Minimising Leachate Wastewater Generated from NaOH-Catalysed Biodiesel Synthesis from Methanol

: Currently, energy generation based on fossil fuels is producing negative environmental impacts; two of the main symptoms of these impacts are water pollution and climate change. Consequently, the search for new technologies to satisfy the energy demand must have the goal to minimise possible impacts to the environment. There are alternatives with biofuels and, among them, biodiesel. The cheapest reaction pathway for biodiesel production is the transesterification of tri-glycerides by methanol in the presence of sodium hydroxide; however, this option can contaminate large volumes of water used in the final leach of the biodiesel product. Therefore, a feasible way of producing this biofuel while simultaneously minimising leaching water will be environmentally friendly and will improve economical savings. The present study developed an experimental design in order to minimise the addition of NaOH during biodiesel production by the basic homogeneous pathway. The best operating conditions were 46 °C, methanol in situ 7.5% v / v and NaOH 0.035 M. These conditions allowed to reduce the leaching water amount by 25% compared to techniques reported in the literature; however, the yield to biodiesel decreased from 98 wt.% to 87 wt.% when a model waste oil was used instead of virgin oil.


Introduction
The popularity of biodiesel (BD) as a renewable fuel alternative to the use of petroleum diesel is growing rapidly, due to the increase in the price of fossil fuels and the environmental awareness that has been strengthening over recent years. The development of BD in some countries has focused on the need to reduce the emissions of greenhouse gases, which is one of the main problems worldwide. The burning of fossil fuels generates environmental issues, such as carbon emissions and global climate change [1]. Biodiesel is a biodegradable fuel, constituted by a mixture of fatty acid alkyl esters, which can be produced from several raw materials such as vegetable oils, animal fats, residual cooking oils, residual oils from the vegetable oil refinement and even sanitary sewage [2]. Compared to some renewable biofuels (bioethanol, e.g.), biodiesel do not need further processes for the purification step; the needless of additional energy will help in the total energy balance of the process. Additionally, there are many routes to produce this biofuel, and the technological cost is very low in the homogeneous pathway.
Biodiesel is quite attractive as alternative fuel, complementing fossil oil diesel, because of its oxygen content that improves the combustion efficiency while reducing the emission of unreacted hydrocarbons (HCs) by 68%, carbon dioxide (CO2) and carbon monoxide (CO) by 44%, sulphur dioxides (SO2) up to 100% and nitrogen oxides (NOx) and the emission of polycyclic aromatics in the range of 80-90% [3]. Hence, in this work is presented a reason to reduce the amount of leaching water in the process of biodiesel production, obtaining a minor environmental impact and the usage of less reagents (e.g., catalysts).

Literature Review
The main production route used in the industry around the world nowadays is the homogeneous alkaline transesterification of alkyl esters with a short-chain alcohol, generally methanol [2,4]. However, in a continuous search for more economical and sustainable solutions, alternative feedstocks, such as microalgae [5,6] and waste cooking oil [7], and alternative technologies have emerged, and their economic and environmental benefits are being extensively studied.
From the total cost of BD production, it is considered that between 60% and 80% is due to the raw material used. This is why the correct selection of the oil is of vital importance. In addition, the reaction yield and the properties of the products will be slightly different according to the type of oil selected [8]. In the case of virgin soybean oil, some authors have reported that the cost of the oil corresponds to 88% of the total estimated cost of production [9,10]. Nevertheless, to avoid the competition of cultivation areas, it is needed to switch to second-generation feedstocks [11].
Waste cooking oil (WCO) is not suitable for human consumption. However, the quality of the WCO causes concerns, due to its physical and chemical properties that could complicate the biodiesel production [12,13]. Waste cooking oils (WCO) obtained from restaurants and food companies exhibit a wide variety of modified features with respect to virgin oils [14]. During the frying process, the oil is exposed to high temperatures in the presence of air and humidity; under these conditions, there will be important changes in its composition due to hydrolytic, thermal and oxidative reactions. These changes in the main fats are known; however, it is not easy to determine the ratio of degradation with respect to the virgin oil because of the large number of variables involved in the frying process. Some of them are related to the process itself, such as temperature, burn duration, heating pattern (continuous or intermittent) and if it is in motion, among others, in addition to external factors such as the type of oil to be used, degree of unsaturation, what type of food will be subject to frying, the lipid composition that contains the food to be fried and additional ingredients such as condiments [15]. The new products formed during the frying process are carbonaceous oligomers and dimers, oxidised triglycerides and their derivatives such as diglycerides and free fatty acids [16].
The WCO contains a high amount of water and free fatty acids (FFAs) content, because during the frying operation, water mixes with oil at a high temperature. This water in oil causes the hydrolysis of triglycerides and form more FFAs. It increases the viscosity and acidity of WCO [17]. Sedimentation is the most preferred pre-treatment method for the WCO [18]. After the pre-treatment, the production of biodiesel from WCO is done with the transesterification approach in the presence of alcohol and a catalyst [19].
Nevertheless, a negative impact of second-generation feedstocks is the high alcohol requirement for biodiesel production from these feedstocks [20]. Thus, it is quite important to explore third-generation feedstocks for biodiesel production; in this case, oils and algae oils are also included. The use of WCO as a biodiesel feedstock can mitigate issues such as water contamination and drainage system blockages, which require additional cleaning [13]. Nonedible oils (Jatropha, Rubber, Karanja, etc.) have been studied as feedstocks for biodiesel production; however, there is a competition on the use of the land for farming, energy versus food production. Then, the use of algae for biodiesel production is a feasible alternative compared to feedstocks used for first-and second-generation biodiesel [21,22] due microalgae having the potential to produce 15-300 times more than the traditional yield from other oil sources in relation to their plantation area [23].
There are many ways to produce biodiesel from these feedstocks, including a homogenous acid catalyst, heterogeneous acid catalyst, homogenous base catalyst, heterogeneous base catalyst and enzyme catalyst and nano-catalyst processes that have been explored recently. Heterogeneous catalysts offer many advantages over homogeneous catalysts, such as simple catalyst recovery, catalyst reusability, simple product purification, less energy and water consumption, less added cost of purification and simple glycerol recovery [24].
The most important advantages from homogeneous basic catalysts are the operation conditions (low temperature and room pressure). Furthermore, the leaching water amount is high due the interaction of the alkali ions, usually after the transesterification process; a volume of biodiesel will need about 3-6 volumes of water until the pH of the light phase (biodiesel) lower to a neutral pH; then, the biodiesel is heated up to 65 °C for at least 10 min to remove the water remanent in the final product. Many countries have prepared their standards for biodiesel fuel. The biodiesel standards developed by the European Union (EN 14214) and the USA (ASTM D 6751) are the internationally accepted standards to maintain the quality of biodiesel [25]. Pure biodiesel is termed as B100, while the mixing of biodiesel with diesel fuel is termed as a biodiesel blend (BX), where X is always less than 100. ASTM D 6751 standards are for pure biodiesel (B100), not for biodiesel blends [17].

Triglyceride Source Selection
A convenient oil for BD production must fulfil some requirements, such as being widely available, containing a high proportion of monounsaturated fatty acids (C16:1 and C18:1), a low proportion of polyunsaturated acids (C18:2 and C18:3) and an adequate content of saturated fatty acids (C16:0 and C18:0) [26]. The primary sources of edible oil are palms, soybeans, rapeseeds, sunflowers, peanuts, cottonseeds, coconuts, olive oil, etc. [27]. Since soybean oil is widely available in Mexico and it fulfils most of the convenient features (Table 1), in this work, it was chosen as the raw material. Soybean oil consists primarily of linoleic acid (55.53%), oleic acid (23.26%) and palmitic acid (11.75%).

Basic Reagent Selection
Among the basic reagents used to prepare biodiesel is sodium hydroxide, followed by potassium hydroxide (Table 2). Due to its low cost and moderate reaction conditions [29], in this work, sodium hydroxide was selected as the basic reagent.

Experimental Designs
The classic batch preparation of biodiesel by the homogeneous basic route follows, as the first step, dissolving the basic reagent in the alcohol, in order to obtain the alkoxide blend (Equation (1)). This blend is added to the triglyceride in order to promote the transesterification reaction (Equation (2)). Every step is run under agitation and temperature control. At the end of the reaction time, a two-phase solution is separated into biodiesel (light phase) and glycerol (heavy phase). Then, biodiesel is leached with fresh water, measuring the pH of the water after each leach until a neutral value is reached [28,32].
The molar ratio of alcohol to triglyceride is recognised as one of the main factors affecting the yield of biodiesel [29,33]. The theoretical ratio for the transesterification reaction is 3 moles of alcohol for 1 mol of triglyceride, producing 3 moles of fatty acid esters (BD) and 1 mol of glycerol. Excess alcohol is used in biodiesel production to ensure that the oils or fats will be completely converted to esters; therefore, higher alcohol triglyceride ratios (3:1 to 6:1) are used. The problem arises because accompanying this alcohol is sodium hydroxide, which affects the process by increasing the leaching water due to the affinity of the ions of sodium in water and making the posterior treatment of this water more complicated. In this study, the addition of methanol is proposed in two ways: ✓ As part of the alkoxide blend (sodium methoxide), called the MetOH/TG ratio. ✓ Combined into the triglyceride system, called MetOH in situ. This addition is bounded by the values 10 vol.% and 30 vol%, because a higher amount of MetOH in situ could cause an increase in the leaching water amount [34].
The other very important factor is the reaction temperature, which is bounded between the limits that decrease the reaction time and accelerates the saponification reaction of the triglycerides [35,36]. Both bounds consider that the maximum reaction temperature must be lower than the boiling point of MetOH in order to ensure that the alcohol will not be lost through vaporization [30].

Design for Virgin Oil Experiments
The first design proposed for this case study was 2 4−1 , which showed information on the effect of the factors up to their highest values (temperature and reagent ratios with respect to the oil). This design (Figure 1) was used for the transesterification of virgin soybean oil to monitor the development of the reaction in an ideal and controlled environment. A detailed pathway of the reaction was available by [37]. The parameters for the experiment design were chosen using data from the literature. High temperatures and bigger amounts of sodium hydroxide for the transesterification process led to saponification.
As the response variables, the biodiesel/glycerol and biodiesel/leach water ratios were recorded as the main objective of the project; additionally, the pH of the leaching water was monitored using a potentiometer. The experiments included 5 central points for error estimations; in order to minimise the bias, the experiments were randomised using the statistical program Minitab 17™, obtaining 13 experimental runs ordered as {6, 11, 9, 4, 8, 3, 13, 5, 1, 10, 7, 12, 2}. Some researchers reported that a ratio of biodiesel/glycerol up to 3 was considered as a good indicator of the success of the process; additionally, the ratio of biodiesel/leach water was important to fulfil the most important objective of this work, and the measurement of the pH indicated that the biodiesel produced was ready for the last step of the process (according to standards from the USA and the European Union) and storage.

Design of Experiments for Model Waste Cooking Oil Preparation
The quality of the biodiesel obtained by the transesterification reaction pathway depends, mainly, on the composition of the oil [12,15,16]. Frying is the operation in which food is immersed in oil at 150-200 °C temperatures in the presence of moisture, antioxidant and prooxidant contents [38]. There are various reactions associated with the frying process, such as hydrolysis, oxidation, polymerization, decomposition and isomerization of oil [39]. Therefore, an artificial process for aging the original virgin oil was designed to obtain a "model" waste cooking oil; it followed a 2 4−1 design of experiments with 5 replicas at the central point, considering the burning time and content of salt, sugar and water present in the raw material as the factors (Figure 2). The preparation of biodiesel using the model waste cooking oil was performed at the best operating conditions found for the virgin oil without any pre-treatment.

Virgin Soybean Oil
The first part of the measurements was used to corroborate that different points of the design of the experiments required different amounts of leaching water. The pH of the first leaching water, in all the experiments, reached values between 11 and 12 (Figure 3), and it could be noted that it was required to use four leaches to achieve pH values close to 7. A statistical analysis of the data was carried out using a Pareto chart (Figure 4), which showed the effect of the different combinations of factors on this pH response.   As can be seen in the Pareto diagram, the amount NaOH did not have a significative influence over the first pH of the leaching water. Nevertheless, the reaction temperature clearly influenced this pH. The statistical relationship between these factors and their impact on the pH confirmed these results (Equation (3)). The second leach did not show any effect of the factors on this pH ( Figure 5). The responses of the design of experiments, the pH of the first leaching water, biodiesel density, biodiesel/glycerol ratio and leaching water/biodiesel were analysed at each experimental point. The ratio of the glycerol/biodiesel products suggested in the literature (1:6) was corroborated; consequently, at these operating conditions, high yields to biodiesel were achieved ( Figure 6). Following the present design of the experiments, the highest yield of biodiesel was obtained in the experiment labelled number 9, which was the point at which the factors were at their minimum values. It is noteworthy that, in experiment 11, saponification occurred; therefore, the yield of biodiesel was zero. The guiding methodology for conducting these experiments was the same technique applied in the first part of this project, where there was a reaction time of half an hour. The amount of soybean oil used in each experiment as a source of triglycerides was kept constant, as in the first design of the experiments. The temperature was kept at 48 °C for all the experiments, in order to prevent the combination of high temperature and inadequate amount of NaOH that favoured the saponification reaction (experiment labelled 11).

Leaching Stage
The amount of leaching water needed to obtain a value of the pH close to 7, after biodiesel leaching in each experiment, showed that the experiment labelled 5 corresponding to the central point of the design of experiments required a volume ratio of 1:1 of leaching water to triglycerides ( Figure 7); this is of paramount importance, since the literature suggests ratios larger than 3. It is worth mentioning that, for some experiments (for example, 7), the reaction might have been compromised (Figure 8), since the impact of reducing the reagents fed into the yield of biodiesel was observed. Even so, the reduction of almost 10% in the yield means a saving of up to 300% of leaching water, according to what was reported in the experiments and in the literature [32,36,40]; therefore, it would be convenient to analyse the viability of the project since the yields at the industrial level are usually close to 90%, with higher times than the ones proposed in the present work.
Additionally, it should be mentioned that the conditions that minimise the volume of the leaching water with respect to the biodiesel of 1:1 mean feeding only 50% of the NaOH suggested in the literature, which corresponds to 3.5 × 10 −4 moles/batch. Additionally, at this point, methanol in situ was added in 7.5 vol.% with respect to the triglycerides, which clearly favoured the yield of biodiesel and the saving of the leaching water.

Model Waste Cooking Oil
The operating conditions obtained for the point that minimised the amount of leaching water of virgin oil were used in the transesterification of 10 mL of the model waste oil: 3.5 × 10 −4 moles/batch of NaOH, 7.5 vol.% MetOH in situ, 48 °C and 0.037 moles of Me-tOH/batch were applied to all the experiments.

Leaching Water
The amount of leaching water increased dramatically for the model waste oil in comparison with the virgin oil ( Figure 9); this happened for the minimum aging conditions and continued as the aging conditions became more severe. The leaching water increased up to 500% in comparison with the best point obtained using virgin oil. This behaviour showed that even a small aging of the oil used as the raw material provoked a high impact on the necessary amount of leaching water.

Changes of pH of Leaching Water
Comparing the pH of water after leaching the products prepared with the model waste oil, it was observed that it persistently presented high values. Therefore, high concentrations of alkaline species, sodium hydroxide and sodium methoxide were still abundant after the transesterification of the oil. The abundance of these alkaline species should be a consequence of the inability of methoxide to react with free fatty acids; additionally, slight soap formation was observed, which means that some sodium hydroxide followed the saponification of the free fatty acids. Therefore, the amount of free fatty acids strongly influenced the decrease of pH of the leaching water and, consequently, increased the demand of this leaching water ( Figure 10).

Yield to Biodiesel
Applying the conditions found to minimise the leaching water necessary for virgin cooking oil, the yields observed using the model waste oils were lower (Figure 11), from 5% to 17%, approximately. Therefore, the use of aged oils is still a feasible option; however it is necessary to use some adequate pre-treatments to control free fatty acids content, avoiding the pollution of the available triglycerides. Figure 11. Yield comparison in biodiesel production for the model oil.

Comparative Leaching Water/Biodiesel
Finally, the consumption of leaching water was increased by the use of model waste oils (Figure 12), since, in the most environmental attractive scenario (experiments 1-4, 7, 10 and 12), the ratio of the volume of the leaching water consumed with respect to the biodiesel produced was 4:1. Therefore, composition changes caused by the aging of the

Yield (%)
raw oil significantly affected the consumption of the leaching water, increasing it to 400% with respect to using virgin oil.

Figure 12.
Wastewater ratio in biodiesel production.

Discussion
According to the experimental data obtained by both the model virgin and waste cooking oils, the best efficiency points (BEP) are compared in the table below (Table 3). Table 3. Comparing different parameters in biodiesel production.

Monitored Parameter
Virgin Oil BEP WCO * Leaching water volume 4:1 1:1 5:1 pH after 4th leaching 7.12 7.10 * 8.36 Yield to biodiesel production 98% 87% 80% * Lowest point from the design of the experiment for a waste cooking oil prepared in a laboratory.
As can be seen, the minimization of the leaching condition will carry a reduction in the yield of the reaction. Nevertheless, the leaching water required can be about 25% of the original amount used for the process. Furthermore, the waste cooking oil is one of the hardest parameters to control due the source of the feedstock. In this way, this work focused on the creation of one kind of "residential feedstock", the yield of the transesterification reaction was reduced to levels about 80% and the leaching water increased approximately 25% versus the use of the first-generation feedstock (virgin cooking oil).

Conclusions
Following a design of experiments, it was found that it is possible to minimise the amount of leaching water for biodiesel prepared by the transesterification of triglycerides in soybeans using methanol in the presence of sodium hydroxide. This reduction was accompanied by a reduction in the amount of reagents, which favours the feasibility of the production. Thus, the best operating conditions to produce biodiesel are at 46 °C, methanol in situ 7.5 vol.% and the addition of NaOH was reduced to 0.035 M; the yield to biodiesel was 98.5%.
Nevertheless, the use of a model waste oil exhibited a high impact on the amount of leaching water required. Even the point with the least severe conditions increased the amount of leaching water; it could be related to the amount of free fatty acids, which also Data Availability Statement: The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.