A Proposed Stereochemical Mechanism for the Improved Preparation of Maleic Anhydride Cycloadduct of CLA

The fatty acid derivatives, prepared from renewable natural oils, can be used as highly promising and potential substitutes for petrochemicals. The study of process improvement and stereochemical mechanism for preparing these derivatives would be beneficial for their industrial production. Conjugated linoleic acid (CLA) containing 9cis-11trans (9c, 11t) and 10trans-12cis (10t, 12c) isomers was prepared from Salicornia herbacea seed oil. Maleic anhydride cycloadduct of CLA (MAC) was prepared by an improved process, and it was characterized by FTIR, 1H and 13C NMR, etc. A new method to calculate conformers-ratio of CLA or MAC was also developed. Furthermore, the stereochemical mechanism for the improved preparation of MAC was proposed primarily by the calculation method above. The following observations were made: 1) The yield of MAC could reach as high as 96.7% under mild reaction conditions and with an easy and efficient product separation;2) The trans-trans CLA in the s-cis conformation acted as a predominant reactant to Diels-Alder [4 + 2] cycloaddition of maleic anhydride, which was the main reaction occurred simultaneously with catalytic configurational isomerizations of CLA in one step;3) From all studied CLA conformers, the most stable conformation was the s-trans conformation of trans-trans CLA, while the s-cis conformation of trans-trans CLA had the most favorable structural parameters for cyclohexenyl ring formation;4) Four MAC conformers derived from 9c, 11t- and 10t, 12c-CLA, were obtained as final main products that were determined to be cis-cycloadducts;5) The endo/exo ratios of the cis- cycloadducts derived from 9c, 11t- and 10t, 12c-CLA, were 2.14:1 and 1.99:1, respectively;and 6) The results obtained from the calculation method above were in excellent accordance with those from our experiments.

Diffusion Analysis and Modeling of Potassium Sorbate in Gelatin Based Antimicrobial Film

The diffusion of potassium sorbate incorporated into gelatin based antimicrobial film was measured. Fick’s second law was applied to investigate the mechanism of potassium sorbate released from the films. The mathematical model was established. The result showed that the diffusion coefficient D increased with the increase of potassium sorbate concentration. The effects of temperature 5?C, 25?C and 35?C on diffusion were investigated. The mechanisms of potassium sorbate diffusion through gelatin films were mainly Fickian and determined by the power law model Mt/M∞ = k × tn. A decrease in temperature from 35?C to 5?C resulted in a reduction of diffusion coefficients from 5.20 × 10?12 to 1.36 × 10?12 m2/s. The diffusion coefficient of potassium sorbate is influenced by receiving solution pH values.

A Novel Deep Hybrid Recommender System Based on Auto-encoder with Neural Collaborative Filtering

Due to the widespread availability of implicit feedback(e.g., clicks and purchases), some researchers have endeavored to design recommender systems based on implicit feedback. However, unlike explicit feedback,implicit feedback cannot directly reflect user preferences. Therefore, although more challenging, it is also more practical to use implicit feedback for recommender systems. Traditional collaborative filtering methods such as matrix factorization, which regards user preferences as a linear combination of user and item latent vectors, have limited learning capacities and suffer from data sparsity and the cold-start problem. To tackle these problems,some authors have considered the integration of a deep neural network to learn user and item features with traditional collaborative filtering. However, there is as yet no research combining collaborative filtering and contentbased recommendation with deep learning. In this paper, we propose a novel deep hybrid recommender system framework based on auto-encoders(DHA-RS) by integrating user and item side information to construct a hybrid recommender system and enhance performance. DHA-RS combines stacked denoising auto-encoders with neural collaborative filtering, which corresponds to the process of learning user and item features from auxiliary information to predict user preferences. Experiments performed on the real-world dataset reveal that DHA-RS performs better than state-of-the-art methods.