Effect of freeze-thaw and PEF pretreatments on the kinetics and microstructure of convective and ultrasound-assisted drying of orange peel

Orange peel, the main by-product of the juice industry, contains valuable biocompounds (fibers, vitamins, polyphenols) but its high moisture content makes it prone to microbial and enzymatic deterioration. Hot-air drying is widely used to stabilize peels, extending shelf life and reducing handling needs, but prolonged exposure to heat entails high energy consumption and quality degradation. Pretreatments that modify tissue structure (e.g., osmotic, blanching, freeze-thaw, cold plasma, ultrasound in liquid media, electro-infrared, pulsed electric fields) can facilitate moisture transport during subsequent drying. Freeze-thaw can damage cell walls via ice crystals, with crystal size and distribution depending on freezing rate; PEF can induce electroporation, enhancing mass transfer depending on treatment intensity. Airborne ultrasound during drying can intensify moisture transport via mechanical stresses and interfacial micro-stirring, with effects dependent on product structure. A gap exists regarding how freeze-thaw pretreatment influences ultrasound-assisted convective drying of orange peel; only limited work has examined PEF combined with ultrasound for quality attributes. The study aimed to evaluate how freeze-thaw (fast vs slow) and PEF (low vs high intensity) pretreatments affect the kinetics of conventional and ultrasound-assisted drying of orange peel at 50 °C and to assess the resulting microstructural changes.

Prior studies report that airborne ultrasound can reduce drying time through mechanical effects and micro-stirring, with efficacy dependent on internal structure. Combinations of pretreatments with ultrasound-assisted drying have been explored: blanching/brining of apple, osmotic pretreatment of pineapple, and ethanol pretreatment of apple prior to US-assisted drying. Other combinations before conventional drying include freeze-thaw plus ultrasound in liquid prior to microwave drying and different freeze–ultrasound thawing schemes for vacuum freeze-drying of okra. However, no studies had assessed freeze-thaw effects on ultrasound-assisted convective drying. For PEF, previous research shows mixed outcomes dependent on product, temperature, and drying method; in some cases PEF accelerated convective or vacuum drying, in others it had negligible or adverse effects. The authors previously studied PEF+US effects on antioxidant properties of orange peel and kiwifruit. Literature also shows freezing pretreatments can either reduce or not affect convective drying times depending on product and microstructure, reflecting complex structure–transport relationships.

Raw material: Valencia Late oranges (Citrus sinensis) were purchased locally (Valencia, Spain), stored at 4 °C, washed, and hand-peeled. Slices of peel (5.0 × 2.5 × 0.3 cm; flavedo + albedo) were prepared. Initial moisture content was 2.9 ± 0.2 kg water/kg dry matter. Pretreatments: - Freeze–thaw (FT): Two freezing rates. • Fast freezing (FF): Samples wrapped in plastic film; blast chiller at −35.0 ± 0.3 °C for 2 h. • Slow freezing (SF): Samples wrapped in plastic film; domestic freezer at −18.0 ± 0.7 °C for 24 h; air turbulence minimized to delay freezing. Both FF and SF samples were thawed at 25 ± 1 °C to uniform internal temperature before drying. - Pulsed Electric Field (PEF): EPULSUS-PM1 system delivering high-intensity monopolar pulses. Conditions: 1.25 kV/cm, 10 Hz, 25 μs pulse width; samples in tap water (conductivity 1.04 ± 0.03 mS/cm) as conductive medium. Two intensities: Low (LiPEF) 8 pulses, 0.33 kJ/kg; High (HiPEF) 24 pulses, 0.98 kJ/kg. Unpretreated (UP) controls included. Drying experiments: Conducted in an ultrasound-assisted convective dryer. For each run, 18 slices were randomly placed to ensure homogeneous airflow. Conditions: 50 °C air temperature, 1 m/s air velocity; sample mass recorded every 10 min. Termination criterion: 60% loss of initial weight. Two drying modes were applied for each pretreatment: conventional hot air (AIR) and ultrasound-assisted convective drying (US). Airborne ultrasound was applied through the chamber’s aluminum vibrating cylinder excited by a piezoelectric transducer at 21.8 ± 0.4 kHz with constant electrical power of 50 W. All experiments performed at least in triplicate. Experiment codes: UP-AIR/US; FF-AIR/US; SF-AIR/US; LIPEF-AIR/US; HIPEF-AIR/US. Modeling: Drying kinetics were modeled mechanistically using a diffusion-based model for an infinite slab with unidirectional moisture flow: ∂W/∂t = D_eff ∂²W/∂x². Boundary condition accounted for internal diffusion and external convection: −D_eff ρ_s ∂W/∂x|_(x=L) = k (a_w(L,t) − a_air). L = 0.005 m (thickness); ρ_s from literature; equilibrium via sorption isotherms of orange materials. Finite differences numerical solution; kinetic parameters (D_eff, k) identified by minimizing squared deviations between experimental and calculated moisture content using the SIMPLEX algorithm in Matlab 2015B. Model performance evaluated by percentage of variance explained (%Var). Microstructure: Dried samples were cut (≈3 mm³), chemically fixed (25 g/L glutaraldehyde, 0.025 M phosphate buffer, pH 6.8, 4 °C, 24 h), post-fixed (20 g/L OsO4, 1.5 h), dehydrated through graded ethanol (300–1000 g/kg), contrasted in 40 g/L uranyl acetate in ethanol, embedded in epoxy resin (Durcupan). Ultramicrotomy produced 1.0 μm sections, stained with 2 g/L toluidine blue (implicit from coloration), and observed by light microscopy to assess parenchyma structure (flavedo, albedo).

- Baseline and ultrasound effect: UP-AIR required 300 ± 18 min to reach ≈0.55 ± 0.02 kg water/kg d.m.; UP-US required 218 ± 3 min, a 27.2% reduction. - Pretreatments alone (AIR): SF-AIR (297 ± 21 min) slightly faster than UP-AIR; FF-AIR significantly slower (363 ± 10 min). LiPEF-AIR (281 ± 12 min) and HiPEF-AIR (287 ± 6 min) were modestly faster than UP-AIR. - Pretreatments + US: Significant further acceleration versus UP-US. • SF-US: 195 ± 13 min (−35.0% vs UP-AIR). • LIPEF-US: 180 ± 8 min (shortest; −40.0% vs UP-AIR). • HIPEF-US: 193 ± 6 min (−35.6% vs UP-AIR). • FF-US: 245 ± 6 min; slower than UP-US. - Modeling fit: %Var ≥ 99.7% for all cases; excellent agreement between experimental and model moisture contents. - Kinetic parameters (representative values; means ± SD): • UP-AIR: D_eff = 6.6 (0.7) ×10⁻¹⁰ m²/s; k = 8.5 (0.5) ×10⁻⁴ kg/m²·s. • UP-US: D_eff = 8 (1) ×10⁻¹⁰ m²/s (+21% vs UP-AIR); k = 15 (3) ×10⁻⁴ (+76% vs UP-AIR). • FF-AIR: D_eff 4.1 (0.1) ×10⁻¹⁰ (significantly lower); k 8.9 (0.5) ×10⁻⁴. • SF-AIR: D_eff 5.4 (0.2) ×10⁻¹⁰ (lower); k 11 (2) ×10⁻⁴ (higher). • LIPEF-AIR: D_eff 5.8 (0.4) ×10⁻¹⁰ (lower); k 10.1 (0.8) ×10⁻⁴ (higher). • HIPEF-AIR: D_eff 5.3 (0.3) ×10⁻¹⁰ (lower); k 11.0 (0.1) ×10⁻⁴ (higher). • FF-US: D_eff 6.4 (0.3) ×10⁻¹⁰ (lower than UP-US); k 11.9 (0.5) ×10⁻⁴. • SF-US: D_eff 8.3 (0.8) ×10⁻¹⁰; k 17 (2) ×10⁻⁴ (highest k among US cases). • LIPEF-US: D_eff 11 (2) ×10⁻¹⁰ (highest D_eff overall); k 14.6 (0.6) ×10⁻⁴. • HIPEF-US: D_eff 8.0 (0.6) ×10⁻¹⁰; k 15.1 (0.8) ×10⁻⁴. - Microstructure: • Conventional drying caused marked structural degradation in UP-AIR; difficult to distinguish albedo/flavedo; large intercellular spaces. • Freeze–thaw AIR: degraded structure with large intercellular spaces but better-defined walls than UP-AIR; FF-AIR showed more defined but blurred walls; SF-AIR showed swollen walls and disrupted middle lamella. • PEF AIR: deformed or lysed cells, large pores; more blurred walls at higher intensity (HiPEF-AIR), indicating greater dispersion of wall components. • Ultrasound-assisted drying preserved tissue organization: flavedo and albedo remained distinguishable; UP-US cells nearly intact with dense, continuous walls. FF-US showed warped, thickened walls and lowest D_eff and k among US; SF-US exhibited some wall breaks facilitating external mass transfer (highest k). LIPEF-US showed better-defined, compact cells than HIPEF-US, aligning with highest D_eff. - Overall: Pretreatments alone tended to decrease internal moisture diffusivity (higher internal resistance) but could increase external mass transfer (higher k), while ultrasound reduced both internal and external resistances. The combination of slow freezing or low-intensity PEF with ultrasound produced the most pronounced process intensification.

Freezing rate critically shaped ice crystal size/distribution and thus structural outcomes: fast freezing generated many small intracellular crystals that swelled cell walls and dispersed wall components, hindering diffusion (FF-AIR slower; FF-US lower D_eff and k). Slow freezing produced fewer, larger extracellular crystals that partially broke cell walls and the middle lamella, easing diffusion and external transfer (slight acceleration in SF-AIR; higher k and faster drying in SF-US). PEF likely enhanced mass transfer via electroporation, but excessive intensity can soften and collapse the matrix, impeding diffusion; hence small benefits in AIR and a clear optimum when combined with ultrasound, where moderate electroporation (LiPEF) reduced acoustic impedance and improved ultrasound energy coupling. Airborne ultrasound enhanced drying by the sponge effect (cyclic compression/expansion), microcracking, and boundary-layer thinning, increasing both D_eff and k. The magnitude of ultrasound effects depended on pretreatment-induced structural changes, which modulated acoustic impedance and coupling. Consequently, optimized pretreatments (slow freezing; low-intensity PEF) synergized with ultrasound to maximally shorten drying while better preserving microstructure. The modeling confirmed these mechanisms: pretreated AIR samples showed reduced D_eff (increased internal resistance) and, except FF-AIR, increased k (reduced external resistance), whereas ultrasound increased both parameters, with the greatest gains for LiPEF-US and SF-US. Microstructural observations corroborated kinetic behavior: preserved organization under US, more severe damage at high PEF intensity, wall breaks aiding external transfer in SF-US, and wall warping hindering transport in FF-US.

Integrating airborne ultrasound with properly tuned pretreatments can significantly intensify orange peel convective drying while preserving tissue microstructure. Ultrasound alone reduced drying time by 27% versus conventional drying; combining ultrasound with slow freezing or low-intensity PEF further shortened drying (35–40% vs control), with LiPEF-US yielding the fastest kinetics and highest effective diffusivity among all conditions. Mechanistic modeling and microscopy jointly evidenced that pretreatments alone increase internal resistance (lower D_eff) but may reduce external resistance (higher k), whereas ultrasound decreases both; structural alterations that improve acoustic coupling (e.g., moderate electroporation, controlled wall disruption) maximize ultrasound benefits. Future work could optimize pretreatment intensity and ultrasound power/frequency across broader matrices and geometries, evaluate energy efficiency and scale-up, and quantify impacts on bioactive retention and downstream functionality of upcycled peel ingredients.