The spectroscopic analysis of complex liquids has widespread applications in the food and beverage industry, the chemical industry, the pharmaceutical industry, healthcare and defence. The last few years have seen an increasing demand for in-situ, rapid and cost-effective authentication and quality control in these fields, with a premium placed on highly sensitive, accurate and, if possible, non-invasive through-container detection techniques.
Fake and faulty products affect producers, suppliers, merchants, and end consumers, and they lead to a financial loss for all parties along the supply chain. The costs incurred are not only direct but also indirect (e.g. shipping costs) and intangible (e.g. reputation), in addition to the healthcare costs associated to the worst cases of contamination and adulteration.
VeriVin is developing a tool which allows for the unique spectroscopic ID tags of a given batch of complex liquids (oil, honey, wine, whisky and so on) to be uploaded onto a database, analysed and used for quality control and authentication. The spectroscopic ID tag of a sample can be contrasted with the ID tags of other similar samples, aiding to verify its expected quality and provenance. Differences in the fingerprint of the same sample over time can be used to gather information about the status and evolution of the sample – or its possible adulteration.
The results of this analysis could go beyond validation and quality control and eventually also lead to more accurate pricing and provenance certification. They could even serve as a comparative purchasing tool to help consumers make choices based on physical data.
Our aim is to develop the largest database of spectroscopic data in the food and beverage industry (wines, spirits, olive oils, honey etc.) and be able to track the movement and authenticity of beverages all along the supply chain. We mean to extract meaningful conclusions from the spectral data acquired with our devices by breaking down and analysing millions of spectra using chemometrics and machine learning techniques. So far, VeriVin has focused on the food and beverage industry, but its technology is translatable to any industry in which the through- barrier analysis of complex liquids is applicable.
Differentiating between the spectra of two different complex liquids requires sophisticated statistical analysis and machine learning techniques. Ascertaining the presence of any particular compound by spectral methods is, in layman’s terms, like looking for a needle in a haystack. Most techniques for liquid analysis in industry are therefore heretofore invasive, ultimately relying on gas chromatography mass spectrometry (GCMS), which is accurate but requires sending an open sample to a lab. This means that the price to pay is the loss or spoilage of the sample. Accurate spectroscopic methods for analysing liquids without breaking the seal are therefore highly desirable. Our company’s two-pronged approach aims to address these two needs: increased accuracy in the spectroscopic analysis of complex liquids and improved through-container spectroscopic analysis.