(65f) Development of a Cross Flow Micro Heat Exchanger: Design and Analysis

Abstract Micro heat exchangers (?ÝHEx) form an important component of the modular microplants that usually includes micromixers, microreactors, etc. Recently the advantages of miniaturization are seen to realize efficient process development and process intensification. Smaller dimensions help in achieving higher effective transport area, practically isothermal conditions in the reaction system and also suitability towards extreme conditions. Being smaller in size, these devices handle very small volumes of reacting liquids. As a result, they offer a relatively smaller time scale and hence are more suitable for the fast reactions. Being lightweight and compact, they facilitate rapid heat and mass transport, offer extremely precise control of process conditions, and deliver high performance. In comparison to the relatively a large body of literature available on the micromixers, micro reactors, very few papers deal with the micro heat exchangers. The micro heat exchangers (Alm, 2005; Schubert, 2001) are different from the conventional compact heat exchangers (Shah, 2006). In general, the micro heat exchangers are usually used for heating/cooling the reacting media by passing it directly through the heat exchanger, which is similar to the conventional heat exchanger applications in chemical plants. Although it offers extremely high heat transfer area density (> 5,000 m2/m3), due to the smaller dimensions they offer relatively high pressure drop and hence the internal design of heat exchanger and the fluid volume capacity limits the range of operational flow rates. The limited range of flow rates mainly restricts the operation to be laminar and hence structuring of the flow area becomes one of the very few ways of generating local turbulence. The most important design aspects of a ?ÝHEx are the design of flow area, distribution of fluid on the flow area and the accuracy in the connection of alternate plates that minimizes the risk of mixing of two fluids. First two of these issues raise the need of the detailed CFD simulation of the flow on a heat exchanger plate and identify better plate geometries that would help in enhancing the heat transfer coefficient (U). Although a few design configurations for the micro heat exchangers exist, here we have chosen the cross flow type of heat exchanger which is relatively simple to build and has potential to be a microreactor. A brief account of the experiments and the CFD simulations of the micro heat exchanger are given below. The performance of the in-house designed and fabricated heat exchanger was further compared with a commercially available heat exchanger for identical experimental conditions. The designed ?ÝHEx was fabricated in SS316 using precision micro-machining. The design of a single heat exchanger plate included three sections: a raised portion on the top of the plate with a recessed volume for flow of heat transfer fluid, a flat section that mainly acts as a flange for stacking of the plates and finally a recessed portion at the back of the raised portion that makes it sure that the raised portion of the next plate would fit in it. This approach helped in avoiding the mixing of fluids between the alternate plates. Typical heat exchanger plate is shown in Figure 1A. On each plate, the size of the region for fluid flow is 20mm x 20mm x 0.5mm. Thickness of a typical plate with heat transfer fluid is 2mm, while the end plates are 8mm thick to facilitate suitable projections for the standard Swagelok fittings. Experiments were carried out by passing the process fluid (40 - 80C) and the cooling fluid (26C) through the inlet ports for at different flow rates. Four thermocouples RS-32 were used for online temperature monitoring, one each at each of the inlet and outlet ports. The LMTD was estimated using suitable correction factor for cross flow and the heat transfer coefficient was obtained for different experimental conditions. The variation in the estimated heat transfer coefficient with inlet flow rate of the cold fluid for a heat exchanger with 5 plates is shown in Figure 2 and compared with the performance from a commercially available heat exchanger (IMM-CR). The range of heat transfer coefficients observed here was similar to that reported in the literature and higher than that for conventional heat exchangers. CFD simulations were carried out to study the flow pattern over the heat exchanger plates and also to identify the strategy to simulate the heat transfer in the microscale dimensions. In order to achieve the above objectives, geometry of the system and subsequently the grid was generated using GAMBIT 3.6. In order to facilitate the suitable meshing of the geometry, the entire geometry was divided in different volumes having height of 0.5mm each. The flow volume and the solid volumes were meshed appropriately with suitable grid on the common surfaces. A typical grid for the flow volume is shown in Figure 1B. The simulations were carried out using FLUENT 6.3.9. The flow was solved assuming the laminar viscosity model and the conjugate heat transfer between the liquids (cooling and process fluids on alternate plates) was solved to obtain the temperature profiles. Simulations were carried out for a range of flow rates and inlet temperatures of the two fluid streams. Pressure drop across the fluid streams and also the temperatures of the outlet streams of both the fluids was monitored. Typical simulation results in terms of the velocity distribution and the temperature distribution for a heat transfer plate are shown in Figure 3. The difference in the experimental and simulated heat transfer coefficient for the case of ?ÝHEx with just 2 plates obtained at different inlet flow rates was less than 4.2%. Further simulations for different inlet flow rates, fluid temperatures, and different fluids and with more number of plates will be carried out. References: Alm, B.; Knitter, R. and Haubelt, J. (2005) Chem. Eng. Tech., 28(12), 1554 Schubert, K. et al. (2001) Microscale Thermphysics, 5, 17 Shah, R. (2006) Heat Trans. Eng. 27(5), 3