Molecular design of antioxidant lubricating oil additives via QSPR and analysis dynamic simulation method

Abstract

Alternatives antioxidant lubricant additives have been proposed by many researchers to replace long-time use of multifunctional lubricant additive, Zinc-dialkyl-dithiophosphate (ZDDP). Computational methods (QSPR and MD) were successfully used to design five novel anti-oxidant lubricating oil additives with improved properties and dynamic binding energies. The five novel antioxidant lubricant additives with improved properties and without sulfated ash, phosphorus, and sulfur (SAPS) were successfully designed. These group of newly designed additives were better than other similar research from the literature and could stop or terminate complete oxidation of the lubricant. Moreover, the result of molecular dynamics simulations (MD) in which 3-(2-(3-amino-4,5-dihydroxyphenyl)-3-chloro-4-oxoazetidin-1-yl)-2-argioquinazolin-4(3H)-one with the most promised dynamic binding energy of -1487.68 kcal/mol was found to be dynamically bound better on the simulated steel coated surface than the DLC coated surface and was also revealed to be excellently good when compared with commercially sold multifunctional additives, ZDDP (197.143 kcal/mol). These groups of five newly designed additives could be easily synthesized in the wet laboratory by adding –OH and or NH2 around the ortho, meta and para position of the phenyl group of the structure template. This research will help designing new oxidation resistance lubricating oil additives with improved properties that will reduce the capacity of base oil to oxidize and form sludge during the autoxidation process of the lubricating oil.

Machine-learning-assisted molecular design of phenylnaphthylamine-type antioxidants

“…1 (numbers in brown font). Based on the research literature on antioxidants and the typical characteristics of chemical spaces, 7,17,18,27,29 15 fragments (listed in Table 1) were selected as substituent groups (e.g., hydroxyl, amino, various alkyl groups). Then, a series of phenylnaphthylamine antioxidants were designed and modeled by an exhaustive method, that is, each fragment group was introduced into ten substituent positions on the skeleton structure.…”

“…27,58 The compatibility and interaction between antioxidants and matrix also have significant roles in the antioxidation process. 29,33 Incompatibility and weak interaction result in the migration of antioxidants, which is not conducive to continuous antioxidation. Hence, another two factors, i.e., d 47 and E binding , 57 were selected to evaluate the interaction between antioxidants and matrix.…”

“…27,28 At the same time, the substituent position in aromatics also affects the BDE, which further impacts AA. 29 As mentioned previously, studies suggest that AA can be improved by introducing electron-donating groups at the ortho and para positions of the phenyl group. [30][31][32] In addition to these studies on chemical factors, Abdulfatai et al 29 and Luo et al 33 studied the effects of physical factors on AA by an MD method.…”

“…29 As mentioned previously, studies suggest that AA can be improved by introducing electron-donating groups at the ortho and para positions of the phenyl group. [30][31][32] In addition to these studies on chemical factors, Abdulfatai et al 29 and Luo et al 33 studied the effects of physical factors on AA by an MD method. They concluded that a higher bonding strength between antioxidants and the matrix leads to lower antioxidant mobility, which is conducive to a better antioxidant effect.…”

Molecular Mechanics-Based Quantitative Structure-Activity Relationship Study on the Inhibitory Activity of Some Schiff Bases against Escherichia coli

“…To create QSAR models, the selected descriptors were subjected to regression analysis [12] using Genetic Function Approximation (GFA) in material studio software [22] with empirically determined activities as the dependent variable. The models were evaluated using a "lack of fit" (LOF) score, which was calculated using a modest version of Friedman's original formula, with the best model receiving the highest fitness score [23,24]. The original Friedman formula is used to calculate the LOF [25] shown in equation 1.…”