Optimizing the Extraction Efficiency of Flaxseed Gum Using a Response Surface Methodology Approach

The extraction of gum from natural raw materials is of increasing importance in various industries, including food, pharmaceuticals, and cosmetics, particularly due to their emulsifying properties and potential applications as stabilizers and thickeners. This study presents an insight on the influence of changing parameters like reagents and operating condition on yield and some properties of the flax (Linum usitatissimum L.) seed gum. The extraction conditions were meticulously examined using a full factorial design, highlighting the significant impact of pretreatment, seed preparation, and solvent selection on the extraction yield. A response surface methodology (RSM) was then applied to optimize the water/benzoic acid ratio of the pretreatment step, the ethyl alcohol/water ratio, and the medium pH of the extraction method, resulting in a maximum yield of 14.47%. Furthermore, detailed analyses of the chemical and emulsifying properties of the gum were conducted showing emulsifying capacities over 94%, offering promising application prospects, particularly in the food industry.


Introduction
Hydrocolloids extracted from plant sources, especially polysaccharide gums, have become indispensable ingredients in the food, pharmaceutical, and cosmetic industries.Teir versatility as thickeners, stabilizers, gelling agents, and emulsifers in food products, their role in the manufacture of controlled-release drugs, and their use in cosmetic products make them valuable components.Teir appeal lies in their natural origin, availability, sustainability, and their ability to meet the growing consumer demand for natural and healthy ingredients [1].
Among the potential sources of these valuable gums, fax or linseed, from the botanical name Linum usitatissimum L. (Linnaeus, C., 1857), has attracted particular interest due to their abundance and rich polysaccharide composition.Flaxseeds, already widely recognized for their nutritional benefts because of their omega-3 fatty acids and dietary fber content, ofer a unique opportunity for the full utilization of plant-based raw materials [2].Teir polysaccharide content holds considerable potential for the production of highquality polysaccharide gums.Tis abundant and economically viable source contributes to addressing the increasing demand for high-quality products [3].
Beyond its physical attributes which can be used as a thickening and emulsifying agent in food and other industries like pharmaceutic and cosmetic, faxseed gum (FG) also known as faxseed or linseed mucilage ofers several health benefts [2].It can be utilized as a dietary fber source in human nutrition or act as a prebiotic, supporting the healthy growth of gut fora.In fermented dairy products, FG promotes the growth of lactic acid bacteria while positively infuencing the product's texture additionally to the antimicrobial properties against pathogenic bacteria and fungi [4].Moreover, in vivo studies have demonstrated the antiulcer activity of FG, reducing the number and length of gastric ulcers induced by ethanol in rats [5].Chemical compounds derived from faxseed exhibit antitumor properties due to their antioxidant activity, preventing the oxidation of proteins, lipids, or DNA, which are potential causes of cancer [6].
In Morocco, the faxseeds are generally used as ingredients for herbal teas or served as gourmet dishes.Lahsissene et al. mentioned in the catalogue of medicinal plants used in the region Zaër (Western Morocco) that the species L. usitatissimum is highly recommended against urinary tract infections in the Zaër region [7].For example, it is used at night in a mixture with other herbal teas, after decoction in water, as laxative and to treat constipation and urolithiasis [8,9].In a recent survey of medicinal plants used to treat hypercholesterolemia in Casablanca, it was found that the majority of herbalists frequently prescribe L. usitatissimum seeds, followed by Coriandrum sativum L., due to their higher use value.In addition, L. usitatissimum is reported to be one of the most marketed medicinal plants for its hypocholesterolemic properties, due to its richness in metabolites derived from the fatty acid, sterol, and lignan families [10].Te chosen extraction method is crucial to target the appropriate substances [4].According to the herbalists, a convergence was noticed in the rate of methods used to prepare the plant extract, from infusion in 33% of cases to decoction or as powder in 30.3% of cases each.In agreement, the majority of surveys have reported the wide use of infusion and decoction for traditional drug preparation [11].In popular pharmacopoeia, it is commonly prescribed as a laxative, emollient, and cough suppressant.Flaxseed is used to treat diabetes, heart disease, hypertension, and gastrointestinal and kidney disorders.A study on the efect of FG consumption in fasting subjects showed a lowering of total cholesterol (TC) and LDL-cholesterol, with an increase in fecal fat excretion [12].
Flaxseed carbohydrates are primarily concentrated in the hull, with a minimal 1-2% being digestible carbohydrate, mainly in the form of soluble sugar.Te majority of faxseed carbohydrates are indigestible, comprising both soluble and insoluble fber.FG located in the outermost layer of the hull accounts for approximately a quarter of the total carbohydrate in faxseed, making up about 7-10% of the total seed composition.Te insoluble part of faxseed carbohydrate is composed mainly of nonstarch polysaccharides, including cellulose and lignan [13,14].Structural studies reveal that FG consists of heterogeneous polysaccharides with a stif random coil structure [15].Tis mucilage is a complex mixture of polysaccharides, encompassing L-galactose, Dxylose, L-arabinose, L-rhamnose, and D-galacturonic acid, with a trace of D-glucose.Further division of FG reveals a neutral fraction (83%) and an acidic fraction (17%).Te neutral fraction includes arabinoxylans (56%) and galactoglucans (44%), while the acidic fraction contains pectinlike molecules composed of L-rhamnose, D-galactose, Lfucose, D-xylose, L-arabinose, and D-glucose [16,17].Reportedly, the extraction conditions of FG signifcantly afect its composition, particularly its monosaccharide and protein contents.At high-temperature extraction range (70-90 °C) an increase in the yield and the acidic polysaccharide is observed associated with the inactivation of microorganisms and enzymes, protein denaturation, and a decrease in the production of other compounds [18].
Nowadays, the extraction of FG presents a multifaceted challenge, requiring a balance between yield, purity, and costefectiveness.Te current state of the art in FG extraction techniques involves a spectrum of methodologies aimed at optimizing yield and quality [19].Conventional methods encompass solvent-based approaches, temperature variations, and pH adjustments, while more recent innovations explore enzymatic treatments or microwave-assisted extraction.Some studies focus on specifc aspects like chemical composition, while others strive for environmentally friendly processes.Despite the diversity of approach, common threads emerge among these techniques.Tey all seek to disrupt the structural integrity of the faxseed matrix to release the gum components efectively.Safdar et al. reported that the traditional hot water extraction technique (HWE), which is easy to control and requires no special equipment or conditions, achieved a higher yield (8.92%) at around 80 °C compared to other methods like ultrasound-assisted (UAE) and alkaline-acidic extraction (AAE).However, UAE is more conducive to FG purity [20].Nevertheless, these methods require a lot of solvent and other reagents despite their efciency in terms of yield and feasibility on a small scale.Te need for a costefective and efcient extraction method is crucial for the utilization of FG.
Statistical approaches are recommended to address the complexities of optimizing extraction conditions.Te response surface methodology (RSM) designs, specifcally central composite design (CDD), have proven to be invaluable tools in scientifc research in various felds, from chemistry to engineering when a more comprehensive exploration of the experimental space is required [21].However, the Box-Behnken design provides a balanced compromise between accuracy and practicality, particularly suitable for exploring complex relationships among continuous variables with a structured approach to identify optimal conditions while minimizing experimental trials, thus saving time and resources [22].Its three-level design provides precise insights into process responses based on multiple factors [23].
Tis study aims to expand our understanding of the impact of changing pretreatment and extraction conditions, as well as the nature of the solvents, on two conventional fax gum extraction processes in order to reduce the amount of solvents used while maximizing yield using response surface methodology (RSM).In addition, it assesses the chemical and emulsifying properties of the 2 Journal of Analytical Methods in Chemistry gums obtained, giving an insight into their potential applications in various industries.[24], using a water/benzoic acid mixture instead of ethanol because their ethanol-based dehulling procedure resulted in signifcant losses and caused the formation of two layers of gum in the subsequent extraction steps: an upper layer that is typically easier to extract and a lower layer containing oily impurities, which posed extraction challenges.To illustrate the pretreatment impact, two separate treatments were applied to diferent groups of seeds.One portion of the seeds (designated as pretreatment � "Yes") was soaked in a 3 : 1 (v/v) solution of water and benzoic acid (1N) at room temperature for 12 hours.Te oily components were subsequently removed by washing the seeds three times with a 1 : 3 (v/v) water/acetone solution.In contrast, the remaining portion (designated as pretreatment � "No") had the seeds steeped in water alone for 12 hours at room temperature before being cleaned with sterile water.Subsequently, all seeds were dried and stored until further use.

Seed Preparation.
Te seeds were either thoroughly crushed slightly with a hammer (intact), ground to a powder (blender), or whipped (glass beads).For the whipped procedure, the whole seeds were placed in a plastic container holding water at a 10 : 1 (v/w) ratio, along with a few glass beads, and briskly shaken before being let to stand for around 1 min.Te operation is done three times (with a 1 min cycle) until a gelatinous mass appears.It should be noted that no purifcation step was performed on the powders acquired from the seed preparation step, and this will be the case throughout the remainder of this study.

Solvents and Extraction Methods.
Two diferent procedures of separation on and splitting were chosen to isolate the gum with some modifcations while respecting the operating parameters such as time, temperature, and pH given by the authors.

Method A.
Method A is based on the procedure described by Kaushik et al., with some modifcations [25].
Te dried crushed/powdered or whipped seed sample of L. usitatissimum (50 g) was boiled for 2 h at 80 °C with continuous stirring into 1 L of deionized water or mixed with a solution of ethyl alcohol/water 1 : 1 (v/v), and the pH was adjusted to 7. Te extract was fltered through mousseline cloth and concentrated to approximatively 30% of the volume using rotary evaporator at 40 °C, and then the concentrated extract obtained was precipitated with 1 : 1 volume of ethanol 95% solution (v/v).Te precipitate was collected by centrifugation at 6000 rpm for 5 min followed by lyophilization.

Method B.
Method B that was used to extract gum from faxseed hulls was described by Qian et al., [26].Te sample (50 g) was dispersed into deionized water (700 mL) or ethyl alcohol/water at 70 °C for 3 h with stirring.Te recovered supernatant was then centrifuged at 4000 rpm for 15 minutes to remove insoluble residues.Tese residues were washed twice by 5 : 1 volume of deionized water, and the extraction process was repeated as before until the pomace was exhausted (2 to 3 times).Te supernatants were recovered using a vacuum pump, combined, were carefully poured into an acetone solution until no obvious gel precipitate appeared, and then left to settle for 12 h at room temperature.Ten the obtained precipitate was centrifuged and washed with deionized water three times.
After adding a 1:1 amount of ethanol 95% solution (v/v) to the mixture and repeating three washings, the crude precipitate was produced.Te resulting sample was then left to dry in Petri dishes at room temperature before being freeze-dried for later use.

Experimental Design and Statistical Analysis.
In order to investigate the impact of the selected variables on fax gum yield, it was decided for the rest of the study to work under high-temperature conditions and with the normal time that each method can take as mentioned in several works [27,28].

Full Factorial Design.
In order to investigate the impact of the pretreatment, the preliminary screening related to the preparation of the seed before use, and prepared the solvent used in the extraction procedures chosen.For this purpose, a completely randomized 3 × 2 × 2 × 2 factorial experimental design was applied in the frst stage to exhibit the infuence of the independent qualitative variables at diferent levels, while the yield (%) was considered as an experimental response for a total of 25 combinations [21].
Te selected factors are represented in Table 1.

Response Surface Methodology (RSM).
In the second stage of the study, a response surface methodology (RSM) was used twice to fnd the efect of continuous variables on the yield (%) and to predict a formula with highest responses as target.A Box-Behnken design with three center points was employed twice due to lack of ft on the frst attempt.
Tere were two levels for each independent factor: pH (4-10), water to pretreatment solvent ratio ( where M g is the mass (g) of extracted gum and M s is the mass (g) of L. usitatissimum seed.Subsequently, the moisture content was determined from weight loss after heating 0.5 g of the extracted gum for 24 h at 105 °C in an oven and the ash content was quantifed after dry mineralization (3 g) at 550 °C for 5 h in a mufe furnace according to the AOAC method [30].
Te total protein content was estimated using the Bradford method (1976) [31] by adding 0.5 ml of Bradford reagent (BRADFORD REAGENT-B6916 from Sigma-Aldrich) to 0.5 ml of sample.After a 30 min dark rest at room temperature, absorbance was measured at 595 nm (Spectrophotometer UV-Visible CPS-240A, Shimadzu) against a bovine serum albumin (BSA) standard range of 0 to 0.1 g/L.
Te total carbohydrate content of the polysaccharide hydrolyzate was determined according to phenol sulphuric acid method of Dubois et al. [32] by shaking 0.1 g of the samples with 1 mL of concentered H 2 SO 4 for 30 min in a water bath at 45 °C.Ten, 5% of phenol was added and the mixture was heated for 5 min at 90 °C.After cooling at room temperature, the absorbance was measured at 490 nm.A series of glucose concentration were used as standard.

Fourier-Transform Infrared Spectra (FT-IR) Analysis.
Te principal functional groups were detected by analysis of infrared spectra using a Bruker Tensor II FT-IR Spectrometer system (Bruker Optics GmbH).About 5 mg of purifed polysaccharide sample was mixed with 100 mg of anhydrous potassium bromide and pressed into a 13 mm disc.Te IR spectrum was recorded in the range 400 to 4,000 cm −1 with a resolution of 4 cm −1 .

Emulsifying Properties.
Oil-in-water emulsions were prepared by adding 5 ml of sunfower oil to 45 ml of hydrocolloid suspensions (60 mL) with a 0.5% (w/v) concentration previously prepared by dispersing the freeze-dried samples 2 h into distilled water with continuous stirring at room temperature until complete dissolution.Te mixture was stirred at 1200 rpm for 5 min and homogenized at 9800 rpm for 2 min using high shear homogenizer system (CAT M. Zipperer GmbH Drive motor Unidrive X 1000D, Germany) and then sonifed for 5 min by using an ultrasonic where e v is the emulsion volume and t v is the total volume.Te emulsion stability (ES) against high temperatures was determined by heating in a water bath at 80 °C for 30 min, followed by centrifugation at 3500 rpm for 5 min.Te emulsion stability was calculated as [34] ES (%) � where f ev is the fnal emulsion volume and i ev is the initial emulsion volume.

Full Factorial Design
3.1.1.Statistical Analysis.Te high coefcient of determination R 2 of 0.97 indicates that 97% of the variance in the yield can be explained by the model (Figure 1).Te analysis of variance for the least-squares ft of the extraction yield variation model (%) shows an F-ratio of 25.8136 which indicates signifcant variation in the model relative to the error, while the probability associated with the F-ratio (Prob.> F) is very close to zero ("<0.0001"),meaning that the model has a signifcant efect relative to the error with no lack of ft.

Efects of the Categorical Variables on the Extraction
Yield.Te signifcant diferent factors and interactions affecting the extraction yield (%) were investigated, and the summary is given in Figure 2; notably, "Seed preparation," "Pretreatment," and "Solvent" emerged as key factors with highly signifcant efects, all boasting p values well below the conventional threshold of 0.05.Tis signifes their substantial infuence on the dependent variable, underscoring the critical importance of seed preparation methods, pretreatment procedures, and solvent choices in our study.Furthermore, two interactions, "Seed preparation * Pretreatment" and "Seed preparation * Solvent," were also deemed statistically signifcant, suggesting that the impact of seed preparation is contingent on both pretreatment and solvent selection.Also, the Method * Solvent can be considered a small-scale efect interaction.However, the "Method" factor and the remaining interactions did not exhibit signifcant efects on the dependent variable, as indicated by their higher p values.Tese results contribute to a more nuanced understanding of the variables at play in our study, aiding in the refnement and fne-tuning of future experimental procedures.Figures 2(a)-2(e) show the comparisons between various factor levels using ANOVA and Hsu's MCB test, highlighting the maximum and minimum p values for each comparison.
Te thickness of the gray circles is proportional to the mean diference from the selected group (red circles), and the outside angle of intersection indicates whether the group means are signifcantly diferent as shown in Figure 2(e).
Te analysis of variance (ANOVA) for the "Seed preparation" factor, which has three levels, revealed signifcant results regarding its infuence on the dependent variable.Te data show a statistically signifcant diference among the sample preparation levels ("Intact," "Blender," and "Glass Beads").Te calculated F-ratio is 4.2041 with a p value of 0.0284, indicating that the choice of sample preparation method has a substantial impact on the studied variable.
Furthermore, comparisons suggest that the "glass beads" and "blender" levels, with p values of 0.7738 and 0.5380, respectively, do not have a signifcant efect compared to the best level ("Intact").However, the "intact" level shows a signifcant efect compared to the worst level in terms of the studied variable.Tis means that using "Glass Beads" or "Blender" for sample preparation has a diferent and signifcant efect compared to "Intact."Nevertheless, the choice between "Glass Beads" and "Blender" appears to have no signifcant impact on one another, but both signifcantly difer from "Intact." Similarly, the "Pretreatment" factor (p value � 0.0238) and the "Solvent" factor (p value � 0.0027) have efects less than 0.05, signifying signifcance on the dependent variable.Te comparison between "Yes" and "No" levels for the "Pretreatment" factor demonstrates a statistically signifcant diference, with a diference greater than −3.2974 at a 95% confdence level.Likewise, the comparison between "Ethanol" and "Water" levels of the "Extraction Solvent" factor shows a statistically signifcant diference greater than −3.6701 at a 95% confdence level.However, neither the analysis of variance (ANOVA) nor the direct comparison between "A" and "B" levels of the "Method" factor indicates a signifcant diference at the 95% confdence level.Consequently, the "Method" of separation does not have a signifcant efect on the dependent variable.
Following these results and due to availability of material and efciency, for the subsequent stages of the study, For the  Journal of Analytical Methods in Chemistry extraction of FG, we chose technique "A," which involves "pretreatment," "glass bead sample preparation," and "ethanol" usage as solvent.Tis approach will yield highquality results by optimizing the extraction conditions.

RSM Results.
In order to optimize the FG extraction process and identify the optimal circumstances for maximal output, our research was divided into two crucial stages.We carefully selected variables in the frst phase that had the greatest potential to enhance the extraction procedure.Te second phase involved precisely determining the optimal extraction conditions while accounting for the following variables: the medium pH of extraction technique A, the water/benzoic acid ratio of the pretreatment step, and the ethyl alcohol/water ratio.It is important to remember that diferent combinations of these factors could produce the best FG production.

6
Journal of Analytical Methods in Chemistry <0.0001 underscores the statistical robustness of our model, suggesting that the relationship between our independent variables and extraction yield is highly signifcant (Figure 3).

Efect of Diferent Variables on Yield. Sequential testing following the analysis of variance (ANOVA) results
shows that the factors water/benzoic acid ratio, ethyl alcohol/water ratio, pH, and some of their interactions have signifcant efects on the response (Yield %) (Table 3).Te water/benzoic acid ratio factor exhibits an F-ratio of 31.5662 with a probability (Prob.> F) of 0.0014, indicating a significant efect on the response (Yield %).Similarly, the ethyl alcohol/water ratio factor has an F-ratio of 20.1605 with a probability of 0.0041, signifying that its variation has a statistically signifcant impact on yield.However, the very high F-ratio of 181.2324 with an extremely low probability, less than 0.0001, indicates that the pH factor has an extremely signifcant efect on the response.Additionally, interactions between the water/benzoic acid ratio and ethyl alcohol/water ratio, water/benzoic acid ratio and pH, ethyl alcohol/water ratio and pH, and water/benzoic acid ratio * ethyl alcohol/water ratio have signifcant probabilities, revealing how these factors interact to infuence the response.
Te three 3D response surface curves (Figure 4) illustrate the impact of the water/benzoic acid ratio, ethyl alcohol/ water ratio, and pH factors, highlighting the complex interactions between the factors and their infuence on yield.In general, the observed curvilinear efect on these curves results from quadratic interactions.Te frst curve reveals a concave relationship between water/benzoic acid ratio and ethyl alcohol/water ratio concerning yield, emphasizing a signifcant efect on yield, as confrmed by sequential ANOVA testing.Te yield increases as the water/benzoic acid ratio increases, reaching a peak at around 3.262.However, this curve also indicates the importance of quadratic interactions, ethyl alcohol/water ratio * ethyl alcohol/ water ratio, with a probability of 0.0101.Likewise, the observed curvilinear efect on the second curve is due to the signifcance of the pH * pH quadratic efect on the response, with a very low probability (<0.0001) and an optimal point at around 3.001 for ethyl alcohol/water ratio, suggesting that yield is optimal at this value.However, this relationship is modulated by pH, with the maximum yield achieved at a pH of approximately 8.731.Furthermore, the third curve confrms the signifcant impact of pH on yield, with its concave shape and a peak, suggesting that yield depends on these two factors, and there exists an optimal combination of ethyl alcohol/water ratio and pH values that maximizes yield.

Optimization.
Te model we developed to predict the relationships between independent variables and the dependent variable, yield, is based on a complete quadratic equation.Tis equation accounts for interactions and quadratic efects of factors, providing precise modeling of the expected response.Te prediction equation for yield is as follows: where (1) X 1 (pH): the level of pH.
(2) (X 2 1 ): the square of the level of pH.(3) X 2 (water/benzoic acid ratio): the ratio of water to benzoic acid.(4) X 3 (ethyl alcohol/water ratio): the ratio of ethyl alcohol to water.(5) (X 23 ): the square of the ethyl alcohol/water ratio.(6) X 1 × X 2 : the interaction between pH and the water/ benzoic acid ratio.
(7) X 2 × X 3 : the interaction between the water/benzoic acid ratio and the ethyl alcohol/water ratio.
Te positives coefcients X 1 (1.2474) and X 3 (0.5252) suggest that increased pH levels and the ethyl alcohol/water ratio generally result in larger yields, whereas X 2 exhibits a negative infuence (−0.5206), implying that higher water to benzoic acid ratios lead to lower yields.Furthermore, it appears from the negative coefcients −0.3260 and −0.3566 for the interaction terms X 1 × X 2 and X 2 × X 3 that certain combinations of pH levels and water-to-benzoic acid ratios, as well as some combinations of water/benzoic acid and the ethyl alcohol/water ratios lead to lower yields.Underscoring the signifcance of optimizing these variables to enhance the extraction process and improve overall yield.
Utilizing this response surface model, we plan to achieve a maximize yield of 14.51% and minimize resource consumption by conducting experiments and analyzing key      8 Journal of Analytical Methods in Chemistry variables such as water/benzoic acid ratio at 3.2 to 1, ethyl alcohol/water ratio at 2.17 to 1, and pH at 8.7.Table 4 summarizes the predicted and experimentally validated values for the factors.
Results derived from the prediction model were experimentally validated, and the predicted values closely matched the experimental measurements.For instance, the predicted yield was computed as 14.51087%, while experimental measurements yielded 14.47% with a margin of error of ±0.14 after fve replicates.Te strong match between predicted values and experimental outcomes demonstrates our model's robustness and reliability.
By changing the extraction process conditions as provided by the optimization model, we can potentially achieve a yield that exceeds 14%, which is higher than that found by Hu et al. (12.73%) at higher temperatures (70 °C-90 °C), using more resources and requiring much longer periods of time [27].Te precision of predictions, coupled with the minimal diference between expected values and experimental measurements, suggests that our model can be a valuable tool for the FG extraction industry, enabling the optimization of production conditions to achieve high and consistent yields.

Characterization of Flaxseed Gum
3.4.1.Determination of Extraction Yield, Moisture, Ash, and Protein Contents.Te chemical compositions of faxseed gums obtained by the optimization process (OFS) in comparison with a fraction that underwent a purifcation step after washing and precipitation three times with isopropyl alcohol (PFS) are presented in Table 5. Te yield of the OFS was 14.47% of the seed weight, while the yield of the PFS was 12.44 equivalent to 85.97% of the unpurifed gum.Total sugars, moisture, and ash levels increased drastically after purifcation, while total nitrogen content decreased slightly from 51.02 mg/g to 50.74 mg/g of gum.
3.5.FT-IR Spectra Analysis.Te results of FT-IR spectroscopy reveal characteristic bands in the fngerprint region of gum, as reported in previous studies [3,35].A broad intense band at 3392 cm −1 is attributed to the stretching vibrations of hydroxyl groups (−OH), indicating the presence of hydrogen bonding in the structure [36].Another noteworthy band is the asymmetric stretching vibration at 2923 cm −1 , which is due to C-H stretching.Te band at 1716 cm −1 is consistent with the vibration of a nonconjugated carbonyl group (C�O).Furthermore, the bands observed between 1630 cm −1 and 1400 cm −1 suggest the absorption of carbonyl groups (C�O) originating from the carboxyl function of galacturonic acid, with an elongation vibration observed at 1373 cm −1 , as well as carboxylates (-COO-) corresponding to uronic acids [37].Bands located between 1151 cm −1 and 962 cm −1 indicate the presence of C-O and C-O-H bonds, suggesting the presence of glycosidic xyloglucan [38].
Additionally, the bands at 742 cm −1 and 675 cm −1 can be attributed to the anomeric confguration linkage units in the pure crystallinity of gum structure [39].
However, our fndings were slightly diferent from those reported by Ren et al.Despite showing typical polysaccharide absorption peak areas [18], specifc vibrations associated with the C-H bond were observed in the frequency range of 2900 cm −1 to 2800 cm −1 (Figure 5).While this may suggest the presence of benzyl groups (C 6 H 5 -CH 2 -), the presence of vibrations of the C�C bond around 1600 cm −1 , along with the absence of the characteristic N-H bending of the amide II band of proteins, precisely the absence of the double highfrequency stretching C�O bands and low-frequency N-H bending, favors the possibility of traces of benzoic acid resulting from the pretreatment step.Tis is in line with the low protein content previously found in the chemical composition of the OFS and PFS, demonstrating the success of the deproteinization process.Further purifcation may be required to confrm this hypothesis.

Emulsifying Properties.
Te emulsifcation properties of fax gum solutions were evaluated before and after purifcation treatment.As shown in Figure 6, both OFS and PFS had excellent emulsifying capacities (EC) well in excess of 94%, with mean values of 97.40 ± 0.49% and 94.05 ± 0.10%, respectively.Tis can be explained by the low partial denaturation of lipoproteins in the hydrocolloid chains, which prevents the formation of nanogels at pH 8 and consequently greater aggregation of emulsion droplets at the oil/water interface [40].Additionally, the pretreatment procedure may play a signifcant role in the release of particular bioactive components that efectively improve FG's apparent viscosity.Tis improvement is important for FG's rheological and functional properties, which direct its production and practical uses as a stabilizer, thickener, gelling agent, texture modifer, and suspending agent [24].
However, emulsifcation stability (ES) decreased drastically, particularly after isopropyl alcohol treatment, with value of 86.23 ± 0.68%, while 78.45 ± 0.35% for PFS (optimization process).A study of the steric and mechanical behavior of high molecular weight hydrophilic polysaccharide chains in galactomannans reported the formation of layers around oil droplets, preventing aggregation of the emulsifer [41].Te slight decrease in emulsifcation stability suggests that the enhanced level of polymerization products Journal of Analytical Methods in Chemistry inhibits the mobility of the molecules and consequently slows down absorption at the oil/water interface [27].

Conclusions
Tis study uses two hot water procedures to illustrate the critical necessity of extraction conditions for faxseed gum yield.Te results of the factorial design demonstrated that pretreatment, seed preparation, and solvent choice play a vital role in extraction yield.Furthermore, the use of a response surface methodology (RSM) focusing on continuous variables allowed us to determine the optimal parameters for maximizing gum yield, achieving a key value of 14.47%.Additionally, this study conducted an in-depth analysis of the chemical and emulsifying properties of the gum, ofering promising prospects for its use as an ingredient in various products, especially in the food industry.However, it is worth noting that the optimization of other parameters such as time and additional purifcation steps remains a key factor to meet the specifc requirements of certain industrial-scale applications.

Figure 1 :
Figure 1: Scatterplot of ftting model for the full factorial design.

Figure 3 :
Figure 3: Scatterplot of observed values versus predicted values for RSM.
y l A lc o h o l/ W a t e r r a t i o ( 1 ,3 ) E th y l A lc o h o l/ W at er ra ti o (1 ,3 ) W at er /B en zo ic A ci d ra ti o (3 ,5 ) W a te r / B e n z o ic A c id r a ti o ( 3 er /B en zo ic A ci d ra ti o (3 ,5 ) W a te r / B e n z o ic A c id r a ti o ( 3

5 E
th y l A lc o h o l/ W at er ra ti o (1 ,3 ) E th y l A lc o h o l/ W a te r r a ti o ( 1

Figure 4 :
Figure 4: Graphical representation of 3D response surface plots illustrating the interactive efects of key variables on yield (%) of faxseed gum.(a) Surface plot for variation in water/benzoic acid ratio and ethyl alcohol/water ratio.(b) Surface plot for variation in water/benzoic acid ratio and pH.(c) Surface plot for the variation in ethyl alcohol/water ratio and pH.

Table 2 :
Box-Behnken central composite design for the continuous variables.

Table 1 :
Te independent factors and their levels for the full factorial design.
[33] (model JP Selecta Ultrasons-HD, 3000867)[33].Te dispersions were then fnally centrifuged at 4000 rpm for 10 min.Te emulsifying capacity (EC) was calculated as EC (%) � e v t v   × 100, Model Fitting.Te response surface analysis indicates that the generated response surface curves accurately predict the extraction yield as a function of variations in the independent variables: water/benzoic acid ratio, ethyl alcohol/water ratio, and pH.Te results reveal a high-quality model with a high coefcient of determination R 2 of 0.98277, indicating that our model explains nearly 98% of the observed variation in yield.Te low Root Mean Square Error (RMSE) of 0.2621 strengthens the precision of our predictions.Furthermore, the extremely low p value of

Table 3 :
Analysis of variance (ANOVA) for the predictive model of extraction yield (%).

Table 4 :
Predicted and experimentally validated values for the factors.

Table 5 :
Chemical comparison of two fax gum solutions.
10 Journal of Analytical Methods in Chemistry A.O., and K.A. were responsible for formal analysis.R.A.M., O.M.N., and H.I. were responsible for resources.A.E. and H.I. were responsible for data curation.H.I., R.A.M., O.M.N., and M.T. were responsible for review and editing.M.O. was responsible for supervision.All authors have read and agreed to the published version of the manuscript.