Use of X-ray tomography to map crystalline and amorphous phases in frozen biomaterials

Ann Biomed Eng. 2007 Feb;35(2):292-304. doi: 10.1007/s10439-006-9176-7. Epub 2006 Nov 29.


The outcome of both cryopreservation and cryosurgical freezing applications is influenced by the concentration and type of the cryoprotective agent (CPA) or the cryodestructive agent (i.e., the chemical adjuvants referred to here as CDA) added prior to freezing. It also depends on the amount and type of crystalline, amorphous and/or eutectic phases formed during freezing which can differentially affect viability. This work describes the use of X-ray computer tomography (CT) for non-invasive, indirect determination of the phase, solute concentration and temperature within biomaterials (CPA, CDA loaded solutions and tissues) by X-ray attenuation before and after freezing. Specifically, this work focuses on establishing the feasibility of CT (100-420 kV acceleration voltage) to accurately measure the concentration of glycerol or salt as model CPA and CDAs in unfrozen solutions and tissues at 20 degrees C, or the phase in frozen solutions and tissue systems at -78.5 and -196 degrees C. The solutions are composed of water with physiological concentrations of NaCl (0.88% wt/wt) and DMEM (Dulbecco's Modified Eagle's Medium) with added glycerol (0-8 M). The tissue system is chosen as 3 mm thick porcine liver slices as well as 2 cm diameter cores which were either imaged fresh (3-4 h cold ischemia) or after loading with DMEM based glycerol solutions (0-8 M) for times ranging from hours to 7 days at 4 degrees C. The X-ray attenuation is reported in Hounsfield units (HU), a clinical measurement which normalizes X-ray attenuation values by the difference between those of water and air. NaCl solutions from 0 to 23.3% wt/wt (i.e. water to eutectic concentration) were found to linearly correspond to HU in a range from 0 to 155. At -196 degrees C the variation was from -80 to 95 HU while at -78.5 degrees C all readings were roughly 10 HU lower. At 20 degrees C NaCl and DMEM solutions with 0-8 M glycerol loading show a linear variation from 0 to 145 HU. After freezing to -78.5 degrees C the variation of the NaCl and DMEM solutions is more than twice as large between -90 and +190 HU and was distinctly non-linear above 6 M. After freezing to -196 degrees C the variation of the NaCl and DMEM solutions increased even further to -80 to +225 HU and was distinctly non-linear above 4 M, which after modeling the phase change and crystallization process is shown to correlate with an amorphous phase. In all tissue systems the HU readings were similar to solutions but higher by roughly 30 HU, as well as showing some deviations at 0 M after storage, probably due to tissue swelling. The standard deviations in all measurements were roughly 5 HU or below in all samples. In addition, two practical examples for CT use were demonstrated including: (1) glycerol loading and freezing of tissue cores and, (2) a mock cryosurgical procedure. In the loading experiment CT was able to measure the permeation of the glycerol into the sample at 20 degrees C, as well as the evolution of distinct amorphous vs. crystalline phases after freezing to -196 degrees C. In the mock cryosurgery example, the iceball edge was clearly visualized, and attempts to determine the temperature within the iceball are discussed. An added benefit of this work is that the density of these frozen samples, an essential property in measurement and modeling of thermal processes, was obtained in comparison to ice.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Animals
  • Biocompatible Materials / chemistry*
  • Cryopreservation / methods*
  • Crystallography / methods*
  • In Vitro Techniques
  • Liver / chemistry*
  • Phase Transition
  • Radiographic Image Interpretation, Computer-Assisted / methods*
  • Swine
  • Tomography, X-Ray Computed / methods*


  • Biocompatible Materials