2.1. Material
Quince (Cydonia oblonga Mill.), sourced locally from the village of Smilets near Pazardzhik city, Bulgaria, undergoes a meticulous preparation process. The procedure involves washing, peeling, and drying the fruits. These fruits are sliced into small pieces, approximately 3.0mm to 5.0mm in thickness, with the removal of seeds. The freshly sliced quince is then arranged in a single layer on a Food Dehydrator (Dryer) Machine and subjected to a 10-hour drying period at 45°C, or until the moisture content falls within the range of 9% to 13%, ensuring they attain complete dryness and brittleness. Following the drying phase, the quince slices are finely milled using a Nutri-bullet blender into a powdered form. This entire preparation process takes place at the Technical University of Sofia, branch Plovdiv, Bulgaria.
2.2. Methods
The analysis of quince powder involved the assessment of moisture, protein, fat, ash, and dietary fiber content in accordance with established food analysis methods [
13]. Specifically, protein content was determined by multiplying the quantified nitrogen using the conversion factor 6.25 (AOAC 920.152). The fat, obtained through Soxhlet, was determined gravimetrically after solvent removal (AOAC 920.85). Ash content was acquired by burning at approximately 550 °C (AOAC 940.26), and dietary fiber was obtained through the enzyme-gravimetric method (AOAC 985.29). Carbohydrates were calculated as the residual fraction. Results are expressed as g/100 g dry weight (dw). The nutritional energy value (kcal/100 g fw and dw) was computed using the following conversion factors: 9 kcal/g for fat, 4 kcal/g for protein and carbohydrates, and 2 kcal/g for fiber.
Color measurements were conducted employing a colorimeter (model PCE-CSM, Germany) with a viewing angle of 0° and a pulsed xenon lamp as the light source. The instrument provides readings in terms of color coordinates, where L* represents whiteness to darkness, a* denotes redness to greenness, and b* signifies yellowness to blue. Instrument calibration was executed using a standard white plate, and samples were positioned on a petri dish for each measurement.
The hue angle (H°) and chromaticity (C) were calculated using the following equations [Eq. 1 and Eq. 2]:
The antioxidant activity of quince powder assessed through DPPH (1,1-diphenyl-2-picrylhydrazyl rad-ical), ABTS (2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid)), FRAP (Ferric Re-ducing Antioxidant Power Assay), and CUPRAC (CUPric Reducing Antioxidant Ca-pacity), was conducted through a two-step triple extraction. Approximately 1g of the sample was mixed with 10 ml of 70% ethanol and conducted in a water bath at 80 °C. Additionally, using an ultrasonic bath, ultrasonic extraction was carried out at a temperature of 50 °C, repeated three times in 20-minute intervals. After the extraction process, the sample was centrifuged at 6000 rpm for 15 min. The supernatant was transferred to a new tube, and another 10 ml of ethanol was added to the precipitate for the second extraction. After the third extraction, the supernatants were mixed and stored in a refrigerator. The protocol for the methods adhered to the description comprehensively detailed in two articles by Ivanov et al. (2014) and Bogoeva et al. (2017) [
14,
15].
The characteristics of sorption of Bulgarian quince powder (equilibrium moisture content and monolayer moisture content) are conducted following the specification of the static-gravimetric method (at 10°C, 25°C and 40°C and water activity of 0.1 to 0.9) and the detailed steps describe in the article of Bogoeva, 2020 [
16]. For the analysis, the sample undergoes preparation by placing a portion in a desiccator over distilled water for desorption, while another portion is placed in a desiccator over CaCl
2 for adsorption. Following 20 days of hydration of one part and dehydration of the other part, the resulting powder is transferred to aluminum weighing plates and measured (1.0000g ± 0.0050g) using the analytical balance. Hygrostats, consisting of borosilicate glass jars with acrylic plastic lids featuring silicone rings, are prepared for use with saturated salt solutions derived from LiCl, CH
3COOK, MgCl
2, K
2CO
3, MgNO
3, NaBr, NaCl, KCl, creating conditions for water activity (a
w) ranging from 0.1 to 0.9. Thymol crystals are introduced into each hygrostat with a water activity exceeding 0.5 to prevent microbiological growth. The weighed samples are then placed in the prepared jars, positioned in three distinct thermostats set at temperatures of 10°C, 25°C, and 40°C. These samples remain in these environments until they reach equilibrium moisture content, typically around one month [
16].
Several mathematical models are available for predicting the equilibrium moisture content. To analyze the obtained equilibrium sorption data, four modified three-parametrical models, namely Oswin [Eq. 3], Henderson [Eq. 4], Ching-Pfost [Eq. 5], and Halsey [Eq. 6], were chosen.
The equations [Eq. 3, Eq. 4, Eq. 5 and Eq. 6] involve parameters such as moisture content (M in % d.b.), water activity (a
w in decimal), and coefficients A, B, and C, with temperature (t in °C) playing a role. The fitting of these modified models was executed using the computer program StatSoft’s
STATISTICA 12. Evaluation, estimation, and comparison of the models were performed based on three criteria: mean relative error (P%) [Eq. 7], standard error of moisture (SEM) [Eq. 8], and randomness of residuals [Eq. 9], according to the equations used of Durakova, 2020 [
17].
The monolayer moisture content represents the maximum amount of moisture that the powder can retain on its surface at given temperatures (10°C, 25°C, and 40°C) and relative air humidity ranging from 11% to 87%. The powder remains dry when the moisture content is below the monolayer level, but it can become sticky, clump together, or spoil when it exceeds this level. Calculation of the monolayer moisture content involves the linearization of the Brunauer-Emmett-Teller equation [Eq. 10], where parameters include monolayer moisture content (M in % d.b.), water activity (a
w in decimal), and the coefficient C [
16,
17].