Molecular models of brown coal containing inorganic species

Abstract

Molecular models of inorganic solution species, brown coal and brown coal containing inorganic species have been optimised using the computer-aided molecular package for Windows, CAChe, and the general-purpose semi-empirical molecular package MOPAC. The coal models were based on elemental composition and the experimentally measured distribution of aliphatic carbon, aromatic carbons, and oxygen functional groups. The calculated partial charges and bond polarities of the model are consistent with the hydrophilic properties of brown coal. Water molecules within the model exert the major stabilisation through H-bonds and electrostatic interactions. Calculated values of bond lengths and angles of the inorganic models were in good agreement with reported values, but variations occur because they are modelled as isolated structures. Brown coal models with NaCl, Na⁺, Mg²⁺ and Ca²⁺ are stabilised when the respective cations have been located in spaces that reduce steric hindrance and are surrounded by oxygen functional groups, carboxyl anions and water molecules. Coal models with octahedral Fe(III) or Ni(II) complexes may form stable structures bonded to carboxyl groups, but mono-nuclear complexes with bi-dentate carboxyl ligands usually form distorted structures. Polynuclear complexes form energetically favoured structures in coal, especially when carboxyl ligands form mono-dentate bonds to the metal centres. Reported experimental data on iron and nickel complexes, and for polymeric iron species in brown coal, are consistent with the modelling results.

Introduction

Coal is a substance derived from a complex accumulation of plant remains in swamps and its heterogeneous nature precludes a definitive molecular structure. As a result, numerous attempts have been made to provide a simpler molecular model for the insoluble organic matter. Models for bituminous coals have included aromatic/hydroaromatic structures in which clusters of aromatic groups have been connected by various ether or aliphatic links. Models for low-rank coals are of greater heterogeneity, especially when humic acids, waxes and other materials are considered as components of brown coal. Ohkawa et al. (1997) considered various approaches for constructing coal molecular structure and proposed a knowledge and partial structure evaluation method. Ad hoc models of low-rank coals, based on elemental composition, carbon distribution and ratio of one-, two-, and three-ring fused aromatics, and oxygen distribution, have been developed, as for example by Hüttinger and Michenfelder, 1987, Domazetis, 2001. Discussions of the structures of lignin and the coalification process by Stout et al., 1988, Levine, 1993 point to central characteristics, such as the distribution of p-hydroxyphenylpropane, guaiacyl, syringyl units, and linkages such as arylglycerol-β-aryl, phenylcoumaran and ether, as reported by Dorrestijn et al. (2000). The degradation processes of wood modify functional groups and split inter-unit linkages, while the gradual degradation of polysaccharides is followed by a slow biotransformation of the lignin macromolecule by depolymerisation, demethylation, de-methoxylation and further defunctionalisation. The formation of carboxyl functional groups during coalification may occur as a result of easier oxidation of the C-1 carbon in lignin-like structures. Models derived from structural studies of coalified wood found in brown coals deposits have been reported by Hatcher, 1990, Hatcher and Clifford, 1997, Hatcher and Faulon, 1994, Hatcher et al., 1982, Hatcher et al., 1988, Faulon et al., 1994. A helical template for low rank coal models has been proposed by Hatcher and Faulon (1994).

Our major interest has been to develop a model of brown coal suitable for studies of the interaction of inorganic species with oxygen functional groups in the coal. To this end, we have developed and compared two models; one that encapsulates experimentally measured properties and another based on studies of coalified wood similar to that reported by Hatcher and co-workers.

Studies of the chemical interactions of various inorganic species have often been underpinned by a notion that an ‘ion-exchange’ takes place between aqueous inorganic species and oxygen functional groups effective in coal. The interactions between coal functional groups and various inorganic species, however, are far more complex than suggested by the concept of ion exchange. The inorganic species that may be added to the coal matrix span almost the entire range of metal complexes and such species may well be modified by their environment within the coal. Thus, salts such as NaCl and MgCl₂ may be added as a solution of dissolved anions and cations to the wet coal matrix, while acid–base chemistry may be used to add cations to the coal matrix containing carboxyl anions. Transition metal complexes can be added to brown coal, such as for example, the mononuclear and polymeric iron species reported by Domazetis et al., 2005a, Domazetis et al., 2005b. Davies et al. (1997) have discussed sites in humic acids having different affinities for metal complex formation and brown coal similarly is expected to contain different binding sites and affinities. As a result, numerous chemical interactions involving a variety of chemical species need to be invoked when considering the addition of inorganic species within brown coal.

The interactions of inorganics with hydrophilic low-rank coals are relevant to a number of coal utilisation processes, including the removal of inorganics to produce ultra-low ash coal, subsequent addition of selected inorganics, including Fe, Ni, species and studies of reaction pathways of metal mediated pyrolysis and low temperature catalytic gasification. This paper presents results of computer molecular modelling of brown coal, and studies of the interactions of the aqua complexes of Na⁺, Mg²⁺, Ca²⁺, Fe³⁺ and Ni²⁺ with carboxyl and phenoxy functional groups of the brown coal model.