Adsorption behavior of selected monosaccharides onto an alumina interface

Journal of Colloid and Interface Science 270 (2004) 21–28 www.elsevier.com/locate/jcis

Adsorption behavior of selected monosaccharides onto an alumina interface Kaman Singh ∗ and Sudhanshu Mohan Physical Chemistry Department, National Sugar Institute, Ministry of Food and Consumer Affairs, Kanpur 208 017, India Received 25 November 2002; accepted 5 May 2003

Abstract The adsorption of glucose and fructose from their aqueous solutions onto an alumina interface has been carried out spectrophotometrically at room temperature. The adsorption isotherms are characterized as typical L-type and an adsorption mechanism on the basis of dipolar interactions has been suggested. In addition to this, a partial role of metal–saccharide interactions as found in organometallic complexes (OMCs) for the observed adsorption cannot be ruled out. Various kinetic and thermodynamic parameters of the adsorption process have been evaluated. The effects of variation in experimental conditions of the system have also been investigated. The adsorption exhibited a typical response to the pH effect and maximum adsorption was found near the isoelectric point of alumina (pH 9.0). The anionic addition 3− to the suspension affects the adsorbed amount and Cl− , SO2− 4 , and PO4 affect the adsorption quantitatively. The addition of similar concentration of cations was found to reduce the adsorbed amount. The presence of cationic and anionic detergents was found to influence both the adsorbed amount and the adsorption rate. The temperature was found to have an inverse effect on adsorption. Adsorptive kinetic parameters have revealed that fructose tends to be a better adsorbate than glucose. This is found to be consistent with the chelation behaviour of monosaccharides as found in the OMC of monosaccharides. The thermodynamics of the adsorption model indicates its spontaneous and exothermic nature. The negative values of entropy are an indication of the probability of a favorable nature of adsorption.  2003 Elsevier Inc. All rights reserved. Keywords: Adsorption kinetics; Alumina–monosaccharide model; L-type isotherms; Separation of carbohydrates; Chromatographic method

1. Introduction The adsorption phenomenon has acquired many applications in technological and biological fields since its discovery. The physical chemistry of removal of colorants or even colloids involves an adsorption mechanism. The chromatographic separation of azoyl derivatives of sugar has been studied by Colemen and McCloskey [1]. The azoyl esters, in solution, may be adsorbed onto a column of appropriate inert material such as alumina, silica, silicic acid, or magnesium silicate, and a colored band is obtained. If a solvent is passed through the column, the band will move down the column. The adsorbed azoates of a mixture of sugars may be separated by passing solvent through the column. The bands representing the least easily adsorbed substances * Corresponding author. Present address: Department of Chemistry,

Government Autonomous Postgraduate Leading, College, Tikamgarh 472001, India. E-mail address: [email protected] (K. Singh). 0021-9797/$ – see front matter  2003 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2003.05.002

pass down the column first, and under good conditions a separate, distinct band is obtained for each component of the mixture. The bands may be separated mechanically by breaking the column; the azoates are recovered by elution from the adsorbent by extraction with a solvent. The method has been successfully applied [1] to the separation of mixtures such as glucose and fructose; glucose and cellobiose; and arabinose, glucose, trehalose, and collobiose. Adsorption analyses have been applied [2] directly to sugar mixtures for their identification and separation. Much progress has recently been made in the study of the kinetics of adsorption of polymers [3–7], proteins [3–5], dyes [8–10], and drugs [11] onto various surfaces. Evidently, less experimental work has been done on investigations directly on carbohydrates, which play an essential role in the metabolism of living organisms. However, bauxite [12] has been used in the clarification of cane juice for the removal of color in the sugar industry. The use of bauxite in clarification [13] of cane juice helps not only in the removal of color but also in diminishing the ash content of juice, i.e., removal

22

K. Singh, S. Mohan / Journal of Colloid and Interface Science 270 (2004) 21–28

of Ca2+ , Fe3+ , and K+ . The adsorption behavior of a disaccharide is relatively complex and it seems more reasonable to start with a monosaccharide. A preliminary attempt on adsorption of monosaccharide was initiated by Bajpai and co-workers [14,15]. However, adsorption of numerous monosaccharides onto metallic surfaces is not evident at all. The present study is intended to explore the future possibility of utilizing these data (adsorbent) in various separation process in the sugar industry in addition to demineralization and decolorization of sugar cane juice. In order to have a better understanding of the adsorptive kinetic aspect of various sugars an extensive comparative study is essential. In the present investigation, therefore, fructose and glucose have been chosen for the adsorption studies as a representative selection of monosaccharides. The present work consists of an experimental kinetic study of adsorption of the selected monosaccharides, glucose and fructose, onto alumina with the object of exploring optimum conditions for separating carbohydrates chromatographically. As regards the potential importance of this work, not only are carbohydrates extremely important as food, but they are also vital molecules in the biological field which hold the key to heredity and life itself. Thermodynamic parameters of adsorption were also evaluated to predicting the nature of the adsorption.

2. Experimental Alumina of chromatographic grade (B.D.H.) was employed as adsorbent with surface area 18.0 m2 g−1 , estimated employing the BET technique using nitrogen gas as an adsorbate. Samples of glucose and fructose (B.D.H) were used to make at aqueous solution of pH 6.52. Solutions were prepared in conductivity water. In order to quantify the adsorbed amount, the spectrophotometric method was employed, using a UV–VIS (Shimadzu-260) spectrophotometer which displays a signal on a CRT screen. It covers wavelengths from 190 to 900 nm. The light sources is a 50-W tungsten halogen lamp and a deuterium lamp. It has a resolution of 0.1 nm, wavelength display is readable to 0.1 nm, wavelength accuracy is ±0.3 nm, and wavelength reproducibility is ±0.1 nm. 2.1. Determination of surface charge In order to determine the point of zero charge (PZC), the procedure [16] reported in the literature was employed. The PZC of alumina in the presence/absence of Cl− and SO2− 4 ions was determined employing potentiometric titrations of a suspension containing alumina (0.2 g) and 1.0 × 10−3 l−1 mol KNO3 as the supporting electrolyte in 50 ml solution. The PZC was found to be 9.0. It was observed that even in the presence of Cl− and SO2− 4 ions at the studied concentration range PZC did not change. A similar type of independence of the PZC of various oxides from the added anion has been widely reported in the literature [17].

2.2. Study of adsorption kinetics Batch-mode adsorption experiments were conducted. In a series of stoppered flasks, an accurately weighed amount of alumina (0.2 g) was mixed with a known volume (25 ml) of monosaccharide solutions of known concentration and pH (6.52) were shaken till the equilibrium adsorption was obtained. After definite intervals of time, the adsorbent was separated and the clear solutions were analyzed spectrophotometrically [18] after filtration for the residual monosaccharide content at λmax 550 and 570 nm for fructose and glucose, respectively, employing a UV-265 Shimadzu spectrophotometer. For kinetic measurements similar experiments were repeated and the progress of the reaction was recorded at definite time intervals as described above. The amount of adsorbates was then calculated. 2.3. pH measurements Monosaccharide solutions of known concentrations were adjusted to a known pH value (6.5) using 0.1 M HCl/NaOH solutions. To these known volumes of solutions alumina (0.2 g) was added. Immediately after separation of the adsorbent, the pH of the clear solution was measured and it was reported as the solution pH. It is to be noted that the pH of the solution varied after mixture of the adsorbent with adsorbate solutions. The extent of variation depends on the initial pH. However, in the entire pH range studied, variation were found to be very small. In addition, at the end of the equilibrium, pH was recorded and the difference was found negligible.

3. Results and discussion 3.1. Influence of concentration and adsorption isotherms The influence of increasing concentrations of monosaccharide solutions on adsorption has been investigated by varying it in the range of 5.56 to 55.6 × 10−4 mol l−1 . The results shown in Fig. 1 illustrate that the amount of adsorption increases with increasing concentration of adsorbates and finally becomes constant (levels off). This is due to the fact that with increasing concentration of monosaccharides, the availability of carbohydrates at the interface increases, which in turn enhances the adsorption; finally, when all the available sites on the adsorbent surface are occupied, it becomes difficult for monosaccharides to interact with the alumina surface, resulting in saturated adsorption. Adsorption isotherms shown in Fig. 1 illustrate that in the present case the adsorption isotherms resemble a typical Langmuirian plots [7,19] and indicates that the adsorption process follows a Langmuir isotherm. In fact, in the L2 type adsorption isotherm adsorption should be linear up to 2 h. In the present case, however, it is not so.

K. Singh, S. Mohan / Journal of Colloid and Interface Science 270 (2004) 21–28

23

constant-rate region [22] the rate constants for adsorption (k1 ) and desorption (k2 ), may be calculated using Langmuir’s isotherm equation [21], according to which the rate of adsorption (Rad ) is given by Rad = k1 C(1 − θ ) − k2 θ,

(2)

where k1 , k2 , C, and θ have their usual significance. Now, in order to calculate the rate constant for adsorption (k1 ), the following assumptions [15] can be adopted: (i) The rate of adsorption may be approximated by the rate of decrease in concentration of the adsorbate solutions; i.e., Fig. 1. L-type adsorption isotherms illustrating adsorption behavior of monosaccharides onto an alumina interface: Al2 O3 = 0.2 g, pH 6.52, temp. 27 ± 1 ◦ C.

It is a fait accompli [20] that the progress of adsorption is a two-regime phenomenon. At initial stages, the substrate surface is bare and the kinetics of adsorption is governed by diffusion of the molecules from bulk to the surface. All the molecules that arrive at the surface are assumed to be immediately adsorbed. In the later stages, an energy barrier against adsorption exists which slows down the adsorption rate considerably. 3.2. Adsorption coefficients (K = k1 /k2 )

Rad = −dC/dt.

(3)

(ii) In the initial (obviously linear) portion of the adsorbed amount versus time curve, the rate of desorption may be neglected as the adsorption is dominant in the initial periods, i.e., k2  k1 , and, as a consequence, the second term on the right-hand side of Eq. (2) can be ignored in comparison to the first term; i.e., k1 C(1 − θ ) k2 θ . (iii) The surface coverage (θ ) may approximated as θ = q(t)/qeq = (C0 − C)/(C0 − Ce ),

(4)

where q(t) and qeq are the amount adsorbed at time t and at equilibrium and C0 is the initial bulk concentration of adsorbate solutions. Thus, following the above simplification, we can write Eq. (2) as C − Ce dC = k1 C . dt C0 − Ce

Adsorption coefficient K was evaluated using the well known Langmuir equation [21],

−

KCe as , 1 + KCe KCe as 1 + KCe = a or 1 Ce Ce = + , a as K as

If equilibrium adsorption arrives at a much later stage then the numerical value of Ce may be ignored in comparison to those of C0 and C (as an approximation), and Eq. (5) is reduced to

a=

− (1)

where Ce = equilibrium concentration of the adsorbates, K = adsorption coefficient, a and as = adsorbed amount of adsorbate (mg g−1 at equilibrium and at saturation, respectively). From the ratio of slope to intercept of the linear plots drawn between Ce /a and Ce (Eq. (1)), the values of adsorption coefficients have been calculated to be 1.49 × 102 and 1.99 × 102 l mol−1 for glucose and fructose, respectively.

4. Kinetics of adsorption Adsorption kinetics was studied by monitoring the progress of the adsorption process at different time intervals and the rate of adsorption was found to be almost constant up to nearly 60 min. It then decreased gradually with time and finally attained a constant value near 100 min. Thus, from the

dt dC = k1 . C0 C2

(5)

(6)

Upon complete integration of Eq. (6), we get k1 1 1 = t+ , C C0 C0

(7)

where C = concentration of adsorbate solution at any time t, C0 = initial concentration of adsorbate. The adsorption coefficient (K) was first calculated by using the linearized Langmuir equation (1). The ratio of the slope to the intercepts of the linear plots between Ce and Ce /a (figures not shown) also furnishes the numerical values of the adsorption coefficients. Obviously, from the plot between 1/C and t, the value of k1 may be calculated using Eq. (7). Once k1 is known, the value of k2 may be found, since k1 /k2 has already been evaluated by Eq. (1). Many other investigators [16,23,24] have also used the value of K as (k1 /k2 ), as adopted in the present study. The superiority of this method for evaluating the kinetic parameters lies in the fact that no

24

K. Singh, S. Mohan / Journal of Colloid and Interface Science 270 (2004) 21–28

Table 1 Adsorption and kinetic parameters for the adsorption of selected monosaccharides onto an alumina interface Al2 O3 = 0.2 g, pH 6.52, temp. 27 ± 1 ◦ C Glucose

Fructose Rate const. ×104

Adsorption capacity Adsorption coeff. (l mol−1 )

Expt.

Graphical

(Mg g−1 )

1.49 × 102

22.4

Rate const. ×104

Adsorption capacity

Desorption k2 (s−1 )

Adsorption coeff. (l mol−1 )

Expt.

Graphical

(Mg g−1 )

Adsorption k1 (ml mg−1 s−1 )

(mg g−1 )

21.6

17.31

0.1172

1.99 × 102

26.2

complicated mathematical computations [25] (such as in the Runge–Kutta method or Marquardt’s optimization routine) are required and a simple linear plot will serve the purpose well. The adsorption and kinetic parameters for the adsorption of the said monosaccharides onto an alumina interface have been summarized in Table 1. 4.1. Mode of adsorption It has been reported [26] that in solution, monosaccharides behave as weak acids: pKa

+ C6 H12 O6 −→ C6 H11 O− 6 +H , Glucose 12.87 pKa

+ C6 H12 O6 −→ C6 H11 O− 6 +H . Fructose 12.60

Since the point of zero charge (PZC) of the alumina used in the present investigation is 9.0 and the experimental pH is 6.52, under these experimental conditions, on the surface of alumina, protonated aluminol groups (–AlOH+ 2 ) must be present with unprotonated aluminol groups (–AlOH): − AlOH + H2 O = AlOH+ 2 + OH .

(8)

Under these experimental conditions, the occurrence of different types of interactions between alumina surface and monosaccharide molecules may be assumed if the structures of monosaccharides are taken into considerations: 1. The partially negatively charged oxygen atoms of the sugar hydroxyls (–OH groups) interact with the protonated aluminol groups (–AlOH+ 2 ) of the alumina surface by electrostatic forces. δ− δ+ δ− AlOHδ+ 2 + 6 OH12 C6 → AlOH2 . . . 6 OH12 C6 .

(9)

↑ Electrostatic attraction 2. The partially negatively charged oxygen atom of the monosaccharide molecules may establish hydrogen bonding (HB) with the AlOH groups on alumina surfaces, resulting in the adsorption (amount) of monosaccharide onto the alumina surface: AlOH + 6 OH12 C6 → AlOHδ− . . . δ+ 6 OH12 C6 . ↑ Hydrogen bonding (HB)

(10)

(mg g−1 )

Adsorption k1 (ml mg−1 s−1 )

Desorption k2 (s−1 )

25.8

19.8

0.9932

3. Beyond pH (9.0), the surface of alumina become negatively charged due to formation of AlO− groups on its surface: AlOH = AlO− + H+ .

(11)

There will be repulsion between the AlO− and negatively charged oxygen atoms of these monosaccharides. This consequently reduces the amount of adsorption: AlO− + 6 OH12 C6 → AlO− . . . − 6 OH12 C6 .

(12)

↑ Electrostatic repulsion The hydrogen atoms in the –OH groups of monosaccharides bear a positive partial charge which would also result in an electrostatic repulsion between monosaccharides and positively charged alumina surface. Potentiometric [27] determination of the ionization constant has revealed that the glucose and fructose can only be treated as a monobasic not as a dibasic acid, and the values of the ionization constant (K) of the monosaccharide were of the same order as found for other aliphatic compounds. The average values of pK for glucose and fructose have been reported [27] as 12.87 and 12.60, respectively. It is instructive to note that the value of pK a obtained from the modified Debye–Hückel equation for glucose and fructose are in close agreement with potentiometric data [27]. Hence pK a values clearly implied that sugars have to be partially dissociated at the pH of the study (6.5), and above (9.0) there should be an appreciable ionization of monosaccharides, which strengthens the belief that the electrostatic attraction represented by Eqs. (10) and (11) are the correct working mode of interaction. In addition to the above possible interactions, other interactions may also be responsible for the observed adsorption. Aside from the fact that the hydrogen atoms in the OH groups of monosaccharides bear a positive partial charge, which would also result in an electrostatic repulsion between the monosaccharide and the positively charged alumina surface, such forces are barely sufficiently strong to explain the adsorption phenomena. Other concepts have to be considered carefully to explain the adsorption behavior. A structure such as AlOH+ 2 simply corresponds to water coordinated to Al+ , and glucose and fructose can basically coordinate

K. Singh, S. Mohan / Journal of Colloid and Interface Science 270 (2004) 21–28

to surface aluminum atoms by substitution of the coordinated water. In such structures, the coordinating hydroxyl groups of monosaccharides can be deprotonated as found in organometallic complexes (OMCs) of monosaccharide. Thus, metal–saccharide complex interaction/formation may be another possibility for the occurrence of adsorption of monosaccharides onto an alumina surface. Such visualization appears at a glance a bit unreasonable. However, the gustatory property of the adsorbed monosaccharide (material) on the alumina interface revealed that the material is slightly salty with no metallic taste whatever, which gives evidence for metal–saccharide interaction. The well established [28] OMCs of monosaccharides further give confidence in the metal–saccharide interaction responsible for the occurrence of adsorption. This fact has also been verified with a reappraisal of the spectral evidence for chelate formation [29,30]. A characteristic shift in absorbance occurs when partial chelation has taken place, and results from such studies must be used as an indication of metal–saccharide interaction. Charley et al. [28] described the preparation of these the chelates using ferrous and ferric irons with fructose. They also investigated some properties of chelate and proposed a possible structure for it, which they represented in the following way:

25

Fig. 2. Plots showing the variation of surface coverage with time at various initial concentrations of D-fructose: Al2 O3 = 0.2 g, pH 6.52, temp. 27 ± 1 ◦ C.

4.2. Surface coverage (θ ) The fraction of the surface occupied by the adsorbate at different time intervals may be calculated employing the equation θ = (C0 − C)/(C0 − Ce ), where Ce and C have the same meaning as in Eq. (4). The amount of surface coverage for different initial concentrations of adsorbates has been calculated up to 60 min. Results with fructose are shown in Fig. 2 as a representative selection of monosaccharides. The results clearly illustrate that the surface coverage increases progressively with time. Obviously, as time increases, a greater number of monosaccharide molecules interact with the binding sites on the alumina surface resulting in increased amount of adsorption. Figure 2 also illustrates that the surface coverage (min−1 ) also increases with increasing concentration of fructose.

In fact, it seems doubtful that such a model could exist due to the bulky and mainly cyclic nature and ring size of the monosaccharide molecule, which prevent the accommodation of such molecules in place of relatively smaller hydroxyls (–OH groups) on the surface of alumina. However, monosaccharides in their open chain forms would be unlikely to form such a structure. Angyal [31] described the formation of chelates between carbohydrates and metal ions and found that carbohydrates with an axial–equatorial– axial sequence of three oxygen atoms, on a six-membered ring, in the open chain structure, formed good sites for complex formation with cations. Aasa et al. [32] claim to have confirmed this structure by ESR and NMR studies of the chelate. Hence, the contribution of metal–saccharide interaction to observed adsorption cannot be ruled out. The photoelectron spectroscopy (PES) and X-ray crystallography of the adsorbed mass could be used to obtain this information; however, this is beyond the scope of the present investigation.

4.3. Factors affecting adsorption 4.3.1. Effect of pH In the titled adsorption model, since the adsorbates are nonpolar and the surface is polar in nature, the effects of variation in pH of the suspension will hardly affect the charge profile of the monosaccharides. However, a remarkable change in charge density over the alumina surface definitely occurs which consequently affects the adsorbed amount. In the present study, the effect of pH has been investigated by varying the pH in the range 2.0 to 12.3. The results reveal that the amounts of adsorbed adsorbates increase with increasing pH of the suspension and attain a maximum adsorption (amount) at pH 9.0. However, there is a decrease in adsorbed amount beyond pH 9.0. At the experimental pH (6.52), which is quite below the isoelectric point of alumina [33] (9.0), the alumina surface acquires a net positive charge due to the presence of –AlOH+ 2 groups on its surface. The partially negatively charged oxygen of the

26

K. Singh, S. Mohan / Journal of Colloid and Interface Science 270 (2004) 21–28

Table 2 Adsorption and kinetic parameters at various pH for the adsorption of selected monosaccharides onto an alumina interface, Al2 O3 = 0.2 g, pH 6.52, temp. 27 ± 1 ◦C pH 2.0 3.5 4.0 6.9 9.0 12.3

Glucose

Fructose

k1 (×104 ml mg−1 s−1 )

k2 (×105 s−1 )

k1 (×104 ml mg−1 s−1 )

k2 (×105 s−1 )

10.93 16.31 17.22 18.20 24.2 19.83

7.31 11.08 11.48 12.38 16.45 13.48

11.2 18.2 19.6 20.1 26.1 21.8

5.71 9.24 9.94 10.15 12.89 10.98

–OH groups in monosaccharide molecules interact with the positively charged surface. Hence the interionic attraction between the monosaccharide molecule and alumina surface increases which continues up to isoelectric point and at this point, the maximum quantity of adsorption is found. Kinetic parameters shown in Table 2 also support our experimental findings. However, above pH 9.0, the alumina surface acquires a negative charge due to AlO− groups and at this stage an electrostatic repulsion begins to operate between the AlO− groups of alumina and O−δ –H+δ groups of the monosaccharide molecules. Thus, with further increase in the pH, the amount of adsorption decreases. 4.3.2. Effect of low-molecular-weight ionic salts Salts when added to an adsorption system often respond via two mechanisms, viz., screening of charges in the solution or preferential adsorption of added ions to the surface. In the present study, the influence of the addition of salts on the amount and rate of the adsorption was investigated by adding different salts of the Na+ ion in the concentration range 0.01–0.10 M. The results are shown in Fig. 3 for

Fig. 3. Plots showing the effect of added anion concentration on the adsorbed amount of monosaccharides at fixed concentration (0.8 × 10−4 mol l−1 ).

fructose as a representative plot of monosaccharides, and they clearly demonstrate the increase in the amount of adsorbed monosaccharides with increasing salt concentration and obey the following sequence: 3− Cl− < SO2− 4 < PO4 .

Similarly, the cations were added in the same concentration range as anions and the effectiveness was found to obey the sequence Na+ < Ca2+ < Fe3+ . The observed increase in the adsorption of anions is surprising. In fact, there would be no specific adsorption because PZC did not change in the presence of anions such as Cl− and SO2− 4 . Earlier work [17] gives us confidence in the sense that the PZC of various oxides on the added anion remained unaltered. In a case study similar behavior was also observed for disaccharide (sucrose) adsorption under identical conditions. Bajpai and Choubey [15] have also reported similar findings. However, the results can be explained as follows: Salts when added to an adsorption system often respond via two mechanisms: the screening of charges in the solution or the preferential (selective) adsorption of added ions onto the surface. Since the surface is negatively charged above the PZC of alumina due to AlO− groups, there is a lower probability of the adsorption of the added ions onto the surface. Moreover, in that case a decrease in adsorption would be observed, which is not found in the present case. Therefore, it is the shielding effect of the added ions that influences the adsorption. The explanation may be that in the case of addition of anions, the electrostatic repulsion between δ+ the AlOH+ 2 groups of the alumina surface and the –OH groups of monosaccharides are screened by the added anions and the adsorbed amount increases accordingly. It seems that these anions compete with monosaccharides for the positively charged surface sites. It is also clear that the adsorption increases with the increasing anionic charge and, consequently, the PO3− 4 ions appear to be the most effective ones. Thus the observed order of effectiveness of anions is also justified. In the case of cationic addition, a decrease in the amount of adsorption has been observed, which is quite obvious as in the presence of cations, the electrostatic repulsion between the monosaccharides (–OHδ+ ) and the positively charged aluminol groups (AlOH+ 2 ) of the surface increases, which, in turn, results in a fall in the adsorbed amount. The observed

K. Singh, S. Mohan / Journal of Colloid and Interface Science 270 (2004) 21–28

27

Table 3 Effect of addition of salts on the rate constants of adsorption (k1 ) Salt

NaCl

Na2 SO4

Na3 PO4

No salt

Conc.

Glucose

Fructose

(M)

Rate const. (k1 × 104 ml mg−1 s−1 )

Adsorption rate (mg g−1 min−1 ) × 104

Rate const. (k1 × 104 ml mg−1 s−1 )

Adsorption rate (mg g−1 min−1 ) × 104

0.02 0.04 0.10 0.02 0.04 0.10 0.02 0.04 0.10 –

18.7 30.7 39.0 31.8 46.1 48.7 62.9 87.8 93.9 10.92

4.76 6.99 8.99 7.19 14.8 19.9 18.9 11.1 23.8 2.8

19.8 32.9 43.6 33.4 49.6 54.2 66.8 90.3 100.2 12.8

10.05 06.04 45.64 59.58 36.71 29.79 22.03 19.86 55.46 4.56

Fig. 5. Plots illustrating the exothermic nature of adsorption of monosaccharide onto an alumina interface at fixed concentration, 0.6 × 10−4 mol l−1 . Fig. 4. Representative plots illustrating the effect of surfactant on the plateau adsorption of monosaccharide (fructose) onto alumina at fixed concentration = 0.6 × 10−6 mol l−1 .

sequence of adsorption also seems to be consistent with the chelating power [28] of these ions with monosaccharide as found in OMCs. For a certain concentration of salts, values of rate constants and adsorption rate were evaluated and the results are summarized in Table 3. 4.3.3. Surfactant effect The effect of the cationic surfactant on the adsorption of monosaccharides was studied by adding cetyl trimethylammoniumbromide (CTAB) in the concentration 2.0–6.0 × 10−4 mol l−1 . The results with fructose are shown in Fig. 4; they illustrate that both the adsorbed amount and adsorption rate increase with the increasing concentration of the CTAB. The reason for the observed increase is quite apparent, as the added cationic surfactants may bind to the negatively charged sites on the monosaccharide molecules and thus reduce the electrostatic repulsion between the monosaccharide molecules and the alumina surface, which finally increases the adsorbed mass. To study the effect of the anionic surfactant, sodium oleate was added in the concentration range 2.0–6.0 × 10−4 mol l−1 and results are shown in Fig. 4. It is clear from

Fig. 4 that both the adsorbed mass and the adsorption rate decrease with increasing concentration of sodium oleate in the studied range. The results may be attributed to the fact that the presence of anionic surfactant molecules in the carbohydrate solution causes the electrostatic repulsion to increase, which reduces adsorption amount and rate. 4.3.4. Effect of temperature The effect of temperature on the adsorption behavior of the studied adsorption model has been studied in the range 5–45 ◦ C. In order to illustrate the result, a graph has been plotted of the adsorbed amount of monosaccharides against the temperature of the adsorption medium. Figure 5 demonstrates the exothermic nature of the process. The decrease of adsorption with rise of temperature may be due to the enhanced escaping tendency of the monosaccharide molecules from the surface of the adsorbent. With increasing temperature, the binding forces between the monosaccharide molecules and the surface are weakened and the adsorption decreases. Thermodynamic parameters were calculated as follows: (i) G0 was calculated using the equation [34] G0 = −RT ln b.

28

K. Singh, S. Mohan / Journal of Colloid and Interface Science 270 (2004) 21–28

Table 4 Thermodynamic parameters for adsorption of monosaccharides onto an alumina interface G0 H 0 S 0

Glucose

Fructose

−5.56 kcal mol−1 −6.67 kcal mol−1 −3.72 kcal mol−1

−7.10 kcal mol−1 −5.50 kcal mol−1 −5.33 cal g K−1

(ii) Apparent heat of adsorption enthalpy, H 0 , was evaluated using the equation ln(k2 /k1 ) = H 0 /2.303R × (1/T1 − 1/T2 ). (iii) Entropy, G0 , was estimated using the equation G0 = H 0 − T S 0 . The exothermic nature of the adsorption was again confirmed by the thermodynamic parameters as shown in Table 4. The negative value of G0 is an indication of the spontaneous nature of the process. The apparent enthalpy of adsorption, H 0 , also confirms the exothermic nature of the adsorption process. The negative value of S 0 suggested the probability of a favorable and complex nature [35] of adsorption for the titled model.

5. Conclusions The kinetical study of adsorption in an alumina–monosaccharide model resembles typical L-type adsorption isotherms and indicates that the adsorption process follows a Langmuir isotherm. Adsorption is supposed to occur via metal–saccharide interaction and electrostatic interactions between the aluminol groups of the alumina surface and oxygen atoms of monosaccharide molecules. The adsorption exhibits a typical response to the pH effect and maximum adsorption was observed near pH 9.0. The addition of anions to the suspension brings about an increase in the adsorbed amount. The alumina seems to be an effective adsorbent for the removal of glucose and fructose from their aqueous solutions. Ceteris peribus, adsorption of glucose onto alumina was found to be less than half obtained for the adsorbed amount of fructose and glucose onto alumina. The effectiveness of the added anions follows the 3− sequence Cl− < SO2− 4 < PO4 . The addition of the same concentration of cations was found to increase the adsorption in the sequence Na+ < Ca2+ < Fe3+ . The presence of cationic and anionic detergents was found to influence both the adsorbed amount and the adsorption rate. The thermodynamics of the studied adsorption model is an indication of the favorable and exothermic nature of adsorption.

Acknowledgments The authors are indebted to the Director, National Sugar Institute, Kanpur for facilities. The authors are thankful to the Head, Department of Chemistry, for providing some chemicals. One of the authors (K.S.) thanks Professor R.P. Sahu, Principal, Government Postgraduate Leading College, Tikamgarh for his moral support. References [1] G.H. Coleman, C.M. McCloskey, J. Am. Chem. Soc. 67 (1945) 381, 65 (1943) 1588. [2] B.W. Lew, M.L. Wolfrom, R.M. Goepp Jr., J. Am. Chem. Soc. 68 (1946) 1449. [3] J.D. Aptel, J.C. Voegel, A. Schmitt, Colloids Surf. 29 (1988) 359. [4] J.C. Dijit, M.A. Cohen Stuart, J.E. Hofman, G.J. Fleer, Colloids Surf. 51 (1990) 141. [5] A.V. Elgersma, R.L.J. Zsom, J. Lyklema, W. Norde, Colloids Surf. 65 (1992) 17. [6] S. Chibowski, J. Colloid Interface Sci. 140 (1990) 444. [7] C.H. Giles, T.H. MacEwan, S.N. Nakuhwa, D. Smith, J. Chem. Soc. 4 (1960) 3973. [8] G. McKay, M.S. Otterburn, A.G. Sweeney, Water Res. 15 (1981) 327. [9] G. McKay, M.S. Otterburn, J.A. Aga, Water Air Soil Pollut. 24 (1985) 307. [10] S.D. Khattri, M.K. Singh, J. Indian Chem. Soc. 76 (1999) 389. [11] A.K. Bajpai, R. Dengre, J. Appl. Polym. Sci. 60 (1996) 2219. [12] P. Honig, Principles of Sugar Technology, Elsevier, London, 1992. [13] K.S.G. Doss, N.S. Jain, J. Sci. Ind. Res. 5 (1946) 4. [14] A.K. Bajpai, M. Rajpoot, Adsorption 6 (2000) 349. [15] A.K. Bajpai, A. Choubey, J. Indian Chem. Soc. 77 (2000) 238. [16] A.K. Bajpai, M. Rajpoot, D.D. Mishra, J. Colloid Interface Sci. 187 (1997) 96. [17] S. Chibowski, J. Colloid Interface Sci. 140 (1990) 444. [18] C.A. Browne, F.W. Zerban, Physical and Chemical Methods of Sugar Analysis, Wiley, New York, 1949, p. 263. [19] J.A. Veith, G. Sposito, Soil Sci. Soc. Am. J. 41 (1997) 697. [20] D.F. Siqueira, J. Reiter, U. Breiner, R. Stadler, M. Stamm, Langmuir 12 (1996) 972. [21] I. Langmuir, J. Am. Chem. Soc. 40 (1918) 1361. [22] U.D.N. Bajpai, A.K. Bajpai, Polym. Int. 32 (1993) 43. [23] W.H. Thomas, V.K. La Mer, J. Colloids Sci. 19 (1964) 323. [24] J. Arnebrant, T. Nylander, J. Colloids Sci. 111 (1986) 529. [25] D.D. Ravethar, V.D. Ambeskar, R.A. Mashelkar, J. Appl. Polym. Sci. 39 (1990) 769. [26] T.S. Britton, Hydrogen Ions, Chapman & Hall, London, 1955, p. 216. [27] N.A. Ramesh, S.S. Katiyar, Sugar Tech. Assoc. India 29 (1961) 77. [28] P.J. Charley, B. Sarkar, C.F. Stitt, P. Saltman, Biochim. Biophys. Acta 69 (1963) 313. [29] M.W. Kearsley, G.G. Birch, Food Chem. 2 (1977) 209. [30] K. Singh, J. Indian Chem. Soc. 76 (1999) 49. [31] S.J. Angyal, Pure Chem. 35 (1973) 131. [32] R. Aasa, B. Malmstrom, P. Saltman, T. Vangard, Biochim. Biophys. Acta 80 (1964) 430. [33] I.G.A. Parks, Chem. Rev. 65 (1965) 177. [34] M. Doula, A. Joannou, Dimirkou, Commun. Soil Sci. Plant Anal. 1749 (1996) 27. [35] T.W. Biggar, M.W. Cheung, Soil Sci. Soc. Am. Proc. 37 (1973) 863.