(659c) Optimisation of Solar-Aided Hydrogen Production Process Integrated Carbon Capture Using Methyl Diethanolamine/Piperazine with Carbon Dioxide Utilisation

The continuous development of the world economy and human activities give incredible rise to the energy demand [1]. Fossil fuels, as the main energy source, its utilisation have caused a large amount of CO₂ emissions which is the main reason of global warming [2]. Thus, decarbonisation of the energy supply by replacing fossil fuel with clean and sustainable energy is crucial for the future development globally. Renewable energy would be an important part in this system. However, the main challenge for a 100% renewable energy supply system is the intermittent of solar, wind etc. It takes time for technical transition to develop a mature energy supply system capable of dealing with varying energy supply and demand [3]. Therefore, the concept of hydrogen economy has been proposed and considered as an effective solution during this transition stage [4]. There are different pathways and technologies for hydrogen production using renewable energy [5]. Among all these renewables, solar hydrogen has gained increasing attention and has been studied largely [6].

Wang et al. has carried out an optimisation study for solar steam methane reforming using molten salt (SSMR-MS) to reduce TAC and CO₂ emissions [7]. In their work, the CO₂ removal model is at constant separation efficiency which leads to inaccurate evaluation. Considering the coke formation in process, the lower bound of steam to methane ratio should also be adjusted. What’s more, carbon capture and utilisation is increasingly mentioned in the context of cost-intensive carbon capture technologies application [8]. Converting CO₂ into a wide variety of value-added chemical products becomes more and more popular. Therefore, integrated rate-based CO₂ removal model using methyl diethanolamine (MDEA) and piperazine (PZ) as different solvents in SSMR-MS along with CO₂ utilization for polypropylene carbonate and ethylene glycol (EG) production are investigated in this work. This is the main novelty of this work.

In this work, the employment of machine learning in process optimisation from [7] is extended for optimal design of SSMR-MS with integration of CO₂ capture and utilization. Surrogate model artificial neural network (ANN) has been constructed as a black box to describe the accurate function of TAC, hydrogen production rate, molten salt duty and gas flowrate from CO₂ capture unit because of its attractive computational simplicity. A rigorous rate-based CO₂ removal model for CO₂ capture is firstly developed in Aspen Plus V8.8. ANN model is employed for gas flowrate prediction in this unit and these constructed ANN surrogate models are implemented in the rigorous SSMR-MS process by using the user model in Aspen Plus through Excel Link [9]. Then, this new SSMR-MS process is used to construct a new surrogate model for optimisation. Latin hypercube sampling method is used for sample generation. For each sample generation, the new SSMR-MS process in Aspen Plus would call Visual Basic Application (VBA) in Excel to transfer data from the ANN surrogate model constructed for CO₂ removal model. A hybrid global optimisation algorithm is employed to solve the developed optimisation problem and generate the optimal design, which is then validated in Aspen Plus V8.8 and SAM. This hybrid algorithm combines the advantages of the stochastic optimisation algorithm and the deterministic optimisation method.

The computational results demonstrate that when using MDEA as solvent for CO₂ capture, an annual production of 33.59 kt EG leads to a decrease of Levelised Cost of Hydrogen Production (LCHP) from 2.40 $ kg^-1 to 0 $ kg^-1. With an annual production of 485.73 kt poly, LCHP decreases from 2.40 $ kg^-1 to 1.25 $ kg^-1. When using piperazine as solvent for CO₂ capture, the production rate for these two chemicals is similar with using MDEA as solvent. However, the process using piperazine shows great advantage for its lower capital and operating cost. The TAC of the CO₂ capture unit using piperazine is 31.62 % lower than the process using MDEA. The most striking result is that the LCHP is largely reduced compared with the conventional steam methane reforming process.

References

1. IEA, 2015.World Energy Outlook 2015, OECD.
2. Miltner, A., Wukovits, W., Pröll, T. and Friedl, A., 2010. Renewable hydrogen production: a technical evaluation based on process simulation. Journal of Cleaner Production, 18, S51-S62.
3. Gielen D, Boshell F, Saygin D, Bazilian MD, Wagner N, Gorini R. The role of renewable energy in the global energy transformation. Energy Strategy Reviews 2019;24:38-50.
4. Abe, J., Popoola, A., Ajenifuja, E. and Popoola, O., 2019. Hydrogen energy, economy and storage: Review and recommendation. International Journal of Hydrogen Energy, 44(29), 15072-15086.
5. Ishaq, H. and Dincer, I., 2021. Comparative assessment of renewable energy-based hydrogen production methods. Renewable and Sustainable Energy Reviews, 135, p.110192.
6. Koumi Ngoh, S. & Njomo, D., 2012, An overview of hydrogen gas production from solar energy, Renewable and Sustainable Energy Reviews, 16, 6782-6792.
7. Wang, W., Ma, Y., Maroufmashat, A., Zhang, N., Li, J. and Xiao, X., 2021, Optimal design of large-scale solar-aided hydrogen production process via machine learning based optimisation framework. Applied Energy, 305, 117751.
8. Mac Dowell, N., Fennell, P., Shah, N. and Maitland, G., 2017. The role of CO₂ capture and utilization in mitigating climate change. Nature Climate Change, 7(4), 243-249.
9. Fontalvo Alzate, J., 2014. Using user models in Matlab® within the Aspen Plus® interface with an Excel® link. Ingeniería e Investigación, 34, 39-43.