Noteworthy Chemistry

May 18, 2014


Easily resolve E- and Z-Julia–Kocienski reaction products. Monofluorinated alkenes are used as isosteres (replacements for peptidic bonds) in medicinal chemistry. The Julia–Kocienski reaction can be used to make these compounds, but it produces E- and Z-stereoisomers that are difficult to separate even with chromatographic methods.

Y. Zhao, F. Jiang, and J. Hu* at the Shanghai Institute of Organic Chemistry developed a convenient method for isolating the stereoisomers. When they studied the reaction between a fluorinated 2-pyridyl benzyl sulfone (1 in the figure) and 2-naphtaldehyde (2), they found that acidic workup conditions gave a mixture of stereoisomers, whereas neutral conditions yielded only the Z-isomer.

Let us know what you think about Noteworthy Chemistry!

Preparation and isolation of E- and Z Julia–Kocienski products

With the use of 19F-NMR spectroscopy, they concluded that the major products of the reaction are two diastereomeric sulfinates, which hydrolyze to form the alkenes. Because the sulfinates hydrolyze at different rates and only one is water-soluble, the alkenes can be separated via kinetic resolution.

The entire process consists of

  1. a Julia–Kocienski reaction under basic conditions,
  2. adding water and extracting the Z-isomer with ether,
  3. acidifying the aqueous phase, and
  4. extracting it with ether to remove the E-isomer.

The authors ran the reaction with several aldehydes and sulfones. E- and Z-products were made with excellent selectivities, including fluorinated analogues of the anticancer compounds combretastatin-A-4 and DMU-212. (J. Am. Chem. Soc. DOI: 10.1021/jacs.5b02112; José C. Barros)

Back to top

Which rearrangement best converts an acid to an isocyanate? J. Zhao, C. S. Siegel, and colleagues at Sanofi–Genzyme (Waltham and Cambridge, MA) developed a kilogram-scale synthesis of a glucosylceramide synthase inhibitor for potential use against Fabry disease, a rare genetic lysosomal storage disease. One of the key steps is the conversion of a thiazole carboxylic acid to a carbamate through an isocyanate intermediate.

In the authors’ first synthesis, they treated the acid with ethyl or isobutyl chloroformate to form the mixed anhydride. Sodium azide (NaN3) was used to form the acyl azide which was then heated in toluene to effect a Curtius rearrangement.

Differential scanning calorimetry studies on the acyl azide showed that it begins to decompose at only 40 ºC with a sharp exotherm of 200 kJ/mol. The authors then modified the synthesis by replacing NaN3 with diphenylphosphoryl azide (DPPA). They added DPPA to the acid in toluene at 100 ºC so that the acyl azide was generated in situ in a dose-controlled manner.

The authors then devised an even safer option: a Lossen rearrangement in which the starting acid is converted to a hydroxamic acid by treating it with hydroxylamine and carbonyl diimidazole (CDI). A second portion of CDI converts the hydroxamic acid to the dioxazolone, which upon heating rearranges to the isocyanate. (S)-Quinuclidinol traps the isocyanate to produce the target carbamate. (Org. Process Res. Dev. DOI: 10.1021/op500379a; Will Watson)

Back to top

Dissolve gold without using aqua regia or cyanide. Conventional wisdom dictates that gold, one of the least reactive metals, dissolves nitric acid–hydrochloric acid (better known as aqua regia) or strongly basic cyanide solutions. Both processes involve oxidative dissolution; the products are AuCl4 and Au(CN)2, respectively. For this reason, although many gold compounds have been prepared, very few can be made directly from elemental gold.

S. N. Britvin* at Saint Petersburg State University and the Russian Academy of Sciences (Apatify) and A. Lotnyk at the Leibniz Institute of Surface Modification (Leipzig, Germany) prepared a water-soluble phosphine ligand (3 in the figure) that can dissolve elemental gold. This finding may evolve into a new way to prepare gold-based complexes.

Phosphine ligand synthesis and its reaction with Au(0)

The authors synthesized tricyclic phosphine compound 3 by capping the cyclic triamine 1,4,7-triazacyclononane (1) with tris(hydroxymethyl)phosphine (2) in one step via a Mannich-type condensation. Compound 3 has excellent stability in air and elevated temperatures. It is soluble in water up to 0.1 M and can be protonated at all three basic nitrogen atoms.

Treating a bright yellow aqueous solution of HAuCl4 with 4, the triprotonated form of 3, changes the color to dark purple, indicating the formation and aggregation of gold nanoparticles. When an excess of 4 is added, however, the color disappears because 4 mediates the air-oxidation of gold to form the cationic Au(I) species 5. In addition, ligand 3 replaces cyanide anion, the strongest known Au(I) ligand, in KAu(CN)2 to form the anhydrous salt [33Au][Au(CN)2].

The authors demonstrated that a strongly coordinating phosphine ligand can promote the oxidative-dissolution of elemental gold, thus offering an alternative method for preparing gold complexes directly from Au(0). From the perspective of phosphine ligand development, this type of ligand may fill the gap between classical tertiary phosphines and aminophosphines because of its strong electron-donating ability and low steric demand. (J. Am. Chem. Soc. DOI: 10.1021/jacs.5b01851; Xin Su)

Back to top

QSPR study links molecular structures to critical constants. Process design requires an understanding of the thermodynamic properties of the materials involved. When data on these properties are unavailable, a quantitative structure–property relationship (QSPR) regression method can establish relationships between molecular properties and the thermodynamic properties of interest. These relationships can be updated and refined as new data become available.

W. H. Carande and colleagues at the National Institute of Standards and Technology (Boulder, CO) and the University of Colorado Boulder used support vector regression (SVR) to estimate critical properties and acentric factors (a measure of nonsphericity) for 900 pure compounds. They used experimental data on three-dimensional geometry and connectivity to calculate >500 molecular descriptors for each compound. These were used to define the input vectors for the SVR calculations.

Optimal SVR parameters produced results that were as close as possible to target values based on experimental data for critical temperatures, critical temperature/critical pressure ratios, and reduced saturation pressures (related to the acentric factor). The authors showed that the effective use of the available information in large experimental datasets requires many variables: 11 to 33 in the current study. Descriptors based on charge distributions were especially influential in the calculations, as were some descriptors related to vibrational and spectroscopic analysis.

The methods used in this study calculated critical property values that were closer to the observed values than those generated in a previous study by the same research team. The authors attribute the improvement to expansions and corrections to the molecular descriptor dataset and better model generation.

The dataset for the current study used only experimental values. This eliminated the overfitting problem observed in other recent studies that used predicted and observed data. (J. Chem. Eng. Data DOI: 10.1021/je501093v; Nancy McGuire)

Back to top

This polymer’s fluorescence is stimulated by adding an acid and quenched via basification. By using noncovalent interactions, researchers have developed a variety of supramolecular polymers with fluorescence that changes in response to stimuli. When most of these polymers are exposed to external stimuli, their fluorescence is quenched or turned off.

W. Bu and co-workers at Lanzhou University (China) used an elegant molecular design to prepare a polymer (1 in the figure) that works in lighting-up and dimming-down modes. Its emission intensity is enhanced in the presence of acids and decreases when it is exposed to bases.

Polymer whose fluorescence is enhanced via acid complexation

Polymer 1 in solution is weakly fluorescent because the intramolecular rotations of the phenyl groups in its tetraphenylethylene (TPE) unit nonradiatively dissipate its exciton energy. Adding a dumbbell-shaped diacid (2 or 3) to the polymer solution forms noncovalent networks. The supramolecular complexation restricts intramolecular rotation and enhances the polymer fluorescence (lighting up).

Adding a base disassembles the complex structure and returns the polymer to its weakly fluorescent state (dimming down). The lighting–dimming cycle can be repeated several times. (Chem. Commun. DOI: 10.1039/C5CC00934K; Ben Zhong Tang)

Back to top