Effects of combined ultrasonic and microwave vacuum drying on drying characteristics and physicochemical properties of Tremella fuciformis.

This study analyzes the effects of ultrasonic waves on the drying kinetics of Tremella fuciformis during microwave vacuum drying. The physicochemical properties and structural characteristics of T. fuciformis polysaccharides (TFPs) were studied by drying tremella samples using hot air drying (HAD), microwave vacuum drying, ultrasonic pretreatments with microwave vacuum drying (US + MVD), and air-borne ultrasonic pretreatments combined with microwave vacuum drying (USMVD) under acoustic energy densities of 0.14, 0.28, and 0.42 W/mL. The results showed that USMVD and US + MVD accelerated the mass transfer process of T. fuciformis. Compared with HAD treatment, TFP samples obtained by USMVD and US + MVD had a reduced molecular weight to a certain extent, and they had stronger shear thinning ability. In addition, USMVD-TFPs at 0.42 W/mL retained higher total sugar, reducing sugar, and uronic acid, and the degree of reduction in the monosaccharide component content was small.

Introduction

Tremella (Tremella fuciformis Berk.; Tremellaceae), is a high quality medicinal and edible fungus [1]. Tremella fuciformis polysaccharides (TFPs) are the product of T. fuciformis spores by deep fermentation and have a wide range of pharmacological activities. In recent years, many reports have indicated that TFPs have a variety of biological activities, including antioxidant, anti-tumor, anti-aging, immune regulation, and hypoglycemic effects [2]. Therefore, TFPs have broad application prospects in the health food and medicine industries.

The high moisture content of fresh tremella accelerates the degradation of its quality, so processing is crucial for the consumption and use of tremella [3]. Drying is one of the traditional methods of processing tremella. Among the traditional drying methods, hot air, freeze, vacuum, and microwave drying have been used to produce dehydrated tremella. These drying methods often have problems such as requiring a long drying time, having high heat loss, being inefficient, resulting in a low quality product [4]. However, when microwaving is combined with vacuum drying technology this can reduce the local scorch phenomenon caused by uneven microwave drying of edible fungi. In addition, based on the characteristics of ultrasound, ultrasonic-assisted microwave vacuum drying technology can effectively strengthen and improve the drying process. Ultrasound has been gradually applied in food dehydration processes due to its significant mass transfer and enhancement effects. However, before ultrasound reaches the sample surface, the influence of different media on ultrasound-assisted drying is worth paying attention to. The formation of microchannels in ultrasonic pretreatment increases the effective diffusivity of water so that the total drying time is reduced [5]. In addition, the application of air-borne ultrasound can effectively reduce the resistance of tremella to external mass transfer, which is affected by drying conditions and the interface between the material and the gas phase [6]. Therefore, using ultrasonic pretreatment and air-borne ultrasound-assisted drying is the focus of this study.

The evaporation of water causes the main change in fungi during drying, and it also causes changes in their physical and chemical properties. Different drying methods have been reported to affect the physicochemical properties of polysaccharides, such as their chemical and monosaccharide composition along with their molecular weight, structure, and conformation. For example, freeze, hot air, and infrared drying can change the content of neutral sugars, uronic acid, and the protein of bitter gourd polysaccharides [7]. Similarly, Gan et al. [8] showed that the drying process may cause changes in monosaccharide conformation, resulting in changes in monosaccharide composition and the proportions of polysaccharides, but did not cause significant changes in the molecular weight and size of the main polysaccharides in longan. In addition, polysaccharides have a variety of functions due to their complex structure and certain molecular weights. Fu et al. [9] showed that the molecular weight and apparent viscosity of polysaccharides in loquat leaves changed during drying, which proved that the shear thinning behavior of polysaccharides may be related to the untangling of molecular chains. As far as we know, previous studies on the physicochemical properties and biological activities of T. fuciformis polysaccharides have mainly focused on the extraction and purification of polysaccharides. However, the effects of microwave vacuum drying and ultrasound combined with microwave vacuum drying on the properties of polysaccharides have not been studied.

Therefore, the purpose of this study was to study the effects of ultrasound on the drying kinetics of T. fuciformis during microwave vacuum drying, and to explore the effects of different drying methods on the structure and rheological properties of T. fuciformis polysaccharides. The chemical composition, polysaccharide structure, monosaccharide composition, molecular weight, and rheological properties of polysaccharides from T. fuciformis were evaluated by hot air drying (HAD), microwave vacuum drying, ultrasonic pretreatments (0.14, 0.28, and 0.42 W/mL) with microwave vacuum drying and air-borne ultrasound (0.14, 0.28, and 0.42 W/mL) combined with microwave vacuum drying. These results provide an important reference for the preparation and application of tremella polysaccharides.

Materials and methods

Raw materials

Tremella fuciformis was provided by the company (Jianhong Agricultural Development Co., Ltd., Fujian, China). The subbody is milky white, colloid translucent, and 10–12 cm in diameter. The tremella samples were stored in a refrigerator at 4 °C prior to the experiments. The auricular part was removed before the experiments. The initial moisture content of T. fuciformis was 84.05 ± 1.26% (on wet basis, w.b.).

Drying methods

Air-borne ultrasound combined with microwave vacuum drying (USMVD)

Air-borne ultrasound combined with microwave vacuum drying (USMVD) was carried out in an innovative combination drier (KL-2D-2ZG, Kailing Microwave Equipment Co., Ltd., Guangdong, China), which scheme is presented in Fig. 1. It allows drying with the application of microwave, vacuum and ultrasound technique separately or simultaneously. High-power ultrasound was generated by the air-borne ultrasonic generator. The combination drier was equipped with a vacuum pump (minimum absolute pressure 10 kPa), drying compartment, microwave generator (frequency 2,450 MHz, power capacity 2000 W), ultrasonic scraping unit (frequency 20 kHz, power capacity 500 W), and other system components. The control panel was adjusted to the relevant parameters (vacuum degree of − 80 kPa, microwave power of 1000 W, and a drying temperature of no>60 °C); then fresh tremella samples (100 g) were placed in the microwave chamber for drying. The energy input was controlled by setting the amplitude of the sonicator device. Effective ultrasonic power applied to the sample at a specific amplitude level was calculated by calorimetry [10]. The ultrasound power was then transformed to acoustic energy density, which is expressed as W/mL [11]. As shown in Table 1, the acoustic energy density involved in air-borne ultrasound combined with microwave vacuum drying (USMVD) were 0.14, 0.28, and 0.42 W/mL, respectively; the dried tremella samples were designated as US1MVD, US2MVD, and US3MVD, respectively. Considering the shrinking phenomenon, the height of sample holder was periodically adjusted to ensure the contact between tremella and an ultrasonic probe continued during the whole drying process.

Fig. 1

Schematic diagram and experimental setup for the air-borne ultrasound combined with microwave vacuum dryer: 1. vacuum pump, 2. vacuum gauge, 3. transducer, 4. drying compartment, 5. microwave generator, 6. magnetron, 7. ultrasonic transducer, 8. ultrasonic generator, 9. temperature and humidity transmitter, 10. water tank, 11. refrigeration, 12. drain valve, 13. sample holder, 14 . programmable logic controller, 15. computer.

Table 1

Drying schedule.

Abbreviation	Drying methods	Microwave power 1000 W, vacuum − 80 kPa	Calculated acoustic energy density /W·mL−1	Acoustic energy density /W·mL−1
MVD	Microwave vacuum drying	√	0	0
US1 + MVD	Ultrasonic pretreatment assisted by microwave vacuum drying	√	0.134	0.14
US2 + MVD		√	0.278	0.28
US3 + MVD		√	0.413	0.42
US1MVD	Air-borne ultrasound combined with microwave vacuum drying	√	0.142	0.14
US2MVD		√	0.288	0.28
US3MVD		√	0.422	0.42

Microwave vacuum drying (MVD)

Microwave vacuum drying (MVD) adopted dryer mentioned above (KL-2D-2ZG, Kailing Microwave Equipment Co., Ltd., Guangdong, China). The vacuum was set to − 80 kPa, the microwave power was 1000 W, and a drying temperature of no>60 °C. Then the ultrasonic control system was closed, and the samples (100 g) were dried in the microwave vacuum dryer.

Ultrasonic pretreatment assisted by microwave vacuum drying (US + MVD)

Samples containing fresh tremella (100 g) were vacuum-packed in polypropylene bags, and then the bags were placed in an ultrasonic water bath at frequency of 20 kHz (KQ-300VDE, Jiangsu Kunshan Shumei Ultrasonic Instrument Co., Ltd., China; internal dimensions: 300 × 240 × 150 mm). In Table 1, the acoustic energy density of 0.14, 0.28 and 0.42 W/mL (calculated through the calorimetric method) were used in this study and pretreatments were carried out at 40 °C for 20 min, then the above MVD was performed. The dried tremella samples were denoted as US1 + MVD, US2 + MVD, and US3 + MVD, respectively.

Hot air drying (HAD)

The samples were hot-air dried following the methods provided in Li et al. [3]. The fresh tremella was placed in in an electro-thermostatic blast oven (DHG-9053A, Jinghong Laboratory Equipment Co., Ltd., Shanghai, China) at 60 °C and dried at a constant temperature 15 h under the maximum air speed.

Measurement of related drying parameters

Measurement of moisture content

According to the Chinese national standard (GB5009.3–2016), the moisture content of the sample was measured at 105 °C in accordance with the oven drying method until each sample was dried to constant weight. According to the Chinese National Standard (GB7096-2014), all drying tests were carried out at least three times to ensure that the moisture content of each tremella material<15% (w.b.). The moisture content (MC) of the material in the drying process was calculated based on the method of Zhao et al.[12] and Eq. (1):

where m is the mass of material at time t (g); and m (g) is the constant mass after drying at 105 °C.

A drying curve of the change in the moisture ratio (MR) over time was drawn based on the detection of water loss in the drying process of samples and was calculated using Eq. (2):

where M (g/g, d.b.) is the moisture content at time t, M (g/g, d.b.) is the moisture content at equilibrium, and M0 (g/g, d.b.) is the initial moisture content. Because the equilibrium water content M of the material is very small, it can be ignored in the MR calculation [13].

Calculation of drying rate

The drying rate (DR) refers to the dry base moisture content reduced per unit time in the drying process, as shown in Eq. (3):

where Mt and Mt+dt are dry base moisture content (g/g, d.b.) of the material at time t and t + dt, respectively, and dt is the time interval (min).

Calculation of effective moisture diffusivity (D)

Although the shape of the fruiting structure of tremella do not exhibit a regular geometry, they were considered spherical. Fick's second diffusion equation was used to calculate effective moisture diffusivity (D) in the drying process. In this work, it is assumed that the D remains constant during the drying process, and the solids are symmetrical with respect to the unidirectional diffusion coordinates. The D was calculated according to Eq. (4):

where D is effective moisture diffusivity (m2/s), n is a positive integer, t is drying time (s), and r is the sphere radius (m) [14].

Calculation of the energy consumption and specific energy consumption

The energy consumption includes two parts, namely, energy consumption of ultrasonic treatment and of MVD. The energy consumption (EC) and specific energy consumption (SEC) was evaluated by Eqs. (5–9):

where Q is the energy consumption for heating water (kWh), C is the specific heat capacity of the water (4.187 kJ/kg·K), M is the mass of the water (kg), ΔT is the increase in the water temperature (K), P1 is the effective power output of ultrasonic power (kW), P2 is the effective power output of microwave waves (kW), t is the processing time (s), EC, E, and E (kWh) are the energy consumption as well as the energy consumption the microwave vacuum drier and the ultrasonic system, respectively, SEC (kWh/kg) is the energy required to evaporate 1 kg of water inside the tremella by drying, and m (kg) is the mass of water removed during drying process [5].

Mathematical modeling

The mathematical models used to assess the drying parameters for tremella are shown in Table 2. The coefficient of determination (R), chi-square test (χ), root-mean-square error (RMSE), sum of squares error (SSe), Akaike information criterion (AIC), and Bayesian information criterion (BIC) were used to evaluate the fitness of each model. Higher R values and lower RMSE and χ values indicated that the model provided a better fit for the data. These values were calculated using Eqs. (10–15) respectively:

Table 2

Mathematical models applied to the experimental drying curves.

Model	Mathematical equation	References
Page	MR = exp(−ktn)	[50]
Wang and Singh	MR = 1 + at + bt2	[51]
Two term	MR = a·exp(−kt)+(1–a)·exp(−kat)	[52]
Logarithmic	MR = a·exp(−kt) + b	[53]
Midilli	MR = a·exp(−ktn) + bt	[54]

Note: MR is the moisture ratio; t is the drying time; a, b, n, and k are model correlation coefficients.

where MR, MR, and MR are the ith experimentally observed, predicted, and average moisture content, respectively; meanwhile, N and n represent the number of observations and parameters, respectively.

Preparation of polysaccharides

Dried tremella were pulverized with a grinder (HC-350Y, Wuyi Haina Electric Appliance Co., Ltd., Zhejiang, China) and passed through an 80-mesh sieve for later use. Refer to the extraction method of Fu et al.[15]; some modifications were made. Next, 5 g tremella powder was added into 200 mL distilled water, and was stirred in constant temperature (85 °C) water bath for extraction (DF-101S, Lichen Bongxi Instrument Technology Co., Ltd., Shanghai, China) for 5 h, centrifuged at 4000 × g for 12 min (L550, Xiangyi Laboratory Instrument Development Co., Ltd., Hunan, China), and the supernatant was removed. The residue was repeatedly extracted twice, and the filtrate was concentrated. Then, 95% anhydrous ethanol was added to the supernatant for precipitation for 24 h, followed by centrifugation (4000 × g, 12 min) to collect alcohol-insoluble substances. The white filamentous substance was fully dissolved in water, and the excess organic solvent was removed by a rotary evaporator (W2-100S, Shensheng Technology Co., Ltd., Shanghai, China), and concentrated to an appropriate volume; next, the T. fuciformis polysaccharides (TFPs) were prepared by freeze drying. The cold trap and pressure of drying chamber were set at − 40℃ and 50 Pa, respectively. Crude polysaccharides were obtained from tremella by drying with HAD, MVD, US1MVD, US2MVD, US3MVD, US1 + MVD, US2 + MVD and US3 + MVD. Then, the corresponding codes were H-TFPs, M-TFPs, US1M-TFPs, US2M-TFPs, US3M-TFPs, US1 + M-TFPs, US2 + M-TFPs, US3 + M-TFPs, respectively, and the fresh tremella polysaccharides (Fre-TFPs) were compared.

Structural characterization of polysaccharide

Chemical components

Total sugar content was determined based on a total sugar content kit (Keming Biotechnology Co., Ltd., Suzhou, China). Reducing sugar content was determined based on a reducing sugar content kit (Keming Biotechnology Co., Ltd.). The determination of protein content was based on a kit using the bicinchoninic analysis method of protein content determination (Keming Biotechnology Co., Ltd.). The extraction and content determination of uronic acid were as follows: add 1 mL distilled water into 0.1 g tremella powder for mixed extraction, centrifuge the mixture at 10,000 × g for 10 min, remove precipitant, and the residue was repeatedly extracted twice. Use a galacturonic acid kit for determination (Keming Biotechnology Co., Ltd.).

Determination of molecular weight

The molecular weight of TFPs was determined by a gel permeation chromatography instrument (PL-GPC50, Varian, Inc., Shropsire, UK) based on the method of Wang et al.[16]. Chromatographic conditions were set as follows: a chromatographic column (7.5 × 300 mm, PLgelOlexis, Varian, Inc., Shropsire, UK), a protection column (7.5 × 50 mm, PLgelOlexis), a mobile phase of ultrapure water, a flow rate of 1.0 mL/min, and a column temperature of 40 °C. The weight average molecular weight (M), number average molecular weight (M), and polydispersion index (PDI = M/M) of each TPF sample were calculated by gel permeation chromatography software.

Determination of monosaccharide composition

Monosaccharides were precolumn derivatized by 1-phenyl-3-methyl-5-pyrazolone and separated and identified by reversed-phase high performance liquid chromatography (HPLC) using the method of Liu et al.[17]. The standard sugars were mannose (Man), galactose (Gal), glucose (Glc), arabinose (Ara), rhamnose (Rha), xylose (Xly), and fucose (Fuc). Trifluoroacetic acid (1 mL) was added to each 50 mg sample and hydrolyzed at 110 °C for 5–6 h. The supernatant was extracted after centrifugation. Next, 100 μL 1-phenyl-3-methyl-5-pyrazolone methanol solution (0.5 mol/L) was added, and the reaction was performed at 70 °C for 100 min under dark conditions. After cooling, 100 μL HCl solution (0.3 mol/L) and 400 μL water was added and monosaccharides were extracted with 700 μL chloroform three times. The upper solution was passed through a 0.45 μm needle filter, and then the derivatives were assayed using a Rigol L-3000 HPLC system (Puyuan Precision Electric Technology Co., Ltd., Suzhou, China) equipped with a C18 column (250 mm × 4.6 mm, 5 μm, Shimadzu, Japan). The isocratic elution procedure was 80% A (phosphoric solution) and 20% B (acetonitrile) with a flow rate of 1.0 mL/min and the detected wavelength was 250 nm.

Fourier transform infrared spectrometer analysis

Each 1.0 mg freeze-dried sample of TFPs was mixed and ground with 100 mg dried KBr powder. After pressing, a Fourier transform infrared spectrometer (VERTEX 70, Bruker Optics, Ettlingen, Germany) was used to scan the spectra in the wavelength range of 400–4000 cm−1. The spectra were processed using Origin 9.0 software.

Rheological analysis

Rheological analysis was conducted based on modifications of the method Wang et al.[16]. A mass concentration of 20 mg/mL TFP solution was prepared and measured by a rheometer (MCR301, Anton PaarTrading Co., GmbH, Graz, Austria) equipped with a 40 mm parallel plate. 1 The linear viscoelastic region was determined under the conditions of temperature 25 °C and frequency 1 Hz; the strain was scanned in the range of 0.01–100%, and the scanning strain is determined to be 1%.

Shear rheology measurement was done at 25 °C; the changes in apparent viscosity (η) and stress (τ) of samples at shear rates of 0.1–1000 s−1 were measured and fitted by a power-Law model as shown in Eq. (16):

where η is the apparent viscosity (Pa · s), γ is the shear rate (s−1), k is the consistency coefficient (Pa · s), and n is the flow behavior index (pseudoplastic index) [18].

Dynamic viscoelastic measurement: temperature 25 °C, frequency 1 Hz, scanning strain 1%; the change of the sample energy storage modulus (G') and loss modulus (G“) were measured at an angular frequency of 0.1–100 rad/s and fitted to the model. The equation is shown as Eqs. (17–18):

where G' and G'' represent energy storage modulus and loss modulus, respectively, n' and n'' represent correlation coefficients, ω is angular frequency (rad/s), k' and k'' represent the energy recovered and energy lost in each sinusoidal shear deformation cycle, respectively [18].

Statistical analysis

All data were expressed as means plus standard deviation, and each experiment was conducted three times simultaneously. Statistical analyses were performed by Origin 9.0 software (Origin Lab Corporation, Northampton, MA, USA). The data were analyzed using SPSS 18.0 software. Duncan's method was used to analyze the significance of differences among multiple groups of samples (p < 0.05).

Results and analysis

Drying kinetics

Figure 2 (a–d) illustrates the drying curve and drying rate curve of tremella under different acoustic energy densities of USMVD and US + MVD, respectively. During the drying process, with an increase of drying time, the water ratio of tremella showed an exponential downward trend; the final water content decreased to 0.027 ± 0.004 (g/g, d.b.) (Fig. 2a and 2c). Under the same acoustic energy density, the water content of USMVD tremella decreased the fastest, followed by US + MVD and MVD, indicating that two ultrasonic treatment had positive effects on mass transfer. This was in line with the study of Wang et al. [19], who suggested that this phenomenon may be related to the characteristics ultrasonic waves, which can change the diffusion boundary layer of the medium due to the internal effect of mass transfer and the change of pressure and oscillating viscosity. The application of ultrasonic pretreatment promotes mass transfer between the material and the liquid medium (such as distilled water), thus increasing the water loss. However, when compared with US + MVD, USMVD accelerates the heat transfer process more effectively under the condition of MVD, which may be related to the energy loss caused by the medium resistance [20]. Compared with the drying time of MVD alone, drying times of US1MVD, US2MVD, and US3MVD were reduced by 28.57%, 35.71%, and 42.86% with ultrasonic treatment, respectively. Similarly, ultrasonic pretreatment with a water bath can shorten the MVD time of tremella by 14.29% to 32.14%, indicating that this treatment can significantly improve the drying rate of tremella in the drying process. In addition, in the range of acoustic energy density selected in the experiment, with an increase in acoustic energy density, the drying time of tremella was shortened due to stronger cavitation and mechanical effects, indicating that the drying effect of tremella was improved with higher acoustic energy density. This result is consistent with a research report related to combined drying methods using ultrasound [21].

Fig. 2

Drying curves and drying rates of microwave vacuum drying (MVD), ultrasonic pretreatment assisted MVD (US + MVD) and air-borne ultrasound combined with MVD (USMVD) under different acoustic energy densities. Ultrasonic pretreatments with MVD were US1 + MVD, US2 + MVD, and US3 + MVD with MVD acoustic energy density of 0.14, 0.28, 0.42 W/mL, respectively; US1MVD, US2MVD, and US3MVD indicate air-borne ultrasound combined with MVD acoustic energy density of 0.14, 0.28, 0.42 W/mL.

Generally speaking, the drying process can be divided into three stages: periods of increasing, constant, and falling drying rates. According to the drying rate curve (Fig. 1b, d), MVD, US + MVD, and USMVD drying methods can be divided into increasing and decreasing rates, of which the decreasing rate is the main stage, which is similar to previous study on lotus seeds [12]. The enhancement effect of ultrasonic waves is closely related to water content. In the initial drying stage of US + MVD and USMVD, the inside of the material had a high free water content, with a rapid transmission of ultrasonic waves. At this time, the attenuation coefficient and internal transmission resistance of ultrasonic waves are very small, which is conducive to the penetration of ultrasonic waves through the material producing a strong strengthening effect [22]. When drying enters the final stage, the effect of ultrasonic waves on the drying rate becomes smaller, which may be caused by the sharp attenuation of ultrasonic energy in the later stage of drying, which leads to a decrease of the strengthening effect of ultrasonic waves on the drying process [23]. With an increase in acoustic energy density, the internal vibrations of drying materials become more intense, and the corresponding cavitation effect is strengthened. The kinetic energy caused by the cavitation effect is greater than the adsorption force of strongly attached water in the sample micro-channels, so that the mobility of adsorbed water molecules enhances and the drying rate increases significantly [24]. In addition, Jiang et al. [5] reported that the condensed substances generated by ultrasonic pretreatment were conducive to collecting water in organelles and cell walls under the condition of high microwave power (865 W), so that the drying process was mainly controlled by the evaporation of water.

In order to study the effect of ultrasonic waves on mass transfer, five models were used to fit the drying curve of tremella. The fitting results of five models all had the highest coefficient of determination R2 (0.9895–0.9997) as well as the lowest root-mean-square error RMSE (0.0059–0.0327), and chi square χ2 (4.29E − 05–1.20E − 03) (Table S1). In general, a model with more predictors provides more accurate predictions, but is also more likely to overfit the data [25]. Therefore, an additional criterion was introduced to evaluate the analytical models, i.e. the Akaike Information Criterion (AIC) or Bayesian Information Criterion (BIC), combined with model parameters, R2, χ2, RMSE value, and Midilli model can better characterize the moisture content of a fruiting body of tremella during drying. This is consistent with the findings of ultrasound-assisted vacuum drying techniques for garlic slices [26]. Meanwhile, we noticed that the drying constant k of this model increased with an increase in acoustic energy density.

Effective moisture diffusivity (D), energy consumption (EC) and specific energy consumption (SEC)

Effective moisture diffusivity (D) is an important parameter used to describe water transport during the falling rate period. Compared with MVD alone, when the acoustic energy density increased from 0.14 to 0.42 W/mL, the D in tremella dehydrated by US + MVD and USMVD increased by 13.95% to 36.63% and by 32.56% to 62.79%, respectively (Table 3). The results show that ultrasonic waves can achieve the purpose of increasing the drying rate of samples. Previously, some studies have reported similar results for samples of goji berry pre-treated by water bath US and apple treated by airborne US [27], [28]. The alternating compression and expansion caused by ultrasonic waves produce a “sponge effect”, which is conducive to the formation of microchannels in solid materials and promotes the migration of water inside the material to the surface, which was manifested in the increase in D [29]. Nadery Dehsheikh and Taghian Dinani [30] reported, when compared with drying using the power level 0 W, 500 W, and 1000 W ultrasonic pretreatment can improve the drying rate of convection dried banana chips; this was attributed to the effects of ultrasound and other mechanisms, such as a turbulent boundary layer effect that caused increased porosity in cells; in this case, the samples within the mass transfer rate and water to the surface of the diffusion rate increased. The energy consumption and specific energy consumption under USMVD, US + MVD, and MVD are presented in Table 3. The energy consumption of MVD was 0.24 kWh, and the specific energy consumption was 2.86 kWh/kg. The specific energy consumption for the US3MVD-treated tremella sample was the lowest (1.81 kWh/kg). Because ultrasonic treatment has a positive effect on water removal, the drying time is shortened and energy consumption is reduced [31]. With regard to the specific energy requirement, the values for some of the treated samples (US1 + MVD, US2 + MVD, and US3 + MVD) are higher than those for the untreated samples. This may occur because of the consumption of the energy spent on heating the water. Similarly, Ni et al. [28] used a water bath ultrasonic pretreatment combined with electrohydrodynamic drying to treat goji berry. Through specific energy consumption, they found that ultrasonic pretreatment at too high or too low temperature is not conducive to saving energy. Compared with MVD, the specific energy consumption of USMVD was reduced by 24.48–36.71%. This indicates that energy consumption tends to decrease with increasing acoustic energy density. Therefore, airborne ultrasonic assisted treatment based on MVD has the characteristics of having a short drying time and low energy consumption, which can reduce the cost of drying tremella and ensure the product quality.

Table 3

Drying time, effective moisture diffusivity (D), energy consumption, and specific energy consumption of the dried Tremella fuciformis.

Drying methods	Time (min)	Deff × 10−5 (m2/s)	Energy consumption (kWh)	Specific energy consumption (kWh/kg)
MVD	14	1.72 ± 0.06e	0.24 ± 0.01a	2.86 ± 0.07c
US1 + MVD	12	1.96 ± 0.05d	0.27 ± 0.02a	3.30 ± 0.08a
US2 + MVD	10.5	2.06 ± 0.06d	0.26 ± 0.01a	3.15 ± 0.08ab
US3 + MVD	9.5	2.35 ± 0.09c	0.25 ± 0.02a	3.05 ± 0.06b
US1MVD	10	2.28 ± 0.12c	0.17 ± 0.03b	2.16 ± 0.12d
US2MVD	9	2.58 ± 0.10b	0.16 ± 0.01bc	2.00 ± 0.12de
US3MVD	8	2.80 ± 0.11a	0.14 ± 0.02c	1.81 ± 0.11e

Data presented as mean values ± standard deviation.

Note: D, effective moisture diffusivity calculated using; MVD, microwave vacuum drying; US1 + MVD, US2 + MVD, and US3 + MVD, ultrasonic pretreatments with microwave vacuum drying acoustic energy densities of 0.14, 0.28, and 0.42 W/mL, respectively; US1MVD, US2MVD, and US3MVD, air-borne ultrasonic combined with microwave vacuum drying acoustic energy densities of 0.14, 0.28, and 0.42 W/mL, respectively.

Chemical component of TFPs

The chemical components of polysaccharides greatly affect their physicochemical and processing properties. Table 4 shows the chemical composition of polysaccharides in fresh and dried tremella samples. The total sugar content (13.25 g/100 g) and reducing sugar content (14.21 mg/g) in Fre-TFPs were higher than those in dried samples, indicating that drying treatment affected the total sugar and reducing sugar content in TFP samples. After drying, the retention of total sugars and reducing sugars retained by H-TFPs and M-TFPs were significantly lower than those retained by acoustic energy density drying at 0.28 W/mL and 0.42 W/mL (p < 0.05), which was due to the degradation of some polysaccharides into monosaccharides and oligosaccharides caused by the Maillard and caramelization reactions [32]. However, polysaccharides obtained by ultrasonic technology are not prone to bubbling, have a small contact area with air, and are not prone to browning, which reduces the degree of the Maillard reaction, so the total sugar loss of USMVD and US + MVD was minimal. Lei et al.[33] also drew a similar conclusion that the application of ultrasound could reduce the generation of the Maillard reaction, while increased acoustic energy density could prevent the sugar content in shiitake mushroom from decreasing more than that in the control group. In the dried samples, the protein content was controlled between 5.37% and 5.93%, indicating that MVD with or without ultrasonic treatment had no significant effect on the protein content (p > 0.05), but the protein content of US3M-TFPs was significantly higher than that of H-TFPs (p < 0.05). The results showed that high acoustic energy density USMVD could effectively prevent protein denaturation of tremella. The US3M-TFPs (6.75%) showed significantly higher uronic acid content than other dried samples (p < 0.05), which may occur because MVD with high acoustic energy density airborne ultrasound can reduce the drying time and the amount of oxygen present during the drying process, thus inhibiting the oxidation of uronic acid. Another possible explanation is that the uronic acids in polysaccharides were exposed after ultrasonic treatment [4], [34].

Table 4

Chemical components, molecular weights, polydispersion indices, and monosaccharide mole ratios of Tremella fuciformis polysaccharides (TFPs) under different drying methods.

Sample	Fre-TFPs	H-TFPs	M-TFPs	US1 + M-TFPs	US2 + M-TFPs	US3 + M-TFPs	US1M-TFPs	US2M-TFPs	US3M-TFPs
Chemical components
Total sugar (g/100 g)	13.25 ± 0.55a	7.78 ± 0.34f	8.26 ± 0.29ef	8.91 ± 0.35cde	9.09 ± 0.24 cd	9.53 ± 0.18bcd	8.93 ± 0.34cde	9.67 ± 0.20b	10.12 ± 0.38b
The reducing sugar (mg/g)	14.21 ± 0.65a	9.82 ± 0.26e	10.46 ± 0.35de	10.94 ± 0.29 cd	11.29 ± 0.25c	11.60 ± 0.23c	11.11 ± 0.20 cd	12.56 ± 0.27b	13.63 ± 0.34a
Protein content (%)	8.13 ± 0.47a	5.37 ± 0.23c	5.39 ± 0.17bc	5.47 ± 0.19bc	5.54 ± 0.21bc	5.57 ± 0.31bc	5.60 ± 0.38bc	5.91 ± 0.16bc	5.93 ± 0.11b
Uronic acid (%)	9.42 ± 0.31a	5.80 ± 0.10e	5.87 ± 0.12de	6.07 ± 0.14cde	6.15 ± 0.14cde	6.25 ± 0.26 cd	6.16 ± 0.22cde	6.35 ± 0.04c	6.75 ± 0.22b
Molecular weight and polydispersion index
Mn (kDa)	163.70 ± 3.55f	296.88 ± 2.85a	187.92 ± 2.93d	172.22 ± 2.87e	194.27 ± 3.16d	215.92 ± 3.27bc	177.68 ± 3.41e	210.32 ± 2.83c	220.89 ± 3.19b
Mw (kDa)	282.16 ± 2.93d	359.67 ± 3.70a	281.52 ± 3.96d	257.40 ± 3.39f	286.03 ± 2.53d	304.20 ± 2.17c	273.24 ± 5.36e	320.54 ± 5.48b	315.14 ± 4.23b
PDI	1.73	1.21	1.49	1.51	1.47	1.41	1.53	1.49	1.45
Monosaccharide composition and mole ratio
Fucose	13.69	5.11	6.08	5.48	6.42	7.22	6.81	6.92	8.22
Arabinose	1.29	0.39	0.59	0.47	0.51	0.58	0.61	0.56	0.71
Rhamnose	1	1	1	1	1	1	1	1	1
Glucose	0.11	0.07	0.08	0.07	0.08	0.11	0.07	0.09	0.09
Mannose	4	1.49	1.69	1.71	2.02	2.23	2	2.14	2.52
Galactose	N.D.	N.D.	N.D.	N.D.	N.D.	N.D.	N.D.	N.D.	N.D.
Xylose	3.25	1.2	1.34	1.49	1.75	1.84	1.69	1.75	2.14

Data presented as mean values ± standard deviation. Different letters represent significant difference between the values of different drying methods (p < 0.05).

Note: M, number average molecular weight; M, weight average molecular weight; PDI, polydispersity index; N.D., No content detected. The following preparations of Tremella fuciformis polysaccharides were used: Fre-TFPs, fresh polysaccharides; H-TFPs, polysaccharides under hot air drying; M-TFPs, polysaccharide under microwave vacuum drying; US1 + M-TFPs, US2 + M-TFPs, and US3 + M-TFPs, polysaccharides under ultrasonic pretreatments with microwave vacuum drying acoustic energy densities of 0.14, 0.28, 0.42 W/mL, respectively; US1M-TFPs, US2M-TFPs, and US3M-TFPs, polysaccharides under air-borne ultrasound combined with microwave vacuum drying acoustic energy densities of 0.14, 0.28, 0.42 W/mL, respectively.

Molecular weights of TFPs

Molecular weight and its distribution are important factors affecting the physical and chemical properties of polymers. Some studies have shown that the molecular weight of polysaccharides was significantly affected by the type of drying treatment used. Generally speaking, the higher the molecular weight of a polymer, the greater its tensile strength and elasticity [35]. The TFP elution spectrogram showed a primary peak with a single overlapping peak and similar peak times, indicating that all samples were homogeneous and their molecular weights were similar (Figure S1). Table 4 shows the molecular weight (M, M) and polydispersion index (PDI) of TFPs under different drying methods. The M of H-TFPs (359.67 kDa), US3 + M-TFPs (304.20 kDa), US3M-TFPs (315.14 kDa), and US2M-TFPs (320.54 kDa) were significantly higher than that of Fre-TFPs (282.16 kDa). The results showed that the polysaccharide chains tended to aggregate and the molecular weights increased with the rapid removal of bound water, especially under hot air conditions. Similar results have been reported by Gan et al. [8]. The change in the molecular weight of polysaccharides has been attributed to a three-step mechanism of polysaccharide aggregation, β-elimination depolymerization, and reaggregation [36]. Due to the presence of proteins in TPFs, TFPs gradually form aggregates through similar mechanisms [37]. In contrast, ultrasonic assisted MVD can reduce intermolecular aggregation to a certain extent. The M of US1 + M-TFPs and US1M-TFPs was lower when compared with that of Fre-TFPs. This may be related to the degradation and aggregation ability of TFPs under 0.14 W/mL ultrasonic drying treatment. So far, various reports have been published on the effects of drying on the distribution of molecular weight in polysaccharides from different raw materials. Yuan et al.[35] found that the molecular weights of okra polysaccharides obtained by HAD and vacuum drying was higher than those of microwave drying and freeze drying. Studies on Angelica sinensis polysaccharides have reached similar conclusions. However, Yan et al.[7] showed that freeze drying pretreatment had a slight effect on the molecular chains of bitter gourd polysaccharides, while high temperature and thermal degradation during HAD and infrared drying promoted a significant reduction in the molecular weight of the polysaccharides. A polydispersion index (PDI) was used to clarify the width of the molecular weight distribution of the polymers. The larger the PDI is, the wider the molecular weight distribution is. When the chain length of the monodisperse polymer is the same, PDI is equal to 1 [38]. The PDI of TFPs obtained by all drying treatments was close to 1 in the range of 1.21–1.73, indicating that the molecular weight distribution of TFPs was narrow.

Constituent monosaccharides of TFPs

The monosaccharide composition of T. fuciformis polysaccharides is one of the main indices used to determine the primary structure of T. fuciformis polysaccharides. The HPLC of standard monosaccharides included fucose (RT = 7.91 min), arabinose (RT = 10.55 min), rhamnose (RT = 14.40 min), glucose (RT = 15.89 min), mannose (RT = 16.73 min), galactose (RT = 17.44 min), xylose (RT = 19.75 min). Figure S2 shows the monosaccharide chromatograms of Fre-TFPs, H-TFPs, M-TFPs, USM-TFPs (0.14, 0.28, 0.42 W/mL) and US + M-TFPs (0.14, 0.28, 0.42 W/mL) samples. Tremella fuciformis polysaccharides are mainly composed of fucose (Fuc), arabinose (Ara), rhamnose (Rha), glucose (Glc), mannose (Man), and xylose (Xly). Table 4 summarizes the molar ratios of the monosaccharide components of Tremella fuciformis. Combined with Figure S2, we found that different drying methods did not affect the monosaccharide composition of TFPs, and ultrasound generally did not change the primary structure of polysaccharides; nevertheless, the molar ratios were different, which may be caused by the conformational changes of monosaccharides caused by different drying methods [39]. Fuc, Man, and Xly were the main monosaccharides of TFPs, which was consistent with the results of other researchers. The monosaccharide group of TFPs extracted from hot water was divided into 1.59:1:0.85 (Man: Xyl: Fuc) [1]. However, the main chain of TFPs was mainly composed of mannose residues, while the composition of the secondary chain was different. Ma et al.[40] showed that the monosaccharide composition of TFPs varied with different strains, and was not limited to the structure of the main chain of mannose. Compared with H-TFPs, in this case the MVD, USMVD, and US + MVD treatments reduced monosaccharide content to a small degree, which may be due to the fact that when polysaccharides were dried in hot air or oxygen-rich environments, hydroxyl oxidation and intermolecular hydrogen bond breakage easily occurred, thus affecting the resulting monosaccharide content [41].

FT-IR spectra of TFPs

The differences of functional groups, chemical composition, and molecular conformation of TFPs on the surface were investigated by infrared spectroscopy. The infrared spectra of TFPs under different drying methods and conditions are shown in Fig. 3. Fig. 3 shows that the infrared spectra of different TFPs have a good overlapping effect and no obvious differences were observed, but some peak shape differences may occur in the measurement process. These results indicated that they had similar structural characteristics, which further indicated that different drying methods and conditions had no significant effect on the structure of TFPs. Stretching vibration peaks of O − H and C − H were observed at 3408.6 cm−1 and 2927.1 cm−1 [38]. In addition, a group of peak values at 1250.5 cm−1 were C − H variable angle vibration, indicating that TFPs have three characteristic peaks of carbohydrates. The characteristic peak at 1727.1 cm−1 was the C = O tensile vibration in − COOH, the absorption peak at 1414.3 cm−1 was the C = O symmetric tensile vibration in − COOH, and the characteristic peak of carboxyl group indicated the presence of uronic acid. The absorption peak at 1614.7 cm−1 was a “typical protein band” (1651–1555 cm−1), indicating that TFPs contained a small amount of protein [9]. The absorption peak at 1067.2 cm−1 was caused by the C − O − C tensile vibration and C − O − H angular vibration, indicating that TFPs have an acidic heteropolysaccharide containing uronic acid [16].

Fig. 3

Fourier transform infrared spectra of Tremella fuciformis polysaccharide (TFPs) with different drying methods. Fre-TFPs, H-TFPs, and M-TFPs indicate preparation of fresh TFPs, TFPs under hot air drying, and TFPs under microwave vacuum drying (MVD), respectively; US1 + M-TFPs, US2 + M-TFPs and US3 + M-TFPs indicate preparation of TFPs under ultrasonic pretreatments with MVD acoustic energy densities of 0.14, 0.28, 0.42 W/mL, respectively; US1M-TFPs, US2M-TFPs, and US3M-TFPs indicate preparation of TFPs under air-borne ultrasound combined with MVD acoustic energy densities of 0.14, 0.28, 0.42 W/mL, respectively.

Rheological properties of TFPs

Shear rheology of TFPs

At 25 °C, the curves of apparent viscosity change and stress change between shear rates (1000–0.1 s−1) of TFP solutions obtained by different drying methods are shown in Fig. 4(a–d). Due to the presence of polymer entanglement and hydrogen bonding, the solution of tremella polysaccharide presents a partially associated conformation and thus exhibits high viscosity at a low shear rate [35]. With an increase in the shear rate, the apparent viscosity of all samples decreased. The internal reason may be that the external shear force destroyed the intermolecular forces of polysaccharides, resulting in the reduction of the number of chain entanglements at a high shear rate and the orientation of polymer chains along the direction of flow [42]. The shear stress of TFP solutions increased with an increase in the shear rate, and the flow characteristic is that TFP solutions flow under the action of an external force. Both show the non-Newtonian fluid characteristic of thinning by shear. In Fig. 4(a, c), compared with Fre-TFPs, all dried samples showed lower apparent viscosity and shear stress, while M-TFPs showed the lowest apparent viscosity and shear stress, which may be related to their molecular weight and polydispersity. The smaller the molecular weight of polysaccharides, the less the chain flexibility and pseudoplasticity. According to Ma et al. [43], linear molecules require more rotational space than highly branched molecules, and solutions with highly branched structures are usually less viscous than linear molecules with the same molecular weight. Therefore, H-TFPs that showed lower apparent viscosity at a highest molecular weight may contain more branching polysaccharide chains. The apparent viscosity of US3M-TFPs and US3 + M-TFPs was higher than that of H-TFPs; then, under the same acoustic energy density, the apparent viscosity of the polysaccharide solution obtained by the air-borne ultrasound treatment was better than that obtained by water bath ultrasound pretreatment. The viscosity of the polysaccharide dispersion system changed because of ultrasound. This may be related to the degree to which polysaccharides are heated. Xiao et al. [44] investigated the effect of three drying methods on the rheological properties of exopolysaccharides. The results showed that the chain expansion of polysaccharides increased the volume of molecules when polysaccharides were heated, which strengthened the thermal movement between molecular chains, weakened the interaction between molecules, and led to a decrease in viscosity. In Fig. 4(b, d), differences in apparent viscosity of samples assisted by water bath ultrasound pretreatment and air-borne ultrasound treatment (0.14, 0.28, and 0.42 W/mL) were observed with MVD. With an increase in acoustic energy density, the apparent viscosity also increased, which is consistent with the trend of their molecular weight. During the flow process, the polysaccharide molecular chains intertwine with each other to form a network structure. With the increase in the molecular weight of the polysaccharides, the mutual intertwining effect becomes stronger, resulting in higher flow resistance and increased viscosity of the system. The shear stress of TFP solutions also showed the same characteristics.

Fig. 4

Apparent viscosity versus shear rate plot (a–b) and shear stress-shear rate curves(c–d) of Tremella fuciformis polysaccharide (TFP) solutions. Fre-TFPs, H-TFPs, and M-TFPs indicate preparation of fresh TFPs, TFPs under hot air drying, and TFPs under microwave vacuum drying, respectively; US1 + M-TFPs, US2 + M-TFPs, and US3 + M-TFPs indicate preparation of TFPs under ultrasonic pretreatments with MVD acoustic energy densities of 0.14, 0.28, 0.42 W/mL, respectively; US1M-TFPs, US2M-TFPs, and US3M-TFPs indicate preparation of TFPs under air-borne ultrasound combined with MVD acoustic energy densities of 0.14, 0.28, 0.42 W/mL.

The power-law model performed well in describing the rheological behavior of polysaccharide solutions produced by different drying methods and conditions; this equation can explain the shear thinning behavior of TFP solutions. The fitting results are shown in Table S2. The fitting degree R2 of the flow curve fitting equation of TFP solutions obtained by different drying methods was above 90%; that is, the obtained model can reflect the rheological properties of the actual polysaccharide solutions within a 90% confidence interval. Compared with the dried samples, the k value of Fre-TFP solutions was the highest (48.81). Among the dried samples, the k value of M-TFP solutions was the lowest (8.79), followed by H-TFPs (11.05) and US3M-TFP solutions which was the highest (19.63); this was consistent with the change trend of apparent viscosity. In addition, the n values of all samples were less than 1, indicating that TFP solutions obtained by various drying methods are pseudoplastic fluids. The smaller the n value is, the closer the sample solution is to non-Newtonian fluid, indicating that the apparent viscosity decreases with the increase of shear rate, which is a typical shear thinning phenomenon. Among them, the TFP solutions obtained by airborne ultrasonic treatment at 0.42 W/mL had a small n value, indicating that US3M-TFP solutions have higher pseudoplasticity and show stronger shear thinning ability. The phenomenon of shear thinning has an important influence on food processing. The polysaccharides obtained by ultrasonic assisted MVD have better rheological properties, which is suitable for processing at a high shear rate, improving material transportation and infusion technology, and making it easier to process products [45].

Dynamic viscoelastic analysis

The linear viscoelastic region of TFP solutions was initially determined in order to study the dynamic viscoelastic properties of TFP solutions obtained by different drying methods; frequency scanning was carried out in the linear viscoelastic region. As shown in Fig. 5 (a–c), the storage modulus (G′) and loss modulus (G″) of TFP solutions increased with the increase of angular frequency within the test frequency range; the TFP samples underwent a solution-gel transition. At a low angular frequency where G′ < G″, the system is mainly viscous, indicating that the solution cannot form a gel at a low angular frequency. With the increase of angular frequency, G′ > G″ after the curve intersects in the high frequency region, the system is mainly elastic and presents a certain weak gel structure. This is consistent with previous research [46]. The transition from a low frequency liquid structure to a high frequency weak gel structure is related to the change of polysaccharide properties; the structure tends to the gel induction point due to frequency-related chain interactions [47]. In addition, compared with the dried sample, the G′ in the whole region of Fre-TFP solutions was higher than G″, and there was no crossover point, indicating that the system was mainly elastic, which may be due to the weak gel structure formed by interchain coupling between tremella polysaccharide molecules [48]. In dried samples, the intersection of G′ and G″ curves was in the range 1.11 to 6.95, suggesting a drying dependence of the crossing frequency [41]. The cross values of US3M-TFPs (1.43) were lower than those of other TFP solutions, indicating that ultrasonic treated TFP solutions showed excellent gel properties and system stability.

Fig. 5

Dynamic rheological curves of Tremella fuciformis polysaccharide (TFP) solutions prepared by the eight drying methods. Fre-TFPs, H-TFPs, and M-TFPs indicate preparation of fresh TFPs, TFPs under hot air drying, and TFPs under microwave vacuum drying (MVD); US1 + M-TFPs, US2 + M-TFPs, and US3 + M-TFPs indicate preparation of TFPs under ultrasonic pretreatments with MVD acoustic energy densities of 0.14, 0.28, 0.42 W/mL, respectively; US1M-TFPs, US2M-TFPs, and US3M-TFPs indicate preparation of TFPs under air-borne ultrasound combined with MVD acoustic energy densities of 0.14, 0.28, 0.42 W/mL, respectively.

The power law parameters of G′ and G″ for different TFP solutions are shown in Table S2. As it turns out, the k′ and k″ values of TFP solutions obtained by drying treatment were within the range from 5.55 to 14.98 and from 7.96 to 15.82, respectively. The n′ and n″ values were within the range from 0.40 to 0.48 and from 0.24 to 0.31, respectively. Among them, G′= k′·ω (n′ > 0) and G″= k″·ω (n″ > 0) for entangled weak gels with n′ > n″ [49]. The k′ values of US2 + M-TFPs and US3 + M-TFPs were higher than the k″ values, indicating that the solution was mainly elastic rather than viscous, while the results of other dried samples were opposite. For n′ and n″, except US1 + M-TFPs, the n′ and n″ values of ultrasonic treatment were all lower than H-TFPs and M-TFPs, indicating that ultrasonic assisted drying may play a role in reducing the frequency sensitivity of polysaccharide solutions [18].

Conclusions

The results of this study show that the application of air-borne ultrasound in the MVD process is beneficial to accelerating the drying process, and has a higher D and the lowest energy consumption of the methods tested here. With an increase of acoustic energy density, the drying time of tremella was shortened due to stronger cavitation and mechanical effects. The strengthening effect of a water bath ultrasound pretreatment was lower than that of air-borne ultrasound treatment, but the drying time was shorter than that of MVD alone. Five models were used to fit the drying curve of tremella; the Midilli model had the best fitting result. The structure and monosaccharide composition of polysaccharides were not affected by different drying methods, but some differences were observed in total sugar content, reducing sugar content, uronic acid content, molecular weight, molar ratio of monosaccharides, and rheological properties. In the dried samples, US3M-TFPs retained higher total sugar, reducing sugar, and uronic acid, and the degree of reduction in the monosaccharide component content was small. Compared with HAD treatment, TFP samples obtained by USMVD and US + MVD had a reduced molecular weight to a certain extent. In addition, US3M-TFPs samples had the strongest shear thinning ability, which was suitable for the transportation and inculcation process of food materials. The purpose of this study is to provide a high efficiency and high-quality drying method for the edible fungi drying industry in the future.

CRediT authorship contribution statement

Jingxin Xu: Conceptualization, Investigation, Methodology, Data curation, Software, Writing – original draft, Writing – review & editing. Danni Wang: Writing – review & editing. Yanping Lei: Writing – review & editing. Lujie Cheng: Writing – review & editing. Weijing Zhuang: Writing – review & editing. Yuting Tian: Conceptualization, Funding acquisition, Supervision, Writing – review & editing.

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.