Development of a new, continuous, inline, Aqua-gas drying system and its application to drying perilla leaves while retaining the functional components and antioxidant activities

Abstract

We tested a continuous inline food-drying system that consists of a conveyor belt passing through three chambers containing Aqua-gas^® (AQG), superheated steam (SHS) in the near-anaerobic state, and cooling nitrogen or carbon dioxide gas. Common viable and coliform bacteria were sterilized by the AQG, SHS, AQG→SHS, and steaming treatments. The SHS, AQG→SHS, hot air (HA) drying, and steaming→HA drying treatments each produced dried perilla leaves with a final wet weight-based water content of <10 %. Most of the treatments preserved the α-linolenic acid content. The rosmarinic acid content was greatly increased by SHS and AQG→SHS treatment but greatly decreased by HA drying treatment. The AQG treatment produced significantly higher antioxidant activity, which was greatly increased by the SHS and AQG→SHS treatments due to the increase in the total soluble polyphenol content by the blanching effect, but the activity was greatly decreased by HA drying treatment. Radical-scavenging activity was well correlated with the total soluble polyphenol content. This prototype continuous inline AQG drying system both retains functional components and enables more efficient processing.

Introduction

The effects, types, and amounts of functional components in vegetables and other agricultural crops have been well investigated, and both consumer interest in safe domestic agricultural products and the importance of vegetables in health-promoting diets have been increasing. Many agricultural crops with high functionality are under-utilized and discarded, however. Egoma (Perilla frutescens (L.) Bitt. var. frutescens) is one such example. Its seeds are cultivated mainly for oil, while the leaves are rarely used (some are used in the production of kimchi [Korean pickle]). Egoma leaves contain a-linolenic acid, one of the n-3 essential fatty acids, and rosmarinic acid, a type of polyphenol. Alpha-linolenic acid has been reported to have an anti-obesity effect (Chicco et al., 2009), to improve insulin resistance (Chicco et al., 2009; Muramatsu et al., 2010), and to be partly converted in the body to eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) (Ezaki et al., 1999), both of which are reported to maintain/improve brain function (Okuyama, 1995). Rosmarinic acid has been reported to have high antioxidant (Meng et al., 2008) and anti-allergic effects (Takano et al., 2004).

In order to maintain the functionality of agricultural materials, establishing better preservation methods is necessary. For example, drying treatment is widely used to improve the stability of crops and food during storage. Hot air (HA) drying is commonly used to preserve tomatoes (Orikasa et al., 2006) and spinach (Hirota et al., 2000), as well as other crops. However, HA drying is not effective when the functional components of the crop or food are susceptible to oxidation (Luning et al., 1995). Freeze-drying (FD) methods are used to improve storage stability under low-oxygen and low-temperature conditions (Miyasaka, 1987; Yamaguchi et al., 2012); however, this method is time consuming and costly.

To address the shortcomings of the HA drying and FD methods, we have investigated the applicability of vacuum-microwave drying to egoma leaves, and we observed that this method produced dried materials that retained their functional components (Ogawa et al., 2016). Another method, superheated steam (SHS) drying, has been tested for various materials. In the SHS protocol, saturated steam is heated to >100 °C with an external heater under a low-oxygen state. SHS-dried potatoes retain their color and vitamin C content (Shibata and Mujumdar, 1994), and SHS-dried burdock, carrots, and lotus roots not only retain their color, but characteristics such as sweetness can also be enhanced (Miyatake, 2008). In contrast, Ichige et al. (2009) reported that, compared to HA drying, SHS drying reduced the carotene content of carrots (presumably due to the high treatment temperature), which indicates that the SHS method is difficult to control.

Our research group developed an improved variant of the SHS method, Aqua-gas^® (AQG), which is superheated steam containing micro-droplets of hot water. As a pretreatment technology, AQG can be performed before primary processing, such as during sterilization, without reducing the material yield and while maintaining qualities such as functionality (Sotome et al., 2005; Isobe, 2006). However, AQG has generally been considered to be less suitable as a drying medium, as it contains fine water droplets. Nevertheless, we investigated the use of the AQG method to dry egoma leaves. The results demonstrated that the method produces dry materials with more practical treatment conditions compared to the conditions necessary for SHS treatment (Chikashige et al., 2019). We also observed that the AQG method exerted a bactericidal effect and maintained the egoma leaves' functional components. The AQG method thus showed promise as an efficient approach that can complete the multiple steps that are generally required for drying leafy vegetables (blanching, drying, and sterilization) in a single step, based on the effect of superheated steam containing micro-droplets of hot water. Drying by a combined use of AQG and SHS is also expected to require less processing time in actual operations.

However, since this method was a combination of batch processes in our previous research, we attempted to achieve a more effective drying treatment in this study. We therefore developed a continuous inline AQG drying system that is capable of conveyor-transfer, with AQG and SHS performed in a nearly anaerobic state. The treatment effects of an inline SHS treatment device on various materials have been described previously (Ono et al., 2007; Abe and Miyashita, 2006), but the treatment effects of our continuous inline AQG drying system have not yet been clarified. The effectiveness of a drying system that uses a combination of AQG and SHS has also not yet been established. We therefore examined the applicability of this combined method to the treatment of perilla, a plant in the Lamiaceae family that contains the same components as egoma, using this prototype device.

Materials and Methods

Materials tested  We purchased perilla leaves that were available year-round from a retail shop in Hamada, Japan on February 28, 2022. The perilla plants were cultivated in Aichi Prefecture. After harvesting, the leaves might have been stored at <4 °C in a precooling chamber, but the precise conditions are unknown. The leaves were used after washing and dehydration.

A prototype continuous inline AQG drying system

The configuration of the prototype system  Inline AQG and SHS treatment devices from different manufacturers are already available,^{i, ii)} but a continuous inline drying system combining AQG and SHS has not yet been developed. We therefore developed a new continuous inline AQG drying system, the prototype of which is shown in Figures 1 and 2. Its basic structure uses a small SV Roaster^® (Nakanishi Manufacturing Co., Osaka, Japan). The first chamber provides the AQG treatment; it is mounted with four Aqua-gas generators (IR type Aqua-gas generator AQGE-150i, Taiyo Seisakusho Co., Hokkaido, Japan). Two of the Aqua-gas nozzles are positioned vertically at the entrance-side of the conveyor and at the exit-side under the conveyor, and the other two Aqua-gas nozzles are placed horizontally at the conveyor inside the front entrance. Deionized water is supplied to an Aqua-gas generator by an electromagnetic quantitative pump (EHN-C21VC1YN, Iwaki Co., Tokyo). AQG steam (a mixture of superheated steam and fine water droplets) at about 120 °C is generated by the Aqua-gas generator and is released from each nozzle at a rate of 72.0 g/min.

Fig. 1.

A continuous inline Aqua-gas drying system.

Fig. 2.

Equipment diagram including the internal structure and gas pipeline.

We first speculated that (a) the AQG steam that was reheated in the heating device could be sent back to the AQG chamber through the blower, and (b) the AQG environment could be controlled simultaneously by the subsequent increase in the chamber temperature. However, we observed that the AQG steam was changed into dry steam, and the chamber temperature was not increased by the reheating of AQG steam in the heating device. To address this problem, we made the AQG environment secure by sending boiler steam at 0.05 MPa to the heating device, and then heating the steam to 140 °C in the heating device and circulating it back into the chamber using a 40 Hz blower. The heating chamber has an internal volume of 0.12 m³.

The new system's second chamber is a small SV Roaster that provides the SHS treatment at 120 °C–320 °C, because the SHS environment is secured by sending boiler steam at 0.2 MPa to a heating device, and the steam that is thus reheated in the heating device at 5 kW is circulated into the chamber through the 40 Hz blower. The heating chamber has an internal volume of 0.12 m³. This prototype device has a 300-mm-long connecting structure between the first and second chambers that is heated by an incorporated heater at 1 kW to prevent condensation due to cooling.

The AQG and SHS treatments are conducted using separate conveyors for which specific times can be set. The two conveyors are connected with 5 mm of free space at the boundary. The conveyor speed in the AQG treatment chamber can be set at 20-300 sec, which is a short time that is capable of blanching and sterilization treatment. The conveyor speed in the SHS treatment chamber can be set at 2–30 min, providing a sufficiently long time for drying treatment. The AQG chamber has casters capable of carrying and operating independently of the SHS chamber. The cooling unit (a third chamber) has an internal volume of 0.0689 m³ and is capable of cooling materials by jetting nitrogen gas or carbon dioxide gas at 0.075 MPa to prevent oxidation due to exposure to the air when the processed materials are still hot.

Temperature measurement in the AQG treatment chamber  To determine whether the temperature in the AQG treatment chamber was sufficient, we placed type K thermocouples connected to a data logger (midi Data Logger GL820, Graphtec Co., Yokohama, Japan) at 17 sites at 71-cm intervals on the mesh tray on the conveyor and at one site at the steam blowout port of the blower to extend over a 75-cm length of the AQG treatment chamber, and we then conducted an AQG treatment trial run. The temperatures registered by the sensors were recorded by the data logger for 120 min, covering the starting operation and the AQG treatment. Based on the data obtained, we confirmed the temperature profile in the treatment chamber.

Temperature profile in the AQG treatment chamber  To evaluate the performance of this prototype, we first examined the temperature rise in the AQG treatment chamber. As shown in Figure 3, the results showed that almost all of the atmosphere temperature in the chamber exceeded 115 °C (within about 20 min after the power was turned on) by circulating the boiler steam that was superheated by the heating device and transferred into the chamber through the blower. The perilla leaves' moisture content data confirmed that there was no decrease after AQG treatment (as described below), indicating that the AQG atmosphere in the AQG treatment chamber was secure.

Fig. 3.

Temperature profile in the Aqua-gas treatment chamber.

Treatment procedures  The treatment conditions are summarized in Table 1. The treatment methods used were: (a) Aqua-gas (AQG) treatment; (b) superheated steam (SHS) drying; (c) AQG followed by SHS drying (AQG→SHS); (d) hot air (HA) drying for the experimental plots, and freeze drying (FD) for the control plot.

Table 1 Treatment methods and conditions tested.

No treatment (Freeze-dried [FD] product = control)

(a) Aqua-gas (AQG)-treated product   Test No. 1: AQG (120 °C, 72.0 g/min, 30 sec)

(b) Superheated steam (SHS)-dried product   Test No. 2: SHS drying (150 °C, 10 min)   Test No. 3: SHS drying (180 °C, 7 min)

(c) AQG treatment followed by SHS-dried product   Test No. 4: AQG (120 °C, 72.0 g/min, 30 sec) → SHS drying (150 °C, 10 min)   Test No. 5: AQG (120 °C, 72.0 g/min, 30 sec) → SHS drying (180 °C, 7 min)

(d) Hot air (HA)-dried product   Test No. 6: HA drying (55 °C, 5 h)   Test No. 7: Steaming (30 sec) treatment → FD   Test No. 8: Steaming (30 sec) treatment → HA drying (55 °C, 5 h)

72.0 g/min indicates the amount of micro-droplets of hot water.

In the AQG treatment, the gas-phase temperature of the heating medium was set to 120 °C; the total amount of superheated steam and fine water droplets supplied was 72.0 g/min, and the treatment time was 30 sec; we set these parameters based on those that were most stable at about 115 °C, as determined by Sotome et al. (2006). The AQG environment was secured by circulating boiler steam reheated by the heating device, which generated the superheated steam into the chamber through the blower described above. The SHS treatment chamber was run at 100 °C to prevent condensation in the chamber pass-through (Table 1, Test No. 1).

In the SHS drying treatment, we set the steam temperatures to 150 °C and 180 °C and the corresponding drying times to 10 and 7 min, respectively, as these conditions provided completely dried materials in preliminary experiments at the 130 °C–150 °C values used by Ono et al. (2006) and 120 °C–180 °C used by Abe and Miyashita (2006). The SHS environment was secured by circulating the boiler steam reheated in the heating device to produce superheated steam which was transferred to the chamber through the blower. In this experiment, the AQG treatment chamber pass-through was not performed (Table 1, Test Nos. 2 and 3).

For SHS drying after AQG treatment (AQG→SHS), the sample was first treated in the AQG heating chamber at 120 °C for 30 sec with a steam–water droplet flow rate of 72.0 g/min. The blanched sample was then quickly transferred to the SHS treatment chamber and dried at 150 °C and then subjected to treatment with superheated steam at 180 °C for 10 and 7 min, respectively (Table 1, Test Nos. 4 and 5).

In the HA drying treatment, a vacuum equilibrium-heating dryer (BCD-2000U; Yahiro Industry Co., Tokyo) was used. The drying temperature was set to 55 °C, and the drying duration was 5 h; we set these conditions based on the 40 °C–70 °C setting used by Orikasa et al. (2008), the 50 °C–75 °C setting used by Morifusa et al. (2012), and the 50 °C setting used by Hirota et al. (2000). Samples that were dried without blanching (Table 1, Test No. 6) and samples that were dried after steam-blanching treatment (Table 1, Test No. 8) were prepared. In this treatment, a reference sample subjected to FD following steam-blanching treatment was also prepared (Table 1, Test No. 7). In the steam-blanching treatment, saturated steam generated by a once-through boiler (SZ160; Miura Co., Ehime, Japan) was supplied to a steamer using a pressure-reducing valve, and the pressure of the saturated steam supplied from the boiler was adjusted to 98 kPa with a pressure-reducing valve; the temperature was set to about 100 °C (which is generally used for steam-blanching treatment). A 30-g sample was placed on a cloth, laid in a basket steamer, and processed in the steamer.

All experiments were conducted in triplicate, and the data were subjected to the following analyses.

Counting of viable and coliform bacteria  Viable and coliform bacteria were counted by standard methods. Briefly, 90 g of sterilized physiological saline was added to 10 g of a treated wet sample (FD, Test Nos. 1 and 7), and the mixture was stirred for 1 min using a Stomacher^® circulator to yield a sample stock solution (10¹ diluted solution). Next, 99 g of sterilized physiological saline was added to 1 g of a dried sample (Test Nos. 2–6, 8), and the mixture was subjected to the same treatment as that described above (10² diluted solution). For the measurement of the viable bacteria counts, 1 ml of each serial dilution of this sample stock solution was seeded on a plate, followed by the addition of 15 ml of sterilized standard agar medium. After incubation at 30 °C for 3 days, the numbers of colonies that had formed were counted.

To measure the coliform counts, twofold-concentrated brilliant green lactose bile (BGLB) medium was prepared, 10 ml of which was added to a test tube. A Durham tube was then inserted into the test tube, and the entire unit was sterilized. The same amount of sample stock solution was added. After incubation at 35 °C for 2 days, the sample was judged to be coliform-positive or -negative based on the presence or absence of gas formation in the Durham tube.

Measurement of moisture content  The moisture content of the treated samples was measured as follows. Approximately 2 g of a wet sample (FD, Test Nos. 1 and 7) or ∼0.5 g of a dried sample (Test Nos. 2–6, 8) was collected on a glass petri dish, and its precise weight was measured. After the sample was dried to a constant weight in an isothermal drier (DX 300; Yamato Scientific Co., Tokyo) at 105 °C, the weight of the dried sample was subtracted from that of the original sample, and then the difference was divided by the weight of the original sample to give the wet weight-based water content.

Next, wet samples were lyophilized, and all of the samples were finely pulverized with a Vibrating Sample Mill TI-100 (CMT Co., Greenville, SC, USA). All dried samples were dried at 105 °C as described above, and the wet weight-based water content of all of the dried samples was measured. The amount of dry matter in each dried sample was determined in this manner, and all of the following measurements were calculated per dry matter.

Measurement of the α-linolenic acid content  We quantified the α-linolenic acid in the samples as described by Hashimoto et al. (1999) and our report (Chikashige et al., 2019), with the following modifications.

Equipment: gas chromatography detector (Nexis GC-2030, Shimadzu, Kyoto, Japan); column: TR-FAME (0.25 mm × 60 m; Thermo Fisher Scientific, Waltham, MA); column temperature: held at 50 °C for 1 min, increased at 25 °C/min to 175 °C, at 2 °C/min to 230 °C, and then held at 230 °C for 2 min. The detector was a flame ionization detector (FID), the detector temperature was 280 °C, the temperature in the sample vaporization chamber was 250 °C, split (split ratio: 5.0), and the sample injection volume was 1.0

Measurement of the rosmarinic acid content  Rosmarinic acid in samples was quantified as described previously (Chikashige et al., 2019), with the following modifications.

Equipment: liquid chromatography pump (Nexera LC-20AD, Shimadzu); column: Shim-pack XR-ODS (4.6 × 100 mm) (Shimadzu); eluent: solution A: 20 % aqueous solution of acetonitrile; solution B: 80 % aqueous solution of acetonitrile (both containing 0.2 % formic acid). The solution was held at 100 % of A for 2 min, and then the eluent was changed from 100 % of A to 100 % of B for 10 min, held at 100 % of B for 3 min, and then held at 100 % of A for 3 min to complete the gradient elution. Column temperature: 40 °C; flow rate: 1 ml/min; detection: UV 280 nm.

Measurement of the total soluble polyphenol content  The total soluble polyphenol levels were measured by the Folin-Ciocalteu method (Singleton and Rossi, 1965). Ten milliliters of 70 % (v/v) ethanol were added to 0.2 g of a sample, and the mixture was stirred and extracted overnight. The supernatant of the extract was collected by centrifugation at 12 000 g, 4 °C for 10 min. To 80 µl of serial dilutions of the extract, we added 80 µl each of phenol reagent diluted five times with water and a 10 % (w/v) aqueous solution of sodium carbonate. The resultant mixtures were allowed to stand for 60 min, and the absorbance values at 650 nm were then determined with a microplate reader (Spark Control, Tecan Japan, Kanagawa, Japan). The measured values are expressed as gallic acid equivalents per 1 g of dried perilla leaf powders.

Evaluation of antioxidant activity  The 1,1-diphenyl-2-picrylhydrazyl (DPPH) radical scavenging activity was measured as follows. An 800 µM DPPH solution in ethanol was prepared as needed during the analysis. The activity was determined by adding 50 µl of 30 % ethanol, 50 µl of 200 mM MES buffer, and 50 µl of 800 µM DPPH solution to 50 ̅l of the perilla leaf extract used for the measurement of the total soluble polyphenol content. The resulting mixture was stirred for 20 min, and the absorbance at 520 nm was then measured with a microplate reader. The results are expressed as Trolox equivalents per 1 g of dried perilla leaf powder.

Statistical analyses  The component levels and antioxidant properties of the samples were determined, and statistical significance tests were performed using the mean values of duplicate measurements of triplicate experiments (n=3). Bactericidal effects were also determined in duplicate, and the data are expressed as mean values of viable cell counts. The coliform bacteria results were assessed in triplicate experiments (n=3). All of the statistical analyses were performed using the Tukey-Kramer method with R ver. 3.5.0 for Windows (R Core Team, R Foundation for Statistical Computing, Vienna, Austria).

Results and Discussion

Bactericidal effects of the different treatments on the perilla leaves  Table 2 presents the bactericidal effects observed in perilla leaves treated under the conditions listed in Table 1. A viable bacterial count of initial material (No treatment, FD, control) at the order of 10⁴ was observed, and the control was also positive for coliform bacteria. These microorganisms were not detected after treatment with AQG, SHS, AQG followed by SHS, or steaming. However, viable bacteria were detectable after the HA drying without steaming pretreatment, and viable bacteria were also detectable after the post-steaming-treatment HA drying due to the moderate-temperature conditions, allowing the growth of bacteria. These results were probably obtained because in the AQG treatments, the heat-transfer coefficients to the materials were extremely high at almost 120 °C, and in the SHS treatments, the temperature was extremely high at >120 °C.

Table 2 Comparison of viable microbial cell counts in perilla leaves after the various treatment methods.

Test No.	Bacteria numbers, cfu/g	Coliforms
FD (control)	5.2 × 104	+
1	<300	−
2	<300	−
3	<300	−
4	<300	−
5	<300	−
6	2.2 × 105	−
7	<300	−
8	7.5 × 102	−

The treating methods and conditions tested (Test No.) are indicated in Table 1. All data are means, or the result were confirmed (n = 3).

cfu: colony-forming units, +: positive, −: negative.

In the steam-blanching treatment, the heat-transfer coefficients of steam to materials were comparable to those in the AQG treatment, and the temperature was almost 100 °C. However, it is possible that viable bacteria proliferated from secondary contamination under drying in the post-steaming-treatment HA drying. The AQG, SHS, and steaming treatment data indicate that for effective microbial inactivation, treatment with a rapid temperature increase to ≥100 °C is necessary in wet-heat conditions.

Together these results demonstrated that treatment with AQG alone or AQG combined with subsequent SHS can effectively kill microorganisms on the surface of perilla leaves due to the rapid heating of the sample surface in the AQG treatment described above, and also due to the wet-heat conditions.

Moisture content following the different treatments of perilla leaves  The moisture content obtained with the different treatments is shown in Table 3. The moisture content was preserved during the treatment with AQG or steaming at levels comparable to those produced by No treatment (FD), due to the wet-heat conditions. This result also indicates that the AQG atmosphere was secure in the AQG treatment chamber, because the materials did not dry out under AQG treatment. These findings also confirmed that the SHS, AQG followed by SHS, HA drying, and steaming followed by HA drying treatments each produced dried perilla leaves with a final wet weight-based water content of <10 %, due to the dry-heat conditions.

Table 3 Comparison of moisture contents in perilla leaves after the various treatment methods.

Test No.	Wet weight basis, %
FD (control)	88.7
1	91.3
2	7.8
3	3.2
4	6.4
5	1.8
6	3.7
7	88.8
8	3.7

The treating methods and conditions tested (Test No.) are indicated in Table 1. All data are mean (n = 3).

Effects of the different treatments on the functional component levels in perilla leaves  Figures 4 and 5 illustrate the measured levels of α-linolenic and rosmarinic acid as the main functional components in perilla leaves subjected to different treatment methods including AQG. The SHS drying treatment of raw perilla leaves reduced the yield of α-linolenic acid to 92 % of that in the control sample (FD); this difference was significant, but not large. We observed that treated perilla leaves in which the original levels of α-linolenic acid were nearly completely retained could be prepared by AQG treatment, AQG treatment followed by SHS treatment (combined use), HA drying, steaming, and HA drying after steaming treatment. In light of these results, the effects of the treatment temperature and time on the residual amount of α-linolenic acid did not show a clear trend among all experimental groups, including AQG, SHS, and HA drying; we speculate that this is because (i) α-linolenic acid is relatively resistant to heat-treatment temperature and duration, and (ii) due to air oxidation.

Fig. 4.

Effects of treating methods on the a-linolenic acid contents in perilla leaves. The treating methods and conditions tested (Test No.) are indicated in Table 1. All data are mean ± SE (n=3). Different letters: p<0.05 differences.

Fig. 5.

Effects of treating methods on the rosmarinic acid contents in perilla leaves. The treating methods and conditions tested (Test No.) are indicated in Table 1. All data are mean ± SE (n=3). Different letters: p<0.05 differences.

In contrast, the rosmarinic acid content after AQG treatment was slightly decreased compared to that in the control (FD); the decreased content after AQG treatment was smaller than that after the steaming treatment and the HA drying post-steaming treatment. However, the content of rosmarinic acid after HA drying alone decreased levels by almost 10 % compared to the control. This was because, unlike in the other heat-treatment methods, oxidation progressed in the HA drying treatment due to the oxidase that was present. The rosmarinic acid content following the SHS treatment and that of the AQG treatment followed by SHS treatment were both significantly increased compared to the control value.

These results revealed that the AQG treatment is an effective method for preparing dried perilla leaves, as evidenced by the retention of high levels of a-linolenic and rosmarinic acids resulting from their reduced oxidation. In addition, drying with a combination of AQG and SHS treatments was more effective compared to AQG treatment alone, in terms of quality preservation. In addition, the combination of AQG+SHS is a more efficient approach as its use shortens the treatment time compared to AQG treatment alone by batch processing, and it is a simpler treatment process, thus improving continuous inline drying treatment.

Effects of different treatments on the antioxidant activity of perilla leaves  We next examined the effects of treating raw perilla leaves under the conditions listed in Table 1 on the total soluble polyphenol content and DPPH radical scavenging activity. The results are shown in Figure 6.

Fig. 6.

Effects of treating methods on antioxidant activities in perilla leaves. The treating methods and conditions tested (Test No.) are indicated in Table 1. All data are mean ± SE (n=3). Different letters: p<0.05 differences (polyphenol data).

When raw perilla leaves were directly subjected to the HA drying treatment, the residual amount of total soluble polyphenol decreased to a level that was 56 % of the control (lyophilized) product. In contrast, the amount of total soluble polyphenol increased significantly after treatment with AQG alone and in comparison to the steaming treatment. These results confirmed the effectiveness of the short-time AQG treatment, i.e., the higher heat transfer coefficient associated with this method. The total soluble polyphenol content also tended to be increased after drying with AQG followed by SHS, and it was significantly higher than that obtained by treatment with AQG alone. The content after the drying treatment with SHS alone was significantly increased.

We speculate that the reason why the total soluble polyphenol content was increased by this treatment, and possibly the rosmarinic acid content too (Fig. 5), is that the increase occurred because a blanching effect was obtained by a near-AQG state in the AQG treatment chamber, although the AQG treatment chamber was not operated under the SHS treatment conditions.

We conducted this study to determine whether we could obtain a sterilizing effect while retaining the functional components by using SHS treatment after AQG treatment; the results indicate that this is an efficient approach to achieve this objective. On the other hand, an increased polyphenol content was produced by the SHS-only treatment, for the reason described above. This may be inevitable because of the structure of this prototype device, but it is also an advantage in that the same effect produced by AQG treatment followed by SHS treatment can be obtained by SHS treatment alone.

The total soluble polyphenol content and DPPH radical scavenging activity of the treated perilla leaves correlated well (r=0.980), demonstrating that continuous inline AQG drying treatment is an effective means of preserving and increasing the original total soluble polyphenol content and DPPH radical scavenging activity of raw leaves. A likely cause of the increased total soluble polyphenol content and DPPH radical scavenging activity after blanching with AQG or SHS is that internal polyphenols and other antioxidants are released following disruption of the cell wall, as reported by Jimenez-Montreal et al. (2009) and Dewanto et al. (2002).

Conclusion

In a previous study we observed that it is possible to perform bactericidal and drying steps while retaining the functional components of foodstuffs by AQG treatment followed by SHS treatment, but that method employed a combination of batch processing. In the present study, in an attempt to obtain a more effective drying treatment for agricultural products, we tested a continuous inline AQG drying system that is capable of conveyor- transfer with AQG and SHS in a near-anaerobic state. The results suggest that drying leafy vegetables in particular with our continuous inline AQG drying prototype can not only retain functional components and obtain the sterilizing effect with AQG and SHS, but the system also provides a more efficient and simplified treatment process.

Acknowledgements  We thank Mr. N. Takiyama (Shimane Institute for Industrial Technology) for technical assistance with revisions of the manuscript.

Conflict of interest  There are no conflicts of interest to declare.

References

•  Abe  T., and  Miyashita,  K. (2006). Surface sterilization of dried fishery products in superheated steam and hot air. J. Jpn. Soc. Food Sci. Tecnol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 53, 373–379 (in Japanese).
•  Chicco,  A. G.,  D'Alessandro,  M. E.,  Hein,  G. J.,  Oliva,  M. E., and  Lombardo,  Y. B. (2009). Dietary chia seed (Salvia hispanica L.) rich in a-linolenic acid improves adiposity and normalises hypertriacylglycerolaemia and insulin resistance in dyslipaemic rats. Br. J. Nutr., 101, 41–50.
•  Chikashige,  K.,  Ogawa,  T.,  Nakajima,  M.,  Sotome,  I., and  Isobe,  S. (2019). Effects of Aqua-gas drying conditions on functional components and antioxidant activities in egoma (Perilla frutescens (L.) Bitt. var. frutescens) leaves. Food Sci. Technol. Res., 25, 39–18.
•  Dewanto,  V.,  Xianzhomg,  W.,  Kafui,  K. A., and H. I. (2002). Thermal processing enhances the nutritional value of tomatoes by increasing total antioxidant activity. J. Agric. Food Chem., 50, 3010–3014.
•  Ezaki,  O.,  Takahashi,  M.,  Shigematsu,  T.,  Shimamura,  K.,  Kimura,  J.,  Ezaki,  H., and  Gotoh,  T. (1999). Long-term effects of dietary alpha-linolenic acid from perilla oil on serum fatty acids composition and on the risk factors of coronary heart disease in Japanese elderly subjects. J. Nutr. Sci. Vitaminol., 45, 759–772.
•  Hashimoto,  M.,  Gamoh,  S.,  Tanabe,  Y.,  Hossain,  M. S.,  Masumaru,  S.,  Shinozuka,  K.,  Kworn,  T. M.,  Hata,  N., and  Misawa,  Y. (1999). The hypotensive effect of docosahexaenoic acid is associated with the enhanced release of ATP from the caudal artery of aged rats. J. Nutr., 129, 70–76.
•  Hirota,  T.,  Tahata,  K.,  Inoue,  Y., and  Nakagawa,  K. (2000). Conditions for making dry spinach by heated-air drying. Bull. Hyogo Pre. Agri. Inst. (agriculture), 48, 54–57 (in Japanese).
•  Ichige,  M.,  Yasuda,  S.,  Haba,  Y.,  Kano,  H., and  Kimura,  Y. (2009). Properties of the carrots dried by super-heated steam. Bulletin of Food Research Center in Aichi Industrial Technology Institute, 8, 98–100 (in Japanese).
•  Isobe,  S. (2006). Development of a new superheated steam system containing micro-water droplet and its application for food processing. J. Integr. Stud. Diet. Habits (Nihon Shokuseikatsu Gakkaishi), 17, 193–197 (in Japanese).
•  Jimenez-Montreal,  A. M.,  Garcia-Diz,  L.,  Martinez-Tome,  M.,  Mariscal,  M., and  Murcia,  M. A. (2009). Influence of cooking methods on antioxidant activity of vegetables. J. Food Sci., 74, H97–H103.
•  Luning,  P. A.,  Ebbenhorst-Seller,  T., and  De Rijk,  T. (1995). Effect of hot-air drying on flavour compounds of bell peppers (Capsicum annuum). J. Sci. Food Agric., 68, 355–365.
•  Meng,  L.,  Lozano,  Y. F.,  Gaydou,  E. M., and  Li,  B. (2008). Antioxidant activities of polyphenols extracted from Perilla frutescens varieties. Molecules, 14, 133–140.
•  Miyasaka,  M. (1987). Dried vegetables. Cookery Science, 20, 97–103 (in Japanese).
•  Miyatake,  K. (2008). Possible application of superheated steam characteristics for the food processing. Food Processing and Ingredients, 43, 4–7 (in Japanese).
•  Morifusa,  S.,  Orikasa,  T.,  Muramatsu,  Y., and  Tagawa,  A. (2012). Effect of solution spraying during hot air drying on surface hardening and browning of freshly cut apple pulp. J. Jpn. Soc. Food Sci. Tecnol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 59, 583–590 (in Japanese).
•  Muramatsu,  T.,  Yatsuya,  H.,  Toyoshima,  H.,  Sasaki,  S.,  Li,  Y.,  Otsuka,  R.,  Wada,  K.,  Hotta,  Y.,  Mitsuhashi,  H.,  Matsushita,  K.,  Murohara,  T., and  Tamakoshi,  K. (2010). Higher dietary intake of alpha-linolenic acid is associated with lower insulin resistance in middle-aged Japanese. Prev. Med., 50, 272–276.
•  Ogawa,  T.,  Chikashige,  K.,  Araki,  H.,  Kitagawa,  M.,  Katsube,  T.,  Ohta,  Y.,  Yamasaki,  Y.,  Hashimoto,  M., and  Azuma,  K. (2016). Effects of drying methods and pre-treatment conditions on the functional component contents and antioxidant activities in egoma (Perilla frutescens (L.) Bitt. var. frutescens) leaves. J. Jpn. Soc. Food Sci. Tecnol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 63, 217–224 (in Japanese).
•  Okuyama,  H. (1995). Nutritional regulation of brain functions-docosahexaenoic acid. Nippon Nogeikagaku Kaishi, 69, 583–585 (in Japanese).
•  Ono,  K.,  Endo,  H.,  Inatsu,  Y., and  Miyao,  S. (2006). Sterilizing effect of superheated steam on microbes in chinese cabbage. J. Jpn. Soc. Food Sci. Tecnol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 53, 172–178 (in Japanese).
•  Ono,  K.,  Endo,  H.,  Yamada,  J.,  Shoji,  I., and  Isobe,  S. (2007). Effect of superheated steam treatment on the preservation and sensory characteristics of buckwheat noodles. J. Jpn. Soc. Food Sci. Tecnol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 54, 320–325 (in Japanese).
•  Orikasa,  T.,  Tagawa,  A., and  Ogawa,  Y. (2006). Water absorption characteristics of dried tomato and surface softening of samples during soaking. J. Jpn. Soc. Food Sci. Tecnol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 53, 522–525 (in Japanese).
•  Orikasa,  T.,  Long  Wu,  Shiina,  T., and  Tagawa,  A. (2008). Drying characteristics of kiwifruit during hot air drying. J. Food Eng., 85, 303–308.
•  Shibata,  H., and  Mujumdar,  A. S. (1994). Steam drying technologies: Japanese R&D. Drying Technology, 12, 1485–1524.
•  Singleton,  V. L., and  Rossi,  J. A. (1965). Colorimetry of total phenolics with phosphomolibdic-phosphotungstic acid reagents. Am. J. Enol. Viticult., 16, 144–158.
•  Sotome,  I.,  Koseki,  S.,  Suzuki,  K.,  Isobe,  S.,  Yamanaka,  S.,  Ogasawara,  Y., and  Nadachi,  Y. (2005). Sterilization of vegetables without deterioration in their quality by a combined treatment with superheated steam and hot water spray. J Antibact. Antifung. Agents (Bokin Bobai), 33, 523–530 (in Japanese).
•  Sotome,  I.,  Suzuki,  K.,  Koseki,  S.,  Sakamoto,  K.,  Takenaka,  M.,  Ogasawara,  Y.,  Nadachi,  Y., and  Isobe,  S. (2006). Blanching of potato with superheated steam containing micro-droplets of hot water. J. Jpn. Soc. Food Sci. Tecnol. (Nippon Shokuhin Kagaku Kogaku Kaishi), 53, 451–458 (in Japanese).
•  Takano,  H.,  Osakabe,  N.,  Sanbongi,  C.,  Yanagisawa,  R.,  Inoue,  K.,  Yasuda,  A.,  Natsume,  M.,  Baba,  S.,  Ichiishi,  E., and  Yoshikawa,  T. (2004). Extract of Perilla frutescens enriched for rosmarinic acid, a polyphenolic phytochemical, inhibits seasonal allergic rhinoconjunctivitis in humans. Exp. Biol. Med., 229, 247–254.
•  Yamaguchi,  A.,  Nishi,  R.,  Hirose,  J.,  Urabe,  K., and  Nadamoto,  T. (2012). Changes in food items because of processing with different drying methods. Food Preservation Science (Nihon Shokuhin Hozo Kagaku Kaishi), 38, 169–176 (in Japanese).
• i)  https://www.nakanishi.co.jp/product/heating/sv-loaster.html (accessed Sept. 27, 2022) (in Japanese)
• ii)  https://www.taiyo-seisakusho.co.jp/products/aqua-cooker/ac_spec/ (accessed Sept. 27, 2022) (in Japanese)