Online monitoring of transesterification reaction by medium-resolution benchtop 1H NMR and NIR spectroscopy
Graphical abstract
Introduction
Nowadays, biodiesel is one of the major biofuels worldwide consumed in transportation vehicles, with a significant consumption increase in the last decade, boosted mainly by United States, Brazil, and some European countries [1].
Biodiesel production is mostly based on the transesterification reaction of triglycerides from vegetable oils or animal fats, with a short-chain alcohol. This reaction occurs by a homogeneous or heterogeneous (acid or basic) catalyst, with an excess of alcohol to shift the reaction equilibrium to the product formation, producing fatty acid esters (biodiesel) and glycerol [[2], [3], [4]].
The reaction consists of a set of three consecutive and reversible steps, where a triglyceride (TG) molecule reacts successively with three alcohol molecules, e.g., methanol (MeOH) (Eq. 1). Firstly, one diglyceride (DG) and one fatty acid methyl ester (FAME) are formed (Eq. 2). The DG molecule could be a 1,2-DG or 1,3-DG. Sequentially, the DG molecule reacts with a MeOH molecule to produce a monoglyceride (MG) molecule and another FAME molecule (Eq. 3). The MG could be a 1-MG or 2-MG. Finally, MG reacts with MeOH to produce another FAME molecule and a glycerol (GL) molecule (Eq. 4), summing three FAME molecules overall in the reaction process [5,6].
Besides the fact that this reaction is well established, some concerns remain about the final product quality, related to some parameters of the transesterification reaction, e.g., incomplete reactions can lead to a poor biofuel product due to the presence of GL, MG, and DG in the final product. These compounds bring huge problems to the fuel storage/preservation and also in vehicles fuel injection system [7,8]. In this way, it is of paramount importance to monitor the transesterification reaction process to ensure the final product quality and to minimize the process costs [9].
Several works have reported different analytical techniques to monitor the transesterification reaction of biodiesel through the determination of the major components present in the reaction medium. Nevertheless, few works were able to quantify minority compounds as MG and DG. Among those works, in which were able to perform this important task, the great majority are based on chromatographic techniques, presenting several limitations to be applied to online monitoring of the biofuel producing reaction [5,[10], [11], [12], [13], [14]].
The present work proposes a new application of low-field proton nuclear magnetic resonance (LF-1H NMR) and near-infrared (NIR), for the online monitoring of the transesterification reaction, in terms of online quantification of FAME, TG, DG and MG, directly in the reaction medium.
Section snippets
Instrumentation
LF-1H NMR spectra were acquired on a benchtop low-field (LF) NMR spectrometer (Spinsolve, Magritek) equipped with permanent magnets to generate a homogeneous magnetic field of 1.0 T (43 MHz for proton Larmor frequency). For the acquisition of LF-1H NMR spectra, a glass-flow cell, previously described in Killner et al. [15], was used. The glass-flow cell has adequate volume and length to ensure the sample pre-polarization. Each spectrum was acquired immediately after the application of a single
Conversion rates determination by HF-1H NMR
All molar conversion rates, for different compounds present in the reaction medium during the transesterification reaction, were determined, based on the areas of the peaks showed in the HF-1H NMR spectra. Different algebraic equations were applied based on the work of Nieva-Echevarría et al. [17].
For those compounds that generate specific non-overlapped signals in the spectrum, such as 1-MG, 2-MG, and 1,2-DG, the molar conversion rates (C) were determined using Eqs. (5), (6), (7):
Conclusion
The present work demonstrated the online application of LF-1H NMR and NIR spectroscopic techniques, to monitor the progress of the transesterification reaction in terms of major compounds, as fatty acid methyl esters and triglycerides, and also minor compounds as monoglycerides and diglycerides. The studied compounds presented acceptable errors of prediction applying developed PLS calibration models for the LF-1H NMR and the NIR techniques, except for 2-MG. Thereby, this approach allows a
CRediT authorship contribution statement
Diego Galvan: Conceptualization, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing - review & editing. Leticia Magalhães de Aguiar: Conceptualization, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing - review & editing. Jarbas José R. Rohwedder: Methodology, Supervision, Writing - review & editing. Dionisio Borsato: Funding acquisition, Supervision, Writing - review & editing. Mario Henrique M. Killner:
Acknowledgements
The State University of Londrina (UEL), University of Campinas (UNICAMP) for the technical support and CAPES for granting the scholarship. The authors also thank the Spectroscopy Laboratory (LABSPEC) – UEL and Instituto Nacional de Ciências e Tecnologias Analíticas Avançadas (INCTAA).
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (21)
- et al.
History, evolution, and environmental impact of biodiesel in Brazil: a review
Renew. Sust. Energ. Rev.
(2017) - et al.
Determination of physicochemical properties of biodiesel and blends using low-field NMR and multivariate calibration
Fuel.
(2019) - et al.
Monitoring the mechanism and kinetics of a transesterification reaction for the biodiesel production with low field 1H NMR spectroscopy
Fuel.
(2019) - et al.
Modelling phase behavior of triglycerides, diglycerides and monoglycerides related to biodiesel transesterification in mixtures of alcohols and CO2 using a polar version of PC-SAFT
Fluid Phase Equilib.
(2020) - et al.
Online monitoring of enzymatic hydrolysis of marine by-products using benchtop nuclear magnetic resonance spectroscopy
Food Control
(2020) - et al.
Kinetics of sunflower oil methanolysis at low temperatures
Bioresour. Technol.
(2008) - et al.
Monitoring of the ester production by near-near infrared thermal lens spectroscopy
Fuel.
(2019) - et al.
Compact NMR spectroscopy for real-time monitoring of a biodiesel production
Fuel.
(2015) - et al.
A PLS regression model using NIR spectroscopy for on-line monitoring of the biodiesel production reaction
Fuel.
(2011) - et al.
A method based on 1H NMR spectral data useful to evaluate the hydrolysis level in complex lipid mixtures
Food Res. Int.
(2014)
Cited by (14)
Successful combination of benchtop nuclear magnetic resonance spectroscopy and chemometric tools: A review
2023, Analytica Chimica ActaGreen and fast biodiesel production at room temperature using soybean and Jatropha curcas L. oils catalyzed by potassium ferrate
2022, Journal of Cleaner ProductionCitation Excerpt :The characteristic peak area corresponding to the methyl ester protons (OCH3) appears at 3.62–3.63 ppm. These peaks can be observed in both JCO and SBO FAMEs but are not detected in the vegetable oil (Galvan et al., 2020). Likewise, Fig. 7 (b) and 7 (d) reveal that the glyceride signal (CHCH2 proton coupling) usually appearing at 4.07–4.23 ppm is completely absent (referenced from the SBO and JCO oils NMR spectra in Fig. 7), indicating that yield is similar to conversion described in Table 2, with high purity of the reaction FAMES for both oils (Ng and Yung, 2019).
Fluorescence spectroscopy in tandem with chemometric tools applied to milk quality control
2022, Journal of Food Composition and AnalysisCitation Excerpt :Thus, it was decided to employ a discriminant classifier technique and another one-class classifier to verify the classification performance of milk samples adulterated with whey. The PLS-DA algorithm emerged from the PLS regression, and the main difference between PLS and PLS-DA is the Y matrix (Galvan et al., 2020b; Vieira et al., 2020). The Y matrix in PLS contains the values of the property of interest, while in the PLS-DA matrix, Y includes information on the sample classes.
Calibration with virtual standards for monitoring biodiesel production using a miniature NIR spectrometer
2022, TalantaCitation Excerpt :Several analytical techniques can be applied to quantify the compounds of the reaction medium during transesterification. With the aim of enhancing process monitoring and control, taking advantage of analytical sensors that provide fast and reliable responses, spectroscopic methods such as infrared [6,11–14], low-field proton nuclear magnetic resonance (1H NMR) [10,15] and Raman [16,17] have been applied to monitor transesterification reactions. These methods are non-destructive and suitable for in situ applications through the use of fiber optic probes, and easily adapted for process analysis.