3.2. Technological Properties of Industrially Produced APF
Technological properties of fine and coarse APF were assessed (Table 1
). Since APF was produced with the intention to be used in bakery and confectionery products, some of the parameters determined were compared with literature data available for wheat flour as well as commonly used gluten-free flours.
Average particle size of 0.16 and 0.5 mm for fine and coarse APF was determined by sieving, in the intervals of 0.06 to 0.30 and 0.06 to 1 mm, respectively. The maximum weight percentage for coarse flour is between 500 μm < n < 1 mm (28.32%), while for fine flour the distribution is slightly different and the maximum weight percentage is 27.78% for particle size between 160 μm < n < 300 μm.
Bulk density of fine and coarse APF was 435 ± 16 and 459 ± 20 g/L, and packed density 632 ± 25 and 591 ± 24 g/L, respectively. Bulk and packed density of sugar-depleted AP powder dominantly consisting of particles smaller than 150 μm was 557 and 447 g/L respectively [36
]. Lowering of moisture content decreased the bulk density of apple powders obtained using various drying methods. The same trend was observed for tapped density of AP that ranged from 430 to 580 g/L [37
]. Based on data available for other flours, it can be seen that while packed density values of APF are in concordance to those found in wheat flour (640 g/L) [38
], bulk density is lower compared to other flours including wheat [39
], rice [41
] and corn [42
]. According to Oladapo et al. [43
], low values of bulk densities make the flour more suitable for the baking process.
Water holding capacity, referring to the ability of material to bind and hold water within the matrix, depends on the content and chemical structure of DF [44
]. Reported AP WHC ranged from 1.62 g/g [45
] to 6.34 g/g [46
]. Values obtained for WHC of fine and coarse APF (4.69 ± 0.19 and 4.79 ± 0.18 g/g) were in concordance with previously reported data [36
]. The main factors affecting WHC are related to the composition of AP, possible pre-treatment such as washing or bleaching, and drying conditions. Breakdown of cell wall polysaccharides does not occur at low dehydration temperatures, allowing for a high WHC of APF.
Since the amount of water needed to hydrate flour components to produce dough with optimum consistency is one of the most fundamental quality parameters of flour [49
], values obtained for APF were compared with reported data for both wheat and other flours. The WHC values obtained for APF were found higher than those reported for fine (1.14 g/g) [50
], full fat (1.85 g/g) and defatted (1.92 g/g) wheat flour [39
], as well as commercial (0.88 g/g) [41
], full fat (1.26 g/g) and defatted (1.56 g/g) rice [39
] and corn flour (1.57 g/g) [51
]. WHC of APF was also found higher than the WHC reported for oat bran (2.1 g/g) [52
]. It is known that flours with a high WHC are widely used in foods like meat products, custards and soups to enhance thickening and viscosity, and in baked products to improve freshness and handling characteristics [10
The ability of a powder to dissolve in water indicates its capacity of rehydration. Solubility affects the functional characteristics of powders in food systems. Lower values for fine and coarse APF solubility (27.9% ± 0.9% and 29.1% ± 0.7%) corresponded to high content of DF. Apple pomace powder solubility of 37.5% (total fiber content 26.5%) obtained by flash blanching, freeze-drying was reported [53
High hydrated density of fine and coarse APF (0.50 ± 0.02 and 0.63 ± 0.03 g/mL) in comparison to reported data [28
] can also be attributed to conditions of water removal that did not cause cell wall material shrinkage.
Swelling capacity, as well as WHC and OHC, provide insight regarding DF behavior during food processing and gut transit [54
]. SWC, referred to the amount of insoluble fiber, was 5.5 ± 0.2 and 7.0 ± 0.3 mL/g for fine and coarse APF, respectively. SWC of sugar-depleted AP powder was found to be 7.0 mL/g [55
]. Porous structures developed within the cell wall matrix during the process of water removal at low temperature allow for easy and complete rehydration.
Oil holding capacity of 1.27 ± 0.04 and 1.40 ± 0.05 g/g ascribed to fine and coarse APF, respectively, are in accordance with values reported for laboratory produced AP powder [47
] and other plant raw materials [39
]. Higher values were reported for boiled (1.69 g/g) [48
] and washed AP (2.24 g/g) [55
]. On the other hand, lower values were found in different rice flours, full fat (0.75 g/g) and defatted (1.1 g/g) [39
], dry-milled (0.8 g/g), wet-milled (0.58 g/g) and commercial (0.5 g/g) [41
]. OHC is related to the presence of lignin, its structure and surface properties, overall charge density, thickness, hydrophobic nature and size of particles [45
]. Increase in drying temperature decreases the OHC value.
All values determined decreased with the decrease in APF particle size. Distinct effect of particle size on hydration capacity, oil absorption and emulsifying properties is in accordance with reported data [58
]. Grinding may affect the hydration properties of DF as a result of an increase in surface area, leading to faster hydration [59
]. However, it may cause alternation and collapse of the fiber matrix that traps water resulting in a water retention decrease. Both AP dehydration and dry AP grinding to fine and coarse APF in our study was obviously performed with no impact on the hydration characteristics of wet AP. Technological properties of APF confirmed its effectiveness in fortifying and development of DF-rich food products as well as in low calorie food.
3.5. Sensorial Properties of Fortified Cookies
In several researches, bakery and confectionery products various shares of AP incorporated were assessed for sensory properties and consumer acceptance. While products with the AP share of up to 20% usually had a pleasant apple flavor and good acceptance at the consumer level [19
]—the sensory parameters more often decreased with a further increase in the AP ratio. Lower or moderate acceptability of cookies with 30%, 40% and 50% of AP was reported [24
]. In concordance with previously published studies, cookies produced within the scope of this study with 25% of APF were best rated. From the obtained results (Table 2
), it can be seen that in terms of overall sensory quality, the best rated cookies were CC25. They received the highest scores for all tested attributes. They were distinguished by typical appearance, corresponding texture and distinctive aroma. However, CC50 were also in the category of ‘excellent quality’ while all other samples had ‘very good quality’, including the control. Due mostly to an insufficiently baked layer in the structure of the cookies and a lack of crunchiness, CC75 received the lowest scores.
The use of laboratory produced AP powder increased the ratings for fruit and baking flavor of short dough biscuits, according to Alongi et al. [47
]. While finer particle size of AP performed better than larger ones when used for cakes [63
], this study revealed that cookies containing coarse APF possess better sensory properties along with superior nutritional value.
No differences in panel scores were observed previously for cookies with 30%, 40% and 50% of AP [25
]. They all had lower scores in comparison to cookies produced and tested within the scope of this study. The average score for appearance was 3.1 ± 1.1; flavor, 3.0 ± 1.0, texture, 3.1 ± 1.1 and overall acceptability, 3.0 ± 1.0 (8-point hedonic scale of 1 = like extremely and 8 = dislike extremely was used) [25
3.7. Estimation of the Optimal Share and Particle Size of APF
To obtain a more comprehensive comparison between the produced cookies and to achieve as reliable a recommendation as possible regarding which ratio of APF to use in cookie fortification, functional and sensory characteristics were subjected to multivariate analysis by principal component analysis (PCA) followed by a Varimax rotation. Therefore, the following variables: TPC, TFC, ABTS, DPPH, appearance, structure, chewiness, odor, taste and overall quality were considered. The results showed that the first two principal components (PC) explain 94.86% of the total variance of initial data (PC1 explains 47.79%, and PC2 explains 47.07%). Projection of the initial variables in the PC1 vs. PC2 plane is shown in Figure 4
It can be seen that the PC1 is strongly related to variables associated with polyphenolic (TPC) and flavonoid (TFC) content and scavenging capacity (ABTS and DPPH). In the case of PC2, the most important variables are related to the sensory perception: structure, chewiness, odor, taste and overall quality. Thus, one can interpret the two first components, as follows: PC1—physicochemical characteristics (antioxidative property) and PC2—sensory characteristics of samples.
Samples at the beginning (prefix ”0-“) and after 12 months (prefix “12-“) in the PC1 vs. PC2 plane are shown in Figure 4
B. The samples that are close to each other have similar properties, overall, and samples that are far apart are very different. To examine the influence of storage time on the change in functional and sensory characteristics of the samples, the distance between the points representing the sample at the beginning and after 12 months were observed. A larger distance indicates more intensive changes in the analyzed characteristics of sample over the observed time period. It can be seen that CC50 and CF25 changed the least with time in terms of functional and sensory properties. PCA results unequivocally confirm the better performance of coarse APF in cookie formulations. It allows CC50 to be reliably recommended for further production. Hence, consumers’ acceptance of cookies with 50% of coarse flour was further tested by hedonic tests.
3.8. Consumer Acceptability of Cookies with 50% of Coarse APF
The results of the mean drop analysis for CC50 are shown in Figure 5
. A point in the plot that shows a statistically significant mean drop and the percentage of consumers above the cut-off point is a cause for concern and suggests that the product be modified in the appropriate direction [64
]. Two sufficiently large consumer groups (≥20%) with significant mean drops (p
< 0.05) were identified for the tested product. One felt the product was ‘not sweet enough’ (32.2% of respondents), and the other that product had ‘too weak apple odor’ (41.7%). Average hedonic scores for the latter group were 4.3 ± 1.9 for ‘odor acceptance’ and 5.9 ± 1.5 for ‘overall acceptance’, i.e., more or less within the range of ‘neither like nor dislike’ category. The ‘apple-odor’ feature could be enhanced and intensified by proper selection of apple cultivars to be used in the production of this kind of product [65
]. This property can also be affected by the degree of ripeness of the used fruits [66
Looking at the total number of respondents, ‘overall acceptance’, ‘texture acceptance’ and ‘flavor acceptance’ were scored with the average hedonic scores above 6 (6.2 ± 1.8), indicating that the tested respondents liked the product.