Thermodiffusion in multicomponent mixtures. Thermodynamic, algebraic, and neuro-computing models

DOI10.1007/978-1-4614-5599-8MaRDI QIDQ2276432zbMATH OpenOpenAlexFDO

Authors Seshasai Srinivasan, M. Ziad Saghir

Publication date 5 November 2012

Published in SpringerBriefs in Applied Sciences and Technology (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/978-1-4614-5599-8

zbMATH Keywords

algebraic correlationscomputational fluid dynamicscomputational molecular dynamicsequation of statelayered neural networksliquid multilayered mixturesLudwig-Soret effectneurocomputing modelsnonequilibrium thermodynamics of fluidsthermodiffusion

Mathematics Subject Classification ID

Diffusion processes (60J60) Composite and mixture properties (74E30) Boundary-layer theory, separation and reattachment, higher-order effects (76D10) Forced convection (76R05) Diffusion (76R50) Research exposition (monographs, survey articles) pertaining to classical thermodynamics (80-02) Neural networks for/in biological studies, artificial life and related topics (92B20)

Review text

This short monograph refers to an important transport phenomenon in multicomponent mixtures, which stems from a competition between the Soret effect (thermodiffusion in the case of a nonuniform temperature distribution) and the Fickian diffusion process due to concentration gradients. Its manifestations can be observed in many processes occurring in nature, including biological applications and neural network computing analogies. The authors present three basic theoretical approaches aimed at quantifying thermodiffusion, one derives from linear response theory in nonequilibrium thermodynamics of multicomponent mixtures, the second refers to a development of various algebraic simplifications that quantify a coupling of extensive thermodynamic data to an appropriate equation of state, and, in the third, neural networks are employed to predict the thermodiffusion data based on the principles of associative thinking. The resultant estimates of enthalpy or fugacity give a quite accurate quantitative description of thermodiffusion processes in specific liquids. Layered neural network computing algorithms are advanced as reliable tools for the description of thermodiffusion in multilayered mixtures. The book is written from the physicist's point of view and its addressees are professionals in thermal engineering and material science, plus postdoctoral researchers and graduate students of physics and applied science faculties.

Recommendations

• Theoretical and experimental comparison of the Soret effect for binary mixtures of toluene and n-hexane, and benzene and n-heptane
• Modeling of thermotransport phenomenon in metal alloys using artificial neural networks
• An improved thermodiffusion model for ternary mixtures using Fujita's free volume theory
• Effects of concentration and temperature on the coupled heat and mass transport in liquid mixtures
• Hydrodynamic theory of multicomponent diffusion and thermal diffusion in multitemperature gas mixtures

Cited in

(5)

• An improved thermodiffusion model for ternary mixtures using Fujita's free volume theory
• Thermodiffusion in binary and ternary nonpolar hydrocarbon + alcohol mixtures
• Mass effect on thermodiffusion using molecular dynamics
• Theoretical prediction of thermal diffusion in water-methanol, water-ethanol, and water-isopropanol mixtures using the PC-SAFT equation of state
• Modeling of thermotransport phenomenon in metal alloys using artificial neural networks

This page was built for publication: Thermodiffusion in multicomponent mixtures. Thermodynamic, algebraic, and neuro-computing models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2276432)