Effect of cold atmospheric plasma pretreatment on phytochemical content and microstructure of beetroot during intermittent microwave hot-air drying

Abstract

Thermal degradation of phytochemicals in heat-sensitive foods necessitates development of innovative drying strategies. This study investigated the application of dielectric barrier discharge (DBD) plasma pretreatment on intermittent microwave–hot air (IMHA) drying for beetroot slices. Scanning electron microscopy showed that DBD-plasma pretreatment generated surface microchannels and increased porosity by 27%, thereby facilitating moisture removal. The optimal conditions depended on the processing objective: the highest drying efficiency was achieved with 5 min of DBD-plasma pretreatment at 600 W and a pulse ratio (PR) of 1, reducing the total drying time by 34%, whereas the highest phytochemical retention was obtained at 180 W and PR of 2, yielding 55.2% higher betacyanin content, 33.8% greater antioxidant activity, and 43.7% higher total phenolic content. These findings highlight the potential of integrating DBD-plasma pretreatment with IMHA drying to improve drying efficiency and preserve product quality. From an industrial perspective, this approach may support the production of dried beetroot with improved functional quality and shorter processing time. A limitation of this study is that the process was evaluated at laboratory scale; therefore, further pilot-scale validation is required to confirm scalability.

Highlights

• •DBD–IMHA minimized thermal stress, enhancing color retention and pigment extraction.
• •DBD-IMHA method enhanced drying efficiency, reducing total drying time by 53%.
• •Optimal DBD–IMHA treatment raised betaxanthin 59%, betacyanin 57%, and phenolics 44%.
• •DBD-plasma pretreatment increased Tg by reducing moisture plasticization.
• •Pulse ratio reduced porosity but improved phytochemical retention.

1. Introduction

Beetroot (Beta vulgaris L.) has attracted considerable attention as a functional food because of its high content of bioactive compounds and its potential health-promoting properties (Mudgal et al., 2022). Its characteristic red-violet color is primarily derived from betalain pigments, including betacyanins, which account for approximately 75%–95% of total betalains, and betaxanthins, which account for approximately 5%–25% (Chhikara et al., 2019). However, fresh beetroot has a high moisture content, approximately 86%–88%, which makes it susceptible to microbial spoilage and enzymatic deterioration during storage. Therefore, effective preservation methods (such as drying) are required to extend shelf life while maintaining product quality.

Drying is widely used to improve the stability of fruits and vegetables; however, conventional hot-air drying often involves prolonged exposure to elevated temperatures, resulting in extended drying times, high energy consumption, structural collapse, color degradation, and loss of heat-sensitive nutrients and compounds (Dhiman et al., 2021; Reis et al., 2022). Microwave drying has emerged as an alternative method because it promotes volumetric heating, accelerates moisture removal, and can improve product porosity properties compared with conventional hot-air drying (An et al., 2024; Zielinska et al., 2020). Nevertheless, continuous microwave application can cause non-uniform heating and localized overheating. Intermittent microwave operation mitigates these limitations by allowing temperature and moisture redistribution during off-periods, thereby improving drying uniformity (Song et al., 2024). To overcome the limitations of standalone microwave drying, including its dependence on dielectric properties and sample geometry, intermittent microwave–hot air (IMHA) drying has been proposed as a hybrid approach that combines rapid volumetric heating with convective moisture removal, thereby improving drying efficiency and product quality (Maftoonazad et al., 2022). Despite these advantages, moisture removal during hybrid drying may still be constrained by internal mass-transfer resistance and cellular barriers in plant tissues (Bassey et al., 2021; Boateng, 2024). Even so, moisture migration during hybrid drying may still be limited by internal mass-transfer resistance and cellular barriers. Therefore, non-thermal pretreatments have been increasingly explored as a means of modifying tissue structure and facilitating moisture migration prior to drying.

Cold plasma (CP), particularly cold atmospheric plasma, has attracted growing interest as a non-thermal pretreatment because it can modify food surfaces and near-surface structures without causing substantial thermal damage (Gholamazad et al., 2022). CP is generated by gas ionization under an electric field and contains reactive oxygen species (ROS), reactive nitrogen species (RNS), energetic electrons, ions, and UV photons. These reactive components can interact with plant tissues to modify cell walls and membranes, increasing permeability and forming microchannels that enhance moisture transport during drying (Ramezan et al., 2023). Among CP systems, dielectric barrier discharge (DBD) plasma is widely used in food processing, particularly for heat-sensitive materials, because it operates at atmospheric pressure while maintaining the gas temperature near ambient conditions. In DBD systems, the dielectric layer limits discharge current and prevents arc formation, enabling the generation of reactive species with minimal thermal load (Ashtiani et al., 2023). In drying applications, DBD-plasma pretreatment can modify the surface morphology and microstructure of plant tissues by partially disrupting waxy layers and forming micropores or microchannels. These changes may reduce resistance to moisture migration, accelerate drying, shorten processing time, and improve selected quality attributes such as color, rehydration capacity, and retention of bioactive compounds (Chanchula et al., 2025; Seelarat et al., 2024).

Previous studies have demonstrated that CP-pretreatment can enhance drying efficiency and improve selected quality attributes of dried plant products. However, most studies have focused on its application before conventional drying methods, such as hot-air and infrared drying, whereas its combination with microwave-based drying systems has received limited attention (Bao et al., 2021; Khudyakov et al., 2022). To the best of our knowledge, no previous study has comprehensively investigated DBD-plasma pretreatment prior to IMHA drying of beetroot slices. In particular, the effects of CP-induced structural modifications and their interaction with key microwave-drying variables, such as microwave power and Pulse Ratio (PR), remain unclear. In addition, limited information is available on how these factors influence drying behavior and quality attributes, including moisture transport, thermal behavior, microstructure, color, and phytochemical retention. Therefore, the novelty of the present study lies in integrating DBD-plasma pretreatment with IMHA drying of beetroot slices and evaluating their combined effects on drying performance and product quality. Specifically, this study investigated the interactive effects of DBD-plasma exposure time, microwave power, and PR on drying behavior, moisture transport, thermal properties, microstructure, color, rehydration capacity, and bioactive compound retention. It was hypothesized that DBD-plasma pretreatment would induce structural modifications that reduce mass-transfer resistance, thereby enhancing IMHA drying efficiency while preserving the physicochemical quality of dried beetroot.

DBD: Dielectric barrier discharge	ROS: Reactive Oxygen Species
HA: Hot air	RNS: Reactive Nitrogen Species
PR: Pulse ratio	ΔCp: Heat-capacity change
IMHA: Intermittent microwave–hot-air	OES: Optical emission spectroscopy
MR: Moisture ratio	Ton/Toff: MW on/off times
CP: Cold plasma	AOA: Antioxidant activity
Tg: Glass transition temperature	TPC: Total phenolic content
MC: Moisture content (g water/g dry solids)	TAC: Total anthocyanin content
△E: Total Color Change

2. Materials and methods

2.1. Raw materials preparation

Fresh beetroot (Beta vulgaris L.) was purchased from a local market and stored at 4 °C to extend storage stability, minimize microbial spoilage, and maintain physicochemical quality before experimentation. During refrigerated storage, the samples were kept in packaging containing immobilized zinc oxide nanoparticles (ZnO NPs), which served solely as a protective packaging material to reduce potential microbial contamination and oxidative deterioration (Javad et al., 2022). The ZnO nanoparticles were embedded within the packaging polymer matrix and were neither directly applied to the beetroot surface nor incorporated into the food matrix. Therefore, ZnO NPs were not considered an experimental treatment factor in this study. In addition, all samples, including control and CP-pretreated groups, were stored under identical packaging and refrigeration conditions prior to drying experiments. Consequently, any potential packaging-related effects would be uniform across all experimental groups and would not influence the comparative evaluation of drying kinetics, antioxidant activity, phenolic compounds, betalain stability, or other physicochemical properties investigated in this work. The ZnO NP-containing packaging was used exclusively to maintain sample integrity during short-term refrigerated storage before processing. Beetroots (Beta vulgaris L.) of the same variety, similar size, and uniform color were selected, and only defect-free samples were used. The samples had an average weight of 125 ± 10 g and an initial moisture content of 7.71 ± 0.64 g water/g dry solids. The initial moisture content was determined according to AOAC Official Method 925.10 using the air-oven method. Specifically, 2.000 ± 0.001 g of the homogenized sample was dried at 105 °C until constant weight was achieved, approximately 24 h. After immersion in cold water, the samples were washed, peeled, and sliced into uniform dimensions of 3 cm × 1 cm × 1 cm using a mold and cutter. The beetroot samples were packaged in ZnO NP-containing packaging and refrigerated at 4 ± 1 °C to minimize moisture loss.

2.2. Cold plasma pretreatment system

DBD-plasma generator with dimensions of 95 cm × 60 cm × 60 cm (Plasmatech-15B, Kavoshyaran Fan Pouyan, Iran) was used to generate cold atmospheric plasma under atmospheric-pressure conditions. The system operated using ambient air as the working gas at a controlled flow rate of 5 mL/min, regulated by a mass-flow controller. The discharge voltage and frequency were set to 25 kV and 50 Hz, respectively, corresponding to an input power of approximately 75 W (Afshar et al., 2022). The generated CP was non-thermal, with the gas temperature remaining close to ambient conditions. For each experiment, 180 ± 10 g of fresh beetroot samples was placed inside the treatment chamber, which had a diameter of 22 cm. The electrode gap in the DBD reactor, defined as the distance between the high-voltage and grounded electrodes, was fixed at 4 cm, while the distance between the high-voltage electrode and the sample surface was maintained at 2 cm throughout the treatment. All experiments were conducted under controlled laboratory conditions at 25 ± 2 °C and 50 ± 5% relative humidity (Li, Li, et al., 2019; Yi et al., 2022). Untreated samples were used as the control.

2.3. Drying process

A hybrid dryer (SolarDOM LG, model MS94WCR, Korea) was used for the IMHA drying process. The device operated at a voltage of 230 V and a frequency of 50 Hz. Microwave power levels were set at 180, 360, and 600 W. The system was equipped with an adjustable temperature-control sensor operating over a range of 40–230 °C and a rotating tray operating at 2.5 rpm. DBD-plasma pretreated samples were initially dried using the IMHA dryer at different microwave power levels and PR values, followed by hot-air drying at 40 °C and an air velocity of 1 m/s. PR was defined as the ratio of the total pulse cycle time to the microwave active time, as expressed in Eq. (1):

$$\mathrm{PR}=\frac{{\mathrm{t}}_{\mathrm{on}}+{\mathrm{t}}_{\mathrm{off}}}{{\mathrm{t}}_{\mathrm{on}}}$$ (1)

t_on and t_off represent the microwave on-time and off-time, respectively, and were determined by preliminary trials to prevent hot spots, sample charring, and excessive shrinkage (Table 1). Microwave drying was terminated before visible charring occurred. Subsequently, the beetroot samples were placed on a mesh tray and dried. Moisture content was monitored at 45 min intervals using a digital balance with an accuracy of ±0.0001 g (Entris BCA224i-1S, Sartorius AG, Göttingen, Germany). The samples were dried to a final moisture content of 0.16 ± 0.03 g water/g dry solids and then sealed in ZnO NP-containing bags (Kamkari et al., 2024).

Table 1.

On/off times at different Microwave power and PR.

Pulse ratio	180 W	360 W	600 W
	Time (min)
1	21 min on (21 min total)	34 min on (34 min total)	52 min on (52 min total)
2	1 min on/1 min off (42 min total)	1 min on/1 min off (68 min total)	1 min on/1 min off (104 min total)

2.4. Microstructural observations

The effect of DBD-plasma pretreatment on the surface morphology of fresh beetroot samples was investigated using scanning electron microscopy (SEM). Microstructural and porosity analyses of DBD-plasma-pretreated samples dried by IMHA were performed using SEM and micro-computed tomography (micro-CT), respectively.

2.4.1. SEM

Before immersion in liquid nitrogen, small sections were cut from the surface and center of each fresh and dried sample. The samples were evenly distributed on carbon tape attached to an aluminum holder. Before morphological observation, the samples were coated with a thin conductive layer of gold using a high-vacuum sputter coater (Q150R S, Quorum Technologies Ltd., East Sussex, UK) operated at an emission current of 20 mA for 60 s. SEM images were acquired at magnifications corresponding to scale bars of 100, 200, and 500 μm under an accelerating voltage of 20 kV.

2.4.2. Micro-computed tomography (Micro-CT)

The microstructural characteristics of the dried beetroot samples were examined using non-destructive X-ray micro-computed tomography (micro-CT) (LOTUS inVivo, Behin Negareh Co., Iran). Images were acquired at an X-ray tube voltage of 40 kV and current of 130 μA, with a frame exposure time of 1 s and a twofold magnification setting. Three-dimensional datasets were reconstructed using the standard Feldkamp–Davis–Kress (FDK) algorithm implemented in the LOTUS inVivo-REC software. The resulting Micro-CT images were subsequently processed to quantify porosity and assess the microstructural attributes of the samples.

2.5. Optical emission spectroscopy (OES)

The reactive species generated by the DBD-plasma device were identified using optical emission spectroscopy (Avio 220 Max ICP-OES).

2.6. Glass transition temperature

The glass transition temperature (Tg) of the dried beetroot samples was measured using a differential scanning calorimeter (DSC 214 Polyma, NETZSCH, Germany). The coating was applied using a high-vacuum sputter coater (Q150R S, Quorum Technologies Ltd., East Sussex, UK) operated at an emission current of 20 mA for 60 s, producing a uniform conductive coating on the sample surface suitable for SEM analysis. The samples were first ground into powder, and approximately 10–12 mg of each powdered sample was transferred to an aluminum pan. The pan was placed in the DSC chamber and cooled to −50 °C at a rate of 10 °C/min The samples were held at −50 °C for 1 min, after which the temperature was increased to 80 °C at a heating rate of 10 °C/min to analyze the thermal characteristics of the dehydrated beetroot samples. Nitrogen was used as the purge gas during DSC analysis (Kahraman et al., 2021).

2.7. Moisture content

The moisture content (MC) of fresh, CP-pretreated fresh, and microwave-dried beetroot samples was determined using the oven method described by Luo et al. (2019) using Eq. (2):

$$\mathrm{MC}\left(\mathrm{d}\mathbf{.}\mathrm{b}\mathbf{.}\right)=\frac{{\mathrm{M}}_{\mathrm{w}}}{{\mathrm{M}}_{\mathrm{s}}}$$ (2)

M_w represents the mass of water (g), whereas M_s denotes the mass of dry solids in the sample (g). The water mass was calculated as the difference between the initial sample weight and the final dry weight, measured using an analytical balance with an accuracy of ±0.001 g. During the hot-air drying phase, moisture loss was determined from sample weight changes measured at fixed time intervals.

2.8. Color measurement

A white, non-reflective trapezoidal photography light box was constructed with a large base of 84.5 cm, a small base of 28.5 cm, and a height of 54 cm. Images were captured using a 20-megapixel Canon EOS R6 camera mounted on a holder fixed to a 20 cm square door with an 8 cm circular window. Illumination was provided by eight 50 cm daylight 12 V SMD-LED strips connected to a 100 W AC–DC power supply to ensure uniform image acquisition. Images were captured after camera calibration using a color chart and white-balance adjustment. RGB images were converted to the CIE Lab* color space using the “Color Space Converter” plugin in ImageJ version 1.54f (Jaubert-Garibay et al., 2023). The L* (lightness), a* (redness/greenness), and b* (yellowness/blueness) values of the samples were measured. In addition, ΔE (total color difference), hue angle, and chroma were calculated. The following equations were used (Darvish et al., 2022; Zhang et al., 2020):

$$∆\mathrm{E}=\sqrt{{\left({{\mathrm{L}}^{\ast }}_{\mathrm{ref}}-{\mathrm{L}}^{\ast }\right)}^2+{\left({{\mathrm{a}}^{\ast }}_{\mathrm{ref}}-{\mathrm{a}}^{\ast }\right)}^2+{\left({{\mathrm{b}}^{\ast }}_{\mathrm{ref}}-{\mathrm{b}}^{\ast }\right)}^2}$$ (3)

$${\mathrm{C}}^{\ast }=\sqrt{{{{\mathrm{a}}^{\ast }}_{\mathrm{t}}}^2+{{{\mathrm{b}}^{\ast }}_{\mathrm{t}}}^2}$$ (4)

$$\left\{\begin{array}{c}\text{for}{\mathrm{a}}^{\ast }>0\text{and}{\mathrm{b}}^{\ast }\ge 0\\ \mathrm{Hue}\text{angle}={\tan }^{-1}\left[\frac{{\mathrm{b}}^{\ast }}{{\mathrm{a}}^{\ast }}\right]\\ \text{for}{\mathrm{b}}^{\ast }<0\\ \mathrm{Hue}\text{angle}=360+{\tan }^{-1}\left[\frac{{\mathrm{b}}^{\ast }}{{\mathrm{a}}^{\ast }}\right]\end{array}\right)$$ (5)

where, L^⁎_ref, a^⁎_ref and b^⁎_ref are values for fresh beetroot and L^⁎, a^⁎ and b^⁎ are values for dried beetroot.

2.9. Chemical analyses extraction procedures

2.9.1. Betalain extraction

For each treatment, 2.5 g of control or fresh DBD-plasma-pretreated sample, or 1 g of whole dried sample, was mixed with 10 mL of 50% ethanol, shaken for 10 min, and centrifuged at 6000 rpm for 10 min. The extraction was performed twice, and the supernatants were combined to maximize betalain recovery. The combined extracts were filtered through a 0.45 μm membrane filter, and the filtrate was used for spectrophotometric analysis according to Nistor et al. (2017).

2.9.2. Antioxidant activity, Total phenolic and anthocyanin content extraction

A methanol–water extraction procedure was performed according to the protocol described by Chen et al. (2017). For each treatment, 2.5 g of control or fresh DBD-plasma pretreated sample, or 1 g of whole dried sample, was mixed with 30 mL of methanol–water solution (80,20, v/v) and refrigerated at 4 °C for 16 h. After ultrasonic extraction at 40 kHz for 30 min, the extract was centrifuged at 9000 ×g for 10 min. The supernatant was filtered through a 0.45 μm membrane filter and stored at 4 °C.

2.10. Total phenolic content (TPC)

TPC was determined according to the method described by Chong et al. (2013).v.

2.11. Antioxidant activity (AOA)

A DPPH stock solution with a concentration of 1 × 10⁻³ M was prepared by dissolving 22 mg of DPPH in 50 mL of methanol and stored at −20 °C until use. To prepare the working solution, 6 mL of the stock solution (6 × 10⁻⁶ M) was diluted with 100 mL of methanol to obtain an absorbance of 0.80 ± 0.02 at 515 nm. Subsequently, 0.1 mL of extract was mixed with 3.9 mL of DPPH solution, vortexed for 30 s, and allowed to react for 30 min before absorbance was measured at 515 nm. A control DPPH solution without extract was analyzed under the same conditions. The process was conducted in darkness (Ravichandran et al., 2013). AOA was calculated using Eq. 5:

$$\%\text{Inhibition}=\left[1-\frac{{\mathrm{A}}_{\text{sample}}}{{\mathrm{A}}_{\text{control}}}\right]\times 100$$ (6)

2.12. Total anthocyanin content

Total anthocyanin content (TAC) was assessed using the pH differential method, at a pH range of 0.5 to 0.8, as defined by Lachman et al. (2012). One milliliter of the extract was mixed separately with 10 mL of 2% HCl (pH 0.8) and 10 mL of citric acid buffer (pH 3.5), prepared from 0.2 M Na₂HPO₄ and 0.1 M citric acid. After mixing, the absorbance of each solution was measured at 520 nm against an 80:20 methanol–water blank. Using Eq. 6, the TAC was calculated as cyanidin (${\epsilon }_{1\mathit{cm}}^{1\%}$=300; 523 nm).

$$\mathit{TAC}\left(\frac{\mathit{mg}}{\mathrm{L}}\right)=\left(A_1-A_2\right)\times \mathrm{ƒ}$$ (7)

TAC denotes total anthocyanin content, whereas A₁ and A₂ represent the absorbance values in 2% HCl and citric acid buffer, respectively. TAC was expressed as mg cyanidin equivalents per 100 g sample.

2.13. Experimental design

The experimental design and optimization were performed using Design-Expert software version 11.0.3.0 (Stat-Ease Inc., Minneapolis, MN, USA). Because the investigated factors were categorical and had unequal numbers of levels, the study was designed as a randomized general full factorial experiment rather than a conventional response surface methodology design. Three independent variables were considered: CP-pretreatment time at four levels (0, 1, 3, and 5 min), microwave power at three levels (180, 360, and 600 W), and PR at two levels (1 and 2). The complete 4 × 3 × 2 factorial structure generated 24 unique treatment combinations, each performed in three replicates, resulting in 72 randomized experimental runs. The design matrix therefore comprised all possible combinations of the three factors, and the run order was randomized by the software to reduce systematic error. The data were analyzed using a two-factor interaction model to estimate the main effects and pairwise interactions among the processing variables. The measured responses included microwave drying time (MICT), hot-air drying time (HAT), total drying time (TT), MC, AOA, TAC, betaxanthin content, betacyanin content, TPC, L*, a*, b*, total color difference, hue angle, and chroma. Multi-response optimization was conducted using the desirability function approach. The processing factors were kept within their experimental ranges; MICT, HAT, TT, and total color difference were minimized; AOA, TAC, betaxanthin content, betacyanin content, TPC, L*, and a* were maximized; and MC, b*, hue angle, and chroma were constrained within their experimental ranges. The optimum condition was selected based on the highest overall desirability value, representing the best simultaneous compromise between drying efficiency and product-quality retention. The optimal samples selected by the full factorial design model were used for SEM imaging, thermal analysis, and color characterization.

2.14. Statistical analysis

A completely randomized full factorial design was used in this study. Statistical analyses were performed using SPSS software version 22.0. All experiments were performed in triplicate, and mean values were compared using Duncan's multiple range test at a significance level of p < 0.05.

3. Results and discussion

3.1. Optical emission spectroscopy (OES)

Optical emission spectroscopy (OES) is commonly used to identify reactive species in DBD-plasma. In DBD-plasma, high-energy electrons generated by electrical discharge transfer energy to gas molecules, causing ionization, excitation, and dissociation, which result in a complex mixture of ROS and RNS that play critical roles in the chemical reactivity of the CP. Fig. 1 depicts the emission spectrum of DBD-plasma at 25 kV over the wavelength range of 295.90–777.63 nm. Nitrogen (N₂) emissions were observed at 315.07, 352.62, 356.48, and 433.36 nm, indicating vibrational and rotational transitions within the C³Π_u → B³Π_g second positive system. Nitrogen ion (N₂⁺) emission at 425.94 nm indicates ionization and the production of RNS, while the strong nitric oxide (NO) peaks at 352.62 nm further confirm the generation of RNS, with a crucial role in surface interactions and oxidative processes. Oxygen-based species may play a significant role in CP chemistry, with oxygen (O) emission lines at 393.30 and 777.63 nm reflecting oxygen molecule dissociation and ROS generation; these species are highly reactive and contribute to oxidative membrane damage. Furthermore, the presence of hydroxyl radicals, revealed by the strong emission at 295.90 nm, highlights their importance in oxidative reactions due to their high reactivity and short lifetime. These findings demonstrate that DBD-plasma is an effective source of both ROS and RNS (Thana et al., 2026).

Fig. 1.

The optical emission spectrum of the Dielectric barrier discharge (DBD) cold plasma device.

3.2. Microstructural observations

After DBD-plasma pretreatment and IMHA drying, the beetroot microstructure was examined to facilitate interpretation of the following results. The optimal treatment condition was identified based on the highest overall desirability value, representing the most favorable compromise between drying efficiency and product quality obtained through full-factorial-design-based optimization. Fig. 2 depicts the surface morphology of fresh beetroot samples (A) and DBD-plasma pretreated samples (B–D). All DBD-plasma pretreated samples exhibited small, irregular surface fragments and holes, whereas the fresh sample (A) remained intact. These holes enlarged, resulting in an increase in surface area as the DBD-plasma exposure time increased from 1 to 5 min. The formation of microscopic holes and gaps, as illustrated, has been attributed to bombardment of the sample surface by reactive species and weakening of the cell wall network through rupture of surface bonds (Kamkari et al., 2024).

Fig. 2.

SEM and Micro-CT analysis of fresh pretreated samples versus the control, coupled with the effects of varying microwave power, cold atmospheric plasma duration, and pulse ratios on the microstructure of dried beetroot samples.

E=MW Power: 600 W- CP Treatment time: 5 min- PR: 1.

F=MW Power: 600 W- CP Treatment time: 5 min- PR: 2.

G=MW Power: 360 W- CP Treatment time: 5 min- PR: 2.

H=MW Power: 600 W- CP Treatment time: 0 min- PR: 2.

I=MW Power: 180 W- CP Treatment time: 5 min- PR: 2.

In Fig. 2, samples F and H were compared to evaluate the effect of DBD-plasma pretreatment. Under constant microwave power and PR conditions, increasing the DBD-plasma exposure time increased tissue porosity by approximately 27%. As illustrated in Fig. 3, the increase in porosity appears to result from multiple CP-induced effects acting simultaneously on the beetroot tissue. First, energetic particles, ROS, and RNS bombard the tissue surface, weakening intermolecular interactions such as hydrogen bonding, hydrophobic interactions, and van der Waals forces within structural biopolymers, including cellulose, hemicellulose, and pectin. First, energetic particles, ROS, and RNS bombard the tissue surface, weakening intermolecular interactions such as hydrogen bonding, hydrophobic interactions, and van der Waals forces within structural biopolymers, including cellulose, hemicellulose, and pectin. This disruption loosens the lamellar structure and promotes the formation of microholes and cavities (Jafari et al., 2025). Second, DBD-plasma generated in clean dry air produces reactive species such as ozone (O₃), hydroxyl radicals (•OH), and hydrogen peroxide (H₂O₂), which accelerate oxidative degradation of cell wall and membrane components. This process contributes to partial depolymerization and surface etching, leading to the development of porous channels (Kim et al., 1999). Third, electrostatic disruption takes place when accumulated charges within the CP flow exert stresses greater than the tensile strength of the membrane, causing membrane rupture and further enlargement of the resulting fissures (Bermudez-Aguirre, 2019; Lunov et al., 2015). Fourth, hydrogen peroxide is formed when hydroxyl radicals react with superoxide anions and protons in the high-moisture environment of beetroot tissue. As a secondary ROS, hydrogen peroxide contributes to lipid peroxidation, membrane damage, and possible electroporation (Pedrow et al., 2020). Finally, UV irradiation generated by energetic collisions within the DBD-plasma enhances oxidative degradation by increasing hydroxyl radical production and inducing photolysis, which cleaves chemical bonds in membrane and surface macromolecules. Longer exposure times intensify these effects because of the higher concentration of reactive species (Kamkari et al., 2024).

Fig. 3.

Schematic representation of the dielectric barrier discharge (DBD) plasma pretreatment system and its proposed mechanism of action on red beetroot. A: Closer look into the reaction chamber B: How cold plasma affect the cell membrane. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

As shown in Fig. 2, samples F, G, and I exhibited an 87% increase in porosity and reduced shrinkage as microwave power increased, while DBD-plasma pretreatment time and PR were kept constant. This effect may be attributed to the continuous alternation of the microwave electromagnetic field, which rapidly reorients polar molecules (e.g., water) and charged particles (e.g., ions) (Staniszewska et al., 2020). Microwave drying may also influence the orientation of polar bonds within plant tissues, including those in cellulose (β-1 → 4-d-glucose units), hemicellulose (bearing hydroxyl groups), and pectin (rich in carboxyl groups). Such reorientation can weaken hydrogen bonds and disrupt cell structure, thereby enhancing porosity (Chen et al., 2022). As shown in Fig. 2, comparison of samples E and F indicates that increasing PR reduced porosity by 7% and induced greater shrinkage. The extended drying time associated with higher PR decreased effective moisture diffusivity and vapor pressure because of the lower temperature, thereby limiting moisture removal and promoting structural degradation (Dehghannya et al., 2021).

3.3. Thermal analysis

Tg is an important indicator of the physical state and molecular mobility of amorphous food matrices. In the present study, the increase in Tg after DBD-plasma pretreatment indicates reduced molecular mobility and improved physical stability of the dried beetroot matrix. This effect is most likely associated with the combined influence of CP-induced microstructural modification, reduced water plasticization, increased solute concentration during drying, and possible reorganization of cell-wall polymers (Li, Wang, et al., 2023).

As shown in Fig. 4B, C, and E, Tg increased from −25.6 °C to −3.3 °C as microwave power increased, while the DBD-plasma pretreatment time and PR were kept constant. Fruits and vegetables undergo shrinkage when the viscoelastic matrix collapses into air-filled spaces that were previously occupied by water molecules before drying. As shown in Fig. 2, extended thermal processing intensifies this shrinkage, leading to structural deterioration of the sample (Lin et al., 2020). Higher microwave power enhanced internal heating and vapor-pressure-driven water migration, thereby shortening the drying time and limiting the time available for severe shrinkage. This can help to preserve a more porous structure and reduce water plasticization, leading to a higher Tg, which is supported by the microstructural observations showing greater porosity and less collapse in higher microwave power treated samples. Moreover, as microwave power increases, structural changes such as protein denaturation and starch gelatinization may also occur, further reducing the heat absorption capacity of the sample and increasing ∆Cp. In line with these findings, It has been reported that Tg decreases during prolonged drying, which may result in structural damage (Kahraman et al., 2021).

Fig. 4.

Glass transition temperature values of beetroot slices dried at different microwave powers (MW), DBD-plasma pretreatment time and pulse ratio (PR).

Fig. 4A and B show that, at identical microwave power and DBD-plasma pretreatment duration, Tg decreased as PR increased, from 5.5 °C at PR 1 to −3.3 °C at PR 2. This effect may be attributed to the lower internal temperature at higher PR values, together with structural changes associated with greater shrinkage (Dehghannya et al., 2021). As PR increases, heat transfer decreases because the internal temperature becomes lower than that achieved under continuous microwave application. This reduction in heat transfer, in turn, limits internal moisture migration and slows the drying rate. Under these conditions, the food matrix has sufficient time to collapse into the void spaces created by moisture removal, resulting in greater shrinkage (Fig. 2) and lower Tg values. In contrast, a lower PR promotes faster drying, although it may increase the risk of overheating. It also leads to a higher Tg, more rapid thermal transitions, and, ultimately, greater structural stability. As PR increased, ∆Cp decreased, likely because of differences in energy delivery. At lower PR, microwave energy is supplied more continuously, enhancing energy absorption and promoting structural changes such as protein denaturation and starch gelatinization.

Fig. 4 shows that DBD-plasma pretreatment increased Tg from −15.8 °C in the untreated sample (B) to −3.3 °C after 5 min (C) of pretreatment. A likely explanation is that DBD-plasma generated surface erosion, microcracks, and pores, which facilitated the removal of free and weakly bound water from the tissue. Because water is a strong plasticizer in carbohydrate-rich plant materials, its removal reduces free volume and limits molecular mobility in the amorphous phase, resulting in a shift of Tg toward higher temperatures. This finding is consistent with the more porous and disrupted microstructure observed after treatment and with previous reports linking lower moisture plasticization to higher Tg (Kahraman et al., 2021). In addition, faster moisture removal during DBD-plasma assisted drying may concentrate soluble solids such as sugars and organic acids in the remaining matrix. This concentration effect can promote a denser amorphous phase with lower free volume and greater resistance to molecular relaxation, which also contributes to the observed increase in Tg (Chaple et al., 2020; Wongsagonsup et al., 2014). Another possible explanation involves DBD-plasma–induced modifications to the surface chemistry and molecular structure of beetroot tissues. Reactive species generated during CP exposure may oxidize or partially remove waxy surface constituents, thereby increasing surface polarity and wettability and thereby facilitating moisture migration during drying. At the molecular level, the observed increase in ΔCp following DBD-plasma pretreatment may result from CP-induced oxidation that disrupts C—H, C—C, and C C bonds, leading to polysaccharide depolymerization and the formation of smaller, more hydrophilic fragments with a greater number of hydrogen-bonding sites capable of interacting with water molecules. Concurrently, oxidation-driven cross-linking of macromolecules can produce a more rigid molecular structure that requires greater energy for molecular motion and thermal transitions; collectively, these chemical and structural modifications increase the energy required for thermal rearrangements, thereby contributing to the higher ∆C_p (Chen et al., 2023; Miao et al., 2020).

3.4. Drying kinetics

Fig. 5(A–D) and Fig. 5(1–3) illustrate the drying kinetics, showing that the initial drying phase exhibited a higher moisture removal rate under IMHA than under conventional HA drying across all treatment conditions. This phenomenon can be attributed to several factors. First, during the microwave stage, moisture-rich samples absorb more energy and facilitate enhanced heat transfer through molecular friction. Simultaneously, vapor escape may be partially restricted by intact cell structures and limited surface permeability, increasing internal vapor pressure and promoting pressure-driven moisture transport toward the surface (Li, Hu, et al., 2019; Macedo et al., 2022). As presented in Fig. 5(1–3), when microwave power was increased at PR of 1, the MC of control samples decreased markedly from 0.90 to 0.87 g water/g dry solid following the microwave stage. At constant PR and DBD-plasma pretreatment time, higher microwave power intensifies dipole reorientation in the microwave electric field, resulting in more frequent collisions among polar molecules and ions as they respond to electromagnetic fluctuations. This increase in molecular collision promotes heat transfer and accelerates moisture evaporation (Destandau & Michel, 2022). A similar acceleration in drying at higher microwave power was also reported by, Yildiz and İzli (2019).

Fig. 5.

The effect of different microwave powers (180, 360 and 600 W) and cold atmospheric plasma pretreatment time (0, 1, 3, and 5 min) on the drying kinetics of beetroot samples treated at Pulse ratio (PR) = 1 (A–D & 1–3) and PR = 2 (E–H & 4–6).

Moisture removal efficiency decreased noticeably with increasing PR under identical microwave power and DBD-plasma pretreatment time. As illustrated in Fig. 5(E-H) Fig. 5(4–6) continuous microwave application removed moisture faster than intermittent microwave treatment. This difference is likely due to the higher internal temperature and vapor pressure generated during continuous microwave exposure, which enhances moisture migration. By comparison, higher PR values prolonged drying, possibly because of structural hardening or shrinkage that restricted internal moisture movement. Comparable findings were reported by Polat et al. (2019), who noted that increasing the PR reduced the moisture evaporation rate.

The effect of DBD-plasma pretreatment is shown in Fig. 5 (except A and E), where longer pretreatment times significantly enhanced moisture removal (P < 0.05) under both continuous and intermittent drying conditions. This improvement can be attributed to the interaction of RNS and ROS, ions, charged particles, and UV photons with the beetroot tissue, causing surface etching and the formation of irregular microchannels that facilitate moisture diffusion (Ashtiani et al., 2020). Additionally, electrostatic forces generated during CP exposure may disrupt cell membranes, thereby enhancing vapor release (Lunov et al., 2015). Oxidative reactions involving reactive species may further may contribute to membrane degradation and promote overall water loss (Kim et al., 1999). Similar results have been reported by Huang et al. (2019) and Khudyakov et al. (2022) for CP-pretreated white grapes and apple slices.

Overall, increasing microwave power significantly shortened the drying time (P < 0.05), whereas higher PR values extended the drying duration. Prolonged DBD-plasma pretreatment further reduced the total drying time. The longest drying time (464 min) was observed in control samples dried at 180 W with a PR of 2, whereas the shortest time (220 min) was recorded at 600 W for samples pretreated for 5 min with DBD-plasma. This combination enhanced the drying rate by 52% during the microwave stage, 33% during the HA stage, and 34% overall, compared with the control.

3.5. Betalain content

The conjugated chromophore structure of betalains makes them highly sensitive to external processing conditions, including heat, electromagnetic radiation, oxygen exposure, and pH changes. As illustrated in Fig. 7A and B, elevated microwave power levels caused a significant reduction in betaxanthin concentrations from 3.26 mg/g to 1.66 mg/g, and betacyanin from 5.33 mg/g to 3.81 mg/g (P < 0.05). These findings differ from those reported by Seremet et al. (2020) likely due to the employed a substantially shorter microwave exposure time. This difference highlights that prolonged thermal and electromagnetic stress promotes betalain degradation, as high microwave power causes intense localized heating in pigment-containing tissues and rapidly exceeds their thermal stability thresholds. Microwave heating at 600 W for 8 min can raise the medium to 50 °C, a critical threshold at which betalains begin to deteriorate (Yuan et al., 2016). Additional studies indicate near-complete loss of betanin at 99 °C after 60 min and significant degradation at 45 °C, highlighting the thermal sensitivity of these pigments (Fu et al., 2020; Muramatsu et al., 2023). Heating promotes C—C bond cleavage at the C-15 or C-17 carboxyl positions, releasing CO₂ and forming resonance-stabilized decarboxylated derivatives. Simultaneously, thermal degradation through pathways such as oxidation and dehydrogenation can convert betanin to neobetanin by increasing unsaturation within the chromophore. These reactions alter π-electron delocalization, shift visible absorption, and reduce the red-violet color intensity (Kujala et al., 2000; Stintzing & Carle, 2004). Concurrently, under thermal conditions, the O-glycosidic bond of betanin might undergo acid-catalyzed hydrolysis. Protonation of the glycosidic oxygen increases the susceptibility of the C—O bond to cleavage, releasing glucose and the aglycone betanidin. Because betanidin lacks the stabilizing glycosyl group, it is more chemically sensitive and highly susceptible to subsequent oxidation, leading to further degradation and loss of color intensity (Miyagawa et al., 2021). Furthermore, oxidative cleavage of the core aldimine bond produces colorless or pale-yellow byproducts, specifically cycloDOPA-5-O-glycoside and betalamic acid, resulting in a severe loss of the red-violet hue. Lastly, elevated temperatures drive the reversible C-15 epimerization of betanin into isobetanin, an isomerization process that alters light absorption profiles and shifts the matrix's chromatic equilibrium (Otálora González et al., 2020). In contrast, microwave heating significantly accelerates betalain degradation through electromagnetic stress generated by high-frequency oscillating fields, which induce continuous forced dipolar rotation of water molecules and ions. This interaction intensifies rapid internal heating and molecular agitation, thereby accelerating degradation reactions, facilitating molecular breakdown in a synergistic thermal-electromagnetic degradation pattern consistent with the kinetics reported by Liu et al. (2021).

Fig. 7.

Betalains, Total Phenolic Content, Antioxidant Activity and Total Anthocyanin Content of beetroot samples under the influence of different cold atmospheric plasma pretreatment times of 0 (NP), 1 (P-1), (P-3) 3, and (P-5) 5 min in different microwave powers of (180, 360 and 600 W) with pulse ratio of 1 a and 2. * Different letters show a statistically significant difference at a confidence level of 95% (p < 0.05). * Capital letters indicate a significant difference in cold atmospheric plasma pretreatment times. * Lowercase letters indicate significant differences in MW power. * Pulse 1 and Pulse 2 have a significant difference from each other at a confidence level of 95% (p < 0.05).

Increasing PR from 1 to 2 proved to be an effective means of limiting thermal degradation under the same microwave power and DBD-plasma pretreatment time. This adjustment led to an 18% rise in betaxanthin content (3.26 to 3.85 mg/g) and a 13% increase in betacyanin content (5.22 to 5.88 mg/g), highlighting the strong protective effect of pulsed microwave application on betalain stability. This preservation effect is attributed to the temporal control of heat transfer provided by a higher PR. By introducing magnetron rest periods, localized heat redistributes uniformly throughout the cellular matrix via conduction (Duc Pham et al., 2019). This more uniform heat redistribution prevents thermal hotspots, charring, excessive temperature rise, and the attainment of degradation thresholds during drying, thereby limiting hydrolytic and decarboxylation reactions. Consequently, this finding demonstrates that a higher PR might be beneficial for preserving heat-sensitive bioactive compounds. It has been reported that increasing the PR from 1 (continuous mode) to 6 during microwave drying of potato resulted in the lowest color change and the highest color quality. This improvement was attributed to shorter microwave on-times, reduced hot-spot formation, lower pigment degradation and isomerization, and slower Maillard browning reactions (Dehghannya & Habibi-Ghods, 2025).

DBD-plasma pretreatment significantly increased the measured betalain content in both fresh and dried samples (P < 0.05). Extending exposure time from 1 to 5 min increased betaxanthin and betacyanin concentrations in fresh samples by 43.93% and 55.20%, respectively, compared with the control. Although the same extraction procedure was applied to all samples and the control was not subjected to thermal treatment, the higher betalain content observed in the DBD-plasma pretreated sample may be attributed to increased cell membrane permeability, which likely enhanced mass transfer and improved pigment extraction efficiency. Furthermore, combining 5 min of DBD-plasma pretreatment with microwave power of 180 W produced a synergistic effect, increasing betaxanthin by 59% and betacyanin by 57.32% compared with untreated dried control samples (Fig. 7A and B). Taken together, the data indicate that DBD-plasma pretreatment improved betalain retention mainly by altering tissue microstructure and facilitating moisture transport during drying. During CP-exposure, ROS, RNS, ionized particles, and UV photons interact with the tissue surface, disrupting cell wall components such as hemicellulose–pectin complexes, cell wall proteins, and the middle lamella, thereby weakening intermolecular bonds within the cellular matrix and promoting the formation of micropores and microchannels. These microstructural changes increase cell membrane permeability, create a more porous structure, and reduce resistance to moisture and solute transport, thereby facilitating faster vapor release during drying, shortening thermal exposure, and consequently reducing tissue shrinkage and limiting thermal degradation of betalains. At the same time, the enhanced structural accessibility improves solvent penetration and pigment release during extraction, which can increase betalain extraction efficiency. This combined effect of CP-induced microstructural modification and enhanced mass transfer during drying has been reported previously (Kamkari et al., 2024; Kashfi et al., 2020). In addition to its physical effects on tissue structure and mass transfer, DBD-plasma pretreatment has been reported to influence betalain stability through changes in pH. CP-generated reactive species may react with surface moisture to form acidic compounds, including nitric, nitrous, and carbonic acids, along with hydrogen radicals, thereby lowering the pH of the medium (Ramezan et al., 2024). Under mildly acidic conditions, betalains may be stabilized to some extent; however, excessive or localized acidification can also trigger degradation pathways. These include acid-catalyzed deglycosylation and epimerization, which generate unstable aglycones prone to further degradation, as well as hydrolysis of the Schiff-base linkage between cycloDOPA and betalamic acid, leading to the formation of colorless precursors (Khan, 2016). Conversely, acidification may also promote re-condensation between cycloDOPA-5-O-glucoside and betalamic acid, reorganizing the conjugated chromophore system and producing a bathochromic shift that changes the absorption profile and shifts the visible color from red to violet (Sadowska-Bartosz & Bartosz, 2021). Overall, the influence of CP-induced acidification on betalains appears to depend on its severity. Mild acidification may help preserve pigments, whereas excessive localized acidification may promote structural transformation or degradation. At the same time, CP-induced microchannels and enhanced mass transfer can improve extraction efficiency and reduce heat-induced pigment loss.

3.6. Color

Color in fruits and vegetables is primarily attributed to natural pigments like flavonoids, betalains, and carotenoids with notable biological activity as secondary metabolites (Fallah & Stich, 2024). Owing to their safety and vibrant hues, these plant-derived compounds are favored over synthetic colorants and significantly influence consumer acceptance in food selection (Oplatowska-Stachowiak & Elliott, 2017; Zhang et al., 2024). The statistical analysis presented in Table 2 and Fig. 6(A–E) demonstrates that microwave power, PR, and DBD- plasma pretreatment significantly affected the colorimetric parameters of the dried beetroot samples, including L*, a*, b*, hue angle, chroma, and ΔE. These macroscopic color changes are intrinsically linked to the chemical stability of the betalain molecules during dehydration, as detailed mechanistically in Section 3.5.

Table 2.

Color parameters characteristics resulted from cold atmospheric plasma pretreatment applied at 0, 1, 3, and 5 min with 180, 360, and 600 W of microwave power and pulse ratios of 1 and 2.

Cold plasma time (min)	MW power (W)	Pulse ratio	L⁎	a⁎	b⁎	ΔE	Hue	Chroma
0 (Control)	0	0	34.04ᴰ ± 0.28	27.70ᴰ ± 0.05	3.53ᴰ ± 0.10	0.00ᴬ	7.26ᴰ ± 0.18	27.92ᴰ ± 0.06
1 min	0	0	29.79ᶜ ± 1.84	28.52ᶜ ± 0.08	-0.42ᶜ ± 0.30	5.18ᴮ ± 1.80	359.15ᶜ ± 0.59	28.52ᶜ ± 0.83
3 min	0	0	26.68ᴮ ± 0.19	29.69ᴮ ± 11.16	-4.78ᴮ ± 0.02	10.65ᶜ ± 3.07	350.85ᴮ ± 40.15	30.07ᴬ ± 9.20
5 min	0	0	22.31ᴬ ± 3.03	32.15ᴬ ± 2.61	−8.64ᴬ ± 1.29	16.82ᴰ ± 3.90	344.95ᴬ ± 1.03	33.29ᴮ ± 2.91
0 (Control)	180	1	31.11ᴰᵃ ± 1.00	29.34ᴰᵇ ± 0.89	1.57ᴰᵃ ± 0.39	3.20ᴬᵇ ± 0.38	3.06ᴰᵃ ± 0.37	29.38ᴰᵇ ± 0.97
1 min	180	1	28.09ᶜᵃ ± 0.83	30.03ᶜᵇ ± 1.34	−2.92ᶜᵃ ± 0.64	3.38ᴮᵇ ± 1.68	354.43ᶜᵃ ± 5.01	30.17ᶜᵇ ± 0.65
3 min	180	1	24.53ᴮᵃ ± 1.27	32.74ᴮᵇ ± 0.78	−7.56ᴮᵃ ± 1.26	4.65ᶜᵇ ± 1.10	346.98ᴮᵃ ± 3.31	33.60ᴬᵇ ± 1.23
5 min	180	1	19.65ᴬᵃ ± 0.84	34.79ᴬᵇ ± 0.73	−12.66ᴬᵃ ± 0.39	5.49ᴰᵇ ± 1.04	340.00ᴬᵃ ± 0.89	37.02ᴮᵇ ± 0.76
0 (Control)	360	1	33.87ᴰᵇ ± 1.39	28.36ᴰᵃ ± 1.33	2.57ᴰᵃ ± 0.60	1.33ᴬᵇ ± 0.75	5.53ᴰᵃ ± 2.61	28.49ᴰᵃ ± 1.20
1 min	360	1	31.67ᶜᵇ ± 1.16	28.38ᶜᵃ ± 0.67	−0.11ᶜᵃ ± 0.54	1.91ᴮᵇ ± 1.04	359.77ᶜᵃ ± 2.01	23.38ᶜᵃ ± 0.72
3 min	360	1	28.24ᴮᵇ ± 1.33	28.21ᴮᵃ ± 0.21	−5.61ᴮᵃ ± 1.71	2.30ᶜᵇ ± 0.50	348.74ᴮᵃ ± 2.85	28.76ᴬᵃ ± 1.62
5 min	360	1	24.10ᴬᵇ ± 0.22	30.53ᴬᵃ ± 0.27	−9.75ᴬᵃ ± 0.16	2.65ᴰᵇ ± 0.11	342.28ᴬᵃ ± 0.14	32.04ᴮᵃ ± 0.31
0 (Control)	600	1	36.81ᴰᶜ ± 0.98	27.75ᴰᶜ ± 1.34	4.03ᴰᵇ ± 0.40	3.11ᴬᵃ ± 0.63	8.26ᴰᵇ ± 1.12	28.04ᴰᵇ ± 1.32
1 min	600	1	32.33ᶜᶜ ± 0.06	27.26ᶜᶜ ± 0.00	0.74ᶜᵇ ± 0.31	3.05ᴮᵃ ± 0.40	1.55ᶜᵇ ± 0.09	27.27ᶜᵇ ± 0.31
3 min	600	1	30.19ᴮᶜ ± 0.95	27.42ᴮᶜ ± 0.44	−3.01ᴮᵇ ± 1.08	4.53ᶜᵃ ± 1.10	353.72ᴮᵇ ± 2.84	27.59ᴬᵇ ± 0.97
5 min	600	1	25.42ᴬᶜ ± 0.46	28.72ᴬᶜ ± 0.38	−8.02ᴬᵇ ± 0.36	4.65ᴰᵃ ± 0.60	344.39ᴬᵇ ± 2.34	29.81ᴮᵇ ± 0.21
0 (Control)	180	2	29.92ᴰᵃ ± 0.24	29.88ᴰᵇ ± 1.59	0.34ᴰᵃ ± 0.69	5.41ᴬᵇ ± 1.38	359.38ᴰᵃ ± 0.00	29.88ᴰᵇ ± 1.72
1 min	180	2	27.67ᶜᵃ ± 0.43	31.25ᶜᵇ ± 0.39	−4.05ᶜᵃ ± 0.76	5.01ᴮᵇ ± 0.30	352.61ᶜᵃ ± 0.55	31.51ᶜᵇ ± 0.85
3 min	180	2	22.88ᴮᵃ ± 0.85	32.62ᴮᵇ ± 1.42	−8.37ᴮᵃ ± 0.93	5.99ᶜᵇ ± 1.50	345.60ᴮᵃ ± 0.30	33.68ᴬᵇ ± 1.70
5 min	180	2	17.75ᴬᵃ ± 1.27	34.70ᴬᵇ ± 1.05	−13.73ᴬᵃ ± 0.15	7.29ᴰᵇ ± 1.44	338.40ᴬᵃ ± 1.07	37.32ᴮᵇ ± 1.02
0 (Control)	360	2	32.63ᴰᵇ ± 0.99	28.16ᴰᵃ ± 1.10	1.33ᴰᵃ ± 0.22	2.31ᴬᵇ ± 1.38	2.64ᴰᵃ ± 0.45	28.18ᴰᵃ ± 1.12
1 min	360	2	29.26ᶜᵇ ± 0.63	29.05ᶜᵃ ± 0.57	−2.37ᶜᵃ ± 0.29	2.08ᴮᵇ ± 0.89	355.33ᶜᵃ ± 2.34	29.15ᶜᵃ ± 0.18
3 min	360	2	27.37ᴮᵇ ± 0.18	31.31ᴮᵃ ± 0.15	−6.40ᴮᵃ ± 0.24	2.38ᶜᵇ ± 0.16	348.43ᴮᵃ ± 0.30	31.94ᴬᵃ ± 0.28
5 min	360	2	25.11ᴬᵇ ± 0.07	33.02ᴬᵃ ± 0.39	−10.39ᴬᵃ ± 0.64	3.41ᴰᵇ ± 0.30	342.53ᴬᵃ ± 0.53	34.62ᴮᵃ ± 0.74
0 (Control)	600	2	35.61ᴰᶜ ± 0.83	26.43ᴰᶜ ± 0.73	2.27ᴰᵇ ± 0.00	3.13ᴬᵃ ± 0.93	4.90ᴰᵇ ± 0.11	26.52ᴰᵇ ± 0.73
1 min	600	2	31.42ᶜᶜ ± 0.18	26.00ᶜᶜ ± 0.04	−1.23ᶜᵇ ± 0.59	3.10ᴮᵃ ± 0.04	357.29ᶜᵇ ± 0.00	26.02ᶜᵇ ± 0.59
3 min	600	2	30.77ᴮᶜ ± 0.41	29.48ᴮᶜ ± 0.39	−4.58ᴮᵇ ± 1.76	4.10ᶜᵃ ± 1.09	351.16ᴮᵇ ± 2.52	29.83ᴬᵇ ± 1.66
5 min	600	2	27.57ᴬᶜ ± 1.62	31.15ᴬᶜ ± 0.40	−8.67ᴬᵇ ± 3.37	5.35ᴰᵃ ± 1.14	344.44ᴬᵇ ± 1.99	32.33ᴮᵇ ± 0.28

^⁎Different letters show a statistically significant difference at a confidence level of 95% (p < 0.05).

^⁎Capital letters indicate a significant difference in cold plasma pretreatment times.

^⁎Lowercase letters indicate significant differences in MW power.

^⁎ Pulse 1 and Pulse 2 have a significant difference from each other at a confidence level of 95% (p < 0 0.05).

Fig. 6.

Schematic representation of color changes in beetroot samples under the influence of different processing variables (MW power, cold atmospheric plasma duration and PR = 1, 2) and proposed mechanisms governing the observed pigment alteration.

As shown in Fig. 6(B and D), increasing the pretreatment duration caused pronounced darkening, with L* decreasing from 36.32 to 20.07, and a distinct tonal shift, with a* increasing from 26.38 to 31.82 and b* decreasing from 5.06 to −9.16 (P < 0.05). This shift is consistent with the reduced betacyanin and betaxanthin contents observed at high microwave power (Sadowska-Bartosz & Bartosz, 2021; Zannou et al., 2024), resulting in loss of red-purple intensity and a lighter yellowish appearance. The quantitative shift in hue angle from 3.06 (180 W) to 8.26 (600 W) and the decrease in chroma visually confirm this pigment degradation. Furthermore, higher ΔE values at 600 W underscore a substantial visual variation from the reference sample, primarily reflecting the pigment losses quantified in Section 3.5.

Conversely, increasing the PR effectively mitigate visual degradation by improving thermal control. As shown in Fig. 6 (A and B), higher PR settings resulted in significantly improved color retention, characterized by higher a* values and lower L* and b* coordinates (P < 0.05). This effect is consistent with the higher betalain retention observed under pulsed microwave operation, which reduced the overall color difference compared with continuous or more intense heating conditions (Section 3.5) (Aghilinategh et al., 2015).

DBD-plasma pretreatment introduced modification to the beetroot matrix, significantly altering its colorimetric profile. Considering Fig. 6 (B and D), as pretreatment duration increased, samples exhibited pronounced darkening (L* decreasing from 36.32 to 20.07) and a distinct shift in tone, with a* increasing from 26.38 to 31.82, and b* moving from 5.06 to −9.16 (P < 0.05). This shift toward the blue-violet region is further confirmed by the hue angle transition from 7.26° to 344.95°. These changes correspond well with the increased measurable betalain content in CP-pretreated samples, indicating that CP-induced structural modification affected both pigment extractability and visual color expression (Ramezan et al., 2024). DBD-plasma has also been investigated in wolfberries, where it increased the a* value and enhanced redness. However, unlike the response observed in beetroot, both L* and b* also increased in wolfberries, suggesting a lighter and more yellowish appearance (Du et al., 2024). This discrepancy may be attributed to differences in pigment composition and tissue-level responses to CP-treatment. In berries, the predominant pigments are anthocyanins and/or carotenoids (Olas, 2018), whereas beetroot color is primarily governed by betalains. Because betalains are highly sensitive to pH changes, acidification, and CP-induced structural modifications, DBD-plasma pretreatment in beetroot may alter pigment stability as well as the surrounding cellular environment.

3.7. Total phenolic content

Fig. 7C shows that, under constant DBD-plasma pretreatment and PR conditions, increasing microwave power reduced the TPC of the control samples from 4.28 to 3.86 mg GAE/g. This reduction may be attributed to microwave-induced dielectric heating, in which dipole rotation and ionic conduction can promote thermal degradation and structural alteration of phenolic compounds (Dudley et al., 2015; Galanakis et al., 2013). The main classes of phenolic compounds reported in red beet include flavonoids, phenolic acids, and phenolic amides, many of which contain hydroxyl groups and exhibit polar characteristics. Phenolic compounds reported in beetroot include caffeic acid, catechin, coumaric acid, vanillic acid, ferulic acid, cinnamic acid, protocatechuic acid, and syringic acid, many of which contain hydroxyl groups that contribute to their polarity and AOA (Alara et al., 2021; Boskou et al., 2006). Phenolic compounds exhibit chemical sensitivity to elevated temperatures caused by microwave energy, resulting in degradation through decarboxylation, dehydroxylation, and oxidation, which produces quinones and low molecular weight phenolics such as vanillic and protocatechuic acids. High-temperature drying or extended exposure leads to the hydrolysis of phenolic esters and glycosides. Concurrently, free radicals (e.g., •OH, O₃) generated through thermal or CP-induced processes facilitate ring cleavage and polymerization (Salazar-Orbea et al., 2023; Santhirasegaram et al., 2015).

As shown in Fig. 7C, increasing the PR from 1 to 2 enhanced the measured TPC across all treatments. Specifically, at constant microwave power of 180 W and a fixed DBD-plasma pretreatment time, the TPC of the control sample increased by 17.89%, from 4.28 to 5.05 mg GAE/g, as the PR increased. This improvement may result from intermittent microwave drying intervals, which promote more uniform heat distribution and mitigate the detrimental effects of excessive temperature rise (Pham & Karim, 2022). In contrast previous findings for apple slices dried by intermittent microwave drying (IMWD) showed that TPC increased significantly with increasing microwave power and air velocity, while higher drying temperature and PR led to a marked decrease in TPC. The reduction in TPC at higher temperatures and longer PR was attributed to extended exposure to heat and atmospheric oxygen, which can accelerate degradation of phenolic compounds (Aghilinategh et al., 2015).

As shown in Fig. 7C, increasing DBD-plasma exposure initially increased the TPC of fresh samples by more than twofold. Compared with the control, samples treated with DBD-plasma for 3 min and dried at 180 W with PR = 1 showed a statistically significant 43.70% increase in TPC, from 4.28 to 7.60 mg GAE/g (P < 0.05). The observed increase can be attributed to DBD-plasma–induced structural and biochemical modifications within the beetroot tissue. During CP-pretreatment, ROS and RNS may partially oxidize cell walls and membrane structures, cleave polysaccharide linkages, disrupt lipid bilayers, and weaken interactions between phenolic compounds and structural polymers such as cellulose, hemicellulose, lignin, and pectin. These oxidative modifications destabilize the plant matrix, increasing tissue permeability and facilitating the release of phenolic compounds that are originally bound to, or entrapped within, the cellular network, thereby enhancing their extractability and measured concentration (Shirazi et al., 2025).

In addition to these structural effects, DBD-plasma-induced abiotic stress may activate phenylalanine ammonia-lyase (PAL) and stimulate the phenylpropanoid biosynthetic pathway, thereby further contributing to phenolic accumulation in treated tissues (Tabibian et al., 2020). TPC decreased in both fresh and dried samples when the DBD-plasma exposure time reached 5 min, indicating that excessive CP-pretreatment promoted phenolic degradation. This decline may occur through at least two pathways. First, highly reactive species generated by DBD-plasma, particularly hydroxyl radical and singlet oxygen, can oxidize phenolic compounds by abstracting hydrogen atoms from hydroxyl groups and destabilizing aromatic structures. The susceptibility of individual phenolics depends on their molecular structure; for example, glycosylated derivatives may show greater resistance because steric hindrance and sugar substitution reduce hydrogen abstraction and retard oxidative transformation. Second, CP-generated ozone can react with phenolic molecules through ozonide formation, followed by decomposition into reactive carbonyl oxides and other intermediates. These reactions alter aromatic ring integrity, generate hydroxylated derivatives, and ultimately decompose phenolic compounds into simpler oxidation products (Kumar et al., 2023). Therefore, the effect of DBD-plasma on TPC appears to be dose-dependent, with moderate pretreatment enhancing phenolic release and extractability, whereas excessive pretreatment promotes oxidative losses. Similarly, Zhou et al. (2020) reported a dose-dependent results of TPC during hot-air drying of wolfberry after atmospheric CP-pretreatment. They noted that the phytochemical content of the product may be influenced by several factors, including input gas type, applied voltage, CP pretreatment time, and sample variety. In their study, the highest TPC was observed after 30 s of CP exposure (15 mg GAE/g DM), whereas longer pretreatment durations reduced TPC to 14.5 and 13.5 mg GAE/g DM at 45 and 60 s, respectively.

3.8. Total anthocyanin content

Fig. 7D shows that, under constant DBD-plasma pretreatment and PR conditions, increasing microwave power significantly reduced TAC in the control samples from 103.10 to 78.10 mg cyanidin/100 g (P < 0.05). In response to heat stress, plants may regulate anthocyanin levels through changes in biosynthesis, accumulation, and degradation. Exposure to elevated temperatures can promote anthocyanin degradation, particularly when tissue temperature exceeds critical stability thresholds of above 15 °C(Zhao et al., 2021). Increased microwave power amplifies thermal and electromagnetic effects, thus expediting the degradation of highly polar anthocyanins via enhanced dielectric heating. This process involves electromagnetic oscillations generated by microwave energy, which repeatedly reorient polar and charged molecules. Molecular reorientation leads to collisions and rotational stresses, which weaken intramolecular bonds, especially those susceptible to dielectric heating mechanisms (Dudley et al., 2015; Teng et al., 2020). Thermal hydrolysis can cleave glycosidic linkages in anthocyanins, leading to the formation of anthocyanidins, or aglycones. Because of the lack of sugar moieties, these aglycones are more chemically unstable and prone to oxidative destruction. These aglycones participate in oxidative reactions, ring-opening mechanisms, and polymerization, leading to decreased color intensity and AOA. Major degradation pathways include C-ring oxidative cleavage, formation of chalcone intermediates, and breakdown into phenolic acids, such as protocatechuic acid, particularly in cyanidin-based anthocyanins (Charmongkolpradit et al., 2021; Enaru et al., 2021). A study on microwave-assisted drying of blueberry showed that all drying treatments caused substantial anthocyanin degradation, reducing total monomeric anthocyanins by approximately 70%–95% relative to the fresh fruit. Among the evaluated methods, hot-air convective drying at 90 °C followed by microwave vacuum drying (MWVD) was identified as the most effective treatment, yielding the highest anthocyanin retention and AOA. In contrast, direct MWVD shortened drying time but adversely affected anthocyanin stability (Zielinska & Michalska, 2016).

As shown in Fig. 7E, TAC increased by 26.61%, from 103.30 mg cyanidin/100 g in the control sample dried at 180 W and PR = 1 to 130.79 mg cyanidin/100 g at PR = 2, under constant microwave power and DBD-plasma pretreatment conditions. Intermittent microwave application reduces heat-induced damage by incorporating successive off-periods that allow partial heat dissipation and temperature equilibration (Duc Pham et al., 2019). In comparison with CMD, IMD has been reported to provide better retention of total anthocyanins, particularly at a PR of 1/6, corresponding to 5 s microwave-on and 30 s microwave-off periods. Under these conditions, the TAC of dried samples ranged from 5.73 for CMD to 12.22 mg/100 g dry basis for IMD. This improved preservation was attributed to the pulsed microwave regime, in which the extended off-time reduced heat accumulation and minimized localized overheating. Consequently, degradation of heat-sensitive anthocyanins was limited, leading to greater pigment stability during drying (Zhao et al., 2016).

All fresh samples showed a significant increase in TAC (P < 0.05) as DBD-plasma pretreatment time increased, with TAC rising by 78.08% from 1 to 5 min of exposure. The TAC of dried samples pretreated with DBD-plasma for 5 min and dried at 180 W with PR = 1 was 62.07% higher than that of the corresponding control, increasing from 103.30 to 167.42 mg cyanidin/100 g. The significant increase in TAC in DBD-plasma pretreated beetroot samples may be attributed to CP-induced rupture of cell membranes and vacuolar structures, which facilitates the release of anthocyanins, phenolic compounds, and flavonoids. This structural disorder increases intracellular diffusion and solvent penetration, improving mass transfer and anthocyanin extraction during analysis (Shishir et al., 2020; Zhang et al., 2019). It has been reported that anthocyanins are more stable and exhibit a vivid red color at pH < 3, become less intensely colored or nearly colorless at pH 3–6, and show instability with a bluish hue at pH > 6(Chung et al., 2017). DBD-plasma reactive species may interact with moisture present on beetroot slices to generate acidic compounds, thereby lowering the medium's pH; the resulting acidification, intensified by longer CP-pretreatment times, can enhance anthocyanin stability (Ramezan et al., 2024). CP can enhance the extraction of bioactive compounds, including alkaloids, flavonoids, polysaccharides, amino acids, and hydroxycinnamic acids, by disrupting cell walls and increasing membrane permeability through reactive species and localized electric fields. It also indirectly stabilizes anthocyanins by promoting co-pigmentation with extracted compounds like flavonoids and hydroxycinnamic acid. These non-covalent interactions (π–π stacking, hydrogen bonding, electrostatic forces) protect the flavylium cation from degradation, enhance color intensity, shift hue toward blue, and improve thermal and pH stability. CP further facilitates co-pigmentation by modifying molecular structures and surface charge (Ramezan et al., 2024). Moreover, the enhanced stability of anthocyanins under extended DBD-plasma pretreatment may be associated with glycosylated forms, which generally degrade more slowly than aglycone counterparts; however, compounds such as rutin and quercetin glycosides are flavanols rather than anthocyanins and should not be presented as anthocyanin examples (Grzegorzewski et al., 2010; Kovačević et al., 2016). Moreover, extended DBD-plasma pretreatment may also stimulate the phenylpropanoid pathway, enhancing metabolic flow and augmenting anthocyanin accumulation. Increased UV exposure during prolonged DBD-plasma treatment may activate PAL, an enzyme involved in the early stages of the phenylpropanoid pathway, thereby increasing the availability of intermediates required for flavonoid and anthocyanin biosynthesis (Tabibian et al., 2020), thereby intensifying the availability of intermediates required for flavonoid and anthocyanin synthesis (Chouhan et al., 2017; Chung et al., 2017). The effect of CP treatment on anthocyanin extraction from haskap berries has also been investigated. CP-treatment increased the oxygen content on the haskap surface, leading to greater surface roughness, degradation of cell wall components, and improved hydrophilicity and hydration properties. These changes enhanced interactions between the haskap matrix and the extraction solvent, thereby promoting anthocyanin solubilization and increasing extraction efficiency by 3.35%–20.47% (Li, Li, et al., 2023).

3.9. Antioxidant activity

Fig. 7D demonstrates that, at constant PR and DBD-plasma pretreatment duration, AOA in dried control samples decreased significantly from 52.70% at 180 W to 23.02% at 600 W (P < 0.05). Beetroot's phenolic constituents, hydroxycinnamic acids (caffeic, ferulic, p-coumaric), gallic acid, and flavonoids (rutin, quercetin), enhance AOA and color stability via radical scavenging and metal chelation, but might be degraded under high drying temperatures, diminishing their functional properties (Carrillo et al., 2019; Oliveira et al., 2016). microwave drying can affect antioxidant compounds through both thermal and non-thermal mechanisms. Thermally, elevated temperatures accelerate the degradation of antioxidant compounds; non-thermally, dipole rotation and electromagnetic field interactions may disrupt hydrogen bonding and induce conformational changes in polar molecules, thereby destabilizing bioactive compounds (Zhou et al., 2021). Additionally, increasing microwave power increases the porosity of dried beetroot tissue (Fig. 2), enhancing oxygen diffusion and thus exposing released antioxidants to oxidative degradation (Saha et al., 2019). Similar thermal effects were reported in dried cherry, where elevated microwave power led to phenolic breakdown and diminished AOA (Lech et al., 2015).

As indicated in Fig. 7D, increasing PR significantly increased AOA by 8.62% under constant microwave power and DBD-plasma pretreatment duration (P < 0.05). The AOA of dried control samples at 180 W ranged from 52.70% at PR = 1 to 57.24% at PR = 2. The retention of AOA during intermittent microwave drying might be attributed to periodic heat dissipation and moisture redistribution, which reduce thermal stress and protect heat-sensitive compounds like polyphenols and flavonoids. Rest intervals minimize localized overheating and oxidative damage, creating a milder environment than continuous microwave exposure (Li et al., 2021). Compared with CMD of sweet potato, IMD at a PR of 1/6 was reported to achieve superior radical scavenging activity. Under these conditions, the radical scavenging activity of dried samples ranged from 45.80% for CMD to 53.02% for IMD. This effect was attributed to the longer off-periods of IMD, which promoted more uniform heat distribution and reduced cumulative thermal and oxidative stress during drying, thereby limiting phenolic degradation and enhancing radical scavenging activity (Zhao et al., 2016).

Prolonged DBD-plasma pretreatment significantly increased AOA in both fresh and dried samples (P < 0.05; Fig. 7D). Compared with the fresh control samples, increasing DBD-plasma exposure time to 5 min resulted in a 28.96% increase in AOA in fresh treated samples. At the optimal condition tested, namely drying at 180 W with 5 min DBD-plasma pretreatment and PR = 1, AOA increased by 33.80% compared with the corresponding dried control sample. SEM analysis reveals that DBD-plasma reactive species cause etching on cell wall surfaces, resulting in the creation of holes and microchannels that facilitate the oxidation of cell wall bonds and lead to structural rupture; prolonged exposure exacerbates these changes, thereby augmenting the release of secondary metabolites including flavonoids and phenolic compounds, ultimately elevating the AOA. This structural disruption may also increase AOA by releasing phenolic compounds that were previously bound or poorly extractable within the cell-wall matrix. Once released, these redox-active molecules become more accessible in the extract, and their hydroxyl groups can more effectively donate electrons or hydrogen atoms in antioxidant assays. Moreover, DBD-plasma pretreatment provide porous structure which might enhanced extractability of phytochemical compounds (Jafari et al., 2025). Numerous studies have explored the effect of wound-induced responses on AOA, and the increase in AOA after 3 min of DBD-plasma pretreatment may be linked to the accumulation of specific polyphenols in response to tissue stress (Fernando Reyes et al., 2007; Surjadinata & Cisneros-Zevallos, 2012), which these findings were align with those reported by Tappi et al. (2018). Although prolonged DBD-plasma pretreatment generates oxidative species, such as ROS, RNS, ozone, nitric oxide, and nitrogen dioxide, which may promote phenolic oxidation, AOA increased with longer DBD-plasma pretreatment in the present study. This could be explained by reactive species oxidizing catechins, leading to the formation of procyanidins, such as trimers, which exhibit greater AOA (Sruthi et al., 2022; Tappi et al., 2018). Betalains' antioxidant properties is primarily mediated via proton-coupled electron transfer (PCET) processes facilitated by the 1,7-diazaheptamethinium scaffold, which allows for the simultaneous proton and electron donation to neutralize free radicals like DPPH. Within this class, betaxanthins contribute through their Schiff base (—C N—) structures and adjacent hydroxyl or amino groups, while betacyanins rely on the hydroxyl (—OH) groups present on their aromatic rings (Bastos & Gonçalves, 2017; Nakagawa et al., 2018). Like to betalains, the measured anthocyanin content increased after 5 min of DBD-plasma pretreatment, which may partly explain the observed increase in AOA. Similarly, Du et al. (2026) Similarly, Du et al. (2026) observed that, among various physical pretreatments, including ultrasound and pulsed electric field treatment, DBD-treated goji berries exhibited the highest overall AOA. DBD markedly enhanced ABTS scavenging capacity and preserved DPPH activity. This was primarily attributed to the enrichment of TPC and flavonoids content, a reduction in drying time, and the inhibition of enzymatic degradation, which collectively facilitated the release and retention of bioactive compounds.

4. Conclusion

This study demonstrated that DBD-plasma pretreatment can be effectively integrated with IMHA drying to intensify beetroot dehydration while influencing the retention of quality-related compounds. The results suggest that CP-generated reactive species induced surface etching, pore formation, and microstructural modification, thereby improving moisture diffusion and reducing drying resistance. Beyond shortening drying time, DBD-plasma pretreatment also affected the preservation and extractability of bioactive pigments and antioxidant compounds, indicating that this technology can be used to regulate the balance between drying efficiency and product quality. The findings further showed that increasing microwave power markedly shortened drying time, whereas increasing PR improved the retention of heat-sensitive compounds. The shortest drying time was achieved at 600 W with 5 min DBD-plasma pretreatment, reducing the total drying time from 464 to 220 min. Therefore, the most suitable operating condition depends on the intended application: high microwave power drying may be preferred when rapid dehydration is the main industrial goal (5 min DBD-plasma pretreatment combined with 600 W and PR = 1), while milder microwave conditions combined with a higher PR are more appropriate when retention of color and functional compounds is prioritized (5 min DBD-plasma pretreatment at 180 W with PR = 2). The broader significance of this work is that DBD-plasma assisted IMHA drying provides a promising strategy for producing dried beetroot with improved drying efficiency and potentially enhanced functional quality. Nevertheless, the study was limited to a specific set of operating conditions and one product geometry; therefore, further validation under broader conditions and larger-scale systems is still required.

The following are the supplementary data related to this article.

Supplementary Fig. S1.

A schematic of microwave drying mechanism.

Supplementary Fig. S2.

Red beetroot arrangement in cold atmospheric plasma dielectric barrier discharge device (Side view angle).

Supplementary Fig. S3.

Red beetroot arrangement in cold atmospheric plasma dielectric barrier discharge device (Top view angle).

Supplementary Video S1

Video demonstrating the operation of the cold atmospheric plasma dielectric barrier discharge (CAP-DBD) device in red Beetroot pretreatment.

Future research directions

While this study establishes the high efficacy of DBD-plasma IMHA hybrid drying, future research must address key areas for industrial scaling. First, advanced analyses are needed to elucidate the structural alterations and degradation pathways of betalains caused by DBD-plasma induced acidification. Furthermore, the efficacy of alternative working gases, such as noble gases, should be evaluated and compared to the ambient air used here. Finally, assessing long-term interactions with active packaging and conducting comprehensive sensory evaluations are crucial to ensure commercial viability.

CRediT authorship contribution statement

Amir Kamkari: Writing – original draft, Visualization, Software, Methodology, Investigation, Formal analysis, Data curation. Saeed Dadashi: Writing – review & editing, Visualization, Validation, Supervision, Software, Resources, Project administration, Funding acquisition, Conceptualization. Maryam Khakbaz Heshmati: Writing – review & editing, Software, Resources, Methodology, Investigation. Jalal Dehghannya: Visualization, Resources. Yousef Ramezan: Writing – review & editing, Resources, Investigation, Funding acquisition, Conceptualization.

Funding declaration

No funding was received for conducting this study.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.