Deuterium exchangeability in modern and fossil plant resins

Accepted Manuscript Deuterium exchangeability in modern and fossil plant resins Gabriela González Arismendi, Ralf Tappert, Ryan C. McKellar, Alexander P. Wolfe, Karlis Muehlenbachs PII: DOI: Reference:

S0016-7037(18)30419-8 https://doi.org/10.1016/j.gca.2018.07.033 GCA 10864

To appear in:

Geochimica et Cosmochimica Acta

Received Date: Accepted Date:

17 May 2018 24 July 2018

Please cite this article as: Arismendi, G.G., Tappert, R., McKellar, R.C., Wolfe, A.P., Muehlenbachs, K., Deuterium exchangeability in modern and fossil plant resins, Geochimica et Cosmochimica Acta (2018), doi: https://doi.org/ 10.1016/j.gca.2018.07.033

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Deuterium exchangeability in modern and fossil plant resins Gabriela González Arismendia* Ralf Tapperta Ryan C. McKellar b Alexander P. Wolfec* Karlis Muehlenbachsa a

Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Alberta, T6G 2E9, Canada b Royal Saskatchewan Museum, 2445 Albert St., Regina, SK S4P 4W7, Canada c Department of Biological Sciences, University of Alberta, Edmonton, AB T6G 2E9, Canada *Corresponding author: email: [email protected] Present address: Department of Geoscience, University of Calgary, 2500 University Drive NW, Calgary, Alberta T2N 1N4, Canada

Keywords: amber, conifer resin, deuterium, isotopic exchange, paleoenvironment

1

Abstract Deuterium exchange experiments on modern and fossil plant resins (amber) were conducted to assess to what extent diagenetic alteration can overprint the stable hydrogen (δ2H) and carbon (δ13C) isotopic composition of these materials. Pairs of resins and amber fragments were placed together with deuterated water in sealed quartz-glass tubes to assure that all samples were exposed equally to the chosen experimental conditions. Experiments lasting up to one year were carried out at temperatures of 50 °C and 90 °C using water artificially enriched in deuterium (2H concentrations between 2,861 and 8,845 µg/g), which resulted in deuterium concentrations measured in samples well above natural concentrations (i.e., 2H ~143 µg/g). Modern resins recorded a mean deuterium exchange with deuterium-enriched water of 2.0 ± 0.1 %, compared to 1.3 ± 0.2 % observed in their fossil counterparts. No significant changes were observed in the δ13C of modern and fossil resins during the experiments. Overall, these results indicate that the extent of 2H isotopic exchange in resins is minor and apparently occurs prior to polymerization. Consequently, the stable isotopic composition of fossil resins has the potential to serve as a useful proxy for paleoenvironmental reconstructions, given the widespread distribution of ambers in the Mesozoic and Cenozoic fossil record.

2

1. Introduction Fossil plant resins (including their gem variety amber) are expected to have a high potential for retaining pristine chemical and isotopic signatures over geological timescales. This is due to their highly polymerized macromolecular structures (Nissenbaum and Yakir, 1995). Their stable isotopic composition may, therefore, provide useful information concerning the paleoenvironments associated with resin-producing forest ecosystems. Prior studies of the stable isotopic composition of fossil resins have largely focused on δ13C, showing considerable promise in part because diagenetic overprinting appears minimal (Nissenbaum and Yakir, 1995; Murray et al., 1998; McKellar et al., 2008; Tappert et al., 2013; Dal Corso et al, 2017). Only few studies, however, have addressed the potential applicability of the stable hydrogen isotopic composition (δ2H) of resins as a paleoenvironmental proxy. This is because the extent of post-depositional hydrogen isotopic exchange remains poorly understood (Nissenbaum et al., 2005; McKellar et al., 2008; Wolfe et al., 2012, 2016). The present study aims to rectify this situation by providing experimental results that directly address both C and H isotopic exchangeability in modern and fossil resins, thereby testing their viability as a substrate that retains information about the environmental conditions during their formation. Resins consist largely of terpenoids and acids, which are produced as a defense mechanism by a range of higher plants. The hydrogen that is incorporated into the resin during biosynthesis is primarily derived from environmental waters, for which the isotopic composition is largely controlled by regional precipitation patterns and temperature (Terwilliger and DeNiro, 1995; Chikaraishi et al., 2004; Sessions, 2006). 3

Since the early work of Epstein et al. (1976), deuterium fractionation factors between environmental waters and a range of plant biomarkers, including n-alkanes, lipids (fatty acids), and terpenoids, have been determined (e.g., Yakir and DeNiro, 1990; Sessions et al., 1999; Sessions, 2006; Chikaraishi et al., 2009; Yang and Leng, 2009). For bulk resins, the deuterium fractionation factor relative to source waters is approximately -200 ‰ (Nissenbaum et al., 2005; Chikaraishi et al., 2004; Wolfe et al., 2012). However, recent research conducted on modern plants suggests that various factors, such as evapotranspiration, relative humidity, and soil moisture, can influence the deuterium fractionation between waters and plant secondary metabolites, thus resulting in a relatively wider δ2H range (Polissar and Freeman, 2010; Sachse et al., 2006; Smith and Freeman, 2006; Sachse and Billault , 2012). In order to use the stable hydrogen isotopic compositions of fossil resins as a paleoenvironmental indicator, it is necessary to test whether or not measured δ2H values are overprinted by diagenetic alteration, or if they capture the primary environmental δ2H of environmental waters. In this study, deuterium exchange experiments were conducted to measure the degree of isotopic exchange between 2H-enriched water and both modern and fossil resins. Temperature effects on the extent of isotopic exchange were evaluated by conducting experiments at 50 and 90 °C, which spans a typical temperature range for most burial diagenetic systems. In addition, molecular rearrangements and physical changes were assessed by Fourier-transform infrared (FTIR) spectroscopy and thermal gravimetric analysis (TGA) before and after the isotope exchange experiments.

4

2. Materials and Methods 2.1. Samples Modern and fossil resin samples produced by both gymnosperms and angiosperms were used for the exchange experiments, given their markedly different compositions (Tappert et al., 2011). Modern resins from Metasequoia glyptostroboides (hereafter: MG), a deciduous cupressaceous conifer (LePage et al., 2005; Nugue, 2005) were collected from a cultivated specimen at the United States National Arboretum in Washington, D.C., USA. Commercial copal resins, a hardened resin at the early stage of polymerization, from Hymenaea courbaril (HC), a tropical fabalean legume that is widely distributed in Central and South America (Langenheim, 1995), were obtained from native specimens growing in Brazil. Fossil resins for the experiments included Cretaceous amber from Grassy Lake, Alberta, Canada (GLA), and mid-Oligocene to Miocene ambers from the Dominican Republic (DA). The former material is associated with coal seams of the Foremost Formation (Campanian) at the southern margin of the Western Canadian sedimentary basin (McKellar et al., 2008; McKellar and Wolfe, 2010). GLA amber samples used were dark yellow with reddish brown flow lines, a feature typical of mature amber samples (McKellar et al., 2008). The botanical source of GLA is thought to be Parataxodium, an extinct cupressoid genus that appears to be directly ancestral to Metasequoia (McKellar and Wolfe, 2010). Dominican amber specimens originate from the northeast Santiago area. These are thought to be mid-Oligocene to Miocene in age, and are found associated with laminated sands, silts, and clays (Grimaldi, 1995). The color of DA is typically yellow to deep red, suggesting a high degree of maturity (Lambert et al., 1985). 5

Hymenaea protera, an extinct species closely related to the extant Hymenaea courbaril, is considered to be the botanical source of DA (Cunningham et al., 1987; Grimaldi, 1995; Penney, 2010) (Table 1). All samples of modern and fossil resins used in the experiments were translucent with no visible organic or inorganic inclusions. Prior to analysis, freshly-broken resin and amber samples, each weighing 7.5 ± 2.5 mg, were cleaned with distilled water and subsequently air-dried without any additional chemical or physical treatment. 2.2. Deuterium exchange experiments The deuterium exchange experiments were carried out by placing pairs of modern and fossil resins into water with different deuterium concentrations. Modern resin samples were placed into inner quartz glass tubes (25 cm length, 6 mm diameter), which were subsequently inserted into outer quartz glass tubes (40 cm length, 9 mm diameter) containing a fossil resin sample. The outer tubes were later fitted with an Ultra-Torr™ vacuum metal fitting that allowed the experimental system to be evacuated to 3.99 kPa (30 mTorr). By introducing modern and fossil resin samples in different parts of a single system (Fig. 1), it was possible to ensure that both samples were subjected to identical experimental conditions without the possibility of cross-contamination. Following the addition of 0.5 ± 0.1 ml of deuterium-enriched water ([2Hw] = 8,845; 3,196; and 2,861 µg/g) using a syringe, the resin samples, and water were frozen by placing the outer tubes into liquid nitrogen. Simultaneously, all parts of the tubes that were not immersed in liquid nitrogen were heated to release deuterated H2O(g) adsorbed to the tube surface. After cryogenic evacuation, the outer tubes were sealed. Experiments were conducted at 50 °C and 90 °C (± 1 ºC), which simulates burial to a depth of ~3 km assuming a 6

geothermal gradient between 20 °C and 30 °C per km. The experiments were carried out over 5, 15, 30, 240, and 365 days. In addition to the experiments with deuterated water, runs with non-deuterated deionized water (2H =142.8 µg/g) were performed as controls. A total of fifty-six samples were prepared by this method. These samples included duplicates for each temperature, water deuterium concentration and experiment duration. Additional samples, specifically intended for thermal gravimetric (TGA) and Fouriertransform infrared (FTIR) analysis, were run at 90ºC in water containing 8,845 µg/g deuterium for 15, 30, and 240 days. At the end of each experiment, samples were dried under vacuum (3.99 kPa) at room temperature for 72 hours, and subsequently analyzed (Fig. 1). 2.3. Stable isotope analysis After the experiments, individual resin samples (~7.5 mg) were combusted with CuO (~1 g), Cu (~50 mg), and Ag (~50 mg) at 800 ºC in vacuum-sealed quartz glass tubes for 4 h. Upon cooling the combustion products, CO2 and H2O were extracted cryogenically. Water was reduced to H2 (g) using Indiana Zinc at 500 ºC for 25 min (Coleman et al., 1982). The isotope ratios of both H 2(g) and CO2(g) were measured off-line using a dual inlet Finnigan MAT-252 isotope-ratio mass spectrometer. Isotopic results are reported relative to VSMOW (Vienna Standard Mean Ocean Water) for deuterium (δ2H), and VPDB (Vienna Pee Dee Belemnite) for carbon (δ13C) with normalization conducted according to standard methods. Deuterium results (in per mil, ‰) are expressed in the conventional δ-notation: δ(Sample)= ((Rsample/Rstandard) - 1)

Eq. 1

7

where R is the 2H/1H ratio of the sample and the standard, respectively (e.g. Coplen, 2011). Because δ2H is a nonlinear scale relative to the 2H/1H ratio, a small error on values corresponding to large fractionation factors can be introduced (Sessions and Hayes, 2005). To prevent this, a conversion factor based on the relation 2H/1H for 2

HVSMOW of water = 1.56·10-4 was used for the calculation of deuterium concentrations in

µg/g, [2H], before and after the exchange experiments: [2H](µg/g) = 1.5576·10-1 (δ2H + 1000)

Eq. 2

The accuracy of the analytical method, based on laboratory standard waters and the analysis of replicate core samples, was better than 0.3 ‰ for δ18O and 0.8 ‰ for δ2H relative to VSMOW. 2.4. Thermal gravimetric analyses (TGA) In order to discriminate between molecular water adsorbed to the surface of resins, and exchangeable hydrogen within the resins, selected run-products from the 240day experiments were subjected to TGA. This technique was also used to elucidate the potential influence of water on the transformation of modern resins to amber, and to assess the percentage of exchangeable hydrogen in the samples more precisely. Selected resin samples were studied using a TGA-Q500 thermal gravimeter at the University of Alberta. The furnace temperature was programmed to equilibrate at 30 °C at the beginning of the experiment. Each sample was heated at a constant rate of 5 °C/min starting at 30 °C up to 250 ± 0.1 °C, with isotherms at 50 and 90 °C set up for 60 min at each temperature. The

8

purge gas used was N2 (99% pure), flowing at 60 mL per minute (Table S1). Measurements were conducted using ~7.5 mg of either modern resins or amber, and weight retention/temperature curves were recorded. Activation energies were calculated using the non-isothermal data. 2.5. Fourier-transform infrared (FTIR) microspectroscopy Samples were characterized using FTIR analyses in order to identify potential chemical changes that occurred during the experiments. The spectroscopic analyses were conducted before (t = 0) and after (t = 30, and 240 days) the isotopic exchange experiments with 8,845 µg/g-deuterated water, and another time after TGA (TGA – Tmax = 250 °C) (Fig. 1). FTIR analyses were conducted in transmittance mode using a Thermo-Nicolet Nexus 470 spectrometer fitted with a Nicolet Continuum IR microscope, and liquid nitrogen cooled MCT (MCT=mercury cadmium telluride) detector. Samples were first examined with a stereomicroscope, and inclusion-free fragments of crushed samples (<5 μm thickness) were selected and placed directly on a NaCl(s) plate. Beam size was set to 100 x 100 μm (square aperture) and the spectral resolution to 4 cm-1. OMNIC© software was used for data acquisition and processing. Liquid nitrogen was used to cool the detector, and dry, purified (CO2-free) air was used as a purging gas. The background was analyzed before each measurement, and no further spectrum correction was performed after data acquisition. Further technical details are provided in Tappert et al., (2011). Spectra were taken between 700 and 4000 cm-1 (2.5–14.3 μm), and for each analysis, 200 individual interferograms were co-added. Also, absorbance values were normalized, and spectral regions between 3700–3500 cm-1 and 2750–2580 cm-1 were specifically 9

monitored to identify potential changes as a response to deuterium incorporation. 3. Results 3.1. Deuterium exchange in resins at 50 and 90 °C The results of the deuterium exchange experiments are summarized in Table 2 and Figure 2. Within the time series plots, two main observations can be made: (1) the deuterium concentration of modern and fossil resin samples increases with time, but stabilizes to a constant [2H] value after around 30 days, at both 50 and 90 °C; (2) after 365 days, some of the modern and fossil resin samples have [2H] values that markedly deviate from the consistent trend of the samples with shorter experiment durations (Fig. 2). Resins from M. glyptostroboides and Dominican amber, for example, show a decrease in [2H], from a maximum of 340.7 ± 4.0 and 342.2 ± 2.0 µg/g at 240 days (50 °C, [2H]=8845 µg/g experiments), to about 261.0 ± 4.0 and 298.4 ± 0.6 µg/g at 365 days, respectively (Table 2). During the same time interval, H. courbaril resin and Grassy Lake amber gained deuterium relative to original values, reaching 283.9 ± 5.6 and 167.0 ± 8.3 µg/g, respectively. Control samples that were exposed to non-deuterated water for 5 days show no variation in [2H] (Table 2; Fig. 2). Figure 3 illustrates that the increase in deuterium concentration within modern and fossil resin samples during the experiments is directly proportional and strongly correlated (R2 > 0.90) with the deuterium concentration in the waters used in the experiments. The intercept corresponds to the initial [2Hresin] concentration of the untreated material, while the slope gives the magnitude of deuterium exchange between water and samples. Accordingly, the percentage of total hydrogen from deuterium-

10

enriched waters exchanged for resin from M. glyptostroboides, both at 50 and 90 °C, were 1.9 ± 0.3 % and 2.1 ± 0.2 %; while resin from H. courbaril exchanged 1.9 ± 0.3 % at 50 °C, and 1.7 ± 0.2 % at 90 °C. Dominican amber, on the other hand, showed a deuterium exchange of about 1.5 ± 0.2 % and 2.1 ± 0.2 % at 50 and 90 °C, respectively. Grassy Lake amber showed an exchange of 0.7 ± 0.2 % at 50 °C, and 0.6 ± 0.2 % at 90 °C both, which is notably lower than deuterium exchange capacity determined for modern resins or DA (Table 2; Fig. 3). 3.1.1. Physical changes recorded after the experiments After the experiments, the following physical changes in the resin samples were observed: (1) Morphology—Most of the modern resins changed from their original tabular shape to spherical forms initially exhibiting noticeable water inclusions, which consistently disappear after the thermal treatment at 90 °C, whereas fossil resin samples preserved their original shape, and showed no inclusions (Fig. 4). (2) Color—H. courbaril resins changed from milky white to yellow-orange; Grassy Lake amber developed a yellowish surface coloration and also the characteristic milky bands described by McKellar et al. (2008). By contrast, resins from M. glyptostroboides, and Dominican amber did not show color changes (Fig. 4). 3.2. Thermal gravimetric analysis (TGA) One of each of the M. glyptostroboides resins and one sample of Grassy Lake ambers were chosen for TGA analyses from experimental runs that were carried out at 90 °C for 240 days in water with a [2H] of 8,845 µg/g. Figure 5 shows their thermal

11

gravimetric mass loss and corresponding differential thermogravimetric (DTG) curves. The following observations were made during TGA: (1) Modern M. glyptostroboides resin (MG-240) exhibited relatively minor mass losses at 50.15 °C and 149.95 °C; and two well-defined mass loss steps at 95.2 °C, and 170 °C. The total loss was about 3.25 % of the original mass (Fig. 5a). (2) Fossil resin samples (Grassy Lake amber) (GLA-240) exhibited minor mass losses at lower temperatures: 49.51 °C, 89.9 °C, and 120.06 °C. The total loss was 2.02 % of original mass, which is slightly lower compared to the modern resin (Fig. 5b). Isothermal experiments up to ~100 °C imply mass changes likely relating to water desorption. Non-isothermal experiments (e.g., Trejo et al., 2010) conducted at a heating rate of 5 °C/ min. allow estimations of the activation energy. For the modern resins, our estimates of activation energy are around 19.21 kJ mol-1 (Fig. 5c), whereas those for fossil resins were around 3.60 kJ mol-1 (Fig. 5d). Interestingly, the deuterium isotope concentrations measured after TGA analyses of the same sample were lower by 115.0 ± 7.3 µg/g compared to those measured before the analyses, which is attributable to the loss of adsorbed deuterated water (Table 2). Comparison of the [2H] of samples before and after TGA measurements indicate a deuterium depletion of about 44 % (305 µg/g before TGA, 171 µg/g after TGA). This result is consistent with an approximately 8 % water desorption in resins and ambers (Table 2). The carbon isotopic composition of the modern and fossil resins, on the other hand, did not change during the TGA experiments, within measurement error (Table 2).

12

3.3. FTIR spectroscopy The FTIR spectra of resin and amber samples are sensitive not only to changes in their terpenoid constituents during alteration (Murae et al., 1995), but also to deuterium addition in various functional groups (e.g., Flakus and Chemecki, 2002). Table 3 summarizes and compares the main absorbance bands observed in modern and fossil resins, their shift after the isotope exchange experiments and TGA measurements, and the tentative wavelength assignment to organic functional groups. This information complements Figs. 6 and 7, which show the FTIR spectra of the samples and the effects of the isotope exchange experiments after 30 and 240 days at 50 °C and 90 °C, respectively. The IR spectra of the untreated material closely resemble those previously documented for GLA (McKellar et al., 2008; McKellar and Wolfe, 2010), and DA (Langenheim and Beck, 1965; Beck, 1986), HC, and MG modern resins (Tappert et al., 2011) (Tables 2 and 3; Fig. 6, Fig. S1). In general, the spectra of these samples are characterized by a broad peak between 3700 and 3200 cm-1, which is typically linked to O-H vibrational modes. Strong absorption features in the spectral range 3000–2800 cm-1 are attributed to stretching modes of terminal methyl and methylene groups. Additional absorption features in the range 1800–1600 cm-1 are attributed to carbon-carbon or carbon-oxygen double bonds (C=C, C=O). The spectral range below 1500 cm-1 typically contains the highest number of absorption features in resin spectra, and it is most useful in fingerprinting resin types in relation to botanical source. The absorption features in this range are generally linked to carbon and oxygen single bonds of the various acid and ester components that are present in resins, and can be tentatively assigned to skeletal

13

vibrations of double bounded carbon (Langenheim 1969, Wolfe et al., 2009, Tappert et al., 2011, Pastorelli et al., 2011). Figure 7 shows the effects of water sorption-desorption processes on MG and GLA spectra after the isotope exchange experiments and after the TGA measurements (i.e., after heating to 250 °C). These spectroscopic changes are discussed sequentially below. The 3600 – 3080 cm-1 spectral region After the exchange experiments, the spectra showed an increase in absorbance and a broadening of the hydroxyl peak, which is accompanied by a shift of the primary absorbance band towards higher wavenumbers. This region is related to hydroxyl groups that are associated with free alcohols (3650 cm-1), phenols (3600 – 3200 cm-1), and intramolecular hydrogen bonds of carboxylic acids (3500 – 2400 cm-1) (Figs. 6 and 7). The increase in the broadness of the hydroxyl band with experimental run-time is related to an increase in water sorption by the modern and fossil resin samples. However, a medium intensity peak at 3076 cm-1 was identified in resin samples (HC, and MG) as well as DA, but was absent or weak in GLA, possibly due to its greater geological maturity (Figs. 6 and S1; Table 3). The 3000 – 2800 cm-1 spectral region The spectral features in this range are characteristic of sp3 and sp2 C-H hybridization. Figure 7 shows the resulting spectra of product samples after the exchange experiments and TGA measurements. The intensity of the absorption bands after TGA measurements in GLA-240 spectra decreases and, at the same time, a broadening occurs. These changes suggest desorption of deuterated molecular water bonding, which is

14

consistent with the deuterium measurements (Table 3). Other IR spectral regions Other IR spectral changes in samples from long-term exposure exchange experiments include the appearance of relatively high absorption peaks at 1645 and 887 cm-1, and the appearance of distinctive peaks at 1092, and 815 cm -1, which are most likely related to an increase in desaturation with respect to pendant CH2 groups. 4. Discussion As with other organic polymers, modern and fossil resins incorporate deuterium either by water sorption, which implies physicochemical resin–water interactions, or by chemical organic hydrogen substitution (Zhou and Lucas, 1999; Schimmelmann et al., 1999). Three factors must be considered in order to evaluate the physical and chemical changes observed in our experiments: (1) the primary chemical composition of the resins; (2) the molecular water sorption/desorption mechanisms; and (3) the rates of deuterium exchange between water and resins. Some of the analyzed resin samples (e.g., HC and GLA; Fig. 4) showed an increase in their deuterium concentration after long-term experiments (365 days), accompanied by pronounced morphological alteration (shape, color). These changes were most pronounced in experiments conducted at 90°C, and may indicate an increase in water bonding with resin constituent molecules (Zhou and Lucas, 1999). On the other hand, constancy in δ13C throughout the experiments indicates that the carbon isotopic composition of modern and fossil resins does not change during diagenesis (Table S2). A similar result was recently obtained by Diefendorf et al. (2015) while investigating the

15

role of selective thermal alteration and degradation of plant n-alkane and terpenoids. They found that the δ13C values of these biomarkers remain largely unaltered, with only minor 12C depletion due to preferential 12C-12C bond breakage, at temperatures above 220 °C (see also Bjorøy et al., 1992). This is consistent with previous observations, which suggest that chemical reactions with the potential to influence the thermal degradation processes in modern and fossil resins only occur at temperatures above 250 °C (Murae et al., 1995). Such changes are likely produced through dehydrogenation and decarboxylation that transformed diterpenoid acids to aromatic diterpenes (Diefendorf et al., 2015). Therefore, mass loss during TGA can be attributed first and foremost to water loss. It also appears that modern resins are more prone to water loss than fossil resins. This indicates that water sorption–desorption mechanisms only play a role during the early maturation stage of resins. 4.1. Mechanisms of molecular water sorption-desorption in resins Two models previously applied to synthetic polymers may be invoked to explain the molecular exchange between water and natural resins: (1) free volume water diffusion, in which water molecules reside in the free volume of the resinous material, and (2) water molecule interactions with hydrophilic functional groups, such as hydroxyl and carboxyl, which may also lead to deuterium exchange (Jeffrey and Jeffrey, 1997; Zhou and Lucas 1999). Water sorption in resins can also be classified according to their activation energies into: (1) weakly-bound hydrogen molecular arrangement, which requires only a small amount of energy (4 to 15 kJ mol-1) to remove the single hydrogen water-resin bond (Type I); and (2) moderately-bound hydrogen, which requires a higher activation 16

energy (50 to 60 kJ mol-1) to remove inter-connected hydrogen-bound water from resin molecules (Type II) (Jeffrey and Jeffrey, 1997). Activation energies for modern and fossil resins derived from non-isothermal TGA measurements (19.21 kJ mol-1 and 3.60 kJ mol1

, respectively) indicate that most water molecule-resin interactions occur through weak

hydrogen Type I-bonding (Fig. 5). Nonetheless, both mechanisms may have an influence on the diagenetic transformation of plant resins into polymerized amber, and may also be associated with the morphological transformation from chips to spheres observed in some resins at high temperatures. The polymeric structure of fossil resins is based on multicyclic saturated and partially aromatized structures that lead to a gradual decrease of electronegative functional groups (Grimalt et al., 1988; Clifford et al., 1997). Our results show that the deuterium concentration is proportional to water sorption, and thus inversely proportional to the maturity of the samples as evidenced by a relatively higher deuterium absorption of DA when compared with GLA (see Table 1, Figs. 2 and 3). Although residual moisture may exert an impact in our interpretations, the difference in the percentage of exchangeable 2H of resins subjected to 50 vs. 90 °C is encouragingly low (Fig 3). As previously mentioned, the resins subject to higher temperatures were free of water inclusions upon the end of the long-term experiments, while the resins subject to unrealistic burial temperatures of 250 °C during TGA show a difference of ~8% compared with those not subjected to such elevated temperatures. Such consistently low differences among our experiments point to the 2H content of natural fossil resins as a useful proxy data for paleoenvironmental interpretations, and

17

also suggest that most of the resin’ 2H signature apparently forms prior to polymerization and in isotopic equilibrium with ambient waters. 4.2. Post-experimental chemical characterization Most of the information regarding hydrogen bond dynamics, such as inter- and intra-molecular interactions and vibrational structure, can be deduced from infrared absorption spectra, particularly in the spectral region 3600 to 2500 cm-1 (He et al., 2004) (Figs. 6, S1). In this region, all resin samples showed increases in absorbance and broadness, which are proportional to the time they were exposed to experimental conditions. Even though the physical state of the samples (e.g., thickness variability) and environmental conditions during measurement (e.g., humidity) might have some effect on the resulting spectrum (Derrick et al., 1999), their combined effects can be considered minor when compared to the effects of water sorption related to Type I-bonding with water. Water bonding in modern and fossil resins has been identified to change the response in the 3600-3080 cm-1 spectral region. This effect suggests interactions with different hydrophilic functional groups within the resin macromolecules. The amount of surface-bound water molecules increases with time and temperature, and is more prevalent in modern resins compared to fossil resin samples, which may lead to an environmental assessment of diagenetic conditions with further study. The variability of other features in the infrared spectra of modern and fossil resins is likely linked to thermal degradation after long-term exposure to the experimental conditions. This assumption is supported by changes in absorption features at 3070 cm -1, and 880-890 cm-1. An increase in absorption within these spectral regions, which correspond to the sp3 stretching and sp2 disubstituted bending of the =C–H groups in 18

olefines, respectively, reflect desaturation and conversion of single bonds into double bonds (Anderson et al., 1992). In most compounds with a labdane skeleton, the 3080 cm-1 band can be assigned to the stretching vibration of the C-H bond in exomethylene (Grimalt et al., 1988; Anderson et al., 1992; Wolfe et al., 2009; Tappert et al., 2011). The appearance of this band in GLA after TGA treatment, and the observed increase in its intensity in younger ambers and resins, together with an increase in the peak located at ~1645 cm-1 (C=C stretches within exocyclic methylene) (Grimalt et al. 1988), together with a decrease in the peak at ~1450 cm-1 (in C-H bending modes), are both consistent with a decrease in saturation (Fig. 6). As previously observed, mature ambers show an inverse relationship between the peaks listed above, and an absence of absorption peak at 3080 cm-1 that result from their depletion of exocyclic methylene during postdepositional polymerization (e.g., Brody et al., 2001; Tappert et al., 2011). In mature ambers, non-variability or slight increases on the out-of-plane C-H deformation at 880– 890 cm-1 peaks were also observed. However, the lack of peaks in the 2000–2500 cm-1 region suggests that changes due to molecular deuterium exchange (Flakus and Chemecki, 2002) may not be readily detectable using FTIR spectroscopy. In summary, TGA and FTIR analysis demonstrated that even though water sorption-desorption processes may play a role during the diagenetic alteration process of resins, the bottom line of this study shows that they do not overprint the deuterium concentration of the samples (Table 2). This conclusion is also consistent with the remarkable stability of 13C values for the same materials during the experiments, a feature which has been also observed in secondarily redeposited ambers (Mänd et al., 2018). Therefore, well-

19

preserved amber deposits can be considered as reliably archives of paleoenvironmental hydrogen water isotope composition. 4.3. Further implications The slight degree of deuterium exchange observed here indicates that rock interactions with thermally active pore fluids produce negligible deuterium incorporation. Thus, the degree of diagenetic deuterium exchange would be insufficient to overprint the primary environmental signals contained in the δ2H of angiosperm and gymnosperm resins. Given a biosynthetic fractionation factor of about 200 ‰, the maximum deuterium overprint in resins would be around 20 ‰. Since, reworked fossil resins may undergo additional degradation processes, including photo-oxidation and hydrolysis processes outside the scope of the present analyses (see Pastorelli et al., 2012, 2013), visually degraded resin samples should be avoided, or their data interpreted with caution. 5. Conclusions The main observations and conclusions from the deuterium exchange experiments on modern and fossil resins are: (1) Secondary deuterium incorporation under isothermal conditions at 50 and 90 °C is insufficient to overprint the primary hydrogen isotopic composition of modern and fossil resins, which is inherited from metabolic processes during resin biosynthesis and attendant environmental conditions. (2) Modern angiosperm (HC) and gymnosperm (MG) resins incorporate slightly more deuterium (MG: 1.9 ± 0.3% at 50 °C and 2.1 ± 0.2 % at 90 °C; HC: 1.9 ± 0.3 % at 50 °C

20

and 1.7 ± 0.2 % at 90 °C) relative to their fossil counterparts (DA: 1.5 ± 0.2 % at 50 °C and 2.1 ± 0.2 % at 90 °C; GLA: 0.7 ± 0.2 % at 50 °C, and 0.6 ± 0.2 % at 90 °C). (3) Secondary, post-metabolic, deuterium incorporation in resins is mainly governed by water sorption-desorption processes that mainly occur in the early stages of maturation. An additional contribution of deuterium can come from surface-bonded (adsorbed) water. Overall, this study demonstrates that unaltered resin specimens can preserve their original stable isotopic signatures (δ2H and δ13C) throughout burial and diagenesis. The stable isotopic composition of fossil resins is therefore well suited to be used as a paleoenvironmental proxy.

ACKNOWLEDGEMENTS We thank Thomas Stachel (University of Alberta) for access to FTIR facilities, Shiau-Yin Wu for technical assistance and expertise during Thermal Gravimetric Analysis at the Nanofabrication and Characterization Facility (nanoFAB–University of Alberta). Many thanks also to Olga Levner for providing technical expertise and instrumental training. This research was funded by the Natural Sciences and Engineering Research Council of Canada (Discovery awards to APW and KM, postdoctoral fellowship to RCM and RT). Finally, we would like to thank three anonymous journal reviewers and Dr. Sarah Feakins for their editorial contribution that improved an early version of this manuscript.

21

Table and figure captions Table 1. Modern and fossil resin samples used in this study. Table 2. Deuterium concentration in modern and fossil resins after exchange experiments with deuterated waters ([2Hw] = 2,861 µg/g; 3,196 µg/g; and 8,845 µg/g) at 50 °C and 90 °C and run-times of 5, 15, 30, 240, and 365 days (represented by number in sample name). Highlighted samples were analyzed using TGA, and FTIR spectroscopy. See details in text. Table 3. Intensity analysis and functional-group assignments of spectral features in FTIR spectra of modern and fossil resins used in the deuterium exchange experiments. Greyunderlain fields indicate bands that showed changes. A= absent, W= wide, V= variable, S= strong, M= medium, (sh)= sharp, (br)= broad, (w)= weak. Figure 1. Flowchart summarizing the individual steps in the experimental and analytical protocol applied in this study. The inset (upper left corner) schematically illustrates the setup for the deuterium exchange experiments as well as subsequent physicochemical characterization. Figure 2. Time-series plots showing the changes in deuterium concentrations in modern and fossil resins exposed to deuterated waters with different deuterium concentrations ([2Hw] = 8,845 µg/g (■); 3,196 µg/g (♦); 2,861 µg/g (●)) at 50 °C (a–d) and 90 °C (e–h). Dashed lines connect data-points from experiments exceeding 240 days that were likely affected by physical degradation that was so severe that it resulted in isotopic deviations. Figure 3. Plot of the concentration of deuterium [2H] in the deuterated waters against the concentration of deuterium in the resin samples after experiments conducted at 50 °C (a–

22

d) and 90 °C (d–g). Each symbol represents the deuterium concentration of samples per days and their mean value (●) (Days= 5 (□), 15 (○), 30 (∆), 60 (+), 240 (x), and 365 (◊)). The intercept in the linear equation is the primary deuterium concentration of raw samples; while the slope represents the percentage of deuterium transferred from the waters to the specimens. The standard error bars of the slopes and intercepts show the uncertainties of the regression analyses. Figure 4. Physical changes observed in resins and ambers after 60 and 365 days at 90 °C. HC (a) and MG (b) specimens showed significant morphological changes from chips to elongate forms after 60 days, and from elongate forms to spheres after 365 days. In addition, HC samples showed a change in color from pale-white to dark orange. DA(c), and GLA (d) samples developed a foam-covered surface, which is more conspicuous in GLA samples. The color of the amber specimens generally became darker after the experiments. Figure 5. TGA (black) and DTG (blue) curves (a, b), and calculated activation energies (Ea) from non-isothermal experiments (c, d) for samples MG-240 and GLA-240. The magnitude of the changes in mass during the TGA measurements is accentuated in the DTG curve. Activation energies were calculated from Arrhenius plots in the temperature interval 30 °C to 50 °C. Figure 6. FTIR spectra of modern resins (1) and fossil resins (2) before and after exchange experiments conducted at 90 °C and with run-times of 30 and 240 days. For tentative peak identification refer to Figure 7 and Table 3. Figure 7. Comparison of FTIR spectra of MG and GLA before and after exchange experiments with run-times of 240 days (a-b), and after TGA measurements, i.e., after 23

heating to 250 °C (c). Residual from the subtraction of the pre-experimental from the post-TGA spectrum (d). The residual provides some indication of the spectral changes that occurred during the experiments. However, differences in thickness and surface morphology of the individual sample fragments that were analyzed also influence the residual.

24

APPENDICES Tables and figures caption Table A1. Conditions during TGA analyses. Isothermal (steps 3, 4) and non-isothermal (steps 2, 4) analyses of resin and amber samples after the exchange experiments. Table A2. δ13C of resins and ambers after the exchange experiments and TGA measurements. Figure A1. FTIR spectra of modern resins (1) and fossil resins (2) before and after exchange experiments conducted at 50 °C and with run-times of 30 and 240 days. For tentative peak identification refer to Figure 7 and Table 3.

25

References Anderson K., Winans, R.E. and Botto, R.E., (1992) The Nature and fate of natural resins in the geosphere. II. Identification, classification and nomenclature of resinites. Org. Geochemistry, 18, 829–841. Beck C.W. (1986) Spectroscopic investigations of amber. Appl. Spectrosc. 22, 57–200. Bjorøy M., Hall P. B., Hustad E. and Williams J. A. (1992) Variation in stable carbon isotope ratios of individual hydrocarbons as a function of artificial maturity. Org. Geochemistry 19, 89–105. Chikaraishi Y., Naraoka H. and Poulson S. R. (2004) Carbon and hydrogen isotopic fractionation during lipid biosynthesis in a higher plant (Cryptomeria japonica). Phytochemistry, 65, 323–330. Chikaraishi Y., Tanaka R., Tanaka A. and Ohkouchi N. (2009) Fractionation of hydrogen isotopes during phytol biosynthesis. Org. Geochemistry, 40, 569–573. Clifford D. J., Hatcher P. G., Botto R. E., Muntean J. V., Michels B. and Anderson K. B. (1997) The nature and fate of natural resins in the geosphere—VIII. NMR and PyGC-MS characterization of soluble labdanoid polymers, isolated from Holocene class I resins. Org. Geochemistry, 27, 449–464. Coleman M. L., Shepherd T. J., Durham J. J., Rouse J. E. and Moore G. R. (1982) Reduction of water with zinc for hydrogen isotope analysis. Anal. Chem., 54, 993– 995.

26

Coplen, T.B., (2011) Guidelines and recommended terms for expression of stableisotope-ratio and gas-ratio measurement results. Rapid Commun. Mass Spectrom., 25, 2538–2560. Cunningham A., Gay I. D., Oehlschlanger A. C. and Langenheim J. H. (1987)

13

C NMR

and IR analyses of the structure, aging and botanical origin of Dominican and Mexican ambers. Phytochemistry, 22, 965–968. Dal Corso, J., Schmidt, A. R., Seyfullah, L. J., Preto, N., Ragazzi, E., Jenkyns, H. C., Delclòs, X., Néraudeau, D. and Roghi, G. (2017) Evaluating the use of amber in palaeoatmospheric reconstructions: the carbon-isotope variability of modern and Cretaceous conifer resins. Geochimica et Cosmochimica Acta, 199, 351–369. Derrick M. R., Stulik D. and Landry J. M. (1999) Infrared spectroscopy in conservation science. Getty Publications. pp 236. Diefendorf A. F., Sberna D. T. and Taylor D. W. (2015) Effect of thermal maturation on plant-derived terpenoids and leaf wax n-alkyl components. Org. Geochemistry, 89, 61–70. Epstein S., Yapp C.J. and Hall J.H. (1976) The determination of D/H ratios of nonexchangeable hydrogen in cellulose extracted from aquatic and land plants. Earth Planet. Sc. Lett., 30, 241–251. Flakus H. T. and Chelmecki M. (2002) Infrared spectra of the hydrogen bond in benzoic acid crystals: temperature and polarization effects. Spectrochim. Acta. Part A, Mol. Biomol. Spectrosc., 58, 179–96.

27

Grimaldi D. A. (1995) The age of Dominican amber. In Anderson, K. B. and J. C. Crelling, editors. Amber, resinite, and fossil resins. American Chemical Society Symposium Series 617. pp. 203–217.Washington, DC: American Chemical Society. Grimalt J., Simoneit B., Hatcher P. and Nissenbaum A. (1988) The molecular composition of ambers. Org. Geochemistry, 13, 677–690. He Y., Zhu B. and Inoue Y. (2004) Hydrogen bonds in polymer blends. Prog. Poly. Sci., 29, 1021–1051. Jeffrey G. A. and Jeffrey G. A. (1997) An introduction to hydrogen bonding. 12, pp. 303, Oxford University Press, New York. Lambert J.B., Frye J.S. and Poinar G.O. (1985) Amber from the Dominican Republic: analysis by nuclear magnetic resonance spectroscopy. Archaeometry, 27, 43–51. Langenheim J. H. (1969) Amber: a botanical inquiry. Science, 163, 1157. Langenheim J. H. (1995) Biology of amber-producing trees: focus on case studies of Hymenaea and Agathis. In Amber, resinite and fossil resins (eds K. B. Anderson and J. C. Crelling). pp. 1–31. Washington, D.C.: American Chemical Society. Langenheim J.H. and Beck C.W. (1965) Infrared spectra as a mean of determining botanical source of amber. Science, 149, 52–55. LePage B., Yang H. and Matsumoto M. (2005) The evolution and biogeographic history of Metasequoia. In The Geobiology and Ecology of Metasequoia (eds LePage, B. A., Williams, C. J. and Yang, H.), 22, 3–114, Springer, Dordrecht.

28

Mänd, K., Muehlenbachs, K., McKellar, R. C., Wolfe, A. P. and Konhauser, K. (2018) Distinct origins for Rovno and Baltic ambers: Evidence from carbon and hydrogen stable isotopes. Palaeogeogr Palaeoclimatol Palaeoecol., 505, 265–273. McKellar R. C., Wolfe A.P., Tappert R. and Muehlenbachs K. (2008) Correlation of Grassy Lake and Cedar Lake ambers using infrared spectroscopy, stable isotopes, and palaeoentomology. Can. J. Earth Sci., 45, 1061–1082. McKellar R. C. and Wolfe A. P. (2010) Canadian Amber. In Biodiversity of fossils in amber from the major world deposits (ed. D. Penney), pp. 96–113, Siri Scientific Press, Manchester. Murae T., Shimokawa S. and Aihara A. (1995) Pyrolytic and spectroscopic studies of the diagenetic alteration of resinites. In Amber, resinite and fossil resins (eds K. B. Anderson & J. C. Crelling), pp. 76–91. Washington, DC: American Chemical Society. Murray A. P., Edwards D., Hope J. M., Boreham C. J., Booth W. E., Alexander R. A. and Summons R. E. (1998) Carbon isotope biogeochemistry of plant resins and derived hydrocarbons. Org. Geochemistry, 29, 1199–1214. Nissenbaum A. and Yakir D. (1995) Stable Isotope Composition of Amber. In Amber, resinite and fossil resins (eds. K. B. Anderson & J. C. Crelling), pp. 34–42. Washington, DC: American Chemical Society. Nissenbaum A., Yakir D. and Langenheim J. H. (2005) Bulk carbon, oxygen, and hydrogen stable isotope composition of recent resins from amber-producing Hymenaea. Die Naturwissenschaften, 92, 26–29.

29

Nugue C. (2005) Cultivars of Metasequoia glyptostroboides. In The Geobiology and Ecology of Metasequoia (eds LePage, B. A., Williams, C. J. and Yang, H.), 22, pp. 361–366, Springer, Dordrecht. Pastorelli G., Richter J. and Shashoua Y. (2011) Photoageing of Baltic amber – Influence of daylight radiation behind window glass on surface colour and chemistry. Polym. Degrad. Stabil., 96, 1996–2001. Pastorelli G., Richter J. and Shashoua Y. (2012) Evidence concerning oxidation as a surface reaction in Baltic amber. Spectrochim. Acta Mol. Biomol. Spectrosc., 89, 268–269. Pastorelli G., Shashoua Y. and Richter J. (2013) Hydrolysis of Baltic amber during thermal ageing - an infrared spectroscopic approach. Spectrochim. Acta Mol. Biomol. Spectrosc., 106, 124–128. Penney, D. (2010) Dominican Amber. In Biodiversity of fossils in amber from the major world deposits (ed. D. Penney), pp. 22–41, Siri Scientific Press, Manchester. Polissar P. J. and Freeman K.H. (2010) Effects of aridity and vegetation on plant-wax δD in modern lake sediments. Geochim. Cosmochim. Acta, 74, 85–97. Sachse, D., Billault, I., Bowen, G.J., Chikaraishi, Y., Dawson, T. E., Feakins, S.J., Freeman, K.H., R. M., Clayton, McInerney, F.A., van derMeer, M.T.J., Polissar, P., Robins, R.J., Sachs, J.P., Schmidt, H-L., Sessions, A.L., White, J.W.C., West, J.B., and Kahmen, A., (2012) Molecular paleohydrology: interpreting the hydrogenisotopic composition of lipid biomarkers from photosynthesizing organisms. Annu. Rev. Earth Planet. Sci, 40, 221–249.

30

Sachse D., Radke J. and Gleixner G. (2006) δD values of individual n-alkanes from terrestrial plants along a climatic gradient—implications for the sedimentary biomarker record. Org. Geochemistry, 37, 69–83. Schimmelmann A., Lewan M. D. and Wintsch R. P. (1999) D/H isotope ratios of kerogen, bitumen, oil, and water in hydrous pyrolysis of source rocks containing kerogen types I, II, IIS, and III. Geochim. Cosmochim. Acta, 63, 3751–3766. Schimmelmann A., Sessions A. L. and Mastalerz M. (2006) Hydrogen isotopic (D/H) composition of organic matter during diagenesis and thermal maturation. Annu. Rev. Earth Pl. Sc., 34, 501–533. Sessions A. (2006) Seasonal changes in D/H fractionation accompanying lipid biosynthesis in Spartina alterniflora. Geochim. Cosmochim. Acta, 70, 2153–2162. Sessions A. L., Burgoyne T. W., Schimmelmann A. and Hayes J. M. (1999) Fractionation of hydrogen isotopes in lipid biosynthesis. Org. Geochemistry, 30, 1193–1200. Sessions A. L. and Hayes J. M. (2005) Calculation of hydrogen isotopic fractionations in biogeochemical systems. Geochim. Cosmochim. Acta, 69, 593–597. Smith F. and Freeman, K. (2006) Influence of physiology and climate on δD of leaf wax n-alkanes from C3 and C4 grasses. Geochim. Cosmochim. Acta, 70, 1172–1187. Tappert R., McKellar R. C., Wolfe A. P., Tappert M., Ortega-Blanco J. and Muehlenbachs K. (2013) Stable carbon isotopes of resin exudates record responses of C3 plants to changes in atmospheric oxygen since the Triassic. Geochim. Cosmochim. Acta, 121, 240–262.

31

Tappert R., Wolfe A. P., McKellar R. C., Tappert M. and Muehlenbachs K. (2011) Characterizing modern and fossil gymnosperm exudates using micro-Fourier transform Infrared Spectroscopy. Int. J. Plant Sci., 172, 120–138. Terwilliger V.J. and DeNiro M.J. (1995) Hydrogen isotope fractionation in woodproducing avocado seedlings: biological constraints to paleoclimatic interpretations of δD values in tree ring cellulose nitrate. Geochim. Cosmochim. Acta, 59, 5199– 5207. Trejo F., Rana M. S. and Ancheyta J. (2010) Thermogravimetric determination of coke from asphaltenes, resins and sediments and coking kinetics of heavy crude asphaltenes. Catal. Today, 150, 272–278. Wolfe A.P., Csank A.Z., Reyes A.V.E., McKellar R.C., Tappert R. and Muehlenbachs, K. (2012) Pristine Early Eocene wood buried deeply in kimberlite from northern Canada. PloS One, 7, e45537. Wolfe A. P., McKellar R. C., Tappert R., Sodhi R. N. and Muehlenbachs K. (2016) Bitterfeld amber is not Baltic amber: Three geochemical tests and further constraints on the botanical affinities of succinite. Rev. Palaeobot. Palynol., 225, 21–32. Wolfe A. P., Tappert R., Muehlenbachs K., Boudreau M., McKellar R. C., Basinger J. F. and Garrett A. (2009) A new proposal concerning the botanical origin of Baltic amber. Proc R. Soc. A., 276, 3403–3412. Yakir D. and DeNiro M. J. (1990) Oxygen and hydrogen isotope fractionation during cellulose metabolism in Lemna gibba. Plant Physiol., 93, 325–332.

32

Yang H. and Leng Q. (2009) Molecular hydrogen isotope analysis of living and fossil plants—Metasequoia as an example. Prog. Nat. Sci., 19, 901–912. Zhou J. and Lucas J. P. (1999) Hygrothermal effects of epoxy resin. Part I: the nature of water in epoxy. Polymer, 40, 5505–5512.

33

Table

Table 1. Modern and fossil resin samples used in this study.

Sample

Sample type

Source tree

Location

Age

(1.1)

HC

Resin

Hymenaea courbaril - Copal (commercial sample)

(1.2)

MG

Resin

Metasequoia glyptostroboides

National Arboretum, Washington DC. United States

Modern

(2.1)

DA

Amber

Hymenaea protera (extinct specimen)

Dominican Republic

MidOligocene – Miocene

GLA

Amber

Cupressaceae genera. Parataxodium.

Grassy Lake, Alberta. Canada

Campanian, Cretaceous

(2.2)

(1.1)

Brazil

Modern

Hymeneae courbaril, (1.2)Metasequoia glyptostroboides, (2.1)Dominican Amber, (2.2)Grassy Lake Amber

Table 2. Deuterium concentration in modern and fossil resins after exchange experiments with deuterated waters ([2Hw] = 2,861 ꭒg/g; 3,196 ꭒg/g; and 8,845 ꭒg/g) at 50°C and 90°C and run-times of 5, 15, 30, 240, and 365 days (represented by number in sample name). Highlighted samples were analyzed using TGA, and FTIR spectroscopy. See details in text. Temperature Deuterated waters [2HW] (µg/g) Sample HC-BE HC5.DW HC-5 HC-15 HC-30 HC-60 HC-240 HC-365 MG-BE MG5.DW MG-5 MG-15 MG-30 MG-60 MG-240 MG-365 DA-BE

50 ± 1 °C 2,861

3,196

8,845

[2Hresin] (µg/g) 121.4

[2Hresin] (µg/g) -

[2Hresin] (µg/g) -

[2Hresin] (µg/g) -

119.4

-

-

-

112.5

135.9 151.0 154.6 155.4 155.4 244.1 -

161.1 175.5 208.7 210.4 300.6 259.9 -

209.2 308.1 313.1 327.0 328.9 305.9 -

108.0

-

-

-

115.8

120.0 124.7 124.8 125.1 125.0 183.6 -

157.7 167.0 174.8 175.2 176.1 182.5 -

256.8 277.6 316.8 320.2 323.9 306.4 -

DA5.DW DA-5 DA-15 DA-30 DA-60 DA-240 DA-365 GLA-BE GLA5.DW GLA-5 GLA-15 GLA-30 GLA-60 GLA240 GLA-365

HC-5 HC-15 HC-30 HC-60 HC-240 HC-365 MG-5 MG-15 MG-30 MG-60

120.9

-

-

-

110.7

134.9 140.7 152.0 152.8 174.9 200.8 -

163.8 168.7 180.9 186.2 202.5 217.0 -

230.7 260.7 272.0 273.4 292.7 222.9 -

115.1

-

-

-

Temperature

121.5 137.6 139.3 139.9 140.3 243.4

135.6 144.8 148.4 152.1 152.9 245.3 90 ± 1 °C

159.6 164.7 165.1 168.8 172.4 248.2

TGA** [2Hresin] (µg/g)

2,861

3,196

8,845

187.9 -

121.7 137.4 146.9 155.4 156.2 208.5 137.6 151.0 155.8 156.1

154.6 180.7 188.1 199.2 203.2 229.6 182.7 198.8 202.7 210.8

232.9 276.3 285.1 285.1 290.5 283.9 295.3 303.9 304.6 307.9

MG-240 MG-365 DA-5 DA-15 DA-30 DA-60 DA-240 DA-365 GLA-5 GLA-15 GLA-30 GLA-60 GLA-240 GLA-365

171.3 132.2 156.5 -

160.9 222.7 145.6 151.1 161.6 163.2 171.5 178.7 127.8 137.0 141.2 146.5 146.7 131.3

222.9 215.5 171.1 180.4 190.7 192.2 241.1 245.8 137.6 150.2 157.6 157.3 157.5 162.8

305.6 298.4 222.7 330.4 340.7 342.2 342.5 261.0 166.2 168.4 168.8 170.4 169.7 167.0

Table 3. Intensity analysis and functional-group assignments of spectral features in FTIR spectra of modern and fossil resins used in the deuterium exchange experiments. Grey-underlain fields indicate bands that showed changes. A= absent, W= wide, V= variable, S= strong, M= medium, (sh)= sharp, (br)= broad, (w)= weak. Raw Functional Group O-H

C-H

C=O

C=C

Type Frequency** HC Alcohols, phenols 3650 A(w) free O-H Hydrogen bonded 3400 V(br) Carboxylic acids 3076 S(w) R(C=O)O-H sp2 hybridization 2960 S(sh) hydrogen sp3 hybridized 2935 V(w) (R3-CH) 2858 S(sh) 2848 A Shoulder Esters (R(C=O)R) 1775 (w) 1730 Shoulder 1715 A Acids (R(C=O)H) 1697 S(sh) R2C=CH2C 1648 M(sh) C=C-H2 1460 M(sh) Aromatic ring C1448 A H, C=C 1417 M(w) 1382 S(sh)

MG

DA

GLA

A

A

V(w)

V(br)

V(br)

V(br)

S(w)

S(w)

A

S(sh)

S(sh)

S(sh)

A

A

A

M(sh) M(sh)

S(w) M(sh) Shoulder A (w) Shoulder Shoulder S(sh) A A S(sh) M(sh) M(sh) A M(w)

S(w) V(sh)

S(sh)

A

S(sh)

A Doublet

M(sh) S(sh)

A Doublet

A Shoulder S(sh) A M(w) A

Stretching CD2 sp2 hybridized CH(C=CH2) C-D in plane bending vibration

1092

W

W

W

W

887

S(sh)

Doublet

S(sh)

M(sh)

815

A

A

A

A

T= 50 °C, t= 240 days O-H

C-H

C=O

C=C

Alcohols, phenols free O-H Hydrogen bonded Carboxylic acids R(C=O)O-H sp2 hybridization hydrogen sp3 hybridized (R3-CH)

Esters (R(C=O)R)

Acids (R(C=O)H) R2C=CH2C C=C-H2 Aromatic ring CH, C=C

Stretching C2H2 sp2 hybridized C-

3650

A

A

V(w)

V(w)

3400

W(br)

W(br)

V(br)

W(br)

3076

S(w)

S(w)

S(w)

A

2960

V(sh)

S(br)

S(sh)

S(br)

2935

M(sh)

A

A

A

2858 2848

V(w) M(sh)

1730 1715 1697 1648 1460

M(w) A Shoulder - V(w) Shoulder A S(sh) M(sh) M(sh)

S(w) M(sh) Shoulder A (w) Shoulder M(sh) S(sh) A A S(sh) M(sh) M(sh) A M(sh)

M(sh) V(w) Shoulder (br) S-M(sh) A S-M(sh)

1448

A

S(sh)

A

S(w)

1417 1382 1092 887

M(w) S(sh) W S(sh)

A Doublet W Doublet

M(sh) S(sh) W S(sh)

A M(w) M(w) M(sh)

1775

Shoulder(w) A

H(C=CH2) C-2H in plane bending vibration

815

M

M

M

M

T= 90 °C, t= 240 days O-H

C-H

C=O

C=C

Alcohols, phenols free O-H Hydrogen bonded Carboxylic acids R(C=O)O-H sp2 hybridization hydrogen sp3 hybridized (R3-CH)

Esters (R(C=O)R)

Acids (R(C=O)H) R2C=CH2C C=C-H2 Aromatic ring CH, C=C

Stretching C2H2 sp2 hybridized CH(C=CH2) C-2H in plane

3650

V(w)

A

V(w)

A

3400

V(br)

W(br)

V(br)

W(br)

3076

M(w)

M(sh)

M(sh)

V(w)

2960

M(sh)

M(sh)

A

shoulder

2935

S(sh)

S(sh)

S(br)

S(sh)

2858 2848

M(sh) M-S(sh)

1730 1715 1697 1648 1460

M(sh) A Shoulder - V(w) Shoulder A S(sh) S(sh) M(sh)

M(sh) V(w) Shoulder A (w) Shoulder M-V(w) S(sh) A A M-V(w) M(w) S-M A M(w)

M(sh) V(w) Shoulder (br) S-M(sh) A S-M(sh)

1448

M(sh)

S

A

S(br)

1417 1382 1092

M(br) S(sh) M

A Doublet M

M(w) S(sh) M

M(sh) M(sh) M

887

S(sh)

M(sh)

S(sh)

V(w)

815

M

M

M

M

1775

Shoulder(w) M(br)

bending vibration

Deuterated waters: = 8,845, 3,196, and 2,861 µg/g

Temperatures: 50˚C and 90˚C

1

2

Time series: Days: 5, 15, 30, 60, 240 and 365

Drying process Time: 72 hrs at ~4 Pa

Isotope analyses

Chemical and physical characterization Samples: Modern and fossilized resins Under thermal treatment at 50˚C and 90˚C [2Hw] = 8,845 µg/g

FTIR Time intervals: 0, 30 and 240 days

Thermal Gravimetric Analyses (TGA) Isotope analyses

I. Deuterium exchange experiments

[2Hw]

II. Samples characterization

Experimental exchange sample treatment

300

a

400

e

300

[2HW] = 3,196 µg/g

200

200

400

After TGA

[2HW] = 2,861 µg/g

100 0

[2H samples] (µg/g)

[2HW] = 8,845 µg/g

100

Hymenaea copal 0

100

200

300

400

0

100

200

300

b

0

400

400

f

300

300

200

200

100

100

0

M. glyptostroboides 0

100

200

300

400

0

100

Time (days)

200

300

[2H samples] (µg/g)

[2H samples] (µg/g)

400

90 °C

Resins

400

[2H samples] (µg/g)

50 °C

0

Time (days)

Ambers 300

200

200

100

100

400

Dominican Amber 0

100

200

300

400

0

100

200

300

d

400 h

0 400

300

300

200

200

100

100

0

Grassy Lake Amber 0

100

200

Time (days)

300

400

0

100

200

300

Time (days)

400

[2H samples] (µg/g)

400

300

0

[2H samples] (µg/g)

g

c

0

[2H samples] (µg/g)

[2H samples] (µg/g)

400

[2H samples] (µg/g)

[2H samples] (µg/g)

Resins 400 300

50 °C

Hymenaea copal

[2HHC Ex 50°C] = 0.019 [2HHCBE] + 127; R2 = 0.93

90 °C a

[2HHC Ex 90°C] = 0.017 [2HHCBE] + 120; R2 = 0.96

e

400 300

200

200

100

100

400 300

M. glyptostroboides

b

[2HMG Ex 50°C] = 0.019 [2HMGBE] + 101; R2 = 0.95

f [2HMG Ex 90°C] = 0.021 [2HMGBE] + 116; R2 = 0.96

400 300

200

200

100

100 0

0

2000 4000 6000 8000 [2Hwater] (µg/g)

2000 4000 6000 8000 [2Hwater] (µg/g)

[2H samples] (µg/g)

[2H samples] (µg/g)

Ambers 400 300

Dominican Amber

g

c [2HDA Ex 90°C] = 0.021 [2HDABE] + 116; R2 = 0.97

[2HDA Ex 50°C] = 0.015 [2HDABE] + 120; R2 = 0.97

400 300

200

200

100

100

400 300

Grassy Lake Amber [2HGLA Ex 50°C] = 0.007 [2HGLABE] + 125; R2 = 0.78

d

h [2HGLA Ex 90°C] = 0.006 [2HGLABE] + 120; R2 = 0.82

400 300

200

200

100

100 0

2000 4000 6000 8000 [2Hwater] (µg/g)

0

2000 4000 6000 8000 [2Hwater] (µg/g)

Figure

3500

2100

2800

1400

700

2800

3500

2100

1400

700

2

1

2858 cm-1

2858 cm-1 2848

2848 cm-1

cm-1 3462 cm-1

3076

cm-1

3462 cm-1

3076 cm-1

Hymenaea protera Mid-Oligocene v Miocene (DA)

Hymenaea courbaril modern resin (HC)

0 days

Absorbance (arbitrary units)

Absorbance (arbitrary units)

0 days

30 days 30 days

3421 cm-1

240 days

Metasequoia glyptostroboides modern resin (MG)

240 days

3650 cm-1

Cupressaceae; cf. Parataxodium Cretaceous (GLA) 0 days

0 days

30 days

30 days

240 days

240 days

4000

3500 3000 2500 2000 1500 1000 wavenumber (cm-1)

4000

3500

3000

2500

2000

1500

wavenumber (cm-1)

1000