Pectic hydrocolloids from steam‐exploded lime pectin peel: Effect of temperature and time on macromolecular and functional properties

Abstract Previously, we showed the weight average molecular weight (M w) and intrinsic viscosity ([ƞ]) of pectic hydrocolloids recovered from steam‐exploded citrus peel were low, suggesting fragmentation due to process temperature and/or time‐at‐temperature. We have tested this hypothesis on a commercial lime pectin peel, washed to remove soluble sugars and dried for stabilization, using a static steam explosion system. We examined temperatures of 120–150°C at 1–3 min hold times. Galacturonic acid recovery and M w ranged from 22% to 82% and 142–214 kDa, respectively. Recovery of most major pectic sugars increased concomitantly with galacturonic acid as temperature and time‐at‐temperature increased. [ƞ] ranged from 1.75 to 6.83 dl/g. The degree of methylesterification ranged from 66.5% to 72.1%. Tan (δ) (Loss modulus/Storage modulus; G″/G′) values of sugar–acid gels for 120–140°C treatments were <1.0. Ideal optimization analysis, where time, [ƞ], and percent recovery were maximized, identified processing conditions that favor either increased [ƞ] or percent recovery. The results presented here support our hypothesis that temperature and time‐at‐temperature affect M w and [η] of the recovered pectic hydrocolloids. These results also demonstrate that manipulating either temperature or time‐at‐temperature enables the production of structurally varied populations of pectic hydrocolloids. Based on optimization analysis, commercially viable values of [ƞ] can be obtained while recovering approximately 50% of the pectic hydrocolloids.

application sectors (Technavio, 2018). Nearly 63.5 × 10 6 kg of pectin were estimated to have been produced in 2017 (IMARC, 2018), of which approximately 85% is estimated to have been from citrus (Staunstrup, 2018). Despite this increasing demand for pectin, the potential supply far exceeds these enthusiastic predictions as approximately 10 × 10 6 Mg of oranges for processing were forecast from Brazil and 3 × 10 6 Mg from the United States for 2017-2018 (Anonymous, 2018) with a pectin content of approximately 2% of fruit fresh weight.
Supporting evidence has been provided that RG I regions are interspersed between HG regions (Coenen et al., 2007) and that the degree of polymerization for HG regions in citrus pectin is between 81 and 117 GalAs (Yapo et al., 2007). Non-HG-associated RG I has been reported from highly soluble fruit parenchyma cells (Cornuault et al., 2018) and mucilage from Arabidopsis seeds (Poulain et al., 2019). There is also an RG II region that is more complex but less frequent in general (Caffall & Mohnen, 2009), accounting for between 1% and 4% of the primary cell walls of dicots, nongrass monocots, and gymnosperms (O'Neill et al., 2004).
The most common method for extracting commercial grade pectin is via the use of a hot acid process (Rolin et al., 2002). More recently, numerous alternative technologies have been proposed for pectin extraction, but none have been commercialized (Adetunji et al., 2017;Fishman et al., 2006;Kaya et al., 2014). We have developed a continuous process to release pectic hydrocolloids from their intracellular entrapment using steam explosion (Cameron et al., 2016(Cameron et al., , 2017a(Cameron et al., , 2017bDorado et al., 2017), which has the benefit of enabling pectin extraction from steam-exploded citrus fruit peel using a simple water wash. Data from these previous studies suggest that the observed low M w and [ƞ] may have resulted from the processing variables of temperature and time-at-temperature.
To test this hypothesis, we utilized a static, batch steam explosion system (Grohmann et al., 2013) to process a stabilized lime pectin peel at temperatures ranging from 120 to 150°C and for hold times of 1, 2, and 3 min. Data on the recovery of the major pectic sugars, macromolecular properties, and functionality of the recovered pectic hydrocolloids were obtained.

| Steam explosion
Stabilized lime pectin peel (milled, washed, pressed, and dried peel from juice-extracted fruit) was provided by CP Kelco. The stabilized pectin peel was rehydrated overnight with deionized water at 4°C using six weight equivalents of the peel mass. Steam explosion on 600 g of rehydrated pectin peel was performed with a static benchscale system (Dorado et al., 2019;Grohmann et al., 2013;Widmer et al., 2010). Steam was introduced into the reaction vessel, and the pressure was maintained to produce temperatures of 120, 130, 140, and 150°C by a thermocouple. The temperature is monitored using a McMaster-Carr type thermocouple attached to a stainless steel cap. The thermocouple is inserted into the vessel at the top of the steam gun. Once the thermocouple is fully inserted, the cap and a silicone gasket at the top of the thermocouple are secured with a clamp to the steam gun. The thermocouple is attached to a custombuilt temperature monitoring box with a digital display. After obtaining the desired temperature and pressure, which took approximately 45 s (Grohmann et al., 2013), the pressure and temperature were maintained for 1, 2, or 3 min. Three replicates were performed for each treatment, and the three replicates were then pooled. The pooled samples were transferred to plastic bags, sealed, and stored at −20°C until analyzed.

| Extraction of pectic hydrocolloids from steamexploded pectin peel
Pectic hydrocolloids were extracted from frozen, steam-exploded pectin peel as previously described (Cameron et al., 2016). Briefly, equal masses of water and steam-exploded tissue (100 g each) were mixed together and placed on a wrist shaker for 30 min. Two replicates were prepared for each treatment. A total of three washes were performed on each replicate. After each wash, the mixture was centrifuged at a relative centrifugal force (RCF) of 15,000 g for 20 min at 4°C. Following centrifugation, the wash liquids for each replicate were pooled and residual insoluble solids were removed from the supernatants by filtration through 1.2μm glass filter fiber (GF/C, Whatman/GE Healthcare Life Sciences Ltd.). An aliquot of the pooled supernatant was enzyme-digested, and the pectic sugars present were quantified by HPAEC-pulsed amperometric detection as previously described (Cameron et al., 2016). The percent recovery was calculated by dividing the amount of the sugar in the steamexploded biomass by the amount of the sugar in the pooled water extract. All estimates were made on a dry weight basis. Pectic hydrocolloids were recovered by precipitation with acidified ethanol at 4°C overnight (Cameron et al., 2016;Kertez, 1951). Following centrifugation, as described above, the pellets were frozen in liquid nitrogen and lyophilized as previously described (Cameron et al., 2016). Lyophilized pectic hydrocolloids were made into 2% (w/v) solutions in deionized water and then extensively dialyzed against multiple changes in deionized water using 6,000-8,000 Da MWCO dialysis tubing (Spectra/Por) overnight. After dialysis, the retentate was precipitated and lyophilized as previously described (Cameron et al., 2016). Dialyzed and lyophilized pectic hydrocolloids were stored at −80°C.

| Recovery of pectic hydrocolloids
Percent recovery of the major citrus pectic hydrocolloid sugars (GalA, rhamnose, galactose, and arabinose) was estimated by the concentration present in the steam-exploded peel and the recovered pectic hydrocolloids. The concentrations of these pectic sugars were estimated as previously described following enzymatic hydrolysis and high-performance anion-exchange chromatography (Cameron et al., 2016).
The DM of the pectic hydrocolloids was determined by titration according to a method modified from that found in the United States Pharmacopeia (Pharmacopeia US, 1995). The prepared solution of pectic hydrocolloids was titrated against sodium hydroxide of known molarity using bromothymol blue as an indicator and saponified for 15 min at room temperature with an excess of base. After the excess base was neutralized, the solution was titrated a second time, and the DM was calculated as in Equation (1).

| High-performance size-exclusion chromatography of pectic hydrocolloids
Dialyzed, lyophilized pectic hydrocolloids were chromatographed as previously described (Dorado et al., 2019). A dn/dc value of 0.132 was used (Fishman et al., 2003). Electronic outputs from all the scattering angles measured by the multiangle light scattering detector (MALLS), differential pressure detector (DP), and differential refractive index detector (dRI) were processed with ASTRA software (Ver. 6.1.1.17; Wyatt Technology). Each sample was replicated a minimum of three times. The Astra software enables the estimation of several macromolecular parameters, including M w , M n, and [η].

| Standard acid in glass
USA-SAG (standard acid in glass) values were determined using a Ridgelimiter according to methods detailed by the International Pectin Producers Association (Anonymous, 2017;Cox & Higby, 1944;Joseph & Baier, 1949). Gels were formulated assuming 150 °SAG. Briefly, gels are prepared to contain 650 g of total soluble solids (sucrose plus pectin). Assuming a 150 °SAG, the amount of pectin added would equal 650/150 = 4.33 g pectin, which is rounded down to 4 g, plus 646 g sucrose. The pectin is mixed with 20-30 g of the sugar and solubilized in 410 ml deionized water. After solubilization of the sugar plus pectin, the solution is heated to boiling, and the remaining sugar is added in two portions. Then, the solution is heated until a weight of 1,015 g is reached. The heated solution is allowed to rest for 1 min and then poured into prepared glass cups of standardized size and shape, which contain 2 ml of a 48.8% (w/v) tartaric acid solution.
The jellies are then stored for 20-24 hr at constant temperature (25°C ± 3°C) before being removed from the glass. The amount of sag is measured after 2 min using a Ridgelimiter.

| Rheology
Sugar-acid gels made from recovered pectic hydrocolloids were prepared by the method of Yoo et al. (2003) with only slight modifications. Pectic hydrocolloids (0.2 g) were solubilized in 7.3 g of 0.1 M citrate buffer (pH 3.0) by stirring overnight. Subsequently, they were centrifuged for 30 min at an RCF of 12,100 g. The supernatant was brought to 60% sugar, which was dissolved thoroughly in a 98°C water bath for 30 min. The sugar gel was placed on the Peltier of the rheometer (AR1000; TA Instruments), and the geometry (parallel plate, 500 µm gap) was lowered into place. The excess gel which extruded from under the geometry was removed. Gel conditioning was done at 20°C for 2 min prior to measuring the rheological properties of the pectin-sugar mixtures in dynamic shear. Dynamic shear data were obtained from frequency sweeps over the range of 0.08-628 rad/s at a 2% strain, which was in the linear viscoelastic region.
Storage modulus (G′) and loss modulus (G″) were obtained, and tan δ, which indicates more gel-like than liquid-like properties, was calculated dividing G″ by G′.

| Statistical analysis of data
An optimization analysis was performed using Design-Expert (ver-

| Recovery of pectic hydrocolloids from steamexploded pectin peel
The measured recovery percentage of the major pectic sugars generally increased with increasing temperature and time-at-temperature ( Figure 1). GalA recovery, the dominate sugar in citrus pectin, ranged from 22% to 82% (Figure 1d). The response surface and contour plot illustrating the predicted relationship between time and timeat-temperature for GalA recovery percent from steam explosion, indicates that time, time-at-temperature, or both can be manipulated to obtain desired levels for GalA recovery ( Figure S1A). For percent recovery of GalA, ANOVA indicated that the quadratic model was significant at p = .01 level. Only temperature was a significant factor at the p = .001 level (Table 1).
Previous results from either continuous or single-batch steam explosion of citrus peel produced recovery percentages (based on GalA) ranging between 58% and 78% for raw, unwashed orange fruit peel (Cameron et al., 2016;Grohmann et al., 2013). At a treatment tempera-

| Degree of methylesterification
There were significant differences among DM values with the various treatments, for both time-at-temperature within a  (May, 1990).

| Intrinsic viscosity
Functionality, as measured by intrinsic viscosity [η], was significantly affected by the experimental parameters ( Figure 4, Table 1). The model was significant at p = .001, temperature was significant at p = .001, time-at-temperature was significant at p = .05, and temperature 2 was significant at p = .01. Two-way ANOVA revealed numerous significant differences within a temperature group for the various times and between temperature groups for time-attemperature ( Figure 4). Figure S1B is the response surface and contour plot illustrating the predicted relationship between time and time-at-temperature [η] = intrinsic viscosity (dl/g); G′ = storage modulus; p = probability value; R 2 = regression coefficients. *, **, and *** represent significance at p < .05, p < .01, and p < .001, respectively; X 1 and X 2 represent temperature (°C) and time-at-temperature (min), respectively. TA B L E 1 ANOVA statistics for results that demonstrated a significant (p > .05) statistical model F I G U R E 2 (a) Representative SEC-MALLS-RI-DP chromatogram. The shaded box represents the area of the chromatogram designated as the pectic hydrocolloid peak. LS = laser, RI = refractive index, DP = differential pressure. (b) Weight average molecular weight of recovered pectic hydrocolloids. Error bars represent the standard error of the mean for of at least three replicates. Bars with different lower case letters indicate a statistically significant difference (p > .05) within a temperature group, and bars with different upper case letters indicate a statistically significant difference (p > .05) for different temperatures within a treatment time

| Rheology
The functionality of the recovered pectic hydrocolloids also was explored by examining the rheology of sugar-acid gels. Figure 5 shows Tan (δ) quantifies the balance between energy loss and storage.
A value for tan (δ) greater than unity indicates more liquid properties, whereas one lower than unity means more solid properties, regardless of the viscosity. Samples prepared at 150°C for 2 and 3 min manifested more viscous properties than viscoelastic, gellike properties which were indicated by tan (δ) values higher than 1.0 ( Figure S2). The loss of gelling properties in these samples could be derived from their lower M w compared with samples from other treatments. For G′, the ANOVA model was significant at p = .05, and temperature was a significant factor at p = .01 level ( Figure S1C and Table 1), manifesting higher G′ at lower temperatures.
The magnitudes of G′ and G″ of pectin-sugar mixtures increased as ω increased, showing that G′ was much higher than G″ at all values of ω with high frequency dependency. A similar trend has been reported with other high DM pectin containing sugar gels (Evageliou et al., 2000;Silva et al., 1995). Plots of ln G′ and G″ versus ln ω of true gels typically displays a slope of zero, and G′ is higher in magnitude than G″ over broad ranges

| Standard acid in glass (SAG)
Functionality, as based on USA-SAG testing, was limited as pectic hydrocolloids from the 120°C-3 min, and all of the 130°C treatments did not produce a gel that could be measured using the Ridgelimiter.
The °SAG could only be measured on the 120°C-2 min gel ( Figure 6) as there was insufficient material to test the 120°C-1 min pectic hydrocolloids. The 120°C-2 min gel had a °SAG value of 180. Since the 130°C samples did not form sufficiently strong gels, the 140 and 150°C samples were not tested.

| Optimization analysis for intrinsic viscosity and percent recovery
Using the historical data capabilities of the Design-Expert software, we performed an optimization analysis to estimate the optimal temperature and time-at-temperature to maximize both [ƞ] and percent recovery. The goal for temperature was set within the range, and time, [ƞ], and % recovery were set to maximize their value. Time was maximized because longer time-at-temperature values are more realistically feasible than very short ones. The models were reduced by removing terms with p values greater than p > .1 as recommended by Design-Expert. The Desirability function using these parameters was 0.708. Using coded values (−1, 0, +1, etc.) for time and time-at-temperature, the equation for predicting [ƞ] was: (2) [ ] = 5.24 − 1.79 × temperature − 0.46 × time − 1.10 × temperature 2 F I G U R E 6 Gel produced from pectic hydrocolloids recovered from the 120°C-2 min treatment used for °SAG determination and for predicting percent recovery, it was: Contour plots for the optimization analysis (Figure 7) show the surfaces resulting from these equations. Figure

| CON CLUS IONS
The results presented here and by Dorado et al. (2019) (Widmer et al., 2010) support our hypothesis that temperature and time-attemperature used in our previous studies (Cameron et al., 2016(Cameron et al., , 2017bGrohmann et al., 2013) Muzamal et al. (2015) has shown that the explosion step alone is not necessary to disintegrate the biomass studied. Brownell et al. (1986) also showed that pressure drop itself was not necessary for the release of glucose. These results also demonstrate that manipulating either temperature or time-at-temperature would enable the production of structurally varied populations of pectic hydrocolloids. Using steam explosion coupled to simple water extraction to obtain pectic hydrocolloids from juice-extracted or culled citrus fruit could open new applications for inexpensive, environmentally friendly pectic hydrocolloids where rheology modification, ion-capture, or hydration control functionalities are needed.

ACK N OWLED G M ENTS
We would like to thank Peiling Li and Sandra Matlack for expert technical support.

DATA AVA I L A B I L I T Y S TAT E M E N T
Data available on request from the authors. The data that support the findings of this study are available from the corresponding author upon reasonable request.