OTHER DESIGNATIONS: Soft x-ray imaging, photoelectron emission microscopy (PEEM), scanning transmission x-ray microscopy (STXM), full-field microscopy, x-ray diffraction imaging (XDI), x-ray tomography, computer-aided tomography (CAT) scans.

PURPOSE: The wavelengths of soft x-ray photons (1–15 nm) are very well matched to the creation of nanoscopes capable of probing the interior structure of biological cells and inorganic mesoscopic systems. Problems addressed by soft x-ray imaging techniques include:

• Cell biology
• Nanomagnetism
• Environmental science
• Soft matter, polymers

HOW THE TECHNIQUE WORKS: The fine spatial resolution needed in soft x-ray microscopy can be attained by use of photon optics or electron optics.

1. Photon optics. Fresnel zone plates perform the same function for x-rays that lenses do for visible light. With STXM, the sample is scanned through the illuminated spot to build up an image. With the use of two zone plates (a condenser and an objective), it is possible to create a full-field image.
2. Electron optics. In PEEM, a smallish spot on the sample is illuminated and the emitted photoelectrons are passed through an electron microscope column to produce a magnified full-field image.

UNIQUENESS: The tunability of synchrotron radiation is absolutely essential for the creation of contrast mechanisms. Cell biology CAT scans are performed in the “water window” (300–500 eV). Nanomagnetism studies require the energy range characteristic of iron, cobalt, and nickel (600–900 eV).

EXAMPLES:

CAT Scans of Single Cells Show Details Invisible to Light Microscopy
Nanomagnetism Dynamics
Imaging Without a Lens

Tomographic reconstruction of
Saccharomyces cerevisiae (yeast).

X-ray tomography is the first high-throughput imaging technology that generates images of whole, hydrated cells at better than 60-nanaometer resolution. With it, researchers have obtained three-dimensional views of the internal structure of whole, hydrated Saccharomyces cerevisiae (yeast) cells, bridging the mesoscale resolution “gap” — the middle area between light (200 nm) and electron microscopy (3 Å). With the ALS transmission x-ray microscope, data collection is fast (under 3 minutes) and relatively easy (like light microscopy), producing high-resolution, absorption-based images (like electron microscopy) that provide contrast between cellular structures and allow for discernment of individual structures. After data collection, tomographic techniques are used to reconstruct the original information into quantifiable three-dimensional views of the entire cell. Through the use of computer algorithms, the researchers then process the reconstructed data to create made-to-order images of whole cells and their internal structures.

C.A. Larabell and M.A. Le Gros, “X-ray tomography generates 3-D reconstructions of the yeast, Saccharomyces cerevisiae, at 60-nm resolution,”Molecular Biology of the Cell 15, 957 (2004).

Time-resolved PEEM images of square and rectangular vortex patterns.

Any desktop or laptop computer user knows how fast the storage capacity of hard disks is growing. To maintain this pace, we require new tools to study magnetic materials in smaller areas and over shorter times. Researchers have used a new time-resolved x-ray photoemission imaging technique to resolve the motion of magnetic vortices, peculiar magnetic structures that appear in micron-size magnetic patterns in response to an excitation field pulse. They are of considerable technological interest because a low stray magnetic field leads to a magnetic stability and minimizes the cross-talk between adjacent vortices—two prerequisites for high storage densities. For the microscopic study of such ultrafast magnetization dynamics, researchers developed a novel technique based on 70-picosecond-long synchrotron x-ray pulses that can be used like light flashes from a strobe to freeze the dynamics and acquire a snapshot of the motion. Analysis of the observed gyrating trajectory of the core on such short time scales suggests that the precession is induced by a left-right handedness, or chirality, in the magnetization pattern, thereby demonstrating that handedness plays an important role in the dynamics of microscopic magnets.

S.-B. Choe, Y. Acremann, A. Scholl, A. Bauer, A. Doran, J. Stöhr, and H.A. Padmore, “Vortex-driven magnetization dynamics,” Science 304, 420 (2004).

Image reconstruction of two clusters of gold balls showing convergence of both the reconstructed image (on the left of each panel) and its boundary (on the right of each panel) as the number of iterations increases from 1 to 1000.

For samples consisting of a very large number of identical objects in a regular array (e.g., atoms in a crystal), it is mathematically possible to construct an image of the object from its diffraction pattern. Researchers are now taking image reconstruction a big step further by studying nonperiodic samples using x-ray diffraction imaging (XDI), or “lensless imaging.” Like crystallography, XDI is based on the analysis of diffraction patterns, but it uses iterative algorithms to extract the phase information needed to reconstruct the object and requires that the diffraction intensity be zero outside the object’s boundary. The better this boundary is known, the faster the iterations converge to an accurate image. Most researchers have relied on x-ray microscopy or other techniques to supply this information. Now, researchers have done away with this requirement through the use of a new "shrink-wrap" algorithm, which uses a transform of the diffraction pattern itself to provide initial information about the boundary. The technique has demonstrated the ability to make two-dimensional images of clusters of gold balls 50 nm in diameter. The researchers anticipate that a three-dimensional resolution of 10 nm will be possible for life-science samples, where radiation damage is an issue, and 2 nm for solids.

S. Marchesini, H. He, H.N. Chapman, S.P. Hau-Riege, A. Noy, M.R. Howells, U. Weierstall, and J.C.H. Spence, “X-ray image reconstruction from a diffraction pattern alone,” Phys. Rev. B 68, 140101(R) (2003).