Mass transfer measurements for syngas fermentation

Abstract

A general scale-up correlation is developed for the volumetric gas-liquid mass transfer coefficient (kLa) in stirred-tank reactors (STRs) for oxygen transport using results from the literature and the current study. Power data for the dispersion of air in water by a six-bladed Rushton impeller are obtained by measuring electrical power, and a gassed power correlation is proposed. A correlation for gas-liquid mass transfer based on the energy input criteria of the form kL a = cl (Pg Iv) Cl u: fit the data well, but, in general, is dependant on flow regime and the vessel size. The volumetric mass transfer coefficient (kLa) is correlated on the basis of the relative dispersion parameter (NIN co) for similar impeller hydrodynamics and operating regimes. A stirred-tank reactor scale-up technique from bench-scale (T = 0.211 m and D/T = 0.35) to industrial-scale (up to T = 2.7 m) is proposed using a normalized hydrodynamic flow regime map and shown to be useful in understanding the range of operational conditions for the successful scale-up of stirred tank reactors. The experiments show that the scale-up model can be applied for a single and or multi-gas component mass transport to liquids if the STR is operated in the same hydrodynamic regime. A method for determining dissolved carbon monoxide concentrations from artificial syngas is developed and the volumetric gas-liquid mass transfer rates are determined. The gas-liquid scale-up model is used in carbon monoxide mass transfer scale-up for syngas fermentation in a STR. Carbon monoxide mass transfer rates obtained in this work are predicted from those of air correlated from literature. A STR scale-up model is extended to other reactor types such as air lift reactors and bubble columns in mixing applications, and the corresponding gas-liquid scale-up model is suggested.

Carbon Monoxide Mass Transfer in a Syngas Mixture

“…Furthermore, at constant syngas flow rate in Figure 5, the k L a values for dissolved CO increase in an approximately linearly fashion, which agrees with observations for air and pure O 2 . 10 The higher k L a values at higher impeller speeds and gas flow rates in Figure 5 are most likely caused by a pH change due to the formation of carbonic acid (H 2 CO 3 ), involving a simultaneous chemical reaction and CO 2 mass transfer. The initial pH value for all experiments in pure tap water was ∼9.2, and by the time all liquid samples had been withdrawn, the pH had dropped to ∼5.7.…”

“…Figure 4 shows that the k L a trends are similar to those observed for air and pure O 2 for similar operating conditions. 10 The error bars represent the standard error from the mean k L a values typically obtained from three runs. Furthermore, at constant syngas flow rate in Figure 5, the k L a values for dissolved CO increase in an approximately linearly fashion, which agrees with observations for air and pure O 2 .…”

“…If CO 2 is excluded from the artificial syngas mixture, the pH remains constant throughout the experiment. 10 Thus, it can be assumed that the change in liquid acidity had an instantaneous effect on the CO solubility, resulting in greater sensitivity of the CO mass-transfer rates to increases in impeller speed and gas flow rate as shown in Figure 5. The higher impeller speeds can also cause liquid recirculation in the STR.…”

“…More detailed information of the equipment and the STR geometry is given by Kapic. 10 The k L a values for dissolved CO from syngas were determined by introducing a step change in CO concentration and then measuring the dissolved CO concentration as a function of time. The specified syngas flow rate of interest was adjusted prior to the start of the experiment.…”

“…A more detailed explanation of preparing and collecting liquid samples and measuring the dissolved CO concentration can be found elsewhere. 10,12 Assuming that the liquid phase is well mixed, that CO mass transfer from the gas phase to the liquid phase is liquid-phasecontrolled, and that the CO concentration in the liquid at the syngas-liquid interface is in equilibrium with the CO concentration in the syngas, the volumetric mass-transfer coefficient (k L a) can be determined from where C is the dissolved CO concentration in the liquid at time t and C i is the CO concentration at the syngas-liquid interface assumed to be in equilibrium with the bubble. The solution of eq 1 yields where C o is the initial dissolved concentration.…”