“Ice Formation Processes in Upper Tropospheric Conditions”*
 

*Sponsored by the National Science Foundation (NSF-ATM9632917 and NSF-ATM0071321). Much of the material on this page has been reproduced from annual and final reports on these grants. Additionally, we fold in some results from related research supported by the now defunct NASA Atmospheric Effects of Aviation Program (ice formation by carbonaceous aerosol particles). Any conclusions given here are those of the author (Paul J. DeMott).

Background

A scientifically sound assessment of the potential that global change effects may have on the earth’s climate must include information about ice formation in tropospheric clouds. Presently, cloud properties and their feedbacks provide some of the greatest uncertainties in understanding the forcing of global climate and predictions of climate change. Our laboratory is involved in ongoing studies to investigate ice formation processes by aerosol particles of representative compositions in conditions typical of the cold regions of the troposphere. These studies are trying to gain information about the indirect effects of aerosols on cold clouds through homogeneous and heterogeneous ice nucleation processes. The experimental results have provided data for use in improving theoretical calculations of ice formation in cirrus clouds, parameterizing ice formation processes in numerical models of various scales, and giving inferences to ice formation in the real atmosphere.

The primary technique we have used for investigating ice formation processes is the continuous flow diffusion chamber (CFDC). This device permits the exposure of aerosol particles, at relevant atmospheric sizes, to constant conditions of temperature and relative humidity for a specified time. Ice crystals nucleating on the aerosols are separately detected. A new continuous flow ice thermal gradient diffusion (CFDC) chamber was constructed early in this proposal work. The design incorporated advances used in a current aircraft unit, described by Rogers et al. (2001), and the new CFDC was optimized for operation to temperatures as low as -65°C. Systems were also designed for monodisperse aerosol particle generation and characterization. Such studies inevitably also require information on particle water uptake and so investigations have been performed using a humidified tandem differential mobility analyzer (Brechtel and Kreidenweis, 2000) and a cloud condensation nucleus counter.

Activities

Laboratory experiments were first designed to quantify homogeneous freezing nucleation of pure sulfate aerosol particles as a function of temperature (-40 to -65°C), relative humidity (RH), and the weight percent composition of solute in particles. Sulfate particles (H2SO4-H2O-(NH4)2SO4) comprise a significant part of upper tropospheric aerosols and are often speculated to be the particles on which most ice forms in cirrus clouds. First experiments examined the freezing conditions and rates of liquid and crystalline (or partially crystalline) forms of sulfates A particle preconditioning system was constructed to control the temperature and RH exposure of aerosol particles in order to fix the phase state of aerosols just prior to entry into the CFDC. Studies were later extended to examine freezing of pure dicarboxylic acid particles and then to mixtures of these with sulfate particles. New studies are beginning to investigate ice nucleation in other relevant organic and organo-sulfate aerosol particle systems.

Investigations of heterogeneous freezing nuclei initially focused on carbonaceous soot particles in association with other work funded during the late 1990’s by the NASA (Atmospheric Effects of Aviation Program). Our laboratory experiments were the first rigorous measurements of heterogeneous ice nucleation by aerosol particles made at temperatures below -40°C. Systems were designed to aerosolize commercial carbon powders and to generate and collect aerosol particles from the high temperature combustion of jet fuel. Commercial carbon particles were treated with sulfuric acid in varying amounts. The weight percent of H2SO4 deposited on the particles was estimated based on CCN measurements (static thermal gradient diffusion chamber) of the treated particles combined with Köhler theory. Black carbon particles were generated at sizes across the aerosol accumulation mode, while the combusted jet fuel aerosol particles were size-selected to be representative of the sizes of carbon-containing particles measured in aircraft exhaust trails. Later, investigations turned to heterogeneous nucleation by mineral dust surrogates. Most recently, we have begun to explore re-dispersed soil dust particles as heterogeneous ice nuclei.

A schematic of the generalized experimental system used in the various studies is shown in Figure 1.

Findings

Homogeneous freezing of sulfate particles

Studies of homogeneous freezing of pure sulfate aerosol particles are presented in Chen et al. (2000). Binary H2SO4/H2O particles of sizes up to 0.1 mm require RH above about 90% with respect to water in order to dilute sufficiently to freeze at temperatures down to -60°C. Our experiments agreed with theoretical predictions that smaller particles freeze at the lowest temperatures and highest relative humidities. Results suggest that particles at sizes characteristic of the H2SO4 particles in commercial aircraft exhaust (<0.01mm) would not freeze until above 100% relative humidity. A best fit to the data suggests a supercooling, below the homogeneous freezing temperature of similar size pure water droplets, that scales by a factor of 1.98±0.26 times the melting point depression associated with the sulfuric acid solution droplet composition (i.e., weight % H2SO4). This result is in good agreement with studies of larger droplet freezing (Koop et al. 1998) and Fourier transform infrared spectroscopy (FTIR) studies of freezing of sulfuric acid drops in a chilled flow tube (e.g., Prenni et al. 2001a).

We have found that initially dry (NH4)2SO4 particles do not form ice until high relative humidities, approaching water saturation at upper tropospheric temperatures. Studies of aerosol response to increasing RH in these cases suggest a slower than expected dissolution and growth to equilibrium sizes at lower temperatures. For aerosol particles already in the liquid state, the freezing point depressions of the sulfate drops were below those of equivalent size pure water drops by as much as 1.75±0.35 times the melting point depression of bulk solutions. This result is reasonably consistent with flow tube studies reported by Prenni et al. (2001a) and studies of larger droplet freezing reported by Bertram et al. (2000), but is not in agreement with the behavior of sulfate particles found in some other FTIR studies (e.g., Cziczo and Abbatt, 1999) for reasons that are still not clear. Liquid ammonium bisulfate particles were found to freeze at temperatures of 1.41±0.36 times the melting point depression below the homogeneous freezing temperature of equivalent size pure water drops. The apparent preference for freezing in this state of ammoniation was not significant within measurement uncertainties.

The ice formation conditions of liquid sulfate aerosol particles were thus found to be rather similar and not distinguishable within measurement uncertainties (see Figure 2). The results imply that particle size may be a more important factor than the exact sulfate composition in determining the onset RH for cirrus formation. Koop et al. (2000) have speculated that the reason for this result is the singular dependence of homogeneous freezing on water activity. Any of the compositions investigated, can explain the higher ice supersaturations found to be required for forming some cirrus clouds (e.g., Heymsfield et al. 1998; Jensen et al. 2001). Nevertheless, homogeneous freezing of sulfates cannot explain the onset RH conditions of many other cirrus (same references). Other solute effects on freezing or heterogeneous nucleation must be invoked to explain those observations.

Ice formation in systems of pure organic particles and mixed sulfate/organic particles

Initial studies of the effects of organic aerosol constituents on upper tropospheric ice formation focused on the impacts of dicarboxylic acids, representing a range of solubility characteristics. These compounds are present as constituents in tropospheric aerosols due primarily to photo-oxidation of olefins (or alkenes) emitted from natural plant matter and released in small quantities from automobile exhaust, wood smoke,  and meat cooking. The results of studies of water uptake and ice formation (below -40°C) by pure adipic, succinic, oxalic, glutaric and malonic acid aerosol particles were described by Prenni et al. (2001b). Water uptake characteristics were determined using a humidified tandem differential mobility analyzer (HTDMA) system and cloud condensation nuclei counter (static thermal gradient diffusion chamber).  Results were consistent with the limited previously published information, but represented the first study of the complete series of such organic acids. The HTDMA results gave new inferences to how such organics affect deliquescence, dilution and evaporation of solution droplets at relative humidities below water saturation. Such information is vital for understanding the effects of pure organic aerosols or organic components on ice formation mechanisms at upper tropospheric temperatures, where concentrated solution drops will freeze homogeneously and cirrus clouds form in water subsaturated conditions. For example, adipic acid shows virtually no water uptake to high relative humidities, so its action in coordinating water molecules into ice embryos must be very different than for soluble species. Interestingly, adipic acid was the most active of the series in catalyzing ice formation at low temperatures, probably via a heterogeneous deposition nucleation mechanism. For the more soluble acids (glutaric, oxalic), comparison of observed water uptake to predictions of a thermodynamic (phase equilibria) model for organic aerosols demonstrated that most retained water more effectively than sulfates during evaporation. This indicates a potential role of organic components in keeping aerosols in a liquid phase state in the upper troposphere.

 

Homogeneous freezing of the more soluble pure dicarboxylic aerosol particles, while similar in form to results for liquid sulfate aerosol particles, was determined by Prenni et al. (2001b) to be not as effective (offset to lower temperatures and higher relative humidities) as for sulfates. More recent experiments (Brooks et al., 2004b) have now elucidated the role of aerosol phase state and size effects to liquid particle freezing in accounting for those first results. When a liquid phase state was clearly maintained for mixtures with the more soluble organic acids, no quantitative differences in freezing conditions were discerned for any mixture compared to pure ammonium sulfate (see Figure 3). Using a new theoretical/experimental method our group has now developed for inferring both water activity and surface tension dependence on composition from HTDMA and CCN activity results (Kreidenweis et al. 2004; Koehler et al. 2004), we find that the freezing conditions are generally as predicted by the water activity based homogeneous freezing theory of Koop et al. (2000). At present, we have reported these results only at the Fall 2003 AGU meeting and are preparing them for formal publication.

 

Recent studies have begun to investigate the water uptake and ice nucleating ability of other important classes of organic compounds such as humic and fulvic acids (Brooks et al. 2004a). Humic-like substances have been shown to represent a large fraction of the water-soluble organic component of atmospheric aerosols (Decesari, et al., 2001). Our preliminary studies show that the water uptake varies greatly with the source location and method of isolating the humic matrix from samples.  However, in most cases very little water is taken up, indicating that the presence of humic material in atmospheric particles will reduce the water uptake, relative to pure ammonium sulfate particles of the same size. These substances also appear to be surface-active in liquid aerosols and so may affect homogeneous freezing of aerosols quite differently than dicarboxylic acids. They may influence homogeneous freezing if it preferentially occurs near the surface of drops (e.g., Tabazadeh et al. 2002) and they appear to strongly alter surface tension (Brooks et al. 2004c) as well as water activity and thereby affect droplet composition in a manner that will alter freezing rates. Again, the combination of a means for deriving both the water activity and surface tension of organic solutions and mixtures with sulfates from HTDMA and CCN data (Kreidenweis et al., 2004; Koehler et al. 2004) with a homogeneous freezing model (e.g., Koop et al. 2000) provides a means of predicting expected homogeneous freezing effects. In the case of solutions containing humic substances, lowered liquid/air surface tensions dominate over reductions in water uptake and lead to predictions of somewhat earlier onset freezing of solutions. We hope to perform new studies to measure freezing rates of humic mixtures and compare these to predictions in a follow on program.

There is ample motivation to continue to investigate organic aerosol effects on cirrus formation and atmospheric ice formation in general. DeMott et al. (2003) describe the association of homogeneous freezing inhibition with the presence of high organic content in aerosols at a high altitude laboratory sampling site. Additionally, Cziczo et al. (2004a) now note the preferential fractionation of aerosols with higher organic content into the interstitial aerosol population within cirrus clouds. Cziczo et al. (2004b) provide additional new results for organic impacts on cirrus formation and note that the mechanism and/or interfering components have not been identified.

Heterogeneous ice nucleation by insoluble particles

Carbonaceous particles

Commercial black carbon aerosol particles at sub-micron sizes displayed ice-forming ability at upper tropospheric temperatures in the CFDC even without H2SO4 uptake, apparently reflecting the operation of a deposition/sorption ice nucleation mechanism on hydrophobic soot. Multi-layer coatings of H2SO4 on particles led to onset conditions for ice formation that could only be explained by a heterogeneous freezing process. These onset conditions trend with those inferred for continental cirrus by Heymsfield and Miloshevich (1995). Other results and implications for cirrus and contrail formation are described in DeMott et al. (1999).

CCN supersaturation spectra measurements for 0.05 mm combusted jet fuel aerosol particles suggested that these particles contained around 10% soluble mass. Other experiments indicated that the soluble component was probably acidic sulfate, although condensed organic compounds may also have been present in aerosols. Unlike the studies on the coated black carbon particles, the smaller combustion particles did not show evidence of being active heterogeneous ice nuclei in cirrus conditions (see attached Figure 2). The results were most consistent with homogeneous freezing of the liquid component of the particles, assuming this was sulfuric acid. The implication for aircraft contrail formation is that exhaust particles containing insoluble cores are probably the first to catalyze ice formation, but still require RH near water saturation to do so. These results suggest a negligible effect of exhaust aerosols on cirrus clouds in aircraft corridors in comparison to background sulfates. The potentially higher ice nucleation efficiency of larger exhaust residues that result from ice contrail persistence and processing remains to be investigated.

Surrogates for mineral dust particles

We have conducted studies of heterogeneous ice nucleation by dust-like particulates in cirrus conditions. The M.S. studies of Cassie Archuleta (2003), soon to be submitted as a refereed publication (Archuleta et al. 2004), examined heterogeneous ice nucleation by manufactured metal oxide and alumina-silicate particles in cirrus conditions. These particles were studied in their pure state and after condensing sulfuric acid coatings. Additionally, a sample of dust from China that has been used as a reference material for Asian dust, was re-dispersed and studied for heterogeneous ice nucleation in cirrus conditions. The results may be summarized as follows:

-              CFDC results at temperatures down to -60°C indicate that relatively pure mineral oxide particles larger than about 100 nm nucleate ice via a deposition or sorption nucleation process at lower relative humidity than required to homogeneously freeze sulfuric acid drops of the same size (see Figure 4). The relative efficiency of ice formation increases at larger particle sizes. Most dust aerosols of this type will reside in the natural aerosol accumulation mode between 0.1 and 1mm sizes, so when such relatively pure mineral dusts are present at cirrus levels they could control the conditions of cirrus formation.

-              Coating these same particles with sulfuric acid to support a condensation freezing mechanism did not usually result in more effective ice formation. We found that modified homogeneous freezing parameterizations may work quite well for quantifying ice formation by the coated mineral dust surrogates in numerical models.

-              Natural mineral dust particles obtained from dispersion of the reference Asian dust sample were very effective deposition/sorption ice nuclei. Particles of 200 nm size activated heterogeneous ice formation at an ice relative humidity of around 133%, irrespective of temperature (see Figure 5).

Very recently we have begun to explore ice nucleation by other natural dust samples. Kirsten Koehler (M.S. candidate) has been studying water uptake and ice nucleation by selected dust particle samples from the United States and has begun seeking samples from other major desert locales as the basis for research that will be proposed as part of the basis of a new proposal. Initial results suggest that pure mineral dusts of many types are very effective ice nuclei and possess similar size dependencies for ice nucleation as found in our studies of Asian dusts. Kirsten wishes to extend these studies into the precipitating cloud regime of supercooled temperatures in order to explore the full range of dust particle impacts on atmospheric ice formation. She solicited and received academic support for this endeavor in a successful application for a NASA Global Change Graduate Fellowship. We hope to provide additional support to this project by way of a new proposal to NSF.

Other New Directions

Biomass burning aerosol particles

Preliminary experiments were done during the no-cost extension period to investigate the ice nucleation behavior of aerosol particles produced from biomass combustion. Short burns of different types of fuels were done within a large storage vessel and aerosol particles were selected by a DMA for CFDC processing to measure ice formation at temperatures below -35˚C. Certain grasses appear to be sources of more effective heterogeneous ice nuclei than burned wood.  While the ice nucleation active fraction of the total particle number may be only 0.1% or less, the high concentrations of smoke particles lofted from fires could result in a large regional source of ice nuclei to cirrus levels. Other preliminary inferences were that increasing age of smoke and higher sample relative humidity are associated with decreased ice nucleation activity for all fuel types examined. These preliminary results were reported at the AGU Fall conference (Prenni, A.J., P.J. DeMott, and S.M. Kreidenweis, Biomass burning particles as potential ice nuclei, Eos. Trans. AGU, 84(46) Fall Meet. Suppl., Abstract A22B-1066, 2003). We plan to pursue additional studies through a new proposal.

 

Collaborative Studies

Single particle mass spectroscopy of ice nuclei

Collaboration was fostered early in this study with Dr. Daniel Murphy and colleagues of the NOAA Aeronomy Laboratory’s (Boulder, CO) Meteorological Chemistry group. This collaboration was based on the desire to be able to quantify the physical and chemical properties of individual ice nuclei in real time. The system designed for this study uses a counterflow virtual impactor to separate ice crystals from the outlet flow of the CFDC, evaporate the residues of nucleated ice crystals and send these on for analysis by single particle laser ionization mass spectrometry. Proof-of-concept laboratory studies (Cziczo et al. 2003) were done with modest use of funds from the above-noted NSF grants. Follow on proposals (Single Particle Mass Spectroscopy of Ice Nucleating Particles; Physical and Chemical Impacts on the Ice Nucleating Properties of Atmospheric Particles in Springtime) expanded on this initial research in order to make atmospheric measurements.

 

Numerical modeling of aerosol-ice cloud interactions

Results from the laboratory studies have provided guidance for sensitivity simulations done as part of the Cirrus Cloud Parcel Modeling Project of the GEWEX Cloud Systems Studies Working Group II.

 

We have also recently joined in planning efforts for a proposed NSF Science and Technology Center (Center for Multiscale Modeling of Atmospheric Processes) led by Dr. David A. Randall. This project plans to address the current inability to simulate the interactions of clouds with large-scale atmospheric circulations in climate models through a revolutionary new approach in which two-dimensional fine-grid “Cloud-System Resolving Models” (CSRMs) are embedded within the much larger grid cells of an atmospheric general circulation model (GCM) (Randall et al. 2003). In an MMF, the CSRM takes the place of the single-column “parameterizations” that are used in conventional GCMs. This approach will allow straightforward and physically realistic parameterization of aerosols and cloud microphysics, including ice nucleation processes, in global models. We have contributed to plans for using the multi-scale modeling framework to explore general circulation model predictions of clouds, precipitation and radiative forcing to included details on ice nucleation processes gained in laboratory studies.

 

REFERENCES

 

Archuleta, C.A., P.J. DeMott, and S.M. Kreidenweis, 2004: Ice nucleation by surrogates for atmospheric mineral dusts and mineral dust/sulfate particles at cirrus temperature. In preparation for submission to Atmospheric Chemistry and Physics.

Archuleta, C.M., 2003: Ice nucleation by surrogates for atmospheric mineral dust and mineral dust/sulfate particles at cirrus temperatures. M.S. Thesis, Dept. of Atmos. Sci., Colorado State University, Ft. Collins, 107 pp.

Brooks, S.D., P.J. DeMott and S.M. Kreidenweis, 2004a: Water Uptake by Particles Containing Humic Materials and Mixtures of Humic Materials with Ammonium Sulfate. Atmospheric Environment (in press).

Brooks, S.D., A.J. Prenni, K. Koehler, P.J. DeMott, and S.M. Kreidenweis, 2004b: Ice nucleation by internally mixed particles containing ammonium sulfate and dicarboxylic acids. In preparation for submission to J. Geophys. Res.

Brooks, S.D., P.J. DeMott, and S.M. Kreidenweis, 2004c: Investigation of aerosol particles containing humic materials as cloud condensation nuclei. In preparation for submission to J. Geophys. Res.

Brechtel, F., and S.M. Kreidenweis, 2000: Predicting particle critical supersaturation from hygroscopic growth measurements in the humidified TDMA. Part II. J. Atmos. Sci., 57, 1872-1887.

Bertram, A.K.,T. Koop, L.T. Molina, and M.J. Molina, 2000: Ice formation in (NH4)2SO4-H2O particles. J. Phys. Chem. A., 104, 584-588.

Chelf, J.H., and S.T. Martin, 1999: Laboratory measurements of H2O vapor pressures and equilibrium freezing temperature of aqueous NH4HSO4 solution from -30°C to 20°C, Geophys. Res. Lett., 26, 2391-2394.

Chen, Y., P.J. DeMott, S.M. Kreidenweis, D.C. Rogers and D. Eli Sherman, 2000: Ice formation by sulfate and sulfuric acid aerosol particles under uppertropospheric conditions, J. Atmos. Sci., 57, 3752-3766.

Chen, Y., S.M. Kreidenweis, L.M. McInnes, D.C. Rogers and P.J. DeMott, 1998: Single particle analyses of ice nucleating particles in the upper troposphere and lower stratosphere, Geophys. Res. Lett., 25, 1391-1394.

Clegg, S.L., P. Brimblecombe and A.S. Wexler, 1998: A thermodynamic model of the system H+-NH4+-SO42--NO3--H2O at tropospheric temperatures. J. Phys. Chem., 103, 2137-2154.

Cziczo, D.J., D.M. Murphy, P.K. Hudson, and D.S. Thomson, 2004a: Single Particle Measurements of the Chemical Composition of Cirrus Ice Residue During CRYSTAL-FACE. J. Geophys. Res. (in press).

Cziczo, D.J., P.J. DeMott, S.D. Brooks, A.J. Prenni, D.S. Thomson, S.M. Kreidenweis, and D.M. Murphy, 2004b: Observations of organic species and atmospheric ice formation. Submitted to Nature.

Cziczo, D.J., P.J. DeMott, C. Brock, P.K. Hudson, B. Jesse, S.M. Kreidenweis, A.J. Prenni, J. Schreiner, D.S. Thomson, and D.M. Murphy, 2003: A Method for Single Particle Mass Spectroscopy of Ice Nuclei. Aerosol Sci. and Technol., 37, 460-470.

Cziczo, D.J. and J.P.D. Abbatt, 1999: Deliquescence, efflorescence and supercooling of ammonium sulfate aerosols at low temperature: Implications for cirrus cloud formation and aerosol phase in the atmosphere. J. Geophys. Res., 104, 13781-13790.

DeMott, P.J D.J. Cziczo, A.J. Prenni, D.M. Murphy, S.M. Kreidenweis, D.S. Thomson, R. Borys, and D.C. Rogers, 2003: Measurements of the concentration and composition of nuclei for cirrus formation.  Proc. Natnl. Acad. Sci. , 100, No. 25, 14655-14660.

DeMott, P.J., 2002: Laboratory studies of cirrus cloud processes, Chapter 5 in Cirrus, D.K. Lynch, K. Sassen, D.O.C Starr, G. Stephens Eds., Oxford University Press, New York, pp. 102-135.

DeMott, P.J., 2001: Laboratory studies of cirrus cloud processes, In Cirrus, David K. Lynch, Ed., Oxford University Press, London (in press).

DeMott, P.J., Y. Chen, S.M. Kreidenweis, D.C. Rogers and D. Eli Sherman, 1999: Ice formation by black carbon particles, Geophys. Res. Lett., 26, 2429-2432.

Decesari, S., M.C. Facchini, E. Matta, F. Lettini, M. Mircea, S. Fuzzi, E. Tagliavini and J.P. Putand, 2001. Chemical features and seasonal variation of fine aerosol water-soluble organic compounds in the Po Valley, Atmos. Envron., 35, 3691-3699.

Heymsfield, A.J., and R.M. Sabin, 1989. Cirrus crystal nucleation by homogeneous freezing of solution droplets. J. Atmos. Sci., 46, 2252-2264.

Heymsfield, A.J. and L.M. Miloshevich, 1995: Relative humidity and temperature influences on cirrus formation and evolution: Observations from wave clouds and FIRE II. J. Atmos. Sci., 52, 4302-4326, 1995.

Heymsfield, A. J., L.M. Miloshevich, C. Twohy, G. Sachse and S. Oltmans, Upper tropospheric relative humidity observations and implications for cirrus ice nucleation. Geophys. Res. Lett., 25, 1343-1347, 1998.

Jensen, E.J., O.B. Toon, S.A. Vay, J. Ovarlez, R. May, T.P. Bui, C.H. Twohy, B.W. Gandrud, R.F. Pueschel, and U. Schumann, 2001: Prevalence of ice-supersaturation regions in the upper troposphere: Implications for optically thin ice cloud formation. J. Geophys. Res., 106, 17253-17266.

Koehler, K., S.M. Kreidenweis, and P. J. DeMott, 2004: Cloud condensation activity of atmospheric aerosols. Part II: Application to organic species. In preparation for submission to Atmospheric Chemistry and Physics.

Koop, T., B. Luo, A. Tsias, and T. Perter, 2000: Water activity as the determinant for homogeneous ice nucleation in aqueous solutions. Nature, 406, 611-614.

Koop, T., H.P. Ng, L.T. Molina, and M.J. Molina, 1998: A new optical technique to study aerosol phase transitions: The nucleation of ice from H2SO4 aerosols, J. Phys. Chem. A, 102, 8924-8931.

Kreidenweis, S. M., K. Koehler, and P. J. DeMott, 2004: Cloud condensation activity of atmospheric aerosols. Part I: Water activity estimates from hygroscopicity data. In preparation for submission to Atmospheric Chemistry and Physics.

Prenni, A.J., P.J. DeMott, and S.M. Kreidenweis, 2003: Water uptake of internally mixed particles containing ammonium sulfate and dicarboxylic acids. Atmos. Environ., 37, 4243-4251. Rogers, D.C., P.J. DeMott, S.M. Kreidenweis and Y. Chen, 2001: A continuous flow diffusion chamber for airborne measurements of ice nuclei, J. Atmos. Oceanic Technol. (in press).

Prenni, A.J., M.E. Wise, S.D. Brooks, and M.A. Tolbert, 2001a: Ice nucleation in sulfuric acid and ammonium sulfate particles. J. Geophys. Res., 106, 3037-3044.

Prenni, A.J., P.J. DeMott, S.M. Kreidenweis, D.E. Sherman, L.M. Russell and Y. Ming, 2001b: The effects of low molecular weight dicarboxylic acids on cloud formation. J. Phys. Chem. A., 105, 11240-11248.

Randall, D., M. Khairoutdinov, A. Arakawa, and W. Grabowski, 2003: Breaking the cloud parameterization deadlock. Bull. Amer. Meteor. Soc., 84, 1547-1564.

Tabazadeh, A., Y.S. Djikaev, and H. Reiss, 2002: Surface crystallization of supercooled water clouds. Proc. Natnl. Acad. Sci., 99, No. 25, 15873-15878.

 

FIGURES

Figure  1.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1. Schematic of particle generation and experimental systems used in this study. Solutes or agitated suspensions of sulfates, dicarboxylic acids, carbonaceous particles, nanosized-powders of metal oxides, or natural dust samples were atomized to produce particles. Carbonaceous and metal oxide particles were either sent directly to measurement instruments or first via a pathway (temperature controlled flow tube or “coater”) where H2SO4 coatings were induced. Combusted jet fuel particles, produced at high temperature with a camp stove system, were stored in a large metal bin outside the laboratory and sampled over a two-hour period. Aerosol particles are size selected with a differential mobility analyzer (DMA), measured for concentration by a condensation nucleus counter (CNC), measured for CCN activity with a thermal gradient diffusion chamber (TGDC) and thermodynamically preconditioned prior to ice nucleus measurement in a continuous flow diffusion chamber (CFDC).

Figure 2.

Figure 2. Average relative humidity (RHice: w.r.t. ice, RHw: w.r.t. water) and temperature conditions at which 1% of different aerosol particle types were observed to form ice in the CFDC residence time (~12 s). Results for homogeneous freezing of sulfates are from Chen et al. (2000). Results for coated black carbon are from DeMott et al. (1999). Results for particles from combustion of jet-A fuel are unpublished.

Figure 3

 

Figure 3 Relative humidity and temperature conditions for ice formation (1% activation in 10 s) by 100 nm (at 5% RH) malonic cid aerosols compared to the homogeneous freezing conditions of 100 nm (at 5% RH) liquid ammonium sulfate particles in the CSU CFDC instrument. Strongly drying the malonic acid particles can affect their phase state (e.g., causing crystallization or partial crystallization), leading to a slight offset in ice formation toward higher ice supersaturation as found previously for sulfate particles by Chen et al. (2000).

Figure 4

 

Figure 4. Left panel: The relative humidity and temperature conditions for ice formation on 1% of untreated hematite (iron oxide) particles of different sizes in the CFDC residence time compared to the required conditions for freezing, homogeneously, the same proportion and sizes of pure sulfuric acid aerosol particles (colored curves). Sulfuric acid freezing conditions are based on Chen et al. (2000). Right panel: Conditions required for freezing 1% of hematite particles of different sizes coated by ~4 weight % sulfuric acid as compared to the homogeneous freezing conditions of the diluted acid coating on the particles (again, based on Chen et al. 2000).

Figure 5

Figure 5. The relative humidity and temperature conditions for ice formation on 1% (during CFDC residence time) of size-selected populations of dust particles from a Chinese desert dust sample, similar in composition to airborne Asian (kosa) dust. Particles were atomized from water suspensions and strongly dried prior to ice nuclei processing in the CFDC. TEM micrographs of two particles are shown at right.