Ufa State Petroleum Technological University

Organic compounds have the widest practical application, and the methodology of their synthesis is often developed specifically to solve applied problems in medicinal chemistry, catalysis, materials chemistry, agriculture, food industry, and the creation of electronic and sensor devices. This is due to the huge variety of properties and functional capabilities of organic substances, as well as the ability to fine-tune their structure to impart certain practically useful characteristics. Currently, the number of organic compounds used in various fields of industry, medicine, and agriculture is in the hundreds of thousands, and their number continues to grow steadily due to the rapid development of synthetic organic chemistry and predictive methods for determining properties, including using artificial intelligence. This collective review is devoted to the achievements of Russian chemists in the field of practically oriented organic chemistry over the past 5–10 years. The review presents the achievements of leading research teams representing both RAS institutes and Russian universities, from Kaliningrad to Siberia.

The injection of carbon dioxide (CO2) into coal seams simultaneously facilitates geological sequestration to reduce atmospheric CO2 levels and enhances methane recovery, albeit with potential implications for coal strength and the long-term geomechanical stability of the reservoir. In this research, a collection of 159 experimental data sourced from reputable literature was utilized to evaluate the changes in coal strength as exposed to CO2. To model the unconfined compressive strength (UCS) of coal specimens during CO2 injection, four robust deep learning and machine learning models were used: deep neural network (DNN), convolutional neural network (CNN), light gradient-boosting machine (LGBM), and random forest (RF). The novelty of this work lies in the collection of extensive data compared to similar studies and the use of advanced artificial intelligence (AI) algorithms to model mechanical transformations of coal interacting with CO2. The findings indicated that the DNN model accurately predicted coal UCS with a mean absolute percent relative error (MAPRE) of 2.98% (training), 3.09% (testing), and 3.01% (overall). Both statistical and graphical analyses confirmed that deep learning models (DNN and CNN) outperformed other machine learning models in capturing coal strength variations during CO2 sequestration. Specifically, the DNN model accurately predicts trends, showing that coal UCS decreases with rising CO2 saturation pressure, longer CO2 exposure time, and lower CO2 soaking temperature. By computing Pearson, Spearman, and Kendall correlation coefficients, sensitivity analysis revealed varying impacts on coal UCS during CO2 sequestration, with coal rank and CO2 interaction time showing significant contributions. Coal rank exhibited a strong positive correlation with UCS, highlighting the importance of intrinsic material characteristics in maintaining coal strength. Conversely, CO2 interaction time showed negative correlations, particularly in non-linear terms, indicating that extended exposure to CO2 tends to degrade the mechanical integrity of coal. CO2 interaction temperature minimally influences UCS, and CO2 pressure weakly correlates with a slight decrease in coal strength. Also, SHAP analysis showed that coal rank has the largest impact on prediction variability. The leverage approach identified two suspected data points and four outliers, confirming the reliability of the experimental dataset, while the proposed DNN model demonstrated high accuracy in predicting the UCS of CO2-exposed coals.

Syntheses are reported for secondary and tertiary amines as well as ammonium salts containing cycloacetal fragments. The inhibitory activity of these compounds for St3 steel in 3

The results of the experimental study of the electroplastic effect on coarse-grained Grade 2 titanium under tension are presented. A comparison is made of the deformation behavior without current and with current of various modes differing in density and duty cycle. The structure evolution of the samples is investigated for different parameters of the pulsed electric current. It is shown that increasing the intensity of the electric pulse treatment results in the reduction of both yield stress and plasticity. Tension without current activates two mechanisms of plastic deformation - dislocation sliding and twinning at relatively high tensile stress. Pulsed electric current activates dislocation sliding, resulting in reduction of tensile stress and suppression of twinning. As a result, dislocation pileups initiate fracture earlier than when both plastic deformation mechanisms are active. The results demonstrate the possibility of significantly reducing the flow stress of titanium with a slight decrease in elongation at break.

The preparation of secondary amines, amides, and quaternary ammonium salts (QAS) containing 1,3-dioxolane and gem-dichlorocyclopropane fragments based on N,N-dimethylpropylenediamine under heating or microwave irradiation (MW) conditions is described. Optimal conditions for the preparation of the target products in quantitative yields using MW are determined.