A High-Throughput 3-D X-ray Microtomography System with Real-Time 3-D Reconstruction

A high-throughput x-ray microtomography system (XMS) that can acquire, reconstruct, and interactively display rendered 3-D images of a sample at micrometer-scale resolution within minutes has been developed at Advanced Photon Source (APS) beamline 2-BM, which is managed by the Synchrotron Radiation Instrumentation Collaborative Access Team (SRI-CAT). This system could bring better understanding of an array of scientific and technological problems, ranging from failure in microelectronic devices to structures in biological samples.

A rendered image from a three-dimensional tomographic reconstruction of a cricket, obtained with the x-ray microtomography system. The image has been digitally cut at different planes to show the internal structures of the cricket's head.  
Left: A rendered image from a three-dimensional tomographic reconstruction of a cricket, obtained with the x-ray microtomography system. The image has been digitally cut at different planes to show the internal structures of the cricket's head.

The XMS uses a unique combination of the high-brilliance x-rays from the APS; a fast, high-resolution x-ray detector system; a high-speed data network; a parallel computer cluster; and a parallel processing software for reconstruction and display. This combination of advanced technologies gives a user the ability to quickly study the full 3-D microstructure of a sample, to study in real time the evolution of a dynamically varying sample, or to study a large number of samples with unprecedented speed and imaging resolution. The XMS can process 3-D images of millimeter-sized samples with micrometer-scale resolution on time scales of minutes.

With this system, a complete tomographic data set can be collected, reconstructed, and displayed in as little as 3 minutes. Immediately after each tomographic projection is received by a local desktop computer, it is transferred via a high-speed network to a dedicated cluster-based, parallel computing system based on the Message Passing Interface (MPI) standard, where the reconstruction calculation is performed concurrently with the data acquisition. Images of the reconstructed data are available for viewing by the user just minutes after all the data have been received.

Because of its unprecedented data-acquisition and processing speed, the XMS has the unique capability to study the evolution of dynamically evolving samples in three dimensions at the micrometer scale in real time. Because of the high throughput of the system, it is also well suited to rapid 3-D inspection of a large number of samples. Key potential applications that take advantage of the unprecedented acquisition rate and reconstruction speed include:

  • Failure mechanisms in microelectronic devices due to electromigration, thermal breakdown, nonuniformities, and radiation
  • Material transport in porous media, such as soils, rock, and across insulation barriers
  • Material failure in metallurgical and ceramic samples, such as that due to effects of induced strain, crack propagation, and corrosion on simple and composite materials and those with bonded and welded interfaces
  • Dynamic processes in composite and complex materials and fluids, such as curing of adhesives, cements, coatings, and paint
  • Growth and depletion processes in biological samples, such as tissue and bone sections, including healing, or collateral damage following injury or microsurgery, and environmental effect on growth
  • Study of transgenic mutation expression and correlation in large samples of rapidly maturing organisms, such as fruit flies

The x-ray microtomography system is based on an earlier approach first demonstrated jointly by Argonne's Advanced Photon Source, Experimental Facilities Division; Argonne's Mathematics and Computer Sciences Division; and the University of Southern California, Information Sciences Institute. Research funding was provided by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under contract W-31-109-ENG-38; and the Office of Advanced Scientific Computing Research, Mathematical, Information, and Computational Sciences Division.

F. De Carlo,1 D. C. Mancini,1 B. Teiman,1 Y. Wang,2 I. McNulty1

1 Experimental Facilities Division, Argonne National Laboratory; 2 X-radia Corporation

See a related article at the ANL Technology Transfer Web site describing the earlier approach.