Adopting Green Solvents through Predicting Reaction Outcomes with AI/Machine Learning

从传统的有机溶剂（其中许多是危险的、挥发性的或不可持续的）转换为现代的绿色溶剂是高价值化学品制造的关键可持续发展目标之一。目前，绿色溶剂的使用通常在工艺开发阶段而不是发现阶段进行探索。由于产率、选择性、杂质谱和纯化的变化，这需要重新优化工艺。这些导致更长的开发时间、成本和额外的不确定性。另一方面，及早选择合适的溶剂可以提高化学选择性，避免额外的反应步骤，并简化产物的纯化。预测这些变化是绿色溶剂在制造业中更广泛适应的重要基础能力。不幸的是，在绿色溶剂中的反应数据的缺乏是发展这种能力的关键障碍。因此，有一个迫切需要的ML模型，预测在绿色溶剂中的反应性的基础上，在传统的溶剂中的可用数据。除了解决早期工艺开发时间短的问题外，这将增加在发现阶段使用绿色溶剂的信心，支持考虑副产物、杂质和纯化方法的复杂合成路线规划工具，并作为评估有害杂质的宝贵监管工具。本项目将通过以下目标解决这些挑战：O 1通过整理具有可靠反应时间的反应数据和纳入速率定律，解决文献中反应性数据的稀缺问题。O2为绿色溶剂开发依赖于溶剂的反应性和反应选择性预测模型。O3根据工业相关反应的化学信息学分析，生产一套标准底物，并收集它们在绿色溶剂中的反应性数据。这些产出将对化学制造行业产生变革性影响，通过更短的开发时间提供快速、更可持续和更好的质量控制工艺，以及预测绿色溶剂中反应结果的置信度。该项目将在化学信息学和人工智能/机器学习领域的工业合作伙伴的支持下进行，例如Lhasa Ltd.和Molecule One。其产出将由高价值化学品制造领域的最终用户合作伙伴指导和利用：阿斯利康，CatSci和概念生命科学。

The switch from traditional organic solvents, many of which are hazardous, volatile or non-sustainable, to modern green solvents is one of the key sustainability objectives in High Value Chemical Manufacture. Currently, the use of green solvents is often explored at process development stage, instead of discovery stage. This necessitates re-optimisation of processes, due to changes in yield, selectivity, impurity profile and purification. These lead to longer development time, cost, and additional uncertainty. On the other hand, selecting the right solvent early may enhance chemoselectivity, avoid additional reaction steps, and simplify purification of the products.Predicting these changes is an important underpinning capability for wider adaptation of green solvents in manufacturing. Unfortunately, the scarcity of reaction data in green solvents is a key obstacle in developing this capability. Thus, there is an urgent need for ML models which predict reactivity in green solvents based on available data in traditional solvents. In addition to addressing the short time-scale of early-stage process development, these will increase the confidence in utilising green solvents at discovery stage, support sophisticated synthetic routes planning tools which takes into account side products, impurity and purification methods, and act as valuable regulatory tools for assessing hazardous impurities.This project will address these challenges through the following objectives: O1 Addressing the scarcity of reactivity data in the literature through curation of reaction data with reliable reaction time and inclusion of rate laws. O2 Developing solvent-dependent reactivity and reaction selectivity prediction models for green solvents.O3 Producing a set of standard substrates based on cheminformatics analysis of industrially relevant reactions and collecting their reactivity data in green solvents.These outputs will have transformative impacts in the chemical manufacture industry, delivering rapid, more sustainable and better quality-controlled processes through shorter development time, and confidence in predicting reaction outcomes in green solvents. The project will be carried out with support from industrial partners working in the field of cheminformatics and AI/Machine learning, e.g. Lhasa Ltd. and Molecule One. Its outputs will be guided and exploited by partners who are end-users in the High Value Chemical Manufacturing sectors: AstraZeneca, CatSci, and Concept Life Science.