Environ. Sci. Technol. 2010, 44, 7096–7101
Uptake of Methacrolein and Methyl Vinyl Ketone by Tree Saplings and Implications for Forest Atmosphere A K I R A T A N I , * ,† S E I T A T O B E , ‡ A N D SACHIE SHIMIZU‡ Institute for Environmental Sciences, University of Shizuoka, 52-1 Yada, Shizuoka 422-8526, Japan, and School of High-Technology for Human Welfare, Tokai University, 317 Nishino, Numazu 410-0395, Japan
Received May 24, 2010. Revised manuscript received August 9, 2010. Accepted August 10, 2010.
Methacrolein (MACR) and methyl vinyl ketone (MVK) are oxygenates produced from isoprene which is abundantly emitted by trees. The uptake rate of these compounds by leaves of three different Quercus species, Q. acutissima, Q. myrsinaefolia, and Q. phillyraeoides, at typical concentrations within a forest (several part per billion by volume) were determined. The rates of uptake of croton aldehyde (CA) and methyl ethyl ketone (MEK) were also investigated for comparison. The rates of uptake of the two aldehydes MACR and CA were found to be higher than those of the two ketones. In particular, the rate of MEK uptake for Q. myrsinaefolia was exceptionally low. The ratio of intercellular to fumigated concentrations, Ci/Ca, for MACR and CA was found to be low (0-0.24), while the ratio for the two ketones was 0.22-0.90. To evaluate the contribution of tree uptake as a sink for the two isoprene-oxygenates within the forest canopy, loss rates of the compounds due to uptake by trees and by reactions with hydroxyl radicals (OH radicals) and O3 were calculated. The loss rate by tree uptake was the highest, followed by the reaction with OH radicals, even at a high OH concentration (0.15 pptv) both for MACR and MVK, suggesting that tree uptake provides a significant sink.
Introduction Large amounts of anthropogenic and biogenic volatile organic compounds (VOCs) are continuously emitted into the atmosphere. Many of these compounds, such as aromatic hydrocarbons, aldehydes, and ketones, are toxic to humans, animals, and plants. Secondary products of VOCs are also formed in reactions with reactive species such as hydroxyl radicals (OH radicals) in the atmosphere. Methacrolein (MACR) and methyl vinyl ketone (MVK) are major secondary products in the reactions of isoprene with OH radical and O3. Their formation yields in the reaction of isoprene with OH radical and O3 are reported in the range of 0.16-0.4 (1, 2). These two compounds are precursors of photochemical pollutants and their behavior has been intensively investigated in laboratories and in the field (3–7). The major atmospheric sink of MACR and MVK has been considered to be provided by reactions of these compounds with the OH radicals and O3 (5, 8). * Corresponding author phone: +81-54-264-5788; fax: +81-54264-5788; e-mail:
[email protected]. † University of Shizuoka. ‡ Tokai University. 7096
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 18, 2010
It is known that plants can remove gases from the atmosphere, both by absorption through the stomatal apertures and also by adsorption on the cuticle or outer surface of the leaf. It has been shown that certain plants can remove linear (9) and branched (10) aldehydes and ketones (10–12) from the atmosphere, and it has even been suggested that plants may be used to clean indoor air of VOC pollutants (10, 13). However, the fate of MACR and MVK in a forested area and the contribution of forest trees in the removal of these VOCs have not been characterized. Most previous plant uptake studies have been conducted at unrealistically high concentrations of VOCs (∼10-6) due to the sensitivity limitations of the analytical methods used, such as gas chromatography. No data are available for uptake of VOCs by trees at ambient levels (concentration of 10-9). Our previous report showed that the use of a measurement system based on proton transfer reaction mass spectrometry (PTR-MS) can significantly increase the certainties (precision of 1-1.5%) in measuring VOC concentrations at 101-102 part per billion by volume (ppbv) in humid air (12). We successfully determined the uptake rates of C3-C5 aldehydes, benzaldehyde, and C4-C6 ketones by houseplants, using the PTRMS-based measurement system (10). In the present study, we investigate the ability of tree saplings to absorb the isoprene oxidation products MACR and MVK at their ambient ppbv concentrations. On the basis of the measured uptake rates, we evaluate the contribution of the sink strength provided by plant uptake with respect to the fate of MACR and MVK in a forested area.
Experimental Section Plant Materials. Saplings (3 year-old) of Quercus acutissima, Q. myrsinaefolia, and Q. phillyraeoides were obtained from a local nursery in Shizuoka city, Japan. Q. acutissima is a deciduous broad leaved tree and the other two species evergreen broad leaved trees. Five saplings were prepared for each plant species, and each was transplanted into a pot with a volume of 18 L in soil composed of 30% leaf mold, 30% Kanuma soil, 20% andosol, and 20% sharp sand. The plant heights were 60-80 cm. The plants were used for experiments 2 months after transplantation. VOC Exposure and Measurement System. The VOC exposure and measurement system has been described previously (12), and the schematic diagram is given in the Supporting Information. The system consists of three parts: an air flow system for containing known and constant concentrations of specific VOC species, a fumigation chamber, and an analytical system. An attached branch of the test tree, with a projected leaf area of 200-300 cm2, was enclosed in a transparent fluorinated ethylene-propylene copolymer (FEP) bag (20-40 L volume). The open side of the bag was tightly closed with a cable tie, and the enclosed volume was less than 20 L. VOC-rich air was introduced into the bag via the inlet port at a flow rate of 1.3 L min-1. The time required to reach 99% equilibration of VOC concentrations was less than 20 min. VOC and water vapor concentrations were measured in the inlet and outlet lines with a PTR-MS instrument (Ionicon GmbH, Innsbruck, Austria). The PTRMS operating conditions are described in the Supporting Information. The carbon dioxide (CO2) concentration of the inlet and outlet air was measured with an infrared gas analyzer (ZYF; Fuji Electric Systems, Japan), and the net photosynthetic rate of the enclosed leaves was calculated. Two three-port solenoid valves automatically changed the gas flow every 5 min. These valves were operated synchronously but in 10.1021/es1017569
2010 American Chemical Society
Published on Web 08/18/2010
FIGURE 1. Changes in MA and MVK concentrations of the inlet and outlet air streams of the bag with and without the plant. Panels A and B show background measurement data collected before the start of MACR and MVK fumigation measurements. Solid and dashed lines indicate inlet and outlet concentrations of the target VOCs and mass. opposite directions, i.e., when the inlet air was directed into the PTR-MS, the outlet air was directed to the CO2 analyzer. Plants were illuminated with a 400 W metal halide lamp (D400; Toshiba LiTec, Japan). Photosynthetic photon flux density (PPFD) was kept at 200-400 µmol m-2 s-1 at the top of the measured branch. The leaf temperature in the bag was monitored with two T-type fine-wire thermocouples and maintained at 24-26 °C. The VOC uptake measurement was conducted for three different saplings for each plant species. In a series of the measurements, the concentration of fumigated VOCs was raised up step by step from zero to 100-200 ppbv by changing the dilution ratio. At the beginning of the measurements, the emission of target VOCs from the enclosure bag was carefully checked using the purified air without the VOCs added. In order to investigate the relationship between VOC uptake and plant physiological parameters such as stomatal conductance, PPFD was often decreased step-by-step using light-shading sheets and finally set to approximately zero by switching off the irradiation lamp and room lights. The VOC uptake rate A, stomatal conductance gS, and intercellular VOC concentration Ci were determined according to our previous study (10). Fumigated VOCs. Compounds investigated in this series of experiments are the isoprene oxidation products methacrolein (MACR) and methyl vinyl ketone (MVK). Croton aldehyde (CA) and methyl ethyl ketone (MEK) were also used to compare the plant uptake capacity difference between
compounds. CA is an isomer of MACR and MVK. MEK has a structure similar to MVK but does not have a double bond. Chemical characteristics of these compounds are given in Table S1 in the Supporting Information. An aliquot (∼10 µL) of the standard liquid (purities >98%) was injected into the glass tube (1.5-2 mm diameter and 10 cm long) and installed into the permeator. The inflow rate to the permeator was 1 L min-1, and the air space in the permeator was maintained at 30 ( 0.1 °C. A flow of 10 mL min-1 of the outflow was diluted in order to produce several ppbv of the VOC air, and the remaining outflow was vented outside. The proton transfer reaction produces a protonated molecular ion. In this case, mass 71 was monitored for the MACR, CA, and MVK concentrations and mass 73 for the MEK concentration. Since the liquid level in the diffusion tube becomes lower as time progresses, the VOC concentration gradually decreases, at most by 15% after 24 h. The purities of the prepared VOC were analyzed by gas chromatography-mass spectrometry (GC-MS) according to the method of Tani et al. (14) and found to be greater than 98%. Most of the impurities had molecular weights different from the molecular weights of the target VOC compounds and therefore did not interfere with the PTR-MS measurement of the target VOC compounds.
Results and Discussion Blank Measurements. Before exposure of the plant to VOC, blank measurements were conducted for the enclosure bag (panels A and B in Figure 1). The detection limit of this PTRVOL. 44, NO. 18, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7097
TABLE 1. Normalized Uptake Rate, gS, and Ci/Ca for the Corresponding Aldehydes and Ketonesa chemical
species
normalized uptake ACA (mmol m-2 s-1)
gS (mmol m-2 s-1)
Ci/Ca for VOC
MACR
Q. myrsinaefolia Q. acutissima Q. phillyraeoides
23.1 ( 8.1 ab 24.9 ( 7.6 ab 21.0 ( 4.2 ab
54.6 ( 39.0 94.9 ( 38.2 74.6 ( 24.6
0.05 ( 0.08 c 0.18 ( 0.10 bc 0.24 ( 0.12 bc
CA
Q. myrsinaefolia Q. phillyraeoides
29.5 ( 11.1 a 25.4 ( 6.6 ab
53.4 ( 28.3 44.6 ( 11.2
0c 0c
MVK
Q. myrsinaefolia Q. acutissima Q. phillyraeoides
11.2 ( 6.1 bc 16.3 ( 0.8 abc 12.5 ( 4.3 abc
46.8 ( 39.0 84.7 ( 22.8 53.0 ( 15.4
0.22 ( 0.11 bc 0.36 ( 0.17 b 0.36 ( 0.14 b
MEK
Q. myrsinaefolia Q. phillyraeoides
2.9 ( 0.8 c 14.3 ( 4.3 abc
81.3 ( 38.5 53.5 ( 23.0
0.90 ( 0.06 a 0.35 ( 0.14 b
(average: 23)
(average: 13)
a Different letters indicate significant difference between compounds (P < 0.05; Tukey multiple comparisons test). gS and Ci/Ca indicate stomatal conductance and ratio of intercellular VOC concentration Ci to fumigated concentration Ca, respectively.
MS system was 0.01 ppbv, and the calculated background mass 71 concentration was less than 0.2 ppbv. The outlet concentration was found to be slightly but significantly higher than the inlet concentration (107, implying that the precision is greater than that of the PTR-MS used in our previous study (12). The overall relative error, of which calculation procedure is shown in the Supporting Information, ranged from 0.8 to 2.2%. Hence a >2.2% difference between VOC influx (massin) and efflux (massout) per second (moles seconds-1), relative to massin, provides a detectable VOC uptake rate. The PTR-MS precision 7098
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 18, 2010
can also be improved with longer integration times, but this is not appropriate for monitoring a real-time change in VOC concentrations (17). Nevertheless, the precision of PTR-MS made it possible to measure this small difference and quantify the plant VOC uptake rate even at ambient VOC levels (10-9) for the first time. VOC Uptake and Ci. The calculated uptake rates for the four compounds were divided by the outlet concentration to normalize them to concentration (Table 1). The normalized uptake rate Aca of MACR was in the range of 21.0-24.9 mmol m-2 s-1 for the three tree species, with no significant differences between them. The other aldehyde CA, an isomer of MACR, had Aca values of 25.4 and 29.5 mmol m-2 s-1 for Q. myrsinaefolia and Q. phillyraeoides, respectively. These values are not significantly different from the Aca values for MACR. However, Aca of the two ketones MVK and MEK tended to be lower and was in some cases significantly lower than the Aca values of the two aldehydes. In particular, Aca of MEK for Q. myrsinaefolia had the very low value of 2.9 mmol m-2 s-1. In some measurements, the VOC concentration or PPFD was varied step by step. When the MACR concentration was increased step by step in the fumigation experiment using Q. acutissima, the outlet concentration was always lower than the inlet concentration (panel E in Figure 1). The uptake rate was found to increase with an increase in the fumigated concentration, but net photosynthetic and transpiration rates were almost constant (panel A in Figure 2). A linear relationship between the uptake rate and the fumigated concentration was observed (panel B in Figure 2). Similar results were observed for the remaining compounds and for the other two tree species in the concentration-varying experiment. When PPFD was decreased step by step in the fumigation experiment using Q. acutissima, the concentration difference between the inlet and outlet became smaller (panel F in Figure 1). The difference was finally less than 0.1 ppbv under dark conditions. All three of the gas exchange parameters, VOC uptake, net photosynthetic, and transpiration rates, were found to decrease with a decrease in PPFD. Although VOC uptake had positive values even under dark conditions, the net photosynthetic rate was negative, indicating dark respiration (panel A in Figure 3). A linear relationship between the uptake rate and gS was observed (panel B in Figure 3). The positive value of the y-axis intercept indicates that MVK is being adsorbed onto the leaf surface. Similar results were observed for the remaining compounds and for the other
FIGURE 2. Changes in VOC uptake, transpiration, and net photosynthetic rates of Q. acutissima (A) and the relationship between MACR concentration and the uptake rate (B). Tr and Pn indicate transpiration rate and net photosynthetic rate, respectively. Data were calculated from those in panel E of Figure 1.
FIGURE 3. Changes in VOC uptake, transpiration, and net photosynthetic rates of Q. acutissima (A) and the relationship between stomatal conductance and the uptake rate (B). Tr, Pn, and gS indicate transpiration rate, net photosynthetic rate, and stomatal conductance, respectively. Data were calculated from those in panel F of Figure 1. two tree species in the PPFD varying experiment. The rate of adsorption onto the leaf surface was less than 5% of the uptake rate under full light conditions (PPFD ) 500 µmol m-2 s-1) for all of the measurements. A very low or zero rate of adsorption onto the leaf surface has been reported for two tree species, Populus nigra and Camellia sasanqua, which were exposed to 0.5-1 ppmv MEK and acrolein (11), and for houseplants Spathiphyllum clevelandii and Epipremnum aureum, which were exposed to 20-500 ppbv aldehydes and ketones (10). Our result and previous studies (10, 11) show that Aca is regulated by stomatal conductance gS. As shown in Table 1, the averaged gS was found to be in a similar range between compounds and between tree species. To normalize the uptake capability to gS, the intercellular VOC concentration Ci was calculated. The ratio of Ci to fumigated concentration Ca, Ci/Ca, is the most appropriate measure to compare plant VOC uptake capacity among plant species and among VOC species (10). A lower Ci/Ca value indicates that a plant leaf has a higher uptake capacity. Ci/Ca for CA was 0 for Q. myrsinaefolia and Q. phillyraeoides and the same result has been observed for houseplants (10). Ci/Ca for MACR was also found to be low (0-0.24) for the three tree species investigated in this study. The value reported for Spathiphyllum clevelandii was 0.42 ( 0.07 (10). This is higher than the values determined in the present study. Ci/Ca for MVK and MEK ranged from 0.22 to 0.36, except for the MEK value of Q. myrsinaefolia. It was exceptionally high (0.9) because of the low plant uptake rate. These differences in Ci/Ca between plants may be brought about by different uptake capacities of individual plant species. The uptake rate of VOCs may depend on the solubility of VOCs into the leaf, the metabolic availability of the compounds, the translocation rate of VOCs from the fumigated
leaf to other organs, and storage of the VOCs in vacuoles and cell walls (11, 18). Dissolution of VOCs into leaf water is the first step of the uptake process if the uptake is mainly regulated by stomata (11, 12). In our previous study (10), we observed a clear relationship between Henry’s law constants and Ci/Ca for both the C3-C5 aldehydes and C4-C6 ketones. Although the Ci/Ca value was found to decrease with an increase in water solubility, the aldehydes and ketones showed different dependencies on Henry’s law constants, i.e., Ci/Ca tended to be lower for the aldehydes than for the ketones. We also showed that water-soluble benzaldehyde had a greater deviation from the relationship (10), having a higher Ci/Ca (0.4). In the present study, Ci/Ca for the two aldehydes was also lower than that for the ketones. Henry’s law constant of MVK and MEK was much higher than that of MACR (Table S1 in the Supporting Information), and water solubility cannot explain the difference in the observed Ci/Ca values between compounds. This may be attributed to different plant scavenging abilities with respect to the ketones and aldehydes. Our previous study showed that in longer-term fumigation measurements, the total uptake amounts of aldehydes and ketones by houseplants are 30-100 times as much as the amounts dissolved in the leaves (10). This suggests that the VOCs are scavenged in the plants. The VOC dissolved in the leaf water is then metabolized, translocated, or stored inside the plant (18). Transport of contaminants in the direction from the leaf to stem may occur via the phloem flow (19). However, the phloem flow is not fast (10-100 cm/h), and little is known about the ability of the phloem flow to remove contaminants from the leaf. It is known that many chemicals, such as toxic polycyclic aromatic compounds, can be metabolized and transformed in plant leaves (20). Cytochrome P-450 monooxygenases and glutathione-S-transferases (GSTs) are VOL. 44, NO. 18, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
7099
TABLE 2. Integrated Loss Rate (0-10 m) of MACR and MVK by Tree Uptake and in the Reaction with the O3 and OH Radicalsa MACR loss rate (pmol m-2s-1)
MVK loss rate (pmol m-2s-1)
O3 (ppbv)
OH (pptv)
plant uptake
OH reaction
O3 reaction
plant uptake
OH reaction
O3 reaction
30
0.05 0.10 0.15 0.05 0.10 0.15 0.05 0.10 0.15
138 138 138 138 138 138 138 138 138
33.7 (24) 67.5 (49) 101.2 (73) 33.7 (24) 67.5 (49) 101.2 (73) 33.7 (24) 67.5 (49) 101.2 (73)
0.7 (0) 0.7 (0) 0.7 (0) 1.1 (1) 1.1 (1) 1.1 (1) 1.6 (1) 1.6 (1) 1.6 (1)
78 78 78 78 78 78 78 78 78
18.9 (24) 37.9 (49) 56.8 (73) 18.9 (24) 37.9 (49) 56.8 (73) 18.9 (24) 37.9 (49) 56.8 (73)
2.8 (4) 2.8 (4) 2.8 (4) 4.6 (6) 4.6 (6) 4.6 (6) 6.4 (8) 6.4 (8) 6.4 (8)
50 70
a The reaction rate constant for the reactions of OH radical with MACR and MVK are 3.35 and 1.88 × 10-11 cm3 molecules-1 s-1, respectively (23). The reaction rate constant for the reactions of O3 with MACR and MVK are 1.14 and 4.56 × 10-18 cm3 molecules-1 s-1, respectively (23). Normalized uptake rates ACA were assumed to be 23 and 13 mmol m-2 s-1 for MACR and MVK, respectively. MACR and MVK concentrations were both assumed to be 2 ppbv. OH radical and O3 concentrations were assumed according to previous papers (4, 6–8). The values in parentheses indicate the percentage of loss rate in the reaction with respect to that induced by plant uptake.
enzymes which may play roles in detoxification of the compounds (18). Although there is no direct evidence to date that the four compounds used in our experiment are metabolized inside the leaf, the difference in Ci/Ca between the aldehydes and ketones may be attributed to different metabolic availability. The aldehydes may be easily oxidized inside the leaf to produce carboxylic acids that would be expected to be more quickly utilized in metabolic pathways. Ci/Ca for MEK of Q. myrsinaefolia was found to be exceptionally high (0.90 ( 0.06), suggesting that metabolic pathways of Q. myrsinaefolia slowly degrade MEK inside the leaf. It is known that xenobiotics and their metabolites, such as glutathione S-conjugates, are stored within trees (21). Vacuoles and cell walls are storage sites for soluble and insoluble conjugates, respectively. The compounds used in the present study are all water-soluble, and therefore some of their metabolites might be stored in vacuoles. Fates of the Isoprene Oxygenates within Forests. Our results suggest that a forest may function as a sink for MACR and MVK, which are mainly produced in the reaction of biogenic VOC isoprene with the OH radicals and O3. OH radicals and O3 also react with MACR and MVK (1), and these reactions can be major sinks for the two compounds. Gierczak et al. (5) calculated the rate of atmospheric loss of these compounds and showed that the process of primary loss for both MVK and MACR is the reaction with OH radicals throughout the troposphere. They reported that photolysis is a minor sink. To evaluate the contribution of tree uptake as a sink of the two oxygenates within a forest canopy, the loss rates by tree uptake and reactions with OH radicals and O3 were calculated using a simple reaction model, which is briefly described in the Supporting Information. The values of the normalized uptake rate ACA were assumed to be 23 and 13 mmol m-2 s-1 for MACR and MVK, respectively; these values are listed in Table 1 as average values. They are considered to be representative averages for the whole daytime because they were obtained under the PPFD corresponding to one-fourth to one-fifth full sunlight. The canopy height and leaf area index (LAI) in a typical Quercus forest in temperate region were assumed to have the typical values of 10 and 3 m2 m-2, respectively. The OH radical concentration was assumed in three different scenarios with high (0.15 pptv), medium (0.10 pptv), and low (0.05 pptv) values and O3 concentrations with 70, 50, and 30 ppbv (4, 6–8). The loss rate integrated from ground to the canopy top (from 0 to 10 m) was calculated on a ground area basis (Table 2). 7100
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 18, 2010
Tree uptake had a higher loss rate than the two reactions, for both MACR and MVK in all cases. The O3 reaction provided the least contribution to their loss within the canopy (
