Analysis of multi-wavelength measurements of diffusive properties via dispersion dependence of optical properties

Do ternary liquid mixtures exhibit negative main Fick diffusion coefficients?

Measured main Fick diffusion coefficients are throughout positive. However, they may appear to be negative after transformation to the molar reference frame, if the excess volume is significant and the experimental uncertainties of the cross diffusion coefficients are large.

Interplay of structure and diffusion in ternary liquid mixtures of benzene + acetone + varying alcohols

The Fick diffusion coefficient matrix of ternary mixtures containing benzene + acetone + three different alcohols, i.e., methanol, ethanol, and 2-propanol, is studied by molecular dynamics simulation and Taylor dispersion experiments. Aiming to identify common features of these mixtures, it is found that one of the main diffusion coefficients and the smaller eigenvalue do not depend on the type of alcohol along the studied composition path. Two mechanisms that are responsible for this invariant behavior are discussed in detail, i.e., the interplay between kinetic and thermodynamic contributions to Fick diffusion coefficients and the presence of microscopic heterogeneities caused by hydrogen bonding. Experimental work alone cannot explain these mechanisms, while present simulations on the molecular level indicate structural changes and uniform intermolecular interactions between benzene and acetone molecules in the three ternary mixtures. The main diffusion coefficients of these ternary mixtures exhibit similarities with their binary subsystems. Analyses of radial distribution functions and hydrogen bonding statistics quantitatively evidence alcohol self-association and cluster formation, as well as component segregation. Furthermore, the excess volume of the mixtures is analyzed in the light of intermolecular interactions, further demonstrating the benefits of the simultaneous use of experiment and simulation. The proposed framework for studying diffusion coefficients of a set of ternary mixtures, where only one component varies, opens the way for further investigations and a better understanding of multicomponent diffusion. The presented numerical results may also give an impulse to the development of predictive approaches for multicomponent diffusion.

Soret forced Rayleigh scattering instrument for simultaneous detection of two-wavelength signals to measure Soret coefficient and thermodiffusion coefficient in ternary mixtures

We describe an instrument for the measurement of the Soret and thermodiffusion coefficients in ternary systems based on the transient holographic grating technique, which is called Soret forced Rayleigh scattering (SFRS) or thermal diffusion forced Rayleigh scattering (TDFRS). We integrated the SFRS technique and the two-wavelength detection technique, which enabled us to obtain two different signals to determine the two independent Soret coefficients and thermodiffusion coefficients in ternary systems. The instrument has been designed to read the mass transport simultaneously by two-wavelength lasers with wavelengths of λ = 403 nm and λ = 639 nm. The irradiation time of the probing lasers is controlled to reduce the effect of laser absorption to the sample with dye (quinizarin), which is added to convert the interference pattern of the heating laser of λ = 532 nm to the temperature grating. The result of the measurement of binary benchmark mixtures composed of 1,2,3,4-tetrahydronaphthalene (THN), isobutylbenzene (IBB), and n-dodecane (nC12) shows that the simultaneous two-wavelength observation of the Soret effect and the mass diffusion are adequately performed. To evaluate performance in the measurement of ternary systems, we carried out experiments on the ternary benchmark mixtures of THN/IBB/nC12 with the mass fractions of 0.800/0.100/0.100 at a temperature of 298.2 K. The Soret coefficient and thermodiffusion coefficient agreed with the ternary benchmark values within the range of the standard uncertainties (23% for the Soret coefficient of THN and 30% for the thermodiffusion coefficient of THN).