Coherent Diffraction Microscopy and Its Applications in Nanoscience and Biology

The discovery of X-ray diffraction from crystals by van Laue, William Bragg, and Lawrence Bragg nearly a century ago opened up a new era for visualizing the arrangement of atoms in three dimensions. Indeed, X-ray crystallography has since made revolutionary impacts in physics, chemistry, materials sciences, biology and medicine, and a number of Nobel prizes has been awarded to this field. It has now reached a point at which it can determine almost any structure, as long as good-quality crystals are obtained. Many samples, however, cannot be accessed by this approach, such as amorphous and disordered materials including glasses, polymers, strains and defects in crystals, quantum dots and wires, dislocation and deformation structures, and some inorganic nanostructures. In biology, structures such as whole cells, organelles, some viruses, and many important protein molecules cannot or are difficult to crystallize; hence their structures are not accessible by X-ray crystallography. Overcoming these limitations requires the employment of different techniques and methods. One very promising approach currently under rapid development is coherent diffraction microscopy (Fig. 1), also called coherent diffraction imaging or lensless imaging, which was first demonstrated in 1999 (1). Since then, approximately 30 groups worldwide have applied this technique to imaging a variety of samples, ranging from nanoparticles, nanocrystals, biomaterials and whole cells to carbon nanotubes by using X-rays, electrons and tabletop lasers. Numerous national and international workshops and conferences have been organized to discuss the current progress and the future potential of this burgeoning field. Below we present a few recent research highlights to illustrate our activities in this fast-growing field.

Fig. 1 Schematic layout of coherent diffraction microscopy. The oversampled diffraction intensities are measured from a finite specimens, and then directly phased to obtain a high-resolution image.

i). We developed a general method to solve the missing data problem (i.e. the intensities at the center of diffraction patterns can’t be experimentally measured), which was the bottleneck for wider applications of coherent diffraction microscopy. This work in principle cleared the way for single-shot imaging experiments by using X-ray free electron lasers (2).

ii). We applied coherent X-ray diffraction microscopy to 3D imaging of materials science samples at the nanoscale resolution, and for the first time revealed the internal GaNGa2O3 core shell structure in three dimensions (Fig. 2). This work opens the door for non-destructive and quantitative imaging of 3D morphology and 3D internal structure of a wide range of materials at the nanometer scale resolution that are amorphous or possess only short-range atomic organization (3).

Fig. 2 Iso-surface renderings of the reconstructed image from a GaN quantum dot nanoparticle, showing (a) the front view, (b) the back and (c) the side view. (d) 3D internal structures of the nanoparticle.

iii). We developed a novel oversampling scheme to distinctively improve the quality of phase retrieval. This work will contribute to high-quality image reconstruction of materials science samples and biological structures using X-ray diffraction microscopy (4). We also developed an iterative algorithm that combines the concept of optimization with the traditional hybrid input-output algorithm for phase retrieval of oversampled diffraction intensities (5).

iv). We carried out the first experimental demonstration of resonant X-ray diffraction microscopy for element specific imaging of buried structures with a pixel resolution of ~ 15 nm by exploiting the abrupt change in the scattering cross-section near electronic resonances. We performed nondestructive and quantitative imaging of buried Bi structures inside a Si crystal by directly phasing coherent X-ray diffraction patterns acquired below and above the Bi M5 edge (Fig. 3). Resonant X-ray diffraction microscopy can in principle be applied to element and chemical state specific imaging of a broad range of systems including magnetic materials, semiconductors, organic materials, biominerals and biological specimens (6).

Fig. 3 Elemental mapping of Bi structure showing (a) reconstructed image at E=2.550 keV and

(b) E=2.595 keV, respectively. (c) Distribution of the Bi structure obtained by taking the difference of the two images, which represents a 2D projection of the 3D Bi distribution and therefore does not contain the depth information of the Bi distribution in the sample. (d) SEM image of the same sample.

v). We applied coherent X-ray diffraction microscopy to nondestructive imaging of mineral crystals inside biological composite materials -intramuscular fish bone -at the nanometer scale resolution. We identified mineral crystals in collagen fibrils at different stages of mineralization. Based on the experimental results and biomineralization analyses, we proposed a dynamic mechanism to account for the nucleation and growth of mineral crystals in the collagen matrix (Fig. 4). The results obtained from this study not only further our understanding of the complex structure of bone, but also demonstrate that coherent X-ray diffraction microscopy will become a useful tool to study biological materials (7).

Fig. 4 Schematic illustration of the suggested dynamic model for the nucleation, growth and orientation of mineral crystals in the collagen matrix at different stages of mineralization. The model was proposed based upon the results obtained by using X-ray diffraction microscopy.

vi). We recorded and reconstructed coherent X-ray diffraction patterns from single, unstained viruses, for the first time. By separating the diffraction pattern of the virus particles from that of their surroundings, we performed quantitative and high-contrast imaging of a single virion (Fig. 5). The structure of the viral capsid inside a virion was visualized. This work opens the door for quantitative X-ray imaging of a broad range of specimens from protein machineries and viruses to cellular organelles. Moreover, our experiment is directly transferable to the use of X-ray free electron lasers, and represents an experimental milestone towards the X-ray imaging of single macromolecules (8).

Fig. 5 (a) X-ray diffraction imaging of single herpesvirus virions at a resolution of ~20 nm. (a) X-ray diffraction pattern obtained from a single, unstained virion. (b) High-contrast image reconstructed from (a) where the background and the surroundings of the virion were completely removed. (c) SEM image of the same virion. (d) Negative stain TEM image of a similar herpesvirus virion.

vii). In collaboration with Murnane and Kapteynʼs group at the University of Colorado, Boulder, we performed the first experimental demonstration of soft X-ray diffraction microscopy with a resolution as high as 70 nm using table-top coherent soft x-ray sources (Fig. 6). We have also implemented field curvature correction to the diffraction patterns, which allows high numerical aperture imaging and near-diffraction-limited resolution of 1.5λ. Due to its simple optical design, high resolution, large depth of field, 3D imaging capability, scalability to shorter wavelengths and ultrafast temporal resolution, tabletop soft x-ray diffraction microscopy will find broad applications in biology, nanoscience, and materials science (9,10).

Fig. 6 Lensless imaging using coherent soft x-ray laser beams at 47 nm. (a) SEM image of a waving stick figure sample (scale bar = 1 micron). (b) Coherent soft x-ray diffraction pattern after curvature correction. (c) Reconstructed image with curvature correction. (d) Line-out of the image along the legs (shown in the inset), verifying a resolution of 71 nm.

Due to the broad applications of coherent diffraction imaging, our work has been featured in Nature, Science, Physics Today, Discovery Channel, Nature Biotechnology, Nature Nanotechnology, Nature Photonics, Nature Science Update, Nature Physics Portal, Physical Review Focus, Proc. Natl. Acad. Sci. USA and other media. Finally, X-ray free electron lasers are currently under rapid development in the U.S., Europe, Japan and other nations with the construction cost of hundreds of millions of dollars apiece (Fig. 7). Theoretical studies have shown that, by using extremely intense and ultrashort X-FEL pulses, one can record a diffraction pattern from a single biomolecule before it is destroyed (11). The combination of coherent diffraction imaging and XFEL may thus open a new horizon of imaging single large protein molecules without the need of crystallization (12). This explains why coherent diffraction imaging is a major justification for the construction of these large-scaled facilities. As Prof. Miao is a coteam leader of the coherent imaging experiment of the Linac Coherent Light Source (LCLS) at SLAC, we will be actively involved in 3D imaging reconstruction of noncrystalline specimens (both materials and large biomolecules) at the near atomic resolution by using the LCLS.

Fig. 7 The Linac Coherent Light Source (LCLS) at SLAC.