Computer-designed repurposing of chemical wastes into drugs.

As the chemical industry continues to produce considerable quantities of waste chemicals1,2, it is essential to devise 'circular chemistry'3-8 schemes to productively back-convert at least a portion of these unwanted materials into useful products. Despite substantial progress in the degradation of some classes of harmful chemicals9, work on 'closing the circle'-transforming waste substrates into valuable products-remains fragmented and focused on well known areas10-15. Comprehensive analyses of which valuable products are synthesizable from diverse chemical wastes are difficult because even small sets of waste substrates can, within few steps, generate millions of putative products, each synthesizable by multiple routes forming densely connected networks. Tracing all such syntheses and selecting those that also meet criteria of process and 'green' chemistries is, arguably, beyond the cognition of human chemists. Here we show how computers equipped with broad synthetic knowledge can help address this challenge. Using the forward-synthesis Allchemy platform16, we generate giant synthetic networks emanating from approximately 200 waste chemicals recycled on commercial scales, retrieve from these networks tens of thousands of routes leading to approximately 300 important drugs and agrochemicals, and algorithmically rank these syntheses according to the accepted metrics of sustainable chemistry17-19. Several of these routes we validate by experiment, including an industrially realistic demonstration on a 'pharmacy on demand' flow-chemistry platform20. Wide adoption of computerized waste-to-valuable algorithms can accelerate productive reuse of chemicals that would otherwise incur storage or disposal costs, or even pose environmental hazards.

A robotic platform for flow synthesis of organic compounds informed by AI planning.

The synthesis of complex organic molecules requires several stages, from ideation to execution, that require time and effort investment from expert chemists. Here, we report a step toward a paradigm of chemical synthesis that relieves chemists from routine tasks, combining artificial intelligence-driven synthesis planning and a robotically controlled experimental platform. Synthetic routes are proposed through generalization of millions of published chemical reactions and validated in silico to maximize their likelihood of success. Additional implementation details are determined by expert chemists and recorded in reusable recipe files, which are executed by a modular continuous-flow platform that is automatically reconfigured by a robotic arm to set up the required unit operations and carry out the reaction. This strategy for computer-augmented chemical synthesis is demonstrated for 15 drug or drug-like substances.

Decomposition of Olefin Metathesis Catalysts by Brønsted Base: Metallacyclobutane Deprotonation as a Primary Deactivating Event.

Brønsted bases of widely varying strength are shown to decompose the metathesis-active Ru intermediates formed by the second-generation Hoveyda and Grubbs catalysts. Major products, in addition to propenes, are base·HCl and olefin-bound, cyclometalated dimers [RuCl(κ2-H2IMes-H)(H2C═CHR)]2 Ru-3. These are generated in ca. 90% yield on metathesis of methyl acrylate, styrene, or ethylene in the presence of either DBU, or enolates formed by nucleophilic attack of PCy3 on methyl acrylate. They also form, in lower proportions, on metathesis in the presence of the weaker base NEt3. Labeling studies reveal that the initial site of catalyst deprotonation is not the H2IMes ligand, as the cyclometalated structure of Ru-3 might suggest, but the metallacyclobutane (MCB) ring. Computational analysis supports the unexpected acidity of the MCB protons, even for the unsubstituted ring, and by implication, its overlooked role in decomposition of Ru metathesis catalysts.

Towards More Efficient, Greener Syntheses through Flow Chemistry.

Technological advances have an important role in the design of greener synthetic processes. In this Personal Account, we describe a wide range of thermal, photochemical, catalytic, and biphasic chemical transformations examined by our group. Each of these demonstrate how the merits of a continuous flow synthesis platform can align with some of the goals put forth by the Twelve Principles of Green Chemistry. In particular, we illustrate the potential for improved reaction efficiency in terms of atom economy, product yield and reaction rates, the ability to design synthetic process with chemical and solvent waste reduction in mind as well as highlight the benefits of the real-time monitoring capabilities in flow for highly controlled synthetic output.

A General Decomposition Pathway for Phosphine-Stabilized Metathesis Catalysts: Lewis Donors Accelerate Methylidene Abstraction.

Sterically accessible Lewis donors are shown to accelerate decomposition during catalysis, for a broad range of Grubbs-class metathesis catalysts. These include benzylidene derivatives RuCl2(NHC)(PCy3)(═CHPh) (Ru-2: NHC = H2IMes, a; IMes, b; H2IPr, c; IPr, d; H2ITol, e) and indenylidene complexes RuCl2(NHC)(PCy3)(═C15H10) (NHC = H2IMes, Ru-2f; IMes, Ru-2g). All of these precatalysts form methylidene complex RuCl2(NHC)(═CH2) Ru-3 as the active species in metathesis of terminal olefins, and generate RuCl2(NHC)(PCy3)(═CH2) Ru-4 as the catalyst resting state. On treatment with a 10-fold excess of pyridine, Ru-4a and Ru-4b decomposed within minutes in solution at RT, eliminating [MePCy3]Cl A by net loss of three ligands (PCy3, methylidene, and one chloride), and a mesityl proton. In comparison, loss of A from Ru-4a in the absence of a donor requires up to 3 days at 55 °C. The σ-alkyl intermediate RuCl2(13CH2PCy3)(NHC) (py)2 resulting from nucleophilic attack of free PCy3 on the methylidene ligand was undetectable for the H2IMes system, but was spectroscopically observable for the IMes system. The relevance of this pathway to decomposition of catalysts Ru-2a-g was demonstrated by assessing the impact of pyridine on the in situ-generated methylidene species. Slow initiation (as observed for the indenylidene catalysts) did not protect against methylidene abstraction. Importantly, studies with Ru-4a and Ru-4b indicated that weaker donors (THF, MeCN, DMSO, MeOH, and even H2O) likewise promote this pathway, at rates that increase with donor concentration, and severely degrade catalyst productivity in RCM, even for a readily cyclized substrate. In all cases, A was the sole or major 31P-containing decomposition product. For DMSO, a first-order dependence of decomposition rates on DMSO concentration was established. This behavior sends a warning about the use of phosphine-stabilized metathesis catalysts in donor solvents, or with substrates bearing readily accessible donor sites. Addition of pyridine to RuCl2(H2IMes)(PCy3)(═CHMe) did not result in ethylidene abstraction, indicating that this decomposition pathway can be inhibited by use of substrates in which the olefin bears a ß-methyl group.

Olefin Metathesis at the Dawn of Implementation in Pharmaceutical and Specialty-Chemicals Manufacturing.

The recent uptake of molecular metathesis catalysts in specialty-chemicals and pharmaceutical manufacturing is reviewed.

The divergent effects of strong NHC donation in catalysis.

Strong σ-donation from NHC ligands (NHC = N-heterocyclic carbene) is shown to have profoundly conflicting consequences for the reactivity of transition-metal catalysts. Such donation is regarded as central to high catalyst activity in many contexts, of which the second-generation Grubbs metathesis catalysts (RuCl2(NHC)(PCy3)([double bond, length as m-dash]CHPh), GII) offer an early, prominent example. Less widely recognized is the dramatically inhibiting impact of NHC ligation on initiation of GII, and on re-entry into the catalytic cycle from the resting-state methylidene species RuCl2(NHC)(PCy3)([double bond, length as m-dash]CH2), GIIm. Both GII and the methylidene complexes are activated by dissociation of PCy3. The impact of NHC donicity on the rate of PCy3 loss is explored in a comparison of s-GIIm, vs. u-GIIm, in which the NHC ligand is saturated H2IMes or unsaturated IMes, respectively. PCy3 loss is nearly an order of magnitude slower for the IMes derivative (a difference that is replicated, albeit smaller, for the benzylidene precatalysts GII). Proposed as an overlooked contributor to these rate differences is an increase in the Ru-PCy3 bond strength arising from π-back-donation onto the phosphine ligand. Strong σ-donation from the IMes ligand, coupled with the inability of this unsaturated NHC to participate in significant π-backbonding, amplifies Ru → PCy3 π-back-donation. The resulting increase in Ru-P bond strength greatly inhibits entry into the active cycle. For s-GII, in contrast, the greater π-acceptor capacity of the NHC ligand enables competing Ru → H2IMes back-donation (as confirmed by NOE experiments, which reveal restricted rotation about the Ru-NHC bond for H2IMes, but not IMes). Ru → PCy3 back-donation is thus attenuated in the H2IMes complexes, accounting for the greater lability of the PCy3 ligand in s-GIIm and s-GII. Similarly inhibited initiation is predicted for other metal-NHC catalysts in which a π-acceptor ligand L must be dissociated to permit substrate binding. Conversely, enhanced reactivity can be expected where such L ligands are pure σ-donors. These effects are expected to be particularly dramatic where the NHC ligand has minimal π-acceptor capacity (as in the unsaturated Arduengo carbenes), and in geometries that maximize NHC-M-L orbital interactions.

Chemical plants: high-value molecules from essential oils.

As society faces a future of dwindling petrochemical supplies at increasing cost, much attention has been focused on methods to degrade biomass into renewable commodity-chemical building blocks. Reported here is a powerful complementary approach that amplifies the complexity of molecular structures present in plant materials. Essential-oil phenylpropenoids are transformed via acrylate cross-metathesis into potent antioxidants that are widely used in perfumery and cosmetics, and in treating disorders associated with oxidative damage.

Ethylene-promoted versus ethylene-free enyne metathesis.

The role of ethylene in promoting metathesis of acetylenic enynes is probed within the context of ring-closing enyne metathesis, using first- and second-generation Grubbs catalysts. Under inert atmosphere, rapid catalyst deactivation is observed by calibrated GC-FID analysis for substrates with minimal propargylic bulk. MALDI-TOF mass spectra reveal a Ru(enyne)(2) derivative that exhibits very low reactivity toward both enyne and ethylene. Under ethylene, formation of this species is suppressed. Enynes with bulky propargylic groups are not susceptible to this catalyst deactivation pathway, even under N(2) atmosphere.