The Study of Adsorption Kinetics of Flavan-3-Ols, Dihydrochalcones and Anthocyanins onto Barley β -Glucan

: Polyphenols can interact with dietary fibers and these interactions can affect polyphenols bioactivities. The interactions ca n be studied with the adsorption process, and adsorptions of flavan-3-ols (procyanidin B1, procyanidin B2), dihydrochalcones (phloretin, phloretin-2-glucoside), and anthocyanins (cyanidin-3-glucoside, cyanidin-3-galactoside) onto β -D- glucan from barley were studied in this work. The intention was to reveal the kinetics of the adsorption process. Adsorpti on was carried out using model solutions at room temperature. The results showed that in the flavan -3-ol group procyanidin B1 showed higher adsorption capacity than procyanidin B2. Phloretin showed a higher adsorption capacity than phloretin -2-glucoside in the dihydrochalcone group and anthocyanins showed similar adsorption capacities. Parameter k 1 for all polyphenols was in the range from 0.30 h − 1 to 0.93 h − 1 . Adsorption capacity q e for all polyphenols ranged from 2.90 mmol g − 1 to 9.76 mmol g − 1 , parameter k 2 was in the range 70− 225 g mol − 1 h − 1 , and adsorption capacities q e ranged from 3.70 mmol g − 1 to 9.90 mmol g − 1 .


INTRODUCTION
OLYPHENOLS are a group of organic compounds that naturally occur in plants.They can be classified into phenolic acids, flavonoids, lignans, and stilbenes. [1]Many potential bioactivities of polyphenols have been reported among which we highlight interactions between polyphenols and food constituents such as dietary fibers.These interactions can affect polyphenols' accessibility for absorption (bioaccessibility) in the digestive tract as well as the amount of polyphenols that can be absorbed and reach various organs (bioavailability). [2]-glucan is a dietary fiber that can be found in various foods and food supplements. [3]−6] However, β-glucan is one of the fibers that can interact with polyphenols. [7,8] better understanding of interactions between polyphenols and dietary fibers can be achieved by exploring the adsorption process and testing the suitability of different kinetics models.−11] The kinetic study is widely used in different applications such as the adsorption of catechin onto cellulose, [12] the adsorption of tea polyphenols on starch, [13] or the adsorption of various adsorbates from aqueous solution on different adsorbents. [14,15]Furthermore, the adsorption of polyphenols on different adsorbents can be investigated using the various kinetic models.
The aim of this research was to study kinetics of adsorption of various polyphenols onto β-D-glucan.The selected polyphenols (Figure 1) belong to the flavan-3-ols (procyanidin B1 and procyanidin B2), dihydrochalcones (phloretin, phloretin-2-glucoside), and anthocyanins (cyanidin-3-glucoside and cyanidin-3-galactoside).The experimentally obtained results were fitted by non-linear pseudo-first order and non-linear pseudo-second order P model in order to understand the adsorption kinetics of polyphenols onto β-D-glucan.To the best of our knowledge, only few papers dealt with modeling of adsorption kinetics of polyphenols, especially if adsorbent was β-D-glucan [9,16] Also, there are no papers of kinetic study with the selected polyphenols like the ones selected for our study, and β-D-glucan.Therefore, this work will help in better understanding the adsorption kinetics of flavan-3-ols, dihydrochalcones and anthocyanins on β-D-glucan and give useful data about the interactions between polyphenols and β-D-glucan.

MODELING
Models of adsorption kinetics can be classified into diffusion models and adsorption models.Diffusion models are considering the diffusion as the slowest step of the process which limits the rate of the entire process.The diffusion can be related to the mass transfer through the liquid film which surrounds particles of the adsorbent or to the mass transfer through the liquid that fills the adsorbent pores.Adsorption models are considering adsorption at the surface of the adsorbent as the slowest step of the process. [17]he most widely used adsorption kinetic models are pseudo-first and pseudo-second order model. [10,11,18]inetics of pseudo-first order reaction is following [Eq.(1)] while [Eq.( 2)] describes pseudo-second order behaviour [19] .
e 1 e k t q q (1) In [Eq.( 1)] and [Eq.( 2)], qt and qe represent amounts (mol) of an adsorbate adsorbed per mass (g) of adsorbent at reaction time t and at the equilibrium, respectively.Each equation has reaction rate constant, k, included.In order to facilitate later discussion, we marked pseudo-first-order rate constant as k1 and the pseudo-second order one as k2.
Pseudo-first order equation can be linearized [Eq.( 3)] Pseudo-second-order model can also be linearized [Eq.( 4)] = + 2 t e 2 e 1 1 t t q q k q (4) Two kinetics models: pseudo-first [Eq.( 1)] and pseudo-second order model [Eq.( 2)], were applied in this study in order to investigate the adsorption kinetic.The parameters of the models: qe, k1, and k2, were obtained by non-linear regression using tool Solver in MS Excel (Microsoft Corporation, Redmond, USA).
Solid β-D-glucan (95 %) obtained from barley was purchased from Sigma-Aldrich (St. Louis, USA).The stock solution was prepared in distilled water at a concentration of 190 mg L -1 .The solution was heated for 15 min at 80 °C and afterward stored in the refrigerator at 4 °C.
In order to retain the constant pH value of the solutions, a phosphate buffer of pH 5.5 was applied.Phosphate buffer was prepared with 0.1 M sodium phosphate dodecahydrate and 0.1 M sodium dihydrogen phosphate dihydrate.For the determination of total anthocyanins, solutions of pH 1.0 and pH 4.5 were used.pH 1.0 was 0.025 M potassium chloride, while the second one was 0.4 M sodium acetate trihydrate.

Performing the Adsorption Experiments
Adsorption of two flavan-3-ols (procyanidin B1 and procyanidin B2), two dihydrochalcones (phloretin and phloretin-2-glucoside), and two anthocyanins (cyanidin-3-glucoside and cyanidin-3-galactoside), was performed onto β-D-glucan.Total volumes of model solutions were 500 µL.Solutions consisted of β-D-glucan in the concentration of 5 mg L −1 and related polyphenol in the concentration of 100 mg L −1 (cinitial, mg L −1 ); the rest was the buffer solution of pH 5.5.All solutions were homogenized on vortex mixer and put in an incubator for 1, 2, 5, 8, and 16 hours at room temperature.At the end of the adsorption experiment, the solutions were filtered on Eppendorf Minispin centrifuge (Eppendorf, Hamburg, Germany) by using 100−500 µL Vivaspin 500 centrifugal concentrators with polyethersulfon membrane (Sartorius, Goettingen, Germany).Afterwards, the content of unadsorbed polyphenols was analyzed (c, mg L −1 ).Flavan-3-ols and dihydrochalcones were determined by Folin-Ciocalteu method for the determination of total phenolic content, while anthocyanins were analyzed by using the method for total anthocyanins. [20]Blank experiments considered experiments under the same conditions but without β-D-glucan, were conducted as well.Dihydrochalcones, flavan-3-ols and anthocyanins passed through the membrane in the blank experiment and they were determined by the use of spectrophotometric Folin-Ciocalteu method and by method for determination of total anthocyanins (co, mg L −1 ).The concentration of adsorbed dihydrochalcones, flavan-3-ols and anthocyanins (cadsorbed, mg L −1 ) was calculated as: The concentration of unadsorbed dihydrochalcones, flavan-3-ols and anthocyanins at equilibrium (ce, mg L −1 ) was calculated as: The amount of adsorbed dihydrochalcones, flavan-3ols and anthocyanins in millimoles per gram of β-D-glucan, q, was calculated according to [Eq. ( 7)].
where Vm is the total volume of the model solution (L), γa is the mass concentration of β-D-glucan (g L −1 ) and Va is the volume of β-D-glucan (L).

Determination of Flavan-3-ols and Dihydrochalcones
Solutions for the spectrophotometric determination of flavan-3-ol and dihydrochalcone contents by the Folin-Ciocalteu method were prepared in a glass tube by adding 1580 μL of distilled water, 20 μL of sample (polyphenol) solution after adsorption, 100 μL of Folin-Ciocalteu reagent, and 300 μL of 200 g L −1 solution Na2CO3.Solutions were mixed on a vortex mixer (Grant Bio, Cambridgeshire, England) and incubated 30 minutes at 40 °C.Afterwards, the solutions were analyzed on spectrophotometer UV-2005 (Selecta, Barcelona, Spain).The absorbance was measured at 765 nm against the blank solution. [20]Calibration curves were constructed for each polyphenol in concentration range 1-1000 mg L −1 for dihydrochalcones, 1−600 mg L −1 for procyanidin B1, and 1−300 mg L −1 for procyanidin B2 using the same procedure.Two replicate samples were prepared and analyzed three times for each polyphenol concentration.

Determination of Total Anthocyanins
For the determination of anthocyanin content after the adsorption, two solutions were prepared, each containing 500 µL of the solution after adsorption supplemented with 1500 µL of solution with different pH.The pH value of the first solution was 1.0 (0.025 M potassium chloride), while the second one had a pH value of 4.5 (0.4 M sodium acetate trihydrate).Both solutions were incubated for 15 min at a dark place.Afterwards, the absorbance was measured at 510 nm and 700 nm with a UV-Vis spectrophotometer against the blank solution.The true absorbance of anthocyanin (A) was calculated according to [Eq. ( 8)], where A510 and A700 were absorbances of prepared solutions measured at 510 and 700 nm, respectively.
The calibration curves were constructed for each anthocyanin in the concentration range 1−200 mg L −1 using the same procedure.Two replicate samples were prepared and analyzed three times for each polyphenol concentration.

RESULTS AND DISCUSSION
During the adsorption processes, polyphenols were adsorbing onto the surface of β-D-glucan.The adsorption process was monitored at different times in order to reveal the kinetics of the adsorption.
Different behavior in the adsorption process between flavan-3-ols, dihydrochalcones and anthocyanins onto β-D-glucan can be seen (Figures 2−4).Comparing flavan-3-ols, procyanidin B1 showed higher adsorption capacity (Figure 2).Phloretin showed higher adsorption capacity among the tested dihydrochalcones (Figure 3).The tested anthocyanins showed similar adsorption capacities (Figure 4).Different adsorption capacities of polyphenols can be related to the chemical structures of polyphenols. [21]t is possible that the special arrangement of OH groups in the case of flavan-3-ols affected the adsorption capacity.
Furthermore, glycosylation was less favored in the case of dihydrochalcones.In the case of anthocyanins, adsorption was similar.
Adsorption is a complex process [22] and for a better understanding, different kinetic models should be applied.Kinetic study can provide the information about the   reaction pathways and the mechanism of the adsorption reactions. [23]Pseudo-first and pseudo-second order models have been widely used to describe the kinetics of the adsorption process, [9,17,24] especially kinetic model of pseudo-first order for the description of the adsorption process of solute from the liquid. [10]The pseudo-first order model is one of the probably earliest kinetic models for the adsorption process.Pseudo-second order model assumes that the rate of adsorption or desorption process controls the overall kinetics.Pseudo-first and pseudo-second order models were also used in our study.Generally, linear or non-linear forms of these models can be applied. [17,19,25,26]owever, the linear forms of the kinetic models can lead to erroneous results due to the modelling error.Also, parameters obtained from non-linear and linear models sometimes do not show good agreement. [19]Therefore, we decided to apply non-linear form of the models using nonlinear regression with tool Solver.Figures 2−4 present the non-linear pseudo-first and pseudo-second order models for the adsorption of the tested flavan-3-ols, dihydrochalcones, and anthocyanins, respectively.The models were obtained by regression analysis.The analysis provided values of the model parameters (Table 1).Parameters k1 and qe obtained with pseudo-first order models were in the range from 0.30 h −1 to 0.93 h −1 , and from 2.90 mmol g −1 to 9.76 mmol g −1 , respectively.Parameter k2 and qe obtained with pseudo-second order model were from 70 g mol −1 h −1 to 225 g mol −1 h −1 , and from 3.70 mmol g −1 to 9.90 mmol g −1 , respectively.Adsorption capacities obtained from kinetic models for flavan-3-ol group was higher for procyanidin B1, then for phloretin in dihydrochalcones group and anthocyanins showed similar adsorption capacities.This is in accordance with experimentally obtained adsorption capacities (Figures 2−4).In order to see how the adsorption capacities of polyphenols obtained from kinetic models agree with experimental adsorption capacities, they were put in the correlation (Figure 5).We assumed that the adsorption equilibrium was reached after 16 hours for all tested polyphenols and, accordingly, took experimental and model adsorption capacities after 16 h into correlation.Due to the determination of the coefficient R 2 , pseudo-second order model showed better agreement with experimental

Figure 5 .
Figure 5.The correlation between adsorption capacities (qe) determined experimentally and from kinetics models of: a) pseudo-first order; b) pseudo-second order reactions.

Table 1 .
Parameters of pseudo-first and pseudo-second order models describing adsorption kinetics of six polyphenols onto β-glucan.Mark S represents standard error.

Table 2 .
Half-reaction time for pseudo-first and pseudosecond order model for adsorption of six polyphenols onto β-glucan.