Three wood isotopic reference materials for δ²H and δ¹³C measurements of plant methoxy groups

Highlights

• Three wood isotopic reference materials were prepared to serve as long-term reference materials for plant methoxy groups.

• Methoxy groups from homogenized wood samples were quantitatively converted to iodomethane.

• δ²H_OCH3 and δ¹³C_OCH3 values of generated iodomethane were calibrated against international reference substances.

• δ²H_OCH3 values of the three wood samples span a relatively wide range (~−280 to −190 mUr).

• The three investigated wood materials HUBG3–5 are ideally suited for long-term usage and inter-laboratory comparison.

Abstract

Methoxy groups (OCH₃) of plants show specific stable carbon and hydrogen isotope patterns that are used for applications in biogeochemical, atmospheric, paleoclimatic and food research. The method of choice for determining stable hydrogen and carbon isotope values of methoxy groups (δ²H_OCH3 and δ¹³C_OCH3 values) is the conversion to gaseous iodomethane (CH₃I) and subsequent measurement by stable isotope ratio mass spectrometry. However, comparative measurements particularly for stable hydrogen isotopes of plant methoxy groups are limited due to the lack of suitable reference materials. We have prepared three batches of powdered wood samples (birch HUBG3, beech HUBG4, and tineo HUBG5) collected from different geographical locations to serve as long-term reference materials for normalization of δ²H_OCH3 but also δ¹³C_OCH3 values. Methoxy contents of the three wood samples range between 4.7 and 5.4%. Methoxy groups from subsamples of the three homogenized wood samples were quantitatively converted to CH₃I and δ²H_OCH3 and δ¹³C_OCH3 values of this CH₃I were calibrated against international reference substances by high-temperature conversion- and elemental analyzer isotope ratio mass spectrometry. The δ²H_OCH and δ¹³C_OCH3 values of HUBG3 and HUBG4 at 1σ uncertainty calibrated to the VSMOW and VPDB isotopic δ-scale, respectively are −272.9 ± 1.5 mUr and −29.40 ± 0.13 mUr; and −239.1 ± 1.4 mUr and −30.17 ± 0.13 mUr, respectively. In addition, the calibrated δ²H_OCH value of HUBG5 is −191.7 ± 0.8 mUr. Whilst the δ²H_OCH3 values of the three wood samples span a relatively wide range (~−280 to −190 mUr) suitable for normalization of δ²H_OCH3 values of most plant samples that have been reported so far, δ¹³C_OCH3 values are restricted to a composition (~−30 mUr) typical of wood methoxy groups. We suggest that the three investigated wood materials HUBG3–5 are ideally suited for long-term usage, inter-laboratory comparison, and together with the two recently reported methyl sulfate salts (HUBG1 and HUBG2) complete a new set of solid methoxy reference materials that cover almost the full range of plant methoxy groups reported so far. Therefore, we recommend replacing liquid CH₃I and instead use the solid reference materials set HUBG1–5 for normalizing δ²H_OCH3 and δ¹³C_OCH3 values to the respective δ-scales.

Introduction

Methoxy groups (OCH₃) of lignin and pectin constitute a significant fraction of the biospheric C₁ methyl (CH₃) pool of plant origin accounting for ~4–6% of wood dry mass (Galbally and Kirstine, 2002; Keppler et al., 2004) and ~1–3% of leaf dry mass (McRoberts et al., 2015). In the past 15 years several studies have shown that methoxy groups of plants have both distinct stable hydrogen (δ²H_OCH3) and carbon (δ¹³C_OCH3) isotope values (Keppler et al., 2004) with reported δ²H_OCH3 and δ¹³C_OCH3 values ranging from −149 mUr (Greule et al., 2010) to −405 mUr (Anhäuser et al., 2018) and −7.1 mUr (Greule et al., 2010) to −77.2 mUr (Keppler et al., 2004), respectively. These specific δ²H and δ¹³C isotope patterns have considerable potential for application as tools for investigations in biogeochemical, atmospheric, paleoclimatic and food research (Greule et al., 2019). For a more detailed overview of its applications in environmental research we refer to previous studies by Anhäuser et al. (2014), Feakins et al. (2013a), Greule et al., 2012, Greule et al., 2010, Hepp et al. (2017), Keppler et al. (2004) and Lee et al. (2019). Please also note that we follow the suggestion by Brand and Coplen (2012) and express isotope delta values in milli-Urey [mUr] (after Urey, 1948) instead of per mil [‰] (1 mUr = 1‰).

Most measurements of site-specific isotopic abundances of organic molecules are accomplished by nuclear magnetic resonance (NMR) spectroscopy, which requires relatively large masses of chemically pure samples. This complicates the application of this technique to most natural samples. By combining a long-established technique for the derivatization of plant methoxy groups to gaseous iodomethane (CH₃I) using hydriodic acid (Zeisel, 1885) with conventional methods for continuous flow isotope ratio mass spectrometry (CF-IRMS), it is possible to determine the ¹³C and ²H isotope compositions of plant methoxy groups with high precision and no apparent isotopic fractionation on milligram quantities of bulk tissue (Greule et al., 2009, Greule et al., 2008).

Stable isotope analyses require reference materials to normalize stable isotope ratios on the respective δ-scale (Brand et al., 2014; Carter and Barwick, 2011; Meier-Augenstein and Schimmelmann, 2019; Werner and Brand, 2001). Highest accuracy is achieved when workflows adhere to the principle of “identical treatment of sample and reference material” (IT) in preparation and analysis, which requires that samples are chemically analogous to standards of known isotopic composition (Meier-Augenstein and Schimmelmann, 2019; Schimmelmann et al., 2016; Werner and Brand, 2001). When selecting reference materials for IRMS analyses further criteria have to be considered as outlined in detail by several previous studies, such as isotopic homogeneity, unchanging stable isotopic composition over time, chemical similarity to the samples, easy preparation, storage and handling as well as no health risk (Brand et al., 2014; Carter and Barwick, 2011; Meier-Augenstein and Schimmelmann, 2019; Schimmelmann et al., 2016; Werner and Brand, 2001). Most published stable isotope values of methoxy groups analyzed by IRMS were normalized to the respective δ-scale using liquid CH₃I as the reference material. However, this practice violates the above principles because liquid CH₃I, unlike the sample, is typically not heated at 130 °C with HI for derivatization (Greule et al., 2019). Moreover, δ²H and δ¹³C values of commercially available CH₃I have restricted ranges (δ²H: −66 to −179 mUr; δ¹³C: −46 to −70 mUr) (Feakins et al., 2013b; Greule et al., 2019; Keppler et al., 2007) that do not bracket typical δ²H_OCH3 and δ¹³C_OCH3 values of plant methoxy groups.

For these reasons, we have recently investigated two methyl sulfate salts (HUBG1 and HUBG2), for their suitability to serve as methoxy reference materials (Greule et al., 2019). These methyl sulfate salts have been demonstrated to be highly suitable to serve as reference materials to normalize δ¹³C_OCH3 values, as they are treated in an identical manner as the sample and span a relatively wide range of δ¹³C values (HUBG1: −50.31 ± 0.16 mUr; HUBG2: +1.60 ± 0.12 mUr) covering most of the natural δ¹³C_OCH3 values of terrestrial plant methoxy groups that have been reported so far (Greule et al., 2019). However, these methyl sulfate salts have a restricted range of δ²H_OCH3 values (HUBG1: −144.5 ± 1.2 mUr; HUBG2: −102.0 ± 1.3 mUr) and thus are of limited use for normalizing δ²H_OCH3 values of plant methoxy groups, which have δ²H_OCH3 values in the range of −150 mUr to −300 mUr (Anhäuser et al., 2015; Greule et al., 2015, Greule et al., 2012; Keppler and Hamilton, 2008; Riechelmann et al., 2017). To properly bracket δ²H values of plant methoxy groups, additional reference materials, with lower δ²H_OCH3 values, are needed. To meet the requirements for stable isotope analysis (Brand et al., 2014; Carter and Barwick, 2011; Schimmelmann et al., 2016; Werner and Brand, 2001), these reference materials must be available in sufficiently large quantities to be applied on a long-term basis and be readily distributable for inter-laboratory comparisons.

To fulfill this need, we have investigated the suitability of three woods from different geographical locations as long-term reference materials for normalization of δ²H_OCH3 values. Wood contains relatively high concentrations of methoxy groups in the range of 4–6% (dry weight basis), primarily originating from lignin. The δ²H values of these lignin methoxy groups are highly depleted in ²H compared to the δ²H values of the source water due to a large, uniform apparent biosynthetic isotopic fractionation of −213 ± 17 mUr (Anhäuser et al., 2017b; Keppler et al., 2007). Since the stable hydrogen isotope values of tree source waters generally range from ~0 to −150 mUr, depending on the geographical origin, δ²H_OCH3 values of wood lignin can be found in the range of −200 to −350 mUr. Hence, woods from different climatic and geographic regions are ideally suited to serve as reference materials for plant methoxy groups. Wood meets additional requirements for a suitable reference material because it is a non-toxic, solid, stable substance amenable to long-term storage and safe distribution, and can be treated identically to the samples when analyzing stable isotope signatures of methoxy groups.

Here we describe the preparation and analysis of three wood samples (HUBG3, HUBG4 and HUBG5) and make them available as interlaboratory reference materials for δ²H_OCH3 and δ¹³C_OCH3 measurements of plant methoxy groups. We detail the procedures (Fig. 1) used to verify the homogenization of these materials, and the techniques used to determine their δ²H_OCH3 and δ¹³C_OCH3 values with high confidence.