Perspective of reactive separation of levulinic acid in conceptual mixer settler reactor

Levulinic acid is a carboxylic acid present in industrial downstream. It is an important chemical and can be transformed into various important chemicals such as 1,4-pentanediol, aminolevulinic acid, succinic acid, gamma valarolactone, hydoxyvaleric acid, and diphenolic acid. It is considered one of the top ten most important building block chemicals and bio-derived acids. Levulinic acid can be directly produced using biomass, chemical synthesis, and fermentation processes at industrial and laboratory scales. The biomass process produces the char, whereas the fermentation process generates waste during the production of levulinic acid, leading to an increase in the production cost and waste streams. The separation of levulinic acid from the waste is expensive and challenging. In the present study, reactive extraction was employed using trioctylamine in i-octanol for the separation of levulinic acid. The experimental results were expressed in terms of performance parameters like distribution coefficient (0.099–6.14), extraction efficiency (9–86%), loading ratio (0.09–0.7), and equilibrium complexation constant (11.34–1.05). The mass action law model was also applied and found the predicted values were in close agreement with the experimental results. The mixer settler extraction in series was used to achieve more than 98% separations of acid. Furthermore, the conceptual approach for separation of levulinic acid using a mixer settler reactor scheme was discussed and presented various design parameters including extraction efficiency, diffusion coefficient, equilibrium complexation constant, and loading ratio. The study is helpful in recovering the valuable chemicals present in industrial downstream and reducing their environmental impacts if any.


Introduction
Many industries discharge their downstream wastes into the environment, leading to the pollution of soil, water, and air. Some of the valuable chemicals and biochemicals are also present in the industrial waste stream and fermentation broth. Therefore, the recovery of these chemical compounds becomes important and necessary due to their harmful effects on the environment. The fermentation broth and industrial downstream contain a variety of carboxylic acids including levulinic, lactic, glutaric, aspartic, vanillic, and protocatechuic acid. These acids are very important due to their significant applications in fuel, medicine, food, plastics, cosmetics, chemicals, and fertilizer.
Levulinic acid (LA) is used in the manufacture of various daily need items. It can be used as a raw material in a variety of industries, including plastics, polymers, pharmaceuticals, cosmetics, food, agriculture, biofuel, chemical solvents, and rubber manufacturing (Andersson-Engels et al. 1995;Rackemann and Doherty 2011;Pasquale et al. 2012;Zhang et al. 2012;Mthembu 2015). It can also be employed as a platform chemical for the conversion of high-value products like methyltetrahydrofuran (MTHF) as fuel additive; δ-aminolevulinic acid (DALA) as a broad-spectrum herbicide/pesticide; and diphenolic acid (DPA) as an alternative to bisphenol A in the polymer industry.
LA known as keto acid (C 5 H 8 O 3 ) is a ketonic group compound with good potential of replacing the petroleum-based stocks in the chemical and biofuel industries (Zion Market Research Report). It is a colorless, small-chain fatty acid soluble in acidic ethanol, diethyl ether, and water. LA is a common substrate in chemical reactions such as condensation, esterification, halogenation, hydrogenation, oxidative dehydrogenation, and salt formation (Zhang et al. 2020). It is a versatile and potential building block chemical in the synthesis of various organic products (Çelebican et al. 2020). Various fuel and chemical products can be obtained using LA. It is considered one of the top twelve organic chemicals by the Energy Department of the USA (Holladay et al. 2004). Its permeation enhancers are very safe and effective. Levulinic acid can also be obtained from cellulosic biomass, grass, and wood chips, and has been regarded as a basic premium material due to its high chemical reactivity. GF Biochemicals Ltd. is the word's largest producer of levulinic acid. Hebei Yanuo, Zibo Changlin Chemical, Heroy Chemical Industry, Hefei TNJ Chemical, and Langfang Triple well Chemical have also started the production of levulinic acid using biomass as a raw material in recent times. The rice straw, sugarcane bagasse, rice husk, wheat straw, corn straw, and other agricultural wastes contain sucrose, glucose, xylose, and fructose that can be converted to levulinic acid (Kumar et al. 2019).
According to the industry report, the market for levulinic acid can be predicted to increase at a CAGR of 6% from 2019 to 2024. Increasing demand for levulinic acid in the plasticizer, pharmaceutical, and cosmetic sectors is the primary factor for its larger market. According to economic projections, the levulinic acid market will be worth USD 71.85 million by 2027, with an annual growth rate of 14.10% from 2020 to 2027.
Various industries are using direct biomass as a raw material for the production of valuable carboxylic acid. They use many toxic chemicals such as sulphuric acid, hydrochloric acid, and nitric acid as the catalysts during the production of biochemicals and discharge them downstream causing environmental pollution. Therefore, it is inevitable to separate them to decrease environmental pollution.
To remove and recover the carboxylic acids from aqueous streams, a variety of separation techniques have been used, including adsorption, ion exchange, extraction, ultrafiltration, precipitation, electrodialysis, distillation, membrane separation, reverse osmosis, and nanofiltration (Inci and Aydin 2003;Lalikoʇlu et al. 2015). Among these techniques, extraction (Brouwer et al. 2017), nanofiltration, electrodialysis (Kim et al. 2013), and adsorption (Datta and Uslu 2017) have been employed to separate LA from aqueous solutions. These processes are expensive, cumbersome, and produce waste material during the process. Some advantages and disadvantages of various separation processes are listed in Table 1. Comparatively, the commercial separation of LA is a less expensive, simple, and environmentally and ecologically benign process as it produces less waste material in the reject streams. In reactive extraction, acid molecules in the organic phase form complexes with organic molecules, allowing for enhanced acid separation.
Several studies have looked into the reaction of trioctylamine (TOA) with active or inert diluents to extract levulinic acid Ismail 2008a, 2008b;Uslu and Ismail 2008b;Kumar et al. 2011Kumar et al. , 2015Kumar et al. , 2021Eda et al.2018). Datta et al. (2016) studied the separation of LA using TBP and TOA with 1-octanol and found that TOA is the best extractant compared to the TBP. Kumar et al. (2015) studied and used various diluents like decane, decanol, decane + decanol, toluene, MIBK, and DCM with TOA for the separation of LA.
The goal of this research is to remove levulinic acid from industrial waste to reduce pollution. In this work, i-octanol was combined with TOA to discover the best conditions for maximising LA extraction efficiency. This optimal data was used to develop a continuous levulinic acid separation Reactive extraction -Control of the pH in the reactor -Solvents can be used -Higher concentration product -Required addition separation method Li et al. 2006) process that resulted in lower energy usage, waste creation, and higher extraction efficiency. The mixer settlor reactor was employed in series for a continuous process, and the recovery of LA from the aqueous phase was determined to be 98%. In this situation, mass action law was also used to elucidate the mechanism of the liquid-liquid equilibrium. The present study results can be used to design recovery equipment and reduce the environmental effect.

Experimental procedure
In the experimental studies, the initial concentration of LA in the aqueous phase was 0.1-1.0 mol.L −1 . Here, a lower concentration of acid was selected because of the occurrence in the fermentation broths and industrial waste streams. The organic phase was prepared using TOA as a reactive extractant (0-80% v/v) with i-octanol. The sample was taken an equal volume of both phase solutions in a 250-ml conical flask and shaken in an orbital shaker (S-24, REMI) for 4 h at a constant temperature of 298.15 K. After 4 h, the sample was taken out and centrifuged at 4000 rpm for 5 min in Remi Centrifuge R-4C to separate the aqueous and organic phase.

High-performance liquid chromatography
After phase separation, the aqueous phase was filtered by a syringe filter (0.45-micron) and diluted 100 times with double distilled water. The aqueous phase samples were analysed by using an Agilent 1200 High-performance liquid chromatography HPLC using a C-18 column with Refractive Index Detector (RID). The analysis was carried out using 5 mM H 2 SO 4 as the mobile phase, with a flow rate of 0.6 mL/min at 35 °C column temperature. The HPLC analysis of levulinic acid is shown in Fig. 1.

Fourier transform infrared analysis
Fourier transform infrared (FTIR) (Shimadzu, IRAffinity-1) was used to examine interpret the acid-extractant complex formation in the organic phase. All the measurements were carried out in a potassium bromide window cell. The acidextractant complexity was observed and examined in levulinic acid reactively extracted using TOA in i-octanol. In the initial aqueous phases, the comparative IR spectrum was obtained and analysed in conjunction with the organic phases of the balance condition concerning stoichiometry in the acid-extractant complex. The vibration peak at 1791 cm −1 represents the C = O stretching, strong carbonyl group (carboxylic acid); 1153 and 1053 cm −1 confirm the presence of C-N stretching of amine; and 728, 1352, 2866, and 2922 cm −1 confirm the C-H stretching of alkyl group. The FTIR spectra of the range 500-4000 cm −1 are shown in Fig. 2.

Results and discussion
Levulinic acid occurs in a dilute concentration in the waste streams after its production through the fermentation broth and synthesis process. Therefore, the sample solutions of levulinic acid were prepared with a 0.1-1 mol.L −1 concentration of LA in the aqueous phase. The organic phase containing TOA as reactive extractant with i-octanol was used.

Physical extraction
The un-dissociated molecules of LA among the aqueous and organic phases can be characterised as follows: Therefore, the partition coefficient (P) can be defined as: The dimerization constant of LA can be represented as: where [LA] 2.org and [LA org ] 2 are the LA concentration in the organic phase and dimerized form respectively.
The overall distribution coefficient is influenced by the concentration of H + ions in the solution as well as its ionic strength. Overall distribution coefficient can be written as: As the concentration of LA (0.1-1 mol.L −1 ) is very dilute, term K LA ∕[L + ] aq in denominator of Eq. (5) can be neglected. Hence, Eq. (5) becomes: Equation (6) can be expressed as follows: The results of physical extraction were found and represented in terms of distribution coefficient (0.099-0.681), dimerization coefficient (226), partition coefficient (0.051), and extraction efficiency (9-40.5%), respectively. The experiments were performed and found the maximum extraction efficiency of 40.5% with a distribution coefficient of 0.681 was observed at equilibrium. The partition coefficient (P) and dimerization coefficient (D) were evaluated with the linear regression method using Eq. (7).

Chemical extraction
The efficiency of physical extraction is not sufficient for separating LA from the industrial downstream. To improve K D values, the reactive extractant was used along with the lowdensity non-reactive diluent. During the experimental work, trioctylamine was used as a reactive extractant in i-octanol to form the organic phase. The trioctylamine (TOA) is less soluble in water and provides a higher separation efficiency of the acid. In TOA, the N-H bond carries more polarity than the C-H bond. Therefore, TOA has the potential to improve extraction efficiency by separating more acid molecules from the aqueous phase into the organic phase. The equilibrium isotherms between aqueous and organic phases are shown in Fig. 3.
As per the reaction mechanism, n molecules of TOA per molecule of levulinic acid take part in the reaction to form the acid-extractant complex, represented as: Theoretically, the mass action law can be used to estimate the equilibrium complexation constant (K E ). The equilibrium complexation constant, K E , can be written as: In the aqueous phase, LA dissociates upon reaching equilibrium, and is represented as: where K a is the dissociation constant. Concentration of LA in aqueous phase (C LA ), can be expressed as: Using Eqs. (11) and (12) The distribution coefficient, K D, and extraction efficiency, % can be determined as: Using Eqs. (13) and (14), K D can be written as: K D can be written as: [LA] (12) and n can be found using the intercepts, as well as the line's inclination.
The distribution coefficients (K D ) were determined and found in the range of 0.47-6.14 using TOA (20-80 vol %) in i-octanol. It was decreased with increase in the initial concentration of LA in aqueous phase. At lower concentrations of LA, a greater K D was achieved (0.1 mol.L −1 ). In the organic phase, a higher acid concentration results in a larger loading ratio. More acid molecules could interact with the extractant available for chemical extraction and forum a 1:1 and 2:1 as acid-extractant complex in the organic phase. The acid-extractant complexes were formed due to the strong interaction between amine and acid, resulting in higher distribution coefficients (Yang et al. 1991). The analysis of K D values reveals that TOA is a good reactive extractant in i-octanol for the separation of LA.
The extraction efficiency (η) was observed between 32 and 86% when using TOA in i-octanol. The η value was observed to increase with TOA concentration up to 40% TOA in i-octanol, and thereafter, it gradually decreases when the TOA increases in the organic phase (Fig. 4). It attributed to the governing synergistic effect of hydrogen bonding and polarity of i-octanol with acid molecules (Wasewar et al. 2004). Separation efficiency is dependent on the diluted physicochemical properties like solubility, molecular weight, dielectric constant, and refractive index (Kumar et al. 2020). As a result, i-octanol was combined with the reactive extractant TOA to increase the efficiency of LA separation.
The loading ratio, Z, is defined as the ratio of acid concentration in the organic phase at equilibrium to the initial concentration of reactive extractant in the organic phase. It was The Z values were observed between 0.06 and 0.7 using 20-80% TOA in i-octanol. The loading ratio depends on the strength of the acid-extractant interaction as well as the overall extraction equilibrium's stoichiometry (Kertes and King 1986;Keshav et al. 2009;Wasewar et al. 2011). The loading ratio decreases with the increase of reactive extractants in the organic phase because it enhances the availability of more favorable solvating agents (Yang et al. 1991). The loading ratio was found to be greater than 0.5, including acid-extractant complex formation of 1:1 and 2:1 in the organic phase. The equilibrium complexation constant, K E , affects the complex formation between LA and reactive extractant. It depends on the loading ratio, i.e. complex form (1:1) or (2:1) in the organic phase (Fig. 5). The following equation can be used to determine equilibrium complexation constant; K E : The complexation constant (K E ) was observed between 11.34 and 1.05 using TOA (0.457-1.83 mol.L −1 ) in i-octanol. The various estimated values of K E are specified in Table 3.
The values of n and K E were calculated by plotting log K D + log (1 + K LA /[L + ]) vs log [TOA] in which results in a straight line with a slope of n and an intercept of log K E . The graphical regression can be used to calculate the values of K E as n for various extraction systems (Fig. 6). K E defines the extent of separation of acid; if K E is less than one, it indicates that the product has a poor separation efficiency compared to the reactant, but if K E is greater than one, it indicates that the acid has a high separation efficiency. The range of values of K E and n was found to be (12.40-0.99) and (0.35-0.99) respectively, with different concentrations of TOA and LA.
The experimental K E values are near to the mass action law predicted values. Thus, the mass action law can be considered a good model for representing the separation of levulinic acid.

Design of mixer settlor reactor
According to the experimental results and the analysis, the process needs to be designed with process intensification. The mixer settler reactor can be used for the separation of 100% LA from the aqueous phase. The optimum experimental data was used for designing the conceptual mixer settler reactor in series to attain 98% recovery of LA from the organic phase.   Figure 7 represents the conceptual mixer settlor reactor connected in series. The inlet concentrations of levulinic acid are [LA] inlet1 and [LA] inlet2 for the first and second reactors respectively. The outlet concentration of acid in the organic phase can be calculated using the mass balance around the first mixer settler reactor given as follows: [LA] aqoutlet1 Similarly, the material balance around the second mixer settlor could provide the [LA] aq 0.0019 mol.L −1 . Thus, the overall efficiency of levulinic acid could be achieved as 98%.
The first mixer settler reactor provides a distribution coefficient of 6.14 with an extraction efficiency of 86%. The overall distribution coefficient and extraction efficiency could be achieved as 58.02 and 98%, respectively, with the second mixer settler reactor. The conceptual mixer settler reactor and the recommended parameter are described in Fig. 7. Thus, the mixer settlor reactor increases the separation efficiency, leading to a negligible acid concentration in the waste downstream.

Back extraction
The separated levulinic acid needs to be back-extracted from the loaded organic phase through the different regeneration process like using HCl, NaOH, temperature swing, and diluent swing methods Wang 1991a, 1991b;Tamada et al.1990;Tamada and King 1989). In the regeneration process, the acid-laden aqueous phase and the acid-free organic phase can be contacted with a fresh aqueous stream to produce the acid-laden aqueous phase and the acid-free organic phase for recycling to the extractor.

Environmental issues
The industries use toxic acids as catalysts, viz., H 2 SO 4 , HCl, and HNO 3 in the manufacturing of valuable levulinic acid. In the production of acid, the chemical industry affects the environment and lifestyle as well. Industrial activities result in wastewater pollution, gas emissions, and pollution of underground water and natural ponds, etc. Therefore, an industrial downstream requires the separation and recovery of such valuable solutes and their management to control such types of environmental pollution. The mixer settler reactor and its design can play an important role in recovery of the LA and preventing environmental pollution.

Conclusion
The main obstacles in dealing with industrial waste streams, fermentation broth, and aqueous streams can be addressed using the reactive extraction method to separate the LA. The experimental data suggest that reactive separation is an effective and flexible process to recover the valuable LA compared to the other methods like membrane, adsorption, and distillation. The overall results conclude that an enhanced separation efficiency of 86% was achieved using the optimum 40% TOA in i-octanol with an initial concentration of 0.1 mol.L −1 of LA. The predicted values were found to be in good agreement with the experimental values when the mass action law model was applied to the experimental results. Hence, the mass action law model can be applied to predicate the various parameters for designing the system of reactive extraction for the separation of LA. The mixer settler reactor was found to be an excellent option in the design of the extraction system to achieve the recovery of LA as high as 98%. Moreover, the conceptual mixer settler reactor in series carries with zero waste and minimum energy usage during the recovery process of LA. The present study can be very useful to the scientists and engineers in designing the extraction system for separation of LA and reducing its environmental impact.