Hygroscopic Properties of Two Model Humic-like Substances and

Oct 22, 2003 - are studied using two natural FA: the Nordic Aquatic. Fulvic Acid (NAFA) and the Suwannee River Fulvic Acid. (SRFA) as model compounds ...
0 downloads 0 Views 154KB Size
Download PDF
Environ. Sci. Technol. 2003, 37, 5109-5115

Hygroscopic Properties of Two Model Humic-like Substances and Their Mixtures with Inorganics of Atmospheric Importance MAN NIN CHAN AND CHAK K. CHAN* Department of Chemical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong

Water-soluble macromolecular polyacids can play a potentially important role in the hygroscopic properties of atmospheric aerosols. These acids have molecular structures similar to natural fulvic acids (FA) (or humic acids) and are referred to as humic-like substances (HULIS). In this study, the hygroscopicity of HULIS and the mixture of HULIS and sodium chloride (NaCl) and that of HULIS and ammonium sulfate (AS) aerosols at a mass ratio of 1:1 are studied using two natural FA: the Nordic Aquatic Fulvic Acid (NAFA) and the Suwannee River Fulvic Acid (SRFA) as model compounds in an electrodynamic balance. NAFA and SRFA both absorbed and desorbed water reversibly without crystallization and retained water at a relative humidity (RH) < 10%. NAFA and SRFA have a mass growth ratio of 1.25 and 1.45 from RH ) 10% to RH ) 90%, respectively. However, these results are different from those of another natural FA (the Nordic River Fulvic Acid Reference) in the literature. The differences are possibly due to the differences in the chemical composition of the natural FA, which depends on their sources and the isolation methods. These results suggest that a standardization of the isolation methods of HULIS is needed for better understanding of their atmospheric properties and environmental impacts. In general, the deliquescence and crystallization RH of FA-inorganic mixtures are comparable with those of their respective pure inorganic species. Since FA are less hygroscopic than NaCl and AS, all mixtures absorb less water compared to their respective pure inorganic species of equal particle mass. The FA-AS mixtures have a larger water uptake than the sum of those of the FA and AS individually following a simple additivity rule as noninteracting species at RH ) 90%. This enhancement effect increases as the RH decreases. There is no such enhancement effect for the FA-NaCl mixtures until RH is below 90%. These results reveal that the effect of the interactions between FA and inorganic species on the water uptake of the mixtures, in general, is a function of RH.

Introduction The hygroscopicity of atmospheric aerosols depends on their chemical composition and determines the size and water content of the aerosols. It has influence on aerosols’ environmental impacts, including those on global climate, * Corresponding author phone: (852)2358-7124; fax: (852)23580054; e-mail: [email protected]. 10.1021/es034272o CCC: $25.00 Published on Web 10/22/2003

 2003 American Chemical Society

and has been extensively studied in laboratory and field measurements using an electrodynamic balance (EDB) or a Tandem Differential Mobility Analyzer (TDMA). It is generally acknowledged that the inorganic constituents alone cannot fully explain the observed hygroscopicity of atmospheric aerosols in laboratory and field measurements. Organic compounds constitute a significant mass fraction of fine atmospheric aerosols. The role of organic compounds in the hygroscopicity of atmospheric aerosols has been the subject of a number of investigations (1-4). Water-soluble organic compounds (WSOC) are predominantly found in fine atmospheric aerosols and can contribute to a substantial portion of organic compounds (5-7). Dicarboxylic acids, in particular, oxalic acid, malonic acid, and succinic acid, are the first class of WSOC identified in the atmospheric aerosols. In general these acids constitute less than 10% of WSOC (5, 6, 8). More recently, a new class of WSOCsmacromolecular polyacidsshas been detected in aerosols (9-15), snow (16), and fog samples (17, 18) through the use of spectrometric methods and size exclusion chromatography. Because of their molecular similarity to natural fulvic acids (FA) or humic acids (HA), they are referred to as humic-like substances (HULIS) in the literature. HULIS have been found to account for a substantial amount (up to 50%) of WSOC in fine aerosols (7, 19) and thus have the potential to affect aerosol hygroscopicity. Atmospheric HULIS have molecular weights in the order of several hundred Dalton, which is close to that of natural FA rather than natural HA (7). Natural FA have been suggested as representative substances for atmospheric HULIS (20). Recently, the effect of organic species on the deliquescence behavior and hygroscopic growth of their mixtures with inorganic species (e.g., sodium chloride (NaCl) and ammonium sulfate (AS)) has been studied (21-24). Organic species have been found to lower or to have no effect on the deliquescence RH (DRH) of the inorganics. In general, organic species are less hygroscopic than typical inorganic species (25-28). On a per unit particle mass basis, mixtures of organic and inorganic species have a smaller hygroscopic growth than the respective pure inorganic species (21-24). In addition to their individual contributions to hygroscopic growth following additivity rules, organic species interact with inorganic species, which in turn affects the hygroscopic growth of the mixtures. Assuming that only the inorganic fraction of the mixtures absorbs water and the organic fraction acts as inert materials, Cruz and Pandis (22) and later Choi and Chan (24) have quantified the effect of some organic species on the water uptake of their mixtures with NaCl and AS at RH ) 85%. However, these analyses did not take into consideration the water absorbed by the organic species, and both studies focused on quantifying the effect at a single high RH (RH ) 85%) only. To better understand the effect of inorganic and organic species on the hygroscopic growth of atmospheric aerosols, it is thus necessary to study the effect over a range of RH during growth and evaporation, which is more relevant to the atmospheric environment. Despite these recent measurements of the hygroscopicity of organic species and their mixtures with inorganic species, the role of WSOC, in particular HULIS, in the hygroscopicity of atmospheric aerosols is far from certain. The objective of this study is to (1) study the hygroscopicity of HULIS using two natural FA, Suwannee River Fulvic Acid (SRFA), and Nordic Aquatic Fulvic Acid (NAFA), as model compounds and (2) investigate how natural FA affect the hygroscopicity of their mixtures with NaCl and AS as a function of RH at a mass ratio of 1:1. The results presented here provide a first VOL. 37, NO. 22, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5109

estimation on how atmospheric HULIS absorb water and influence the hygroscopicity of inorganic aerosols at different RH values.

Experimental Methods The hygroscopic measurements were carried out using an EDB, which has been described by Peng et al. (26). The RH inside the EDB was changed in discrete steps and was controlled by the RH of an inlet gas stream, which was determined by measuring the dew point and ambient temperature. The precision of the dew point and ambient temperature measurement were 0.2 °C and 0.1 °C, respectively. The experiments were conducted at a temperature ranging from 22.5 °C to 23.8 °C and RH ranging from 10% to 90%. The NAFA and SRFA were obtained from the International Humic Substances Society (29) and used without further purification. The isolation procedure of the two natural FA has been discussed in detail (30, 31). The dry samples of the natural FA had low contents (∼0.45-0.98 wt %) of inorganic residues. These FA were completely dissolved in ultrapure water to make stock solutions of about 1 wt % FA and were used to generate particles of 10-20 µm in diameters by a piezoelectric droplet generator (Uni-Proton Inc., NY, U.S.A., Model 201). In the hygroscopic measurements using the EDB, a reference state is needed to calculate the composition of the particle as a function of RH. Bulk solutions of a known composition and water activity (aw ) RH/100) are a possible choice. However, the two FA are extremely soluble that their water activities were over 0.99 (equal to RH ) 99% at equilibrium) at 20 wt % solution. Hence, the bulk data cannot be used to identify a reference state for the single particle measurements because the single particle measurements were conducted at RH below 90%. Another choice of reference state is the dry particle, which resulted from the evaporation of a droplet at low RH (RH ) 0%). However, some species have been found to retain water at low RH, and it cannot be assumed a priori that the dried particle is water-free (26). In this paper, instead of using the absolute composition, such as the mass fraction of solute (mfs) used in our previous EDB measurements, the results are presented in the form of the mass ratio as a function of aw. The mass ratio is defined as the ratio of particle mass, m, at a given RH to particle mass, mo, at reference RH (RH ) 10%) at equilibrium:

m mass ratio ) mo

(1)

The uncertainty in the mass ratio measurement for the droplets is about (0.03 in m/mo, and for partially crystallized and solid particles it is about (0.06 in m/mo. When crystallization or deliquescence occurs, the particle loses and absorbs water abruptly, resulting in a relatively sharp change in particle mass between the two RH. However, since the RH was changed in discrete steps (i.e., multiple distinct steps, not continuously), only the ranges of crystallization RH (CRH) and DRH are reported.

Results and Discussion (1) The Hygroscopicity of Nordic Aquatic Fulvic Acid (NAFA) and Suwannee River Fulvic Acid (SRFA). In Figures 1 and 2, the mass ratio is plotted as a function of aw for NAFA and SRFA, respectively. The NAFA and SRFA showed a similar trend in hygroscopicity in that there was no abrupt change in the particle mass upon evaporation and growth, suggesting the absence of a phase transition. The NAFA and SRFA absorbed and desorbed water reversibly and behaved like “nondeliquescent” species. Like NAFA and SRFA, many organic species including some water-soluble organic salts (25), dicarboxylic acids (26), multifunctional acids (26), and 5110

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 22, 2003

FIGURE 1. The hygroscopicity of Nordic Aquatic Fulvic Acid.

FIGURE 2. The hygroscopicity of Suwannee River Fulvic Acid.

TABLE 1: Crystallization RH, Deliquescence RH, Mass Ratio, ξw and ξw′ of the NRFA and SRFA, and Their Mixtures with NaCl and AS at a Mass Ratio of 1:1 system

CRH

DRH

SRFA SRFA-AS SRFA-NaCl NAFA NAFA-AS NAFA-NaCl

no 39.8-43.6 47.1-50.2 no 38.1-41.5 49.5-52.6

no 79.8-82.5 74.2-76.8 no 78.6-81.4 74.8-77.4

m/moa m/mo,ZSRb 1.25 2.62 4.19 1.45 2.91 4.02

ξw c

ξw′d

2.21 4.31

1.49 1.34 1.00 0.96

2.31 4.41

1.76 1.46 0.95 0.89

a All m/m , m/m b Mass o o,ZSR, ξw, and ξw′ are determined at RH ) 90%. ratio is predicted by ZSR model. c ξw is calculated using eq 2. d ξw′ is calculated using eq 3.

pharmaceutical aerosols (27) have been found to absorb and desorb water continuously. An important issue is whether NAFA and SRFA retain water at low RH (RH < 10%) or not. At RH < 10%, nondeliquescent particles either form anhydrous particles or retain some residual water. For example, citric acid, tartaric acid, and malic acid retain water at low RH (mfs * 1) due to the strong interactions between the polar functional groups and water molecules at high degrees of supersaturation (26). The natural FA possess a range of functional groups such as carboxylic and hydroxyl groups (32). It is thus possible that the NAFA and SRFA particles may retain water at low RH. The dependence of the mass ratio of the NAFA and SRFA on aw can be empirically fitted by m/mo ) (1-aw)-0.152 and m/mo ) (1-aw)-0.091, respectively. The two FA show a very similar trend in their water absorption at low RH (e.g., RH ) 20%). However, NAFA becomes more hygroscopic than SRFA at a higher RH. As shown in Table 1, NAFA increases its mass by 45%, but the SRFA increases its mass by 25% only when the RH is increased from 10% to 90%. Gysel et al. (33)

have studied the hygroscopicity of another natural FA, the Nordic River Fulvic Acid Reference (NRFA), using a TDMA. They reported that the NRFA particles deliquesced at RH ) 75-80% during growth and had a diameter growth ratio (Gf ) Dp(RH)/Dp(RHo), the ratio of the particle diameter at a high RH to the dry particle diameter at reference RHo) of 1.15 at RH ) 90%. The mass growth ratios of SRFA and NAFA are 1.25 and 1.45, at RH ) 90% respectively. To compare these mass growth ratios to the diameter growth ratios reported by Gysel et al. (33), the mass growth ratio is converted to the diameter growth ratio, Gf ) {(FFA/Fw) * [(m/mo)-1]+1}1/3, where FFA is the density of the FA, and Fw is the density of water and the volume of mixing is assumed to be zero. Jone et al. (34) have reported that the partial specific volume of humic material ranges from 0.45 to 0.71 cm3 g-1. Taking the averaged density of 0.58 cm3 g-1 for the two FA, the Gf of SRFA and NAFA is 1.13 and 1.21, respectively, at RH ) 90%, which are close to those reported by Gysel et al. (33). The difference can be partly explained by the uncertainty of the density effect and shape effect in their TDMA measurement. However, it should be noted that in addition to the hygroscopic growth, the deliquescence and efflorescence behaviors of the three FA are different. In general, the natural FA absorb less water than the typical atmospheric inorganic species such as NaCl (Gf ) 2.36) and AS (Gf ) 1.72) at RH ) 90%. Composition of Natural FA. Although there are only small differences between our results and those of Gysel et al. (33) in diameter growth ratios, there is a significant difference in the efflorescence behavior. Gysel et al. (33) found that the NRFA particles effloresced in the diffusion dryer (residence time ) 180s) at RH < 5%, but we did not observe the efflorescence of NAFA and SRFA at RH ∼ 3% for 12 h, which was much longer than the 60 min typically required for the particle to achieve equilibrium in this study. Since sufficient time was allowed for water evaporation from the particles in the EDB and in the diffusion dryer, mass transfer delay is not a likely explanation of the difference in the measurements. Natural FA is a mixture of compounds, whose compositions depend on the sources and the isolation methods. The composition of FA is believed to be the key factor to account for these differences. In addition to the natural FA, trace amounts of metal and inorganic ions were also isolated. The presence of impurities can promote crystallization of aerosols that are difficult to effloresce such as NH4NO3 (35). In the natural FA, metal and inorganic ions balance the charges of the FA molecules and reduce the intramolecular and intermolecular charge repulsion, which in turn enhance the molecular contraction and association, respectively. The multivalent cations (e.g., calcium) promote precipitation at a lower FA and ionic concentration than monovalent ions (e.g., sodium) except proton (36). The presence of impurities might be the reason for the observed efflorescence of NRFA by Gysel et al. (33). In our measurements, the effect of the metal and inorganic ions on our hygroscopic measurement is not significant, even at the lowest RH. Overall, the differences between Gysel et al. (33) and our measurements suggest that it is important to have a standardized isolation method of atmospheric HULIS for understanding their properties and environmental impacts. Mass Transfer Limitations. The EDB has a distinct advantage over TDMA for hygroscopic measurements in laboratories: there is virtually no practical limitation on the residence time in the EDB to allow the particles to achieve equilibrium with the gas phase. TDMA measurements can be compromised by mass transfer effects (22, 37). It is important to assess if equilibrium has been achieved in the TDMA measurements, especially for species of large molecular weights such as FA and HULIS. Gysel et al. (33)

FIGURE 3. The hygroscopicity of mixtures of Nordic Aquatic Fulvic Acid with AS at a mass ratio of 1:1.

FIGURE 4. The hygroscopicity of mixtures of Suwannee River Fulvic Acid with AS at a mass ratio of 1:1. reported that the NRFA particles did not effloresce and retained a small amount of water at RH ∼ 5%. However, they observed efflorescence of the NRFA particles in the diffusion dryer (residence time ∼ 180s; RH < 5%) before these particles were grown to droplets to effect the evaporation measurements. During the evaporation measurements, the particles had a residence time of the order of seconds (∼8 s), which is significantly shorter than that in the diffusion dryer. The NRFA particles may not have sufficient time to effloresce, even at the lowest RH (RH ∼ 4%) during evaporation. Since Gysel et al. (33) only reported the evaporation measurements at a single residence time, the effect of the residence time on their evaporation and growth measurements was not known. It is possible that there is a nucleation rate limitation in their measurements. In this study, we found that 60 min were sufficient for the particles to obtain equilibrium measurements in the EDB measurement although mass transfer effects have been found in some other species such as MgSO4 and glutaric acid, which take significantly longer to achieve their equilibrium state in hygroscopic measurements (26, 38, 39). (2) The Hygroscopicity of Mixtures of FA and AS. Figures 3 and 4 show the hygroscopicity of the mixtures of NAFA and SRFA with AS at a mass ratio of 1:1, respectively. The NAFAAS and SRFA-AS mixtures had similar deliquescence and efflorescence behaviors. The NAFA-AS mixture crystallized at RH ) 38.1-41.5%, and the SRFA-AS mixture crystallized at RH ) 39.8-43.6%, which is close to that of pure AS (RH ) 37-40%) (40). The NAFA-AS mixture deliquesced at RH ) 78.6-81.4%, and the SRFA-AS mixture deliquesced at RH ) 79.8-82.5%, which is also close to that of pure AS (RH ) 80 ( 0.4%) (41). Here the DRH of mixture refers to the point where the evaporation and growth curves overlap, i.e., the disappearance of solid state, and the deliquescence is VOL. 37, NO. 22, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5111

FIGURE 5. The hygroscopicity of mixtures of Nordic Aquatic Fulvic Acid with NaCl at a mass ratio of 1:1.

FIGURE 6. The hygroscopicity of mixtures of Suwannee River Fulvic Acid with NaCl at a mass ratio of 1:1. complete. Overall, the NAFA and SRFA have little effect on the CRH and DRH of AS. The evaporation and growth curves of the FA-AS mixtures are slightly different from those of pure AS. Two observations can be made from the measurements: (1) After crystallization, the NAFA-AS and SRFA-AS mixtures continued to decrease in the particle mass because of the loss of water. This suggests that the mixtures contain water after crystallization, presumably due to the presence of FA. (2) During growth, the NAFA-AS and SRFA-AS mixtures absorbed a small amount of water before the deliquescence of the mixtures. At RH below the CRH of the mixtures, the water contents during evaporation and growth are similar. (3) The Hygroscopicity of Mixtures of FA and NaCl. Figures 5 and 6 show the hygroscopicity of the mixtures of NAFA-NaCl and SRFA-NaCl at a mass ratio of 1:1, respectively. The NAFA-NaCl and SRFA-NaCl mixtures have similar efflorescence and deliquescence behaviors. The NAFA-NaCl mixture crystallized at RH ) 49.5-52.6% and the SRFA-NaCl mixture crystallized at RH ) 47.1-50.2%. These CRH are comparable with the CRH of pure NaCl (CRH ) 47%) (42). The NAFA-NaCl mixture and SRFA-NaCl mixture deliquesced at RH ) 74.8-77.4% and at RH ) 74.276.8%, respectively, which are also similar to the DRH of NaCl (70-75%) (43). Similar to the FA mixtures with AS, the FA-NaCl mixtures retain water, and the mass of the mixtures decreases with decreasing RH due to the evaporation of water from the particles after crystallization. In the growth measurements, the mixtures start to absorb water at low RH. The mass of the mixtures increases with increasing RH before the deliquescence of the mixtures and the NaCl. In summary, the deliquescence and efflorescence behaviors of the 1:1 FA-inorganic mixtures are dominated by 5112

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 22, 2003

the inorganic species. The DRH and CRH of the FA-inorganic mixtures are comparable with those of their respective pure inorganic species. During evaporation, all mixtures crystallize at an RH close to the CRH of their respective inorganic species and may retain water at low RH after crystallization. During growth, all mixtures absorb water before the deliquescence of the mixtures and deliquesce at an RH close to the DRH of their respective inorganic species. All mixtures absorb less amounts of water, compared to their respective pure inorganic species of equal particle mass (mass growth ratio of pure NaCl and AS is 7.37 and 3.17, respectively) at RH ) 90% (see Table 1). This finding is consistent with previous findings that organic species typically reduce the total hygroscopic growth of their mixtures with inorganic species (22-24). It should be emphasized that the composition of the FAinorganic mixtures cannot be directly inferred from their mass ratios at low RH, and the value of m/mo equal to unity does not necessarily mean that the particles are water-free. The m/mo values would be slightly larger if mo were defined as the water-free particle mass. Like the NAFA and SRFA particles, the FA-inorganic mixtures may also retain water at low RH because of the existence of nondeliquescent FA. However, since the FA-inorganic mixtures only absorb a small amount of water before deliquescence, they retain only a small amount of water at low RH, when compared to the water contents after they are completely deliquesced. Hence the effect of the residual water in the “dried” particles on m/mo is not expected to be significant. Monovalent ions have been found to enhance the aggregate formation of FA molecules in the aqueous solution (36). Irregular light scattering patterns were observed for all particles of the FA-inorganic mixtures at RH higher than the respective CRH of the pure inorganic species in the mixtures through the visual inspection with a microscope. This observation is not likely to be due to the phase transition because there is no sharp change in the particle mass till the RH reaches the CRH of the pure inorganic species. It is possible that “large” aggregates formed in the aqueous droplets of the FA-inorganic mixtures before the crystallization of the inorganic species. (4) Mass Growth Ratios of Organic-Inorganic Mixtures. Because of the interactions between the inorganic and organic species, simple mixing rules may not provide an accurate estimate of the water absorption of their mixtures. Cruz and Pandis (22) quantified the effect of organic species on the water uptake of inorganic species at RH ) 85%. The ratio of the water uptake in mass of a mixed particle to the water uptake of the inorganic fraction (i.e., the organic fraction is treated as inert), ξw, is calculated as

ξw )

(m/mo) - 1 finorg[(m/mo)inorg - 1]

(2)

where (m/mo) and (m/mo)inorg are the measured mass ratio of the mixture and the pure inorganic species at a given RH, and finorg is the mass fraction of the inorganic species in the mixture, respectively. Equation 2 is the same equation that appears in ref 22 by Cruz and Pandis, who obtained the mass ratios from water uptake measurements on a volume basis using TDMA. In eq 2, organic species are assumed to be inert, and the water contributed by organic species to the hygroscopic growth of the mixtures has not been incorporated. The two FA studied here and some organic species are hygroscopic and absorb water in growth. It is thus necessary to take the water uptake of organic species into consideration when quantifying the effect of chemical interaction between inorganic and organic species on the water uptake of the mixtures.

TABLE 2: Growth Factor, ξw and ξw′ of Some Water-Soluble Organic Compounds, and Their Mixtures with NaCl and AS species

Gf (RH ) 85%)

malonic acida malonic acid-NaClb malonic acid-ASb succinic acida succinic acid-NaClb succinic acid-ASb glutaric acida glutaric acid-NaClb glutaric acid-ASb citric acida citric acid-NaClb citric acid-ASb glycerold glycerol-NaClb glycerol-ASb pinonic acidc pinonic acid-NaClc pinonic acid-ASc

1.40 1.59 1.45 1.00 1.57 1.43 1.22 1.49 1.38 1.28 1.42 1.41 1.36 1.61 1.45 1.00 1.71 1.41

ξw (RH ) 85%)e

ξw′ (RH ) 85%)f

0.75 1.99

0.65 1.00

0.61 1.66

0.61 1.66

0.51 1.36

0.49 1.01

0.33 1.01

0.32 0.87

0.75 2.07

0.67 1.08

1.00 1.60

1.00 1.60

FIGURE 7. The ξw′ value as a function of aw for NAFA-AS and SRFA-AS mixtures.

a Growth diameter ratio from RH < 5% to RH ) 85% by Peng et al. (26). b Mixtures were at a molar ratio of 1:1 except glutaric-NaCl which was at a mass ratio of 1:1 and a growth diameter ratio from RH ) 10% to RH ) 85% by Choi and Chan (24). c Mixtures were at a mass ratio of 1:1 and the growth diameter ratio from RH < 10% to RH ) 85% by Cruz and Pandis (22). d Growth diameter ratio from RH ) 10% to RH ) 85% by Choi and Chan (48). e ξw from Choi and Chan (24) and Cruz and Pandis (22) using eq 2. f ξw′ is calculated using eq 3.

To take this into account, we modify eq 2 assuming that both inorganic and organic species absorb water independently by introducing ξw′, which is defined as

ξw ′ )

(m/mo) - 1 finorg[(m/mo)inorg - 1] + forg[(m/mo)org - 1]

(3)

where forg ) 1 - finorg is the mass fraction of organic species in the mixture and (m/mo)org is the measured mass ratio of pure organic species at a given RH. Equation 3 yields eq 2 when (m/mo)org ) 1 (i.e., the organic species does not absorb water). The denominator in eq 3 is the Zdanovskii, Stokes, and Robinson (ZSR) estimation of water uptake of mixtures. The ZSR equation, which assumes that organic and inorganic species absorb water independently, has been used for estimating the growth ratio of organic-inorganic mixtures (44, 45). As seen in Table 1, the ZSR model overpredicts the mass ratios of the FA-NaCl mixtures by less than 10% but underpredicts those of the FA-AS mixtures by about 20%. In eq 3, enhancement (ξw′ > 1) and reduction effects (ξw′ < 1) indicate that the water uptake of the mixture is larger and smaller than the sum of water uptake of the inorganic and organic fractions following the ZSR equation, respectively. One can interpret these conditions as that the interactions between the inorganic and the organic fractions enhance (ξw′ > 1) or reduce (ξw′ < 1) water uptake of the mixtures compared to the water uptake of the mixtures predicted by the additivity rule. Neutral effect (ξw′ ∼ 1) indicates that the water uptake of the mixtures is close to the water uptake of mixtures predicted by the ZSR equation. The values of ξw′ of all FA-inorganic mixtures calculated using eq 3 are shown in Table 1, with the uncertainty estimated to be (0.2. The chemical interactions between FA and inorganic species have no effect in the FA-NaCl mixtures but have an enhancement effect in the FA-AS mixtures at RH ) 90%. An overestimation of the effect, especially for the NAFA-AS mixtures, results if the water uptake of the FA is not considered (i.e., eq 2 is used), as shown in Table 1.

FIGURE 8. The ξw′ value as a function of aw for NAFA-NaCl and SRFA-NaCl mixtures. In Table 2, ξw and ξw′ for the mixture data of Cruz and Pandis (22) and Choi and Chan (24) are compared. Comparing the values of ξw and ξw′ of the mixtures, reduction effect (ξw′ < 1) remains in the organic-NaCl mixtures. However, a neutral effect (ξw′∼1) is observed for the mixtures of AS with glycerol, malonic acid, and glutaric acid using eq 3, but an enhancement effect (ξw > 1) results from eq 2. These watersoluble organic species have a hygroscopic growth ratio comparable to that of pure AS. When their growth is not properly considered, the effect of the interactions of organic and inorganic species is overestimated. The above analysis focuses on data measured at a single RH (RH ) 85%). It is possible that such interaction effect may change at different RH, because the interactions between organic and inorganic species differ as a function of solution concentration. Here we investigate the effect for the FAinorganic mixtures for the whole range of RH (down to the RH that the FA-inorganic mixtures effloresced) using eq 3. In the calculations, both evaporation and growth data are used after the mixtures have completely deliquesced, but only the evaporation data are used in the hysteresis regime. It is clear from Figures 7 and 8 that ξw′ increases as RH decreases for both the SRFA-NaCl and NAFA-AS mixtures, but ξw′ has reached a maximum at an RH of ∼60-65% for the NAFA-NaCl and SRFA-AS mixtures. For both FA-NaCl and FA-AS mixtures, the enhancement effect increases as RH decreases. The RH dependence of ξw′ for the organic-inorganic mixture data in the literature is shown in Figures 9 and 10. Neutral, reduction, and enhancement effects have been observed for the organic-AS mixtures at high RH (RH ) 80%) in Figure 9. The effect observed for some organic-AS mixtures (e.g., malonic acid, succinic acid, and pinonic acid) changes as a function of RH. For example, a neutral effect VOL. 37, NO. 22, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5113

with multivalent cations (e.g., calcium) that are different from those with monovalent cations such as sodium (36, 46, 47). It is recommended that much more research efforts in understanding atmospheric HULIS and their interactions with typical inorganic species are needed to elucidate the impacts of HULIS on atmospheric aerosols.

Acknowledgments This work was funded by the Earmarked Grant from the Research Grants Council of the Hong Kong Special Administrative Region, China (HKUST6056/02P).

Literature Cited FIGURE 9. The ξw′ value as a function of aw for organic-AS mixtures.

FIGURE 10. The ξw′ value as a function of aw for organic-NaCl mixtures. is observed at high RH, but a reduction effect is observed at low RH for the malonic acid-AS mixtures. Figure 10 shows that the value of ξw′ of all organic-NaCl mixtures is less than 1, i.e., the reduction effect is observed for all organic-NaCl mixtures in the whole RH range studied. Although the changes in ξw′ for the organic-NaCl and some organic-AS mixtures are within the uncertainty, there appears to be a trend indicating that the reduction effect increases as the RH decreases, contrary to the effects observed for the FAinorganic mixtures. Based on the results of NAFA and SRFA, it appears that atmospheric aerosols containing FA and HULIS may not be water-free at a low RH. This is consistent with suggestions from the previous TDMA field measurements that water associated with organic species becomes more significant at a low RH. However, our measurements, in particular, the efflorescence properties, are different from the literature data of NRFA (33). The composition of the natural FA, which depends on the sources and the isolation methods, is the key factor to account for these differences. Different approaches have been used to isolate and quantify the HULIS in atmospheric field measurements (7, 9, 17). A better understanding of the properties of atmospheric HULIS requires a standardization of the isolation methods. Ideally, the isolated products should not contain inorganic species and other macromolecular compounds such as proteins and polysaccharides, which have been detected in atmospheric aerosols. The distinct trend of the dependence of ξw′ on the RH for FA, compared with the simple organic species, further suggests that the chemical interactions between the FA and inorganic species are likely to be very different from those between the simpler organic species and inorganic species. Natural HA and FA are known to have strong interactions 5114

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 22, 2003

(1) Dick, W. D.; Saxena, P.; McMurry, H. P. J. Geophys. Res. 2000, 105, 1471. (2) Andrews, E.; Larson, S. Environ. Sci. Technol. 1993, 27, 857. (3) Saxena, P.; Hildemann, L. M.; McMurry, P. H.; Seinfeld, J. H. J. Geophys. Res. 1995, 100, 18, 755. (4) Hameri, K.; Charlson, R. J.; Hansson, H. C.; Jacobson, M. J. J. Aerosol Sci. 1997, 28, S153. (5) Sempere, R.; Kawamura, K. Atmos. Environ. 1994, 28, 449. (6) Sempere, R.; Kawamura, K. Atmos. Environ. 1996, 30, 1609. (7) Krivacsy, Z.; Gelencser, A.; Kiss, G.; Meszaros, E.; Molnar, A.; Hoffer, A.; Meszaros, T.; Sarvari, Z.; Temesi, D.; Varga, B.; Baltensperger, S.; Nyeki, S.; Weingartner, E. J. Atmos. Chem. 2001, 39, 235. (8) Yao, X. H.; Fang, M.; Chan, C. K. Atmos. Environ. 2002, 36, 2099. (9) Andracchio, A.; Cavicchi, C.; Tonelli, D.; Zappoli, S. Atmos. Environ. 2002, 36, 5097. (10) Decesari, S.; Facchini, M. C.; Fuzzi, Z.; Tagliavini, E. J. Geophys. Res. 2000, 105, 1481. (11) Decesari, S.; Facchini, M. C.; Matta, E.; Mircea, M.; Fuzzi, S.; Chughtai, A. R.; Smith, D. M. Atmos. Environ. 2002, 36, 1827. (12) Gelencser, A.; Hoffer, A.; Molnar, A.; Krivascy, Z.; Kiss, G.; Meszaros, E. Atmos. Environ. 2000, 34, 823. (13) Gelencser, A.; Sallai, M.; Krivascy, Z.; Kiss, G.; Meszaros, E. Atmos. Res. 2000, 54, 157. (14) Havers, N.; Burba, P.; Lambert, J.; Klockow, D. J. Atmos. Chem. 1998, 29, 45. (15) Havers, N.; Burba, P.; Klockow, D.; Klockow-Beck, A. Chromatographia 1998, 47, 619. (16) Cini, R.; Innocenti, N. D.; Oppo, C.; Orlandi, G.; Stortini, G.; Tesel, U.; Udisti, R. Int. J. Environ. Anal. Chem. 1996, 63, 15. (17) Facchini, M. C.; Fuzzi, S.; Zappoli, S.; Andracchio, A.; Gelencser, A.; Kiss, G.; Krivacsy, Z.; Meszaros, E.; Hansson, H.-C.; Alsberg, T.; Zebuhr, Y. J. Geophys. Res. 1999, 104, 26821. (18) Facchini, M. C.; Decesari, S.; Mircea, M.; Fuzzi, S.; Loglio, G. Atmos. Environ. 2000, 34, 4853. (19) Zappoli, S.; Andracchio, A.; Fuzzi, S.; Facchini, M. C.; Gelencse´r, A.; Kiss, G.; Kriva´csy, Z.; Molnar, A.; Meszaros, E.; Hansson, H.-C.; Rosman, K.; Zebuhr, Y. Atmos. Environ. 1999, 33, 2733. (20) Fuzzi, S.; Decesari, S.; Facchini, M. C.; Matta, E.; Mircea, M. Geophys. Res. Lett. 2001, 20, 4079. (21) Lightstone, J. M.; Onasch, T. B.; Imre, D.; Oatis, S. J. Phys. Chem. 2000, 104, 9337. (22) Cruz, C. N.; Pandis, S. N. Environ. Sci. Technol. 2000, 34, 4313. (23) Hameri, K.; Charlson, R.; Hansson, H.-C. AlChE 2002, 48, 1309. (24) Choi, M. Y.; Chan, C. K. Environ. Sci. Technol. 2002, 36, 2422. (25) Peng, C. G.; Chan, C. K. Atmos. Environ. 2001, 35, 1183. (26) Peng, C. G.; Chan, M. N.; Chan, C. K. Environ. Sci. Technol. 2001, 35, 4495. (27) Peng C. G.; Chow A. H. L.; Chan C. K. Aerosol Sci. Technol. 2001, 35, 753. (28) Prenni, A. J.; DeMott, P. J.; Kreidenweis, S. M.; Sherman, E.; Russell, L. M.; Ming, Y. J. Phys. Chem. A 2001, 105, 11240. (29) International Humic Substances Society. Web page: http:// www.ihss.gatech.edu/. (30) Aiken, G. R. In Humic Substances in Soil, Sediment, and Water; Aiken, G. R., McKnight, D. M., Wershaw, R. L., MacCarthy, P., Eds.; Wiley: New York, 1985; pp 363-385. (31) Stevenson, F. J. Humic Chemistry: Genesis, Composition, Reactions, 2nd ed.; Wiley: New York, 1994. (32) Leenheer, J. A.; McKnight, D. M.; Thurman, E. M. In Humic Substances in the Suwannee River, Georgia: Interactions, Properties and Proposed Structures; USGS Open File Report 87557; Averett, R. C., Leenheer, J. A., McKnight, D. M., Thorn, K. A., Eds.; U.S. Geological Survey: Denver, CO, 1989; pp 331359.

(33) Gysel, M.; Nyeki, S.; Weingartner, E.; Baltensperger, U.; Galambos, I.; Kiss, G. Abstract of papers; The Sixth International Aerosol Conference, Taipei, Taiwan, 2002; p 859. (34) Jone, M. N.; Birkett, J. W.; Wilkinson, A. E.; Hesketh, N.; Livens, F. R.; Bryan, N. D.; Lead, J. R.; Hamilton-Taylor, J.; Tipping, E. Anal. Chim. Acta 1995, 314, 149. (35) Han, J. H.; Hung, H. M.; Martin, S. T. J. Geophys. Res. 2002, 107, 3-1-3-9 10.1029/ 2001JD001054. (36) Swift, R. S. In Humic Substances II: In Search of Structure; Hayes, M. H. B., MacCarthy, R., Malcolm, R. L., Swift, R. S., Eds.; Wiley: New York, 1989; pp 449-465. (37) Ristovski, Z. D.; Morawska, L.; Hitchins, J.; Barron, W. J. Aerosol. Sci. 1998, 29, S327. (38) Chan, C. K.; Ha, Z.; Choi, M. Y. Atmos. Environ. 2000, 34, 4795. (39) Zhang, Y. H.; Chan, C. K. J. Phys. Chem. A 2002, 106, 285. (40) Chan C. K.; Flagan R. C.; Seinfeld J. H. Atmos. Environ. 1992, 26, 1661. (41) Tang, I. N.; Wong, W. T.; Munkelwitz, H. R. Atmos. Environ. 1981, 12, 2463.

(42) Tang, I. N.; Tridico, A. C.; Fung, K. H. J. Geophys. Res. 1997, 102, 23269. (43) Tang, I. N.; Munkelwitz, H. R. J. Geophys. Res. 1994, 99, 18801. (44) Zdanovskii, A. B. Zh. Fiz. Khim. 1948, 22, 1475. (45) Stokes, R. H.; Robinson, R. A. J. Phys. Chem. 1966, 70, 2126. (46) van Dijk, H. In Soil Biochemistry; McLaren, A. D., Skujins, J., Eds.; Dekker: New York, 1971; Vol. 2, Chapter 2. (47) Mingelgrin, U.; Gerstl, Z. In Organic Substances in Soil and Water: Natural Constituents and Their Influence on Contaminant Behavior; Beck, A. J., Jones, K. C., Hayes, M. H. B., Mingelgrin, U., Eds.; Royal Society of Chemistry: London, 1993; pp 102-127. (48) Choi, M. Y.; Chan, C. K. J. Phys. Chem. A 2002, 106, 4566.

Received for review March 26, 2003. Revised manuscript received August 27, 2003. Accepted September 9, 2003. ES034272O

VOL. 37, NO. 22, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

5115