Vacuum transparent insulation for visualization of solidification phenomena A. Ashkenazi, J. Tanny Abstract Experiments on solidi®cation of aqueous solu- tions are sometimes carried out in tanks with transparent sidewalls to allow visualization of the ¯ow and crystal structure. The tank sidewalls must be thermally insulated and condensation of water vapor on the outer sidewalls must be prevented for clear visualization. In this note a system is presented which is capable of eliminating the water vapor condensation, providing thermal insulation and thus facilitating continuous ¯ow visualization of the solidi®cation process. Some visualization photographs are presented. 1 Introduction In recent years many research works were focused on the ¯ows associated with the solidi®cation processes as re- viewed by Huppert (1990). The interest in these processes has motivated extensive experimental work, most of which devoted to ¯ow visualization using various methods. The shadowgraph technique was utilized, for example, by Turner (1980), Chen and Turner (1980), Turner and Gustafson (1981) and Tanny (1995). Dye injection meth- ods were used by Thompson and Szekely (1988), Huppert (1990) and others. Chen and Chen (1991) used the com- puted tomography method to determine the porosity of the mush. Visualization of the crystal structure was carried out using direct observation by e.g. Huppert (1990) and Chen and Chen (1991). One of the technical dif®culties associated with ¯ow visualization in solidi®cation experiments is the insulation of the sidewalls. The insulation has two purposes: (i) to avoid undesired temperature gradients in the system and (ii) to prevent condensation of water vapor on the outer surface of the tank sidewalls. The latter problem arises when the outside wall temperature is lower than the am- bient dew point temperature, and it hinders clear and continuous visualization of the crystallization process and the associated ¯ows. The most simple insulation method of covering the tank sidewalls by styrofoam plates has two disadvantages. First, in a humid environment, condensation can take place even during the short time where the plates are removed for observation and photography. Secondly, the opaque in- sulation does not allow continuous visual recording of the processes. An improved insulation method is using a transparent insulation box ®lled with dry nitrogen (N 2 ) gas (C.F. Chen, private communication). This method was useful in reducing condensation during solidi®cation ex- periments. In this short communication we suggest a new method, the vacuum transparent insulation (VTI). A transparent box made of one optical glass plate and a plexiglass frame is attached to the sidewall of the test tank through which the ¯ow is visualized. The box is evacuated to simulta- neously provide thermal insulation and prevention of water vapor condensation on the tank sidewalls. This method is preferable over the nitrogen gas method since besides eliminating condensation, the evacuated gap sig- ni®cantly improves the thermal insulation. 2 Experimental setup and procedures The experiment was conducted in a square box with di- mensions of 10 cm ´ 10 cm ´ 10 cm (see Fig. 1). Two opposite sidewalls of the box were made of brass plates in which passages for temperature-controlled water were provided. The brass plates were covered with stainless steel plates to eliminate corrosion. A constant-temperature circulator provided the water for the cooled sidewall while the other was insulated. The other two sidewalls were made of optical glass to facilitate ¯ow visualization. These sidewalls were insulated by the vacuum transparent insu- lation. The bottom and top of the box were insulated by styrofoam plates. Each insulation unit consisted of an open rectangular box with an optical glass plate and a plexiglass frame (see Fig. 1). The required level of vacuum was estimated such that the dew point temperature of the air left within the evacuated box would be lower than the expected sidewall temperature. Consequently, no condensation would occur at the glass sidewall of the tank. For normal room condi- tions (25 °C and 55% RH), and for a typical sidewall temperature of )20 °C, the required pressure within the insulation unit was calculated as 6.5 kPa. The glass win- dow of the insulation box was designed so as to withstand the pressure difference imposed between the inner evac- Short Communication Experiments in Fluids 28 (2000) 482±484 Ó Springer-Verlag 2000 Received: 19 April 1999 / Accepted: 30 July 1999 A. Ashkenazi, J. Tanny (&) Center for Technological Education Holon Holon Institute of Technology Arts and Sciences P.O.B. 305, Holon 58102, Israel The authors thank Mr. Shay Meiri of Weizmann Institute of Science for his help with the vacuum system. 482