Rendering of Beef Tallow for Biodiesel Production: Microwave versus Boiling Water and Acetone Fat Extraction

Abstract

Biodiesel can substitute for conventional diesel fuel and contribute to the decarbonization of the transportation sector. To improve biodiesel sustainability and decrease production costs, low-grade fats such as non-edible animal fats must be used. Animal fats are mixed with tissues which must be removed before alcoholysis to avoid biodiesel contamination with nitrogen and phosphorus-containing compounds. Biodiesel was produced by the methanolysis of beef tallow and beef tallow/soybean oil mixtures over calcium heterogeneous catalysts obtained by the calcination of scallop shells. The tallow from fatty bovine tissues was extracted using boiling water, dry microwave treatment, and acetone extraction. The thermal stability and the moisture content of the extracted fats were evaluated by thermogravimetry. The thermograms of fats revealed that microwave treatment, which was faster (3 min instead of 40 min for boiling water and 240 min for acetone extraction) and had the lowest energy consumption, led to a dry fat with a thermal stability analogous to that of fats extracted with boiling water and acetone. All the extracted fats behaved similarly in the methanolysis reaction over calcium catalyst, with biodiesel yield (61–62%) being 30% lower than the analogous obtained from soybean oil (88%). Co-processing the extracted tallow with soybean oil overcomes the drawback related to the low-grade fats.

1. Introduction

The mitigation of anthropogenic carbon can be partially attained by substituting fossil fuels with renewable ones. Biodiesel, a mixture of alkyl esters, has been pointed out as a feasible renewable fuel in the transportation sector [1,2], which is the main contributor to anthropogenic CO₂. First-generation biodiesel, derived from vegetable oils, presents several sustainability issues [3]. The occupation of arable land with oil crops dedicated to fuel production is seen as a threat to food production capacity, which triggers strong opposition to biodiesel due to fuel/food competition [4]. The price of vegetable oils is another negative factor as it makes the production of biodiesel expensive [5] and uncompetitive with the fossil fuel congener, except for very high prices of crude oil [6]. Bashir et al. [7] emphasized, in a recent review, that the choice of biodiesel fat feedstock is the most important factor in obtaining a sustainable and economically-competitive renewable transportation fuel. According to these researchers, the third- (algae and waste oils) and fourth- (advanced solar oils) generation fat feedstocks are preferred in biodiesel production at a commercial scale.

Biodiesel production from animal fats has great potential because this feedstock does not compete with the food industry and helps with waste management [8]. Animal fat is viewed as a promising biodiesel raw material due to its low cost and easy availability [9]. Despite the advantages of biodiesel derived from animal fat, this raw material only makes up 7% of the fat in biodiesel produced worldwide and only 4% of the biodiesel produced in EEC [10]

Beef tallow, pork lard, chicken fat, and grease are some examples of animal fats that can be employed [11,12,13,14,15,16,17]. Animal fat feedstocks lead to a biodiesel with higher cetane number than that of vegetable oils. The high content of saturated fatty acids is responsible for this improved characteristic of animal fat biodiesel [16]. Furthermore, the high cetane number of animal fat biodiesel allows for a reduction in NO_X emissions [17]. Considering that raw materials have the greatest weight in the cost structure of biodiesel production (Figure 1), the use of animal fats instead of vegetable oils contributes to the reduction in production costs. Animal fats are cheaper than vegetable oils and used cooking oils (

Table 1. Fats raw materials prices (www.alibaba.com, accessed on 1 March 2022).

Fat	EUR per Metric Ton
Soybean oil	184–450
Rapeseed oil	230–583
Palm oil (crude)	138–238
UCO	137–200
Tallow (category 1)	100–148

). According to the literature, animal fats (AF) contain high levels of free fatty acids (FFA), which makes conventional biodiesel production impracticable due to soap formation. Complex processes to convert AF into biodiesel are reported in the literature. Ikura et al. [18] combined thermal cracking and acid-catalyzed esterification to produce biodiesel from animal fats. The process, after the thermal cracking, included a distillation process to separate the FFA for the esterification step. Simpler procedures, as schematized in Figure 1, always include the pretreatment of animal fat to decrease the fat acidity [19].

Raw animal fat has protein-rich tissues which must be removed before the alcoholysis reaction. The fats are recovered from waste fat tissues by the rendering processes, mainly wet or dry rendering [20,21]. In wet rendering, the fat is recovered through heating in the presence of water. Boiling in water and/or steam at high temperatures can be employed [21,22]. In dry rendering, the fat tissues are cooked in their “own juices” with dry heat [22].

Several processes are used in industry, all of them involving the application of heat, the extraction of moisture, and the separation of fat.

The boiling water rendering presents some issues: high energy and time consumption, a fat with a high humidity, and increased free fatty acid content due to the long contact of fat with water [21]. Additionally, considering that the main purpose of the rendering is the separation of fat from other materials including residual moisture, the introduction of additional moisture is seen as unproductive [21]. For these reasons, some processes were replaced by dry ones.

In recent years, microwave technology (MW) has received a lot of attention from researchers who use it in a wide range of processes such as methane fermentation [23,24], sludge treatment [25], lipid extraction from microalgae [26], MW-assisted catalyzed reactions [27], and even non-catalyzed biodiesel production [28]. According to researchers, MW technology consumes less energy than traditional technologies and is thus more cost-effective. Microwave (MW) technology can also be used for animal fat rendering. This process is quicker, energy efficient, and the fat-rendering yield obtained is higher [29]. Lin et al. [30] reported that, among all the rendering methods used to recover chicken fat from broilers, the MW rendering was the most promising one, obtaining the highest rendering yields and quality. Sheu and Chen [31] reported similar results: MW produced the highest rendering yields and the lightest-colored broiler skin fats. Recently, Hewavitharana et al. [32] compared several methods of extracting fat from alimentary products and concluded that microwave-assisted fat extraction is an effective methodology when compared with conventional methods, including the Folch method (liquid extraction with chloroform plus methanol mixture). The authors underlined the short processing time for microwave-assisted fat extraction. The literature on the microwave-assisted extraction of animal fat is not abundant and the existing literature does not explain how microwaves act in the solvent-free extraction of fat. Taking into account the mechanism described for the extraction of microalgae oils [33], it is assumed that microwaves heat the water molecules found in the adipose tissue, which promotes localized heating and increased pressure, causing the breakdown of adipocyte, and releasing triglycerides.

Conventional methods of fat extraction involve the use of solvents. Such methods are often used in the extraction of oil from oilseeds [34,35]. In a previous study, the use of acetone as a co-solvent in the methanolysis reaction not only improved components’ miscibility but also avoided catalyst deactivation by the adsorption of oily species [36]. For these reasons, the rendering process using acetone as the solvent was also evaluated.

The following sections give data regarding the manufacture of biodiesel using beef tallow in order to contribute to the body of knowledge in the area of sustainable biodiesel production from non-edible animal fat. To increase the fat yield that allows biodiesel production by methanolysis over heterogeneous basic catalysis (CaO catalyst), different methods of extracting beef tallow were investigated (boiling water, extraction with acetone, and microwave-assisted). Low energy consumption, low moisture content, and low acidity of the extracted fat were the goals of the fat-rendering studied procedures. The methanolysis of alimentary grade soybean oil (SBO) was used as a standard. Calcium-based heterogeneous catalyst, obtained from waste materials, was used due to its recognized catalytic performance in biodiesel production [37].

2. Materials and Methods

The experimental work followed the steps outlined in Figure 2, starting with the extraction of beef tallow, which was then used in the production of biodiesel, and ending with the quantification of biodiesel and the characterization of the co-produced glycerin.

The beef tallow collected from a local butcher was subject to several rendering processes (Figure 2) to separate fat from protein-rich tissues. In the microwave-assisted (MW) process, a plastic container with 100 g of beef tallow was placed in a microwave oven and heated at maximum power (1400 W) for 1, 2, and 3 min, separately. After removing the solid residues, the liquid was cooled at room temperature.

For boiling water (BW) fat extraction, the fat tissues were boiled with water (enough to cover the fat material) in a pressure cooker for 40 min (1.5 atm; 112 °C). After removing the residues, the liquid was cooled at room temperature and the fat layer was removed with a drainer. The cooking time was optimized to extract the maximum fat.

Acetone fat extraction, at reflux temperature, was accomplished in a glass 500 mL round flask equipped with a reflux column heated by a nest-shaped heating mantle. The fat tissues were boiled with acetone (1.28 w_fat/w_acetone ratio) for 4 h. After removing the solid material, the liquid was cooled at room temperature and the fat layer was removed with a drainer. All the extraction methods were replicated 3 times to evaluate their reproducibility.

The thermal stability and moisture content of the extracted fats was assessed by thermogravimetry. The thermograms were acquired, under airflow (synthetic, dry), for samples of 60–100 mg in alumina crucibles, using a Netzsch TG–DTA-DSC thermobalance and a heating rate of 20 °C/min. The thermal degradation profile of soybean oil was used as a reference. The free fatty acid (FFA) content of the fat feedstocks was assessed by alkalimetric titration according to the standard EN 14214:2003 [38,39].

Biodiesel was produced by the methanolysis of beef tallow and beef tallow/soybean oil (1:1 in weight basis) over calcium oxide catalyst using standard conditions to test calcium-based catalysts [40]. The methanolysis tests, at methanol reflux temperature, were carried out in a batch reactor, as previously described [41]. Briefly, the catalytic tests were carried out in a 3-neck 500 mL glass reactor using a nest-shaped heating mantle and a magnetic stirrer (orbital magnet). The slurry containing the methanol (methanol/fat = 12 molar ratio) and the catalyst (5%, fat weight basis) was heated to the reflux temperature of methanol, remaining at this temperature for 1 h to promote the activation of methanol molecules on the catalyst surface [42]. Acetone-assisted methanolysis tests (V_methanol/V_acetone = 2.8) were carried out as reported by Dias et al. [36], to avoid the catalyst deactivation by fat acidity and moisture. At the end of this period, the fat, previously heated to 67 °C, was added, starting the counting of the reaction time. After the reaction period (2.5 h), the catalyst was separated from the reaction mixture by filtration (paper filter grade 1, 11 μm). The glycerin and the oily phases were separated by gravitational settling in a funnel. Both liquid phases were characterized by ATR-FTIR and the methyl esters were quantified as described by Dias et al. [43]. Briefly, the FAME yield computation was made using the infrared spectral features in the 1480–1410 cm⁻¹ range, attributing the band centered at 1436 cm⁻¹ to FAME. A calibration curve was previously obtained for oil/FAME mixtures with different FAME contents. The infrared spectra were collected with a resolution of 4 cm⁻¹, using an FT-MIR Perkin Elmer Spectrum Two IR Spectrometer. A horizontal total attenuated reflection accessory (ATR), from PIKE Technologies, with a ZnSe crystal, was used. The catalytic test using MW tallow, which was randomly selected, was replicated 3 times to evaluate the reproducibility of the catalytic test.

The calcium catalyst (mainly CaO) was obtained by the calcination of calcium-rich seafood shells [44,45]. The scallop shells, which had been household waste, were washed using tap water and dried naturally for several days (approximately a week) while in contact with the laboratory’s atmospheric air. The dried shells were coarsely crushed and calcined for 3 h in a muffle furnace at 800 °C (heating rate 5 °C/min) in a static air atmosphere. The calcined material, after cooling to room temperature, was finely ground using a mechanical agate mortar and sieved to remove the fraction above 175 μm. The as-prepared catalyst was characterized by XRD to identify crystalline phases and the degree of crystallinity, and by ATR-FTIR (using the above-described equipment and conditions) to characterize the surface Ca species as previously described [46]. The diffractogram was recorded with a Rigaku Geigerflex diffractometer with Cu K_α radiation at 40 kV and 40 mA (2°/min). The XRD pattern, for Bragg angle 2θ in the range 5–70°, was compared with JCPDS standard files to identify the crystalline phases. The morphology of scallop shells, raw and calcined, was examined by scanning electron microscopy (SEM) using Joel JSM7001F FEG-SEM equipment. The powdered samples were spread over a carbon adhesive tape and covered with a thin metallic film of Pd-Au to improve their electronic conductivity.

3. Results and Discussion

The SEM images (Figure 3) of the scallop shells, before and after calcination, show that the initial morphology, agglomerates of thin calcite lamellae, transforms into a more disorganized morphology with cubic calcite crystallites, mixed with material, with irregular (in terms of both shape and size) crystallites belonging to the lime. Observed morphologies are similar to those reported in the literature for scallop shell-derived materials [47].

Characterization of the prepared catalyst by XRD revealed patterns belonging to lime contaminated with calcite (Figure 4), which agrees with the crystallite morphologies observed by SEM. The XRD data show that the calcination temperature was too low to fully decompose the calcium carbonate, which is the main component of scallop shells [45]. No XRD lines of portlandite (JCPDS 01-072-0156) were detected.

The FTIR spectrum (Figure 5) of the fresh catalyst, with the most intense band centered at 3641 cm⁻¹, shows that the catalyst surface is dominated by hydroxyl species [46], which are the active sites of methanol activation, forming surface methoxy species [42]. The less intense bands, for wavenumbers below 1600 cm⁻¹, belong to adsorbed CO₂ and non-decomposed calcite (CaCO₃) [46].

All the extraction methodologies allowed for a clear and homogeneous fat, but with different fat yields. The energy consumption and fat yield of each rendering process are summarized in

Table 2. Energy consumed in rendering processes, fat yield, and fat acidity.

Rendering Process	Equipment	Power (W)	Time (min)	Energy (kWh/100 g Raw Fat)	Fat Yield (g/100 g Raw Fat)	Fat Acidity (%)
Microwave	MW oven	1400	3	0.070	60 ± 3	0.81
Boiling water	Stove	2800	40	1.868	30 * ± 3	1.96
Acetone extraction	Heating mantle	325	240	1.300	15 ± 1	2.39

* 15% of moisture.

. The microwave-assisted method was the most performant methodology in the extraction procedure since it presented the lowest power consumption per gram of extracted fat. The high potential of microwaves in the production of biodiesel, including the fat extraction steps, has been previously reported [48]. The authors also report that microwaves have lower energies than Brownian motion, thus are unable to break chemical bonds, which prevents the degradation of raw materials by unwanted reactions.

The boiling water procedure presented high energy consumption and led to fat with a high moisture content, which is a disadvantage for biodiesel production by alcoholysis over calcium catalyst [46].

The fat extracted using boiling water and acetone had a greater acidity than that extracted by MW. The boiling water can aid in the hydrolysis of triglycerides, resulting in a greater FFA level (Equation (1), [49]). For acetone-extracted fat, the high acidity is justified by the duration of the operation and the fact that acetone is not miscible with the glycerin produced during the hydrolysis of triglycerides, which prompts the hydrolysis.

$$\mathrm{Triglyceride}+3 \mathrm{Water}\to 3\mathrm{FFA}+\mathrm{Glycerol}$$

(1)

The thermal stability of extracted beef tallow was assessed by thermogravimetry under airflow. The thermograms in Figure 6 show that all the characterized tallows have a higher thermal stability than soybean oil with maximum thermal degradation rates at temperatures higher than those observed for soybean oil. The acetone-extracted tallow showed the lowest thermal stability. The thermal instability of fats, and their derived biodiesel, is related to the degree of unsaturation of aliphatic carbon chains which are converted into hydroperoxides [50]. The highest thermal stability of animal fat biodiesel is well documented in the literature [51]. The duration of the microwave extractive process did not affect the thermal stability of the extracted fat.

The degree of unsaturation of fat is relevant for the cetane number of the produced biodiesel but it is also important for the fat oxidation during the rendering process. A higher degree of unsaturation implies lower oxidative stability [52] of fat and biodiesel [51] because the oxidation process starts with the oxygen attack on the -C=C- bonds [53]. The major products of fat oxidation contain hydroxyl, perohydroxyl, aldehydes, ketones, carboxylic acids, and trans double bonds, which can be characterized by FTIR. The formation of carboxylic acids during rendering is highly undesired because acids will react with the basic catalysts, leading to soap formation [38]. ATR-FTIR was used to characterize the extracted fats and soybean oil and bands were assigned based on data from the literature. The extracted fats showed intense reflectance bands in the range 1780–1690 cm⁻¹, which are ascribable to the -C=O ester (≈1746 cm⁻¹, [54]) and -C=O carboxylic acid (1710–1690 cm⁻¹, [55]) vibration modes (Figure 7).

For soybean oil, the band in this spectral range is simple and centered at 1744 cm⁻¹ (ester -C=O) with a slight displacement towards smaller wavenumbers, which is compatible with its low acidity (0.66%). The boiling water-extracted tallow shows an analogous band around 1744 cm⁻¹ with a shoulder around 1725 cm⁻¹ coming from FFA content (Table 2), which is higher than that of soybean oil. The fats extracted with MW show a C=O band broader than soybean oil and boiling water tallow and slightly displaced towards lower wavenumbers, which seems to indicate the formation of ketone and aldehyde species during rendering. Increasing the MW processing time, the spectral shoulder due to carboxylic acids becomes more intense, thus confirming the oxidizing effect of microwaves. The acetone-extracted fat shows a complex FTIR feature in the 1780–1690 cm⁻¹ range, with an intense band at 1710 cm⁻¹, which is compatible with its higher acidity (Table 2). Moreover, the acetone-extracted tallow presents a displacement toward smaller wavenumbers in the 2990–2810 cm⁻¹ range (-CH₂- and -CH₃- vibration modes) (Figure 7), which could be related to their higher degree of unsaturation and a different FFA chain length [56]. The FTIR band around 722 cm⁻¹, belonging to cis-substituted -C=C-, almost vanishes for MW-extracted tallows.

The extracted fats were used to produce biodiesel by methanolysis over the Ca catalyst described above. As previously reported [36], to diminish the undesired fat acidity and moisture effects, co-processing with alimentary grade soybean oil and solvent-assisted methanolysis (acetone) tests were carried out.

The data in Figure 8 show that the MW- and BW-extracted fats behave similarly in the methanolysis reaction but lead to a lower FAME yield than that of soybean oil (61.4–61.7% instead of 88.4%). This activity decay when tallows were processed is due to their acidity and moisture (Table 2). Co-processing BW tallow with SBO (50% W/W) allows the partial mitigation of the fat acidity (FAME yield raises from 61.4% to 74.5%). The maximum catalytic activity using tallow was attained in solvent-assisted (acetone) methanolysis tests for equimixtures of tallow and soybean oil. The acetone-extracted tallow shows similar behavior to the other extracted fats since a mixture containing 25% of acetone-extracted and microwave-extracted fats led to a FAME yield of 80%. The methanolysis reaction seems to be indifferent to the unsaturation degree of the processed fat. This result disagrees with that reported by Yang et al. [57]. For homogeneous catalysis with KOH, the authors state that unsaturated fats behave better during methanolysis because the limiting step is the conversion of diglycerides into monoglycerides, which is faster for polyunsaturated fat molecules. It seems that, in a heterogeneous catalyzed process, the limiting step of the mechanism is different to that identified by Yang et al. [57]. Dias et al. [42] emphasize methanol activation as the rate-limiting step in the oil methanolysis catalyzed by heterogeneous Ca catalysts.

The data in Figure 8 show that calcium catalysts are promising for biodiesel production from tallow. Data from the literature (

Table 3. FAME yields from tallow using different catalysts (data collected from [19]).

Catalyst	%Wcat (Fat Basis)	Methanol/Fat (Molar Ratio)	Reaction Temperature; Time	FAME Yield (%)
H2SO4	1.25	30	60 °C; 24 h	93.2
KOH	0.8	6	60 °C; 2 h	90.8
Immobilized lipase	20	12	50 °C; 48 h	89.7

) [19] show, for different catalysts, reaction conditions, either similar or more drastic, to those used to obtain similar FAME yields, even using homogeneous catalysts that allow higher reaction rates but which cannot be recovered and reused after the reaction.

All the processed fats allowed the co-production of glycerin with a low level of contamination with species from the reaction medium. The glycerin FTIR spectra in Figure 9 show a low content of MONG (Matter Organic Non-Glycerin [58]) for all the characterized samples, thus indicating a null effect of the fat extraction procedure on the co-produced glycerin purity.

4. Conclusions

Biodiesel was produced from beef tallow over a calcium-based heterogeneous catalyst. The raw fat, collected at a local butcher, was subjected to different rendering processes to separate the fat from the tissues. Fat rendering was accomplished using boiling water, microwave-assisted, and acetone extraction at reflux temperature. The boiling water process led to a fat material with a high moisture content (15%) and with the highest energy consumption. The microwave-assisted rendering promoted the slight oxidation of the extracted fat but with high yields (60% instead of 30% for boiling and 15% for acetone processes, by weight) and short processing times (3 min instead of 40 min in the boiling water process). The acetone-extracted fat showed the highest acidity which can arise from fat hydrolysis during the extraction process. All the extracted fats displayed similar behaviors in the methanolysis reaction (with FAME yields slightly higher than 60%, instead of 88% for pure SBO) over calcium-based catalysts with a decay of catalyst performances due to fat acidity and moisture. Such undesired effects were decreased by co-processing the extracted beef tallow with soybean oil and using acetone solvent. The data show that microwave-assisted rendering has a high potential for animal fat processing in biodiesel production. Moreover, since the separated tissues are free of chemicals, they can be used in animal food production.