|
Description
|
Recovering these carboxylic acids from the fermentative streams in a sustainable, green, and economical way is a significant challenge. This work assessed hydrophobic eutectic solvents (HES) – water-immiscible – for the selective recovery of carboxylic acids via liquid-liquid extraction. Different trioctylphosphine oxide (TOPO) mixtures with menthol and thymol were studied and deeply characterized by 1H and 31P NMR to yield stable eutectic solvents including, a novel experiment of 31P-NMR at variable temperatures for the first time. Those stable eutectic solvents were tested in the liquid extraction of complex aqueous mixtures containing C2-C6 carboxylic acids and simple sugars (glucose and xylose). The back-extraction of the carboxylic acids for the recovery of the HES was optimized, being necessary in three stages for the complete cleaning of the eutectic solvent using NaOH 0.1 M. The eutectic mixture of TOPO and thymol in a molar ratio of 1:2 exhibited an overall recovery of C5 and C6 carboxylic acids over 70 %, allowing its selective extraction from the rest of the compounds in the complex mixture. Likewise, this HES (after back extraction) was successfully reused in a second extraction cycle, keeping the performance of the fresh one. Therefore, in this study it was demonstrated that the use of HES, it was able to extract with a high selectivity carboxylic acids of (≥ C5), besides being very stable solvents these allowed reusability reducing the environmental impact and process costs. (2025-02-05)
Methodology Materials All chemicals were used as received without further purification from the supplier Sigma Aldrich. Trioctylphosphine oxide (TOPO) ≥ 99.0%, Thymol ≥ 98.5%, and Menthol 99% were used to synthesize the HES. Acetic acid 99.0%, Propionic acid ≥ 99.0%, Butyric acid ≥ 99.0%, Valeric acid ≥ 98.0%, Hexanoic acid ≥ 99.0%, Fumaric acid ≥ 99.0%, Glucose ≥ 99.0% and Xylose ≥ 99.0% were used to prepare synthetic mixtures of carboxylic acids. Synthesis and characterization of HES In this study, six mixtures were synthesized with a different molar ratio of TOPO to menthol and thymol: 1TOPO:1Thymol (1T:1T), 1TOPO:2Thymol (1T:2T), 2TOPO:1Thymol (2T:1T), 1TOPO:1Menthol (1T:1M), 1TOPO:2Menthol (1T:2T) and 2TOPO:1Menthol (2T:1M). Thymol and menthol were selected because of their availability, natural source, and good performance in extracting single components 33. The following synthesis method was a thermal procedure as described in literature 25. The temperature selected was 40 ºC, and if the mixture did not reach the liquid state, the temperature was raised to a maximum of 100 ºC, and the temperature was kept for 1 h. Afterward, the resultant mixtures were allowed to cool naturally at room temperature and kept isolated from the light to prevent degradation for 24 hours. Those mixtures that remain in a liquid state after cooling and storing for 24 hours were selected for the extraction assays. Otherwise, they were discarded. Because of the aqueous nature of the extraction feed phase, it is essential to ensure that HES is hydrophobic. To demonstrate the hydrophobic nature of the synthesized hydrophobic eutectic solvents (HESs), they were subjected to a one-hour mixing with water, followed by decantation. Subsequently, using the Karl Fisher titration method, the water content in each HES was measured. The results showed values of less than 1 g/L of water, indicating the immiscibility of HESs in water. Furthermore, analysis of the extracted aqueous samples using HPLC, in both the extraction and the back-extraction, revealed no signals of TOPO, menthol, or thymol. This observation provides further evidence of the absence of HES degradation into the aqueous phase during extraction processes. The formation of the HES was monitored using nuclear magnetic resonance (31P and 1H-NMR). NMR has also been used to explore the interactions of HES molecules with the different compounds to extract. The experiments were conducted using a Bruker Advance NEO 9.4 T spectrometer (400 MHz 1H resonance frequency) with a 5 mm probe. The 31P resonance frequency was 161.99 MHz, and a single pulse sequence with 1H decoupling. The Press delay (D1) was 30 seconds for all the experiments. In the case of the experiments at variable temperatures, variations of 10 ºC were used. 31P reference was triphenyl phosphate in acetone-d6 (TPP) at -17.59 ppm. 31P-NMR experiments at variable temperatures were conducted under two modes: (1) an experiment increasing the temperature from 10 ºC to the temperature formation established when the synthesis was optimized, i.e., 40 ºC (in intervals of 10 ºC), and final increase of the temperature up to 60 ºC to check the stability of the molecule’s interaction; and (2) decreasing the temperature from 60 ºC to 20 ºC (in intervals of 10 º C) to prove the stability of the molecule’s interaction with the temperature variation again. In addition, a control experiment of TOPO was also carried out to make sure the signal offset was not due to the temperature variation, as is usually the case in NMR techniques at variable temperatures 43. Liquid-Liquid Extraction Aqueous solutions of acetic, propionic, butyric, valeric, hexanoic, pyruvic, and fumaric acids were prepared individually and all together with a concentration of 1 g/L each. Additionally, glucose and xylose were added at 1 g/L for some experiments to make the mixture more representative of the composition of a real fermentation effluent. In real fermentation effluents, the concentration of each acid can vary significaly based on various operating conditions including the substrate type, inoculum, pH, temperature and retention time 44–46. To maintain consistency and avoid potential saturation of components in hydrophobic eutectic solvents (HESs) , it was opted to establish an average concentration of 1 g/L for all the compounds in the mixture. Furthermore, it’s essential to note that the primary focus of this study is to understand and determine the key factors influencing the interaction of individual components or their collective presence. HES and aqueous solution were introduced in a 20 mL Ace pressure glass and stirred at 300 rpm at 30 ºC for different times, with feed/HES volume ratio of 1:1. Both phases were separated by pouring in a settling funnel and allowing a decantation time until a clear separation between phases was observed. The concentration of different compounds in the aqueous phase (raffinate) was analyzed by High-Performance Liquid Chromatography (HPLC) in an Agilent 1260 Infinity apparatus equipped with a refractive index (RID)-G13662A as detector and a Hi-Plex H+ column (300 x 7.7 mm) from Phenomenex. The extraction yield (YE) and distribution coefficient (Ki) were calculated as shown in equations (1) and (2). Y_E (%)=(C_i^F-C_i^R)/(C_i^F )×100 Eq. 1 Where C_i^F refers to the initial concentration of the compounds in the aqueous phase (feed) and C_i^R is the component concentration in the aqueous phase (raffinate) after the extraction. The distribution coefficient was defined as the concentration of solute in the organic solvent divided by the concentration of solute in the aqueous phase 47, when the equilibrium is reached. K_i=(C_i^E)/(C_i^R ) Eq. 2 Where C_i^E, refers to the concentration of the compounds in the HES phase after the extraction (extract), and C_i^R is the compound concentration in the aqueous phase (raffinate) after the extraction and under equilibrium conditions. On the other hand, Fourier-Transform Infrared Spectroscopy (FTIR) analyses of the HES samples after extraction were carried out to determine whether the structure of the solvents is maintained after extraction and whether water is being absorbed by the HES Fourier-Transform Infrared Spectroscopy (FTIR) analysis were carried out of the HES after use. A Nicolet 5700 spectrometer apparatus equipped with an iS50 ATR device was used for measurements in the range of 500-4000 cm -1. 2.4 Recovery of carboxylic acids from the HES by liquid Phase Extraction (Back-extraction) The recovery of the previously extracted acids was done through a new extraction stage by putting the HES enriched in the acids of interest in contact with an alkaline solution, which was enriched with the acids, leaving the HES "clean" for its subsequent reuse. This second extraction stage was called "back-extraction" . The back-extraction yields (YBE) were calculated as shown in Equation 3. Y_BE (%)=(C_i^E')/(C_i^F' )×100 Eq. 3 Where, C_i^E', refers to the compound’s concentration in the alkali phase after the back- extraction (extract’) and C_i^F', refers to the compound’s concentration in the HES before the back-extraction (feed’) which is the same as the C_i^E (Figure 1). Different types of alkali solutions (NH4, NaOH, and Na2CO3) were studied for the back extraction step since basic agents have previously been used successfully in the extraction of carboxylic acids in both DES 13 and ILs 24. Since we were interested in obtaining a HES that is as "clean" as possible to be suitable for reusing, this "back-extraction" step must be optimized to ensure that practically no carboxylic acid remains. For this purpose, we optimized the following extraction variables: the type of alkaline agent, the concentration of the alkaline agent, the extraction time, the alkali/HES volume ratio, and the number of required steps. Finally, the overall extraction yield (Y) was calculated considering the initial loading of compounds in the raw feed (CiF) and that obtained in the recovered aqueous solution after the back extraction (CiE`), as shown in equation 4 (Figure 3). Figure 3. Complete extraction system for the recovery of carboxylic acids. Y (%)=(C_i^(E^' ))/(C_i^F )∙100 Eq. 4
|
