The effect of column length on separation in thermogravitational thermal diffusion columns

In the past a number of investigators (1 to 6) have shown interest in the use of the thermal-diffusion column to make specialized separations. The customary thermaldiffusion column presents construction difficulties because the distance between the hot and cold walls must be veiy small, on the order of 0.02 in., to obtain reasonable separations. Large plate spacing may be tolerated, however, if an obstruction to flow is placed in the space between the walls. A number of investigators have experimented with baffles; Washall and Melpolder (7) recently did work on a spiral-wrapped column. The first packed column was reported in 1948 by Debye and Bueche ( 8 ) , and the only work on the effects of operating variables on separations in packed columns done to date was reported by Sullivan et al. in 1957 (9).In a previous paper ( l o ) , the transport equation for the packed thermal-diffusion column was developed by modifying the differential equation basic to the convective velocity distribution. This equation contains one term representative of the driving force for convection, namely gravity and the horizontal density gradient, one term for the transmission of drag force through the fluid by viscous shear, and one term for the drag of the packing on the fluid. The first two terms are the starting point for the column without packing.The same kind of transport equation resulted for the packed column as for the column without packing, but the coefficients H and K vary in a different manner with plate spacing, and a new variable, the packing permeability, is involved. In the work reported here these transport equations were tested in batch operation by varying these parameters. EXPERIMENTAL WORKThe thermal-diffusion column used consisted of parallelvertical plates, the working space of which was about 2 ft. high by 4 in. wide ( 1 1 ). Heat was supplied by a steam jacket ALDEN H. EMERY, JR., and MAURICE LORENZPurdue University, Lafayette, Indiana on the hot plate and removed by cold water flowing in a jacket on the cold plate. The plates included three large ports to fill and drain, two pressure taps in the center portion to measure permeabilities, five hypodermic needles for withdrawing samples, and nine thermocouples. The plates were spaced by a solid steel plate, the center of which was removed to form the working space. Plastic tape formed the gasket between the spacer and the plates.Three packing materials were used: steel wool, coarse glass wool of about 20-p diam., and fine glass wool of about 2-fi diam., the variety usually found in laboratories. The permeability of each packing could be varied by changing the density of packing. This was limited at the low-permeability end by the maximum amount that could be compressed into the working space without warping the plates and at the high-permeability end by the amount that would maintain its shape in the working space without forming channels. In addition, these three materials cover different ranges of permeability. The total permeability range covered was from 2 ...

Thermal diffusion in a packed column

EMERY¹,

Lorenz²

1963

A theoretical and experimental investigation of the effects of horizontal barriers in thermal diffusion columns

Fleming

Powers

1963

A theoretical and experimental investigation of the effects of horizontal barriers on the separation of binary mixtures attained in thermogravitational thermal diffusion columns was undertaken in an attempt to further the understanding of these effects. The presence of horizontal barriers serves to reduce the internal convective flow and t o divide tht column into a number of smaller columns with interconnecting end feeds. Equations developed from such a model serve to predict the effect of the number of barriers, temperature difference, barrier diameter, and other parameters on the steady state and transient behavior of a batch column and on the manner in which bulk flow through the column influences the steady state separation in continuous-flow columns.Data were taken in both batch and continuous columns to test the theory. Parameters varied experimentally (with an ethanol-water system) were number of barriers, (N = 0, 2, 4, 8, 16, 50).temperature difference, and diameter of the cylindrical barriers. It was found that the theoretical developments were entirely adequate to explain the observed influence o f number of barriers for both types of column operation. The slight dependence of steady state batch separation on temperature difference that was observed is consistent with data of other investigations, and the independence of this type of separation on barrier diameter is in agreement with theoretical predictions. The theoretical predictions with respect to changes in temperature difference and a semitheoretical analysis of the effect of barrier diameter making use of isothermal hydrodynamic determinations proved satisfactory in predicting the influence of changes in these two parameters on both the transient batch and steady state continuous-flow column operation.Thermal diffusion has been the subject of much theoretical and experimental investigation ever since Clusius and Dickel ( 5 ) first introduced their thermogravitational column in 1938. In contrast to an apparatus utilizing the static method of thermal diffusion where there are no convection currents, the thermogravitational column multipIies the separation by means of convection currents in a manner similar to the way a countercurrent extraction cascade produces concentration differences many times greater than the difference for a single stage. As a result, although the degree of separation obtainable in an apparatus utilizing the static method is usually small, a thermogravitational column can produce separations approaching 100% Thus, it is the thermogravitational column that has been of primary interest for the aforementioned theoretical and experimental work.Despite the amount of theoretical and experimental work which has appeared in the literature, industry has not yet applied thermal diffusion as a method for separations. Frazier (11) has developed a number of novel endfeed systems for feeding a group of thermal diffusion columns, and Frazier's co-workers have recently reported (13) versible process and requires a relatively large amount ...

Separation in a thermoelectrogravitational electrophoresis column without reservoirs: Part I. Theoretical development of the transport equation and its application to the analysis of steady state batch operation

Crosser¹,

Powers²,

Prabhudesai³

1973

The thermoelectrogravitational electrophoresis column without reservoirs has been conceived following the principle of the Clusius-Dickel thermal diffusion column. The transport equation approach in thermal diffusion developed by Jones and Furry to explain the behavior of a conventional thermogravitational thermal diffusion column has been applied to describe the electrophoretic separation in the thermoelectrogravitational electrophoresis column. Equations have been established for a system in which one single or one hypothetically single component is mobile and the other species are at their isoelectric points.Theoretical calculations for the velocity profiles, temperature distribution, and the steady state batch separation factors have been made with field strength, temperature difference, membrane spacing, and electric mobility as variables, and expected trends of the results are discussed using the steady state solution of the transport equation for batch operation. SCOPEThe objective of these studies was the experimental definition of a mathematical theory for a continuous thermoelectrogravitational liquid diffusion column without reservoirs in order to provide guidelines for the design of separation systems using these diffusional fluxes. The work is an extension of analysis for identical geometry for the thermal diffusion column without reservoirs. CONCLUSIONS AND SIGNIFICANCEExperimental results for the separation of bovine albumin from buffer showed that the & e o v gave qualitative predictions of the influence of design variables on separation and capacity consistent with previous results for liquid thermal diffusion columns of the same type. The preliminary theory suggests that moderate laboratory data will suffice for the design of larger scale continuous separations using both thermal and electro diffusion with convection.The purpose of this work was to develop suitable mathematical theory and to explore the possibility of constructing thermoelectrogravitational electrophoresis columns without reservoirs and to obtain sufficient experimental data to establish the theory. In the present paper, a basic mathematical theory for theoretical analysis of such columns is developed. Two subsequent papers will be devoted to the presentation of further theoretical development and experimental work and a discussion of the results obtained. A summary of previously reported experimental apparatus is presented in Prabhudesai ( 1965), and 0. K. Crosser is at the University of Missouri, Rolla, Missouri. J. E.