(36b) A Model-Based Approach for the Shift from Batch to Continuous Production of Succinic Acid from Glycerol

Developing a bioprocess can be a very cumbersome task, with a multitude of experiments required to affirm its feasibility. Usually, such a bioprocess is initially developed in batch mode which is sufficient most of the times.

Nevertheless, there are scenarios, where a batch bio process is not fit for purpose and continuous operation mode should be considered, e.g. to efficiently integrate the bioprocess with continuous downstream processing or within an integrated biorefinery.

The production of succinic acid form glycerol [1] fits well such a scenario.

Glycerol is the main by-product of biodiesel production, hence the biological production of succinic acid serves two purposes: a. the production of a significant added value product (succinic acid) and b. the valorisation of a biofuel byproduct, which can help to make the biofuel production more profitable [1].

In our bioprocess, we are using and adapted A. succinogenes strain, which naturally overproduces succinic acid, that it is able to grow by consuming glycerol [3]. Initially the production of succinic acid from glycerol with A. succinogenes was developed as a batch process. For this process we constructed a robust kinetic model which was used for bioprocess design and optimisation [3, 4].

Nevertheless, to improve commercialisation potential, and to be better integrated into a biodiesel biorefinery, the bioprocess should have higher productivity, than that of the batch and should be able to handle continuously and efficiently all of the glycerol substrate. To achieve this, the process should move from a batch to continuous mode of operation. This can be efficiently achieved, through the use of a predictive bioprocess model, which can compute the behaviour of the process under different conditions. As with the batch model the development of a continuous model requires experimental data, in order to calibrate and validate it. Here we used the intrinsic kinetics of an existing batch process model [4], to develop our continuous process model.

In this work, two continuous bioprocesses were developed: a. through the design of a continuous stirred tank reactor (CSTR) with cell recycling and b. through the design of a packed bed bioreactor (PBBR) It should be noted that it is the first time succinic acid is successfully produced in a continuous manner using a packed-bed bioreactor. A. succinogenes was entrapped in sodium alginate beads, which then were used as the packing material for the bioreactor. The biomass concentration inside the beads remained stable and the beads were used for prolonged fermentation time, without showing any substantial reduction in process yield. A multiscale diffusion-advection-reaction model was developed, using the batch process parameters. The model is able to compute concentration profiles of the substrate and the products inside the catalyst beads and in the bulk phase of the bioreactor.

The developed models for both processes are able to successfully simulate both the dynamic and the state behaviour of the process. The model predictions are validated with experimental data. The computed error between the predicted concentration profiles and the actual experimental results varies from 0 % to 6 % for the CSTR and 2 % to 4% for the packed bed reactor.

Both continuous processes outperform the batch process in terms of succinic acid productivity, with the PBBR being the best between them. The PBBR process is 4 times more productive than the CSTR and 6.6 time more productive than the batch process. Furthermore, the PBBR is able to achieve complete conversion of the substrate in contrast to the CSTR case.

This in conjunction with the fact that the bacterial culture is immobilised, results in an effluent consisting mostly of succinic acid, leading to reduced separation costs.

Continuous processing can make this succinic acid bioproduction easily integrable with a biodiesel biorefinery, helping to significantly improve both its economic potential and its environmental impact.

References

[1] A. Vlysidis, M. Binns, C. Webb, and C. Theodoropoulos, \A techno-economic analysis of biodiesel biore_neries: Assessment of integrated designs for the co-production of fuels and chemicals," Energy, vol. 36, 4671-4683, 2011.

[2] H. W. Tan, A. R. A. Aziz, and M. K. Aroua, Glycerol production and its applications as a raw material: A review, Renewable and Sustainable Energy Reviews, vol. 27, pp. 118-127, 2013.

[3] A. Vlysidis, M. Binns, C. Webb, and C. Theodoropoulos, Glycerol utilisation for the production of chemicals: Conversion to succinic acid, a combined experimental and computational study," Biochemical Engineering Journal, vol. 58-59, pp. 1-11, 2011.

[4] A. Rigaki, C. Webb, and C. Theodoropoulos, Double Substrate Limitation Model for the Bio-Based Production of Succinic acid from Glycerol, Biochemical Engineering Journal, p. 107391, 2019.