Scheme 1 LSO omega-3 PUFA-rich thermal aging process
Numerous chemical and physical analytical methods have been developed to asses lipid oxidation such as conjugated diene value, peroxide value (PV), alcohols, epoxides, p -anisidine assay, HBR titration, iodometric titration, xynol orange, total polar components (TPC), high performance liquid chromatography (HPLC), fatty acid composition determined by gas chromatography-mass spectrometry (GC-MS), Fourier transformation infrared spectroscopy (FTIR), volatile product determination by gas chromatography, dimer/polymers by size exclusion chromatography (SEC), and electron spin resonance (ESR) (Jacobsen, 2015; Hwang et al., 2017, Velasco et al., 2005). There is however a lack of consistency in many of the results, because most of these analytical methods are designed to detect one type of oxidation product while lipid oxidation is a very complicated process producing numerous products at different times of oxidation. Hence, as suggested in Hwang et al. (2017) comprehensive review, the development of methods that combine the concomitant detection of different types of oxidation products is necessary for the consistent assessment of lipid oxidation. Two important questions were raised by this researcher: which oxidation product best represents a given stage of the lipid’s oxidation? And which analytical method should be used? In this respect, as described below LF-1H-NMR spectroscopy technology has significant potential, as shown in this paper, in elucidating molecular structures of oxidation products from lipids and in revealing the mechanisms of lipid oxidation.
The field of 1H LF-NMR energy time relaxometry is a powerful tool for identifying molecular species and to study their dynamics even in complex materials (Berman et al., 2013a; 2015; Wiesman et al., 2018; Resende et al., 2019ab; Rudszuck et al., 2019). This relates to the measurement of energy time relaxation values as a consequence of interactions among nuclear spins and between spins and their surroundings (matrix). Longitudinal magnetization returns to equilibrium following application of a radio frequency pulse because of energy transferred to the lattice (spin-matrix interactions), and transverse relaxation arises from spin-spin interactions following a 90° pulse. The time constants for longitudinal and transverse energy relaxations are T 1 and T 2 respectively (Berman et al., 2013b). Relaxation time distribution experiments range from simple and rapid one dimensional (1D) tests (T1 or T2) to more complicated multidimensional ones (e.g., T1 vs. T2). 1D tests use constant intervals between pulses, allowing for either longitudinal or transverse relaxation to be evaluated, whereas in multidimensional experiments, the signal is measured as a function of two or more independent variables, allowing the spin system to evolve under different relaxation mechanisms (Song et al., 2002, Berman 2013b). By assuming a continuous distribution of exponentials, a relaxation time distribution of exponential coefficients is achieved with components appearing as peaks. This is an ill-posed Inverse Laplace Transform (ILT) problem. The common mathematical solution implemented today, for both 1D and 2D data, is based on L2  -norm regularization (Song et al., 2002; Graham, et al., 1996; Berman et al., 2013b, Campisi-Pinto et al., 2018).
Current technologies are not effective in characterizing the morphological and chemical structural domains of saturated, monounsaturated fatty acids (MUFA) and PUFA materials, or how the morphological structures of fatty acids, at the meso, nano, and molecular levels, affect their oxidation mechanisms.1H LF-NMR energy relaxation time technology consisting of L1/L2 norm regularization (Campisi et al. 2018, 2019; Resende et al., 2019a,b), is proposed as a tool to analyze PUFA oils undergoing thermal oxidation. This technology can generate two‐dimensional (2D) chemical and morphological spectra using a recently modified and developed primal‐dual interior method for the convex objectives (PDCO) optimization solver for computational processing of the energy relaxation time signals T1 (spin–lattice) and T2 (spin–spin): With carefully chosen reconstruction parameters, the data signals can be reconstructed into 2D graphics of the different energy relaxation times assigned to the mobility of different chemical structures, and their adjacent environments (Wiesman et al., 2018). This reconstruction of LF‐NMR signals into two and three dimensional (2D and 3D) T1 vs. T2 graphs is able to effectively characterize the chemical and morphological domains of complex materials (Wiesman et al., 2018). The 2D graphical maps of T1 vs. T2 generated for butter, rapeseed oil, soybean oil, and linseed oil show that the different degrees of unsaturation of fatty‐acid oils affect their chemical and morphological domains, which influences their oxidative susceptibility (Resende et al., 2019a). The technology of the 1H LF‐NMR energy relaxation time proved to be an effective tool to characterize and monitor PUFA oxidation (Resende et al., 2019a). The use of 2D graphic reconstruction of T1 vs. T2 as compared to only one dimensional (1D) has the ability to increase peak separation on the diagonal (T1 = T2) and when new polymerized oxidation products new peaks appear below the diagonal (same constant T1 but decreased T2) during later stages of the oxidation process (Resende et al., 2019a,b).
Methods using high field 1H NMR relaxation were found by Sun and Moreira (1996), Hein et al (1998), Sun et al. (2011) and Bakota et al (2012) to correlate well with various parameters associated with lipid oxidation (e.g., free fatty acid; polar materials in heated oils; solid fat content (SFC approved as AOCS Cd 16b-93)). It was proposed by Hwang et al. (2017) that ”there are molecular structure and composition changes in oil during oil oxidation and degradation process affecting the chemical environment surrounding the protons. Thus the proton mobility affecting the NMR energy relaxation time values changes as oil degrades”.
High field 1H NMR was also used to analyze aldehydes produced in various heated oils (Guillen and Uriarte, 2009) . These researchers reported on the ability to analyze a list of aldehyde products in linseed oil heated at 190oC for 20 h, and also determined acyl groups’ iodine value and polar compounds. Merkx et al. 2018 reported a broad band selective 1H NMR method for determination of both hydroperoxides and aldehydes in oxidized oils. Furthermore, based on electron spin resonance (ESR) system, combined with free radical standard and trapping agents (TEMPO and PBN) was released for determination of peroxides in the early fast initiation phase of oil oxidation (Velasco et al., 2005). Blumich (2016) developed compact 1H LF -NMR systems and Guilleux et al. (2016) developed an automate LF-NMR system. However, one of the remaining problems of 1H LF -NMR and especially 2D T1-T2 systems is the relatively long experimental and data processing time required to finalize the results. Therefore these systems are not yet suitable for high throughput applications such as real-time reaction monitoring or rapid screening of oil oxidation (Hwang et al., 2017).
In the present study the objective is to develop a non-sample modifying1H LF-NMR energy time relaxation sensitive application supported with other techniques (e.g., HF NMR, GC-MS and viscosity) to evaluate LSO oxidative aging processes, based on monitoring the main chemical and structural changes occurring during thermal oxidative reactions. As for example by following the alkyl tails T2 values it is possible to correlate the degree of olefin functionality and the ratio of the olefin components with the different degrees of functionality that control the onset of crosslinking polymerization and gel structure formation. We demonstrate below the capability of using the rapid 1H T2 energy relaxation time technology to monitor LSO molecular segments mobility. In particular the monitoring of the LSO aliphatic tail’s relaxation was used to follow the chemical and structural changes in all the autoxidation aging phase; starting from the initiation phase (abstraction of hydrogen, fatty acid chain rearrangement and oxygen uptake yielding of hydroperoxides production), the propagation phase (chain reactions resulting in tail cleavage to form alkoxy radicals, and alpha, beta-unsaturated aldehydes formation), and the termination phase (cross linking formation of polymerized end products).