PERGAMON Carbon 38 (2000) 29–45 A methodology for analysis of 002 lattice fringe images and its application to combustion-derived carbons a a, b * Hong-Shig Shim , Robert H. Hurt , Nancy Y.C. Yang a Division of Engineering, Box D, Brown University, 182 Hope Street, Providence, RI 02912, USA b Sandia National Laboratories, Livermore, CA 94550, USA Received 10 July 1998; accepted 15 April 1999 Abstract A technique is presented for semi-quantitative digital analysis of 002 lattice fringe images of carbon materials. The automated technique yields statistics on fringe length and tortuosity, as well as new quantitative parameters describing the mode, degree, and length scale of orientational order among graphene layers. The technique is applied to combustion-derived carbons with special emphasis on the nanostructural characterization of a variety of solid fuel chars as a function of residence time. During the combustion of pulverized coal and biomass (120 ms at 1700–1900 K), the main features of char nanostructure are established early, near the end of the rapid, in-flame pyrolysis. Only for the high-volatile bituminous coals is the nanostructural order in the young chars significantly altered by the subsequent char combustion process. A noteworthy general observation is the presence of two distinct length scales for orientational order in most chars. At short length scales ˚ ( ,30 A) a high degree of orientational order is observed reflecting the presence of small molecular orientation domains. At ˚ larger length scale (.30 A), a lesser but significant degree of orientational order is still present among the distinct domains. Quantitative analysis of this low-grade long-range order using Maier–Saupe theory indicates that it cannot arise by a liquid crystal mechanism. Long-range orientational order in these samples can thus be classified as ‘mesophasic’, exhibiting a high degree of order arising through liquid crystal formation in the fluid phase of pyrolysis (or retention of anisotropy in very high rank coals), and ‘statistical’, perhaps arising from elongational strain during carbonization.  1999 Elsevier Science Ltd. All rights reserved. Keywords: A. Char; B. Combustion; C. TEM; D. Crystal structures, Textures 1. Introduction 002 lattice fringe imaging mode (002 LF). Lattice fringe imaging has been widely applied to fibers [14], soot and Carbon nanostructure, defined here as the local spatial carbon black [13,15], pyrolytic carbons [16], polymeric arrangement and orientation of nanoscale graphene layers, carbons [12], and coal chars [14,17–20], yielding im- affects many important properties of carbon materials, portant structural information and insight into bulk materi- including mechanical strength and modulus [1,2], coeffi- al properties. cient of thermal expansion [2], electrical properties [3] and Most of the results from TEM studies have been their directional dependencies; pore size distribution [4,5]; qualitative or semi-quantitative. Exceptions to this are the and reactivity of cokes to oxidizing gases [5]. Several determination of mean interlayer spacing from the electron techniques provide information on carbon nanostructure diffractogram, dark field studies [5], which can yield size (referred to by some researchers as ‘texture’, or ‘mi- distributions of molecular orientation domains, and several crotexture’), including X-ray diffraction [6–8], magneto- recent attempts to apply digital image analysis to 002 resistance [9], and high resolution TEM [4,10–13], allow- lattice fringe images [19,22,23]. Extraction of quantitative ing the identification of ordered regions by dark field information from 002 LF images is challenging due to techniques, and direct visualization of graphene layers in contrast variations, twisting and overlapping of the aro- matic layers, and amorphous interference [4]. More work is needed before reliable techniques are available to extract *Corresponding author. Tel.: 11-401-863-2685. E-mail address: Robert Hurt@brown.edu (R.H. Hurt) all the information embedded in lattice fringe images. ] 0008-6223 / 00 / $ – see front matter  1999 Elsevier Science Ltd. All rights reserved. PII: S0008-6223(99)00096-2