Food authentication goes green: Element profile-driven differentiation between organic and conventional apple juices

Abstract

Apples and apple juice are among the most popular fruit products in Germany. Rising demand and comparatively higher prices for organically produced apples and apple juices make them susceptible for adulteration with cheaper raw materials to achieve higher profit margins. Therefore, sustainable experimental methods are required to detect such fraud regarding the cultivation method. Element profiling using inductively coupled plasma – mass spectrometry (ICP-MS) is such an experimental approach to distinguish between cultivation methods. Using an environment-friendly sample preparation strategy and a ratio-based evaluation approach in combination with a random forest classification model, it was possible to distinguish between the cultivation methods of processed apples from northern Germany with an accuracy of 94.2 % ± 2.3 %. The results were verified by analyzing samples from local supermarkets. Furthermore, the detection of admixtures of conventional juice to organic juice was studied using a regression analysis (5–50 % admixture). Admixtures could reliably be detected from a proportion of 20 % with a support vector machine regression model (R²: 0.8 ± 0.005, RMSE: 6.9 % ± 0.1 %). Thus, falsification of the cultivation method can be detected even in mixtures. The study shows great potential towards sustainability, reducing sample preparation time, hazardous chemicals and energy consumption.