Kinetics of melt crystallization of organic eutectic forming binary mixtures in non-flow systems

https://doi.org/10.1016/S0255-2701(00)00113-6Get rights and content

Abstract

The kinetics of melt crystallization of binary eutectic forming organic mixtures in non-flow systems with cooling from below for the preferential crystallization of one component alone have been studied experimentally. The effect of subcooling, superheating and the initial concentration of the melt on the crystallization kinetics have been studied. Based on the experimental data, a correlation for the instantaneous height of the crystal (volume fraction of the solidified mass) has been proposed. The experimental results were also analyzed based on a model from the literature, which considered simultaneous heat and mass transport. The analysis shows that in the present experiments, the melt crystal interface was very close to the initial melt composition, i.e. concentration equilibrium was maintained. This is attributed to the very low cooling rates used, even though the thermal diffusivity is much greater than the mass diffusivity of the melt.

Introduction

Melt crystallization is a process for the separation of a binary mixture of organic compounds. It is the process where a molten mixture is cooled to a little below its freezing point when some of the material solidifies. In binary mixtures, that form a eutectic, this will be a pure component. The remaining melt, called the residue, will contain some of the unsolidified pure (desired) component. The purified product is recovered by separating the solid from the residue and remelting it. It is an attractive method for separation of binary organic mixtures where the boiling points of the two components are very close to each other and distillation is not so easy. This is distinct from the solid solution systems where the phase diagram is similar to the vapor liquid equilibrium diagram of ideal systems. Unlike the eutectic forming systems, the crystal phase growing in equilibrium with the liquid of a given composition is not a pure substance and hence multistage operations are required to attain high purities.

Wynn [1] has classified melt crystallization processes into two categories — suspension processes and progressive freezing. In suspension processes the crystals are suspended in the melt and the density difference between crystals and melt causes a relative counter-current motion between the two. Progressive freezing, a technique unique to melt crystallization, involves the growth of the crystal layer on a cold surface immersed in the melt. The rather limited literature on melt crystallization has been reviewed by Ulrich [2] and Rittner and Steiner [3].

The kinetics of melt crystallization (rate of crystal growth) is a basic characteristic of the process. The present study focuses on the separation of binary organic eutectic mixtures by melt crystallization in non-flow or stagnant systems. The experimental data obtained in the present study is used to examine the effect of subcooling, superheating and the initial concentration of the melt on the crystal growth rates. The experimental data was also used to develop empirical correlations for estimating the melt crystallization kinetics. The experimental results are also examined using a model from the literature based on heat and mass transfer theory.

Fig. 1 shows the phase change diagram of binary organic eutectic forming mixtures. If the initial composition of the melt is below the eutectic point, as it cools, it will reach the liquidus line when pure naphthalene (in the benzene–naphthalene system shown in Fig. 1a) will crystallize out. As further cooling proceeds the composition will move along the liquidus line, while pure naphthalene continues to crystallize out, until the eutectic point is reached. At this point no more separation is possible and the solid forming will be a mixture of the binary at the eutectic composition. Similar behavior is observed with other two systems.

Section snippets

Experimental

The experimental set-up consists of a test section and two auxiliary devices. These auxiliary devices are a circulating cooling bath with a temperature controller and a power supply unit. The test section is a vertical glass cylinder covered with a 50-mm-thick styrofoam insulation. The insulation could be easily removed to facilitate the visual observation of crystallization process as and when required. The plate at the bottom which serves as the heat transfer surface is made of brass and was

Morphological stability

The subcooling, Tl(C0)−Tw, used in the experiments was carefully controlled to ensure a flat and planar surface. The surface temperature has to be kept below the crystallization temperature of the desired component but above that of the other component and well above the eutectic temperature of the binary mixture. This helps in the preferential crystallization of one component alone without freezing of the undesired component entrapped within the crystal interstices. Furthermore, the

Transient crystal height

Using the experimental data on the kinetics of crystallization from the present study, the solid–liquid phase equilibrium data, the transport and thermodynamic properties of the binary melt, a correlation in non-dimensional form for the height of the crystallized solid (transient crystal growth) for eutectic forming binary organic mixtures for low subcoolings (also called undercooling) has been developedS=1.77(Ste.τ)0.5(1−Ste*)0.88(C′o)−0.31where, Ste, Ste* and C0′ are the Stefan number, a

Theoretical analysis of experimental data

Consider a binary melt of sub-eutectic composition at a temperature above its equilibrium temperature at that composition in a cylindrical container as used in the experimental part of this study and described earlier. The co-ordinates of the physical system are shown in Fig. 8. It is assumed that the solid/melt interface remains flat and parallel to the bottom surface. The transport of heat and solute takes place by molecular diffusion only. Heat is conducted through the liquid melt towards

Summary

The crystallization of a sub-eutectic binary mixture has been studied both experimentally and analytically. Under the conditions investigated, the systems were thermally stable and buoyancy induced motion in the melt zones was not noticed. The crystal/melt interface was flat, planar and morphologically stable. A one-dimensional moving boundary model for the solidification of pure substances from the literature was used to describe the process. In comparison to the experimental data, it was

Nomenclature

Ccomposition of the melt (mol%)
Cpspecific heat (J/(kg K))
Dmass diffusivity (m2/s)
Fparameter in Eq. (21) (−)
GrGrashof number (heat transfer), R3(TiTbl)/ν2 (−)
GrGrashof number (mass transfer), R3(xixbl)/ν2 (−)
kthermal conductivity (W/(m K))
lheight of melt (initial) (m)
PePeclet number (defined by Eq. (7)) (−)
PrPrandtl number, ν/α (−)
Rradius of the cylinder (m)
Rpseudo steady state velocity of the interface (m/s)
sheight of the interface (m)
Sdimensionless height of the interface (m)
Sc

References (13)

There are more references available in the full text version of this article.

Cited by (5)

  • A new melt crystallization process for efficient purification of chlorobromobenzene

    2020, Separation and Purification Technology
    Citation Excerpt :

    When the temperature was raised to 60 °C, p-chlorobromobenzene was completely dissolved. It has been demonstrated that the cooling rate, crystallization temperature and crystallization time affect the crystallization process and the product yield significantly [7,12,13]. Hence, the effects of the above three operation parameters on the melting crystallization process of chlorobromobenzene isomers were studied.

View full text