High-intensity, Low-frequency Ultrasound Treatment as Sustainable Strategy to Develop Innovative Biomaterials from Agri-food Byproducts and Wastes

Bioconversion is an important avenue for finding value from biomass waste produced by the agricultural industry. One avenue of conversion is the development of upcycling byproducts and waste from food processing to value-added products. This includes degradable biomaterials which have real potential to reduce waste, improving economic, social and environmental impacts. As such, this research paper was focused on exploring two avenues of bioconversion from waste products of tomato skin, hemp meal and hops vines: identification of phytochemicals and the development of bioplastic. Combined to these researches, the effect of Ultrasound as a green technology was studied in both contexts. It was found that Ultrasound treatment reduced extraction time for saponin and phenolic acid from tomato skin, hemp meal and/or hops vines from 24h to 30 min. However, Ultrasound Assisted Extraction (UAE) was shown to affect the phenolic acid and saponin profiles of certain extracts. Ultrasound treatment was shown to positively impact the overall microscopic structure and qualities of bioplastic such as water activity, percentage moisture, hardness, cohesiveness, resilience, and springiness index. This study suggests that Ultrasound can be used as sustainable non-thermal method for extraction of active saponins and phenolics but also in bioplastic formulation to enhance physico-chemical characteristic.


Introduction
Agriculture is a pillar of human activities which allows for production of most if not all the food consumed.However, the agricultural sector is also responsible for the production of different wastes (Obi et al., 2016).The food/agricultural industry produces large amounts of plant materials a portion of which through improper management, overproduction or simply necessity ends up as waste.In Canada, it is estimated that about 40% of the food produced in farms is not consumed every year (Ishangulyyev et al., 2019).Furthermore, the number doesn't account for the by-products linked to the processing of the other 60 percent.
Finding value in waste is a sound economical proposition.Creating and implementing systems to process waste could lead to the development of new economic sectors (Romero-Herná ndez & Romero, 2018).By utilizing waste, the mass of waste produce diminished which is an important aspect of developing more sustainable practices.Biorefineries offer a solution by producing cellulose, lignin, hemicellulose, and other bio-product from the biomass.However, they can be complex processes with limitation in their outputs and further transformation of certain of these products when other than biofuels would need to happen (Clark & Deswarte, 2015).As such, the important of exploring other methods of bioconversion should not be dismissed.A perfect example of this mindset can be seen in the various researches dedicated to utilizing fruit waste for the recovery of pectin (Kluczkovski et al., 2020;Sarangi et al., 2023).
Creating value-added products like packaging products made directly from grinded left-over biomass could offer a simple but effective method (Menon & Rao, 2012).Similarly, Heinz recently started to find innovative ways to upcycle peels, stems and seeds from more than two million tons of tomatoes to develop sustainable composite materials for diverse utilizations (Ontario Processing Vegetable Growers, 2023).Other companies such as Ford joined up with Heinz and other companies such as Coca-Cola Company, NIKE Inc., to use the above composite materials to be used in vehicle manufacturing and alternative to traditional polyethylene terephthalate (PET) plastic bottles.To fill these huge gaps, developing degradable biomaterials containing phenolic antioxidants, in addition to finding new use for Ultrasound application as a green technique were the major aim of this study.
As this need to find value from waste is clearly identified, it is also important to note the importance of finding alternative sustainable converting processes (Vilkhu et al., 2008).
Ultrasound assisted extraction is a sustainable technique that has become of interest for its capacity to increase extraction efficiency while having a relatively low energy use (Herrera & Luque De Castro, 2004;Mazza et al., 2019).High-Power Low-Frequency Ultrasound application has been shown to decrease the time needed for extraction of a variety of samples through use of cavitation and detexturation of samples (Chemat et al., 2017).Through application of micro jet forces, Ultrasound breaks down the surface of plant matrix allowing for the different compounds to dissolve in the solvent in a more efficient matter than with regular stirring.Furthermore, Ultrasound treatment has been shown to allow the development of stronger double emulsion (Dornan, 2021).
On the other hand, plastic production and packaging are intrinsically linked as plastic is one of its mainstay components.In 2015, the global production of plastic was estimated at 350 million tons (Geyer et al., 2017).Although plastics are so commonly used and produced in enormous quantities, proper disposal systems are rarely put in place.Up to 2015, up to 55% of plastics were still disposed of mainly through landfills and although trends show a slow increase in the proportion of plastics properly disposed of through incineration and recycling, a lot of environmental damage has/is still taking place (Eriksen et al., 2014;Okunola et al., 2019) Plastics have slow degradation processes and the by-product of these breakdowns are often toxic for the environment (Arvanitoyannis, 2013;Gewert et al., 2015;Lebreton et al., 2017;Webb et al., 2013) Plastic are almost always made of non-biodegradable polymers making them non-biodegradable materials.Biopolymers and biomaterials, which are both by definition biodegradable present themselves as a powerful area of scientific exploration as they could potentially replace all together traditional plastic (Kanmani & Rhim, 2014).
It is in this optic that the focus of this study was to work on finding different avenues of utilization of cellulosic waste material.With this goal in mind, this study had a two pronged approached to further extract value from waste.The first aspect of the research focused on extracting valuable plant component such as saponins and phenolic acids which find use in the cosmetic industry or in the pharmaceutical industry.This was performed utilizing the previously discussed green technology of HPLF-US to increase the efficiency of these extractions.
The second aspect of our research focus on using the left over cellulosic material resulting from the UAE, as bulking component to develop bioplastics made entirely of biodegradable component.as to find alternative solution to improving sustainable production and future commercialization.To develop analytical protocols that can be used on different plant-based materials, it was important to assess the experimental methods on different plant sources.Therefore, working on diverse plant materials permitted to develop a suitable methodology and analytical protocols.Raw materials were not purchased from commercial sources but were obtained through collaboration with Agriculture and Agri-Food Canada (AAFC) from whom three available plants waste materials including tomato skin, hemp meal, and hops vines were provided.

Materials
Tomato skin, Hemp meal and hops vines were collected from a tomato canner, a hemp farmer and a hop farmer, respectively that collaborated with AAFC.Acetonitrile, HPLC grade methanol and acetic acid were obtained from Fisher Chemical (Ontario, Canada).Ethanol (100%) was from Commercial Alcohols (Ontario, Canada).Acetic acid was from Aldrich Chemical Company INC.Flavonoid, Phenolic compound standard, Aluminum Chloride powder and Folin-Ciocalteu's reagent were ordered from Sigma-Aldrich (Ontario, Canada) as was Sodium Acetate and Sodium Carbonate Anhydrous were from Bioshop (Ontario, Canada).Saponin standard of Tomatine was purchased from Biopurify (Sichuan, China).Glycerol, Phosphoric acid, and Calcium hydroxide were obtained from Fisher Chemical (Ontario, Canada).Gelatin was received from Sigma-Aldrich (Ontario, Canada).Sodium Alginate was sourced from Landor Trading Co. (Quebec, Canada).Casein Sodium was ordered from MP biomedicals (CA, USA).

Traditional Extraction of Tomato Skin, Hemp Meal and Hops Vines
A modified version of Dornan et al., 2020 was used to perform extractions.Grinded tomato skin, hemp meal or hops vines were extracted (1/20 w/w) using mechanical stirring or for 30 minute using UAE, set with a power of 90W and a frequency of 20 kHz in 5% acetic acid in ethanol.

Total Phenolic Content
The total phenolic content was measured using Folin-Ciocalteu assay (Gunenc et al., 2015).Absorbance was read at 725nm using Biotek Cytation Hybrid Multi-Mode Reader (Biotek, Winooski, VT, USA).The TPC values were expressed as mg of gallic acid equivalent per 100g of samples.

Total Flavonoid Content
The Total flavonoid content was measured through aluminium chloride colorimetric assay (Yang et al., 2012).Standard concentration of quercetin was prepared from a stock solution of 1mg/mL (0.01mg/mL, 0.02mg/mL, 0.04mg/mL, 0.05mg/mL, 0.06mg/mL, 0.08mg/mL and 0.1mg/mL).A volume of 1 mL of standard or sample were mixed with 50 mL of aluminum chloride solution (0.1mM) and 50 mL sodium acetate solution (0.1nM).

Total Saponin Content
Total saponin content was measured using a modified version of the method (Makkar et al., 2007;Navarro del Hierro et al., 2018) A solution of vanillin (80mg/mL) and standard solution of diosgenin of 0.1 mg/mL, 0.2 mg/mL, 0.3 mg/mL, 0.4 mg/mL and 0.5mg/mL were prepared.100mL of standard or samples were mixed with 100mL of vanillin solution and 1000mL of a 70% sulfuric acid solution.The solution was vortexed and incubated at 60°C for 10 minutes.Absorbance was read at 544 nm using Biotek Cytation Hybrid Multi-Mode Reader (Biotek, Winooski, VT, USA).

Oxygen Radical Absorbance Capacity
Oxygen Radical Absorbance Capacity assay were performed as outlined in Huang et al. (2002).Standard Trolox solutions (50 µM, 25 µM, 12.5µM and 6.25µM) in a K2HPO4 buffer (0.75M) were prepared.On a microplate 20µLof Trolox standard or diluted sample solutions were platted with 120µLof fluorescein solution.After incubating at 37°C for 30 minutes, 60µLof AAPH solution (153 mM).Absorbance was recorded each second for 60mins to create a curve of the absorbance in relation to time using a Biotek Cytation Hybrid Multi-Mode Reader (Biotek, Winooski, VT, USA).This functionality is partially resulting from the presence of phenolic compounds which were measured as described in section 2.4, by the total phenolic content/TPC assay.

Developing Bioplastic Formulation
Bioplastic were formulated based on previous research (Freire, 2019).Mass and role of each ingredient are outlined in Table 1.Biowaste materials were mixed casein, calcium hydroxide, canola oil, glycerol phosphoric acid.A previously-prepared mixture of boiling water, sodium alginate and gelatin were added to the biowaste material preparation.The bio-material paste was treated with US before being extruded or extruded directly.All resulting material were left to dry for 24h.

Texture Profile Analysis
Texture Profile Analysis was performed on the biomaterials using a CT3 Brooksfield Texture Analyzer set with a TA3/100 probe.Targer deformation was set at 0.15cm, trigger load at 6.8g, test speed and return speed at 0.05cm/s and recovery time was set a 60s.Hardness, Springness, Cohessiness, Resilience and Adhessiveness were measured as described by (Breene, 1975;Roopa & Bhattacharya, 2008).Each tests were performed in triplicates.

Water Activity and % Moisture
Biomaterial samples were analyzed using an Aqualab Model Series 3 water activity meter (Decagon Devide Inc, WA, US) and % moisture was also obtained after proper drying time.Each sample was measured as to obtain triplicates values.

Scanning Electron Microscopy
SEM images Biomaterial samples were obtained using a Tescan Scanning Electron Microscope at the Carleton Nano Imaging Facility.The most representative picture was selected from each triplicate.

Polarized Light Microscopy
PLM images of Biomaterial samples were determined by polarized light microscope (Axioplan 2 imaging and Zeiss Axiophot 2 universal microscope, Carl Zeiss Inc., Jena, Germany).The images were taken with a Retiga 1300 camera linked to Northern eclipse software.The distributed size from the images was analyzed via Image J software.

Statistical Analysis
Results were analyzed using IBM SPSS Statistics software (IBM corp, Armonk, New York, USA).Significant difference between triplicate values was determined using a statistical analysis of variance ANOVA.A P-value inferior to 0.05 indicated a significant difference between the triplicate values.The mean values were compared using Duncan's Multiple Range

Effect of UAE on yield of extract
The percentage yield of final extracts using mechanical stirring for 24h and 30 minutes HPLF-US treatment as extraction methods were compared for each of the samples obtained as seen in Table 2.When looking at the difference between the sample types, there was no significant difference in final yield between tomato skin, hemp meal and hops vines indicating that for this extraction methodology, yield was independent of sample type.Despite slightly lower yield of samples for both tomato skin and hops vines, extraction methods did not affect the TPC, TSC and the TFC as will be discussed in the next sections.

Effect of UAE on TPC
The Total phenolic content (TPC) of Ultrasound extracted (US 30 min) and traditional mechanically extracted (24h), tomato skin, hemp meal and hops vines are shown in Table 2.There was no significant difference (p>0.05) between the US treatment (US 30 min) and 24h extraction (24h) for any of the three samples (tomato, hemp, and hops).However, there was a significant difference in average TPC between the hop vines extraction and the tomato and hemp extractions.No significant difference was found between tomato skin TPC and hemp meal TPC.As such, hops vines resulted in higher TPC than both hemp meal and tomato skin.Furthermore, from the lack of significant difference mentioned between the TPC associated to US and traditional extraction of any of the sample, it can be mentioned that UAE doesn't influence the TPC when compared to a 24h traditional extraction.

Effect of UAE on TFC
The total flavonoid content in tomato skin, hemp meal and hops vines are shown in Table 2. Values for the TFC of US Samples, 24h samples for both hemp meal and tomato skin were observed to be respectively 19.92±2.24mg of QE/g of sample, 15.08±1.37mg of QE/g of sample, 31.75±6.30mg of QE/g of sample and 51.31±3.52 mg of QE/g of sample.Tomato skin and hemp meal extractions showed no statistical difference (p> 0.5).However, both hop sample extraction had significantly higher TFC, 171.10±2.66mg of QE/g of sample for the US extraction and 131.80±13.79mg of QE/g of sample for the 24h extraction, when compared to all other extractions.This is in accordance with the results in Table 2, which showed significantly higher results for the hops extraction.Samples with significantly higher phenolic concentration also showed a significantly higher concentration of flavonoids.There was no significant difference in TFC between 24h and US extracts of any of the sample tomato, hemp or hops samples indicating that TFC was not influenced by the difference in extraction techniques.

Influence of UAE on Phenolic Acid Profile
Protocatechuic acid, hydroxybenzoic acid, chlorogenic acid, syringic acid, p-coumaric and ferulic acid were all present in both extracts.Furthermore, no statistical difference (p>0.05) can be noted for the concentration of any of the phenolic present in both samples.Hemp extract on the other hand show significant difference in both composition and concentration of phenolic acids.24 hours extracted hemp sample (Hemp 24h) shows a higher diversity of phenolic acid when compared to the 30 minutes US extracted samples.Both hemp seed TPC and phenolic profile is also in accordance to previously obtained results such as the one presented in (Teh et al., 2014).As such, two more phenolic acid are present in the 24h extract vanillic acid and syringic acid.In both samples, chlorogenic acid, p-coumaric, and ferulic acid were present in concentration that were not shown to be statistically different.On the other hand, the concentration of protocatechuic acid and hydroxybenzoic acid was shown to be statistically different (p<0.05) in both hemp extract.The 24 hours extraction showing higher average concentration of both these compounds.Hops vines extract had similar phenolic acid profile.In both extracts the following phenolic acid were detected: gallic acid, protocatechuic acid, hydroxybenzoic acid, chlorogenic acid, vanillic acid, syringic acid, p-coumaric acid, ferulic acid and o-coumaric acid.Gallic acid, p-coumaric acid, syringic acid and ferulic acid were all found to be similar concentration in Hops 24h and Hops US 30 min, meaning there was no statistical difference between the concentration resulting from the analysis.Alternatively, protocatechuic acid, hydroxybenzoic acid, chlorogenic acid and o-Coumaric acid were present in statistically higher concentration in both the 24h hop extract and the US extract.As such hops showed the most diversity in phenolic acid composition between all 3 plant samples.UAE extraction for both tomato skin and hops vines increases the concentration of phenolic compound extracted.However, the same cannot be said for the extraction of hemp where traditional method showed overall higher diversity and concentration of phenolic.This could indicate a need to increase time of extraction for hemp sample as to increase the efficiency of US as has been previously reported in (Medina-Torres et al., 2017).

Influence of UAE on Flavonoid Profile
The flavonoid profile of the various tomato skin, hemp meal and hop vines extracts are summarized in Table 2.
Tomato skin extracts showed the same flavonoid profile on both Tomato US 30 min or Tomato 24h.Catechin, Rutin and Quercetin were all found in both tomato extracts.These results are in accordance to Dumitrash et al., 2016.Where UAE extraction was shown to decrease the time of extraction in tomato seeds.The concentrations of each of those phenolic compounds were not statistically different.Hemp US 30 min and Hemp 24h showed a significant difference when it came to the concentration of the various flavonoid identified.Both extracts showed the presence of the same 4 flavonoids: catechin, rutin, quercetin and kaempherol.All these compounds were found in higher concentration in the 24h extract aside from kaempherol.Kaempherol concentration was not found to be statistically different between US extracts and the 24h extracts.Hops vines extracts showed difference in the flavonoid profile.As such, Hops US 30 min had a broader diversity of flavonoid showing the presence of pyrogallol, catechin, rutin, myricetin, quercetin and kaempherol.Comparatively, pyrogallol was not found to be present in the 24h extracts.The concentration of catechin and quercetin was found to statistically higher in the US extract.Rutin concentration appeared to be significantly lower in the UEA extracted sample than the 24h extracts.The concentrations of myricetin and kaempherol of both hop extracts were not found to be statistically different.Additionally, Hops was found to have the highest diversity of flavonoids of the three plant samples.As such, the total flavonoid content were not affected by UAE for both tomato and hops extract slight difference could be noticed in hemp extracts.This could be remedied by performing longer US extraction as to ensure further cavitation and sonoporation of the sample which would logically lead to further leaching of bioactive compounds still present in the plant matrix.

Effect of UAE on TSC
The total saponin content was measured for each of the samples (tomato skin, hemp meal and hops vines) extracts.).There was no significant difference found between the TSC of US extracts and the 24h extracts (p >0.05) in tomato extracts.Hemp extracts also did not show any significant difference in TSC when extracted with US vs the traditional method, the measured TSC was respectively found to be 1511.

Effect of UAE on Saponin Profile
Table 3 shows the concentration of Tomatine present in the different extracts.Tomatine was found only in tomato extracts.Tomato extracts show the presence of tomatine which is in accordance with previous literature (Friedman, 2002).Furthermore, US treatment leads to a decrease in the time necessary for extraction from 24h to 30 minutes to extract saponins.

Influence of UAE on ORAC
The Oxygen radical absorbance capacity (ORAC) was measured for each extraction method and each sample.(Khan et al., 2010) where similar results were observed when comparing the standard and Ultrasound assisted extraction of orange peels.

Overall Appearance of Bioplastics
In Figure 1, little difference between No US samples and US samples.However, US treated samples of both Hemp and tomato were slightly oily toward the center of the sample surface when compared to the No US samples.This oily sheen could potentially be canola oil separating from the bioplastic structure.As canola oil was used as an external plasticizer it is possible that US treatment increased the rate of separation of canola oil from the biopolymer structure (Tyagi & Bhattacharya, 2019).Hops-based bioplastics were much drier and rougher in appearance Overall, US treatment appeared to influence the overall appearance of both hemp and tomato sample only slightly.Hops-based bioplastic had a different appearance from the other samples.No bubbles were caught in the matrix of the biopolymer and the overall appearance looks jagged and rougher.These finding can also be noted in the SEM pictures of hops-based samples.US treatment increases the roughness of the biopolymer surface.Small crater like structures are present on surface of the biopolymer gel hinting at US induced cavitation.This indicate the lack of biopolymer to coat completely the hops vines particles as uncoated fiber of cellulosic material as can be seen in the 40X pictures of most samples.In US samples, small tunnel like structure are present.These results support the hypothesis that sonoporation/cavitation influences the surface structure of the developed bioplastics.
These results indicate that US treatment influence both the internal structure of the gel emulsion and the surface texture of our samples.

Figure 1 .
Figure 1.Overall appearance of bioplastic made from tomato, hemp and hops with Ultrasound treatment and no Ultrasound treatment

Figure 2 .
Figure 2. Polarized Light Microscopy images of biomaterial with Ultrasound treatment and no Ultrasound treatment under a magnification of 10x

Table 1 .
Overall formulation of bioplastic per 100g

Table 2 .
Phenolic Profile and Antioxidant activity of plant extracts

Table 3 .
TSC and tomatine content in mg/g of dried grinded tomato skin, hemp meal and hops vines after extraction Results are represented as mean values ± SEM with differing subscript letter representing statistical difference.* DE: Diosgenine equivalent Tomato extracts showed values of 733.44±161.73mg of Trolox/g of sample for UAE and 1070.28±166.83mg of Trolox/g of sample and the traditional extraction.Hemp US and Hemp, No US had values of 1273.16±240.06 and 1142.86±240.06mg of Trolox/g of sample.Hops on the other hand showed values of 1624.59±432.66 and 1649.57±306.86 mg of Trolox/g of sample for the US extracts and 24h extracts.The method of extraction did not influence ORAC as there was no significant difference between US extracted sample results and 24h extracted sample results.As ORAC values are related to the antioxidative activity of phenolic and flavonoid compounds, these results correspond to the data measured in section 3.3 to 3.5.Other studies come to support that UAE does not lead to lower antioxidant activity of extracts as long as these values correlate with the phenolic content of a sample extracted in