The reaction of 2‐amino‐3‐cyano‐4,5,6,7‐tetrahydrobenzo[b]‐thiophene with diethyl malonate: synthesis of coumarin,pyridine, and thiazole derivatives

The reaction of 2‐amino‐3‐cyano‐4,5,6,7‐tetrahydrobenzo[b]thiophene ( 1 ) with diethyl malonate ( 2 ) gave two products: 3 and 4 . The reactivity of 3 toward a variety of chemical reagents was studied to give azoles, azines, and their fused derivatives. © 2001 John Wiley & Sons, Inc. Heteroatom Chem 12:168–175, 2001     

Gas‐free initiators for high‐temperature free‐radical polymerization

Quaternary ammonium persulfates as free‐radical initiators for high‐temperature polymerization are synthesized and their shelf‐life stability investigated. These initiators do not have gaseous byproducts and are therefore ideal for frontal polymerization, a process in which polymeric materials are produced via a thermal front that propagates through the unreacted monomer/initiator solution. Quaternary ammonium persulfate initiators offer additional qualities such as high solubility in organic media and low volatility, which are desirable for frontal polymerization. The initiators are synthesized using two procedures, and the initiating efficacy of the respective products is compared to a peroxide initiator in the frontal polymerization of 1,6‐hexanediol diacrylate. Of all the quartenary ammonium persulfates synthesized, tricaprylmethylammonium (Aliquat) persulfate (APSO) is the best initiator for frontal polymerization because it is soluble in organic media, is very reactive, and does not produce volatile byproducts under decomposition. A study of the decomposition kinetics of APSO is performed, and frontal polymerization is proposed as a quicker analytical technique to assay the purity. © 2000 John Wiley & Sons, Inc. J Polym Sci A: Polym Chem 38: 3984–3990, 2000     

Sampling the Variational Posterior with Local Refinement

Variational inference is an optimization-based method for approximating the posterior distribution of the parameters in Bayesian probabilistic models. A key challenge of variational inference is to approximate the posterior with a distribution that is computationally tractable yet sufficiently expressive. We propose a novel method for generating samples from a highly flexible variational approximation. The method starts with a coarse initial approximation and generates samples by refining it in selected, local regions. This allows the samples to capture dependencies and multi-modality in the posterior, even when these are absent from the initial approximation. We demonstrate theoretically that our method always improves the quality of the approximation (as measured by the evidence lower bound). In experiments, our method consistently outperforms recent variational inference methods in terms of log-likelihood and ELBO across three example tasks: the Eight-Schools example (an inference task in a hierarchical model), training a ResNet-20 (Bayesian inference in a large neural network), and the Mushroom task (posterior sampling in a contextual bandit problem).     

Thermoplastic elastomer gels. I. Effects of composition and processing on morphology and gel behavior

Thermoplastic elastomer gels (TPEGs) composed of a poly[styrene-b-(ethylene-co-butylene)-b-styrene] triblock copolymer and a low-volatility, midblock-compatible mineral oil have been investigated at different oil concentrations to ascertain the effect of composition on TPEG morphology and mechanical properties. The impact of thermal processing is also examined by comparing gels thermally quenched to 0°C or slowly cooled to ambient temperature. Transmission electron micrographs reveal that gels with 70 to 90 wt % oil exhibit styrenic micelles measuring ca. 24 nm in diameter. Variation in composition or cooling rate does not have any perceivable effect on micelle size or shape, whereas the rate at which the gels are cooled influences the extent of microstructural order and the time to rupture (t_R) at constant strain. Dynamic rheological testing confirms the presence of a physically crosslinked network at TPEG compositions ranging from 70 to 90 wt % oil, independent of cooling rate. Results presented here suggest that the dynamic elastic shear modulus (G′) scales as t^α_R where α varies from 0.41 to 0.59, depending on cooling rate. © 1998 John Wiley & Sons, Inc. J. Polym. Sci. B Polym. Phys. 36: 2379–2391, 1998     

A method to evaluate the Sachs formula

A rule-based method for evaluating the Sachs formula is introduced. For simple molecules, such as linear chains and monocycles, the Sachs formula can be evaluated by directly using the rules. For complex molecular systems, the Sachs formula can be evaluated by first breaking up the molecule into constituent pieces of sufficient simplicity so that data for them are available in the data base and then constructing the molecule from the known data by adding all pieces together, one at a time, via a series of binary rules. The method changes the way of evaluating the Sachs formula from a trial-and-error-type numerical search to a purely algebraic manipulation and thus tremendously reduces the computing times. © 1997 John Wiley & Sons, Inc. Int J Quant Chem 65 : 19–36, 1997     

A new Na/Mg inverse crown ether

A `missing' member of the inverse crown ether family, namely μ₄‐oxo‐tetrakis(μ‐2,2,6,6‐tetra­methyl­piperidinido)­di­mag­nes­ium­(II)­disodium(I), [Na₂Mg₂O(C₉H₁₈N)₄], has been synthesized by blocking the alternative aromatic metallation route via the use of sterically hindered 1,3,5‐mesityl­ene as a solvent. [Na₂Mg₂O(NR₂)₄] (NR₂ is 2,2,6,6‐tetra­methyl­piperidinide) is shown to form a cationic planar eight‐membered ring with alternating metal and N atoms, which captures at its core an oxide guest that lies on an inversion centre [principal dimensions: Na—O = 2.2405 (11) Å, Na—N = 2.445 (3) and 2.572 (3) Å, Mg—O = 1.8673 (9) Å, and Mg—N = 2.032 (2) and 2.063 (2) Å].