Hybrid Drying of Apples: A Comparison of Continuous and Intermittent Process Modes

Abstract

In recent years, microwave and ultrasound technology has been under extensive development in drying technologies. Researchers are constantly searching for improved solutions or alternatives to hot air drying. The goal of this work was to determine the intermittent action of ultrasound and microwaves on convective drying. An examination of five specific cases of stationary and nonstationary drying processes was conducted. The evolution of moisture content and drying rate over process time was discussed, and the average drying rate and time, drying constant, effective diffusion coefficient, and specific energy consumption were also compared. To identify the differences between the dried products, the quality characteristics such as: water activity, color, shrinkage, rehydration, polyphenol content, odor, and flavor of apples were analyzed. The results indicate that intermittent drying provides a good alternative to convective drying, including when combined with microwave and ultrasound treatments. Applying microwaves or ultrasound intermittently resulted in an increase in the effective diffusion coefficient (by 68%) and drying rate (by 117%) and a reduction in drying time (by 53%), compared to convective drying. This processing method resulted in lower energy consumption by up to 13% and well-preserved quality attributes—this could be very promising for the production of healthy, ready-to-eat apple snacks.

1. Introduction

One of the easiest and most efficient ways to preserve agricultural products and extend their shelf life is drying. These materials have a high moisture content and water activity in their natural state, which makes them vulnerable to undesirable chemical, biochemical, and enzymatic reactions. Examples of such reactions include lipid oxidation, enzymatic browning, non-enzymatic browning, hydrolytic reactions, and enzyme reactions. These processes accelerate deterioration and diminish product quality [1]. To prevent or reduce such negative effects, the necessary steps should be taken, i.e., proper storage conditions followed by water removal under elevated temperature by drying operation.

Although air drying is readily accessible and relatively simple to utilize, it unfortunately requires a significant amount of energy and time. Products obtained by this technique are often characterized by a lower overall quality as compared to advanced drying technologies. One of the solutions used to accelerate the air drying can be a combination of different methods, such as microwave radiation or acoustic waves in microwave- or ultrasound-assisted air drying, also known as a hybrid technique. Numerous studies have confirmed that this drying method increases the efficiency and efficacy of the process. As a result, energy consumption can be reduced while preserving the quality of the product [2,3].

Microwaves penetrate dielectric materials, generating heat internally rather than just at the surface. Water molecules, being dipolar, oscillate rapidly in response to the alternating electromagnetic field. This oscillation and ionic polarization increase molecular friction, converting kinetic energy into thermal energy, which enhances internal heating. An increased vapor pressure within the product “pumps” moisture from the interior to the surface (higher effective diffusion coefficient), accelerating the drying rate and reducing the overall processing time [3].

Zahoor and Khan [4] showed that microwave assistance in the air drying made the process rate greater, shortened the process duration, and retained the vitamin A and color of bitter gourds, due to more rapid and uniform moisture transfer. In the work of Man et al. [5], the intermittent application of microwaves in air drying of apples contributed to a significantly shorter drying time (up to 81%), and improved the drying uniformity by about 30%, in comparison with air drying. The amount of energy used in the operation (specific energy consumption) typically decreases when drying time is substantially reduced [6]. Unfortunately, the quality of the final product may be negatively impacted by microwaves, depending on the type of food being processed, the processing parameters, and other variables. Most often, however, the adverse effects, such as vitamin loss, change in color, flavor, and aroma, nutrient and antioxidant loss, structural and textural changes, are caused by improperly selected power or radiation time [7].

The use of airborne ultrasound (US) technology also accelerates the drying process. The following ultrasonic application techniques are frequently discussed in the literature reports: US-assisted air drying and US pre-treatment. Ultrasound can trigger numerous phenomena, including mechanical, thermal, diffusive, and chemical processes. The effects of airborne ultrasound in convective drying are mostly associated with its mechanical effect. This impact serves to reduce internal and/or external mass transfer resistance without generating excessive thermal energy during the process. The ultrasound may generate pressure fluctuations, oscillations in the air velocity, and microstreaming. Such mechanisms have been demonstrated to enhance moisture evaporation, reduce the boundary layer, improve mass transfer, and stimulate diffusion. When an ultrasound passes through water within a solid matrix, then may induce acoustic cavitation, which facilitates the release of tightly bound moisture. Thus, the integration of airborne ultrasound has the potential to enhance the efficiency of air drying [8].

There is a lot of data concerning the subject of ultrasonic technology solutions in the drying of fruits and vegetables. As reported by Huang et al. [9], ultrasound typically has a positive impact on the dehydration process. However, the extent of improvement depends on various factors, including material properties and process variables such as temperature, air velocity, and ultrasound power. For example, the research of Magalhães et al. [10] indicated a higher water effective diffusivity (up to 93%), and mass transport (up to 30%), and even a 58% reduction in drying duration of apple cubes than in air drying.

Considering the way of microwave and ultrasound supply in air drying, usually a continuous mode is investigated. However, intermittent mode may promote the drying process effectiveness and has the potential to improve the quality of dry products. Intermittent drying is an advanced technique that involves alternating periods of active drying with periods of tempering (rest), where the drying process is paused or slowed [11]. The temperature, airflow, humidity, energy input, or pressure, are thus periodically changed. The presence of the cycles: an active drying phase, followed by a tempering phase, allowing moisture to redistribute within the food, which can enhance subsequent drying rates. Such an approach is designed to optimize both product quality and energy efficiency [12].

Intermittent drying can be applied using various energy sources: hot air, microwave, vacuum, infrared, convection, and radiation. For instance, microwave-intermittent drying is particularly effective for preventing overheating and ensuring uniform moisture distribution, especially in fruits like bananas and apples. However, the optimal intermittent drying scheme (timing, temperature, energy input) depends on the specific food product and desired quality outcomes [12].

Chua et al. [13] described a general classification for the methods used in intermittent drying and also presented an interesting scheme for the types of intermittent drying modes for various dryers. It was realized that IT drying reduced energy consumption and improved the quality of thermolabile products when compared to continuous drying, but, on the other hand, that it may prolong the overall drying time. The latter was probably the reason for the lower interest in intermittent drying [14].

In recent times, microwave- and/or ultrasound-assisted convective drying has been extensively studied for different food products, for example, carrot [15], zucchini [16], kale [17], strawberry [18], raspberry [19], apple [10], sour cherry [20], nut [21], and mushroom [22]. While the utilization of microwaves in the drying of heat-sensitive food products has been widely studied and adopted by agri-food industry sectors worldwide, the use of ultrasonic waves is still being tested in laboratories. Moreover, there is little available literature on convective drying using periodic phases (intermittent drying mode) in which microwaves and/or ultrasound are applied.

In view of these findings, there is a need to explore the option of using intermittent MW and/or US applications in hot air drying, which could be another attractive solution for highly perishable products. Many different types of bacteria can grow on raw, unprocessed foods; therefore, food must be processed to prevent spoilage [23]. One product that belongs to this group is apple. The fruit is an essential element of a diet due to high content of important minerals, fiber, vitamins, carotenoids, and antioxidative compounds. The good flavor and odor of apples make them an attractive raw snack, available year-round. The annual production of apple was approximately 84.32 million metric tons in 2023/24. China was the largest producer (48 million metric tons), followed by the EU (11.01 million metric tons) and the US (4.89 million metric tons). Considering the 10-year average of 79.15 million metric tons from 2014 to 2024, it can be assumed that production of these fruits is increasing [24].

The purpose of the experiments presented in this paper was to establish the microwave and ultrasound action on the dehydration of apple fruit. The hybrid processes were analyzed under two conditions: continuous (constant drying conditions during drying) and non-stationary (variable drying conditions). The evaluation of the processes included kinetic analysis, energy intensity assessment, and a quality analysis of the obtained products.

2. Methodology of the Study

2.1. Experimental Material

The study utilized apples (Malus domestica) of the ‘Ligol’ cultivar as the experimental material. The fruits were kept in the refrigerator after purchase. Consistently sized and mature fruits were taken out of the refrigerator and allowed to reach room temperature one hour before drying. Then, the apples were cleaned, peeled, and cut into 15 mm cubes using an RG-100 slicer (Hällde, Kista, Sweden). Each drying trial was conducted with 30 uniformly shaped cubes, corresponding to approximately 60 g of raw material. The initial moisture content (MC₀) averaged 0.8955 kg/kg (wet basis) and was determined using an XM120 moisture analyzer (Precisa, Dietikon, Switzerland) with an accuracy of 0.0001 kg/kg. Drying was carried out until the moisture content reached 0.05 kg/kg (wet basis). This final moisture level was established based on preliminary experiments to ensure sufficient shelf life of the dried apple.

2.2. Drying Processes

The drying operations were conducted in the laboratory-scale hybrid dryer (PROMIS-TECH, Wrocław, Poland), illustrated in Figure 1.

The dryer is able to perform the drying process in two different ways: through forced convection (CV) or through microwaves (MW). The air heating system consists of a fan (Figure 1 (1)) and a heater (3). Microwave energy is supplied by a magnetron (10) operating at a frequency of 2.45 GHz, with adjustable power in the range of 100–500 W. Additionally, the drying process could be intensified through the application of airborne ultrasound from AUS (PUSONICS, Madrid, Spain). The system comprises a generator and preamplifier (9), and a transducer (4) that is positioned directly above the perforated pan (7) that holds the material (6). Throughout the entire drying process, the samples remained static on the rotating pan. The ultrasound frequency generated is approximately 21 kHz, the intensity is approximately 160 dB, and the amount of power consumed by the power supply is 200 W. To prevent the material from overheating and the formation of hot spots, the pan is rotated at a constant speed of 10 RPM using a drive mechanism. The drying process is automated and regulated by a controller (11), which is programmed via a personal computer (12). The mass and the temperature of the samples are measured automatically at specified time intervals (every 300 s) using a balance (8), model PS 6000.R2 with a precision of 0.01 g (RADWAG, Radom, Poland), and a pyrometer (5), model CT LT15 with a precision of 0.1 °C (OPTRIS, Berlin, Germany), respectively. The velocity, temperature, and relative humidity of the air are measured continuously during the process using sensors (2), models HD29371TC1.5 and HD4817ETC1.5, with a precision of 0.01 m/s, 0.1 °C, 0.01%, respectively (DLETAOHM, Selvazzano Dentro, Italy), and are located at the inlet and outlet ducts. All data is recorded and stored on the PC (12).

Five drying operations were studied experimentally. The baseline program utilized forced convection (CV) at a temperature of 50 °C and an airflow velocity of 2 m/s. Subsequently, two stationary hybrid programs were examined, wherein the base program was enhanced with microwaves at 100 W (CVMW) or ultrasound at 200 W (CVUS). The final two drying programs were intermittent (non-stationary) schemes (IT1, IT2) in which the base program was amplified with microwave pulses of variable power. The structure of these intermittent programs is illustrated in Figure 2.

The results presented in this paper are a continuation of research on hybrid apple drying processes that was presented, among others, in the work of Musielak et al. [25]. This study analyzed the hybrid drying of apples (cv. Ligol) under forced convection conditions (air temperature 70 °C; flow rate 2 m/s) with microwave assistance at 100 and 200 W, as well as ultrasonic assistance at 100 W. The drying experiments were conducted in a hybrid drum dryer. The process parameters were kept constant, and the material was agitated dynamically during operation. The analysis covered the process kinetics (including mathematical modeling using thin-layer models), energy consumption during drying, and polyphenolic compound content and profile (as a quality parameter).

2.3. Kinetics and Energy Intensity of Drying Processes

The moisture ratio (dimensionless moisture content) was defined according to the following equation [26]:

$$MR\left(t_i\right)={\displaystyle \frac{MC\left(t_i\right)-M{C}_{eq}}{M{C}_0-M{C}_{eq}}}$$

(1)

where MC (t_i) denotes the dry basis moisture content at the i-th time of the process, MC_eq is the equilibrium dry basis moisture content, and MC₀ is the initial dry basis moisture content.

Equation (1) is usually simplified as [27]:

$$MR\left(t_i\right)={\displaystyle \frac{MC\left(t_i\right)}{M{C}_0}}$$

(2)

The dry basis moisture content was calculated using the following equation:

$$MC\left(t_i\right)={\displaystyle \frac{m\left(t_i\right)-m_s}{m_s}}$$

(3)

where m (t_i) is the mass of sample at the i-th time of the process and m_s is the mass of the dry matter.

The instantaneous drying rate was calculated using the following equation:

$$DR\left(t_i\right)=-{\displaystyle \frac{m\left(t_i\right)-m\left(t_{i-1}\right)}{t_i-t_{i-1}}}$$

(4)

While the average drying rate was calculated using the following formula:

$$D{R}_a=-{\displaystyle \frac{\Delta m}{\Delta t}}={\displaystyle \frac{m_f-m_0}{DT}}$$

(5)

where m₀ is the initial mass of sample, m_f is the mass of sample at the end of drying, and DT is the drying time.

The drying constant (k) and the effective diffusion coefficient (D_eff) were calculated according to the methodology proposed by Lopez et al. [28], with slight modifications [17]. The equation of Fick’s second law of diffusion under the assumption of a constant diffusion coefficient is as follows [29]:

$${\displaystyle \frac{\partial MC}{\partial t}}=D_{eff}{\nabla }^{2}MC$$

(6)

For a one-dimensional problem (unidirectional diffusion of moisture in the x direction), this equation takes the form [30]:

$${\displaystyle \frac{\partial MC}{\partial t}}=D_{eff}{\displaystyle \frac{{\partial }^{2}MC}{\partial x^2}}$$

(7)

Two boundary conditions are used for an infinite slab [29,30]:

• It is assumed that moisture diffusion is symmetrical relative to the center of the slab, which means that the moisture concentration gradient at the center is zero:

$${\left.{\displaystyle \frac{\partial MC}{\partial x}}\right|}_{x=0}=0$$

(8)

• On the surface of the slab (half the thickness L), there is a constant moisture concentration value corresponding to the equilibrium between the moisture content of the material and the environment, known as equilibrium moisture content:

$${\left.MC\right|}_{x=L}=M{C}_{eq}$$

(9)

The solution of the diffusion Equation (7) and boundary conditions (8) and (9), as given by Sherwood [31] and Crank [32], is as follows:

$$MR={\displaystyle \frac{8}{{\pi }^2}}{\displaystyle \sum _{n=0}^{\infty }\frac{1}{\left(2n+1\right)}\cdot \exp \left(-\frac{{\left(2n+1\right)}^2\cdot {\pi }^2}{4}\cdot \frac{D_{eff}}{L^2}\cdot t\right)}$$

(10)

where L is slab half-thickness if the evaporation occurs on both sides of slab.

For long drying times, Equation (10) can be simplified to only the first term of a series, resulting in an exponential bi-parametric equation [28]:

$$MR=a\cdot \exp \left(-k\cdot t\right)$$

(11)

where a is the model constant (a = 8/π²) and k is the drying constant, defined as follows [28]:

$$k={\displaystyle \frac{{\pi }^2\cdot D_{eff}}{4\cdot L^2}}$$

(12)

The value of the drying rate, k, was determined by approximating the experimental data with the model (11). Next, the D_eff was determined using the rearranged Equation (12). The thickness of the sample, L, was measured by caliper.

The energy intensity of each process was assessed based on the direct electricity consumption of the entire drying system (EC), which was measured using a standard electricity meter with an accuracy of 0.01 kWh. Next, the specific energy consumption (SEC) was calculated by referring the EC to the mass of evaporated moisture (∆m) [33]:

$$SEC=3.6\cdot {\displaystyle \frac{EC}{\Delta m}}$$

(13)

where 3.6 is 1 kWh recalculated into MJ.

2.4. Quality of Products

Color measurements were performed using a CR400 colorimeter (KONICA MINOLTA, Tokyo, Japan) equipped with a CIE D65 illuminant and a CIE 2° standard observer. Mean values of the color components in CIELAB color space were determined and the color difference index (dE) was subsequently calculated according to the equation [34]:

$$dE=\sqrt{{\left(L_0^{*}-L_d^{*}\right)}^2+{\left(a_0^{*}-a_d^{*}\right)}^2+{\left(b_0^{*}-b_d^{*}\right)}^2}$$

(14)

where L* is lightness, a* is the green-red chromatic coordinate, b* is blue-yellow chromatic coordinate, indexes 0 and d denote raw and dried material, respectively.

Water activity (a_w) was determined using a LabMaster-aw analyzer (NOVASINA, Lachen, Switzerland) with a precision of 0.001, at 25 °C until a constant value was reached. The reading was considered constant if the measured value did not change by more than 0.001 units within 5 min. For the dried samples, water activity was assessed after a 24 h equilibration period in a desiccator (without desiccant) to allow for uniform moisture distribution within the material.

The volumetric shrinkage of the apple samples (VS) was determined by the Mayor and Sereno [35] liquid displacement method. The average volumetric shrinkage (VS) coefficient was calculated using the following equation:

$$VS={\displaystyle \frac{V_0-V_d}{V_0}}$$

(15)

where V₀ is the volume of raw sample and V_d is the volume of the dry sample. The sample volume (raw/dried) was determined by measuring the volume of toluene displaced after immersing the sample.

The rehydration of apple samples was carried out by dipping them in deionized water, and measuring the weight change with a precision of 0.001 g. The rehydration ratio (RR) was defined as follows [36]:

$$RR={\displaystyle \frac{m_r}{m_d}}$$

(16)

where m_d is the weight of dried sample and m_r is the weight of rehydrated sample.

The retention of polyphenols (RPF) was assessed using a Folin–Ciocalteu method [37]. The polyphenol content was expressed as gallic acid equivalents (mg GAE/kg dry matter) based on a calibration curve constructed with analytically pure gallic acid (AKTYN, Poznań, Poland). The retention of polyphenols (RPF) was calculated using the following relationship:

$$RPF={\displaystyle \frac{C_d}{C_0}}$$

(17)

where C₀ is the polyphenol content in raw samples and C_d is the polyphenol content in dry samples.

The sensory profile analysis was performed by a specially trained team of evaluators in accordance with ISO 8586:2023 [38]. The tasting team was composed of members who regularly perform sensory evaluations of food and underwent standardized training to ensure consistency in scoring. Descriptors (i.e., individual distinguishing features) of the product’s odor and flavor were assessed, as well as the overall quality of the odor and flavor profile (i.e., the overall sensory impression of how well all the taste and smell descriptors were harmonized). The list of descriptors was selected and established using the profiling analysis procedure in accordance with ISO 11035:1994 [39]. The intensity of each descriptor was quantified using a 10 cm graphic scale (linear unstructured scales were used) with appropriate boundary markings. The evaluators, taking into account the boundary terms, marked their assessment on this scale. The obtained results were then converted into numerical values. Values from 0 to 10 were adopted, where imperceptible/disharmonized was 0 (left edge of the scale), and very intense/highly harmonized was 10. The obtained individual results in the form of numerical data were used as the basis for calculating the mean values and standard deviation for each descriptor. The sensory profile analysis was carried out in the Department of Gastronomy Science and Functional Foods at Poznan University of Life Sciences, which complies with the requirements of the ISO 8589:2007 [40]. The assessments were performed in two repetitions.

2.5. Statistical Analysis

The results are presented as mean values ± standard deviations. One-way ANOVA, Tukey’s post hoc test, and Pearson’s correlation were performed for the statistical analysis. Differences indicated by different letters were deemed statistically significant at p < 0.05. All calculations were performed using Statistica software, version 12 (StatSoft, Tulsa, OK, USA). The adjusted coefficient of determination (adj. R²), the reduced chi-squared (red. Χ²), and the root mean squared error (RMSE) were used to assess the degree of fit between the model (11) and the experimental data [41]:

$$adj. R^2=1-{\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\left(M{R}_{\exp ,i}-M{R}_{pred,i}\right)}^2/\left(n-p\right)}}{{\displaystyle \sum _{i=1}^n{\left(M{R}_{\exp ,i}-M{R}_{\exp ,aver}\right)}^2/\left(n-1\right)}}}$$

(18)

$$red. {\chi }^2={\displaystyle \frac{{\displaystyle \sum _{i=1}^n{\left(M{R}_{\exp ,i}-M{R}_{pred,i}\right)}^2}}{n-p}}$$

(19)

$$RMSE=\sqrt{red. {\chi }^2}$$

(20)

where MR_exp,i and MR_pred,i are the i-th experimental and predicted with the model values of the dimensionless moisture content, respectively, MR_exp,aver is the average value of the experimental dimensionless moisture content, n is the total number of observations and p is the number of model parameters.

3. Results and Discussion

3.1. Drying Kinetics

Figure 3a illustrates the evolution of the averaged curves for dry basis moisture content (MC_db) over time for each process. The shape of the curves is very consistent across all drying procedures; however, the total duration of the processes differs significantly. The longest process was for forced convection (CV), which took approximately 7 h (25,200 s). In contrast, the shortest process was microwave-assisted convection (CVMW), which took only 1.5 h (5400 s). Ultrasound-assisted (CVUS) and the intermittent (IT1, IT2) processes exhibited comparable trends in the MC_db curves and had a total duration of approximately 3.2–3.5 h (11,580–12,600 s).

Figure 3b presents the drying rate curves as a function of time. The plotted curves indicate that the drying processes occurred entirely within a period of falling drying rate. During the initial phase of the process, the rate increased to a maximum value, after which it began to decrease—this decrease in the DR value was observed until the end of the drying process. In the intermittent processes, peaks in the drying rate resulting from the application of microwave pulses can be observed after 2 h (7200 s) of the process.

The graphs in Figure 4 illustrate the average values for the drying rates and times for each process. The convective-microwave process (CVMW) exhibited the highest average drying rate, while the convective process (CV) demonstrated the lowest drying rate (Figure 4a). The ultrasound-assisted convection (CVUS) and the intermittent (IT1, IT2) processes exhibited comparable values to the average drying rate. These findings are consistent with the average drying times. The shortest drying time was observed for the CVMW process, while the longest was observed for the CV process. The CVUS, IT1, and IT2 processes exhibited similar durations, with the ultrasound-assisted convective process demonstrating a slightly longer drying time than the intermittent processes (Figure 4b).

Figure 5 presents a comparison of the experimental (exp) and predicted (fit)—with the simple exponential model (11), dimensionless moisture content for each process. It can be seen that the model precisely predicts the experimental data. This finding is corroborated by the high values of the adjusted R-squared and the low values of the reduced chi-squared and root mean squared error presented in

Table 1. The mean values for model parameter, a, drying constant, k, effective diffusion coefficient, D_eff, reduced chi-squared, red. χ², adjusted R-squared, adj. R² and root mean square error, RMSE, determined for each particular process during approximation.

Process	a (SD) –	k × 104 (SD) 1/s	Deff × 109 (SD) m/s2	red. χ2	adj. R2	RMSE
CV	0.99 (0.00) c	1.61 (0.05) c	3.67 (0.11) c	4.73 × 10−5	0.9993	0.0069
CVUS	1.01 (0.01) b	2.65 (0.14) b	6.05 (0.33) b	6.78 × 10−5	0.9990	0.0081
CVMW	1.08 (0.01) a	4.81 (0.06) a	10.97 (0.13) a	1.32 × 10−3	0.9848	0.0363
IT1	1.03 (0.00) b	2.68 (0.06) b	6.11 (0.15) b	2.99 × 10−4	0.9948	0.0172
IT2	1.03 (0.01) b	2.75 (0.12) b	6.27 (0.27) b	4.45 × 10−4	0.9925	0.0208

Mean (standard deviation); means in the same column followed by the same letter are not significantly different according to Tukey’s HSD test (p < 0.05).

.

An analysis of the drying constant (k) and the effective diffusion coefficient (D_eff), given in Table 1, demonstrates that these parameters exhibit a high correlation with the drying rate (Figure 4a). The highest values of k and D_eff were observed in the microwave-assisted convective process (CVMW). The lowest values of the parameters were observed for the forced convection (CV). For the ultrasound-assisted convection (CVUS) and the intermittent processes (IT1, IT2), similar values of k and D_eff were noted (there were no statistically significant differences), which were higher than for CV but lower than for CVMW.

On the basis of Figure 3b, it was earlier stated that drying occurs during a period of falling drying rate for each of the analyzed schedules. During this period, the rate of the process is primarily determined by the diffusive movement of moisture within the material [26]. Therefore, it is evident that the effective diffusion coefficient presented in Table 1 is of significant importance in evaluating the kinetics. The lowest D_eff value for the forced convection (CV) is consistent with the lowest average drying rate (Figure 4a) and the longest drying time (Figure 4b).

The increase in the diffusion coefficient in the microwave-assisted forced convection (CVMW) could be attributed to the volumetric heating of the samples resulting from the interaction of electromagnetic waves with water molecules. Consequently, moisture evaporation within the material occurs, whereby the vapor phase diffuses more readily to the surface and subsequently to the core of the drying medium. The increase in pressure within the material, resulting from the formation of vapor, can facilitate the transfer of liquid moisture to the surface, thereby additionally enhancing the mass exchange. As a result, the CVMW process had the highest drying rate (Figure 4a) and shortest drying time (Figure 4b) of all the processes carried out.

The application of airborne ultrasound has also been demonstrated to intensify the heat and mass exchange. Kowalski et al. [42] identified two primary mechanisms responsible for this observed intensification: the heating effect and the vibration (mechanical) effect. It was also found that although the heating effect is more readily discernible, it typically plays a less significant role than the vibration effect. Nevertheless, both effects contribute to an increase in the average drying rate and a reduction in the duration of the process, as evidenced by the findings of the presented research (Figure 4).

The lack of discernible changes in the kinetic parameters of the intermittent programs (IT1, IT2) relative to CVUS may be attributed to the high structural similarity between these processes. Indeed, the intermittent processes constituted a substantial portion of the ultrasound-assisted forced convection, accounting for c.a. 60% of the total drying time. Only during the final stage of the CVUS process, after 2 h, was the drying process enhanced with the application of microwave impulses of variable power. The results indicate that the pulses had no discernible effect on the investigated kinetic parameters.

The obtained results are consistent with those reported in previous studies on the drying of apples in a rotary dryer [25] and with those presented by Şen and Aydin [43] for a tunnel dryer. Slight differences between the presented values may be attributed to the manner in which the drying process was conducted and the determination of the process parameters. Nonetheless, it was found that microwaves effectively influenced the diffusion coefficient, leading to a significant increase in the drying rate and a reduction in drying time. Ultrasound also positively affected all parameters, however, the advantages over convective drying are smaller compared to microwaves.

3.2. Energy Intensity of Drying

Figure 6 presents the average value for specific energy consumption (SEC) for each process. This value indicates the amount of energy required to evaporate a unit mass of moisture. A higher value indicates a more energy-intensive process. The results indicate that the least energy-intensive process was the microwave-assisted forced convection. The other processes showed a significantly higher SEC.

The energy intensity of the drying process is influenced by two key factors: the instantaneous energy consumption and the duration of the process. The severity of the process conditions and the complexity of the drying program are positively correlated with the value of the instantaneous energy consumption. This is because increasing the temperature and air velocity during convective drying, or using additional sources of energy such as ultrasound or microwaves, necessitates the use of more energy to power the system. If the implementation of more rigorous drying conditions or supplementary energy sources does not result in a notable reduction in drying time, it will lead to the process having an elevated energy intensity. This phenomenon was observed in the ultrasound-assisted forced convection (CVUS) and the intermittent processes (IT1, IT2), in which a reduction in drying time relative to forced convection (CV) (Figure 4b) did not result in a significant decrease in specific energy consumption (Figure 6). In contrast, microwave-assisted forced convection (CVMW) was characterized by a meaningful reduction in SEC. In this case, the substantial reduction in drying time (c.a. 77%) more than compensated for the higher instantaneous energy consumption resulting from the necessity to power the microwave generation system.

Similar outcomes were documented in a study on apple drying in a tumble dryer [25]. The application of ultrasound during the convective process exhibited a slight drop in the SEC value, whereas a notable reduction in this parameter was discerned in the microwave-assisted convection processes. Papers by Kowalski et al. [42] and by Kroehnke and Musielak [44] documented elevated energy consumption in ultrasound-assisted convection processes, substantiating the fact that this technique can negatively influence the energy intensity of the process.

3.3. Product Quality

The values of the analyzed quality parameters are summarized in

Table 2. Mean values for water activity, a_w, color difference index, dE, volumetric shrinkage, VS, rehydration ratio, RR, and retention of polyphenols, RPF, for each particular process.

Process	aw (SD) –	dE (SD) –	VS (SD) –	RR (SD) –	RPF (SD) –
CV	0.227 (0.035) a	14.28 (0.15) a	0.758 (0.023) ab	4.080 (0.017) b	0.79 (0.00) a
CVUS	0.273 (0.040) a	10.08 (1.19) b	0.763 (0.021) a	4.330 (0.151) ab	0.82 (0.01) a
CVMW	0.172 (0.008) b	11.56 (1.90) ab	0.746 (0.027) abc	4.083 (0.047) b	0.73 (0.03) b
IT1	0.251 (0.024) a	11.78 (0.07) ab	0.716 (0.017) bc	4.570 (0.155) a	0.84 (0.00) a
IT2	0.245 (0.015) a	10.96 (0.04) ab	0.713 (0.008) c	4.187 (0.083) b	0.80 (0.02) a

Mean (standard deviation); means in the same column followed by the same letter are not significantly different according to Tukey’s HSD test (p < 0.05).

.

One of the most crucial parameters concerning food products is water activity (a_w). This is a thermodynamic parameter that provides information on the moisture status of a given material. A higher value of this parameter indicates a higher amount of free water available in the material. Free water is an excellent environment for the growth of pathogenic microorganisms, such as bacteria, fungi, yeast and molds. Moreover, it can act as both an environment and a substrate for many adverse chemical reactions that can result in the spoilage of food products. The rate of progression for most adverse processes that lead to spoilage are notably inhibited when the a_w value is below 0.6 [45]. In the current study, water activity values for all dried products were found to be below the critical value, thus ensuring the requisite safety. Furthermore, no discernible differences were observed for a_w values across the various processes. However, in the case of microwave-assisted forced convection (CVMW), a statistically significant lower value of water activity was recorded. A similar effect was observed by Kroehnke et al. [46] for carrot and by Nawirska-Olszańska et al. [47] for golden berry. In both cases the application of microwaves resulted in a lower water activity value for the final products in comparison to the convective process. Pham and Karim [48] observed that an increase in microwave power also resulted in a lower water activity value for dried papaya. This may have resulted from the effective interaction between the electromagnetic waves and water molecules. Microwave heating has a significant impact on the rate of bound water removal, which is considered a rate-limiting factor in conventional hot-air drying methods. As a result, moisture content is reduced and the product’s shelf life is increased while maintaining the desired quality characteristics [48].

Another important parameter is the color of the final product. The color difference indexes (dE) presented in Table 2 indicate the range of difference between the raw and the dried samples, in three-dimensional CIELAB color space. An increased dE value indicates a greater color difference. It is postulated that a normal observer can discern differences between colors when dE exceeds 4. The dE values obtained indicate a significant color change for all of the processes investigated. Additionally, a notable distinction was observed between the various processes. The highest color difference (corresponding to the highest dE value) was observed for the forced convection (CV), while the lowest was observed for the forced convection assisted by ultrasound (CVUS). In the convective-microwave (CVMW) and intermittent (IT1, IT2) processes, the dE value was of a similar value, falling between those observed for CV and CVUS. Kroehnke and Musielak [44] also observed a reduction in color difference for ultrasound-assisted convection. Matys et al. [49] stated that they observed a positive correlation between the power of the microwaves being applied during drying and the color difference index (dE) of the obtained samples. The higher the MW power, the greater the value of dE, indicating more pronounced changes.

In addition to the color changes, the shape and size of the material also undergoes alterations during the drying process. These changes can be advantageous in certain cases, for example, a reduction in weight and size can result in lower costs associated with packaging, handling, and transportation [50]. However, if the changes are too extensive, they can also result in a loss of product value. As illustrated in Table 2, the volumetric shrinkage (VS) demonstrates that the samples subjected to the disparate drying techniques exhibited comparable volume reductions, within the range of 71% to 76%. However, it is notable that the processes conducted under intermittent conditions (IT1, IT2) exhibited slightly less shrinkage (VS = 71.60% and 71.26% for IT1 and IT2, respectively) compared to those conducted under continuous conditions (CV, CVMW, CVUS). Such a result could be due to gradual moisture transport from the internal core of the apple to the surface during MW and US intermittent drying, and could also follow from the so-called “puffing effect” [51]. Lower values of VS mean the dry apple sample had a higher total volume, and thus less shrinkage after dehydration. It must also be emphasized that the best shape retention (i.e., the original shape of a cube) after processing was observed for the IT samples.

Furthermore, the drying process results in alterations to the internal structure of the product. These effects can be beneficial (e.g., an increased porosity leading to accelerated diffusion), neutral, or detrimental (e.g., destruction of the internal structure, leading to cell collapse) [36]. The rehydration ratio (RR) can be used as an indirect measure of changes to the internal structure. In samples in which cellular structures have been destroyed, water absorption is significantly hindered during rehydration, resulting in a lower rehydration ratio compared to samples with preserved cellular structure [36]. The data presented in Table 2 demonstrate comparable structural alterations for the examined samples. The weight gain resulting from water absorption (equal to RR-1) ranged from 308% to 357%. CV, CVMW, and IT2 were characterized by less absorption of water, c.a. 308% for CV and CVMW, and 318% for IT2. The ultrasound-assisted forced convection (CVUS) and intermittent process (IT1) exhibited slightly elevated RR-1 values (333% and 357% for CVUS and IT1, respectively). The reduced RR values may be attributed to structural damage and cellular shrinkage resulting from the continuous application of microwaves (CVMW), as well as prolonged convective drying (CV). Furthermore, these findings are consistent with those of Tepe and Tepe [52] for convective and microwave drying, and İzli & Polat [53] for intermittent microwave drying, who also observed a higher rehydration ratio for dried apple slices at lower or intermittent microwave power compared to hot air drying.

In evaluating the drying of food products, as well as other quality parameters it is equally important to assess the retention of nutritional and functional compounds. Elevated temperatures, atmospheric oxygen, moisture loss, light, the presence of enzymes and catalyzers, and the pH of the environment, among other factors, can lead to the degradation of important compounds. The preservation of polyphenolic compounds in dried fruits is of significant nutritional importance, as these bioactive constituents contribute to antioxidant capacity and overall health benefits. They contribute, for example, to a reduction in the risk of cardiovascular disease and cancer [54]. According to Pandey and Rizvi [55], epidemiological evidence suggests that long-term consumption of polyphenol-rich diets provides protective effects against diabetes, osteoporosis, and neurodegenerative disorders.

The values for the polyphenol retention coefficient (RPF) presented in Table 2 leads to the conclusion that a significant amount of this compound was retained (with reference to the raw material), with an average level of 79% (ranging from 73% to 83%). The lower values of RPF for CVMW and CV samples are due to the high temperature and longer drying time at which these antioxidant compounds unfortunately degrade [56]. The highest polyphenol retention in the dried apples was observed for the IT1 drying program. One can see that intermittent MW and airborne-US assistance in CV drying allows for the preservation of large quantities of these valuable ingredients. Kroehnke et al. [46] had similar findings for convective drying of carrots with microwave and ultrasound assistance. In that investigation, microwave activity in CVMW processes caused significant damage to cells and cell structures, culminating in the oxidation of phenolic molecules and higher polyphenol losses in the samples. In turn, applying US at 200 W in CVUS enhanced the level of phenolic compounds when compared to CV and CVMW. When MW and US were used at the same time (CVMWUS), the total polyphenol concentration was higher than in the CV; however, the level of change was highly dependent on the drying conditions.

In the case of food, the sensory characteristics of the products have a direct influence on consumer judgment and decision. In addition to appearance, odor and flavor are the most important parameters evaluated during this assessment [57]. Therefore, all products were evaluated by a sensory panel and the scores for the odor and flavor descriptors of apples dried using the different processes were determined. Then, radar charts were developed to characterize each apple sample, as shown in Figure 7.

The results show that the odor characteristics of the dried apple have distinct differences (Figure 7a). According to the assessment, the highest intensity of apple and sweet odors was found for CV and IT2. A lower intensity of odors characteristic of apple fruit (i.e., sweet and apple) was found for CVMW, CVUS, and IT1. Wang et al. [58] reported a similar phenomenon concerning maintaining the original aroma of raw apple for microwave-vacuum dried (MVD) apple slices with ultrasound treatment. The overall acceptability (or quality) of the dried apple products is similar to the ratings presented in this study. The highest intensity of unpleasant odors such as sour and musty were detected in the CV and CVUS samples, respectively.

The results demonstrated that apples dried using CVUS sample differed to the greatest extent in terms of both flavor and aroma profiles when compared to the reference CV sample: its flavor and aroma profiles were poorly harmonized (the mean scores were 5.0 ± 0.5 and 9.0 ± 0.1 for CVUS and CV samples on a 10-point scale showing the overall impression of how well the taste and aroma were harmonized). In this sample, the lowest intensity of sweet and apple taste was found, and the highest intensity of strange and bitter flavor was detected, which were not present in the CV sample (Figure 7b). The CVMW samples’ flavor and aroma profile was well harmonized (7.1 ± 0.2). CVMW exhibited moderate sweet, apple, and sour tastes; a bitter taste was detected, but with a lower intensity compared to CVUS, and no strange taste was noted. For the IT1 and IT2 samples, the flavor and aroma profiles were also well harmonized (7.1 ± 1.1 and 7.6 ± 0.6, respectively). A rather higher intensity of sweet and apple flavor was discerned, and in the IT2 sample, no bitter or strange taste was detected. Similar observations in the sensory evaluation of apple chip snacks were found by Zhu et al. [59], where sensory characteristics such as the aroma profile differed across the drying methods (including hot air drying—HAD), but the taste profile was very comparable between the apple samples.

Based on the analysis of the sensory profile of dried apples using different processes, it was concluded that intermittent drying positively affects the taste and aroma profile of this product, as the sensory profiles of taste and aroma for IT2 and IT1 samples showed a greater harmony compared to those obtained using the CVUS method. Good harmonization of these product characteristics may attract consumer attention and influence their preferences. However, this requires further research. According to Petikirige et al. [60], all sensory characteristics undergo modification during the drying process, with the extent of change influenced by factors such as drying method, temperature, airflow rate, energy input, fruit composition, cultivar, sample thickness, and applied pre-treatments.

4. Conclusions

The findings of this study indicate that the application of microwaves and ultrasound affects both the kinetics and energy consumption of drying processes, as well as the quality of the resultant products.

It was found that the drying of raw apple proceeded completely within the falling drying-rate period, in which the rate of the process is mainly controlled by the internal transport of moisture (inside the material). A comparison of the results to the reference process, forced convection (CV), revealed that microwave enhancement (CVMW) led to a substantial increase in the average drying rate and effective diffusion coefficient, consequently reducing the total drying time. Ultrasound (CVUS) also had a positive effect on the evaluated kinetic parameters, but the level of change was significantly lower than that observed for microwaves. Intermittent processes (IT1 and IT2) were characterized by values of kinetic parameters similar to those observed for the ultrasound-assisted convective process.

It was shown that the microwave assistance (CVMW) contributed to the greatest reduction in specific energy consumption, which resulted from meaningful reduction in drying time. Ultrasound-assisted (CVUS) or intermittent (IT1 and IT2) processes were characterized by a visibly smaller reduction in energy intensity, which correlates with less pronounced effects of drying conditions on kinetic parameters in these processes compared to CV.

Research into product quality revealed that microwaves had the greatest impact on the assessed parameters. In the CVMW process, microwaves positively affected water activity and color, had no effect on shrinkage and rehydration, and negatively affected polyphenol content. Ultrasound (CVUS) had a smaller impact on product quality and was found to have a positive effect on color, polyphenol content and rehydration; a neutral effect on shrinkage; and a negative effect on water activity. The processes carried out under intermittent conditions (IT) had a positive effect on all the parameters evaluated. Furthermore, well-harmonized flavor and aroma profiles for IT-dried samples proved that the impact of intermittency on the drying of apples was favorable. Thus, intermittent mode appears to be the most successful and sustainable in view of the apparent positive effects of the drying conditions on the kinetics and energy intensity, as well as the product quality, which was well-preserved.