Soret diffusion

Lesher C.E. (1986) Effects of silicate liquid composition on mineral-liquid element partitioning from Soret diffusion studies. /. Geophys. Res. 91, 6123-6141. 

W. Holstein, The roles of ordinary and soret diffusion in the metal-catalysed formation of filamentous carbon,/. Catal. 152, 42-51 (1995). 

In general, the diffusive mass flux is composed of diffusion due to concentration gradients (chemical potential gradients), diffusion due to thermal effects (Soret diffusion) and diffusion due to pressure and external forces. It is possible to include the full multicomponent model for concentration gradient driven diffusion (Taylor and Krishna, 1993 Bird, 1998). In most cases, in the absence of external forces, it is... 

The effect of thermal diffusion or Soret diffusion was originally predicted by Soret for liquid solutions. In 1917 the effect was also predicted for mixtures of gases by Enskog, and by Chapman and Dootson. Basically, thermal diffusion can provoke a separation of the components of a mixture under influence of a thermal gradient. There is some relationship between thermal diffusion and the movements of particles in a thermal gradient, known as thermophoresis. A particle suspended in a fluid subjected to a thermal gradient will exhibit collisions from hot molecules at one side and cold molecules from the other side. Thus there will be a net force of the... 

Rosner, D. E., Thermal (Soret) Diffusion Effects on Interfacial Mass Transport Rates, Physicochem. Thermodyn., 1, 159-185 (1980). 

This inequality in (26-11) stems from the fact that the determinant of the 2 x 2 matrix of phenomenological transport coefficients [, y] mnst be positive to ensure a positive-definite quadratic form for sq- The contribution from thermal Soret diffusion in the final expression for Ja (see equations 25-76 and 25-77) provides a definition of P in terms of the thermal diffusion coefficient kr and the temperature dependence of cpA-... 

The relevance of interphase gradients distinguishes between two different classes of problems, and this is reflected on the type of boundary condition at the pellet s surface. It is known that specifying the value of the concentration (or temperature) at the surfece (Dirichlet boundary condition) may not be realistic, and thus finite external transfer effects have to be considered (in a Robin-type boundary condition) [72]. Apart from these, a large number of additional effects have also been considered. Some examples include the nonuniformity of the porous pellet structure (distribution of pore sizes [102], bidisperse particles [103], etc.), nonuniformity of catalytic activity [104], deactivation by poisoning [105], presence of multiple reactions [106], and incorporation of additional transport mechanisms such as Soret diffusion [107] or intraparticular convection [108]. 

The coefficients, L., are characteristic of the phenomenon of thermal diffusion, i.e. the flow of matter caused by a temperature gradient. In liquids, this is called the Soret effect [12]. A reciprocal effect associated with the coefficient L. is called the Dufour effect [12] and describes heat flow caused by concentration gradients. The... 

Figure C3.1.7. Time-resolved optical absorjDtion data for the Soret band of photo lysed haemoglobin-CO showing six first-order (or pseudo-first-order) relaxation phases, I-VI, on a logaritlimic time scale extending from nanoseconds to seconds. Relaxations correspond to geminate and diffusive CO rebinding and to intramolecular relaxations of tertiary and quaternary protein stmcture. (From Goldbeck R A, Paquette S J, Bjorling S C and Kliger D S 1996 Biochemistry 35 8628-39.)...

The e.m.f. of a thermogalvanic cell is the result of four main effects (a) electrode temperature, (b) thermal liquid junction potential, (c) metallic thermocouple and (d) thermal diffusion gradient or Soret. 

The mass flux vector is also the sum of four components j (l), the mass flux due to a concentration gradient (ordinary diffusion) jYp), the mass flux associated with a gradient in the pressure (pressure diffusion) ji(F), the mass flux associated with differences in external forces (forced diffusion) and j,-(r), the mass flux due to a temperature gradient (the thermal diffusion effect or the Soret effect). The mass flux contributions may then be summarized ... 

Diffuse reflectance spectroscopy (DRS) of VO-porphyrins on reduced and sulfided catalysts exhibit shifts in the porphyrinic electronic spectra (Soret, a, (3 bands) to higher frequencies. Adsorption results in modification of the delocalized electronic resonance structure not observed on the oxide form of the catalyst. X-ray photoelectron spectroscopy reveals shifts to higher Mo binding energies on reduced and sulfided catalysts following VO-porphyrin adsorption, consistent with transfer of electrons from Mo electron donor sites to the V02+ ion. Interaction at the electron donor sites is stronger than interaction at electron acceptor sites typical of the oxide catalyst. This gives rise to the possibility of lower VO-porphyrin diffusion rates on sulfided catalysts, but this effect has not been experimentally demonstrated. 

Thermal diffusion, also known as the Ludwig-Soret effect [1, 2], is the occurrence of mass transport driven by a temperature gradient in a multicomponent system. While the effect has been known since the last century, the investigation of the Ludwig-Soret effect in polymeric systems dates back to only the middle of this century, where Debye and Bueche employed a Clusius-Dickel thermogravi-tational column for polymer fractionation [3]. Langhammer [4] and recently Ecenarro [5, 6] utilized the same experimental technique, in which separation results from the interplay between thermal diffusion and convection. This results in a rather complicated experimental situation, which has been analyzed in detail by Tyrrell [7]. 

All sample specific quantities are found within the last term, rj is the solution viscosity, D the diffusion coefficient, Ks the thermal conductivity, ST the Soret coefficient, and (dn / <)c)rp the concentration derivative of the refractive index at constant temperature and pressure. 

Since the interferometer used for (dn / dT)c>p measurement is heated completely, and not just the cuvette, it has been made out of Zerodur (Schott, Mainz), which has a negligible thermal expansion coefficient. Precise values of the refractive index increments are crucial for the determination of the thermal diffusion coefficient and the Soret coefficient. The accuracy achieved for (dn / dc)ftP is usually better than 1 %, and the accuracy of (dn / dT)rp better than 0.1 %. 

Concentration grating Due to the Ludwig-Soret effect, the temperature grating is the driving force for a secondary concentration grating, which starts to build up and is superimposed upon the thermal one. Its temporal and spatial evolution is obtained from the one-dimensional form of the extended diffusion equation... 

We have outlined how TDFRS not only provides a useful tool for the study of the Ludwig-Soret effect in multicomponent liquids, but can also contribute valuable pieces of information towards solving the puzzles encountered in polymer analysis. Though TDFRS is conceptually simple, real experiments can be rather elaborate because of the relatively low diffraction efficiencies, which require repetitive exposures and a reliable homodyne/heterodyne signal separation. As an optical scattering technique it has much in common with PCS, and the diffusion coefficients obtained in the hydrodynamic limit (q —> 0) for monodisperse solutions are indeed identical. 

Keywords Cahn-Hilliard model Diffusion Nonlinear dynamics Pattern selection Polymer blends Soret effect Spinodal decomposition Thermal diffusion... 

Since there had not been any measurements of thermal diffusion and Soret coefficients in polymer blends, the first task was the investigation of the Soret effect in the model polymer blend poly(dimethyl siloxane) (PDMS) and poly(ethyl-methyl siloxane) (PEMS). This polymer system has been chosen because of its conveniently located lower miscibility gap with a critical temperature that can easily be adjusted within the experimentally interesting range between room temperature and 100 °C by a suitable choice of the molar masses [81, 82], Furthermore, extensive characterization work has already been done for PDMS/PEMS blends, including the determination of activation energies and Flory-Huggins interaction parameters [7, 8, 83, 84],... 

The diffusion, thermal diffusion, and Soret coefficients for nine different PDMS concentrations from c = 0.09 to c = 0.9 have been measured between the binodal temperature and approximately 368 K. Figure 8 shows on the left side the diffusion and thermal diffusion coefficients. The temperature dependences of the latter are very well described as thermally activated processes according to (11) with a common activation temperature Ta = 1,395 K, which is very close to the 1,460 K obtained for the critical blend in Sect. 2. 

Fig. 8 Diffusion (D) and thermal diffusion (Dj) coefficient of PDMS/PEMS (16.4/48.1) left) and Soret coefficient right) for different PDMS mass fractions given in the legends. Binodal points mark the intersection with the binodal. The dashed line segments are extrapolations into the two-phase regime. Figures from [100], Copyright (2007) by The American Physical Society...

In contrast to the critical temperature Tc, the spinodal temperature Tsp is well below the binodal temperature for off-critical mixtures and can hardly be reached due to prior phase separation. The diffusion coefficients in the upper left part of Fig. 8 have been fitted by (23) with a fixed activation temperature determined from Dj. The binodal points in Fig. 8 mark the boundary of the homogeneous phase at the binodal. The spinodal temperatures Tsp are obtained as a fit parameter for every concentration and together define the (pseudo)spinodal line plotted in the phase diagram in Fig. 7. The Soret coefficient is obtained from (11) and (23) as... 

The diffusion, thermal diffusion, and Soret coefficients of this system are shown in Fig. 8. Samples of two different off-critical compositions (c = 0.3 and c = 0.9) were prepared. The temperature was set to a value of a few degrees above the bin-odal. Hence, the sample was entirely within the homogeneous phase and one would expect that heating could only drive the blend further into the stable one-phase region. 

A modified Cahn-Hilliard (CH) model [114] is used for the theoretical analysis of the impact of thermal diffusion on phase separation by taking into account an inhomogeneous temperature distribution, which couples to a concentration variation via the Soret effect. The Flory-Huggins model is used for the free energy of binary polymer-mixtures. The composition is naturally measured in terms of volume fraction 0 of a component A, which can be related to the weight fraction c by... 

Sj = Dj/D and D = (MkBTc b )/v are the Soret and the diffusion coefficient, respectively. In the absence of thermal diffusion, (49) reduces to the well known Cahn-Hilliard equation, which belongs to the universality class described by model B [3], In fact, (49) gives a universal description of a system in the vicinity of a critical point leading to spinodal decomposition. 

---