Evaluation Of Filter Cake Oil As Antifoam In Yeast Production. New Use For This By-Product Of The Sugarcane Derivatives Industry

Purpuse: Fodder yeast is obtained in an aerobic fermentation process where foaming is a major problem to be solved. In this article, the antifoam property of crude and puried lter cake oil is evaluated in order to use this residual as an alternative to replace the import of commercial antifoam agents and to reduce the production costs of fodder yeast Method: Knock down test and the comparisons with two commercial antifoam agents were done. Blackstrap molasses medium at 20 and 40 g/L of total reducing sugar was used. All products were studied in their pure form and commercial ones also in dilutions 1:2 and 1:5 v/v. Hansen's solubility parameters (HSPs) to analyze the anity of each defoamer for yeast were determined. Results: It was obtained the crude and puried lter cake oil showed similar behavior to commercial defoamers with an immediate antifoam effect, removing between 40 and 60% of the initial foam at both sugar concentrations in the rst 5 minutes. The regression model showed puried lter cake oil has the greatest knockdown effect (Ca = 55.85 and 74.32) and with greater foam suppression stability the commercial defoamer Quimifoam Máster (C b = -0.69 and -1.38). Ra values obtained in HSPs test, indicated the anity of defoamers to the medium. Conclusion: Puried lter cake oil is an effective product for its use as an antifoam with the best knock down effect for both concentrations of sugars in the medium. The determination of HSPs corroborates the effectiveness of this product to suppress foam. replacement of commercial defoamers, products that are imported and with high prices in the market.


Introduction
In fermentation processes, foam formation is inevitable and occurs as a consequence of the introduction of large masses of air into the process and the presence of colloidal substances or long-chain organic compounds (soluble proteins, alcohols, etc.) [1]. There are different ways to control and/or remove foam, such as the use of mechanical devices, removal of foaming agents from process units that prevent their formation, spreading water over it or breaking down by applying pressured ows over foam [2]. However, the most widely used method is applying antifoam agents of different natures.
Candida utilis fodder yeast, commonly named Torula yeast, is the most widely used microorganism in the production of unicellular protein as an established product with good acceptance in feed formulation for animal nutrition [3,4]. It is obtained from heavy aerobic fermentation of a sugar-rich medium where foam formation occurs, which is a major problem that must be addressed.
Generally, for foam control, a wide range of antifoam commercial products with dissimilar properties and compositions have been used, but most contain oils (silicone, mineral, uorocarbons), solid particles (silica, calcium, plastic beds), or mixtures of the latter with amphiphilic particles and molecules (alcohols, fatty acids, fatty esters) [5]. The defoamers used in yeast production are mainly based on fatty acids, fatty acid esters and aliphatic alcohols, among others [6].
Use of some types of vegetable oils such as jojoba, sun ower, coconut oil, soybeans and rapeseed, have been reported as defoamers in different productions to replace commercial ones with good results [6-9].
Filter cake oil is a by-product of sugarcane wax re ning process. Its use as an antifoam agent in fodder yeast production would be of interest to sugar and derivatives industry because it offers two options: the possibility of a new use of this by-product and also it would have an impact on reduction of Torula yeast cost production by substituting commercial defoamers, products that are imported and of high prices.
In consulted bibliography by the authors, no recent reports of the use of lter cake oil as antifoam were found. In the reports of Montano et al [11,16], antifoam effect of lter cake oil as a fatty vehicle in mixtures with different esters (monopalmitate, monomyristate, monotannate, monooleate and monostearate) was tested compared with sun ower oil. In these studies, best results were obtained when lter cake oil was used, with a similar procedure of commercial defoamers. Taking these results as a starting point, the objective of this work is to evaluate antifoam property of pure lter cake oil (without mixing with other compounds) in the production of Candida utilis yeast.

Materials
Two types of lter cake oil, one with high content of wax or crude (CACHA) (average dry base composition of 79% oil, 16 % wax and 5% impurities) obtained from the sugarcane wax re ning factory annexed to Majibacoa sugar mill (Las Tunas Province, Cuba) and other a puri ed lter cake oil (CACHAP) were tested (processed) at The Cuban Research Institute of Sugar Cane Derivatives (ICIDCA). Also two commercial defoamers: Silicone antifoam B30 (FV) (non-ionic emulsion system) (Carini Chem SRL, Italy) and Quimifoam Máster (C10) (anionic emulsion system) (Zucker S.A, México), both specially used for fermentation processes in alcohol distilleries and yeast plants, were evaluated.
Commercial defoamers were studied in their pure form and in dilutions of 1: 2 and 1: 5, as this is the way they are used in industry. In all cases, the behavior obtained from them was compared with lter cake oils. Methods

Defoaming Potential
To study the defoaming potential was made a Knock down physical test, a 1000 mL cylinder with an air diffuser was used. Air ow was provided by a compressor (Oks Otto Klein Gmbh, Germany) and controlled by a ow meter. A stopwatch (class 0.2 seconds) was used to measure time interval in the experiments and a heating plate was employed to increase the temperature of lter cake oil when applyed to the medium.
Defoamers were studied in the laboratory to know their potential to suppress foam on a foamy medium made up of a molasses solution at 20 and 40 g/L of total reducing sugar (TRS). These sugar concentration was chosen because it is a substrate being used in many fermentation processes, they were likely to provide a foaming system broadly representative of industrial production of Torula yeast. As indicator parameter of antifoam character was considered the effect it produces over dynamic foam.
Experiments were carried out at room temperature (25 ° C).
Procedure was done according to the method reported by Montano et al [11] and Kato et al [17]. A 500 mL of molasses solution was poured into the cylinder. Air ow was adjusted at 4 vvm to promote foam formation of the medium volume.
Mollasses solutions at the concentrations used had different uid properties and material content (sugars and impurities), so the surface tension was not the same and this caused the volume of the initial foam to be different.
When foam volume reached 600 mL for the solution of 20 g/L TRS and 650 mL for 40 g/L, a drop of the antifoam to be evaluated (volume equivalent to 0.05 g) was dropped with the aid of a micropipette. In the case of crude lter cake oil was dropped 0.5 ml. Volume of foam generated was measured every minute for a period of 15 minutes; this way a "variable of response" was obtained, as "the relative foam removal as a function of time".
Relative foam removal (RFR) was calculated measuring foam volume at minute 0 as the initial minute (ini) and at the other time intervals (t) as shown in Eq. 1.

Hansen Solubility Parameters Determination (HSPs)
HSPs determination is a tool to express a nity degree between a solute and any solvent, based on the Hansen theory, which remarks "likes attracts likes". Hansen parameters of different solutes and solvents (δ D , δ H , δ P ) may be plotted in a three-dimensional graph plot x,y,z. Closer the solute-solvent pair is in a three-dimensional space, more soluble they are. A nity of defoamers with surfactant medium (yeast) was determinated quantitatively calculating the distance (R a ) of Hansen factors obtained for each one (Eq. 2).
where: Hansen's parameters were determined by Hansen solubility sphere method, from the a nity test with organic solvents of known HSPs. Hansen parameters of foamers studied were obtained experimentally according to a methodology described in detail by Kato et al [17] and Hernández et al [18]. Through the Yamamoto Molecule Breaking (Y-MB) methods of contribution of functional groups and a knowledge of their expression SMILE (Simpli ed Molecular Input Line Entry Speci cation), developed by Abbott and Yamamoto H [19] and available in HSPiP software version 5.2.0. Hansen parameters of Candida utilis yeast amino acids were determined because they were the soluble components of the medium causing the foam.

Statistical Analysis
In knock down tests each experimental condition was performed by triplicate and data were statistically analyzed in Statgraphics Centurion XVII.2 program, at time 5, 10 and 15 minutes.
One-way analysis of variance (Simple ANOVA) and the "least signi cant difference" (LSD) test was used to compare statistical differences between means values. Differences were considered signi cant at p value < 0.05. Also, to determine activity of each defoamer a regression model adjustment of relative foam removal as a function of time for the defoamers in pure form in each medium concentration studied was performed.

Results And Discussion
Defoaming potential Table 1 presents foam height values for all defoamers tested in this study during the 15 minutes time of the test and Fig. 1 shows the behavior of foam removal percentage (RFR) versus time, obtained in knock down test. Four tested defoamers had an immediate effect over foam when they were added, decreasing its volume rapidly and keeping foam level controlled below the initial value throughout the test; this indicates the compounds are able in changing interfacial properties of a liquid, resulting in an extensive foam suppression. Table 1 Foam height (mean ± standard deviation) vs time for all the experimental conditions studied In the rst 5 minutes, it was possible to remove between 40 and 60% of the initial foam at both concentrations (20 and 40 g/L of TSR), although higher values were sometimes reached for the latter concentration. These results have coincide with the proposed by McClure et al [9], who suggested compounds present in molasses can "displace" the defoamer of the interface, implying a decrease in its effect after a given period of time and also indicates a need to add another defoamer dose.
For a concentration of 20 g/L TSR ( Fig. 2A), it was observed that from the two commercial defoamers (FV and C10) C10 has highest stability of foam removal. Removal capacity of non-commercial defoamers (CACHA and CACHAP) compared to commercial ones -at 5 minutes of its addition to the system -was similar, because no signi cant differences were detected with a P-value (0.3813) greater than 0.05, with a 95% con dence level. After 10 minutes, commercial defoamers had a more stable effect on foam control and CACHA shows better results than CACHAP.
After 10 and 15 minutes of analysis at both total sugar concentrations in the medium (20 and 40 g/L), differences were detected between commercial and non-commercial defoamers with values lower than 0.05 (0.0173 and 0.0014 respectively), with 95% con dence. Commercial defoamers FV and C10 had a similar behavior throughout the analysis without showing signi cant differences between them in "foam removal percentage"(RFR). Figure 2A shows in comparison between commercial defoamers and noncommercial defoamers, after 10 minutes of starting the test, foam removal percentage is completely differently signi cant, since FV and C10 continued removing around 20% more than the non-commercial ones. For CACHA and CACHAP defoamers, signi cant differences were only detected between them at 15 minutes of addition when they reached values of 22.5 and 10% of foam removal respectively. These results show, at 20 g/L TSR concentration for 15 minutes, CACHA can be considered a better option than CACHAP. During the test, the removal values of all the defoamers evaluated were around 50%, which is considered a value below the expected result.
At a concentration of 40 g/L TSR (Fig. 2B), puri ed lter cake oil (CACHAP) exhibits a foam removal behavior similar to commercial defoamers and superior to crude oil (CACHA). The behavior of the commercial defoamers (FV and C10) at this concentration was similar to that obtained for a concentration of 20 g/L TSR where C10 had the greatest in uence on the stability of the foam.
In general, among non-commercial defoamers, the most unfavorable behavior corresponds to CACHA with the lowest foam removal at all times evaluated with signi cant differences, with P values (0.0131, 0.0079, 0.0059) less than 0.05 for 5, 10 and 15 minutes respectively, with 95% con dence. CACHAP presented similar values to commercial defoamers after 5 minutes of their addition, detecting signi cant differences only after 15 minutes, time in which their behavior is similar to that obtained for CACHA.
The behavior of the commercial defoamers diluted 1: 2 and 1: 5 was compared with the non-commercial ones in their pure form under the same working conditions. For the concentration of 20 g/L TRS in the medium at a 1: 2 dilution (Fig. 3A), signi cant differences were found when comparing commercials with CACHA and CACHAP at 10 and 15 minutes, with a con dence level of 95% and P values less than 0.05 (0.013 and 0.0096 respectively). Greater instability was observed for commercial defoamers when working with a 1: 5 dilution as shown in Fig. 3B, mainly for C10, which, after 15 minutes, could no longer remove the foam from the system. Figure 4 represents the same behavior, but for concentrations of 40 g/L TSR. Signi cant differences were detected between commercial defoamers, especially for the more dilute condition (Fig. 4B). CACHAP showed a favorable response, with better foam removal behavior and higher values than diluted commercial defoamers and CACHA.

Regression model analysis
Taking into consideration what was proposed by Montano et al [11], to determine the antifoam with the best knock down effect and the highest stability of foam removal, a linear regression model was adjusted. Adjustment was made for pure defoamers and at both concentrations of TRS in the medium.  When comparing values obtained, it was observed better results were achieved by commercial defoamers compared to non-commercial ones (CACHA and CACHAP). However, results of Simple ANOVA analysis did not detect signi cant differences between defoamers evaluated with a P value greater than 0.05 (0.527) with 95% con dence, indicating any of them can be employed.
In addition, analyzing C b value of "foam suppression stability" determination, results indicate C10 was more favorable because it shows highest slope and a p value less than 0.05 (0.0106) was obtained, indicating presence of signi cant differences of this parameter, with a 95% con dence. When multiple range test was applied with Fisher's "least signi cant difference (LSD) procedure", differences were detected between this defoamer and the non-commercial ones (CACHA and CACHAP), which had no differences between them; so suppressive effect of the latter is similar.
In case of 40 g/L TSR if the highest value of the slope calculated, when regression equation is considered, the one with greatest stability was crude lter cake oil (CACHA), however considering the low correlation coe cient R 2 , it was discarded, since result only explained 62% of its behavior as antifoam. Therefore, the one with highest stability (C b ) was C10. Regarding knock-down effect (C a ), best result was achieved with CACHAP. At this concentration, results of ANOVA analysis did not detect signi cant differences with C b but signi cant differences for C a , with P values of (0.0435 and 0.0627 respectively) and a 95% con dence level; differences of C b were established between CACHA and the rest of evaluated defoamers.
Based on the results obtained, the puri ed lter cake oil can be used successfully as an antifoam in this medium. Its application in raw form is not recommended since its effect in suppressing the foam was inferior. In addition, due to its high wax content, it can cause incrustations in the equipment and its application to the environment is di cult, since because it is very dense (pasty) it must be added at a temperature close to 60 ° C.
Filter cake oil is a by-product of the sugarcane industry, which is available and its use as a defoamer can be an interesting and economical option for the production of Candida utilis; since it will allow the substitution of commercial defoamers that have a high cost in the market.

Hansen Solubility Parameters (HSPs)
A medium where these defoamer should be used is a mixtures of different compounds and where yeast has a concentration of 10 g/l. For Hansen solubility parameters calculation of the medium, the amino acids that characterize Candida utilis yeast were taken as a reference and coincide with those present in either molasses or vinasse, which are the carbon sources normally used as a culture medium in this process; amino acid composition for Candida utilis yeast reported by Yañez [20] y Otero and Almazán [21] was assumed. From mass percent and densities of each amino acid, the volumetric percent were obtained by calculating mixture density by trial and error. HSPs of the mixture were calculated from the sum of Hansen´s parameters of each amino acid, affected by the volumetric contribution of each one to the medium. Table 2 shows the amino acids present in the medium and their respective HSPs.  Important is to clarify Hansen parameters were not obtained for crude lter cake oil because it has a high wax content and a variable concentration, so it is expected these factors (δd, δp, δh) will be less than those of puri ed oil, therefore Ra will be higher and its a nity lower to the medium. Figure 5 shows relationship between foam volume and Ra values of 5, 10 and 15 min test time. In this case, it coincides with one with lowest value of Ra. C10 was the defoamer that maintains lowest foam volume, which also complements the criterion that the lower the value of Ra, the greater the a nity with the medium. This aspect corroborates Kato et al [21] proposition where they state "use of the Ra calculation could be another criterion for selection of best antifoam". Another aspect corroborated in this test is existing similarity in foam suppressing behavior between defoamers studied and use of puri ed lter cake oil, meaning it may be an option to consider for its use in SCP technology.

Conclusions
The puri ed lter cake oil was determined to be an effective product for use as antifoam. From the analysis carried out, it was obtained that this oil presented the highest knock down effect and C10