Flux-matching theory of particle charging

A new flux-matching theory is formulated and applied to the study of particle charging by ions. Assuming that the ion-particle interaction includes the Coulomb + polarization forces the collisionless kinetic equation is solved and the ion concentration profile in the free-molecule zone (at the distances less than the ion mean free path) is found. This profile is then matched to that derived from the solution of the diffusion equation, which describes the ion transport outside the free-molecule zone. Three matching parameters are introduced: the ion flux, the matching distance, and the ion density at the matching distance; and three conditions are formulated for fixing these parameters: (i) the constancy of the total ion flux, (ii) the continuity of the ion concentration profile, and (iii) the continuity of the derivative of the ion concentration profile. The charging efficiencies are expressed in terms of their free-molecule values, the ion diffusivity in the carrier gas, and the ion thermal velocity. This approach is applied for calculating the efficiencies of particle charging in the transition regime (the particle size is comparable to the ion mean free path and the Coulomb length). The corrections due to ion-carrier gas interaction to the particle-ion recombination rate are shown to remain finite even for very small particles, whereas in the case of particle-ion repulsion the contribution of ion-molecular collisions to the rate of particle charging is suppressed in the free-molecule regime.

Modelling binary homogeneous nucleation of water-sulfuric acid vapours: parameterisation for high temperature emissions

Particles formed in the automobile exhaust might form a significant fraction of fine particles in urban air. We have developed a model and produced parametrizations for predicting the particle formation rate at exhaust conditions. We studied the formation in the mixture of water and sulfuric acid vapors and at temperatures between 300 and 400 K. A thermodynamically consistent version of the classical binary homogeneous nucleation model was used. The needed thermodynamical input data (vapor pressures, chemical activities, surface tensions, densities) are carefully investigated and utilized in thermodynamically consistent way. The obtained nucleation rates are parametrized in order to be able to use this nucleation model in aerosol dynamic models, exhaust models, or other process models. The parametrization reduces computational time at least by a factor of 500.

Effects of fine and ultrafine particles on cardiorespiratory symptoms in elderly subjects with coronary heart disease: the ULTRA study

The ULTRA Study, a study investigating the association between fine and ultrafine particulate air pollution and cardiorespiratory health, was conducted during the winter of 1998-1999 in Amsterdam, the Netherlands; Erfurt, Germany; and Helsinki, Finland. At each study center, a panel of elderly subjects with coronary heart disease recorded cardiac and respiratory symptoms in a diary. Exposure to ambient air pollution was characterized by measuring daily mass concentrations of particles smaller than 10 micro m (PM(10)) and 2.5 micro m (PM(2.5)), number concentrations of ultrafine particles (NC(0.01-0.1)), and gases. Odds ratios for the relation of symptoms to air pollution, adjusted for time trend, respiratory infections, and meteorologic variables, were mostly homogeneous across the centers. No association was found between air pollution and chest pain. A 10- micro g/m(3) increase in PM(2.5) was positively associated with the incidence of shortness of breath (odds ratio (OR) = 1.12, 95% confidence interval (CI): 1.02, 1.24) and with avoidance of activities (OR = 1.09, 95% CI: 0.97, 1.22). NC(0.01-0.1) was only associated with the prevalence of avoidance of activities (OR = 1.10, 95% CI: 1.01, 1.19). In conclusion, PM(2.5) was associated with some cardiac symptoms in three panels of elderly subjects. PM(2.5 )was more strongly related to cardiorespiratory symptoms than ultrafine particles were.

Nucleation probability in binary heterogeneous nucleation of water-n-propanol vapor mixtures on insoluble and soluble nanoparticles

Nucleation probabilities for binary heterogeneous nucleation have been measured quantitatively. Heterogeneous nucleation of binary n-propanol-water vapor mixtures on partially soluble 6.9-nm NaCl particles and on nonsoluble oxidized and nonoxidized 8-nm Ag particles at a constant nucleation temperature of 288 K (NaCl) or 285 K (Ag) has been investigated experimentally and theoretically. An expansion chamber was applied to generate supersaturated vapor mixtures. Number concentrations of particles activated to condensational growth were determined optically for various vapor phase activities at constant temperature. Nucleation probabilities have been measured as functions of the vapor phase activities for several n-propanol-water vapor mixing ratios. Beyond certain vapor phase activities a sharp onset of heterogeneous nucleation was observed both for NaCl and for Ag particles. For insoluble particle surfaces the theoretical slopes of the nucleation probability versus vapor phase activity curves were found to be somewhat steeper as compared to the experiment. On the other hand, for soluble particles the experimental slopes were found to be in satisfactory agreement with theory. The onset of the nucleation process is defined at an activity value where 50% of the particles were activated to condensational growth. Onset activities were obtained for various mixing ratios of the binary vapor mixtures. In the case of NaCl nuclei for n-propanol rich mixtures, the heterogeneous nucleation calculations based on a thermodynamically consistent version of Fletcher theory and an experimentally determined contact angle provide a reasonable approximation of the experimental data. However, it appears that the Fletcher theory is not applicable in the region of transition from n-propanol rich to water rich mixtures. Based on the Köhler theory of activation of soluble particles, a theory was formulated accounting for the presence of two condensable vapors and limited solubility of NaCl in water-propanol liquid mixtures. This approach provides a satisfactory description of the transition from activation of soluble particles to nucleation of vapors on insoluble particles. In the case of silver particles Fletcher theory with macroscopic experimental contact angles as well as with contact angle zero results in strong deviations from experimental data. A significant improvement was achieved by introducing fit contact angles, which are substantially lower than the macroscopic experimental values and may be considered as approximate values of the microscopic contact angles.

Singular self-preserving regimes of coagulation processes

The late stages of the time evolution of disperse systems when either coagulation alone governs the temporal changes of particle mass spectra or simultaneous condensation complicates the evolution process are studied under the assumption that the condensation efficiencies and coagulation kernels are homogeneous functions of the particle masses, with gamma and lambda being their homogeneity exponents, respectively. In considering the asymptotic behavior of the particle mass distributions the renormalization-group approach is applied to three types of coagulating systems: free coagulating systems in which coagulation alone is responsible for disperse particle growth; source-enhanced coagulating systems, where an external spacially uniform source permanently adds fresh small particles, with the particle production being a power function of time; and coagulating-condensing systems in which a condensation process accompanies the coagulation growth of disperse particles. The particle mass distributions of the form N(A)(g,t)=A(t)psi(gB(t)) are shown to describe the asymptotic regimes of particle growth in all the three types of coagulating systems (g is the particle mass). The functions A(t) and B(t) are normally power functions of time whose power exponents are found for all possible regimes of coagulation and condensation as the functions of lambda and gamma. The equations for the universality function psi(x) are formulated. It is shown that in many cases psi(x) proportional, variant x(-sigma) (sigma > 1) at small x, i.e., the particle mass distributions are singular. The power exponent sigma is expressed in terms of lambda and gamma. Two exactly soluble models illustrate the general theoretical consideration.

Soluble-insoluble transition in binary heterogeneous nucleation

In this Letter, we report the first experimental study of binary heterogeneous gas-liquid nucleation. Onset activities were obtained for nucleation of various n-propanol-water vapor mixtures on 6.9 nm NaCl particles soluble in water and practically insoluble in n-propanol. The Fletcher theory of vapor nucleation on insoluble particles provided a reasonable approximation for n-propanol rich mixtures only, whereas the Köhler theory of activation of soluble particles worked only for water rich mixtures. A new theory was formulated providing a satisfactory description of the transition from activation of soluble particles to vapor nucleation on insoluble particles.

Nonsingular self-preserving regimes of coagulation-condensation process

Growth of disperse particles is considered assuming that the preexisting disperse particles coagulate and grow simultaneously by condensing a low volatile substance (vapor, in what follows) whose concentration is permanently refreshed by a spatially uniform and constant in time source. The kinetics of the condensation-coagulation process is studied under the assumption that the condensation rate and coagulation kernels are homogeneous functions of the particle masses. The power exponents characterizing these functions define the asymptotic self-preserving regimes of the particle growth. Four such regimes are detected: (i) the mass of the disperse phase consumes all vapor and grows linearly with time, while the vapor concentration grows (or even drops) with time as its power s<1; (ii) the mass of the disperse phase grows slower than a linear function of time, while the vapor concentration grows asymptotically as time; (iii) the mass of disperse phase remains finite; and (iv) both, the mass of disperse phase and the vapor concentration grow linearly with time. For all above regimes the equations are derived defining the shape of the asymptotic mass distribution. The latter is shown to depend on a combination of the particle mass and time. The theory is illustrated by two exactly soluble models, and numerical results for the condensation-coagulation growth of aerosol particles in free molecular regime.

Kinetics of nucleation controlled formation and condensational growth of disperse particles

The kinetics of nucleation controlled formation and condensational growth of disperse particles is considered under the assumptions that: (i) only a small amount of condensable substance nucleates and forms the particles that grow by condensing the rest of the substance. (ii) The condensation efficiency is a power function of the particle mass. A nontrivial perturbation theory with respect to the smallness parameter mu= (the mass of nucleated matter)/(the total mass of condensable matter) is developed allowing one to describe the source-enhanced and free (no source) condensation processes in terms of universal functions: the particle-mass spectrum and the concentration of condensable matter. The theory relies upon a scaling transformation that removes at all the smallness parameter from the evolution equations (if the nucleation rate is a power function of the concentration of condensable matter) or leaves it in the expression for the nucleation rate where this parameter defines only a concentration scale of the nucleation process (for the nucleation rates of general form). The theory is illustrated by the exact analytical solutions of the nucleation-condensation kinetic equations for three practically important cases: (i) gas-to-particle conversion in the free-molecular regime, (ii) formation and diffusion controlled condensational growth of islands on surfaces, and (iii) formation and diffusion controlled growth of disperse particles in the continuum regime. The analytical expressions for the mass spectra of growing particles are found in the case of free condensing particles. The final mass spectra in free condensing systems display rather unusual behavior: they are either singular at small particle masses or not, depending on the value of the power exponent in the mass dependence of the condensation rate.

Nucleation burst in a coagulating system

The source-enhanced formation and growth of disperse particles is considered assuming the particles born by nucleation grow then by coagulation and condensation of a low volatile vapor onto their surfaces. After formulating the basic equations governing the particle-formation-growth process a realistic process is considered: nucleation-coagulation growth of aerosol particles in a free molecular regime. The kinetics of this process is studied under the assumption that the particle mass spectrum has a log-normal form whose parameters are expressed in terms of three moments of particle mass distribution: particle number concentration, and the moments of the orders 1/3 and 2/3. These three moments together with condensable vapor concentration are shown to meet a set of four first-order nonlinear differential equations that contain a small parameter: relative vapor concentration spent to the disperse particle production. This parameter, however, does not permit a direct application of the perturbation theory: only after two consequent rescalings it becomes possible to remove the small parameter and describe the particle-formation-growth process in terms of universal functions, depending on a specially defined nondimensional group playing the role of time. It is shown that the particle-formation-growth process can be naturally separated into two stages: (i) formation by nucleation and condensational growth of particles, and (ii) growth of formed particles by coagulation and condensation. Each stage is described by its own set of universal functions which are found from the solution of respective differential equations. The asymptotic stage of the process is shown to be described by a self-preserving distribution depending only on two moments: particle number concentration and the moment of particle-mass distribution of the order of 2/3.

Effect of Acids on Water Vapor Uptake by Pyrogenic Silica

Effect of gaseous HCl and HNO3 on the water vapor uptake by pyrogenic silica was studied at different relative humidities (RH) for pure water and different compositions of binary and ternary vapor mixtures. Experiments showed that the ability of silica to uptake water strongly depends on RH and on the type of acids and their concentration in the vapor mixtures. At low acid concentration in the binary mixtures the influence of acids is probably small. Water uptake by silica does not change monotonically with acid concentration: at first it decreases and then starts to grow. However, the presence of acids promotes water uptake, and the effect is very significant at low RH. HCl seems to be more effective acid to enhance water uptake than HNO3 . In the case of ternary mixtures the adsorbed weight of water is a bit larger than that adsorbed from the binary mixtures. Acids are accumulated by silica surface, and the accumulation is larger for nitric acid.

NMR Study of Phase Transitions in Pure Water and Binary H(2)O/HNO(3) Films Adsorbed on Surface of Pyrogenic Silica

Pyrogenic silica (aerosil) was employed as host within which the phase transitions in the adsorbed pure water and binary H(2)O/HNO(3) films have been studied with NMR spectroscopy. The median freezing temperature and freezing temperature region were shown to be highly sensitive both to the average thickness of the adsorbed films and to the amount of adsorbed nitric acid. The molar concentration of nitric acid in the adsorbed films was found to be very small, on the order of 10(-3)-10(-2) (M/liter). The concentration was found to be greater in the layers adjacent to the surface of silica and sharply decreases with distance from the surface. The difference between the median freezing temperatures for adsorbed pure water and for the binary system was found to be about 9 K for films of equal thickness. This is about 150 times greater than the difference between the freezing temperatures of bulk pure water and a solution with the same concentration of nitric acid.