CCER seminar

Speaker:
Thijs Vlugt (TUD, 3ME)
Personal Page: homepage.tudelft.nl/v9k6y/


Abstract:
Hydrogen is one of the most popular alternatives for energy storage. Because of its low volumetric energy density, hydrogen should be compressed for practical storage and transportation purposes. Recently, electrochemical hydrogen compressors (EHCs) have been developed that are capable of compressing hydrogen up to P = 1000 bar at much lower costs. As EHC compressed hydrogen is saturated with water, the maximum water content in gaseous hydrogen should meet the fuel requirements issued by the International Organization for Standardization (ISO) when refuelling fuel cell electric vehicles (< 5ppm). Knowledge on the vapor liquid equilibrium of H2O–H2 mixtures is crucial for designing a method to remove H2O from compressed H2. To the best of our knowledge, the only experimental high pressure data (P > 300 bar) for the H2O–H2 phase coexistence is from 1927 [J. Am. Chem. Soc., 1927, 49, 65–78]. Molecular simulation and thermodynamic modeling was used to study the phase coexistence and thermodynamic properties of the H2O–H2 system. Special simulations using so-called fractional molecules are needed for these simulations. It was found that the presence of water has a significant effect of the properties of compressed hydrogren, and that the water content is generally much larger than 5 ppm so that a drying step is needed. The electro osmotic drag of water inside the membrane of the electrochemical hydrogen compressor is also studied. At the end of my presentation, I present to recent cases of the use of Machine Learning (ML) in thermodynamics: (1) for the computation of partial molar properties from simulations and (2) the combination of thermodynamics and ML to solve arson cases.

MS Teams link to join:
teams.microsoft.com/l/meetup-join/19%3ameeting_Yzg0OWQ2YzQtYTJjMC00MzRjLTg2NzEtOWE5YTNjNjZhYTQ3%40thread.v2/0