Environ. Sci. Technol. 2008, 42, 6837–6842
Abiotic Methyl Bromide Formation from Vegetation, and Its Strong Dependence on Temperature A S H E R W I S H K E R M A N , * ,† SARAH GEBHARDT,† COLIN W. MCROBERTS,‡ JOHN T. G. HAMILTON,‡ JONATHAN WILLIAMS,† AND FRANK KEPPLER† Max Planck Institute for Chemistry, J. J. Becherweg 27, 55128 Mainz, Germany, and Agri-Food and Biosciences Institute, Newforge Lane, Belfast BT9 5PX, U.K.
Received February 10, 2008. Revised manuscript received July 5, 2008. Accepted July 6, 2008.
Methyl bromide (CH3Br) is the most abundant brominated organic compound in the atmosphere. It is known to originate from natural and anthropogenic sources, although many uncertainties remain regarding strengths of both sources and sinks and the processes leading to its formation. In this study a potential new CH3Br source from vegetation has been examined, analogous to the recently discovered abiotic formation of methyl chloride from plant pectin. Several plant samples with known bromine content, including ash (Fraxinus excelsior), saltwort (Batis maritima), tomato reference material (NIST-1573a), hay reference material (IAEA V-10), and also bromine enriched pectin, were incubated in the temperature range of 25-50 °C and analyzed for CH3Br emission using gas chromatography/mass spectrometry. All plant samples inspected showed an exponential increase in CH3Br emission as a function of temperature increase, i.e., emissions were observed to approximately double with every 5 °C rise in temperature. Next to temperature, it was found that emissions of CH3Br were also dependent on the bromine content of the plants. The highest CH3Br release rates were found for the saltwort which contained the highest bromine concentration. Arrhenius plots confirmed that the observed emissions were from an abiotic origin. The contribution of abiotic CH3Br formation from vegetation to the global budget will vary geographically as a result of regional differences in both temperature and bromide content of terrestrial plants.
Introduction Methyl bromide is the most abundant brominated organic compound in the atmosphere with a global average tropospheric mixing ratio of ca. 8 pptv. Its abundance and relatively long lifetime, around 0.7 years (1), make it the largest carrier of bromine to the lower stratosphere. On a per-atom basis, bromine is 50-60-fold more effective than chlorine in destroying ozone (2), and CH3Br is presently thought to be responsible for approximately 15% of halogen-catalyzed ozone destruction in the upper atmosphere. Methyl bromide * Corresponding author phone: +49 6131 305263; fax: +49 6131 305388; e-mail:
[email protected]. † Max Planck Institute for Chemistry. ‡ Agri-Food and Biosciences Institute. 10.1021/es800411j CCC: $40.75
Published on Web 08/13/2008
2008 American Chemical Society
is known to have natural and anthropogenic sources, including biomass burning, ocean emissions, terrestrial ecosystems emissions, and industrial and agricultural usage (1). Due to its high ozone destruction potential its use has been restricted under the Montreal Protocol. However, many uncertainties persist regarding the strengths of both sources and sinks and the processes leading to its formation. Thus the atmospheric budget still remains out of balance with approximately 25% of the sources not yet accounted for. Recent global modeling studies suggest that emissions from tropical vegetation and biomass burning might make a more significant contribution to the atmospheric CH3Br load than previously thought (3). Plants are known to act, in varying degrees, as both a source and sink of methyl bromide (4, 5). Results from a number of field studies show that diverse ecosystems such as, temperate grasslands (6), coastal tundra (7), mangroves (8), rice paddies (9, 10), salt marshes (11, 12), and peatlands (13-15) can emit CH3Br to the atmosphere. Thus in view of the range of vegetation now known to produce CH3Br and the uncertainties attached with the estimation of their strength, it is considered that an abiotic source from dry plant material and litter could provide a significant contribution toward balancing the atmospheric budget. While field observations indicate the net effect occurring in various ecosystems, the underlying mechanism of CH3Br production in plants and its dependence on parameters such as ambient temperature and moisture, remains poorly understood. Methyl transferases have been suggested to be the main enzyme-mediated reaction for the biosynthesis of methyl halides by plants (16, 17), by transferring a methyl group from S-adenosylmethionine (SAM) to a halide ion (18). An abiotic route for CH3Br formation from methoxy phenols in soil has been postulated (19) although its environmental importance remains unknown. Another process, described by Hamilton et al. (20), demonstrated abiotic release of CH3Cl from senescent or dead plant material in the temperature range of 30-350 °C. In that study, it was suggested that a widespread plant structural component, pectin, was the methyl donor, reacting readily with chloride ion to form CH3Cl. Pectin is a polysaccharide composed primarily of partially esterified R-(1-4) linked galacturonic acid units and normally comprises 7-23% of the plant leaf cell wall (21). Using stable isotope techniques Keppler et al. (22) confirmed the central role of plant methoxyl groups in abiotic methyl halide formation. Moreover, in a later study by the same group (23) it was suggested that abiotic methylation of chloride in terrestrial ecosystems may be the largest source of atmospheric CH3Cl. In that study it was also suggested that pectin and lignin methoxyl groups might be the organic precursor compound for CH3Cl formation during biomass burning. Studies thus far on the abiotic process of methyl halide production from vegetation (20, 22, 23) have revealed that the most crucial parameters for emissions are temperature and plant halide content. Natural bromine concentrations in plants generally vary between 1 and 40 mg/kg (24) although in halophytes they can reach up to 740 mg/kg (25). Although bromine is not an essential element for plant growth, it is easily absorbed by plants and is found in almost all plant tissue (26-29). Bromine uptake can occur via roots and leaves (30-33). Dry and wet deposition is an important source of halogens to terrestrial ecosystems in coastal environments (34). Bromine concentration in seawater is ∼65 mg/L, and the seawater Cl/Br molar ratio is ∼650, whereas in precipitation the ratio is ∼845 (25). Marine aerosols, derived from sea spray, are known to contain VOL. 42, NO. 18, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6837
FIGURE 1. Methyl bromide emission from pectin enriched in bromine.
FIGURE 2. Temperature dependence of CH3Br and CH3Cl emissions for saltwort material (Batis maritima). Error bars show SD for triplicate samples ((1σ).
TABLE 1. Content of bromine, chlorine and methoxyl groups in plant material and pectina sample
bromine (mg/kg dw)
hay (IAEA V-10) ash tomato (NIST 1573a) saltwort pectin + KBr
8b 34.4c 1335d 872d 632d
d
a ND, not determined. XRF. e Becker et al. (51).
b
chlorine (mg/kg dw)
% methoxyl
ND 7800b 6540e 186000d ND
IAEA laboratory data
2.38 1.17 0.89 0.81 7.95 c
INAA.
halogens in different relative proportions from seawater, with a typical Cl/Br molar ratio of ∼294 (35). The aim of this study was to investigate abiotic formation of CH3Br from dried plant matter with differing bromine content across the ambient temperature range of 25-50 °C. This range was selected on the basis of possible future climate scenarios. By monitoring CH3Br emission rates as a function of temperature, halogen content, and organic functional group availability (methoxyl group) in plant material, we attempt to evaluate the potential of the abiotic process as it affects the production of atmospheric CH3Br.
Materials and Methods Sample Preparation. Ash (Fraxinus excelsior) leaves were collected in the vicinity of Belfast, Northern Ireland, and Saltwort (Batis maritima) was grown under greenhouse conditions. Air-dried ash and freeze-dried saltwort leaves were ground to a powder. Tomato reference material (NIST1573a) was acquired from NIST (Gaithersburg, MD), and hay 6838
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 18, 2008
powder reference material (IAEA V-10) was purchased from IAEA (Vienna, Austria). Pectin (Sigma-Aldrich, St. Louis, MO) was mixed with potassium bromide solution to form a gel, followed by freeze-drying, and then ground to pass through a 1 mm sieve. Sample Incubation. Samples (0.1-1.5 g, dependent on halogen content) were incubated in 40 mL vials (Fisherbrand, Pittsburgh, PA) sealed with PTFE/silicon septa (SigmaAldrich, St. Louis, MO). In the initial experiment with bromide enriched pectin (632 mg kg-1 dry weight) samples were incubated for 17 h at temperatures of 30, 40, and 50 °C. For all other experiments samples were incubated in the temperature range of 25-50 at 5 °C intervals over the same time period. Samples were allowed to equilibrate at 25 °C for 30 min prior to sampling of headspace for determination of CH3Br and CH3Cl. Determination of Bromine and Chlorine. Bromine and chlorine content of plant samples were measured using X-ray fluorescence (XRF) as described in Cheburkin and Shotyk (36), or Instrumental Neutron Activation Analysis (INAA) as described in Wa¨nke et al. (37). Determination of Methoxyl Content. Sample, in triplicate, (∼10 mg) was placed in a glass vial (14 mL) and ethyl cellulose (50 µL, 6 mg/mL in chloroform) added as internal standard. After evaporation of the solvent, hydriodic acid, 57% (1 mL) and distilled water (100 µL) were added and the vial was sealed with a screw cap containing a PTFE/ silicone septum. For quantification, a series of standards were prepared in a similar fashion to the samples except the distilled water was replaced by aqueous methanol solutions of known concentration. For quality control purposes, samples of methyl D-galactopyranoside with known methoxyl content were analyzed with each sample batch. Samples were heated at 100 °C for 30 min, and the headspace (50 µL) was monitored for methyl and ethyl iodide content by gas chromatograph/mass spectrometry (GC/MS) using a 6890GC/ 5973N system (Agilent Technologies, Palo Alto CA). The MS was employed in the selected ion monitoring (SIM) mode measuring ions currents at m/z 142 and 156. The GC oven was equipped with a 12.5 m × 0.25 mm × 8 µm Chrompack CP-Poraplot Q capillary column (Varian, Oxford, UK) and programmed at 80 °C for 1 min and then ramped at 10 °C min-1 to a final temperature of 200 °C. Helium (1.5 mL/min) was employed as the carrier gas with an injection split ratio of 100:1. Methyl Bromide Analysis. CH3Br was measured by GCMS using a modified 6890GC/5973N system (Agilent Technologies, Palo Alto CA) described by Gros et al. (38). Sample (2-10 mL) was injected into a continuous helium gas flow (30 mL/min) through a septum over 1 min and followed by a 10 min flush to ensure quantitative transfer to the cryotrap (-70 °C, on glass beads in a 1/16” stainless steel tube). All lines prior to the cryo-trap were heated to 60 °C to prevent analyte adsorption to the tubing walls. Target compounds were desorbed from the trap, by rapid heating to 200 °C, and transferred directly to the GC. The separating column was a 60 m × 0.25 mm × 1 µm DB-5 capillary column (J&W Scientific, Agilent Technologies, Palo Alto CA). Helium gas (2.7 mL/min constant flow) was employed as carrier gas, and the GC was programmed to hold at 35 °C for 4 min, ramp at 15 °C min-1 to 120 °C, hold this temperature for 3 min, then ramp at 120 °C min-1 to 250 °C and hold for 3 min. The MS was operated in the SIM mode measuring ion currents at m/z 94 and 96. Between measurements the cryo-trap was conditioned by flushing with helium and heating at 200 °C. Calibrations were performed against a gravimetrically prepared calibration gas (Apel-Riemer Environmental, Denver CO). The detection limit, defined as a signal exceeding the background by at least three times the standard deviation of the noise (for the specific ion at its specific retention time)
TABLE 2. Ratios of Released CH3Cl/CH3Br at 30, 40, and 50 °C and Cl/Br in the Plants
sample saltwort ash tomato (NIST 1573a) a
temp (°C)
CH3Cl (ng g-1 dw d-1)
CH3Cl (nmol g-1 dw d-1)
CH3Br (ng g-1 dw d-1)
CH3Br (nmol g-1 dw d-1)
CH3Cl:CH3Br (Molar ratio)
Cl/Br (Molar ratio) b
Cl:Br/ CH3Cl:CH3Br (Molar ratio)
30 50 30 40 50 50
440 5380 13a 64a 115a 398
8.71 106 0.26 1.27 2.3 7.9
18.6 175 0.23 0.9 4.6 156
0.19 1.84 0.002 0.009 0.05 1.64
44.5 57.8 106 134 47 4.8
475 475 505 505 505 11
10.7 8.2 4.7 3.8 10.7 2.3
Data from Hamilton et al. (20).
b
Derived from chlorine and bromine content in plant.
FIGURE 4. Conversion of bromine to CH3Br in temperatures ranging from 25 to 50 °C. that the initial oven temperature was set at 30 °C and the split ratio at 10:1. The mass spectrometer was operated in the SIM mode monitoring ion currents at m/z 50, for quantification, and 52, as a qualifier ion. For quantification purposes, calibration was against samples of headspace above standard solutions of known concentrations equilibrated at 25 °C (39).
Result and Discussion
FIGURE 3. Methyl bromide emissions from different plant materials. (a) hay, (b) ash, and (c) tomato leaves. (S/N ) 3), was found to be 0.2 ppb. The coefficient of variation (CV) for the GC/MS analysis was 7.2%. Methyl Chloride Analysis. CH3Cl was measured by GC/ MS using the system described above for determination of methoxyl content. The GC conditions were identical except
Methyl Bromide Formation from Dried Plant Material. Hamilton et al. (20) investigated abiotic production of CH3Cl from plant leaf material using pectin as a model methyl donor substance. In that study it was observed that pectin methoxyl groups reacted with chloride ion to form methyl chloride with emission rates dependent upon both temperature and chloride content. Therefore our first approach to establish the effect of temperature on release of CH3Br from plant material was to incubate pectin enriched with bromide ion (632 mg kg-1) at 30, 40, and 50 °C. As can be seen from Figure 1 CH3Br emission rates show an exponential increase with increasing temperature. Results obtained from this experiment showed that the emission rate of CH3Br at 30 °C was 26.4 ng g-1 dw d-1 and this rate rose by a factor of ∼11 when pectin was incubated at 50 °C. The next step was to examine emissions from plant leaf material including saltwort, ash, hay, and tomato. In initial experiments the effect of sample preparation was investigated by comparing emissions from whole and ground leaf samples when incubated at temperatures of 30, 50, and 100 °C. The results from these experiments showed that, in general, regardless of sample preparation, emissions for each plant type were in the same range and increased similarly with temperature. However, with all sample types and at all temperatures the standard deviations of the measured VOL. 42, NO. 18, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6839
FIGURE 5. Arrhenius dependency of reaction rate constant and temperature for the formation of CH3Br and CH3Cl in different plant materials. emissions from whole leaf samples were considerably higher than those from ground leaf samples. From these observations we concluded that for future experiments it would be more suitable to use ground samples as this would simplify the interpretation of results without significantly changing production rates and thus the overall outcome of this study. Saltwort is a typical saltmarsh plant known to emit both CH3Br and CH3Cl (12, 40). Therefore, as saltwort had a similar bromine content to pectin, 872 mg kg-1 bromine, and a high chlorine content, 186 g kg-1 chlorine, it was decided to use this plant to compare the emission profiles of CH3Br and CH3Cl (Figure 2). Samples were incubated at temperatures ranging from 25 to 50 °C in 5 °C increment steps and the amounts of CH3Br and CH3Cl formed measured. The methyl halide emission patterns were analogous, exhibiting an exponential increase with temperature, indicative of a similar mechanism of formation for both compounds. CH3Br and CH3Cl emission rates at the lowest temperature investigated (25 °C) were 12 ng g-1 dw d-1 and 173 ng g-1 dw d-1, respectively. When samples were incubated at 50 °C, CH3Br and CH3Cl emissions increased by approximately 15 and 31 fold, respectively. The observed differences in the emission ratio (Table 2) is almost certainly due to the higher chlorine content of the plant matter compared to bromine. Interestingly, CH3Br emissions from pectin were higher than those observed for saltwort although it contained less bromine. This can be explained by the higher methoxyl content of pectin (∼8%) compared with that of saltwort (∼1%) (Table 1). Samples of ash, hay, and tomato leaf material were further examined in order to study the effect of temperature on other vegetation typical of temperate forest, meadows and agriculture areas. Hay and ash contain natural levels of bromine (Table 1) but the tomato sample had a higher than expected concentration of bromine which might reflect its growth conditions. Agricultural vegetables grown on bromine enriched soils can contain high concentrations of bromine as shown by Mino and Yukita (41). This enrichment may occur due to CH3Br fumigation or residual amounts of bromine 6840
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 18, 2008
from fertilizer usage. Yuita (42) reported a 20-fold difference in bromine content between plants grown on CH3Br treated soil and untreated soil. Hay, ash, and tomato leaf material were incubated under the same conditions as specified for saltwort and emissions of CH3Br measured (Figure 3). CH3Br emission rates of these samples showed an exponential increase with temperature similar to pectin and saltwort (R2 ) 0.98). Hay containing 8 mg kg-1 (dw) bromine and ash with a concentration of 34 mg kg-1 (dw) bromine emitted the lowest amounts of CH3Br. Emission rates at 25 °C were observed to be 0.009 ng g-1 dw d-1 and 0.12 ng g-1 dw d-1 for hay and ash, respectively, and these increased dramatically by factors of 23 (hay) and 38 (ash) when the temperature of incubation was raised to 50 °C. On average, a 5 °C temperature rise doubled the amount of CH3Br released. Tomato leaf material exhibited a similar emission profile but emitted higher concentrations of CH3Br, up to 171 ng g-1 dw day-1 at 50 °C. Figure 4 shows the fraction of bromine that was converted to CH3Br within one day. Bromine conversion to CH3Br was most efficient with pectin and is followed by saltwort, tomato, ash, and then hay. As previously mentioned the higher efficiency of conversion observed with pectin is most likely due to its high methoxyl content. However, despite having the highest methoxyl content of all the plant species investigated, hay with a methoxyl content of 2.4%, showed the lowest bromine conversion rates. This may be due to a higher lignin content in the hay sample compared to the other species. As has been reported by Hamilton et al. (20) lignin methoxyl groups would not be available as a methyl donor over the temperature range used in this study but might become of interest at higher temperatures, for example those occurring during biomass burning. However, the employed analytical method does not distinguish between methoxyl groups derived from pectin or lignin, and total methoxyl content may therefore not be an accurate measure of methylating potential at these low temperatures. Conversion of Bromine to CH3Br and Comparison to Conversion of Chlorine to CH3Cl. From Table 2 it can be
seen that the molar chlorine content of the studied plants is 1-3 orders of magnitude higher than the molar bromine content. Both saltwort and ash had a similar Cl/Br molar ratio as shown in Table 2 (475 and 505, respectively), reflecting the similar Cl/Br molar ratios which are usually found in seawater and precipitation while tomato had a low Cl/Br ratio (11) presumably due to anthropogenic origin of bromine in this sample. Strikingly, the molar ratios of CH3Cl to CH3Br emitted from all investigated plant species were in the range of 5-134 and thus were much lower than the Cl/Br molar ratios shown for the plant material (Table 2). Since the vast majority of halogens in plant material exist as the halide ion this clearly indicates that the bromide ion is more efficiently methylated than chloride ion in plant material. The observation that bromide is preferentially used (in comparison with chloride) for methylation reactions in organic systems, both bioticandabiotic,hasbeenpreviouslyreported(7,19,40,43,44). In our study the molar ratios of CH3Cl to CH3Br for saltwort samples at 30 and 50 °C were 45 and 58, respectively. These values are higher than those recently reported from field studies of coastal marsh plants. In those studies Manley et al. (40) observed saltwort emission molar ratios of 12 ( 0.78, whereas Rhew et al. (12) found ratios of 17 ( 14 for various coastal marsh plants. The CH3Cl to CH3Br molar ratios of ash leaves at 30 and 40 °C were 106 and 134, respectively and thus higher than in the saltwort samples. Interestingly, in a recent study by Saito and Yokouchi (44) molar ratio values in the range of 60-220 for two tropical fern species were reported. In most previous studies, where living plants were monitored for release rates of CH3Cl and CH3Br, an activity of methyl transferases was assumed as the main methylation process. However, our data show that variations in CH3Cl to CH3Br molar ratios can also be explained by an entirely chemically driven process. This raises the question of whether methyl transferases are the sole source of methyl halide in living plants or whether methyl halide formation in living plants is also a result of a metabolic “accident” as recently suggested by Manley (45). Calculation of the Activation Energy for CH3Br Formation. In our study, for all sample types and at all incubation temperatures, employed only a very small percentage of halogen, and methoxyl group are converted to the halomethane. Thus these reactions can be classed essentially as pseudofirst order, and the observed rates are effectively rate constants. An Arrhenius plot of the experimental data would therefore allow the activation energy (Ea) to be determined for each plant material. As can be seen in Figure 5 all Arrhenius plots for CH3Br formation followed a straight line at temperatures ranging from 25 to 50 °C. Ea values, calculated from the slope of the curves, for CH3Br formation for hay, ash, tomato, pectin, and saltwort were 103, 110, 120, 97, and 85 kJ/mol, respectively. In general, Ea values reported for enzyme-catalyzed reactions are much lower (usually
