Dataset for stable carbon isotope analysis of commercial vegetable oils in Brazil

Description

This dataset contains stable carbon isotope composition (δ¹³C) data used in the study “Stable carbon isotope analysis for adulteration screening of commercial vegetable oils in Brazil”. The dataset includes anonymized commercial samples of refined maize, sunflower, soybean, and canola oils collected in Brazil, controlled C3:C4 mixtures prepared with sunflower and maize oils, and repeatability data from replicate EA-IRMS analyses. Commercial brands and manufacturers were anonymized to preserve confidentiality. Carbon isotope values are reported in milliurey (mUr).

Steps to reproduce

Commercial refined vegetable oil samples were collected from supermarkets and wholesale retailers in São Paulo city and inland municipalities of São Paulo State, Brazil, between 2021 and 2022. Samples were selected to represent commercially available maize, sunflower, soybean, and canola oils and were anonymized using coded identifiers to preserve brand and manufacturer confidentiality. Controlled binary mixtures of sunflower oil (C3 source) and maize oil (C4 source) were prepared by mass directly in tin capsules for EA-IRMS analysis, using a predefined gradient of C3:C4 proportions. The reported mixture proportions correspond to the actual weighed fractions. Carbon stable isotope composition (δ¹³C) was determined by continuous-flow isotope ratio mass spectrometry (CF-IRMS), using aliquots of 50–100 µg of oil weighed into tin capsules. Samples and reference materials were prepared and analyzed under the same conditions. Isotope ratios were expressed relative to the VPDB standard and reported in milliurey (mUr). The dataset was cleaned and organized into separate files containing commercial sample data, controlled C3:C4 mixture data, repeatability measurements, and a variable dictionary. Brand and manufacturer information was removed from the public dataset. To reproduce the analyses, import the CSV files, use the variable dictionary to identify variables and units, and calculate descriptive statistics for each oil type. Differences among oil groups can be evaluated using the Kruskal-Wallis test followed by Dunn’s post hoc pairwise comparisons with Holm adjustment. The relationship between δ¹³C values and mixture proportions can be evaluated by simple linear regression, and the binary mixing model can be applied to estimate apparent C3 or C4 contributions in samples with atypical isotopic values. The statistical analyses reported in the associated manuscript were performed in Python 3.12 using NumPy, pandas, and SciPy.

Institutions

• Universidade Estadual Paulista (Unesp)

  São Paulo, São Paulo

Categories

Funders

• Coordenação de Aperfeicoamento de Pessoal de Nível Superior

  Ministry of Education

  Brasília

  Grant ID: Finance Code 001

Licence