Environ. Sci. Technol. 2005, 39, 5722-5728
Laboratory Investigation of Heterogeneous Interaction of Sulfuric Acid with Soot DAN ZHANG AND RENYI ZHANG* Department of Atmospheric Sciences, Texas A&M University, College Station, Texas 77843
The internal mixing state of soot with sulfuric acid is believed to significantly impact the optical, cloud-forming, and chemical properties of soot-containing aerosols, but little is known about the interaction between soot and sulfuric acid. We report the first measurements of the uptake of H2SO4 on three types of soot generated from methane, hexane, and kerosene combustion. H2SO4 loss on soot is found to be irreversible. The measured uptake coefficients are 0.018 ( 0.007 for kerosene soot, 0.012 ( 0.006 for methane soot, and 0.0076 ( 0.0016 for hexane soot at a total pressure of 1-2 Torr and 298 K assuming a geometric surface area, likely corresponding to the upper limits. Additional experiments using the differential mobility analysis and Fourier transform infrared spectroscopy techniques are carried out to further characterize the interaction of H2SO4 with soot. The results suggest that uptake of H2SO4 takes place efficiently on soot particles, representing an important route to convert hydrophobic soot to hydrophilic aerosols.
Introduction Atmospheric aerosols play an important role in regulating cloud formation and solar radiation intake in the earthatmosphere system. Soot particles, formed as a result of incomplete combustion of fossil fuels and biofuels, and outdoor biomass burning, have an average global source estimated to be as high as 24 Tg yr-1(1). Soot particles are of special interest because of their role in atmospheric heterogeneous reactions, cloud formation, and global radiative balance and climate. The direct effect lies in the large absorbing efficiency of solar and terrestrial radiation by soot (2). In addition, they may serve as cloud condensation nuclei (CCN) (3). In particular, the activation of soot particles as CCN is thought to be enhanced by the uptake of water-soluble inorganic species such as H2SO4 and HNO3 (4-6). Sulfuric acid plays a key role in atmospheric new particle formation. Gaseous H2SO4 in the atmosphere is produced by the oxidation of SO2, which is from either direct industrial and automobile emission or oxidation of reduced biogenic sulfur compounds. H2SO4 in the gas phases can have two fates: homogeneous nucleation with H2O, NH3, or organic vapors, leading to new particle formation (see, e.g., refs 7-9), and heterogeneous scavenging by preexisting aerosols, resulting in particle growth. In the latter case, soot particles provide heterogeneous surfaces for H2SO4 in the urban and regional atmosphere, forming internally mixed aerosols. Because of the abundance and the climatic impact of soot and H2SO4 * Corresponding author phone: (979) 845-7656; fax: (979) 8624466; e-mail:
[email protected]. 5722
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 15, 2005
in the atmosphere, it is of importance to understand their heterogeneous interaction to provide quantitative information for current models involving aerosol microphysics and aerosol/cloud interaction. Numerous efforts have been made in laboratory, field, and modeling studies to determine the sources, distribution, heterogeneous reactivity, and radiative properties of soot in the atmosphere. Typically, the mixing state of soot in the atmosphere is characterized by two extreme scenarios: soot partially or entirely coated within other organic/inorganic constituents (internally mixed) and distinct soot particles mixed externally with other aerosol types (externally mixed). Model calculations indicated that soot could exert higher “positive” direct radiative forcing when associated with other scattering aerosols, e.g., sulfate, nitrate, and organic carbon (10). In addition, an internal mixture of soot with other aerosol ingredients appears significantly more absorptive than the external mixture counterpart (10). Because of the high positive forcing of soot-containing aerosols, it has been suggested that the warming effect from black carbon may nearly balance the net cooling effect of other anthropogenic aerosol constituents (11). Several field measurements have been carried out to investigate the mixing state of atmospheric soot particles in urban and rural areas. For example, Hasegawa and Hota (12) examined samples of individual soot-containing particles using the dialysis of water-soluble components and electron microscopy. The size distribution and morphology of ambient mixed soot particles were studied using differential mobility analysis (DMA) and transmission electron microscopy (13). A recent study by Mallet and co-workers (14) employed a novel approach to estimate the mixing state of soot in urban areas by comparing the measured single scattering albedo with values calculated for both internal and external mixing cases. Field measurement results revealed that soot particles in the urban zone were mostly externally mixed, while internal mixtures were more prevalent at rural sites. Numerous experiments have been performed in the laboratory to investigate uptake of atmospheric species on soot, such as water (see, e.g., refs 15 and 16), ozone (see, e.g., ref 17), and nitrogen-containing compounds (see, e.g., refs 18-23). The reported uptake coefficients of the various chemical species on soot vary by several orders of magnitude, depending on the soot type, soot deactivation due to exposure to the reactant, and concentrations of the gaseous species used. Other previous studies have reported measurements of H2SO4 mass accommodation and uptake coefficient on aqueous H2SO4 as well as sea-salt aerosol surfaces (24-26). Despite the considerable interest in the interaction between soot and H2SO4, to date no kinetic data are available on uptake of gas-phase H2SO4 on soot, which is crucial in understanding the role of soot particles in cloud formation, radiative transfer, and heterogeneous chemistry. In this paper we report measurements of the uptake coefficients of H2SO4 on the surfaces of soot generated from methane, hexane, and kerosene combustion. The kinetic experiments are conducted with a low-pressure, coated-wall laminar flow reactor apparatus equipped with an ion driftchemical ionization mass spectrometer. We also investigate the potential of soot deactivation by exposing soot to H2SO4 over an extended time period. Two additional experiments are performed to further characterize the interaction of H2SO4 with soot. The particle size and number distributions of soot with and without exposure to H2SO4 vapor are measured by a differential mobility analyzer coupled to a condensation particle counter (CPC). A Fourier transform 10.1021/es050372d CCC: $30.25
2005 American Chemical Society Published on Web 06/21/2005
FIGURE 1. Schematic representation of the flow reactor-ID-CIMS system setup. infrared (FTIR) spectrometer is employed to collect IR spectra of soot and soot-H2SO4 aerosols.
Experimental Section Heterogeneous Uptake of Sulfuric Acid on Soot. The uptake experiments were conducted in a low-pressure, coated-wall laminar flow reactor coupled to an ion drift-chemical ionization mass spectrometry (ID-CIMS) system shown schematically in Figure 1. The flow reactor was a cylindrical Pyrex glass tube 35 cm in length. Soot produced from burning three different types of fuels (methane, hexane, and kerosene) was evenly deposited on the inner wall of a 20 cm long, 1.65 cm i.d. Pyrex tube suspended above the flame, which was then inserted inside the flow reactor. The total mass of soot deposited was measured to be in the range of 1-10 mg for all experiments with a typical amount of 2 mg. An O-ring seal valve was installed between the flow reactor and the ID-CIMS instrument so that replacement of the soot-coated tube was easily facilitated without breaking the vacuum of the mass spectrometer system. The methane flame was produced with a Santoro-type coflow laminar diffusion burner, which consisted of two concentric tubes for fuel and air to flow through the inner and outer tubes, respectively (27). The kerosene and hexane flames were generated using a commercial alcohol burner. A N2 carrier gas flow of ∼100 sccm was passed through a bubbler containing ∼1 mL of commercial 96 wt % H2SO4. Sulfuric acid was introduced into the flow reactor through a central movable injector. The H2SO4 partial pressure in the flow reactor was estimated to be on the order of 10-7 to 10-8 Torr using the ID-CIMS method (9, 28). Dry N2 entrained into the reactor via a side port near the entrance was used as the main carrier gas, reaching a flow velocity characteristic of laminar flow conditions (i.e., the Reynolds number Re ) 2auF/µ < 2000, where a is the internal radius of the flow reactor (cm), F the density of the gas (g cm-3), u the flow velocity (cm s-1), and µ the absolute viscosity of the gas (g cm-1 s-1)). The Reynolds number was in the range of 30-60 for all experiments in this study. All carrier flows were monitored with calibrated electronic mass flow meters (Millipore Tylan 260 series). The flow reactor operated at low pressures (about 1-2 Torr) and room temperature, with typical flow velocities of 1600-2200 cm s-1. All experiments reported in this work were performed at 298 ( 2 K. Details of the ID-CIMS instrumentation were described previously (28). Briefly, the ID-CIMS system consisted of an ion source, a drift tube, and a mass spectrometer. Gas flow from the flow reactor was entrained directly into the drift tube region. A custom-made corona discharge was used as the ion source. For H2SO4 detection, nitrogen mixed with SF6 at the parts per million level was admitted through the corona charge while a high negative voltage (-5 kV) was applied to produce SF6- reagent ions. The primary ion-molecule reaction between H2SO4 and SF6- occurred at a rate near the collision limit, leading to the formation of HSO4- (m/e ) 97) (24). The drift tube was constructed with a 50 cm length Pyrex tube containing a set of 50 stainless steel rings
connected in series with 1 MΩ resistors between the rings. The rings had an i.d. of 1.4 cm through which the flow passed. The chemical ionization reaction between the reagent ion and H2SO4 occurred in this region. A voltage was applied to the rings to enhance the ion flow. At the downstream end of the drift tube, the majority of the flow was diverted to a mechanical pump while a small portion of the flow was diverted through a pinhole into the mass spectrometer. The uptake coefficient (γ) is defined as the ratio of the molecules removed from the gas-phase to the total gassurface collisions. The uptake process can be either reversible or irreversible. In some survey experiments, the movable injector was withdrawn to expose H2SO4 vapor to the soot film and then pushed back to the original position. Changes in the H2SO4 signal intensity with the injector position clearly indicated that uptake of H2SO4 on soot was irreversible on the time scale of our experiments. The uptake coefficient or reaction probability (γ) was determined by observing the loss of sulfuric acid on the soot surface when the H2SO4 vapor was exposed to the soot film and was calculated from the measured first-order rate constant (k) corresponding to the signal loss (29-31):
γ ) 2rk/(ω + rk)
(1)
where r is the radius of the flow tube and ω is the mean thermal speed. The uptake coefficient was determined using the geometric surface area of the soot-coated glass tube. To account for the H2SO4 radial gradients in the flow tube in the presence of a large reactant wall loss, the observed firstorder loss rate constants (kobs) were corrected for gas-phase diffusion with the method suggested by Brown (32). The H2SO4 gas-phase diffusion coefficient in N2 in the Brown calculations was adopted to be 69.1 Torr cm2 s-1 at 298 K, an average of two literature values (24, 33). The corrections for gas-phase diffusion were 20-40% for small γ values (γ < 0.01) and as large as a factor of 1.7 for a γ value of 0.03. The γ value was investigated for fresh soot as well as soot that had been exposed to H2SO4 over an extended time period (about 20 min). Characterization of the H2SO4-Soot Interaction with DMA and FTIR Spectroscopy. A schematic diagram of the experimental apparatus for soot generation, H2SO4 uptake, and characterization with DMA and FTIR spectroscopy is illustrated in Figure 2. Soot particles were produced by a gas or liquid burner and sampled using a device similar to that described by Kasper et al. (34). A stainless steel tube (3/8 in. o.d.) was horizontally mounted with a 1 mm orifice drilled at the bottom. A filtered N2 flow was used as a carrier gas to dilute the soot particles at a flow rate of 650 sccm. The sootladen flow was subsequently introduced into a diffusion drier containing anhydrous CaSO4 to reduce the relative humidity of the flow to below 5%. To coat soot with H2SO4, the dry soot particle flow was entrained into a 50 cm long, 3 cm i.d. Pyrex reservoir containing 96% H2SO4. Alternatively, in some experiments the soot particles bypassed the H2SO4 reservoir to obtain data without H2SO4 coating (i.e., fresh soot). VOL. 39, NO. 15, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5723
FIGURE 2. Schematic diagram of the experimental apparatus for soot generation, H2SO4 uptake, and characterization using DMA and FTIR spectroscopy. Two types of experiments were conducted to characterize H2SO4 coating on soot. In one type of experiment, the size distributions of soot-H2SO4 particles were measured using a differential mobility analyzer (TSI 3081) coupled to a CPC (TSI 3760A). Filtered N2 gas was used as the sheath flow with a flow rate 10 times the sample flow rate. During the experiment, the polydisperse soot-laden gas flow first passed through a 210Po diffusion charger prior to entering the DMA instrument to establish an equilibrium charge distribution. Number concentrations of the particles were counted by the CPC over the whole size range scanned by the DMA instrument. The size distributions were subsequently produced through a data acquisition system controlled by a personal computer. In another type of experiment, the IR spectra of the sootH2SO4 particles were measured with an FTIR spectrometer (Nicolet Magna 560) equipped with an MCT detector to analyze the aerosol chemical composition. The aerosols were sampled in situ in a 116.6 m path-length multipath optical cell with two KBr windows. Spectra were recorded by averaging 24 scans in the typical wavenumber range from 4000 to 750 cm-1. All IR spectra were measured at 4 cm-1 resolution.
Results and Discussion Uptake Coefficient of Sulfuric Acid on Soot. Figure 3a shows a typical experiment illustrating that uptake of H2SO4 on soot is irreversible. The experiment was conducted at 1.6 Torr. A steady-state flow of H2SO4 in N2 was first established with the movable injector passing the soot film. At about 85 s, the injector was quickly withdrawn to expose a 7 cm length of the soot film to H2SO4 vapor while the HSO4- signal was monitored in the mass spectrometer. A sharp decrease in the signal intensity was observed upon exposure of H2SO4 to soot. No recovery in the H2SO4 signal was observed during the exposure. At about 230 s, the injector was moved downstream (i.e., the soot film was no longer exposed to H2SO4) and the HSO4- signal returned to its original level. This type of measurement suggested an irreversible uptake of H2SO4 vapor on the soot surface under our experimental conditions. The uptake coefficient measurements were conducted by observing the decay of the HSO4- (m/e ) 97) signal as a function of the injector position as the injector was successively pulled upstream over the soot film. The HSO4- signal decay was found to be exponential according to the equation (35, 36)
It ) I0 exp(-kobst) 5724
9
(2)
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 15, 2005
FIGURE 3. Time evolutions of the HSO4- (m/e ) 97) signal. In (a) the injector was withdrawn to expose a 7 cm length of methane soot film to the H2SO4 vapor and then returned to its original position. Experimental conditions: Ptotal ) 1.55 Torr, U ) 1515 cm s-1, and T ) 298 K. In (b) the injector was successively withdrawn for a total of 12 cm length over kerosene soot. Experimental conditions: Ptotal ) 1.5 Torr, U ) 2149 cm s-1, and T ) 298 K.
FIGURE 4. HSO4- (m/e ) 97) signal as a function of injector position. Experimental conditions: Ptotal ) 1.4 Torr, U ) 1864 cm s-1, and T ) 298 K. The filled circles and solid line represent the measurement on fresh soot; the open circles and dashed line represent the measurement on soot after about a 20 min exposure to H2SO4. where I0 represents the initial HSO4- signal without uptake on soot, It is the signal at a given reaction time (or corresponding to a given injector position), and kobs is the observed first-order loss rate constant. Figure 3b illustrates a typical HSO4- signal decay on kerosene soot at various injector positions. The experiment was conducted at Ptotal ) 1.5 Torr, U ) 2149 cm s-1, and T ) 298 K. The injector was successively withdrawn for a 3 cm distance over the soot surface between 130 and 470 s. At 470 s, the injector was pushed downstream to the original position, and a full recovery of the signal was observed. At the end of the experimental run, the N2 carrier gas bypassed the H2SO4 bubbler to obtain a background signal. The first-order decay of the HSO4- signal was plotted against the reaction distance for each uptake experiment. Figure 4 shows typical HSO4- signal decays on kerosene soot as a function of injector position. The experiments were conducted at Ptotal ) 1.4 Torr and U ) 1864 cm s-1. The filled circles represent data measured on fresh soot, while the open circles correspond to a measurement on soot after 20 min
of exposure to H2SO4. Both decays followed the first-order kinetics. Linear regression analysis of each data set was performed to obtain the observed first-order rate constant kobs. The loss of H2SO4 on fresh soot was found to be slightly higher than that on soot that had been exposed to H2SO4 for about 20 min. Accordingly, the γ value decreased slightly from 0.028 for fresh soot to 0.022 for the H2SO4-exposed soot. Similar behaviors were observed for H2SO4 uptake on all three types of soot (i.e., methane, hexane, and kerosene) over the pressure range of 1.3-1.7 Torr. Significant deactivation of soot surfaces by chemical adsorption was previously reported for the uptake of NO2 (19) and N2O5 and O3 (20). In contrast to the previous studies, our measured decrease of γ was relatively minor for the H2SO4 uptake, usually within 20% for a ∼20 min exposure time. Using an estimated H2SO4 partial pressure (about 5 × 10-8 Torr) in the flow tube under our experimental conditions, the amount of H2SO4 taken up by ∼78 cm2 (geometric surface area) of soot was approximately 1 × 1015 molecules. The fractional soot surface coverage was roughly 3% for a 20 min exposure, assuming a monolayer deposition of H2SO4 molecules. During each kinetic run, the soot surface coverage was typically less than 1%. Previous studies postulated possible chemical mechanisms for the irreversible uptake of certain inorganic species on soot. For example, the soot surface was proposed to serve as a catalytic site for reactions such as NO2 + H2O, which did not cause soot deactivation in the atmosphere (19). In other cases, soot was suggested to be oxidized by the adsorbed species, leading to soot deactivation. Choi and Leu (18) proposed a bimolecular HNO3 decomposition mechanism on soot surfaces, resulting in formation of NO, NO2, and H2O, as well as soot oxidation. No reaction product was detected in the mass spectra during our experiments. In addition, the lack of a signature for significant soot deactivation during the 20 min time period suggested that the loss of H2SO4 on soot might involve little chemical reaction. Rather, it was likely explained by the sticky nature of H2SO4 and the porous structure of soot. If the uptake corresponds to a physical adsorption process, the γ value will eventually approach that of H2SO4 uptake on liquid H2SO4 when the soot surface is fully covered by H2SO4 (24). Uptake coefficient measurements of H2SO4 on soot performed in this study are summarized in Table 1, along with the experimental conditions and the R2 value for each linear fit. Since the decrease in uptake coefficients due to soot deactivation was within the error bars of our reported γ values, we did not differentiate the uptake coefficients on fresh soot from those on soot after H2SO4 exposure. For experiments conducted on three types of soot, small differences in the uptake coefficients were observed. The measured γ value was 0.018 ( 0.007 for kerosene soot at a total pressure of 1-2 Torr and 298 K, while smaller γ values were determined for methane (0.012 ( 0.006) and hexane (0.0076 ( 0.0016) soot under the same experimental conditions. The uncertainty represents the scatter in the data at the 1 standard deviation level and is not an estimate of systematic errors. We estimated that systematic uncertainty in our measured rate constants was within (10%, considering the possible sources of error in the measurements (i.e., gas flows, temperature, detection signal, and pressure) and in the gas-phase diffusion corrections. The uncertainty, however, did not include that associated with the assumption using the geometric area of deposited soot, to be discussed below. The uptake of other inorganic species on soot was found to vary considerably for different soot types. For example, in a recent study of the interaction of water vapor with soot, initial uptake coefficients of H2O on toluene, acetylene, and diesel soot were determined, while no measurable interaction
TABLE 1. Summary of Uptake Coefficient Measurements of H2SO4 on Soota Ptotal (Torr)
U (cm/s)
R2
γ
methane
1.56 1.56 1.51 1.51 1.51 1.51 1.51
1515 1815 1971 1971 1971 1494 1494
0.987 0.982 0.979 0.990 0.979 0.995 0.992
hexane
1.50 1.56 1.51 1.51 1.50 1.56 1.56 1.56 1.54 1.54
1342 1394 1748 1748 1715 1720 1720 2246 2393 2393
0.993 0.994 0.989 0.999 0.978 0.997 0.996 0.988 0.987 0.993
kerosene
1.51 1.51 1.51 1.54 1.30 1.30 1.71 1.51 1.51 1.51 1.50 1.40 1.40 1.50 1.60 1.70 1.50 1.51 1.51 1.51
2149 2149 2149 2402 1768 1768 2461 1951 1951 1946 2010 1864 1864 2005 2148 2307 2006 1971 1971 1971
0.997 0.978 0.984 0.995 0.994 0.991 0.974 0.974 0.994 0.976 0.991 0.995 0.989 0.994 0.990 0.998 0.973 0.966 0.982 0.970
0.010* 0.014* 0.020* 0.013 0.015 0.0053 0.0043 avb ) 0.012 ( 0.006 0.0083* 0.0080* 0.011* 0.0054 0.0080* 0.0062* 0.0056 0.0079* 0.0084* 0.0068 avb ) 0.0076 ( 0.0016 0.017* 0.014 0.014 0.011 0.014* 0.0096 0.0097 0.011* 0.0088 0.010 0.030* 0.028* 0.022 0.027* 0.021* 0.015* 0.024* 0.027* 0.024 0.022 avb ) 0.018 ( 0.007
soot type
a Values with an asterisk correspond to measurements on fresh soot, whereas values without an asterisk correspond to measurements on soot after about 20 min of exposure to H2SO4. b Averages are given with the range of 1 standard deviation of the mean.
between H2O and decane soot was observed at ambient temperature (16). Similarly, the uptake coefficient of HNO3 was found to vary considerably on FW2 carbon black, graphite, hexane, and kerosene soot (18). In a study of NO2 uptake on methane, hexane, propane, and kerosene soot, the uptake coefficient was found to vary within an order of magnitude for the different soot types (19). Our calculated γ values were based on the geometric surface area assumption, i.e., taking the surface area coated by soot in the flow tube as the soot surface area for reaction. In reality the deposited soot film might be highly porous and exhibit a large internal surface area. Using the Brunauer, Emmett, and Teller (BET) surface area of 25 m2 g-1 for methane soot (37), 91 m2 g-1 for kerosene soot, and 46 m2 g-1 for hexane soot (18), γ would be predicted to be approximately a factor of 5-18 less than that assuming a geometric surface area for a typical amount of 2 mg of soot in our experiments, considering the difference in surface area. Our reported γ values would represent the upper limit if the BET surface needs to be taken into account. To investigate the dependence of the uptake coefficient on internal surface areas, we performed additional experiments to measure γ for various total masses of the deposited soot. VOL. 39, NO. 15, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5725
hydrophobic (D. Zhang and R. Zhang, unpublished results). In the previous study of the mass accommodation coefficient of H2SO4 vapor on aqueous H2SO4, it was also concluded that the observed wall loss coefficient of H2SO4 vapor was independent of the gas-phase H2O concentration in the range of 5 × 1012 to 3 × 1016 molecules cm-3, which was correlated with RH (24).
FIGURE 5. Measured size distributions of kerosene soot aerosols using DMA for fresh soot (open circles) and soot after exposure to H2SO4 under the same flow conditions (solid lines). In all experiments, no apparent dependence of the measured γ on soot mass was observed in the mass range of 1-10 mg. Similar mass independence of γ was found for NO2 (19) and HNO3 (18) uptake on kerosene soot and HNO3 uptake on NaCl and KBr (38). There are several possible explanations to account for the lack of the mass dependence of the measured γ. The soot films had many layers of packing, and the H2SO4 vapor likely only sampled a limited portion of micropores. It was also plausible that the sticky nature of H2SO4 made diffusion into internal layers difficult and only the upper layers were reached by H2SO4 molecules, as proposed for HNO3 uptake on NaCl and KBr (38) and HNO3 uptake on kerosene soot (18). Hence, H2SO4 diffusion into the internal surface would be much slower than predicted by simplified diffusion models (18). The effect of internal surface areas of soot on the uptake coefficient of H2SO4 requires further experimental and theoretical studies. Our experiments were conducted under low relative humidity (RH < 1%). We were unable to measure the uptake coefficient at high RH since binary homogeneous nucleation of H2SO4 and H2O occurred (9) and complicated the measurements. A previous study investigating HNO3 uptake on n-hexane soot found little dependence on RH for RH as high as 80% (23); a low surface density of water adsorption sites was suggested as a possible explanation. Our own measurements of hygroscopicity of methane, hexane, and kerosene soot showed that those three soot types were
All uptake experiments in this work were conducted at 298 ( 2 K. In a previous study of the gas-phase adsorption of HNO3 on n-hexane soot, the uptake was found to increase with decreasing temperature between 228 and 298 K, with a negative enthalpy of adsorption (23). Similarly, the adsorption of H2O vapor on soot was found to be exothermic in the temperature range of 190-300 K (16). Nevertheless, recent studies suggested that the reactive uptake coefficients for NO2 and O3 on various hydrocarbon flame soot were independent of temperature over the range of 240-350 K (17, 39). Characterization of the Interaction of H2SO4 with Soot Using DMA and FTIR Spectroscopy. (1) DMA Measurements. Using the experimental apparatus shown in Figure 2, size distribution measurements were conducted for kerosene soot and soot-H2SO4 mixed particles to infer the effect of H2SO4 uptake on the size and concentration of soot particles. Figure 5 shows typical size distributions of fresh kerosene soot produced from the alcohol burner, as well as soot particles after being passed through a 96% H2SO4 reservoir. A particle sample flow of 650 sccm and a N2 sheath flow of 6.5 slpm were maintained for all measurements by means of a control valve and a flow controller. The residence time of the soot flow in the H2SO4 reservoir was estimated to be ∼20 s. Particle sizes were scanned from 10 to 450 nm by the DMA instrument. Figure 5 shows an approximate Gaussian distribution for both cases, with the highest concentration occurring at 200 nm for the fresh soot particles and 300 nm for the H2SO4-exposed soot particles. The increase in the soot particle sizes after exposure to H2SO4 vapor confirmed efficient uptake of H2SO4 on soot. Despite the size shift caused by H2SO4 uptake, the peak concentration remained nearly unchanged compared with that of the fresh soot particles. This suggests that new particle formation under the current experimental conditions was negligible. The
FIGURE 6. IR spectra of fresh methane soot aerosols (a) and methane soot-H2SO4 mixed aerosols (b) measured in the multipath optical cell. Spectra are baseline corrected, intensity normalized, and shifted for clarity. The blank between 2300 and 2400 cm-1 was originally a peak due to CO2 from the combustion and was removed from the spectra. 5726
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 15, 2005
irreversible adsorption of H2SO4 on soot caused significant condensational particle growth and resulted in internally mixed soot-H2SO4 particles, in qualitative agreement with the uptake coefficient measurement. (2) FTIR Spectroscopy Measurements. To characterize the chemical composition of soot particles with H2SO4 adsorpsorption and to further investigate the mechanism of H2SO4 uptake on soot, we performed FTIR spectroscopic measurements for soot particles with and without H2SO4 coating. Figure 6 shows the FTIR spectra of methane soot without (spectrum a) and with (spectrum b) H2SO4 exposure. The pressure inside the multipath cell was kept at 1 atm for all FTIR experiments due to the relatively low sensitivity of the technique. The absorption features of soot were attributed to the carbon skeleton, such as the -CtC-H (2180, 2120 cm-1), aromatic -C-H (3016 cm-1), and -O-H (3320, 3270 cm-1) groups. In a previous FTIR study using filter-deposited soot samples, much broader bands at 800-1600, 2800-3000, and 3200-3700 cm-1 were observed compared with the IR peaks from this study (40). This is likely due to the large absorption caused by deposited soot samples compared with the suspended soot aerosols in this work, although the authors attributed the broader bands to possible reaction products during the soot generation processes. In two early spectroscopic studies of hexane soot, absorption bands at 1260, 1440, 1590, 1700-1800, and 3040 cm-1 were identified associated with the carbon skeleton, oxygenated groups, or -OH (41, 42). Distinct S-O features were observed after the soot particles were exposed to H2SO4. Several absorption peaks at 1260, 1100, and 810 cm-1 appeared for the sample of H2SO4-coated aerosols. Those IR bands are close to those previously reported for liquid sulfuric acid, characteristic of the band locations of SO42- and HSO4- (43, 44). The sulfate bandwidths at half-height for the H2SO4-coated soot aerosols are typically in the range of 15-25 cm-1, much smaller than those collected using bulk liquid solutions, which are generally 100-200 cm-1 (43, 44). The narrow absorption bands are typical for S-O characteristics in the aerosol phase, as observed for (NH4)2SO4 aerosols where the widths for the SO42- bands were in the range of 14-44 cm-1 (45). As discussed above, our experiments were performed under low RH conditions, and hence, homogeneous nucleation of H2SO4/H2O to form an external mixture of soot and sulfate aerosols was precluded. This was particularly exemplified in the DMA experiments, showing an increase in the particle sizes without a change in the particle concentrations. The sulfate features in the IR spectra reflected internal H2SO4 mixing with soot, consistent with the uptake and DMA measurements. Two other observations were made on the basis of analysis of the IR spectra. First, the absorption bands of soot particles attributed to the carbon skeleton, such as the -CtCsH, aromatic, and -O-H groups, did not change appreciably after exposure to H2SO4 vapor. Also, if chemical reactions did occur on the soot surfaces, the IR spectrum of sootH2SO4 mixed aerosols might exhibit more oxygenated functionalities, such as -C-O and -CdO, as a result of oxidation of the soot surface. Such oxygenated functional groups were not evident in the IR spectra of the H2SO4-coated soot aerosols. It is likely, though, that the change in the absorption bands of soot due to H2SO4 coating and trace product formation were not adequately resolved in the IR spectra for small H2SO4-containing soot aerosols. The results of H2SO4 uptake on soot particles are crucial to understand how the soot-H2SO4 aerosols influence the atmospheric radiative balance, heterogeneous chemistry, and their CCN-forming potential. The present study reveals efficient uptake of H2SO4 for three types of soot, which is relatively fast compared with the uptake of other inorganic atmospheric species (such as HNO3 or NO2) on soot. This
suggests that the adsorption of soot by H2SO4 likely represents an important way to convert hydrophobic soot into CCN, considering the large abundance of soot and H2SO4 in the atmosphere from combustion, industrial sources, and automobile and aircraft emissions. Summary. We have reported the first measurements of uptake coefficients of H2SO4 on three types of soot at a temperature of 298 K and in the pressure range of 1-2 Torr. H2SO4 loss on soot was found to be irreversible. A slight decrease in the uptake coefficient was observed with extended exposure of soot to H2SO4. Additional experiments using the DMA and FTIR spectroscopy techniques confirmed the interaction of H2SO4 with soot. The results imply that the heterogeneous reactions between soot and sulfuric acid proceed efficiently and that the activation of soot by H2SO4 represents an important way to convert hydrophobic soot into CCN.
Acknowledgments This work was supported by the Robert A. Welch Foundation (Grant A-1417) and U.S. Environmental Protection Agency (EPA) (Grant R03-0132).
Note Added after ASAP Publication Due to a production error the heading for column 5 in Table 1 was presented incorrectly in the version published ASAP June 21, 2005; the corrected version was published ASAP July 15, 2005.
Literature Cited (1) Penner, J. E.; Eddleman, H.; Novakov, T. Towards the development of a global inventory for black carbon emissions. Atmos. Environ. 1993, 27A, 1277-1295. (2) Horvath, H. Atmospheric light absorptionsa review. Atmos. Environ. 1993, 27A, 293-317. (3) Jensen, E. J.; Toon, O. B. The potential impact of soot particles from aircraft exhaust on cirrus clouds. Geophys. Res. Lett. 1997, 24, 249-252. (4) Hallett, J.; Hudson, J. G.; Rogers, C. F. Characterization of combustion aerosols for haze and cloud formation. Aerosol Sci. Technol. 1989, 10, 70-83. (5) Brown, R. C.; Miake-Lye, R. C.; Anderson, M. R.; Kolb, C. E.; Resch, T. J. Aerosol dynamics in near-field aircraft plumes. J. Geophys. Res. 1996, 101, 22939-22953. (6) Schumann, U.; Strom, J.; Busen, R.; Baumann, R.; Gierens, K.; Krautstrunk, M.; Schroder, F. P.; Stingl, J. In situ observations of particles in jet aircraft exhausts and contrails for different sulfur-containing fuels. J. Geophys. Res. 1996, 101, 6853-6869. (7) Weber, R. J.; Marti, J. J.; McMurry, P. H.; Eisele, F. L.; Tanner, D. J.; Jefferson, A. Measured atmospheric new particle formation rates: Implications for nucleation mechanisms. Chem. Eng. Commun. 1996, 151, 53-64. (8) Viisanen, Y.; Kulmala, M.; Laaksonen, A. Experiments on gasliquid nucleation of sulfuric acid and water. J. Chem. Phys. 1997, 107, 920-926. (9) Zhang R.; Suh, I.; Zhao, J.; Zhang, D.; Fortner, E. C.; Tie, X.; Molina, L. T.; Molina, M. J. Atmospheric new particle formation enhanced by organic acids. Science 2004, 304, 1487-1490. (10) Jacobson, M. Z. A physically-based treatment of elemental carbon optics: Implications for global direct forcing of aerosols. Geophys. Res. Lett. 2000, 27, 217-220. (11) Jacobson, M. Z. Strong radiative heating due to the mixing state of black carbon in the atmospheric aerosols. Nature 2001, 409, 695-697. (12) Hasegawa, S.; Ohta, S. Some measurements of the mixing state of soot-containing particles at urban and non-urban sites. Atmos. Environ. 2002, 36, 3899-3908. (13) Okada, K.; Heintzenberg, J. Size distribution, state of mixture and morphology of urban aerosol particles at given electrical mobilities. J. Aerosol Sci. 2003, 34, 1539-1553. (14) Mallet, M.; Roger, J. C.; Despiau, S.; Putaud, J. P.; Dubovik, O. A study of the mixing state of black carbon in urban zone. J. Geophys. Res. 2004, 109, D04202, doi: 10.1029/2003JD003940. (15) Seisel, S.; Lian, Y.; Keil, T.; Trukhin, M. E.; Zellner, R. Kinetics of the interaction of water vapour with mineral dust and soot surfaces at T ) 298 K. Phys. Chem. Chem. Phys. 2004, 6, 19261932. VOL. 39, NO. 15, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
5727
(16) Alcala-Jornod, C.; Rossi, M. J. Chemical kinetics of the interaction of H2O vapor with soot in the range 190 K e T g 300 K: A diffusion tube study. J. Phys. Chem. A 2004, 108, 10667-10680. (17) Lelievre, S.; Bedjanian, Y.; Pouvesle, N.; Delfau, Vovelle, J. L.; C.; LeBras, G. Heterogeneous reaction of ozone with hydrocarbon flame soot. Phys. Chem. Chem. Phys. 2004, 6, 1181-1191. (18) Choi, W.; Leu, M. T. Nitric acid uptake and decomposition on black carbon (soot) surfaces: Its implications for the upper troposphere and lower stratosphere. J. Phys. Chem. A 1998, 102, 7618-7630. (19) Longfellow, C. A.; Ravishankara, A. R.; Hanson, D. R. Reactive uptake on hydrocarbon soot: Focus on NO2. J. Geophys. Res. 1999, 104, 13833-13840. (20) Longfellow, C. A.; Ravishankara, A. R.; Hanson, D. R. Reactive and nonreactive uptake on hydrocarbon soot: HNO3, O3, and N2O5. J. Geophys. Res. 2000, 105, 24345-24350. (21) Al-Abadleh, H. A.; Grassian, V. H. Heterogeneous reaction of NO2 on hexane soot: A Knudsen cell and FT-IR study. J. Phys. Chem. A 2000, 104, 11926-11933. (22) Prince A. P.; Wade, J. L.; Grassian, V. H.; Kleiber, P. D.; Young, M. A. Heterogeneous reactions of soot aerosols with nitrogen dioxide and nitric acid: atmospheric chamber and Knudsen cell studies. Atmos. Environ. 2002, 36, 5729-5740. (23) Aubin, D. G.; Abbatt, J. P. Adsorption of gas-phase nitric acid to n-hexane soot: Thermodynamics and mechanism. J. Phys. Chem. A 2003, 107, 11030-11037. (24) Poschl, U.; Canagaratna, M. J.; Jayne, T.; Molina, L. T.; Worsnop, D. R.; Kolb, C. E.; Molina, M. J. Mass accommodation coefficient of H2SO4 vapor on aqueous sulfuric acid surfaces and gaseous diffusion coefficient of H2SO4 in N2/H2O. J. Phys. Chem. A 1998, 102, 10082-10089. (25) ten Brink, H. M. Reactive uptake of HNO3 and H2SO4 in sea-salt (NaCl) particles. J. Aerosol Sci. 1998, 29, 57-64. (26) Jefferson, A.; Eisele, F. L.; Ziemann, P. J.; Weber, R. J.; Marti, J. J.; McMurry, P. H. Measurements of the H2SO4 mass accommodation coefficient onto polydisperse aerosol. J. Geophys. Res. 1997, 102, 19021-19028. (27) Santoro, R. J.; Semerjian, H. G.; Dobbins, R. A. Soot particle measurements in diffusion flames. Combust. Flame 1983, 51, 203-218. (28) Fortner, E. C.; Zhao, J.; Zhang, R. Development of ion driftchemical ionization mass spectrometry. Anal. Chem. 2004, 76, 5436-5440. (29) Keyser, L. F.; Moore, S. B.; Leu, M.-T. Surface reaction and pore diffusion in flow-tube reactors. J. Phys. Chem. 1991, 95, 54965502. (30) Zhang, R.; Jayne, J. T.; Molina, M. J. Heterogeneous Interactions of ClONO2 and HCl with Sulfuric Acid Tertahydrate: Implications for the Stratosphere. J. Phys. Chem. 1994, 3, 867-874. (31) Zhang, R.; Leu, M. T.; Keyser, L. F. Heterogenous reactions involving ClONO2, HCl, and HOCl on liquid sulfuric acid surfaces. J. Phys. Chem. 1994, 98, 13563-13574.
5728
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 15, 2005
(32) Brown, R. L. Tubular flow reactions with first-order kinetics. J. Res. Natl. Bur. Stand. (U. S.) 1978, 83, 1-8. (33) Hanson, D. R.; Eisele, F. Diffusion of H2SO4 in humidified nitrogen: Hydrated H2SO4. J. Phys. Chem. A 2000, 104, 17151719. (34) Kasper, M.; Siegmann, K.; Sattler, K. Evaluation of an in situ sampling probe for its accuracy in determining particle size distributions from flames. J. Aerosol Sci. 1997, 28, 1569-1578. (35) Zhang, R.; Leu, M. T.; Keyser, L. F. Hydrolysis of N2O5 and ClONO2 on the H2SO4/HNO3/H2O ternary solutions under stratospheric conditions. Geophys. Res. Lett. 1995, 22, 1493-1496. (36) Zhang, R.; Leu, M. T.; Keyser, L. F. Heterogeneous chemistry of HONO on liquid sulfuric acid: A new mechanism of chlorine activation on stratospheric sulfate aerosols. J. Phys. Chem. 1996, 100, 339-345. (37) Tesner, P. A.; Shurupov, S. V. Some physico-chemical parameters of soot formation during pyrolysis of hydrocarbons. Combust. Sci. Technol. 1995, 105, 147-161. (38) Fenter, F.; Caloz, F.; Rossi, M. J. Heterogeneous kinetics of N2O5 uptake on salt, with a systematic study of the role of surface presentation (for N2O5 and HNO3). J. Phys. Chem. 1996, 100, 1008-1019. (39) Lelievre, S.; Bedjanian, Y.; Laverdet, G.; Le Bras, G. Heterogeneous reaction of NO2 with hydrocarbon flame soot. J. Phys. Chem. A 2004, 104, 10807-10817. (40) Kirchner, U.; Scheer, V.; Vogt, R. FTIR spectroscopic investigation of the mechanism and kinetics of the heterogeneous reactions of NO2 and HNO3 with soot. J. Phys. Chem. A 2000, 104, 89088915. (41) Akhter, M. S.; Chughtai, A. R.; Smith, D. M. The structure of hexane soot-I: spectroscopic studies. Appl. Spectrosc. 1985, 39, 143-153. (42) Akhter, M. S.; Chughtai, A. R.; Smith, D. M. The structure of hexane soot-II: extraction studies. Appl. Spectrosc. 1985, 39, 154-167. (43) Zhang, R.; Wooldridge, P. J.; Abbatt, J. P. D.; Molina, M. J. Physical chemistry of the H2SO4/H2O binary system at low temperatures: Implications for the stratosphere. J. Phys. Chem. 1993, 97, 7351-7358. (44) Middlebrook, A. M.; Iraci, L. T.; McNeill, L. S.; Koehler, B. G.; Wilson, M. A.; Saastad, O. W.; Tolbert, M. A.; Hanson, D. R. Fourier transform-infrared studies of thin H2SO4/H2O filmss Formation, water uptake, and solid-liquid phase-changes. J. Geophys. Res. 1993, 98, 20473-20481. (45) Weis, D. D.; Ewing, G. E. Infrared spectroscopic signatures of (NH4)2SO4 aerosols. J. Geophys. Res. 1996, 101, 18709-18720.
Received for review February 23, 2005. Revised manuscript received May 11, 2005. Accepted May 20, 2005. ES050372D
