Microwave-Assisted Extraction of Phenolic Antioxidants from Potato Peels

A response surface method was used to optimize the microwave-assisted extraction parameters such as extraction time (t) (min), solvent (methanol) concentration (S) (v/v) and microwave power level (MP) for extraction of antioxidants from potato peels. Max. total phenolics content of 3.94 mg g−1 dry weight (dw) was obtained at S of 67.33%, t of 15 min and a MP of 14.67%. For ascorbic acid (1.44 mg g−1 dw), caffeic acid (1.33 mg g−1 dw), ferulic acid (0.50 mg g−1 dw) max contents were obtained at S of 100%, t of 15 min, and MP of 10%, while the max chlorogenic acid content (1.35 mg g−1 dw) was obtained at S of 100%, t of 5 min, and MP of 10%. The radical scavenging activity of the extract was evaluated by using the DPPH assay and optimum antioxidant activity was obtained at S of 100%, t of 5 min, and MP of 10%.


Introduction
The potato (Solanum tuberosum L.) processing industry is one of the largest food processing industries in Canada and the United States. In 2008, some 21.8 million tonnes of potatoes were produced, and over 50% of were processed, mostly for French fries (60%) and chips (22%). Waste generation from potato processing ranges from an estimated 33% to 35% of the original potato fresh weight [1]. At present, a large proportion of this waste is used as an animal feed and for the production of bio-fuels. Recent studies, however, have also described the use of potato peels as a source of natural antioxidants to increase the shelf life of several food products, including soybean oil [2] and processed lamb meat [3]. concentration (S) and time (t) significantly influenced total phenolics content (p ≤ 0.0001 and p ≤ 0.005, respectively) as did their quadratic terms (S 2 and t 2 ; p ≤ 0.0001 and p ≤ 0.05, respectively) ( Table 1), whereas neither microwave power (MP) nor its quadratic term (MP 2 ), nor any parameter interactions (e.g., S × t, MP × S...) had any effect (p > 0.05). The predicted regression model for total phenolics (Phen tot ) can be described in terms of coded factors, with S and t as the significant factors: 2 A plot of measured vs. predicted values of Phen tot (Figure 1) shows the close agreement between these values, suggesting that the model and resulting response surface can be used to predict total phenolics under different experimental conditions.  The results obtained were in accordance with those of Rodriguez de Sotillo et al. [16], who studied extraction and stability of phenolics extracted from waste potato peel using aqueous solutions of methanol, and observed that with an increase in temperature, the total phenolic yield increased significantly, however it also changed the composition of the phenolics obtained. Similar results were obtained in the present study where the total phenolic yield and composition of the phenolics varied with changes in experimental parameters.
Similar regression models were developed for ascorbic acid and other individual phenolics using the software. Only the significant factors were used to obtain the predicted concentration for the individual phenolics. For ascorbic acid (AA) only MP showed a significant and inverse effect on ascorbic acid content (p ≤ 0.01), while all other parameters, interactions and quadratic terms were not significant (p > 0.05). Equation 2 shows the relationship between MP and AA content: This result indicates that changes in solvent concentration and extraction time had no significant effect on the concentration of the AA obtained, but power had considerable effect on the response when MAE was used to extract AA from freeze dried potato peel. A plot of predicted vs. actual AA content ( Figure 2) shows that experimental values are widely scattered, likely because AA is heat sensitive and the temperature reached as a result of the microwave power levels and solvent concentrations used was high enough to oxidize it.

Radical
The well extracts. On while all oth shows the re

Respons
Regressio ChloA, Caf ndependent The respo phenolics (F ncreased wi A, en al ct The linear and squared terms of the solvent concentration had a significant impact on the model. Similar results were obtained when extracting phenolics from peanut skin [24]. In this study, Phen tot content increased with an increase in methanol concentration up to 65-70% and then declined. The optimum extraction condition for total phenolics was estimated to be 67.33% aqueous methanol, 15 minutes extraction time, and 14.67% power. The maximum predicted Phen tot content under such conditions would be 3.94 ± 0.21 mg GAE g −1 . Microwave power had no significant impact on the Phen tot , content though at higher power levels Phen tot increased to predicted value by reducing the extraction time and increasing solvent concentration. The Phen tot value obtained under the experimental conditions used ranged from 1.2 ± 0.8 to 3.9 ± 0.21 mg GAE g −1 dry potato peel powder. Several researchers using conventional method of extraction reported a range of 1.51 ± 0.17 to 3.32 ± 0.12 mg GAE g -1 [11], 0.48 mg g -1 [16]. Al-Weshahy et al. [11] reported that the highest Phen tot values were found in coloured potato varieties. Other researchers have also shown that coloured potato varieties have more GAE total phenolic content [4,28]. In our study we used the brown skin variety 'Russett Burbank' for MAE optimization. Our results indicate that MAE exhibited a better efficiency than conventional methods of extraction (Table 2). The response surface plot for AA content ( Figure 4) shows that solvent concentration was highly significant, whereas time and microwave power may play only minor role in the extraction process within the ranges tested. Microwave power affects the temperature, i.e., an increase in power level will increase the temperature of the extraction process. AA heat-sensitivity is reflected in a significant decrease in its concentration in the extracts achieved with increasing power. Extending the extraction time resulted in a linear increase in the AA. Increasing the solvent concentration increased the AA content in the extract. Further investigation with lower power levels and use of modified MAE techniques such as lowtemperature vacuum MAE and nitrogen-protected MAE are needed to maximize the yield of AA from potato peel samples [30]. Similar response plots (not shown) were obtained for the individual phenolic compounds measured using HPLC. Optimized experimental conditions derived from the plots, and the resultant predicted individual phenolic compound levels are given in Table 3. For ChloA and CafA the solvent concentration played a significant role, with their concentration rising with an increase in solvent concentration. The linear relationship of factors such as extraction time and power on the content of an individual phenolic compound might be attributable to the selection of a limited range for these treatment parameters in this study. Total phenolic yield increased with increase in power level, but the composition of the individual phenolic compounds varied significantly. The values obtained for individual phenolics like ChloA and CafA were in accordance with the literature and were significantly higher than those obtained by conventional extraction methods (e.g., [4,11,[31][32][33]).
The DPPH method was used to measure the free radical scavenging activity of the potato peel extract. The assay is a simple, rapid and convenient method, independent of sample polarity. These advantages make DPPH an ideal method that can be used to model the reaction between antioxidants and lipids in food system [34,35]. The optimum extraction condition that provided maximum free radical inhibition (74%) were predicted to be 100% (v/v) methanol concentration, power level of 10% and extraction time of 5 minutes. Experimental results suggested that the inhibition increased with increase in concentration up to 65% (v/v) and then remained constant up to 100% (v/v).
To the best of our knowledge, there have been no previous reports addressing the optimization of the MAE process for potato peel. We were able to optimize the phenolic content in the MAE extract while considering the efficiency, economy and feasibility of the process. Predictions achieved through the regressions developed indicated that different combinations of independent variables are required to maximize each dependent variable in the extract. For individual phenolics, the solvent concentration plays a major role because different aqueous methanol mixes have different dielectric properties [36][37][38]. This change in the dielectric property with varying methanol fractions has a profound effect on the extraction of phenolics from potato peel.

Materials and reagents
Potatoes (cv. 'Russet Burbank,') were obtained from the Elite Potato Centre at Bon Accord (NB, Canada). Tubers were washed and peeled with a mechanical peeler to obtain uniformity in thickness of the peel. The peel samples were lyophilized in a laboratory freeze-dryer (Thermo Savant Modulyod-115, NY, USA) until a constant weight was obtained. After drying, the samples were ground to pass a standard 150 µm sieve, thus ensuring uniformity and symmetry of particle size. The freeze-dried powder was kept in closed opaque containers at −20 °C until analysis. All reagents and solvents used were of HPLC grade (Fisher Scientific, Ottawa, ON, Canada)

Equipment and apparatus
MAE extraction was carried out with a focused open-vessel microwave system (Star System 2, CEM Matthews, USA) operating at 800 W maximum power and a frequency of 2.45 GHz. Power level used for the experiments were expressed as a percent of the power supplied within the microwave cavity. The maximum microwave output power in the cavity was calibrated using the protocol developed by Cheng et al., [39]. The regression equation (Equation 7) obtained from the calibration was used to convert % power level into Watts: The mode of microwave power applied was intermittent with power on for 30 s min −1 .

Preparation of potato peel extracts
The phenolic compounds were extracted from freeze dried potato peels using 30, 65 or 100% (v/v) methanol (MeOH). Solvent (40 mL) was added to dried potato peel (2 g) in a Pyrex vessel and placed inside the microwave, where phenolics were extracted at different power levels (% W) over different periods of time (min). The extract was allowed to cool at room temperature and was then centrifuged at 10,000 rpm for 15 min. The supernatant was collected and used for total phenolics determination, free radical scavenging activity assay and HPLC analysis of the different phenolic compounds.

Determination of total phenolic compounds
Total soluble phenolics were determined using Folin-Ciocalteu reagent [40,41]. Potato peel extract solution (1 mL) was mixed with double distilled water (7.5 mL) and Folin-Ciocalteu reagent (0.5 mL) followed by 5% Na 2 CO 3 solution (1 mL). The mixture was incubated at room temperature for 90 min and its absorbance was measured at 765 nm using a spectrophotometer (Ultrospec 2100pro, Biochrom Ltd., Cambridge, UK). A standard curve was plotted using different concentrations of gallic acid and    The central composite rotatable design uses least-squares regression to fit the experimental data to a quadratic model [24,43]. The quadratic model describes the relationship between response (Y) and the process parameters (X i ,X j ,X k ….)is as follows: β β β β (9) where β 0 is the constant coefficient , β i is the linear coefficient, β ii is the quadratic coefficient for main process parameters and β ij is the second order interaction coefficient of variables i and j, respectively. The 3D response surface graphs for the predicted value were plotted using the software's tools.

Conclusions
Utilization of potato peel for extraction of beneficial phytonutrients such as phenolic antioxidants not only provides health benefits, but also adds value to the waste generated by the potato processing industries. In our study RSM proved to be effective in estimating the effect of three independent variables on the extraction of ascorbic acid and selected phenolics. Methanol concentration and extraction time played significant roles in extraction of individual phenolics. We were able to extract higher levels of phenolics from dried potato peel than the values reported number of previous studies, while using less solvent and considerably reducing the extraction time. Future work could focus on optimization and large scale extraction of selected phytonutrients from waste potato peels and their use as food additives.