Temperature-Dependent Henry's Law Constants of Atmospheric

Sep 10, 2013 - Department of Chemistry, University of Colorado Denver, Denver, Colorado ... Science, School of Chemistry, Beijing Institute of Technol...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/JPCA

Temperature-Dependent Henry’s Law Constants of Atmospheric Organics of Biogenic Origin Chunbo Leng,†,‡ J. Duncan Kish,† Judas Kelley,† Mindy Mach,† Joseph Hiltner,† Yunhong Zhang,‡ and Yong Liu*,† †

Department of Chemistry, University of Colorado Denver, Denver, Colorado 80217, United States The Institute of Chemical Physics, Key Laboratory of Cluster Science, School of Chemistry, Beijing Institute of Technology, Beijing 100081, China

‡

S Supporting Information *

ABSTRACT: There have been growing interests in modeling studies to understand oxidation of volatile organic compounds in the gas phase and their mass transfer to the aqueous phase for their potential roles in cloud chemistry, formation of secondary organic aerosols, and fate of atmospheric organics. Temperature-dependent Henry’s law constants, key parameters in the atmospheric models to account for mass transfer, are often unavailable. In the present work, we investigated gas−liquid equilibriums of isoprene, limonene, α-pinene, and linalool using a bubble column technique. These compounds, originating from biogenic sources, were selected for their implications in atmospheric cloud chemistry and secondary organic aerosol formation. We reported Henry’s law constants (KH), first order loss rates (k), and gas phase diffusion coefficients over a range of temperatures relevant to the lower atmosphere (278−298 K) for the first time. The measurement results of KH values for isoprene, limonene, α-pinene, and linalool at 298 K were 0.036 ± 0.003; 0.048 ± 0.004; 0.029 ± 0.004; and 21.20 ± 0.30 mol L−1 atm−1, respectively. The fraction for these compounds in stratocumulus and cumulonimbus clouds at 278 K were also estimated in this work (isoprene, 1.0 × 10−6, 6.8 × 10−6; limonene, 1.5 × 10−6, 1.0 × 10−5; α-pinene, 4.5 × 10−7, 3.1 × 10−6; and linalool, 6.2 × 10−4, 4.2 × 10−3). Our measurements in combination with literature results indicated that noncyclic alkenes could have smaller KH values than those of cyclic terpenes and that KH values may increase with an increasing number of double bonds. It was also shown that estimated Henry’s law constants and their temperature dependence based on model prediction can differ from experimental results considerably and that direct measurements of temperature-dependent Henry’s law constants of atmospheric organics are necessary for future work.

1. INTRODUCTION Secondary organic aerosols (SOAs), a major component of atmospheric aerosols, are typically generated during gas phase oxidation of volatile organic compounds (VOCs) to form semivolatile products, which undergo condensation and/or absorption into aerosol phase. In the past decade, there has been a surge of interest in the mechanistic understanding of their formation processes. Considerable efforts have been devoted to detailed measurements in both field and laboratory environments to improve our capability to accurately predict their impacts on atmospheric chemical balance and global radiation budget. However, gas phase oxidation mechanisms alone are unable to reproduce SOA yields in the laboratory.1−3 Lately, growing evidence has indicated that oxidation of many secondary species in condensed phase could contribute to the formation of SOAs,1,3−15 and aqueous phase chemistry could account for up to 20% of the total SOA yield.9 These low vapor pressure condensable organics include carbonyl species, alcohols, carboxylic acids, and nitrogen containing species, and they are usually generated by gas phase oxidation of VOCs © 2013 American Chemical Society

from anthropogenic and biogenic sources. Among all the nonmethane VOCs, isoprene has received the most attention8,16−22 as it has the largest emission rate.23 Also, recent work has revealed that in-cloud oxidation of water-soluble isoprene can directly contribute to an increase in SOA mass.17 In addition to isoprene, other VOCs could also partition into aqueous phase and undergo aqueous photolysis reaction and/or reaction with dissolved OH, in turn, contributing to the SOA formation.24 Moreover, one recent study has suggested that dissolution of VOCs can be greatly enhanced in dew water.25 As such, recently, VOCs’ partition in aqueous phase and participation in aqueous phase chemistry have drawn increasing attention. In addition to experimental studies,26−28 many modeling approaches have also been applied to understand the atmospheric organic cloud chemistry and the formation of SOA.5,19−22 For example, Mouchel-Vallon et al.19 developed a Received: April 12, 2013 Revised: August 12, 2013 Published: September 10, 2013 10359

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A

Article

Figure 1. Schematic diagram of bubble column system.

from the bottom of a column containing a liquid into the headspace, where changes in gas phase concentrations of the solute are then monitored. Unlike the static equilibrium method, only relative concentrations of the solute in the gas phase, over time, are needed for measurements of Henry’s law constants. This avoids uncertainties resulting from determining absolute concentrations of the solute in both gas and aqueous phases in the static equilibrium method.29 In the present work, we selected three biogenic hydrocarbons (isoprene, limonene, and α-pinene) as they can serve as precursors for many other condensable organics in aqueous phase. We also studied linalool as it is a simple oxidation product of naturally occurring terpene. We measured their Henry’s law constants and temperature dependences over a temperature range relevant to the lower atmosphere (278−298 K), for the first time, using the bubble column technique. In addition, we also are the first to report first-order loss rates and diffusion coefficients of all the species as a function of temperature and evaluate their characteristic time to achieve gas−liquid equilibrium and partition between the gas and aqueous phases.

fully explicit chemical mechanism, GECKO-A (Generator of Explicit Chemistry and Kinetics of Organics in the Atmosphere), to describe oxidation of organics in the gas phase and their mass transfer to the aqueous phase. To demonstrate, they investigated production of water-soluble compounds originating from gaseous oxidation of three hydrocarbons (isoprene, octane, and α-pinene). In their simulation, temperaturedependent Henry’s law constants (KH(T)) were key parameters for phase transfer. Unfortunately, the current collection of Henry’s law constants especially temperature dependence data required to develop detailed models far exceeds the number of species that has experimental data available. As a result, in their paper, if KH values at 298 K were unavailable from literature, an empirical group contribution method was used for estimation. Moreover, because of a lack of enthalpy of solution ΔHsol data, which determines the KH(T), default values of −50 kJ mol−1 were used for thousands of species in their model. Estimated KH(T) values of isoprene were also applied to other recent modeling studies.20−22 Apparently, to reduce uncertainties in atmospheric modeling studies and improve our current knowledge state of organic cloud chemistry and SOA formation, there is a scientific need of temperature-dependent Henry’s law constants for not only the volatile precursors but also their oxidation products. Henry’s Law describes the relationship between the solubility of a gas and its partial pressure above the liquid, and it plays a central role in regulating the partition of myriads of VOCs between the gas phase and the aqueous phase. Consequently, it affects their participation in both aerosol and cloud formation processes, and their contribution to a variety of environmental issues including air pollution, climate change, and public health. At present, Henry’s law constants of many VOCs, including those closely related to atmospheric chemistry, are not directly measured. Instead, they are only estimated from thermodynamic data, and more importantly, temperature dependences of Henry’s law constants are often unavailable.29,30 The bubble column technique, based on a dynamic equilibrium, has been commonly used for determination of Henry’s law constants.31−35 In the measurements, the solute of interest rises

2. EXPERIMENTAL SECTION 2.1. Materials. Toluene (Fisher), α-pinene (98%, Acros), isoprene (99%, Alfa Aesar), limonene (96%, Acros), and linalool (97%, Alfa Aesar) were used as delivered without further purification. 2.2. Methods. Henry’s law constants were measured at five different temperatures (278, 283, 288, 293, and 298 K) using a temperature-controlled bubble column system (Figure 1). Details about the system have been described elsewhere.31−33,36 In brief, it was largely composed of two double-jacketed bubble columns: the first one (35 cm long) was used for generating organic vapor and the second one (80 cm long) was used for absorbing organic vapor. Organic vapors were generated by bubbling organic liquids in the first column with nitrogen gas (zero grade). The exiting gaseous mixture was then passed through water in the second column to establish an equilibrium between the solutes in the exiting gas and the solutes in the 10360

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A

Article

aqueous phase. Low flow rates of N2 (50−100 mL min−1) and high liquid depth (500−1000 mL, 40−80 cm depth) were used to attain the gas−liquid equilibrium. Two 3-way valves were employed to switch the N2 so it either went through the first column or bypassed it. When the 3-way valves were set to the bypass position, the dissolved organic species in the second column were purged with N2 at various flow rates (500−1000 mL min−1). Concentrations of organic solutes in the headspace were monitored as a function of time. The concentration decay of organic solutes in the gas phase can be utilized to derive Henry’s law constants based on the following equation:31 ln(C0/C t) = [Φ/(KHRTV ) + k]t

(1)

where C0 and Ct are the concentration of the organic vapor at equilibrium and at time t (sec) after the equilibrium, respectively; Φ is the gas flow rate (mL min−1); KH is the Henry’s law constant (M atm−1); R is the gas constant (0.082 atm K−1 M−1); T is the temperature (K); V is the water volume (mL); and k is the first-order loss rate (s−1). Throughout the experiments, the entire system was under atmospheric pressure and both columns were temperaturecontrolled by a refrigerated circulator (VWR 1160S). Concentrations of toluene, isoprene, limonene, and α-pinene in the headspace were measured by a GC-FID (HP5890). Since no separation is needed, column was not used in this work. Instead, a six-port valve and a two-position switch (VICI Valco instruments) were used to control gas sample injection with volumes ranging from 5 to 40 μL. The oven temperature was maintained at 473 K throughout the experiments. Peak areas of FID signals, which were proportional to gas phase concentrations of organics, were acquired at every 50−60 s for about 10−15 min and used to derive the Henry’s law constants. Concentrations of isoprene and linalool were also determined by a FTIR spectrometer with a DTGS detector (Thermo Avatar 360) coupled to a Teflon gas cell (10 cm path length and equipped with CaF2 windows). In these cases, Henry’s law constants were acquired based on integrated absorbance peaks (960−1020 cm−1) for isoprene and (1180−1240 cm−1) for linalool as a function of time. The concentration of linalool was typically recorded for about 20−30 min with a collection interval of 2 min. For each reaction condition, experiments were repeated at least 3 times and uncertainties are obtained from standard deviation of the measurements.

Figure 2. Measured ln(C0/Ct) vs time for the toluene system at 298 K with Φ from 500 to 1000 mL min−1 and water volume of 750 mL.

Figure 3. Measured d ln(C0/Ct)/dt vs Φ/V for the toluene at 298 K for determination of KH. Vertical bars indicate standard deviation values.

linearity is expected, and 1/(KHRT) and k can be acquired from the slope and the intercept, respectively. Our measurement of toluene’s KH at 298 K was 0.14 ± 0.03 mol L−1 atm−1, and it is in excellent agreement with literature values ranging from 0.13 to 0.21 mol L−1 atm−1.31,38,39 The first-order loss rate was (8.3 ± 2.0) × 10−4 s−1 at 298 K, consistent with the recently reported value of (5 ± 1) × 10−4 s−1.31 ΔHsol (molar enthalpy of solution; details about ΔHsol determination will be discussed in next section) for toluene was determined in the present work to be −42.1 ± 3.9 kJ mol−1, in reasonable agreement with the literature value of −38.06 kJ mol−1.40 Such agreement has indicated that our experimental conditions have probably met all the requirements for accurate determination of Henry’s law constants. 3.2. Henry’s Law Constants of Isoprene, Limonene, and α-Pinene. The Henry’s law constants and first-order loss rates of isoprene, limonene, and α-pinene were measured at five temperatures; all results are listed in Table 1. Exponential decays of the measured relative concentration of all compounds over time are included in the Supporting Information. The measured isoprene Henry’s law constant at 298 K was 0.036 ± 0.003 mol L−1 atm−1, in reasonable agreement with the most recently measured value of 0.028 mol L−1 atm−1.41 The Henry’s law constants of limonene and α-pinene at 298 K were 0.048 ± 0.004 and 0.029 ± 0.004 mol L−1 atm−1, respectively. To the best of our knowledge, no measured Henry’s law constants of

3. RESULTS AND DISCUSSION 3.1. Measurement Validation. Henry’s Law is a limiting law. To accurately measure the Henry’s law constants, in addition to the gas−liquid equilibrium requirement, it is pivotal to keep the concentration of the dissolved organic solution sufficiently dilute; the partial pressure of organic vapor small compared to the total system pressure; and the liquid under constant volume and temperature conditions. In this work, our system was first validated by measuring the Henry’s law constant of toluene at 298 K, which has been relatively well studied.31,34,37 For example, Lee et al.31 and Hoff et al.37 determined KH values of 0.13 mol L−1 atm−1 and 0.17 mol L−1 atm−1, respectively, using the bubble column technique, and Mackay et al.34 calculated a KH value of 0.15 mol L−1 atm−1 based on a thermodynamics model. Figure 2 shows linear plots of ln(C0/Ct) versus time for toluene at 298 K with different flow rates, Φ, as predicted by the eq 1. With the slopes derived from curve fitting of the plots in Figure 2, d ln(C0/Ct)/dt versus Φ/V plots can be obtained (Figure 3). According to eq 1, 10361

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A

Article

Temperature in the atmosphere spans a wide range from ∼220 K at the tropopause to close to room temperature near sea level. Temperature dependence of Henry’s law constants affects both seasonal and altitudinal dependence of atmospheric organic partition between gas and liquid phases. In the present work, we reported the temperature dependences of Henry’s law constants for isoprene, limonene, and α-pinene for the first time (Table 1). Unsurprisingly, the KH of all three organics increased with decreasing temperatures, conceivably due to lower solubility at a higher temperature. The Henry’s law constants as a function of temperature can be described by the following equation:

Table 1. Summary of Henry’s Law Constants, First-Order Loss Rates, and Enthalpies of Solution for Limonene, Isoprene, α-Pinene, and Linalool in Water

KH(T ) = KH(298) × exp[(−ΔH /R )(1/T − 1/298)] (2)

where KH(T) is Henry’s law constant for a given temperature and ΔHsol is the molar enthalpy of solution (kJ mol−1). Here, the temperature dependence is −d[ln KH(T )]/d(1/T ) = ΔH /R

(3)

Figure 4 showed the temperature dependent KH of isoprene. According to eq 3, the slope of the linear plot of −ln(KH) vs 1/

limonene and α-pinene are available, and the estimated Henry’s law constants derived from extrapolated vapor pressure for limonene and α-pinene are 0.0034 and 0.031 mol L−1 atm−1.42 As seen, experimentally measured and calculated KH values can differ by 1 order of magnitude. First-order loss rates of isoprene, limonene, and α-pinene were on the order of 10−3 s−1 at 298 K, slightly larger than that of toluene. In the aqueous phase, dissolved organic compounds could undergo different reactions including oxidation by radicals in water (e.g., OH and NO3), photolysis, hydration, dissociation, esterification, and aldol condensation.24 Given the chemical nature of hydrocarbons, their aqueous phase losses are likely attributable to their interaction with trace oxidants such as OH radicals in water. It should be pointed out that although the temperaturedependent loss rate data present an unusual U-shape, given the complex nature of aqueous phase reaction loss and the much larger errors in the measurements, the observed fluctuation probably should be treated as no temperature dependence. Note the KH value of limonene, a cyclic terpene, was slightly larger than that of isoprene, which is a noncyclic alkene. Also the measured KH value of α-pinene, a monoalkene, was smaller than those of isoprene and limonene, both of which are dienes. Despite very limited data from the current study, the pattern observed here was consistent with observations from Yaws and Yang43 in which many straight chain alkenes, cyclic terpenes, monoalkenes, and dienes were investigated. The results showed Henry’s law constants of noncyclic chain alkenes were generally smaller than those of cyclic terpenes, and the Henry’s law constants increased with an increasing number of double bonds. This may be explained by difference in polarizability of the molecules. Polarizability generally increases as volume occupied by electrons increases, and both the ring and diene structure would favor higher polarizability in molecules, resulting in stronger interaction with water. Henry’s law constants represent the solubility of a solute in a solution, and our results may indicate that dienes and cyclic alkenes could have a slightly higher propensity to stay in water than monoalkenes and straight chain alkenes.

Figure 4. Measured d ln(C0/Ct)/dt vs Φ/V for isoprene at five different temperatures. Vertical bars indicate standard deviation values.

Figure 5. Measured ln KH vs 1/T for isoprene, limonene, α-pinene, and linalool. ΔH values were derived from curve fitting. Vertical bars indicate standard deviation values. 10362

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A

Article

method was 0.035 ± 0.006 mol L−1 atm−1 at 298 K, in excellent agreement with the obtained value from the GC-FID method (0.036 ± 0.003 mol L−1 atm−1). This has demonstrated the feasibility of the bubble column system coupled with an FTIR/ gas cell for determination of Henry’s law constants of atmospheric organics. The measured linalool Henry’s law constant and first-order loss rate at different temperatures were also given in Table 1. The Henry’s law constants of linalool ranged from 21.20 to 62.08 mol L−1 atm−1 from 298 to 278 K, nearly 2 orders of magnitude higher than that of isoprene. Linalool is an unsaturated alcohol originating from isoprene. The addition of the OH functional group allows for greater interaction with water via hydrogen bonding and considerably enhances its solubility in water. This observation was consistent with enhanced hygroscopicity of atmospheric organics when they became more oxygenated during chemical aging. Our measured linalool Henry’s law constant at 298 K was smaller than the literature value of 46.51 mol L−1 atm−1 reported by Altschuh et al.45 The disagreement likely came from differences in experimental methods. In their studies of Henry’s law constants for 86 compounds at 298 K, for substances with relatively large solubility such as linalool, they did not use a vapor generation column; instead, they added a small amount of pure solute directly to water. This method may have resulted in a less equilibrated gas−liquid system and led to larger uncertainty. ΔHsol of linalool was reported here for the first time (−34.3 ± 2.5 kJ mol−1), similar to those of isoprene and limonene. Note the loss rate for linalool is markedly different from those of hydrocarbons, which are on the same order of magnitude. Such difference is probably because linalool has a higher tendency to be solvated in the water, lowering its probability to interact with other trace oxidants in water.

T yielded ΔH (Figure 5). The measured molar enthalpy of solution for isoprene was −35.7 ± 2.8 kJ mol−1, close to the ΔHsol value of 1,3-butadiene, −37.4 kJ mol−1,44 which has a similar molecular structure to isoprene. Following the same procedure, we obtained temperature dependent Henry’s law constants of limonene and α-pinene and sequentially the molar enthalpies of solution. Our measurement results were ΔHlimonene = −33.7 ± 3.1 kJ mol−1 and ΔHα‑pinene = −12.9 ± 2.7 kJ mol−1. For a typical dissolution process of soluble ionic compounds, three key steps are involved. It starts with an endothermic step to break ionic bonds (lattice energy) holding the solute ions together, followed by solvation process during which the solvent surrounds the solute ions forming intermolecular or ionic bonds between the solute and solvent. Solvation is an exothermic process. Afterward, the clusters of solute and solvent undergo an endothermic step to be redistributed evenly throughout the mixture. However, for covalent compounds, since no ions are formed in the dissolution process, the first step only needs little energy to break intermolecular forces. As a result, overall enthalpy change in a dissolution process for covalent compounds is expected to be dominated by the solvation. ΔHα‑pinene is significantly smaller than the others, probably due to its two ring bulky and rigid structure in the molecule, and the steric effect makes solvation more difficult for α-pinene. It should be pointed out both ΔHsol values for isoprene and α-pinene are quite differently than the default ones of −50 kJ mol−1 used in the GECKO-A model.19 3.3. Henry’s Law Constant of Linalool. The FID detector is particularly suitable for hydrocarbon. In this work, to minimize uncertainty resulting from the low sensitivity of FID toward linalool, an alterative DTGS detector in an FTIR spectrometer coupled to a Teflon gas cell was used. To validate the approach, we also measured Henry’s law constant of isoprene here. Figure 6 shows the exponential decay of the integrated absorbance of the peaks (960−1020 cm−1), which was used to measure relative concentrations of the solute in the gas phase over time and in turn determine the Henry’s law constant. The KH value of isoprene by the FTIR/gas cell

4. ATMOSPHERIC IMPLICATIONS For each atmospheric species, its partition into cloud droplets followed by aqueous phase reactions involves multiple steps including gas phase species diffusion to the air−water interface; interfacial transport; hydrolysis/ionization of the species in aqueous phase; aqueous phase diffusion of ionic and nonionic species inside the cloud droplets; and chemical reaction in the droplet.46 To assess the fate of atmospheric organics and their potential contribution to cloud chemistry and SOA formation, effects of all mass transport limitations must be taken into account. The interfacial transport of the species offers a key resistance for mass transfer, and the characteristic time scale associated with establishing phase equilibrium at the gasparticle interface can be estimated by the following equations:19,30,33,47 τH = [(r 2 /(3Dg )) + (4r /(3αν))]R gTKH

(4)

Dg = (M ref /M )1/2 Dg,ref

(4.1)

ν = (8R gT /πM )1/2

(4.2)

where r is the radius of cloud droplet; α is the mass accommodation coefficient; KH is the Henry’s Law constant (M atm−1); ν is the mean thermal velocity of gas molecules; Dg is the gas phase diffusion coefficient (cm2 s−1); M is the molecular weight (g mol−1); and “ref” subscripts denote values for a reference species. Water was used as a reference (D = 0.214 cm2 s−1 in air at 298 K) to estimate the diffusion coefficients of others.48 A mass accommodation coefficient of

Figure 6. Exponential decay of integrated absorbance of isoprene peaks (960−1020 cm−1) was used to measure relative concentrations of the solute in the gas phase over time. 10363

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A

Article

Table 2. Summary of Characteristic Time to Achieve Interfacial Equilibrium and the Fraction in Two Different Clouds fraction in condensed phase (×106) −1

−6

substance

T (K)

Dg (cm s )

τ/10

isoprene

298 293 288 283 278 298 293 288 283 278 298 293 288 283 278 298 293 288 283 278

0.110 0.106 0.103 0.100 0.096 0.078 0.075 0.073 0.071 0.068 0.078 0.075 0.073 0.071 0.068 0.073 0.071 0.069 0.066 0.064

1.27 1.55 2.01 2.83 3.62 2.38 3.32 4.17 5.72 7.53 1.42 1.71 1.82 2.16 2.29 1.12 1.39 1.81 2.34 3.39

limonene

α-pinene

linalool

2

0.05 for all organics49 and a cloud droplet radius of 5.6 μm were used here.50 Under these conditions, the characteristic times required to achieve gas−droplet equilibrium at 298 K were 1.27 × 10−6 s for isoprene, 2.38 × 10−6 s for limonene, 1.42 × 10−6 s for α-pinene, and 1.12 × 10−3 s for linalool (Table 2). According to eq 4, the characteristic time is a function of T, Dg, ν, and KH, all of which were dependent on temperature. To look into the effect of temperature on the characteristic time, we also needed temperature-dependent diffusion coefficients Dg(T) of all the species. Theoretical values for the temperaturedependent diffusion coefficients can be obtained using the 6− 12 Lennard-Jones potential model based on gas kinetic theory.51−54 The calculation method and the individual and binary collision parameters of H2O used in the calculation have been described in details elsewhere.48,55 In brief, within the 6− 12 Lennard-Jones potential model, the temperature-dependent diffusion coefficient can be determined as follows:48,51−55 D(T ) = 0.002628T1.5/[(2μ)0.5 σ 2 Ω(1,1)(θ )]

stratocumulus 0.389 0.461 0.585 0.803 0.999 0.514 0.700 0.857 1.15 1.47 0.307 0.361 0.373 0.433 0.448 2.28 × 2.74 × 3.49 × 4.41 × 6.22 ×

103 103 103 103 103

cumulonimbus 2.65 3.14 3.99 5.48 6.81 3.50 4.78 5.85 7.82 10.0 2.09 2.46 2.54 2.95 3.05 1.55 1.87 2.38 3.00 4.23

102 102 102 102 102

× × × × ×

103 103 103 103 103

temperature from a known experimental value by the following equation: D(T2) = D(T1)(T2/T1)1.5 [Ω(T1)/Ω(T2)]

(7)

With eq 7, we calculated the diffusion coefficients of water molecules in air at different temperatures (Table 3). The Table 3. Summary of Calculated Diffusion Coefficients of Water in Air at Different Temperatures and Parameters for the Calculations T

θ

Ω(T)

D (water/air cm2 s−1)

298 293 288 283 278

1.604 1.577 1.550 1.523 1.496

1.166 1.175 1.183 1.191 1.200

0.214 0.207 0.200 0.194 0.187

(5)

calculated diffusion coefficients of water in air were 0.207 cm2 s−1 at 293 K, 0.200 cm2 s−1 at 288 K, 0.194 cm2 s−1 at 283 K, and 0.187 cm2 s−1 at 278 K. Our calculated values are approximately 20% lower compared to the values from Bolz and Tuve,57 but within 5% difference compared to the values from Nellis and Klein.58 They were then used as references to obtain diffusion coefficients of isoprene, limonene, α-pinene, and linalool at lower temperatures (Table 2). On the basis of the diffusion coefficients of isoprene, limonene, α-pinene, and linalool at lower temperatures, characteristic times for each species to achieve interfacial equilibrium at lower temperatures can be estimated. As revealed from the calculated characteristic times in the Table 2, the gas−liquid equilibrium can be rapidly established under typical atmospheric conditions and the change in temperature from 298 to 278 K may affect the time scale by up to a factor of 3. At thermodynamic equilibrium, the distribution of a species between the gas phase and the aqueous phase can be estimated by the following equation:

where T is temperature, μ is the reduced mass of the colliding species; σ is the collision diameter, a parameter for LennardJones potentia,l and characteristic of the colliding molecules, in Angstroms; ⟨Ω(1.1)*(θ)⟩ is the collision integral normalized to its rigid sphere value; θ = κT/ε is the reduced temperature, κ is Boltzmanǹs constant, and ε is the energy of molecular interaction. The ε used for calculations of binary diffusion coefficients can be approximated with the parameters of individual species within the Lennard-Jones potential model according to the combination rule56 εAB/κ = [εA(H2O)/κ × εB(air)/κ ]1/2

× × × × ×

(s)

(6)

where εA(H2O)/κ = 356 K, εB(air)/κ = 97 K,51,52 and εAB/κ = 186 K are the individual and binary collision parameters. Equation 5 reveals that the diffusion coefficient of the same system varies with the temperature as a function of T1.5/Ω. By simplifying eq 5, we can predict the diffusion coefficients at atmospheric 10364

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A x = Na /(Na + Ng) = [1 + 1/(KHRTL)]−1

Article

Notes

(8)

The authors declare no competing financial interest.

■

where Na and Ng are its number concentrations in the aqueous and gas phases, respectively (mol mL−1 of air); L is the liquid water content (mL liquid water per mL air). Using eq 8 and the temperature-dependent Henry’s law constants, we obtained the partition of isoprene, limonene, α-pinene, and linalool in stratocumulus clouds (L ≈ 0.44 g/m3) and cumulonimbus clouds (L ≈ 3 g/m3)59−61 at different temperatures (Table 2). Our results have illustrated that although lower temperatures slightly enhance partitioning of all the organics in cloudwater, they are still predominantly present in the gas phase. Nevertheless, the fact that the fraction of linalool in the aqueous phase is nearly 3 orders of magnitude higher than that of the unoxidized precursor may imply that gas phase oxidation of VOCs can considerably increase their solubility in water and result in a significant partition of reacted products in the aqueous phase.

ACKNOWLEDGMENTS This work is supported by Research Corporation for Science Advancement (Grant # 20192). Y.Z. acknowledges the support from National Natural Science Foundation of China (20933001). We are also grateful to the anonymous reviewers for their valuable comments.

■

(1) Carlton, A. G.; Wiedinmyer, C.; Kroll, J. H. A Review of Secondary Organic Aerosol (SOA) Formation from Isoprene. Atmos. Chem. Phys. 2009, 9, 4987−5005. (2) Ervens, B.; Turpin, B. J.; Weber, R. J. Secondary Organic Aerosol Formation in Cloud Droplets and Aqueous Particles (aqSOA): A Review of Laboratory, Field and Model Studies. Atmos. Chem. Phys. 2011, 11, 11069−11102. (3) Hallquist, M.; Wenger, J. C.; Baltensperger, U.; Rudich, Y.; Simpson, D.; Claeys, M.; Dommen, J.; Donahue, N. M.; George, C.; Goldstein, A. H.; et al. The Formation, Properties and Impact of Secondary Organic Aerosol: Current and Emerging Issues. Atmos. Chem. Phys. 2009, 9, 5155−5236. (4) El Haddad, I.; Liu, Y.; Nieto-Gligorovski, L.; Michaud, V.; Temime-Roussel, B.; Quivet, E.; Marchand, N.; Sellegri, K.; Monod, A. In-Cloud Processes of Methacrolein under Simulated Conditions. Part 2: Formation of Secondary Organic Aerosol. Atmos. Chem. Phys. 2009, 9, 5107−5117. (5) Ervens, B.; Volkamer, R. Glyoxal Processing by Aerosol Multiphase Chemistry: Towards A Kinetic Modeling Framework of Secondary Organic Aerosol Formation in Aqueous Particles. Atmos. Chem. Phys. 2010, 10, 8219−8244. (6) Ervens, B.; Carlton, A. G.; Turpin, B. J.; Altieri, K. E.; Kreidenweis, S. M.; Feingold, G. Secondary Organic Aerosol Yields from Cloud-Processing of Isoprene Oxidation Products. Geophys. Res. Lett. 2008, 35, L02816. (7) Fu, T.; Jacob, D. J.; Heald, C. L. Aqueous-Phase Reactive Uptake of Dicarbonyls As a Source of Organic Aerosol Over Eastern North America. Atmos. Environ. 2009, 43, 1814−1822. (8) Lim, H.; Carlton, A. G.; Turpin, B. J. Isoprene Forms Secondary Organic Aerosol through Cloud Processing: Model Simulations. Environ. Sci. Technol. 2005, 39, 4441−4446. (9) Lim, Y. B.; Tan, Y.; Perri, M. J.; Seitzinger, S. P.; Turpin, B. J. Aqueous Chemistry and Its Role in Secondary Organic Aerosol (SOA) Formation. Atmos. Chem. Phys. 2010, 10, 10521−10539. (10) Kroll, J. H.; Ng, N. L.; Murphy, S. M.; Varutbankdul, V.; Flagan, R. C.; Seinfeld, J. H. Chamber Studies of Secondary Organic Aerosol Growth by Reactive Uptake of Simple Carbonyl Compounds. J. Geophys. Res. 2005, 110, D23207. (11) Yu, G.; Bayer, A. R.; Galloway, M. M.; Korsharn, J. J.; Fry, C. G.; Keutsch, R. N. Glyoxal in Aqueous Ammonium Sulfate Solutions: Products, Kinetics and Hydration Effects. Environ. Sci. Technol. 2011, 45, 6336−6342. (12) Zhao, J.; Levitt, N. P.; Zhang, R. Y.; Chen, J. M. Heterogeneous Reactions of Methylglyoxal in Acidic Media: Implications for Secondary Organic Aerosol Formation. Environ. Sci. Technol. 2006, 40, 7682−7687. (13) Zhao, J.; Levitt, N. P.; Zhang, R. Heterogeneous Chemistry of Octanal and 2,4-hexadienal with Sulfuric Acid. Geophys. Res. Lett. 2005, 32, L09802. (14) Levitt, N. P.; Zhao, J.; Zhang, R. Heterogeneous Chemistry of Butanol and Decanol with Sulfuric Acid: Implications for Secondary Organic Aerosol Formation. J. Phys. Chem. A 2006, 110, 13215− 13220. (15) Pun, B. K.; Griffin, R. J.; Seigneur, C.; Seinfeld, J. H. Secondary Organic Aerosol 2. Thermodynamic Model for Gas/Particle Partitioning of Molecular Constituents. J. Geophys. Res. 2002, 107, ACC4−1−ACC4−6.

5. CONCLUSIONS In this work, we measured Henry’s law constants of isoprene, limonene, α-pinene, and linalool and their temperature dependence over a temperature range relevant to the lower atmosphere (278−298 K) for the first time. Our measurements in combination with literature results indicated that noncyclic alkenes could have smaller Henry’s law constants than those of cyclic terpenes and that Henry’s law constants may increase with an increasing number of double bonds. It should be noted that temperature-dependent Henry’s law constants for isoprene and α-pinene obtained from this work were fairly different from the estimated values used in several recent modeling studies, especially for ΔHα‑pinene.19−21 Also the limonene result has illustrated that estimated Henry’s law constants for atmospheric organics could differ from experimental measurement over 1 order of magnitude. Furthermore, in a typical atmospheric model, thousands of species are involved in gas and aqueous phase reaction. These species could have markedly different Henry’s law constants and temperature dependence from estimation like α-pinene and limonene. As such, it is necessary to carry out direct experimental measurements of temperaturedependent Henry’s law constants of many atmospheric organics, which play key roles in air quality, climate change, and human health issues. We were also the first to report firstorder loss rates and diffusion coefficients of all the species as a function of temperature and to evaluate their characteristic times to achieve gas−liquid equilibrium and partitions between gas phase and aqueous phase. Results show that interfacial equilibrium can be rapidly established under typical atmospheric conditions, and all these organics would predominantly exist in the gas phase despite the chemical oxidation of VOCs enhancing their solubility in water and resulting in a significant partition of reacted products in the aqueous phase.

■

ASSOCIATED CONTENT

S Supporting Information *

Plots of ln(C0/Ct) vs time. This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*(Y.L.) E-mail: [email protected]. Tel: 303-556-4772. Fax: 303-556-4776. 10365

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A

Article

(16) Kroll, J.; Ng, N.; Murphy, S.; Flagan, R.; Seinfeld, J. Secondary Organic Aerosol Formation from Isoprene Photooxidation. Environ. Sci. Technol. 2006, 40, 1869−1877. (17) Ng, N.; Kwan, A.; Surratt, J.; Chan, A.; Chhabra, P.; Sorooshian, A.; Pye, H.; Crouse, J.; Wennberg, P.; Flagan, R.; et al. Seocndary Organic Aerosol Formation from Reaction of Isoprene with Nitrate Radicals. Atmos. Chem. Phys. 2008, 8, 4117−4140. (18) Carlton, A. G.; Wiedinmyer, C.; Kroll, J. H. A Review of Secondary Organic Aerosol (SOA) Formation from Isoprene. Atmos. Chem. Phys. 2009, 9, 4987−5005. (19) Mouchel-Vallon, C.; Brauer, P.; Camredon, M.; Valorso, R.; Madronich, S.; Herrmann, H.; Aumont, B. Explicit Modeling of Volatile Organic Compounds Partitioning in The Atmospheric Aqueous Phase. Atmos. Chem. Phys. 2013, 13, 1023−1037. (20) Couvidat, F.; Seigneur, C. Modeling Secondary Organic Aerosol Formation from Isoprene Oxidation under Dry and Humid Conditions. Atmos. Chem. Phys. 2011, 11, 893−909. (21) Karl, M.; Tsigaridis, K.; Vignati, E.; Dentener, F. Formation of Secondary Organic Aerosol from Isoprene over Europe. Atmos. Chem. Phys. 2009, 9, 7003−7030. (22) Tilgner, A.; Herrmann, H. Radical-Driven Carbonyl-to-Acid Conversion and Acid Degradation in Tropospheric Aqueous Systems Studied by Capram. Atmos. Environ. 2010, 44, 5415−5422. (23) Guenther, A.; Karl, T.; Harley, P.; Wiedinmyer, C.; Palmer, P.; Geron, C. Estimates of Global Terrestrial Isoprene Emissions Using MEGAN (Model of Emissions of Gases and Aerosols from Nature). Atmos. Chem. Phys. 2006, 5, 3181−3210. (24) Epstein, S. A.; Nizkorodov, S. A. A Comparison of The Chemical Sinks of Atmospheric Organics in the Gas and Aqueous Phase. Atmos. Chem. Phys. 2012, 12, 8205−8222. (25) Okochi, H.; Sato, E.; Matsubayashi, Y.; Igawa, M. Effect of Atmospheric Humic-Like Substances on the Enhanced Dissolution of Volatile Organic Compounds into Dew Water. Atmos. Environ. 2008, 87, 213−223. (26) Crahan, K. K.; Hegg, D.; Covert, D. S.; Jonsson, H. An Exploration of Aqueous Oxalic Acid Production in the Coastal Marine Atmosphere. Atmos. Environ. 2004, 38, 3757−3764. (27) Herrmann, H.; Wolke, R.; Muller, K.; Bruggemann, E.; Gnauk, T.; Barzaghi, P.; Mertes, S.; Lehmann, K.; Massling, A.; Birmili, W.; et al. Febuko and Modmep: Field Measurements and Modelling of Aerosol and Cloud Multiphase Processes. Atmos. Environ. 2005, 39, 4169−4183. (28) Sorooshian, A.; Ng, N. L.; Chan, A. W. H.; Feingold, G.; Flagan, R. C.; Seinfeld, J. H. Particulate Organic Acids and Overall WaterSoluble Aerosol Composition Measurements from the 2006 Gulf of Mexico Atmospheric Composition and Climate Study (GOMACCS). J. Geophys. Res. Atmos. 2007, 112, D13201. (29) Staudinger, J.; Roberts, P. V. A Critical Review of Henry’s Law Constants for Environmental Applications. Environ. Sci. Technol. 1996, 26, 205−297. (30) Betterton, E. A. Henry’s Law Constants of Soluble and Moderately Soluble Organic Gases: Effects of Aqueous Phase Chemistry. Adv. Environ. Sci. Technol. 1992, 24, 41−50. (31) Lee, S. H.; Mukherjee, S.; Brewer, B.; Ryan, R.; Yu, H.; Gangoda, M. A Laboratory Experiment to Measure Henry’s Law Constants of Volatile Organic Compounds with a Bubble Column and a Gas Chromatography Flame Lionization Detector (GC-FID). J. Chem. Educ. 2012, 90, 495−499. (32) Roberts, J. M. Measurement of The Henry’s Law Coefficient and First Order Loss Rate of PAN in n-Octanol. Geophys. Res. Lett. 2005, 32, L08803. (33) Ip, H. S. S.; Huang, X. H. H.; Yu, J. Z. Effective Henry’s Law Constants of Glyoxal, Glyoxylic Acid, and Glycolic Acid. Geophys. Res. Lett. 2009, 36, L01802. (34) Mackay, D.; Shiu, W. Y.; Sutherland, R. P. Determination of Air−Water Henry’s Law Constants for Hydrophobic Pollutants. Environ. Sci. Technol. 1979, 13, 333−337.

(35) Betterton, E. A.; Hoffmann, M. R. Henry’s Law Constants of Some Environmentally Important Aldehydes. Environ. Sci. Technol. 1988, 22, 1415−1418. (36) Kish, J. D.; Leng, C.; Kelley, J.; Hiltner, J.; Zhang, Y. H.; Liu, Y. An Improved Approach for Measuring Henry’s Law Coefficients of Atmospheric Organics. Atmos. Environ. 2013, 79, 561−565. (37) Hoff, J. T.; Mackay, D.; Gillham, R.; Shiu, W. Y. Partitioning of Organic Chemicals at the Air−Water Interface in Environmental Systems. Environ. Sci. Technol. 1993, 27, 2174−2180. (38) Linstrom, P. J. NIST Chemistry Web Book, NIST Standard Reference Database Number 69; National Institute for Standard Technology (NIST): Gaithersburg, MD, 2005. (39) Hartkopf, A.; Karger, B. L. Study of The Interfacial Properties of Water by Gas Chromatography. Acc. Chem. Res. 1973, 6, 209−216. (40) Fioroni, M.; Vogt, D. Toluene Model for Molecular Dynamics Simulations in the Ranges 298 < T (K) < 350 and 0.1 < P (MPa) < 10. J. Phys. Chem. B 2004, 108, 11774−11781. (41) Sander, R. Compilation of Henry’s Law Constants for Inorganic and Organic Species of Potential Importance in Environmental Chemistry, 1999. http://www3.mpch-mainz.mpg.de/∼sander/res/ henry.html. (42) TOXNET Toxicology Data Network. http://toxnet.nlm.nih. gov/. (43) Yaws, C. L.; Yang, H. C. Henry’s Law Constant for Compound in Water. In Thermodynamic and Physical Property Data; Yaws, C. L., Ed.; Gulf Publishing Company: Houston, TX, 1992; pp 181−206. (44) Wilhelm, E.; Battino, R.; Wilcock, R. J. Low-Pressure Solubility of Gases in Liquid Water. Chem. Rev. 1977, 77, 219−262. (45) Altschuh, J.; Bruggemann, R.; Santl, H.; Elchmger, G.; Piringer, O. G. Henry’s Law Constants for a Diverse Set of Organic Chemicals: Experimental Determination and Comparison of Estimation Methods. Chemosphere 1999, 39, 1871−1887. (46) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics: from Air Pollution to Climate Change; Wiley: New York, 2006. (47) Schwartz, S. E. Mass-Transport Considerations Pertinent to Aqueous Phase Reactions of Gases in Liquid-Water Clouds. In Chemistry of Multiphase Atmospheric Systems; Jaeschke, W., Ed.; NATO ASI Series; Springer: Berlin, Germany, 1986; Vol. 6, pp 415−471. (48) Ivanov, A. V.; Trakhtenberg, S.; Bertram, A. K.; Gershenzon, Y. M.; Molina, M. J. OH, HO2, and Ozone Gaseous Diffusion Coefficients. J. Phys. Chem. A 2007, 111, 1632−1637. (49) Davidovits, P.; Kolb, C. E.; Williams, L. R.; Jayne, J. T.; Worsnop, D. R. Update 1 of: Mass Accommodation and Chemical Reactions at Gas-Liquid Interfaces. Chem. Rev. 2011, 111, PR76− PR109. (50) Kim, B. G.; Schwartz, S. E.; Miller, M. A.; Min, Q. Effective Radius of Cloud Droplets by Ground-Based Remote Sensing: Relationship to Serosol. J. Geophys. Res. 2003, 108, 4740. (51) Chapman, S.; Cowling, T. G. Mathematical Theory of NonUniform Gases; Cambridge University Press: London, U.K., 1959. (52) Hirschfelder, J. O.; Curtis, C. F.; Bird, R. B. Molecular Theory of Gases and Liquids; John Wiley and Sons: New York, 1954. (53) Reid, R. C.; Sherwood, T. K. The Properties of Gases and Liquids; McGraw-Hill: New York, 1958. (54) Welty, J. R.; Wicks, C. E.; Wilson, R. E. Fundamentals of Momentum, Heat, and Mass Transfer, 3rd ed.; Wiley: New York, 1984. (55) Liu, Y.; Ivanov, A. V.; Molina, M. J. Temperature Dependence of OH Diffusion in Air and He. Geophys. Res. Lett. 2009, 36, L03816. (56) Marrero, T. R.; Mason, E. A. Gaseous Diffusion Coefficients. J. Phys. Chem. Ref. Data. 1972, 1, 3−118. (57) Bolz, R. E.; Tuve, G. L. Handbook of Tables for Applied Engineering Science, 2nd ed.; CRC Press: Boca Raton, FL, 1976. (58) Nellis, G.; Klein, S. Heat Transfer; Cambridge University Press: Cambridge, U.K., 2008. (59) Hess, M.; Koepke, P.; Schult, I. Optical Properties of Aerosols and Clouds: The Software Package OPAC. Bull. Am. Meteorol. Soc. 1998, 79, 831−844. 10366

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367

The Journal of Physical Chemistry A

Article

(60) Rosenfeld, D.; Lensky, I. M. Satellite-Based Insights Into Precipitation Formation Processes in Continental and Maritime Convective Clouds. Bull. Amer. Meteorol. Soc. 1998, 79, 2457−2476. (61) Thompson, A. Simulating the Adiabatic Ascent of Atmospheric Air Parcels Using the Cloud Chamber; Department of Meteorology, Penn State, 2007.

10367

dx.doi.org/10.1021/jp403603z | J. Phys. Chem. A 2013, 117, 10359−10367