Aerosol flow in a tube furnace reactor of gas-phase synthesised silver nanoparticles

In a previous work, gas-phase synthesis of silver nanoparticles through evaporation of silver powder and subsequent particle nucleation by cooling was shown to be a viable method for achieving high purity silver nanoparticles (Backman et al. J Nanopart Res 4:325–335, 2002). In order to control the size of the produced nanoparticles, careful design of the reactor is required with respect to thermal and flow characteristics. In the present work, the silver nanoparticle reactor is rigorously simulated by means of multidimensional computational fluid and particle dynamics. The CFD-computed flow is input for a combined simulation of the vapour field and particle homogeneous nucleation, growth and coagulation. The results are compared with the experimental data and with the predictions from the usually employed simple model of an idealized plug flow reactor. The multidimensional CFD-based analysis is shown to explain and help understand different aspects of the reactor operation and size distribution of the particles produced. Yet the simple plug flow method is found to provide reasonable accuracy when an appropriate correction factor is used for the nucleation rate. Considering its robustness and computational simplicity, the plug flow method can be qualified as adequate from the engineering practical point of view for the case of silver nanoparticle reactors.     

Data evaluation of laminar flow diffusion chamber nucleation experiments with different computational methods

In order to evaluate the experimental data from laminar flow diffusion chamber (LFDC) experiments on homogeneous nucleation, an extensive postmeasurement computational analysis is required. The present work investigates the influence of the used computational methodology on the derived nucleation curves. To this end a reanalysis is made of previous LFDC experiments of 1-butanol nucleation in helium [D. Brus et al., J. Chem. Phys. 122, 214506 (2005)] using two different methods. The first method is based on single fluid heat and vapor transport in the carrier gas ignoring the aerosol processes, as commonly made in LFDC data evaluations. The second method is more comprehensive as is based on multidimensional computational fluid-particle dynamics. The calculations are made under the usual simplification of one-way coupling between fluid flow and particles, which is a valid approximation in most practical aerosols, while full aerosol dynamical effects are accommodated. Similar results were produced by the two methods. This finding corroborates the usual practice of omitting aerosol calculations in LFDC experimental data evaluation.     

2H NMR and polyelectrolyte-induced domains in lipid bilayers

2H NMR studies of polyelectrolyte-induced domain formation in lipid bilayer membranes are reviewed. The 2H NMR spectrum of choline-deuterated phosphatidylcholine (PC) reports on any and all sources of lipid bilayer surface charge, since these produce a conformation change in the choline head group of PC, manifest as a change in the 2H NMR quadrupolar splitting. In addition, homogeneous and inhomogeneous surface charge distributions are differentiated. Adding polyelectrolytes to lipid bilayers consisting of mixtures of oppositely charged and zwitterionic lipids produces 2H NMR spectra which are superpositions of two Pake sub-spectra: one corresponding to a polyelectrolyte-bound lipid population and the other to a polyelectrolyte-free lipid population. Quantitative analysis of the quadrupolar splittings and spectral intensities of the two sub-spectra indicate that the polyelectrolyte-bound populations is enriched with oppositely charged lipid, while the polyelectrolyte-free lipid population is correspondingly depleted. The same domain-segregation effect is produced whether cationic polyelectrolytes are added to anionic lipid bilayers or anionic polyelectrolytes are added to cationic lipid bilayers. The 2H NMR spectra permit a complete characterization of domain composition and size. The anion:cation ratio within the domains is always stoichiometric, as expected for a process driven by Coulombic interactions. The zwitterionic lipid content of the domains is always statistical, reflecting the systems tendency to minimize the entropic cost of demixing charged lipids into domains. Domain formation is observed even with rather short polyelectrolytes, suggesting that individual polyelectrolyte chains aggregate at the surface to form "superdomains". Overall, the polyelectrolyte bound at the lipid bilayer surface appears to lie flat along the surface and to be essentially immobilized through its multiple electrostatic contacts.