Atmospheric aerosols play a key role in affecting climate, pollution and human health by in- fluencing the properties and lifetimes of clouds and precipitation. Despite this, the molecular mechanisms involved in the nucleation of nanoscale liquid droplets are not yet fully under- stood. In this thesis, we attempt to gain a better understanding of the surface properties and small-scale thermodynamics of nanoscopic clusters of water using molecular dynam- ics simulation to complement the recent surge of scientific interest in atmospheric aerosol nucleation. Existing literature is reviewed on simulation techniques to obtain the surface tension of planar and curved interfaces, as well as studies on vapour-liquid nucleation. This is followed by several in-depth computational investigations into the surface properties of water. In the first, we apply a free-energy perturbation technique, namely the test-area (TA) method, to compute the surface tension of planar, cylindrical and spherical geometries of water. In addition, a mathematical analysis of the TA method is carried out, revealing further insight into the underlying features of perturbation methods as well as their limi- tations. The results are compared to previous literature and the differences discussed. In the second and third, the excess free energy is obtained for small clusters of water using an extension of a non-equilibrium disassembly method originally developed for studies on argon clusters. The results are used to construct free-energy curves that are subsequently fed into classical rate expressions and compared to experimental values. In the final part of the thesis, the formalism generalising the cluster disassembly method to multi-component systems is presented.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:712922 |
| Date | January 2016 |
| Creators | Lau, Gabriel |
| Contributors | Jackson, George ; Haynes, Peter ; Hunt, Patricia |
| Publisher | Imperial College London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://hdl.handle.net/10044/1/45358 |
Page generated in 0.0199 seconds