David Topping

Hygroscopic properties of submicrometer atmospheric aerosol particles measured with H-TDMA instruments in various environments—a review

The hygroscopic properties play a vital role for the direct and indirect effects of aerosols on climate, as well as the health effects of particulate matter (PM) by modifying the deposition pattern of inhaled particles in the humid human respiratory tract. Hygroscopic Tandem Differential Mobility Analyzer (H-TDMA) instruments have been used in field campaigns in various environments globally over the last 25 yr to determine the water uptake on submicrometre particles at subsaturated conditions. These investigations have yielded valuable and comprehensive information regarding the particle hygroscopic properties of the atmospheric aerosol, including state of mixing. These properties determine the equilibrium particle size at ambient relative humidities and have successfully been used to calculate the activation of particles at water vapour supersaturation. This paper summarizes the existing published H-TDMA results on the sizeresolved submicrometre aerosol particle hygroscopic properties obtained from ground-based measurements at multiple marine, rural, urban and free tropospheric measurement sites. The data is classified into groups of hygroscopic growth indicating the external mixture, and providing clues to the sources and processes controlling the aerosol. An evaluation is given on how different chemical and physical properties affect the hygroscopic growth.

Saturation Vapor Pressures and Transition Enthalpies of Low-Volatility Organic Molecules of Atmospheric Relevance: From Dicarboxylic Acids to Complex Mixtures

There are a number of techniques that can be used that differ in terms of whether they fundamentally probe the equilibrium and the temperature range over which they can be applied. The series of homologous, straight-chain dicarboxylic acids have received much attention over the past decade given their atmospheric relevance, commercial availability, and low saturation vapor pressures, thus making them ideal test compounds. Uncertainties in the solid-state saturation vapor pressures obtained from individual methodologies are typically on the order of 50-100%, but the differences between saturation vapor pressures obtained with different methods are approximately 1-4 orders of magnitude, with the spread tending to increase as the saturation vapor pressure decreases. Some of the dicarboxylic acids can exist with multiple solid-state structures that have distinct saturation vapor pressures. Furthermore, the samples on which measurements are performed may actually exist as amorphous subcooled liquids rather than solid crystalline compounds, again with consequences for the measured saturation vapor pressures, since the subcooled liquid phase will have a higher saturation vapor pressure than the crystalline solid phase. Compounds with equilibrium vapor pressures in this range will exhibit the greatest sensitivities in terms of their gas to particle partitioning to uncertainties in their saturation vapor pressures, with consequent impacts on the ability of explicit and semiexplicit chemical models to simulate secondary organic aerosol formation.

Comparative Thermodynamic Studies of Aqueous Glutaric Acid, Ammonium Sulfate and Sodium Chloride Aerosol at High Humidity ⊥

Aerosol optical tweezers are used to simultaneously characterize and compare the hygroscopic properties of two aerosol droplets, one containing inorganic and organic solutes and the second, referred to as the control droplet, containing a single inorganic salt. The inorganic solute is either sodium chloride or ammonium sulfate and the organic component is glutaric acid. The time variation in the size of each droplet (3-7 microm in radius) is recorded with 1 s time resolution and with nanometre accuracy. The size of the control droplet is used to estimate the relative humidity with an accuracy of better than +/-0.09%. Thus, the Kohler curve of the multicomponent inorganic/organic droplet, which characterizes the variation in equilibrium droplet size with relative humidity, can be determined directly. The measurements presented here focus on high relative humidities, above 97%, in the limit of dilute solutes. The experimental data are compared with theoretical treatments that, while ignoring the interactions between the inorganic and organic components, are based upon accurate representations of the activity-concentration relationships of aqueous solutions of the individual salts. The organic component is treated by a parametrized fit to experimental data or by the UNIFAC model and the water activity of the equilibrium solution droplet is calculated using the approach suggested by Clegg, Seinfeld and Brimblecombe or the Zdanovskii-Stokes-Robinson approximation. It is shown that such an experimental strategy, comparing directly droplets of different composition, enables highly accurate measurements of the hygroscopic properties, allowing the theoretical treatments to be rigorously tested. Typical deviations of the experimental measurements from theoretical predictions are shown to be around 1% in equilibrium size, comparable to the variation between the theoretical frameworks considered.

The Kelvin versus the Raoult Term in the Kohler Equation

Abstract A comprehensive sensitivity study was carried out examining the sensitivity of hygroscopic growth and activation as modeled with the Kohler equation. Different parameters in the Kohler equation were varied within the range of their currently known uncertainties. The parameters examined include not only those describing the nature of the soluble substances in a particle/droplet and the surface tension σ of the droplet solution, but also the recently proposed representation of parameters coupling the Raoult and Kelvin terms (i.e., partitioning of solute between the surface and bulk phases, although the recently proposed adsorption to wettable but insoluble material was not considered). The examined variations cause significant changes in both hygroscopic growth and activation. Whereas the hygroscopic growth regime below 95% RH is insensitive toward the surface tension σ, σ has a large influence on the activation, increasing with decreasing particle size. This implies that a cloud condensation nucle...

Ubiquity of organic nitrates from nighttime chemistry in the European submicron aerosol

In the atmosphere night time removal of volatile organic compounds (VOC) is initiated to a large extent by reaction with the nitrate radical (NO3) forming organic nitrates which partition between gas and particulate phase. Here we show based on particle phase measurements performed at a suburban site in the Netherlands that organic nitrates contribute substantially to particulate nitrate and organic mass. Comparisons with a chemistry transport model (CTM) indicate that most of the measured particulate organic nitrates are formed by NO3 oxidation. Using aerosol composition data from three intensive observation periods at numerous measurement sites across Europe, we conclude that organic nitrates are a considerable fraction of fine particulate matter (PM1) at the continental scale. Organic nitrates represent 34% to 44% of measured submicron aerosol nitrate and are found at all urban and rural sites, implying a substantial potential of PM reduction by NOx emission control.In the atmosphere nighttime removal of volatile organic compounds is initiated to a large extent by reaction with the nitrate radical (NO3) forming organic nitrates which partition between gas and particulate phase. Here we show based on particle phase measurements performed at a suburban site in the Netherlands that organic nitrates contribute substantially to particulate nitrate and organic mass. Comparisons with a chemistry transport model indicate that most of the measured particulate organic nitrates are formed by NO3 oxidation. Using aerosol composition data from three intensive observation periods at numerous measurement sites across Europe, we conclude that organic nitrates are a considerable fraction of fine particulate matter (PM1) at the continental scale. Organic nitrates represent 34% to 44% of measured submicron aerosol nitrate and are found at all urban and rural sites, implying a substantial potential of PM reduction by NOx emission control.

Surface tension of mixed inorganic and dicarboxylic acid aqueous solutions at 298.15 K and their importance for cloud activation predictions

The following study extends previous work on modelling multicomponent surface tensions and the impact they have on aerosol and climate modelling. Mixed dicarboxylic acid and ammonium sulfate solution surface tensions have been measured experimentally and were modelled using an additive, a semi-empirical and two thermodynamic methods. A thermodynamic method with parameters fitted to binary solution data reproduced experimental results most closely with average absolute deviation from the experimental data over all measurements of 2.6%, compared with 7.7% for the additive, 8.1% for the semi-empirical and 7.3% for the other thermodynamic method. The choice of surface tension modelling method can lead to differences of up to 50% in the critical saturation ratios of aerosol particles. When applied to a trimodal aerosol distribution this leads to a difference in dry diameter for activation of 30 to 40 nm. This is greater than the difference induced by changing inorganic to organic mass ratios (>10 nm).

Direct comparison of the hygroscopic properties of ammonium sulfate and sodium chloride aerosol at relative humidities approaching saturation.

Holographic optical tweezers are used to make comparative measurements of the hygroscopic properties of single component aqueous aerosol containing sodium chloride and ammonium sulfate over a range of relative humidity from 84% to 96%. The change in RH over the course of the experiment is monitored precisely using a sodium chloride probe droplet with accuracy better than ±0.09%. The measurements are used to assess the accuracy of thermodynamic treatments of the relationship between water activity and solute mass fraction with particular attention focused on the dilute solute limit approaching saturation vapor pressure. The consistency of the frequently used Clegg-Brimblecombe-Wexler (CBW) treatment for predicting the hygroscopic properties of sodium chloride and ammonium sulfate aerosol is confirmed. Measurements of the equilibrium size of ammonium sulfate aerosol are found to agree with predictions to within an uncertainty of ±0.2%. Given the accuracy of treating equilibrium composition, the inconsistencies highlighted in recent calibration measurements of critical supersaturations of sodium chloride and ammonium sulfate aerosol cannot be attributed to uncertainties associated with the thermodynamic predictions and must have an alternative origin. It is concluded that the CBW treatment can allow the critical supersaturation to be estimated for sodium chloride and ammonium sulfate aerosol with an accuracy of better than ±0.002% in RH. This corresponds to an uncertainty of ≤1% in the critical supersaturation for typical supersaturations of 0.2% and above. This supports the view that these systems can be used to accurately calibrate instruments that measure cloud condensation nuclei concentrations at selected supersaturations. These measurements represent the first study in which the equilibrium properties of two particles of chemically distinct composition have been compared simultaneously and directly alongside each other in the same environment.

Including phase separation in a unified model to calculate partitioning of vapours to mixed inorganic–organic aerosol particles

A simple approach to calculate liquid-liquid phase separation has been developed based on the derivation of partitioning coefficients between multiple liquid phases and inclusion in a framework used to partition an arbitrary number of compounds between the vapour and particle phases in an atmospheric aerosol. The representation compares favourably with a more complex and expensive benchmark gas-particle thermodynamic model for simple well-constrained systems. The model has then been applied to consider liquid phase separation in multicomponent particles formed by the equilibration of organic products generated by a near-explicit model of VOC oxidation. Inclusion of phase separation decreases the predicted mass of condensed organic material by -10 to -50%, dependent on the concentration of semi-volatile components and ambient conditions in the model scenario. The current study considers only two liquid phases, but the framework can readily accommodate an arbitrary number, though this is beyond the scope of the current work. Uncertainty introduced by the omission of phase separation is far lower than existing uncertainties in pure component vapour pressures, where orders of magnitude differences in predicted mass are found, though the bias introduced when choosing a particular method for estimating the saturation vapour pressure will influence the magnitude of phase separation. The proposed technique is the first to be used to practically deal with the many hundreds or thousands of components present in the ambient atmospheric aerosol.

Measurements and Predictions of Binary Component Aerosol Particle Viscosity

Organic aerosol particles are known to often absorb/desorb water continuously with change in gas phase relative humidity (RH) without crystallization. Indeed, the prevalence of metastable ultraviscous liquid or amorphous phases in aerosol is well-established with solutes often far exceeding bulk phase solubility limits. Particles are expected to become increasingly viscous with drying, a consequence of the plasticizing effect of water. We report here measurements of the variation in aerosol particle viscosity with RH (equal to condensed phase water activity) for a range of organic solutes including alcohols (diols to hexols), saccharides (mono-, di-, and tri-), and carboxylic acids (di-, tri-, and mixtures). Particle viscosities are measured over a wide range (10-3 to 1010 Pa s) using aerosol optical tweezers, inferring the viscosity from the time scale for a composite particle to relax to a perfect sphere following the coalescence of two particles. Aerosol measurements compare well with bulk phase studies (well-within an order of magnitude deviation at worst) over ranges of water activity accessible to both. Predictions of pure component viscosity from group contribution approaches combined with either nonideal or ideal mixing reproduce the RH-dependent trends particularly well for the alcohol, di-, and tricarboxylic acid systems extending up to viscosities of 104 Pa s. By contrast, predictions overestimate the viscosity by many orders of magnitude for the mono-, di-, and trisaccharide systems, components for which the pure component subcooled melt viscosities are ≫1012 Pa s. When combined with a typical scheme for simulating the oxidation of α-pinene, a typical atmospheric pathway to secondary organic aerosol (SOA), these predictive tools suggest that the pure component viscosities are less than 106 Pa s for ∼97% of the 50,000 chemical products included in the scheme. These component viscosities are consistent with the conclusion that the viscosity of α-pinene SOA is most likely in the range 105 to 108 Pa s. Potential improvements to the group contribution predictive tools for pure component viscosities are considered.

An assessment of vapour pressure estimation methods

Laboratory measurements of vapour pressures for atmospherically relevant compounds were collated and used to assess the accuracy of vapour pressure estimates generated by seven estimation methods and impacts on predicted secondary organic aerosol. Of the vapour pressure estimation methods that were applicable to all the test set compounds, the Lee-Kesler [Reid et al., The Properties of Gases and Liquids, 1987] method showed the lowest mean absolute error and the Nannoolal et al. [Nannoonal et al., Fluid Phase Equilib., 2008, 269, 117-133] method showed the lowest mean bias error (when both used normal boiling points estimated using the Nannoolal et al. [Nannoolal et al., Fluid Phase Equilib., 2004, 226, 45-63] method). The effect of varying vapour pressure estimation methods on secondary organic aerosol (SOA) mass loading and composition was investigated using an absorptive partitioning equilibrium model. The Myrdal and Yalkowsky [Myrdal and Yalkowsky, Ind. Eng. Chem. Res., 1997, 36, 2494-2499] vapour pressure estimation method using the Nannoolal et al. [Nannoolal et al., Fluid Phase Equilib., 2004, 226, 45-63] normal boiling point gave the most accurate estimation of SOA loading despite not being the most accurate for vapour pressures alone.