From Theory to Experiment: Transformer-based Generation Enables Rapid Discovery of Novel Reactions

Deep learning methods have been proven their potential roles in the chemical field, such as reaction prediction and retrosynthesis analysis. However, the de novo generation of unreported reactions using artificial intelligence technology remains not be completely explored. Inspired by molecular generation, we proposed the task of novel reaction generation. In this work, we applied the Heck reactions to train the transformer model, state-of-art natural language process model and obtained 4717 generated reactions after sampling and processing. We then confirmed that 2253 novel Heck reactions by organizing chemists to judge the generated reactions, and adopted organic synthesis experiment to verify the feasibility of unreported reactions. In this process, it only took 15 days from Heck reaction generation to experimental verification, proving that our model learns reaction rules in-depth and can make great contributions in the novel reaction discovery.

FEATURES OF CU - NI NANOPARTICLE SYNTHESIS: EXPERIMENT AND COMPUTER SIMULATION

Сочетание эксперимента и компьютерного моделирования позволили исследовать особенности процесса синтеза наночастиц Cu - Ni. Наночастицы синтезированы методом экзотермического горения в растворах. Рентгено-фазовый анализ полученных материалов показал, что все образцы представляют собой чистые биметаллические нанопорошки с искаженной кубической кристаллической структурой каждого металла. Методом Монте-Карло в температурном диапазоне от 300 до 600 K установлены закономерности формирования манжеты для двух случаев начального расположения наночастиц меди и никеля: непосредственное соприкосновение и относительное смещение на величину 0,2 нм. Показана возможность тесной интеграции кристаллических структур в результате взаимодействия наночастиц Cu и Ni. Combination of experiment and computer simulation made it possible to study the features of the process of Cu - Ni nanoparticle synthesis. Nanoparticles are synthesized by the method of exothermic combustion in solutions. The X-ray phase analysis of the obtained materials showed that all samples are pure bimetallic nanopowders with a distorted cubic crystal structure of each metal. The Monte-Carlo method in the temperature range from 300 to 600 K established regularities of the neck formation for two cases of the initial location of copper and nickel nanoparticles: direct contact and relative displacement by 0,2 nm. The possibility of close integration of crystal structures as a Cu and Ni nanoparticles interaction result is shown.

Zn-Al Layered Double Hydroxide Thin Film Fabricated by the Sputtering Method and Aqueous Solution Treatment

Zn-Al layered double hydroxides (LDHs) were synthesized herein via a simple process. First, Al-doped ZnO film was deposited onto a glass substrate using the facing target sputtering system. Successful synthesis of the Zn–Al LDH was achieved via a treatment process using an aqueous solution which contains NO3− anions. X-ray diffraction analysis confirmed that it was consistent with the previous Zn–Al LDH synthesis experiment data, and the calculated d-value was 9.1 Å. Scanning electron microscopy observations revealed that the as-synthesized sample had a plate-like structure.

Novel K/Mn phosphate hydrates, K2Mn3(H2O)2[P2O7]2 and KMn(H2O)2[Al2(PO4)3]: hydrothermal synthesis and crystal chemistry

Two novel K/Mn phosphate hydrates, namely, dipotassium trimanganese dipyrophosphate dihydrate, K2Mn3(H2O)2[P2O7]2, (I), and potassium manganese dialuminium triphosphate dihydrate, KMn(H2O)2[Al2(PO4)3], (II), were obtained in the form of single crystals during a single hydrothermal synthesis experiment. Their crystal structures were studied by X-ray diffraction. Both new compounds are members of the morphotropic series of phosphates with the following formulae: A 2 M 3(H2O)2[P2O7]2, where A = K, NH4, Rb or Na and M = Mn, Fe, Co or Ni, and AM 2+(H2O)2[M 3+ 2(PO4)3], where A = Cs, Rb, K, NH4 or (H3O); M 2+ = Mn, Fe, Co or Ni; and M 3+ = Al, Ga or Fe. A detailed crystal chemical analysis revealed correlations between the unit-cell parameters of the members of the series, their structural features and the sizes of the cations. It has been shown that a mixed type anionic framework is formed in (II) by aluminophosphate [(AlO2)2(PO4)2]∞ layers, with a cationic topology similar to the Si/Al-topology of the crystal structures of feldspars. A study of the magnetic susceptibility of (II) demonstrates a paramagnetic behaviour of the compound.