Answers are grounded in this paper and cite their sources.
We report a three-step telescoped continuous-flow synthesis of γ-aminobutyric acid (GABA) starting from ethyl 4-hydroxycrotonate (EHC), a stoichiometric byproduct generated during L-carnitine production. The route proceeds through azidation with DPPA/DBU, dual-site catalytic hydrogenation over a Pd/C fixed bed, and alkaline ester hydrolysis. All three steps are conducted in a fully integrated flow system with no intermediate isolation required. A key engineering challenge, salt-induced microchannel clogging during quenching of the azidation step, was resolved with a 22 wt % subsaturated NH4Cl solution, which simultaneously suppresses clogging and maintains stable liquid–liquid phase separation. The hydrogenation step, which must reduce both the C═C bond and the azide group in a single pass, was optimized over a five-variable parameter space using multiobjective Bayesian optimization with Gaussian process surrogate models, guided by inline FTIR spectroscopy. Starting from five Latin hypercube initialization points, the hypervolume appeared to plateau after only 21 total experiments. The three-step telescoped process delivers GABA in 75.1% overall yield. This study demonstrates how a structurally defined industrial byproduct can be valorized through a combination of inherently safe continuous-flow processing and data-driven optimization, with particular relevance to systems operating in uncharted parameter spaces under hazardous intermediate constraints.
No takes yet. Share an insight, caveat, or question.
Shen et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: