Cassia grandis L.f. Pods as a Source of High-Value-Added Biomolecules: Optimization of Extraction Conditions and Extract Rheology

Abstract

High-value-added biomolecules such as phenolic compounds and flavonoids from secondary metabolism and macromolecules such as sugars are the main constituents of several plants. Thus, this work aims to optimize the extraction of these biomolecules present in the pods of Cassia grandis L.f. Initially, the effect of choline-based ionic liquids—ILs (choline chloride [Ch]Cl, dihydrogen citrate [Ch][DHC], and bitartrate [Ch][BIT]) as extracting agents for phenolic compounds and flavonoids was evaluated based on their efficiency and selectivity. Then, a 2³ full factorial design with six central points was performed using the IL concentration, the solid–liquid ratio, and the temperature as independent variables. The extract obtained in the best condition was subjected to pervaporation, after which the concentrates and the crude extract were used to determine the physical properties (density, viscosity, and refractive index). The hydrophobic–hydrophilic balance of the extracting agent and the biomolecules are the extraction process’s driving force. The best extraction condition was formed by [Ch][DHC] at 15 wt%, with a solid–liquid ratio of 1:15, at 45 °C for 30 min, resulting in 153.71 ± 5.81 mg·g⁻¹ of reducing sugars; 483.51 ± 13.10 mg·g⁻¹ of total sugars; 11.79 ± 0.54 mg·g⁻¹ of flavonoids; and 38.10 ± 2.90 mg·g⁻¹ of total phenolic compounds. All the physical properties of the biomolecules are temperature-dependent; for density and refractive index, a linear correlation is observed, while for viscosity, the correlation is exponential. Increasing the temperature decreases all properties, and the extract concentration for 8× presents the highest values of density (1.283 g·cm⁻³), viscosity (9224 mPa·s), and refractive index (1.467).

1. Introduction

The genus Cassia comprises more than 500 species worldwide and belongs to the Fabaceae family. The tree, which belongs to the Fabaceae family, has a height between 15 and 30 m. Their leaves are alternately formed by 15 to 20 pairs of opposite leaflets, and the in-florescence has more than 15 pink flowers [1]. The fruits (pods) reach up to 75 cm long and are rigid. Inside, lobes are found which are separated by transverse partitions, which contain seeds surrounded by a dark red flesh with a thick consistency, pungent odor, sweet taste, and high water solubility [2]. Due to its wide geographical distribution, rich phytochemical profile, and abundance of bioactive compounds, this plant species has become a promising target for pharmacological, medical, and industrial use [3,4].

The occurrence of C. grandis L.f. across tropical regions of Asia (India and China), East Africa, and America (Brazil, Colombia, Mexico, Cuba, El Salvador, Nicaragua, Costa Rica, and Honduras) suggests a high adaptability to diverse ecosystems, which can influence the production of secondary metabolites. Preliminary studies have already identified the presence of cardiac glycosides, flavonoids (105 mg quercetin equivalent per gram), alkaloids (45 mg RU·g⁻¹), phenolic compounds (169.73 mg gallic acid·g⁻¹), stilbenes, and anthraquinones, substances recognized for their biological activity [5]. This chemical diversity gives C. grandis L.f. multiple therapeutic applications, including antiparasitic, antimicrobial, and anti-inflammatory properties and cytotoxicity against cancer cell lines. Additionally, the aqueous extract of its sweet-tasting pulp is traditionally used to treat respiratory diseases, cardiac glycosides, gastric disorders, and skin infections, allowing us to call it a multipurpose plant [6]. Despite these empirical uses, there remains a lack of systematic studies that explore its rheological properties and thermodynamic stability.

Syrups rich in phenolic compounds and sugars from C. grandis L.f. are produced through evaporation and are used in the above-mentioned health applications. To the best of our knowledge, there are no existing studies on the behavior of syrups. However, the rheological behavior of syrups affects sugar refining, provides information on the physical properties of foods, and helps us understand the heat transfer mechanism for their production [7]. The wide variety of high-value compounds resulting from secondary metabolism, combined with macromolecules such as sugars, and their diverse application in treating various diseases put pressure on the scientific community to develop methodologies for extracting these compounds. Nevertheless, research in this field is still quite scarce [8,9].

The literature has been extensive in presenting different techniques for the extraction of biomolecules from plant matrices, highlighting traditional methods such as solid–liquid extraction, liquid–liquid extraction, and maceration, or unconventional techniques such as supercritical extraction, ultrasound-assisted extraction, and microwaves [10]. Conventional methods are simple to perform. However, they are not selective; in most cases, they use volatile solvents and require long processing times [11]. On the other hand, unconventional techniques require more sophisticated and expensive equipment. New approaches, such as ionic liquids, have been proposed as alternative, more sustainable solvents [12].

Ionic liquids (ILs) are salts in the liquid stage below 100 °C formed by organic cations (ammonium, sulfonium, imidazolium, etc.) combined with organic or inorganic anions (chloride, tosylate, acetate, dicyanamide, etc.) through relatively stronger ionic interactions between symmetrically packed ions [13]. The various combination possibilities allow the designation of ionic liquids as designer solvents since their physicochemical properties can be tuned for specific tasks [14]. The interesting properties of ILs include non-volatility, low vapor pressure, non-flammability, wide electrochemical windows, high thermal stability, and good electrical conductivity [15]. Choline (2-hydroxyethyltrimethylammonium)-based ionic liquids belong to the quaternary ammonium salts associated with different anions. Choline salts are biodegradable, relatively inexpensive, and water-soluble [16,17]. According to Ventura et al. [18], choline is a complex B vitamin widely used as a food additive. It presents ecotoxicity to Vibrio fischeri marine bacteria, which is practically harmless and moderately toxic. The use of Aliivibrio fischeri as a bioindicator is advantageous due to its natural bioluminescence, which enables rapid, sensitive, and cost-effective screening of bioactive compounds. Inhibition or modulation of its luminescence by plant extracts may indicate toxic, antimicrobial, or antioxidant properties [8,19].

This study aims to optimize the extraction of high-value-added biomolecules, including phenolic compounds, flavonoids, and sugars (reducing and total) from Cassia grandis L.f. pods using choline-based ionic liquids as sustainable and selective extracting agents. The extraction process was optimized through a full factorial experimental design, evaluating the effects of ionic liquid concentration, solid–liquid ratio, and temperature on the yield of the target compounds. Additionally, the physical and rheological properties of the extracts and their concentrated forms were characterized to assess their potential applications.

2. Materials and Methods

2.1. Materials

C. grandis L.f. pods were collected from different trees (5 at least) in public gardens in Salvador (13°00′08.4″ S and 38°51′81.1″ W). These trees are not part of a collection; however, they were confirmed by a botanist through a plaque informing the genus and species Cassia grandis L.f. The biomass was transported to the Federal University of Bahia laboratory, where the dirt was removed, and the material was sanitized using 20 mg L⁻¹ sodium hypochlorite. The pods were opened manually with a saw, and the internal contents (seeds, molasses-like material that coats the seeds and some of the woody material of the pod) were milled in a knife mill (Usifer, Farropilha, Brazil) and dried in a drying oven (Nova Ética, Vargem Grande Paulista, Brazil) at 55 °C. The particle size distribution profile was determined using a set of sieves of varying pore sizes (Figure S1 of the Supporting Information). Particle sizes between 0.25 and 0.85 mm were homogenized to perform the extraction processes.

The extraction was carried out using different ionic liquids based on choline, such as choline chloride—[Ch]Cl (≥98 wt%), choline bitartrate—[Ch][BIT] (≥98 wt%), and choline dihydrogen citrate—[Ch][DHC] (≥98 wt%), all purchased from Merck. The biomolecules determined in the aqueous extract of the extraction process were total phenolic compounds—TPCs (standard—gallic acid—≥98 wt%), flavonoids—FLA (standard—catechin—≥98 wt%), and total sugars (TSs) and reducing sugars (RSs) (standard—glucose—≥99 wt%), all purchased from Merck. All other chemicals were of analytical grade. The chemical structure of ionic liquids based on choline, as well as the standards, is shown in Figure 1.

2.2. Extraction Protocol

Initially, preliminary extractions were performed to determine the best extraction agent, using ionic liquids based on choline ([Ch]Cl, [Ch][BIT], and [Ch][DHC] under specified conditions such as 10 wt% ionic liquid, a solid–liquid ratio of 1:10, and a temperature of 35 °C, under constant magnetic agitation (Tecnal TE-085, Piracicaba, Brazil) at 100 rpm for 30 min. After choosing the extracting agent, optimization experiments for biomolecule extraction from C. grandis L.f. pods were performed using a full factorial design (FFD). The procedure consisted of 8 experimental runs and six central points. The variables were extracting agent concentration (X₁, wt%), solid–liquid ratio (X₂—g mL⁻¹), and temperature (X₃—°C). The values of the variables were established based on previous works of our research group [20,21].

The dried biomass was mixed with the extracting agent (test volume 10 mL) under experimental conditions, placed in 50 mL Erlenmeyer flasks, and conditioned in a bath (Quimis Q226M1, Diadema, Brazil) at the desired temperature for 30 min. The extract was centrifuged (Centrifuge Q 222T108, Quimis, Diadema, Brazil) at 2000 rpm for 10 min, and the biomass was discarded. Quantification of total phenolic compounds, flavonoids, and total and reducing sugars was immediately performed. The kinetics of the extraction process was determined under the optimized condition by varying the extraction time at intervals of 1, 5, 15, 30, 60, and 90 min.

2.3. Analytical Determination

Total phenolic compound (TPC) content was determined according to the method described by Swain and Hillis [20,22]. The absorbance was measured at 660 nm using a spectrophotometer (Shimadzu UV-3600 plus, Shimadzu, Kyoto, Japan). The TPC content was calculated using a standard curve (Equation (1)) based on gallic acid (0.25–0.50 mg mL⁻¹), and the results were expressed as the mean and standard deviation of mg gallic acid equivalent (GAE) g⁻¹ of biomass.

$$\mathrm{A}\mathrm{b}\mathrm{s}=\left(0.0075\times {\mathrm{C}}_{\mathrm{G}\mathrm{A}\mathrm{E}}-0.1242\right),\left({\mathrm{R}}^2=0.998\right)$$

(1)

where Abs corresponds to the absorbance at 660 nm, C_GAE is the concentration of gallic acid, and R² is the correlation coefficient.

Flavonoid (FLA) content was determined by the methodology described by Sheng et al. [23], which used a standard curve (Equation (2)) prepared from an aqueous solution of catechin (5–60 mg L⁻¹) as a standard to calculate the absorbance at 510 nm and, consequently, the FLA content. The results were expressed as the mean and standard deviation of mg catechin equivalents (CE) g⁻¹ of biomass.

$$\mathrm{A}\mathrm{b}\mathrm{s}=\left(0.0012\times {\mathrm{C}}_{\mathrm{F}\mathrm{L}\mathrm{V}}-0.0039\right),\left({\mathrm{R}}^2=0.991\right)$$

(2)

where Abs corresponds to the absorbance at 510 nm, C_FLV is the concentration of catechin, and R² is the correlation coefficient.

Reducing sugar (RS) was quantified using the 3,5-dinitrosalicylic acid (DNS) method proposed by Miller [24]. The absorbance was measured at 540 nm, and the results were calculated using a standard curve prepared with an aqueous solution of glucose (0.1–1.0 mg mL⁻¹), and the RS concentration was expressed as the mean and standard deviation of mg glucose equivalents (mg g⁻¹ of biomass). Total sugar (TS) was determined by acid hydrolysis (HCl 2.0 M) followed by neutralization with NaOH (2.0 M), and the glucose equivalent was determined as previously described.

2.4. Physical Properties

The C. grandis L.f. extracts (1x—brut extract) obtained under the best process conditions were roto-evaporated (Longen Scientific LSRE52CS-1, São Paulo, Brazil) under the conditions of 5 kPa at 45 °C until obtaining 2×, 4×, and 8× concentrated extracts and subjected to determinations of properties (density, viscosity, and refractive index).

Density (ρ) and dynamic viscosity ($\mathsf{\eta }$) data were simultaneously analyzed using Anton Paar SVM 3000 (Anton Paar, Graz, Austria) equipment based on an adapted Couette measuring principle. A sample of 5 mL was injected into the equipment using a glass syringe. Experimental measurements were performed at a temperature between 20 and 55 °C under atmospheric pressure (101.3 kPa). The density and viscosity measurements had a combined uncertainty of ±5 × 10⁻⁵ g·cm⁻³ and 0.01·$\mathsf{\eta }$, respectively. Refractive indices (nD) were also measured using an Anton Paar RXA170 (Anton Paar, Graz, Austria) refractometer at a temperature range from 20 to 55 °C under atmospheric pressure (101.3 kPa), with a combined uncertainty of ±2 × 10⁻⁵. The measurement was carried out in triplicate.

2.5. Statistical Analyses

All analytical determinations were conducted at least in triplicate. One-way ANOVA was used to analyze the experimental data, and the means were compared by the Tukey test using the SAS 9.0 software (SAS, Cary, NC, USA). Differences of p ≥ 0.05 were considered statistically significant.

Experimental data from a three-factor full factorial design (2³) were analyzed using the software Statistica 7.0 (StatSoft Inc., Tulsa, OK, USA). The experimental condition of the central point was repeated six times to determine the pure error. Linear and interaction effects of the three variables considered on the response variables were calculated. The independent variables (factors) were the ionic liquid concentration (X₁), solid–liquid ratio (X₂), and temperature (X₃). The dependent variables were total phenolic compounds (TPCs), flavonoids (FLA), total sugars (TSs), and reducing sugar (RSs) (Equation (3)). Pareto diagrams were constructed to evaluate the effects of the factors on each of the responses. Factors that were not statistically significant in the Pareto diagram were not considered in the regression model.

$$\mathrm{Y}={\mathsf{\beta }}_0+{\sum }_{\mathrm{i}=1}^{\mathrm{k}}{\mathsf{\beta }}_{\mathrm{i}}{\mathrm{X}}_{\mathrm{i}}+{\sum }_{\mathrm{i}\ne \mathrm{j}=1}^{\mathrm{k}}{\mathsf{\beta }}_{\mathrm{i}\mathrm{j}}{\mathrm{X}}_{\mathrm{i}}{\mathrm{X}}_{\mathrm{j}}$$

(3)

where Y is the response variable; X_i, and X_j are the independent variables; and k is the number of tested variables (k = 3). The regression coefficient is defined as β₀ for intercept, β_I for linear, and β_ij for interaction.

3. Results

3.1. Extraction of Biocompounds from C. grandis L.f.

Initially, the effect of the extracting agent was evaluated by comparing choline-based ionic liquids (at a concentration of 10 wt%, a solid–liquid ratio of 1:10, and a temperature of 35 °C for 30 min) with water as a conventional solvent (Figure 2 and Table S1 of Supporting Information).

Ionic liquids based on cholines have little influence on the extraction of macromolecules formed by carbohydrates (reducing and total sugars) when compared with values obtained for water. These molecules are known to be hydrophilic compounds and are quite soluble in water. Examples of sugars such as sucrose (disaccharide) and glucose (monosaccharide) have pronounced solubilities in water of 66.7% and 47.7% [25].

The decreasing order for the raw averages of sugar is water > [Ch]Cl (log K_ow = −3.70) > [Ch][BIT] (log K_ow = −1.43 > [Ch][DHC] (log K_ow = −1.32), which corroborates the decrease in the logarithm of the octanol–water coefficient (log K_ow); that is, more hydrophilic ionic liquids can remove a greater quantity of sugars. The statistical analysis of the averages using the Tukey test shows that the extraction using water, [Ch]Cl, and [Ch][Bit] as the extracting agent are statistically equal (p ≥ 0.05).

On the other hand, in extracting high-added value biomolecules such as flavonoids and total phenolic compounds, the order is reversed, and the most hydrophobic compounds are the best extracting agents. Phenolic compounds, including flavonoids, are formed by one or more aromatic rings linked to a hydroxyl group. Generally, low-molecular-weight phenolic compounds are water-soluble, while high-molecular-weight ones are insoluble [11]. This observation corroborates the extraction results; water-soluble compounds are extracted by water, while most water-insoluble ones are extracted by choline-based ionic liquids. The increase in hydrophobicity allowed greater extraction of phenolic compounds and flavonoids. Flavonoids and many phenolic compounds are commonly found in glycosylated forms, which means that they are chemically linked to one or more sugar fractions. This glycosylation significantly influences the polarity and solubility of the molecule. The glycosidic bond generally increases these compounds’ hydrophilicity (water solubility) compared to their aglycone (non-glycosylated) forms, which are more hydrophobic [26,27]. This amphiphilic nature (hydrophobic/hydrophilic balance) justifies the extraction results observed. The values differ statistically from each other (p ≥ 0.05). [Ch][DHC] was chosen to continue studies to determine the best conditions for extracting biomolecules from C. grandis L.f.

Table 1. Extraction of reducing and total sugars, flavonoids, and total phenolic compounds from C. grandis L.f. by [Ch]Cl for 30 min using the full factorial design (FFD).

Run	X1 (CIL, wt%)	X2 (RS/L, g mL−1)	X3 (T, °C)	RS (mg g−1)	TS (mg g−1)	FLA (mg g−1)	TPC (mg g−1)
1	−1 (5)	−1 (1:5)	−1 (25)	149.55	401.44	6.96	12.95
2	+1 (15)	−1 (1:5)	−1 (25)	127.46	427.73	11.67	22.57
3	−1 (5)	+1 (1:15)	−1 (25)	151.94	447.44	7.18	15.40
4	+1(15)	+1 (1:15)	−1 (25)	136.69	459.44	14.04	30.93
5	−1 (5)	−1 (1:5)	+1 (45)	157.06	504.40	8.74	16.48
6	+1 (15)	−1 (1:5)	+1 (45)	146.25	499.84	13.54	28.22
7	−1 (5)	+1 (1:15)	+1 (45)	153.34	457.49	10.89	24.05
8	+1 (15)	+1 (1:15)	+1 (45)	153.71	483.51	15.79	38.10
9	0 (10)	0 (1:10)	0 (35)	142.57	481.26	6.67	22.80
10	0 (10)	0 (1:10)	0 (35)	139.78	490.16	8.00	23.42
11	0 (10)	0 (1:10)	0 (35)	143.35	479.04	6.34	22.99
12	0 (10)	0 (1:10)	0 (35)	147.73	491.07	8.31	23.57
13	0 (10)	0 (1:10)	0 (35)	143.29	464.44	6.82	23.22
14	0 (10)	0 (1:10)	0 (35)	145.51	473.31	7.15	23.34

depicts the experimental results of 14 runs for RS, TS, FLA, and TPC and the coded levels of the three independent process variables (X₁—[Ch][DHC]; X₂—solid–liquid ratio; and X₃—temperature). This variable’s effect and interactions were studied using a 2³ full factorial design with six repetitions and a central point. The values of independent variables were fixed according to conditions of extraction found in the preliminary experiments for different biomass carried out in our laboratory.

The biomolecule concentration ranged from 127.46 to 157.06 mg g⁻¹ (RS); 401.44–504.40 mg g⁻¹ (TS); 6.34–14.04 mg g⁻¹ (FLA); and 12.95–38.10 mg g⁻¹ (TPC). Overall, the increase in the values of the independent variables promotes the increase in the values of high-value-added biomolecules such as FLA and TPH. However, this observation is not valid for extracting RS and TS. This variation depends on the effects of positive or negative interaction between the variables aiming to extract biomolecules. The experimental results are shown in the response surface in Figure 3.

The increase in temperature during the extraction process enhances the permeability of cell membranes, thereby facilitating the diffusion of biomolecules into the extracting liquid. Furthermore, the solubility of these biomolecules is improved, which explains the higher extraction yields observed at elevated temperatures [28].

The significance of each variable was determined by F-test and p-values and is presented through the Pareto diagram (Figure 4) and ANOVA Tables S2–S5 of the Supporting Information). According to Atkinson and Donev [29], the independent variables are significant when they reflect higher F and smaller p-values. Therefore, each experimental set has different significant variables such as RS (C_IL; T and the interaction between C_iL vs. T); TS (T and interaction between R_S/L vs. T); FLA (R_R/L and T), and TPC (C_IL, R_S/L, T, and interaction between C_IL vs. R_S/L, and R_S/L vs. T). The results suggested that the independent variables were well chosen for the study of the extraction process since they are statistically significant (p ≥ 0.05). The curvature of curves was significant (p ≤ 0.05) for the extraction of TS and FLA.

According to the 2³ full factorial design, the data could be expressed by the predicted linear equation. The equations are presented as coded values as shown in

Table 2. Predicted equations and regression coefficients for the dependent variables obtained from the 2³ full factorial design (coded value).

Variable	Equation	R2
Reducing Sugar—RS	RS = 147.00 − 5.97X1 + 1.92X2 + 5.59X3 + 2.25X1X2 + 3.36X1X3 − 0.99X2X3	0.9502
Total Sugar—TS	TS = 460.16 + 7.47X1 + 1.81X2 + 26.15X3 + 2.04X1X2 − 2.10X1X3 − 17.62X2X3	0.9276
Flavonoids—FLA	FLA = 11.10 + 2.66X1 + 0.87X2 + 1.14X3 + 0.28X1X2 − 0.23X1X3 + 0.23X2X3	0.9726
Total Phenolic Compounds—TPCs	TPC = 23.59 + 6.37X1 + 3.53X2 + 3.13X3 + 1.03X1X2 + 0.08X1X3 + 0.83X2X3	0.9915

. The predictive equations show good agreement due to the high regression coefficients (R²), which range from 0.9502 to 0.9915, which are statistically acceptable at the 95% significance level (p ≥ 0.05).

According to Patel et al. [30], predictive models permit the theoretical determination of the optimal conditions. However, the predate model found in our work did not present an optimal point. In this case, the alternative would be to search for the optimal path and then use a new experimental matrix. However, the limitations of the independent variables prevent this procedure. An increase in the concentration of [Ch][DHC] raises process costs due to the market value of this extracting agent; an increase in the solid–liquid ratio makes it difficult to concentrate the biomolecule due to the elimination of water, which will require the use of rotary evaporation. Finally, phenolic compounds are thermolabile, and temperatures above 50 °C result in the loss of biological activity of these biomolecules [20,31]. Therefore, determining the optimal path and using axial points are not feasible. The maximum concentrations were observed at 5 wt% [Ch][DHC], 1:5 (R_S/L, g mL⁻¹), and 45 °C for RS (157.06 mg g⁻¹) and TS (504.40 mg g⁻¹); 15 wt% [Ch][DHC], 1:15 (R_S/L, g mL⁻¹), and 25 °C for FLA (14.04 mg g⁻¹); and 15 wt% [Ch][DHC], 1:5 (R_S/L, g mL⁻¹), and 45 °C for TPC (38.10 mg g⁻¹).

Although the optimal conditions for each biomolecule are obtained separately, it is important to highlight that the extraction process is unique, and the concentrations of the biomolecules are determined together in each experiment. Therefore, the decision-making process is to choose the extraction conditions with the best performance. The extraction kinetics of high-value-added compounds was studied using [Ch][DHC] 15 wt%, solid–liquid ratio 1:15 at 45 °C (between 1 and 90 min (Figure 5 and Table S6 of Supporting Information). The extraction of high-value-added biomolecules reached a steady state within 30 min, except for TS, which peaked at 1 min. After this time, the values of each biomolecule’s concentration were statistically similar at 95% confidence.

3.2. Physical Properties of C. grandis L.f. Extract and Its Concentrates

After determining the best conditions for extracting biomolecules from C. grandis L.f. ([Ch][DHC] 15 wt%, solid–liquid ratio 1:15 at 45 °C for 30 min), the extract was concentrated by rotary evaporation under 5 kPa at 45 °C, in order to obtain different concentrated extracts (2×, 4×, and 8× concentrated). The physical properties were measured for each extract. Figure 6 Tables S7, S8, and S9 of Supporting Information) depicts the physical properties of the C. grandis L.f. extract and its concentrates obtained using [Ch][DHC] 15 wt%, solid–liquid ratio 1:15 at 45 °C.

Determining physical properties plays an important role in understanding the design and operation of processes. Therefore, the properties (density, viscosity, and refraction index) were measured between 20 and 60 °C and at 0.1 MPa due to the thermal sensitivity of the biomolecules in the extract.

The correlation between the density of the C. grandis L.f. extract and its concentrates with temperature is linear. Increasing temperature decreases the density from 1.283 g·cm⁻³ (20 °C for the 8× concentrated extract) to 1.252 g cm⁻³ (20 °C for the crude extract), which is due to the higher kinetic energies of biomolecules present in the extract [32]. The linear correlation is associated with Equation (4).

$$\mathsf{\rho }=\mathrm{a}-\mathrm{b}·\mathrm{T}$$

(4)

where ρ is the viscosity (g cm⁻³), T is the temperature (K), and the indices a and b are the linear and angular coefficients.

The viscosity of the C. grandis L.f. extract and its concentrates follows a non-linear relationship. Increasing temperature leads to an exponential decrease in viscosity. Viscosity values range from 0.811 mPa·s (60 °C for the crude extract) to 9224.8 mPa·s (20 °C for the 8× concentrated extract). The behavior is expressed by Equation (5).

$$\mathrm{\eta }={\mathrm{\eta }}_0·{\mathrm{e}}^{-{\displaystyle \frac{{\mathrm{E}}_{\eta }}{\mathrm{R}\mathrm{T}}}}$$

(5)

where η is the viscosity (mPa·s), η₀ is the initial viscosity (mPa·s), E_η (Kj.mol⁻¹) is the activation energy, R is the ideal gas constant (J mol⁻¹K⁻¹), and T is the temperature (K).

The viscosity of the extracts is lower than that observed with pure water (100 mPa·s at 25 °C), except for the values found for the 8× concentrated extracts at all temperatures. Gabsi et al. [7] attributes this to the progressive disentanglement of long-chain molecule arrangements that helps overcome intermolecular resistance to flow.

The increase in temperature also leads to a decrease in the refractive index since this increase in temperature increases the internal energy of the solvent [33,34]. This observation was also reported by Tan et al. [35] in aqueous solutions with different solutes. The values varied from 1.3535 (60 °C for the crude extract) to 1.4672 (20 °C for the 8× concentrated extract). The trend follows Equation (6).

$${\mathrm{n}}_{\mathrm{D}}=\mathrm{c}+\mathrm{d}·\mathrm{T}$$

(6)

where $\mathrm{n}$_D is the (dimensionless) refractive index, T is the temperature, and c and d are linear regression parameters that correspond to the linear and angular coefficients, respectively.

Table 3. Regression parameters for the physical properties of the C. grandis L.f. extracts and their concentrates obtained using [Ch][DHC] 15 wt%, solid–liquid ratio 1:15 at 45 °C for 30 min.

$\mathbf{Density}\text{—}\mathbf{\rho }$
Sample	a	b	R2
Crude	1.230 ± 0.008	6.0 × 10−4 ± 2.0 × 10−5	0.9863
Concentrated 2×	1.283 ± 0.005	6.0 × 10−4 ± 1.5 × 10−5	0.9957
Concentrated 4×	1.386 ± 0.003	7.0 × 10−4 ± 1.0 × 10−5	0.9983
Concentrated 8×	1.510 ± 0.003	8.0 × 10−4 ± 1.1 × 10−5	0.9987
Viscosity—η
Sample	η0	Eη	R2
Crude	1.1 × 10−4 ± 8.8 × 10−5	1.8 × 105 ± 2.1 × 102	0.9991
Concentrated 2×	9.0 × 10−4 ± 1.0 × 10−4	2.0 × 105 ± 3.9 × 102	0.9977
Concentrated 4×	1.0 × 10−4 ± 2.5 × 10−5	3.1 × 105 ± 4.8 × 102	0.9987
Concentrated 8×	7.6 × 10−7 ± 8.8 × 10−8	5.6 × 105 ± 2.8 × 102	0.9999
Refraction Index—nD
Sample	c	d	R2
Crude	1.394 ± 0.002	1.0 × 10−4 ± 5.4 × 10−6	0.9861
Concentrated 2×	1.451 ± 0.001	2.0 × 10−4 ± 3.9 × 10−6	0.9980
Concentrated 4×	1.498 ± 0.002	3.0 × 10−4 ± 8.0 × 10−6	0.9933
Concentrated 8×	1.548 ± 0.001	3.0 × 10−4 ± 1.5 × 10−6	0.9998

presents the values of the adjustment parameters for the physical properties’ models. These models fit the experimental data well since the correlation coefficients are higher than 0.9861.

4. Conclusions

Experimental data for the high-value-added biomolecules (flavonoids and total phenolic compounds) and macromolecules (reducing and total sugars) have been reported, and 2³ full factorial designs evaluated the best condition. The results showed that best extraction of reducing sugars (153.71 ± 5.81 mg·g⁻¹), total sugars (483.51 ± 13.10 mg·g⁻¹), flavonoids (11.79 ± 0.54 mg·g⁻¹), and total phenolic compounds (38.10 ± 2.90 mg·g⁻¹) was achieved using [Ch][DHC] at 15 wt%, with a solid–liquid ratio of 1:15, at 45 °C for 30 min. The interaction between the extracting agent and biomolecules expressed by their hydrophobic–hydrophilic characteristics plays an important role in the extraction process. The kinetics of extraction reached the steady state at 30 min. The crude and concentrate extracts present physical properties depending on the temperature. A linear correlation was observed for density and refraction index, while viscosity had an exponential dependence. Increasing the temperature implies decreased physical properties for all the studied extract concentrations. Thus, extracts from Cassia grandis L.f. pods are promising sources of bioactive compounds with potential applications in pharmaceutical, cosmetic, and food products, adding value to a widely available yet underexplored biomass. Moreover, the selection of this plant species is justified not only by its broad distribution across tropical regions of Latin America but also by its traditional use in folk medicine, where it is recognized for its laxative, antimicrobial, and antioxidant properties. Additionally, the pods were chosen because they represent biomass often discarded after the plant’s reproductive cycle, making them a sustainable and low-cost alternative for obtaining bioactive compounds.