Adsorption isotherms of non-ionic surfactants on Na-bentonite (Iran) and evaluation of thermodynamic parameters

Adsorption isotherms of non-ionic surfactants on Na-bentonite (Iran) and evaluation of thermodynamic parameters

Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113 Adsorption isotherms of non-ionic surfactants on Na-bentonite (Iran) and evalua...

544KB Sizes 10 Downloads 51 Views

Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

Adsorption isotherms of non-ionic surfactants on Na-bentonite (Iran) and evaluation of thermodynamic parameters M. Ghiaci ∗ , R.J. Kalbasi, A. Abbaspour Department of Chemistry, Isfahan University of Technology, Isfahan 84156, Iran Received 14 August 2006; received in revised form 7 October 2006; accepted 13 October 2006 Available online 20 October 2006

Abstract Non-ionic surfactants have been adsorbed on bentonite (Iran) as a relatively good adsorbent in various temperatures (297–323 K). The experimental data obtained from non-ionic surfactants on bentonite have been fitted into the adsorption isotherm equations such as: Langmuir, Freundlich, Langmuir–Freundlich (Modified Langmuir), Hill, Redlich–Peterson, BET, Halsey, Harkins–Jura, Smith, Henderson, Chaung-Pfost, Oswin, and some modified forms of these isotherms, and their goodness of fit have been compared. In addition, the data has been fitted into the Modified BET isotherm which considers the dependency of sorption on the temperature. The correlation coefficient (r2 ) obtained by fitting the experimental data to the Modified Langmuir, Hill, Redlich–Peterson, BET and Modified BET isotherms have been favorably high. The thermodynamic data of Gibbs free energy (G◦ ), enthalpy (H◦ ) and entropy (S◦ ) terms have been determined for the non-ionic surfactant adsorption; the negative values of G◦ obtained indicates that the non-ionic surfactant adsorption process is a spontaneous one. Finally, the intercalation of surfactant in the interlamellar space was followed by X-ray measurements. © 2006 Elsevier B.V. All rights reserved. Keywords: Non-ionic surfactant; Isotherm; Bentonite; Thermodynamic parameters; Modified Langmuir; Modified BET

1. Introduction Adsorption of non-ionic surfactants from aqueous solutions has been studied for years because of their industrial relevance. These surfactants, made of an aliphatic part attached to a hydrophilic and polar chain, play an important role in diverse applications ranging from ore flotation, paint technology, lubrication, paper making, oil recovery, and biochemistry of proteins. Relationships between adsorption layer organization and numerous properties such as colloidal stability or wetting characteristics are actively debated in the literature. The development of technologies for the decontamination of soils and waters polluted by hydrophobic or organic compounds has encouraged research into the use of anionic and non-ionic surfactants as potential agents for the enhanced solubilization and removal of contaminants from soils and sediments [1–3]. However, the effectiveness of a surfactant is attenuated if this is adsorbed by the soil, since the amount available for solubilizing



Corresponding author. Tel.: +98 311 3912351; fax: +98 311 3912350. E-mail address: [email protected] (M. Ghiaci).

0927-7757/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.colsurfa.2006.10.032

the contaminant decreases, and its mobility through the medium to which it is applied is reduced [4–6]. Furthermore, surfactant adsorption will increase the organic carbon content of the soil, favoring the adsorption of hydrophobic organic compounds. As a result of these processes, the adsorption of surfactants by soil components may lead to a significant reduction in their effectiveness to remove the contaminants from the soils. In this sense, in the selection of a surfactant, as well as its capacity to solubilize the contaminant it is important to consider its tendency to be adsorbed by the soil or sediment to be decontaminated. The results of studies on the adsorption of anionic and non-ionic surfactants by soils show that there is little consensuses about the mechanisms by which such compounds are adsorbed. Whereas some authors have reported the existence of a positive relationship between adsorption and the organic matter (OM) content of soils [7–10], others have found a relationship between adsorption and the clay content [11–16]. This means that the behavior of surfactants in the soil is not well known. The difficulty involved in establishing an adsorption–desorption mechanism lies, on one hand, in the complexity of the molecular structure of surfactants, which are composed of two moieties: one hydrophilic, or water-soluble, and the other hydrophobic, or water-insoluble.

106

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

On the other hand, it derives from the formulation of commercial surfactants, which are generally mixtures of homologues. The n-alkyl polyethylene glycol ethers CH3 (CH2 )i−1 (OCH2 CH2 )j OH usually abbreviated as Ci Ej are the most intensively studied non-ionic surfactants. The possibility to vary separately the length of hydrophilic head j and the hydrophobic tail i offers a wide field of applications in different areas as tertiary oil recovery, medicine, cosmetics and nano-science [17]. Recently, Turco Liveri and co-workers [18] have used the n-alkyl polyethylene glycol ethers in the mixed surfactant systems to investigate the rate-retarding effects of such systems on the alkaline hydrolysis of a cationic cobalt complex. Moreover, it might be helpful at this point to mention that in compare to the n-alkyl polyethylene glycol ethers very little attention is paid to the study of n-alkyl polypropylene glycol ethers in the scientific literature [19–25]. Bauduin et al. [26] compared the hydrophilic/lipophilic behavior of the ethylene and propylene glycol alkyl ethers, and as a result of different considerations, probably the n-alkyl polypropylene glycol ethers present an attractive alternative to the n-alkyl polyethylene glycol ethers. Modeling of the adsorption mechanisms of surfactants has traditionally been based on the interpretation of adsorption isotherms [27]. There are many other techniques to understand the adsorption mechanism and the mode of arrangement of the adsorbed layer [28]. However, they all require experimental procedures and instrumentation which are not always available in most laboratories. It is thus desirable to be able to get as much information as possible about the system from the commonest and most easily interpreted source of data-adsorption isotherms. In the work discussed in this paper, the adsorption of various non-ionic surfactants, onto bentonite of Iran was investigated in aqueous solution, in order to obtain the information about the adsorbed layer and the mechanism of its formation at different surfactant concentrations. We employed the information from the adsorption isotherms together with a theoretical evaluation of the surface adsorption properties of adsorbent and adsorbate. The aim of this work was to investigate, experimentally, the potential of natural bentonite to adsorb non-ionic surfactants series characterized by a different polyoxyethylene chain and hydrophobic tail. Isotherm studies were conducted to evaluate the adsorption capacity of natural bentonite. The effects of initial adsorbate concentration and temperature were studied. In order to determine the mechanism of non-ionic surfactant adsorption on the bentonite and to evaluate the relationship between adsorption and the temperature, the experimental data were applied to the Freundlich, Langmuir, Modified Langmuir, BET, Halsey, Harkins–Jura, Smith and Henderson isotherm equations. X-ray diffraction analysis was carried out to study the effect of nonionic surfactant adsorption on the crystalline structure of the bentonite.

bentonite, had the following chemical composition (in wt.%): SiO2 (65.04), Fe2 O3 (1.67), MgO (1.87), Al2 O3 (13.61), CaO (2.01), TiO2 (0.19), Na2 O (2.26), K2 O (0.75). Since XRD analysis revealed the presence of significant amounts of quartz and feldspar, the raw material was purified as explained elsewhere [29]. Potassium bichromate stock solution (1.0N) was prepared by dissolving 49.04 g potassium bichromate in 1.00 L water. 196.08 g ammonium ferrosulfate was added to flask and 15.0 mL concentrated H2 SO4 was added gently, the total mixture was diluted to 1.00 L while stirred continuously. Orthophenthrolin indicator solution was prepared by dissolving 1.48 g of material in 100.0 mL water containing 0.69 g ferrosulfate. Deionized distilled water was used in all experiments.

2. Experimental

2.2.3. Experimental procedure Organic compounds are one of the most important parts of solid phase bentonite. Several methods [31] with a large and powerful group of procedures are available for determination of organic compounds in bentonite. Most of these methods are based on measuring the amount of organic carbon in ben-

2.1. Reagents and materials All reagents were purchased from Hansa chemical company and were of commercial grade. The parent material, salafchegan

2.2. Instruments and methods 2.2.1. Preparation of bentonite A 4 wt.% bentonite slurry in water was prepared and swollen at room temperature under continuous stirring. After 5 h, the concentration of the suspension was diluted to 2 wt.% by addition of known volume of 0.1 M NaCl solution. The suspension was stirred overnight followed by decantation. The ion-exchange procedure was repeated three times with fresh NaCl solution. Finally, the solid was rinsed with water to get free of chloride ions. At the last washing, stirring was stopped; the particles contained in the suspension were separated from the quartz deposits and centrifuged. The cake was dried at room temperature. This treatment resulted in a material with the following composition (in wt.%): SiO2 (64.8), Fe2 O3 (1.82), MgO (2.02), Al2 O3 (13.68), CaO (1.56), TiO2 (0.19), Na2 O (1.89), and K2 O (0.32). The cation-exchange capacity (CEC) measured was 70 meq 10−2 g−1 , as measured by the method of Ming and Dixon [30]. The BET surface area of the unmodified Na-bentonite determined from N2 adsorption (single point method) performed in a Quantasorb equipment (63 m2 g−1 ) [29]. Bentonite suspensions at 20 g dm−3 (2%, w/w) were prepared by dispersing 5 g of dehydrated bentonite (the clay was dried overnight at T = 383 K) in 250 cm3 of bidistilled water to swell the bentonite. This solid/liquid ratio for the dispersion of bentonite was chosen in order to avoid sedimentation in long time range. 2.2.2. Determination of surfactant concentration The concentration of surfactant solutions was measured by shaking the known amounts of the bentonite in 100.0 mL reagent bottles containing 20.0 mL of known concentration of various non-ionic surfactants solutions for 48 h at temperatures (297, 303, 307, 313 and 323 K) using an incubator (200 rpm). The solutions were then filtered and the concentrations were measured.

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

107

tonite. The decomposition process called wet ashing, converts the organic compound to carbon dioxide and water. This process involves the oxidation of the organic constituents of a sample with oxidizing reagents such as HNO3 , H2 SO4 , HClO4 , or the combination of these reagents. In this work we have used the procedure which involves the oxidative decomposition of organic compounds in bentonite by potassium bichromate in the presence of H2 SO4 ; the excess amount of potassium bichromate is titrated with ammonium ferrosulfate in the presence of orthophenanthroline. The consumed potassium bichromate is proportional to oxidation of organic carbon. In this procedure the oxidation of organic carbon results in releasing CO2 according to the following reaction: 2K2 Cr 2 O7 + 3C + 8H2 SO4 ⇒ 2Cr 2 (SO4 )3 + 3CO2 + 8H2 O + 2K2 SO4 Approximately 77–94% of organic carbon is oxidized in this procedure. 1.00 g of grinded bentonite was transferred to a 250.0 mL flask. 10.0 mL of potassium bichromate was added to the flask and 10 mL concentrated H2 SO4 was added to the mixture and the mixture was stirred gently. Caution: do this step under the hood. The CO2 bubbles evolved out of solution, and the attention should be kept not to let the solution sprinkle out of the flask. A total of 100–150 mL water was transferred to the flask and 10 drops of indicator was added to the solution. Then, the solution was titrated by ammonium ferrosulfate (0.5N). The color of solutions would be yellow to light green in accordance with the amount of organic compound of the bentonite. During the addition of ferroammonium sulfate the color of solution would change to light green and final color would be dark green; in the equivalent point, the color of solution would be red. The procedure was performed for blank and bentonite sample; and the percent of organic carbon compound were reported according to the following equation:    S 1.724ANC %O.M. = 1 − (2.1) B 0.088W In which, %O.M. is the percent of organic carbon compound; S and B the volumes of the ammonium ferrosulfate solution for sample and blank titration, respectively. A and N are potassium bichromate volume and normality, respectively. C is carbon equivalent mass and W is the mass of dry bentonite. 1.724 is the coefficient considered for conversion of organic carbon to organic material, and 0.088 is conversion coefficient for the degree of organic carbon oxidation. The proportion of C/N value in rich nitrogen containing bentonite and organic materials is 10 and the value for poor nitrogen containing materials is 20. According to the procedure described, the C/N value has been determined for bentonite.

Fig. 1. X-ray diffraction patterns of derived organobentonites under dry condition: (a) C12 PhE20 ; (b) C12 E7 ; (c) CPB; (d) C12 E6 ; (e) C9 PhE6 ; (f) C12 E4 . Table 1 Interlayer spacing and 2θ degree of derived organobentonite Surfactant (10−2 mol dm−3 )

˚ d spacing (A)



C12 E4 C9 PhE6 C12 E6 CPB C12 E7 C12 PhE20

19.71 17.57 16.5 15.5 13.95 12.17

4.95 5.67 5.69 5.81 6.07 6.34

(λ = 0.154056 nm). The operating condition of XRD is at 40 kV and 30 mA in a step scan mode. To prepare the air-dried samples, the suspensions were placed on glass plates and allowed to dry under atmospheric conditions. To prepare the wet samples, the initial suspensions were centrifuged and the residue was set under paste form on deeper sample support and immediately passed to XRD Studies. The X-ray diffraction patterns of derived organobentonite under dry condition are depicted in Fig. 1. Interlayer spacing and 2θ degree of derived organobentonites are presented in Table 1. The interlayer spacing gradually increased with the amount of non-ionic surfactant adsorbed on bentonite. For example, the interlayer ˚ spacing of C12 E4 and C12 E7 bentonites was 19.71 and 13.95 A, respectively. Apparently, at low surfactant loading conditions, the interlayer spacing of organobentonite is related to the surfactant sorption capacity which in turn may be controlled by using non-ionic surfactants with different molecular structure. 3. Data analyses

2.2.4. XRD studies The instrument for XRD studies was a Philips X pert model MPD analytical diffractometer using Cu K␣ radiation

It is often desirable to analyze surfactant adsorption in terms of a theoretical model in order to obtain a molecular

108

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

interpretation. The parameters from such an analysis can later be used to compare the adsorption behavior of different surfactants and also to predict the adsorption in new systems. Several adsorption isotherms have been studied for adsorption of non-ionic surfactants onto the bentonite (Iran) [32–48]. 4. Results and discussion 4.1. Adsorption of non-ionic surfactants on a solid surface The most important type of non-ionic surfactant is that with an oligo (oxyethylene) group as the polar head. Denoting an oxyethylene group by E, simple non-ionics can be abbreviated as Cm En if we have an alkyl chain as the lipophilic part. We can also have a more complex hydrophobic part with a branched and/or unsaturated group; many non-ionics contain aromatic groups. For the polyoxyethylene surfactants, the volumes of the two parts are of similar size; typically the polar part is larger than the non-polar one. It is, therefore, appropriate and fruitful to consider a non-ionic surfactant as a short AB block copolymer. One conspicuous feature of a non-ionic surfactant is the temperature dependence of physicochemical properties. This may be problematic in applications but can also be turned into an advantage, since temperature triggered systems can be designed. To master and understand the special temperature-dependent (effective) interactions between the polar solvent and the polyoxoethylene chains proves to be most essential for non-ionics. These interactions are obviously not unique for surfactant systems, but have a general bearing also for polymers containing oxyethylene groups. Since the critical packing parameter, CPP, is a reflection of the balance of the interactions between the hydrophobic moieties and the polar parts, we realize that the critical packing parameter plays an important role in surfactant adsorption, regardless of whether the surface is hydrophobic or hydrophilic. It will be illustrated in this article that the adsorption increases as the CPP increases, i.e., as the surfactants are able to pack closer at the surface with a larger energy gain. Since surfactants in an aqueous solution normally are used at conditions where the critical packing parameter is small we will list those conditions that increase the CPP and hence also the adsorption. Increasing the critical packing parameter of a system with a single straight-chain non-ionic surfactant can be accomplished in the following ways: (1) change of a surfactant to one with a longer hydrocarbon chain; (2) Change of a surfactant to one with a branched hydrocarbon chainl; (3) Use of surfactant with two hydrocarbon chains; (4) Change of a surfactant to one with a shorter polyoxyethylene chain; (5) Increase of the temperature; (6) Addition of salt. 4.2. Characterization In Fig. 1, we reported the X-ray spectra for the airdried surfactant complexes with bentonite (air humidity, 59%). The concentration of the surfactants in all cases is the same (8 × 10−3 mol L−1 ). The interlayer spacing of derived organobentonites is presented in Table 1. The interlayer spacing

˚ [29]. As shown, the interlayer of the original bentonite is 12.3 A spacing has changed depending on the size of the lipophilic and hydrophobic parts of the surfactants. For example, the interlayer spacing of the bentonite which modified with C12 PhE20 ˚ or even shrinks a approximately has not changed (12.19 A) little bit. The order of interlayer spacing of organobentonites ˚ > C9 PhE6 (17.57 A) ˚ > C12 E6 (16.50 A) ˚ > CPB [C12 E4 (19.71 A) ˚ > C12 E7 (13.95 A) ˚ > C12 PhE20 (12.17 A)] ˚ is in accor(15.50 A) dance with the surfactant capacity. It seems that by decreasing the size of the oxyethylene group, the interlayer spacing has increased. The structure of the lipophilic chain might also have a meaningful effect on the adsorption and even the mechanism of adsorption. It is understandable that the interaction of the oxyethylene group of the surfactant with surface of bentonite arises from hydrogen bonding of the oxygen atoms of the ethoxyl groups with the Si–OH groups of the bentonite. It has been shown that AEs show strong adsorption on SiO, but not on some oxides such as AlO and FeO [49,50]. Given that AE adsorbs strongly on SiO via hydrogen bonding, it is reasonable to speculate that AE with appropriate molecular weight will intercalate into the interlamellar space of the bentonite, which is composed predominately of a SiO surface. During the transfer of a POE chain inside such a polar corona, some important contributions such as, steric repulsion and stretching deformation, should be considered. Another important point in this regard is the localization of the lipophilic chain of the surfactant in the most favorable location on the surface. As mentioned before, apparently the interlayer spacing of organobentonite is related to the surfactant sorption capacity which in turn may be controlled by using non-ionic surfactants with different molecular weight. It is presumed that C12 PhE20 will adsorb on the external surfaces of the bentonite and it could not penetrate into the interlamellar spaces of the bentonite. Interlayer spacing of the samples with C12 E6 and C9 PhE6 surfactants shows that the one with phenyl moieties could have a better interaction with the lipophilic sites of the surface and probably with each other. 4.3. Adsorption mechanism To evaluate the adsorption process, several non-ionic surfactants have been investigated which are reported as below:

The effect of ethoxylated moities of the non-ionic surfactants on sorption to bentonite is shown in Figs. 2 and 3, using surfactants (C12 E4 , C12 E6 , C12 E7 ) and (C9 PhE6 , C9 PhE20 ) with different number of ethylene oxide units but the same alkyl chain length, respectively. It is clearly seen that the plateau for surfactant adsorption decreases as the ethoxylated part of the surfactant increases. Because hydrogen bonding is the major

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

109

Fig. 2. Plot of Modified Langmuir adsorption isotherm against concentration of surfactants with different number of ethylene oxide units but the same alkyl chain length onto bentonite clay (salafchegan, Iran) at 313.15 K. Experimental points: (䊉) C12 E4 ; () C12 E6 ; () C12 E7 ; and () CPB.

Fig. 3. Plot of Modified Langmuir adsorption isotherm against concentration of surfactants with different number of ethylene oxide units but the same alkyl chain length onto bentonite clay (salafchegan, Iran) at 313.15 K. Experimental points: (䊉) C9 PhE20 ; () C9 PhE6 ; and () CPB.

driving force for adsorption on Si–OH and Al–OH sites on the surface, it has been suggested that the ethylene oxide chains of the adsorbed surfactant molecules are directly adsorbed on active sites of the surface with flat configuration [51,52]. The

decreased chain–chain interaction between hydrocarbon chains, especially for C9 PhE20 , arises from the increased hindrance by the long ethylene oxide chains on the bentonite surface make the formation of double layer aggregates of surfactant unfavorable.

Table 2 Parameter estimation of Hill and Modified Langmuir equations fitted to the adsorption isotherms of non-ionic surfactants onto the bentonite T (K)

Surfactant

Modified Langmuir equation (mol kg−1 )

Hill equation K

r2*

n

M (mmol kg−1 )

K

r2*

n

M

323.15

C12 E4 C9 PhE6 C12 E6 CPB C12 E7 C12 PhE20

1.637 1.724 1.68 – 1.562 1.261

1767 1692 1650 – 1450 1245

7997 14800 11690 – 6626 976.3

0.995 0.993 0.994 – 0.995 0.996

1.636 1.723 1.68 – 1.562 1.261

1.778 1.692 1.65 – 1.45 1.245

242.5 262.8 263.14 – 279.1 235.1

0.995 0.994 0.994 – 0.995 0.996

313.15

C12 E4 C9 PhE6 C12 E6 CPB C12 E7 C12 PhE20

1.624 1.647 1.615 1.7 1.528 1.253

1706 1599 1573 1510 1364 1213

7835 10160 8429 10090 5854 937.6

0.995 0.996 0.996 0.994 0.997 0.995

1.624 1.647 1.615 – 1.528 1.253

1.706 1.599 1.573 – 1.364 1.213

250.3 270.7 269.8 – 292.1 235.2

0.995 0.996 0.996 – 0.997 0.995

307.15

C12 E4 C9 PhE6 C12 E6 CPB C12 E7 C12 PhE20

1.579 1.58 1.533 1.69 1.469 1.227

1672 1572 1527 1410 1307 1200

6177 6764 5273 15320 4268 783.9

0.996 0.996 0.998 0.996 0.998 0.995

1.579 1.58 1.533 – 1.469 1.227

1.672 1.572 1.527 – 1.307 1.2

251.6 265.9 168.2 – 296.2 228.4

0.996 0.996 0.998 – 0.998 0.995

303.15

C12 E4 C9 PhE6 C12 E6 CPB C12 E7 C12 PhE20

1.562 1.48 1.467 1.68 1.451 1.192

1634 1567 1493 1360 1251 1183

5757 3658 3607 15840 4058 619.8

0.997 0.997 0.997 0.995 0.999 0.994

1.561 1.481 1.467 – 1.452 1.191

1.634 1.567 1.493 – 1.251 1.183

255.9 255.1 265.7 – 306.5 220.5

0.997 0.997 0.997 – 0.999 0.994

297.15

C12 E4 C9 PhE6 C12 E6 CPB C12 E7 C12 PhE20

1.52 1.427 1.463 1.67 1.455 1.133

1587 1543 1391 1280 1197 1166

4623 2649 3910 17480 4346 423

0.998 0.996 0.997 0.995 0.998 0.995

1.52 1.427 1.463 – 1.455 1.133

1.587 1.543 1.391 – 1.197 1.166

257.7 250.2 285.2 – 317 208

0.998 0.996 0.997 – 0.999 0.995

*

r2 indicate the correction between experimental and calculated data.

110

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

Fig. 4. Plot of Modified Langmuir adsorption isotherm against concentration of non-ionic surfactants onto bentonite clay (salafchegan, Iran) at 313.15 K. Experimental points: (䊉) C12 E4 ; () C12 E6 ; () C9 PhE6 ; () C12 E7 ; () C9 PhE20 and () CPB.

Consequently, increasing the length of ethylene oxide chains on non-ionic surfactants will generally decrease the amount of non-ionic surfactant adsorbed by bentonite (Figs. 2–4). The adsorption isotherms for surfactants (C12 E4 , C12 E6 , C12 E7 ; C9 PhE6 and C9 PhE20 ) with the same hydrocarbon chain lengths but different degree of ethoxylation on bentonite are compared in Figs. 2–4. It seems that to maximize the adsorption of non-ionic surfactants with the same hydrocarbon tails

on bentonite, the polyoxoethylene chain must be shorter. The adsorption isotherms for surfactants (C12 E6 , C9 PhE6 ) with different hydrocarbon tails and the same polyoxoethylene are compared in Fig. 4. The results show that the surfactant with phenyl moiety has a higher adsorption, which part of it might be due to the lower solubility of the surfactant C9 PhE6 in water. Moreover, the surfactants with phenyl moieties show stronger non-bonded interaction between the hydrocarbon tails. Generally speaking, one could assume that the number of adsorbed surfactant molecules with longer ethylene oxide chains per unit weight of bentonite is smaller than the number of surfactants with short ethylene oxide chains. On the other hand, as mentioned above, CPP has an important role in surfactant adsorption. The higher the value of CPP causes the more packed surfactant and consequently the greater adsorption. As shown in Fig. 4, the higher adsorption of C9 PhE6 rather than C12 E6 is because of better interaction of phenyl groups, which causes to more packed surfactants. If we forget about the C12 E9 and C9 PhE20 , because of not having appropriate oxyethylene moieties, the adsorption of the other non-ionic surfactants, i.e., C12 E4 , C12 E6 , and C9 PhE6 is greater than the cetylpyridinium bromide (CPB). Non-ionic surfactants adsorbed based on hydrogen bond with the surface, while cationic surfactant adsorption mechanism is based on cationic exchange. Fig. 5, obviously shows that the rise in temperature causes the increasing of the adsorption. As mentioned above, having a greater value of CPP causes a higher adsorption, and one of the important factors that affect the CPP is

Table 3 Parameter estimation of Redlich–Peterson and BET equations fitted to the adsorption isotherms of non-ionic surfactants onto the bentonite T (K)

Surfactant

Redlich–Peterson equation

BET equation

n

a

K

r2

323.15

C12 E4 C9 PhE6 C12 E6 C12 E7 C12 PhE20

1.964 1.947 1.955 1.752 1.362

4114 4243 4421 2259 527.8

270.7 226.4 220.4 212.8 188.2

0.998 0.995 0.998 0.998 0.997

313.15

C12 E4 C9 PhE6 C12 E6 C12 E7 C12 PhE20

1.932 1.804 1.79 1.615 1.377

3812 2624 2458 1447 561

218.8 225.1 221.9 217.1 182.5

307.15

C12 E4 C9 PhE6 C12 E6 C12 E7 C12 PhE20

1.796 1.775 1.62 1.442 1.362

2319 2296 1321 917.5 527.6

303.15

C12 E4 C9 PhE6 C12 E6 C12 E7 C12 PhE20

1.734 1.601 1.479 1.442 1.365

297.15

C12 E4 C9 PhE6 C12 E6 C12 E7 C12 PhE20

1.649 1.548 1.406 1.431 No fit

K

r2

2.326 2.215 2.135 1.799 1.414

141.7 154.7 157.9 178.9 175.3

0.975 0.971 0.972 0.976 0.993

0.998 0.997 0.998 0.998 0.997

2.206 2.046 1.997 1.665 1.371

149.4 164.6 166.1 192.1 177.1

0.975 0.976 0.974 0.983 0.993

221.7 219.6 224.9 223.2 178.4

0.998 0.998 0.998 0.998 0.997

2.136 1.983 1.901 1.568 1.341

152.6 165.1 169.5 199.9 175.7

0.983 0.997 0.984 0.985 0.993

1879 1177 802.6 841.5 529.8

223.2 222.5 230.2 223.2 172.2

0.997 0.997 0.996 0.998 0.997

2.071 1.937 1.828 1.485 1.3

157.1 162.7 177.2 210.7 175.8

0.98 0.985 0.987 0.987 0.994

1403 971 674 842.2 No fit

223.7 220.2 236.2 221.5 No fit

0.998 0.997 0.994 0.998 No fit

1.984 1.877 168.3 1.416 1.242

161.6 164 188.8 219.9 176.7

0.987 0.989 0.95 0.987 0.996

M

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

111

Table 4 Parameter estimation of Modified BET and Modified Oswin equations fitted to the adsorption isotherms of non-ionic surfactants onto the bentonite Non-ionic surfactant

C12 E4 C9 PhE6 C12 E6 C12 E7 C12 PhE20

Modified BET

Modified Oswin

M

B

K

r2

1.829 1.61 1.455 1.204 1.106

0.0089 0.011 0.013 0.011 0.064

151.7 161.8 169.3 197.9 176.1

0.982 0.995 0.985 0.983 0.994

temperature. In addition, the increase in temperature causes the increase in surfactant solubility and consequently the adsorption would be decreased. Based on the results shown in Tables 2–4 and Fig. 5, the temperature basically affects the CPP value more than solubility. According to the results shown in Tables 2–4, the experimental data has been fitted into the nonlinear isotherm equations of Langmuir–Freundlich (Modified Langmuir), Hill, Redlich–Peterson and Modified BET while the other isotherm equations are not fitted as well as the equations mentioned above. The surfactant C9 PhE20 is the only surfactant which can be fitted to the Langmuir isotherm (presented in Supplementary data). As it mentioned before, this surfactant probably cannot penetrate into the interlamellar spaces, and because of the long oxyethylene moiety is not able to create strong and sufficient non-bonded interactions between the hydrocarbon tails. Therefore, one would expect that the surfactant adsorb exclusively as a monolayer on the external surface of the support. The results show that none of the linear isotherm equations are appropriate for experimental data. Data reported in the Table 4 are well fitted into Modified BET isotherms. The Modified BET isotherms which consider the effect of temperature, as a surface model can appropriately show the increase of the adsorption with temperature. Based on the good results obtained from Modified BET isotherm in Fig. 5,

Fig. 5. Modified BET isotherm of C12 E4 non-ionic surfactant in various temperatures.

M

B

C

r2

12.16 10.39 9.193 6.96 6.61

0.059 0.071 0.08 0.062 0.037

1.884 1.929 1.961 2.088 2.023

0.923 0.931 0.93 0.922 0.963

we can conclude that the surfactants are adsorbed as monolayer, because the Modified BET isotherm is based on monolayer adsorption. Among the five surfactants investigated, the results for C9 PhE20 fit very well with Langmuir isotherm, and the data shows the monolayer adsorption of this surfactant (Supplementary data). According to the reported values of n in Table 2 when n > 1, there is a favorable adsorption. The values of RL were observed to be in the range of 0–1, which indicate that the adsorption of non-ionic surfactants onto bentonite are favorable for this study. 4.4. Evaluation of thermodynamic data The effect of temperature on non-ionic adsorption on bentonite was investigated over the concentration range (0.0008–0.018) mol dm−3 , at temperatures (297, 303, 307, 313 and 323) K under optimized conditions of shaking time, amount of adsorbent. Fig. 6 shows that value of the semi-logarithmic plot of distribution coefficient Kd against 1000/T. Similar observations have been reported earlier for cationic surfactant such as cetylpyridinium bromide (CPB) [29,53].

Fig. 6. Semi-logaritmic plot of distribution coefficient Kd against reciprocal temperature for the adsorption of C12 E4 onto bentonite. Experimental points for surfactant concentrations/(mol dm−3 ): () 0.018; () 0.015; () 0.0125; (䊉) 0.01; ( ) 0.008r and (—) 0.006.

112

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

Table 5 Thermodynamic parameters for the sorption of non-ionic surfactants on bentonite in various concentration and different temperature Non-ionic surfactant

Concentration (mol dm−3 )

H◦ (kJ mol−1 )

S◦ (J K−1 mol−1 )

G◦ (kJ mol−1 ) 297 K

303 K

307 K

313 K

323 K

C12 E4

0.006 0.008 0.01 0.0125 0.015 0.018

3.0903 3.438 4.45 3.727 3.342 3.324

53.38 53.5 55.267 51.709 49.05 47.528

−12.739 −12.451 −11.944 −11.627 −11.211 −10.783

−13.059 −12.772 −12.275 −11.938 −11.505 −11.068

−13.272 −12.986 −12.496 −12.144 −11.701 −11.258

−13.591 −13.307 −12.827 −12.455 −11.995 −11.543

−14.125 −13.842 −13.379 −12.972 −12.485 −12.018

C9 PhE6

0.006 0.008 0.01 0.0125 0.015 0.018

5.000 6.165 6.893 7.594 4.804 4.356

59.151 61.677 62.612 63.272 53.12 50.247

−12.523 −11.957 −11.521 −11.117 −10.937 −10.494

−12.877 −12.32 −11.893 −11.495 −11.255 −10.794

−13.113 −12.562 −12.141 −11.747 −11.467 −10.994

−13.46 −12.928 −12.513 −12.125 −11.785 −11.294

−14.057 −13.538 −13.133 −12.755 −12.315 −11.794

C12 E6

0.006 0.008 0.01 0.0125 0.015 0.018

5.444 7.817 3.742 4.546 4.273 3.836

60.929 67.064 52.857 53.907 51.666 48.822

−12.376 −12.086 −11.702 −11.195 −10.874 −10.426

−12.736 −12.488 −12.014 −11.513 −11.18 −10.714

−12.976 −12.756 −12.222 −11.725 −11.384 −10.906

−13.33 −13.158 −12.534 −12.043 −11.69 −11.194

−13.936 −13.828 −13.054 −12.573 −12.2 −11.674

C12 E7

0.006 0.008 0.01 0.0125 0.015 0.018

5.254 6.363 5.772 6.489 5.712 5.335

59.082 61.118 57.884 58.658 54.828 52.152

−12.269 −11.754 −11.157 −10.737 −10.326 −10.109

−12.623 −12.12 −11.499 −11.085 −10.65 −10.421

−12.859 −12.364 −11.727 −11.317 −10.866 −10.629

−13.213 −12.73 −12.069 −11.665 −11.19 −10.941

−13.803 −13.34 −12.639 −12.245 −11.73 −11.461

C9 PhE20

0.006 0.008 0.01 0.0125 0.015 0.018

4.621 3.926 3.826 3.511 3.758 3.888

54.882 51.203 49.698 47.305 46.87 45.992

−11.418 −11.221 −10.727 −10.448 −9.904 −9.477

−11.742 −11.527 −11.021 −10.73 −10.18 −9.747

−11.958 −11.731 −11.217 −10.918 −10.364 −9.927

−12.282 −12.037 −11.511 −11.2 −10.64 −10.197

−12.822 −12.547 −12.001 −11.67 −11.1 −10.647

The values of H◦ and S◦ were calculated from the slopes and intercepts of the plot of ln Kd against 1000/T (Fig. 6) by using the equation: ln Kd =

H ◦ S ◦ − R RT

(4.1)

The Gibbs free energy of specific adsorption G◦ is calculated from the equation: G◦ = H ◦ − T S ◦

(4.2)

The calculated values are given in Table 5. The negative values of G◦ for non-ionic surfactant are an indication of a spontaneous process; that is to say that the adsorption forces are strong enough to break the potential barrier. From the data presented in Table 5, it is evident that the principal contribution to the G◦ of negative value is the large positive value of S◦ , whereas H◦ is positive. In addition, from the results given in Table 5 it is found that increasing the temperature cause to more negative G◦ and the value for surfactant C12 E4 is maximum, which shows the maximum value of adsorption.

5. Conclusion In this study the data of the adsorption of non-ionic surfactants on bentonite has obtained and for the first time the fitting of the data onto different linear and nonlinear isotherm equations, have been investigated and the capability of the isotherms has been studied to interpret the results. In addition, the surface modeling based on two-variables of concentration and temperature were studied which shows the proper fitting of the data onto the predicted model. Bentonite high capability adsorptions for non-ionic surfactants C12 E6 , C9 PhE6 have been higher than the cationic surfactant (CPB). The shorter polyoxoethylene-chain surfactants have higher adsorption and the surfactants containing phenyl groups have higher adsorption due to the higher phenyl group interaction and higher CPP rather than ethoxylouryl-chain surfactants. It thus appears that short oxyethylene chain nonionic surfactants are suitable for organobentonite preparation to maximize organic carbon loading on bentonite. Among the different adsorption isotherms Langmuir– Freundlich (Modified Langmuir), Hill, Redlich–Peterson, BET and Modified BET are best fitted to obtain adsorption data. Among the surfactants studied, C9 PhE20 is well fitted into the

M. Ghiaci et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 297 (2007) 105–113

Langmuir and BET. This is obvious that because the Modified BET is a surface model which consider the concentration and temperature simultaneously can interpret the data better than all the other isotherms. Finally the thermodynamic parameters were obtained according to the models. The negative values of G◦ for non-ionic surfactant are an indication of a spontaneous process; whereas H◦ is positive, and show that the spontaneous adsorption is based on the high entropy of the system [29,53]. Acknowledgements The authors thank to Isfahan University of Technology (IUT) Research Council and Center of Excellence in Chemistry department of Isfahan University of Technology for the support of this work. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.colsurfa.2006.10.032. References [1] C.C. West, J.H. Harwell, Environ. Sci. Technol. 26 (1992) 2324. [2] D. Di Cesare, J.A. Smith, Rev. Environ. Contam. Toxicol. 134 (1994) 1. [3] J.D. Rouse, D.A. Sabatini, R.E. Brown, J.H. Harwell, Water Environ. Res. 68 (1996) 162. [4] J.F. Lee, P.M. Liao, C.C. Kuo, H.T. Yang, C.T. Chiou, J. Colloid Interface Sci. 229 (2000) 445. [5] Z. Zheng, J.P. Obbard, Water Res. 36 (2002) 2667. [6] M.S. Rodr´ıguez-Cruz, M.J. S´anchez-Mart´ın, M. S´anchez-Camazano, J. Environ. Qual. 33 (2004) 920. [7] K. Urano, M. Saito, C. Murata, Chemosphere 13 (1984) 293. [8] N. Literz, H.W. Doering, M. Thiele, H.P. Blume, Ecotoxicol. Environ. Saf. 14 (1987) 103. [9] K. Fytianos, E. Voudrias, A. Papamichali, Chemosphere 36 (1998) 2741. [10] I.F. Paterson, B.Z. Chowdhry, P.J. Carey, S.A. Leharne, J. Contam. Hydrol. 40 (1999) 37. [11] M.L. Cano, P.B. Dorn, Environ. Toxicol. Chem. 15 (1996) 684. [12] Z. Ou, A. Yediler, Y. He, A. Kettrup, T. Sun, Chemosphere 32 (1996) 827. [13] B.J. Brownawell, H. Chen, W. Zhang, J.C. Westall, Environ. Sci. Technol. 31 (1997) 1735. [14] Y.H. Shen, Chemosphere 41 (2000) 711. [15] Y.H. Shen, Chemosphere 44 (2001) 989. [16] Y.H. Shen, M.H. Yen, Environ. Technol. 20 (1999) 425. [17] C. Browarzik, D. Browarzik, Fluid Phase Equilib. 235 (2005) 127. [18] R. Lombardo, C. Sbriziolo, M.L. Turco Liveri, Colloids Surf. A: Physicochem. Eng. Aspects 273 (2006) 1.

113

[19] J. Klier, C.J. Tucker, T.H. Kalantar, D.P. Green, Adv. Mater. 12 (2000) 1751. [20] R.M. Stephenson, J. Chem. Eng. Data 38 (1999) 134. [21] K. Lunkenheimer, H.R. Holzbauer, R. Hirte, Prog. Colloid Polym. Sci. 97 (1994) 116. [22] A. Sokolowski, B. Burczyk, J. Colloid Interface Sci. 94 (1983) 369. [23] A. Sokolowski, J. Chlebicki, Tenside Deterg. 19 (1982) 282. [24] L.F. Komarova, Y.N. Garber, Zh. Org. Khim. 7 (1971) 2507. [25] H.L. Cox, W.L. Nelson, L.H. Cretcher, J. Am. Chem. Soc. 49 (1927) 1080. [26] P. Bauduin, A. Basse, D. Touraud, W. Kunz, Colloids Surf. A: Physicochem. Eng. Aspects 270–271 (2005) 8. [27] P.E. Levitz, C.R. Geosci. 334 (2002) 665. [28] G.D. Parfitt, C.H. Rochester, in: G.D. Parfitt, C.H. Rochester (Eds.), Adsorption from Solution at the Solid/Liquid Interface, Academic Press, London, 1983, p. 3. [29] M. Ghiaci, R.J. Kalbasi, H. Khani, A. Abbaspur, H. Shariatmadari, J. Chem. Thermodyn. 36 (2004) 707. [30] D.W. Ming, J.B. Dixon, Clays Clay Miner. 35 (1987) 463. [31] T.S. Ma, Anal. Chem. 62 (1990) 78R. [32] L.K. Koopal, W.H. Van Riemsdijk, J.C.M. De Wit, M.F. Benedetti, J. Colloid Interfaces Sci. 166 (1994) 51. [33] V. Desauziers, N. Castre, P. Le Cloirec, Environ. Technol. 18 (1997) 1009. [34] O. Altin, H. Onder Ozbelge, T. Dogu, J. Colloid Interface Sci. 198 (1998) 130. [35] T. Viraraghavan, F. De Maria Alfaro, J. Hazard. Mater. 57 (1998) 59. [36] N. Calace, E. Nardi, B.M. Petronio, M. Pietroletti, Environ. Pollut. 118 (2002) 315. [37] N. Kannan, M. Sundaram, Dyes Pigments 51 (2002) 25. [38] R. Sivaraj, C. Namasivayam, K. Kadirvelu, Waste Manage. 21 (2001) 105. [39] A.K. Jain, V.K. Gupta, A.B. Suhas, J. Hazard. Mater. B 101 (2003) 31. [40] D.M. Ruthven, Principles of Adsorption and Adsorption Processes, John Wiley, New York, 1980. [41] D. Naden, M. Streat, Ion Exchange Technology, Ellis Horwood, London, 1984. [42] M.S. Chiou, H.Y. Li, J. Hazard. Mater. B 93 (2002) 233. [43] C.R. Cantor, P.R. Schimmel, Biophysical Chemistry, Part III. The Behaviour of Biological Macromolecules, Freeman, San Francesco, 1980 (Chapter 15). [44] R.S. Juang, R.L. Tseng, F.C. Wu, S.H. Lee, J. Chem. Technol. Biotechnol. 70 (1997) 391. [45] M.M. Nssar, Y.H. Magdy, Chem. Eng. J. 66 (1997) 223. [46] D.L. Peterson, O. Redlich, J. Chem. Eng. Data 7 (4) (1962) 570. [47] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (2) (1938) 309. [48] H. Iglesias, J. Chirife, Lebensmittel-Wissenschaft Technologie 9 (1976) 107. [49] L. Huang, C. Maltesh, P. Somasundaran, J. Colloid Interface Sci. 177 (1996) 222. [50] S. Mathur, B.M. Moudgil, Miner. Metall. Process 15 (2) (1998) 24. [51] P. Somasundaran, E.D. Snell, Qu. Xu, J. Colloid Interface Sci. 144 (1991) 165. [52] Qu. Xu, T.V. Vasudevan, P. Somasundaran, J. Colloid Interface Sci. 142 (2) (1991) 528. [53] M. Ghiaci, R. Kia, R.J. Kalbasi, J. Chem. Thermodyn. 36 (2004) 95.