University of Canterbury, Private Bag 4800, Christchurch, 8140, New Zealand.
Nelson Marlborough Institute of Technology, Marlborough Campus, P.O. Box 643, Blenheim, 7240, New Zealand.
Anal Chim Acta. 2021 Oct 16;1182:338944. doi: 10.1016/j.aca.2021.338944. Epub 2021 Aug 13.
We present a fully automated approach for quantitative compositional analysis of fermented beverages using benchtop nuclear magnetic resonance (NMR) spectroscopy. NMR spectroscopy, renowned for its applications in chemical structure elucidation, is gaining attention as a quantitative analytical technique due to its inherent linearity and the ability to obtain comprehensive quantitative information with a single simple experiment. Recently developed benchtop NMR spectrometers offer the quantitative capabilities of NMR to a wide range of potential users in industry, but their applicability has been limited by the reduced effective spectral resolution and the need for more advanced data processing. We address this problem with a model-based algorithm that hinges on the well-understood description of quantum mechanical phenomena in NMR spectroscopy. We demonstrate the effectiveness of our approach on a challenging problem of analysing the composition of wine and related fermented beverages - an important potential niche application of quantitative NMR. We successfully quantify more than 15 major components in the wine matrix and enable the quantification of species whose analysis is generally not possible with established methods. The average discrepancy of the obtained concentrations, when compared to the traditional methods of analysis, usually does not exceed 10% and is lower for the most abundant species (e.g. below 5% for ethanol).
我们提出了一种使用台式核磁共振（NMR）光谱仪对发酵饮料进行定量成分分析的全自动方法。NMR 光谱学以其在化学结构阐明方面的应用而闻名，由于其固有的线性和仅通过单次简单实验即可获得全面定量信息的能力，它正作为一种定量分析技术受到关注。最近开发的台式 NMR 光谱仪为工业界的众多潜在用户提供了 NMR 的定量能力，但由于有效光谱分辨率降低以及需要更先进的数据处理，其适用性受到限制。我们通过基于模型的算法解决了这个问题，该算法依赖于对 NMR 光谱学中量子力学现象的充分理解。我们在分析葡萄酒和相关发酵饮料成分这一具有挑战性的问题上展示了我们方法的有效性——这是定量 NMR 的一个重要潜在利基应用。我们成功地定量了葡萄酒基质中的 15 种以上主要成分，并能够定量分析通常无法用现有方法进行分析的物质。与传统分析方法相比，获得的浓度的平均差异通常不超过 10%，对于最丰富的物质（例如乙醇低于 5%）则更低。
