Experimental Study on Biodiesel Production in a Continuous Tubular Reactor with a Static Mixer

Abstract

This research on biodiesel production aims to improve energy processes to advance towards a sustainable economy. This study focuses on improving the biodiesel conversion efficiency in a helical tubular reactor coupled with a static mixer. A 2³ factorial design was used to evaluate how variables such as the molar ratio of alcohol–oil (4:1–8:1), residence time (4–8 min), and catalyst concentration (0.5–1 wt%) affect the transesterification process. Soybean oil and methanol were used, with NaOH as a catalyst at 60 °C. The results show that the residence time and catalyst concentration are key factors in increasing biodiesel production by up to 10%. An experimental yield of 84.97% was obtained with a molar ratio of 6:1 alcohol–oil, 0.9 wt% NaOH, and a reaction time of 6 min. The experimental design predicted a yield of 91% with a molar ratio of 4:1 alcohol–oil, 1 wt% NaOH, and a reaction time of 8 min, with a deviation of 1.88% from the experimental values. The fit of the experimental model was R² = 0.9632. These findings are valuable for improving the transesterification process and the development of biodiesel in continuous flow reactors.

1. Introduction

Research on biofuel production is gaining increasing attention due to its advantages. They are considered to be sustainable and renewable fuels, as well as economically viable. Furthermore, their appeal lies in the wide variety of resources that can be utilized in their production [1]. One of these alternatives is the use of liquid biofuels, such as biodiesel, which is considered a green fuel, biodegradable, has excellent combustion performance, and very low emissions to the environment [2,3]. Biodiesel production is influenced by the feedstock, the type of catalyst, and operating costs. However, the use of certain feedstocks can create competition with the food industry, as vegetable oils are the most commonly used for biodiesel production [4,5]. The most commonly used method for producing biodiesel is transesterification, which involves a chemical reaction between vegetable oils or animal fats and alcohol, using an alkaline catalyst to accelerate the process [6,7]. The result of the transesterification reaction is a mixture of long-chain fatty acid methyl esters (FAMEs) and glycerol [8].

The transesterification reaction can be performed in different types of reactors for large-scale biodiesel production. A reactor can operate continuously or batch-wise; batch reactors are widely used in the industry, although their production is limited by the reactor size and the extended downtime between batches. Additionally, they face challenges such as low biodiesel conversion rates, high energy consumption, and increased waste generation, which undermine the sustainability of the process [9]. Continuous reactors are an ideal option for large-scale biodiesel production, as they provide greater efficiency compared to batch reactors, while also minimizing downtime and waste generation. However, their main challenge lies in the high initial costs, driven by the specialized design and equipment required for their operation [10]. In recent years, some researchers have reported that the use of continuous tubular reactors improves both the conversion and the quality of the biodiesel produced, and it also reduces operational and energy costs. It is also important to mention that proper instrumentation can facilitate the automation of the process [11,12,13]. In this type of reactor, the transesterification reaction takes place inside the tubes at a constant flow rate. The longer the tube, the greater the mixing and residence time. Additionally, the incorporation of mixers, whether static, mechanical, or other injection devices, ensures a more efficient mixing process, which improves mass and heat transfer, reduces reaction times, and allows for a higher percentage of biodiesel conversion [14,15].

Sungwornpatansakul et al. [16] conducted a comparative study between a mechanical mixer with a water bath, agitation, and a capacity of 3 L and a static helical mixer measuring 26 cm in length and 0.8 cm in internal diameter, with a volumetric flow rate of 0.125 L/s. The transesterification reaction was carried out using pure corn oil and methanol and catalyzed with potassium hydroxide (KOH). The results showed that at a temperature of 30 °C, with a molar ratio of 6:1 of methanol–oil and a KOH concentration of 1.8 wt%, a yield of 98% FAME was achieved in 15 min using the static helical mixer, while the yield in the mechanical mixer was 95% under the same conditions. They concluded that the static mixer proved to be more efficient as it facilitates a greater acceleration in the mixing of methanol droplets and oil at the start of the mixer. However, the tests were conducted exclusively at a temperature of 30 °C, an operational variable that directly impacts the transesterification process, which could represent one of the main limitations of this study. Somnuk et al. [17] conducted an optimization study using response surface methodology to obtain methyl esters, employing helical static mixers as continuous reactors. The process occurs in three stages: stages 1 and 2 correspond to esterification, while stage 3 focuses on transesterification, achieving methyl ester conversions of 70%, 95.94%, and 99.96%, respectively, in each stage. This last conversion percentage was achieved using 115% methanol, 13.5% H₂SO₄, and 5 g/L of KOH. The optimal operating conditions for transesterification are 13% methanol, 5 g/L of KOH, a length of 0.7 m for the helical static mixer, and a reaction time of 8 min, resulting in biodiesel that meets EU standards. However, one of the concerns is the excessive use of alcohol in the transesterification process, as well as the production of wastewater, which would increase operating costs. Huang et al. [18] designed a new bioreactor with a static mixer for biodiesel production through an enzymatic process. The optimization was carried out using computational fluid dynamics analysis. The optimal reaction conditions included a static mixer length of 30 cm, a flow rate of 1.77 m/s, a temperature of 40 °C, and a molar ratio of 3.4:1 alcohol–oil, with 10% by weight of lipases. The biodiesel yield was 81.28% in 1.5 h; in contrast, under the same operating conditions, the bioreactor with the static mixer achieved the same yield in just 30 min, confirming greater mass transfer efficiency thanks to the static mixer. One of the main limitations of bioreactors lies in the use of microorganisms, as they require careful handling to prevent contamination and maintenance issues. Additionally, this increases operational costs and extends the production times for biodiesel.

Regarding continuous production in tubular reactors with static mixers, some authors have reported the integration of micromixers to intensify biodiesel production [19]. A study conducted by Baydir and Aras [20] carried out several experiments to analyze biodiesel production in narrow-channel tubular reactors, using T-type micromixers with different inner diameters. Transesterification was performed with methanol and sunflower oil in a 6:1 ratio, using KOH as the catalyst. They reported a conversion of 99.8% biodiesel with the micromixer with an 8 mm inner diameter and a reactor with a diameter of 1 mm in a time of 2 min. In comparison, the 1.2 mm micromixer, adapted to the 1.5 mm reactor, achieved a conversion of 95%. They concluded that the use of micromixers with small diameters increases the biodiesel conversion by 12% to 15% with a concentration of 1 wt% KOH. Additionally, they observed that the conversion percentage decreases as the residence time increases, which is attributed to the saponification effect. However, microreactors face challenges regarding scalability, with one of their main drawbacks being their limited capacity to produce biodiesel in large volumes. On the other hand, there are several proposals that include different types of mixers and micromixers, both static and mechanical, adapted to continuous tubular reactors. These are investigated to gain a deeper understanding of the transesterification process and to improve the principles of conventional mixing. The hydrodynamics of the fluid, the impact of catalytic activity, and specifically the interaction between the oil and alcohol are also analyzed [21,22].

Based on the literature review, tubular reactors with static mixers optimize or intensify biodiesel production. However, most of them are associated with micro-flows or microstructure geometric designs, which increase investment costs and reduce production capacity. Therefore, this study presents the design of a helical tube reactor, a geometric configuration that has been little studied for biodiesel production. The objective of this study was to generate biodiesel through the transesterification process in a helical continuous tubular reactor coupled with a static mixer. The effects of the residence time, catalyst concentration, and molar ratio between alcohol and oil were also examined, as these factors influence the biodiesel conversion yield. The evaluation of these effects was conducted using a factorial design of experiments.

2. Experimental Setup and Procedure

2.1. Materials and Methods

The experimental setup is shown in Figure 1a. It consists of a 6 m long helical tubular reactor made of stainless steel with an inner diameter of 4 mm. Due to its geometry, it has a compact design that promotes good mixing and enhances mass and heat transfer. Additionally, it is connected to a 2 m long static mixer made of plastic with an inner diameter of 12.7 mm (Nordson brand), which improves mixing and reduces the residence time (Figure 1b). The transesterification was carried out using commercial edible soybean oil provided by the agro-industrial company APECSA, chosen for its wide availability and relatively low cost, as it is a domestically produced product in Mexico; additionally, its low impurity level and high fatty acid content enhance the efficiency of the transesterification reaction. The alcohol used was methanol with a purity of 99.98%, from the J.T. Baker brand (Phillipsburg, NJ, USA), and sodium hydroxide from the Fagalab brand (Sinaloa, Mexico) was used as the catalyst. The flow to the reactor was supplied by two Aquatrol (ADP-01141) dosing pumps, supported by a heating and agitation system from Topline Lab (SH-2). The temperature was measured with Danoplus digital thermometers, and glassware was used, including 1 L separation funnels, graduated cylinders of 100, 250, 500, and 1000 mL, and 1 L beakers from the Kimax brand (Millville, NJ, USA). Finally, the biodiesel samples were analyzed using infrared spectroscopy with a Shimadzu FT-IR IRAffinity-1S spectrometer (Kyoto, Japan).

2.2. Production of Biodiesel

The transesterification process was carried out in a helical reactor coupled with a static mixer, as represented in Figure 2. In the diagram, two tanks with temperature sensors are observed: one contains commercial soybean oil and the other contains methanol catalyzed with NaOH (methoxide). Each tank is connected to a dosing pump that feeds a preheating coil before entering the static mixer. Subsequently, the mixture of methoxide and oil enters the tubular reactor. At the end of the transesterification process, the resulting flow enters a separation funnel, where two immiscible fluids are separated by decantation: biodiesel at the top and glycerol at the bottom. Finally, the sample was weighed to determine the biodiesel yield, evaluated using Equation (1) [23]. The preheating coil and the helical reactor are submerged in a tank of water and placed on a heating and stirring plate that maintains a temperature of 60 °C. This temperature is the most commonly studied in the literature for this type of transesterification process [24].

$$Biodieselyield\left(\%\right)={\displaystyle \frac{weightofproducedbiodiesel\left(\mathrm{g}\right)}{Sampleweight\left(\mathrm{g}\right)}}(100\%)$$

(1)

2.3. Statistical Analysis

All experiments were conducted randomly and in triplicate using a 2³ factorial design (2 levels and 3 factors) with a central point to analyze the influence of operating conditions on the transesterification process. The levels of the factors were selected based on the operating limits established in the preliminary tests, and the factors evaluated were the molar ratio, NaOH concentration, and residence time in the production of biodiesel, variables that, according to other studies, have the greatest impact on the transesterification process, along with temperature [25]. The experimental results were analyzed using an ANOVA (p = 0.05) with Design-Expert^® software (version 13, Stat-Ease Inc., Minneapolis, MN, USA). The factorial design for the percentage of biodiesel conversion is detailed in

Table 1. Results of the 2³ factorial design for the FAME conversion %.

Run	Molar Ratio	Catalyst (wt%)	Residence Time (min)	FAME a Conversion (%)	FAME b Conversion (%)
1	4:1	0.5	4	65.53 ± 2.04	65.81
2	8:1	0.5	4	68.21 ± 1.95	68.62
3	4:1	1	4	73.68 ± 2.99	73.68
4	8:1	1	4	76.57 ± 1.77	76.98
5	6:1	0.75	6	84.26 ± 2.29	84.26
6	4:1	0.5	8	85.83 ± 2.47	86.03
7	8:1	0.5	8	88.42 ± 2.70	88.20
8	4:1	1	8	90.01 ± 1.88	91.17
9	8:1	1	8	89.12 ± 1.22	89.57

^a Experimental data. ^b Estimated data.

.

The evaluation of the influence of each of the factors was carried out through the following equation:

$${\beta }_i=\left(y_i^{+}\right)-\left(y_i^{-}\right)$$

(2)

where ${\beta }_i$ is the effect of the ith factor on the response, and $y_i^{+}$ and $y_i^{-}$ are the mean responses for the upper (+) and the lower (−) levels of the ith factor. The evaluation of the interaction of the factors and the estimated calculation of the response can be determined using the following equation:

$$Y={\beta }_0+\sum {\beta }_i{x}_i+\sum {\beta }_{ij}{x}_{ij}+{\beta }_{ijk}{x}_{ijk}+\epsilon$$

(3)

where Y is the estimated response, ${\beta }_0$ is the general mean, $\sum {\beta }_i{x}_i$ is the sum of the main effects of the factors, $\sum {\beta }_{ij}{x}_{ij}$ is the sum of the effects of the interactions of two factors, ${\beta }_{ijk}{x}_{ijk}$ is the effect of the interaction of the three factors, and ε is the summation of the error in the model.

3. Results

Several preliminary experiments were conducted to evaluate the effect of the catalyst and residence time at a temperature (T_Re) of 60 °C, with a molar ratio of 6:1 methanol–oil, to obtain FAME in a continuous helical tubular reactor coupled with a static mixer. The operating conditions of these experiments are detailed in

Table 2. Operating parameters.

Parameter	Value
Temperature (°C)	60 (±0.5)
Residence time (min)	2, 4, 6, 8, 10
[NaOH] (wt%)	0.3, 0.6, 0.9
Molar ratio	6:1

. To select the appropriate variables in the factorial design, preliminary tests were conducted under the aforementioned conditions, which are widely studied by other authors [26]. Figure 3 shows the percentage of FAME conversion at different catalyst concentrations and reaction times. The highest conversion values, reaching 85%, were obtained at both 6 and 8 min, with a concentration of 0.9 wt% NaOH. In contrast, at 2 min, the conversion was 54.5%, which is attributed to a higher flow rate that limits adequate mixing [27]. On the other hand, at 10 min, a slight decrease in biodiesel conversion is observed, possibly due to the products reacting back with the oil or because the reaction rate decreases due to the depletion of the reactants [28]. The results of the operating conditions evaluated through the factorial design for the percentage of biodiesel conversion are summarized in Table 1. In all tests, samples were taken, removing the glycerol phase, and the FAME phase was washed with distilled water to eliminate impurities. The FAME samples were analyzed using infrared spectroscopy, showing the characteristic spectra of biodiesel. Figure 4 presents an FT-IR spectrum of the biodiesel produced from soybean oil and methanol, using NaOH as the catalyst. It shows the composition of the functional groups present in the FAME. In the region of 2800–3000 cm⁻¹, characteristic peaks of symmetrical and asymmetrical C-H bonds are observed. A peak at 1748 cm⁻¹ is notable, corresponding to the carbonyl group of the ester (C = O). The region between 1400 and 1500 cm⁻¹ indicates the presence of methyl ester groups, while the signal around 1175 cm⁻¹ is associated with the asymmetry of the C-O-C bonds. All these signals are characteristic of the conversion of vegetable oils into biodiesel [29,30,31].

The ANOVA results showed that the effect of the studied factors was significantly different from zero (F test, p ˂ 0.0001), while the lack of fit in the model was not (F test, p ˂ 0.067). The central point does not show curvature in the model, which does not synergistically influence the response; the model is appropriate for this design [32]. The analysis of the most significant effects can be represented by the following equation:

$$\begin{array}{cc}Biodiesel\left(\%\right)=& 33.95-0.032MolarRatio+16.36Catalyst\hfill \\ & +5.41Residencetime+1.79Molarratio\ast Catalyst\hfill \\ & +0.15Molarratio\ast Residencetime-0.39Catalyst\hfill \\ & \ast Residencetime-0.38Molarratio\ast Catalyst\hfill \\ & \ast Residencetime+\epsilon \hfill \end{array}$$

(4)

This equation allows for the analysis of the effects of each parameter and their interactions on the response, as a positive sign in front of the term indicates a synergistic effect, while a negative sign indicates an antagonistic effect.

Figure 5 shows a residual distribution plot, where the values of the residuals are evaluated against those predicted by the model. The residual distribution does not follow a trend and is below 3%, indicating that the model fits the biodiesel conversion within the studied experimental range.

Moreover, the model is adequate to describe the relationship between the values and the response. This can be seen in Figure 6, where the experimental values are compared with the responses predicted by the model, obtaining a coefficient of determination of R² = 0.9632, which indicates that the observed data fit the model [33].

Interaction plots are used to analyze the influence of the effects between each pair of variables on biodiesel production. Figure 7a–c show the average response value (FAME) at the upper (+) and lower (−) levels of the three studied effects. Figure 7a shows the interaction between the molar ratio and the concentration of the catalyst on the biodiesel yield. The yield is virtually unaffected by the molar ratio when the NaOH concentration is high (B+), but when the catalyst concentration decreases to 0.5 wt% (B−), a slight increase in yield is observed. Despite this, the conversion remains higher with a concentration of 1 wt% NaOH (B+). This effect occurs because transesterification is a reversible reaction; therefore, an increase in the amount of alcohol and the concentration of the catalyst can promote the generation of by-products, which reduces the FAME conversion [34]. Figure 7b shows the interaction plot between the molar ratio and residence time. The biodiesel conversion responds significantly to the residence time when it is at its upper level (C+), although no significant variation in the response is observed when the molar ratio is at its upper or lower level. The same effect was observed by Sakthivel et al. [35], who found that the biodiesel conversion does not increase with the molar ratio, but it is influenced by a longer residence time, up to a point, after which the yield begins to decrease. Meanwhile, with a lower residence time (C−), the biodiesel conversion slightly increases when the molar ratio is at its upper level (8:1). This highlights the utility of helical static mixers, which, thanks to their large contact surface, optimize mass and heat transfer, enhance transesterification, and contribute to the intensification of FAME production in relatively efficient timeframes [36]. Figure 7c presents the interaction plot between the catalyst and residence time. When the catalyst concentration is at an upper level (1 wt% NaOH), a very significant increase in biodiesel yield is observed, rising from 75 to 91.17% as the residence time increases from 4 to 8 min. Furthermore, it can be noted that the catalyst has less influence on the yield than the residence time, as when the latter is at an upper level (C+), the biodiesel yield varies from 87 to 91.17% when using 0.5 wt% and 1 wt% NaOH, respectively. This behavior occurs when free fatty acids react with a highly alkaline medium over extended periods, promoting soap formation. This is undesirable in static mixers, as it can cause blockages in the pipelines due to the accumulation of solid residues [37].

In addition to the interaction plots, the analysis uses surface plots to examine how the concentration of the catalyst and the molar ratio affect biodiesel production based on residence time, which is the most determining factor. Figure 8a,b present the response surfaces of biodiesel conversion in relation to the residence time, catalyst concentration, and molar ratio. Figure 8a shows the biodiesel yield as a function of the residence time and NaOH concentration. A very significant effect on biodiesel yield is observed, varying from 67 to 87% when the catalyst concentration is low (0.5 wt% NaOH) and the residence time is increased from 4 to 8 min, respectively. Furthermore, it is indicated that increasing the NaOH concentration (1 wt%) has a slightly positive effect on the biodiesel yield, reaching 91.17% when the residence time is increased to 8 min. The influence of the static mixer is evident, improving mixing, reducing glycerol production, and slightly increasing FAME conversion [36]. This finding has also been reported in other studies, where it is suggested that the use of potassium hydroxide (KOH) has a slightly greater impact on FAME conversion [34,38]. Figure 8b illustrates the response surface of biodiesel conversion in relation to the residence time and molar ratio, confirming that residence time is the factor that most influences biodiesel conversion. It has been observed that a variation in the molar ratio does not affect the chemical equilibrium of the transesterification reaction. Moreover, it has been reported that the mass transfer between alcohol and oil decreases if good mixing is not achieved, and that the reaction rate decreases when the molar ratio of alcohol increases along with high catalyst concentrations, which negatively impacts FAME production [39]. The influence of the static mixer and the reactor geometry enhances the homogenization of the mixture, slowing down the reversibility of the reaction and limiting the formation of by-products such as glycerol [40]. On the other hand, these results underscore the importance of investigating as many effects or interactions related to biodiesel production as possible, since there are studies indicating that both the molar ratio and the concentration of the catalyst, in combination with optimal reaction times at suitable temperatures, directly influence the biodiesel conversion process [41].

According to the analysis, the model suggests the best option to maximize biodiesel yield (91.17%) by operating the reactor at 60 °C, with a molar ratio of 4:1 of alcohol–oil, a concentration of 1 wt% NaOH, and a reaction time of 8 min, which is not significantly different from the experimental value of 90%. However, it was observed that increasing the amount of methanol slightly decreases the biodiesel yield percentage. Otherwise, the results of the interaction graph analysis indicate that there are only two factors that significantly affect biodiesel conversion: residence time, which is crucial for continuous-flow tubular reactors, and catalyst concentration. Additionally, the inclusion of the static mixer improved mixing and, consequently, the yield in biodiesel production.

4. Conclusions

This study was conducted with the aim of producing biodiesel through the transesterification process using soybean vegetable oil and methanol catalyzed with NaOH. For this purpose, a static mixer coupled to a helical tubular reactor was used. The analysis was carried out using a factorial design of experiments 2³, focused on the effect of the operational variables: residence time, molar ratio, and NaOH concentration. It was found that residence time has a greater influence on the biodiesel yield compared to the catalyst concentration and molar ratio. According to the experimental results, the biodiesel yield reached 90% at a temperature of 60 °C, with a molar ratio of 4:1 methanol–oil and a concentration of 1 wt% NaOH, during a residence time of 8 min. In turn, the predictive model suggests a maximum conversion of 91% biodiesel under the same operating conditions. Based on the operational variables analyzed, a regression equation was derived for the production of biodiesel. Consequently, the prototype of the proposed process shows great potential for the optimal production of biodiesel. However, it would be valuable to consider using other types of oils or residual fats, in addition to analyzing the impact of mixer length with preheating. To evaluate the robustness of the system, experiments are planned to be repeated using KOH as a catalyst and varying reaction temperatures, a factor that proved to be a limitation in the experimental design. Optimization of the static mixer design is also proposed, as it could be prone to residue accumulation, potentially affecting process efficiency. Future proposals include integrating sensors to monitor critical variables in real time and improve process control, as well as the characterization of biodiesel and an economic analysis of its production.