Jianwei (John) Miao - US grants

Visualizing the arrangement of atoms has played a crucial role in understanding the microscopic world. There are already a few ways of imaging atomic structures, but each has its limitations. Scanning probe microscopes are limited to imaging atomic structures at the surface. Transmission electron microscopes can resolve atoms but only for samples thinner than ~ 30 nm. X-ray crystallography can reveal the globally averaged 3D atomic structures based on the diffraction phenomenon, but requires crystals. These limitations can in principle be overcome by coherent x-ray diffraction microscopy (or lensless imaging) that is based upon coherent x-ray scattering in combination with a method of direct phase recovery called oversampling. Coherent x-ray diffraction microscopy has been successfully applied to 2D and 3D imaging of nanoscale materials and biological samples, and a highest spatial resolution of 7 nm has been achieved. By using 3rd generation synchrotron radiation, we propose to develop a next-generation coherent x-ray diffraction microscope for 3D imaging and characterization of nanoscale systems. The new microscope will have the following features: i) a 4K x 4K back-illuminated CCD camera will be used to improve the resolution to the 1 nm level; (ii) data acquisition will be automated; iii) 3D images will be directly reconstructed from oversampled diffraction patterns without using lower resolution images; iv) samples will be mounted in helium ambiance which can be at room or liquid nitrogen temperatures; and v) faster and more precise phase retrieval algorithms will be developed for 3D image reconstruction.


X-ray imaging has been used to probe the structure of matters for more than a century. Unlike visible light, however, x-rays are difficult to manipulate and focus. The highest resolution of x-ray images currently achievable is about 100 atomic diameters. For disordered samples x-rays cannot image at such high resolution. A promising approach, which is currently under very active development, uses x-ray and image reconstruction using a computer. This approach has been used to image materials and biological specimens with a resolution of about 14 atomic diameters . By using currently the brightest x-ray source in the nation - the Advanced Photon Source in Chicago, we propose to develop a next-generation x-ray microscope for improving the resolution to around 3 atomic diameters. The new microscope will include a larger detector and better software for quick and accurate 3D image reconstruction. We anticipate the 3D x-ray microscope will find broad applications in many areas of science and structural biology.

This award is for the development of a single-photon-sensitive confocal microscope, capable of true 4D (four dimensional: x, y, z, t) live imaging at video rate. It is based on an Image Intensified CMOS sensor (ICMOS) and a high-speed confocal scanner, which are designed to meet the following specifications: 1) High-speed (< 1 ms/frame), mega-pixel imaging with single-photon sensitivity. This is the same sensitivity as an EMCCD (Electron Multiplying CCD) but at one hundred times faster frame rate. 2) High-speed (< 1 ms/frame) confocal scanning for a single x-y focal plane. In addition, capability to scan in depth (z) up to 100 microns in 30 msec, resulting in a ''true 4D movie'' at video frame rate. 3) High-speed (< 10 ms/frame) FLIM (Fluorescence Lifetime Imaging Microscope) with < 100 psec lifetime resolution for a single x-y focal plane. This is a similar lifetime resolution as conventional scanning confocal microscopes but with a frame rate that is one hundred times faster. It will enable a ''true 4D FLIM movie'' at video frame rate. 4) Video-rate (~30 ms/frame) FLIM with true spectrum analysis for a single x-y focal plane. This new microscope may revolutionize the way millisecond time-scale phenomena are visualized in all biological systems, spanning from single molecules, single cells, and neural networks (such as the brain), to in vivo imaging of tissue in animals.

In addition to the scientific benefit of this new microscope, this award will contribute to multi-disciplinary education of students, at both the graduate and undergraduate level, at the forefront of biology, chemistry, physics, and engineering.

DESCRIPTION (provided by applicant): X-ray protein crystallography is currently the primary methodology used for determining the 3D structure of protein molecules at near-atomic or atomic resolution. However, many biological samples such as whole cells, cellular organelles and other important protein molecules are difficult or impossible to crystallize and hence their structures are not accessible by crystallography. Overcoming these limitations requires employment of different techniques and methods such as nuclear magnetic resonance and cryo-electron microscopy. A very promising approach currently under rapid development is X-ray diffraction microscopy (i.e. X-ray crystallography without crystals) in which the X-ray diffraction pattern of a non-crystalline specimen is measured and then directly phased by an iterative algorithm. Since its first experimental demonstration in 1999, X-ray diffraction microscopy has been successfully applied to 2D and 3D imaging of non-crystalline specimens such as whole cells. The highest resolution achieved thus far is 7 nm, while the ultimate resolution is limited by the X-ray wavelengths and radiation damage to the specimens. Due to its applicability to thick specimens and its high-resolution imaging capability, X-ray diffraction microscopy can be used to bridge the gap between light and electron microscopy and provide quantitative 3D structural information of whole cells at 10 nm resolution. During the next 4 years of this project, we propose to 1) minimize the mean phase error (i.e. the image ambiguity) in phasing the 3D X-ray diffraction patterns of whole cells;2) experimentally study the ultimate 3D resolution attainable from frozen-hydrated whole cells affected by radiation damage;3) test the hypothesis that X-ray diffraction microscopy can image the 3D intracellular structure of frozen-hydrated cells at a resolution of 10 nm;and 4) localize specific multiprotein complexes in Caulobacter and yeast. We are confident that our multidisciplinary team with expertise in physics, biology and instrumentation will help to more rapidly validate this novel imaging technique for quantitative 3D imaging of the intracellular structure at 10 nm resolution and the localization of multiprotein complexes inside whole cells. Cryo X-ray diffraction microscopy will be demonstrated to quantitatively image the 3D intracellular structure of frozen-hydrated yeast cells to a resolution of 10 nm. This novel imaging technique will be applied to the localization of multiprotein complexes inside whole cells.

DMREF: Collaborative Research: Design and Testing of Nanoalloy Catalysts in 3D Atomic Resolution

Non-Technical Description: The project aims at the discovery of ultra-small, nanometer-sized alloy catalysts to improve the efficiency of fuel cells, mobile power generation, and automotive catalytic converters. State-of-the-art laboratory and computer simulation techniques will be engaged to explore the uncharted design space of bimetallic nanostructures for such applications and implement reductions in cost through partial replacement of precious metals such as platinum by cheaper alternatives. The team of three PIs will synthesize new nanocatalysts using biomimetic approaches, image the positions of all atoms in 3D resolution using the world's most powerful electron microscope, and carry out performance tests in fuel cells in a close feedback loop with predictions by multi-scale modeling and simulation. Fundamental understanding of synthesis controls, atomic-scale order, and associated reactivity of the nanoalloys will lead to rational design rules to optimize catalyst performance and enable targeted improvements of promising materials. The development and validation of predictive multi-scale simulation tools will also benefit the broader computational user community. New fundamental insight into alloy synthesis and reactivity controls has further potential benefits to improve catalysts for commodity chemicals, magnetic information storage, batteries, sensors, and nanoelectronic devices. Undergraduate students, high school students, and teachers will be engaged in summer research activities at UCLA and in annual Engineering Career Days at the University of Akron to encourage careers in science and engineering.

Technical Description: The project focuses on the computationally driven, rational optimization of nanoalloy atomic composition and shape for catalytic performance in the Oxygen Reduction Reaction (ORR) in fuel cells. Specific aims include the deterministic synthesis of Pt-M Nanocrystals (M = Fe, Co, Ni, Cu, Cr, Mn), the three-dimensional characterization of nanoalloy catalysts in atomic resolution and model refinements, as well as the prediction, tests, and optimization of the reactivity in the ORR. Methods comprise biomimetic synthesis protocols coupled with molecular dynamics and kinetic Monte Carlo simulations, ORR performance testing by voltammetry and density functional theory calculations, and in-situ monitoring of all reactions. The coordinates of the atoms in the synthesized nanostructures will be monitored by electron tomography to identify atomic ordering, to validate and improve interatomic potentials, and to predict reaction rates in ORR. Detailed understanding of alloy growth mechanism, shape control, and catalytic performance through new polarizable and reactive force fields for alloys and their aqueous interfaces from first principles will close a wide gap between experimental capabilities and missing theoretical understanding. Aided by thorough experimental characterization, predictions with unprecedented accuracy at length scales of 1 to 100 nm appear feasible, far beyond the limits of quantum-mechanical methods and building on previous successful models for interfaces of pure metals (CHARMM-METAL).