EM: Not Just Another Gadget

Joanne Zhang | joanne.zhang@yale.edu May 1, 2010

With its prices running from $250,000 to $1 million, the electron microscope (EM) is not something purchased on a whim. The EM’s price is justified by its wide range of uses, which include cellular biology, forensic analysis, and material science.

Two German engineers, Ernst Ruska and Max Knoll, built the first transmission EM in 1931. Now, nearly eighty years later, the EM is a relatively routine tool in basic research.

While the overall mechanical structure has remained constant, the EM has differentiated into multiple types of microscopes. Currently, there are four major types: transmission (TEM), scanning (SEM), reflective (REM), and scanning transmission (STEM). All focus a high-energy electron beam to produce images.

TEMs image ultra-thin specimens (usually 60-90 nm thick) by passing electrons through the sample. These electrons can then convey the information about the appearance of the specimen to an image projector, such as a computer camera software.

The electron beam of the SEM produces an image in a quite different way. Here, the electrons interacting with the surface of the specimen change in intensity and energy, and this difference provides the SEM with signals that generate an image of the sample’s surface.

In contrast, REM and STEM are directed towards more specific types of samples and imaging styles. REM, used primarily for various surface and interface analyses, interprets results from reflected beams of scattered electrons. STEM, a hybrid of TEM and SEM, is often used as an analytic technique such as spectroscopy (energy dispersive X-ray and electron energy loss) and annular dark-field imaging (ADF produces high contrast images without staining). Such techniques offer both images and quantitative data.

Recent advancements have centered largely on improving image resolution. Good resolution was a difficult goal before computer technology, especially in the TEM. Barry Piekos, Research Associate and Lecturer in the Molecular, Cellular, and Developmental Biology department, explains, “Until there was computer hardware that could correct for chromatic and spherical aberrations, the resolution for a TEM was limited to 1 angstrom. But once these aberrations were fixed, the resolution has approached the theoretical limit below 1 Angstrom.”

Spherical aberrations are optical effects resulting from the refraction of light rays when they pass through a lens. While a perfect lens would focus all incoming rays to a single point, real lenses focus rays differently depending on their origin. Chromatic aberrations result when a lens cannot focus all colors of an image to the same point due to different colors’ different refractive indices. Previously, these aberrations have prevented a clear focus on the object of interest. Today, however, resolution for the TEM ranges from between 0.5 to 1 Angstrom thanks to computers that approximate and fix the aberrations.

Since 2004 the Transmission Electron Aberration-corrected Microscope (TEAM) has endeavored to design a TEM with a resolution below 0.05 nm, about half the size of a hydrogen atom. Last year, TEAM reached the 0.05 nm target resolution and has been able to show carbon atoms breaking and forming bonds with each other in a graphene crystal. Such TEM capabilities aid in understanding molecular structures.

Despite its numerous applications, the EM has its disadvantages. In addition to the hefty price tag, the microscope is expensive to maintain. Stable currents and high voltages are needed to produce the electron beam, yet the beam can only work under a continuously pumped or ultra-high vacuum system that prevents gas atoms from scattering the electron and ruining the beam. Additionally, a constant circulation of cool water is needed to prevent the machine from overheating.

Still, the instrument’s flaws do not detract from the vast amounts of information gained from its use. Applicable to a wide range of research fields, as time passes, the EM will likely become a staple in the research laboratory.