(105c) Organic Solvent Nanofiltration in Continuous Catalytic Reactions | AIChE

(105c) Organic Solvent Nanofiltration in Continuous Catalytic Reactions


Peeva, L. - Presenter, Imperial College London
da Silva Burgal, J., Imperial College London
Livingston, A. G., Imperial College London


Organic solvent nanofiltration (OSN) is an emerging technology for performing membrane separation/purification processes in organic solvents and has been successfully applied for organometallic catalyst recovery. One major advantage of OSN separation is that it does not require any phase transition or biphasic operation. Thus, the development of a continuous catalytic process combined with continuous catalyst recovery by OSN can potentially offer major economic and process efficiency advantages over the conventional batch-based system and/or biphasic operation processes.

The main hurdle towards implementing OSN in catalytic processes has been the compatibility of the existing OSN membranes (particularly the polymeric ones) with the reaction conditions. Typically the transition metal catalyzed reactions are performed at high temperatures (100 °C and above) in aggressive solvents (e.g. DMF) and at high concentrations of base/acid – quite challenging conditions for the polymeric membranes! In this work we report a continuous process for a Heck coupling reaction combined with in situ separation of the organometallic catalyst using polymeric membrane [1,2]. In contrast to the previous works, the reaction and the separation are performed simultaneously in the same vessel, both performed at 80 °C (high temperature) and in dimethylformamide (DMF) containing >0.9 mol.L-1 triethylamine (high base content). This work will also demonstrate a continuous ring closing metathesis (RCM) reaction where Ru catalyst is retained by a polymeric OSN membrane. Both processes were operated in two modes - continuous stirred tank reactor mode (CSTR) and plug flow reactor mode (PFR).


Results and Discussion

Two reactor configurations are investigated to perform the Heck coupling reaction: a continuous single stirred tank reactor/membrane separator (m-CSTR); and a plug flow reactor (PFR) followed by m-CSTR (PFR-m-CSTR). Both configurations showed stable performance for more than 250 hours at conversion rates higher than 90 %. The combined PFR-m-CSTR configuration was found to be the most promising, achieving conversions above 98 % and high catalyst turn-over numbers (TONs) of ~20,000. In addition, low contamination of the product stream (~27 mg Pd per kg of product) makes this process configuration attractive for the pharmaceutical industry.

Some potential adverse effects in the combined continuous reactor-membrane separation units will be also discussed. Apart from membrane stability under the reaction conditions, another important issue which is often neglected in the membrane studies is the effect of the membrane material on the reaction kinetics. This could possibly be a crucial factor for a continuous process and several membrane materials were investigated for their inhibitory effect on the Pd catalyst. The results showed that cross-linked polyimide reduces the reaction rate, while polybenzimidazole seems to completely inhibit the reaction (most likely via catalyst inhibition) [2]. The study also demonstrates the excellent performance of polyether ether ketone membranes as separation materials in high temperature catalytic processes.

Finally a continuous RCM reaction successfully performed in a flow-through nanofiltration membrane reactor/separator unit in CSTR and PFR modes will be demonstrated. The PFR reactor ran continuously for 32 hours showing a stable conversion of ~100% throughout the run. The permeate stream contained ~0.34 mgL-1 Ru suggesting ~0.2% catalyst losses. The plug-flow configuration seemed the most promising and further investigation should be performed in order to evaluate its through potential.


[1]   L. Peeva, J. Burgal, S. Vartak, A. G. Livingston, J. Catal. 306 (2013)190-201.

[2]   L. Peeva, J. Arbour, A.Livingston, Org. Process Res. Dev. 17 (2013) 967-975.