Tailoring Rheological, Viscoelastic, and Starch Structural Properties in Plant-Based Beverages via Homolactic Fermentation of Quinoa and Chickpea Flour Blends

Abstract

This study investigated the effects of homolactic fermentation on the rheological, viscoelastic, and starch structural properties of quinoa–chickpea flour-based beverages. Three formulations with increasing proportions of chickpea flour (10, 25, and 50%) were fermented for 10 h with Lactobacillus acidophilus LA-5. Apparent viscosity, deformation capacity, storage modulus (G′), and pasting behavior were measured along with FTIR-based analysis of the starch molecular structure. All fermented samples reached pH values < 4.5 and exhibited improved rheological properties with significant increases in viscosity and storage modulus (G′), particularly in the 50:50 blend. These enhancements were attributed to the synergistic effects of homolactic fermentation and inherent properties of chickpea starch, particularly its high amylose content, large granule size, and long amylopectin chains. FTIR analysis revealed that the short-range molecular order of starches was preserved after fermentation in all beverages, except for the 50:50 blend, as evidenced by the increased degree of order (DO) and double helix (DD) ratios. Overall, these findings demonstrate that integrating chickpea flour and controlled homolactic fermentation is an effective strategy for tailoring the viscosity and stability of plant-based probiotic beverages, providing a theoretical basis for the development of clean-label and functional fermented plant-based systems.

1. Introduction

The consumption of plant-based products has steadily increased in recent years, especially among people who are aware of the relationship between nutrition and health [1]. Among these products, plant-based beverages (PBB) have presented a significant increase in the last few years [2]. PBBs have gained interest from the scientific community and food industry because they are a good alternative for lactose-intolerant people, those with milk allergies, or people who follow a vegan or vegetarian lifestyle [3,4]. Furthermore, concerns regarding animal welfare and the carbon footprint associated with milk production have influenced many individuals’ decisions to avoid dairy product consumption and to prefer plant-based alternatives [5,6].

PBBs are colloidal suspensions of dispersed plant-based materials such as cereals, legumes, pseudocereals, and nuts [2]. However, these beverages face technological challenges related to their poor physical properties, such as low viscosity, which promotes phase separation and can directly affect their stability [7,8,9]. The composition and physical properties of PBBs depend largely on the botanical source used. Quinoa (Chenopodium quinoa) is rich in starch (55–65%) and contributes to the viscosity of the prepared PBB, while chickpeas (Cicer arietinum) offer high protein solubility, water-holding capacity (WHC), and emulsifying ability [7,10,11,12,13]. Blending quinoa and chickpea flours may improve beverage functionality by enhancing interactions among polysaccharides, proteins, and lipids [10,11], aligning with clean label demands that avoid the addition of synthetic stabilizers [7]. In addition, an interesting process that can be explored to improve the physical characteristics of PBBs is fermentation by lactic acid bacteria (LAB). LAB can metabolize different carbon sources, such as starch, raffinose, and sucrose, and transform them into lactic acid [14]. The fermentation process plays a crucial role in the breakdown of macromolecules due to bacterial metabolism, including the production of amylases (α-amylase, β-amylase, amyloglucosidase, and pullulanase/amylopullulanase) and proteases, which are essential for modifying starch and protein structures in plant-based sources [14,15,16,17,18,19]. This may positively affect and improve the physical characteristics of PBBs. Some studies have reported that fermentation can increase the viscosity (>5%) of lupin-based beverages [18]. Fermentation increases starch granule porosity, partially hydrolyzes amorphous regions, and reduces starch molecular weight, improving the physical properties of PBBs [14]. It also decreases surface proteins, enhances granule swelling, and increases viscosity in fermented PBBs [19].

Quinoa and chickpea flours as substrates for fermented beverages have been explored separately (either quinoa or chickpeas alone) [20,21], or as blends with other plant-based sources, such as quinoa and soybean [10], and chickpea mixed with coconut [11]. Additionally, the effects of quinoa and chickpea flour blends on the stability and physicochemical properties of fermented plant-based beverages during storage have been reported [22]. However, the effect of liquid-state fermentation on rheological characteristics when blends of quinoa and chickpea are used, to our knowledge, has not been well documented.

Accordingly, this study aimed to investigate the effect of homolactic fermentation and chickpea flour incorporation on the rheological properties of quinoa-based beverages. Quinoa and chickpea flours were blended in different ratios and fermented with Lactobacillus acidophilus LA-5. The effects of fermentation on the rheological and physical properties of the beverages, including the water-holding capacity, apparent viscosity, deformation capacity under stress, and viscoelastic behavior, were evaluated before and after fermentation. Additionally, structural modifications of starch were evaluated by FTIR spectroscopy, and the thermal properties of the flours and pasting behavior of the raw mixtures and fermented beverages were also studied. This study addresses the current knowledge gap regarding the combined use of lactic acid fermentation and legume flour addition as an integrated strategy to modulate the rheological behavior and short-range starch order of quinoa-based probiotic beverages. By integrating thermal, rheological, and FTIR analyses that are not commonly applied to plant-based beverage systems, this work offers a comprehensive view of pH changes and LAB growth, including the production of metabolites such as lactic acid in plant-based systems. These findings provide a theoretical foundation for the design and optimization of processing conditions of stable, functional, and clean-label plant-based beverages.

2. Materials and Methods

2.1. Plant-Based Sources

Quinoa seeds (Chenopodium quinoa Willd.) were obtained from Quinoa Fundo San Jose de Cáhuil (Cáhuil, O’Higgins Region, Chile) at 34°28′44″ S, 72°00′14″ W. Chickpea grains (Cicer arietinum L.) were purchased from a local supermarket (Santiago, Metropolitan Region, Chile) at 33°26′29.4″ S, 70°42′45.6″ W.

2.2. Flours Preparation

Quinoa and chickpea flours were prepared as previously described by Hurtado-Murillo et al. (2024, 2025) [22,23]. Quinoa seeds were washed with distilled water at 1:5 (weight: volume). The washed seeds were dried for 24 h at 40 ± 2 °C in a natural convection incubator (IN55; Memmert, Schwabach, Germany). Subsequently, the dried quinoa seeds were milled using a laboratory-scale cross-beater mill stainless-steel grinding insert (Pulverisette 16, Fritsch, Germany) and sifted through a <250 μm sieve. Dehulled chickpea grains were used for flour preparation, following the same grinding procedure. The resulting quinoa and chickpea flours were packed in airtight bags and stored at 4 ± 1 °C until further use.

2.3. Differential Scanning Calorimetry Analysis

Differential scanning calorimetry (DSC) analysis was performed according to Contardo et al. (2020) [24], with some modifications, using a DSC 1 STAR System (Mettler-Toledo, Greinfensee, Switzerland). One gram of flour was hydrated with 40 mL of deionized water for 1 h. Subsequently, 40 mg of the hydrated sample was placed in an aluminum DSC pan (100 μL). The pans were hermetically sealed, and an empty pan was used as a reference. The samples were scanned from 25 °C to 90 °C at a heating rate of 10 °C/min. The onset (To), peak (Tp), and endset (Te) gelatinization temperatures were calculated using the STARe Software (DB V12.10, Mettler-Toledo, Greinfensee, Switzerland). Each sample was examined in triplicate

2.4. Pasting Properties of Quinoa and Chickpea Flour

The pasting properties of quinoa flour (QF) and chickpea flour (CF) were determined using Rapid-Visco-Analysis (RVA 4500, Perten Instruments, New South Wales, Australia) according to the method described by Huang et al. (2021) [25], with some modifications. Briefly, 3 g of the flour sample was weighed into aluminum canisters, and 21 mL of distilled water was added. The canister was then placed in the instrument, and viscosity patterns were obtained as a function of temperature held at 25 °C for 2 min, heated between 25 °C and 95 °C at 13.5 °C/min, held at 95 °C for 5 min, cooled to 25 °C at 13.5 °C/min, and held at 25 °C for 1 min. The analysis was performed with constant stirring (100 rpm). The pasting parameters evaluated were (1) peak viscosity (maximum paste viscosity achieved during the heating phase of the profile), (2) breakdown viscosity (difference between peak and hold viscosity), (3) setback viscosity (difference between final and hold viscosity), and (4) final viscosity (viscosity at the end of the run). All measurements were performed at least in triplicate.

2.5. Preparation of the Plant-Based Beverages

Beverage preparation has been described in previous studies [22]. The flours were mixed at a solid-to-water ratio of 1:7 (weight: volume), and three different proportions of QF and CF were used to produce the beverages: QF (90%) was mixed with 10% CF (QF90-CF10), 75% QF with 25% CF (QF75-CF25), and 50% QF with 50% CF (QF50-CF50) (w/w). These extracts were pasteurized at 75 ± 2 °C for 15 min and homogenized at 14,000 rpm using an OV5 homogenizer (OV5 homogenizer; VELP Scientific, Inc., Deer Park, NY, USA). The bottles and containers used for the PBB preparation were dried and autoclaved at 121 °C for 15 min.

2.6. Fermentation Process

Freeze-dried probiotic cultures of L. acidophilus LA-5 (Chr.Hansen, Hørsholm, Denmark) were grown in de Man Rogosa and Sharpe broth (MRS; Condolab, Madrid, Spain). The MRS broth was incubated at 38 ± 2 °C under anaerobic conditions for 24 h. A sample (1 mL) of MRS broth was cultured in MRS agar at 38 ± 2 °C for 24 h to isolate the LAB. Overnight cultures of L. acidophilus LA-5 were inoculated into plant-based beverages at a concentration of 1 × 10⁶ CFU/mL (10% v/v). The blends were fermented at 38 ± 1 °C and 100 rpm under anaerobic conditions in an incubating orbital shaker (SI500, Richmond Scientific, Chorley, UK) for 10 h until the pH of each beverage reached less than 4.5 units.

2.7. Physicochemical Analysis of the Fermentation Process

Physicochemical parameters for determining the fermentation process included pH, titratable acidity (TTA), viable bacterial count, and °Brix values. The pH of the samples (5 mL) was directly measured using a potentiometer (HI5521-02, Hanna Instruments, Smithfield, RI, USA). For TTA determination, 5 mL of the samples were diluted in 45 mL of deionized water and titrated with 0.1 N NaOH solution, using phenolphthalein (0.1% w/v in 95% ethanol) as an indicator until the pH reached pH 8.3 [20]. The TTA, expressed as the acid lactic constant (0.090), was calculated as follows:

$$\mathrm{TTA} \left(\%\right)= {\displaystyle \frac{ \left(\mathrm{Volume} \mathrm{of} \mathrm{NaOH} \left(\mathrm{mL}\right)\times \mathrm{Normality} \mathrm{of} \mathrm{NaOH} \times 0.90\right)}{\mathrm{Volume} \mathrm{of} \mathrm{the} \mathrm{sample} \left(\mathrm{mL}\right)}} \times 100$$

(1)

Glucose/fructose and lactate concentrations were measured using the Y15 photometric analyzer (Biosystems, Barcelona, Spain), according to the manufacturer’s instructions. Samples (1 mL) of each beverage were centrifuged at 15,000× rpm for 5 min at 8 °C, and 500 μL of the supernatant was used for the measurements. The hexose determination kit used measures the combined concentration of glucose and fructose. Therefore, the lactic acid yield was calculated as the ratio of lactic acid produced (g/L) to hexoses consumed (g/L), based on the difference between initial and residual glucose/fructose concentrations. This approach reflects the actual conversion efficiency during fermentation, rather than the theoretical yield based on the initial sugar content. The °Brix values of the samples were evaluated using a refractometer (HI96800, Hanna Instruments, Smithfield, RI, USA) calibrated with deionized water. The measurement involved placing a drop of the sample on the lens of the refractometer. Viable bacterial counts were determined following the method described by Qiu et al. (2023), with some modifications [26]. Beverage samples (10 mL) were serially diluted (1:10 v/v) using buffered peptone water (1% w/v) (Condolab, Madrid, Spain). Next, 1 mL of each diluted sample was spread-plated onto MRS agar and incubated for 72 h at 38 ± 2 °C under anaerobic conditions. MRS plates containing 25–250 LAB colonies were used to determine bacterial concentration. The results were expressed as CFU/mL.

2.8. Physical Properties of Plant-Based Beverages

2.8.1. Determination of the Water-Holding Capacity

The water-holding capacity (WHC) of the plant-based beverages was analyzed according to Xu et al. (2022) [27]. Falcon tubes containing approximately 50 mL of the sample at 8 ± 2 °C were centrifuged at 5000× rpm for 30 min at 4 ± 2 °C in a Hettich centrifuge (Universal 320 model/Andreas Hettich GmbH & Co. KG, Föhrenstr. 12, D–78532, Tuttlingen, Germany). After centrifugation, the supernatant was removed, and the resulting serum was accurately weighed. WHC was calculated using the following formula:

$$WHC \left(\%\right)= {\displaystyle \frac{m2}{m1}} \times 100$$

(2)

where m₁ is the total weight of the sample and m₂ represents the weight of serum produced after centrifugation. All analyses were performed at least in triplicate.

2.8.2. Rheological Measurements of Plant-Based Beverages

The rheological properties of the samples were measured using a rheometer (Discovery HR2, TA Instruments, New Castle, DE, USA) equipped with a flat parallel plate geometry (stainless steel, 50 mm diameter, 1000 µm gap), following the method described by Hurtado-Murillo et al. (2024) [22]. The samples were carefully placed on a plate and covered with a solvent trap to maintain temperature. The TRIOS software package version 5.1.1 (TA Instruments, New Castle, DE, USA) was used to control the equipment and acquire the rheological parameters. The steady-shear flow measurements were taken at 8 °C in a 0.1–150 s⁻¹ shear rate range. The linear viscoelastic range (LVR) values for the samples were obtained from the plot of the elastic modulus (G′) Vs. Oscillatory strain (%) under oscillatory conditions at 1 Hz and from 0.01 to 20%.

In addition, the viscoelastic behavior of the samples was measured in a frequency sweep test, where the temperature was maintained at 8 °C, and the moduli (G′, G″) response to increasing frequency (0.1 to 100 Hz) at a strain of 0.2% within the LVR was measured. All samples were analyzed before and after fermentation.

2.9. Fourier Transform-Infrared Spectroscopy (FTIR)

FTIR analyses of the samples were performed using a Spectrum One ATR-FTIR spectrometer (PerkinElmer, Waltham, MA, USA). For each measurement, approximately 1 g of freeze-dried sample was compressed into a supporting disc using a Quick Press Hand (PerkinElmer, Waltham, MA, USA) to ensure consistent contact with the ATR diamond crystal. The samples were analyzed at room temperature over a spectral range of 4000 to 400 cm⁻¹, with a spectral resolution of 4 cm⁻¹ and an average of 32 scans per sample. The obtained spectra within the range of 1200–800 cm⁻¹ were processed using the OMNIC 9 software, including baseline correction, ATR correction, deconvolution, and normalization. FTIR was used to evaluate the changes in the short-range-ordered structures of starches in the 995–1047 cm⁻¹ region using second-derivative spectra to enhance resolution [28]. The degree of order (DO; 1047/1022) and degree of double helix (DD; 995/1022) ratios of the deconvoluted spectra were determined for each sample by measuring the amplitudes of the bands located at 1047 cm⁻¹, 1022 cm⁻¹, and 995 cm⁻¹, respectively. Characteristic IR bands were identified in both unfermented and fermented beverages [29].

2.10. Statistical Analysis

All treatments and analyses were performed in triplicate in two independent assays. The results are reported as mean ± standard deviation. Statistical analyses were performed for Windows using Statgraphics Centurion XIX software (Manugistics, Inc., Rockville, ML, USA). To demonstrate normality, the Shapiro–Wilk test was applied. Statistical significance was tested using a one-way analysis of variance (ANOVA). Tukey’s HSD was used to perform multiple comparisons. Differences were considered significant at the 95% confidence level (p < 0.05).

3. Results and Discussion

3.1. Thermal Properties of Quinoa and Chickpea Flour

Starch gelatinization is a crucial parameter that significantly influences the rheological and physical properties of PBBs. Differential Scanning Calorimetry (DSC) was used to analyze the thermal properties of the flours, and the results are presented in

Table 1. Thermal properties of quinoa and chickpea flours and pasting properties of different quinoa-to-chickpea flour ratios from RVA.

Thermal Properties (°C)	Quinoa Flour		Chickpea Flour
Onset (To)	63.6 ± 2.2 a		68.3 ± 1.2 b
Peak (Tp)	66.4 ± 0.7 a		72.38 ± 1.2 b
Endset (Te)	70.6 ± 1.6 a		79.0 ± 0.4 b
	Quinoa-to-Chickpea Flour Ratio
Pasting Properties (Pa.s)	QF90-CF10	QF75-CF25		QF50-CF50
Peak Viscosity (PV)	0.405 ± 0.0 a	0.350 ± 0.0 a		0.410 ± 0.1 a
Breakdown Viscosity (BV)	0.226 ± 0.0 a	0.201 ± 0.1 a		0.297 ± 0.0 b
Setback Viscosity (SV)	1.342 ± 0.1 a	1.081 ± 0.0 a		1.307 ± 0.1 a
Final viscosity (FV)	1.521 ± 0.0 a	1.228 ± 0.1 a		1.420 ± 0.0 a

Values are presented as mean ± standard deviation of at least three replicates. Means with different lowercase letters (a,b) indicate significant differences (p < 0.05) between QF-CF ratios. QF, quinoa flour; CF, chickpea flour.

. The QF sample exhibited an endothermic peak temperature (Tp) of 66.4 °C, an onset temperature (To) of 63.7 °C, and an endset temperature (Te) of 70.6 °C. In contrast, CF showed significantly higher temperatures (p < 0.05) than QF, with Tp at 72.4 °C, To at 68.3 °C, and Te at 79.0 °C. These results can be related to starch gelatinization temperature, as starch is typically the predominant component in QF and CF, with levels ranging from 55 to 65% and 30–45%, respectively [30,31]. The difference in starch gelatinization temperatures between the flours might be due to variations in the starch granule size, amylose-to-amylopectin ratio, crystalline arrangement, and protein content [12,32]. Quinoa starch granules are approximately 1–2 μm [31], whereas chickpea starch granules are significantly larger at ~17–20 μm [33,34]. Larger starch granules, such as those found in chickpeas, often require more energy for water penetration and the disruption of their crystalline structures, which could explain their higher gelatinization temperatures [35,36]. The amylose-to-amylopectin ratio is another critical factor that affects the thermal properties of raw materials. Quinoa starch has a low amylose content (20–27%), which can promote gelatinization at low temperatures [31]. Conversely, chickpeas have a higher amylose content (30–40%), forming a more linear and rigid structure, making them less prone to disorganization and requiring a higher gelatinization temperature [37]. Additionally, the crystalline structure of starch influences its thermal behavior. Quinoa starch typically displays an A- or A-C-type pattern, which is denser and more compact than the B-type pattern found in some legumes [30]. Despite their density, A-type starches are generally associated with lower gelatinization temperatures [38,39]. Furthermore, the difference in starch gelatinization temperatures between the flours can be attributed to the structure of the amylopectin chains and the crystalline arrangement. Quinoa starch generally contains shorter amylopectin chains, which require less energy to dissociate, and has a higher affinity for water because of its more accessible crystalline regions [40]. In contrast, chickpea starch exhibits a B- or C-type pattern, which is less dense but contains longer amylopectin chains. These characteristics often result in higher gelatinization temperatures that disrupt the crystalline structure [33,41]. Protein content also plays a role in gelatinization. Chickpea flour has a higher protein content than quinoa flour [42]. This protein may act as a physical barrier, impeding water absorption by starch granules and increasing the gelatinization temperature of chickpea flour [43]. These results are consistent with those reported by Tiga et al. (2021), who observed onset (To) and endset (Te) gelatinization temperatures of 67.13 °C and 75.80 °C, respectively, for quinoa flour [43]. Similarly, Lu et al. (2022) reported onset (To), peak (Tp), and endset (Te) temperatures of 64.86 °C, 70.44 °C, and 75.02 °C, respectively, for chickpea flour [44].

3.2. Pasting Properties of Quinoa-to-Chickpea Flour Ratios

Rapid Visco Analyzer (RVA) analyses are valuable for monitoring viscosity changes with temperature, which influence the physical and rheological properties of the system. The pasting properties, including the peak viscosity (PV), breakdown viscosity (BV), setback viscosity (SV), and final viscosity (FV), of the three quinoa-to-chickpea flour blends are listed in Table 1, and the corresponding curves are shown in Figure 1. Statistically significant differences (p < 0.05) were observed in pasting properties across different quinoa-to-chickpea flour ratios.

The peak viscosity (PV) is associated with the thickening behavior and water-binding capacity of the starches in the QF-CF blends. Interestingly, no significant differences in the PV were observed between the blends. The QF90-CF10 blend exhibited a PV of approximately 0.405 Pa.s, QF75-CF25 showed a PV of 0.350 Pa.s, and QF50-CF50 had a slightly higher PV (0.410 Pa.s). However, these results show a decreasing PV trend in the QF75-CF25 blend, followed by an increase in the QF50-CF50 blend. The decrease in PV for QF75-CF25 may be attributed to the higher proportion of chickpea, which contains starch with a higher amylose content than QF90-CF10. Amylose forms gel-like structures that limit granule swelling and reduce viscosity [45,46]. Furthermore, some studies have reported that the lipid content in chickpea flour (~5%) may form lipid-amylose complexes, further contributing to the reduction in PV [47]. Although quinoa remained the dominant component in this blend, its ability to promote viscosity may have been diminished by its interactions with chickpea proteins and lipids. These components can interact with quinoa starch through hydrogen bonding and hydrophobic interactions, forming complexes that stabilize the matrix and reduce starch swelling capacity [48,49]. In contrast, the increase in PV observed for QF50-CF50 suggests that, at a greater CF content, the system may experience a shift in behavior. Despite its high amylose content, chickpea starch contains long-chain amylopectin, which can enhance granule swelling and water retention, contributing to higher PV [49,50]. Additionally, protein-starch-lipid interactions may stabilize the starch matrix, supporting the observed increase in PV [34,46,49]. In this context, the contribution of quinoa’s high-swelling starch capacity combined with the structural properties of chickpea starch may have resulted in a synergistic effect that enhanced viscosity, despite the higher amylose content of chickpea starch. These results align with the findings of Mohammed et al. (2014) [46], who reported that wheat-to-chickpea flour blends exhibited similar viscosity trends influenced by interactions such as amylose-lipid complexation, protein-starch associations, and ternary protein-lipid-starch networks [51,52], thereby maintaining starch swelling and contributing to the observed synergistic increase in PV. Furthermore, it has been reported that although protein-starch-lipid interactions are generally associated with starch granule swelling inhibition, the molecular structure of chickpea starch, particularly its long-chain amylopectin, may counteract these effects and contribute positively to the swelling power and peak viscosity [30,34], which could explain the increased PV observed for the QF50-CF50 blend.

The breakdown viscosity (BV) is the difference between the peak viscosity and holding strength [53]. Significant differences (p < 0.05) were observed in BV among the different quinoa-to-chickpea flour ratios. The QF90-CF10 blend presented a BV of 0.226 Pa.s, the QF75-CF25 blend exhibited a BV of approximately 0.201 Pa.s, and the QF50-CF50 blend presented the highest BV value (0.297 Pa.s). The reduction in BV observed in QF75-CF25 suggests a system in which a higher proportion of quinoa flour (75%) dominates the behavior, leading to rapid swelling and subsequent disintegration of the smaller quinoa starch granules. This rapid swelling, which is influenced by the lower amylose content of quinoa starch, could contribute to a less stable system under shear stress. Conversely, 25% chickpea flour could introduce larger starch granules with a higher amylose content and greater thermal stability, which might partially restrict granule swelling and stabilize the viscosity profile. However, this proportion of chickpea flour was insufficient to counterbalance the disintegration of quinoa starch granules, resulting in the lowest BV among the blends. Furthermore, the interaction between chickpea proteins and starch granules can reduce granule swelling by limiting water availability, particularly for quinoa starch [54]. Competition for water between quinoa and chickpea starches, including other components such as proteins and fibers, could also alter hydration dynamics, leading to the incomplete gelatinization of some starch granules. These combined factors seem to create a system in which BV reflects intermediate stabilization from chickpea components but remains dominated by the rapid disintegration of quinoa starch. In contrast, the significantly higher BV observed in the QF50-CF50 blend may result from a combination of structural contributions from chickpea starch and the moderating effect of chickpea proteins on quinoa starch behavior. Higher CF content introduces starch granules with greater thermal stability and proteins that may interact with the starch matrix, limiting excessive granule swelling and delaying breakdown under shear stress [33,55]. This protein–starch interaction may help reinforce the structural integrity of quinoa starch granules during heating, thereby reducing their susceptibility to disintegration and contributing to the higher BV observed in this blend. Some studies have reported that proteins can interact with starch granules through various mechanisms, including the formation of protective networks, hydrophobic bonds, and electrostatic interactions, which might mitigate starch breakdown, limit starch swelling, and stabilize the viscosity profile [12,55,56]. Similar trends have been observed in quinoa–wheat flour blends, where increasing the proportion of quinoa flour led to a reduction in breakdown viscosity, likely owing to the lower thermal and shear stability of quinoa starch granules compared to those of wheat [12].

Setback viscosity (SV) reflects the tendency of starch pastes to retrograde during cooling, a process closely linked to the physical and rheological stability of starch-based systems. SV is strongly influenced by the molecular architecture of starch, including the amylose-to-amylopectin ratio and the chain-length distribution of amylopectin, as well as by the presence of proteins and non-starch polysaccharides [43,57]. Generally, longer amylopectin chains and higher amylose content are associated with increased retrogradation, resulting in higher SV values. However, shorter amylopectin chains and lower amylose contents are associated with decreased retrogradation, resulting in lower SV values [12,49]. No significant differences in PV were observed between the flour mixtures. Nonetheless, the QF90-CF10 blend had the highest SV of 1.342 Pa.s, QF75-CF25 exhibited the lowest SV of approximately 1.081 Pa.s, and QF50-CF50 had an intermediate SV value of 1.307 Pa.s. These results indicate the complex interaction of starch composition, non-starch polysaccharides, and proteins, which could collectively influence retrogradation patterns in quinoa-to-chickpea flour ratio blends [12,57,58]. Typically, a lower amylose-to-amylopectin ratio, as found in quinoa, is associated with delayed retrogradation and, thus, a lower SV [50,59]. However, the higher SV observed for the QF90-CF10 blend suggests that factors other than the amylose content may have contributed to retrogradation. In particular, the structural characteristics of quinoa starch, such as its highly branched amylopectin and shorter chain lengths, may facilitate interactions with proteins and non-starch polysaccharides, promoting molecular re-association during cooling and positively impacting the SV [60]. The QF75-CF25 blend, which had a higher proportion of CF than the QF90-CF10 blend, exhibited the lowest SV. Chickpea starch generally has high amylose content, which promotes retrogradation, but chickpea proteins and non-starch polysaccharides may interfere with the re-association of quinoa amylopectin molecules. This interference could reduce the extent of retrogradation in QF75-CF25 compared to that in the QF90-CF10 blend. Additionally, although quinoa dominates this blend at 75%, its potential contribution to retrogradation appears limited. Proteins and non-starch polysaccharides from chickpea flour may disrupt the alignment of quinoa amylose and amylopectin molecules, thereby reducing their ability to form crystalline structures. Furthermore, the highly branched structure of quinoa amylopectin typically delays retrogradation [61], and its effects may be less pronounced in this blend, owing to competitive interactions with chickpea components. The intermediate SV value observed for the QF50-CF50 blend reflects a balance between the retrogradation tendencies of quinoa and chickpea starches, which could positively impact the physical stability during cold storage and prevent particle sedimentation. Moreover, SV in this blend appears to improve gel strength and stability, benefiting the physical and rheological properties of the system [35]. However, unlike the findings of Wang et al. (2015), who observed no significant changes in SV across different quinoa-to-wheat flour ratios [62], the present study showed that quinoa-to-chickpea flour ratios influence SV, likely due to distinct starch-protein interactions specific to chickpea components [63].

The final viscosity (FV) indicates the ability of a material to form a viscous paste after cooling. No significant differences were observed between the different quinoa-to-chickpea flour ratios. For the QF90-CF10, QF75-CF25, and QF50-CF50 blends, the FV were approximately 1.521 Pa.s, 1.228 Pa.s, and 1.420 Pa.s, respectively. The slightly higher FV observed for the QF90-CF10 blend may highlight the greater ability of quinoa flour to form a stable paste or gel upon cooling, which enhances the gel stability [64]. Interestingly, although quinoa starch is characterized by a lower amylose content than chickpea starch, its strong retrogradation properties may stem from its amylopectin structure. Quinoa amylopectin has shorter branched chains, which are more efficient in forming ordered crystalline regions during retrogradation [12,31,65,66]. This property can compensate for the lower amylose content, contributing to its superior ability to form stable gels upon cooling [64]. In contrast, the QF50-CF50 had an intermediate FV value. The results suggest a balanced interaction between quinoa and chickpea flour components, which affects the physical and rheological properties of the blend in terms of moderate fluidity and sufficient viscosity to resist deformation. Although quinoa starch contributes to gel formation through both amylose re-association and structural support from amylopectin [12,67], chickpea starch, despite having longer amylopectin chains that form less compact crystalline networks, may promote faster retrogradation owing to improved molecular alignment [68]. This observation aligns with the findings of Mohammed et al. (2014) [46], who noted that the aggregation and re-association of amylose molecules play a dominant role in determining the final viscosity and retrogradation behavior of wheat-chickpea flour blends.

3.3. Plant-Based Beverage Fermentation

Table 2. Fermentation and growth kinetics parameters. Changes in pH, acidity, Brix degrees, and LAB growth before and after 10 h of fermentation.

	Beverage
Condition	Parameters	QF90-CF10	QF75-CF25	QF50-CF50
Unfermented samples	pH	6.0 ± 0.0 Aa	6.1 ± 0.0 Aa	6.10 ± 0.1 Aa
	TTA (%)	0.3 ± 0.0 Aa	0.2 ± 0.0 Aa	0.2 ± 0.0 Aa
	Yield (g lactate/g glucose)	0.0 ± 0.0 Aa	0.0 ± 0.0 Aa	0.0 ± 0.0 Aa
	°Brix	10.7 ± 0.2 Aa	9.9 ± 0.1 Aa	8.7 ± 1.3 Ba
	Log CFU/mL	2.3 ± 0.1 Aa	2.3 ± 0.1 Aa	2.5 ± 0.1 Aa
Fermented samples	pH	4.1 ± 0.04 Ab	4.3± 0.0 Ab	4.3 ± 0.1 Ab
	TTA (%)	0.4 ± 0.0 Ab	0.4 ± 0.0 Ab	0.5 ± 0.0 Bb
	Yield (g lactate/g glucose)	1.0 ± 0.0 Ab	0.9 ± 0.0 Ab	1.0 ± 0.0 Ab
	°Brix	10.0 ± 0.4 Aa	9.6 ± 0.3 Aa	7.3 ± 0.2 Ba
	Log CFU/mL	7.9 ± 0.3 Ab	9.1 ± 0.1 Ab	7.9 ± 0.0 Ab
	μmax(log CFU/g/h)	1.0 ± 0.0 a	1.4 ± 0.0 b	1.1 ± 0.11 a
	Growth T10-T0 (log CFU/10 h)	5.6 ± 0.1 a	6.8 ± 0.1 b	5.4 ± 0.7 a

Values are presented as mean ± standard deviation of at least three replicates. Means with different lowercase letters (a,b) indicate significant differences (p < 0.05) between the same QF-CF ratio before and after fermentation. Means with different uppercase letters (A,B) indicate significant differences (p < 0.05) between different QF-CF ratios. QF, quinoa flour; CF, chickpea flour.

presents the pH levels, lactic acid yields, °Brix values, and growth kinetic parameters determined post-fermentation. A significant decrease (p < 0.05) in pH was observed at the end of fermentation, which corresponded to an increase in TTA (%) and LAB growth. The initial pH of each blend was approximately 6.1. At the end of fermentation, the pH value decreased to approximately 4.3, suggesting that LAB effectively metabolized the starch present in the beverages, leading to acidification over the 10 h fermentation period. Moreover, lactic acid yield and TTA (%) were directly linked to the final pH values at the end of fermentation. These changes in low pH and increased acidity were due to lactic acid production by L. acidophilus LA-5 during fermentation. The °Brix values of the beverages remained constant until the end of fermentation, indicating that LAB preferentially metabolize easily fermentable monosaccharides such as glucose and fructose over more complex carbon sources such as sucrose and raffinose, which may be present in quinoa and chickpea flour [69]. Additionally, as noted in previous studies, the QF50-CF50 blend has the highest TTA at the end of fermentation [23]. This could be attributed to the high buffering capacity of chickpeas, which requires LAB to produce more lactic acid to reduce the pH below 4.5 when a greater proportion of legume flour is present in the PBBs [10]. Similar results were reported by Huang et al. (2022) [10], who indicated that quinoa-based plant-based beverages with a higher proportion of legume (soybean) required more extensive acid production to overcome buffering effects and achieve suitable fermentation profiles (pH < 4.5) [10].

3.4. Water-Holding Capacity of Beverages After Fermentation

Water holding capacity (WHC) reflects the ability of solid components in a beverage matrix, such as proteins, starch, and fiber, to retain water [66]. The WHC (%) of different plant-based beverages (PBBs) was evaluated before and after fermentation (Figure 2). No statistically significant differences (p > 0.05) were observed between the fermented and unfermented samples. Prior to fermentation, the QF90–CF10 blend exhibited a WHC of approximately 73%, whereas the QF75–CF25 and QF50–CF50 blends showed values of 68% and 70%, respectively. After 10 h of fermentation with L.s acidophilus LA-5, the WHC values slightly increased to approximately 74%, 72%, and 73% for the same formulations. Although these changes were not statistically significant, they suggest a moderate enhancement in water retention, which could help maintain physical stability and reduce phase separation during storage [10,22,70,71].

In well-stabilized fermented PBBs, the WHC values typically range from 80% to 90% [6]. The absence of significant differences in this study may be attributed to the relatively short fermentation period (10 h) and pH threshold (below 4.5), which may not have been sufficient to stimulate the production of secondary metabolites such as exopolysaccharides by L. acidophilus. These results are consistent with those reported in a previous study [72], where fermentation of a lentil-based beverage using Leuconostoc citreum and Lacticaseibacillus paracasei also resulted in no significant changes in the WHC. This suggests that short-term homolactic fermentation may have a limited effect on WHC in legume-enriched PBBs. For instance, extended fermentation (e.g., ≥24 h, pH < 3.5) may lead to greater enzymatic and acid-mediated breakdown of macromolecules, which could enhance the water-binding capacity of the fermented matrix.

3.5. Effect of Fermentation on the Viscosity of Quinoa and Chickpea Beverages

Viscosity is a fundamental property of PBBs owing to its role in ensuring product stability by preventing phase separation. Higher viscosity supports the suspension of particulates and improves the homogeneity of the beverage, which is particularly important in fermented systems where proteins, polysaccharides, and other components interact dynamically, thereby influencing the rheological properties of the final product [7]. The viscosities of both unfermented and fermented PBBs were measured, as shown in Figure 3A–C. The viscosity decreased with an increase in the shear rate, indicating a shear-thinning behavior typical of non-Newtonian fluids. A significant increase in viscosity (p < 0.05) was observed when L. acidophilus LA-5 was used as the starter culture. After 10 h of fermentation, the PBBs reached viscosity values of approximately 12.0 Pa.s, 14.6 Pa.s, and 18.0 Pa.s at zero shear rates in the QF90-CF10, QF75-CF25, and QF50-CF50 blends. These findings highlight the modifications induced by fermentation, including changes in starch granules and macromolecules such as proteins and the production of peptides and metabolites by LAB. Under acidic conditions (e.g., pH < 4.5), lactic acid gradually penetrates hydrated starch granules and releases amylose and amylopectin, enhancing their interaction with water and other components of the matrix [48]. Furthermore, the hydrolytic activity of bacterial enzymes, particularly exoamylases, may contribute to starch breakdown and release of smaller polysaccharides, which can increase viscosity. Protein hydrolysis during fermentation may further expose hydrophilic and decreased hydrophobic regions of the protein, promoting more protein-water interaction sites and contributing to enhanced water retention and increased viscosity [14,73]. In our previous studies, we reported that the fermentation process under these conditions (e.g., 10 h and L. acidophilus as starter) can increase protein solubility, protein hydrolysis, and changes in the secondary protein structure of quinoa and chickpea beverages [23], which could be related to the improvement of viscosity of the fermented PBB.

The differences in viscosity between the different blends indicated that the quinoa-to-chickpea ratio significantly influenced the rheological properties of the PBBs. The QF50-CF50 blend presented the highest viscosity (18.0 Pa.s) at the end of fermentation. This result suggests that the inclusion of chickpeas positively affects the viscosity of beverages based on pseudocereal-like quinoa and may modulate their physical stability. Several factors may explain the viscosity differences in the PBB, including starch content and granule size, as well as macromolecules such as proteins and fibers. It can be hypothesized that the high amylose content and large starch granules in chickpeas promote greater molecular entanglement and intermolecular interactions during fermentation, which contribute to thickening [34]. In addition, the high protein content in CF may increase the gelation of proteins owing to the low pH (<4.5) and high acidity, favoring a more cohesive protein-starch matrix, creating a network that retains water and other components, such as lipids and carbohydrates, and indirectly increases viscosity [37]. Furthermore, it has been reported that an increase in the soluble fiber content of chickpeas after fermentation could promote more water absorption, thereby increasing the viscosity of PBBs [74]. Similar findings have been reported for quinoa-soybean beverages, where higher legume proportions enhanced apparent viscosity [10]. Bianchi et al. (2015) reported that formulations with greater proportions of legumes, such as soybeans, exhibited higher viscosity than other blends [75].

The production of organic acids and enzymatic activity during fermentation could have mainly degraded quinoa starch because of its low starch size and amylose-to-amylopectin ratio, which allowed chickpea starch with a higher amylose content to dominate the viscosity profile [76]. In addition, RVA analysis was performed on fermented PPBs, as shown in Figure 4 and Table S1. Among the fermented PBBs, the QF50-CF50 blend exhibited the highest peak viscosity (PV) and final viscosity (FV), demonstrating that under acidic pH (<4.5), a higher proportion of CF positively affects the entanglement and interaction of starch, proteins, and polysaccharides. The high amylose content and larger starch granules in CF likely contributed to the formation of a stronger matrix, leading to higher PV and FV values compared with other fermented blends. In addition to amylose content, amylopectin chain length may play a significant role in pasting behavior (PV and FV). Chickpea starch has been reported to contain a high proportion of long-chain amylopectin (~33% of its total amylopectin content), which enhances intermolecular associations and promotes the formation of a more cohesive system during heating and subsequent cooling, as observed in CF-rich fermented PBBs by RVA analysis [33,77]. Furthermore, soluble sugars released during homolactic fermentation, either derived from acid-induced starch hydrolysis or inherent in the raw material, could affect pasting properties by binding water molecules and altering the availability of free water, thereby modifying starch swelling and consequently increasing the viscosity of fermented PBBs [78]. Some studies have reported that the effect of the substrate ratio on apparent viscosity is dependent on the dose of quinoa or legume (soybean) in fermented plant-based beverages [10], directly affecting the rheological properties of plant-based systems.

3.6. Changes in the Viscoelasticity Behavior After the Fermentation

Strain sweep testing is a useful method for indirectly quantifying the extent to which a beverage deforms or stretches when force is applied, providing insight into its texture, cohesiveness, and flow behavior [79]. Figure 3D–F illustrates the effect of homolactic fermentation on the storage modulus (G′) as a function of the oscillation strain (%) in quinoa-to-chickpea flour beverages. As the strain increased, the storage modulus (G′) decreased, suggesting reduced resistance to deformation within the colloidal suspension. This behavior implies that the applied strain significantly influences the deformation resistance of the fermented beverages. In addition, fermented beverages exhibited statistically significant differences (p < 0.05) compared to the unfermented samples, confirming the modification of starches during the fermentation process and suggesting that the production of enzymes such as amylases by LAB affects the integrity of starch granules. Fermentation may induce partial erosion of ungelatinized starch granules that remain intact after the pasteurization of PBBs [70]. Furthermore, the increase in deformation resistance reflected in the increase in the crossover point of the fermented beverages could result from enhanced exposure to amylose and amylopectin fractions in the partially gelatinized starches. The enzymatic activity of amylases produced by LAB may hydrolyze starch granules, thereby reducing their molecular weight while simultaneously increasing the resistance to deformation [80]. Moreover, gelatinized starches may undergo retrogradation during fermentation, forming more ordered and rigid structures that contribute to an increased G′ value [77,81]. Carvalho Alves et al. (2024) reported an increase in shear stress in a plant-based beverage made from Syagrus coronata after 24 h of fermentation using kefir grains, indicating pseudoplastic rheological behavior typical of non-Newtonian systems [82].

The effect of chickpea flour addition on the deformation resistance and elastic behavior of quinoa-based fermented beverages was also observed. After the fermentation process, the QF50-CF50 blend exhibited the greatest difference in storage modulus (G′) as a function of the oscillation strain compared to the other blends. These results indicate that a higher proportion of chickpea flour positively influenced the deformation resistance of fermented PBBs. Previous studies have reported that chickpea starch exhibits better elastic properties than other starches, such as corn, potato, and Turkish bean starch. This advantage is likely due to the higher amylose content in chickpeas, which contributes to stronger molecular entanglement and interactions, and promotes higher resistance to deformation under applied tension [83,84]. Additionally, starches with a higher proportion of long amylopectin chains, as found in chickpeas, tend to have a higher storage modulus due to increased intermolecular association during retrogradation, which positively affects the physical stability of PBBs [85].

The viscoelastic behavior of the PBBs was evaluated based on their storage modulus (G′) and loss modulus (G″) as a function of frequency in the linear viscoelastic region, which reflects the structural and mechanical properties of the liquid system. Figure 5 shows that G′ and G″ increased as the frequency increased. All the samples exhibited weak viscoelastic gel behavior related to the dominance of the storage modulus over the loss modulus (G′ > G″), as shown in Figure 5. Differences in G′ values were observed after fermentation, suggesting improvements in the rheological properties associated with the viscoelastic behavior of PBBs. This result could be attributed to changes in the starch network within the quinoa and chickpea beverages, where a more entangled and cohesive macromolecular configuration with strong component interactions, such as lipids, proteins, fibers, and starches, likely contributed to the increased storage modulus of the PBBs [73]. Huang et al. (2022) [10] demonstrated that the storage modulus increased after fermentation using L. delbrueckii subsp. bulgaricus and Streptococcus thermophilus in a pseudocereal-legume-based beverage.

In addition, a higher proportion of quinoa flour resulted in lower G′ values, whereas increased chickpea flour content enhanced the G′ values of PBBs. A higher amylose and protein content in chickpea flour may promote the formation of starch-starch and starch-protein complexes, enhancing molecular association and improving elastic characteristics. Furthermore, chickpea starch, which has high amylose content and long-chain amylopectin, exhibits distinct retrogradation patterns that likely contribute to the increase in G′ [83]. As shown in Figure 4 and Table S1, fermented QF90-CF10 presented a lower SV (0.023 Pa.s), demonstrating that after exposure to acidic conditions, quinoa flour tended to retrograde slowly, mainly because of its lower amylose content, and the increase in soluble fiber content after homolactic fermentation appeared to retard retrogradation [22,37]. In contrast, QF-50-CF50 presented a higher SV (0.148 Pa.s), indicating that the CF content positively impacted the G′ of the fermented PBBs related to chickpea protein and chickpea starch content, which could increase the strengthening of the protein-protein complexes and chickpea starch with higher amounts of amylose content and long-branch chain amylopectin, enhancing the retrogradation patterns. Some studies have reported that adding legumes to other plant-based beverages, such as quinoa-soybean blends, improves the viscoelastic properties of the fermented system [10]. These results emphasize the pivotal molecular structure of chickpea starch with a high amylose content and long amylopectin chains in enhancing the viscoelastic behavior of fermented PBBs. Although the observed increases in viscosity, storage modulus, and pasting parameters indicate stronger molecular associations between starches and proteins, no direct microscopic or microstructural evidence was obtained to evaluate the specific arrangement of these components in our research. Therefore, future studies should incorporate microstructural characterization to complement the rheological data and better understand the internal organization of PBB systems after fermentation.

3.7. Fourier Transform-Infrared Spectroscopy of Plant-Based Beverages

Macronutrient composition (e.g., carbohydrates, proteins, and lipids) is a critical parameter that can modulate the physical, rheological, and structural properties of formulated PBBs [83]. PBBs made from quinoa and chickpea flour have a high carbohydrate content and, consequently, a high starch content. Quinoa has a starch content of approximately 55–65% [31,65]. The starch content in chickpeas is approximately 40–50% [30]. According to previous studies, fermentation can potentially induce modifications in the starch structure, particularly affecting the crystalline regions, which may influence the rheological and stability properties of plant-based beverages [28]. However, the extent of such structural changes can vary depending on the fermentation time, pH, and microbial enzymatic activity [29,86,87]. Starch crystallinity plays a crucial role in determining the rheological and physical properties of PBBs. Higher crystallinity is generally associated with a more rigid and ordered structure, which can negatively impact viscosity. Conversely, lower crystallinity can enhance viscosity and viscoelastic properties, thereby improving the physical stability of PBB [88].

The short-range ordered structures of quinoa and chickpea starches in the beverages were evaluated using FTIR-ATR, specifically analyzing spectral changes in the region of  1200–800 cm⁻¹. This region includes characteristic absorption bands associated with amorphous starch (~995 cm⁻¹), an intermediate order region (~1022 cm⁻¹), and ordered crystalline domains (~1047 cm⁻¹).

Table 3. Effect of fermentation on the molecular order of starch structure in quinoa-to-chickpea beverages.

Beverage	Fermentation Conditions	Amorphous Starch Region (cm−1)	Intermediate Region (cm−1)	Ordered Starch Region (cm−1)	DO (1047/1022)	DD (995/1022)
QF90-CF10	Unfermented	994 ± 0.0 Aa	1022 ± 0.0 Aa	1046 ± 0.0 Aa	0.73 ± 0.0 Aa	0.95 ± 0.0 Aa
	Fermented	994 ± 0.6 Aa	1022 ± 0.0 Aa	1046 ± 0.0 Aa	0.72 ± 0.0 Aa	0.94 ± 0.0 Aa
QF75-CF25	Unfermented	994 ± 0.0 Aa	1022 ± 0.0 Aa	1046 ± 0.0 Aa	0.70 ± 0.0 Aa	0.94 ± 0.0 Aa
	Fermented	994 ± 0.0 Aa	1021 ± 0.6 Aa	1045 ± 0.1 Aa	0.71 ± 0.0 Aa	0.96 ± 0.0 Aa
QF50-CF50	Unfermented	994 ± 0.0 Aa	1022 ± 0.0 Aa	1046 ± 0.0 Aa	0.65 ± 0.0 Ba	0.93 ± 0.0 Aa
	Fermented	994 ± 0.5 Aa	1022 ± 0.0 Aa	1046 ± 0.0 Aa	0.71 ± 0.0 Ab	0.97 ± 0.0 Ab

Values are presented as mean ± standard deviation of at least three replicates. Means with different lowercase letters (a,b) indicate significant differences (p < 0.05) between the same ratio of QF-CF before and after the fermentation process. Means with different uppercase letters (A,B) indicate significant differences (p < 0.05) between different ratios of QF-CF. QF, quinoa flour; CF, chickpea flour. The FTIR band positions (wavenumbers, cm⁻¹) were determined from the amplitude of negative peaks in the second-derivative spectra within the crystalline region (995–1047 cm⁻¹). The bands were grouped into three regions: the amorphous starch region (~995 cm⁻¹), the intermediate region (~1022 cm⁻¹), and the ordered starch region (~1047 cm⁻¹), which are indicative of starch crystallinity and molecular organization. The degree of order (DO; 1047/1022) and degree of double helix (DD; 995/1022) ratios were calculated from normalized amplitudes.

presents the absorption band positions (wavenumber, cm⁻¹) of these key bands, and Figures S1 and S2 show the second-derivative spectra used to enhance band resolution. The negative peaks obtained from second-derivative processing allowed the identification of subtle changes in the short-range-ordered structure of starch before and after homolactic fermentation, particularly in the regions between 995 and 1022 cm⁻¹ and 1022–1047 cm⁻¹ [29]. Across all blends, non-significant shifts in the FTIR band positions were observed after homolactic fermentation, indicating that the short-range ordered structures of the starches remained unaltered under the studied conditions. Although lactic acid fermentation can reduce starch crystallinity in terms of long-range structures in some plant-based systems, particularly those rich in amylose or subjected to extended fermentation times, as reported by Zhao et al. (2019) [29], the fermentation conditions applied in this study (10 h, pH > 4.5, and absence of enzymatic pretreatment) may have limited extensive changes in the short-range ordered structures of quinoa and chickpea starches [28]. Furthermore, it is important to consider that the sensitivity of FTIR to detect subtle structural modifications may be constrained not only by the spectral resolution used (4 cm⁻¹), but also by potential band distortions caused by the anomalous dispersion of the refractive index, as previously reported by Boulet-Audet et al. (2010) [89]. Therefore, peak positions were identified using second-derivative spectra only to locate the characteristic bands related to short-range molecular order, whereas the structural interpretation focused primarily on intensity-based DO and DD ratios, which are less susceptible to such distortions.

Interestingly, the inclusion of chickpeas appeared to modulate the degree of short-range order in starch, as evidenced by changes in the DO and DD ratios. In the QF50–CF50 blend, homolactic fermentation significantly increased the DO value from 0.65 to 0.71 and the DD value from 0.93 to 0.97 (p < 0.05), suggesting a mild reinforcement of the short-range order and double-helical structures. It could be indicated that the structural response of starch to fermentation is composition-dependent, with chickpea starch contributing to greater resistance to disruption or enhanced reorganization under the tested conditions. Chickpea starch typically contains high levels of amylose and long-chain amylopectin, which may promote partial retrogradation or stabilize the starch conformation through interactions with lactic acid and other fermentation-derived metabolites [37]. Such interactions, primarily hydrogen bonding between the hydroxyl groups of amylopectin and organic acids, could contribute to the preservation or reordering of double helices in the starch structure under moderate fermentation conditions [28,29].

Additionally, the higher chickpea flour content in the QF50–CF50 blend may have altered the starch–protein interactions and increased the water-binding capacity, creating a more hydrated microenvironment that favors molecular mobility and structural reorganization. Homolactic fermentation may also exert a mild hydrolytic effect, preferentially on the amorphous regions of chickpea starch, which might be more susceptible to disruption [73,76]. The partial degradation of these less-ordered domains may have facilitated the reassembly of amylopectin chains into more stable and ordered double-helical structures, thereby increasing the short-range molecular order of the starch. This mechanism has been previously described in mildly fermented or retrograded starch systems, particularly under non-enzymatic and moderately acidic conditions [90]. These structural rearrangements could explain the significant increase in DO (1047/1022) and DD (995/1022) values observed in the QF50–CF50 blend after fermentation. Taken together, these results suggest that chickpea-enriched matrices are more prone to starch reorganization, potentially contributing to the enhanced physical stability and rheological properties of the final product. However, to fully validate the hypothesized starch modifications suggested by the FTIR, rheological, and RVA data, future studies should incorporate complementary techniques such as X-ray diffraction (XRD) to accurately quantify subtle changes in starch crystallinity in quinoa–chickpea fermented beverages.

4. Conclusions

This study demonstrated that homolactic fermentation plays a central role in tailoring starch structure and enhancing the rheological and viscoelastic properties of plant-based beverages. The addition of chickpea flour further influenced pasting behavior, with higher proportions leading to increased peak, breakdown, setback, and final viscosity values, which were attributed to its high amylose content, long-chain amylopectin, and large starch granules. The combined effect of fermentation and chickpea enrichment improved the apparent viscosity and viscoelasticity of quinoa-based probiotic beverages. Interestingly, the water-holding capacity (WHC) remained unaffected by fermentation or chickpea addition. Among all the blends, the 50:50 quinoa–chickpea blend exhibited the highest apparent viscosity, improved storage modulus (G′), and significant increases in DO and DD ratios after fermentation, as confirmed by the FTIR band positions, indicating a subtle but measurable reinforcement of short-range molecular starch organization. These findings highlight the potential of combining controlled homolactic fermentation and legume flour enrichment as a sustainable and functional strategy for designing dairy-free and physically stable probiotic beverages. Future work should investigate extended fermentation times and integrate complementary structural analyses, such as X-ray diffraction, to further elucidate their impact on starch crystallinity and short-range molecular organization.