Diffusion of Tracer in Altered Tonalite: Experiments and Simulations with Heterogeneous Distribution of Porosity

Numerical time-domain-diffusion simulations were used for studying the diffusion behavior of tracer molecules in rock matrix with homogeneous and heterogeneous porosity. As the heterogeneous sample in these simulations, a 3D tomographic image of altered tonalite was used, in which the mineral components and the pores resolved by X-ray microtomography were represented by their respective intragranular porosities determined previously by the ¹⁴C-PMMA method. The apparent diffusion coefficient of a tracer in altered tonalite was determined experimentally, and was then used in the simulations. In the altered tonalite analyzed, inclusion of heterogeneity in the porosity increased the diffusion coefficient by 16 %. Altered and pristine feldspar was the main mineral component in the sample (72 %), and it also provided the dominant contribution to tracer diffusion, explaining alone 52 % of the diffusion coefficient. The large pores resolved by microtomography (6 %) and altered and pristine mica (22 %) gave an equal contribution to the diffusion coefficient. The simulation method applied was also validated by comparing the results to both an analytical and a numerical solution to the diffusion equation in a homogenous medium. In addition, the method was compared to discrete-time random-walk simulations in the case of randomly placed overlapping spheres.     

Solutions to and Validation of Matrix-Diffusion Models

A model transport system is considered in which a pulse of tracer molecules is advected along a flow channel with porous walls. The advected tracer thus undergoes diffusion, matrix-diffusion, inside the walls, which affects the breakthrough curve of the tracer. Analytical solutions in the form of series expansions are derived for a number of situations which include a finite depth of the porous matrix, varying aperture of the flow channel, and longitudinal diffusion and Taylor dispersion of the tracer in the flow channel. Novel expansions for the Laplace transforms of the concentration in the channel facilitated closed-form expressions for the solutions. A rigorous result is also derived for the asymptotic form of the breakthrough curve for a finite depth of the porous matrix, which is very different from that for an infinite matrix. A specific experimental system was created for validation of matrix-diffusion modeling for a matrix of finite depth. A previously reported fracture-column experiment was also modeled. In both cases model solutions gave excellent fits to the measured breakthrough curves with very consistent values for the diffusion coefficients used as the fitting parameters. The matrix-diffusion modeling performed could thereby be validated.     

Kinetics of the reactions of halogenated methyl radicals (CF3, CF2Cl,CFCl2, and CCl3) with molecular chlorine

The kinetics of four gas-phase reactions involving halogenated methyl radicals (R ? CF₃, CF₂Cl, CFCI₂, and CCI₃) with molecular chlorine have been studied using a tubular reactor coupled to a photoionization mass spectrometer. The radicals were homogeneously generated by the pulsed photolysis of precursor molecules at 193 nm. The subsequent decays of the radical concentration were monitored in real-time experiments as a function of Cl₂ concentration to obtain the rate constants of these R + Cl₂ reactions. Where possible, the rate constants were measured as a function of temperature to determine Arrhenius parameters. Apparent discrepancies between these measured rate constants for CF₃ and CCl₃ with Cl₂ and ones obtained in prior indirect studies are explained. The higher activation energies for these R + Cl₂ reactions compared to that of the CH₃ + Cl₂ reaction are attributed in part to the different polarities of the transition states formed.     

High-performance anion-exchange chromatography–mass spectrometry method for determination of levoglucosan, mannosan, and galactosan in atmospheric fine particulate matter

Biomass burning has a strong influence on the atmospheric aerosol composition through particulate organic, inorganic, and soot emissions. When biomass burns, cellulose and hemicelluloses degrade, producing monosaccharide anhydrides (MAs) such as levoglucosan, mannosan, and galactosan. Therefore, these compounds have been commonly used as tracers for biomass burning. In this study, a fast water-based method was developed for the routine analysis of MAs, based on high-performance anion-exchange chromatography with electrospray ionization mass spectrometry detection. This method combines simple sample preparation, fast separation, and the advantages of the selective detection with MS. Analysis run was optimized to the maximum separation of levoglucosan, mannosan, and galactosan with 15-min analysis. The validation results indicated that the method showed good applicability for determination of MA isomer concentrations in ambient samples. The limit of detection was 100 pg for levoglucosan and 50 pg for mannosan and galactosan. Wide determination ranges enabled the analysis of samples of different concentration levels. The method showed good precision, both for standard solutions (3.9–5.9% RSD) and for fine particle samples (4.3–8.5% RSD). Co-elution of internal standard (carbon-13-labeled levoglucosan) and sugar alcohols with levoglucosan decreased the sensitivity of levoglucosan determination. The method was used to determine the MA concentrations in ambient fine particle samples from urban background (Helsinki) and rural background (Hyytiälä) in Finland. The average levoglucosan, mannosan, and galactosan concentrations were 77, 8.8, and 4.2 ng?m^?3 in Helsinki (winter 2008–2009) and 17, 2.3, and 1.4 ng?m^?3 in Hyytiälä (spring 2007), respectively. The interrelation of the three MA isomers was fairly constant in the ambient fine particle samples.     

Fouling dynamics in suspension flows

A particle suspension flowing in a channel in which fouling layers are allowed to form on the channel walls is investigated by numerical simulation. A two-dimensional phase diagram with at least four different behaviors is constructed. The fouling is modeled by attachment during collision with the deposits and by detachment caused by large enough hydrodynamic drag. For fixed total number of particles and small Reynolds numbers, the relevant parameters governing the fouling dynamics are the solid volume fraction of the suspension and the detachment drag force threshold. Below a critical curve in this 2D phase space only transient fouling takes place when the suspension is accelerated from rest by a pressure gradient. Above the fouling transition line, persistent fouling layers are formed via ballistic deposition for low and via homogeneous deposition for large solid volume fractions. Close to the fouling transition line, the flow path between the deposited layers meanders, while necking appears for increasing distance from the transition. Finally, another transition to a fully blocked flow path takes place. As determined by the estimated amount of deposited particles at saturation, both transitions seem to be discontinuous. Large fluctuations and long saturation times are typical of the dynamics of the system.     

Kinetics of the reactions of chlorinated methyl radicals (CH2Cl, CHCl2, and CCl3) with NO2 in the temperature range 220-360 K

The kinetics of the reactions of chlorinated methyl radicals (CH2Cl, CHCl2, and CCl3) with NO2 have been studied in direct measurements at temperatures between 220 and 360 K using a tubular flow reactor coupled to a photoionization mass spectrometer. The radicals have been homogeneously generated at 193 or 248 nm by pulsed laser photolysis of appropriate precursors. Decays of radical concentrations have been monitored in time-resolved measurements to obtain the reaction rate coefficients under pseudo-first-order conditions with the amount of NO2 being in large excess over radical concentrations. The bimolecular rate coefficients of all three reactions are independent of the bath gas (He or N2) and pressure within the experimental range (1-6 Torr) and are found to depend on temperature as follows: k(CH2Cl + NO2) = (2.16 +/- 0.08) x 10(-11) (T/300 K)(-1.12+/-0.24) cm3 molecule(-1) s(-1) (220-363 K), k(CHCl2 + NO2) = (8.90 +/- 0.16) x 10(-12) (T/300 K)(-1.48+/-0.13) cm3 molecule(-1) s(-1) (220-363 K), and k(CCl3 + NO2) = (3.35 +/- 0.10) x 10(-12) (T/300 K)(-2.2+/-0.4) cm3 molecule(-1) s(-1) (298-363 K), with the uncertainties given as one-standard deviations. Estimated overall uncertainties in the measured bimolecular reaction rate coefficients are about +/-25%. In the reactions CH2Cl + NO2, CHCl2 + NO2, and CCl3 + NO2, the products observed are formaldehyde, CHClO, and phosgene (CCl2O), respectively. In addition, a weak signal for the HCl formation has been detected for the CHCl2 + NO2 reaction.