1. INTRODUCTION
⌅The press mud, obtained during the purification of sugarcane juice, is mainly used as an organic fertilizer in sugarcane fields, as well as in biogas production. The mud contains sugarcane wax at 3% w/w of the cake (ICIDCA, 2000ICIDCA. 2000. Producción de cera y grasas a partir de cachaza. Manual de los Derivados de la Caña de Azúcar. Editorial científico-técnica, La Habana, 572-596.). During the second half of the last century in Cuba the American Sugar Co. Swenson (1947)Swenson OJ. 1947. Method of extracting wax from cachaza. US2508002A. developed and introduced a technology for the production of sugarcane wax by extraction with heptane, which was further exported to the United States for refining. During the 90s a wax refining technology was developed, and a wax refining plant was built in Cuba due to the interest in refined wax (WR) as raw material for the production of some pharmaceutical products.
Sugarcane wax has a dry base average composition of 68.53% wax, 28.61% oil and 2.47% impurities, also called resins. The sugarcane wax refining technology consists of a solvent extraction process with 95% v/v ethanol. The refining process begins by bringing the sugarcane wax into contact with cold 95% v/v ethanol for eight hours in a stirring tank. Then the dispersed solid is centrifuged and the heavy phase, rich in refined wax, is subsequently extracted with hot ethanol in an extractor-decanter centrifuge; the refined wax is cooled and separated in a disk centrifugal bowl. After that, the refined wax is dried and packaged for later commercialization. This technology uses 95% v/v ethanol as solvent because it is a renewable solvent of national production. There are several reports about the use of ethanol for vegetable oil extraction, but there is no evidence about how useful it is compared to other solvents.
The extraction of vegetable oils with ethanol is not an exclusive feature of sugarcane wax refining, since literature reports several works about the solubility degree of this solvent with several vegetable oils. During the first half of the last century, Taylor et al., (1936)Taylor TI, Larson L, Johnson W. 1936. Miscibility of Alcohol and Oils. Ind. Eng. Chem. 28, 616-618. https://doi.org/10.1021/ie50317a030 investigated the solubility of 14 different vegetable oils in ethanol, including corn, olive, peanut, peach, soybean and sunflower oils, among others. This study showed that they have different solubility degrees according to their composition and that they vary from 1 to 10 kg of oil per 100 kg of ethanol at 90% w/w and between 5 to 15 kg of oil for 100 kg of absolute ethanol. Measurements were made at 25 ºC. The results showed the influence of small amounts of water on the solubility of vegetable oils in this solvent. In their studies Kaparthi et al., (1955)Kaparthi R, Charti KS. 1955. Solubilities of Vegetable Oils in Aqueous Ethanol and Ethanol- Hexane Mixtures. J. Am. Oil. Chem. Soc. 36, 77-80. https://doi.org/10.1007/BF02540248 concluded that even in the case of absolute ethanol, this solvent is not a useful solvent for vegetable oils, such as sunflower, peanut and cottonseed because concentrations higher than 10 kg of oil per 100 kg of ethanol at temperatures lower than 30 ºC are not achieved, although at 70 ºC the sunflower, peanut and cottonseed oils are soluble.
Rao et al., (1955)Rao RK, Krishna MG, Zaheer RH, Arnold LK. 1955. Alcoholic Extraction of vegetable oils. Part I. Solubilities of Cottonseed, Peanut, Sesame and Soybean Oils in Aqueous Ethanol. J. Am. Oil. Chem. Soc. 32, (7) 420-423. https://doi.org/10.1007/BF02639700 carried out research in the extraction of vegetable oils from cotton, peanut, sesame and soybean seeds using ethanol of various alcoholic degrees and under several temperature conditions. For all the oils, they determined solubility curves and showed that there was a linear relationship between the “critical temperature”, defined as the temperature at which the alcohol solution reached saturation for the oil studied, and the alcohol concentration used.
Given the importance of ethanol in Brazil, Freitas et al., (2010)Freitas SP, Lago RCA. 2007. Equilibrium Data for the Extraction of Coffee and Sunflower Oils with Ethanol. Braz. J. Food. Technol. 10, 220-224. showed the advantages of the use of ethanol instead of hexane for the extraction of coffee and sunflower oil, owing to the ecological nature of ethanol. With its use for the extraction of coffee and sunflower oil, they obtained results of 11.4 and 18.6%, respectively, during their extraction of seeds, although these results are lower than those obtained with hexane. A similar study was developed by Rodrigues et al., (2010)Rodrigues ChEC, Oliveira R. 2010. Response surface methodology applied to the analysis of rice bran oil extraction process with ethanol. Int. J. Food. Sci. Tech. 45, 813-820. https://doi.org/10.1111/j.1365-2621.2010.02202.x for soybean oil and concluded that the use of ethanol for this process is feasible; while Rodrigues et al., (2010)Rodrigues ChEC, Oliveira R. 2010. Response surface methodology applied to the analysis of rice bran oil extraction process with ethanol. Int. J. Food. Sci. Tech. 45, 813-820. https://doi.org/10.1111/j.1365-2621.2010.02202.x evaluated it in the extraction of rice oil. Both studies used experimental designs for their respective purposes and concluded that temperature and the alcoholic degree were the main independent variables, which agreed with the results reported in previous works on the use of ethanol as a vegetable oil solvent.
The solubility of canola, maize, jatropha and maracauba oils in ethanol at temperatures of 25-60.15 ºC was investigated by Da Silva et al., (2010) Silva CAS da, Sanaiotti G, Lanza M, Follegatti-Romero LA, Meirelles AJA, Batista EAC. 2010. Mutual Solubility for Systems Composed of Vegetable Oil + Ethanol + Water at Different Temperatures. J. Chem. Eng. Data 55, 440-447. https://doi.org/10.1021/je900409p ; while Shariati et al., (2013)Shariati A, Azaribeni A, Hajighahramanzadeh P, Loghmani Z. 2013. Liquid Liquid Equilibria of Systems Containing sunflower. Oil, Ethanol and Water. APCBEE. Procedia 5, 486 - 490. https://doi.org/10.1016/j.apcbee determined the solubility curves of sunflower oil in ethanol.
There are only a few studies in which Hansen’s solubility theory was used as a tool to determine the HSPs of vegetable oils to know which would be the most suitable solvent for these oils. Batista et al., (2015)Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 determined the HSPs of used frying oil, coconut oil, palm oil and biodiesel. They confirmed that used frying oil and biodiesel-diesel blends were feasible alternatives that could be used as additives for diesel fuel.
A study by De la Peña-Gil et al., (2016)Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 to develop a specific method to determine the HSPs of complex vegetable oils used group contribution methods with two approaches: the first assumed that the contributions of the 3 forces could be subdivided into larger functional groups (i.e. fatty acids and fatty acid methyl esters), which are additives; while the second approach assumed that vegetable oils are composed of simple triglycerides mixtures in the same mass fractions as fatty acids. They concluded that the HSPs calculated for vegetable oils with the second assumption provided similar values to the HSPs obtained from the HSPiP software.
During studies developed by The Cuban Research Institute of Sugarcane Derivatives (ICIDCAICIDCA. 2000. Producción de cera y grasas a partir de cachaza. Manual de los Derivados de la Caña de Azúcar. Editorial científico-técnica, La Habana, 572-596.) in the 90s about sugarcane oil extraction from sugarcane wax with ethanol at 18-20 ºC, only 17.9% efficiency was achieved with ethanol at 93% v/v and 42.9% with absolute ethanol. Therefore, the use of absolute alcohol in the refining process of sugarcane wax was suggested, but investigating the possibility of using other more efficient solvents for this purpose were recommended, topics which are investigated in this work.
In the current refined wax plant, low efficiencies and high 95% v/v ethanol consumption rates are reported during sugarcane oil extraction. The objective of this work was to verify whether ethanol 95% v/v is a suitable solvent for the sugarcane wax refining process using Hansen’s solubility theory, as well as to determine other solvents that can be suitable for this process.
2. MATERIALS AND METHODS
⌅2.1. Oil purification
⌅Sugarcane oil obtained from the sugarcane wax refining factory annexed to the Majibacoa sugar mill in Las Tunas province, Cuba, was used for this study. HPLC grade acetone at 25 ºC was used for the purification process, according to factory laboratory standards. This impure oil obtained in the factory, with an average dry base composition of 78.9% oil, 15.78% wax and other impurities, was subsequently dissolved in acetone at 25 ºC with an acetone-oil ratio of 4:1 and filtrated under vacuum with a Buchner funnel and fast filtration paper to be purified. Acetone with oil was evaporated in a vacuum rotary evaporator and then the oil was dried in an oven at 100 ºC. The oil obtained was subjected to a second purification step.
2.2. Hansen Solubility Parameters determination (HSPs)
⌅2.2.1. Fundamental
⌅HSPs determination is a useful tool for solvent selection in the extraction of natural chemical products because it expresses the affinity degree between a solute and any solvent based on the Hansen theory, which stipulates that “likes attract likes”. There are two important groups of methods for HSPs determination, experimental and theoretical methods, but only experimental methods are evaluated in this work for the HSPs determination of sugarcane oil.
The basic equation governing the assignment of Hansen parameters establishes that the total cohesion energy between two molecules, E, must be the sum of the individual energies: non-polar interactions or atomic dispersion (ED), molecular di-polar interactions (EP) and hydrogen-bonding interactions (EH) (Hansen C.M. 2007Hansen CM. 2007. Solubility parameters-an introduction. Hansen solubility parameters: a user’s handbook. CRC, Boca Raton, 43-66. ISBN 978-0-9551220-2-6), so:
Dividing this by the molar volume gives the square of the Hildebrand solubility parameter as the sum of the squares of the Hansen D, P, and H components: dispersion (δD), polar (δP), and hydrogen bonding (δH).
The Hansen parameters of different solutes and solvents may be plotted in a tridimensional plot x,y,z (δD,δP ,δH). The closer the solute-solvent pair is in the three-dimensional space, the more soluble they are. The distance between the solvent (a) and solute (b) is usually called Ra and can be determined by the following expression:
Experimental evaluation of a solute with various solvents, generally between 40-50 different solvents, and its classification into “good” or “bad” solvents by visual inspection, helps us to determine the HSPs of the solute (solute’s coordinates within Hansen space). Considering that the “good solvents” surround the solute forming a hypothetical sphere, called “Hansen’s Sphere”, good solvents can also be found on the surface of the Hansen sphere, so for them Ra is equal to the radius of the sphere (R0). By defining the Ra/R0 ratio as the relative energy difference (RED), the good solvents will obtain RED values less than 1 and the bad solvents higher. Both the radius of the sphere and the center (HSPs) are determined by the use of appropriate multi-response optimization algorithms. (Abbott and Yamamoto, 2015Abbott S. and Yamamoto H. (2015) HSPiP Software, 5th edn. 5.2.05; Díaz and Hernández, 2020Díaz de los Ríos M, Hernández Ramos E. 2020. Determination of the Hansen solubility parameters and the Hansen sphere radius with the aid of the solver add-in of Microsoft Excel, SN Applied Sciences 2, 676, https://doi.org/10.1007/s42452-020-2512-y ).
2.2.2. Hansen solubility parameters determination through classic Hansen method
⌅The determination of Hansen’s parameters was executed with 48 different analytical degree solvents as shown in Table 1. Solvent-oil samples in a 9:1 ratio were placed in duplicate test tubes and shaken for 24 hours at 30 0C and then allowed to rest 72 h for a visual inspection of the presence of phase separation (Batista et al., 2015Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 ).
| Noa. | Solvent | Dispersion component (δD) MPa1/2 | Polar component (δP) MPa1/2 | Hydrogen bonding component (δH) MPa1/2 | Scoreh | Relative Energy Difference (RED) | Molecular volume (MVol) cm3/mol |
|---|---|---|---|---|---|---|---|
| 197 | 1-Decanol | 16.0 | 4.7 | 10.5 | 1 | 0.146 | 191.8 |
| 542 | 1-Octanol | 16.0 | 5.0 | 11.2 | 1 | 0.189 | 158.2 |
| 375 | Ethylene Glycol Monobutyl Ether | 16.0 | 5.1 | 12.3 | 1 | 0.256 | 131.8 |
| 306 | 1,4-Dioxane | 17.5 | 1.8 | 9.0 | 1 | 0.295 | 85.7 |
| 930 | 1-Hexanol | 15.9 | 5.8 | 12.5 | 1 | 0.319 | 125.2 |
| 617 | Tetrahydrofuran (Thf) | 16.8 | 5.7 | 8.0 | 1 | 0.334 | 81.9 |
| 429 | Isoamyl Acetate | 15.3 | 3.1 | 7.0 | 1 | 0.359 | 150.2 |
| 328 | Ethyl Acetate | 15.8 | 5.3 | 7.2 | 1 | 0.362 | 98.6 |
| 182 | Cyclohexanol | 17.4 | 4.1 | 13.5 | 1 | 0.371 | 105.7 |
| 545 | Oleic Acid | 16.0 | 2.8 | 6.2 | 1 | 0.392 | 319.7 |
| 552 | 1-Pentanol | 15.9 | 5.9 | 13.9 | 1 | 0.418 | 108.6 |
| 732 | 2-Methyl-2-Butanol | 15.3 | 6.1 | 13.3 | 1 | 0.422 | 109.6 |
| 93 | 2-Butanol | 15.8 | 5.7 | 14.5 | 1 | 0.457 | 92.0 |
| 524 | Methylene Dichloride (Dichloromethane) | 17.0 | 7.3 | 7.1 | 1 | 0.505 | 64.4 |
| 156 | Chloroform | 17.8 | 3.1 | 5.7 | 1 | 0.520 | 80.5 |
| 92 | 1-Butanol | 16.0 | 5.7 | 15.8 | 1 | 0.558 | 92.0 |
| 649 | Trichloroethylene | 18.0 | 3.1 | 5.3 | 1 | 0.572 | 90.1 |
| 58 | Benzyl Alcohol | 18.4 | 6.3 | 13.7 | 1 | 0.581 | 103.8 |
| 431 | Isobutyl Alcohol | 15.1 | 5.7 | 15.9 | 1 | 0.604 | 92.9 |
| 5 | Acetic Acid | 14.5 | 8.0 | 13.5 | 1 | 0.624 | 57.6 |
| 570 | 2-Propanol | 15.8 | 6.1 | 16.4 | 1 | 0.628 | 76.9 |
| 255 | Diethyl Ether | 14.5 | 2.9 | 4.6 | 1 | 0.628 | 104.7 |
| 254 | Diethyl Carbonate | 15.1 | 6.3 | 3.5 | 1 | 0.732 | 121.7 |
| 569 | 1-Propanol | 16.0 | 6.8 | 17.4 | 1 | 0.736 | 75.1 |
| 7 | Acetone | 15.5 | 10.4 | 7.0 | 1 | 0.751 | 73.8 |
| 698 | o-Xylene | 17.8 | 1.0 | 3.1 | 1 | 0.762 | 121.1 |
| 999 | So Mixb | 18.7 | 3.8 | 3.7 | 1 | 0.770 | 86.0 |
| 367 | Ethylene Dichloride | 18.0 | 7.4 | 4.1 | 1 | 0.772 | 79.4 |
| 398 | Formic Acidg | 14.6 | 10.0 | 14.0 | 0 | 0.779 | 37.9 |
| 999 | So Mixc | 18.5 | 4.2 | 2.8 | 1 | 0.823 | 103.0 |
| 598 | Pyridine | 19.0 | 8.8 | 5.9 | 1 | 0.838 | 80.9 |
| 187 | Cyclopentane | 16.4 | 0.0 | 1.8 | 1 | 0.851 | 94.6 |
| 637 | Toluene | 18.0 | 1.4 | 2.0 | 1 | 0.859 | 106.6 |
| 52 | Benzene | 18.4 | 0.0 | 2.0 | 1 | 0.924 | 89.5 |
| 11 | Acetophenone | 18.8 | 9.0 | 4.0 | 1 | 0.929 | 117.4 |
| 325 | Ethanol | 15.8 | 8.8 | 19.4 | 1 | 0.992 | 58.6 |
| 181 | Cyclohexane | 16.8 | 0.0 | 0.2 | 1 | 0.996 | 108.9 |
| 122 | Carbon Tetrachloride (0 Dipole Moment) | 17.8 | 0.0 | 0.6 | 1 | 0.998 | 97.1 |
| 999 | So Mixd | 15.3 | 12.4 | 5.1 | 1 | 1.000 | 98.0 |
| 999 | So Mixe | 16.2 | 12.1 | 4.3 | 1 | 1.000 | 76.0 |
| 297 | Dimethyl Formamide (Dmf) | 17.4 | 13.7 | 11.3 | 0 | 1.002 | 77.4 |
| 540 | Octane | 15.5 | 0.0 | 0.0 | 0 | 1.019 | 163.4 |
| 409 | Heptane | 15.3 | 0.0 | 0.0 | 0 | 1.024 | 147.0 |
| 417 | Hexane | 14.9 | 0.0 | 0.0 | 0 | 1.040 | 131.4 |
| 999 | So Mixf | 16.5 | 8.1 | 20.9 | 0 | 1.081 | 82.0 |
| 670 | 2,2,4-Trimethylpentane | 14.1 | 0.0 | 0.0 | 0 | 1.086 | 165.5 |
| 437 | Isopentane | 13.8 | 0.0 | 0.0 | 0 | 1.107 | 117.1 |
| 303 | Dimethyl Sulfoxide (Dmso) | 18.4 | 16.4 | 10.2 | 0 | 1.289 | 71.3 |
| 456 | Methanol | 14.7 | 12.3 | 22.3 | 0 | 1.425 | 40.6 |
| 10 | Acetonitrile | 15.3 | 18.0 | 6.1 | 0 | 1.442 | 52.9 |
| 368 | Ethylene Glycol | 17.0 | 11.0 | 26.0 | 0 | 1.631 | 55.9 |
| 406 | Glycerol | 17.4 | 11.3 | 27.2 | 0 | 1.750 | 73.4 |
| 696 | Water | 15.5 | 16.0 | 42.3 | 0 | 3.203 | 18.0 |
a Solvent identification number in the HPSiP program.
b Solvent mixture (57% v/v benzene, 43% v/v pyridine) representing 1-Bromonaphthalene
c Solvent mixture (acetophenone 43% v/v, benzene 35% v/v, toluene 22% v/v) representing Chlorobenzene
d Solvent mixture (acetonitrile 54% v/v, diethyl carbonate 39% v/v, oleic acid 7% v/v) representing Butyronitrile
e Solvent mixture (acetonitrile 60% v/v, cyclohexane 25% v/v, acetophenone 15% v/v) representing 2-Nitropropane
f Solvent mixture (glycerol 50% v/v, isobutyl alcohol 40% v/v, 1,4 Dioxane 10% v/v) representing 1,3 Butanediol
g Bad solvent within the Hansen sphere
h Score 1: good solventsScore 0: bad solvents
Visual inspection was performed by each author separately and individual ratings were compared. The samples were divided into three groups: soluble (without phase separation), not soluble (phase separation) and doubtful. The doubtful group included the samples in which their evaluation differed for each author and also those classified directly as such. Due to the nature of the oil used, the solution tended to darken and it was sometimes difficult to visually determine whether or not there was phase separation. Therefore, tests classified as “doubtful” were observed under the microscope or a drop of sample was placed in a filter paper; if a black spot was formed at the sample application point, then it was classified as not soluble, on the other hand, if the sample formed a ring of uniform color, it was soluble according to the established procedure (Redelius, 2004Redelius P. 2004. Bitumen Solubility Model Using Hansen Solubility Parameter. Energ. Fuel 18, 1087-1092. https://doi.org/10.1021/ef0400058 ).
Samples in which the solvent solubilized the solute received a score of 1 and samples in which the solvents did not solubilize the solute the score were labelled 0. Both the HSPiP software using the genetic algorithm and the improved Microsoft Excel workbook from the authors were used for data processing to define the HSPs for sugarcane oil (Abbott and Yamamoto, 2015Abbott S. and Yamamoto H. (2015) HSPiP Software, 5th edn. 5.2.05; Díaz and Hernández, 2020Díaz de los Ríos M, Hernández Ramos E. 2020. Determination of the Hansen solubility parameters and the Hansen sphere radius with the aid of the solver add-in of Microsoft Excel, SN Applied Sciences 2, 676, https://doi.org/10.1007/s42452-020-2512-y ).
2.2.3. Fit improvement of Hansen solubility parameters
⌅The reliability of the parameters was mainly based on the fact that the solvents used in the study would cover the largest possible area of the three-dimensional space in such a way that the edge of the sphere and its center could be correctly defined.
To check the quality of fit, the values called “Core” by HSPiP software were used. This option showed how much the solute´s HSPs could vary in different directions without too high a penalty. Clearly the larger the “Core” values, the less the Sphere was defined, so some additional solvents may be needed to redefine it.
One way to improve the fit is to use the “Sphere Radius Chk option” in HSPiP software. It goes through a list of green solvents that can be edited and finds any that are near the boundary of the Sphere (i.e. they have a RED of 0.9 to 1.1). It then checks each of those solvents for closeness to solvents in your test list. If they are close, then you already have enough data in that area. But if no solvent is nearby then it is likely that this test solvent would provide useful new information to improve the quality of the fit.
The suggested solvents by the software to achieve a better fit were: 2-nitropopane, 1-bromonaphthalene, 1,3-butanediol, butyronitrile and chlorobenzene. Since they were not available the “Solvent Optimizer” option was used. This option calculates solvent mixtures capable of achieving similar characteristics to recommended solvents. The mixtures were named Somix, tested, classified and scored according to the procedure described above. To determine the HSPs of sugar cane oil, a total of 53 solvents were tested, including mixtures.
2.3. Intrinsic viscosity measurements method
⌅One of the more promising methods to evaluate polymer HSPs for limited data is by using intrinsic viscosity (Hansen, 2007Hansen CM. 2007. Solubility parameters-an introduction. Hansen solubility parameters: a user’s handbook. CRC, Boca Raton, 43-66. ISBN 978-0-9551220-2-6).
In this technique, intrinsic viscosities determined with different solvents are normalized by dividing each one by the highest intrinsic viscosity value. These normalized data (≤ 1) are then used in a weighted averaging technique, according to the following equations, to determine the HSPs of the solute (Hansen, 2007Hansen CM. 2007. Solubility parameters-an introduction. Hansen solubility parameters: a user’s handbook. CRC, Boca Raton, 43-66. ISBN 978-0-9551220-2-6).
The members on the left side of each equation represents dispersion, polar and hydrogen bonding forces respectively; the subscript 2 represents the solute. On the right side of each equation is, as appropriate, weighted averaging of the component of the force of attraction according to normalized intrinsic viscosity ([η] i ); the subscript i represents the solvent.
2.3.1. Determination of Hansen solubility parameters using the intrinsic viscosity method
⌅To determine the HSPs using the intrinsic viscosity method, 7 solvents were chosen with the criterion of expanding the exploration zone for the calculation of the HSPs of sugarcane oil. Among them were: isopropanol, 1,4 Dioxane, Ethyl acetate, 1 Decanol, 1 Octanol, 2- Methyl -2- Butanol and o-Xylene.
To calculate the intrinsic viscosity, sugarcane oil solutions were prepared for each solvent chosen from 10; 15; 20; 25; 30 kg / m3. A ubbelohde viscometer was used to measure the relative viscosity (ηre ) for each concentration. Each measurement was performed in triplicate while maintaining a temperature of 25 0C (Stanciu I, 2009Stanciu I. 2009. Study solubility of two copolymers used as viscosity improvers for multi-grade oil. AUDJG - Food Technology, 323-326).
A linear regression of the inherent viscosity was adjusted against the concentrations, where the intercept is the value of the intrinsic viscosity:
Where ηr is the reduced viscosity which was calculated by:
With the HSPs values of the solvents studied and those of the normalized intrinsic viscosity, the HSPs of the solute was estimated.
3. RESULTS AND DISCUSSION
⌅3.1. Hansen solubility parameter determination
⌅The HSPs obtained for sugarcane oil using HSPiP software were δD = 16.56 MPa1/2, δP = 2.78 MPa1/2 and δH = 10.54 MPa1/2. Of all solvents analyzed, 14 were alcohols, among which 12 were aliphatic with two aromatics. For this particular case, methanol was the only one that had a RED greater than 1; other alcohols were good solvents for sugarcane wax oil. 1-decanol was the best of all the solvents analyzed. Usually, according to the international chemical safety sheet, it is used in the manufacture of plasticizers, lubricants, surfactants and as solvent, but its use in this application requires a technical-economic feasibility analysis.
When running the HSPiP program, the HSPs of the sugarcane oil varied their values, thus slightly moving the center of the Hansen sphere in three-dimensional space according to an indicator of the program, mentioned above as the “Core”. At times. when the program was run, the absolute ethanol did not fit inside the sphere, leaving it outside with a RED of 1,009, which is incorrect, since it had been classified as a good solvent. In addition, other solvents that were classified as bad remained within the sphere, including Dimethyl Formamide (Dmf), formic acid, among others, depending on where the algorithm had defined the center of the sphere.
The HSPs presented at the beginning of this section were the best fit that was achieved, with 97.9% fit, in which all the solvents classified as good were within the sphere and only formic acid, which turned out to be classified as a bad solvent was wrongly situated inside.
Of course, accuracy depends on the reliability of the data used, which is indicated by the fit of the Hansen sphere “Core” Values. According to Abbott (2013)Abbott S. 2013. www.hansen-solubility.com , if errors were made during sample evaluation, many solvents that do not dissolve the solute would fall inside the sphere or many that dissolve the solute would be outside. Figure 2 shows that only one solvent that does not dissolve the solute is within the sphere, which validates the reliability of the input data.
3.1.1. Fit improvement in Hansen solubility parameters
⌅For the first fitting, the “Core” was 1.30, 1.60, 1.35. In the program it is defined that if, for the parameter analyzed, the core is greater than 1, it is possible that the adjustment was inaccurate. This tells us that the amount of solvents analyzed was not sufficient to guarantee a result with high reliability.
Table 1 shows the classification of solvents as Good and Bad. Recalculating the HSPs provided: δD = 16.24 MPa½; δP = 3.21 MPa½; and δH = 10.34 MPa½ with a “Core” of 1.00, 0.50, 0.85, respectively. This indicates a more reliable result, although the parameter δD could still be investigated. The results obtained from the use of the GRG nonlinear optimization method of Microsoft Excel Workbook were 16.38, 3.10 and 10.25 for δD, δP and δH, respectively, not quite different from the use of HSPiP software.
Absolute ethanol was located at the limit of the Hansen solubility sphere with a RED = 0,993. this means that any mixture of ethanol-water that is used for the extraction of this oil would not be good. Absolute ethanol is produced in Cuba and is an environmentally friendly solvent capable of decreasing the plant’s consumption rates and increasing extraction efficiency. If used, the technology would not change much, making it a possible substitute for 95% v / v ethanol currently used.
Table 2 compares different types of vegetable oils according to their HSPs and it is observed that absolute ethanol is only reported as a good solvent for sugar cane oil, being the only one with a RED less than 1. In order to know why this oil has a larger δH, an infrared spectroscopy was performed to identify its main structural groups.
| Solute | Dispersion component (δD) MPa1/2 | Polar component (δP) MPa1/2 | Hydrogen bonding component (δH) MPa1/2 | Distance between solute-ethanol (Ra) | Hansen sphere radius (R0) | Relative Energy Difference (RED) | References |
|---|---|---|---|---|---|---|---|
| Sugarcane oil | 16.24 | 3.21 | 10.34 | 10.72 | 10.80 | 0.993 | This work |
| used frying oil | 15.35 | 3.77 | 6.87 | 13.52 | 9.57 | 1.413 | Batista et al., 2015Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 |
| Coconut oil | 14.95 | 4.63 | 6.98 | 13.22 | 9.80 | 1.349 | Batista et al., 2015Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 |
| Palm oil | 17.54 | 3.34 | 4.08 | 16.62 | 7.48 | 2.222 | Batista et al., 2015Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 |
| Soybean oil | 16.50 | 2.00 | 2.70 | 18.08 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Canola oil | 16.50 | 2.10 | 2.60 | 18.14 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Coconut oil | 16.20 | 2.50 | 2.80 | 17.77 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Cocoa butter | 16.20 | 2.10 | 2.20 | 18.47 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Soybean oil (tgas + glycerol) | 16.80 | 1.90 | 9.60 | 12.15 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Canola oil (tgas + glycerol) | 16.80 | 1.90 | 9.60 | 12.15 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Coconut oil (tgas + glycerol) | 16.40 | 2.30 | 11.40 | 10.37 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Cocoa butter (tgas + lycerol) | 16.50 | 1.90 | 9.50 | 12.14 | NR | NR | De la Peña et al., 2016Peña-Gil A de la, Toro-Vázquez JF, Rogers MA. 2016. Simplifying Hansen Solubility Parameters for Complex Edible Fats and Oils. Food. Biophys. 3, 283-291. https://doi.org/10.1007/s11483-016-9440-9 |
| Pine oil | 15.60 | 3.00 | 9.80 | 11.22 | NR | NR | Hansen, 2007Hansen CM. 2007. Solubility parameters-an introduction. Hansen solubility parameters: a user’s handbook. CRC, Boca Raton, 43-66. ISBN 978-0-9551220-2-6 |
| Olive oil | 15.90 | 1.20 | 5.40 | 15.93 | 12.00 | 1.328 | Hansen, 2007Hansen CM. 2007. Solubility parameters-an introduction. Hansen solubility parameters: a user’s handbook. CRC, Boca Raton, 43-66. ISBN 978-0-9551220-2-6 |
| Castor oil | 15.90 | 4.60 | 12.00 | 8.51 | NR | NR | Hansen, 2007Hansen CM. 2007. Solubility parameters-an introduction. Hansen solubility parameters: a user’s handbook. CRC, Boca Raton, 43-66. ISBN 978-0-9551220-2-6 |
| Palm oil | 17.70 | 3.50 | 3.70 | 17.00 | 4.70 | 3.617 | Hansen, 2007Hansen CM. 2007. Solubility parameters-an introduction. Hansen solubility parameters: a user’s handbook. CRC, Boca Raton, 43-66. ISBN 978-0-9551220-2-6 |
| Linseed oil | 14.17 | 3.65 | 3.65 | 16.89 | NR | NR | Burke, 1984Burke, J. 1984. The Book and Paper Group Annual, vol.3, https://cool.culturalheritage.org.bpg |
| Cottonseed oil | 15.14 | 3.39 | 4.07 | 16.31 | NR | NR | Burke, 1984Burke, J. 1984. The Book and Paper Group Annual, vol.3, https://cool.culturalheritage.org.bpg |
NR Not reported
Figure 1a shows the infrared spectrum of crude wax oil, which was very similar to the infrared spectrum of castor oil and pine resin (Figures 1b and 1c, respectively) (Institute of Chemistry University of Tartu, 2019Institute of Chemistry University of Tartu, Estonia http://lisa.chem.ut.ee/IR_spectra/paint/binders/castor-oil/ ½, 11/3/2019.). A wide band of low intensity between 3200-3650, bands between 1540-1850, 2200-3000 and signals in 1200-1420 suggest the presence carboxylic groups. This result allows us to corroborate a possibility that δH is influenced by the presence of carboxylic groups.
Figure 2 shows the plotted Hansen sphere using 3-D axes for sugarcane oil and solvents as shown in Table 1. Blue spheres represent good solvents and red cubes represent bad solvents. The unfilled cube with the red edges (formic acid) indicate a bad solvent that fell inside the Hansen sphere. It could be because the program needs more information for a better fit, such as the evaluation of more solvents. When running the program for different times, the HSPs values of the sugarcane oil did not vary much and reached an adjustment of 98.1%.
According to RED values, sugarcane oil is soluble in 39 of the solvents tested. For Batista et al., (2015)Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 the best solvents for the vegetable oils analyzed were n-butyl acetate, ethyl acetate and o-xylene. In the case of ethyl acetate and o-xylene, which were tested in the present study, they were good solvents, but they were not among the best. Contrary to the results obtained by Batista et al., (2015)Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 , who exposed 1-decanol as a poor solvent, in the current work 1-decanol was the best solvent for crude wax oil, given its similar δH and δP values. This could be due to a different composition between the oils analyzed by Batista et al., (2015)Batista MM, Guirardello R, Krähenbühl MA. 2015. Determination of the Hansen Solubility Parameters of Vegetable Oils, Biodiesel, Diesel, and Biodiesel-Diesel Blends. J. Am. Oil. Chem. Soc. 92, 95-109. https://doi.org/10.1007/s11746-014-2575-2 and the current work, which were expressed in Hansen’s parameters.
A large amount of good solvents was consistent with the R0 calculated by the HSPiP program which adjusted R0 = 10.7. Solvents close to RED = 1, such as benzene, cyclohexane, N, N-dimethylformamide, ethanol, carbon tetrachloride, octane, acetophenone and the Somix used to improve HSPs of sugarcane oil, are those that improved the Hansen sphere by its influence to R0 value.
3.2. Calculation of Hansen’s solubility parameters using the intrinsic viscosity method
⌅This method was used to verify the HSPs of the sugarcane oil obtained from the solvent testing method, since no information about them was found in the literature. The results are shown in Table 3. The HSPs calculated using the intrinsic viscosity method were: δD = 16.16 MPa½; δP = 4.44 MPa½; and δH = 10.40 MPa½. The Results obtained by this method are compared with those reported by the HSPiP software in Table 4, where it may be observed that the values calculated by both methods are similar.
| Solvent | Intrinsic viscosity [η]i, kg /m3 | Normalized intrinsic viscosity[η]i |
|---|---|---|
| Isopropanol | 0.0105 ± 0.0026 | 0.203 |
| 1.4 Dioxane | 0.0107 ± 0.0018 | 0.211 |
| Ethyl Acetate | 0.0034 ± 0.0041 | 0.065 |
| 1 decanol | 0.0494 ± 0.0008 | 0.974 |
| 1-Octanol | 0.0018 ± 0.0005 | 0.036 |
| 2-Methyl-2-Butanol | 0.0507 ± 0.0009 | 1.000 |
| o-Xylene | 0.0260 ± 0.0115 | 0.513 |
| Method | Dispersion component (δD) MPa1/2 | Polar component (δP) MPa1/2 | Hydrogen bonding component (δH) MPa1/2 |
|---|---|---|---|
| Hansen | 16.24 | 3.21 | 10.34 |
| Intrinsic viscosity | 16.16 | 4.44 | 10.40 |
In the intrinsic viscosity method, some inaccuracies in the results may arise due to the choice of solvents for the study, more in this case, where the HSPs of sugarcane oil were unknown; but the result obtained through the method of Intrinsic viscosity strengthens the one obtained using Hansen’s Solubility Theory.
4. CONCLUSIONS
⌅According to the crude wax oil, the HSPs obtained in this work indicated that absolute ethanol could be used as a solvent for its extraction. Although it is not the best solvent, it has the advantages of being of national production, environmentally-friendly with very low toxicity. The good solvents determined in this work must be evaluated from the economic point of view. Sugarcane oil HSPs resulting from the analysis are: δD = 16.24 MPa1/2; δP = 3.21 MPa1/2: δH = 10.34 MPa1/2.