Research Article Received: 20 September 2013
Revised: 14 January 2014
Accepted: 15 January 2014
Published online in Wiley Online Library
Rapid Commun. Mass Spectrom. 2014, 28, 845–854 (wileyonlinelibrary.com) DOI: 10.1002/rcm.6843
A framework for the extraction and interpretation of organic molecules in speleothem carbonate Peter M. Wynn1* and Jochen J. Brocks2 1 2
Lancaster Environment Centre, University of Lancaster, Lancaster LA1 4YQ, UK Research School of Earth Sciences, The Australian National University, Canberra, A.C.T. 0200 Australia
RATIONALE: The organic content of speleothem calcite is a well-recognized component of their chemical composition. To date, the techniques for interpretation of this material include UV fluorescence, FTIR spectroscopy and biomarker analysis using gas chromatography/mass spectroscopy (GC/MS). However, investigation of the minute concentrations of molecules in speleothems demands careful sampling and laboratory controls. METHODS: To be certain extracted molecules were encapsulated at the time of speleothem growth and do not represent contamination, we submitted three pieces of speleothem calcite to a rigorous extraction procedure. Based on sequential digestion and analysis by GC/MS, we measured concentration profiles of individual compounds with increasing distance from sample surfaces. RESULTS: Declining concentrations toward interior extracts identified cholesterol, phthalates, and n-alkanes as surface contaminants. In contrast, iodo organic compounds had homogeneous concentration profiles and were also significantly above laboratory background levels, consistent with an indigenous origin. However, further laboratory testing demonstrated that iodo organics were produced by the reaction of iodine derived from the speleothem with solvent additives and other impurities of the extraction procedure. Sitosterol and some fatty acids demonstrated distributions which were probably indigenous to the speleothem archive, thus recording environmental conditions commensurate with time of growth. CONCLUSIONS: We do not aim to provide an environmental interpretation of extracted molecules, but highlight the caution necessary before doing so. We ultimately establish a framework for differentiating between organic constituents that are introduced to the speleothems during storage, handling and as artifacts of extraction, and those encapsulated in situ at the time of growth. Copyright © 2014 John Wiley & Sons, Ltd.
Biomarkers extracted from organic-rich sediments and sedimentary rocks have proved invaluable for environmental studies.[1] Yet other archives with a much lower molecular content still require development of more sensitive techniques and lower contamination background levels before their archived information can be used with confidence and at high resolution. Speleothems now form some of the most informative and exciting archives available to the palaeo-environmental community.[2–5] Due to accurate dating by U-series techniques and the well-developed science behind analyzing and interpreting the isotopes and trace elements encapsulated within the speleothem archive, they represent excellent potential for the extraction and interpretation of new molecular proxies. Organic matter is transported into a cave environment predominantly via four main pathways; transport by air, drip water and fauna, or autochthonous production.[6] The flux of organic matter along drip water flow pathways is largely controlled by the physical and chemical characteristics of the overlying soil, karst and hydrological flow routes.[7,8]
Rapid Commun. Mass Spectrom. 2014, 28, 845–854
Copyright © 2014 John Wiley & Sons, Ltd.
845
* Correspondence to: P. M. Wynn, Lancaster Environment Centre, University of Lancaster, Lancaster LA1 4YQ, UK. E-mail: [email protected]
Autochthonous production is enabled through cave microbiological communities,[9] and air/fauna transport is capable of moving pollen/spores into the cave system where particulates and volatile organic components may adhere to speleothem deposits.[10,11] Where hydrological flow routes are the principle agents of organic mobilization and transport, incorporation into the stalagmite record is frequently manifest as annual fluorescent laminae,[12,13] and can also be evidenced by trace element distributions associated with organic colloids.[14,15] Techniques including micro-Fourier transform infrared (FTIR) have been used to reveal the presence of principle bonding structures at high resolution,[16] and X-ray absorption near-edge spectroscopy (XANES) enables understanding of mechanisms of organic matter incorporation into speleothem calcite.[17] The characterization of molecules extracted from speleothems is helping to further unravel the various sources of organic matter, yielding a plethora of new palaeo-environmental information. Several pioneering studies have already attempted to use organic compounds extracted from stalagmites as biomarkers of environmental processes. Such compounds have included lipids (alkanes, sterols, alkanols, and fatty acids (FA)), as indicators of vegetation change and microbiological activity,[6,9,18–21] methyl-alkanoates as derivatives of FA under conditions of high pH,[22] phenols
P. M. Wynn and J. J. Brocks associated with lignin precursors,[23] and PAHs (polycyclic aromatic hydrocarbons) as indicators of combustion processes.[24] However, there remains a need to identify organic compounds which are indigenous to the speleothem calcite, as opposed to those which are introduced through laboratory procedures or post-depositional environmental contamination. Based on techniques developed for sedimentary rocks,[25–27] we present a methodology that enables recognition of contaminants which have invaded speleothem calcite. In the simplest cases, recent contaminants should be entirely surficial, while syngenetic hydrocarbons should be homogeneously distributed throughout the rock. However, contaminants may also diffuse into fissures or pore spaces leaving a concentration gradient with decreasing abundance from the surface towards the centre of speleothem calcite.
EXPERIMENTAL Speleothem description and extraction procedure The three speleothems are denoted as BFM-J (a stalagmite from the UK Midlands);[28] SCFS-2 (part of a flow stone core collected from Victoria Fossil Cave (Spring Chamber), South Australia); and a fossil black coloured speleothem from the Nullabor Plain, Australia (Null-1). No further environmental information is provided here as the aim is not to produce a palaeo-interpretation of extracted molecules, but to build and test an extraction protocol to identify molecules which were incorporated at the time of speleothem growth. Each of the speleothems was sectioned using a diamond-tipped cutting wheel pre-cleaned in dichloromethane (DCM) and methanol to produce an obloid shaped calcite block approx 1 cm × 1 cm × 5 cm extracted from close to the central portion of the speleothem to ensure similar growth conditions throughout and avoid variable laminae thickness towards the edge of the sample. Digestion of obloids in 1 M HCl (volume 20 mL, pre-cleaned by 3 × extraction with DCM) allowed the surface coating of calcite to be etched away. This process was repeated up to three times, thus removing three external layers of calcite. Weighing the speleothem obloids between each surface etch yielded the mass of calcite removed (Table 1). The remaining calcite from the interior of each obloid was crushed in an alumina ring mill and extracted using a Dionex accelerated solvent extractor (ASE) with DCM as solvent at 100 °C and 1000 psi. The extracted rock powder from the internal speleothem obloid was then digested in 1 M HCl to release the ’bound’ fraction following Table 1. Mass of calcite material removed during each extract by acid digestion Mass of calcite material digested in each extract (g) Speleothem
846
SCFS-2 BFM-J Null-1
1st extract
2nd extract
3rd extract
Interior extract
1.5 2.5 1.44
3.2 1.8 1.6
2.6 1.3 1.81
10.6 8.1 7.8
wileyonlinelibrary.com/journal/rcm
Blyth et al.[29] As the compositions of the interior ASE extract and the acid digest were similar, they were summed to produce a bulk concentration. Organics released into solution from acid carbonate digestion were removed from acidic solution using a separating funnel and addition of three successive aliquots of DCM. The DCM extracts were bulked together and back-extracted three times with purified water. Extracts were concentrated to 100 μL under a stream of purified nitrogen gas and derivatized using N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA). Standard addition of stearic acid methyl ester enabled quantification by gas chromatography/ mass spectroscopy (GC/MS). Sand baked at 600 °C was crushed in the alumina ring mill three times for 60 s before crushing the speleothem sample. The final aliquot was retained for blank analysis. Crushed sand was subjected to ASE and acid digestion, following procedures outlined above to obtain a comprehensive laboratory blank. The laboratory background concentration of compounds is plotted in all figures as a dotted line. As the laboratory background level is largely independent of the mass of sand that was extracted to obtain the blank, concentrations were computed relative to the mass of the speleothem interior extract. Gas chromatography/mass spectrometry Derivatized extracts were analyzed by GC/MS in full-scan mode at a mass resolution of 1000 on a Micromass AutoSpec Premier equipped with an Agilent 6890 gas chromatograph and a DB-5MS capillary column (60 m × 0.25 mm i.d., 0.25 μm film thickness) using helium as the carrier gas. The source was operated at 260 °C in electron ionization (EI) mode at 70 eV ionization energy and with 8000 V acceleration voltage. Samples were injected in splitless mode at a constant temperature of 300 °C. For full-scan experiments, the GC oven was programmed at 40 °C (2 min), heated to 315 °C at 4 °C/min and a total run time of 90 min. The AutoSpec fullscan duration was 0.7 s plus 0.2 s interscan delay over a mass range of 55 to 600 Da. Based on repeat injection experiments, error quantification (% standard deviation (SD)) for lipid quantification is calculated as 4–5%.
RESULTS AND DISCUSSION Obloids cut from three different speleothems were sequentially digested in HCl. Each digest solution was extracted with solvent, and the molecular content quantified (specifically lipid extracts of n-alkanes, n-alkanoic acids, and sterols; phthalates; and halogenated iodine compounds) by GC/MS. Each extract does not represent any degree of age control, but a time-independent analysis of molecular content from outer to inner extract. The surface extract of all three speleothem samples contained a range of compounds including lipid fractions of n-alkanes, n-alkanoic acids (silylated) and sterols. Figures 1 and 2 show the distribution of key compounds from sequential extracts of speleothem SC-FS-2 and the associated blank. Across all three speleothem samples, detectable n-alkanes range from n-C21 to n-C36. The total sum of all n-alkanes in each speleothem sample was 0.2, 1.9 and 0.02 μg g–1 for SC-FS-2, Null-1 and BFM-J, respectively.
Copyright © 2014 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2014, 28, 845–854
Extracting and interpreting organic molecules in speleothems
Figure 1. m/z 85 partial ion chromatograms of the sequential digestion experiment of speleothem SCFS-2 showing alkanes (n-alkanes are identified by ’•’ or carbon number; MMA = midchain branched monomethyl alkanes, UCM = unresolved complex mixture; ’ ~ ’ are truncated signals of FAs). The y-axis gives approximate concentrations per gram of digested speleothem. The axis is calibrated to n-C28 and has not been corrected for differences in response factors. (A) Comprehensive system blank, (B) 2nd digest, and (C) 1st digest (contaminated exterior of the samples). See Fig. 2 for digestion schematic. Straight-chain FA ranged from C8 to C24 with highest concentrations reached in outer extracts (maximum concentration 20 μg g–1 for hexadecanoic acid in speleothem Null-1). The only identified sterols were cholesterol and sitosterol. Although sitosterol was only observed consistently in extracts from speleothem BFM-J and at concentrations
Revised: 14 January 2014
Accepted: 15 January 2014
Published online in Wiley Online Library
Rapid Commun. Mass Spectrom. 2014, 28, 845–854 (wileyonlinelibrary.com) DOI: 10.1002/rcm.6843
A framework for the extraction and interpretation of organic molecules in speleothem carbonate Peter M. Wynn1* and Jochen J. Brocks2 1 2
Lancaster Environment Centre, University of Lancaster, Lancaster LA1 4YQ, UK Research School of Earth Sciences, The Australian National University, Canberra, A.C.T. 0200 Australia
RATIONALE: The organic content of speleothem calcite is a well-recognized component of their chemical composition. To date, the techniques for interpretation of this material include UV fluorescence, FTIR spectroscopy and biomarker analysis using gas chromatography/mass spectroscopy (GC/MS). However, investigation of the minute concentrations of molecules in speleothems demands careful sampling and laboratory controls. METHODS: To be certain extracted molecules were encapsulated at the time of speleothem growth and do not represent contamination, we submitted three pieces of speleothem calcite to a rigorous extraction procedure. Based on sequential digestion and analysis by GC/MS, we measured concentration profiles of individual compounds with increasing distance from sample surfaces. RESULTS: Declining concentrations toward interior extracts identified cholesterol, phthalates, and n-alkanes as surface contaminants. In contrast, iodo organic compounds had homogeneous concentration profiles and were also significantly above laboratory background levels, consistent with an indigenous origin. However, further laboratory testing demonstrated that iodo organics were produced by the reaction of iodine derived from the speleothem with solvent additives and other impurities of the extraction procedure. Sitosterol and some fatty acids demonstrated distributions which were probably indigenous to the speleothem archive, thus recording environmental conditions commensurate with time of growth. CONCLUSIONS: We do not aim to provide an environmental interpretation of extracted molecules, but highlight the caution necessary before doing so. We ultimately establish a framework for differentiating between organic constituents that are introduced to the speleothems during storage, handling and as artifacts of extraction, and those encapsulated in situ at the time of growth. Copyright © 2014 John Wiley & Sons, Ltd.
Biomarkers extracted from organic-rich sediments and sedimentary rocks have proved invaluable for environmental studies.[1] Yet other archives with a much lower molecular content still require development of more sensitive techniques and lower contamination background levels before their archived information can be used with confidence and at high resolution. Speleothems now form some of the most informative and exciting archives available to the palaeo-environmental community.[2–5] Due to accurate dating by U-series techniques and the well-developed science behind analyzing and interpreting the isotopes and trace elements encapsulated within the speleothem archive, they represent excellent potential for the extraction and interpretation of new molecular proxies. Organic matter is transported into a cave environment predominantly via four main pathways; transport by air, drip water and fauna, or autochthonous production.[6] The flux of organic matter along drip water flow pathways is largely controlled by the physical and chemical characteristics of the overlying soil, karst and hydrological flow routes.[7,8]
Rapid Commun. Mass Spectrom. 2014, 28, 845–854
Copyright © 2014 John Wiley & Sons, Ltd.
845
* Correspondence to: P. M. Wynn, Lancaster Environment Centre, University of Lancaster, Lancaster LA1 4YQ, UK. E-mail: [email protected]
Autochthonous production is enabled through cave microbiological communities,[9] and air/fauna transport is capable of moving pollen/spores into the cave system where particulates and volatile organic components may adhere to speleothem deposits.[10,11] Where hydrological flow routes are the principle agents of organic mobilization and transport, incorporation into the stalagmite record is frequently manifest as annual fluorescent laminae,[12,13] and can also be evidenced by trace element distributions associated with organic colloids.[14,15] Techniques including micro-Fourier transform infrared (FTIR) have been used to reveal the presence of principle bonding structures at high resolution,[16] and X-ray absorption near-edge spectroscopy (XANES) enables understanding of mechanisms of organic matter incorporation into speleothem calcite.[17] The characterization of molecules extracted from speleothems is helping to further unravel the various sources of organic matter, yielding a plethora of new palaeo-environmental information. Several pioneering studies have already attempted to use organic compounds extracted from stalagmites as biomarkers of environmental processes. Such compounds have included lipids (alkanes, sterols, alkanols, and fatty acids (FA)), as indicators of vegetation change and microbiological activity,[6,9,18–21] methyl-alkanoates as derivatives of FA under conditions of high pH,[22] phenols
P. M. Wynn and J. J. Brocks associated with lignin precursors,[23] and PAHs (polycyclic aromatic hydrocarbons) as indicators of combustion processes.[24] However, there remains a need to identify organic compounds which are indigenous to the speleothem calcite, as opposed to those which are introduced through laboratory procedures or post-depositional environmental contamination. Based on techniques developed for sedimentary rocks,[25–27] we present a methodology that enables recognition of contaminants which have invaded speleothem calcite. In the simplest cases, recent contaminants should be entirely surficial, while syngenetic hydrocarbons should be homogeneously distributed throughout the rock. However, contaminants may also diffuse into fissures or pore spaces leaving a concentration gradient with decreasing abundance from the surface towards the centre of speleothem calcite.
EXPERIMENTAL Speleothem description and extraction procedure The three speleothems are denoted as BFM-J (a stalagmite from the UK Midlands);[28] SCFS-2 (part of a flow stone core collected from Victoria Fossil Cave (Spring Chamber), South Australia); and a fossil black coloured speleothem from the Nullabor Plain, Australia (Null-1). No further environmental information is provided here as the aim is not to produce a palaeo-interpretation of extracted molecules, but to build and test an extraction protocol to identify molecules which were incorporated at the time of speleothem growth. Each of the speleothems was sectioned using a diamond-tipped cutting wheel pre-cleaned in dichloromethane (DCM) and methanol to produce an obloid shaped calcite block approx 1 cm × 1 cm × 5 cm extracted from close to the central portion of the speleothem to ensure similar growth conditions throughout and avoid variable laminae thickness towards the edge of the sample. Digestion of obloids in 1 M HCl (volume 20 mL, pre-cleaned by 3 × extraction with DCM) allowed the surface coating of calcite to be etched away. This process was repeated up to three times, thus removing three external layers of calcite. Weighing the speleothem obloids between each surface etch yielded the mass of calcite removed (Table 1). The remaining calcite from the interior of each obloid was crushed in an alumina ring mill and extracted using a Dionex accelerated solvent extractor (ASE) with DCM as solvent at 100 °C and 1000 psi. The extracted rock powder from the internal speleothem obloid was then digested in 1 M HCl to release the ’bound’ fraction following Table 1. Mass of calcite material removed during each extract by acid digestion Mass of calcite material digested in each extract (g) Speleothem
846
SCFS-2 BFM-J Null-1
1st extract
2nd extract
3rd extract
Interior extract
1.5 2.5 1.44
3.2 1.8 1.6
2.6 1.3 1.81
10.6 8.1 7.8
wileyonlinelibrary.com/journal/rcm
Blyth et al.[29] As the compositions of the interior ASE extract and the acid digest were similar, they were summed to produce a bulk concentration. Organics released into solution from acid carbonate digestion were removed from acidic solution using a separating funnel and addition of three successive aliquots of DCM. The DCM extracts were bulked together and back-extracted three times with purified water. Extracts were concentrated to 100 μL under a stream of purified nitrogen gas and derivatized using N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA). Standard addition of stearic acid methyl ester enabled quantification by gas chromatography/ mass spectroscopy (GC/MS). Sand baked at 600 °C was crushed in the alumina ring mill three times for 60 s before crushing the speleothem sample. The final aliquot was retained for blank analysis. Crushed sand was subjected to ASE and acid digestion, following procedures outlined above to obtain a comprehensive laboratory blank. The laboratory background concentration of compounds is plotted in all figures as a dotted line. As the laboratory background level is largely independent of the mass of sand that was extracted to obtain the blank, concentrations were computed relative to the mass of the speleothem interior extract. Gas chromatography/mass spectrometry Derivatized extracts were analyzed by GC/MS in full-scan mode at a mass resolution of 1000 on a Micromass AutoSpec Premier equipped with an Agilent 6890 gas chromatograph and a DB-5MS capillary column (60 m × 0.25 mm i.d., 0.25 μm film thickness) using helium as the carrier gas. The source was operated at 260 °C in electron ionization (EI) mode at 70 eV ionization energy and with 8000 V acceleration voltage. Samples were injected in splitless mode at a constant temperature of 300 °C. For full-scan experiments, the GC oven was programmed at 40 °C (2 min), heated to 315 °C at 4 °C/min and a total run time of 90 min. The AutoSpec fullscan duration was 0.7 s plus 0.2 s interscan delay over a mass range of 55 to 600 Da. Based on repeat injection experiments, error quantification (% standard deviation (SD)) for lipid quantification is calculated as 4–5%.
RESULTS AND DISCUSSION Obloids cut from three different speleothems were sequentially digested in HCl. Each digest solution was extracted with solvent, and the molecular content quantified (specifically lipid extracts of n-alkanes, n-alkanoic acids, and sterols; phthalates; and halogenated iodine compounds) by GC/MS. Each extract does not represent any degree of age control, but a time-independent analysis of molecular content from outer to inner extract. The surface extract of all three speleothem samples contained a range of compounds including lipid fractions of n-alkanes, n-alkanoic acids (silylated) and sterols. Figures 1 and 2 show the distribution of key compounds from sequential extracts of speleothem SC-FS-2 and the associated blank. Across all three speleothem samples, detectable n-alkanes range from n-C21 to n-C36. The total sum of all n-alkanes in each speleothem sample was 0.2, 1.9 and 0.02 μg g–1 for SC-FS-2, Null-1 and BFM-J, respectively.
Copyright © 2014 John Wiley & Sons, Ltd.
Rapid Commun. Mass Spectrom. 2014, 28, 845–854
Extracting and interpreting organic molecules in speleothems
Figure 1. m/z 85 partial ion chromatograms of the sequential digestion experiment of speleothem SCFS-2 showing alkanes (n-alkanes are identified by ’•’ or carbon number; MMA = midchain branched monomethyl alkanes, UCM = unresolved complex mixture; ’ ~ ’ are truncated signals of FAs). The y-axis gives approximate concentrations per gram of digested speleothem. The axis is calibrated to n-C28 and has not been corrected for differences in response factors. (A) Comprehensive system blank, (B) 2nd digest, and (C) 1st digest (contaminated exterior of the samples). See Fig. 2 for digestion schematic. Straight-chain FA ranged from C8 to C24 with highest concentrations reached in outer extracts (maximum concentration 20 μg g–1 for hexadecanoic acid in speleothem Null-1). The only identified sterols were cholesterol and sitosterol. Although sitosterol was only observed consistently in extracts from speleothem BFM-J and at concentrations
