Novel Method for The Valorization of Essential Oils From Orange By-Product


 The authors have requested that this preprint be removed from Research Square.


Background
Citrus production has been considered one of the rst natural production in Lebanon. These fruits grow on the southern half of Lebanese coast and represent half of the total agricultural output in the country. Citrus production increased over the year of 2001 and peaked in 2004 to reach a production of 395,300 tons. Due to the war in 2006, this production has decreased to stand to a range of 230,497 tons in 2013. (Mikhael and Saadeh, 2016) In manufacturing, these fruits are utilized for juices (fresh or commercial) or citrus based production, in consequence a huge amount of wastes, mostly peels, are trashed. 50-60% of yearly processed citrus products are counted as wastes called "pastazzo". Industries seek for a waste management procedure to minimize the costly disposal of these waste materials (Sharma et al., 2017).
Food waste has always been a worldwide problem, where a huge number of by-products annually get lost and un-used. Extracting essential oil from citrus fruits by-products is considered a valid approach to minimize the fruit wastes as well as valorising it through the production of food preservatives, avours, and cosmetics (Negro et al., 2016). Citrus waste contains valuable compounds in their pulp, seeds, and peels 4 , such as avonoids, dietary bres, polyphenols, carotenoids, ascorbic acids and essential oil 2 .
Polyphenols and carotenoids are known to have various health bene ts especially their antioxidant activities, owed to their polyphenols contents presenting a variety of reported biological properties such as skin anti-agent, anti-carcinogenic and anti-allergenicity (Etebu and Nwauzoma, 2014). In addition, they play a major role in the cosmetic and pharmaceutical elds. Essentials oils are combination of many compounds but mainly consist of isoprenoids, monoterpenes, and sesquiterpenes, they are responsible for the aroma of different plants, which can be used in pharmaceutical industries. They can be added to food to enhance avor and can be used as natural antimicrobials 4 . Researchers found that citrus essential oils can control the growth of a broad range of bacteria without causing any detrimental health effect (Martos et al., 2008).
Hydrodistillation is the traditional technique used to extract essential oils. However, this method was shown to be associated with a long extraction time, degradation in the quality and the chemical composition of extracted oil, large amount of used organic solvent and a poor extraction e ciency (Zheljazkov et al., 2014). Ultrasound technique has shown to be a good alternative to decrease extraction time, maintain quality and avoid the use of organic solvent 8 .
In this sense, and due to the disadvantages of the traditional method, and considering that not too many articles have been published on ultrasound pretreatment prior to hydrodistillation, and to cover the lack in literature. Extraction of essential oils from orange by-products, based on ultrasound assisted extraction combined with hydrodistillation will be assessed in this study. Afterward, Gas chromatography analysis coupled with mass spectrometry identi es oil components beside many other analytical tests to monitor the quality and safety measures of the prototype extracted oil. Two commercial orange oils have been tested to create a valid comparison.

Materials And Methods
Plant Materials: The orange peels, resulting from orange pressing, were collected from two different juice stores located in Beirut, Tarik El Jdideh area, Lebanon. Zest from orange peels were removed using a peeler.

Methods of extraction:
Hydrodistillation Method: A 101.5 g of orange zest were immersed in 200 ml of distilled water. The zest was disposed in a balloon heater attached to a condenser to ensure the condensation of E.O. The extraction was carried on for 6 hours at 100˚C. At the end of distillation, two phases were obtained; a water phase (aromatic water) and the E.O, less dense than water (Louaste et al., 2003).
Prototype Method: The prototype developed in our study is a combination between hydrodistillation and ultrasound assisted extraction-method.
A mass of 682.21 g of orange zest was disposed in a large tank with a 3500 ml of distilled water heated to 100˚C and attached to a condenser to ensure condensation of the vapor by cooling. An ultrasonic transducer is attached to the tank and a frequency of 28.9 Hz for the rst 10 minutes was performed. After one hour two phases were obtained: water and E.O.
Yield and extraction time: The yield of extracted orange E.O were expressed in g relative to 100 g of orange peels and it was calculated according to the following equation: 10 The extraction time by the two methods was expressed per hour.

GC-MC analysis:
GC-MC analysis were performed using a gas chromatography coupled with mass spectrometer, equipped with a split injector for separation and quanti cation of E.Os compounds (Koleilatn et al., 2017).

Qualitative Test:
Two commercial orange essential oils and the extracted oil by prototype were used in this study.

Free radicals/DPPH:
Different concentrations of all essential oils (25, 50, 100 μg/ml) were diluted ve times with DPPH solution in 0.4 Mm methanol. The blank will be consisted of 0.4 Mm methanolic solution of DPPH. After 30 minutes of incubation at room temperature, the reduction in the number of free radicals will be measured by reading the absorbance at 517 nm using a spectrophotometer.
The percentage inhibition of DPPH radicals by each E.O will be calculated according to the following formula: % of inhibition= [(AB -AA) /AB] × 100 Where AB absorption of blank sample (t=0) and AA = absorption of tested oil (t=30 min).
IC50 values which represents the concentration of E.O that caused 50% scavenging, will be determine from the plot of inhibition percentage against concentration (Abd et al., 2011). Refractive Index: The refractive index of the 3 orange oil samples were measured using a REF 123® digital refractometer.
Relative Density: According to Cristina Fabine et al., (2008), the relative density of essential oil was calculated using the following formula of the ratio of mass of the liquid sample and the mass of water.
Where: C= density g/ml m= mass per g V= volume per ml Total Phenolic compounds: According to the method described by Taga et al., (1984), 100 ul of each orange essential oil sample will be dissolved in 10 ml of 0.4 Mm methanol, and 2 ml of this solution will be made up with 0.3% HCl to 5 ml.
A 100 ul aliquot of the resulting solution will be added to 2 ml of 2% Na 2 CO 3 and after 2 min, 100 ul of Folin-Ciocalteau reagent (diluted with methanol 1:1) will be added and mixed well. After 30 min incubation, the absorbance of the mixtures will be recorded spectrophotometrically at 750 nm.
The total phenolic contents will be calculated as gallic acid equivalent (GAE) from a calibration curve of gallic acid standard solutions and expressed as mg of gallic acid per 100 ul of essential oil sample (Abd et al., 2013).
Refractive index, relative density, DPPH and total phenolic compounds tests were conducted in triplicate.

Pesticides Residues:
To quantify pesticide residues in oil samples, 10 ml of acetonitrile was added to 10 ml of orange E.O samples and shaken vigorously before adding Magnesium sulfate MgSO 4 (4g) , sodium chloride NaCl (1 g), sodium dibasic citrate (1g) and sodium tribasic citrate (0.5g) and then shaking for 3 minutes. The tube samples were then subjected to a centrifugation at 5000 rpm at 4˚C for 5 minutes. 5 ml of the aqueous supernatant were mixed with dispersive SPE EMR (speci cally designed to remove lipid from high-fat samples) dissolved in 5 ml water then shaken vigorously for 5 minutes. Another centrifugation was conducted at 5000 rpm at 4˚C for 5 min. 5 ml of the aqueous supernatant obtained will be transferred in a new polish tube then shaken vigorously for 3 minutes before being centrifuged at 5000 rpm at 4˚C for 5 min. GC/MS-MS is done to identify the pesticides residues in each oil samples (Lebanese Agricultural Research Institute (LARI)).

Microbiological Analysis:
Total counts: On a plate count agar (PCA) 100 μl of oil were added and rotated gently to ensure uniform mixing of the oil with agar. The plates are incubated at 30˚C for 48 hours. The microbiological analysis of the oil was performed at the Microbiology laboratory, (Lebanese Agricultural Research Institute (LARI), Fanar,

LEBANON)
Total coliforms: On a violet red bile agar, 100 ul of oil were added and spread on the agar. The plates are incubated at 37˚C for 24 hours.
Antimicrobial Effect: Essential oils: Two commercial orange essential oils and the extracted oil by the prototype were used in this study. The oils were rst dissolved in DMSO 1:1 (v/v) to give stock solutions. For the bioassay, the stock solutions of essential oils were sterilized using 0.45 m disposable syringe lters prior to the assessment of their antimicrobial effect. Stock solutions of essential oils were stored in dark bottles in the refrigerator at 4˚C for subsequent use.

Bacterial strains:
The bacterial strains used in this study were provided by the Department of Health Sciences in Beirut Arab University (BAU) and the American University of Beirut Medical Center (AUBMC). Two gram positive bacterial strains (Staphylococcus aureus and Listeria monocytogenes) and two gram negative bacterial strains (E.coli and Pseudomonas aeruginosa) were tested against the two commercial orange essential oils and the extracted oil by prototype and were used to form mono-species bio lms. The cultures of bacteria were maintained in their appropriate agar slants at 4°C throughout the study and used as stock bacterial cultures.

Inoculum standardization:
Bacterial inoculums were obtained from stock cultures and inoculated in Lysogeny broth medium then incubated at 37˚C for 18-24 hours. From the fresh grown cultures, decimal dilutions were made in sterile saline (0.9 %) until reaching a turbidity of 0.5 McFarland (10 8 CFU/mL), for testing the antibacterial effects of essential oils (ClSI, 2013).

Agar Disc Diffusion Method:
The antibacterial activity of essential oils was rst examined by agar disc diffusion method, which is the preliminary assay for screening the antibacterial activity of essential oils and to select between e cient ones (Burt et al., 2004). Agar disc diffusion was performed using an overnight bacterial culture, where 3 colonies of each strains to be tested were picked and inoculated in sterile saline solution and adjusted to approximately 10 8 CFU/ml, then 100 µl of the prepared bacterial suspensions were spread over plates of Mueller-Hinton agar (MHA) and left for 15 min at room temperature. To enhance the diffusion in the agar, essential oils were mixed with 10% aqueous DMSO and sterilized by ltration through a 0.45μm membrane lter (Wikler et al., 2018). Under aseptic conditions, 5 μl of essential oils in DMSO (1:1) were pipetted on the 5 mm sterilized lter paper discs placed on the top of inoculated MHA plates and kept for 30 min in the refrigerator to allow oil diffusion. Filter paper disc impregnated with DMSO was used as a negative control, while a standard disc containing gentamycin (10μg/disc) was used as positive control. All plates were incubated at 37°C for 18-24 h. After overnight incubation, the inhibition zones were recorded (Saleh et al., 2018).

Determination of MIC:
The minimum inhibition concentration (MIC) values were determined for the essential oils showing potent antibacterial effect against tested isolates using agar dilution method as recommended by the National Committee for Clinical Laboratory Standards (NCCLS) with some modi cations 18 . To enhance the solubility of essential oils, a nal concentration of 0.5% (v/v) Tween-20 was added to the MHA medium.
First, stock solutions of each tested essential oil (100 mg/ml) were prepared, followed by series of two fold dilution in Mueller-Hinton broth resulting in six concentrations (50 mg/ml, 25 mg/ml, 12.5 mg/ml, 6.25 mg/ml, 3.12 mg/ml and 1.56 mg/ml), then 1 ml of each prepared serial dilution was added to 9 ml of melted Mueller Hinton agar at 48°C, mixed thoroughly and poured in sterile plates. Plates were dried at room temperature for 30 min prior to spot with 3 μl aliquots of bacterial cultures containing approximately 10 4 CFU/ml of each tested isolate. Inoculated plates were incubated at 37°C for 18-24 h and the MICs values were determined. Plates of MHA without essential oils was used as a negative growth control. The MICs values were considered as the lowest concentration of oil resulting in the inhibition of visible bacterial growth on the agar plate (Hammer et al., 2001).

Mono-Species Bio lm Formation:
To assess the effect of essential oils on bio lm formation, bio lm on glass surface assay was performed, where the adhered bio lm to glass coverslips is visualized under light microscope 20 . Each bacterial strain was grown overnight in nutrient broth and diluted 1:5 in Luria-Bertani broth (LB), diluted cultures were used to immerse a sterile coverslip. In this study, sterile beakers for each bacterial strain containing 2.5 cm coverslip were used, 300μl bacterial suspension were added with equal volume of essential oil showing potent antibacterial effect against each isolate, the untreated beaker was used as reference control. After overnight incubation at 37˚C, each beaker was washed 3 times with distilled water, xed with 95% of ethanol for 30 min and then stained with 0.1% crystal violet for one hour at room temperature. After a nal wash, all coverslips were dried and microscopically visualized for bio lm formation 17 .

Results
A difference in the extraction time was noted between the hydrodistillation and the prototype as seen in Table 1; the hydrodistillation needed 6 hours to extract a volume of 0.35 μl (3.5×10 -4 ml) while the prototype extraction method took a duration of 1 hour to give a volume of 8.8 ml of oil. While assessing the percentage of yield in 100 g of orange peels, a percentage of 34 x 10 -5 (0.00034%) of oil was extracted using the hydrodistillation, though in prototype method in 100 g of orange peels, the percentage of oil extracted was 0.733%. While comparing the composition of the extracted essential orange oil using two different methods of extraction, as shown in Table 2, the total of essential oil was higher for the oil obtained by the hydrodistillation compared to the prototype one, with values of 99.03 and 95.17%, respectively. The compounds including α-Pinene (prototype: 4.20%, hydrodistillation: 2.10%), Sabinene (prototype: 2.24 %, hydrodistillation: 1.13%), β-Myrcene (prototype: 7.55%, hydrodistillation: 5.33%), 1-Octanol (prototype: 0.15%, hydrodistillation: 0.08%), Linalool (prototype: 2.24%, hydrodistillation: 1.28%), Trans-p-Mentha-2,8-dien-1-ol (prototype: 1.20 %, hydrodistillation: 0.04 %), Limonene 1,2-epoxide (prototype: 2.88 %, hydrodistillation: trace), Cis-p-Mentha-2,8-dien-1-ol (prototype: 2.40%, hydrodistillation: trace), α-Terpineol (prototype: 1.10%, hydrodistillation: 0.40%), (E)-Carveol (prototype: 2.20 %, hydrodistillation: 0.03%), (Z)-Carveol (prototype: 1.11%, hydrodistillation: 0.06%), Carvone (prototype: 3.40%, hydrodistillation: 0.05%), Perillaldehyde (prototype: 0.40%, hydrodistillation: 0.08%), p-Mentha-1,8dien-9-ol (prototype: 0.33%, hydrodistillation: 0.05%), Valencene (prototype: 0.38%, hydrodistillation: 0.20%), and α-Farnesene (prototype: 0.99%, hydrodistillation: 0.05%), were found to present higher percentages for the oils obtained by the prototype compared to hydrodistillation. For the prototype method, these values are approximately doubeled or tripeled however, some compounds are detected to be nearly the same (3-Carene, 1-Octanol, Nonanal, Decanal). β-Citral and α-Citral have shown a higher percentage of 0.02% and 0.05% respectively in the hydrodistillation method, while Octanal, Citronellal, α-Copaene, β-Cupepene, Dodecanal, βSinensal and α-Sinensal was identi ed as trace amount of less than 0.01% in the extracted orange 47 oil through prototype method. Nevertheless, Limonene 1,2-epoxide and Cis-p-Mentha-2,8-dien1-ol reveal a trace amount in the hydrodistillation method. According to the ndings of the GC-MC and while comparing energy, time consumption and yield, this new approach using a combination of methods the oil extracted using prototype method showed promising results. The results of this new approach, using the combination of hydrodistillation and ultrasound methods, have showed a promising new technology of essential oil extraction indicating better results in term of constituents, energy, yield and time consumption. From here, further tests should be conducted to assess the quality and safety measures of the orange EO besides the comparison of two others commercial ones. According to the DPPH free radical scavenging rate of the 3 orange E.O in different concentrations: the oil obtained by the prototype and two other commercial essential oils and as shown in Figure 2, the DPPH scavenging activity of oil increase when increasing the concentration of oil. The extracted orange E.O presents the highest % of free radical scavenging rate in all concentrations (36%, 41% and 57% in 25 ug/ml, 50 ug/ml and 100 ug/ml respectively). The commercial orange (CO2) has showed to have the lowest % of DPPH radical scavenging activity (10%, 11.5% and 14.8% in 25 ug/ml, 50 ug/ml and 100 ug/ml respectively). The extracted orange E.O has an IC50 of 87.7 ug/ml, the commercial orange (CO1) have 138. 8 Table 8 represents the MIC of selected E.Os on the three bacterial strains, ranging from 3.125 mg/ml to 50 mg/ml. The results showed that commercial orange E.O (CO2) showed the highest activity on Staphylococcus aureus and Listeria monocytogenes with MIC values 3.125 mg/ml and 12.5 mg/ml respectively followed by the extracted E.O that showed moderate MIC values ranging between 50 mg/ml against E.coli and 25 mg/ml against Staphylococcus aureus and Listeria monocytogenes. The potential of orange E.Os to prevent bio lm formation by the tested gram positive and gram negative was investigated in this study and visualized by the light microscope (Figures 3, 4, 5). The results revealed that the extracted orange E.O showed the capability to prevent the bio lm formation by the tested bacteria, where the highest reduction in bio lm formation was observed against E.coli followed by Listeria monocytogenes then Staphylococcus aureus when compared to control (untreated bacteria). On the other hand, the commercial E.O (CO2) exhibited a strong suppression of bio lm formation by Listeria monocytogenes followed by Staphylococcus aureus when compared to control.

Discussion
In this study, a new technique for essential oils extraction prototype (ultrasound extraction method combined with hydrodistillation method) has been evaluated comparing it with the traditional method (hydrodistillation) aiming to minimize the time and energy consumption, with a maximum yield without affecting the composition of extracted oil. The gas chromatography analysis of the extracted oils, indicated that limonene is the major component present in the oil with a percentage of 81.32% for the oil obtained by hydrodistillation and 60.23% for the one obtained by the prototype. This is in in agreement with the study of Fisher et al., 2008 who showed that limonene is present as a major component of 88.2%. In agreement with another study of Mou da et al., 2003, limonene is the most abundant compounds and its contents vary from a range of 63% to 90% in different orange E.Os. The prototype method reduces the time of extraction, and give a higher yield by 25 folds comparing with hydrodistillation, without affecting the composition of oil since both oil pro les share the same components but with different percentages as showed in our results. In agreement with Pingret et al., 2014, ultrasound technique provide an improvement in time over the conventional method and a higher yield by 15 to 26 folds. According to Teneva et al., 2019 the free radical scavenging activity show a higher ability when limonene is less active and below 80% , the gas chromatographic pro le of the extracted orange E.O using prototype show a less limonene percentage 60.23% than the oil extracted using hydrodistillation 88.2%. We can expect that the oil extracted using prototype to have a better radical scavenging activity than the oil obtained using hydrodistillation method. The qualitative test (antioxidant activity, total phenolic compounds, refractive index and relative density) were performed on the extracted

Conclusion
Orange essential oils have been extracted using two different methods and a comparison between the conventional (hydrodistillation) method and the innovative technique, combining hydrodistillation and ultrasound. The characteristics analyzed were extraction time, yield and essential oil composition. The higher yield of oil was obtained using the prototype technique, without affecting oil composition, with a minimal extraction time and energy consumption, leading to reduced burden on environment. Free radical, total polyphenol, refractive index, relative density, pesticides residues, total count, total coliforms, antibacterial activity and bio lms inhibition are the tests that have been conducted on the extracted orange E.O using prototype and two other commercial ones to create a valid comparison. All of the investigated oil produced a considerable DPPH scavenging activity, the highest one being for the extracted orange E.O. with the highest content of total polyphenol. The refractive index and relative density values of the three oil samples fall in the speci ed range according to EOA and ISO standards, respectively. Commercial oil (2) and the extracted oil contained pesticides residues, most properly due to the mal agricultural practices and the absence of laws and regulations. The three essential oils have shown to be bacteria free. The extracted essential oil shows an antimicrobial activity against both gram positive (Listeria monocytogenes and Staphylococcus aureus) and gram negative bacteria (E.coli), commercial orange oil (CO2) shows an antimicrobial activity only against gram positive bacteria, while the commercial oil (CO1) show no activity on any bacterial strain. None of the three orange essential oils shows an effect on Pseudomonas aeruginosa. The extracted and the commercial orange E.O (CO2), reduced the bio lm formation by all tested bacterial strains. This study has shown that the prototype combing hydrodistillation and ultrasound-assisted extraction to valorize orange essential oil from orange peels could be considerable a suitable alternative to extract essential oil and a replacement for chemical additives for use in the food industry, attending to the needs for safety and satisfying the demands of consumers for natural components. Because of the COVID 19 pandemic and the October revolution that happened while conducting the research work of this study, the antimicrobial activity of orange essential oils were not repeated in a duplicate or triplicate and it was one of the limitations of this study. Another problem was the availability of orange peels since every industry that uses orange fruit discard the peels immediately. As already known that essential oils are very expensive due to the costly and complicated production process, orange essential oils was not easy to nd in stores, and the two commercial ones that we worked on them in this study was of unknown origin, type of extraction or any other information.
For further research, studies should be carried out to assess the antimicrobial effect of orange E.O on a broader range of bacteria, to be able to understand the antimicrobial effect of the oil and how it can be applied to our food. Appling the oil to our daily food consumption like meat, fresh vegetables and fruits salad etc. and monitor if there is an improvement in the shelf life of the product taking into considerations the major sensory aspects of food. It is a well of interest to study the toxicity of some      Microscopic visualization of the effect of E.O on bio lm E.coli compared to control