going industrial

Scanning electron microscopes have made an almost immeasurable impact in research, but what of their industrial uses? Tom Ray charts the rise of industrial SEM and finds that the market researchers got it very wrong when they predicted only six would sell 

FIFTY years have passed since the scanning electron microscope (SEM) was first put to use in a routine industrial application.1 The so-called SEM 3, which was also the first magnetically focussed and fully engineered SEM, completed in 1958, was commissioned by the Canadian Pulp and Paper Research Institute to be used for examining wood fibres in their Ottawa laboratories.

It seemed at the time that very few people believed there was much potential for the SEM to be used in other industrial applications.2 Indeed, one of the pioneers of scanning electron microscopy, Professor Charles Oatley, apparently used to relate that when the SEM was being considered as a commercial product a group of marketing experts were sent out to make an evaluation of the number of SEMs that might be sold. Their conclusion was that between six and ten instruments would saturate the market.3 To be fair, at that time, the SEM was only capable of 20nm resolution, whereas the available transmission electron microscopes (TEM) were producing resolutions below 5nm. The aforementioned market research seemed to have been directed at existing users of a TEM for whom the SEM had little appeal. However, it is clear that very few people were able to imagine the enormous value that scanning electron microscopy might have to research and industry in the future.

The rest, as they say, is history, and there are now thousands of SEMs used in commercial applications across a wide range of industries. Alongside improvements in the resolution of these instruments, there have been other developments that have widened the scope and analytical potential for SEMs. Most significant amongst these have been the addition of stable cryo-stages, which facilitate the examination of wet samples in a frozen state. The more recent introduction of low pressure SEMs - which allow non-conducting samples to be viewed without the need for coating - has also expanded the range and ease with which items can be imaged. In addition, X-ray detectors, which permit elemental analysis of the surface of the sample, make the SEM a powerful analytical tool as well as a powerful imaging tool. Whereas the early SEMs were restricted in use to dry samples that were either already electrical conductors, or that could be coated with a thin conducting layer, this is no longer the case. These days it is possible routinely to examine complex items such as food, pharmaceuticals, paints, textiles, cosmetics, soil samples, and biological specimens and to learn about their elemental composition as well as their micro-structure.

Despite the fact that SEMs have become much more powerful, and much easier to use, there are still challenges involved in sample preparation and optimising the set up of the microscope to ensure sample integrity and image quality. Equally, limitations remain to the SEM’s usefulness as a tool for elemental analysis, and complementary techniques such as X-ray microfluorescence, FT-IR spectroscopy and Raman spectroscopy are increasingly being used alongside the SEM to answer questions that the SEM cannot answer easily, if at all.

As a contract laboratory serving clients spanning every industry from food production to oil surveying, RSSL is routinely confronted with research projects and one-off requests involving a wide range of samples. These may be well-defined samples, such as ice cream shown in Figure 1, and marshmallows shown in Figure 2, or completely unknown, as with the contaminants that are investigated and identified by Reading Scientific Services’ foreign body identification service.

In the case of research projects involving the known samples, it is often the case that structural detail is required to understand physical properties of the sample. In others, it is the elemental distribution that is of interest. What characterises the vast majority of these samples is that some of them would have been more difficult to image by SEM ten years ago, at least to the level of detail and confidence that it is now possible to achieve routinely.

Understanding the microstructure of a product is valuable in process and product development, since microstructure determines many aspects of product performance. It is especially important in products such as food, pharmaceuticals, cosmetics and healthcare because the product needs to perform in a practical sense (e.g. the salad cream must pour, and the toothpaste must squeeze out). It also needs to perform in a sensory capacity, delivering the taste, mouthfeel and/or skin feel that consumers expect. All of these characteristics depend to some degree on microstructure, particle size and particle distribution.

Whereas particle sizing might be thought of as a job for a dedicated particle sizing instrument there are occasions when this gives an incomplete picture. Particle sizers treat particles as spheres, when clearly that is not always the case. The SEM is routinely used on pharmaceutical and other powders to clarify not only the size but also the shape of particles.

The distribution and size of droplets in a liquid phase are just as important as particle size in the solid phase. In the case of a product containing high amounts of water, or one that is frozen already (such as the ice cream mentioned above), the microstructure is best observed in a SEM where the sample has been frozen. Freezing must be done rapidly to preserve the structure and to minimise damage from ice formation. After freezing, the specimen is placed onto a cold stage where it can be fractured to expose the internal surfaces. Etching of the sample is the next process, in which water is sublimated from the surface of the sample exposing the underlying structural features. Sublimation of ice under the microscope vacuum starts at –1000C, but care must be taken to use a temperature appropriate to the sample. Any oils and fats in the sample may melt at lower temperatures and any melting will distort the ultimate image.
In the investigation of unknown samples, such as contaminants found in food or pharmaceuticals, microscopy is frequently applied to get a first look at the contaminant. It may also be appropriate to examine packaging in more detail, and one of RSSL’s investigations involved close inspection of tiny holes in pharmaceutical vials which were found to be empty after shipping to the Far East. In this case, microscopy revealed the distinctive jaw marks of an ant, with the appropriate name of Monomorium destructor, which had gnawed its way through the plastic.

More commonly, RSSL’s Foreign Body Identification service investigates fragments of plastic, glass and metal (as well as animal body parts on occasions) using the full range of SEM capabilities to characterise and identify the contaminant.

In the case of very small fragments, elemental analysis may be useful simply to define the item, for example, as being stainless steel as oppose to nickel, but the specific composition is likely to be much more revealing. Comparing the elemental composition of a glass fragment, for example, against a reference source can usually reveal whether it has come from an item that might be found in the factory, or possibly, the home of the consumer who has reported the contamination incident. 
Similarly, an SEM image of an insect body part may be all that is required by an entomologist to compare against a reference sample and conclude whether the likely origin of infestation is the country where an ingredient was grown, or the country where the end product was processed and packaged.

Whereas the X-ray microanalysis facility of the SEM is very useful for examining small samples, RSSL is one of the few laboratories that has the capability of analysing larger samples using the technique of X-ray microfluorescence spectrometry - XR(M)F. Unlike conventional XRF, which requires relatively large surface areas to analyse, XR(M)F analyses regions of between 50µm and 200µm in diameter, which can be precisely selected and positioned for analysis. The technique is non destructive and permits fast, simultaneous, multi-element detection for all elements from sodium to uranium. It also permits the gathering of qualitative, quantitative and spatial distribution information, not just from the surface of a sample but also from a little beneath the surface.

In addition to assisting RSSL in its foreign body identification work, XR(M)F also permits analysis of paint chips, ink, metal fatigue, and electronic circuitry and components. Recent applications at RSSL have included investigating the effectiveness of different cleaning solutions for removing different stains from a variety of surfaces, and investigations of counterfeit packaging. In the latter case, the technique can rapidly expose differences in the elemental composition of inks used on packaging, and provide conclusive evidence of counterfeiting. The currency imaged in Figure 3 shows how the elemental distribution can be mapped on a relatively large surface area.

It is doubtful whether the pioneers of the SEM could have foreseen the range of applications and samples to which the technique would ultimately be applied. After all, it is still the case today that many businesses are unaware of the potential for using the SEM and XR(M)F to investigate product problems or to better understand how their products perform.

Yet the SEM now has the ability to image a wide range of samples, and those referred to above are just a selection of those that RSSL routinely examines. There are many more applications that currently exist in other specialist laboratories, and still more that might yet emerge over the next fifty years.

References
1.presentation at the 51st Annual Meeting of the Microscopy Society of America, Cincinnati, August 1993. D. McMullan Cavendish Laboratory, University of Cambridge, UK. Correspondence to: Dr D. McMullan, MP Group, Cavendish Laboratory, Madingley Road, Cambridge CB3 0HE, U.K.

2. Atack D, Smith KCA: The scanning electron microscope - A new tool in
fibre technology. Pulp Pap Mag Can 1, 245-251 (1956)

3. The Early History and Development of The Scanning Electron Microscope, Bernie C Breton CUED

By Tom Ray. Tom is a senior scientist at RSSL and has worked in the microscopy lab for just over 7 years.