Drying of butiá pulp by the foam-layer method and characterization of the obtained powder

Abstract The objective of this work was to develop a drying process using the foam-layer method to obtain the powder from butiá (Butia spp.) pulp, as well as to characterize this powder according to its yield and physicochemical and technological characteristics. The foam was optimized for density and stability by varying whipping times and albumin and xanthan gum concentrations. Foam drying was optimized for vitamin C retention and yield using different foam thicknesses and drying temperatures. The optimized foam showed a density of 0.25 g cm-3 and a high stability, being suitable for subsequent drying. The lowest foam thickness (0.50 cm) and the highest drying temperature (80°C) resulted in the highest retention of vitamin C, whereas the increase in drying temperature improved yield. The butiá pulp powder obtained under the optimized condition presented an acid pH of 3.25, a low humidity of 7.97%, a water activity of 0.206, a water retention capacity of 4.90 g H2O per gram of powder, a solubility of 74.40%, a soluble solids content of 61°Brix, and a predominantly yellow color. The foam-layer drying method can be used to obtain butiá pulp powder.


Introduction
Butiá (Butia spp.) is the fruit of a palm tree of the genus Butia, of the Arecaceae family, which occurs naturally in South America and is on the list of endangered species in Brazil, where it shows potential for income generation (Hoffmann et al., 2014).The pulp of this fruit is rich in bioactive compounds with antioxidant activity, such as ß-carotene and vitamin E, and has a high concentration of vitamin C (Barbosa et al., 2021).However, the widespread consumption of the fresh fruit is hindered by its high acidity, fibrous texture, and low sugar content, requiring processing to enhance its quality for a higher consumer acceptance and, consequently, to improve its shelf life throughout the year, which is important since this is a highly perishable seasonal fruit (Hoffmann et al., 2017).
Different methods, such as pasteurization (Hoffmann et al., 2017) and hot-air drying and microwave radiation (Ferrão et al., 2017;Macedo et al., 2020), have been used in recent studies for the preservation of fruit properties.However, further scientific efforts are necessary to find a technology for enhancing fruit shelf-life that is cheaper and easier to use, especially since most of the processing in Brazil is currently carried out by small farmers (Büttow et al., 2009).
The foam-layer drying method is a simple and accessible process, with lower equipment costs than the other methods (de Cól et al., 2021).The quality of the final product is affected by the additives used for foam formation, whipping time, drying temperature, and foam-layer thickness, which shows the importance of determining the ideal conditions for the entire process (Reis et al., 2021).Up to date, there are no known scientific studies on the drying of butiá pulp using this method.
The objective of this work was to develop a drying process using the foam-layer method to obtain the powder from butiá pulp, as well as to characterize this powder according to its yield and physicochemical and technological characteristics.

Materials and Methods
The butiá fruits used in the study were harvested in January 2021 in Vale do Taquari, a region of the state of Rio Grande do Sul, Brazil (29º28'01"S, 51º57'41"W, at 21 m of altitude).Clusters of three palm trees were collected when most of the fruits showed a yelloworange color.When the fruits were considered mature, i.e., detached from the bunch after a light manual pressure, they were selected for sanity and integrity, sanitized, and stored at -18ºC.To obtain the pulp, the fruits were thawed under refrigeration and pulped using the DPT-75 pulp extractor, with a 1.0-mm sieve (Tomasi, Caxias do Sul, RS, Brazil); the pulp was kept at -18°C until use.
To produce the foam, the pulp was thawed under refrigeration and heated in a water bath, at 40°C, until reaching a temperature of 23°C.Albumin (Naturovos, Salvador do Sul, RS, Brazil) and xanthan gum (Foco Alternativo, Santa Cruz do Sul, RS, Brazil) were mixed into the pulp, and, then, the mixture was stirred using the 12-speed Oster Power 750 W household mixer (Newell Brands, Atlanta, GA, USA), at full speed, as described by Pandith & Srivastava (2018).
To evaluate the effect of different albumin (2.0 to 10% w/w) and xanthan gum (0 to 0.5% w/w) concentrations, as well as of whipping times (10 to 20 min), on foam density and stability, 19 assays were performed using a complete 2 3 factorial design (Table 1).Density was calculated by the ratio between foam weight and volume (Ratti & Kudra, 2006).Foam stability was determined by measuring the liquid drained from a 15 g foam sample -placed in a filter paper over a glass funnel -after 120 min at room temperature (Raharitsifa et al., 2006); the greater the amount of liquid drained, the less stable the foam.
Optimal foaming conditions were obtained using a full factorial design through the response surface methodology (RSM) (Pandith & Srivastava, 2018).The following equation was fitted to the experimental data of density and stability (means of triplicates): Y i = X 0 + X 1 G + X 2 A + X 3 T + X 11 G 2 + X 22 A 2 + X 33 T 2 + X 12 G×A + X 13 G×T + X 23 A×T, where Y i is the predicted response (Y 1 is density and Y 2 is stability); X 0 is the independent value; X 1 , X 2 , and X 3 are the linear coefficients; X 11 , X 22 , and X 33 are the squared coefficients; X 12 , X 13 , and X 23 are the interaction coefficients; G is xanthan gum; A is albumin; and T is whipping time.
For the drying process, the foams were placed in a single layer in 16x12 cm stainless steel trays and then dried in the SL-101 and SL 100 static-air ovens (Solab: Equipamentos para Laboratórios, Piracicaba, SP, Brazil).The mass of the samples was recorded every 30 min until no weight changes were detected in three subsequent weighing.
To evaluate the effect of different temperatures (60-80°C) and foam-layer thicknesses (0.50-1.50 cm) on vitamin C retention and yield, 11 assays were carried out in a full 2 2 factorial design (Table 2).Each drying assay was performed in a duplicate of two independent experiments, consisting of two trays with the same foam-layer thickness at the same temperature.Vitamin C content was determined by titration with sodium 2,6-dichloroindophenolate salt (Sigma-Aldrich, São Paulo, SP, Brazil) using method 967.21 of Official Methods of Analysis of AOAC International (Horwitz, 1980).Yield was calculated as the ratio between the dry material weight and time required for a complete drying, in grams per hour (Anesiadis et al., 2008).
To obtain the optimized drying conditions, the RSM was used (Pandith & Srivastava, 2018).For this, the following equation was fitted to the experimental data of vitamin C and yield: , where Y i is the predicted response (Y 1 is vitamin C and Y 2 is yield), X 0 is the independent value, X 1 and X 2 are the linear coefficients, X 11 and X 22 are the quadratic coefficients, X 12 is the interaction coefficient, T is temperature, and E is foam-layer thickness.
For the characterization of butiá powder, humidity was measured by keeping 3.0 g of the sample, at 105ºC, in a static-air oven until reaching a constant mass (Horwitz, 1980), with results expressed in grams of water per 100 g of sample, on a wet basis.Water activity was determined using the LabSwift-aw water activity meter (Novasina AG, Lachen, Switzerland) at room temperature.The pH and total soluble solids (°Brix) of the pulp, foam, and powder were also obtained.Briefly, 5.0 g of the samples were diluted in 50 mL of distilled water and filtered through an 80 g m -2 filter paper, and, then, the pH and total soluble solids were measured, respectively, using the P1000 benchtop pHmeter (PHOX Suprimentos Científicos, Colombo, PR, Brazil) and the 32ATC portable refractometer (Homis do Brasil: Equipamentos Industriais Ltda, São Table 1.Real and coded values (1) for the concentration of xanthan gum and albumin, as well as for whipping time, in the 2 3 factorial experiment, affecting the results obtained for butiá (Butia spp.) foam density and stability. (1)Encoded values are in parentheses. (2)Assays to evaluate the effect of different concentrations of albumin (2.0-10% w/w), xanthan gum (0.0-0.5% w/w), and whipping times (10-20 min) on the foam density and stability of butiá pulp.
Paulo, SP, Brazil).The solubility of butiá pulp powder was calculated by the ratio between the supernatant mass and the dry solid mass obtained after drying as described by Eastman & Moore (1984) -i.e., 1.0 g of powder was mixed with 100 mL of distilled water, kept under magnetic stirring for 5 min, and then centrifuged at 4,000 rpm for 5 min, after which 10 mL of the supernatant was transferred to a glass Petri dish that had been previously dried.In the present study, the samples were dried in the SL-100 oven (Solab: Equipamentos para Laboratórios, Piracicaba, SP, Brazil), at 105°C, until constant weight.
Water-holding capacity was determined according to Rosell et al. (2009).In this case, samples of 0.5 g of the powder were mixed with 10 mL of distilled water.The mixture was left to stand for 24 hours and, then, was centrifuged at 4,000 rpm for 3 min, and the supernatant was filtered through an 80 g m -2 filter paper.The water retained in the filter and the water collected during filtration were weighed, and the obtained amount was subtracted from the initial weight of water, showing the weight of water per gram of powder.
The color of the pulp, foam, and powder was measured with the CR-400 portable colorimeter (Konica Minolta Sensing Americas, Inc, Ramsey, NJ, USA) using the lightness (L*), red (a*), and blue (b*) coordinates.Hue angle (H o ) and chroma intensity (C*) were calculated through classical equations (McGuire, 1992).
For the statistical analysis, RSM data were analyzed by the Experimental Design module of the Statistica, version 10.0, software (TIBCO Software Inc., Palo Alto, CA, USA) and the analysis of variance (Anova).Three-dimensional surface plots were designed to illustrate the main and interactive effects of the independent variables on the dependent variables.Pulp, foam, and powder characteristics were compared through duplicates of two independent experiments by the Anova and, subsequently, Tukey's test, also using the Statistica, version 10.0, software (TIBCO Software Inc., Palo Alto, CA, USA), with differences considered significant at a 95% confidence level (p<0.05).

Results and Discussion
The results of the RSM for albumin concentration, xanthan gum concentration, and whipping times showed that xanthan gum (linear effect), albumin (linear and quadratic effect), and the interaction between albumin and whipping times significantly affected (r 2 =0.98; p<0.05) foam density (Figure 1).The calculated F-value of 45.91 was higher than the critical F-value of 3.18, indicating the significance of the following model: Y 1 = 0.31 + 0.026G -0.13A + 0.094A 2 + 0.041A×T, where G is xanthan gum, A is albumin, and T are whipping times.
Density showed a concave response to increased changes in the dependent variable due to the linear Table 2. Real and coded values (1) for drying temperature and foam-thickness layer in the 2 2 factorial experiments, affecting the results obtained for the yield and vitamin C retention of butiá (Butia spp.) dried foam and pulp. (1)Encoded values are in parentheses. (2)Assays to evaluate the effect of different temperatures (60-80°C) and foam-layer thicknesses (0.50-1.50 cm) on vitamin C retention and yield.Each drying assay was performed in a duplicate of two independent experiments, consisting of two trays with the same foam-layer thickness at the same temperature.
(3) Means of duplicates from the two independent experiments.and quadratic effect of albumin concentration.Although foam density was lower at a higher albumin concentration, the opposite occurred at concentrations above 9.0% (Figure 1 A), which may be explained by the saturation of albumin solubility (Rajkumar et al., 2007).In general, foam density was higher at 2.0-3.6% and lower at 6.0-10% albumin concentration.
The lowest density was related to the highest concentrations of albumin and the shortest whipping time.Longer whipping times lead to an excessive protein denaturation, which is usually associated with the formation of insoluble aggregates that present a low water-retention capacity, causing foam to collapse and its density to increase (Raikos et al., 2007).
For the foam-layer method, densities from 0.3 to 0.6 g mL -1 are indicated since foams with a low density dry faster, resulting in a lower thermal degradation and a higher general quality (Ratti & Kudra, 2006).Other studies presented similar results, with a reduction in density with the addition of albumin to plum (Prunus domestica L.) foam (Sifat et al., 2021) and an increase in density with the addition of xanthan gum to egg white-based protein foam (Zmudziński et al., 2014).In the present study, the increase in foam density due to the addition of xanthan gum (Figure 1 B) is attributed to the ability of this hydrocolloid to increase viscosity even at low concentrations, causing resistance to air incorporation during whipping (Dabestani & Yeganehzad, 2019).
The linear and quadratic effect of albumin concentration and the interaction of albumin concentration and xanthan gum played a significant role (r 2 =0.77; p<0.10) on foam stability (Y 2 ), whereas whipping time did not (p>0.10).The calculated F-value of 3.45 was higher than the tabulated F-value of 3.18, showing the significance of the following model: Y 2 = -0.11A-0.056A 2 + 0.0811G×A, where A is albumin and G is xanthan gum.
A greater foam stability was obtained with higher amounts of albumin (Figure 1 C); however, when 0.25% (w/w) xanthan gum was added, a smaller amount of albumin was required to reach the desired stability.Susanti et al. ( 2021) highlighted that xanthan gum increased the viscosity of the continuous phase and that its interaction with protein led to a more rigid foam microstructure, whereas Dabestani & Yeganehzad (2019) found that xanthan gum increased the stability of pasteurized fresh egg-white foams, with no liquid drainage at the highest concentrations.
Vitamin C preservation was significantly affected by the linear effect of temperature and layer thickness, as well as by their interaction (r 2 =0.91; p<0.05)  (Figure 2).The calculated F-value of 9.92 was higher than the tabulated F-value of 5.050, showing the significance of the following model: Y = 7.48 + 1.99T -1.49E -1.56T×E,where T is temperature and E is foam-layer thickness.
The amount of ascorbic acid was 56.03 mg 100 g -1 in in natura pulp and between 4.10 and 12.35 mg 100 g -1 in the powders, indicating a reduction of 92.68 to 77.96%.In the analyzed intervals, higher drying temperatures and a thinner foam-layer thickness provided a greater retention of vitamin C (Figure 2 A).It has been reported that vitamin C is more susceptible to degradation when exposed to longer drying times and low temperatures than to shorter times and high temperatures (Pandith & Srivastava, 2018).The smaller losses of vitamin C in the shortest drying time may be related to the lowest exposure of the vitamin to factors such as moisture, light, and oxygen, which contribute to its degradation (Freitas et al., 2018).According to Lee & Labuza (1975), the reduction of moisture increases the viscosity of the medium and reduces the mobility of reaction factors, helping to preserve vitamin C.
For yield, only the linear effect of temperature was significant (r 2 =0.85; p<0.05) and foam-layer thickness did not cause significant changes (p>0.05) in the dependent variable.The calculated F-value of 5.66 was higher than the tabulated F-value of 5.050, showing the significance of the following model: Y = 2.40 + 0.54T, where T is temperature.Therefore, increasing the drying temperature increases yield (Figure 2 B), whereas increasing the thickness of the foam layer does not compensate for the time required for drying.Since food drying is characterized by a high consumption of energy (Qu et al., 2022), the conditions that promote a faster and more efficient removal of moisture are those that result in a higher yield.Using the foam-layer drying method, Araújo et al. (2017) found that the increase in temperature reduced the drying time of acerola (Malpighia emarginata DC.) pulp.
The curves obtained for drying (Figure 3), built from the moisture data on a wet basis and the drying rate as a function of time, were influenced by temperature and by the thickness of the foam layer, showing an exponential decay of moisture over time.When only temperature varied, the data from the drying curves of the first (63°C, 0.65 cm) and second (77°C, 0.65 cm) assays showed a strong drop in humidity at the highest temperature due to the increased heat transferred from the heating air to the foam (Araújo et al., 2017).When only the thickness of the foam layer varied, the data of the second (77°C, 0.65 cm) and fourth (77°C, 1.35 cm) assays indicated a more accentuated drop in moisture at the thinnest thickness, which is explained by the shortest path traveled by water and water vapor from the interior to the surface layer.
The foam showed a lighter color than the pulp, with an increase in luminosity and a decrease in a* and b*, which is attributed to the addition of albumin and the air contained in the foam (Shaari et al., 2018).However, drying had a darkening effect (Table 3) due to the heating of fruit sugars and the agents used for foam formation and stability (Quek et al., 2007), as well as to the degradation and/or transformation of phenolic compounds, vitamin C, and carotenoids (Sant'Anna et The foam had a higher pH than the pulp due to the addition of albumin, whose pH is between 7.0 and 8.0 (Mine, 1995).However, although still considered acid (<5.0), the powder had a higher pH than the foam, which is related to the destruction of acids by heat and drying time (Pandith & Srivastava, 2018).
The increase in the total soluble solids content is attributed to the concentrations of albumin and xanthan gum, as well as to the loss of water during drying.Likewise, Freitas et al. (2018) found that the addition of the foaming agent and drying increased the total soluble solids in cajá (Spondias mombin L.) powder, compared with the pulp.
The results obtained for moisture indicate that the applied drying conditions are sufficient for water removal.Foods presenting a water activity lower than 0.6 are considered microbiologically stable, and any deterioration that occurs is induced by chemical reactions and not by microorganisms (Quek et al., 2007).
The water-holding capacity of the powder of butiá can be explained by the fruit's fiber content, which varies from 0.84 to 4.89% (Hoffmann et al., 2014), whereas solubility is related to the solubility of albumin and the foam structure, resulting in a more porous powder and a greater surface area for the interaction with water (Shaari et al., 2018).Table 3.Values obtained for color, pH, °Brix, humidity, water activity (Aw), water-holding capacity (WHC), and solubility of the butiá (Butia spp.) pulp powder (1) .

Conclusions
1.The foam-layer drying method is a viable alternative for obtaining the powder from butiá (Butia spp.) pulp.
2. Albumin (7.0% w/w), xanthan gum (0.25% w/w), and 10 min of whipping time are the optimized conditions to form foam using butiá pulp suitable for drying.
3. A thinner foam thicknesses of 0.50 cm and a higher drying temperature of 80°C retain more vitamin C, whereas only the use of higher temperature results in a higher yield.
4. The butiá pulp powder presents a low humidity and water activity, an acid pH, and a predominantly yellow color.

Figure 1 .
Figure 1.Butiá (Butia spp.) foam density response surface as a function of whipping time and albumin concentration (A), foam density as a function of whipping time and xanthan gum concentration (B), and foam stability as a function of albumin and xanthan gum concentrations (C).

Figure 2 .
Figure 2. Vitamin C content (A) and yield (B) of butiá (Butia spp.) as a function of drying temperature and foamlayer thickness.

Figure 3 .
Figure 3. Drying curves of butiá (Butia spp.) pulp in the first, second, third, fourth, seventh, and eighth assays as a function of different temperatures and foam-layer thicknesses.