ISNet—严格的多维NMR谱解析理论（Ⅱ）

INDEPENDENT SPIN COUPLING NETWORK (ISNet) A RIGOROUS DEDUCTION THEORY FOR STRUCTURE ELUCIDATION FROM MULTI-DIMENSIONAL NMR

ＩＳＮｅｔ（独立自旋偶合网络，ＩｎｄｅｐｅｎｄｅｎｔＳｐｉｎＣｏｕｐｌｉｎｇＮｅｔｗｏｒｋ）是作者为了严格地从多维ＮＭＲ谱推导出溶液状态中复杂有机物的分子结构而建立的新概念。从数学上讲，ＩＳＮｅｔ可被视为分子结构连接性的另一种表达形式，是一种图论表达。与分子结构图的共价键表达不同之处在于，ＩＳＮｅｔ中的每一条边在理论上是二维ＮＭＲ谱图上的一个可观察的峰。本文建立ＩＳＮｅｔ概念，并讨论从实验波谱，运用ＩＳＮｅｔ分析理论导出各类分子结构的原理及其图论、模糊数学基础。

The concept of an ISNet (Independent Spin Network), used to process various molecular connectivities which can be observed from multi-dimensional NMR spectra, is proposed. The conventional one-dimensional spectra-based computer-assisted structural elucidation (CASE) methods are based upon the correlation of subspectra-snbstructure which is often ambiguous. However, modern mnlti-dimensional NMR experiments offer us abundant molecular spin coupling connectivity information. This information can be directly or indirectly mapped to real chemical structure or substructure. This fact is known as "the correlation of spectra patterns and structures" where pattern implies connectivity. ISNets are the general graph theoretical representation for these patterns. By means of systematic graph theory and fuzzy mathematical analysis, a rigorous deduction theory for general purpose structural elucidation from multi-dimensional NMR spectra is presented.