Journal of Chromatography A, 1216 (2009) 7043–7048 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Evidence of 13 C non-covalent isotope effects obtained by quantitative 13 C nuclear magnetic resonance spectroscopy at natural abundance during normal phase liquid chromatography Eliot P. Botosoa a , Virginie Silvestre a , Richard J. Robins a , Jose Manuel Moreno Rojas b , Claude Guillou b , Gérald S. Remaud a,∗ a Elucidation of Biosynthesis by Isotopic Spectrometry Group, CNRS–University of Nantes Unit for Interdisciplinary Chemistry: Synthesis, Analysis, Modelling (CEISAM), UMR CNRS6230, 2 rue de la Houssinière, BP 92208, Nantes 44322, France b Joint Research Centre of the European Commission - Institute for Health & Consumer Protection Physical & Chemical Exposure Unit - BEVABS - TP281 via Fermi, 2 - 21020 Ispra, Italy article info Article history: Received 5 June 2009 Received in revised form 26 August 2009 Accepted 27 August 2009 Available online 31 August 2009 Keywords: Non-covalent isotope effect Carbon isotope fractionation Normal phase chromatography fractionation 13 C NMR 13 C, 2 H and 18 O IRMS Vanillin abstract Quantitative isotopic 13 C NMR at natural abundance has been used to determine the site-by-site 13 C/ 12 C ratios in vanillin and a number of related compounds eluted from silica gel chromatography columns under similar conditions. Head-to-tail isotope fractionation is observed in all compounds at the major- ity of carbon positions. Furthermore, the site-specific isotope deviations show signatures characteristic of the position and functionality of the substituents present. The observed effects are more complex than would be obtained by simply summing the individual effects. Such detail is hidden when only the global 13 C content is measured by mass spectrometry. In particular, carbon positions within the aromatic ring are found to show site-specific isotope fractionation between the solute and the stationary phase. These interactions, defined as non-covalent isotope effects, can be normal or inverse and vary with the substitution pattern present. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Within the broad context of determining isotope ratios for measuring isotope effects, isotope fractionation or isotope incorpo- ration, it is generally recognized that the procedure used to isolate the target molecule(s) must not per se modify the isotope content of the analyte(s). Although techniques such as liquid–liquid extrac- tion [1] or crystallization [2] show negligible isotope fractionation, chromatographic procedures are well known to introduce substan- tial isotope fractionation [3]. Recently, we have introduced quantitative isotopic 13 C NMR spectroscopy at natural abundance as an efficient tool for deter- mining the site-specific distribution of 13 C within a molecule [4]. Provided the acquisition conditions are carefully controlled, the internal long-term repeatability attained is about 1‰, concomitant with a signal-to-noise ratio (SNR) of about 500 [5]. Hence, in these experimental conditions, changes in values of isotope deviation (ı 13 C) <1‰ must be considered as insignificant, those of >1 but <2‰ ∗ Corresponding author. Tel.: +33 2 51 12 57 19; fax: +33 2 51 12 57 12. E-mail address: Gerald.Remaud@univ-nantes.fr (G.S. Remaud). to indicate a tendency, and those ≥2‰ as statistically significant at the 95% confidence limit, if the measurement is performed once. In the case on n replications, the standard deviation is further reduced by a factor of (n) -0.5 . In many processes where isotope deviation is exploited, such as isotope fractionation during reactions, authen- tication, and metabolism, ı 13 C >2‰ is frequently found, enabling the routine measurement of 13 C/ 12 C ratios in the study of these applications. Smaller isotope fractionations could be exploited by improving the measurement precision, which is solely based on the signal-to-noise ratio (SNR) and therefore on the analysis time. All these analytical objectives require the isolation of the target molecule(s) prior to quantitative 13 C NMR spectroscopy. Purifica- tion of the product under investigation has to be performed with great care in respect to possible isotope fractionation during the process: the molecule must show the same isotope profile before and after purification. As discriminatory behavior between heavy and light isotopes during chromatographic separation on solid phase is well documented [3], we have focused particular attention on this technique. Prior studies of both 2 H and 13 C have essentially been based on more or less enriched levels: results from experi- ments at natural abundance are less common [3,6,7]. Furthermore, there are even fewer studies in which site-specific variation has 0021-9673/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.chroma.2009.08.066