Flow behavior, thixotropy, and dynamic viscoelasticity of ethanolic purified basil (Ocimum bacilicum L.) seed gum solutions during thermal treatment

Abstract During processing, foodstuffs may be treated at various thermal operations. Thus, this study investigated the functional properties and flow behavior at operation conditions to ensure safety and improve quality and stability at high temperatures and analyzed the ability of gum to be used in food formulation. The results showed that the purified basil seed gum (PBSG) solutions could beknown as non‐Newtonian liquids with pseudoplastic behavior. Frequency sweep revealed the storage modulus (G′) was higher than the loss modulus (G″) in the treatments. According to stress sweep, frequency sweep, complex viscosity (η *), and loss‐tangent (tan δ) outcomes, mechanical spectra of PBSG were categorized as weak gels. Besides, concentration and temperature were effect on G′ and G″. The results indicated that, in general, 1% PBSG‐121°C had the maximum yield stress, consistency coefficient (k), extent of thixotropy, and the minimum values of flow behavior index. Also, 1% PBSG showed the highest G′, G″, η *, yield stress values at the limit of the LVE range (τy), flow‐point stress (τf), and corresponding modulus G f (G′ = G″), and the lowest value of tan δ. Exhibiting distinctive rheological characteristics of PBSG makes it as a worthy hydrocolloid to use in food products, which use thermal processing.

flow behavior at operation conditions to ensure safety and improve quality and stability at high temperatures and analyzed the ability of gum to be used in food formulation (Naji, Razavi, & Karazhiyan, 2012;Zameni, Kashaninejad, Aalami, & Salehi, 2015). Results showed that in dilute regime, BSG has random coil conformation, but can form ordered conformation at desirable situations like appropriate concentration, existence of binding agents, or alteration in temperature and pH (Naji-Tabasi et al., 2016).
Several studies have been performed to use BSG as a new hydrocolloid in the food products (Farahmandfar, Asnaashari, Salahi, & Rad, 2017;Hosseini-Parvar, Matia-Merino, & Golding, 2015; Hosseini-Parvar, Osano, & Matia-Merino, 2016;Naji-Tabasi & Razavi, 2017b;Osano, Hosseini-Parvar, Matia-Merino, & Golding, 2014;Rafe, Razavi, & Farhoosh, 2013;Rafe, Razavi, & Khan, 2012;Razavi, Shamsaei, Ataye, & Emadzadeh, 2012). But, no research up to the present time has been published on the broad evaluation of the steady and dynamic rheological parameters of PBSG. Thus, the chief goal of this work was to broadly investigate the steady and dynamic rheological parameters of PBSG and their dependency to temperature and concentration. Herein, the influence of concentration and temperature on the several rheological parameters obtained from shear dependency, time dependency, stress sweep, and frequency sweep tests in the linear viscoelastic (LVE) range was investigated. Finally, the ability to run the Cox-Merz relationship between complex and apparent viscosity was evaluated.

| Materials
Seeds of basil (O. bacilicum L.) were purchased from a local market, Sari, Iran. After removing the foreign bodies and matters, seeds were sealed in bags and stored in a dry and cool place before extraction operation. Ethanol (96%) and sodium azide (NaN 3 ) were provided from Merck Chemical Co. (Darmstadt, Germany) and Applichem Inc.

| BSG extraction and purification
Basil seed gum was extracted and purified as stated by Naji-Tabasi et al. (2016), in the conditions: soaking time: 20 min, temperature: 68°C, pH: 7, and water/seed ratio: 20:1. Separation of mucilage from the swollen seeds was achieved by scraping technique. The seeds were passed through an extractor equipped with a rotating plate that scraped the mucilage layer on the seed surface. After filtration of the crude extract, the one volume of the extracted gum solution was mixed with three volumes of 96% ethanol and left overnight at 4°C to precipitate polysaccharide. The precipitate was then recovered using a sieve to allow excess solvent to drain. The final precipitate was dispersed in distilled water with continuous stirring for 25 min. The extracted gum was then dried in a dehydrator at 38°C, ground, and packed for further use.

| Solution preparation
Hydrocolloid solutions (w/w) were prepared by dissolving a suitable amount of the PBSG powder in deionized water, which contains 0.02% (w/w) NaN 3 as antimicrobial preservative, and stirring with a magnetic stirrer to obtain a uniform solution with concentrations of 0.5%, 0.75%, and 1%. The solutions were shaken by a roll mixer for overnight to hydrate thoroughly.

| Thermal treatments
To determine heat stability of PBSG, steady and dynamic shear rheological properties of 0.5%, 0.75%, and 1% PBSG were evaluated after heat treatment of the solutions at four levels including 20 (witness sample), 60, and 100°C for 30 min as well as at 121°C for 15 min (sterilization treatment) according to Hosseini-Parvar et al. (2010). All treatments were prepared in water bath (Memmert, Schwabach, Germany), except 121°C, which was performed in an autoclave (Famco, Tehran, Iran). Before experiments, solutions were cooled to 20°C.

| Rheological evaluations
Rheological measurement of solutions was carried out in a controlled stress/strain rheometer (Physica MCR 301, Anton Paar GmbH, Stuttgart, Germany) equipped with cone-plate system (0.208 mm gap, 50 mm diameter, and 2° angle) with at least duplication measurements. Before experiment, solutions were authorized to equilibrate to 20°C for 15 min. The temperature-controlled system was a Peltier system (Viscotherm VT2) fully equipped with a fluid circulator (Anton Paar, GmbH) with the accuracy ± 0.01°C. Data were analyzed by the Rheoplus software version 3.40 for analysis.

Shear dependence
Due to the time-independent quality of the gum solutions, primarily, a constant shear rate (100 s −1 ) was used until an equilibrium state was perceived. At that time, the steady shear flow properties of prepared PBSG solutions were quantified at 20°C in shear rates of 0.1-300 s −1 .

Time dependence
To study time-dependent rheological behavior, the PBSG solutions were sheared at effective oral shear rate constant (50 s −1 ) (Bourne, 2002) and the τ and the viscosity (η) were measured as a function of shearing time (t) until a balance mode was achieved. The timedependent flow patterns of the solutions were analyzed by follow models: Weltman model: First-order stress decay, with a nonzero stress value: Structural kinetic model (SKM): where A and B characterize the time-dependent behavior, k is the breakdown rate constant, τ 0 is the initial shear stress, τ eq is the equilibrium stress, ′ 0 is the initial apparent viscosity at t = 0, � ∞ is the steady-state apparent viscosity at t→∞, and n is the order of the structural breakdown reaction. Herein, the second-order (n = 2) was used to explain the SKM of PBSG solutions.

Stress sweep
Before making detailed dynamic measurements, the linear viscoelas-

| Data analysis
The data were subjected to one-way analysis of variance (ANOVA) at 95% significance level (p < 0.05), and Minitab 18 (Minitab Inc., Minneapolis, USA) was used to compare the means by Tukey test.
Data fitting was performed by MathWorks' MATLAB (R2016a) software, using the curve fitting toolbox.

| Time-independent rheological properties
Flow curves of shear stress and apparent viscosity as the function of shear rate of the PBSG solutions at different treatments are shown in Figure 1. The pattern of these curves and power law and Herschel-Bulkley parameters (Table 1) showed that PBSG solutions could be known as non-Newtonian liquids with pseudoplastic behavior (n < 1). Vardhanabhuti and Ikeda (2006) (4) stated that shear-thinning demeanor of BSG is due to aggregation of polymers by hydrogen bonds and high molecular weight.
Hydrocolloids with high pseudoplastic behavior are widely used to modify or ameliorate food texture during high-shear processing like filling and pumping, and during swallowing impart a thinner consistency (Vardhanabhuti & Ikeda, 2006). Moreover, high viscosity provides pleasant mouthfeel (Naji-Tabasi & Razavi, 2017b).
According to Table 1, the effect of thermal treatment on the n p values was significant (p < 0.05). At the concentration of 0.75%, the n p values decreased with increasing temperature, but in 0.5% and 1% BSG solutions, firstly, with increase in temperature from 20 to 60°C, n p values were increased and then, from 60 to 121°C, were decreased.
The consistency coefficient (k p ), which is related to viscosity, of power law model varied from 0.301 Pa.s n (0.5%-20°C) to 1.510 Pa. s n (1%-121°C). At the same temperature, k p increased significantly (p < 0.05) with concentration enhancement. At the concentration of 0.75%, the k p values increased with increasing temperature, but in 0.5% and 1% concentrations, heat treatment did not have a distinct effect on k p . However, at the same concentration, the highest k p values were observed at 121°C, which was in agreement with the results reported by Zameni et al. (2015) and Naji-Tabasi et al. (2016) for BSG and cress seed gum, respectively. These results illustrated that the thermal treatment creates an irreversible intermolecular arrangement in BSG that would participate to raise the viscosity. Table 1 shows that Herschel-Bulkley model specified high R 2 (0.9991-0.9999) and low RMSE (0.0265-0.1434). Herschel-Bulkley model presented a small yield stress (τ 0H ) in PBSG, which is a notable parameter when gum is used as binders. The maximum and minimum τ 0H were obtained for 1%-121°C (0.439 Pa) and 0.5%-20°C (0.1 Pa), respectively. The results obtained for k H and n H parameters of the Herschel-Bulkley model were as almost the same as the results determined by the power law model, but lower and higher values were observed for k H (0.240-1.281 Pa.s n ) and n H (0.545-0.649) in comparison with k p and n p , respectively.
Random coil conformation of BSG was stated by Naji-Tabasi et al. (2016). Polysaccharides with random coil conformation have two Newtonian regions at low and high shear rates that called zeroshear viscosity (η 0 ) and infinite-shear viscosity (η ∞ ). The apparent viscosity of the PBSG solutions was investigated by the Carreau model, and its constant parameters are displayed in Table 1  and η ∞ values increased with increasing temperature (similar trend observed for k P and k H , as presented in Table 1). But different treatments did not have a significant effect (p > 0.05) on λ and N of PBSG solutions.

| Time-dependent rheological properties
Time-dependent rheological behavior of PBSG solutions is shown in Figure 2. Results indicate that the viscosity (Figure 2a (Table 2).
Similar to η 0 and k P , under the same temperature, ′ 0 increased significantly (p < 0.05) with increasing concentration, and at the constant concentration, generally, the ′ 0 values increased with increasing temperature. Table 2 shows that the rate constant of thixotropic breakdown, k, of the 1% PBSG-60°C was significantly higher than others indicating weak connection between its chains. The extent of structural breakdown is defined by the ratio of ′ 0 to � ∞ ). The � 0 ∕ � ∞ obtained for 0.5% PBSG-100°C was slightly higher than other samples indicating that the attendance of both fractions outputs in extent of thixotropy. In this condition, it could be expected to have stronger gel structure and fundamental reformation occurred (Naji-Tabasi & Razavi, 2017b).

Weltman model
As shown in Table 2 (Table 3).
The values of A and B were increased significantly (p < 0.05) with increasing concentration. Also, at 0.5% and 0.75% PBSG, different thermal treatments had not any meaningful difference (p > 0.05), but at the 1% solutions, the values of A and B parameters in 60°C were significantly lower than others. The highest A and B values were obtained for the 1% PBSG and thermal treatments of 20, 100, and 121°C samples, indicating higher structural breakdown rate by shearing at higher concentration. These data correspond thoroughly with the equivalent parameter acquired by the structural kinetic model,  (Table 2), and were consistent with the results of Razavi and Karazhiyan (2009) for Balangu and Salep solutions.

First-order stress decay, with a nonzero stress value
There was a good settlement between the fitted results of firstorder stress decay model, with a nonzero stress, and shear stress data for PBSG solutions (0.883 < R 2 < 0.982) ( Table 2). In general, the equilibrium stress (τ eq ) and initial stress (τ 0 ) values increased as concentration and temperature enhanced. The difference of initial to equilibrium shear stresses (τ 0 −τ eq ) is used as a comparative index of the postponed structural breakdown or the amount of thixotropy. The value of (τ 0 −τ eq ) increased at transition concentration from 0.5% to 1%. The decay rate constant, k, is an indication of how fast the hydrocolloid solution through shearing reaches the equilibrium stress value. Herein, k did not show a defined trend with temperature and concentration (Table 2), which is compatible with the results of Karazhiyan et al. (2009) for Lepidium sativum seed gum. 1% PBSG-60°C sample had the highest k value (0.0710 s −1 ).

| Stress sweep
The amplitude sweep was carried out over the stress range 0.01-40 Pa, 1 Hz frequency, and 20°C to determine LVE region. The LVE region could be used as an indicator of gel strength, so that stronger gels in comparison with weak gels have a more extensive LVE region (Steffe, 1996). From Figure 3 and Table 3, two different regions were classified: a LVE region in which G′ and G″ were unchanging with G′ > G″ (solid-like behavior), and a nonlinear region in which G′ and G″ start to reduce with stress enhancement. The LVE region for 0.5% PBSG in all temperatures was more limited which indicates weaker structure of them. After the crossover point, G″ was higher than G′ and the samples indicated liquid-like behavior. Therefore, the PBSG solutions showed a gel-like structure (a weak gel) at 1 Hz and 20°C. As shown in Figure 3a,b, at the same temperature, G′ and G″ were increased as the concentration is enhanced from 0.5% to 1%. Hence, shear stress increased with raise in PBSG concentration and reached the crossover point. At the same concentration, the storage mod- ) and loss modulus (G � (LVE) ) generally increased significantly with increasing temperature (Table 3). As shown in Figure 3 and Table 3, G � (LVE) of 1%-100°C and 1%-121°C solutions was higher than other samples, which reflects that these two samples have higher intermolecular interactions and entanglements.
Loss tangent (tan δ LVE ) showed the ratio of G″ to G′ in each cycle.
Basically, tan δ < 1 indicates elastic behavior and tan δ > 1 indicates viscous behavior. The tan δ > 0.1 means that the samples have a structure between real gel and high concentrated biopolymer. The tan δ values of PBSG solutions (0.31-0.42) were lower than 1, but higher than 0.1 ( Table 3) that showed elastic structure in weak gel (Naji-Tabasi & Razavi, 2017b). Because all PBSG solutions are not true gel, macromolecules connections and chains entanglements are provisional and could be disunited by using high shear rates. These results approved the shear-thinning behavior of the PBSG solutions in the steady-state experiments (See Figure 1 and flow behavior index in Table 1).
As shown in Figure 3, by increasing the shear stress, G′ and G″ started to reduce. The initial rupture (τ y ) and flow point (τ f ) can be assumed as dynamic yield stress (Table 3). The limiting value of LVE range in a stress sweep test is considered as yield stress or stress point (τ y ). τ f is the stress at the crossover point and shows resistance to flow. τ y and τ f parameters exhibited the same trend, so that, both parameters, at the same temperature, increased with increasing concentration, meaning that the gel network got stronger. At constant concentration, the highest values of τ y and τ f parameters were TA B L E 2 The second-order structural kinetics (n = 2), Weltman and first-order stress decay with a non-zero stress value models parameters determined for the PBSG solutions  pened. G f is a beneficial characteristic of gums to show their capacity to maintain the food texture . Treatments significantly influenced on the G f (

| Frequency sweep
The information of frequency sweep declares that dispersions could be categorized into four groups: dilute solution, concentrated solution, weak gel, and strong gel (Steffe, 1996;Vasile, 2009). For dilute solutions, G″ > G′ and is almost similar to each other at higher frequencies. In concentrated samples, G″ is bigger than G′ at low frequency and the crossover of them happens in the middle of frequency span. For gels, always G′ > G″ in frequency range, so that, strong gel is approximately independent of frequency, but in a weak gel, dynamic modulus is largely dependent on frequency (Kutz, 2013).
Dynamic frequency sweep is performed in the LVE limit to specify the frequency dependence of G′, G″, complex viscosity (η * ), tan δ, and the slope of η * of the PBSG solutions. Figure 4 reveals the mechanical spectra achieved by frequency sweep test, and the rheological parameters obtained from this figure are reviewed in Table 4. Results showed that PBSG has characteristically gel-like behavior, with G′ exceeding G″ in all frequency ranges. Moreover, in the frequency range findings, G′ and G″ did not crossover the other one. Moreover, samples with higher concentration exhibited an enhancement in G′ and G″ and a larger gap between G′ and G″ shows the ability to create macromolecular networks. Ross-Murphy (2012) stated that rise in G′ and G″ while they are parallel together is associated with the network defects, that is to mean, in lower concentration, the high intermolecular zones could not take part in noncovalence cross-junctions (Rincón, Muñoz, De Pinto, Alfaro, & Calero, 2009;Ross-Murphy, 2012), but the number of junction zones created at higher concentrations (1%) is maximum. Similar observations have been reported for psyllium gel (Farahnaky, Askari, Majzoobi, & Mesbahi, 2010) and L. perfoliatum seed gum gel (Hesarinejad et al., 2014).
In general, increase in temperature from 20 to 121°C increased the G′ and G″, except at the 1% PBSG that the values of G′ and G″ in 60°C were lower than 20°C, and also at 0.75% PBSG, G′ and G″ values for 121°C were lower than 100°C. Besides, the G′ was constantly higher than G″ in all frequency ranges. This means that solutions could display a weak gel behavior and their structures are not very susceptible to temperature alterations.
The tan δ was in the range of 0.30-0.41 (Table 4) which specifies weak gel structure in samples. At the constant temperature, the tan δ value for samples with high concentration was lower indicating these samples can represent a behavior between a weak and an elastic gel (Yoshimura, Takaya, & Nishinari, 1998). Also, at the same concentration, PBSG solutions at the 121°C had the lowest tan δ which is related to the establishment of stronger intertwined network. Similar results were obtained for psyllium gum (Farahnaky et al., 2010) and L. perfoliatum gum (Hesarinejad et al., 2014). non-Newtonian shear-thinning behavior (Figure 5a,b). Also, the η * increased with concentration enhancement from 0.5% to 1% (Table 4). The η * of PBSG had dependency to temperature; thus, as the temperature increases from 20 to 121°C, enhancement of η * was observed, except sample of 1%-60°C that η * was lower than other samples, and also at 0.75%, η * value for 121°C was lower than that of 100°C.

| Test of the Cox-Merz rule
According to Cox and Merz (1959), the η * and η a parameters should be identical if gum solution do not have any energetic interactions. On the contrary, irregular biopolymers, hydrocolloid solutions with firm conformation, and regular chains show a structured liquid behavior and do not follow this law (Naji-Tabasi & Razavi, 2017b). Figure 5 shows that the apparent viscosity, η a , in steady shear and the complex viscosity, η * , in dynamic shear were plotted against the shear rate, , and the frequency, ω, respectively. According to this figure, the η * was always higher than the η a ; therefore, PBSG solutions did not obey Cox-Merz rule and exhibit gel-like behavior, except of 0.5%-60°C that at higher shear rate or angular velocity, η * became equals to η a , that is to mean, this sample followed Cox-Merz rule (Figure 5b). The deviation from Cox-Merz rule is due to various molecular rearrangements occurring in the flow patterns over the frequency range or practical shear rate (Richardson & Ross-Murphy, 1987). As demonstrated in Figure 5a,b, the difference between η * and η a increased with the increase in the concentration. Similar trend was observed with increasing temperature from 20 to 121°C. Generally, at low frequencies, in all samples, the deviation from the Cox-Merz rule was higher.

| CON CLUS IONS
Herein, steady and dynamic flow characteristics of PBSG in the LVE region as a function of concentration and temperature were studied.
The Herschel-Bulkley and second-order structural kinetic models F I G U R E 5 Combined plot of complex viscosity (η * ) and apparent viscosity (η a ) against angular frequency/shear rate (Cox-Merz plot) for PBSG solutions [Correction added on 3 May 2019, after first online publication: Figure 3, 4, and 5 have been replaced with the correct images.] indicated well the shear dependency and time dependency of PBSG solutions, respectively. Storage (G′) and loss (G″) moduli as a function of frequency showed that the solutions revealed the rheological behavior similar to weak gel-like macromolecular dispersions with G′ higher than G″ in all frequency ranges. Both G′ and G″ behaviors were dependent on temperature and concentration. Above results suggested that this hydrocolloid has a good potential to use in formulation of food products.

ACK N OWLED G M ENT
We are grateful to Sari Agricultural Sciences & Natural Resources University (SANRU) for financial support under Project No. 02-1397-04.

CO N FLI C T O F I NTE R E S T
The authors declare that they do not have any conflict of interest.

E TH I C A L S TATEM ENT
This study does not involve any human or animal testing.