Impact of Ultrasound Pretreatment and Temperature on Drying Kinetics and Quality Characteristics of Blood Orange Slices: Comparison with Different Drying Methods

Abstract

This study aimed to investigate the impact of ultrasonic pretreatment vacuum drying (UAVD) and temperature on drying kinetics and qualitative attributes of blood oranges in comparison to several drying methods: hot air drying (HAD), vacuum drying (VD), and freeze drying (FD). The drying kinetics and modeling, total phenolic content (TPC), antioxidant capability (assessed using DPPH and ABTS tests), individual phenolic profiles, vitamin C concentration, and color factors were meticulously examined. Drying times were recorded as 22.5 h, 12.5 h, and 9 h for HAD; 11.5 h, 9.5 h, and 8.5 h for VD; and 10 h, 8.5 h, and 7.5 h for UAVD at 50, 60, and 70 °C, respectively. The HAD, VD, and UAVD procedures were conducted at 50, 60, and 70 °C, resulting in reduced drying periods with increasing temperature. The integration of ultrasound markedly lowered drying durations. Eleven thin-layer drying models were utilized to recreate the drying process precisely. The D_eff values of the HAD, VD, and UAVD dried samples varied from 9.08 × 10⁻⁶ to 2.82 × 10⁻⁵ m²/s, from 2.60 × 10⁻⁵ to 2.96 × 10⁻⁵ m²/s, and from 2.20 × 10⁻⁵ to 2.99 × 10⁻⁵, respectively. Among the desiccated blood orange slices, the greatest total phenolic content (TPC) was observed in freeze-dried samples (131.27 mg GAE/100 g), followed by those dried using ultrasonic-assisted vacuum drying (UAVD) at 50 °C (128.77 mg GAE/g DM). Dried blood orange slices had a vitamin C content of 29.79 to 49.01 mg/100. The drying process substantially impacted the color parameters L*, a*, and b*. These findings highlight the efficacy of ultrasound-assisted drying in decreasing drying duration while improving the retention of bioactive components in blood orange slices.

1. Introduction

Citrus fruits are commonly consumed around the world, primarily owing to their substantial nutritional value and related health benefits [1]. They constitute a significant portion of the world’s fruit tree crops. During the 2022/2023 marketing year, oranges accounted for 47% of global citrus production, totaling 48 million tons out of 100 million tons. In Turkey, citrus fruit production in 2023 reached 7.9 million tons, including 2.3 million tons of oranges cultivated across 495 thousand hectares. The principal citrus-producing regions in Turkey are Antalya, Adana, Muğla, and Hatay [2].

The blood orange (Citrus sinensis (L.) Osbeck), commonly called pigmented or red-orange, comprises three primary cultivars: Moro, Tarocco, and Sanguinello. Unlike other orange varieties, the defining characteristic of blood oranges is the existence of red pigments in the flesh and, in some cases, the peel. These pigments, known as anthocyanins, develop in response to low night temperatures. As a result, blood oranges are predominantly cultivated in Mediterranean regions with subtropical climates characterized by high daytime temperatures and low nighttime temperatures [3,4]. Additionally, blood oranges are recognized for their elevated levels of total phenolics, anthocyanins, and flavones, which contribute to their superior antioxidant properties [5]. A significant quality component is anthocyanin content, which is ascribed to the antioxidant activity of these compounds. The prevention of oxidation and free-radical chain reactions is the mechanism by which the antioxidant activity of anthocyanin in blood orange fruits benefits human health, with an impact on certain diseases, such as anti-inflammatory, anticancer, and antidiabetic [6].

A prevalent method for food preservation is drying, which provides benefits including prolonged shelf life, diminished packing and shipping expenses, and reduced weight and volume, thereby enhancing logistics. Additionally, drying enhances the retention of nutritional quality while lowering moisture content to levels that inhibit microbial growth [7,8]. Traditional hot-air convection drying is commonly utilized in the food industry because it is inexpensive and simple to implement. However, hot air drying has notable drawbacks, including adverse effects on food quality, environmental sustainability, and nutrient retention. Prolonged exposure to high temperatures during drying can significantly deteriorate bioactive chemicals and compromise their sensory attributes [9]. Novel drying methods that maximize energy economy while maintaining the quality characteristics of food items are therefore becoming more and more necessary [10]. Among modern drying methods, freeze-drying is regarded as the most efficient approach to obtaining good-quality dried fruits and vegetables, as it preserves structural integrity and nutritional composition better than conventional methods. However, despite its advantages, freeze-drying is characterized by prolonged processing times and high operational costs, limiting its industrial application. As an alternative, VD has gained prominence for its ability to achieve faster drying rates at lower temperatures while operating in low-oxygen environments, thereby minimizing oxidative degradation of bioactive compounds [11]. Vacuum drying entails accelerated surface evaporation by reducing the boiling point of water in a vacuum environment. The resistance to mass transfer is diminished at the surface of a food, resulting in an increased drying rate. The drying rate diminishes over extended drying periods due to a decrease in mass transfer from the interior to the surface of the food. Enhancements in the drying rate necessitate a technique that will augment mass transfer [12].

Advancements in drying technology have led to the evolution of integrated drying techniques that combine multiple methods to enhance efficiency, reduce costs, and improve sustainability. One such emerging approach is UAVD, which has gained attention for its potential to significantly accelerate drying while improving efficiency. This technique enhances the dehydration process by increasing the moisture transport rate without excessive thermal exposure. Water removal is accelerated by the cavitation effects produced by ultrasound, which help to create microchannels inside the food matrix. Furthermore, ultrasound promotes the extraction of tightly bound moisture by inducing localized pressure changes, all while maintaining relatively low temperatures, making it particularly beneficial for heat-sensitive food products [12]. A wide variety of food products, including Asian pear [13], beef and chicken meats [14], carrot slices [15], green beans [12], nectarines [16], raspberry fruit [17], red peppers [18], salmon and trout fillets [19], Schisandra chinensis extract powder [20], papaya [21], and Flos Sophorae Immaturus [22], were effectively dried by UAVD. The increasing adoption of UAVD underscores its significance as a modern dehydration technique that effectively addresses the limitations of conventional drying methods while ensuring superior preservation of food quality and nutritional integrity.

The most important difference between blood orange fruit and other oranges is that the anthocyanin pigment is synthesized from the flesh of the fruit and sometimes from the peel [23]. Therefore, in this study, blood oranges were dried with their peels. No study has been reported in the literature on the zero-waste recycling of blood orange (Citrus sinensis [L.] Osbeck) slices with their peels and the drying of them into thin layers by ultrasonic pretreatment vacuum drying, to the best of our knowledge. This study involved the drying of blood orange slices using HAD, VD, UAVD (at 50, 60, and 70 °C), and FD methods. Several mathematical drying models were employed to examine drying kinetics and delineate the moisture removal behavior with time. Furthermore, the concentrations of TPC, DPPH, and ABTS, together with individual phenolic profiles, vitamin C levels, and color characteristics, were evaluated to ascertain the influence of various drying processes on the qualitative attributes of blood orange slices under varied temperature conditions.

2. Materials and Methods

2.1. Material

Blood oranges (Citrus sinensis (L.) Osbeck, cv. Sanguinello) were purchased in 2023 from an organic citrus farm located in Arsuz, Hatay, Turkey. Following procurement, the fresh oranges were transported to the Food Chemistry Laboratory at Yildiz Technical University and kept at 4 °C till further processing. The initial percentage moisture of the fresh blood oranges was measured to be 83.10 ± 0.26%, utilizing an infrared moisture analyzer (Rad-wag, MA 50-R, Radom, Poland). Prior to drying, the orange samples were thoroughly washed, wiped clean, and sliced into uniform 5 mm thick sections. The dried slices were subsequently stored in a desiccator to preserve their structural integrity until further analysis.

In this study, all the chemicals used for TPC, antioxidant capacity analyses (DPPH and ABTS), individual phenolic compounds, and vitamin C content were procured by Sigma-Aldrich (St. Louis, MO, USA) and Merck (Darmstadt, Germany).

2.2. Methods

2.2.1. Drying Procedure

The drying of blood orange slices was conducted utilizing HAD, VD, and UAVD at 50, 60, and 70 °C, and FD. During HAD, VD, and UAVD, the weight losses of orange slices were monitored at 30 min intervals. The blood orange slices were subjected to drying until the moisture content attained 0.1 kg of water per kg of dry matter.

The HAD procedure was performed using a Testo 440 vane probe anemometer (Lutron, AM-4201, Lutron, Taipei, Taiwan) at a constant air velocity of 1.3 m/s. Horizontal airflow was applied over the surface of the slices throughout the drying process [13].

A vacuum drier (Daihan WOV-30, Gangwondo, Republic of Korea) was employed for VD. A vacuum with an ultimate pressure of 60 mbar and a pump speed of 2 L/s was sustained using a vacuum pump (EVP 2XZ-2C, Zhejiang, China) [13].

In the UAVD technique, blood orange slices were subjected to a 30 min ultrasonic water bath (Daihan, WUCD10H, Seoul, Republic of Korea) running at 100% amplitude, with a power intensity of roughly 1 W/cm², a frequency of 40 kHz, and a capacity of 10 L. Subsequently, blood orange slices were dehydrated using a vacuum dryer (DaihanWOV-30, Seoul, Republic of Korea) [13].

FD was performed by utilizing a defined protocol on a laboratory freeze drier (Martin Christ, Beta 1–8 LSC plus, Osterode am Harz, Germany). The samples were cryopreserved at −80 °C and processed over a period of 72 h [13].

2.2.2. Mathematical Modeling

The moisture ratio (MR) was expressed as (M_t − M_e)/(M₀ − M_e), where M_e is the equilibrium moisture content, M_t is the moisture content at time t, and M₀ is the initial moisture content.

The drying rate (DR) curves were expressed as the weight loss per unit of time (g moisture/30 min). The drying rate (DR) of blood orange slices was determined by employing Equation (1):

$$\mathrm{D}\mathrm{R}={\displaystyle \frac{∆\mathrm{M}}{∆\mathrm{t}}}={\displaystyle \frac{{\mathrm{M}}_{\mathrm{t}+∆\mathrm{t}}-\mathrm{M}}{∆\mathrm{t}}}$$

(1)

where ΔM is the weight change, Δt is the time change, M_t and M_t+Δt are the weight at time t and at time t + Δt, respectively. In this study, the weight change was measured every 30 min, so Δt is 30 min.

The drying data collected during the experiments were analyzed by fitting them into eleven independent thin-layer drying models.

Table 1. Thin-layer drying models applied to the blood orange slice drying curves.

Model	Mathematical Equation	Reference
Demir	y = a × exp(−k × x)n + c	[24]
Henderson and Pabis	y = a × exp(−k × x)	[25]
Lewis	y = exp(−k × x)	[26]
Logarithmic	y = a × exp(−k × x) + c	[27]
Midilli and Kucuk	y = a × exp(−k × (xn)) + b × t	[28]
Modified Page	y = exp(−((k × x)n))	[29]
Page	y = exp(−k × (xn))	[30]
Parabolic	y = a + b × x + c × x2	[31]
Two-term exp	y = a × exp(−k × x) + (1 − a)*exp(−k × a × x)	[32]
Wang and Singh	y = 1 + a × x + b × x2	[33]
Weibull	y = exp(−(xa/ba))	[34]

provides a comprehensive description of these models, which were assessed using nonlinear least squares regression analysis.

The model parameters and R² values were ascertained by nonlinear regression analysis performed with the STATISTICA 13.0 software (StatSoft, Tulsa, OK, USA). The acceptability of each model was evaluated using R² and root mean square error (RMSE) values. A higher R² value and a lower RMSE value indicated a well-fitting model. RMSE values were computed by employing Equation (2) [35]:

$$\mathrm{R}\mathrm{M}\mathrm{S}\mathrm{E}={\left[{\displaystyle \frac{1}{N}}\sum _{i=1}^N{\left({MR}_{pre,i}-{MR}_{exp,i}\right)}^2\right]}^{1/2}$$

(2)

$${\mathrm{X}}^2={\displaystyle \frac{\sum _{i-1}^N{({MR}_{pre,i}-{MR}_{exp,i})}^2}{N-m}}$$

(3)

MR_prei represents the predicted moisture ratio, while MR_expi denotes the experimentally determined moisture ratio. N is the total count of observations, whereas m represents the quantity of constants used in the model.

The effective moisture diffusivity (D_eff) of the orange slices was calculated based on Fick’s second law of diffusion (Equation (4)):

$${\displaystyle \frac{M}{t}}=\nabla \left[D_{eff}(\nabla M)\right]$$

(4)

This equation was adapted to accommodate slab geometry and unsteady diffusion conditions, resulting in the following form:

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

(5)

In Equation (5), D_eff represents the effective moisture diffusivity (m²/s), L is the thickness of the orange slice measured at regular intervals (m), and n is the constant derived from the thin-layer drying models. For extended drying periods, this equation can be simplified into the form presented in Equation (6).

$$\ln \left(MR\right)=\ln \left({\displaystyle \frac{8}{{\pi }^2}}\right)-\left({\pi }^2{\displaystyle \frac{D_{eff}}{4L^2}}t\right)$$

(6)

The values of D_eff were determined by utilizing the slope (K) of a straight line derived from plotting the experimental drying data as lnMR against time. This calculation was based on the following equation:

$$\mathsf{{\rm K}}={\displaystyle \frac{{\pi }^2{D}_{eff}}{4L^2}}$$

(7)

2.2.3. Total Energy Consumption

The drying tests were conducted with a digital energy meter (PeakTech 9035, Ahrensburg, Germany) attached to a socket to monitor the energy consumption. Upon completion of the dehydration process, the drying systems’ total energy consumption was measured using the approach outlined by Tekin et al. [12].

2.2.4. Methods of Analyses

Extraction Procedure

Fresh and dried orange samples were extracted with a methanol–water solution (50:50, v/v) at a ratio of 1:10 w/v. The orange sample mixture was homogenized at 10,000 rpm for 2 min using an ultra-turrax homogenizer (Daihan, HG-15D, Seoul, Republic of Korea). Subsequent to homogenization, the mixture was agitated for 2 h at 25 °C. Subsequent to shaking, the mixture was subjected to centrifugation at 2800 g for 10 min. The resultant supernatant was further filtered using a 0.45 μm syringe filter [36].

Determination of Bioactive Compounds

The TPC assay was carried out using Singleton and Rossi’s modified procedure [37]. A UV–VIS spectrophotometer (Shimadzu UV-1800, Kyoto, Japan) measured the absorbance values at 760 nm. TPC was represented as mg GAE/g DM.

The analysis of DPPH was carried out by the methodology outlined by Brand-Williams et al. [38]. Absorbance was quantified at 515 nm, with results expressed as mg of Trolox equivalent (TE) per liter of material.

The ABTS radical cation scavenging capacity was evaluated using the technique established by Arnao et al. [39], with minor changes. The absorbance data were recorded at 734 nm, and outcomes were reported as mg TE/100 g DM.

Individual Phenolic Compounds

Individual phenolic compounds were analyzed by an HPLC system equipped with a diode array detector (SPD-M20A DAD, Shimadzu Corp, Kyoto, Japan) in accordance with the methodology described by Turan et al. [40]. Chromatograms were monitored at wavelengths of 278, 320, and 360 nm, and the flow rate was maintained at 1 mL/min.

Vitamin C

The HPLC-DAD system (Shimadzu Corp, Kyoto, Japan) was injected with 1 mL of the centrifuged blood orange juice after passing through a 0.45 μm filter (Millipore, Burlington, MA, USA). Supelco, Inc., Bellefonte, PA, USA, conducted the separation of ascorbic acid on a Supelcogel TM C 610H column (30 cm × 7.8 mm inner diameter) and a Supelguard column (5 cm × 4.6 mm) and identified the separation using a diode-array detector set at 210 nm. Pure standards of ascorbic acids were quantified using standard curves. The ascorbic acid concentrations are reported as milligrams per 100 g of sample [41].

Color

The color of both fresh and dried orange slices was evaluated utilizing a chromameter (Konica Minolta CR-400, Osaka, Japan). The color parameters have been identified as L* (lightness/darkness), a* (redness/greenness), and b* (yellowness/blueness). The total color change (ΔE) of the slices was determined using the subsequent formula (Equation (8)):

$$\Delta \mathrm{E}=\sqrt[2]{{\Delta \mathrm{L}}^2+{\Delta \mathrm{a}}^2+{\Delta \mathrm{b}}^2}$$

(8)

2.2.5. Multivariate Exploratory Data Analysis by Principal Component Analysis (PCA)

PCA is a multivariate statistical method widely used to summarize and visualize the information in a dataset [42]. Firstly, the implementation of PCA could be an intriguing solution for a more comprehensive understanding of the interactions between kinetic parameters (drying time, DR and D_eff), energy consumption, and phytochemical quality (TPC, DPPH, and ABTS, Vitamin C, Ferulic acid, Catechin, Hesperidin) and color parameters (L, a, and b) of blood orange slices that have been dried using the three processes (HAD, VD, and UAVD) at 50, 60, and 70 °C. Secondly, the application of PCA was preferred to gain a more comprehensive understanding of the interactions between the phytochemical quality (TPC, DPPH and ABTS, Vitamin C, Ferulic acid, Catechin, Hesperidin) and color parameters (L, a, and b) of different drying samples of blood orange samples dried using three treatments (HAD, VD, and UAVD) at 50, 60, and 70 °C, and FD.

2.2.6. Statistical Analysis

The statistical analysis was carried out using the STATISTICA 13.0 software (StatSoft, Tulsa, OK, USA). All experiments were conducted in triplicate, and the results were expressed as the mean values along with their standard errors. A one-way ANOVA was performed to compare the mean values of the test results. Duncan’s multiple comparison test, at a 95% confidence level, was applied to evaluate the impact of different drying methods on bioactive compounds, changes in the phenolic profile, vitamin C levels, and the color characteristics of the orange samples.

3. Results

3.1. Drying Kinetics

Fresh blood orange slices had an initial moisture percentage of 83.10 ± 0.26%. On a wet basis, the slices were dried until they had a final moisture percentage of 10%. Drying was performed using HAD, VD, and UAVD methods at temperatures of 50, 60, and 70 °C, as well as through FD. Figure 1 shows the pictures of fresh and dried blood oranges with HAD, VD, UAVD, and FD. The samples that were dried using HAD were found to have less volume.

Figure 2 indicates the experimental data for the moisture ratio (MR) vs. time curves of blood oranges dried with HAD, UAVD, and VD methods. Regardless of the drying methods, the moisture ratio (MR) continuously diminished during the drying period. There is no constant rate period in these curves; only a decreasing rate period is observed. The MR values were utilized to assess the eleven models delineated in Table 1.

Table 2. Calculated parameters and statistical model parameters according to HAD, VD, and UAVD.

Model	Parameters	HAD			VD			UAVD
		50 °C	60 °C	70 °C	50 °C	60 °C	70 °C	50 °C	60 °C	70 °C
Demir et al. [24]	a	0.918	1.111	1.175	1.149	1.391	1.352	1.149	1.391	2.004
	k	0.330	−0.411	−0.559	−0.419	−0.436	−0.389	−0.419	−0.436	−0.301
	n	0.396	−0.403	−0.480	−0.372	−0.358	−0.392	−0.372	−0.358	−0.304
	c	0.049	−0.144	−0.086	−0.164	−0.371	−0.390	−0.164	−0.371	−0.989
	R2	0.983	0.997	0.998	0.999	0.999	0.998	0.999	0.999	0.999
	RMSE	0.0455	0.0212	0.0192	0.0165	0.0152	0.0161	0.0137	0.0195	0.0114
	χ2	0.0023	0.0005	0.0005	0.0003	0.0003	0.0003	0.0002	0.0005	0.0002
Henderson and Pabis [25]	a	0.943	1.006	1.015	1.027	1.078	1.013	1.027	1.078	1.087
	k	0.112	0.229	0.330	0.221	0.286	0.286	0.221	0.286	0.272
	R2	0.982	0.993	0.995	0.994	0.987	0.998	0.994	0.987	0.982
	RMSE	0.0466	0.0343	0.0286	0.0470	0.0489	0.0399	0.0314	0.0356	0.0576
	χ2	0.0023	0.0013	0.0009	0.0025	0.0027	0.0018	0.0011	0.0014	0.0038
Lewis [26]	k	0.120	0.227	0.325	0.215	0.265	0.282	0.215	0.265	0.248
	R2	0.979	0.993	0.995	0.993	0.983	0.990	0.993	0.983	0.976
	RMSE	0.0499	0.0344	0.0290	0.0532	0.0565	0.0402	0.0327	0.0448	0.0666
	χ2	0.0025	0.0012	0.0009	0.0030	0.0034	0.0017	0.0011	0.0021	0.0047
Logarithmic [27]	a	0.918	1.111	1.075	1.149	1.391	1.352	1.149	1.391	2.004
	k	0.131	0.166	0.330	0.156	0.156	0.152	0.156	0.156	0.091
	c	0.049	0.144	0.186	−0.164	−0.371	−0.390	−0.164	−0.371	−0.989
	R2	0.983	0.997	0.998	0.999	0.999	0.998	0.999	0.999	0.999
	RMSE	0.0455	0.0212	0.0192	0.0165	0.0152	0.0161	0.0137	0.0195	0.0114
	χ2	0.0022	0.0005	0.0004	0.0003	0.0003	0.0003	0.0002	0.0005	0.0002
Midilli and Kucuk [28]	a	0.959	1.035	1.314	0.966	0.969	4.663	0.966	0.969	0.968
	k	0.177	0.191	0.238	0.149	0.135	0.054	0.149	0.135	0.108
	n	0.809	0.842	1.203	1.194	1.449	0.723	1.194	1.450	1.559
	b	−0.00004	0.00003	0.708	0.0000004	−0.000015	2.890	0.0000004	−0.00001	0.00006
	t	−0.00004	0.00003	0.446	0.0000004	−0.000015	−1.268	0.0000004	−0.00001	0.00006
	R2	0.986	0.998	0.998	0.996	0.998	1.000	0.996	0.998	0.997
	RMSE	0.0406	0.0299	0.0183	0.0314	0.0192	0.0313	0.0243	0.0176	0.0233
	χ2	0.0018	0.0011	0.0005	0.0013	0.0005	0.0014	0.0007	0.0004	0.0008
Modified Page [29]	k	0.123	0.224	0.319	0.211	0.259	0.279	0.211	0.259	0.247
	n	0.845	1.083	1.127	1.131	1.374	1.140	1.131	1.374	1.464
	R2	0.986	0.994	0.997	0.996	0.998	0.993	0.996	0.998	0.997
	RMSE	0.0411	0.0319	0.0222	0.0320	0.0213	0.0343	0.0256	0.0207	0.0255
	χ2	0.0018	0.0011	0.0004	0.0011	0.0005	0.0013	0.0007	0.0005	0.0007
Page [30]	k	0.170	0.198	0.276	0.172	0.186	0.233	0.172	0.156	0.129
	n	0.845	1.083	1.127	1.131	1.374	1.140	1.131	1.374	1.464
	R2	0.986	0.994	0.997	0.996	0.998	0.993	0.996	0.998	0.997
	RMSE	0.0411	0.0319	0.0222	0.0320	0.0213	0.0343	0.0256	0.0207	0.0255
	χ2	0.0018	0.0011	0.0006	0.0011	0.0005	0.0013	0.0007	0.0005	0.0007
Parabolic [31]	a	0.877	0.937	0.957	0.957	1.012	0.951	0.957	1.012	1.014
	c	0.002	0.006	0.014	0.006	0.010	0.009	0.006	0.010	−0.178
	b	−0.072	−0.148	−0.224	−0.146	−0.198	−0.186	−0.146	−0.198	0.006
	R2	0.970	0.996	0.998	0.998	0.999	0.998	0.998	0.999	0.999
	RMSE	0.0598	0.0251	0.0163	0.0184	0.0106	0.0187	0.0191	0.0241	0.0104
	χ2	0.0038	0.0004	0.0001	0.0004	0.0001	0.0004	0.0004	0.0007	0.0001
Two Term Exp [32]	a	0.219	1.544	1.625	1.621	1.890	1.636	1.621	1.889	1.943
	k	0.276	0.413	0.435	0.273	0.394	0.364	0.273	0.394	0.386
	R2	0.988	0.994	0.997	0.996	0.997	0.993	0.996	0.997	0.995
	RMSE	0.2646	0.1535	0.0905	0.0341	0.0257	0.0332	0.0250	0.0217	0.0320
	χ2	0.0016	0.0010	0.0005	0.0013	0.0007	0.0013	0.0007	0.0005	0.0012
Wang and Singh [33]	a	−0.094	−0.169	−0.242	−0.160	−0.193	−0.212	−0.160	−0.193	−0.171
	b	0.002	0.007	0.015	0.007	0.009	0.011	0.007	0.009	0.005
	R2	0.955	0.993	0.997	0.996	0.999	0.996	0.996	0.999	0.999
	RMSE	0.0735	0.0339	0.0227	0.0185	0.0115	0.0264	0.0246	0.0245	0.0117
	χ2	0.0057	0.0012	0.0006	0.0004	0.0001	0.0008	0.0007	0.0007	0.0002
Weibull [34]	a	0.845	1.083	1.127	1.131	1.374	1.140	1.131	1.374	1.464
	b	8.114	4.456	3.138	4.732	4.865	3.585	4.733	3.865	4.042
	R2	0.986	0.994	0.997	0.996	0.998	0.993	0.996	0.998	0.997
	RMSE	0.0411	0.0319	0.0222	0.0320	0.0213	0.0343	0.0256	0.0207	0.0255
	χ2	0.0018	0.0011	0.0006	0.0011	0.0005	0.0013	0.0007	0.0005	0.0007

displays the predicted model parameters and the statistical metrics for blood orange slices. The model that most accurately represented the thin-layer drying kinetics was selected based on the greatest R² values and the lowest RMSE and χ² values. Table 2 indicates that the logarithmic model was the most effective for the HAD, VD, and UAVD approaches, with R² values ranging from 0.997 to 0.998, 0.999 to 1.000, and 0.999 to 1.000, respectively. The k values for the logarithmic model were established as 0.131, 0.166, and 0.330 for 50, 60, and 70 °C for HAD drying and 0.156, 0.156, and 0.152 for VD, respectively. Consequently, a vacuum facilitates the transfer of water from the inside to the exterior due to the elevated water vapor pressure at reduced temperatures. Akdas and Baslar [43] identified the logarithmic model as the most effective model for Mandarin. The remaining 10 models exhibited high determination coefficients (R²) ranging from 0.979 to 0.998 for HAD, from 0.983 to 1.000 for VD, and from 0.982 to 1.000 for UAVD methods.

Table 3. Drying time, drying rate, effective moisture diffusivity, and total energy consumption of the blood oranges.

Drying Method	Temperature (°C)	Drying Time (h)	DR * (h−1)	Deff ** (m2/s)	Total Energy Consumption (kW·h)
HAD	50	22.5 Aa	0.044	9.08 × 10−10	0.163
	60	12.0 Ab	0.083	2.32 × 10−9	0.209
	70	9.0 Ac	0.108	2.82 × 10−9	0.266
VD	50	11.5 Ba	0.082	2.60 × 10−9	0.196
	60	9.5 Bb	0.115	2.76 × 10−9	0.235
	70	8.5 Bc	0.130	2.96 × 10−9	0.286
UAVD	50	10.0 Ca	0.104	2.20 × 10−9	0.298
	60	8.5 Cb	0.118	2.90 × 10−9	0.333
	70	7.5 Cc	0.135	2.99 × 10−9	0.372

* DR: drying rate and ** D_eff: effective moisture diffusivity. Different lowercase letter displays significance of the parameters. Different uppercase letter displays significance of the parameters (p < 0.5).

indicates the drying time, drying rate, effective moisture diffusivity, and total energy consumption of the blood orange slices. Drying times were recorded as 22.5 h, 12.5 h, and 9 h for HAD; 11.5 h, 9.5 h, and 8.5 h for VD; and 10 h, 8.5 h, and 7.5 h for UAVD at 50, 60, and 70 °C, respectively. A reduction in drying time was observed with increasing temperatures. Further analysis revealed that ultrasonic pretreatment significantly decreased the drying period. The effectiveness of ultrasound in minimizing drying durations has also been reported in other studies involving various fruits [40,44,45]. The gradual increase in the product’s temperature throughout the hot air-drying process due to external heat transfer impedes internal moisture migration, thereby extending the drying duration. Figure 1 and Figure 2 support this finding. Figure 1 illustrates that significant shrinkage occurred in samples dried with hot air at 50 °C. In Figure 2, a quick drying rate was first noted in samples dried at 50 °C, followed by a marked reduction in drying rate following a specific phase of the process. The product’s microstructure deteriorated, hindering moisture transfer from the micro-capillaries to the surface. In contrast, vacuum drying decreases pressure, which reduces the water boiling point in the food, hence increasing the surface evaporation rate. The combination of ultrasound with vacuum has been shown to significantly reduce drying times. This effect is attributed to cavitation, which creates microscopic voids within the structure of fruits and vegetables, facilitating easier water removal and thereby shortening the drying process. The secondary consequence is the mechanical impact induced by ultrasound, which diminishes the adhesion between moisture and the microtube, hence expediting moisture evacuation [46]. Furthermore, ultrasonic treatment can diminish internal viscosity and enhance the heat transfer coefficient [15]. The last impact of ultrasound is the thermal effect, which induces a marginally accelerated temperature rise, advantageous for the internal diffusion and evaporation of moisture [44].

The drying of various foods under constant conditions typically results in curves with distinct shapes during the falling rate period [47]. Damage to the cell structure of foods may result in deviations during the constant drying rate period [12]. Ultrasonic treatment enhances the drying rate by leveraging the beneficial impacts of heating through attenuation and adsorption in conjunction with the mechanical effects of pressure waves [12,14,15]. Table 3 indicates that the drying rate values for the HAD, VD, and UAVD methods varied with drying temperature. The drying rate values obtained through the UAVD method, as presented in Table 3, exceeded those of the HAD and VD methods during the falling rate period, attributable to the beneficial effects of ultrasound. The drying rate values escalated with all drying techniques as the temperature increased. A shortened period is essential for the material to attain equilibrium moisture content. Figure 3 illustrates the drying rate vs. kg water/kg dry matter.

In the current study, the D_eff rose as the temperature rose in all of the drying processes. The D_eff values of the HAD, VD, and UAVD dried samples varied from 9.08 × 10⁻⁶ to 2.82 × 10⁻⁵ m²/s, from 2.60 × 10⁻⁵ to 2.96 × 10⁻⁵ m²/s, and from 2.20 × 10⁻⁵ to 2.99 × 10⁻⁵, respectively. Table 3 displays the D_eff values for the HAD-, VD-, and UAVD-dried blood orange slices at 50, 60, and 70 °C. The D_eff values of the UAVD-dried samples are higher than those of the HAD-dried and VD-dried blood oranges. Table 3 presents the D_eff values for the HAD, VD, and UAVD methods at 50, 60, and 70 °C. The D_eff values of the UAVD-dried samples exceed those of the HAD-dried and VD-dried blood oranges. The data in Table 3 suggest that D_eff increases as the temperature rises, which is because the hydration content of the blood oranges decreases due to the rapid evaporation of water molecules. As the temperature increased, Huang and Chen [48] noted that the D_eff of sewage sludges increased.

Energy consumption varies with drying methods and temperature, and it decreases as the temperature increases. This study found that total energy consumption varied from 0.163 to 0.266 kWh with HAD, from 0.196 to 0.286 kWh with VD, and from 0.298 to 0.372 kWh with the UAVD method. The total energy consumption of samples dried using the UAVD method exceeded that of other drying methods; however, the drying times with UAVD were shorter, suggesting that UAVD drying may be more economical.

3.2. Total Phenolic Content and Antioxidant Capacity

Citrus fruits are abundant in phenolic acids and flavonoids, two principal categories of natural antioxidants that underlie their functional qualities. The differences in flavonoid content in fruits are primarily due to biological factors [49]. These secondary metabolites serve several functions in the plant; specifically, in the fruit, they are linked to color, sensory attributes (flavor, astringency, texture), nutritional properties, and antioxidant activity [50].

TPC values and antioxidant capacity values determined by DPPH and ABTS methods of fresh blood orange slices and blood orange slices dried with HAD, VD, and UAVD at 50, 60, and 70 °C, as well as through freeze-drying are shown in

Table 4. Total phenolic content, antioxidant activity, and vitamin C values of fresh and dried blood orange.

	TPC (mg GAE/100 g)
Fresh	154.25 ± 4.65 A	154.25 ± 4.65 A	154.25 ± 4.65 A
FD	131.27 ± 0.41 B	131.27 ± 0.41 B	131.27 ± 0.41 B
	50 °C	60 °C	70 °C
HAD	107.08 ± 0.27 Da	87.99 ± 0.41 Db	44.00 ± 1.90 Ec
VD	122.24 ± 1.08 Ca	112.36 ± 1.49 Cb	59.70 ± 0.0.27 Dc
UAVD	128.77 ± 0.95 Ba	108.16 ± 1.08 Cb	65.25 ± 2.03 Cc
	DPPH (mg TE/100 g)
Fresh	7380.01 ± 16.44 A	7380.01 ± 16.44 A	7380.01 ± 16.44 A
FD	7339.09 ± 8.84 A	7339.09 ± 8.84 A	7339.09 ± 8.84 A
	50 °C	60 °C	70 °C
HAD	5083.67 ± 36.81 Da	4705.49 ± 6.69 Db	2249.00 ± 46.85 Dc
VD	5990.63 ± 20.08 Ca	5123.83 ± 36.81 Cb	4434.40 ± 50.20 Cc
UAVD	6907.63 ± 46.85 Ba	5950.47 ± 66.69 Bb	4926.37 ± 53.55 Bc
	ABTS (mg TE/100 g)
Fresh	242.67 ± 7.03 A	242.67 ± 7.03 A	242.67 ± 7.03 A
FD	225.46 ± 0.35 B	225.46 ± 0.35 B	225.46 ± 0.35 B
	50 °C	60 °C	70 °C
HAD	162.93 ± 6.67 Ea	139.74 ± 3.16 Eb	118.31 ± 2.81 Ec
VD	185.76 ± 0.70 Da	153.80 ± 1.76 Db	129.56 ± 1.41 Dc
UAVD	218.08 ± 4.22 Ca	192.79 ± 7.03 Cb	169.60 ± 2.11 Cc
	Vitamin C (mg/100 g)
Fresh	55.32 ± 0.45 A	55.32 ± 0.45 A	55.32 ± 0.45 A
FD	49.01 ± 0.21 B	49.01 ± 0.21 B	49.01 ± 0.21 B
	50 °C	60 °C	70 °C
HAD	36.19 ± 0.14 Da	33.60 ± 0.35 Eb	29.79 ± 0.18 Ec
VD	44.81 ± 0.30 Ca	41.40 ± 0.25 Db	37.55 ± 0.29 Dc
UAVD	46.10 ± 0.08 Ca	44.82 ± 0.44 Ca	41.63 ± 0.20 Cb

Different lowercase letters display significance of the parameters (p < 0.5). Different uppercase letters display significance of the parameters (p < 0.5).

. The TPC value of the fresh blood orange slice was 154.25 mg GAE/100 g. The drying techniques and temperatures substantially influenced the total bioactive phenolic compounds and antioxidant activity levels. The FD dried blood orange slices have the highest TPC value (131.27 mg GAE/100 g) within the dried blood orange slices. The TPC result of the samples dried by UAVD at 50 °C (128.77 mg GAE/100 g DM) is comparable to that of the FD-dried samples (131.27 mg GAE/100 g DM) (p < 0.05). The preservation of TPC is due to the low temperature and vacuum conditions inherent to the FD process. Tekin-Cakmak et al. [16] and Goztepe et al. [35] indicated that some red fruits, when dried using similar procedures, had the greatest TPC values in freeze-dried samples. The TPC of dried blood orange slices diminished as a result of thermal degradation at rising temperatures throughout the drying process. For UAVD, TPC values decreased from 128.77 mg GAE/g DM to 65.25 mg GAE/g DM by increasing the temperature from 50 °C to 70 °C. Similarly, research on Tunisian eggplants examined how different drying techniques impacted their drying properties and bioactive compounds. Chouaibi et al. [51] established that both freeze-drying and ultrasound-assisted drying reduced product deterioration, corroborating results from analogous research. The elevated TPC value of UAVD-dried blood orange slices has been implicated in the occurrence of cavitation, resulting in the extraction of components that are released from cells during the drying process.

The antioxidant activity of many foods has been extensively studied due to its ability to counteract oxidation processes that reduce chronic illnesses associated with oxidative stress in the human body [52]. Various antioxidant chemicals, including ascorbic acid, flavonoids, and phenolic acids, were regarded as natural sources in horticulture goods. The ABTS analysis was employed to assess the antioxidant potential of both lipophilic and hydrophilic antioxidants, encompassing flavonoids (flavones, flavanones, and flavonols) and phenolic acids, particularly ferulic acid and p-coumaric acid [5]. The antioxidant capacity of fresh blood orange slices was determined to be 7380.01 mg TE/100 g DM in the DPPH assay and 242.67 mg TE/100 g DM in the ABTS assay. Consistent with the TPC results, the maximum DPPH and ABTS values recorded were 7330.09 mg TE/g DM and 226.46 mg TE/g DM, respectively, in FD-dried blood orange slices. At 50 °C, the UAVD technique yields 6907.63 mg TE/100 g DM, followed by the FD method at 7339.09 mg TE/100 g DM for DPPH, with the VD method resulting in 5990.63 mg TE/100 g DM thereafter. A comparable pattern observed with DPPH is also evident with ABTS. Thermal treatments and oxidative processes may have caused the breakdown of phenolic compounds and reduced the antioxidant activity of the samples [53]. The antioxidant capacity of fruits dried using the UAVD technique surpasses that of samples dried using HAD, as indicated by the TPC findings. The results indicate that the UAVD approach could be beneficial as an alternative to the HAD method.

3.3. Vitamin C

The concentrations of different chemical constituents, such as vitamins, minerals, and phenolics, which are recognized for their potent antioxidant qualities, impact the quality of citrus fruits [54]. Fresh juice has high levels of vitamin C, which is measured as ascorbic acid among the vitamins [55]. Rapisarda [56] reported that vitamin C is rich in blood oranges; in fact, Moro and Tarocco have a higher concentration of vitamin C than many other blood orange cultivars, with juice containing 0.50 to 0.80 g/kg. The vitamin C content of both fresh and dried blood orange slices is also shown in Table 4. Fresh blood orange slices have a vitamin C level of 55.32 mg/100 g, whereas dried blood orange slices have a vitamin C content of 29.79 to 49.01 mg/100 g, which is quite comparable to what has been found in other studies [50,57].

3.4. Individual Phenolic Compounds

Table 5. Individual phenolic compounds of blood orange slices (µg/100 mL of dry weight).

	FRESH	FD	HAD			VD			UAVD
			50 °C	60 °C	70 °C	50 °C	60 °C	70 °C	50 °C	60 °C	70 °C
4-Hydroxybenzoic acid	698.384	679.661	617.872	530.046	415.985	665.501	548.873	449.317	687.720	563.045	464.280
Gallic acid	694.133	663.086	586.187	506.159	386.882	608.642	563.000	420.385	630.301	585.626	458.424
Protocatechuic acid	90.007	76.274	50.086	34.536	29.728	60.823	52.608	51.665	81.743	87.068	59.782
Syringic acid	419.783	313.852	151.958	127.316	73.126	135.800	122.398	118.860	137.515	112.840	111.849
Caffeic acid	773.416	741.618	645.529	566.257	412.475	665.606	537.631	402.346	696.803	545.505	405.051
Chlorogenic acid	874.063	828.764	807.190	629.639	469.726	824.394	719.540	570.502	887.985	733.310	588.280
Ferulic acid	2442.164	2307.669	1723.559	1586.942	1223.456	1927.890	1884.938	1645.212	2142.062	1922.387	1633.186
o-coumaric acid	315.635	300.019	236.401	200.192	114.675	276.441	227.648	130.081	272.344	231.669	143.211
p-coumaric acid	902.509	910.730	835.561	726.757	622.768	853.823	773.700	651.752	876.964	757.289	683.279
Sinapic acid	569.698	547.795	439.647	387.185	265.075	459.391	354.217	295.903	498.377	339.098	259.853
Ellagic acid	1137.644	1075.601	814.108	756.400	529.044	911.640	844.927	708.886	980.226	860.551	731.712
Catechin	1796.558	1407.770	1046.529	885.411	874.530	1264.428	1088.181	892.293	1279.250	1284.575	856.867
Chrysin	698.898	684.894	604.188	400.100	241.552	630.269	568.110	451.798	670.510	550.835	452.889
Hesperidin	12,201.82	11,766.77	8445.41	6978.38	5811.42	9832.24	9036.76	7814,85	10,710.31	9419.63	7757.89
Quercetin	265.783	219.296	182.930	146.665	127.057	152.290	125.304	108.828	178.698	134.023	106.376
Rutin	380.074	374.675	194.474	179.055	129.941	255.517	186.065	106.053	244.775	150.100	139.097

shows the effect of different drying methods and temperatures on individual phenolic compounds of blood orange slices. The HAD method caused greater reductions in the amount of all phenolic compounds identified than other drying methods. Hirsch [58] indicated that the activation of oxidative enzymes, including polyphenol oxidase, during hot air oven drying results in a reduction of flavonoid concentration. The loss was reduced due to the diminished activity of the polyphenol oxidase enzyme during freeze-drying at lower temperatures [59]. There are six hydroxycinnamic acids found in blood orange slices: caffeic acid, chlorogenic acid, ferulic acid, o-coumaric acid, p-coumaric acid, and sinapic acid. Ferulic acid was the most dominant hydroxycinnamic acid in fresh and dried blood orange slices, accounting for the largest proportion of the total hydroxycinnamic acid contents. Ferulic acid is found mostly in fresh blood oranges (2442.164 mg/100 g DM). Among dried blood oranges, freeze-dried blood oranges contained the most ferulic acid (2307.669 mg/100 g DM), followed by blood oranges dried with UAVD at 50 °C with a ferulic acid content of 2142.062 mg/100 g KM. Ellagic acid is a hexahydroxydiphenic acid abundantly found in fruits, pomegranates, cranberries, and other plant foods. This study found that it is the most abundant compound in blood orange after ferulic acid. Ellagic acid exhibits a wide range of biological properties, such as playing an active role in anticancer treatment [60]. Flavonoids isolated from citrus fruits are a group of natural compounds with phenolic structures. This study found six flavonoids: chrysin and rutin were flavon, catechin, myricetin, and quercetin were flavanol, and hesperidin was flavanone. Flavanone is the major flavonoid in orange varieties. Table 5 shows that hesperidin was a flavanone and the most abundant individual phenolic compound in fresh and dried blood orange slices. Hesperidin is found mostly in fresh blood oranges (12,201.82 mg/100 g DM), followed by freeze-dried blood oranges (11,766.77 mg/100 g DM).

3.5. Color

The critical factor that influences consumer acceptance of dried products is product color.

Table 6. Color parameters of blood orange slices.

	FRESH	FD	HAD			VD			UAVD
			50 °C	60 °C	70 °C	50 °C	60 °C	70 °C	50 °C	60 °C	70 °C
L*	54.07 ± 0.61 a	52.70 ± 0.16 a	44.48 ± 0.25 d	43.20 ± 0.10 d	40.44 ± 0.07 f	49.75 ± 0.13 b	47.57 ± 0.31 c	42.51 ± 0.09 de	46.73 ± 0.16 c	43.42 ± 0.26 d	42.14 ± 0.09 de
a*	18.35 ± 0.14 ab	19.38 ± 0.41 a	14.80 ± 0.10 d	12.46 ± 0.19 e	9.90 ± 0.18 f	17.64 ± 0.04 b	16.34 ± 0.02 c	14.97 ± 0.09 d	19.35 ± 0.23 a	18.72 ± 0.08 ab	17.57 ± 0.31 b
b*	16.11 ± 0.02 a	15.38 ± 0.26 ab	12.17 ± 0.15 c	9.98 ± 0.54 d	5.11 ± 0.12 f	14.34 ± 0.28 b	12.26 ± 0.15 c	6.74 ± 0.20 ef	14.14 ± 0.15 ab	12.71 ± 0.18 c	7.51 ± 0.12 e

Different lowercase letters display significance of the parameters (p < 0.5).

displays the color characteristics of fresh and dried blood orange slices. The L*, a*, and b* values of fresh blood orange slices are 54.07, 18.35, and 16.15, respectively. In contrast, the L* values of dried blood orange slices vary from 40.44 to 52.70, the a* values range from 9.90 to 19.38, and the b* values range from 5.11 to 16.15. The L*, a*, and b* values of the FD dry blood orange slices were the highest among all dried blood orange slices and were closer to the L*, a*, and b* values of fresh blood orange slices. The L*, a*, and b* values of blood orange slices dried using the freeze-drying method are then compared to those dried using the UAVD method at 50 °C. According to certain researchers [12,13,40], the UAVD approach can avoid color changes throughout the drying process. Turan et al. [40] indicated elevated L*, a*, and b* values for UAVD-dried goji berries in comparison to HAD-dried samples. The reduction in the L* value of dried samples is mostly attributable to Maillard reactions and nonenzymatic browning, which are inevitable during drying operations [61]. Furthermore, drying may diminish essential substrates for the Maillard reaction, including sugars and soluble pigments, resulting in a decrease in the a* value of the dried samples [62].

3.6. PCA

Figure 4 shows that the implementation of PCA could be an intriguing solution for a more comprehensive understanding of the interactions between kinetic parameters (drying time, DR, and D_eff), energy consumption, and phytochemical quality (TPC, DPPH, ABTS, Vitamin C, Ferulic acid, Catechin, Hesperidin) and color parameters (L, a, and b) of blood orange slices that have been dried using the three processes (HAD, VD, and UAVD) at 50, 60, and 70 °C. The first two principal components (PC1 and PC2) accounted for 78.3% of the total variability in the studied parameters: PC1 of the PCA accounted for 61.6% of the variance and was expressed mainly by TPC (0.323), DPPH (0.326), ABTS (0.311), vitamin C (0.304), ferulic acid (0.332), catechin (0.308), hesperidin (0.334), and color parameters (L (0.279), a (0.296) and b (0.323)), which had good correlations among them. PC2, which accounted for 28.2% of the variance, is exclusively represented by drying time (0.468), DR (0.483), D_eff (0.443), and total energy consumption (TEC) (0.458). The score plot clearly illustrates the clustering of samples according to their similarities.

Figure 5 shows that the application of PCA was preferred to gain a more comprehensive understanding of the interactions between the phytochemical quality (TPC, DPPH and ABTS, vitamin C, ferulic acid, catechin, hesperidin) and color parameters (L, a, and b) of different drying samples of blood orange samples dried using three treatments (HAD, VD, and UAVD) at 50, 60, and 70 °C, and FD. The first two principal components (PC1 and PC2) accounted for 78.3% of the total variability in the studied parameters: PC1 of the PCA accounted for 89.5% of the variance and was expressed mainly by TPC, DPPH, ABTS, vitamin C, ferulic acid, catechin, hesperidin and color parameters (L, a and b), which had good correlations among them. PC2 accounted for 5.2% of the variance. The score plot clearly illustrates the clustering of samples according to their similarities.

It is clearly seen from the graphs that phytochemical quality decreases with increasing temperature. Different temperature applications of the same drying methods are on the same line. Numerous researchers have discovered similar findings [63,64,65,66]. However, Zheng and Wang [67] proposed that the antioxidant capacity of plant extracts could be influenced by not only the content of antioxidants but also a synergy between them and other plant constituents. This may elucidate the reason why the antioxidant activity of samples with comparable concentrations of total phenolics can vary significantly [68]. This is a result of the synergistic and antagonistic interactions between antioxidants.

4. Conclusions

This study investigated the effects of different drying methods and temperatures on drying period, phytochemical quality, vitamin C levels, and color alterations in blood orange slices. UAVD demonstrated decreased drying time and improved retention of bioactive constituents, such as vitamin C, compared to VD and HAD. Moreover, UAVD exhibited fewer color changes and shrinkage than VD and HAD. Despite the samples dried by freeze-drying (FD) demonstrating the highest retention of bioactive substances, their total phenolic content (TPC) and CUPRAC recovery values were lower than those of the other samples. The implementation of PCA is an intriguing solution for a more comprehensive understanding of the interactions among kinetic parameters (drying time, DR, and Deff), energy consumption, phytochemical quality (TPC, DPPH, and ABTS, Vitamin C, Ferulic acid, Catechin, Hesperidin), and color parameters (L, a, and b) of blood orange slices that have been dried using the three processes (HAD, VD, and UAVD) at 50, 60, and 70 °C. This study demonstrates that UAVD is efficacious for drying blood orange slices, leading to decreased drying duration, negligible color and physical changes in samples, and improved retention of bioactive components.