Hygroscopic Properties of Internally Mixed Particles of Ammonium

6124

J. Phys. Chem. A 2010, 114, 6124–6130

Hygroscopic Properties of Internally Mixed Particles of Ammonium Sulfate and Succinic Acid Studied by Infrared Spectroscopy Lorena Min˜ambres, Marı´a N. Sa´nchez, Fernando Castan˜o, and Francisco J. Basterretxea* Departamento de Quı´mica Fı´sica, Facultad de Ciencia y Tecnologı´a, UniVersidad del Paı´s Vasco/Euskal Herriko Unibertsitatea, Campus de Leioa, B. Sarriena, s/n, Leioa 48940, Spain ReceiVed: February 5, 2010; ReVised Manuscript ReceiVed: April 23, 2010

Atmospheric aerosols typically consist of inorganic and organic material. Although the organic fraction can affect the behavior of mixed organic/inorganic particles, their physical properties are not well-understood. In this work, infrared spectra of internally mixed submicrometer particles of ammonium sulfate and succinic acid have been studied at ambient temperature in an aerosol flow cell. The spectra of dried particles show distinct features relative to the pure components, as a result of ion-molecule interactions between the inorganic and organic species. The hygroscopic behavior of the particles has been followed for several organic/inorganic mass ratios, showing that around equimolar composition, the mixed particles uptake water in a broad range of relative humidities (30-80%), substantially lower than the deliquescence relative humidity of the mixed system near 80%. Infrared spectra at predeliquescence relative humidities reveal that succinic acid is partially or completely in the liquid phase at much higher concentrations that those corresponding to a saturated solution of succinic acid. This behavior is proposed to arise from the ion-molecule interactions between the organic and inorganic components, which unstabilize the crystal structure relative to the pure solids and cause loss of translational order in the crystal, bringing about an increase in the Gibbs energy of the solid particles and allowing the uptake of water molecules prior to the deliquescence point. The obtained results show that water absorption prior to full deliquescence in this system has to be taken into account because it extends the range of relative humidities at which particles are partially or completely liquid. Introduction Atmospheric aerosols have a considerable impact on air pollution and human health.1,2 They influence visibility in the troposphere and the earth’s radiative energy balance due to their light scattering properties and changes in atmospheric chemical pathways. Atmospheric particles can absorb water and grow with increasing relative humidity (RH), with a possible change of phase, which, in turn, affects their properties. Aerosol particles typically consist of inorganic and organic material and elemental carbon. The most abundant water-soluble inorganic ions are sulfate, nitrate, calcium, chloride, sodium, and ammonium. Organic matter is composed of a wide variety of organic compounds, which can amount to 30-50% of fine particulate mass, depending on location.3-6 Field studies indicate that 20-70% of condensed-phase organic matter is water-soluble7 and includes dicarboxylic acids such as oxalic, malonic, and succinic, which have been identified at a variety of sources, urban and remote.8 The organic fraction can affect properties of atmospheric aerosol particles such as light scattering, hygroscopicity,phasetransition,solubility,orchemicalreactivity.9-13 Despite its importance, the physical state of mixed organic/ inorganic aerosol particles is not well-characterized.14 In laboratory studies, the most effort has been devoted to carboxylic acids such as formic, acetic, succinic, glutaric, oxalic, pyruvic, and malonic acids.8,9,15-18 There has been some debate about the effect of organics on the hygroscopicity of inorganic aerosols, some data suggesting that most organics have a negative effect on the growth factor or evaporation rate of inorganic aerosols and others pointing out that specific * To whom correspondence should be addressed. Phone: +34 94 601 2532. Fax: +34 94 601 3500. E-mail: [email protected].

organic-inorganic interactions result in both positive and negative effects on the hygroscopic properties.15 The general trend appears to be that water-soluble organics may enhance water uptake or may have little impact, whereas insoluble organics tend to inhibit water uptake by atmospheric aerosol. Succinic acid has a very high deliquescence relative humidity (DRH > 94%) and has been proposed to behave as insoluble species in atmospheric aerosols once they are crystallized.19 The hygroscopic properties of the succinic acid/ammonium sulfate (SA/AS) mixed system have been studied by several investigators. Choi and Chan15 reported that the organic acid enhances water absorption of AS with respect to the pure inorganic salt, although the deliquescence behavior and the DRH of (NH4)2SO4 at 80% are not affected. Similar results were obtained by other authors.20,21 On the other hand, Ha¨meri et al.22 concluded that the deliquescence behavior of mixed succinic acid and ammonium sulfate particles is not completely defined, with particles seeming to exist both as crystalline and solution droplets well below the DRH of pure ammonium sulfate. Marcolli and Krieger18 have proposed that water uptake in mixtures of AS and dicarboxylic acids before full deliquescence shows significant deviations with respect to thermodynamic equilibrium that may be caused by morphological effects. Recently, Ling and Chan23 have reported partial deliquescence of AS/SA particles at RHs of 80 to >90%. Previous studies on internally mixed particles of SA and AS have been carried out mostly by electrodynamical balance or DMA techniques. As far as we know, no detailed infrared spectroscopy studies have been presented for this system up to now. Infrared spectroscopy is a sensitive technique to characterize aerosol composition, water content, and particle phase, and

10.1021/jp101149k  2010 American Chemical Society Published on Web 05/05/2010

Mixed Particles of Ammonium Sulfate and Succinic Acid

Figure 1. Experimental setup for the study of hygroscopic properties of aerosols. CPC: condensation particle counter; APS: aerosol particle spectrometer; IR source: infrared light source; FT-IR: Fourier transform infrared spectrometer.

has been used to study other internally mixed organic/inorganic particles.24 The spectral features of aerosols can also reveal information about molecular interactions and the formation of new species. In the present work, we study the water uptake properties of internally mixed ammonium sulfate/succinic acid particles for different compositions and show that the presence of succinic acid enhances the water uptake of ammonium sulfate at RHs lower than the full deliquescence of AS, changing the phases of particles. We rationalize the observed behavior in terms of ion-molecule interactions that change the energy of the mixed solid with respect to the pure compounds. Experimental Section Figure 1 sketches the experimental arrangement used.25 Mixed aqueous solutions of succinic acid (SA) and ammonium sulfate (AS) are formed by dissolving the corresponding solids (SA: Sigma-Aldrich, g99.0%; AS: Alfa Aesar, 99.95%) into deionized water at concentrations ranging from 0.02 to 0.05 kg of solute per dm3 of water and different solute mixing ratios. Aerosols are formed by adding the solutions to a commercial constant output atomizer (TSI 3076), in which pure N2 (Praxair, 99.990%) is injected at 2.5 bar, yielding submicrometer aerosol particles. The formed aerosol flow either enters or bypasses a diffusion drier (TSI 3062). The relative humidity is controlled by mixing two flows of N2 (Praxair 99.990%), one saturated with water vapor from a temperature-controlled water bubbler and the other dry. Both flows are subsequently added to the aerosol flow. The sum of both flows is kept constant in order to maintain the aerosol concentration unchanged. The highest RHs are obtained by heating the water bath to about 35 °C to increase the amount of water vapor, although this did not affect the temperature in the aerosol cell much, which only increased by 1 °C. The resultant flow enters a 1 m long, 50 mm diameter infrared pyrex absorption cell with ZnSe windows at its ends at ambient temperature. RH values are measured with a digital thermohygrometer (Hanna HI93640N) in the central section of the flow exiting the aerosol cell, with a measurement error of 3% for 50 < RH < 85% and 5% in the rest of the interval. An infrared light source (ORIEL 6580) gives collimated infrared radiation that goes lengthways through the aerosol cell axis. The outcoming radiation is directed to a FT-IR spectrom-

J. Phys. Chem. A, Vol. 114, No. 20, 2010 6125

Figure 2. Infrared extinction spectra of succinic acid particles formed by atomizing an aqueous solution. Top: spectrum of undried aqueous particles. Bottom: spectrum of solid particles after passing through a diffusion drier. Gaseous water and carbon dioxide lines have been removed.

eter (Nicolet Magna 860) to record an extinction spectrum from 700 to 4000 cm-1. In order to reduce ambient water infrared absorption, the optical path is sealed and flushed by a current of dry air. Background spectra are recorded before aerosol spectra with the aerosol cell evacuated by a rotary diaphragm pump. Sample interferograms are averaged by collecting typically 200 scans at 4 cm-1 resolution. A small portion of the aerosol flow is diverted to the entrance of an aerodynamic particle spectrometer (TSI 3321) and a condensation particle counter (TSI 3781) to probe particle aerodynamic diameter distribution26 and total particle concentration, respectively. The particle spectrometer can size particles in the 0.50-20 µm interval, with a resolution of 32 channels per decade. The measured aerodynamic diameters of dry particles fit reasonably well to log-normal distributions, from j a) and the standard which the aerodynamic median diameter (D deviation (σ) can be extracted. Geometric median diameters are jg ) D j a(F0/F)1/2, where F0 is then obtained by the formula27 D the unit density and F is the particle density. A Stokes correction26 has to be applied to the size distribution values for particle densities below 0.9 or above 1.1 g cm-3. As an example, j g ) 0.68 µm and σ ≈ 1.44 are obtained for succinic values of D acid particles formed from a 0.05 kg/L aqueous solution. Results and Discussion A. Infrared Extinction Spectra. Figure 2 shows extinction infrared spectra of succinic acid solid and aqueous particles. The absorption lines due to gaseous water have been removed, although some small features remain around 1500 and 3600 cm-1. The 2300-2400 cm-1 range, in which atmospheric CO2 absorption appears, has been suppressed for clarity. The infrared spectrum of the solid particles matches that available in the literature for the condensed phase.28 It has a broad absorption band in the 2000-3400 cm-1 region with a number of prominent peaks due to O-H stretching in the intermolecular hydrogen-bonding structure and intramolecular C-H stretching.29 In crystalline succinic acid, the molecular unit is rather planar and forms linear chains via hydrogen bonds.30 In the 800-1800 cm-1 region, there are several narrow

6126

J. Phys. Chem. A, Vol. 114, No. 20, 2010

Figure 3. Infrared absorption spectra of internally mixed SA/AS dry particles at several compositions, where f(SA) stands for dry mass fraction of succinic acid. The stars show the position of weak bands. The spectrum of pure ammonium sulfate particles is also shown at the bottom.

features, of which the most intense is centered at 1700 cm-1 and originates from a CdO stretching vibration, whereas the bands in the 700-1000 cm-1 range have been assigned to crystal chain vibrations.31 However, the broad band centered at 937 cm-1 and the low intensity one at 812 cm-1, which also appear in the spectrum of gaseous succinic acid,28 can be assigned to intramolecular vibrations. These two bands are also present in the spectrum of the dissolved acid, although somewhat displaced (at 958 and 846 cm-1, respectively). The band at 1700 cm-1 is also found in the spectrum of the aqueous particles, overlapping with nearby liquid water absorption, but the bands between 800 and 1600 cm-1 in the solid disappear or transform in the aqueous phase. These bands will be used to follow the phase transformation of succinic acid from solid to aqueous solution, especially the lower wavenumber region, as it is free of interference from gaseous and liquid water absorption. Finally, the broad and intense liquid water band centered at ∼3400 cm-1 and the succinic acid broad band centered at ∼3000 cm-1 are overlapped. The infrared spectrum of solid ammonium sulfate particles in the 800-4000 cm-1 region (not shown) is well-known32 and consists of the ν3(T2) stretching mode of the SO42- ion at 1115 cm-1 and two modes due to the NH4+ ion, the broad ν3(T2) stretching between 2800 and 3400 cm-1 and the ν4(T2) bending at 1430 cm-1. For aqueous ammonium sulfate particles, the obvious changes are the moderate shifts of the ν4(NH4+) and ν3(SO42-) bands to higher and lower wavenumbers, respectively. The ν3(NH4+) band is extensively overlapped with the liquid water absorption band. Figure 3 presents infrared absorption spectra of internally mixed SA/AS dry particles in the 800-1200 cm-1 range obtained for drying particles formed from a 0.05 kg of solute per liter at various f(SA) compositions, where f(SA) stands for dry mass fraction of succinic acid (e.g., f(SA) ) 0.20 means 20% SA and 80% AS). For comparison, the spectrum of pure AS particles at the same total concentration is also shown. The solid mixture produces three noticeable spectral changes with respect to their separate components. The first is the gradual

Min˜ambres et al. decrease of the ν3(SO42-) band center wavenumber from the pure sulfate at 1115 to 1097 cm-1 for f(SA) ) 0.9. The second is the broadening of the same band, which increases from a fwhm of 34 cm-1 for the pure sulfate to 77 cm-1 at f(SA) ) 0.6, more than twice the first value, and decreases for f(SA) ) 0.9. The band shape shows an asymmetry (especially for f(SA) ) 0.6 and 0.9) indicative of overlapped components. It is to be noted that the region near 1115 cm-1 is free of overlapping with SA bands; therefore, the spectral changes with composition must arise from the ν3(SO42-) band. No similar effect has been observed in the ν4(NH4+) band. Finally, Figure 3 shows the appearance of two new weak bands most visible at f(SA) ) 0.6 centered at 841 and 975 cm-1, close to succinic acid bands, and absent from the spectra of the separate solid components. Interestingly, these two band positions are approximately coincident with bands of aqueous succinic acid. The previous effects can be attributed to ion-molecule interactions between succinic acid molecules and sulfate ions. The small red shift (maximum 17 cm-1) of the intramolecular ν3(T2) stretching mode of the SO42- ion at 1115 cm-1 when mixed with succinic acid in a solid environment may arise from a decrease of the internal force constants of the sulfate ion due to the appearance of intermolecular bonds with succinic acid molecules.33 The broadening of the sulfate ν3(T2) mode may be due to various reasons,33 among which are transverse optic/ longitudinal optic splitting of infrared-allowed bands, loss of translational symmetry due to the more disordered structure in the mixed solid, and anharmonic coupling of the low-frequency mode due to the formation of hydrogen bonds between the COOH group and the oxygens of the sulfate ion. Band splitting is visible at f(SA) ) 0.6 and 0.9 (Figure 3), and band asymmetry can be seen even at f(SA) ) 0.2. We think that the presence of surrounding succinic acid molecules lowers the local symmetry around the sulfate ion, removing the degeneracy of the T2 (triply degenerate) mode of sulfate. We have reproduced the measured sulfate band contours at several f(SA) by adding up three Gaussian shaped bands, each of approximately the same fwhm of a pure sulfate band at 1115 cm-1 and a different relative intensity for each composition. The results are presented in Figure 4. At f(SA) ) 0.2, the experimental band shape can be fitted to a Gaussian centered at 1115 cm-1 of higher intensity and two weaker sidebands. Ion-molecule interactions are expected to be weak at this composition, and the band shape of sulfate is only slightly perturbed. For f(SA) ) 0.6 and 0.9, the contribution of the sidebands increases, making band splitting more evident, in accordance with a higher number of SA molecules surrounding sulfate ions and originating stronger interactions. The appearance of the new weak bands at 841 and 975 cm-1 for f(SA) ) 0.6 can also arise from vibrational coupling phenomena between succinic acid molecules and sulfate ions. These band positions are almost coincident with two bands of aqueous succinic acid, the result being that the environment of sulfate and ammonia ions surrounding succinic acid molecules produces band shift effects similar to those produced by water molecules. These weak bands have not been detected at lower f(SA) values, probably due to low signal. These low frequencies correspond to the intramolecular chain vibration region of the dicarboxylic acid, and it is proposed that the presence of sulfate and ammonia ions slightly changes the vibrational frequencies of the chain to higher wavenumbers due to repulsive forces in the lattice.33 B. Deliquescence and Efflorescence Behavior of Particles. The deliquescence and efflorescence behavior of SA/AS internally mixed particles was investigated by monitoring their

Mixed Particles of Ammonium Sulfate and Succinic Acid

J. Phys. Chem. A, Vol. 114, No. 20, 2010 6127

Figure 5. Deliquescence curves showing liquid water integrated areas in the 2800-3600 cm-1 region versus the relative humidity for internally mixed particles of ammonium sulfate and succinic acid at various compositions, where f(SA) stands for dry mass fraction of succinic acid.

Figure 6. Relative humidity at which the onset of water uptake is observed in deliquescent particles versus the mole fraction of succinic acid for internally mixed particles of ammonium sulfate and succinic acid.

Figure 4. Sulfate ion absorption bands in mixtures of SA and AS in solid particles at a number of solute mass ratios. The experimental band contours can be reproduced by the sum of three Gaussian curves (named g1, g2, and g3) of different relative intensities.

infrared extinction spectra for different f(SA) as a function of relative humidity. For deliquescence experiments, atomized particles prepared from the 0.020 kg/dm3 starting solution concentration were dried by passing them through the diffusion drier and were mixed with adequate flows of humid nitrogen to get the desired RH. The water content of particles as a function of RH was measured by recording the absorption band area of liquid water from 2800 to 3600 cm-1, after subtracting SA and AS spectral features. The results are presented in Figure 5. For pure ammonium sulfate particles (f(SA) ) 0) the deliquescence curve agrees with the results already published;34,35 no water is taken up by the particles until RH ) 80%, where the particles deliquesce abruptly. At the other extreme, when almost all solute is succinic acid (f(SA) ) 0.90), we get similar results as those for pure SA;19 particles remain completely dry until RH near saturation at ∼98%, when they begin to take up

water. At f(SA) ) 0.25, the onset of water uptake is located near that of pure ammonium sulfate. However, at compositions approaching a 1:1 solute mass ratio (f(SA) ) 0.40 and 0.60), the behavior of the mixture changes qualitatively; particles begin to take up water at very low RHs. The starting values of the relative humidity at which we observe liquid water in the extinction spectrum for f(SA) ) 0.40 and 0.60 are near 40%, much lower than those for any of the separate components. We have performed additional measurements at other compositions that have not been included in Figure 5 for clarity, but they follow the same trend. Figure 6 shows the variation of the onset of water uptake as a function of the succinic acid mole fraction x(SA) for all of the compositions studied. To plot Figure 6, it has been taken into account that succinic acid can evaporate partially upon atomizing with respect to ammonium sulfate,22 thus altering the composition of the original mixture. To quantify this effect, we have recorded the infrared spectra of an aqueous solution of succinic acid and ammonium sulfate at a 1:1 solute mass ratio and also that of atomized particles from the previous solution, without further drying or diluting. Calculating the integrated absorbance ratio A˜(SA)/ A˜(AS) for two given bands of SA and AS in both bulk solution and atomized particles, the change in relative composition upon atomization was obtained. For a 1:1 mass ratio for the solution, we measured a ratio of 0.8:1 (SA/AS) in the particles, indicating near to 20% evaporation of succinic acid relative to ammonium sulfate. In Figure 6, it can be seen that the points near x(SA) ) 0.38 do not follow a smooth curve. We think that this result comes from measurement errors in the RH of the water uptake

6128

J. Phys. Chem. A, Vol. 114, No. 20, 2010

Min˜ambres et al.

Figure 7. Efflorescence curves showing liquid water integrated areas in the 2800-3600 cm-1 range versus the relative humidity for internally mixed particles of ammonium sulfate and succinic acid at various compositions, where f(SA) stands for dry mass fraction of succinic acid.

onset, as it can be seen that the corresponding f(SA) ≈ 0.5 water uptake curves in Figure 5 have small slopes near the onset, resulting in RH determination error. The efflorescence behavior of internally mixed particles of succinic acid and ammonium sulfate atomized from a 0.020 kg/L solution was investigated for several compositions, and the results are shown in Figure 7. Aqueous particles were mixed with a flow of dry nitrogen, and the integrated area of liquid water in the particles was recorded as a function of the relative humidity. The data in Figure 7 are in accordance with the results already reported;8,15,19 particles crystallize at RH ≈ 50%, and no discernible differences are observed with the composition of the particles. Information about the particle phase in the deliquescence experiments has been obtained from the spectral features of the mixed particles. The band at 1200 cm-1 has been chosen to monitor solid succinic acid as it does not overlap with ammonium sulfate, aqueous succinic acid, or liquid water bands. Furthermore, dissolved succinic acid can be followed by observing the nearby band at 1179 cm-1. The phase distinction for ammonium sulfate is more involved; the sulfate ion band at 1115 cm-1 in the pure solid is perturbed by succinic acid, and it is obscured within the features of the aqueous sulfate band at 1105 cm-1. Ammonium ion bands centered at 1425 (solid) and 1455 cm-1 (solution) are not affected by mixing with succinic acid, but these bands are broad and extensively overlapped. Although spectral features of the former bands have been processed to extract the solid and aqueous separate components, no quantitative results of AS phases have been obtained. The results for the phase analysis obtained from the deliquescence spectra can be summarized as follows. For 0 e f(SA) e 0.3, succinic acid and ammonium sulfate are in the solid phase for the entire range of RH values up to 80% (spectral features are not so clear for higher RHs), which is in agreement with no water uptake observed until high RH. For 0.3 < f(SA) e 0.6, spectral lines of both solid and dissolved succinic acid appear at RHs as low as 13%, the coexistence of both phases continues up to RH ≈ 50%, and only bands of dissolved SA are present at higher RHs. These results are in accordance with the onset of water uptake at low RHs observed in this interval. Figure 8 shows an example of the observed phase changes for f(SA) ) 0.6 particles. Bands of solid succinic acid and sulfate are present at the lowest RH, but at RH ) 35%, the band of solid succinic acid

Figure 8. Infrared spectra of internally mixed particles of succinic acid and ammonium sulfate at f(SA) ) 0.6 showing solute phase changes. The absorption band of pure ammonium sulfate is also shown for comparison.

at 1200 cm-1 is no longer neat, whereas the band of the liquid at 1179 cm-1 emerges. At RH ) 68%, all succinic acid is in the liquid phase. The spectral bands above 1200 cm-1 and at RH ) 68% are due to imperfect subtraction of gaseous water lines. Regarding the AS phase, in Figure 8, the band shape at RH ) 1% is different from those at RH ) 35 and 68%, which are indistinguishable. This indicates that at least part of AS is in solution. Finally, for 0.7 e f(SA) e 0.9, the results indicate that solid and dissolved phases of SA coexist for a wide range of RHs. At f(SA) ) 0.70, only solid succinic acid appears up to RH ≈ 30%, followed by coexistence of solid and dissolved phases for RH in the 30-75% interval. At higher humidities, only the bands of the liquid are visible. At f(SA) ) 0.90, solid and liquid coexistence of SA is observed starting at RH ≈ 65% until the highest RH values measured, although the solid is predominant. No attempt has been made to follow AS phases in this composition range since ammonium sulfate absorption is low at this mole fraction. Particle composition can be calculated from their infrared spectra. The number of molecules of a given species i per unit volume of aerosol sample Ni is obtained by measuring its integrated band absorbance and applying the Beer-Lambert law32

A˜i )

σ¯ iNiz 2.303 × 102

j i the where A˜i is the integrated band absorbance (cm-1), σ integrated absorption cross section per molecule (m molecule-1), and z the optical path length (m). For liquid water, the integrated absorption cross section from 2800 to 3600 cm-1 has been j H2O ) 1.3 × 10-18 m computed from pure water data36 to be σ molecule-1. To calculate integrated band absorbances from aerosol extinction spectra, one needs to subtract the scattering component first, which can be calculated by Mie theory26 if the optical constants and size distribution of the particles are known.

Mixed Particles of Ammonium Sulfate and Succinic Acid As the optical constants of succinic acid are not available, the scattering contibution of the liquid water extinction spectrum has been removed by simply subtracting the sloping baseline present at high wavenumbers to obtain integrated absorbances A˜H2O. We have checked that this method yields approximate results from our previous Mie calculations in aqueous aerosols.25 The absorption cross section for dissolved succinic acid has been computed from the integrated absorbance of an aqueous solution of succinic acid of known composition in an infrared cell for liquid samples in the 1118.6-1305.06 cm-1 interval. The absorption cross section of solid succinic acid is then calculated by measuring the integrated absorbances of solid and aqueous particles separately in the aerosol cell at the same flow to ensure the same particle number density and applying the formula σs ) σaqA˜s/A˜aq. The relative composition of particles on average can be calculated under these conditions from their spectra and the corresponding absorption cross sections. As an example, at f(SA) ) 0.6 and RH ) 68%, in which all succinic is liquid, NH2O/ NSA,aq ≈ 9. This value is in contrast with the solubility of succinic acid in water of 7.2 g/100 mL at ambient temperature,37 corresponding to a molecule ratio of NH2O/NSA ≈ 90, which is the proportion expected at the deliquescence RH, about 10 times bigger than that in our experiment. For f(SA) ) 0.90, solid and dissolved succinic acid phases coexist, although the solid phase is predominant; for example, at RH ≈ 98%, the liquid to solid ratio of succinic acid is calculated to be NSA,s/NSA,aq ≈ 10. The perturbations in the spectra of internally mixed solid particles of SA and AS can be related to their water uptake properties. For particles consisting mostly of SA, the magnitude of the intermolecular interactions is weak; therefore, the hygroscopic behavior of the particles is very similar to that of pure AS, with no water uptake until full deliquescence. As the mixing ratio approaches an equimolar mixture, the crystal structure order of AS will be lowered, and the NH4+-SO42Coulombic interaction energy will diminish as a result of the inclusion of succinic acid molecules. These interactions cause the system to raise its Gibbs free energy, with the result of admitting water molecules in the 35-80% RH range before the complete deliquescence of ammonium sulfate in order to lower its energy. Water uptake will likely start at the outer shell of the particles (probably polycrystalline) which are gradually dissolved, although the particles have not taken up enough water to deliquesce completely. For particles consisting mainly of succinic acid (e.g., f(SA) ) 0.90), solid and aqueous phases coexist, possibly in the form of a solid core containing most of the particle mass, surrounded by an external shield of dissolved solute, according to one of the morphological structures proposed for aerosol particles.38 We have compared the observed deliquescent behavior with the predictions of the extended AIM aerosol thermodynamics model, which is one of the models widely used to predict hygroscopic properties of aerosol particles and can be used online via its Web site (http://www.aim.env.uea.ac.uk/aim/ aim.php). The method can be applied to mixed inorganic/organic aqueous solutions.39 For a 1:1 mol ratio of SA and AS, the model predicts identical results as those in pure AS; no water uptake until RH ) 80%, when prompt deliquescence takes place and the solid abruptly incorporates a large amount of water to form a saturated salt solution. However, delayed deliquescence over a range of relative humidities has been observed previously in this and other systems.18,22 It has been suggested40 that the coexistence of core and film in micrometer-sized crystals cannot be due to surface effects and must be attributed to some other

J. Phys. Chem. A, Vol. 114, No. 20, 2010 6129 factor such as the physical or chemical state of the initial soluble particle. Several laboratory studies indicate that, for nearequimolar mixtures of AS and SA particles, there is no significant water uptake prior to the AS deliquescence RH.21,23 Prenni et al.21 reported a DRH of 76.3%, practically equal to that of pure AS. Ling and Chan23 observed partial deliquescence of AS/SA internally mixed particles at a 1:1 mol ratio and RHs from 80 to >90%, with AS being observed to dissolve at 80% RH while SA remained as a solid for RHs as high as 90%. Although our results indicate that the amount of uptaken water at RH < 80% is rather small, its presence is clearly revealed by infrared absorption spectroscopy, which is a sensitive technique for detecting liquid water. Water uptake at low RHs in AS/SA internally mixed particles can arise from the influence of particle formation conditions in different experiments. It may be the case that the internally mixed solid particles formed in our experiment after drying may be influenced by kinetic effects rather than by thermodynamic ones, leading to the formation of metastable solid phases that affect the particle hygroscopic properties. Raman spectra of AS/SA particles23 have manifested the formation of such metastable forms upon crystallization. Infrared spectroscopy data presented in this work help to complement the information about the AS/SA system by indicating the presence of ion-molecule interactions in internally mixed solid particles. Conclusion and Atmospheric Implications Our results indicate that the internal mixing of ammonium sulfate and succinic acid in submicrometer particles produces ion-molecule interactions (manifested in the infrared spectra of the particles) that reduce cation-anion Coulombic attractive forces, analogous to the role of a dielectric reducing the attraction between two opposite charges, and increases structural disorder in the crystals with respect to pure components. These changes will produce a slight increase in the Gibbs free energy of the mixed system, which has the effect of admitting more liquid water in the particles in order to lower the energy of the system, forming particles in which all SA and at least part of AS are in the liquid phase at mixing ratios approaching equimolarity, although full deliquescence of AS particles is largely unaffected by the presence of SA. Water absorption prior to full deliquescence is important because it extends the range of RH values at which liquid-phase reactions can take place.16 The presence of water-soluble organics internally mixed with ammonium sulfate aerosol can increase the range of conditions under which the aerosol is a solution, which in turn will influence the gas uptake properties and chemical reactivity of the particles. Our results suggest that this effect, mainly present in water-soluble organic acids, also manifests in some low-solubility acids due to specific physicochemical interactions. For succinic acid and ammonium sulfate, the effect is stronger at 50 wt % fraction, at which particles will be liquid (at least partially) for a broad range of RHs. This behavior, if generally applicable, would have a noticeable impact on the phase of chemically mixed aerosol in the atmosphere. Acknowledgment. The authors are grateful to Ministerio de Ciencia e Innovacio´n (Madrid) for grant-in-aids (CGL200806041/CLI and Consolider CSD-2007-00013), to Gobierno Vasco/Eusko Jaurlaritza (Vitoria-Gasteiz) for general support through a Consolidated Research Group grant, and to Universidad del Paı´s Vasco/Euskal Herriko Unibertsitatea (UPV/EHU). L.M. thanks UPV/EHU for a research grant.

6130

J. Phys. Chem. A, Vol. 114, No. 20, 2010

References and Notes (1) Pachauri, R., Reisinger, A., Eds.; Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the IntergoVernmental Panel on Climate Change; IPCC: Geneva, Switzerland, 2007. (2) Russell, A. G.; Brunekreef, B. A Focus on Particulate Matter and Health. EnViron. Sci. Technol. 2009, 43, 4620–4625. (3) Duce, R. A.; Mohnen, V. A.; Zimmerman, P. R.; Grosjean, D.; Cautreels, W.; Chatfield, R.; Jaenicke, R.; Ogren, J. A.; Pellizzari, E. D.; Wallace, G. T. Organic material in the global troposphere. ReV. Geophys. Space Phys. 1983, 21, 921–952. (4) Chow, J. C.; Watson, J. G.; Fujita, E. M.; Lu, Z. Q.; Lawson, D. R.; Ashbaugh, L. L. Temporal and spatial variations of PM2.5 and PM10 aerosol in the Southern California Air Quality Study. Atmos. EnViron. 1994, 28, 2061–2080. (5) Seinfeld, J. H.; Pandis, S. N. Atmospheric Chemistry and Physics: From Air Pollution to Climate Change; Wiley-Interscience: New York, 1998. ´ .; Me´sza´ros, E.; Hansson, H. C.; Karlsson, H.; Gelencse´r, (6) Molna´r, A A.; Kiss, G. Y.; Kriva´csy, Z. The importance of organic and elemental carbon in the fine atmospheric aerosol particles. Atmos. EnViron. 1999, 33, 2745–2750. (7) Saxena, P.; Hildemann, L. M. Water-soluble organics in atmospheric particles: a critical review of the literature and application of thermodynamics to identify candidate compounds. J. Atmos. Chem. 1996, 24, 57–109. (8) Prenni, A. J.; DeMott, P. J.; Kreidenweis, S. M.; Sherman, D. E.; Russell, L. M.; Ming, Y. The Effects of Low Molecular Weight Dicarboxylic Acids on Cloud Formation. J. Phys. Chem. A 2001, 105, 11240–11248. (9) Lightstone, J. M.; Onasch, T. B.; Imre, D.; Oatis, S. Deliquescence, Efflorescence, and Water Activity in Ammonium Nitrate and Mixed Ammonium Nitrate/Succinic Acid Microparticles. J. Phys. Chem. A 2000, 104, 9337–9346. (10) Saxena, P.; Hildemann, L. M.; McMurry, P. H.; Seinfeld, J. H. Organics alter hygroscopic behavior of atmospheric particles. J. Geophys. Res. 1995, 100, 18755–18770. (11) Cruz, C. N.; Pandis, S. N. The effect of organic coatings on the cloud condensation nuclei activation of inorganic atmospheric aerosol. J. Geophys. Res. 1998, 103, 13111–13123. (12) Cruz, C. N.; Pandis, S. N. Deliquescence and Hygroscopic Growth of Mixed Inorganic-Organic Atmospheric Aerosol. EnViron. Sci. Technol. 2000, 34, 4313–4319. (13) Donaldson, D. J.; Vaida, V. The Influence of Organic Films at the Air-Aqueous Boundary on Atmospheric Processes. Chem. ReV. 2006, 106, 1445–1461. (14) Ciobanu, V. G.; Marcolli, C.; Krieger, U. K.; Weers, U.; Peter, T. Liquid-Liquid Phase Separation in Mixed Organic/Inorganic Aerosol Particles. J. Phys. Chem. A 2009, 113, 10966–10978. (15) Choi, M. Y.; Chan, C. K. The effects of organic species on the hygroscopic behaviors of inorganic aerosols. EnViron. Sci. Technol. 2002, 36, 2422–2428. (16) Wise, M. E.; Surratt, J. D.; Curtis, D. B.; Shilling, J. E.; Tolbert, M. A. Hygroscopic growth of ammonium sulfate/dicarboxylic acids. J. Geophys. Res. 2003, 108, AAC4/1–AAC4/8. (17) Parsons, M. T.; Knopf, D. A.; Bertram, A. K. Deliquescence and Crystallization of Ammonium Sulfate Particles Internally Mixed with WaterSoluble Organic Compounds. J. Phys. Chem. A 2004, 108, 11600–11608. (18) Marcolli, C.; Krieger, U. K. Phase Changes during Hygroscopic Cycles of Mixed Organic/Inorganic Model Systems of Tropospheric Aerosols. J. Phys. Chem. A 2006, 110, 1881–1893.

Min˜ambres et al. (19) Peng, C.; Chan, M. N.; Chan, C. K. The hygroscopic properties of dicarboxylic and multifunctional acids: measurements and UNIFAC predictions. EnViron. Sci. Technol. 2001, 35, 4495–4501. (20) Ming, Y.; Russell, L. M. Thermodynamic equilibrium of organicelectrolyte mixtures in aerosol particles. AIChE J. 2002, 48, 1331–1348. (21) Prenni, A. J.; DeMott, P. J.; Kridenweis, S. M. Water uptake of internally mixed particles containing ammonium sulfate and dicarboxylic acids. Atmos. EnViron. 2003, 37, 4243–4251. (22) Ha¨meri, K.; Charlson, R.; Hansson, H.-C. Hygroscopic properties of mixed ammonium sulfate and carboxylic acids particles. AIChE J. 2002, 48, 1309–1316. (23) Ling, T. Y.; Chan, C. K. Partial crystallization and deliquescence of particles containing ammonium sulfate and dicarboxylic acids. J. Geophys. Res. 2008, 113, D14205. (24) Brooks, S. D.; Garland, R. M.; Wise, M. E.; Prenni, A. J.; Cushing, M.; Hewitt, E.; Tolbert, M. A. Phase changes in internally mixed maleic acid/ammonium sulfate aerosols. J. Geophys. Res. 2003, 108, 4487. (25) Min˜ambres, L.; Sa´nchez, M. N.; Castan˜o, F.; Basterretxea, F. J. Infrared spectroscopic properties of sodium bromide aerosols. J. Phys. Chem. A 2008, 112, 6601–6608. (26) Hinds, W. C. Aerosol Technology. Properties, BehaVior, and Measurement of Airborne Particles; John Wiley & Sons: New York, 1999. (27) Baron, P. A., Willeke, K., Eds. Aerosol measurement: Principles, Techniques and Applications, 2nd ed.; Wiley-Interscience: New Jersey, 2005. (28) National Institute of Standards and Technology Chemistry Webbook. http://webbook.nist.gov/chemistry (2008). (29) Krishnan, S.; Raj, C. J.; Robert, R.; Ramanand, A.; Das, S. J. Growth and characterization of succinic acid single crystals. Cryst. Res. Technol. 2007, 42, 1087–1090. (30) Leviel, J.-L.; Auvert, G. Hydrogen Bond Studies. A Neutron Diffraction Study of the Structures of Succinic Acid at 300 and 77 K. Acta Crystallogr. 1981, B37, 2185–2189. (31) Wehrli, M. Chain vibrations in the infrared spectrum of solid dicarboxylic acids. HelV. Phys. Acta 1941, 14, 516–524. (32) Weis, D. D.; Ewing, G. E. Infrared spectroscopic signatures of (NH4)2SO4 aerosols. J. Geophys. Res. 1996, 101, 18709–18720. (33) Lutz, H. D.; Haeuseler, H. Infrared and Raman spectroscopy in inorganic solids research. J. Mol. Struct. 1999, 511-512, 69–75. (34) Tang, I. N. On the equilibrium partial pressures of nitric acid and ammonia in the atmosphere. Atmos. EnViron. 1980, 14, 819–828. (35) Tang, I. N.; Munkelwitz, H. R. Composition and temperature dependence of the deliquescence properties of hygroscopic aerosols. Atmos. EnViron. 1993, 27A, 467–473. (36) Downing, H. E.; Williams, D. Optical constants of water in the infrared. J. Geophys. Res. 1975, 80, 1656–1661. (37) Lide, D. R. CRC Handbook of Chemistry and Physics, 73 ed.; CRC Press: Boca Raton, FL, 1993. (38) Weis, D. D.; Ewing, G. E. Water content and morphology of sodium chloride aerosol particles. J. Geophys. Res. 1999, 104, 21275–21285. (39) Clegg, S. L.; Seinfeld, J. H. Thermodynamic models of aqueous solutions containing inorganic electrolytes and dicarboxylic acids at 298.15K. I. The acids as non-dissociating components. J. Phys. Chem. A 2006, 110, 5692–5717. (40) Djikaev, Y. S.; Bowles, R.; Reiss, H.; Ha¨meri, K.; Laaksonen, A.; Va¨keva¨, M. Theory of Size Dependent Deliquescence of Nanoparticles: Relation to Heterogeneous Nucleation and Comparison with Experiments. J. Phys. Chem. B 2001, 105, 7708–7722.

JP101149K