The use of tetrahydrofuran (THF) as mobile phase in size exclusion chromatography (SEC) has been found to lead to partial loss of sample and to give anomalous results in the characterization of a liquefaction extract and its hydrocracking products. The problem has been resolved by using NMP (1-methyl-2-pyrrolidinone) as mobile phase in SEC, showing significant fractions of sample eluting at the exclusion limit of an identical SEC column. This fraction has not previously been observed in SEC chromatograms obtained in THF. Comparison of SEC chromatograms obtained by UV-absorption and UV-fluorescence detection (in NMP) suggests that the material observed at the exclusion limit of the column corresponds to larger, more complex polynuclear aromatic ring systems than those present in material separated by the column. In NMP, samples produced during progressively higher temperature hydrocracking experiments eluted, as expected, at longer times, indicating progressive molecular size reduction with increasing intensity of the reaction. These data are consistent with the UV-fluorescence spectra and TGA-derived boiling point distributions of the set of samples. A twofold mechanism for loss of material in THF-based SEC may be proposed: (i) not all the sample dissolves in THF and (ii) some of the larger/more polar molecules apparently soluble in THF tend to deposit on column packings and do not elute through the column. Considerable caution therefore appears necessary in using THF as mobile phase in SEC work for the characterization of complex coal-derived liquids.

Size exclusion chromatography (SEC) has been used extensively in the characterization of coal‐derived liquids to estimate molecular masses (MM), indicating the presence of considerably greater upper mass ranges compared to mass spectrometric methods based on heated probes and field ionization. In the present study, the use of tetrahydrofuran (THF), a solvent commonly used in SEC, has been shown to lead to partial loss of sample on column packings and to allow some separation by an adsorption mechanism rather than by size exclusion. In this paper, we have compared these results with data from SEC, where 1‐methyl 2‐pyrrolidinone was used as the mobile phase, indicating that the largest molecules (not observed in THF‐based work) appear at the exclusion limit, with a continuum of material down to the total permeation limit of the column. In terms of polystyrene standard masses, this is equivalent to a range from 100 u, at the lower limit, up to in excess of 30–40 000 u and possibly over 2 000 000 u. Our results from SEC in NMP suggest that limitations exist in much solution‐based work attempting to identify distributions of MMs in coal‐derived liquids. Matrix‐assisted laser desorption/ionization (MALDI) spectra of the fractions show similar features to those seen by SEC in NMP, with a continuum from high to low mass, and a significant shift to higher mass in the insoluble fraction compared with the soluble fraction. The new data are in line with observations by laser desorption mass spectrometry (LD‐MS) and MALDI‐MS, which have shown much greater upper‐MM limits than either SEC (in THF) or mass spectrometric techniques relying on thermal evaporation of coal‐derived materials from heated probes. The work would appear to have implications for other polydispersed natural polymer systems.

Two propositions relating to the interpretation of size-exclusion chromatograms (SEC) of coal-derived materials in 1-methyl-2-pyrrolidinone (NMP) have been examined. These were (i) that signal peaks showing up at exclusion (short retention time) limits of SEC columns are due to sample polarity alone and (ii) that shifts in SEC chromatograms to longer retention times, observed upon addition of LiBr to the eluent (NMP), are due to dissipation of ionic binding forces, causing disaggregation of polar clusters that would otherwise have appeared at retention times appropriate to larger molecular masses. In our experiments, effects due to polarity and molecular mass have been isolated by using two nonpolar samples (a naphthalene mesophase pitch and a mixture of fullerenes). In the presence of LiBr, precipitation of solute out of solution and shifts of chromatograms to longer retention times, unrelated to sample polarity, have been observed. A partial breakdown of the size exclusion mechanism was identified by the observed extension of chromatograms beyond the permeation limit of the column, similar to those observed when using eluents of insufficient solvent strength (e.g., THF, chloroform). Dosing LiBr into NMP sharply reduces the solvent power of NMP for coal-derived solutes. In the absence of LiBr, SEC chromatograms of the fullerene mixture, the naphthalene mesophase pitch, and its fractions separated by planar chromatography clearly showed significant signal under the “excluded” peak, entirely due to nonpolar material. The damage caused to the SEC column arising from precipitation of sample, in the presence of LiBr, was not permanent as had originally been feared. The balance of the evidence suggests that polarity of some molecules may cause shifts in their elution times to shorter values (larger apparent molecular masses) and that these may overlap with signal from large molecular mass material.