ACERC
Abstracts - 1991 |
ACERC
Abstracts - 1991 |
|
|
Thrust Area 1: Fuel Structure and Reaction Mechanisms |
Meuzelaar, H.L.C.; Statheropoulos,
M.; Huai, H. and Yun, Y.
Computer-Enhanced Analytical Spectroscopy, 3:185-214, (P.C. Jurs, ed.),
Plenum Publishing Co., New York, 1991. Funded by ACERC and US Department of
Energy.
The following sections of this chapter will demonstrate the application of Canonical Correlation Analysis (CCA) to multisource fossil fuel data generated by the application of three different spectroscopic methods: Pyrolysis Low Voltage Electron Ionization Mass Spectrometry (LVMS); Pyrolysis Field Ionization Mass Spectrometry (FIMS); and Photo-acoustic Fourier Transform Infrared Spectroscopy (FTIR) to a single suite of eight U.S. coals obtained from the Argonne National Laboratory Premium Coal Sample Program (ANL-PCSP). Consequently, the three data sets obtained describe the same suite of samples but are markedly different with respect to number as well as type of variables.
Chang, H.-C.K.; Bartle,
K.D.; Markides, K.E. and Lee, M.L.
Advances in Coal Spectroscopy, 141-164, (H.L.C. Meuzelaar, ed.), Plenum
Publishing Corp., New York, 1991. Funded by Gas Research Institute and ACERC.
In this chapter, results from the analysis of the low-molecular-weight organic constituents in six vitrinite-rich coals of ranks ranging from lignite to low volatile bituminous are reported. Aliphatics, neutral aromatics, nitrogen-containing aromatics, and sulfur-containing aromatics were first isolated from the coal extracts using column absorption and complexation chromatography, and then they were resolved and identified using capillary column gas chromatography coupled with sulfur- and nitrogen-selective detectors, and mass spectrometry.
Orendt, A.M.; Solum, M.S.;
Sethi, N.K.; Pugmire, R.J. and Grant, D.M.
Advances in Coal Spectroscopy, (H.L.C. Meuzelaar, ed.), Plenum Publishing
Corp., New York, 1991 (in press). Funded by Pittsburgh Energy Technology Center/Consortium
for Fossil Fuel Liquefaction, US Department of Energy and ACERC.
Techniques in C-13 nuclear magnetic resonance spectroscopy applied in the study of coal and coal chars are discussed along with details of the analysis of the spectral results. The results are compared for various methods of analysis: cross polarization with magic angle spinning (CP/MAS), dipolar dephasing (DD), MAS with block decays (BD), and chemical shielding anisotropy (CSA) measurements. Results of the CP/MAS and DD experiments on the Argonne premium coals as well as other coals and coal chars are reported in terms of twelve structural parameters, including aromaticity. Methods used to determine average cluster size and molecular weight are discussed. Models of coal structure and devolatilization processes are presented along with an analysis of the information obtained from the C-13 NMR experiments.
Solomon, P.R.; Fletcher,
T.H. and Pugmire, R.J.
Fuel, 1991 (in press). Funded by US Department of Energy and ACERC.
The heterogeneous nature of coal and the complexity of the pyrolysis process has made it very difficult to perform unambiguous experiments to determine the rates and mechanism in coal pyrolysis. The last several years have, however, provided a number of new experimental and theoretical approaches that shed new light on the subject. This paper will consider the recent progress on the topics of: kinetics, the formation of volatile products, network models, crosslinking, rank effects, and the "two-component model of coal structure." In kinetics, recent experiments that measure coal particle temperatures at high heating rates provide reasonable agreement on kinetic rate constants. The rates also agree with those derived from low heating rate experiments. In tar formation and transport, a consensus is being reached on the central role of the tar molecule's volatility in explaining the variation with operating parameters (pressure, heating rate, particle size, etc.) in the tar's amount and molecular weight distribution. Progress in the quantitative prediction of tar and char is being made by recently developed models for the fragmentation of the macromolecular network. These models, which provide quantitative description of the relationship between the chemical structure of the coal and the physical and chemical properties of the resultant pyrolysis products (gas, tar, soot, and char), are an exciting advancement in the understanding of the pyrolysis process. Such models are linking the occurrence of the coal's plastic phase with the "liquid" fragments formed during pyrolysis. On the subject of retrogressive crosslinking reactions, both solvent swelling and NMR measurements confirm important rank dependent differences in reaction rates. These appear related to the oxygen functionalities. Reasonable agreement is also seen for rank variations of kinetics rates derived from low heating rate experiments. Experiments suggest that the recently revived "two-component hypothesis" of coal structure has application to low rank coals that are a mix of polymethylenes and a more aromatic network. Bituminous coals, however, appear far more homogeneous. These coals appear to behave in a manner that is described by the network decomposition models. The presentation will provide a brief report on these topics.
McDonald, K.M.; Hyde, W.D.
and Hecker, W.C.
Fuel, 1991 (in press). Funded by ACERC.
Chars derived from Beulah-Zap (lignite A) and Dietz (sub-bituminous B) coals were prepared by three different methods utilizing three different reactor systems. These included a high heating rate method achieved in a methane flat flame burner, a moderate heating rate method achieved in a drop tube reactor, and a slow heating rate method achieved in a muffle furnace. The flat flame char was produced in a flame environment with excess oxygen, while the drop tube and muffle furnace chars were produced in inert environments. Low temperature oxidation rates and kinetic parameters were determined using isothermal thermogravimetric analysis (TGA) at temperatures between 550 and 950 K. Reactivities at different oxidation burnout levels (10%-75%) were compared on both an initial mass and an available mass basis. Using the available mass basis, rates in the intrinsic regime were found to be nearly identical for the different burnout levels. It was also found that the lower burnout levels are more highly influenced by diffusion effects. This was manifest by a decrease in the slope of the Arrhenius plot that began at a temperature of about 750 K for the char at 10% burnout compared to a temperature of nearly 900 K for the char at 75% burnout. In comparing the chars produced by the three different methods, reactivities in the reaction control regime showed that, for both coals, the drop tube char was more reactive than either the flat flame or muffle furnace char. Further tests indicated that the drop tube chars had a hydrogen to carbon ratio that was 2.5 to 5 times greater than the char from either of the other reactors and the devolatilization conversion was significantly less. The activation energies for all three Beulah-Zap chars and for the Dietz muffle furnace and flat flame chars were found to be 28.1±0.6 kcal/mol. A comparison of the reactivities for the flat flame burner chars of the lignite and the sub-bituminous showed that the lignite chars were more reactive by a factor of two. This was consistent over all burnout levels. Further work with the Dietz flat flame char showed the dependency on oxygen concentration, yielding an apparent reaction order of 0.67±0.03. This is in excellent agreement with data found in the literature.
Foster, M.P.; Mayne, C.L.;
Dunkel, R.; Pugmire, R.J.; Grant, D.M.; Kornprobst, J.-M.; Verbist, J.-F.; Biard,
J.-F. and Ireland, C.M.
Journal of the American Chemistry Society, 1991 (in press). Funded by
National Institutes of Health and US Department of Energy.
The two-dimensional structure of bistramide A (a.k.a. bistratene A) has been revised. The bis-amide polyether was isolated from a Lissoclinum sp. ascidian collected in the Fiji islands. The structure was established by a combination of modern NMR methods including 2D INADEQUATE, PS-DQF-COSY, HMQC and HMBC experiments. Critical to establishing the carbon-carbon backbone was CCBond, a new program for automated analysis of 2D INADEQUATE spectra. Bistramide A demonstrated in vitro cytotoxicity towards human colon tumor HCT-116 and murine leukemia L1210 cell lines with an IC50 of 0.1 µg/mL.
Orendt, A.M.; Sethi, N.K.;
Facelli, J.C.; Horton, W.J.; Pugmire, R.J. and Grant, D.M.
Journal of the American Chemistry Society, 1991 (in press). Funded by
US Department of Energy and ACERC.
The principal values of the C-13 chemical shift tensor were measured for the aromatic carbons of 1,4,5,8-tetramethylnaphthalene, 1,2,3,6,7,8-hexadropyrene, and pyracene using the variable angle spinning technique. Ab initio calculations of the complete shielding tensors are also reported and provided the orientation of the principal values in the molecular frame. Good agreement between theory and experiment is obtained. The chemical shift principal values and orientations of these substituted naphthalenes correspond to those found in naphthalene; the observed differences can be rationalized in terms of the strain introduced by the alkyl substituent. In the naphthalenic derivatives studied, the bridgehead carbons exhibit relatively extensive pi-electron delocalization not found in some bridgeheads in more highly condensed aromatic systems.
Yun, Y. and Meuzelaar, H.L.C.
Fuel Processing Technology, 27:179-292, 1991. Funded by US Department
of Energy and Utah Power and Light.
A simple slurry pH titration method was developed as a means of measuring the degree of mild air oxidation in coal, especially for low rank, non-caking coals. The results show that this pH titration method, with prior rehydration at 150ºC, is a practically useful way of determining the degree of air oxidation of a coal sample of known origin but uncertain weathering status. For coals of a lower rank than high volatile bituminous, this method can be simplified to direct measurement of the slurry pH without the rehydration step. Among the variables influencing the rate of mild air oxidation as measured by the slurry pH, temperature is the most important whereas humidity exhibits little or no effect on the initial slurry pH although it has a marked effect on the pH titration profiles. Use of Curie-point pyrolysis-low voltage mass spectrometry on weathered, subbituminous coal samples from a coal storage pile reveals that the main effect of humidity on the composition of the pyrolyzate appears to involve certain sulfur compounds. Application of the initial slurry pH as an oxidation/weathering index yields an activation energy of oxidation/ weathering in the 20-150ºC range of 50-65 kJ/mol (12-16 kcal/mol) and indicates the existence of several different oxidation mechanisms in the temperature regimes between 60 and 150ºC.
Page, S.H.; Goates, S.R.
and Lee, M.L.
The Journal of Supercritical Fluids, 4:109-117, 1991. Funded by Gas Research
Institute.
Mixed mobile phases have often been used improperly in supercritical fluid chromatography (SFC) and supercritical fluid extraction (SFE). Phase separation is unavoidable at certain conditions where conventional SFC and SFE are performed, resulting in inhomogeneous mobile phases. We describe here a relatively rapid and accurate approach to determining transition pressures at different temperatures and modifier concentrations. There is close agreement between the experimental values determined in this work and those published by others for methanol and carbon dioxide. The results have been used to test the predictions of several common methods for estimating the critical pressures of mixtures. The methods failed to predict the critical pressure to within 20 atm between approximately 4-20 mol % of modifier. Chromatographic performance was shown to be severely degraded when the pressure in the column is below the gas-liquid transition pressure. The effect of pressure drop in the column on phase behavior is discussed.
Buchanan, R.M.; Holbrook,
K.M.; Meuzelaar, H.L.C. and Leibrand, R.
IRD Application Brief, Hewlett Packard, 23:5091-2022E, 1991.
Combined TG/IR and, to some extent, TG/MS techniques are finding increased application for structure/reactivity studies as well as for characterization and quality control of synthetic polymers, natural products, and fossil fuels. Unfortunately, the high cost of commercially available integrated TG/IR and TG/MS (let alone TG/IR/MS) systems has kept these techniques well out of reach for most analytical laboratories.
The advent of low cost bench-top FTIR and MS systems, however, has opened up the possibility of designing and assembling an affordable combined TG/FTIR/MS system. Such a system, developed at the University of Utah, Center for Micro Analysis, consists of a standard HP GC/IRD/MSD system utilizing an HP 5890 gas chromatograph, an HP 5965A IRD, and an HP 5971A MSD coupled to a Perkin Elmer Model 7 TG with a high temperature (1600? K max.) furnace. A specially constructed heated transfer line assembly allows direct coupling of the TG system to the GC injection port. The GC oven acts as a convenient heated coupling and flow distribution module. In this experiment no chromatography is performed with only sample transfer from the TG directly to the detectors.
Meuzelaar, H.L.C.; Huai,
H.; Lo, R. and Dworzanski, J.P.
Fuel Processing Technology, 28:119-134, 1991. Funded by Pittsburgh Energy
Technology Center/Consortium for Fossil Fuel Liquefaction, ACERC and Advanced
Coal Technology Center.
In order to arrive at a more detailed chemical description of fossil coal associated resins we need to distinguish between micropetrographic, organic geochemical and process technological definitions, each of which may encompass varying quantities of constituents unrelated to fossil tree resins. New information on composition and origin of Utah Wasatch Plateau coal resins obtained by Curie-point pyrolysis/evaporation in combination with iso-butane chemical ionization mass spectrometry, as presented in this paper, points to the presence of four more or less distinct resin components: (1) a sesquiterpenoid polymer; (2) sesqui- and triterpenoid monomers and dimers; (3) a suite of triterpenoid alcohols, ketones and acids; and (4) a series of increasingly aromatized hydrocarbons with naphthalene and picene type skeletons. Moreover, a strong similarity is found between the composition of recent dammar resin and fossil. Wasatch Plateau coal resins indicating a possible Angiosperm (fam. Dipterocarpaceae) origin of these Upper Cretaceous coal resins. Some of the technological implications of these findings and the consequent need for a more precise chemical definition and nomenclature are discussed.
Pugmire, R.J.; Solum, M.S.;
Grant, D.M.; Critchfield, S. and Fletcher, T.H.
Fuel, 70:414-423, 1991. Funded by Pittsburgh Energy Technology Center
and ACERC.
Solid-state C-13 and H-1 nuclear magnetic resonance (NMR) spectroscopy techniques are used to investigate the relationship between chemical structures of coal and the char particles and condensed tar vapors produced from coals of various ranks at rapid heating conditions. The C-13 NMR analysis of the coal chars indicate that significant amounts of aliphatic material is released from the coal during devolatilization with little initial change to the aromatic cluster size or number of cross links per cluster. The evolution of the char structure following tar release is a function of the time/temperature history of the char. The structures of the primary tars are compared to the parent coal and the gas phase evolution of the tar structure is followed with time.
Crelling, J.C.; Pugmire,
R.J.; Meuzelaar, H.L.C.; McClennen, W.H.; Huai, H. and Karas, J.
Energy & Fuels, 5:688-694, 1991. Funded by US Department of Energy and
ACERC.
The objective of the present study is to examine the chemical structure and composition of Utah Wasatch Plateau coal resinite. Macerals were separated from the coal matrix by hand picking, sink-float treatments, and/or density gradient centrifugation (DGC). DGC separation was found to produce highly purified resinite fractions. Resinite-rich Wattis Seam coal samples were collected from fresh mine faces and, after varying degrees of concentration, subjected to C-13 magic angle spinning NMR, Curie-point pyrolysis MS, and Curie-point pyrolysis GC/MS in addition to petrographic analysis and fluorescence microscopy as well as conventional (e.g., ultimate analysis) characterization methods. The data obtained confirm recent findings indicating that the abundant blue/green fluorescing resinite component is a polymeric substance composed of sesquiterpenoid repeat units with a median size of 204 Da, corresponding to the empirical formula C15H24. The monomeric sesquiterpenoid units obtained during pyrolysis appear to represent different degrees of unsaturatization ranging from C15H26 sesquiterpenes to the aromatic C15H18 cadalene. Clearly, sesquiterpenoids constitute the bulk of extractable resinite in Wasatch Plateau field coal and are likely to be important precursors of the abundant extractable alkynaphthalene moieties in such coal.
Wells, W.F. and Smoot, L.D.
Fuel, 70:454-458, 1991. Funded by Pittsburgh Energy Technology Center.
Causal relationships were determined for the chemical and physical properties and reaction characteristics of three coals and two types of chars derived from these coals. The reactivities of the virgin coals and extract chars correlated well with the fuel rank. The higher proximate volatiles contents in these prepared fuels resulted in reactivities higher than those measured for the low volatile pyrolysis chars. The reactivities of the pyrolysis chars were strongly correlated with fuel properties, whose relative order of importance was: catalytic elements > porosity > hydrogen types > cluster size.
Yun, Y.; Meuzelaar, H.L.C.;
Simmleit, N. and Schulten, H.R.
Energy & Fuels, 5:22-29, 1991. Funded by ACERC.
Three different vacuum pyrolysis mass spectrometry techniques, viz. Pyrolysis-Field Ionization Mass Spectrometry, Thermogravimetry/Low Voltage Electron Ionization Mass Spectrometry and Curie-point Pyrolysis-Low Voltage Electron Ionization Mass Spectrometry were used to analyze samples of Pittsburgh #8 coal obtained from the Argonne National Laboratory Premium Coal Sample Program. The primary objective was to assess the effects of differences in experimental techniques and conditions, e.g., with regard to heating rates, pyrolysis methods and soft ionization procedures (FI vs. Low Voltage EI) on the mass spectral patterns. A second objective was to further characterize and study the pyrolysis behavior of Pittsburgh #8 coal. The results indicate that the distribution and the type of the primary pyrolysis products are largely independent of marked differences in heating rate (10-2 K/s - 104 K/s range) and sample size (2.5 x 10-5g - 5.0 x 10-2g range) as well as overall vacuum pyrolysis MS configurations and conditions used. All three vacuum pyrolysis MS techniques produce remarkably similar mass spectral patterns when analyzing Pittsburgh #8 coal. The results show that Pittsburgh #8 coal contains a significant amount of low temperature (<380ºC) evolving "bitumen" consisting primarily of alkyl-substituted aromatic components. The bitumen evolution step is followed by a partially overlapping "bulk pyrolysis" step characterized by the evolution of abundant hydroxy- and dihydroxy substituted aromatic compounds, thought to be primarily derived from vitrinitic components. During the bitumen evolution step the average MW of the compounds increases with temperature while maintaining a relatively narrow distribution. By contrast, during the bulk pyrolysis step, the average MW tends to decrease while exhibiting a much broader distribution.
Sherwood, M.H.; Facelli,
J.C.; Alderman, D.W. and Grant, D.M.
Journal of the American Chemistry Society, 113:750-753, 1991. Funded
by US Department of Energy.
Carbon-13 chemical shift tensors have been determined in single crystal nephthalene. The high field component of every tensor is oriented perpendicular to the molecular plane, as in other aromatic compounds. The bridgehead carbon tensor is nearly axially symmetric, with its low field component approximately perpendicular to the central bridging bond, in agreement with theoretical predictions. Protonated carbon tensors in naphalene are more asymmetric, with their low-field components directed approximately along the C-H bonds. The Pople model of chemical shielding with MNDO wave functions reproduces the experimental in-plane shielding components. These calculations show that the in-plane components of the carbon shift tensors are governed by the bond orders of the adjacent bonds, and shift tensors thus provide valuable information relating to aromaticity. The measured tensors deviate significantly from the symmetry of an isolated naphthalene molecule because of the lower symmetry of the molecular site in the crystal.
Maswadeh, W.; Huai, H. and
Meuzelaar, H.L.C.
ASC Preprints, Division of Fuel Chemistry, 36(2):733-740, 1991 (201st
ACS National Meeting, Atlanta, GA, April 1991). Funded by ACERC.
A major bottleneck in the development of novel coal characterization methods, such as laser pyrolysis GC/MS, capable of analyzing individual coal particles, is the unavailability of suitable standard samples. Although carefully homogenized and characterized standard coals are now available through the Argonne National Laboratory Premium Coal Sample Program (ANL-PCSP) such "statistically homogenous" coal powders are of limited value as reference materials for single particle analysis methods. Even if it would be feasible to prepare particles of closely similar chemical composition and size, e.g., by using highly concentrated coal maceral fractions and careful sieving, remaining variations in shape, density, porosity or thermal conductivity could still introduce an unacceptably high level of uncertainty for most optimization and calibration purposes.
In an attempt to find model coal char particles with well-defined chemical and physical properties, e.g., for the purpose of modeling char oxidation reactions, several authors have used Spherocarb® particles. Flagan et al have even prepared spherical char particles spiked with mineral matter components in order to more closely mimic actual coal char particles. Although Spherocarb particles still show considerable variability with regard to size (rel. s.d. ~20% on a volume basis), other characteristics such as shape, density, porosity, thermal conductivity and chemical composition are assumed to be quite constant.
Unfortunately, Spherocarb particles are of little value for modeling coal devolatilization reactions due to their very low volatile matter yields. This prompted us to think of ways to increase volatile matter yields by introducing a variety of model compounds, ranging from low molecular weight, bitumen-like components to polymeric materials such as soluble lignins or resins. To the best of our knowledge this article represents the first reported use of bitumen and polymer impregnated Spherocarb particles for modeling devolatilization processes in individual coal particles.
Fletcher, T.H.; Grant, D.M.
and Pugmire, R.J.
ACS Division of Fuel Chem. Preprints, 36(1):250-257, 1991 (201st ACS
National Meeting, Atlanta, GA, April 1991). Funded by Pittsburgh Energy
Technology Center and ACERC.
Models of coal pyrolysis have progressed from simple one or two step empirical Arrhenius expressions that correlate total mass release during devolatilization, as reviewed by Anthony and Howard, to detailed descriptions of hydrocarbon chemistry and mass transport. These models describe the yields and compositions of pyrolysis products from coal under a wide range of heating conditions and ambient pressures. During pyrolysis of softening coals, a liquid phase appears that is referred to as metaplast. Release of pyrolysis gases and tar vapors inside the particle cause bubble formation in the softened coal particle, followed by swelling (increase in the particle diameter) with large internal voids (cenosphere formation). The softened state is followed by crosslinking or repolymerization which solidify the char matrix. As the coal particle is heated to sufficiently high temperatures, the light species in the metaplast are released as hydrocarbon vapors, along with light gases. Coal tar is generally defined to consist of those species which are released from the coal during pyrolysis which condense at room temperature and pressure. Low rank coals and lignite generally give low tar yields, and do not exhibit much softening or swelling behavior; this non-softening behavior may be caused by early crosslinking reactions. High rank coals (i.e., anthracites and low volatile bituminous coals) contain low amounts of volatile matter, and hence coal particles remain relatively intact during pyrolysis unless fragmentation occurs.
Mass transport affects coal pyrolysis in two ways: (1) as the ambient pressure increases, the tar yield decreases, and (2) as particle size increases, the tar yield decreases. However, there seem to be regions where the two mass transport effects are not controlling. For instance, in vacuum, the small pressure generated inside the pyrolyzing coal particle from the release of light gases and tar vapors may control the process. Also, total volatiles yields from a lignite were observed to remain constant with increasing ambient pressure, although this is probably due to the low tar yield of the lignite. Changes in coal pyrolysis yields as a function of particle size for diameters less than 200 µm are small.
Hecker, W.C.; McDonald,
K.M.; Jackson, C.D. and Cope, R.F.
Proc. of the 1991 International Conference on Coal Science, 263-266,
(IEA Coal Research, ed.), Butterworth-Heinemann, London, September 1991. Funded
by ACERC.
To accurately model the combustion of coal, it is necessary to characterize both the coal devolatilization and char oxidation processes. To characterize char oxidation, the heterogeneous reaction of devolatilized coal with oxygen, it is important to understand how the kinetics of char oxidation vary with the burnout (or conversion) level of the char. For example, as a char particle burns, does its rate (normalized to the instantaneous amount of combustible material) vary or does it remain constant? It is also important to know how the degree of devolatilization achieved in preparing a char affects the oxidation kinetics of the resulting char. The objective of this study was to determine the effects of 1) degree of devolatilization, 2) extent of oxidation burnout, and 3) type of oxidation burnout on intrinsic char oxidation rates. Types of oxidation burnout include both those achieved at high temperatures (in a drop tube reactor) and those achieved at low temperatures (in a thermogravimetric analyzer, TGA). The intrinsic (i.e. free of transport limitations) rate and activation energy of each char was determined using isothermal TGA. Hydrogen/carbon ratio and N2 surface area were also measured, and attempts were made to relate these properties to differences in the observed char oxidation rates.
Dworzanski, J.P.; Buchanan,
R.M.; Chapman, J.N. and Meuzelaar, H.L.C.
ASC Preprints, Division of Fuel Chemistry, 36(2):725-732, 1991 (201st
ACS National Meeting, Atlanta, GA, April 1991). Funded by Pittsburgh Energy
Technology Center/Consortium for Fossil Fuel Liquefaction, ACERC, Hewlett Packard
Corp. and US Department of Energy.
Thermal analytical methods have been widely used during the last two decades in the study of biomass thermochemical conversion processes. Biomass, which represents a renewable energy resource, consists primarily of plant cells differentiated into characteristic tissues and organs. Lignins, hemicelluloses and cellulose, as the main components of the cell walls, were therefore extensively analyzed, especially from the point of view of their thermochemical reactivity, which is of basic importance for industrial processing of biomass.
All types of cellulose microfibrils are composed of linearly linked b-(1-->4)-D-glucopyranose units and differ only by the degree of polymerization. The remaining polysaccharides are known collectively as hemicelluloses and exhibit species related composition. These amorphous, complex heteropolymers characterized by a branched molecular structure exhibit a lower degree of polymerization than cellulose. Xylan is the predominant hemicellulose component of angiosperms ("hardwoods") whereas mannan forms the main hemicellulose of gymnosperms ("softwoods"). The third principal component of biomass, viz. lignin, is an irregular, high MW polymer formed by enzyme-initiated, free-radical polymerization of coniferyl alcohol (in hardwoods), coniferyl plus sinapyl alcohols (in softwoods), or coumaryl alcohol plus both above mentioned alcohols (in grasses). Lignins act as binding agents for the cellulose and hemicellulose fibers through a variety of linkages involving ether and carbon-carbon bonds of aromatic rings and propyl side chains.
Thermochemical conversion processes of lignocellulosic materials have been studied using mainly thermogravimetry (TG) or flash pyrolysis (Py) followed by gas chromatographic (GC) separation and identification of the reaction products. Modern analytical techniques based on coupled Py-GC/mass spectrometry (Py-GC/MS) or direct Py-MS as well as TG/MS or TG/infrared spectroscopy (TG/IR) have proved to be especially useful for elucidating the relationships between biomass structure and pyrolysis/devolatilization mechanisms.
A novel TG/(GC)/FTIR/MS system developed at the University of Utah, Center for Micro Analysis and Reaction Chemistry provides the opportunity for combining accurate weight loss measurements with precise information about composition and evolution rates of gaseous and liquid products as a function of temperature. In this paper, the usefulness of TG/FTIR/MS, TG/GC/MS and TG/GC/FTIR for thermochemical characterization of wood, lignins and cellulose will be discussed.
Cope, R.F.; Hecker, W.C.;
Monson, C.R. and Germane, G.J.
Western States Section/The Combustion Institute, Los Angeles, CA, October
1991. Funded by Advanced Fuel Research, Morgantown Energy Research Center and
ACERC.
Early attempts to determine the high-temperature reactivity of coals and chars were hampered by the inability to measure a burning particle's temperature and residence time. Researchers have typically approximated these values with the average temperature and residence time of a cloud of burning particles. Average values, however, do not account for particle-to-particle variations or their possible causes. In 1984 researchers at Sandia National Laboratories developed an optical technique to simultaneously measure the temperature, velocity and diameter of individual particles burning in a flat flame facility. This work reports modifications to the Sandia technique that allow measurement of smaller particles (ca. 30-200 µm) and its application to particles burning in atmospheric and elevated pressure drop-tube reactors.
The modified pyrometer is applied to drop-tube reactors, rather than flat flame burners, to allow a broad range of well-controlled gas environments and operating pressures for char oxidation experiments. Electrical wall heating allows good control of particle temperature histories, however glowing reactor walls present some unique challenges. Diffuse radiation emitted from the reactor walls enters the pyrometer's optical path, producing excess noise in the particle signal. This noise has been minimized by optimizing the design of the reactors' optical ports and the alignment of the pyrometer. The possibility of temperature measurement error caused by wall emissions reflecting off of the particle is shown to be minimal by Maxwell's electromagnetic equations.
Particle properties are obtained by focusing the burning particle's image onto a coded aperture that, as re-designed at Brigham Young University, contains a series of carefully sized blackouts and windows. Temperature is measured by conventional 2-color pyrometry, while velocity is obtained from the particle's transit time. Particle size is extracted from the signal by means of a unique geometric/statistical fitting technique developed during this work. Proper operation of the modified pyrometer has been verified during oxidation of both Spherocarb and coal char in the drop-tube reactors.
McDonald, K.M.; Reade, W.C.;
Cope, R.F. and Hecker, W.C.
Western States Section/The Combustion Institute, Los Angeles, CA, October
1991. Funded by ACERC.
The development and validation of accurate coal combustion modeling requires separate characterization of the coal devolatilization and char oxidation processes. The characteristic time for heterogeneous oxidation of a char particle in a commercial combustor is 1-2 orders of magnitude greater than the time required for devolatilization and homogeneous oxidation of the volatiles. Thus the rate of combustion and hence the efficiency of the coal combustion process will be governed by the rate of oxidation of the nonvolatile coal char.
The effects of both extent of burnout and type of burnout on char oxidation rates and rate parameters have been investigated for chars prepared from Dietz (subbituminous B) coal. Intrinsic rates and rate parameters (reaction order, activation energy, and pre-exponential factor) were determined using isothermal thermogravimetric analysis (TGA). N2BET and CO2DP surface areas were measured, as was hydrogen to carbon ratio. Three types of burnout were studied and compared. Devolatilization mass loss (DML) was studied by devolatilizing the Dietz coal to various extents in a flat-flame methane burner (FFB) and then comparing the oxidation rates and other properties of the resulting chars. High-temperature oxidation burnout was studied by taking a given FFB Dietz char and oxidizing it to various conversion levels in a drop tube reactor at high temperatures characteristic of industrial combustors. The oxidation rates and kinetics of these partially burned out char samples were then determined using TGA. Low-temperature oxidation burnout was studied by oxidizing the given FFB Dietz char to a continuum of burnout levels in the TGA at low temperature (550 to 750 K) and obtaining the instantaneous oxidation rates, which are inherent in the TGA experiment.
The rate of oxidation was found to decrease with increasing devolatilization residence time, even after devolatilization mass loss has become constant. Increasing N2 and CO2 surface areas with devolatilization residence time are inverse to the decreasing rates, and a constant difference between N2 and CO2 surface areas indicates dramatic changes in the mesopore surface area during devolatilization, but not in the micropore surface area. Intrinsic rates of chars oxidized at high temperatures were found to decrease with burnout level, while those of chars oxidized at low temperatures were essentially constant with burnout level.
Hecker, W.C.; McDonald,
K.M.; Reade, W.C.; Jackson, C.D. and Cope R.F.
1991 Annual Meeting of the American Institute of Chemical Engineers,
Los Angeles, CA, November 1991. Funded by ACERC.
The development and validation of accurate coal combustion modeling requires separate characterization of the coal devolatilization and char oxidation processes. The characteristic time for heterogeneous oxidation of a char particle in a commercial combustor is 1-2 orders of magnitude greater than the time required for devolatilization and homogeneous oxidation of the volatiles. Thus the rate of combustion and hence the efficiency of the coal combustion process will be governed by the rate of oxidation of the nonvolatile coal char.
The effects of both extent of burnout and type of burnout on char oxidation rates and rate parameters have been investigated for chars prepared from Dietz (subbituminous B) coal. Intrinsic rates and rate parameters (reaction order, activation energy, and pre-exponential factor) were determined using isothermal thermogravimetric analysis (TGA). N2BET and CO2DP surface areas were measured, as was hydrogen to carbon ratio. Three types of burnout were studied and compared. Devolatilization mass loss (DML) was studied by devolatilizing the Dietz coal to various extents in a flat-flame methane burner (FFB) and then comparing the oxidation rates and other properties of the resulting chars. High-temperature oxidation burnout was studied by taking a given FFB Dietz char and oxidizing it to various conversion levels in a drop tube reactor at high temperatures characteristic of industrial combustors. The oxidation rates and kinetics of these partially burned out char samples were then determined using TGA. Low-temperature oxidation burnout was studied by oxidizing the given FFB Dietz char to a continuum of burnout levels in the TGA at low temperature (550 to 750 K) and obtaining the instantaneous oxidation rates, which are inherent in the TGA experiment.
The rate of oxidation was found to decrease with increasing devolatilization residence time, even after devolatilization mass loss has become constant. Increasing N2 and CO2 surface areas with devolatilization residence time are inverse to the decreasing rates, and a constant difference between N2 and CO2 surface areas indicates dramatic changes in the mesopore surface area during devolatilization, but not in the micropore surface area. Intrinsic rates of chars oxidized at high temperatures were found to decrease with burnout level, while those of chars oxidized at low temperatures were essentially constant with burnout level.
Pugmire, R.J.; Orendt, A.M.;
Facelli, J.C. and Grant, D.M.
International Conference on Coal Science, University of Newcastle-upon-Tyne,
U.K., 1991. Funded by US Department of Energy and ACERC.
The measurement of C-13 chemical shift tensor components in organic compounds has been shown to be valuable in gaining insight into the relationship between structure and the individual shift components, in a manner similar to earlier work on correlations between structural features and average or solution chemical shifts. In this work, the tensor components are measured for the aromatic carbons in several polycyclic aromatic hydrocarbons (PAHs); the compounds were chosen due to their incorporation of structural features that are believed to be important in the aromatic clusters in coals. The aromatic carbons in these compounds, as in coals can be divided into four groups: protonated, alkyl substituted, oxygen substituted, and bridgehead or condensed.
While there have been many solid state C-13 NMR measurements on both coals and PAHs using the anisotropy averaging technique of magic angle spinning (MAS), which measured only the average or isotropic chemical shift values, only a limited amount of data exists on the tensor components on either group of samples. For aromatic carbons, the majority of available data is on substituted benzenes; the only polycyclic aromatic compounds for which tensorial data exists in the literature are naphthalene (I), pyrene (V), coronene, and most recently buckminsterfullerene (C60). Therefore, before any correlations between the tensorial components and structure can be explored, much more data on this class of compounds must be obtained. Recently, the shielding data on 1,4,5,8-tetramethynaphthalene (II), 1,2,3,6,7,8-hexahydropyrene (III), pyracene (IV) (the alpha protonated carbons only), and triphenylene (VII) have been obtained on our laboratory, and the results are discussed below.
The chemical shielding tensor is a measure of the effect that the electronic environment has on changing the strength of the magnetic field experienced at the nucleus. Depending on the orientation a given molecule has with respect to the applied magnetic field, the observed resonance frequency changes. Therefore, the chemical shielding tensor is a reflection of the three dimensional electronic environment of the nucleus and, as such, it contains much potential information about structure and bonding. If one studies a powder sample, (where all possible orientations of the molecule can exist), a broad pattern is obtained from which the three principal values of the shielding tensor can be extracted. Each of the four classes of aromatic carbons has distinctive lineshape reflective of the differences in their electronic surroundings.
Experimentally, there are a number of methods that can be applied to obtain these tensor components. For the results presented below, measurement was either made on a static sample or the technique of variable angle sample spinning (VASS) was used. In the case of phenanthrene, isotopic C-13 labeling was used in the positions 1, 4, and 9, in order to obtain the tensor data on the three protonated carbons with very similar shielding tensors. Quantum mechanical calculations of the complete shielding tensor were also completed on these and other aromatic molecules of interest. These calculations are essential in providing the orientations of the tensor components in the molecular frame, information not obtained experimentally. Past experience has indicated that the calculated orientations are quite reliable (e.g. within ±7 º) in the cases where orientations have been determined experimentally either by single crystal or dipolar spectroscopy.
Solomon, P.R.; Charpenay,
S.; Yu, Z.-Z.; Serio, M.A.; Kroo, E.; Solum, M.S. and Pugmire, R.J.
8th Annual International Pittsburgh Coal Conference, Pittsburgh, PA,
October 1991. Funded by US Department of Energy and ACERC.
Coal pyrolysis is a complicated combination of chemical and physical processes in which coal is transformed at elevated temperatures to produce gases, tar, and char. These processes are described in the Functional Group - Depolymerization, Vaporization, and Crosslinking (FG-DVC) model of coal pyrolysis. An important aspect of this model is that crosslinking is rank dependent. This is based on solvent swelling experiments on chars made from coals of different rank. Low rank coals start to loose their solvent swelling ability prior to significant depolymerization at temperatures as low as 200ºC. Including such crosslinking in the FG-DVC model leads to predictions for low rank coals of a highly crosslinked network (exhibited by low solubility and low fluidity in chars) and low tar amounts.
While the model is in good agreement with a variety of data, it is difficult to find experiments to validate the predicted behavior of the network. In this paper we have used CP-MAS, NMR with dipolar dephasing and other techniques to examine the chars changing functional group and network characteristics. The changes in the char composition have been modeled using the FG-DVC model and the results compared with the data for Pittsburgh Seam bituminous coal and Zap lignite.
Solomon, P.R.; Fletcher,
T.H. and Pugmire, R.J.
8th Annual International Pittsburgh Coal Conference, Pittsburgh, PA,
October 1991. Funded by US Department of Energy and ACERC.
The heterogeneous nature of coal and the complexity of the pyrolysis process has made it very difficult to perform unambiguous experiments to determine the rates and mechanism in coal pyrolysis. The last several years have, however, provided a number of new experimental and theoretical approaches that shed new light on the subject. This paper will consider the recent progress on the topics of: kinetics, the formation of volatile products, network models, crosslinking, rank effects, and the "two-component model of coal structure." In kinetics, recent experiments that measure coal particle temperatures at high heating rates provide reasonable agreement on kinetic rate constants. The rates also agree with those derived from low heating rate experiments. In tar formation and transport, a consensus is being reached on the central role of the tar molecule's volatility in explaining the variation with operating parameters (pressure, heating rate, particle size, etc.) in the tar's amount and molecular weight distribution. Progress in the quantitative prediction of tar and char is being made by recently developed models for the fragmentation of the macromolecular network. These models, which provide quantitative description of the relationship between the chemical structure of the coal and the physical and chemical properties of the resultant pyrolysis products (gas, tar, soot, and char), are an exciting advancement in the understanding of the pyrolysis process. Such models are linking the occurrence of the coal's plastic phase with the "liquid" fragments formed during pyrolysis. On the subject of retrogressive crosslinking reactions, both solvent swelling and NMR measurements confirm important rank dependent differences in reaction rates. These appear related to the oxygen functionalities. Reasonable agreement is also seen for rank variations of kinetics rates derived from low heating rate experiments. Experiments suggest that the recently revived "two-component hypothesis" of coal structure has application to low rank coals that are a mix of polymethylenes and a more aromatic network. Bituminous coals, however, appear far more homogeneous. These coals appear to behave in a manner that is described by the network decomposition models. The presentation will provide a brief report on these topics.
Carlson, R.E.; Vorkink,
W.P.; Lee, M.L.; Zhang, Y. and Shabtai, J.S.
1991 Pittsburgh Coal Conference and Exposition on Analytical Chemistry and
Applied Spectroscopy, Chicago, IL, March 1991. Funded by ACERC.
Elucidation of the organic structure of coal is vital to designing and optimizing processes for direct coal usage (e.g., combustion or coal conversion), and the efficient utilization of this abundant natural resource. Due to the extremely complex macromolecular structure of coal, present structural information is sparse. The analysis of coal and coal derived materials has been approached by numerous analytical and chemical techniques. However, many of those methods fail to yield specific information concerning the macromolecular network, or they significantly alter the network. Solvent extracts have provided valuable information concerning the "easily" extractable portion of the coal matrix, again leaving the macromolecular skeleton virtually unexplored. Structural investigations of coal carried out at high temperatures (>300ºC) lead to free radical rearrangement, crosslinking and thermal cracking of the sample.
A mild depolymerization method has been developed which yields a tetrahydrofuran (THF) soluble product that represents 50 to >70% of the original material, depending on coal rank. Detailed analysis of these products was performed using high-resolution gas chromatography and gas chromatography/mass spectrometry. The qualitative and quantitative results of four depolymerized Argonne coals (Beulah Zap, Illinois No. 6, Utah Blind Canyon, Pocahontas No.3) will be discussed and compared to room temperature solvent extracts (THF) of the same coals.
Maswadeh, W. and Meuzelaar,
H.L.C.
Proceedings of the 39th ASMS Conference on Mass Spectrometry and Allied Topics,
192-193, Nashville, TN, May 1991. Funded by ACERC.
Laser pyrolysis GC/MS using an electronically pulsed cw CO2 laser focused at the center of a Paul trap type particle levitation device connected to a Paul trap type spectrometer (Finnigan MAT ITMS) by means of a short, ballistically heated capillary GC column has proved to be a valuable technique for studying the devolatilization (pyrolysis + desorption) behavior of single, 100-150 µm sized coal particles at very high heating rates (105-106 K/s).
A modified experimental set-up, has been designed in which an electrodynamic balance cell is replaced with a simple electron microscopy (EM) grid (400 mesh, 78% open) to support and stabilize particles while providing markedly improved collection efficiency for laser pyrolysis and desorption products. Comparison of the two GC/MS profiles shows a high degree of qualitative similarity while illustrating the higher yield and improved signal to noise ratio of the EM grid technique. The improved signal quality is especially important for detecting and identifying minor components, reflecting subtle changes in pyrolysis mechanisms as a result of the high laser heating rate.
Orendt, A.M.; Solum, M.S.;
Sethi, N.K.; Hughes, C.D.; Pugmire, R.J. and Grant, D.M.
Magnetic Resonance of Solid Carbonaceous Fuels, ACS Symposium Series,
(R.E. Botto and Y. Sanada, eds.), 1991 (in press). Funded by US Department of
Energy and ACERC.
The methods available in nuclear magnetic resonance to obtain the principal values of the shielding tensor are discussed. Applications to coal and to compounds that model structures that might be important in coal are presented. The composition of aromatic carbons in coal as determined by chemical shielding powder patterns is compared to results obtained by cross polarization with magic angle spinning (CP/MAS) and dipolar dephasing (DD).
|