Industry needs have driven SEMs to become integrated, systems-based solutions that provide robust, automated tools for application-specific capabilities.

The scanning electron microscope (SEM) has long been a powerful research tool that permits a human operator to explore the morphology and composition of material over a wide range of microscopic scales. One area in which SEM use has been growing steadily is industry, where the advent of new manufacturing processes and close tolerance designs has only intensified emphasis on quality control and improved evaluation methods. 

Contaminants and defects in industrial manufacturing present a tremendous process and quality control problem. The speed, accuracy and certainty with which these problems can be detected, identified, characterized and eliminated—or adjusted, based on the specific industry’s needs—have a direct and enormous impact on profitability.

For instance, what would it take to analyze more than 30,000 particles/hr, 24 hr/day, 7 days/wk using an SEM installed in the melt-shop of a steel mill? At a mixing station in a pharmaceutical processing plant? On an automotive assembly line? Rather than just providing high-quality imaging and characterization in controlled environments—i.e. laboratory settings—operators at the industrial level not only require higher throughput and have less time—as well as less trained personnel—they also have to mine and make sense of the data acquired. Therein lies the problem.

Solution-driven approach

Traditionally, quality assurance departments have invested in various types of equipment for inspecting materials, ranging from optical microscopes to high-end SEMs. Optical microscopes offer a low-cost, effective and traditionally robust option, but are limited in their magnification and resolution capabilities. High-end SEMs offer excellent imaging and resolution, while also allowing identification of materials based on their characteristic emissions visualized through energy dispersive X-ray spectroscopy (EDX). These, however, can be expensive, and the addition of the EDX functionality is almost always an add-on technology requiring additional hardware and software. 

To meet the changing needs of industry, manufacturers of microscopy solutions began to consider ways of combining the high-end features of an electron microscope with the affordability of an optical microscope. In addition, the expectation and need for large quantities of quality data that can be provided quickly and consistently, and easily read and interpreted by a variety of personnel each with varying levels of expertise and sophistication, has spurred a solution-driven approach to SEM performance. This new approach has fostered the development of a new, integrated “system” design in SEM technology.

The system approach to microscopy implies that all aspects of the instrument are optimized for efficient performance. Whereas instruments designed for research applications are typically characterized by technical performance specifications, such as attainable resolution, a solution-driven tool is most meaningfully characterized in terms of its ability to assess key specimen metrics with the necessary speed and precision. For true optimization, all components of the system—the SEM, the EDX detector and the software—must function together as an integrated whole. When it does, it allows bulk analysis for contaminant and defect detection, identification, distribution and characterization of features both rapidly and automatically.

Evolution of EDX

The process of integrating the EDX detector has been a decades-long journey of continued improvements, many of which were made only in the last decade or so. The conventional construction of an EDX detector can be described as a “sensor on a stick,” where the “stick” is a rod that conducts heat from the sensor located inside the vacuum chamber of the SEM to a heat sink located external to the specimen chamber. This construction, first devised for the older lithium-drifted silicon, or Si(Li) detectors, was the standard technology for the first three decades of the EDX detector. This type of sensor requires cryogenic operating temperatures and an external reservoir of liquid nitrogen (LN) for cooling. 

In the last 10 years, however, the technology of the EDX detector has changed radically as the Si(Li) sensor has given way to the silicon drift detector (SDD) sensor, which no longer requires LN cooling and sustains excellent energy resolution at much higher counting rates. To translate this capability into higher throughput, however, more X-rays must be delivered to the detector with higher beam currents, but that also means larger beam diameter, which eventually compromises spatial resolution. Thus, increased emphasis now is directed toward increasing the collection efficiency of the detector itself. Two viable strategies are the use of larger detectors and multiple detectors, and though these are effective approaches, they also have a substantial cost impact, as well as other practical limitations.

The most cost-effective means of improving detection efficiency is to place the X-ray sensor as close as possible to the point where the X-rays originate from the specimen. In practice, however, such optimization has traditionally been circumscribed by the conventional construction of the C-ray detector, in which the active sensor element is located at the end of a long tube that is inserted through a port in the microscope chamber. Though this modular “tube mount” or “sensor on a stick” configuration has served the industry well for decades, it has also restricted how closely and flexibly the sensor element can be positioned relative to the specimen.  

This limitation has recently been addressed by the introduction of OmegaMax technology, in which the SDD sensor is built directly into the structure of the microscope. This has permitted the effective solid angle (omega) of the detector to be increased by several times over what had been achieved with conventional tube-mount detectors. In fact, the improvement in detection efficiency is so dramatic that smaller-area sensors can be employed that still deliver substantially enhanced detection efficiency, but with the additional benefit of improved energy resolution. This dramatic advance was made possible by considering the X-ray detector not as a separate entity, but as an integrated component of a system whose objective is to provide practical industrial solutions.

Although the OmegaMax technology supports the use of larger sensors (up to 30 mm2) either individually or in arrays of up to four sensors, one of the practical benefits is that it permits the use of smaller detectors with smaller electron traps, packaged in smaller cases that can be placed closer to the specimen without image distortion and with good resolution at higher counting rates.   

Performance gains

By integrating sensor modules directly into the SEM, the system is able to provide more than four times the solid angle using the same sensor size. The sizable increase in solid angle afforded by the design makes it possible to employ a smaller premium-performance sensor, while still maintaining more than a two-fold increase in detection efficiency.  

When improved counting efficiency and energy resolution are combined, the result is a significant reduction in measurement time. The three diagrams above were collected with a conventional EDX detector using X-ray collection times of 0.2, 0.5 and 1.0 sec/inclusion, from left to right. The distribution “tightens” as the counting time is increased, indicating improved precision in the individual characterizations. The ternary diagram in Figure B was acquired with the OmegaMax detector at 0.2 sec/inclusion with otherwise identical conditions.  

The OmegaMax distribution is at least as good as that acquired at 1.0 sec with the conventional detector; thus, the same data is available at an analysis speed that is five times faster. These performance gains can be further multiplied up to a projected 25x by employing up to four sensors of larger available sizes, which results in up to a 100-fold increase in analysis speed over conventional EDX. This ability to perform quality X-ray analysis at ms/spectrum rates opens a world of new possibilities in time-critical applications.  

Scanning electron microscopes have evolved in the last two decades to produce robust automated tools that support industrial applications. Industry needs have driven manufacturers of these machines to look beyond the mere manufacture of metal boxes to create integrated, systems-based solutions in which the SEM is integrated with complimentary hardware (EDX), allowing application-specific resolution, speed and reporting capabilities. These innovations have ushered in a new era of microscopy and microanalysis that will continue to push the limits of applications in numerous industries for years to come.