Development of dry acerola and ciriguela mixed pulp obtained by spouted bed drying and freeze-drying¹

ABSTRACT

During the drying process, fruit pulps may undergo changes in their quality. This study aimed to select the optimal drying conditions for acerola-ciriguela mixed pulps using spouted bed drying and freeze-drying methods. For spouted bed drying, the influence of drying temperature, feed rate, and maltodextrin concentration was investigated using a 2³ central composite experimental design. A similar design was applied to freeze-drying, but drying temperature, vacuum chamber pressure, and maltodextrin concentration were selected as the independent variables. In both cases, moisture content, water activity, hygroscopicity, and ascorbic acid retention were used as response variables. Optimisation of both processes was performed according to the desirability function. Higher temperatures and maltodextrin concentrations resulted in lower moisture content in the final product obtained through spouted bed drying. Furthermore, increasing these variables significantly reduced the water activity of both products. Ascorbic acid retention was maximised in the spouted bed-dried product when combining low drying temperatures with a slow suspension feed rate, whereas in the freeze-dried product, higher temperature and chamber pressure achieved the best results. These findings highlight the great potential of freeze-dried mixed acerola-ciriguela powder as an ingredient for vitamin C enrichment in foods.

Keywords:
Desirability function; Water activity; Moisture content; Ascorbic acid retention; Maltodextrin.

INTRODUCTION

The food industry has been forced to create new food products, due to the growing consumer demand for healthier foods and natural products. In this context, fruits are increasingly being used as ingredients in the formulation of new products, meeting the current consumer demand (Germer et al., 2018). Moreover, the addition of fruits can improve the sensory aspects of new products (taste, colour, texture, aroma), increase their fibre content and enrich them in vitamins and minerals (Silva et al., 2016).

Although native to Central America, ciriguela (Spondias purpurea L.) has adapted well to the climate of the Northeast region of Brazil. Being rich in carbohydrates, pro-vitamin A, and phosphorus, its fruit has excellent sensory and nutritional qualities, for which it is widely consumed either fresh or processed (Todisco et al., 2014). Acerola (Malpighia emarginata D. C.), which is rich in antioxidant compounds such as vitamins, carotenoids, and polyphenols, stands out for the high vitamin C content of its fruit (ranging from 342.47 to 2649.72 mg/100 g), is of great economic importance, and shows agroindustrial and nutritional potential (Jaeschke; Marczak; Mercali, 2016; Ribeiro et al., 2018a).

The proposal to mix acerola and ciriguela aims to improve the characteristics of one pulp by complementation with nutrients and phytochemicals contained in the other. Mixed fruit juices and pulps are trends observed on the market that have some advantages compared to their separate constituents, such as improved nutritional and sensorial qualities along with new flavours and aromas (Ribeiro et al., 2018b). Despite all their benefits, these fruits are highly perishable due to their high moisture content, which requires some processing to reduce post-harvest losses. Drying is one of the most used techniques to increase food stability thanks to the reduction of moisture content and water activity (Ramírez; Giraldo; Orrego, 2015).

The spouted-bed dryer with inert particles is considered a promising alternative for drying fruit pulps and producing powders from liquid and paste-like foods due to its flexibility and versatility, lower costs and temperature compared to other drying techniques, along with its potential to drive granulation (Chielle et al., 2016; Dantas et al., 2019).

For processing of fruit pulps on spouted bed drying, it has been shown that the high content of low-molecular-weight sugars tends to reduce process efficiency. To minimise these problems, adjuvants are usually added to protect the active principles during drying; among them, maltodextrin with different dextrose equivalent values is widely used thanks to its high solubility (Moreno et al., 2016). Another technique is freeze-drying, based on the removal of water by sublimation, consisting mainly of freezing and vacuum dehydration (Harnkarnsujarit et al., 2016). It was developed to prevent losses of compounds responsible for food aromas, which are very sensitive to operations at high temperatures (Ramírez; Giraldo; Orrego, 2015).

Water activity, moisture content and hygroscopicity are critical parameters for quality and stability of dry foods, because water availability allows spoilage microorganisms to grow (Morais et al., 2020), while retention of ascorbic acid is desirable due to its beneficial effects on human health. The objective of this work was to determine the optimum processing parameters for spouted bed drying and freeze-drying of acerola-ciriguela mixed pulp, looking for a product with high ascorbic acid retention in conjunction with low water activity, moisture content and hygroscopicity.

MATERIALS AND METHODS

Materials

Acerola and ciriguela fruits were obtained at the Supply and Logistics Centre of Pernambuco (CEASA), Recife, PE, Brazil. The carrier agent was maltodextrin (MOR-REX^® 1910) with 10 dextrose equivalent (DE) acquired from Corn Products (Mogi Guaçu, SP, Brazil). Inert particles of type 849 (1 kg), made of high-density polyethylene, were used to form the spout inside the drying chamber.

Acerola-ciriguela mixed pulp preparation

The fruits were purchased at the Supply and Logistics Centre of Pernambuco (CEASA-PE). Acerola and ciriguela fruits were selected based on their stage of maturation with the characteristic colouration of each one, and green and spoiled fruits were discarded. They were washed, sanitised in 2.5% sodium hypochlorite solution and pulped (Bonina Compacta - Itabuna/BA). The pulp was packed in low-density polyethylene bags, immediately frozen and stored at -18 °C (Figure 1). Acerola-ciriguela mixed pulp (ACMP) was prepared by mixing 60% (w/w) acerola and 40% (w/w) ciriguela, because this proportion had previously received the best sensory evaluation (Ribeiro et al., 2018b).

Figure 1
Flowchart for obtaining acerola-ciriguela mixed pulp (ACMP)

Physicochemical characterisation of fresh and mixed pulps

The fresh pulps of acerola, ciriguela and acerolaciriguela mixed pulp (60% / 40%) were characterised according to the content of soluble solids (SS), titratable acidity (TA), SS/TA ratio, pH, water activity, moisture and ascorbic acid content. Soluble solids content (SS) was determined using a digital refractometer, and the results were expressed in °Brix. Titratable acidity (TA) was determined titrimetrically, and the results were expressed in g citric acid per 100 g. These parameters were analysed according to the Association of Official Analytical Chemists (2016), making it possible to calculate the SS/TA ratio. The pH was measured with a digital pH meter. Water activity (a_w) was determined using an electronic dew point water activity meter, model 4TE (Decagon, Pullman, WA, USA) at 25 °C. Moisture content was determined by drying at 105 ºC for 45 minutes using an infrared scale, model ID 50 (Marte, Piracicaba, SP, Brazil), and the results were expressed in % (w/w). Ascorbic acid content was determined by the Tillmans titration method using 2,6-dichlorophenol indophenol and analysed according to the Association of Official Analytical Chemists (2016).

Drying methods

Acerola-ciriguela mixed pulp (ACMP) was subjected to spouted bed drying and freeze-drying.

Spouted bed drying: ACMP was subjected to drying in spouted bed, model FBDJ 1.0 (Labmaq do Brasil, Riberão Preto, SP, Brazil). Experiments were conducted using temperature (70, 75 and 80 °C), suspension flow rate (4, 5, and 6 mL/min), and maltodextrin concentration (15, 20, and 25% w/w) as independent variables. On the other hand, the air pressure (4 bar) and the amount of polystyrene particles (1 kg) were kept constant in all the runs.

Freeze-drying: 20 g of ACMP were placed in polyethylene Petri dishes (90 mm x 15 mm) and frozen at -22 °C. Plates were then subjected to drying in a pilot freeze dryer, model VirTis Genesis 25L (SP Scientific, Warminster, PA, USA). Experiments were conducted using temperature (-20, -30, and -40 °C), vacuum chamber pressure (300, 400, and 500 mTorr), and maltodextrin concentration (15, 20, and 25% w/w) as the independent variables.

Experimental design

A 2³-central experimental design composed of 8 factorial points (levels ± 1) and 3 central points (level 0) was adopted. For both drying methods, the response variables were water activity, moisture content (%), hygroscopicity (g.100 g^-1) and ascorbic acid retention (%). The results obtained were fitted to the general Eq. (1):

(1) $$Y={\beta }_0+{\beta }_1{X}_1+{\beta }_2{X}_2+{\beta }_3{X}_3+{\beta }_4{X}_1{X}_2+{\beta }_5{X}_1{X}_3+{\beta }_6{X}_2{X}_3$$

where Y is the response, β₀ is the constant regression coefficient, β₁, β₂ and β₃ are the linear regression coefficients, X₁, X_2, and X₃ are the independent variables for each drying method, and β₄, β₅ and β₆ are the coefficients of interaction effects.

Characterisation of ACMP powders

Water activity and moisture content: Water activity (a_w) and moisture content (M) were determined as previously described for pulps.

Hygroscopicity: Samples (1 g) of powders were kept for one week in wide-necked jars sealed inside a desiccator containing a NaCl saturated solution (75.29% relative humidity) and then weighed. The results were expressed as the weight of hygroscopic moisture per g of dry solid (Islam et al., 2016).

Ascorbic acid content and retention: Ascorbic acid retention (AAR) was calculated according to Eq. (2):

(2) $$AAR(\mathrm{\%})=B/A\mathrm{X}100$$

where A is the mass of ascorbic acid present in ACMP (g dry weight) and B is the mass present in the dehydrated pulp (g dry weight).

Statistical analysis

Analysis of variance (ANOVA) and the lack of fit test (F test) were performed, and regression coefficients were estimated using Statistica 10.0 software (StatSoft, Tulsa, OK, USA) at a 5% level of significance. Optimisation of both processes was done according to the desirability function (Derringer; Suich, 1980), whose values vary between 0 and 1, with 0 representing a completely undesired fitting and 1 the most desirable one (Barros Neto; Scarmínio; Bruns, 2010).

RESULTS AND DISCUSSION

Physicochemical characterisation of fresh and mixed pulps

The soluble solids content (SS) was significantly different among acerola, ciriguela, and mixed pulps, with that of ciriguela pulp (17.33 ºBrix) being the highest one (Table 1). Acerola pulp showed values of pH, SS and acidity within the standards of identity and quality for this product (Brasil, 2002). Although there are still no established identity and quality standard for ciriguela pulp, SS and acidity values are close to those found by Maldonado-Astudillo et al. (2014). The physicochemical parameters evaluated are critical indicators of the sensory quality, stability, and consumer acceptance of fruit pulps.

Table 1
Physicochemical characterisation of the pulps of acerola and ciriguela, and of the mixed pulp (60% acerola and 40% ciriguela)

Parameter	Acerola	Ciriguela	Mixed pulp
Soluble solids - SS (ºBrix)	7.23 c ± 0.32	17.33 a ± 0.20	11.06 b ± 0.05
pH	3.33 a ± 0.04	2.93 b ± 0.00	3.25 a ± 0.04
Titratable acidity - TA (g/100 g citric acid)	1.05 a ± 0.01	0.72 c ± 0.05	0.93 b ± 0.00
SS/TA ratio	6.88 c ± 0.24	24.02 a ± 1.41	11.85 b ± 0.07
Water activity	0.99 a ± 0.00	0.98 b ± 0.00	0.99 a ± 0.00
Moisture content (%)	91.61 a ± 0.67	75.59 c ± 0.07	85.48 b ± 0.54

• ^*
  Mean of three replicates ± standard deviation. Mean values in the same row followed by the same superscript letter do not differ significantly from each other (p > 0.05) by the Tukey test

Soluble solids (SS) reflect the sugar content and contribute directly to sweetness, with acerola having lower SS (7.23 ºBrix) and ciriguela significantly higher values (17.33 ºBrix), while the mixture showed an intermediate level (11.06 ºBrix), indicating dilution of sweetness from ciriguela. pH values influence flavour perception and microbial stability, with both acerola (3.33) and the mixture (3.25) being less acidic than ciriguela (2.93). Titratable acidity (TA), which affects sourness, was highest in acerola (1.05 g/100 g), lowest in ciriguela (0.72 g/100 g), and balanced in the mixture (0.93 g/100 g). The SS/TA ratio, a key index for sweetness-acidity balance, increased substantially in ciriguela (24.02) compared to acerola (6.88), with the mixture (11.85) showing a moderate flavour profile. Water activity and moisture content remained high and similar among samples, indicating the pulps’ susceptibility to microbial growth and the need for appropriate preservation. Overall, mixing acerola and ciriguela pulps resulted in intermediate physicochemical characteristics, potentially enhancing sensory balance and consumer appeal.

Dehydration of ACMP by spouted bed drying

The coded levels of the independent variables and the results of quality parameters selected for ACMP dehydration by spouted bed drying are shown in Table 2, while Table 3 lists the results of ANOVA applied to each response. Non-significant variables were omitted, and the other coefficients were used in the final predictive linear equations.

Table 2
Coded levels of the independent variables and results of the responses obtained for acerola-ciriguela mixed pulp powders prepared by spouted bed drying

Run	X1	X2	X3	ɑw	M (%)	H (g/100 g)	AAR (%)
1	-1 (70)	-1 (4)	-1 (15)	0.31 cd ± 0.01	6.26 a ± 0.39	17.65 a ± 0.10	67.12 b ± 0.47
2	+1 (80)	-1 (4)	-1 (15)	0.24 f ± 0.01	4.46 cde ± 0.44	18.17 a ± 0.32	50.86 f ± 0.42
3	-1 (70)	+1 (6)	-1 (15)	0.30 de ± 0.00	5.72 abc ± 0.54	14.23 ef ± 0.26	62.34 c ± 0.43
4	+1 (80)	+1 (6)	-1 (15)	0.27 fe ± 0.01	5.07 abcde ± 0.51	17.03 ab ± 0.02	76.67 a ± 0.42
5	-1 (70)	-1 (4)	+1 (25)	0.30 de ± 0.00	4.39 de ± 0.30	14.58 ef ± 0.09	57.85 e ± 0.47
6	+1 (80)	-1 (4)	+1 (25)	0.36 a ± 0.00	4.04 e ± 0.78	14.26 ef ± 0.38	39.56 h ± 0.88
7	-1 (70)	+1 (6)	+1 (25)	0.32 bcd ± 0.00	4.86 bcde ± 0.38	15.06 de ± 1.01	39.12 h ± 0.49
8	+1 (80)	+1 (6)	+1 (25)	0.34 ab ± 0.01	5.62 abcd ± 0.23	13.58 f ± 0.10	44.27 g ± 0.72
9	0 (75)	0 (5)	0 (20)	0.31 cd ± 0.02	5.48 abcd ± 0.05	16.34 bc ± 0.21	59.71 d ± 0.72
10	0 (75)	0 (5)	0 (20)	0.32 bcd ± 0.00	5.97 ab ± 0.42	15.79 cd ± 0.52	59.55 d ± 0.47
11	0 (75)	0 (5)	0 (20)	0.34 abc ± 0.00	6.00 ab ± 0.45	15.07 de ± 0.28	59.86 d ± 0.54

• ^*
  Mean of three replicates ± standard deviation. X₁₌ temperature (ºC); X₂₌ suspension flow rate (mL/min)_; X₃₌ maltodextrin concentration (%); a_w = water activity; M = moisture content; H = hygroscopicity; AAR = ascorbic acid retention. Mean values in the same column followed by the same superscript letter do not differ significantly from each other (p > 0.05) by the Tukey test

Table 3
Results of ANOVA applied to each response of acerola-ciriguela mixed pulp powders prepared by spouted bed drying

Source	Df	ɑw	M (%)	H (g/100)	AAR (%)
Regression	6	0.026421ns	12.20959*	61.35947*	3874.005*
Residual	26	0.008162	8.80423	10.60768	197.664
Lack of fit	2	0.004460ns	3.83628*	4.61222*	190.066*
Pure error	24	0.003702ns	4.96796ns	5.99457ns	7.598ns
Total	32	0.034583	21.01382	71.96715	4071.669
R2		0.76	0.58	0.85	0.95

The a_w values of ACMP dehydrated by spouted bed drying varied between 0.24 and 0.36 (Table 2), being lower than those of individual tropical fruits (acerola, red pitanga, and jambolan) treated in the same way (Borges et al., 2016) and close to those reported for acerola pulp by Dantas et al. (2019). According to Leong et al. (2011), a_w values lower than 0.6 like these highlight microbiological stability and safety of products. Moreover, such a response was shown to be positively influenced by maltodextrin concentration (^X₃) and its interaction with temperature (^X₁ ^X₃) according to Eq. (3), which showed a satisfactory coefficient of determination (R² = 76%) (Table 3):

(3) $${\alpha }_W=0.316+0.023X_3+0.022X_1{X}_3$$

As observed by Alves et al. (2017), an increase in inlet air temperature is responsible for a higher temperature gradient between the fed material and the drying air, hence resulting in a higher heat transfer rate, better water evaporation, and lower a_w values.

Moisture content (M) ranged from 4.04 to 6.26% (Table 2) in ACMP dehydrated by spouted bed drying. The influence of the independent variables on this response is described by Eq. (4):

(4) $$M=5.26-0.25X_1+0.26X_2-0.32X_3+0.28X_1{X}_2+0.35X_1{X}_3+0.24X_2{X}_3$$

Although the regression coefficients of the model for M were significant, R² (Table 3) was too low (0.58); therefore, this model may only be used to get an idea of results, but not for predictive purposes. Such a response was not only negatively influenced by X₁ and X₃ and positively influenced by the suspension flow rate (X₂), but also positively influenced by interactions among all three variables, with the strongest one being that between X₁ and X₃. A similar decrease in the final moisture content of the product with increasing temperature was reported by Alves et al. (2017) for probiotic orange juice. It has been suggested that the higher the temperature, the higher the rate of mass and heat transfer to the particle, thus the stronger the conduction force for moisture evaporation (Costa et al., 2015; Lourenço; Moldão-Martins; Alves, 2020). The M values obtained in this study are not so far from those reported for açaí powders (3.22-4.79%) (Costa et al., 2015; Lucas; Zambiazi; Costa, 2018), as well as for acerola (8.5%), red pitanga (8.3%) and jambolan (9.7%) powders (Borges et al., 2016), all obtained by spouted bed drying.

The hygroscopicity (H) of ACMP powders obtained in the spouted bed ranged from 13.58 to 18.17 g/100 g (Table 2) and was linearly reduced by an increase in maltodextrin concentration according to Eq. (5):

(5) $$H=15.61-0.59X_2-1.19X_3-0.64X_1{X}_3+0.54X_2{X}_3$$

which accounted for 85% (R² = 0.85) of its variation (Table 3).

This trend may be ascribed to the fact that maltodextrin, a material with low hygroscopicity, reduces the water adsorption capacity of dehydrated products (Ferrari; Ribeiro; Aguirre, 2012). The H values obtained in this study are close to those reported for cajá microencapsulated by spray drying (16.16-20.86 g/100 g) using 10 DE maltodextrin (Morais et al., 2020) and for powdered acerola prepared by spouted bed drying (14.10-18.88 g/100 g) (Dantas et al., 2019).

Ascorbic acid retention (AAR) ranged from 39.12 to 76.67% (Table 2) in ACMP dehydrated by spouted bed drying, with the highest value being close to that (72.90%) reported for powdered acerola-containing formulations prepared by the same method using powdered milk as an adjuvant (Dantas et al., 2019).

Although the lack of fit was significant for the AAR model (Table 3), since the pure error value was low, i.e., very good reproducibility, there was a false result of lack of fit because the F calculated value was high due to the very low value at the denominator; therefore, the model can be considered to have statistical and predictive significance with no lack of fit.

Similarly to moisture content, AAR was negatively influenced by X₁ and X₃, but the interactions among variables were more complex, in that, contrary to X₁X₂ interaction, those of X₃ with the other variables had a negative impact. Such behaviour is depicted by Eq. (6), which had excellent fit accounting for no less than 95% of the variation of this response (R² = 0.95):

(6) $$AAR=56.14-1.86X_1-9.50X_3+6.77X_1{X}_2-1.38X_1{X}_3-4.36X_2{X}_3$$

However, X₃ was by far the variable that most affected AAR, i.e., the higher the maltodextrin concentration, the lower the AAR in produced powders, confirming the negative effect of maltodextrin addition on this response. A similar effect was obtained in camucamu pulp dehydrated by this method using maltodextrin as a protectant of bioactive compounds (Fujita et al., 2013). Tonon, Brabet and Hubinger (2010) stated that increasing the maltodextrin content causes some “dilution” of the product’s pigments, which explains the low AAR values in acerola-ciriguela mixed pulp powders prepared by spouted bed drying.

Dehydration of acerola-ciriguela mixed pulp by freeze-drying

Similarly to spouted bed drying, Table 4 lists the coded levels of the independent variables and the results of quality parameters selected for ACMP freeze-drying, while Table 5 shows those of ANOVA applied to the same responses omitting non-significant variables and using the other coefficients in the final predictive linear equations.

Table 4
Coded levels of the independent variables and results of the responses obtained for acerola-ciriguela mixed pulp powders prepared by freeze-drying

Run	X1	X2	X3	ɑw	M (%)	H (g/100 g)	AAR (%)
1	-1 (-40)	-1 (300)	-1 (15)	0.26 de ± 0.01	5.03 bc ± 0.58	16.15 ab ± 0.59	65.40 e ± 1.31
2	+1 (-20)	-1 (300)	-1 (15)	0.20 b ± 0.02	5.73 cd ± 0.21	16.08 a ± 0.10	59.93 d ± 2.01
3	-1 (-40)	+1 (500)	-1 (15)	0.35 a ± 0.00	6.97 ab ± 0.31	16.00 bc ± 0.21	62.13 c ± 2.29
4	+1 (-20)	+1 (500)	-1 (15)	0.33 a ± 0.01	6.62 a ± 0.51	14.97 ab ± 0.74	73.48 de ± 1.33
5	-1 (-40)	-1 (300)	+1 (25)	0.20 e ± 0.00	6.04 cd ± 0.60	14.03 c ± 0.03	63.98 f ± 0.91
6	+1 (-20)	-1 (300)	+1 (25)	0.17 de ± 0.00	5.29 abc ± 0.16	14.15 c ± 0.27	54.84 de ± 0.95
7	-1 (-40)	+1 (500)	+1 (25)	0.26 de ± 0.01	5.69 d ± 0.20	14.57 c ± 0.26	62.46 bc ± 1.02
8	+1 (-20)	+1 (500)	+1 (25)	0.20 bc ± 0.01	4.62 bc ± 0.24	13.95 c ± 0.35	77.20 de ± 1.64
9	0 (-30)	0 (400)	0 (20)	0.22 cd ± 0.01	5.02 cd ± 0.16	15.98 ab ± 0.31	69.10 a ± 1.91
10	0 (-30)	0 (400)	0 (20)	0.23 bcd ± 0.00	5.67 bc ± 0.15	16.56 a ± 0.39	69.00 a ± 1.68
11	0 (-30)	0 (400)	0 (20)	0.21 de ± 0.01	5.32 cd ± 0.23	16.05 ab ± 0.35	67.22 ab ± 0.97

• ^*
  Mean of three replicates ± standard deviation. X₁ = freeze-drying temperature; X₂ = vacuum chamber pressure; X₃ = maltodextrin concentration; a_w = water activity; M = moisture; H = hygroscopicity; AAR = ascorbic acid retention. Mean values in the same column followed by the same superscript letter do not differ significantly from each other (p > 0.05) by the Tukey test

Table 5
Results of ANOVA applied to each response of acerola-ciriguela mixed pulp powders prepared by freeze-drying

Source	Df	aw	M (%)	H (g/100 g)	AAR (%)
Regression	6	0.084595ns	12.8554*	19.26365*	359.300*
Residual	26	0.010862	4.62730	13.38490	34.7501
Lack of fit	2	0.006490*	1.29767*	9.57621*	32.6405*
Pure error	24	0.004372ns	3.32962ns	3.80869ns	2.1123ns
Total	32	0.095457	17.48270	32.64855	394.0501
R2		0.88	0.73	0.59	0.91

The a_w values of freeze-dried ACMP (0.17-0.35) (Table 4) were even lower than those obtained by the spouted bed drying process (Table 2). These values are close to those reported for different freeze-dried fruit pulps (Barbosa et al., 2015; Caliskan; Dirim, 2016; Ramírez; Giraldo; Orrego, 2015). The influence of the independent variables on this response is described by Eq. (7), which explained 88% of its variation (R² = 0.88) (Table 5):

(7) $${\alpha }_W=0.244-0.021X_1+0.036X_2-0.038X_3-0.015X_2{X}_2$$

Based on the features of this equation, the formation of particles with low a_w is expected by a decrease in X₂ as well as increases in X₁ and X₃, with only a weak negative interaction of X₂ and X₃. Increasing the maltodextrin concentration led to a reduction in the a_w of freeze-dried ACMP, as maltodextrin is a high-molecular-weight carbohydrate with low hygroscopicity. Higher maltodextrin levels increase the proportion of dry solids in the formulation, thereby diluting the free water fraction and decreasing the amount of water available for microbiological and chemical reactions (Caliskan; Dirim, 2016; Igual et al., 2014). Similarly, Bednarska and Janiszewska-Turak (2020) and Janiszewska-Turak et al. (2017) observed that the greater the addition of encapsulant, the lower the a_w values of chokeberry and carrot powder juices, respectively.

The moisture content of freeze-dried ACMP, ranging from 4.62 to 6.97% (Table 4), was only a little higher than that of ACMP powders prepared by spouted bed drying and was influenced by all variables according to Eq. (8) (R² = 0.73) (Table 5):

(8) $$M=5.63-0.18X_1+0.22X_2-0.33X_3-0.17X_1{X}_2-0.27X_1{X}_3-0.48X_2{X}_3$$

In addition to the negative influence of drying temperature (X₁) on this response already observed for powders prepared by spouted bed drying, X₂ had a positive effect, that is, the higher the vacuum chamber pressure, the higher the moisture content of particles produced by this drying method. On the other hand, maltodextrin concentration (X₃) had a negative effect, which has also been previously observed for sumac extract (Caliskan; Dirim, 2016).

The hygroscopicity of freeze-dried ACMP ranged from 13.95 to 16.56 g/100 g (Table 4) and was negatively influenced by ^X₁, ^X₃ and ^X₁^X₂ interaction and positively influenced by X₂X₃ interaction, as shown by Eq. (9), which showed, however, poor fit (R² = 0.59) (Table 5):

(9) $$H=15.32-0.20X_1-0.81X_3-0.21X_1{X}_2+0.19X_2{X}_3$$

Maltodextrin concentration had greater negative impact on hygroscopicity than freeze-drying temperature, similarly to what was observed for freeze-dried pulp of lulo (Solanum quitoense L.) (Igual et al., 2014).

The ascorbic acid retention in freeze-dried ACMP, ranging from 54.84 to 77.20% (Table 4), was, on average, higher than that ensured by the spouted bed drying process (Table 3). This response was positively influenced only by X₂ and X₁X₂ interaction, according to Eq. (10), which accounted for 91% of its variation (R² = 0.91) (Table 5):

(10) $$AAR=65.87+3.89X_2+5.08X_1{X}_2$$

The Pareto charts depicted in Figures 2 and 3 illustrate the statistical effects of the independent variables that exerted statistically significant influence on water activity, moisture content, hygroscopicity, and ascorbic acid retention of dry ACMP prepared by spouted bed drying and by freeze-drying, respectively.

Figure 2
Statistical analysis of preparation of dry acerola-ciriguela mixed pulp by spouted bed drying. Pareto charts showing the influence of the independent variables on the responses, namely water activity (A), moisture content (B), hygroscopicity (C), and ascorbic acid retention (D)

Figure 3
Statistical analysis of preparation of dry acerola-ciriguela mixed pulp by freeze-drying. Pareto charts showing the influence of the independent variables on the responses, namely water activity (A), moisture content (B), hygroscopicity (C), and ascorbic acid retention (D), respectively

Optimisation of operating conditions by the desirability function

Even though the desirability function was applied to all responses, only the plots referring to water activity and ascorbic acid retention of powders obtained by both methods are illustrated in Figure 4 because they were the most influenced responses. In addition, since the objective of this study was to obtain a juice powder with the highest possible ascorbic acid content and the lowest possible water activity, maximisation of AAR and minimisation of a_w were imposed as constraints on the desirability function. According to Figure 4A, desirability for a_w of ACMP powder prepared by spouted bed drying can be maximised by reducing drying temperature and suspension flow rate and increasing, at the same time, maltodextrin concentration. On the other hand, the AAR desirability (Fig. 4B) can be increased by simultaneously raising drying temperature and suspension flow rate or by simultaneously reducing drying temperature, suspension flow rate and maltodextrin concentration.

Figure 4
Plots of the desirability function applied to water activity (A and C) and ascorbic acid retention (B and D) of dry acerola-ciriguela mixed pulp prepared by spouted bed drying and freeze-drying, respectively

As for ACMP powders obtained by freeze-drying, desirability for a_w can be maximised by increasing freezedrying temperature and maltodextrin concentration but reducing, at the same time, chamber pressure (Fig. 4C), while that for AAR can be maximised by simultaneously increasing all three independent variables (Fig. 4D).

The statistical analysis of the central experimental design (11 runs) and the desirability analysis generated critical values that could be used as optimal operating conditions to obtain ACMP powders by spouted bed drying and freeze-drying (Figure 4). These conditions for spouted bed drying are drying temperature of 75 ºC, suspension flow rate of 5 mL/min and maltodextrin concentration of 15%, while those for freeze-drying are drying temperature of -20 ºC, vacuum chamber pressure of 500 mTorr and maltodextrin concentration of 25%.

CONCLUSIONS

In this work, dehydration of acerola-ciriguela mixed pulp (ACMP) by spouted bed drying and freeze-drying was studied by evaluating the influence of operating conditions on moisture content, water activity, hygroscopicity, and ascorbic acid retention. The increase in temperature caused a decrease in moisture content and water activity of powders obtained by both drying methods. An increase in maltodextrin concentration caused a decrease in water activity of freeze-dried ACMP and a decrease in hygroscopicity of powders prepared by both methods. As for the ascorbic acid retention in powders, this response was reduced by an increase in maltodextrin concentration when spouted bed drying was used as dehydration method, while it increased with increasing the vacuum chamber pressure in freeze-dryer. Even though both methods ensured very good retention of ascorbic acid, freeze-drying showed better performance in this aspect; therefore, the ACMP powder produced by freeze-drying may be a good source of ascorbic acid. Nevertheless, that prepared by spouted bed drying also had satisfactory physicochemical characteristics, thus being an alternative to freeze-drying according to the circumstances. In conclusion, ACMP can be dehydrated by both methods in order to prolong the shelf life and can find several applications in the development of functional foods with high ascorbic acid content.

• 1
  Dissertation

ACKNOWLEDGEMENTS

This work was supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001, the National Institute of Science and Technology of Tropical Fruits, and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq).

DATA AVAILABILITY STATEMENT

Data-available-upon-request to the corresponding author.

REFERENCES

• ALVES, N. N. et al Spouted bed as an efficient processing for probiotic orange juice drying. Food Research International, v. 101, p. 54-60, 2017.
• ASSOCIATION OF OFFICIAL ANALYTICAL CHEMISTRY. Official methods of analysis of the Association of Official Analytical Chemistry 20 th ed. Washington, DC: AOAC, 2016.
• BARBOSA, J. et al Comparison of spray drying, freeze drying and convective hot air drying for the production of a probiotic orange powder. Journal of Functional Foods, v. 17, p. 340-351, 2015.
• BARROS NETO, B.; SCARMÍNIO, I. S.; BRUNS, R. E. Como fazer experimentos 4. ed. Porto Alegre: Bookman, 2010.
• BEDNARSKA, M. A.; JANISZEWSKA-TURAK, E. The influence of spray drying parameters and carrier material on the physico-chemical properties and quality of chokeberry juice powder. Journal of Food Science Technology, v. 57 n. 2, p. 564-577, 2020.
• BORGES, K. C. et al. Physicochemical characterization and bioactive value of tropical berry pomaces after spouted bed drying. Journal of Food Quality, v. 39, n. 3, p. 192-200, 2016.
• BRASIL. Ministério da Saúde. Secretária de Políticas de Saúde. Alimentos regionais brasileiros Brasília: Ministério da Saúde, 2002.
• CALISKAN, G.; DIRIM, S. N. The effect of different drying processes and the amounts of maltodextrin addition on the powder properties of sumac extract powders. Powder Techonology, v. 287, p. 308-314, 2016.
• CHIELLE, D. P. et al. Spouted bed drying of papaya seeds for oil production. LWT - Food Science and Technology, v. 65, p. 852-860, 2016.
• COSTA, R. G. et al. Effect of operating conditions on the yield and quality of açaí (Euterpe oleracea Mart.) powder produced in spouted bed. LWT - Food Science and Techonology, v. 64, n. 2, p. 1196-1203, 2015.
• DANTAS, S. C. D. M. et al. Spouted-bed drying of acerola pulp (Malpighia emarginata DC): effects of adding milk and milk protein on process performance and characterization of dried fruit powders. Journal of Food Process Engineering, v. 42, n. 6, e13205, 2019.
• DERRINGER, G.; SUICH, R. Simultaneous optimization of several response variables. Journal of Quality Technology, v. 12, n. 4, p. 214-219, 1980.
• FERRARI, C. C.; RIBEIRO, C. P.; AGUIRRE, J. M. Spray drying of blackberry pulp using maltodextrin as carrier agent. Brazilian Journal Food Technology, v. 15, n. 2, p. 157-165, 2012.
• FUJITA, A. et al. Impact of spouted bed drying on bioactive compounds, antimicrobial and antioxidant activities of commercial frozen pulp of camu-camu (Myrciaria dubia Mc. Vaugh). Food Research International, v. 54, n. 1, p. 495-500, 2013.
• GERMER, S. P. M. et al. Influence of process variables on the drum drying of mango pulp. Drying Technology, v. 36, n. 12, p. 1488-1500, 2018.
• HARNKARNSUJARIT, N. et al. Effects of freezing on microstructure and rehydration properties of freeze-dried soybean curd. Journal of Food Engineering, v. 184, p. 10-20, 2016.
• IGUAL, M. et al Optimization of spray drying conditions for lulo (Solanum quitoense L.) pulp. Powder Technology, v. 256, p. 233-238, 2014.
• ISLAM, M. Z. et al Effect of vacuum spray drying on the physicochemical properties, water sorption and glass transition phenomenon of orange juice powder. Journal of Food Engineering, v. 169, p. 131-140, 2016.
• JAESCHKE, D. P.; MARCZAK, L. D. F.; MERCALI, G. D. Evaluation of non-thermal effects of electricity on ascorbic acid and carotenoid degradation in acerola pulp during ohmic heating. Food Chemistry, v. 199, p. 128-134, 2016.
• JANISZEWSKA-TURAK, E. et al The influence of carrier material on some physical and structural properties of carrot juice microcapsules. Food Chemistry, v. 236, p. 134-141, 2017.
• LEONG, S. L. et al The extreme xerophilic mould Xeromyces bisporus: growth and competition at various water activities. International Journal of Food Microbiology, v. 145, n. 1, p. 57-63, 2011.
• LOURENÇO, S. C.; MOLDÃO-MARTINS, M.; ALVES, V. D. Microencapsulation of pineapple peel extract by spray drying using maltodextrin, inulin, and arabic gum as wall matrices. Foods, v. 9, n. 6, 2020.
• LUCAS, B. F.; ZAMBIAZI, R. C.; COSTA, J. A. V. Biocompounds and physical properties of açaí pulp dried by different methods. LWT - Food Science and Techonology, v. 98, p. 335-340, 2018.
• MALDONADO-ASTUDILLO, Y. I. et al. Postharvest physiology and technology of Spondias purpurea L. and S. mombin L. Scientia Horticulturae, v. 174, p. 193-206, 2014.
• MORAIS, B. D. M. et al Optimization of red mombin juice spray drying: physicochemical and sensory analyses of powders. Chemical Engineering & Technology, v. 43, n. 9, p. 1832-1840, 2020.
• MORENO, T. et al. Spray drying formulation of polyphenols-rich grape marc extract: evaluation of operating conditions and different natural carriers. Food and Bioprocess Technology, v. 9, n. 12, p. 2046-2058, 2016.
• RAMÍREZ, M. J.; GIRALDO, G. I.; ORREGO, C. E. Modeling and stability of polyphenol in spray-dried and freeze-dried fruit encapsulates. Powder Technology, v. 277, p. 89-96, 2015.
• RIBEIRO, C. M. C. et al. Fruit development and ripening of acerola - Okinawa cultivar. Acta Horticulturae, v. 1198, p. 199-204, 2018a.
• RIBEIRO, C. M. C. M. et al Optimization of the spray drying process conditions for acerola and seriguela juice mix. Food Science and Technology, v. 39, n. 1, p. 48-55, 2018b.
• SILVA, L. B. et al Chewy candy as a model system to study the influence of polyols and fruit pulp (açai) on texture and sensorial properties. LWT - Food Science and Technology, v. 65, p. 268-74, 2016.
• TODISCO, K. M. et al The use of headspace solid-phase microextraction (HS-SPME) to assess the quality and stability of fruit products: an example using red mombin pulp (Spondias purpurea L.). Molecules, v. 19, n. 10, p. 16851-16860, 2014.
• TONON, R. V.; BRABET, C.; HUBINGER, M. D. Anthocyanin stability and antioxidant activity of spray dried açai (Euterpe oleracea Mart.) juice produced with different carrier agents. Food Research International, v. 43, p. 907-914, 2010.

• Editor-in-Chief:
  Profa. Regilda Saraiva dos Reis Moreira-Araújo - [e-mail]

Publication Dates

• Publication in this collection
  08 May 2026
• Date of issue
  2026

History

• Received
  02 Apr 2024
• Accepted
  23 Oct 2025