Аннотация

Introduction. The existing methods of animal fat obtaining have certain disadvantages, hence fat extraction study highly is relevant. Electrochemically activated solutions are known to have a great potential for animal fat extraction. The present paper introduced a new advanced fat obtaining technology based on the principle of electrochemical activation.
Study objects and methods. The research featured ostrich fat obtained by wet rendering in water and in an electrochemically activated solution (catholyte) using various processing methods and technological parameters. Standard methods helped define the physical and chemical parameters of the obtained fat samples.
Results and discussion. The paper introduced a technological and hardware setup of an ostrich fat production line with the necessary equipment specifications. The research made it possible to define the optimal parameters for fat extraction: the salt concentration for the catholyte = 4 g/100 cm³, voltage = 40–42 V, pH = 11, and redox potential of the catholyte = between –600 and –700 mV. During the fat processing, cell membranes in the electrolyte were destroyed, which inactivated the enzyme system. The obtained combination of physical and chemical factors resulted in ostrich fat of high quality. Fat extraction in an electrochemically activated solution (catholyte) catalyzed the process and increased the fat yield, regardless of the processing temperature. The fat yield exceeded 58% at 55°C and catholyte pH of 11.0. At 95–100°C and pH of 9.5–10.6, it exceeded 95%.
Conclusion. The new technology increased the fat yield, maintained its high quality, and reduced the processing cost. Therefore, the developed production line could be recommended for fat extraction of farm animals, depending on the intended use.

Ключевые слова

Fatty acid, fats, raw material, electrolysis, catholyte, food production, water, solutions

ВВЕДЕНИЕ

Fat rendering technology should take into account the specific character of the technical solutions. Wet rendering produces a three-phase system, which includes fat, broth, and greaves. However, the process is slow and renders no fat at low processing temperatures. Dry rendering involves high temperatures and results in a two-phase system of greaves and fat. Chemical rendering is fast and highly productive, but acid or alkali treatment limits the intended use of the final product. Hydrolytic rendering destroys the protein component of fat with hot alkali. Unfortunately, the resulting fat is instable, and the greaves cannot be used for feed production. Hydromechanical and electric pulse methods are most effective for glue and gelatin production, as they destroy the connective tissue of the fat cell of tubular bones. Extraction methods based on organic solvents are extremely rare in the meat industry: not only are they fire-hazardous and labor-consuming, but they also require sophisticated equipment. It is fat rendering with electrochemically activated solutions that gives the most positive results [1]. The last decades have seen a growing scientific interest in the use of water as a potential source of unconventional chemical reactions [2–5].

Electrochemically activated solutions make various production processes less energy-, time-, and materialconsuming [6]. Electrolysis of water has a number of advantages over mechanochemical, magnetic, and thermal activation methods [7]. Even at low concentrations, electrochemically activated solutions demonstrate properties similar to concentrated acid and alkaline solutions [8, 9]. Domestic and foreign research confirmed their effectiveness in various fields and resulted in a number of patents [10–18]. Some studies featured advanced production methods for electrolyzers and activated water with improved properties [19–22]. An efficient electrocatalyst possesses a number of certain properties. For instance, it has low hydrogen overpotential at industrial current densities with a constant potential value that remains stable over time. It has a good chemical and electrochemical resistance and is capable of a long-time service. In addition, an effective electrocatalyst does not emit harmful products during electrolysis [23–27].

The present analysis of modern scientific research in the field of animal fat extraction indicated the relevance of new technical solutions in this sphere.

The research objective was to create an ecologically safe fat rendering production line based on the principle of electrochemical activation. The study focused on intensifying the process of fat extraction and increasing the fat-water separation.

ОБЪЕКТЫ И МЕТОДЫ ИССЛЕДОВАНИЯ

Ostrich fat became a research object as a novel raw material resource. In addition, ostrich fat is rich in unsaturated fatty acids, easy to separate into fractions, and similar with traditional poultry fats in terms of physical and chemical properties. Ostrich fat has low melting and chilling points. For subcutaneous fat, the melting point is 33–35°С, and the chilling point is 25–27°С. For internal fat, the melting and chilling points are 30–32 and 21–23°С, respectively. As for the iodine value, ostrich fat surpasses that of other farm animals and is close to that of chicken: J₂ of subcutaneous fat is 56.6 ± 0.5 g/100 g; J₂ of internal fat is 61.7 ± 0.5 g/100 g. Table 1 illustrates a comparative analysis of the fatty acid composition of subcutaneous and internal ostrich fat.

The process of in-lab ostrich fat rendering started with preliminary grinding of the refined raw fat to a particle size of 0.5 mm. The wet rendering in water (control sample) and in an electrochemically activated solution, namely catholyte (experimental sample), was performed in an HM-220VM vacuum reactor (Japan – Taiwan). The equipment provided uniform heating and temperature control, as well as vacuum and fat mixing. The water was used at a ratio of 8:1 to the mass of raw materials and met the requirements of State Standard R 51232-98. Extraction was carried out at 55, 75, and 100°C for 60 min. The stepwise thermal processing time was 20 min per stage.

Filtration was used to separate the raw materials into solid and liquid fractions. Vacuum under the filtering surface intensified the process. The fat was separated from the water-protein phase by centrifugation at 6.000 rpm/min for 5 min. The resulting fat with different melting points was dehydrated in a PE-8910 rotary evaporator (Russia) at a residual pressure of 2666.44 Pa. The remaining solid phase underwent subsequent melting at higher temperatures. The new technical approach appeared to have a number of advantages over the existing ones (Patent No. 2382072 RU, February 20, 2010). It proved to be less water- and time-consuming, produced less waste, and rendered fat with a controlled melting point.

The electrochemically activated solution was obtained in an electrolyzer by electrolyzing a 10% aqueous NaCl solution at the following modes: DC power = 0.5–0.6A, voltage = 40–42V, pH = 9–11, redox potential = –(400–700) mV, mass fraction of sodium chloride = 4 g/100 cm³.

Table 2 describes the methods and modes of the fat rendering process.

The conditions and parameters of fat rendering were established both experimentally and by calculation using the Box-Wilson orthogonal central composite design, where each factor was fixed at five levels (–α, –1, 0, 1, +α). The number of design points equaled:

where N₁ – number of points in complete factorial experiment 2n; 2n – number of paired points on the coordinate axes; N₀ – number of experiments in the central design.

The matrix rendered the following equation:

where b₀ – absolute term; x₁, x₂… x_n – factors that determine the level of the performance parameter; b₁, b₂… b_n – regression coefficients for factor indicators that describe the effect of each factor on the performance parameter in absolute terms.

The melting time X₁(Z₁), the pH of the electroactivated medium X₂(Z₂), and the processing temperature of the raw fat X₃(Z₃) were selected as primary factors. The response function was the output of rendered ostrich fat (Y₃). The significance of the coefficients of the mathematical model was established based on the Student’s criteria. The tabulated value of the Student’s test t_p(f) for P = 0.05 and the number of degrees of freedom f = 3 was 3.18. The adequacy of the regression equation was controlled by calculating the error square and the error mean square. The Fisher criterion was determined using the Table as F_1–p = (f₁, f₂). The equation was adequate if the resulting F-ratio was less than the tabular F < F_1–p. (f₁, f₂) for the selected significance level (P = 0.05) and the numbers of degrees of freedom.

Gas chromatography helped define the fatty acid composition of ostrich fat. The methyl ester extraction and chromatography were based on State Standard 32150-2013 under the following conditions: gas chromatograph – Kristall 5000.2, chromatographic column – Omegawax-320 (30 m×0.32 mm×0.50 μm), carrier gas – nitrogen, evaporator gas pressure – 85 kPa, split ratio – 1:40, and evaporator temperature – 250°C. The temperature of the column thermostat was programmed as follows: it was maintained at 120°C for 9 min, and then it was brought up to 210°C at 3°C/min, where it stayed the same for 30–40 min.

The data processing was conducted as follows. The mass fraction of fatty acids was calculated as the ratio of the chromatographic peak area of a particular fatty acid to the sum of all peak areas in the chromatogram. The experiment did not take into account the corrections for the detector sensitivity to the particular fatty acid. It also left out the difference in the molecular weight of the methyl ester and that of the fatty acid itself. The reason was that these corrections are significant only for short-chain fatty acids with the number of carbon atoms between C4 and C10.

The mass fraction of water and volatile substances was determined according to the methods described in State Standard R 50456-92 (ISO 662-80). The iodine number was calculated according to State Standard R ISO 3961-2010. The melting and chilling points were based on State Standard 32189-2013. The control of acid and peroxide numbers was based on State Standard R ISO 27107-2010. The peroxide number was determined using the potentiometric method according to the end point. The density was measured according to State Standard 18329-2014 (ISO 1675:1985). The anisidine number was measured by the optical density of the test solution after its reaction with an acetic acid solution of para-anisidine according to State Standard 31756-2012 (ISO 6885:2006). A rotary digital viscometer was used to determine fat viscosity as specified in State Standard 1929-87. The experiments were done in triplicates and processed statistically.

РЕЗУЛЬТАТЫ И ИХ ОБСУЖДЕНИЕ

The safety and intended use of the product depend on the physicochemical indicators of the rendered fats that are regulated by technical documentation and standards.

The physicochemical indicators of the quality of ostrich fat clearly showed the advantage of the proposed method (Table 3). The fat rendered using the experimental technology had low acid, peroxide, and anisidine values. Processing raw materials in an electrochemically activated solution inactivated the enzyme system (lipase, phospholipase) even at low temperatures (39 and 55°C). The catholyte mechanism proved to be effective, as it lowered and, in some cases, prevented the hydrolytic processes of triacylglycerols. Wet rendering in tap water demonstrated the opposite tendency. The acid numbers revealed a significant amount of free fatty acids and peroxide compounds in the rendered fat. Unlike the experimental technology, water rendering resulted in a characteristic smell.

Fat viscosity is known to affect the rate of heat transfer, separation, and settling. Therefore, this indicator was considered significant for the new technology and equipment. Table 3 shows that internal fat had a lower viscosity than subcutaneous fat, which was due to its unsaturation degree and an increase in the number of double bonds. According to A.V. Gorbatov, liquid products have no ultimate shear stress, and the flow occurs at any minimum shear stress [28]. After cooling, the ostrich fat did not lose all its viscosity. In contrast, pork fat develops a solid crystallization framework at temperatures below 27°C and loses its viscosity.

Table 4 demonstrates that the effectiveness of the experimental technology was fully confirmed by the yield.

Electrochemical activation method produced a significantly higher yield than wet rendering in water (Table 4). Second-order orthogonal designs helped establish the effect of technological parameters on fat yield. Table 5 shows the levels of fat extraction factors, while Table 6 demonstrates the matrix of the experiment. Mathematical model 2^k took into account both the number of star points on the coordinate axes of the factor space and points in the center design. Star shoulder α indicated the distance from the center design to the star point.

The dependence of the output on the set parameters is given as a regression equation (Eqs. (3) and (4)).

After checking the coefficients of the equation, only nine proved significant. Therefore, the final equation was as follows:

The Fisher’s criterion calculated F = (f₁, f₂) = (6; 3) = 7.39 proved the obtained Eq. (4) to be adequate for experiment (S_reprod.= 0.207, S_resid. = 1.53), as condition F < F_1–p. (f₁, f₂); 7.39 < 8.94 holds at P = 0.05.

The obtained equation made it possible to establish an equidirectional positive effect of temperature, rendering time, and catholyte pH on fat yield. In addition, the cross impact of the hydrogen index and the processing temperature was insignificant, while the joint action mechanism of “processing time – catholyte pH” and “processing time – melting temperature” proved to be insignificant. Nevertheless, a significant increase in the abovementioned factors would have a negative effect. Thus, simulating the technological conditions of fat extraction in an electrochemically activated liquid could control the properties of the raw material and its productivity. The identified dependencies were taken into account in order to improve the technological conditions of fat extraction (Fig. 2).

The experimental data made it possible to build a technological line for ostrich fat rendering [29]. It owed its innovative character to a higher fat yield and a more efficient fat extraction process based on electrochemical activation.

Figures 1 and 2 illustrate the sequence of technological operations and hardware design of the fat production line. Table 7 contains the necessary equipment specification.

The ostrich fat production line with the specified properties operated as follows [29]. Raw fat was loaded into the receiving funnel of the AVZh-400 centrifugal machine for primary grinding to 5–6 mm. Simultaneously, hot water (70°C) was added info the grinder in the amount of 20% of the raw material mass (Table 7, 1). After that, the raw material was fed into the IN-20 worm inactivator (2), where it was treated with both indirect and direct heating steam under a pressure of ≤ 0.3 MPa. The electrochemically activated solution (catholyte) flowed in a ratio of 4:1 to the raw material mass through the pipeline that connected the inner working area of the inactivator with the catholyte tank (19). Unlike the traditional wet rendering, the fat was watered in an electrolyte (catholyte). A prefabricated 10% aqueous solution of sodium chloride (25) was poured into the electrolyzer (21), which was connected to the catholyte tank (19). To increase the rendering, the medium for electrochemical activation was applied precisely during the relaxation period. At that stage, its activity and technological properties were at their maximum. The screw rotation frequency was 60 rpm. To sum it up, the raw material was ground and heated with direct heating steam to a predetermined temperature, depending on the melting point of fat rendering.

The obtained combination of mechanical and physicochemical factors made it possible to intensify fat extraction without losing in the yield and quality of fat. The essence of this phenomenon lied in the unipolar processing of sodium chloride solution in the cathode chamber of the electrolyzer. As a result, the solution acquired a metastable state with unique relaxing physicochemical properties. The structural components of the interionic energy interaction underwent a significant change. The active concentration of ions changed because the activity coefficients had changed, not the concentration. The electrolyte is known to reach stable values over a period of time that ranges from minutes to hundreds of hours [27].

The technology made it possible to increase the fat yield without high temperatures or chemicals. Cell membranes were destroyed despite the low processing temperatures.

After that, the processed raw fat was fed to an extra disintegrator (3) to obtain a finely ground pulp suspension (particle size = 0.6 mm). The centrifugal force of the ABZh-245 machine pumped the ground fat to the height of 3 m.

At pressure of 0.3 atm, the suspension was fed by gravity to an extra device (Table 7, 5) in order to separate the pulp into a fat-water fraction and greaves. The fat-water fraction entered the first FCA fat separator (10) via the first collection vessel (6) through the pipeline and pump (7). There it was continuously separated into water-protein glue and fat-water fraction.

The first separation concentrated and purified the fat-water suspension. It entered the rotating drum at the rotation speed of 5,000 min^–1 and dispersed into thin layers between the conical trays. Under the centrifugal force, protein and other solid particles moved along the bottom surface of the trays and were thrown into the sludge space. Partial sludge discharge occurred at specified time intervals in the operating mode. The processing time depended on the concentration of solid particles in the raw material. The average processing temperature in the separators directly depended on the preset melting temperature in the inactivator; the processing time was 5 sec. The water-protein fraction entered the tank (9) via the first outlet of the separator with the help of a pressure disc along the pipeline under a pressure of 0.3 MPa. The mass fraction of the waste water from the first separator was about 70%.

Through the second outlet, the remaining fat-water fraction was fed into the second tank (11). After that, it went through the second separator (FSTsP-1) with a onemillimeter inter-plate gap for further separation (13), resulting in the target product with a preset melting point (14, 15, 16).

To reduce fat loss, the remaining fat-water phase returned into the machine for primary grinding of raw fat after the second separation (1). From there, it went into the IN-20 worm inactivator for processing (2), together with a new batch of raw fat.

The designed line was sustainable and provided not only complete fat rendering, but also its fractionation depending on the melting temperature. A stepwise approach to the processing of raw materials with a gradual increase in temperature made it possible to control the process and process the various types of animal and poultry fats. The new technical solution returned the fat-water phase and suspension through a helical conveyor (4), which was connected to the centrifuge (5), back to the inactivator (Fig. 2).

The new production line provided a closed technological cycle that minimized the loss of valuable fat product low temperature rendering.

Table 8 demonstrates the technical result of the production line.

ВЫВОДЫ

Electrochemically activated solution introduced in the process of fat rendering made it possible to adjust its properties depending on the intended use of the final product. The developed method of fat rendering and the new sustainable fat production line produced high quality fat and increased the yield of the product compared to the traditional wet rendering in water. The method was based on an inexpensive and safe sodium chloride reagent, which reduced costs and losses in the technological cycle. As a result, the new technology increased the productivity of fat extraction and the degree of fat-water fraction separation. The developed fat production line based on electrochemical activation can be recommended for farm animal fat rendering depending on the intended use.

КОНФЛИКТ ИНТЕРЕСОВ

The authors declare that there is no conflict of interest related to the publication of this article.

БЛАГОДАРНОСТИ

The authors would like to express their sincere gratitude to the management of Kuban State Technological University for promoting this research.

ФИНАНСИРОВАНИЕ

The research was part of agreement No. 14.607.21.0161 with the Ministry of Science and Higher Education of the Russian Federation (Minobrnauka), Applied Research and Exploratory Development Program No. RFMEFI60716X16.

СПИСОК ЛИТЕРАТУРЫ

1. Method of preparing ostrich rendered fat. Russia patent RU 2382072C1. 2010.
2. Liato V, Labrie S, Aïder M. Electro-activation of potassium acetate, potassium citrate and calcium lactate: impact on solution acidity, Redox potential, vibrational properties of Raman spectra and antibacterial activity on E. coli O157:H7 at ambient temperature. Springer Plus. 2016;5(1). https://doi.org/10.1186/s40064-016-3453-1.
3. Aïder M, Gnatko E, Benali M, Plutakhin G, Kastyuchik A. Electro-activated aqueous solutions: Theory and application in the food industry and biotechnology. Innovative Food Science and Emerging Technologies. 2012;15:38–49. https://doi.org/10.1016/j.ifset.2012.02.002.
4. Jiménez-Pichardo R, Regalado C, Castaño-Tostado E, Meas-Vong Y, Santos-Cruz J, García-Almendárez BE. Evaluation of electrolyzed water as cleaning and disinfection agent on stainless steel as a model surface in the dairy industry. Food Control. 2016;60:320–328. https://doi.org/10.1016/j.foodcont.2015.08.011.
5. Thorn RMS, Lee SWH, Robinson GM, Greenman J, Reynolds DM. Electrochemically activated solutions: evidence for antimicrobial efficacy and applications in healthcare environments. European Journal of Clinical Microbiology and Infectious Diseases. 2012;31(5):641–653. https://doi.org/10.1007/s10096-011-1369-9.
6. Bakhir VM. Ehlektrokhimicheskaya aktivatsiya [Electrochemical activation]. Moscow: RSRIME; 1992. 657 p. (In Russ.).
7. Krasavtsev BE, Tsaturyan AS, Simkin VB, Alexandrov BL, Alexandrova EA. Industrial device for the electrochemical activation of water. Polythematic online scientific journal of Kuban State Agrarian University. 2015;(110):786–800. (In Russ.).
8. Janoschek R, Weidemann EG, Pfeiffer H, Zundel G. Extremely high polarizability of hydrogen bonds. Journal of the American Chemical Society. 1972;94(7):2387–2396. https://doi.org/10.1021/ja00762a032.
9. Plutakhin GA, Mohammed A, Koshchaev AG, Gnatko EN. Theoretical fundamentals of electrochemical treatment of water solutions. Polythematic online scientific journal of Kuban State Agrarian University. 2013;(92):72–83. (In Russ.).
10. Gerzhova A, Mondor M, Benali M, Aider M. A comparative study between the electro-activation technique and conventional extraction method on the extractability, composition and physicochemical properties of canola protein concentrates and isolates. Food Bioscience. 2015;11:56–71. https://doi.org/10.1016/j.fbio.2015.04.005.
11. Osadchenko IM, Filatov AS, Chamurliev NG. Process for the preparation of minced meat by the use of electroactivated solutions. Proceedings of Lower Volga Agro-University Complex: Science and Higher Education. 2017;45(1):109–114. (In Russ.).
12. Osadchenko IM, Gorlov IF, Sivkov AI, Nikolaev DV, Mosolova NI. Innovative approach to electrical processing of dairy drinks. Storage and Processing of Farm Products. 2018;(1):20–22. (In Russ.).
13. Semenenko SYa, Lytov MN, Chushkin AN, Chushkina EI. Agroecological assessment of the effectiveness of irrigation of tomato using electrolyzed water. South of Russia: ecology, development. 2019;14(3):61–70. (In Russ.). https://doi.org/10.18470/1992-1098-2019-3-61-70.
14. Semenenko SYa, Lytov MN, Chushkin AN, Chushkina EI. Determination of the parameters of technological process of application of electrolized water under drip irrigation of vegetable crops. Proceedings of Lower Volga Agro-University Complex: Science and Higher Education. 2019;54(2):322–330. (In Russ.). https://doi.org/10.32786/2071-9485-2019-02-38.
15. Krasavtsev BE, Tsaturjan AS, Simkin VB, Aleksandrov BL, Aleksandrova EhA. Vegetable oil refining method (versions). Russia patent RU 2525269C2. 2014.
16. Pasko OA. Metabolism in Amaranthus L. seeds after their treatmentwith electrochemically activated water. Agricultural Biology. 2013;38(3):84–91. (In Russ.).
17. Kitanovski VD, Vlahova-Vangelova DB, Dragoev SG, Nikolov HN, Balev DK. Effect of electrochemically activated anolyte on the shelf-life of cold stored rainbow trout. Food Science and Applied Biotechnology. 2018;1(1):1–10. https://doi.org/10.30721/fsab2018.v1.i1.2.
18. Pogorelov AG, Suvorov OA, Kuznetsov AL, Panait AI, Pogorelova MA, Ipatova LG. Disintegration of bacterial film by electrochemically activated water solution. Bulletin of Experimental Biology and Medicine. 2018;165(4):493–496. https://doi.org/10.1007/s10517-018-4202-y.
19. Liato V, Labrie S, Benali M, Aïder M. Ion exchange membrane-assisted electro-activation of aqueous solutions: Effect of the operating parameters on solutions properties and system electric resistance. Process Safety and Environmental Protection. 2015;93:124–138. https://doi.org/10.1016/j.psep.2014.04.005.
20. Pshenko EB, Shestakov IYa, Shestakov VI. Features of electroactivated water production at a coaxial electrode location. Siberian Journal of Science and Technology. 2019;20(1):119–125. https://doi.org/10.31772/2587-6066-2019-20-1-119-125.
21. Semenenko SYa, Chushkin AN, Lytov MN. The law of relaxation of water with electrochemically modified redox potential. Proceedings of Lower Volga Agro-University Complex: Science and Higher Education. 2017;46(2):263–271. (In Russ.).
22. Arakcheev EN, Brunman VE, Brunman MV, Konyashin AV, Dyachenko VA, Petkova AP. Complex technology for water and wastewater disinfection and its industrial realization in prototype unit. Hygiene and Sanitation. 2017;96(2):137–143. (In Russ.). https://doi.org/10.18821/0016-9900-2017-96-2-137-143.
23. Gurbanova UM, Huseynova RG, Tahirli HM, Dadashova SD, Aliyev ASh, Tagiyev DB. Electrodeposition of Ni-Mo alloys from ammonium electrolytes. Azerbaijan Chemical Journal. 2015;(3):25–31. https://doi.org/10.32737/0005-2531-2019-3-25-31.
24. Jaccaud M, Leroux F, Millet JC. New chlor-alkali activated cathodes. Materials Chemistry and Physics. 1989;22(1–2):105–119. https://doi.org/10.1016/0254-0584(89)90033-3.
25. Wu L, He Y, Lei T, Nan B, Xu N, Zou J, et al. Characterization of porous Ni3Al electrode for hydrogen evolution in strong alkali solution. Materials Chemistry and Physics. 2013;141(1):553–561. https://doi.org/10.1016/j.matchemphys.2013.05.061.
26. Bodner M, Hofer A, Hacker V. H2 generation from alkaline electrolyzer. Wiley Interdisciplinary Reviews: Energy and Environment. 2015;4(4):365–381. https://doi.org/10.1002/wene.150.
27. Bakhir VM, Zadorozhniy YuG, Leonov BI, Panicheva SA, Prilutskiy VI, Sukhova OI. Ehlektrokhimicheskaya aktivatsiya: istoriya, sostoyanie, perspektivy [Electrochemical activation: history, state, and prospects]. Moscow: RSRIME; 1999. 256 p. (In Russ.).
28. Gorbatov AV, Machikhin SA, Maslov AM, Tabachnikov VP, Machikhin YuA, Kosoy VD. Strukturno-mekhanicheskie kharakteristiki pishchevykh produktov [Structural and mechanical characteristics of food products]. Moscow: Legkaya i pishchevaya promyshlennost’; 1982. 296 p. (In Russ.).
29. Gorbacheva MV, Tarasov VE, Tarasov SV, Sapozhnikova AI, Gordienko IM. Fat production line. Russia patent RU 2679711C1. 2019.