• authenticity;
  • deuterium;
  • geographical origin;
  • lignified material;
  • nonexchangeable hydrogen;
  • paleoclimate proxy;
  • stable hydrogen isotopes;
  • wood


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  • • 
    Stable isotope ratios of organic compounds are valuable tools for determining the geographical origin, identity, authenticity or history of samples from a vast range of sources such as sediments, plants and animals, including humans.
  • • 
    Hydrogen isotope ratios (δ2H values) of methoxyl groups in lignin from wood of trees grown in different geographical areas were measured using compound-specific pyrolysis isotope ratio mass spectrometry analysis.
  • • 
    Lignin methoxyl groups were depleted in 2H relative to both meteoric water and whole wood. A high correlation (r2 = 0.91) was observed between the δ2H values of the methoxyl groups and meteoric water, with a relatively uniform fractionation of –216 ± 19‰ recorded with respect to meteoric water over a range of δ2H values from –110 in northern Norway to +20‰ in Yemen. Thus, woods from northern latitudes can be clearly distinguished from those from tropical regions. By contrast, the δ2H values of bulk wood were only relatively poorly correlated (r2 = 0.47) with those of meteoric water.
  • • 
    Measurement of the δ2H values of lignin methoxyl groups is potentially a powerful tool that could be of use not only in the constraint of the geographical origin of lignified material but also in paleoclimate, food authenticity and forensic investigations.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References

Stable hydrogen isotope analysis is recognized as a powerful tool in global climate research. Variations in the δ2H and δ18O values of water in ice cores are commonly used to reconstruct past climatic temperature fluctuations, and detailed paleoclimate records were published as early as the 1960s and 1970s (Dansgaard et al., 1969; Epstein et al., 1970)

Furthermore, δ2H values of wood in the annual growth rings of trees can provide the information necessary to reconstruct past climates and to assist in ecophysiological research (Schiegl & Vogel, 1970; Schiegl, 1974; Epstein et al., 1976; Switsur et al., 1996; McCarroll & Loader, 2004; Filot et al., 2006). Site- and compound-specific δ2H values of biomarkers accumulated in sediments are increasingly employed as paleoclimatic and paleohydrological proxies (Sternberg, 1988; Andersen et al., 2001; Sauer et al., 2001; Huang et al., 2002; Sachse et al., 2004).

The general principles governing hydrogen isotopic fractionation are now well delineated. The primary hydrogen source of all organic compounds in the biosphere is water and, in the case of plant biomass, plant leaf water. Hydrogen incorporated into photosynthetic products during primary reduction steps is highly depleted in 2H. However, a significant proportion of these hydrogen atoms are exchanged with hydrogen atoms of water during subsequent metabolism, leading to a secondary enrichment. Hayes (2001) has reviewed fractionation of carbon and, to a lesser extent, hydrogen isotopes in organic compounds produced by a single organism, principally in relation to their enrichment or depletion relative to the total biomass of carbon or hydrogen. This work did not attempt to consider in any depth the isotopic pattern within individual compounds. However, recently an extensive overview and discussion of both intermolecular and intramolecular nonrandom 2H distributions in natural compounds was provided by Schmidt et al. (2003), who also considered their importance in the elucidation of biosynthetic pathways and their potential to assist in assigning an origin to organic compounds in plants.

Natural archives such as ice cores, peatlands and sediments are already widely utilized in climate research, but trees appear to offer an additional very promising method for reconstructing precisely detailed annual climatic histories, not only from living but also from subfossil trees (Schiegl, 1974; Mayr et al., 2003). Indeed, by careful sampling of wood within annual growth rings it may be possible to extract climatic information at a much higher temporal resolution (Barbour et al., 2002; Loader et al., 1995). Early studies on tree rings analysed whole wood but, when Epstein et al. (1976) and Wilson & Grinsted (1977) demonstrated that the wood components lignin, cellulose and hemicellulose differed significantly in isotopic composition, investigations focused on cellulose. One of the main advantages of using cellulose is that it is measured as cellulose nitrate and entirely reflects nonexchangeable hydrogen in this plant component (Epstein et al., 1976). From that time forward, many researchers settled on nonexchangeable hydrogen in cellulose as the best proxy for source water. However, a general problem associated with the determination of the δ2H values of marker compounds for the study of climate and environmental conditions, as well as for investigation of food authenticity investigations, is the isolation of the pure compound for analysis by isotope ratio mass spectrometry (IRMS). Exploitation of components of wood as markers, in particular, has been restricted by the very labour-intensive and time-consuming preparation of samples (e.g. cellulose nitrate). Any improvements to the efficiency of sample preparation would be of immense value as these would not only allow an increase in the number of sampling points within an individual series but also permit replication of time series. Ideally, for accurate determination of the hydrogen isotope signature, the following criteria should apply to the compound or the chemical moiety within a compound:

  • • 
    hydrogen atoms that are nonexchangeable throughout the sample history and during sample preparation and analysis, so that the isotope signature measured reflects the pristine isotopic fractionation of the compound;
  • • 
    high natural abundance in samples;
  • • 
    simple extraction method;
  • • 
    rapid and straightforward sample preparation;
  • • 
    rapid and reliable analysis of the compound;
  • • 
    no isotopic fractionation during any stage of sample processing or analysis.

On the basis of these criteria, we suggest that for 2H analysis of wood the methoxyl groups of lignin offer great potential as target chemical moieties. Lignin, a major component of wood (up to 31%), is produced by secondary metabolic processes and laid down in cellulose cell walls, imparting strength and rigidity to the structure. It can be considered a polymer of three different precursors, termed monolignols, that differ in the degree of methylation of the aromatic ring. The monolignols are ρ-coumaryl alcohol, which has no methoxyl groups, coniferyl alcohol, which has a methoxyl group attached to C-3 of the aromatic ring, and synapyl alcohol, which has methoxyl groups attached to both C-3 and C-5 of the aromatic ring. For more information on lignin structures and biosynthesis, we refer readers to the review by Boerjan et al. (2003). Overall, methoxyl groups can constitute up to 20% of lignin and whole wood can possess up to 6% methoxyl content. Methylation of hydroxyl groups attached to the aromatic ring is catalysed by O-methyltransferases using S-adenosylmethionine (SAM) as the methyl donor. This biochemical origin from SAM has interesting consequences as regards the isotopic composition of the methoxyl groups of lignin. Recent work has shown that the methoxyl groups of lignin and pectin, which together constitute the bulk of the C1 plant pool, have a carbon isotope signature significantly depleted in 13C (Keppler et al., 2004). The depletion between bulk plant biomass and plant methoxyl pools ranges from –11 to –46‰, with the pectin C1 pool generally more depleted than the lignin C1 pool.

Lignin methoxyl groups are considered to be stable; that is, the hydrogen atoms of the methoxyl moiety do not exchange with those of plant water during ongoing metabolic reactions in the plant. Thus, the initial δ2H value of the methoxyl groups of lignin in woody tissue at formation is retained throughout the lifetime of the tree and in preserved tissue. The methoxyl content of lignin in wood is usually determined by the Zeisel method (Zeisel, 1885) using the reaction between methyl ethers and hydroiodic acid (HI) to form methyl iodide (CH3I). Exploiting this reaction (Fig. 1) for the measurement of δ2H values of lignin methoxyl groups ensures that the isotope signal is preserved throughout the analytical procedure, as no isotopic exchange occurs between the methyl groups and HI, and no isotopic fractionation in the course of CH3I formation is observed. In this paper, we report measurements of δ2H values of both whole wood and lignin methoxyl groups from wood samples sourced from various geographical locations which establish a relationship between their isotopic signatures and that of the local precipitation. We demonstrate that the methoxyl groups of lignin meet all the criteria listed above for ideal target chemical moieties for stable hydrogen isotope measurements in wood.


Figure 1. Mechanism of the reaction between lignin methoxyl and hydroiodic acid (modified after Goto et al., 2006).

Download figure to PowerPoint

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References

Plant material and field sites

Study of annual variations of early and late woods was conducted on a section (Fig. 2) of an oak (Quercus robur) tree from Killarney, Ireland (close to one of the International Atomic Energy Agency (IAEA) measurement stations) which had been felled in 1994. A small chisel was used for separation of annual early and late woods.


Figure 2. Sections of an oak (Quercus robur) tree collected from Killarney, Ireland in 1995. (a) Annual growth rings from 1968 to 1993. (b) Enlarged section showing 3 yr of growth from 1985 to 1987; early and late wood formation clearly visible.

Download figure to PowerPoint

In 2004, samples of wood were collected from a variety of species growing in their natural environments at a number of geographical locations around the world (Table 1). Samples in Scandinavia were collected from living trees using a 8-mm-diameter borer and usually obtained across a 10–40-yr period. Samples from other regions were collected as wood slices. Sites were selected on the basis of sample availability and also to cover a broad range of δ2H values for meteoric water. All samples were ground to pass through a 2-mm sieve.

Table 1.  Wood collection sites and related isotopic composition of meteoric water
Common name (species)Country and site codeGeographical locationAltitude (m)δ2H meteoric water (OIPC)95% CIδ2H meteoric water (GNIP)
  1. Longitude, latitude and altitude were determined using a topographical map except for those marked with superscript ‘a’, which are approximations, as the geographical origin of the samples was not precisely known. Hydrogen isotope ratios (δ2H values) for meteoric water and 95% confidence intervals (CIs) were calculated using the Online Isotope Precipitation Calculator (OIPC), version 2.1. In the instances where wood sampling locations were in the broader vicinity of an International Atomic Energy Agency (IAEA) measurement station, the weighted annual values from that station using the International Atomic Energy Agency–Global Network of Isotopes in Precipitation (IAEA-GNIP) data set (International Atomic Energy Agency, 2001) are shown. All δ values are given in ‰ vs VSMOW.

Paper birch (Betula papyrifera)Canada CA146°01′N; 66°40′W105 –72 3
Sweet osmanthus (Osmanthus fragans)China CH131°N; 116°Ea450 –45 8–54.9 Nanjing (32°18′N; 118°18′E); –46.8 Wuhan (30°62′N; 114°13′E)
European birch (Betula pendula)Finland FI169°11′N; 20°75′E330–10812–118 Naimakka (68°41′N; 22°21′E)
European birch (Betula pendula)Finland FI268°49′N; 20°24′E450–10911–118 Naimakka (68°41′N; 22°21′E)
European birch (Betula pendula)Finland FI3, FI467°92′N; 24°15′E610–10810–118 Naimakka (68°41′N; 22°21′E)
Norway spruce (Picea abies)      
Cherry (Prunus sp.)Germany GE151°N; 7.5°E330 –61 3–52.1 Koblenz (51°93′N; 10°25′W) –60 Weil (47°3′N; 7°4′E)
Dark red meranti (Shorea sp.)Indonesia IN11°S; 114°Ea500 –4010 
English oak (Quercus robur)Ireland IR152°N; 9°33′W 50 –5111–36.2 Valentia (50°35′N; 7°58′E)
Croton (Croton socotrana)Yemen YE112°30′N; 53°50′Ea 50  816 
Tamarisk (Tamarisk aphylla)Yemen YE214°25′N; 48°59′E100  1617 
Geronggang (Cratoxylum sp.)Malaysia MA16°N; 100°Ea150 –36 5–27.8 Ko Samui (13°17′N; 100°03′E) –46.9 Singapore (1°35′S; 103°9′E)
English oak (Quercus robur)Northern Ireland54°2′N; 6°W 40 –59 7 
European ash (Fraxinus excelsior)NI1, NI2     
European birch (Betula pendula)Norway NO159°2′N; 10°28′E100 –7712 
Scots pine (Pinus sylvestris)Norway NO260°48′N; 11°04′E120 –8212 
European birch (Betula pendula)Norway NO361°58′N; 9°33′E400 –8813 
European birch (Betula pendula)Norway NO462°22′N; 9°40′E980 –9913
Norway spruce (Picea abies)Norway NO562°42′N; 9°55′E650 –9513
Scots pine (Pinus sylvestris)Norway NO663°60′N; 11°30′E 50 –9012
Norway spruce (Picea abies)Norway NO764°13′N; 11°55′E280 –9512
Norway spruce (Picea abies)Norway NO864°57′N; 13°09′E400 –9812
Norway spruce (Picea abies)Norway NO9,65°04′N; 13°25′E310 –9812
European birch (Betula pendula)NO10     
European birch (Betula pendula)Norway NO1165°58′N; 13°27′E 60 –9512
Norway spruce (Picea abies)Norway NO1266°24′N; 14°39′E120 –9812
European birch (Betula pendula)Norway NO1366°33′N; 15°18′E620–10511
Norway spruce (Picea abies)Norway NO1467°53′N; 15°58′E 75–10012
European ash (Fraxinus excelsior)Norway NO1568°27′N; 17°29′E 30–10112
European birch (Betula pendula)Norway NO1668°40′N; 17°45′E150–10312
Norway spruce (Picea abies)Norway NO1769°00′N; 18°33′E 80–10412
European birch (Betula pendula)Norway NO1869°21′N; 20°06′E370–10912

Treatment of wood with HI: conversion of lignin methoxyl groups to CH3I

Hydrogen isotope signatures of lignin methoxyl groups were assessed by measuring δ2H values of CH3I released upon treatment of wood samples with HI. A volume of 1 ml of 57% HI was added to the sample (10–20 mg of milled material) in a reaction vial (5 ml). The vials were sealed with caps containing polytetrafluoroethylene (PTFE)-lined silicone septa and incubated for 30 min at 105°C. After the heating cycle, the vials were placed at room temperature for 15 min before a sample of headspace was removed and directly injected into the analytical system. Standard solutions of CH3I were treated similarly to the wood samples in order to check for isotope fractionation. No measurable chemical fractionation of hydrogen isotopes in CH3I was observed during the analytical procedure and no significant changes in δ2H values of CH3I were observed when a series of wood samples were incubated for varying periods from 5 to 180 min (Fig. 3). It is worthy of mention that during treatment of wood with HI there is some potential for CH3I production from wood components other than lignin, which may interfere with the lignin methoxyl signal. For example, Goto et al. (2005) demonstrated that the methoxyl-free compound cellulose can form CH3I when treated with HI. However as the amounts produced were some three orders of magnitude lower when compared with wood samples, we would not expect CH3I formation from cellulose to have any measurable effect on the δ2H values of CH3I reported in this study. However, when plant material with a very much lower lignin content is used, CH3I formation from carbohydrates should be considered.


Figure 3. Hydrogen isotope ratios (δ2H values) of CH3I released from oak (Quercus robur) wood (NI1) over time following treatment with hydroiodic acid (HI). Data show the mean of two measurements, except for those with error bars, which are the means of at least three independent replicate measurements (n = 3–15, 1σ).

Download figure to PowerPoint

Isotope ratio monitoring mass spectrometry for determination of δ2H values of CH3I

System configuration  Hydrogen isotopic ratios of CH3I were measured using gas chromatography pyrolysis isotope ratio mass spectrometry analysis (GC-P-IRMS). A Hewlett Packard HP6890 GC (Agilent Technologies, Palo Alto, CA, USA) was interfaced with a Micromass Isoprime isotope ratio mass spectrometer via a 1050°C pyrolysis reactor (GV Instruments, Manchester, UK). A tank of ultra-high-purity hydrogen with a known δ2H value of –718.7‰ (Vienna Standard Mean Ocean Water; VSMOW) was used as an internal reference gas standard. The H3+ factor determined daily during the 2-month measurement period was in the range 9.1–10.4. The internal precision of δ2H was ±4‰ (1σ).

Throughout this paper, the conventional ‘delta’ notation, which expresses the isotopic composition of a material relative to that of a standard on a per mil (‰) deviation basis, is used. Values of δ2H (‰) relative to that for VSMOW are defined by the equation:

  • δ2H (‰) = [(2H/1H)sample/(2H/1H)standard – 1] × 1000‰ (Eqn 1)

Sample analysis  Headspace gas (500 µl) was injected into the gas chromatograph, which was equipped with a DB5-MS column (30 m, internal diameter (ID) 0.32 mm, film thickness 0.5 µm). The injection port temperature was set at 250°C in the split mode (10 : 1). The oven was maintained for 4 min at 30°C and then heated at 15°C min−1 to 180°C. Column flow was held constant at 0.8 ml min−1 throughout the analysis. Following GC separation, CH3I was transferred to a 1050°C pyrolysis reactor which contained a 0.65-mm ID quartz tube packed with chromium pellets sieved to a size of ≤ 0.25 mm (Elemental Microanalysis, Devon, UK). As the sample gas passed through the furnace, a pyrolytic and catalytic reaction resulted in its quantitative conversion to H2 (Morrison et al., 2001). All δ2H values were normalized relative to VSMOW using a CH3I standard. The δ2H values of CH3I were calibrated against international reference substances (IA-R002, IAEA-CH-7 and NBS-22) using the offline temperature conversion–elemental analyser (TC-EA) technique (Iso-Analytical Ltd, Cheshire, UK). The calibrated δ2H value in ‰ vs VSMOW for CH3I was –174 ± 1‰ (n = 4, 1σ). The standard was measured after every third sample injection. If necessary, a drift correction was applied. This methodology achieved a precision, expressed as the average standard deviation for all samples, of 6‰ (1σ).

Bulk δ2H isotope analysis by TC-EA/IRMS

A DeltaPlus-XP High Temperature Conversion/Elemental Analyser (Thermo Electron Corporation, Bremen, Germany) was used for 2H:1H isotope ratio measurement of ground wood samples. Typically, 0.2 mg of sample was weighed into a silver capsule and placed in a desiccator for a week before the samples were introduced into the TC-EA by means of a solid Costech Zero-Blank autosampler (Pelican Scientific Ltd, Alford, UK). The reactor tube was self-packed and comprised an AlsintTM ceramic tube containing a glassy carbon tube, filled with glassy carbon granulate, silver and quartz wool (SerCon, Crewe, UK). Reactor temperature was set to 1425°C while the postreactor GC column was maintained at 85°C. Helium (99.99% purity; Air Products Plc, Walton-on-Thames, UK) pressure was set to 1.45 bar. Data were processed using Isodat NT software, version 2.0 (Thermo Electron Corporation, Bremen, Germany). The run time per analysis was 350 s. Measured 2H:1H isotope ratios are expressed as δ values in [‰] relative to VSMOW.

Isotopic calibration and quality control of TC/EA-IRMS measurement  The working reference gas, H2 (BOC Gases, Guildford, UK), was calibrated against VSMOW using an international reference material, IAEA-CH-7 (IAEA, Vienna, Austria) polyethylene (δ2HVSMOW = –100.3‰), and checked against the international reference materials VSMOW and Standard Light Antarctic Precipitation (SLAP). Cross-checking of the working δ2HVSMOW value of H2 against the international reference material Greenland Ice Sheet Precipitation (GISP) yielded a δ2HVSMOW value for GISP of –194.6‰ (expected: –189.5‰). The H3+ factor was determined on reference H2 gas pulses of different signal size and was found to be 4.43. A batch analysis, comprising 10 samples analysed in triplicate, was preceded and followed by a set of standards, as reported previously (Farmer et al., 2005). Reproducibility of 2H isotope analysis as monitored using the IRMS and laboratory standards was 1.15‰ or better. Each batch was preceded and followed by a blank capsule. Measured δ2H values were normalized according to the method described by Coplen (1988), with Z factors (or ‘stretch’ factors) typically being of the order of 1.08.

δ2H of water samples and the Online Isotope Precipitation Calculator

The Online Isotope Precipitation Calculator (OIPC; accessible at, employing the IAEA database and interpolation algorithms developed by Bowen & Wilkinson (2002) and refined by Bowen & Revenaugh (2003) and Bowen et al. (2005), was used to calculate annual δ2H values of meteoric water from wood sampling sites. Long-term precipitation isotope data were taken from the International Atomic Energy Agency–Global Network of Isotopes in Precipitation (IAEA-GNIP) database (International Atomic Energy Agency, 2001), accessible on the Internet at

Calculation of isotopic fractionation ɛ

The isotopic difference between the δ2H value of the methoxyl groups and the δ2H value of precipitation was calculated using

  • image(Eqn 2)
  • Throughout the paper we use the short form ɛm/w for ɛmethoxyl/water.

Results and Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References

2H values of lignin methoxyl groups in early- and late-season wood from annual growth rings of oak from Killarney in Ireland

Annual δ2H values of lignin methoxyl groups of early-season oak wood for the 25-yr period 1968–1993 (Fig. 2) were in the range of –209 to –242‰, with a mean value of –228‰, while annual values of late-season wood were slightly more depleted, ranging from –211 to –255‰, with a mean value of –237‰ (Fig. 4). Variation (expressed as SD) in the annual δ2H values for early- and late-season wood for the 25-yr period was 10 and 12‰, respectively. The δ2H value measured for lignin methoxyl groups of a composite wood sample representing the entire 25-yr period was –233 ± 4‰ (n = 5, 1σ). The oak tree sampled grew in close vicinity to an IAEA-GNIP sampling station (Valentia, Ireland) where δ2H values of precipitation have been recorded on a nearly monthly basis since the early 1960s. The weighted annual δ2H value for precipitation at this site (International Atomic Energy Agency, 2001) is –36.2‰, and over the last 40 yr weighted monthly values have ranged from ∼–10 to –65‰. The annual variation (expressed as SD) of δ2H values is approximately 11‰, a similar range to that observed for the δ2H values of lignin methoxyl groups in both early- and late-season wood. If we assume –36.2 and –233‰ as mean values of precipitation and lignin methoxyl groups, respectively, we calculate the isotopic fractionation, ɛm/w, to be –205‰. It should be borne in mind that, as δ2H values for precipitation during the growing season are enriched in 2H relative to the weighted annual δ2H value, the actual ɛm/w may be slightly larger. The apparent discrepancy between the observation that the δ2H value of early-season wood is more positive than that of late-season wood and the fact that the primary hydrogen source, leaf water, is relatively enriched in 2H in the summer can be resolved by bearing in mind that the first wood in spring is mainly synthesized from reserve starch transported from the tree roots (Hill et al., 1995; Helle & Schleser, 2004). This starch is also the source of NADPH via the oxidative pentose phosphate cycle before photosynthesis begins. Secondary hydrogen exchange with xylem water results in reserve starch possessing a more positive δ2H value than the immediate products of photosynthesis (McCarroll & Loader, 2004).


Figure 4. Yearly record of hydrogen isotope ratios (δ2H values) of lignin methoxyl groups from early (diamonds) and late (squares) wood of an Irish oak (Quercus robur) tree covering a 25-yr period. Data are single measurements except for those with error bars, which are the means of three independent replicate measurements (< ± 5‰, 1σ). Solid lines, linear regression of early and late wood; dashed line, composite wood.

Download figure to PowerPoint

δ2H values of lignin methoxyl groups in wood samples from various geographical locations and their relation to δ2H values of local precipitation

To further test our hypothesis we collected wood samples from several locations around the world (Table 1, Fig. 5) in order to compare the δ2H values of lignin methoxyl groups with those of local precipitation. Sampling sites were selected in order that a broad range of δ2H values for meteoric water was covered. δ2H values of precipitation were assessed by two methods, and are given in Table 1 together with detailed information on the locations at which samples were collected. In instances where sampling points were in the vicinity of an IAEA measurement station, we show weighted annual values for that station. However, there is only one station (Valentia, Ireland) situated in close proximity to one of our sample points and only a few in the broader vicinity of our other sampling points. Therefore, we decided to use the OIPC ( to calculate δ2H values of local precipitation at collection sites. Sachse et al. (2004) found a good correlation between measured δ2H values for water in lakes on a north–south European transect and those calculated for local precipitation using the OIPC. We make the simplifying assumption that the predicted mean annual δ2H values of the OIPC are a valid proxy for tree source water. However, as trees mainly incorporate hydrogen during the growing season and have highest productivity in spring and late summer, particularly in temperate regions, the δ2H values obtained from the OIPC for these regions may be slightly lower than the actual average values over the growing season. In this work we do not attempt to account for this effect. As can be seen in Table 1, the annual weighted δ2H values of precipitation in the areas where wood was collected ranged from ∼–110 (close to the North Cape, Norway) to ∼+20‰ (Yemen).


Figure 5. Sites of collection of wood samples (n = 30) and the mean annual hydrogen isotope ratio (δ2H value) of precipitation. Because of limited space, not all locations in Scandinavia can be shown. The map was provided by G. J. Bowen and made using the Bowen & Wilkinson (2002) method. VPDB; Vienna Peedee Belemnite.

Download figure to PowerPoint

δ2H values for lignin methoxyl groups and those for bulk wood of samples used in our investigations are presented in Table 2 together with those for local precipitation. Also shown for each location is the isotopic fractionation between water and lignin methoxyl groups. The data are presented graphically in Fig. 6. It is clear that, while lignin methoxyl groups are relatively depleted in 2H with respect to bulk wood, the δ2H values of the methoxyl groups are highly correlated with the OIPC δ2H values of meteoric water (r2 = 0.91). By contrast, the δ2H values for whole wood are only relatively poorly correlated with those of meteoric water (r2 = 0.47).

Table 2.  Hydrogen isotope ratios (δ2H values) of source water (Online Isotope Precipitation Calculator (OIPC) data) and lignin methoxyl groups, bulk biomass and isotopic fractionation, ɛmethoxyl/water m/w), of wood samples from around the world
Country and site codeδ2H source wateraδ2H bulk woodδ2H methoxylɛm/w
  • All δ2H values presented are in ‰. δ2H values for methoxyl groups and bulk wood are means of two samples unless otherwise stated.

  • a

    δ2H values of source water and 95% confidence intervals were calculated using the OIPC, except for IR1, which was taken from the International Atomic Energy Agency–Global Network of Isotopes in Precipitation (IAEA-GNIP) data set.

Norway NO18–109 ± 12–143–292–205
Finland FI2–109 ± 11–145–309–225
Finland FI1–108 ± 12–313–230
Finland FI3–108 ± 10–110–324 ± 4 (3)–243
Finland FI4–108 ± 10–148–320–230
Norway NO13–105 ± 12–139–287–203
Norway NO17–104 ± 12–118–316–237
Norway NO16–103 ± 12–126–313–234
Norway NO15–101 ± 12–108–288–208
Norway NO14–100 ± 12 –96–298–220
Norway NO4–99 ± 13–142–294–216
Norway NO8–98 ± 12 –99–325 ± 8 (3)–251
Norway NO9–98 ± 12 –91–309–234
Norway NO10–98 ± 12–135–287–210
Norway NO12–98 ± 11 –94–289–211
Norway NO5–95 ± 13 –96–309–236
Norway NO7–95 ± 12 –80–311–239
Norway NO11–95 ± 12–131–284–209
Norway NO6–90 ± 12 –77–283–212
Norway NO3–88 ± 13–153–301–234
Norway NO2–82 ± 12 –79–262–196
Norway NO1–77 ± 12–122–270–209
Canada CA1–72 ± 3–132–285–230
Germany GE1–61 ± 3–122–269–222
Ireland NI1–59 ± 7 –89–264 ± 2 (12)–218
Ireland NI2–59 ± 7 –97–244 ± 2 (15)–197
Ireland IR1–51 ± 11 –44–233 ± 4 (5)–205
China CH1–45 ± 8 –76–240–204
Indonesia IN1–40 ± 10 –94–244 ± 14 (3)–213
Malaysia MA1–36 ± 5 –97–220–191
Yemen YE18 ± 16 –66–177 ± 5 (3)–183
Yemen YE216 ± 17 –29–141 ± 8 (3)–155

Figure 6. Hydrogen isotope ratios (δ2H values) for wood lignin methoxyl groups, bulk wood and meteoric water derived from the Online Isotope Precipitation Calculator (OIPC). (a) Measured δ2H values of wood lignin methoxyl groups and bulk wood relative to δ2H values of meteoric water. Error bars for δ2H values of water from the OIPC reflect the calculated 95% confidence interval. Error bars for methoxyl groups are the average standard deviation of all measurements (±6‰, 1σ). Error bars for δ2H values of bulk wood are within the symbols. Solid lines show linear regression. For reference, the 1 : 1 δ2H values (dashed line) of meteoric water are displayed. (b) Measured δ2H values of lignin methoxyl groups relative to δ2H values of meteoric water. The solid line represents constant fractionation, ɛmethoxyl/water m/w), between the OIPC δ2H values of meteoric water (dashed line) and methoxyl groups.

Download figure to PowerPoint

Although early studies used whole wood (Schiegl & Vogel, 1970; Schiegl, 1974), it must be pointed out that nowadays δ2H measurement of bulk wood is not the method of choice in paleoclimate research using tree rings because bulk wood contains a significant portion of exchangeable hydrogen atoms. When Epstein et al. (1976) and Wilson & Grinsted (1977) demonstrated that the main components of wood – lignin, cellulose and hemicellulose – differed in isotopic composition, investigations began to focus on cellulose. This shift has opened up the debate on whether, if whole wood contains the same isotopic record as cellulose, the analysis of individual plant components such as cellulose can be justified given the onerous nature of the extraction process (Barbour et al., 2001; Loader et al., 2003). However, while this argument may be persuasive for 13C and 18O isotope analysis, the relatively poor correlation observed in this study between the δ2H values of whole wood and meteoric water suggests that this approach is not suitable for analysis of hydrogen isotopes. Nevertheless, a complex extraction procedure to isolate a single component, and, in the case of cellulose, nitration to replace exchangeable hydrogen atoms might introduce unquantifiable uncertainties, even if improved correlation coefficients were achieved. The use of lignin methoxyl groups for determination of the environmental δ2H signal in wood as discussed here circumvents such problems because of the simplicity of the sample preparation procedure. In addition, the use of lignin methoxyl groups is a highly specific wood component, presumably because of a more restricted metabolic source of hydrogen.

The mean isotopic fractionation ɛmethoxyl/waterm/w) between methoxyl groups and the modelled water δ2H values of the OIPC was –216 ± 19‰. Assuming constant isotope fractionation between methoxyl groups and meteoric water, we obtain an average fractionation factor (αm/w) of 0.784 (αm/w ≡ (δ2Hm + 1000)/(δ2Hw + 1000), where δ2Hm is the hydrogen isotope ratio of wood methoxyl groups and δ2Hw is the hydrogen isotope ratio of meteoric water). The constant fractionation line obtained using this equation is shown in Fig. 6(b), and it is clear that almost all samples lie within the analytical uncertainty of this line except for a sample from mainland Yemen (YE2), for which the measured δ2H value was slightly higher than that predicted by the equation. The region has a hot, arid tropical climate, with temperatures in summer reaching up to 50°C. From the GNIP map, δ2H values in precipitation show a broad range from c. –20 to +30‰ throughout the year and the weighted annual value (16 ± 17‰) employed in our study from the OIPC fits quite well within that range. However, a plausible explanation for fractionation being lower than calculated is that, in the low relative humidity of this extreme climate, evaporative enrichment of leaf water is much greater than normal. It is known that the leaf-to-air vapour pressure deficit is the dominant control on the degree of leaf enrichment for δ2H (McCarroll & Loader, 2004). Significantly, the sample from the Yemeni island of Socatra, several hundred miles offshore in the Indian Ocean, showed the predicted fractionation, possibly reflecting the higher relative humidity associated with a maritime environment. Interestingly, recent paleoclimate studies using biomarkers from aquatic sediments have found isotope fractionations in a similar range to those we report for lignin methoxyl groups in wood. For example, the isotope fractionation between sedimentary lipids (Sauer et al., 2001), palmitic acid (Huang et al., 2002) and n-alkanes (Sachse et al., 2004) and environmental water was found to be –201 ± 10, –171 ± 15 and –157 ± 13‰, respectively.

Origin and δ2H values of H atoms in O-CH3 groups

The methoxyl groups of the phenylpropanoids originate from position C-3 of serine, and hence from carbohydrates, perhaps even directly from 3-phosphoglycerate. The CH2OH group of serine is transferred by serine hydroxymethyltransferase to tetrahydrofolic acid (FH4) to form N5,N10-methylene-FH4 (‘active formaldehyde’), which is then reduced by N5,N10-methylene-FH4 reductase to N5-methyl-FH4. From here, the methyl group is transferred to homocysteine to form methionine and, after activation of this amino acid to SAM, to the phenolic OH groups of the phenylpropanoids (Boerjan et al., 2003)

The δ2H values of ∼–107‰ (Rinne et al., 2005), and –52‰ (Betson et al., 2006) that have been reported for cellulose extracted from whole wood grown in northern Finland (Pinus sylvestris) and Northern Ireland (Quercus robur), respectively, directly reflect the δ2H values of the carbohydrate pool of these species at these locations. As the C-3 position of serine is generally slightly more depleted (20–50‰) than that of carbohydrates of C3 plants (Zhang et al., 2002; Augusti et al., 2006), we assume δ2H values of around –127 and –72‰ for this position at these locations. N5,N10-methylene-FH4 reductase is a flavoprotein, and Schmidt et al. (2003) have demonstrated that hydrogen transfer by certain flavoproteins is accompanied by large isotope discriminations, explaining the alternating 2H abundance in the even and odd positions, respectively, of fatty acids and the extreme 2H depletion at defined positions of isoprenoids from the deoxyxylulose phosphate pathway. Depletions down to –580‰ in the first case (Billault et al., 2001) and down to –780‰ in the second case (Martin et al., 2004) have been reported. If a depletion between these values (–680‰) is assumed for the third H atom of the O-CH3 groups of the phenylpropanoids, δ2H values of –311 and –275‰ are thus expected for methoxyl groups of wood from northern Finland and Northern Ireland, respectively. These values are in good agreement with mean data reported here for CH3I derived from lignin of whole wood from northern Finland (–317‰) and Northern Ireland (–254‰).


In summary, the methoxyl group of lignin exhibits all the features required of the ideal target moiety for stable hydrogen isotope studies listed in the Introduction. The method employed in measurement of the δ2H values of lignin methoxyl groups, which requires only a small quantity of sample (10–20 mg of wood) and minimal and straightforward sample preparation, clearly has major benefits. However, the greatest advantage for isotope studies is derived from the fact that lignin methoxyl groups contain nonexchangeable hydrogen atoms which can be quantitatively converted to CH3I, avoiding isotopic fractionation. Thus, lignin methoxyl groups of wood may be considered a plant component that truly records δ2H values of meteoric water, with a mean isotopic fractionation that is uniform (–216 ± 19‰) over a range of δ2H values for meteoric water from +20 to –110‰. However, our work should be regarded as a pilot study providing the first δ2H values for lignin methoxyl groups for a limited number of wood samples from around the world. Many more measurements at a higher geographical resolution, both regionally and globally, are essential to provide us with a clearer understanding of the exact relationship between the isotopic compositions of meteoric water and lignin methoxyl groups. Direct measurements of the δ2H values of environmental water available to the plants and comparison with the δ2H values of their lignin methoxyl groups should further help to improve our knowledge. We envisage the δ2H values of lignin methoxyl groups becoming a powerful tool for use not only in the constraint of the geographical origin of lignified material but also in paleoclimatic, origin authenticity and forensic studies.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References

We are grateful to Simone Keppler for her support in collecting wood samples from Scandinavia. We thank Peter and Susan Christie, Peter Silk and H. F. Schöler for providing wood samples from Yemen; Fredericton, Canada and Hau, Germany, respectively. We also thank Gabriel Bowen for providing the global map used in Fig. 4. The European Commission is acknowledged for a Marie Curie Research Training Grant (MCFI-2002-00022) awarded to FK.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgements
  7. References
  • Andersen N, Paul HA, Bernasconi SM, McKenzie JA, Behrens A, Schaeffer P, Albrecht P. 2001. Large and rapid climate variability during the Messinian salinity crisis: Evidence from deuterium concentrations of individual biomarkers. Geology 29: 799802.
  • Augusti A, Betson TR, Schleucher J. 2006. Hydrogen exchange during cellulose synthesis distinguishes climatic and biochemical isotope fractionations in tree rings. New Phytologist 172: 490499.
  • Barbour MM, Andrews T, Farquhar GD. 2001. Correlations between oxygen isotope ratios of wood constituents of Quercus and Pinus samples from around the world. Australian Journal of Plant Physiology 28: 335348.
  • Barbour MM, Walcroft AS, Farquhar GD. 2002. Seasonal variation in delta C-13 and delta O-18 of cellulose from growth rings of Pinus radiata. Plant, Cell & Environment 25: 14831499.
  • Betson TR, Augusti A, Schleucher J. 2006. Quantification of deuterium isotopomers of tree-ring cellulose using nuclear magnetic resonance. Analytical Chemistry 78: 84068411.
  • Billault I, Guiet S, Mabon F, Robins R. 2001. Natural deuterium distribution in long-chain fatty acids is nonstatistical: a site-specific study by quantitative H-2 NMR spectroscopy. Chembiochem 2: 425431.
  • Boerjan W, Ralph J, Baucher M. 2003. Lignin biosynthesis. Annual Review of Plant Biology 5: 519546.
  • Bowen GJ, Revenaugh J. 2003. Interpolating the isotopic composition of modern meteoric precipitation. Water Resources Research 39: 13.
  • Bowen GJ, Wassenaar LI, Hobson KA. 2005. Global application of stable hydrogen and oxygen isotopes to wildlife forensics. Oecologia 143: 337348.
  • Bowen GJ, Wilkinson B. 2002. Spatial distribution of delta O-18 in meteoric precipitation. Geology 30: 315318.
  • Coplen TB. 1988. Normalization of oxygen and hydrogen isotope data. Chemical Geology 72: 293297.
  • Dansgaard W, Johnsen SJ, Moeller J, Langway CC Jr. 1969. One thousand centuries of climatic record from Camp Century on Greenland ice sheet. Science 166: 377380.
  • Epstein S, Sharp RP, Gow AJ. 1970. Antarctic ice sheet: stable isotope analyses of Byrd Station cores and interhemispheric climatic implications. Science 168: 15701572.
  • Epstein S, Yapp CJ, Hall JH. 1976. Determination of D-H ratio of non-exchangeable hydrogen in cellulose extracted from aquatic and land plants. Earth and Planetary Science Letters 30: 241251.
  • Farmer NL, Meier-Augenstein W, Kalin RM. 2005. Stable isotope analysis of safety matches using isotope ratio mass spectrometry – a forensic case study. Rapid Communications in Mass Spectrometry 19: 31823186.
  • Filot MS, Leuenberger M, Pazdur A, Boettger T. 2006. Rapid online equilibration method to determine the D/H ratios of non-exchangeable hydrogen in cellulose. Rapid Communications in Mass Spectrometry 20: 33373344.
  • Goto H, Koda K, Tong GL, Matsumoto Y, Meshitsuka G. 2005. Formation of methyl iodide from methoxyl-free compounds by hydriodic acid treatment. Journal of Wood Science 51: 312314.
  • Goto H, Koda K, Tong GL, Matsumoto Y, Meshitsuka G. 2006. Interference of carbohydrates in the determination of the methoxyl content of lignin in woody samples. Journal of Wood Chemistry and Technology 26: 8193.
  • Hayes JM. 2001. Fractionation of carbon and hydrogen isotopes in biosynthetic processes. In: Valley JW & Cole D, eds. Stable isotope geochemistry. Washington, DC, USA: Mineralogical Society of America, Vol. 43, 225277.
  • Helle G, Schleser GH. 2004. Beyond CO2-fixation by Rubisco – an interpretation of 13C/12C variations in tree rings from novel intra-seasonal studies on broad-leaf trees. Plant, Cell & Environment 27: 367380.
  • Hill SA, Waterhouse JS, Field EM, Switsur VR, Aprees T. 1995. Rapid recycling of triose phosphates in oak stem tissue. Plant, Cell & Environment 18: 931936.
  • Huang YS, Shuman B, Wang Y, Webb T. 2002. Hydrogen isotope ratios of palmitic acid in lacustrine sediments record late quaternary climate variations. Geology 30: 11031106.
  • International Atomic Energy Agency. 2001. GNIP maps and animations.
  • Keppler F, Kalin RM, Harper DB, McRoberts WC, Hamilton JTG. 2004. Carbon isotope anomaly in the major plant C1 pool and its global biogeochemical implications. Biogeosciences 1: 123131.
  • Loader NJ, Robertson I, McCarroll D. 2003. Comparison of stable carbon isotope ratios in the whole wood, cellulose and lignin of oak tree-rings. Palaeogeography Palaeoclimatology Palaeoecology 196: 395407.
  • Loader NJ, Switsur VR, Field EM. 1995. High-resolution stable isotope analysis of tree rings: implications of ‘microdendroclimatology’ for palaeoenvironmental research. Holocene 5: 457460.
  • Martin GJ, Lavoine-Hanneguelle S, Mabon F, Martin ML. 2004. The fellowship of natural abundance 2H isotopomers of monoterpenes. Phytochemistry 65: 28152831.
  • Mayr C, Frenzel B, Friedrich M, Spurk M, Stichler W, Trimborn P. 2003. Stable carbon- and hydrogen-isotope ratios of subfossil oaks in Southern Germany: methodology and application to a composite record for the Holocene. Holocene 13: 393402.
  • McCarroll D, Loader NJ. 2004. Stable isotopes in tree rings. Quaternary Science Reviews 23: 771801.
  • Morrison J, Brockwell T, Merren T, Fourel F, Phillips AM. 2001. On-line high-precision stable hydrogen isotopic analyses on nanoliter water samples. Analytical Chemistry 73: 35703575.
  • Rinne KT, Boettger T, Loader NJ, Robertson I, Switsur VR, Waterhouse JS. 2005. On the purification of alpha-cellulose from resinous wood for stable isotope (H, C and O) analysis. Chemical Geology 222: 7582.
  • Sachse D, Radke J, Gleixner G. 2004. Hydrogen isotope ratios of recent lacustrine sedimentary n-alkanes record modern climate variability. Geochimica et Cosmochimica Acta 68: 48774889.
  • Sauer PE, Eglinton TI, Hayes JM, Schimmelmann A, Sessions AL. 2001. Compound-specific D/H ratios of lipid biomarkers from sediments as a proxy for environmental and climatic conditions. Geochimica et Cosmochimica Acta 65: 213222.
  • Schiegl WE. 1974. Climatic significance of deuterium abundance in growth rings of Picea. Nature 251: 582584.
  • Schiegl WE, Vogel JC. 1970. Deuterium content of organic matter. Earth and Planetary Science Letters 7: 307313.
  • Schmidt HL, Werner RA, Eisenreich W. 2003. Systematics of 2H patterns in natural compounds and its importance for the elucidation of biosynthetic pathways. Phytochemistry Reviews 2: 6185.
  • Sternberg LDL. 1988. D/H ratios of environmental water recorded by D/H ratios of plant lipids. Nature 333: 5961.
  • Switsur VR, Waterhouse JS, Field EM, Carter AHC. 1996. Climatic signals from stable isotopes in oak tree rings from East Anglia, Great Britain. In: DeanJS, MekoDM, SwetnamTW, eds. Tree rings, environment and humanity. Tucson, AZ, USA: Radiocarbon, 637645.
  • Wilson AT, Grinsted MJ. 1977. C-12-C-13 in cellulose and lignin as palaeothermometers. Nature 265: 133135.
  • Zeisel S. 1885. Über ein Verfahren zum Quantitativen Nachweis von Methoxyl. Monatshefte für Chemie 6: 989996.
  • Zhang BL, Billault I, Lo XB, Mabon F, Remaud G, Martin ML. 2002. Hydrogen isotopic profile in the characterization of sugars. Influence of the metabolic pathway. Journal of Agricultural and Food Chemistry 50: 15741580.