A Representation Method for Substructures Associated with
1H-NMR Spectra--HYPER Code--
Organic Synthesis Research Laboratory, Sumitomo Chemical Co., Ltd.; Tsukahara,
Takatsuki-shi 569-1093 Japan
A representation method for substructures associated with 1H-NMR spectra has been developed. The method, HYPER code system, is based on concentric layers of nearest neighbors around a central atom. We also developed a program, “HYPERGEN”, which creates HYPER codes for substructures from the structural information of a compound. Since substructure-subspectrum correlation information is requisite for data oriented structure elucidation systems, HYPER code plays a significant role in analyzing 1H-NMR spectra.
For creating a correlation table it is required that not only the structures and spectra are input, but also information on their assignment. We have therefore also developed a representation method for the assignment. An NMfile format has been defined for this purpose. It consists of structural information, chemical shifts with the corresponding atom numbers, and auxiliary information.
The HYPERGEN system accepts input data in the NMfile format and from this creates HYPER codes. The HYPER codes provide information on the number of hydrogen atoms attached to each carbon. Such information is also provided for the last sphere of a substructure, even though there is no information about nodes and bonds connected to it. The codes for the geminal protons, which are connected to the same carbon atom but have different chemical shifts, are also produced in order to distinguish the difference between chemical shifts.
Although the HYPER codes are theoretically able to represent an infinite number of spheres as substructures, for the purpose of this study we have employed up to the sixth sphere. Larger substructures can be represented through a combination of them in a structure generation process of the structure elucidation system.
Using the HYPER codes we have constructed a correlation table between 1H-NMR subspectra and the corresponding substructures. This table will provide an indispensable constituent for structure elucidation systems.
[Contents (In Japanese) ]
[Contents (In English) ]