Controlled counter-current contacting between a gas and a liquid forms the basis for a wide range of chemical separation operations, including absorption, stripping and distillation. A radical technique based on the rotating of a spiral channel can achieve segregation of any two immiscible phases into two parallel layers, allowing for a detailed control of phase flow rates and phase layer thicknesses. This work experimentally and theoretically studied the mass transfer process of gas-liquid contacting in this novel contactor, aiming for understanding, evaluating and demonstrating its performance.
In order to establish the spiral performance, experiments were conducted over a wide range of contacting conditions. In the experiments, desorption of four different organic solutes from water was studied separately at dilute concentrations, using air as a sweeping gas. The solutes were ethanol, acetonitrile, acetone and 2-butanone (MEK). This collection of solutes at different spiral temperatures (24, 30 and 49°C) gives a range of solute equilibrium distributions (f) from 0.232 to 5.5 (the mole fraction ratio of solute in the two phases). Thus, the performance of the rotating spiral channel was explored using phase and solute systems having different equilibrium characteristics (f) and solute transferring properties. The other contacting conditions were the pressure and the rotation rate, which were fixed to 1.8 bara and 3200 rpm, respectively. For each phase and solute system (f), the amount of solute desorbed was measured over a range of phase flow rates. Interestingly, the results showed that a fixed channel design can process a variety of systems at any desired conditions, producing solute-free water when operating at the appropriate phase flow rate ratio. Furthermore, the experimental results showed a universal peak in the mass transfer coefficient at a liquid layer thickness between 80-90 µm. The peak occurred independently of the gas phase flow rate and appeared prominently for the systems with large f, where the mass transfer was much affected by the liquid phase. This finding indicates that independent adjustment of the liquid phase flow rate could determine the optimum contacting and this optimum can be tailored to occur at any desired phase flow rates ratio by changing only the gas phase flow rate. Thus, simultaneously, the optimal contactor size and solvent usage could be achieved with rotation spiral contacting.
In addition, the spiral performance, based on the extensive data of the current work, was compared to the performance for the conventional packed column, rotating packed beds and the membrane microchannel using data from the literature for these contactors. The normalised total specific throughput (molar flow rate of the treated stream divided by the contactor volume) was developed here and used as the comparison criterion. The maximum of this measure corresponds to a minimum contactor volume to achieve a given separation task. The comparison showed that the rotating spiral was able to operate in the appropriate range of phase flow rate ratios and gave the highest specific throughput. This suggests that the contactor size for this method can be many times smaller than that of the other methods considered.
A two-dimensional (2-D) computational model was adopted in this work to study the detail of the flow and species fields that determine the mass transfer process. This model is based on a novel combination of the governing equations and an existing interface model to capture accurately the Coriolis acceleration effects and phase interactions. The 2-D model effectively predicted a wide range of experimental conditions, demonstrating that Coriolis secondary motion could double the mass transfer performance. A parametric study was also conducted using the 2-D model, where desorption of acetone was taken as a reference case. The purpose of this study was to examine the role of three key parameters (rotation rate, channel aspect ratio and flow rate of both phases) that were not tested experimentally. The results demonstrated that by adjusting the rotation rate, the contacting process could be optimised. For a range between 1000 and 20,000 rpm, it was found that the rotation rate of 16,000 rpm gave a maximum mass transfer coefficient. Furthermore, the data showed that the spiral performance was enhanced considerably by changing the channel aspect ratio. Reducing the channel width from 4 mm to 1 mm increased the mass transfer coefficient by a factor of two. Finally, at a given rotation rate and channel aspect ratio, an improvement in mass transfer was observed by adjusting the flow rates of the contacting phases. Increasing the flow rate of both phases increased the mass transfer coefficient also by a factor of two.
In general, the experimental and theoretical work in this thesis demonstrates the potential of rotating spiral contacting and establish a useful foundation underpinning its future development.
