Separation of Rapeseed Oil Transesterification Reaction Product Obtained Under Supercritical Fluid Conditions Using Heterogeneous Catalysts

Abstract

Rapeseed oil transesterification reaction with ethanol under supercritical fluid conditions was performed either in the presence of catalysts or without them. The catalysts were Al₂O₃ and AlOOH, obtained after Al₂O₃ hydrothermal processing, and CaO/Al₂O₃ and CaO/AlOOH, obtained after permeation. The obtained product was measured for dynamic viscosity and density. Based on these data, kinematic viscosity was calculated. Biodiesel fuel was separated via centrifugation to extract more viscous ethyl esters of saturated fatty acids and unreacted triglycerides in order to comply with the standards for biodiesel fuel. Analyses have found that the maximum content of obtained ethyl esters of fatty acids in a reaction product before separation is reached, in the case of using the CaO/AlOOH catalyst, is in the amount of 93.34% by mass; and none of the samples’ kinematic viscosity values comply with the standards for biodiesel fuel. Performing centrifugation allowed us to reduce viscosity and increase biodiesel fuel concentration to reach the EN14214 standard requirements. Also, a significant deterioration of the initial catalysts’ strength after the singular experiment has been observed: Al₂O₃ by 22.4%, AlOOH by 13.89%, CaO/Al₂O₃ by 25.13%, and CaO/AlOOH by 17.27%.

1. Introduction

The fast-paced depletion of global oil resource deposits, as well as the increased threat of environmental crisis caused by atmosphere pollution, motivate the global community to look for ways to replace hydrocarbon fuel with alternative energy sources. For example, an option in liquid fuel for vehicles can be the environmentally friendly biodiesel fuel, which is composed of fatty acid esters obtained through transesterification of vegetable oils or animal-based fats with low alcohols (methanol, ethanol) [1]. Exhaust fumes from vehicles powered by biodiesel fuel are significantly safer than fumes from engines powered by fossil oil-based diesel fuel [2]. In addition, biodiesel is fireproof; it decomposes within a month to form harmless products, and it has a high lubricating ability [1].

Technology for obtaining biodiesel fuel includes commercial developments using homogeneous alkaline or acidic catalysts [3,4]. However, this technology has disadvantages related to quality in multiple stages, which makes it necessary to separate reaction mix components: biodiesel, excessive alcohol, saponification products, glycerin, etc. Inefficient separation and washing of biodiesel cause serious problems for diesel engines, including filter clogging, nozzle gumming-up, more carbon residue, increased engine wear, flingers sticking, engine knock, lubrication oil thickening, and jellification [5]. The final product may also have esters with high viscosity. These esters include the ones obtained from transesterification of saturated fatty acids (stearinic, palmitimic, etc.), found in large amounts in palm oil, shea butter, etc. Their viscosity is slightly higher than the viscosity of unsaturated fatty acid esters obtained through transesterification of, for example, rapeseed or sunflower oil with high unsaturated fatty acid content. This leads to the understanding that the obtained fuel can have higher viscosity than that set by global biofuel standards EN 14214 [6] and ASTM 6751 [7]. The kinematic viscosity at 40 °C and atmospheric pressure, following the EN 14214 standard, is within the range of 3.5–5.0 mm²/s, following the ASTM 6751 standard, and is within the range of 1.9–6.0 mm²/s. Obtaining biodiesel with acceptable viscosity values is an important task for improving biodiesel quality.

There are methods of separating biodiesel from reaction byproducts; glycerin using ionic liquids [8], membrane separation [9], ultrafiltration [10], and gravity sedimentation [11]. All these methods are limited in terms of obtaining the final product, washing efficiency, the time needed for separation, and complex equipment. A method that does not have these drawbacks is biodiesel separation from unwanted elements through centrifugation. Centrifuges are used in medicine, agriculture, the food industry, etc. [12,13,14]. A large number of studies are devoted to the extraction of fatty acids present in plant-based and animal-based fats, oils, and waxes [15,16,17]. There are also studies on oil extraction from microalgae, which are new generation materials for obtaining biodiesel fuel [18,19,20]. Nevertheless, the use of centrifuges to separate unwanted products from biodiesel is limited by the way glycerin is extracted [11] and does not impact other components of transesterification products.

Transesterification under supercritical fluid (SCF) conditions can be considered the most advanced technology for obtaining biodiesel fuel, which allows for the acquisition of biodiesel without free glycerin [21,22,23]. In addition to a lack of glycerin, an advantage of obtaining biodiesel fuel through SCF transesterification is the high yield at minimal process time without the use of catalysts. Moreover, there are no saponification products, and there is no need for some post-reaction steps, including neutralization, washing, and drying. Thus, the use of water and power related to its disposal is minimized.

Table 1. Conditions on non-catalytic transesterification reaction under SCF conditions.

Initial Oil	Reaction Conditions				Biodiesel Yield, % by Mass	Reference
	Alcohol to Oil Molar Ratio	t, °C	P, MPa	Process Time, min
Rapeseed Castor	42:1 43:1	350 300	45 21	4 90	95.00 96.50	[24] [25]
Sunflower	9:1	280	In dimethylcarbonate SCF environment	9	90.60	[26]
Tobacco seeds	43:1	300	40	90	92.80	[27]
Rapeseed	15:1	280	20	10	96.00	[28]
Used soybean oil	42:1	350	35	-	95.00	[29]
Shea	42:1 42:1	350 400	30 30	30 30	95.70 97.00	[30]
Rapeseed	16:1 18:1	380 365	30 30	30 30	92.81 94.35	[31]
Rapeseed	10:1 12:1	365 365	30 30	30 30	91.41 91.49	[32]
Rice bran oil	7:1	280	-	19	93.00	[33]

shows data on oil transesterification under SCF conditions as well as process parameters, which allow it to achieve biodiesel yield over 90% by mass.

Considering Table 1 results, it can be noted that achievement of high biodiesel yield requires either high molar ratios of initial products or high temperatures. Both of these factors cause high costs and hinder SCF technology implementation on an industrial scale. One of the ways to optimize oil transesterification parameters can be through the use of heterogeneous catalysts.

The transesterification reaction often involves zeolite-based heterogeneous catalysts and their modified forms, with the application of alkaline-earth metals’ alkali and oxides onto their surface [34,35,36], which is caused by their lower solubility and corrosion resistance [37,38]. Also, there are studies on transesterification using catalysts based on magnesium oxides and magnesium-aluminum hydrocarbonates [39,40,41,42], cerium [43], molybdenum [44], strontium [45], zirconium [46].

Table 2. Conditions on catalytic transesterification reaction under SCF conditions.

Initial Oil	Reaction Conditions						Biodiesel Yield, % by Mass	Reference
	Alcohol to Oil Molar Ratio	t, °C	P, MPa	Used Catalyst	Catalyst Share, % by Mass	Process Time, min
Rapeseed	12:1	350	30	ZnO/Al2O3 MgO/Al2O3 SrO/Al2O3	2	30	97.45 94.78 97.46	[47]
Sunflower	41:1 41:1	252 252	9 9	CaO CaO	3 3	6 27	98.00 98.00	[48]
Rapeseed	7:1	310	-	MgO/γ-A12O3	2.1	15	97.00	[49]
Soybean Safflor	12:1 39.6:1	180 300	10 25	MgO CaO	5 1.29	60 10	90.00 91.00	[38]
Rapeseed	4:1	250	10.5	ZnO	1	10	95.00	[50]
Mahua	40:1	250	10	ZnO	1	8	90.00	[51]
Jatropha	40:1	300	9	ZnO/γAl2O3	6	3	95.64	[52]
Rapeseed	42:1	270	15	ZnO/CaO	1.3	60	93.00	[53]
Rapeseed	12:1 18:1	350 380	30 30	Al2O3 Al2O3	2 2	30 30	90.94 92.31	[54]
Rapeseed	18:1	365	30	BiO/Al2O3	2	30	95.02	[55]

summarizes the data on catalytic SCF oil transesterification.

According to the data in Table 1 and Table 2, the use of catalysts provides a temperature fall by 42 °C on average, a lower pressure by 13 MPa, and a shorter reaction time by 9 min, compared to the non-catalytic reaction under SCF conditions. Generally, most studies on catalytic transesterification under SCF conditions deal with the use of various metal oxides on the Al₂O₃ carrier (Table 2), but studies on the use of metal oxides on the boehmite (AlOOH) carrier are limited by the activities carried out in traditional transesterification [56,57]. The use of AlOOH as a catalyst carrier instead of Al₂O₃ is explained by high hydrothermal stability at higher temperatures and pressures typical for SCF conditions in reactions carried out in organic homogeneous solutions, including water-containing ones. The catalysts’ work stability under the SCF conditions of transesterification is also important for the further scale-up of the studied technology.

Despite significant progress in SCF transesterification, high reaction speed over a short time is not often able to provide reaction completion, and apart from biodiesel, the reaction product may include unreacted fatty acids. Efficient methods for such fatty acid and saturated fatty acid ester separation, particularly using centrifugation, remain unstudied.

The object of the study is the centrifugal separation of catalytic and non-catalytic transesterification reaction products obtained under SCF conditions from high-viscosity esters of saturated fatty acids and unreacted fatty acids. The presented results with CaO/AlOOH as a catalyst are also novel. Additionally, there is the elemental composition and compressive strength of the catalysts used before and after experiments, which are needed to estimate the prospects of industrial use under working conditions with high state parameters, typical for SCF conditions.

2. Experimental Part

2.1. Initial Raw Materials

Raw material for transesterification was refined food-quality rapeseed oil from Oily, Belarus, with 99.8% purity and 96% ethanol ($n_D^{25}$ = 1.36242, ${\rho }^{25}$ = 797.1 kg/m³). The use of ethanol in experiments is explained by its lower toxicity compared to methanol when working on a laboratory bench. Choice of rapeseed as an oil base is caused by its high crop yield in moderate climate of Eurasian countries. Oil’s content of fatty acids from manufacturer is given in

Table 3. Fatty acid content of rapeseed oil.

Fatty Acid	Fatty Acid Mass Fraction (% of Total Fatty Acids)
Tetradecanoic (myristic)	0.02
Pentadecanoic (pentadecylic)	0.01
Hexadecanoic (palmitinic)	6.46
Hexadecenoic (palmitoleic)	0.17
Octadecanoic (stearinic)	3.12
Octadecenic (oleinic)	77.67
Octadecadienoic (linoleic)	11.69
Eicosanoic (arachic)	0.22
Eicosenoic (gondoinic)	0.59
Docosanoic (behenic)	0.03
Docosenoic (erucic)	0.02
Tetracosanoic (lignoceric)	<0.01
Tetracosanoic (nervonic)	<0.01
Total	100 ± 0.01

.

Catalyst carrier was granulated Al₂O₃ (mixture of gamma- and chi-oxides) based on reference review, low price and availability [58], and AlOOH. Boehmite was obtained through Al₂O₃ hydrothermal processing in distilled water at water to catalyst mass ratio 2:1 for 9 h at temperature t = 190 °C and corresponding water vapor pressure. For preparation of synthesized catalysts CaO/Al₂O₃ 2% by mass and CaO/AlOOH 2% by mass, single permeation with permeation mixture excess to carrier method was used. Permeation solution was prepared the following way: calculated amount of Ca(NO₃)₂·4H₂O was mixed on a magnetic mixer until true solution formation and then diluted with distilled water in a measuring flask to 100 cm³. Carrier aliquot ~30 g was placed into permeation solution and exposed for ~24 h at room temperature. Then, obtained solution was evaporated for 1 h. Next, vacuum impulse drying was carried out at 373 K and atmospheric pressure until stable sample mass was reached. Then, the catalyst was calcinated in a muffle furnace in air atmosphere at 673 K for 4 h and cooled down.

The use of 2% by mass solution was reasoned in previous studies by the article authors in research with similar catalysts [47,55].

2.2. Experiment Equipment and Techniques

2.2.1. Transesterification Under SCF Conditions

Continuous process was used to obtain biodiesel fuel under supercritical fluid condition of reaction mixture. Equipment unit overview is shown in Figure 1. In order to provide intense mixing of hard-miscible reagents under atmospheric conditions and increase phase contact area, an ultrasonic emulsifier UIP1000HD by Hielscher (Teltow, Germany) was included in the process layout (item 6) for process optimization. Emulsifier working frequency was 20 kHz at emulsifying power of 1000 W. Given parameters recommended by manufacturer, mixture treatment with ultrasound for maximum efficiency was performed at 81 µm oscillating amplitude and under 3.5 bar excessive pressure. Previously, our colleagues [59] proved stability of rapeseed oil and ethanol emulsion obtained through ultrasonic dispersion and revealed effectiveness of emulsification for biodiesel fuel obtaining under SCF conditions using palm oil [60].

In addition, the unit also received catalytic reactor with stationary catalyst layer (item 11). The catalytic reactor volume is calculated to be 5% by mass of the total reaction mixture feed. Catalyst layer takes volume of 160 cm³, 0.8 m high. Amount of catalyst used in experiments was 2.5% of initial raw material mass. For complete reactor filling, contact area increase and additional process intensification catalyst is charged with mixed hollow ceramic cylinders. Pumped mixture rate was 300 mL/min, pressure overfall was taken from readings difference of manometers’ installed before and after catalytic reactor and did not exceed 8 KPa. Required temperature control and maintenance are implemented using a cable with magnesial insulation and a nickel–chromium core that receives power through a thermoregulator coaxially rolled onto a catalytic reactor. Temperature taking is performed by chromel–alumel thermocouples at the input and output of the catalytic reactor. Thermal insulation is provided to minimize heat loss.

2.2.2. Chromatography

Obtained transesterification reaction product was evaluated using chromatography performed at quadrupolar mass-spectrometer Agilent 6890N/5973 by Agilent, Santa Clara, CA, USA. Analysis conditions were the following: electronic ionization at ionizing electrons energy 70 eV, ion source temperature 230 °C. Capillary column TG-5MS was used in the process. This column is 30 m long and 0.25 mm wide. Phase layer thickness was 0.25 µm. Helium was used as a carrier gas. Mass-spectral data processing was performed using MSD ChemStation software, Version D.02.00.275 (Agilent Technologies, Santa Clara, CA, USA). Sample was preliminarily dissolved in isopropyl alcohol at 1:10 ratio.

Chromatogram receiving conditions were set as follows: injector temperature—280 °C, split ratio—1:15; column heating was performed in program mode by areas: initial temperature—100 °C (10 min); heating rate 10 °C/min to 280 °C (2 min); heating rate 10 °C/min to 300 °C (10 min); carrier gas flow rate through the column—1 mL/min; temperature of communication unit with mass-spectrometer—320 °C; sample volume—0.5 µL.

2.2.3. Measuring Thermophysical Properties

Density and dynamic viscosity were measured using Stabinger viscometer SVM3000 (AntonPaar, Graz, Austria). This unit allows us to measure dynamic viscosity within 0.2–104 MPa·s, density within 650–2000 kg/m³. Measuring relative error limits as follows: for viscosity—±0.35%; for density—±0.5%; for temperature—±0.02 °C. The working temperature range is the following: 0–100 °C. Kinematic viscosity was calculated as dynamic viscosity to density ratio.

2.2.4. Centrifugation

Reaction product separation was performed in a cooled centrifuge Hanil Scientific Smart 17 Plus, South Korea, equipped with an imperforate basket. Centrifuges of this type are most often used for separation of concentrated suspensions and emulsions. Centrifugation on this unit is based on different particle behavior in centrifugal field. Suspension is put into a glass tube and set into rotor installed on centrifuge drive shaft.

This unit allows to perform separation at top rotation speed, which is 17,000 rpm, within working temperature range of −10–+40 °C. The unit work cycle time can reach 100 min.

Before the experiment, the sample was put into a thermostat set at 0 °C for 12 h for thermal stabilization. This temperature choice is caused by different crystallization temperatures of fatty acids’ ethyl esters and impurities present in biodiesel.

Then, sample was put into Eppendorfs (24 pcs) of 2 mL volume.

The process was performed at the following parameters:

-

Temperature—4 °C;

-

Rotation speed—6000 rpm;

-

Centrifugation time—10 min.

The choice of parameters is explained by preliminary experiments when, at higher temperatures, lower rotations, and longer times, there was no separation and sedimentation even after preliminary cooling.

2.2.5. Catalysts Analysis

Catalysts’ elemental analysis was performed using X-ray fluorescent spectrometer RMS-02 RehnPV/HLF by NTC ExpertCenter (Moscow, Russia).

Since transesterification under SCF conditions runs at high temperatures and pressures, mechanical strength of initial and used catalysts was also analyzed. Catalyst strength before and after experiments was found using LinteL PK-21 by JSC BSKB Neftekhimavtomatika (Ufa, Republic of Bashkortostan, Russia), unit designed for testing catalysts for mechanical strength under static conditions through compression method. Testing of the catalysts in question was performed automatically. The following method was used to find granules’ static strength: 12 granules of a catalyst were taken, and an average granule cross-section was found; then, the granules were placed into special cassettes, and the force was applied to the catalysts using a piston rod. The strength measuring unit takes the value of the strength limit for each sample. Then, the average value (P_op, N/cm²) is calculated according to the following formula:

$$P_{op}={\displaystyle \frac{1}{N}}{\sum }_{I=1}^N{\displaystyle \frac{X_i}{S}}$$

(1)

where N—number of tests; S—average granule cross-section area set before series of tests, cm²; X_i—strength limit value during the test N; static relative error of strength limit measurement is 1%, piston rod lowering speed during catalyst compression is 4.4 N/s.

2.3. Control Measurements

In order to estimate the influence of ultrasonic impact on the reactive environment, authors took control measurements for studied systems without catalysts. Cases of obtaining biodiesel without stirring the initial mixture, with preliminary mechanical stirring, and with preliminary ultrasound impact have been considered. Experiment (t = 334 °C, P = 30 MPa, alcohol to oil molar ratio—25:1, τ = 36 min) results based on biodiesel yield and kinematic viscosity of obtained product are given in

Table 4. Test measurement results in estimation of ultrasound impact on biodiesel yield.

Stirring Type	Biodiesel Yield, % Mass	Kinematic Viscosity (t = 40 °C, P = 0.1 MPa), mm2/s
No stirring	70.54	6.025
Mechanical stirring	73.58	5.737
Ultrasound emulsification, 20 kHz	80.25	5.072

Control measurement results show that preliminary stirring with ultrasound impact is preferable.

.

3. Results and Discussion

Non-catalytic and catalytic transesterification were performed experimentally under SCF conditions in a continuous mode. Reaction conditions were chosen based on the average values in Table 1 and Table 2: for non-catalytic reaction—t = 334 °C, P = 30 MPa, alcohol to oil molar ratio—25:1, time—36 min; for catalytic reaction in the presence of Al₂O₃, CaO/Al₂O₃ (2% by mass), AlOOH and CaO/AlOOH (2% by mass)—t = 274 °C, P = 18 MPa, alcohol to oil molar ratio—25:1, with catalyst amount 2.5% by mass, time—25 min. After product separation from excess alcohol, dynamic viscosity and density of obtained samples were measured at atmospheric pressure within the temperature range of t = 20–40 °C. The experiment results are shown in

Table 5. Biodiesel fuel samples’ dynamic viscosity.

Sample, No.	Catalyst	Dynamic Viscosity, mPa·s
		t = 20 °C	t = 30 °C	t = 40 °C
1	None	13.759	8.066	4.463
2	Al2O3	14.499	8.498	5.309
3	CaO/ Al2O3 (2% mass)	14.304	8.410	5.107
4	AlOOH	13.981	8.370	4.980
5	CaO/ AlOOH (2% mass)	13.878	8.130	4.544

and Figure 2.

Uncertainty value for dynamic viscosity reaches ±1%, for density—±0.1%.

Table 6. Biodiesel fuel samples’ kinematic viscosity.

Sample, No.	Catalyst	Kinematic Viscosity, mm2/s
		t = 20 °C	t = 30 °C	t = 40 °C
1	None	15.342	9.072	5.072
2	Al2O3	16.135	9.531	6.009
3	CaO/ Al2O3 (2% mass)	15.929	9.437	5.784
4	AlOOH	15.573	9.394	5.643
5	CaO/ AlOOH (2% mass)	15.468	9.130	5.152

shows estimated values of kinematic viscosity for biodiesel at atmospheric pressure and various temperatures obtained through dynamic viscosity to density values ratio.

Considering Table 5 and Table 6, viscosity decreases with measurement temperature rise. Viscosity measurement results for non-catalytic biodiesel production are slightly lower than the values of samples obtained through a catalytic reaction. This is explained by higher temperatures and longer reaction times for non-catalytic procedures. However, none of the samples complies with international standard EN14214 in terms of kinematic viscosity. Rashid et al. [61], at similar technological parameters but using methanol at T = 313 K, received a kinematic viscosity of 4.85 mm²/s and a density of 0.882 g/cm³ at T = 298 K. Demirbas [62,63] at supercritical transesterification using methanol and vegetable oils received kinematic viscosity within the range of 3.59–4.63 mm²/s. Nascimento et al. [64] measured the properties of biofuel obtained from soybean oil at ethanol to oil molar ratio 39:1 and temperatures between 533 and 573 K and pressure over 10 MPa. Density values at T = 293 K varied within the range of 0.90484–0.91699 g/cm³.

3.1. Transesterificationunder SCF Conditions

Results of biodiesel composition chromatographic analysis for main esters for analyzed samples are shown in Figure 3 and Figure 4.

Considering the chosen reaction conditions, the performed supercritical transesterification experiment showed that a non-catalytic reaction does not allow us to obtain predictable FAEE amount in the reaction product at 90% (Figure 3 and Figure 4). One of the reasons is that the obtained unsaturated oleic fatty acid ethyl esters are less when compared to the content in samples obtained using catalysts (Figure 3). This is the result of this ethyl ester decomposition under supercritical transesterification reaction conditions at temperatures over 300–325 °C, which is also proven by our previous study results and other authors’ studies [54,65,66]. It should also be noted that thermal decomposition is not seen for saturated fatty acids’ (palmitinic, stearinic, etc.) ethyl esters due to even higher decomposition temperatures for these esters.

During catalytic reaction analysis, the declared 90% was achieved using all four catalysts. Insignificant final product changes are seen in the case of using boehmite and synthesized boehmite-based catalyst compared to industrial Al₂O₃ and CaO/Al₂O₃ synthesized on its basis.

The maximum content of the obtained FAEE in the reaction product is reached by using CaO/AlOOH in the amount of 93.34% by mass. Meanwhile, none of the samples has reached level of 96.5% by mass in fatty acids esters content in reaction product, which is required by EN 14214 standard.

Also, traces of glycerin, a reaction byproduct, were not found during the reaction either visually or using chromatography. During the reaction process at these temperatures, glycerin decomposes. During the process, it reacts with alcohol:

C₃H₅(OH)₃ + 3C₂H₅OH = C₃H₅(OC₂H₅)₃ + 3H₂O

(2)

Water formed in the reaction interacts with fatty acid triglycerides with formation of fatty acids diglycerides and free fatty acids:

C₃H₅(OCOR)₃ + H₂O = C₃H₅(OCOR)₂OH + RCOOH.

(3)

Then, free fatty acids turn into fatty acids’ compound esters:

RCOOH + C₂H₅OH = RCOOC₂H₅ + H₂O.

(4)

Such esters are ethyl esters of caprylic, pelargonic, and hendecoic acids formed in minor amounts in reaction products.

3.2. Biodiesel Samples Centrifugation

The obtained biofuel samples do not comply with the EN14214 standard on total FAEE content on reaction products or on kinematic viscosity values. Due to this, biodiesel separation was performed in order to extract unreacted triglycerides and saturated ethylpalmitate and ethylstearate, which have a higher viscosity than ethyloleate.

For this, transesterification reaction product centrifugation was performed. Particles of the products having different densities, shapes, and sizes and charged in 24 Eppendorfs are sedimented at different speeds in a centrifugal field. In the case of the same density, larger particles are settled much faster than smaller ones. The higher the environment viscosity, the slower particles settle down. During the centrifugation process, the centrifugal force pushed heavier biodiesel components toward the Eppendorf walls, thus improving sedimentation. After the experiment, white crystals, which did not separate during centrifugation, stayed on the Eppendorf walls. The repeated mixture run in a centrifuge did not show any separation effect.

One sample separation result is shown in Figure 5. Initial biodiesel (Figure 5A) was separated into two fractions (Figure 5B,C). In the Eppendorfs (Figure 5C), thickened residue traces as white crystals can be seen on the walls. The remaining more liquid fraction (Figure 5B) visually looks more clear compared to the pre-separation mixture (Figure 5A).

Chromatographic analysis results for biodiesel samples after separation (Figure 6 and Figure 7) and the remaining sediment (Figure 8 and Figure 9) were obtained. Figure 8 and Figure 9 results are found based on average values of analyzed samples taken from five Eppendorfs.

According to analysis results (Figure 7), the reduction of unreacted triglycerides and saturated fatty acids (palmitinic, stearinic, etc.) and ethyl esters are compared to Figure 4. Meanwhile, ethyl oleate concentration in the overall mixture (Figure 6) rises compared to Figure 3. Its amount combined with other FAEEs for samples 2–5 (Figure 6 and Figure 7) exceeds 96.5% by mass, which, according to EN14214 standard, makes the fuel obtained after centrifugation suitable for use in terms of the required amount of esters in the reaction product. Unseparated residue results (Figure 9) show a high percentage of saturated fatty acids (palmitinic, stearinic, etc.), ethyl esters, and unreacted triglycerides compared to the mixture’s mass composition.

Samples’ dynamic viscosity and density were measured at t = 40 °C and P = 0.1 MPa, then kinematic viscosity ratios were found. Results are shown in

Table 7. Density, dynamic, and kinematic viscosity of biodiesel fuel samples after separation at t = 40 °C and P = 0.1 MPa.

Sample, No.	Dynamic Viscosity, mPa·s	Density, g/cm3	Kinematic Viscosity, mm2/s
1	3.817	0.879	4.342
2	4.502	0.883	5.098
3	4.412	0.882	5.002
4	4.358	0.882	4.941
5	3.913	0.881	4.442

.

Table 7 suggests that biodiesel samples after separation have kinematic viscosity equal to or even under 5.0 mm²/s within experiment error.

Also, transesterification product sediment sample density was measured after separation at t = 40 °C and atmospheric pressure (

Table 8. Transesterification product sediment samples density after separation at t = 40 °C and P = 0.1 MPa.

Sample No.	1	2	3	4	5
Density, g/cm3	0.885	0.892	0.891	0.890	0.890

). Density was measured by the gravimetric method using a glass pycnometer GP 2–10 KSh 7/16 by EximLAB (Khimki, Moscow Region, Russia) (GOST (State standard of the Russian Federation) 22524-77 [67]). Weighing was carried out on analytical balances model BLA-200 (Balances laboratory analytical) and electronic balances “Mettler” model PM600 from “Mettler-Toledo GmbH” company, Greifensee, Switzerland. The use of the Stabinger viscometer is limited by sample size (minimum 10 mL). The sediment sample volume was just over 2 mL. Sediment density is slightly higher than reaction product density after separation (Table 6) due to high fatty acid triglyceride content.

3.3. Catalysts Characteristics

The elemental composition of some used catalysts before and after experiments is shown in

Table 9. Elemental composition of some used catalysts before and after experiments.

Element	Before Experiment		After Experiment
	CaO/Al2O3	CaO/AlOOH	CaO/Al2O3	CaO/AlOOH
Al	98.03	97.90	94.35	94.31
Ca	1.95	2.07	1.85	2.01
Cr	-	-	0.04	0.03
Fe	0.02	0.03	1.20	1.20
Zn	-	-	2.14	2.12
Ni	-	-	0.03	0.02
Cu	-	-	0.32	0.31

.

The supercritical transesterification process causes aluminum and calcium to leach from the catalyst’s surface. The presence of nickel, chrome, as well as zinc and copper in catalyst samples after the reaction is explained by the presence of these metals in parts of reactors made of stainless steel, brass valves, and zinc-coated connection pipes.

The arithmetic mean strength values for catalysts were calculated according to formula (1) for 12 tested samples of each catalyst and are given in

Table 10. Arithmetic mean values of catalyst strength, N/cm².

Al2O3		AlOOH
Before Experiment	After Experiment	Before Experiment	After Experiment
756.7208	587.2030	1038.6954	894.3652
CaO/Al2O3		CaO/AlOOH
Before Experiment	After Experiment	Before Experiment	After Experiment
693.5874	519.2569	889.2547	735.6520

.

Measurement results (Table 10) show significant deterioration of catalysts’ strength after the singular experiment: Al₂O₃ by 22.4%; AlOOH by 13.89%; CaO/Al₂O₃ by 25.13%; CaO/AlOOH by 17.27%. The catalyst AlOOH obtained through hydrothermal processing showed better strength characteristics both before and after the experiment compared to Al₂O₃. The same is true about the catalyst based on it. The catalysts’ strength decline after the experiment is caused by the high permeability of the supercritical fluid, which is able to penetrate catalysts deeply.

4. Conclusions

Transesterification reaction was performed under supercritical fluid conditions in a flow-through mode for the non-catalytic method (t = 334 °C, P = 30 MPa, alcohol to oil molar ratio—25:1, time—36 min), and in the presence of the catalysts Al₂O₃, CaO/Al₂O₃ (2% by mass), AlOOH, and CaO/AlOOH (2% by mass), it was performed at t = 274 °C, P = 18 MPa, an alcohol to oil molar ratio of 25:1, a catalyst amount of 2.5% by mass, and a time of 25 min. For these reaction conditions, the obtained values of kinematic viscosity and biodiesel fuel concentration in the reaction product do not comply with the standard requirements. Sufficiently high parameters for non-catalytic reaction conditions caused ethyloleate decomposition. The performed centrifugation allowed us to separate biodiesel from unreacted triglycerides and highly viscous saturated fatty acid esters. As a result, the reaction product’s kinematic viscosity decreased to a value equal to or under 5.0 mm²/s. The overall content of the main fatty acid ethyl esters increased to values over 96.5% by mass, which allowed us to obtain biodiesel fuel that complied with the EN14214 standard requirements. The initial catalysts’ strength declines after the singular experiment: Al₂O₃ by 22.4%, AlOOH by 13.89%, CaO/Al₂O₃ by 25.13%, and CaO/AlOOH by 17.27%. This is caused by the high permeability of supercritical fluid. Based on the present data for the lowest kinematic viscosity, the highest FAEE content in the reaction product, and high enough structural characteristics after the reaction, the CaO/AlOOH catalyst (2% by mass) can be recommended for transesterification reaction under SCF conditions.