Nowadays, MIS structures based on silicon dioxide films have become the most widely used. However, the desire to reduce the channel length and the nano-scaling in electronics showed, that SiO2 can’t provide the specified MIS devices parameters and characteristics. This is due to the superfine insulator films problem. First of all, this can be solved by replacing SiO2 with another insulator that meets certain requirements. Oxides of transition and rare-earth metals meets almost all of these requirements. However, the principle issue of insulator and semiconductor compatibility remains unsolved in terms of ensure theirs interface quality. Therefore, the researching purpose is to create a general approach to the selection of insulator which is an alternative to silicon dioxide. This insulator must provide a high quality of insulator-semiconductor interface.In MIS structures, the main indicator of semiconductor-insulator interface is its density of the electric charge. Low values of the charge density at the semiconductor-insulator interface are corresponds to high quality of the interface. This is due to the fact that low interface charge density has low affecting on threshold voltage and provide higher its stability. In the first approximation, the surface states density is identically equal to the incomplete bonds density at semiconductor-insulator interface. Therefore, this parameter is selected as the basis of the insulator choosing criterion for a particular semiconductor substrate.It is shown that the classical theory of crystalline lattice (site aspect) can't allow to obtain an analytical expression for the insulator choosing criterion for MIS structures. Therefore, a new form of description of a crystal lattice (intersite aspect) is proposed. Since the crystals differs not only by symmetry but also by the crystallographic orientation of their interface with the vacuum, then an infinite number of solutions about value of incomplete bonds density at the interface correspond to each pair of materials. For obtaining only the one solution by using insulator choosing criterion, a new characteristic parameter of the crystalline substance is introduced - the average bond length. To confirm the practical significance of the average bond length, the relationship between this parameter and electron work function is obtained. The experiment based on literary data of both electron work function and lattice constants was made. The experimental results proves both validity of the relationship and practical significance of the average bond length.On the basis of the intersite aspect and the average bond length, general insulator choosing criterion for MIS structures on certain semiconductor substrate was obtained.For replacement of silicon dioxide in silicon MIS devices, insulator was selected from suitable materials namely oxides of transition and rare-earth metals. The incomplete bonds density at the insulator-silicon interface was calculated for each of these materials. Analysis of the calculation results showed that, in accordance with the insulator choosing criterion, the most promising material is CeO2.The insulator choosing criterion for MIS structures on certain semiconductor substrate is proposed. Its allows, under other equal conditions, to choose the suitable insulator which will provide the least electric charge density at the insulator-semiconductor interface, in comparison with other pretenders. In case of a silicon substrate, such insulator is CeO2. A new characteristic parameter of a crystalline materials is introduced - the average bond length. The analytical expression, which get the relationship between this parameter and the electron work function, is obtained.The obtained results provide an opportunity to make choices insulator for MIS devices based on not only silicon substrates, but also any other semiconductor materials. The proposed insulator choosing criterion can serve as the basis for build new criteria. This ones can help in solving more complicated problems like choosing insulator for MIS devices based on semiconductor substrate with a certain crystallographic orientation.Bibl. 10, Fig. 9., Рассматриваются вопросы общего подхода к поиску диэлектрика для кремниевых МДП-структур в качестве альтернативного диоксиду кремния (SiO2).Показано, что на основе классической теории описания кристаллической решётки (узловой аспект) невозможно получить критерий выбора диэлектрика для МДП-структуры. Поэтому предложен новый –межузловой аспект – описания кристаллической решётки на основе которого получен общий критерий выбора диэлектрика для любой полупроводниковой подложки. Благодаря введению нового характеристического параметра кристаллического вещества – средней длины связи – получено аналитическое выражение для работы выхода электрона из кристалла.Установлено, что наиболее пригодным для кремниевых МДП-структур является диоксид церия CeO2.Библ. 10, рис. 9., Розглядаються питання загального підходу до пошуку діелектрика для кремнієвих МДН-структур як альтернативного діоксиду кремнію (SiO2).Показано, що на основі класичної теорії опису кристалічної решітки (вузловий аспект) неможливо одержати критерій вибору діелектрика для МДН-структур. Тому запропоновано новий – міжвузловий аспект – опису кристалічної решітки, на основі якого одержано загальний критерій вибору діелектрика для будь-якої напівпровідникової підкладки. Завдяки введенню нового характеристичного параметру кристалічної речовини – середньої довжини зв’язку – одержано аналітичний вираз для роботи виходу електрона з кристалу.Встановлено, що найбільш придатним для кремнієвих МДН-структур є діоксид церію CeO2.Бібл. 10, рис. 9.