Noteworthy Chemistry

July 2, 2012

This NMR technique distinguishes rotamers from diastereomers. Equilibrating isomers such as rotamers can give complex 1H NMR spectra. These isomers are conventionally distinguished from diastereomers by methods such as variable-temperature NMR and switching NMR solvents. D. X. Hu, P. Grice, and S. V. Ley* at the University of Cambridge (UK) used long-overlooked chemical-exchange NMR as a tool to differentiate rotamers from diastereomers.

The authors demonstrated their technique with the HATU-assisted coupling of (R)-α-hydroxyvaline (1) and (R)-N-tert-butoxycarbonyl-N-methylvaline (2) to make depsipeptide product 3. [HATU is O-(7-azabenzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexafluorophosphate.] They observed two peaks of equal intensity at 4.59 and 4.28 ppm in the CDCl3 1H NMR spectrum of 3 and identified both as proton HA. They proposed that the two peaks could come from equilibration of 3 with rotamer 4 or epimerization to diastereomer 5.

The carbamate rotation rate would be much faster than epimerization in neutral CDCl3 at room temperature. If the two peaks were caused by rotameric isomerization, they would behave as though they were chemically exchanging on the NMR time scale. Epimerization would not exhibit this behavior.

The authors selectively irradiated the peak at 4.59 ppm in a 1-D gradient nuclear Overhauser effect (NOE) experiment. They observed a peak at 4.28 ppm in the same reverse phase as a result of saturation transfer, which implies that the protons corresponding to these two peaks undergo significant chemical exchange. To confirm this conclusion, they prepared a mixture of the diastereomers 3 and 5, each of which showed two sets of equally intense peaks for proton HA. Separate 1-D gradient NOE experiments identified 3 and its rotamer and 5 and its rotamer. (J. Org. Chem. 2012, 77, 5198–5202; Xin Su)

Using a continuous process tames the Reformatsky reaction. The Reformatsky reaction is a zinc-mediated preparation of β-hydroxyesters from α-haloesters and aldehydes. This reaction is frequently unpredictable because

  • the metal must be preactivated,
  • the danger of a runaway reaction must be minimized by using large volumes of solvent to act as a heat sink, and
  • the presence of unreacted starting material induces side reactions.

G. Loh and coauthors at the Institute of Chemical and Engineering Sciences (Singapore) and Shionogi & Co. (Iwate, Japan) developed a continuous Reformatsky process that overcomes these problems.

The authors optimized the reaction parameters under batch conditions before designing the continuous process. They used reaction and adiabatic calorimetry to obtain thermochemical information, assess hazards, and simulate potential worst-case scenarios.

The continuous process consists of three steps:

  1. In a continuously stirred tank reactor, metallic zinc is activated by diisobutylaluminum hydride (DIBAL-H) and then treated with ethyl bromoacetate to generate a bromozinc dimer.
  2. The organometallic reagent from step 1 is coupled with benzaldehyde (PhCHO) in two sequential plug-flow reactors. Using two PhCHO feeds improves heat control and avoids side reactions with unreacted PhCHO.
  3. The reaction mixture is quenched in citric acid in a static mixer and put in a holding tank. Care is taken to avoid ZnBr2 precipitation in the tubes.

The authors tested the process with various reagent concentrations to assign robustness. The reactor was designed for a product flow of 100 mL/h and can be scaled up further. The continuous process makes heat control easier than traditional batch operations, and the purity of the β-hydroxyester is higher than obtained with batch operation.

The authors believe that solvent recycling can be added to the process. Solvent recycling also can be used in other processes that involve organometallic species, such as Grignard reactions. (Org. Proc. Res. Dev. 2012, 16, 958–966; JosÉ C. Barros)

Synthesize “copolymacromers” with solid-state click reactions. Few solid-phase techniques for polymer synthesis are available, and most of them focus on preparing lower molecular weight oligomers rather than polymers. Although these investigations have produced excellent results, they are just the precursors of solid-phase polymer synthesis.

J. T. Koberstein and coauthors at Columbia University (New York) and the National Institute of Standards and Technology (Gaithersburg, MD) developed a modular technique for polymer synthesis in which a variety of preformed molecular building blocks can be linked in the solid state (much like Lego building blocks) to form a class of polymers called polymacromers.

In the authors’ method, a click reaction between the azide terminus of a macromonomer (the red chain in the figure) and an alkyne group on the functionalized substrate surface couples the macromonomer to the surface with a triazole linkage. The result is a substrate coated with a covalently bound brush of the red polymer with trimethylsilyl (TMS)–protected alkyne groups at the surface.

Removing the TMS groups with K2CO3 in CH2Cl2–MeOH regenerates an alkyne-functionalized surface, and a second macromonomer (blue chain) is attached by using another click reaction. The coupling–deprotection cycle can be applied multiple times to produce the desired copolymacromer sequence. The figure illustrates the preparation of a copolymacromer with three distinct segments. (Macromolecules 2012, 45, 3866−3873; Ben Zhong Tang)

Improve your process by understanding impurity formation. S. J. Bell and coauthors at Almac (Craigavon, Northern Ireland) and AstraZeneca R&D (Mölndal and Södertälje, Sweden) improved the synthesis of a key intermediate in the preparation of P2Y12 inhibitors. The intermediate, ethyl 5-cyano-2-methyl-6-oxo-1,6-dihydropyridine-3-carboxylate, can be synthesized by condensing DMF dimethyl acetal with ethyl acetoacetate followed (without isolation or workup) by a reaction with cyanoacetamide under basic conditions (NaOEt in EtOH). Two main impurities are formed: 6-hydroxy-5-acetyl-3-cyano-2-pyridone (A) and ethyl 5-carboxamido-2-methyl-6-oxo-1,6-dihydropyridine-3-carboxylate (B). Impurity A is formed by cyclization onto the ester group rather than onto the ketone.

The authors originally believed that impurity B forms because the presence of water leads to nitrile hydrolysis. This hypothesis, however, was not supported because impurity B increases at lower temperatures and in the presence of weaker bases. An alternative explanation involves the participation of HNMe2 (a byproduct from the DMF used in the first step) and cyclization onto the nitrile group to form a pyran, which is further transformed to pyridone amide B.

On this basis, the authors modified the reaction by using malononitrile in place of cyanoacetamide and replaced stoichiometric NaOET with catalytic NEt3. The reaction proceeded more smoothly and improved the yield of the target compound from 50% to 82%. (Org. Process Res. Dev. 2012, 16, 819–823; Will Watson)

Use redox chemistry to make a chiral catalyst a “switch-hitter”. The helical chirality of some redox-responsive coordination complexes inverts when the oxidation state of the metal ion is changed. Using this finding, S. Mortezaei, N. R. Catarineu, and J. W. Canary* at New York University developed an “ambidextrous” chiral catalyst that changes its enantioselectivity as a result of helicity inversion when the metal center is oxidized or reduced.

The authors synthesized a chiral ligand from L-methioninol and coordinated it to Cu(ClO4)2 or Cu(MeCN)4PF6 to form the “right-handed” 1 or “left-handed” 2, respectively. The circular dichroism spectra of 1 and 2 are mirror images of each other, which indicates a chirality inversion between Cu(II) (1) and Cu(I) (2).

With the urea groups in the ligand acting as acid catalysts, the authors tested 1 and 2 in the Michael addition of diethyl malonate (3) to trans-β-nitrostyrene (4). Catalyst 1 formed addition product (S)-5 with 72% ee in an overall yield of 55%, whereas 2 gave (R)-5 in 70% ee with 40% yield under optimized conditions.

The authors found that MeCN is the best solvent and that less nucleophilic bases increased enantiopurity values and yields. The catalyst loading could be lowered to 0.5 mol% without significantly decreasing enantiopurity or yields.

Catalyst 1 can be reduced by ascorbate in situ to 2; the product has almost the same catalysis capability as that produced from Cu(MeCN)4PF6. The mechanism of the enantioselective catalysis is not clear, but 1H NMR data suggest that the two urea groups do not act concertedly in the transition state. (J. Am. Chem. Soc. 2012, 134, 8054–8057; Xin Su)

What do you think of Noteworthy Chemistry? Let us know.