Acid and Deep Eutectic Solvent (DES) extraction of pectin from pomelo (Citrus grandis (L.) Osbeck) peels

Author’s Accepted Manuscript Acid and Deep Eutectic Solvent (DES) extraction of pectin from pomelo (Citrus grandis (L.) Osbeck) peels Shan Qin Liew, Gek Cheng Ngoh, Rozita Yusoff, Wen Hui Teoh www.elsevier.com/locate/bab

PII: DOI: Reference:

S1878-8181(17)30425-5 https://doi.org/10.1016/j.bcab.2017.11.001 BCAB650

To appear in: Biocatalysis and Agricultural Biotechnology Received date: 11 August 2017 Revised date: 22 October 2017 Accepted date: 1 November 2017 Cite this article as: Shan Qin Liew, Gek Cheng Ngoh, Rozita Yusoff and Wen Hui Teoh, Acid and Deep Eutectic Solvent (DES) extraction of pectin from pomelo (Citrus grandis (L.) Osbeck) peels, Biocatalysis and Agricultural Biotechnology, https://doi.org/10.1016/j.bcab.2017.11.001 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Acid and Deep Eutectic Solvent (DES) extraction of pectin from pomelo (Citrus grandis (L.) Osbeck) peels Shan Qin Liew1, Gek Cheng Ngoh1*, Rozita Yusoff1, Wen Hui Teoh1 1

Department of Chemical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia. Email: [email protected] (Shan Qin Liew) *[email protected] (Gek Cheng Ngoh) [email protected] (Rozita Yusoff) [email protected] (Wen Hui Teoh) Corresponding author should be addressed:

1

Gek Cheng Ngoh1* Department of Chemical Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia. Tel: +60 3 79675301 Fax: +60 3 79675371 Email: [email protected] (Gek Cheng Ngoh)

Abstract Pectin was extracted from pomelo peel using citric acid. A Box–Behnken design was employed to optimize the yield and the Degree of Esterification (DE) of pectin. The effects of pH, temperature, extraction time and liquidsolid ratio on the yield were investigated. It was found that under optimized conditions at pH 1.80, extraction time of 141 min, temperature of 88 °C and a liquid-solid ratio of 29:1 mL/g, a pectin yield of 39.72% and a DE value of 57.56% were obtained. The results indicate that the pectin extracted from pomelo peel is a slow set high methoxyl type of pectin. pH was found to have the greatest influence on pectin yield and DE. Varying pH at a narrow range between 1 and 2 interestingly showed the formation of diverse pectin functional groups with different structural modifications. The findings suggest that extraction conditions could influence pectin extraction performance, chemical structure, as well as morphological and gelling properties. This study also explored various acid based Deep Eutectic Solvents (DESs) in pectin extraction. The lactic acid−glucose−water DES with a ratio of 6:1:6 gave the highest pectin yield of 23.04%. Citric acid was found to have a better yield performance and more energy saving as compared to DES in pectin extraction. Keywords Pomelo peel; Pectin; Citric acid; Extraction and optimization; Physicochemical properties; Deep Eutectic Solvent

1. Introduction Pectin is a complex heteropolysaccharide consisting mainly of α-(1,4)-linked D-galacturonic acid as the backbone with different Degrees of Esterification (DE) (Mohnen, 2008; Yuliarti et al., 2015). Depending on the DE, pectin can be classified as high methoxyl pectin (HM) with DE ≥ 50 or low methoxyl pectin (LM) with DE < 50. HM pectins can be further divided into rapid set (RS) and slow set (SS) pectin, with respective commercial applications. The rate of setting is specific in such a way that enables RS pectins to be used for different jams and jellies involving processes that require suspension of matter. It is reported that some waste materials such as peels of citrus (e.g. orange, lime and lemon) as well as apple pomace and sugar-beet pulp are good sources of pectin (Arslan and Kar, 1998). Pomelo (Citrus grandis (L.) Osbeck) being the largest citrus fruit, with its peel approximately 40% of its fruit weight, can be considered for pectin extraction from an industrial view point as it may mitigate disposal problem. Pectin is commonly extracted using strong mineral acids as the acidic extraction agents allow for reasonable extraction yield and are time efficient (Lim et al., 2012). However, pectin extracted through this technique is susceptible to degradation. In addition, high acidity accelerates corrosion of equipment and triggers

1

water pollution problems. Moreover, pectin is commonly used in the food sector, where the use of strong mineral acids may leave consumers with a negative impression. Several research works have been conducted to address this issue in which mineral acids were replaced by organic acids for pectin extraction (Koubala et al., 2008; Oliveira et al., 2015). Literature also reported that biodegradable Deep Eutectic Solvents (DESs) are suitable for extraction processes due to their negligible volatility at room temperature, water miscibility and other excellent properties (García et al., 2016). The reported suitability encouraged the application of DESs in the present study. The aims of the present study are to determine the optimum conditions for the extraction of pectin from pomelo peel using citric acid, and to investigate the effects of pH, extraction temperature, time and liquid-solid ratio on the extraction yield and the Degree of Esterification. The physicochemical composition of the extracted pectin was also investigated. The effect of pH on pectin yield was determined through examination of the chemical structure, morphological and gelling properties of the extracted pectin. The scope of the present study was further extended by applying different lactic acid based DESs in pectin extraction and then comparing the performances of all the obtained extractions. 2. Materials and methods 2.1. Materials Pomelo fruit at the same stage of ripeness was obtained from a local pomelo nature park (Malaysia). The peels of the fruit were cut and dried in a hot air oven (Memmert 600, Schwabach, Germany) at 60 °C until a constant weight was attained. The peel was then grinded and sieved into 250–355 µm. The pomelo peel powder was stored in an air tight container and kept in desiccators prior to use. Lactic acid, glucose, glycine, sodium nitrate, sodium azide, sucrose, citric acid, ethanol (EtOH, 95%), and ethanol (EtOH, 70%) used in the present study are of analytical grades, and purchased from R&M, Malaysia. D-galacturonic acid (D-GalA, > 98%) and dextran (~95%) were purchased from Sigma-Aldrich, Germany. 2.2. DESs preparation The DESs listed in Table 1 were prepared based on reported methodologies (Bakirtzi et al., 2016; Dai et al., 2013; Francisco et al., 2012; Paradiso et al., 2016; Rozema et al., 2015; Van den Bruinhorst et al., 2016). The multi-component mixture was placed in a bottle with a magnetic stirrer and capped to prevent any loss in mass during heating on a stirrer-hot plate. When a clear liquid was formed, the heating was stopped and the DES was allowed to cool down to room temperature prior to being used for extraction. 2.3. Pectin extraction (i) Citric acid extraction Ten (10) g of dried pomelo powder was weighed and mixed with distilled water based on the desired liquid-solid ratio. pH of the mixture solution was adjusted by using citric acid. The pomelo mixture was extracted at different extraction temperatures (65–90 °C) at varying pH values (1.50–2.50) and varying liquidsolid ratios (20:1–30:1 mL/g) and extraction times (40–180 min). The extraction was performed in triplicates under different conditions as listed in Table 2. During extraction, the desired temperature was set and controlled using a water bath (Julabo TW20, Seelbach, Germany). (ii) DES extraction Five (5) g of dried pomelo powder was weighed and mixed with 145 mL of DES. The extraction was repeated 3 times in a water bath for 141 min at 88 ºC according to the optimized acid extraction conditions obtained from Section 3.2. Upon completion of the extraction processes using both citric acid and DES, the mixture solution was centrifuged (Sigma 3-15P, Osterode am Harz, Germany) at 4000 rpm for 10 min. The supernatant was decanted and then filtered using a filter cloth. The supernatant was precipitated with 250 mL and 125 mL of 95% (v/v) ethanol for citric acid and DES extraction respectively. The samples were then stored in the dark at room temperature for 24 hours to allow pectin flotation to take place. The floated sample was subsequently separated by filtration and washed twice using 70% (v/v) ethanol to remove impurities. The wet pectin was dried in hot air oven (Memmert 600, Schwabach, Germany) at 65 °C until a constant weight was reached. The pectin extraction was determined based on the method described by Woo et al. (2010). The percentage of pectin yield was calculated using Eq. (1):

2

(%) =

₀ 

× 100

(1)

where; m₀ is the weight of dried pectin (g) and m is the weight of dried pomelo powder (g). 2.4. Experimental design In this study, four factors, at three levels (-1, 0, 1) Box-Behnken response surface design (Table 2) were employed to investigate and optimize the effect of process variables on pectin yield and the extracted DE from pomelo peel using acid extraction. The variables were: pH (X1), time (X2), temperature (X3) and liquidsolid ratio (X4). A total of 29 experiments including 5 center points were designed (N = 2K(K−1) + C0, where N is the total number of experiments, K is the number of independent variables and C0 is the number of centre points). The statistical package Design Expert 6.0.6 (State-Ease Inc., Minneapolis, USA) was used to construct the experimental design and analyze the experimental data. Experimental data were fitted to a second-order polynomial equation to establish the relationship between the independent variables and the responses. The generalized form of the equation is:  = "₀ + ∑,$./ "$ &$ + ∑,$./ "$$ &$' + ∑* ∑,-$.' "*$ &* &$

(2)

where Y represents the response variable, Xi and Xj are the independent variables affecting the response, and β0, βi, βii, and βij are the regression coefficients for intercept, linear, quadratic and interaction terms. Analysis of variance (ANOVA) was used to analyze the effects of process variables statistically. The model equation was validated for its adequacy in predicting the optimum response values by comparing with the experimental results. 2.5. Physicochemical and structural characteristics of pectin 2.5.1. Proximate analyses The moisture, ash, protein and fat of the pectins were determined by approved AOAC methods (1995). A half-gram of extracted pectin was mixed with 50 mL distilled water and the pH was measured by a pH meter (827 pH lab, Metrohm, Switzerland). 2.5.2. Degree of Esterification The Degree of Esterification (DE) of pectin was analyzed using FT-IR (Bruker Tensor 27, Massachusetts, USA) spectroscopy. FT-IR spectra of the samples were recorded from 600 to 4000 cm-1 with 32 scans. The measuring resolution was 2 cm-1 and the resultant spectra were smoothened to remove noise. DE is a ratio of esterified carboxyl group to the number of total carboxyl groups present which can be calculated using the absorbance intensities at 1630 cm-1 and 1745 cm-1; corresponding to the non-methyl-esterified carboxyl groups and the methyl-esterified carboxyl groups, respectively. The sum of the bands absorbance intensities at 1630 cm-1 and 1745 cm-1 corresponding to the total carboxyl groups was measured by an OPUS software (Opus, Chicago, USA). The percentage of DE was determined according to Eq. (3) (Manrique and Lajolo, 2004). 23(%) =

4₁₇₄₅ 4₁₇₄₅94₁₆₃₀

× 100

(3)

2.5.3. Galacturonic acid The galacturonic acid (GalA) content was determined using a colorimetric method (Filisetti-Cozzi and Carpita, 1991; Kliemann et al., 2009) using D-galacturonic acid (> 98%, Sigma-Aldrich, Germany) as standard. The absorbance reading on the test samples was performed in a spectrophotometer (PRIM SECOMAM, Alès, France) at a wavelength of 520 nm. All experiments were performed in triplicates and the absorbance values were compared with the galacturonic acid calibration curve (R2 = 0.97). 2.5.4. Molecular weight of pectin The molecular weight of the extracted pectin was determined using High Performance Size Exclusion Chromatography (HPSEC) coupled with a Dawn Heleos Multiangle Laser Light-Scattering (MALLS) detector (Wyatt Technology, USA) and an OptilabreX Differential Refractometer (RI) at 633 nm wavelength at 25 °C. The extract was dissolved in distilled water (1.5 mg/mL) and was passed through a 0.45 mm membrane filter (Milipore Co., Milford, USA). This was followed by manually injecting the extract through a 100 mL loop using a PL aquagel-OH MIXED-H 8 mm column (Agilent Technologies, USA). The mobile phase was 0.1 M sodium nitrite (NaNO2) solution containing 0.5 g/L sodium azide (NaN 3) as a bactericide which was carried out at 25 °C with a flow rate of 0.6 mL/min (Muhammad et al., 2014). Monodisperse dextrans (~95%, Sigma-

3

Aldrich, Germany) were used as standard and the data were collected by the Astra software (Wyatt Technology, USA). 2.5.5. Solubility The solubility of the extracted pectin was measured using the method described by Jiaxing et al. (2015). A half-gram of the sample was mixed with 50 mL distilled water. The mixed solution was then stirred evenly and incubated at 40 °C for 30 min. The solution was centrifuged for 20 min at 25 °C and at 4200 rpm. The supernatant was then transferred to a beaker and allowed to evaporate in a water bath at 90 °C followed by oven drying at 105 °C until a constant weight was achieved. The solubility was determined as follows; <>?@AB (%) =

C D

× 100

(4)

where; m1 is the constant weight of dried supernatant (g) and m2 is the weight of sample (g). 2.5.6. Color The color of the pectin sample was measured by a colorimeter (WF30, iWAVE, China). The CIELab coordinates (L*, a*, b*) of the pectin samples were directly obtained from the colorimeter. The L* value represents lightness, ranging from 0 (black) to +100 (white); a* value ranges from −100 (green) to +100 (red) and b* value ranges from −100 (blue) to +100 (yellow). The hue angle (H*ab) and chroma (C*) were calculated by equations (5) and (6): G∗

∗ EFG = arctan ( ∗)

(5)

I ∗ = (J ∗' + @ ∗' )/⁄'

(6)

F

2.5.7. Surface morphology analysis Pectin extracted from the pomelo peels was mounted onto a specimen stub with a double-sided tape. The morphological structure of the pectin samples were observed using a Scanning Electron Microscope (SEM) (Quanta 200 FESEM, FEI, USA) with an accelerating voltage of 10 kV at magnification ×500. 2.5.8. Rheological measurement The flow behaviors of pectin solutions were investigated using a controlled-stress rheometer (PaarPhysica MC 301, Anton Paar, Graz, Austria) with a 50 mm parallel plate. The pectin gels for rheological tests were prepared using the procedure described by Jiang et al. (2012) with a slight modification. To form pectin gels, sucrose at 60% (w/v) was added to 3% (w/v) pectin solutions, with the pH of the mixture adjusted to 2. They were then heated and stirred at 80 °C for 30 min. The test samples were subject to steady-shearing at 25 °C, with the shear rates ranging from 1 to 500 sˉ¹, and a geometry gap of 0.150 mm. The findings were fitted to Ostward–DeWaele equation, η = Kγn-1, where K is the consistency index and n is the flow behavior index. To study the effect of heating time on the viscosity at 95 °C (Li et al., 2015), 0.5 g of a pectin sample was mixed with 100 mL distilled water in a sealed beaker. Next, the solution was dried in an oven at 95 °C and allowed to stand for 0, 5, 10, 15, 20, 30, 60 min. The relative viscosities of the sample solutions under different drying periods were determined using a viscometer (DV-II, Brookfield, USA). 2.5.9. Energy consumption determination The energy requirement for both acid and DES extraction methods of this study was computed using the specific heat equation. The calculated values were the respective energies used to heat the solvent from 28 °C (room temperature) to 88 °C (extraction temperature). The heat capacities of the mixture solvents were calculated using the Kopp’s rule. 3. Results and discussion 3.1. Effect of extraction parameters Two important responses that can reflect accurately on the performance of the extraction process and the specific application of pectin are yield and DE respectively. The responses can be influenced by certain variables such as pH, temperature, and time, liquid-solid ratio, etc. The yield and DE of the pectin obtained from the present study are presented in Table 2, while Table 3 tabulates the ANOVA findings of the study. It can be seen that the yield of the extracted pectin varies greatly from 4.47 to 39.57% and the values of DE were found to be ranging between 49.71 and 67.50%.

4

[Table 2] According to the p-values shown in Table 3, pH posed the most significant effect on pectin yield, followed by temperature and liquid-solid ratio, while time had the least effect. [Table 3] Reasonable yield performance was achieved when low pH was combined with either long extraction time or high temperature or high liquid-solid ratio. This is due to the fact that low pH causes a concentration difference between the extraction medium and the plant matrix, which enables the extraction solvent that was in contact with the insoluble pectin to easily induce hydrolysis of the insoluble pectin into soluble pectin (Prakash Maran et al., 2013). Moreover, longer extraction time was needed for the diffusion of pectin across the plant structure to the extraction solvent, as higher temperatures provide the thermal energy required to soften the plant structure. As a result, the target compound can diffuse much easier into the extraction medium which in turn would speed up the extraction process. High liquid-solid ratio could also promote the diffusion process of the target compounds (Radojkovic et al., 2012). [Table 4] In order to study the effects of the extraction parameters more thoroughly, the extraction conditions of pectin extracted from different citrus family sources were compared and summarized in Table 4. Temperatures ranging between 65 °C and 90 °C were found to be suitable for pectin extraction for various citrus sources since thermal degradation of the pectin is unlikely to be triggered at these temperatures. From Table 4, it was noted that there was no specific trend of parametric effects in pectin extraction for all the citrus sources. This implies that investigation on each individual parameter for extraction is essential even though pectin is being extracted from the raw materials of the same citrus family sources. The findings showed that the optimized extraction pH for all the citrus sources were between 1.5 and 2.5, except for lemon, which required a higher extraction pH of 3.5. The disparity in the optimized pH between lemon and other citrus sources may be due to the acidic nature of the fruit itself. Lemon is highly acidic which might prevent it from creating a substantial concentration gradient between the extraction medium and the plant matrix. This has the effect of reducing the diffusion ability of the pectin into the extraction medium. The optimum pectin yield obtained from the current work is comparatively higher than the work of Methacanon et al. (2014) with similar DE values. In both cases, pectin was extracted from pomelo although Methacanon et al. (2014) used nitric acid as the extraction acid. The better yield performance in the present work commensurate with the observation that yields obtained from organic acids tend to be higher than those obtained through mineral acids. Extractions of pectin from various citrus sources showed that citric acid gave a better yield performance in the range of 36.71%−67.30% as compared with mineral acids such as nitric acid (19.80%−27.63%) and hydrochloric acid (19.16%−21.10%) (Arslan and Toğrul, 1996; Bagherian et al., 2011; Elizabeth Devi et al., 2014; Kanmani et al., 2014; Koffi et al., 2013; Methacanon et al., 2014; Zanella and Taranto, 2015). The optimum amount of solvent required and the optimum extraction temperature for citric acid-based extraction on pomelo were marginally lower as compared to the extraction using nitric acid. Additionally, the batch extraction time using citric acid was approximately three-quarter of the time required using nitric acid. In a large scale extraction, the lower operating temperature and solid-liquid ratio together with a shorter extraction time may lead to substantial energy and material savings. The optimal pectin yield of the current study (39.72%) is also comparable to that extracted from 'Pera' sweet orange (38.21%) by Zanella and Taranto (2015) and to that extracted from lemon (36.71%) by Kanmani et al. (2014). It is however, comparatively lower than the yield extracted from orange (67.30%) by Elizabeth Devi et al. (2014). In all four studies, citric acid was used for extraction. The pectin extracted from pomelo (present work) and the pectin extracted from ‘Pera’ sweet orange both yielded HM pectin with DE values higher than 50 although the DE value obtained from the present work is lower by 12.65%. In general, the obtained DE values of the extracted pectin showed small deviations. The present study showed that the DE values of all the extracted pectin at varying extraction factors were fairly consistent, with an average value of 58.98%, indicating that pomelo is a potential source for extracting slow set HM pectin. The highest DE was obtained experimentally at pH 2.0, 77.5 °C, with a solid-liquid ratio of 1:25 and an extraction time of 110 min. The response surface plot shown in Fig. 1g-l indicates that pH poses the most significant effect on DE with p < 0.005, followed by liquid-solid ratio (p < 0.05). Extraction time and temperature (p > 0.5) were found to have insignificant effects on DE. The DE values obtained in the present study are similar to those obtained using nitric acid on pomelo (Methacanon et al., 2014). Nonetheless, the DE values obtained in this study are higher than those extracted from orange (Elizabeth Devi et al., 2014) and lemon (Kanmani et al.,

5

2014) but lower than those extracted from grapefruit using hydrochloric acid (Jiang et al., 2012; Liu et al., 2010). As shown in Table 4, DE values vary with citrus sources. Hence, the class of pectin obtained, whether it being LM or HM pectin, may depend on the source of citrus materials. The interactive effects of the parameters on both pectin yield and DE values can also be observed from the surface plots shown in Fig. 1a-f. [Figure 1] 3.2. Process optimization and validation of predictive models The extraction conditions stated above were optimized and a second order polynomial equation was developed by relating the process variables with the responses in terms of coded factors as shown in Eq. (7) and Eq. (8). Pectin yield (%) = 14.31 − 13.11X1 + 3.58X2 + 5.89X3 + 3.90X4 + 2.72X1² − 0.50X2² + 5.73X3² + 2.27X42 − 4.13X12 + 0.42X13 − 2.45X14 + 2.85X23 + 1.24X24 + 1.18X34 (7) DE (%) = 61.03 − 3.40X1 − 0.67X2 − 0.32X3 − 2.16X4 − 4.04X12 − 0.30X22 − 0.40X32 − 0.20X42 − 1.28X12 + 0.48X13 + 0.47X14 − 0.36X23 + 2.54X24 − 2.75X34 (8) The obtained quadratic models can be reduced by taking out all insignificant terms (p > 0.05), given by Eq. (9) and Eq. (10). Pectin yield (%) = 14.31 − 13.11X1 + 3.58X2 + 5.89X3 + 3.90X4 + 5.73X3²

(9)

DE (%) = 61.03 − 3.40X1 − 0.67X2 − 2.16X4 − 4.04X12

(10)

The analysis of variance (ANOVA) was applied to examine the statistical significance of the model terms. The results are listed in Table 3. The low p-values (< 0.05) indicated respectively by pectin yield and DE model, demonstrated that the developed model is significant and well fitted. The lack of fit for both models with high p-value (> 0.05) further emphasizes that the models can be used to predict the responses. The high value of R2 (0.94) and adj-R2 (0.87) of the yield models further signify that the response and independent variables were well correlated. An optimum pectin yield of 39.72% and DE of 57.56% for the full models, and pectin yield of 37.26% and DE of 59.72% for the reduced models were successfully predicted using the quadratic model obtained at extraction conditions of pH 1.80, time 141.38 min, at 88 °C with a liquid-solid ratio of 29:1. The model was validated 3 times with pectin yields of 39.88%, 40.54% and 36.98% while the validation results for DE were 58.55%, 60.14% and 58.99%. The average yield of 39.13% and average DE value of 59.23% are consistent with the predicted values. This further confirms that the model is well fitted with the experimental data. 3.3. Characterization of pomelo pectin Based upon the optimum extraction conditions obtained, the extracted pomelo pectin was characterized and the findings are presented in Table 5. The DE values obtained in this work being greater than 50% indicate that the pectin obtained were HM pectin. The GalA content confirmed that the extracted pectin is of good quality and within the acceptable limit of (GalA ≥ 65%) stipulated by the Food and Agriculture Organization specifications for pectin (Herbstreith, 2005). The solubility of pectin obtained in this work (76.27%) is higher than the commercial citrus pectin (54%) but is comparable with okra pectin (82%); once again suggesting that pomelo pectin has good solubility (Jiaxing et al., 2015). The color of the pectin is bright yellow which is similar to that obtained from grapefruit pectin (Wang et al., 2015). As illustrated in Fig. 2, in the initial 10 min of heat treatment at 95 °C, chain disaggregation occurred resulting in a rapid decrease in the viscosity of the pectin. However, the viscosity of pectin (1.92 mPa.s) was stabilized when the heat treatment time exceeded 30 min. During the entire high temperature treatment at 95 °C, the viscosity of the pectin changed slightly with the viscosities ranging from 1.84 to 2.24 mPa.s. The small variation observed in the viscosity suggests that the heating duration has very little effect on the viscosity of pomelo pectin. [Table 5] [Figure 2] 3.4. Effect of pH on the pectin extraction and morphology

6

The influence of low pH (~2) on pectin extraction was reported in a number of studies (Wai et al., 2010; Xue et al., 2011). In order to elucidate the effect of pH on pectin yield and DE within a low pH range, a series of experiments at pH of 1.00, 1.25, 1.50, 1.75, 2.00 were conducted with the other extraction parameters i.e. temperature, time, and liquid-solid ratio fixed at their centre points. The effect of pH on the yield and DE are illustrated in Fig. 3. At pH 1.50, pectin yield was at its maximum and the DE value was considerably high. At pH below 1.50, DE values were observed to markedly increase as pH decrease. DE is a ratio of the number of esterified carboxyl group to the number of total carboxyl groups present. Lowering pH increases the concentration of the H+ ions which in turn allows more free-carboxyl groups to be ionized to esterified carboxyl groups. Ionization causes the free carboxyl groups i.e. the non-methyl-esterified carboxyl groups (~1630 cm-1) to decrease as more methyl-esterified carboxyl groups (~1745 cm-1) are formed. Therefore, lowering the pH results in an increase in DE value as illustrated in the FT-IR spectra in Fig. 6 (Section 3.5). Galacturonic acid (GalA) content in food affects the chemical and sensorial characteristics of the matrix (pH, total acidity, microbial stability, sweetness, global acceptability) and provides information on the wholesome quality of the food (Manuela M. Moreira et al., 2010). The GalA contents of pectin extracted at different pHs are shown in Fig. 3. The GalA content for a good quality pectin is recommended by the Food and Agriculture Organization (FAO) to be above 65% (Zanella and Taranto, 2015). The pectin samples extracted at pH of 2.00 and 1.50 have both exceeded the 65% threshold. However, pectin sample extracted at pH 1.00 has an extremely low GalA content which is likely caused by degradation triggered by acid hydrolysis under low pH (de Oliveira et al., 2015; Garna et al., 2004). [Figure 3] The macroscopic examination of the extracted pectin is shown in Fig. 4. The effectiveness of the extraction process also depends on the ease of filtration of the extracted pectin. The pectin extracted at pH 1.00 was very viscous, which was difficult to be filtered and dried into solid form (Fig. 4a). The pectin extracted at pH 1.50 had a smooth and compact texture (Fig. 4b) whereas the pectin extracted at pH 2.00 showed rough and fragmented surfaces (Fig. 4c). The micrographs of pectin extracted at different pH values are shown in Fig. 4d-f. The pectin extracted at pH 1.50 (Fig. 4e) had smoother and more compact surface structure with fewer fragments than those obtained at pH 1.00 (Fig. 4d) and pH 2.00 (Fig. 4f). Theoretically, particles of smaller size would provide greater contact area for mass transfer, which can speed up the dissolution process. Thus, pectin extracted at pH 2.00 can be considered as a potential food additive powder. [Figure 4] Since the viscosity of pectin can affect the sensory performance of food products when used as food additives, the rheological properties of the extracted pectin were evaluated in this study. The rheological analyses of pectin gel samples are presented in Fig. 5. The calculated n values shown in Fig. 5a, ranged from 0.97 to 1.03, whereby the shear rate increased under steady-shear conditions. The flow behaviors of all pectin solutions under steady-shear conditions are shown in Fig. 5b. The pectin gel samples in the present study exhibited a shear-thinning (pseudo plastic) behavior. The pseudo plastic property of the polysaccharides allows liquid foods to be pumped more easily and further imparts a thinner consistency with better mouth feel (Chen et al., 2014; Liu et al., 2010). The viscosity of pectin decreased rapidly with increasing shear rate from 1 to 50 sˉ1 and then stabilized at shear rates of 100 sˉ 1. Moreover, the viscosities of all the samples (~0.01 Pa.s) of this study were similar to those obtained from pomelo pectin extractions using tartaric acid (Quoc et al., 2014) and oxalic acid (Quoc et al., 2015). [Figure 5] 3.5. Functional groups of extracted pectin FT-IR spectroscopy was performed to identify the major functional groups of pectin extracted at different pH values as illustrated in Fig. 6. The functional group region between 1600 cm-1 and 1800 cm-1 is useful for the identification and quantification of pectin (Liu et al., 2010). The mixed peaks in this region are caused by vibrations in the ester and carboxylic groups. As shown in Fig. 6, the absorbance intensity of the methyl-esterified carboxyl groups (~1745 cm-1) decreased with an increase in pH from 1−2 while the nonmethyl-esterified carboxyl groups (~1630 cm-1) increased as the pH increased. A clearer peak for non-methylesterified carboxyl groups (~1630 cm-1) was observed at both pH 2.00 and pH 1.75. As pH decreased from pH 1.50 to 1.00, the non methyl-esterified carboxyl groups (~1630 cm-1) gradually disappeared. This suggests that

7

the degradation of extracted pectin might have taken place and the identity of the pectin became unclear or the non-methyl-esterified carboxyl groups (~1630 cm-1) had possibly been transformed to methyl-esterified carboxyl groups (~1745 cm-1). This confirms the significant impact of pH on the functional groups of the pectin. Apart from affecting the methyl-esterified carboxyl groups (~1745 cm-1) and the non-methyl-esterified carboxyl groups (~1630 cm-1), pH also influenced the band around 1014 cm-1 that indicates the ester group C-O stretching region. It has also been reported that pectin sources were related to the fingerprint pattern of the characteristic region between 950 cm-1 to 1200 cm-1 (Kamnev et al., 1998). Within the stated region of wavelengths, the characteristic trend at pH 1.00 was different from the trends exhibited at pH 1.25−2.00 (Fig. 6). The DE obtained in the present study, is in agreement with the reported findings of Singthong et al. (2004) whereby the methyl-esterified carboxyl groups increased with DE values and decreased with non-methylesterified carboxyl groups. The FT-IR spectra of the current work at 1630 cm-1 were found to exhibit similar %Transmittance patterns to the work of Jiang et al. (2012) in which pectin was extracted from Akebia trifoliate var using hydrochloric and citric acids. The chemical structure and functional groups of the pectin extracted in this study also resembled those reported by Jiang et al. (2012); further confirming that the polysaccharide extracted from pomelo peel was pectin. [Figure 6] 3.6. Pectin extraction using Deep Eutectic Solvent (DES) In the present pectin extraction study, Deep Eutectics Solvents (DESs) were also employed to examine their ability to disrupt the structure of pomelo matrix. Since acids are known to be favorable chemical extraction agents, it is therefore preferable to employ acid based or acid−water based DES in extracting pectin. The presence of water in acid−water based DES can lower the viscosity of DES. This implies that a proper selection of mixture component for DES preparation is important as it could directly affect the quantity of chemical usage and also affects the extraction performance. Among the DES combination employed, lactic acid based DES was selected in preference to citric acid. The choice of DES was based on the lower molar mass and melting point of lactic acid at 90.08 g/mol and 16.8 ºC (liquid form) which has an added advantage in economizing the chemical and energy usage for DES preparation. In contrast, citric acid has a molar mass of 192.12 g/mol with a melting point at 153 ºC (solid form). [Table 1] As shown in Table 1, extractions were conducted using DES with varying water contents and different mixture combinations. Extraction was conducted at pH 1.80, extraction time of 141 min, 88 °C with a liquidsolid ratio of 29:1. Comparing the acid−sugar based DES prepared from different molar ratios (i.e. DES 1, DES 2 and DES 3), it was found that lactic acid−glucose−water DES 1 with a molar ratio of 6:1:6 produced the highest yield (23.04%) while lactic acid−glucose DES 3 with a molar ratio of 5:1 yielded the lowest at 7.39%. There are two possible reasons for the observation. DES 3 at pH 0.31 with extremely high acidity might have broken down the structure of pectin after disrupting the pomelo matrix; resulting in a lower yield. Moreover, the low viscosity of DES 1 (68.99 mPa.s) at 13.04 wt% H 2O could have facilitated mass transfer by allowing targeted compounds to be released from the plant matrix; thereby, increasing yield. A similar phenomenon was observed by Van den Bruinhorst et al. (2016) on orange peel using DES. In consideration of the probable causes of low pectin yields, less viscous acid−water based DESs may be considered to promote mass transfer in the extraction of pectin. One way of reducing the viscosity of the DESs is through dilution with water. However, water dilution was reported to weaken the hydrogen bonding interactions in DES components (Cui et al., 2015; Dai et al., 2015); thereby, decreasing the extraction yields of targeted compounds (Cui et al., 2015). The effect of water on DES extraction performance therefore, requires further investigation. In the extraction involving acid−amino based DESs (DES 4 and 5), complete crystallization of the lactic acid−glycine−water DES 4 (3:1:3) and the turning of the lactic acid−glycine DES 5 (9:1) into gel forms were observed when ethanol was added into the extract solution for pectin precipitation. Hence, the presence of water or the lack thereof, in the lactic acid–glycine combination were found to illicit different effects during pectin extraction. The results further demonstrated that the acid−amino based DESs (DES 4 and 5) were not suitable as media for pectin extraction. The possibility of applying a moderate temperature for extracting pectin using DESs was further investigated in the present study. DES 1−3 was applied for pectin extraction at a temperature of 50 °C. At 50 °C, no yield was observed for all the DES solvents used; indicating that the temperature did not supply adequate thermal energy to trigger the extraction process. 3.7. Comparison between citric acid extraction and DES extraction

8

The performances of pectin extractions using citric acid and DES were compared and tabulated in Table 6. The use of citric acid in the extraction is less environmental friendly than DES although its yield is higher than DES by 16%. Moreover, DES solvent preparation is more complicated as it involves mixing several chemicals to attain its deep eutectic point. The pH for DES extraction is also much lower than that used in the conventional acid extraction and it is not recommended to be used for food products. DES with high viscosity could also slow down the diffusion of targeted compounds from pomelo peel matrix. In terms of energy requirement, 1 g of pectin extracted from 10 g of peel powder using citric acid as extraction solvent saved 137.24 J of energy when compared with using DES as the extraction solvent. Taking into account of the quantity of extract, chemical usage, energy requirement and the complexity of the operation, the comparison results collectively suggest that conventional citric acid extraction of pectin is preferable to DES extraction. [Table 6] 4. Conclusions The extraction of pectin from pomelo peel using organic citric acid produced an optimum yield of 39.72% and a DE of 57.56% at pH 1.80 and 88 ºC, with an extraction time of 141 min at a liquid-solid ratio of 29:1. pH posed the greatest impact on pectin yield and DE values compared to other extraction parameters. The quadratic models developed from the optimization study were able to predict both yield and DE values accurately. The pectin yields obtained from pomelo peel are comparable to those obtained from other citrus fruit families with the DE being confirmed as slow set HM pectin. pH of 2.00 greatly modified the morphological structure of pectin and also indicated clearly the functional groups of pectin where as extraction pH of 1.00 gave the least GalA content in pectin. The pectin gel exhibited a pseudo-plastic behaviour with viscosity ~0.01 Pa.s. DES consisting of lactic acid−glucose−water with a ratio of 6:1:6 yielded the highest pectin yield among the investigated DES combinations. Comparatively, citric acid−based pectin extraction is superior to DES in terms of its yield, performance, operational attributes and economical features. Author's contribution The authors Gek Cheng Ngoh, Rozita Yusoff, Wen Hui Teoh have supervised this research work whereas the first author Shan Qin Liew has materially participated in this work. Besides, all authors also contributed in preparation of the manuscript, experimental and interpretation of data. Declaration of interest The authors declare that there is no conflict of interest. Submission declaration The present work has not been published previously in any form and not under consideration for publication elsewhere. Acknowledgement This work was supported by the University of Malaya Research Grant [RP002D-13AET, BKS0092016 and PG182-2015B].

9

References Arslan, N., Kar, F., 1998. Filtration of sugar-beet pulp pectin extract and flow properties of pectin solutions. J. Food Eng. 36, 113-122. Arslan, N., Toğrul, H., 1996. Filtration of pectin extract from grapefruit peel and viscosity of pectin solutions. J. Food Eng. 27, 191-201. Bagherian, H., Zokaee Ashtiani, F., Fouladitajar, A., Mohtashamy, M., 2011. Comparisons between conventional, microwave- and ultrasound-assisted methods for extraction of pectin from grapefruit. Chem. Eng. Process. 50, 1237-1243. Bakirtzi, C., Triantafyllidou, K., Makris, D.P., 2016. Novel lactic acid-based natural deep eutectic solvents: Efficiency in the ultrasound-assisted extraction of antioxidant polyphenols from common native Greek medicinal plants. Journal of Applied Research on Medicinal and Aromatic Plants 3, 120-127. Chen, Y., Zhang, J.-G., Sun, H.-J., Wei, Z.-J., 2014. Pectin from Abelmoschus esculentus: Optimization of extraction and rheological properties. Int. J. Biol. Macromol. 70, 498-505. Cui, Q., Peng, X., Yao, X.-H., Wei, Z.-F., Luo, M., Wang, W., Zhao, C.-J., Fu, Y.-J., Zu, Y.-G., 2015. Deep eutectic solvent-based microwave-assisted extraction of genistin, genistein and apigenin from pigeon pea roots. Sep. Purif. Technol. 150, 63-72. Dai, Y., van Spronsen, J., Witkamp, G.-J., Verpoorte, R., Choi, Y.H., 2013. Natural deep eutectic solvents as new potential media for green technology. Anal. Chim. Acta 766, 61-68. Dai, Y., Witkamp, G.-J., Verpoorte, R., Choi, Y.H., 2015. Tailoring properties of natural deep eutectic solvents with water to facilitate their applications. Food Chem. 187, 14-19. de Oliveira, C.F., Giordani, D., Gurak, P.D., Cladera-Olivera, F., Marczak, L.D.F., 2015. Extraction of pectin from passion fruit peel using moderate electric field and conventional heating extraction methods. Innov. Food Sci. & Emerg. Technol. 29, 201-208. Elizabeth Devi, W., Shukla, R.N., Anitha Abraham, Surender Jarpula, Kaushik, U., 2014. Optimized extraction condition and characterization of pectin from orange peel. Int. J. Res. Tech. 2, 1-9. Filisetti-Cozzi, T.M.C.C., Carpita, N.C., 1991. Measurement of uronic acid without interference for neutral sugars. Anal. Biochem. 197, 157-162. Francisco, M., van den Bruinhorst, A., Kroon, M.C., 2012. New natural and renewable low transition temperature mixtures (LTTMs): screening as solvents for lignocellulosic biomass processing. Green Chem. 14, 2153-2157. García, A., Rodríguez-Juan, E., Rodríguez-Gutiérrez, G., Rios, J.J., Fernández-Bolaños, J., 2016. Extraction of phenolic compounds from virgin olive oil by deep eutectic solvents (DESs). Food Chem. 197, 554-561. Garna, H., Mabon, N., Wathelet, B., Paquot, M., 2004. New method for a two-step hydrolysis and chromatographic analysis of pectin neutral sugar chains. J. Agric. Food Chem. 28, 4652–4659. Herbstreith, F., 2005. The Specialist for Pectin. Confectinery Production. Novenburg. Jiang, Y., Du, Y., Zhu, X., Xiong, H., Woo, M.W., Hu, J., 2012. Physicochemical and comparative properties of pectins extracted from Akebia trifoliata var. australis peel. Carbohydr. Polym. 87, 1663-1669. Jiaxing, L., Chuncheng, S., Lang, M., Yue, W., Xiufeng, X., Yanhui, Z., 2015. Physicochemical properties of Okra pectin. Food Sci. 36, 104-108. Kamnev, A.A., Colina, M., Rodriguez, J., Ptitchkina, N.M., Ignatov, V.V., 1998. Comparative spectroscopic characterization of different pectins and their sources. Food Hydrocoll. 12, 263-271. Kanmani, P., Dhivya, E., Aravind, J., Kumaresan, K., 2014. Extraction and analysis of pectin from citrus peels: Augmenting the yield from citrus limon using statistical experimental design. Iran. J. energy Environ. 5, 303-312. Kliemann, E., De Simas, K.N., Amante, E.R., Prudêncio, E.S., Teófilo, R.F., Ferreira, M.M.C., Amboni, R.D.M.C., 2009. Optimisation of pectin acid extraction from passion fruit peel (Passiflora edulis flavicarpa) using response surface methodology. Int. J. Food Sci. Tech. 44, 476-483. Koffi, K.L., Yapo, B.M., Besson, V., 2013. Extraction and characterization of gelling pectin from the peel of Poncirus trifoliata fruit. Agric. Sci. 4, 614-619. Koubala, B.B., Kansci, G., Mbome, L.I., Crépeau, M.J., Thibault, J.F., Ralet, M.C., 2008. Effect of extraction conditions on some physicochemical characteristics of pectins from “Améliorée” and “Mango” mango peels. Food Hydrocoll. 22, 1345-1351. Li, J.X., Shi, C.C., Ma, L., Wu, Y., Xiao, X.F., Zhou, Y.H., 2015. Physicochemical Properties of Okra Pectin. Food Sci. 36, 104-108. Lim, J., Yoo, J., Ko, S., Lee, S., 2012. Extraction and characterization of pectin from Yuza (Citrus junos) pomace: A comparison of conventional-chemical and combined physical–enzymatic extractions. Food Hydrocoll. 29, 160-165.

10

Liu, L., Cao, J., Huang, J., Cai, Y., Yao, J., 2010. Extraction of pectins with different degrees of esterification from mulberry branch bark. Bioresour. Technol. 101, 3268-3273. Manrique, G.D., Lajolo, F.M., 2004. Cellwall polysaccharide modifications during postharvest ripening of papaya fruit(Carica papaya). Postharvest Biol. Technol. 33, 11-26. Manuela M. Moreira, Luís F. Guido, José M. Cruz, Barros, A.A., 2010. Determination of galacturonic acid content in pectin from fruit juices by liquid chromatographydiode array detection-electrospray ionization tandem mass spectrometry. Cent. Eur. J. Chem. 8, 1236-1243. Methacanon, P., Krongsin, J., Gamonpilas, C., 2014. Pomelo (Citrus maxima) pectin: Effects of extraction parameters and its properties. Food Hydrocoll. 35, 383-391. Mohnen, D., 2008. Pectin structure and biosynthesis. Curr. Opin. Plant Biol. 11, 266-277. Muhammad, K., Mohd. Zahari, N.I., Gannasin, S.P., Mohd. Adzahan, N., Bakar, J., 2014. High methoxyl pectin from dragon fruit (Hylocereus polyrhizus) peel. Food Hydrocoll. 42, Part 2, 289-297. Oliveira, T.Í.S., Rosa, M.F., Cavalcante, F.L., Pereira, P.H.F., Moates, G.K., Wellner, N., Mazzetto, S.E., Waldron, K.W., Azeredo, H.M.C., 2015. Optimization of pectin extraction from banana peels with citric acid by using response surface methodology. Food Chem. 198, 113-118. Paradiso, V.M., Clemente, A., Summo, C., Pasqualone, A., Caponio, F., 2016. Towards green analysis of virgin olive oil phenolic compounds: Extraction by a natural deep eutectic solvent and direct spectrophotometric detection. Food Chem. 212, 43-47. Prakash Maran, J., Sivakumar, V., Thirugnanasambandham, K., Sridhar, R., 2013. Optimization of microwave assisted extraction of pectin from orange peel. Carbohydr. Polym. 97, 703-709. Quoc, L.P.T., Anh, L.T.L., Tien, M.V.T.K., Trang, L.T., 2014. Optimization of the pectin extraction from pomelo peels by oxalic acid and microwave. Banats J. Biotechnol. 9, 67-73. Quoc, L.P.T., Huyen, V.T.N., Hue, L.T.N., Hue, N.T.H., Thuan, N.H.D., Tam, N.T.T., Thuan, N.N., Duy, T.H., 2015. Extraction of pectin from pomelo (Citrus maxima) peels with the assistance of microwave and tartaric acid. Int Food Res. J. 22, 1637-1641. Radojkovic, M., Zekovic, Z., Jokic, S., Vidovic, S., Lepojevic, Z., Milosevic, S., 2012. Optimization of solidliquid extraction of antioxidants from Black Mulberry leaves by response surface methodology. Food Technol. Biotechnol. 50, 167-176. Rozema, E., van Dam, A., Sips, H., Verpoorte, R., Meijer, O., Kooijman, S., Choi, Y., 2015. Extending pharmacological dose-response curves for salsalate with natural deep eutectic solvents. RSC Advances 5, 61398-61401. Singthong, J., Cui, S.W., Ningsanond, S., Douglas Goff, H., 2004. Structural characterization, degree of esterification and some gelling properties of Krueo Ma Noy (Cissampelos pareira) pectin. Carbohydr. Polym. 58, 391-400. Van den Bruinhorst, A., Kouris, P., Timmer, J., de Croon, M., Kroon, M., 2016. Exploring Orange Peel Treatment with Deep Eutectic Solvents and Diluted Organic Acids. Nat. Prod. Chem. Res. 4, 242. Wai, W.W., Alkarkhi, A.F.M., Easa, A.M., 2010. Effect of extraction conditions on yield and degree of esterification of durian rind pectin: An experimental design. Food Bioprod. Process. 88, 209-214. Wang, W., Ma, X., Xu, Y., Cao, Y., Jiang, Z., Ding, T., Ye, X., Liu, D., 2015. Ultrasound-assisted heating extraction of pectin from grapefruit peel: Optimization and comparison with the conventional method. Food Chem. 178, 106-114. Willats, W.G.T., Knox, J.P., Mikkelsen, J.D., 2006. Pectin: new insights into an old polymer are starting to gel. Trends Food Sci. Technol. 17, 97-104. Woo, K.K., Chong, Y.Y., Li Hiong, S.K., Tang, P.Y., 2010. Pectin extraction and characterization from red dragon fruit (hylocereus polyrhizus): A preliminary study. J. Bio.l Sci. 10, 631-636. Xue, Z.-h., Zhang, X., Zhang, Z.-j., Liu, J.-h., Wang, Y.-f., Chen, D.-x., Long, L.-s., 2011. Optimization of pectin extraction from citrus peel by response surface methodology. J. Food Sci. Technol. 32, 128-132. Yuliarti, O., Goh, K.K.T., Matia-Merino, L., Mawson, J., Brennan, C., 2015. Extraction and characterisation of pomace pectin from gold kiwifruit (Actinidia chinensis). Food Chem. 187, 290-296. Zanella, K., Taranto, O.P., 2015. Influence of the drying operating conditions on the chemical characteristics of the citric acid extracted pectins from ‘pera’ sweet orange (Citrus sinensis L. Osbeck) albedo and flavedo. J. Food Eng. 166, 111-118.

11

Figures Captions Fig. 1. Response surface plots showing the effect of process variable on pectin yield and DE. Fig. 2. Viscosity of extracted pectin versus heating time at 95 °C. Fig. 3. Effect of pH on pectin yield, DE and GalA value. Fig. 4. Photographs and scanning electron micrograph of pectin at (a) pH 1.00; (b) pH 1.50 and (c) pH 2.00; at ×500 magnification, 100 μm. Fig. 5. Flow behaviors curves and viscosity curves for the pectin gel at different pH. Fig. 6. FT-IR spectra of pomelo pectin at different pH.

Tables Table 1 DES selected for pectin extraction. Abbreviation

Composition

Mole ratio

Combination

Type

Preparation conditions

DES Appearance

DES pH

DES Viscosity (mPa.s)

DES wt.% H₂ O

Pectin Yield (%)

DE (%)

12

DES1

Lactic acid (D): Glucose (A): Water

(6:1:6)

Acid: Sugar

60 min, 50 °C, 500 rpm

Transparent clear liquid

0.56

68.99

13.04

23.04

79.15

DES2

Lactic acid (D): Glucose (A): Water

(5:1:3)

Acid: Sugar

60 min, 50 °C, 500 rpm

Transparent light yellow

0.48

181.10

7.90

13.34

88.20

DES3

Lactic acid (D): Glucose (A)

(5:1)

Acid: Sugar

60 min, 50 °C, 500 rpm

Transparent light yellow

0.31

416.90

n/d

7.39

90.98

DES4

Lactic acid (D): Glycine (A): Water

(3:1:3)

Acid: Amino acid

45 min, 70 °C, 550 rpm

Transparent light yellow

2.52

112.50

13.53

Crystallizes

n/d

DES5

Lactic acid (D): Glycine (A)

(9:1)

Acid: Amino acid

45 min, 70 °C, 550 rpm

Transparent light yellow

1.85

166.50

n/d

Gelation

n/d

13

Table 2 Experimental conditions from the BBD and the experimental results for pomelo pectin extraction using citric acid. Run 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

x₁ 1 1 0 0 0 -1 0 0 0 0 1 0 -1 1 0 -1 1 0 -1 0 0 0 -1 1 -1 0 0 0 0

(X₁ ) (2.5) (2.5) (2.0) (2.0) (2.0) (1.5) (2.0) (2.0) (2.0) (2.0) (2.5) (2.0) (1.5) (2.5) (2.0) (1.5) (2.5) (2.0) (1.5) (2.0) (2.0) (2.0) (1.5) (2.5) (1.5) (2.0) (2.0) (2.0) (2.0)

Independent variables x₂ (X₂ ) -1 (40) 0 (110) 0 (110) 0 (110) -1 (40) 1 (180) 1 (180) 1 (180) 1 (180) 1 (180) 0 (110) 0 (110) -1 (40) 1 (180) 0 (110) 0 (110) 0 (110) 0 (110) 0 (110) -1 (40) -1 (40) 0 (110) 0 (110) 0 (110) 0 (110) 0 (110) 0 (110) -1 (40) 0 (110)

x₃ 0 -1 0 0 0 0 1 0 0 -1 0 1 0 0 0 1 0 0 0 1 -1 1 0 1 -1 -1 0 0 -1

(X₃ ) (77.5) (65.0) (77.5) (77.5) (77.5) (77.5) (90.0) (77.5) (77.5) (65.0) (77.5) (90.0) (77.5) (77.5) (77.5) (90.0) (77.5) (77.5) (77.5) (90.0) (65.0) (90.0) (77.5) (90.0) (65.0) (65.0) (77.5) (77.5) (65.0)

x₄ 0 0 0 0 1 0 0 1 -1 0 1 1 0 0 0 0 -1 0 1 0 0 -1 -1 0 0 -1 0 -1 1

(X₄ ) (25:1) (25:1) (25:1) (25:1) (30:1) (25:1) (25:1) (30:1) (20:1) (25:1) (30:1) (30:1) (25:1) (25:1) (25:1) (25:1) (20:1) (25:1) (30:1) (25:1) (25:1) (20:1) (20:1) (25:1) (25:1) (20:1) (25:1) (20:1) (30:1)

Dependent variables Yield (%) 4.51 4.47 14.57 12.30 16.84 37.70 31.79 24.45 12.18 12.39 5.22 37.50 18.46 7.24 20.61 39.57 6.87 10.88 36.38 20.75 12.73 24.75 28.25 10.21 35.51 10.39 13.20 9.54 18.41

DE (%) 54.52 51.32 59.70 58.03 58.61 63.13 58.00 59.24 57.81 58.28 49.71 54.22 57.91 54.61 61.01 61.35 54.53 58.90 55.93 60.94 59.77 63.29 62.64 55.00 59.58 62.85 67.50 67.34 64.79

14

Table 3 Analysis of variance (ANOVA) for regression model of pectin yield and DE. Term Model X1-pH X2-time X3-temperature X4-L/S ratio X1² X2² X3² X4² X12 X13 X14 X23 X24 X34 Residual Lack of Fit Pure Error Cor Total R² Adj R²

SS 3217.04 2063.25 153.51 416.19 182.68 48.04 1.65 213.08 33.43 68.15 0.71 23.91 32.38 6.18 5.59 220.31 163.52 56.79 3437.35 0.94 0.87

DF 14 1 1 1 1 1 1 1 1 1 1 1 1 1 1 14 10 4 28

Pectin Yield MS F-value 229.79 14.6 2063.25 131.11 153.51 9.75 416.19 26.45 182.68 11.61 48.04 3.05 1.65 0.11 213.08 13.54 33.43 2.12 68.15 4.33 0.71 0.05 23.91 1.52 32.38 2.06 6.18 0.39 5.59 0.36 15.74 16.35 1.15 14.2

p-value < 0.0001 < 0.0001 0.0075 0.0001 0.0043 0.1025 0.7507 0.0025 0.1670 0.0563 0.8354 0.2380 0.1734 0.5411 0.5606 0.4845

SS 376.25 139.06 5.36 1.20 56.16 105.99 0.59 1.02 0.26 6.58 0.91 0.89 0.53 25.81 30.31 135.51 78.34 57.17 511.77 0.74 0.47

DF 14 1 1 1 1 1 1 1 1 1 1 1 1 1 1 14 10 4 28

DE MS 26.88 139.06 5.36 1.20 56.16 105.99 0.59 1.02 0.26 6.58 0.91 0.89 0.53 25.81 30.31 9.68 7.83 14.29

F-value 2.78 14.37 0.55 0.12 5.80 10.95 0.06 0.11 0.03 0.68 0.09 0.09 0.05 2.67 3.13

p-value 0.0330 0.0020 0.4691 0.7303 0.0304 0.0052 0.8089 0.7498 0.8716 0.4235 0.7634 0.7658 0.8191 0.1248 0.0986

0.55

0.7992

15

Table 4 The effects of different extraction conditions on the qualitative and quantitative characteristics of extracted pectin from various citrus family sources. Parameters

Sources

1:18

citric acid

90

1:25

1.5

60

2.5 85 88

pH

Time (min)

S/L ratio (g/ml)

65

3.5

68

75

1.5

DE (%)

Lemon

36.71

33.77

Kanmani et al. (2014)a

nitric acid

Lime

19.80

77.00

Koffi et al. (2013)b

1:30

citric acid

Orange

67.30

35.85

Elizabeth Devi et al. (2014)a

120

1:70

citric acid

‘Pera’ sweet orange

38.21

70.21

Zanella and Taranto (2015)b

2.5

90

1:25

hydrochloric acid

Grapefruit

21.10

68.20

Arslan and Toğrul (1996)b

1.8

141

1:29

citric acid

Pomelo

39.72

57.56

Present studya

1.5

90

1:30

hydrochloric acid

Grapefruit

19.16

75.60

Bagherian et al. (2011)b

2.0

180

1:30

nitric acid

Pomelo

27.63

55.82

Methacanon et al. (2014)a

90

b

References

Yield (%)

80

a

Responses

Extraction agent

Temperature (°C)

based on optimized results optimization not available – data obtained from highest yield

16

Table 5 Physicochemical composition of the extracted pomelo pectin. Yield (%)a,b Degree of Esterification (%)a,b Galacturonic acid (%)a,b Moisture (%)a,b Ash (%)a,b Protein (%)b Fat (%)b pHa,b Molecular Weight, Mw (Da)b Solubility (%)a,b Cie Lab coordinatesa,c L* a* b* H*ab C* a

39.13±1.89 59.23±0.82 68.54±0.85 14.60±0.31 1.28±0.12 2.11 0.02 2.07±0.01 9.05×10⁴ 76.27±0.51 10.57±0.13 2.62±0.39 11.32±0.14 77.01±1.78 11.63±0.21

data are expressed as means ± standard deviations of triplicate. dried pectin,cwet pectin.

b

17

Table 6 Comparison between pectin extraction using citric acid and DES (lactic acid−glucose−water 6:1:6). Citric acid extraction

DES extraction

Green solvent

Yes

Yes

Chemical use

Low

High

Solvent preparation time

Fast

Slow

Solvent acidity

Low (pH= 1.80)

High (pH= 0.56)

Solvent viscosity

Low (1.38 mPa.s)

High (68.99 mPa.s)

Energy consumption

Low

High (Extra energy needed for prepare solvent)

Highest yield can achieve among the solvent investigated by using same extraction condition

39.13%

23.04%

Energy required per 10 g peel powder to produce 1 g pectin (J)

305.9

443.14

Highlights · Predictive model for pectin extraction from pomelo peels · Interactive effects of pH, time, temperature and liquid-solid ratio on pectin yield · Influence of low pH on morphological structure and functional group of pectin · Impacts of water content and molar ratio on pectin yield in DES extraction · Energy requirement and yield performances between citric acid and DES extraction

18

Fig. 1

Fig. 2

2.30 2.20 Viscosity (mPa.s)

2.10 2.00 1.90 1.80 1.70 1.60 1.50 0

10

20

30

40

Heating time (min)

50

60

70

Fig. 3

90.00 82.00

79.00 80.00

Percentage (%)

70.00

75.06 64.03 57.31

60.00

55.71

52.53

50.13

52.72

50.00 40.70

35.32

36.12

40.00

26.83

30.00 19.18 20.00 9.83 10.00 0.00 1.00

1.25

1.50

1.75

pH Pectin Yield

DE

GalA

2.00

Fig. 4

Fig. 5

Fig. 6