Effect of infrared drying on chemical and microbial properties of Cold‐Hardy grape pomace (Edelweiss and Marquette)

Abstract This study aims to add value to a common wine industry waste by preserving bioactive compounds in cold‐hardy grape pomace (GP) and preventing any microbial growth under the proper drying conditions. Effects of infrared (IR) and hot air (HA) drying on the microbial and physicochemical properties such as color, phenolic compounds, and antioxidant activity of white (Edelweiss) and red (Marquette) GP were studied. The IR heating rapidly reduced the moisture content of GP from 55% wet basis (w.b.) to less than 10%, which resulted in a drying time reduction of 71.9% to 80.2% compared to HA drying. There were no significant differences in color parameters among the IR‐ and HA‐dried pomaces (p > .05). The phenolic content of ‘Edelweiss’ pomace was not significantly affected by both IR and HA drying, whereas the phenolic content of ‘Marquette’ pomace was substantially reduced from 274 mg/g dry extract in raw pomace to 127 mg/g dry extract after HA drying and to 141.9 mg/g dry extract after IR drying. Overall, the microbial load on the fresh pomace samples was dramatically reduced by IR heating, with a reduction of more than 99.9% when the pomaces were dried using IR at a temperature higher than 130°C. However, this high temperature of IR led to a significant reduction of DPPH antiradical scavenge activity for ‘Edelweiss’ pomace (p < .05). This study shows that using the IR approach, cold‐hardy ‘Edelweiss’ and ‘Marquette’ grape pomaces can be efficiently dried with the antioxidant activity maintained, which could be used in a variety of food products as a functional ingredient.

Phenolic compounds, such as tannins and anthocyanins which are antioxidant and anti-inflammatory, reduce the risks of cardiovascular diseases and cancers (Unusan, 2020). During the red winemaking process after pressing, about 50% of tannins and between 15 and 60% of anthocyanins remained in the Vitis vinifera grape seeds and skins, that is, grape pomace (Ky & Teissedre, 2015). In the U.S. Midwest, cold-hardy interspecific hybrid grape cultivars, resistant to harsh cold winter, are used to produce wines. There are differences in grape chemistry between the commonly grown Vitis vinifera grape varieties and cold-hardy interspecific hybrids. It has been observed that in interspecific hybrid wines, fewer phenolic compounds are released in wines than in Vitis vinifera wines (Rice et al., 2017). Therefore, it has been suggested that those compounds are retained in the cold-hardy grape pomace (Rice et al., 2017;Watrelot & Norton, 2020). Thus, there is a tremendous economic opportunity to valorize cold-hardy GP and its compounds in the fields of health and nutraceuticals, as the consumers' demand for bio-sourced products is increasing continuously (Fontana et al., 2013).
Grape pomace contains a significant amount of moisture that can negatively impact its quality during storage if not appropriately dried (Majerska et al., 2019). Conventional drying, such as hot air (HA) drying, is the most common method to reduce the moisture content (MC) below 14% to prevent microbial growth (El-Mesery & Mwithiga, 2014). However, HA drying takes a long time (hours) to reduce the moisture content in the pomace, which promotes microbial development and results in nutritional and bioactive compound breakdown (Martynenko & Kudra, 2016). Establishing a rapid and effective drying approach is essential because of the lengthier drying times and losses in phenolic content associated with traditional drying processes. Infrared (IR) drying is an efficient heating technology that proposes several processing advantages, such as rapid drying and microbial inactivation, while improving the safety and quality of food products .
Hence, in the current study, the development of IR heating takes a further step as an efficient technology to dry cold-hardy GP to a stable product that can prevent microbial growth. It is hypothesized that drying GP using IR heating will provide a sustainable green way to process postwinemaking waste, creating salubrious economic and environmental benefits.
Also, no research has been carried out yet on the impact and optimization of IR heating on cold-hardy interspecific hybrid GP. The objectives of this study were (1) to evaluate the effectiveness of the IR heating in cold-hardy GP valorization and optimize the IR drying parameters, (2) to determine the effects of the IR drying process on the phenolic content and antioxidant properties of cold-hardy GP, and (3) on the microbial properties of the cold-hardy GP.

| Grape pomace samples and preparation
In this study, two types of cold-hardy GP were used: (1) pomace of white grape, cultivar Edelweiss, obtained after crushing and pressing, which was kindly supplied by the Cellar at White Oak Winery in September 2020; and (2) pomace of red grape, cultivar Marquette, obtained after alcoholic, malolactic fermentation and pressing from a research project carried out by Dr. Watrelot, Iowa State University (Ames, IA) in October 2020. The pomace samples were packed in sealed plastic tubs (68 L) and immediately stored in a freezing room (−20°C) at Iowa State University until further use. The pomace samples were thawed at room temperature (25°C) for 24 h before drying.
Stems were manually separated from both pomace samples to collect seeds and skins. Also, the pomace of the 'Edelweiss' grape contained rice husk, commonly used to increase juice yield during pressing. After the drying process, the husk was manually separated prior to the grinding, using five selected U.S. standard sieves (Nos. 6,7,8,12,and 30) and a pan fitted into the last sieve. The depth and outer diameter of the sieves were 2.5 cm and 203 mm, respectively.

| Infrared heating system
A catalytic IR heating system (Catalytic Drying Technologies LLC), assembled in-house in the Department of Food Science and Human Nutrition at Iowa State University, was used for this study. The bench-top (30 cm by 61 cm) catalytic IR heater was powered by natural gas. The unit consists of a heating chamber, gas flow regulator, pressure gauge for gas, metal trays, and gas-powered catalytic IR emitters ( Figure 1).
Preliminary tests helped us to select three IR drying conditions in order to prevent GP from being burned. The IR treatments were performed at the following conditions of gas pressures (kPa) and product-to-emitter gap distances (cm): 0.2 kPa and 14.6 cm for IRlow heating (IR-LH), 0.5 kPa and two gap distances of 14.6 cm (first 8 min) and 22.2 cm (until MC < 10%) for IR-modified heating (IR-MH), and 0.3 kPa and 14.6 cm for IR-high heating (IR-HH). The product-toemitter gap distance and gas pressure change allowed for the variable emitter and product surface temperature ( Table 1).
The emitter and pomace surface temperatures were monitored periodically using a handheld VWR traceable dual laser IR thermometer (VWR International LLC).

| Experimental setup
The IR heating system was first preheated by using the electric heater for about 30 min. Then, the natural gas valve was opened and used to heat the emitter until the surface temperature of the IR emitter reached about 300-400°C. The electric power supply was disconnected after an additional 10-15 min after the gas had been introduced. Pomace samples (220 g) were spread out on nonstick aluminum dishes (40 cm long × 30 cm wide) in a single layer. When the GP sample reached about 40%-50% MC during IR and HA drying processes, the samples were ground and homogenized using a CapressoTM coffee grinder for 10 s.
Then, the drying continued for the homogenized GP until the final MC reached less than 10%. During IR heating, the ground sample was mixed thoroughly to provide uniform heating. Table 1, we devised three IR experimental setups capable of rapidly drying pomace without burning it; these parameters were determined through a series of trial tests. The IR-low heating (IR-LH) was run at a gas pressure of 0.2 kPa, and the product-to-emitter gap distance of 14.6 cm provided the low temperature of the product surface (71-78°C), which is IR treated.

As stated in
IR-high heating (IR-HH) was done at a gas pressure of 0.3 kPa, and the product-to-emitter gap distance of 14.6 cm provided a higher temperature of the product surface (131-141°C). For IR-modified heating (IR-MH), the IR heating system was adjusted at 0.5 kPa gas pressure. Next, the pomace samples were dried at two different product-to-emitter gap distances: drying at the gap distance of 14.6 cm to reach 40% MC, then increasing the gap distance to 22.2 cm, which provided a medium temperature of the product surface. The increased gap distance of 22.2 cm was used in order to avoid overheating or burning of the low-moisture grape pomace at the final stages of drying.
For HA drying, the GP sample of 220 g was spread out on the same nonstick aluminum pan at a single layer and was dried with a HA dryer at 70°C. The sample was ground at 40%-50% MC and then drying was continued until MC was reduced to less than 10%. F I G U R E 1 Infrared (IR) heating system (a); schematic of IR treatment of grape pomace (b). The pomace weight loss over time was monitored intermittently throughout the HA and IR drying by taking out the sample and weighing using an electronic balance (Quintix 412-1S; Sartorius) with a sensitivity of 0.01 g and placing back in the dryer very quickly.

IR treatments
During IR and HA drying, the samples were weighed after every 10 min and 30 min, respectively. The initial MC and initial weight, as well as the pomace weight at time t, were used to calculate the pomace MC at time t during the drying process. The GP samples dried with IR were compared with those dried with the HA as a reference for the conventional drying method.
After IR and HA drying, the GP samples were ground to a particle size of less than 1 mm using a CapressoTM coffee grinder for 1 min.
The dried and ground samples were placed in sealed plastic containers and stored in a freezer at −20°C until further analysis.

| Moisture content analysis
The initial and final MC of pomace samples were determined using a forced air oven (Isotemp Oven model 655F; Fisher Scientific) set at 105°C until reaching consistent weight, and the percentage of weight loss was calculated as wet based (AOAC, 1997). The moisture reduction during drying periods was frequently monitored by measuring pomace weight loss throughout the drying process. All MCs are on a wet basis (w.b.) unless otherwise specified.

| Color analysis
The color parameters, L* (lightness), a*(redness), and b* (yellowness) values of dried GP, were determined using a Hunter Lab ColorFlex EZ colorimeter (Hunter Associate Laboratory Inc.) on 500 mg of dried GP uniformly spread as a layer in a petri dish (60 × 15 mm).

| Total phenolic and tannin contents
Phenolic compounds were extracted from raw pomace and dry ground pomace using a conventional technique, where 10 ml of 70% acetone containing 0.05% TFA was added to 500 mg of GP and flushed with nitrogen before shaking overnight in dark at room temperature. After filtration using a Buchner and paper filter Whatman™ grade 1, the solvents were evaporated using a heating block at 32°C with a nitrogen flush to avoid oxidation. The aqueous extracts were then frozen at −80°C and freeze-dried. The amount of the extract was recorded.
After extraction, the total iron-reactive phenolic content was determined using the Adams-Harbertson assay (Heredia et al., 2006).
Briefly, 5 g/L of dry extract in model wine (13% ethanol, 5 g/L tartaric acid, pH 3.5) was prepared, and 75 μl of this solution was placed in a 1 cm pathlength cuvette, and 800 μl of resuspension buffer (pH 9.4) was added. After vortexing and 10-min incubation at room temperature, the absorbance was measured at 510 nm using a UV-Visible spectrophotometer (Genesys 150, Thermo Scientific). Next, 125 μl of ferric chloride solution was added, and 10 min after vortexing, the absorbance was read at 510 nm. The total iron-reactive phenolic content was then expressed as mg (+)-catechin equivalents per gram of GP extract.
The tannin content from the dry extracts was measured using the protein precipitation method with bovine serum albumin (Heredia et al., 2006). The same solution of dry extracts in model wine was used for this method, and the tannin content was determined and expressed as mg/g of dry extract as (+)-catechin equivalent.

| Antioxidant activity
For the antioxidant activity of the dry pomace extracts, a stable free radical α-diphenylβ-picrylhydrazyl (DPPH) was used, and the scav- were then used to determine the concentration of pomace extract solution required to inhibit 50% of DPPH.

| Microbiological analysis
The number of viable microorganisms for selected microbial groups,  bacteria or Salmonella were below the analytical method's detection limit (10 CFU/g = 1.0 Log CFU/g) and were, therefore, not shown.

| Data analysis
All analyses were done in triplicate, and the results were reported as mean ± standard deviation (SD). Statistical results were significant when p < .05. A one-way ANOVA was carried out on all the samples using a Tukey's test (LSD) with 95% confidence, using XLSTAT 2021 software.

| Moisture removal
The initial MCs of Edelweiss and Marquette pomace were reduced from 59.7 ± 3.6% and 50.1 ± 1.5%, respectively, to a final MC of less than 10%. The total drying rate of both GP samples using catalytic IR heat was faster than that of HA drying (

| Color
The L*, a*, and b* parameters of 'Edelweiss' and 'Marquette' pomaces dried with IR and HA are shown in

| Phenolic and tannin contents
The phenolic and tannin contents of raw and dried GP samples are shown in Figure 2. The phenolic and tannin contents in raw then are pressed to obtain the wine (rich in tannins) and the pomace. Therefore, it was expected that tannins were released in red wine, and a very low content would be in the red GP. The tannin content in 'Edelweiss' GP was lower than in 'Marquette' GP. This result was in agreement with the hypothesis that the low tannin content in interspecific hybrid red wines might be due to a higher retention of those compounds in the cell wall matrix, including proteins and polysaccharides (Springer & Sacks, 2014;Watrelot, 2021).
The drying process can explain the higher phenolics content in dried GP obtained in the current study as it can break and disrupt cell walls, providing large intercellular spaces to extract cellular substances better (Drosou et al., 2015). However, researchers have reported different temperature impacts on phenolic compounds and tannin content. Similar to our study, the thermal treatment of grape seed extract and GP in an electric furnace at 100°C did not change the total extractable polyphenol and tannin contents after 15, 30, and 60 min (Chamorro et al., 2011). However, it was found that the total phenolics content of GP dried in a climatic chamber at 40°C and 10% relative humidity was significantly higher (three times or more) than that of fresh GP due to extractability improvement of phenolic compounds (Carmona-Jiménez et al., 2018). On the contrary, another study indicated that higher drying temperatures caused a higher loss of phenolic compounds as the phenolic content of raw GP (165 mg GAE/g) decreased by 87.0% and 95.4% after drying at 60°C for 26 h and at 85°C for 5 h, respectively (Goula et al., 2016). Room temperature and vacuum oven drying at 70°C also resulted in a significant loss of phenolic compounds because of a longer drying time and thus higher anthocyanin degradation (Yu, 2014). In another study, conventional drying at 40°C, 55°C, and 70°C for 2 h resulted in a significant reduction of bioactive compounds and antioxidant activity compared with intermittent drying followed by 5 and 10 min of tempering. The GPs dried using intermittent drying at 40°C for 5 min showed the lowest loss of bioactive compounds, including 20.6% and 14.5% for total phenolic compounds and total flavonoids, respectively (Borges et al., 2020). There was also more than a 30% reduction of bioactive compounds when two red wine grape pomaces of Pinot Noir and Merlot were dried using 40°C conventional and vacuum drying or ambient air drying (25°C) compared to freeze drying (Tseng & Zhao, 2012). In the current study, IR drying was found to be the practical drying method to preserve the phenolic compounds of 'Edelweiss' pomace but led to a reduction of phenolics content

| Antioxidant activity
A significantly lower antioxidant activity was observed in 'Edelweiss' pomaces dried with IR-HH than dried with HA, as the concentration of 'Edelweiss' pomace dried with IR-HH was higher than HA to inhibit 50% of DPPH ( Figure 3).
On the other hand, the antioxidant activity was higher in 'Marquette' pomace after IR-MH and IR-HH drying than dried with HA. The increase in antioxidant activity could be explained by more accessibility of phytochemical compounds from the cell matrix and higher radical-scavenging activity (Gahler et al., 2003). in the antioxidant activity of GP peel was caused by drying temperatures at 100°C and 140°C, respectively, using an air-circulating oven (Larrauri et al., 1997). On the contrary, thermal furnace and autoclave treatment at 100°C did not change the antioxidant activity of grape seed extract and GP (Chamorro et al., 2011).

| Microbiological quality
The IR and HA drying methods used in the current study reduced the microbial count present on the GP. The IR heating treatments  yeasts (Sui et al., 2014). In the current study, the significantly higher reduction in APC caused by IR-MH and IR-HH than IR-LH could be attributed to the higher temperature received by the product surface. The IR treatment appeared more advantageous than convective heating in killing heat-resistant micro-organisms caused by penetrating cells and damaging the protein inside (Sui et al., 2014).
Other irradiation treatments rather than IR heating have also shown microbial inactivation. For example, gamma irradiation at 6 kGy inhibited bacteria and Y&M growth during the first 3 and 4 days after storage at 24°C, respectively, suggesting that treated GP can be stored for a longer time (Ayed et al., 1999).
Further work is needed and currently focuses on evaluating the chemical and microbial stability of flours obtained from IR-treated GP under various storage conditions.

| CON CLUS ION
The effect of IR heating on the valorization of cold-hardy GP in terms of drying and stabilizing the wine GP was investigated. The effectiveness of IR drying was compared with conventional HA drying. The physicochemical properties, including color, phenolic degradation, and microbial inhibition, were evaluated. All the IR heating treatments could reduce the grape MC in a significantly shorter time than HA. The best IR drying parameters were found to be IR-MH, which resulted in the fastest drying of GP. The HA and IR heating did not cause any significant differences in color and phenolic degradation. However, the drying method impacted the antioxidant activity depending on the type of grape, which could be attributed to the presence of other antioxidants. On an average, a higher microbial inactivation was achieved when the GP was treated with IR heating. Overall, IR heating, mainly IR-MH, was the most suitable processing treatment for GP with shorter drying time, high quality, and safety efficiencies. Establishing a rapid and effective drying approach is essential because of the lengthier drying times and losses in polyphenolic content associated with traditional drying processes.

ACK N OWLED G M ENTS
The authors acknowledge the Cellar Winery at White oak, Cambridge, IA, for providing Edelweiss pomaces after pressing. The author would like to thank Dr. Aubrey Mendonca and his students for the microbiological analysis of grape pomace dry samples. Open access funding provided by the Iowa State University Library.

FU N D I N G I N FO R M ATI O N
This study was supported by the U.S. Department of Agriculture's (USDA) Agricultural Marketing Service through grant SCBG-1427-4.
Its contents are solely the authors' responsibility and do not necessarily represent the official views of the USDA.

CO N FLI C T O F I NTE R E S T
The authors declare that there is no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.