The orange peels were collected from juice producer located in Jimma. It was manually cleaned to remove impurity and dried for three days followed by an electric oven dry at 50°C until similar weight of the peel achieved. The dried peels were washed, mixed in a mixer grinder and sieved to avoid non-finely particles. The result powder was stored in a dark and dry out place for subsequent requirement.

Chemicals and solvents: Totally analytical grade reagent chemicals and solvents [Citric acid, HCl, HS₂O₄, trifluoro acetic acid, NaOH, ethanol, 2, 2-Diphenyl-1-Picrylhydrazyl] were purchased from chemical purchasers.

Experiment design was developed using Stat Ease Design Expert software version 11.1.0. Based on the design of experiment; central composite deign, the second order polynomial regression model was developed to predict the performance of variables. The extraction parameters selected according to previous literatures as shown in appendix; A table 1 with coded and actual level.

The second-order models generated by RSM are often used to determine the critical points and can be written in a general form as (Kleijnen, 2008): (1)Y=β0+∑i=1kβiXi+∑i=1kβiiXiXi+∑i=1k−1∑j=1+1kβijXiXj+ε

Where, Y, x_i and x_j, β₀, β_i, β_i, β_ii, k and are anticipated response, input factors, constant term, linear coefficient, quadratic coefficient, interaction term, number of factors and random error respectively.

Using Eq.(1), if n is the number of repetitions at the cube’s center and R is the total number of run required for four independent variables (irradiation time, pH, Ultrasound power and liquid-solid ratio) is given by Eq.(2): (2)R=2k+(k∗2)+n

MATLAB® 2014a was used for the formulation of Artificial neural network model using feed forward multilayer network contains three primary layers knwon as input (used hyperbolic tangent sigmoid transfer function), hidden and output layers (used pure-linear transfer function) to anticipate the yield (Nazerian, Kamyabb, Shamsianb, Dahmardehb, & Kooshaa, 2018) as indicated in Eq.(4). The developed artificial neural network model constitutes four inputs (irradiation time, pH solution, ultrasound power and liquidsolid ratio) and one output (pectin). The Marquardt–Levenberg back-propagation (MLBP) algorithm was selected for training. The input and output data detected from the actual values are categorized into different parts; (60% - eighteen samples) for training (20% - six samples) for testing and (20% - six samples) for validation. To find out best training efficiency and reduce the effect of larger values in input and output data, input and output were normalized between−1 and 1 (Muthusamy et al. 2019) “ISSN”: “18790003”, “PMID”: “30414904”, “abstract”: In this work, Response Surface Methodology (RSM. The normalized data was forwarded to the artificial neural network model in feed forward multilayer back propagation. The mean square error values between the output neurons and the observational outputs were determined and backward propagated via the network. Then, the individual weights of the neuron, corrected by the algorithm. For cross-validation of parameters, by MATLAB, ANN Tool was developed. After the ANN tool memorizes the data from the training, the cross-validation was applied to prevent over fit of the training. By repetition in testing a number of NN, the best number of neurons in the hidden layer was determined when the mean square error (MSE) value of the output reached its minimum value.For determination of the number of neurons required in the hidden layer, addition of the weighted inputs and the related bias are important data towards for nonlinear situation (Cheok et al. 2012) solid to solvent ratio, and methanol concentration for extracting total phenolic content (TPC; Madadlou et al. 2009) one hidden layer and an output layer with one neuron optimized using response surface methodology (RSM (4)F(x)=tansig(x)=1−ex1+ex

The coefficient of determination; R², Root Mean Squared Error; RMSE, mean average error; MAE, standard error of prediction; SEP, and absolute average deviation; AAD were determined to check the accuracy and predictive ability of ANN and RSM using Eq. (4): (4)R2=1−∑i=1r[(yp−ye)2(yp−m)2] (5)MSE=1−1r∑i=1r(yp−ye)2 (6)RMSE=MSE1/2 (7)SRP=RMSEm (8)AAD=100r∑i=1r(ye−ypyexp) (9)MAE=∑i=1r[ye−ypye]

Where; r = number of run, y_p = predicted values from model, y_e = experimental values and m = average experimental values

By using previous study as reference with modification and it should also be declared that the degrees of parameters were selected depend on previous study (Hosseini et al. 2019; Lin et al. 2018) there is a lack of studies on the extraction of pectin by ultrasound assisted extraction (UAE. In all runs, fifteen gram (15g) of the peel powder was mixed with pH of (13) citric acid solution in 250ml flask. It was then sonicated for (1530 min), in the range of (60120W), ultrasound power, thereafter the mixture was maintained at temperature of 25°C for a certain time and percolated. The percolate carrying pectin was chilled down, for 25 min and centrifuged at 6500 rpm. Ethanol of 96% was utilized, to precipitate the supernatant and impurities, and allowed for one hour for pectin flotation. The floating result was then classified by filtration process and the end result was dried in drier at 50°C to remove moisture. The resulting dried pectin was milled to powdered pectin and stored for analysis and its yield was determined using to Eq. (2) and indicated in table 2. (10) Yield (%)= Amount of dried pectin extracted in gram (g) Amount of sample taken upfor extraction in gram (g)∗×100

The produced pectin contents were analyzed by using prinks Elmer spectrum 65 FT-IR technique with the help of IR correlation. The wavenumber region for the analysis was 4000-400 cm⁻¹ (in the mid-infrared range) and IR spectrum was reported by % transmittance.

The pectin extracted at the optimum processing conditions (the highest yield) was analyzed by determining the following properties.

Esterification Capacity/degree of esterification (EC): It was calculated using the titrimetric method using potassium hydroxide (KOH), instead of sodium hydroxide. Twenty-five (25g) of dried product was moistened with standard ethanol and dissolved in 25 ml of deionized water at forty degree Celsius (40°C). Following full dissolution of product, five pearls of indicators were added to the mixture. The mixture was thereafter volumetrically analyzed with 0.4 M KOH, and the amount of the KOH solution used for color change was recorded as volume (V₁). Subsequently, 15 ml of 0.4 M KOH was added, and the solution was shaken powerfully and allowed to cool down for 10 min. In addition, 15 ml of 0.6 M hydrochloric acid was added and the solution was shaken until the disappearance of pink color. The mixture was volumetrically analyzed with 0.4 M KOH for the end step and the consumed amount was collected as volume (V₂) and the esterification capacity was determined using Eq. (11): (11)EC(%)=V2V1+V2∗100

EC = esterification capacity of dried product, V₁ = initial titer (ml), and V₂ = final titer (ml)

Equivalent weight (EW): This was calculated as method reported by (Mohamed, 2016); 5 g of dried product was put in a 250 ml conical flask, and 15ml ethanol was added. One gram of NaCl and 100 ml of deionized water were added. Finally, five drops of phenol red indicator was added and volumetrically analyzed with 0.1 M KOH. The end of volumetric analysis was represented by pink color and the equivalent weight was calculated according to Eq. (12): (12)EW(%)=ms∗1000Va∗Kn

EW= Molecular mass of dried product, VA = volume of alkali (ml), and K_en= normality of potassium.

Methoxyl content (MC): MC was carried out using the Ranganna’s method as reported by (Fakayode & Abobi, 2018). When the neutral solution was accumulated from determination of equivalent weight, 20 ml of KOH was added. The mixed solution was stirred thoroughly and kept at 25°C for 25 min. Then after, 25 min, 20 ml of 0.3 M HCl introduced and volumetric analysis started with 0.1M KOH. The content of methoxyl was calculated using Eq. (13): (13)MC(%)=Va∗Na∗3.1ms

MC = methoxyl content of dried product

Total anhydrouronic acid content (TAAC): This was carried out using the method adopted by Mohamed, (2016) and determined using Eq. (14): (14)TAAC(%)=1.76∗(a+b)ms

TAAC = total anhydrouronic acid content of dried product a = volume of KOH from equivalent weight determination (ml), and b = volume of KOH from methoxyl content finding (ml) and m_s = mass of sample (g)

Acetyl value (AV);- This was carried out using the Ranganna’s method as reported by (Kliemann et al. 2009); 5 g of dried product was dissolved in 0.2 M KOH solution by stirring and allowed to stand overnight. The contents were diluted to 40 ml with deionized water and an aliquot (15 ml) was located into the distillation equipment. Magnesium sulfate-sulfuric acid mixture (15 ml) was also shifted to distillation equipment and distilled. About 100 ml of distillate was collected and the distillate was titrated with 0.4 M KOH using phenol red indicator. A blank distillation using 15 ml of the Magnesium sulfate- sulfuric acid solution was carried out, and the distillate was volumetrically analyzed. The acetyl content was calculated using Eq. (15): (15)ACV(%)=Va∗Na∗4.3ms

ACV = acetyl value of dried product

A very good correspondence between the observational and anticipated values for pectin yield was obtained from the check bit plot between the anticipated and the observational values as shown in Fig. 2. The anticipated values and observed values are in reasonable agreement, since there is closeness between the both results for pectin yield. Referencing table 2, the expected optimum pectin yield was 26.74% at time of 22.5 min, pH of 1.5, ultrasound power of 155W, and liquid-solid ratio of 22.5:1 mL/g using Eq. (10). Under these conditions, the actual value of pectin was 26.87%, which in corresponding with those expected by computation.

The UAE yield of pectin was less than the result reported by, Hosseini et al. (2019) from microwaveassisted extraction (28.07 ± 0.67 %) and higher than, Hosseini et al. (2016) aqueous extraction (17.95 ± 0.3%) in their optimum extraction conditions. The variation could be due to; plant source, the extraction technique taken for separation (Hosseini et al. 2016b). However, the result indicates that the UAE method is an efficient technique since; this it has minimum consumption of time and energy and in addition, that the extracted yield of pectin in this research was better than result achieved by Wikiera et al. (2016) 19.8% of pectin yield from apple pomace by Endo-xylenes and endo-cellulose-assisted method. Shivamathi et al. (2019) has reported the achievement of 8.93% yield of pectin from custard apple peel at optimized conditions. This implies that the model chosen for anticipation of the pectin yield from orange peel was adequately.

The regression coefficient (interms of coded), variance analysis of the expected model and all colleration coefficients( R², R² adj and R² pre) have been used to test the goodness of the model (Zhang, Chen, Mao, Guo, & Dai, 2014) “ISSN”: “18790003”, “abstract”: The central composite design (CCD listed in table 3. The analysis results proved that the relationship between factors and the response of the expected model was well-correlated. Residual is the difference between the actual and the fitted value of a model and it is used to predict the adequacy of the model. Small residual mean’s; model anticipated is accurate (Samavati, 2013) based on a five level, four variable central composite rotatable design (CCRD. The statistical effluences of all the terms of the model were tested by the F and P-value. The corresponding variables would be more significant if the F-value became higher and the P-value became smaller (Zhang et al. 2014) “ISSN”: “18790003”, “abstract”: The central composite design (CCD. In this study, results showed that, high F-value and p value of developed model were below 0.05, which in turn show the fitness of model was strongly significant (Prakash et al. 2013). The results also proved that, the proposed regression model for pectin yield was satisfactory high with all squared R values table 3, which exhibited a closeness between the actual and the anticipated values by the model.

The Pred.R² of 0.9575 is in reasonable estimate with the Adj. R² of 0.9475; i.e. the distinction is much less than 0.2. The adequate precision ratio of 58.439 shows an adequate signal and this model can be used to navigate the design space since, adequate precision is used to find out the signal to noise ratio and a ratio greater than 4 is desirable. Meanwhile, the very small value of coefficient of variation (CV:7.23%) clearly shown a very strong degree of precision and a great deal of reliability of the experimental values (Prakash et al. (2014). The F-value of 2.95 the lack of fit for this estimation shows that, the model is suitable for anticipation of optimum conditions to extract pectin from orange peel using UAE. The proposed quadratic polynomial model is therefore, perfectly suitable for anticipation of the relationship between independent and dependent variables. The dependent and independent variables are related by the following second-order polynomial equation interms of coded terms by applying multiple regression analysis on the experimental data.

The importance of the independent variables and their effects could be indicated by the magnitude and sign of the coefficients (Zhao-Hui et al. 2011). It could be seen from table 3 and Eq. (16), all the liner, the quadratic and the interaction coefficients (BC, BD, CD) were statistically significant.

The result in table 3 and Eq. (16) separately showed that all of the applied factors were highly effective on extraction yield of pectin except liquid-solid ratio not significant. Citric acid pH solution should be considered more than the other factors, since it was the most influential on extraction process. Similarly the interaction between pH, solution versus ultrasound power and pH solution versus liquid-solid ratio had a high influence on the yield of extraction. Apart from the liquid-solid ratio and pH solution, the interaction between pH and ultrasound power had highly influenced the yield of pectin. The 3D plots were demonstrated below for more exploration in the relationship between response and parameters. At the least pH solution and highest ultrasound power, the maximum extraction yield was attained. This means, high acidic media, brought the pectin leakage from the plant substance and in consequence, increased the extraction efficiency (Hosseini et al. 2019b and Dranca, 2019). Another efficient variable that had a direct influence on the extraction was ultrasound power. Fig. 3a reveals that ultrasound power was better efficient in the study, enhancing the ultrasound power; the extraction efficiency was also enhanced. This result may relate to the cavitation consequences of ultrasound waves and this cavitation bring up the solvent insight into intercellular substance of plant and thereby upgrades the pectin escapees leads to increase the extraction efficiency (Wikiera et al. 2016).

Irradiation time is one of the important process parameters that affect the yield of pectin significantly compared to L-S-R in this study. Due to creation of the cavitation bubble by ultrasound waves that supports the disruption of plant cell wall to improve the extraction efficiency of pectin and produce swelling and hydrate the plant material, in the initial stage, the extraction efficiency was increased up to 22.5 min and the reduction was observed above this value (Shivamathi et al. 2019).

Yield of pectin did not highly influenced with liquid-solid ratio compared to other parameters; an observation that was confirmed by the results of the analysis of variance (p > 0.05) and Eq. (16). It can be assumed that varying the range of study above a SLR of 1:22.5 g/mL might be lead to a different conclusion regarding the involvement of this working parameter in the cavitation effect occurring in an extraction assisted by ultrasound method and consequently its impact on the efficiency of extraction (Dranca, 2019).

The ANN anticipation has been carried out successfully using information shown in indicated in table 1. The ANN-based model was done by; choice of the appropriate training algorithm, determination of optimum value of neuron and validation of the model. Eighteen (80% –18) samples of the data were used for training, six (20%–6) samples for testing and six (20%–6) samples for validation. Using the available actual value, Levenberg–Marquardt (LM) ANN fitting tool and TANSIG Transfer Function 4– X_n–1 (number of input layer, neurons in hidden layer and output layer) model were implemented. Prior training, the network inputs and target were normalized by assigning the maximum (+1) and minimum (–1) levels. Statistical parameters like R; regression coefficient, RMSE; root-mean square error, MAE; mean average error, SEP; standard error of prediction and AAD; absolute average deviation were assumed to determining the higher predictive power of the model techniques. The statistical values of ANN and RSM models are decided using Eq. (4–9) and shown in table 4.

The degree of association or relationship among the parameters in the question was specified by correlation coefficient (R). A unit (1) value of coefficient of correlation implies perfect relationship between variables while zero (0) value is believed to be the absence of linear relationship the parameters. The best solution was chosen based on highest coefficient of correlation and least MSE for training, testing and validation. The correlation coefficients (R) between the actual and the expected values are 0.99943 for training, 0.9863 for testing, 0.99615 for validation and overall correlation was 0.99769. Therefore, the ANN anticipation for training, validation, and testing is highly substantial and meritorious in terms of correlation. As indicated in indicated in Fig. 4 and, the performance value for the final iteration gave a best result.

The analyzed characteristics of the extracted pectins were indicated with their values and summarized in table 5.

To represent the extent to which carboxyl groups in pectin molecules exist as the methyl ester, esterification capacity (EC) is an essential molecular index for pectin classification. Since result indicates, higher degree of esterification above50 percentage, it can be considered as high methoxyl (HM) pectin. Pectin’s has high tendency to form gel rapidly at higher temperature and have a great effective action on the lipid profile, when degree of esterification is above 50 percent (Brouns et al. 2012) molecular weight (MW; Dominiak et al. 2014). However, esterification capacity shows only the ratio between methanol-esterified carboxyl groups and free carboxyl groups, whereas the methoxyl rate represents to the amount of methoxyl groups in a sample (Fakayode & Abobi, 2018). So, the EC could not be judged separately, as it does not show the actual amount of methyl esterification, especially when the galacturonic acid amount is low. High methoxyl pectins (above 50% EC) require a relatively a low pH for gel formation and large concentration of soluble solids, while pectins having methoxyl below 50% EC, establish in flexible gels the using the movement of multivalent cations, which cross-hyperlink the galacturonic acid chains (Garna, Mabon, Wathelet, & Paquot, 2004). In present study, the esterification capacity (EC) of the hydrated pectin was 63.13% similar to previous studies as shown in table 4.

Another important characteristic in classifying the functional behavior of pectin is equivalent weight (EW) since, gelling tendency of individual pectin’s are linked very closely with EW. The higher equivalent weight, has greater gel formation, while the lower equivalent weight, indicates larger partial degradation of the pectin which is non-profitable (Hardy, 1924) strawberries, rhubarb stem and apples, Schryver and Haynies [1916] employed a hot dilute solution of ammonium oxalate to extract pectinogen from the washed and dried material. Subsequently, Farnell [1923], working in conjunction with Schryver, showed that oxalic acid could be substituted for the salt, and indeed, is preferable in so far that it remains in solution when the extracted pectinogen is precipitated by alcohol, whereas ammonium oxalate is strongly adsorbed by the alcohol-gel. Farnell also demonstrated that solutions of ammonium sulphate and of carbonic acid are capable of extracting appreciable amounts of pectinogen from dried turnip, but that water alone is without effect. He suggested that pectinogen is loosely combined with calcium in plant tissues, and is liberated therefrom by any reagent which precipitates the metal. Carre [1922] has, however, shown that dilute hydrochloric acid is an effective extractor of pectinogen, a result which is not concordant with this generalisation. In examining the factors which influence the extraction of pectinogen from the fibre of sugar cane, Farnell (private communication and thus the raise or diminish of the equivalent weight could be subjected to the content of free acid (Mohamed, 2016). The average equivalent weight of orange peel pectin in this study found to be 604.74 kDa closest to data reported by Altaf et al. (2015) at different time, temperature and pH combinations. Extraction was done at temperatures 60°C, 70°C and 80°C for time 30 mins, 45 mins and 60 mins, and pH 2, 2.5 and 3.0 respectively. Using hydrochloric acid, the yield of pectin was higher (ranging from 2.8%-16%, that the equivalent weigh of the pectin extracted form papaya fruits using HCl and Citric acids were 912.17 and 455.1 respectively and stated that, pectin produced at lower pH value, has higher equivalent weight, since lower pH can lead polymerization of pectin into an extended chain, and in turn decrease the amount of free acid..

The methoxyl content (MC) is an instrument tool used to control the setting time, the gel strength and to find the functional properties of pectin (Twinomuhwezi, Godswill, & Kahunde, 2020). Kanmani, (2014) established that, depending on the origin of raw material used, method of extraction, in addition to the method used for determination of methoxyl content, the MC of pectin usually varies from 0.2–12%. The result indicated that 6.23% of methoxyl content was achieved. Since the methoxyl content was below 12%, this pectin has lower ester characteristic, which implies that it is desirable in terms of quality, and in addition due to it has above 50% DE.

As indication of Food Chemical Codex (FCC), Food and Agriculture Organization (FAO), and European Union (EU), pectin must contains at least 65% of galacturonic acid (Willats, Knox, & Mikkelsen, 2006), since anhydrouronic acid amount is used to represent the gelling capabilities of given pectin. The higher value means, the produced pectin has a lower amount of protein. In present study, the result indicated that 68.25% of anhydrouronic acid was achieved. A minimum value of anhydrouronic acid (65%) for commercial pectins has been specified by FAO (Twinomuhwezi et al. 2020). This criterion has limited the potential sources of food and pharmaceutical pectins.

Acetyl value (AV); the gelling capacity of pectin diminished with the raise in the degree of acetylation (Fakayode & Abobi, 2018). The inhibition of gelormation will be created when acetyl formation is found in pectin. Other researcher shown that the gelling capacity of pectin diminished with raise in the degree of acetylation and samples holding 3.5%-4.0% acetyl brings weak gels while gelling capacity restored at levels around 2.4% acetyl (Mohamed, 2016). Based on acetyl value and compared to previous study, the result indicates that orange peel pectin has good gelling capacity (0.371%) indicated in table 5.