Search

The current project describes physical and chemical alterations that result from the thermal exposure of sunflower achenes and seeds. Specifically, achenes and seeds were heated at temperatures ranging from 130 to 600 °C under anoxic conditions for 60 min. Changes were measured in mass, relative percentages of C and N, internal and external morphology, molecular composition by direct temperature-resolved mass spectrometry (DTMS) under EI conditions, and the reflectance on polished specimens. The main constituents of sunflower seeds are lipids and protein, while the pericarps or hulls enclosing the seeds have a ligno-cellulosic structure. The morphology of achenes and seeds remains intact following exposure which makes the identification of the residues possible. The results of the sunflower experiments were compared to previous experiments conducted with peas and wheat grains. Up to approximately 340 °C, the molecular conversion of polysaccharide and protein rich peas and wheat grains follow similar pathways, while the lignin present in sunflower achenes follows its own unique pathway, resulting in a different molecular composition. At higher temperatures the molecular composition of the three propagules becomes identical. In addition the reflectance of the three propagules is identical confirming the similarity in molecular composition. Lipids show a different behaviour compared to the other three biopolymers; from 370 °C no oils are observed in the heated specimens and the DTMS-EI measurements show no lipid markers. Apart from evaporation of the triacylglycerides it is suggested that steroids have a high resistance against thermal degradation and are converted into aromatic moieties.

Peas (Pisum sativum L.) consist mainly of cotyledons and their bulk material is starch and proteins. Their structure is rather confined. Peas were heated at temperatures ranging from 130 to 700 °C under anoxic conditions for maximum 2 h. For each temperature a separate experiment was carried out in a pre-heated oven. Direct temperature-resolved mass spectrometry (DTMS) under electron impact (EI) and chemical ionization (CI, NH3) conditions measured the molecular composition of each solid residue. Fourier-transform infrared (FTIR) spectroscopy was carried out on residues heated at higher temperatures. The resulting solid products of the heating process still show the original markers for polysaccharides and proteins up to 270 °C. Concurrently three phases can be considered, each characterized by its own products. The first phase, from 270 up to 310 °C, shows monosaccharides, protein fragments and aromatic compounds. The second phase, from 310 °C up to a transitional stage from 400 to 440 °C, releases various aromatic and heterocyclic compounds. The third phase at higher temperatures shows a highly C-enriched product that releases CO, CO2, HCN, SO and SO2. The EI and CI experiments fail to discriminate between the pyrolysis products of polysaccharides and proteins between 250 and 400 °C. FTIR shows in this temperature range the development of an aromatic network. An earlier classification based on the changes of the physical and bulk chemical properties as a function of the temperature corresponds well with the classification based on the molecular changes. Residues of heated peas, usually called carbonized peas, are found in the archaeological record. Our study suggests that peas should have been heated up to least 310 °C before their residues survive natural degradation processes.

Direct temperature resolved mass spectrometry (MS) is a microanalytical technique to analyse modern paints by thermal separation and ionisation of organic pigments and polymeric fractions from a platinum/rhodium filament inside the ionisation chamber of the mass spectrometer. Most organic pigments in modern paints are desorbed at lower temperatures. Ethyl acrylate/methylmethacrylate or butyl acrylate/methylmethacrylate copolymers used in acrylic emulsion paints produce mono and oligomeric subunits released from the copolymers at high temperature by pyrolysis. Characteristic low voltage electron ionisation and ammonia chemical ionisation (CI) mass spectra of these copolymers facilitate their identification. DTMS of three different commercial acrylic emulsion paints showed low and high temperature events that could be related to the presence of organic pigments and the acrylic copolymers. Polyethylene glycols with molecular weight up to 2000 Da were identified as additives under ammonia CI conditions. The azo pigments PY3, PY73 and PY74, and the phthalocyanine pigment PG7 reported show molecular ions and a few characteristic fragment ions under direct temperature resolved mass spectrometry (DTMS) analytical conditions. Yellow azo pigments were identified under DTMSMS conditions by their high energy collisionally induced fragment patterns of their parent ions. Laser desorption ionisation mass spectrometry (LDIMS) using a nitrogen laser (337 nm) of an acrylic emulsion paint with four different organic pigments produced radical cations, protonated or sodiated ions of the pigments. The amide bond in the azo pigment PY3 was photolytically cleaved and produced a specific fragment ion. The trace additive polyethylene glycol was observed preferentially while the acrylic copolymers were transparent.

Asphalt has been used since antiquity for various purposes and by painters since the 17th century. Asphalt as a pigment material with a known pre-19th century provenance is extremely rare. The analytical pyrolysis techniques of direct temperature-resolved mass spectrometry (DTMS) and pyrolysis gas liquid chromatography coupled with mass spectrometry (Py-GC/MS) were applied to characterise a unique ‘asphalt’ sample from a painting materials collection dating from the early 19th century. The provenance of this sample was investigated by comparison with a sample from a present day floating asphalt from the Dead Sea. DTMS demonstrated a complex envelope of ions from hydrocarbons ranging past m/z 1000 in both samples. Both samples showed evaporating sets of compounds as well as pyrolysing materials evolving from asphaltenes. Asphaltene network polymers were marked by aliphatic cross-links releasing alkenes and alkanes, and aromatic centres releasing alkylbenzenes, alkylnaphthalenes, alkylanthracenes, alkylphenanthrenes, and alkylbenzothiophenes. Hopanoid hydrocarbons and C-ring monoaromatic steroid hydrocarbons were among the biomarker compounds identified. Particularly striking were homologous series of alkylbenzothiophenes and alkyldibenzothiophenes with long hydrocarbon chains that could be traced with DTMS and confirmed by DTMS/MS. The results indicate that the early 19th century ‘asphalt’ sample from the Hafkenscheid collection is indeed a true asphalt that shows many similarities to the Dead Sea asphalt sample, which is therefore proposed as the possible region of provenance.

The formation of α-methylated and α,α-dimethylated (di)acids during on-line (trans)methylation with tetramethylammonium hydroxide (TMAH) in a Curie-point pyrolysis unit is studied. The reaction of TMAH with nonanedioic, hexadecanedioic and octadecanoic acids is investigated to determine the contribution of the solvent system and the amount of TMAH added to this phenomenon. The results obtained clearly show a different behaviour of diacids in methanolic solutions compared to aqueous systems. At least three different reaction products were identified for the acids when using a methanolic solution of TMAH. Different relative amounts of by-products were formed when varying the amount of TMAH reagent relative to the normal methylated reaction product. A mechanism is postulated for their formation and a hypothesis is drawn up to explain the observed differences in reactivity for both solvents.

The chemical composition of suberin in cork from Quercus suber L. was determined on virgin and reproduction cork from different sites and trees in Portugal. Extractive-free samples were analyzed by pyrolysis-gas chromatography/mass spectrometry in the presence of tetramethylammonium hydroxide (TMAH), which provides in situ hydrolysis and (trans)methylation of hydroxyl and carboxyl groups of the suberin monomers. The average chemical monomeric composition of suberin on virgin and reproduction cork is, respectively, alkanoic acids 5.1, 3.3%; α,ω-diacids 11.2, 10.1%; ω-hydroxyacids 45.0, 48.1%; alkanols 1.9, 1.8%; 9,10-epoxy-18-hydroxyoctadecanoic acid 5.9, 5.8%; 9,10-epoxyoctadecanodioic acid 4.8, 3.6%; 9,10,18-trihydroxyoctadecanoic acid 7.5, 10.4%; 9,10-dihydroxyoctadecanedioic 6.1, 6.8%; ferulic acid 5.8, 4.5% and glycerol 4.3, 4.3%. In general, larger variability was found for the suberin monomers in virgin cork. Total content of ω-hydroxyacids or of C18 or C22 compounds is however similar for all samples.

To date, direct temperature-resolved mass spectrometry (DT-MS, a form of temperature-resolved, in-source pyrolysis mass spectrometry) has generally been used to provide qualitative views of the broad-band molecular-level characteristics of complex organic samples. In this paper low voltage electron impact (EI + ) DT-MS is used to provide semi-quantitative information on molecular-level characteristics of marine particulate organic matter (POM) samples. This class of samples is analytically challenging for two main reasons. Firstly, it is difficult to obtain representative data sets in which sample sizes are sufficient for characterization. Secondly, marine organic matter samples contain many identifiable compound classes in conjunction with a substantial percentage of as-of-yet uncharacterizable material. As a first step toward quantitation, DT-MS analysis of selected standards was used to determine relative response factors of common compound classes identified in marine organic matter. These response factors were then used to quantify the relative contribution of the compound classes in two natural POM samples. The potential of DT-MS as a quantitative technique was then further explored by comparing DT-MS determinations of particulate organic carbon to chlorophyll ratios (POC/chl) in natural POM samples with the same ratios determined by elemental analysis and liquid chromatography (HPLC), a technique currently employed by marine chemists.

An aged intermediate varnish sample from `The Girl with the Pearl Earring' by J. Vermeer was analysed by thermally assisted methylation with tetramethylammoniumhydroxide (TMAH) followed by gas chromatography in combination with mass spectrometry. Using library search of the available mass spectra, only three diterpenoid acids were identified. In addition to methyldehydroabietate, 7-oxo-dehydroabietic acid methyl ester and 7-oxo-15-hydroxy-dehydroabietic acid methyl ester were identified. There was reason to believe that the remaining unidentified compounds might be related to the oxidised derivatives of dehydroabietic acid and are formed upon methylation with TMAH. In order to investigate the behaviour of pure 7-oxo-dehydroabietic acid and 7-oxo-15-hydroxydehydroabietic acid under these specific derivatisation conditions, these compounds were synthesised and isolated. On derivatisation with TMAH, several methylation products, with varying degrees of methylation were formed from each of the pure compounds. Their structures were elucidated using specific fragment ions present in their mass spectra. Unusual products formed by methylation on a carbon atom under these conditions were observed. Based on these comparative studies, 7-oxo-dehydroabietic acid and 7-oxo-15-hydroxy-dehydroabietic acid were identified as the main components of the varnish.

This analytical pyrolysis study investigates the effects of ethanol extraction and of enzymatic protein removal from leaf material of two grass species, Poa annua and Poa pratensis . After pyrolysis the leaf fragments were analysed by low-voltage electron impact (E1) and by ammonia chemical ionization (NH 3 Cl). Pronase-treated and untreated material were compared using multivariate analysis of the PyMS data. The resulting discriminant function spectra among others show a kind of ‘negative image’ of the PyMS spectrum of the plant proteins originally present in Poa leaves. In addition to known protein marker fragment ions several hithertoo unrecognised ones were apparent as well. The nitrogen concentration was determined for several plant fractions. Quantitative comparison of relative intensities of masses found by discriminant analysis led to selection of an additional group of protein marker fragments (m/z 54, 70, 84, 107, 130, 209, 225 and 243 for E1 and 70, 72, 75, 84, 86, 89, 125, 131, 136, 146, 165, 201, 226, 229, 244, 262, 281 and 295 for NH 3 Cl) which showed a significant correlation ( r 2 > 0.5) with the total nitrogen content in the Poa leaves. The origin of grass protein marker fragments was discussed in comparison with reference spectra of two new plant cell-wall proteins, a synthetic polyamine and of albumin. Enzymatic digestion, in addition, yielded a better exposure of the plant cell-wall skeleton, and, therefore, also of the biomacromolecule lignin. Using previously obtained wet chemical data response values were calculated for marker fragments comparative to the analysed content of proteins, polysaccharides and different lignins.

Temperature-resolved in-source pyrolysis—mass spectrometry (Py—MS) and complementary Curie-point pyrolysis—gas chromatography/mass spectrometry (Py—GCIMS) were used to characterize a set of lignin polymers. No sample pretreatment was required and over 100 thermal degradation products were identified in the pyrolysates. Results were compared with whole biomass reference materials. Coniferyl alcohol and sinapyl alcohol, which were formed upon pyrolysis of lignin, could be selectively ionized by ammonia chemical ionization (Cl). Curie-point Py—GC with flame ionization detection was applied to classify the lignin polymers according to syringyl/guaiacyl ratios, the number of preserved phenylpropane structural units and functional group content. Multivariate analysis of the in-source Py—MS data was employed to classify the lignin polymers according to the origin and degree of modification. Combined results showed that enzyme-treated cottonwood-milled wood lignin is a relatively homogeneous polymer with a large fraction of preserved alkyl-aryl ether linkages. Technical lignins, isolated by the steam-explosion, organosolv and Kraft pulping processes, were found to be modified depending on the severity of the isolation procedure employed, and contained large amounts of β — 1-linked stilbene and β — β-linked resinol types of structures. Bagasse lignin shows the most condensed polymer structure with a large proportion of ether bonded p -coumaric acid and ferulic acid.

As a model for polysaccharide char formation in archaeological food residues, bulk samples of microcrystalline cellulose were charred under anoxic conditions at atmospheric pressure and at temperatures ranging from 190 to 310°C. The resulting chars were characterised by pyrolysis—mass spectrometry (Py—MS) and Curie-point pyrolysis—gas chromatography-mass spectrometry (Cu-Py—GC/MS). Although very few polysaccharide moieties were preserved in the experimental chars after heating above 270°C, a characteristic distribution of pyrolysis products representing the condensed char is shown. A selected charred archaeological food residue was analysed using the same analytical techniques. In this way, polysaccharide input can be detected in archaeological residues after approximately 1800 years of burial.

Primary pyrolysis mechanisms for native cellulose are more universally applicable, which is illustrated by the analyses of the characteristic pyrolysis product distributions of O-(2-hydroxyethyl)celluloses (HECs) measured by in-source pyrolysis-ammonia chemical ionisation mass spectrometry (Py-CIMS). The extent of the thermal degradation via transglycosidation, reverse aldolisation and Ei-elimination mechanisms in these cellulose derivatives is strongly influenced by the presence of substituents and by their substitution positions. The transglycosidation mechanism, which is the main thermal degradation pathway of cellulose, is blocked if the 6-O-positions in the cellulose backbone are substituted. The formation of a series of polyethylene glycolaldehydes and polyethylene glycols is explained by the reverse aldolisation and Ei-elimination mechanisms, respectively. The pyrolysis products of HECs have been identified by the determination of the number of hydroxyl groups in the mass peaks, which was done by the comparison of NH3 and ND3 Py-CI mass spectra. The correlation between the pyrolysis product distribution and substituents in HECs is illustrated by a multivariate analysis of the Py-CI mass spectra of 11 HECs, revealing structural information concerning the cellulose character and the average chain length of the substituents in the samples.

Attempts to forge amber inclusions have been known from ancient times, when natural materials such as melted amber or copal were used. At the very beginning of the Twentieth Century these natural materials were mostly replaced by Bakelite and similar synthetic polymers. These materials are not transparent and so can readily imitate certain kinds of milky amber but are useless in terms of transparent ambers with inclusions. In the 1940s two new classes of organic materials became available for amber imitation, i.e. polyester resins and epoxy resins. Both are transparent, able to dissolve organic dyes and can achieve an appropriate degree of hardness, depending on the addition of fillers (e.g. fiberglass), the structure of the original monomers, and the degree of crosslinking. With easy commercial availability and sophisticated “artwork” one can prepare convincing imitations of large transparent amber pieces with a wide variety of “inclusion” (ants, bees, lizards, mosquitos, etc.). Py-GC provides a simple test for the unmasking of such fakes, and Py-GC/MS can identify quite precisely the structures of the materials used for these purposes. A growing number of fossil forgeries produced in the Dominican Republic, and sold there and in New York, have been documented. The results reported are based largely on forgeries acquired by D. Grimaldi specifically for this study and kept at the American Museum of Natural History in New York; and from the Conservation Center's collection of forged artifacts and jewelry made available by New York galleries and private collectors.

Two milled wood lignins (MWLs) were isolated from tobacco leaves (laminae) and midrib material according to the Bjorkman procedure. Prior to MWL isolation, the material was extracted with cyclohexane/ethanol and hot water to remove extractives which could interfere with this isolation. MWLs were obtained with a yield of 0.03% (leaves) and 0.06% (midribs) based on dry pre-extracted material. FT-IR spectroscopy was used for quality control and classification of the isolates. Both preparations fulfilled the spectroscopic lignin criteria of a GS lignin with low syringyl content, although the isolates were not pure. Besides carbohydrate impurities, the lamina material, especially, was contaminated with nitrogen and fatty acid-containing substances. Proteins and cutin are discussed as probable contaminants. Both MWLs were characterized with quantitative pyrolysis—gas chromatography using a resistant heating instrument. Curie-point pyrolysis—gas chromatography/mass spectrometry and pyrolysis—mass spectrometry were used for additional characterization of yboth MWL samples. The results show that the tobacco MWL contains up to 22% carbohydrate-type impurities. Small amounts of alkaloids, proteins and lipids were also detected as contamination. Results of previous studies on mineral acid residues could be corroborated: tobacco lignin contains approximately 10% 4-hydroxyphenylpropane, 78–82% guaiacylpropane and 10–13% syringylpropane moieties.

The combination of laser desorption followed by multiphoton ionization (MPI) with time of flight mass spectrometry (TOF/MS) produces a versatile method of high sensitivity and selectivity. The first step is laser desorption in which molecules are desorbed from the sample intact or with little degradation. Laser ionization is employed as the second step to produce the ions and also induce fragmentation. Soft ionization, which yields primarily the parent ion, and hard ionization, which yields primarily the fragments, are both possible. Instrumental details are described and preliminary results from recent laser desorption MPI-TOF/MS experiments on a perbenzoylated sugar and on a porphyrin are presented. Examples from single photon ionization (SPI) experiments on two alkanes are also shown.

The addition of synthetic polymer standards to geological samples in order to provide quantitative data on pyrolysis product yields of the latter has been shown to be a valuable tool in pyrolysis-gas chromatography (Py-GC). The applicability of this approach for pyrolysis-mass spectrometry (Py-MS) is the focus of the present study. In general, good agreement was found in comparisons of yields of different compound classes determined by Py-MS with data from Py-GC and elemental analysis. Whilst no response factors were determined semi-quantitative data could readily be obtained for the compounds of interest. Examples are given of how different kerogen types can be distinguished not only on the basis of the relative abundance of different compounds, but also their yields. A practical example of the utility of this approach is provided in a study of compositional variations in petroleum asphaltenes. Here, differences were observed not only between petroleum asphaltenes generated from different source rocks, but also within a single oil reservoir (Ula Field, Norwegian North Sea). Whilst Py-MS suffers from several limitations (lack of information on carbon number distribution, compound-specific response factors, multiple contributions to nominal masses) the method has the advantage over Py-GC of being more rapid and providing data in a more readily analysable form. An example of the manner in which to overcome the problem of multiple contribution to nominal masses is shown through the application of higher resolution mass spectrometry.

The suitability of analytical pyrolysis techniques for the chemical characterisation of amorphous (charred) organic residues obtained from prehistoric pottery was studied. No chemical preparation of the sample was needed and the analysis was performed on very small samples (20–30 μg). Curie-point pyrolysis mass spectrometry and Curie-point pyrolysis gas chromatography were used for rapid characterisation of 33 residue-samples obtained from the outside (5) and inside (28) of different types of pottery from the native Roman settlement at Uitgeest-Groot Dorregeest (Netherlands). Discriminant analysis of the data shows several groups of residues of different chemical composition. Four representative residue samples were selected for more detailed analysis by Curie-point pyrolysis—gas chromatography/mass spectrometry. Many bio-organic moieties were detected in the residues situated on the inside of pots including fatty acids and characteristic markers for proteins and polysaccharides. Since no indications could be found to assume severe post-depositional changes in chemical composition, it is clear that the composition of the residues is a reflection of the original use in prehistoric times. The refractory nature of the charred (food) residues causes the preservation of the bio-organic characteristics. Other classes of compounds, such as polycyclic aromatic hydrocarbons, were detected in residues from the outside and were interpreted as smoke condensates from open cooking fires. A regular alkane/alkene pattern found in many residues and in the ceramic material of the pot sherd is interpreted as the pyrolysis product of an aliphatic network that has been formed from foodstuffs under high temperature conditions during cooking.

The condensate from Curie-point pyrolysis of cellulose has been examined as the permethylated derivative and as the borodeuteride-reduced permethylated derivative. The main complete sugar peaks were identified as levoglucosan and cellobiosan. Fragments were also identified which were derived from reverse aldolisation cleavage of a sugar ring to leave acetaldehyde fragments on the reducing and non-reducing ends of intact sugar units. No peaks of significant intensity could be identified as anhydrosugar units with attached ring cleavage fragments.

Macromolecular chlorolignins and lignosulphonates have been analysed in effluents from five pulp mills discharging on the river Rhine. Analytical techniques used were ultrafiltration, pyrolysis—mass spectrometry and pyrolysis—gas chromatography / mass spectrometry. The average organically bound chlorine content was 19 mg/l, with a maximum value of 68 mg/l for the pulp mill Cellulose de Strasburg. The annual discharge of organically bound chlorine (AOX) and of dissolved organic carbon (DOC) of these five mills was calculated to be 2100 tonnes AOX and 33 000 tonnes DOC, respectively. This would account for 90% of the AOX load and 10% of the DOC load of the river Rhine. Evidence for the existence of two chlorolignin classes, aromatic chlorolignins and aliphatic chlorolignins, was obtained. Monochloroguaiacol is the most abundant structural specific aromatic chlorolignin pyrolysis product. Methyl chloride is a major chlorine containing pyrolysis product, probably originating from aliphatic chlorolignins. Sulphur dioxide is a major pyrolysis product of lignosulphonates, and possibly can be used for the specific quantitative determination of lignosulphonates in river water. Radical recombination reactions of methyl radicals and other primary pyrolysis radicals play an important role in these samples, leading to several secondary pyrolysis products. The presence of lignosulphonates in the salt or acid form has an important effect on the pyrolysis process. Methyl chloride can probably be quantified with the reference compound tetramethylammonium chloride. The macromolecular compositions of these effluents show many similarities, but also significant differences, presumably reflecting the use of different wood types, pulping and bleaching conditions at the time of sampling.

The decay of Eucryphia cordifolia wood by white-rot and brown-rot fungi was studied with pyrolysis electron impact mass spectrometry (Py(EI)MS) and pyrolysis ammonia chemical ionization mass spectrometry (Py(CI)MS). Py(EI)MS spectra of the wood after decay by the white-rot fungus Ganoderma australe no longer show mass peaks indicative for lignin. The relative abundance of syringyl (dimethoxy) compounds decreases faster than the coniferyl (monomethoxy) compounds. In the same spectra the relative abundance of hexosans increases whereas the relative amount of pentosans remains constant. The presence of oligomeric sugars in the Py(CI)MS spectrum points to a preservation of some polysaccharides. Brown-rot fungal degradation of similar wood samples analysed with Py(EI)MS reveals no mass peaks for polysaccharides, resulting in a spectrum with mass peaks specific for lignin. The Py(EI)MS, revealing lignin dimers, and the Py(CI)MS spectra suggest that the lignin was not modified by the brown-rot fungus. However, the pyrolysis gas chromatography mass spectrometry (PyGCMS) data show an increase in oxygenated lignin pyrolysis products suggesting a molecular change in the lignin due to brown-rotting. In the brown-rotted wood one sugar pyrolysis product (levoglucosan) was observed by Py(CI)MS and PyGCMS suggesting that part of the cellulose polymer system is inaccessible to fungal degradation. Comparing data obtained from 13C cross polarization/magic angle spinning nuclear magnetic resonance (CP/MAS NMR) with data obtained from PyMS reveals that the latter method provides much more structural information.

An isolation procedure is presented for enrichment of lignin in lamina and midrib of tobacco by digestion with the enzymes pronase (protease) and cellulase (polysaccharidase). Digestion by cellulase was successful only after mechanical fracturing of the wall matrix, extensive removal of cytosolic and extraneous wall components, and after the chemical disclosure of the wall by pronase digestion. The purity of the enzyme digested residues and controls was checked by pyrolysis—mass spectrometry (Py-MS) and Curie-point pyrolysis—gas chromatography / mass spectrometry (CuPy-GC/MS). CuPy-GC/MS showed exclusively markers for lignin and carbohydrates in the pyrolysates of the enzyme digested tobacco samples. Lignin was clearly enriched in the enzyme digested residues. The relative distribution of monomeric and dimeric lignin pyrolysis products in the pyrolysis-mass spectra was comparable in lamina and midrib material. The lignin in both tobacco lamina and midrib material can be classified as a GS lignin with low amounts of coumaryl and syringyl units. Substantial amounts of carbohydrates were shown to be remaining in the residues. Cellulase digestion resulted in a reduction of 60% of the hexose markers in the pyrolysate of the lamina material and of 40% in the midrib material. The MS data on the lignin in enzyme digested cell walls of tobacco lamina material were compared with those from the Bjorkman lignin of the same tobacco sample. The Bjorkman tobacco lignin was a highly purified soluble lignin isolate contaminated with minor amounts of carbohydrates. CuPy-GC/MS indicated a higher abundance of syringyl units in the Bjorkman lignin compared to enzyme digested residues of the lamina material. This difference is presumably caused by a selective (Bjorkman) extraction of lignin from the secondary cell wall. The enzyme prepared lignin not only uses a milder preparation method but also gives a more complete view into the cell wall lignin of tobacco.

The thermal stability and the thermal degradation products of high impact polystyrene / decabromodiphenylether / antimony oxide (HIPS FR) have been studied in situ with derivative thermogravimetry (DTG), temperature resolved pyrolysis—mass spectrometry (Py-MS) and pyrolysis—gas chromatography / mass spectrometry (Py-GC /MS). With in-source temperature resolved Py-MS (negative ions) the thermal degradation processes of HIPS FR have been studied dynamically and antimony (oxy)bromides and brominated higher styrene oligomers upto n = 15 have been detected. During degradation of the HIPS FR polymer matrix several processes take place, such as debromination of the flame retardant decabromodiphenylether to form less brominated diphenylethers, bromination of polystyrene and formation of antimony bromides and antimony oxybromides. The formation of toxic polybrominated dibenzofurans (PBDFs) has been shown to occur in the temperature range in which the HIPS FR polymer matrix degrades (350–400°C). This is explained by debromination of decabromodiphenylether to form less brominated diphenylethers which are much more reactive towards formation of PBDFs.

Minute samples of xylem from healthy and Fusarium oxysporum f.sp. dianthi -infected stems of carnation cultivars differing in degree of resistance to F. oxysporum were investigated by in-source pyrolysis low voltage EI mass spectrometry and by pyrolysis under ammonia CI conditions. Healthy xylem of both cultivars showed PY-MS spectra with markers for cellulose, hemicelluloses, phenolic acids, and a mixed guaiacyl-syringyl lignin. At different stages of infection visibly infected, brown coloured, xylem was separated under the microscope from healthy-looking parts. The PY-MS spectra of the diseased xylem of susceptible ‘Lena’ were entirely different from those of healthy(-looking) areas, showing a drastic change in lignin pyrolysis product distributions. In most cases practically all syringyl constituents, both mono- and dimeric, had disappeared. In some cases also an overall decrease in relative intensity of the lignin pyrolysis products was observed. Invasion of the fungus thus induced a major change of cell wall composition as a result of lignin degradation along different pathways: demethoxylation of syringyl units, the major pathway in ‘Lena’, and propenyl side chain oxidation. The PY-MS spectra of infected ‘Novada’ generally showed the overall oxidation of lignin and less demethoxylation. The ‘Novada’ spectrum also showed m/z 239, 287, and an increase of 257 and 271, which are molecular ions indicative of the phytoalexins dianthalexin and the dianthramides methoxydianthramide S, hydroxydianthramide B and methoxydianthramide B. The compounds appeared comparatively volatile and therefore could be desorbed as a thermal extract before other pyrolysis fragments appear. The compounds were practically absent in adjacent, healthy looking, xylem. Low voltage EI MS of methanol extracts of about 2 mm 3 of infected ‘Novada’ xylem confirmed these data and in a single case also showed evidence for hydroxyanthranilic acid ( m/z 153) which is attributed to degradation of dianthramine. Spectra of thermal and methanolic extracts were quite similar.

Beech wood samples (stem, root and biodegraded wood) as well as beech milled wood lignin, beech polysaccharide fractions (holocellulose, xylan and technical xylan) and microcrystalline cellulose were analysed by pyrolysis-mass spectrometry (Py-MS), i.e. Curie-point pyrolysis low voltage electron impact (EI) MS, as well as by in-ion source platinum filament pyrolysis MS under both ammonia chemical ionisation (CI) and low voltage EI conditions. The mass spectra reveal the presence of monomeric, submonomeric as well as di- and oligomeric pyrolysis products from polysaccharides and/or lignin. The difference in pyrolytic behavior of polysaccharides in the complex wood matrix and of the purified polysaccharides may find its cause in chemical and physical interactions between polysaccharides and lignins in the wood, or may be due to the presence of considerable amounts of inorganic cations in the sample. The pyrolytic behavior of lignin is not significantly influenced by the presence of inorganic cations. The holocellulose and xylan data also exhibit chlorinated lignin marker compounds, which point to the presence of partly chlorinated lignin in the sample. This is probably an artefact due to the isolation procedure of these fractions, which involves treatment with sodium hypochlorite solution. The CI data of the technical xylan fraction indicate the presence of di- and oligomeric pentoses, hexoses and pentose/hexose sugars. These are probably formed during the isolation procedure (which involves extraction of wood chips at 170°C) by partial depolymerisation of the polysaccharide moiety.

Microcrystalline cellulose was analysed by platinum filament pyrolysis desorption chemical ionisation mass spectrometry (DCI-MS) under ammonia chemical ionisation conditions. This polysaccharide was also pyrolysed on a preparative scale as well as by off-line Curie-point pyrolysis and subsequently analysed by ammonia DCI-MS. The DCI-MS data of the native polysaccharide show ion series corresponding to ammonium adduct ions of anhydrocello-oligosaccharides up to anhydrocello-hexaose. In the preparative-scale and the off-line Curie-point pyrolysates up to the pentamer could be detected. Related ion series corresponding to the ammonia adducts of ring fragments left on the saccharide chain were also seen. It is uncertain whether these compounds are formed during pyrolysis directly from the native polymer or arise from the corresponding anhydro-oligosaccharides during the ionisation process. The three pyrolysis methods tested yield comparable results with respect to the pyrolysis products formed, which indicates that the mechanism of formation of anhydrocello-oligosaccharides from cellulose is a generally significant process. Curie-point pyrolysis-gas chromatography data should be interpreted with caution since the oligomeric part of the pyrolysate is not monitored by this method.

Six laboratories currently using pyrolysis-mass spectrometry (Py-MS) on a routine basis have participated in a collaborative exercise to assess the degree of qualitative reproducibility which can be achieved by this technique. Each laboratory has analysed six samples chosen to reflect as far as possible the diversity of samples currently examined by Py-MS. The participants analysed the samples under the conditions they felt appropriate to yield mass programs at both 70 eV and at low (⩽ 20 eV) electron energy. The results have shown that there is a measure of agreement between all laboratories for synthetic polymers but a divergence in the results is apparent as the samples become more complex. It would appear that the nature of the pyrolysis system rather than the mass spectrometer is the major factor affecting reproducibility. Undoubtedly, the problems associated with reproducing low electron energies from instrument to instrument also play a significant role.

Pyrolysis-gas chromatography-mass spectrometry of polyamino acids, proteins and complex proteinaceous material was used to study the flash pyrolytic behaviour of aliphatic amino acid moieties. Two types of highly specific compounds, tentatively identified as 3,5-alkyl-3,4-dihydro-2H-pyrrole-2,4-diones and 3-alkenyl-5-alkyl-pyrrolidine-2,4-diones on the basis of gas chromatographic retention times and high resolution mass spectrometric data, contain sequence information on pairs of aliphatic amino acid moieties (valine, leucine, isoleucine, alanine). The results can be applied to screen the presence or absence of adjacent aliphatic amino acid moieties in complex mixtures containing proteins or in purified proteins. The method is rapid, no treatment of the sample is required and very small amounts of sample are sufficient.

A section from an oligotrophic peat and handpicked subfossil plant fragments from this peat were studied by pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS), an excellent method for the chemical characterisation of very small amounts of plant material. Leaves of peat mosses (Sphagnum section Acutifolia) and rootlets of heathers (Ericaceae), both important contributors to this peat layer, were obtained by micromanipulation. The Py-GC-MS data consist of evaporation products of adsorbed compounds and a large number of pyrolysis products from remnants of the plant macromolecular systems. The Sphagnum pyrolysate consists mainly of compounds derived from polysaccharides, including anhydro sugars from xylose, mannose, galactose and glucose. Only a few phenolic compounds were observed in the peat moss sample. The peat as well as the rootlets of Ericaceae yielded Py-GC-MS traces with many aliphatic and alicyclic compounds and a large number of phenolic pyrolysis products derived from lignin. The suberin of the rootlets of Ericaceae is postulated to be an important source of aliphatic hydrocarbons, methylketones (C17-C31), unsaturated methylketones (C19-C29), 2,4-diketones (C25-C31), long-chain alcohols (C15-C25), fatty acids (C8-C24), 24-ethylcholesterol, 24-ethylcholestanol, 24-ethyl-cholestanone and a number of triterpenoids with friedooleanane skeletons (e.g. taraxerol and taraxerone). The presence in suberin of an insoluble aliphatic polymer, which is characterised by pyrolysate with alkane/alkene doublets ranging from C6-C33, is postulated.

Two extraction procedures to obtain a lignin fraction from tobacco were compared with Curie-point pyrolysis-mass spectrometry and Curie-point pyrolysis-gas chromatography/mass spectrometry. Both fractionation methods resulted in a lignin-enriched residue on which residual non-lignin polymers were attached. The Klason lignin procedure, which involves treatment with 72% sulphuric acid, revealed deficiencies towards the removal of alkaloids, carbohydrates, proteins, and lipids. As a consequence, the weight data on several chemical compound classes, i.e. the protein and lignin content of the Klason lignin residues, are estimated incorrectly. The Willstatter procedure, which involves treatment with hydrochloric acid, resulted in a selective removal of nitrogen-containing compounds, but affected the carbohydrate moiety only partially. The two extraction procedures investigated are not suitable to obtain a pure lignin fraction from tobacco. However, both purification methods resulted in a comparable relative distribution of lignin pyrolysis products in the lamina material of the flue-cured and burley type tobacco. The difference in distribution of lignin markers in both ‘lignin’ preparations of lamina and midrib material of the flue-cured type tobacco suggests a difference in chemical composition in the plant.

In a previous study, selective degradation of carbohydrates versus lignin in buried woods was demonstrated by non-destructive spectroscopy (13C NMR) and destructive wet chemical methods (acid hydrolysis and alkaline CuO oxidation). In this paper, spruce, alder and oak woods deposited in coastal sediments and their recent equivalents are characterized by pyrolysis-mass spectrometry and pyrolysis-gas chromatography-mass spectrometry. Pyrolysis data, wet chemical degradation and NMR data are in good agreement, indicating that flash pyrolysis methods are useful for fast detailed characterization of this type of organic matter.

Six coal samples, intended to be representative of the entire range of coal ranks and including two thermoplastic coals, were analyzed by Curie-point pyrolysis-gas chromatography to characterize parameters related to coal reactivity such as coal rank, thermoplasticity, and the addition of a catalyst to the coal. It is shown that the number and nature of the pyrolysis products, particularly the degree of oxygen functionality of the one-ring aromatic compounds formed, are characteristics of the rank of a coal sample. The presence of a series of long-chain n -alkanes, which are believed to be part of the so-called mobile phase of bituminous coals, is indicative of a thermoplastic coal. Low-temperature pre-oxidation is an effective method of destroying the thermoplastic properties of bituminous coals. The oxidative treatment affects the volatilization of the aliphatic moieties of the thermoplastic coals and significantly reduces the number and yield of pyrolysis products. The evolution of long-chain n -alkanes is almost completely eliminated upon pre-oxidation. Addition of potassium carbonate, which acts as a catalyst in coal gasification and combustion, results in an interaction of the alkali-metal ion with the carboxyl and hydroxyl groups in the coal. Because of this interaction the yield of phenolic compounds upon Curie-point pyrolysis is reduced. On the other hand, the formation of phenolic compounds, originating from ether linkages in the coal matrix, is not influenced by the addition of potassium carbonate.

A Curie-point evaporation/pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS) instrument is described in which a fused-silica capillary column is coupled directly both to the pyrolysis unit and to the ion source of a photoionization mass spectrometer. Photoionization by means of a rare gas resonance lamp was used as the low-energy ionization technique. The performance of the instrument was tested with a Grob test mixture at photoionization energies of 11.8 eV (Ar I) and 10.6 eV (Kr I). Photoionization mass spectra of 1-octanol, methyl octanoate, dimethylaniline and n-dodecane obtained at these energies are discussed. Py-GC-photoionization-mass spectrometry (PIMS) and PIMS data for the (1 → 4)-α-glucan amylose are presented and compared with electron impact (EI) Py-GC-MS and Py-MS results. The cumulative Ar I photoionization mass spectrum of the GC-MS data on amylose closely resembles the 14-eV EI spectrum of amylose. The method is a promising alternative to the existing low-voltage EI ionization mass spectrometry for the pyrolysis analysis of macromolecular materials.

A case study is presented of two uncommon lignite samples, PSOC-975 and PSOC-427, which were studied by microscopic and analytical pyrolysis methods. Pyrolysis-mass spectrometric (Py-MS), pyrolysis-gas chromatographic (Py-GC) and Py-GC-MS data demonstrate that lignite PSOC-975 contains a high abundance of resinous material of some sort, whereas PSOC-427 is characterized by two major unknown pyrolysis products, prist-1-ene and alkylphenols. The chemical data contradict the microscopic observations, which point to a maceral composition of woody and cortical tissues, although a considerable part is unrecognizable. Apparent discrepancies between low-voltage Py-MS and 80-eV Py-GC-MS results were evaluated by a study of standards. It was shown that the aliphatic hydrocarbons are highly underestimated in low-voltage Py-MS. The nature of the disagreement observed between maceral composition and the pyrolysis data needs further study.

A series of Hungarian Miocene brown coals were analysed by pyrolysis-low-energy electron impact mass spectrometry and discriminant analysis. Correlation patterns in the mass peaks of these brown coals coincided with the geothermal history determined from the reflectivity of the macerals. Coal rank and the first discriminant function had a correlation coefficient of 0.96. Discriminant analysis also classified the coals according to their depositional environment. The mass peaks were deconvoluted using pyrolysis-gas chromatography-mass spectrometry data obtained under low-energy photoionization conditions. Several apparently homologous series of mass peaks in the pyrolysis-low-energy electron impact mass spectra were found to represent entirely different types of pyrolysis products. The low-energy fragmentation patterns of isoprenoid and alicyclic hydrocarbons are substantially different from the usual 70 eV electron impact spectra and may cause erroneous interpretation of non-deconvoluted pyrolysis-low-energy electron impact mass spectrometric data. The deconvoluted pyrolysis-low-energy electron impact mass spectrometric data resulted in a more detailed reconstruction of the original environments of deposition. Marine influences were deduced from several types of organic sulphur compounds. Markers for lignified wood, leaf and root remains and resins were present in the pyrolysates.

Cuticles of Agave americana contain a substantial amount of an insoluble, non-hydrolyzable macromolecular substance of a highly aliphatic nature. Based on Fourier transform infrared (FTIR) and 13 C cross-polarization magic angle spinning NMR spectroscopy it was established that the isolated macromolecular substance consists of a polysaccharide and polymethylene moiety in an estimated ratio of 60:40 by weight. After treatment with acid this ratio changed to 37: 63. The isolates obtained before and after the acid treatment were subjected to Curie-point pyrolysis-gas chromatography with flame ionization detection, Curie-point pyrolysis-gas chromatography/mass spectrometry, Off-line pyrolysis in combination with gas chromatography and gas chromatography/mass spectrometry, desorption chemical ionisation mass spectrometry, time/temperature-resolved pyrolysis-field ionization mass spectrometry and heating experiments in a closed system in order to elucidate the molecular structure. Each of the methods applied has its own specifications with respect to pyrolysis conditions and response factors of different classes of pyrolysis products. The data obtained are highly complementary and, in combination with FTIR and 13 C NMR, several new features of the insoluble, non-hydrolyzable aliphatic biopolymer were revealed.

Microcrystalline cellulose was analysed by Curie-point pyrolysis-capillary gas chromatography-mass spectrometry (Py-GC-MS) under electron impact (70 eV) as well as chemical ionisation (isobutane) conditions. Almost all interpretations could be supported by high-resolution mass spectrometric data, obtained by Py-GC-high-resolution MS. The pyrolysis products identified can be classified into several main groups i.e. carbonyl compounds, acids and methyl esters, furans, pyrans, anhydrosugars and hydrocarbons. Altogether 96 compounds have been found, and structural assignments could be given to 94 of them. From these compounds 25 have not been identified earlier in pyrolysates of polysaccharides. Levoglucosan is the major GC amenable pyrolysis product detected. Several formation pathways are proposed to account for the formation of a number of pyrolysis products from levoglucosan and/or cellobiosan. An overview showing the thermal degradation pattern of cellulose through anhydrosaccharides to smaller pyrolysis products is also presented.

Curie-point pyrolysis is applied to the characterization of the carbonaceous material deposited on used hydrotreating catalysts. Basically, the carbonaceous material can be divided in an extractable and non-extractable fraction. The extractable fraction consists of adsorbed molecules from the feed, mainly alkanes and small aromatic compounds. The non-extractable fraction is subdivided into strongly adsorbed feed molecules, a carbon fraction resistant to pyrolysis at 1043 K and a sulphur-containing fraction. The division between the extractable and non-extractable fraction is dependent on the extraction agent. Dichloromethane appears to produce a larger extractable fraction than benzene or hexane.

Total beech wood and three derived fractions (holocellulose, acetylglucuronoxylan and technical acetylglucuronoxylan) were analysed by Curie-point pyrolysis-gas chromatography-mass spectrometry (Py-GC-MS). The pyrolysis products can be divided into several main groups such as non-specific compounds, carbohydrate markers, lignin markers and chlorinated methoxyphenols. The latter group was only detected in the holocellulose fraction and in the (acetylglucurono)xylan fraction. Specific carbohydrate marker compounds detected by Py-GC-MS are anhydrohexoses and -pentoses as well as furan and pyran derivatives. The lignin marker compounds include a variety of mono- and dimethoxyphenols. The chlorinated methoxyphenols are released from chlorolignin residues present in the holocellulose and xylan fractions. The chlorination must have occurred during the delignification process with sodium chlorite. No chlorinated compounds were detected among the pyrolysis products from total beech wood nor among the technical (acetylglucurono)xylan pyrolysis products. The latter fraction was isolated from beech wood using a steam extraction process.