(354f) Discovering Chemical Reaction Pathways and Processes for Net-Zero Greenhouse Gas Emissions

In response to the pressing issue of embodied greenhouse gas (GHG) emissions within the chemical industry, this research underscores the necessity of discovering new pathways including unconsidered chemicals and reactions to innovatively address these environmental challenges [1]. Traditional methods, while producing realistic results, often rely on established pathways and materials, which limits their effectiveness in significantly reducing GHG emissions [2]. Our research seeks to transcend these limitations by identifying and exploring alternative pathways that mitigate GHG emissions and fundamentally transform chemical lifecycle management. Our work diverges by advocating for the exploration and discovery of alternative pathways that not only mitigate GHG emissions but also revolutionize chemical lifecycle management from a foundational chemical perspective [3]. Standard practices in constructing chemical process superstructures tend to neglect the incorporation of novel reactions and chemicals, constrained by the limitations of existing information frameworks.

Our study delves deep into the molecular and reaction level to reconceptualize the production and recycling processes. Employing a double-direction method [4], this approach integrates forward and retrosynthetic analyses to construct an extensive reaction network that covers the entire lifecycle of chemical products, including raw material procurement, manufacturing, usage, and end-of-life stages. The forward analysis collates reaction and chemical data pertinent to both general production and end-of-life pathways. Simultaneously, the retrosynthetic analysis compiles a list of precursor chemicals leading to the target product. This dual analysis facilitates the screening of chemical candidates, enabling the establishment of a circular reaction network that promotes material reuse and minimizes waste. Through retrosynthetic analysis [5], we challenge the conventional boundaries by screening for and integrating innovative chemicals and reactions. This approach not only broadens the superstructure of potential pathways but also uncovers previously unexplored avenues for sustainability.

Using chemical similarity analysis in the development of the chemical reaction network, this strategy not only ensures the identification and integration of chemical candidates that foster material reuse and minimize waste but also addresses the environmental and economic limitations inherent in traditional strategies [6]. To refine this network further, we adopt a hierarchical process design approach. This approach is structured into three distinct levels and scopes, facilitating targeted screening at each scale-up stage—from level 1 (chemistry) to level 2 (reaction), and subsequently from level 2 to level 3 (process). This configuration enables the exclusion of non-viable strategies and the accentuation of promising ones for further development. Moreover, our analysis of Technology Readiness Levels (TRLs) [7] is instrumental in identifying research gaps. By pinpointing these gaps, we set a clear direction for future efforts aimed at realizing processes that are not only cost-effective and low in GHG emissions but also boast a high degree of circularity.

To demonstrate the practical application and validate the effectiveness of our methodology, we initially apply it to a natural gas-based methanol production network. Our approach, employing a double-direction search, we search the tentative chemicals and reactions for methanol production: 453 chemicals and 527 reactions using ASKCOS [8]. Subsequent screening of these chemicals, utilizing Tanimoto similarities, narrows the field to eight chemicals. Integrating this refined chemical set, along with data collated through forward analysis, allows us to construct an extended reaction network from the conventional natural gas based methanol network. Further refinement is achieved by implementing hierarchical process design and screening across three defined scopes and scales. Our findings reveal that the introduction of additional reforming reactions to the conventional methane reforming technology, specifically steam methane reforming, when combined with strategies for recycling CO2 emitted at the end-of-life stage, markedly improves both the economic and environmental performance relative to the prevailing commercialized methanol production networks. This finding sets a foundation for applying our methodology to other chemical products, such as polyethylene terephthalate (PET), potentially offering innovative pathways to significantly reduce embodied GHG emissions.

This research contributes a novel framework to the chemical industry, advocating for a shift towards more environmentally sustainable production methods by focusing on the development of innovative pathways. By addressing the challenge of reducing embodied GHG emissions from a chemical perspective, this study illuminates the path toward a more sustainable and environmentally responsible chemical production paradigm. The novelty of our work lies in its comprehensive approach to rethinking chemical production and lifecycle management through the lens of chemistry.

References

[1] Kula, K., Klemeš, J. J., Fan, Y. V., Varbanov, P. S., Gaurav, G. K., & Jasiński, R. (2023). Environmental footprints and implications of converting GHG species to value-added chemicals: a review. Reviews in Chemical Engineering, (0).

[2] Mion, A., Galli, F., Mocellin, P., Guffanti, S., & Pauletto, G. (2022). Electrified methane reforming decarbonises methanol synthesis. Journal of CO2 Utilization, 58, 101911.

[3] Zhang, C., & Lapkin, A. A. (2023). Reinforcement learning optimization of reaction routes on the basis of large, hybrid organic chemistry–synthetic biological, reaction network data. Reaction Chemistry & Engineering, 8(10), 2491-2504.

[4] Xu, Z., & Mahadevan, R. (2022). Efficient Enumeration of Branched Novel Biochemical Pathways Using a Probabilistic Technique. Industrial & Engineering Chemistry Research, 61(25), 8645-8657.

[5] Lopez, L. M., Zhang, Q., Dollar, O., Pfaendtner, J., Shanks, B. H., & Broadbelt, L. J. (2024). Application of automated network generation for retrosynthetic planning of potential corrosion inhibitors. Molecular Systems Design & Engineering.

[6] Weber, J. M., Guo, Z., Zhang, C., Schweidtmann, A. M., & Lapkin, A. A. (2021). Chemical data intelligence for sustainable chemistry. Chemical Society Reviews, 50(21), 12013-12036.

[7] Roh, K., Bardow, A., Bongartz, D., Burre, J., Chung, W., Deutz, S., ... & Mitsos, A. (2020). Early-stage evaluation of emerging CO 2 utilization technologies at low technology readiness levels. Green chemistry, 22(12), 3842-3859.

[8] ASKCOS, https://askcos.mit.edu/ (Accessed 10March 2024)