pubs.acs.org/Langmuir © 2009 American Chemical Society
Equilibrium and Molecular Mechanism of Anionic Dyes Adsorption onto Copper(II) Complex of Dithiocarbamate-Modified Starch Rumei Cheng,† Shengju Ou,*,†,‡ Bo Xiang,† Yijiu Li,*,† and Qiangqiang Liao† †
Department of Chemistry, Tongji University, Shanghai 200092, China and ‡Key Laboratory for the Chemistry and Molecular Engineering of Medicinal Resources of State Education Ministry, Guangxi Normal University, Guilin 541004, China Received March 26, 2009
The assembly of dyes molecules on metal-polymer complexes is of interest due to their potential applications in photovoltaic cell, separation, and wastewater treatment. In the present work, the interaction of anionic dyes (acid orange 7, acid orange 10, acid green 25, and acid red 18) with the copper(II) complex of dithiocarbamate-modified starch (DTCSCu) was investigated. The sorption studies showed that the interaction mechanism was based on chelating adsorption. The equilibrium data fitted well with Langmuir-Freundlich isotherm, and the capacities followed the order AO7 > AG25 > AR18 > AO10. It was affected by the structure of the dye. The sulfonate groups located on benzene rings favored efficient adsorption. Despite the difference in capacity, the molar n(dye):n(Cu) ratios for acid orange 10, acid red 18, and acid green 25 were approximately 1:2 when the maximum capacities for the dyes were achieved at the optimal pH of 4. It suggested that one dye molecule bound to one dinuclear copper center on DTCSCu. The molar n(dye):n(Cu) ratio for the smallest dye, acid orange 7 (AO7), approached 1:1, demonstrating two AO7 molecules binding to two copper ions of the dinuclear core. The dyes adsorption related to the dinuclear copper core available on the polymer was further verified by electron spin resonance studies. Such interaction resulted in the formation of a ternary dye-metal-polymer complex. The ternary complexes were more stable than DTCSCu, which favored the adsorptions.
1. Introduction The dye effluents of the textile industry and related industries cause serious environmental problems.1,2 Many of the dyestuffs are toxic and carcinogenic.3,4 As a consequence, the treatment of dyeing effluents has long been a major concern. Several treatment technologies for dyestuff removal have been extensively investigated, such as chemical oxidation,5 photodegradation processes,6 activated sludge,7 and adsorption procedures.8-10 Whatever technology is applied, the adsorption is always involved. Recently, there has been growing interest in the transition metal complexes of polymers because of their potential applications in separation and wastewater treatment.11 Some metal-binding polymers have been used as adsorbents for the removal of contaminants. For instance, a transition-metal-incorporated aminosilane material has been used to remove arsenate ions from aqueous solutions.12 The Cu2þ ion can significantly enhance 1-naphthol adsorption onto lignin by the formation of a lignin-Cu2þ *Corresponding authors: Ph þ86-773-3958371; Fax þ86-773-2120958, e-mail
[email protected] (S.O.),
[email protected] (Y.L.).
(1) Pearce, C. I.; Lloyd, J. R.; Guthrie, J. T. Dyes Pigm. 2003, 58, 179–196. (2) Crini, G. Bioresour. Technol. 2006, 97, 1061–1085. (3) Ramakrishna, K. R.; Viraraghavan, T. Waste Manage. 1997, 17, 483–488. (4) Brown, M. A.; De Vito, S. C. Crit. Rev. Environ. Sci. Technol. 1993, 23, 249– 324. (5) Muthukumar, M.; Sargunamani, D.; Selvakumar, N. Dyes Pigm. 2005, 65, 151–158. (6) Zhao, X.; Qu, J. H.; Liu, H. J.; Hu, C. Environ. Sci. Technol. 2007, 41, 6802– 6807. (7) Bahte, S.; Schwarzenbeck, N.; Hausner, M. Bioresour. Technol. 2009, 100, 2902–2909. (8) Crini, G.; Badot, P. M. Prog. Polym. Sci. 2008, 33, 399–447. (9) Blackburn, R. S. Environ. Sci. Technol. 2004, 38, 4905–4909. (10) Kron, D. A.; Holland, B. T.; Wipson, R.; Maleke, C.; Stein, A. Langmuir 1999, 15, 8300–8308. (11) Jang, M.; Min, S. H.; Park, J. K.; Tlachac, E. J. Environ. Sci. Technol. 2007, 41, 3322–3328. (12) Yoshitake, H.; Yokoi, T.; Tatsumi, T. Chem. Mater. 2003, 15, 1713–1721.
752 DOI: 10.1021/la9039489
complex.13 Iron humate, a low-cost industrial residue, is a valuable material in basic dye removal.14 Despite these studies, the metal complexes of polymers have not been well studied, and the mechanism of anionic dye adsorption onto metal-chelating polymers is still not understood. It is crucial to clarify whether the dye forms a strong surface complex or just adsorbs onto the substrate via a weak physical attraction. The most important factors controlling the adsorption of the dye onto the metal complex are the nature of the dye, the type of active adsorption sites available, and the observed adsorption isotherms. The object of this study was to show how the anionic dye interacted with the metal-chelating polymer.
2. Materials and Methods 2.1. Materials. Commercial corn starch, of food-grade quality, was used in this research. Copper(II) chloride (CuCl2 3 2H2O) and other routine reagents were of analytical grade. Four commercially available dyes, acid orange 7 (AO7), acid orange 10 (AO10), acid red (AR18), and acid green 25 (AG25), were purchased from Sigma and used as received. They have been widely studied and are regarded as typical acid dyes.15,16 The molar mass and charges often affect the sorption behavior. All the four dyes have sulfonate groups with negative charges but difference in molar mass and charges. Also, their structures are slightly different. The chemical structures and characteristics of the four dyes are shown in Table 1. 2.2. Techniques. The elemental analyses were performed with a Perkin-Elmer 240C elemental analyzer. The copper contents (13) Wang, X.; Yang, K.; Tao, S.; Xing, B. Environ. Sci. Technol. 2007, 41, 185– 191. (14) Janos, P. Environ. Sci. Technol. 2003, 37, 5792–5798. (15) Jin, X.; Jiang, M. Q.; Shan, X. Q.; Pei, Z. G.; Chen, Z. J. Colloid Interface Sci. 2008, 328, 243–247. (16) Cheung, W. H.; Szeto, Y. S.; Mckay, G. Bioresour. Technol. 2009, 100, 1143–1148.
Published on Web 12/22/2009
Langmuir 2010, 26(2), 752–758
Cheng et al.
Article Table 1. List of Anionic Dyes Used in This Study
were determined by inductively coupled plasma atomic emission spectroscopy (ICP-AES) on POEMS-2. X-band electron spin resonance (ESR) spectra were measured on an EMX spectrometer. Fourier transform infrared spectra (FT-IR) were recorded on a PE Spectrum One spectrometer with KBr pellets in the 4000-450 cm-1 region. Ultraviolet-visible (UV-vis) spectra were measured with a Perkin-Elmer Lamda 35 spectrometer. Thermogravimetric analyses were performed with STA 409 PC/4/H Luxx at a heating rate of 10 C/min under a N2 atmosphere.
the flasks with a 50 mL solution containing a known amount of the respective dye (varying from 300 to 1200 mg L-1). The pH of the solution was adjusted to 4.0 and kept constant. The flasks were agitated for 24 h in a shaking thermostatic bath. The temperature was maintained at 25, 35, and 45 C. After filtration, the concentration of each dye solution was determined. The results were then used to calculate the loading of each dye onto DTCSCu, using eq 1:
2.3. Preparation of Copper(II) Complex of Dithiocarbamate-Modified Starch (DTCSCu). The dithiocarbamate-
qe m ¼ ðc0 -ce ÞV
modified starch (DTCS) was first synthesized as reported previously,17 and the process is shown in Figure 1. Then 10.0 g of dried DTCS was added to a 500 mL solution containing 2.0 g of CuCl2 3 2H2O at pH 5.0. The mixture was stirred for 12 h at 25 C, and the precipitate was separated from the solution. The Cu2þsaturated polymer was washed five times with deionized water. The acquired DTCSCu was kept in a vacuum oven for 1 day and stored in a desiccator. Found: C 39.87%, H 7.62%, N 2.05%, S 2.78%, Cu 2.07%. 2.4. Concentration Measurement Calibration. To determine the dye concentration in the solution, a calibration curve was first obtained using a UV-vis spectrophotometer. The maximum absorbance of each dye was confirmed by scanning the dye aqueous solution over the spectral range of 200-800 nm. A series of dye solutions with various concentrations were used for the measurement of a calibration curve. A linear relationship between the absorbance (at the wavelength of maximum absorbance, λmax) and the dye concentration was obtained. 2.5. Sorption Experiments. Effect of pH. Adsorption studies were performed using a batch technique. The influence of pH was observed by studying the adsorption of each dye over a pH range of 3-8. For these experiments, a series of 100 mL conical flasks were used. Each flask was filled with 50 mL of a dye solution at a concentration of 300 mg L-1 at different pHs. The flasks were shaken for 12 h in a shaking thermostatic bath to reach equilibrium. After filtration, the concentrations of each dye and Cu(II) ions in the aqueous solutions were measured. Equilibrium Experiments. For the equilibrium adsorption studies, a fixed mass of DTCSCu (0.1500 g) was weighed into
where qe is the dye concentration in the solid phase (adsorbent) at equilibrium (mmol g-1), m is the mass of DTCSCu used (g), c0 is the initial dye concentration in the liquid phase (mmol L-1), ce is the dye concentration in the liquid phase at equilibrium (mmol L-1), and V is the total volume of the solution used (L). The models of the isotherms were fitted to equilibrium data using a nonlinear method, with the nonlinear fitting facilities of the Microcal Origin 7.0 software. Desorption Experiments. The dyes release experiments were carried out after an initial incubation time of 24 h under conditions described above. When DTCSCu samples were saturated with individual dyes, they were separated out and placed in 100 mL of various solutions. After shaken for 24 h at 25 ( 1 C, the concentrations of free dyes in solutions were measured. 2.6. Adsorption Isotherms. Langmuir Isotherm. The well-known Langmuir isotherm was originally proposed to describe the adsorption of gas molecules onto metal surfaces.18 The model assumes uniform energy of the adsorption onto the surface and no migration of the adsorbate in the plane of the surface. The Langmuir adsorption isotherm has been applied successfully to many other real situations of monolayer adsorption.19 It is expressed as
(17) Xiang, B.; Li, Y. J.; Ni, Y. M. J. Appl. Polym. Sci. 2004, 92, 3881–3885.
Langmuir 2010, 26(2), 752–758
qe ¼
abce 1 þ bce
ð1Þ
ð2Þ
where qe is the amount of dye adsorbed at equilibrium (mmol g-1) and ce is the adsorbate concentration at equilibrium in aqueous (18) Langmuir, I. J. Am. Chem. Soc. 1918, 40, 1361–1401. (19) Blackburn, R. S.; Harvery, A.; Kettle, L. L.; Payne, J. D.; Russell, S. J. Langmuir 2006, 22, 5636–5644.
DOI: 10.1021/la9039489
753
Article
Cheng et al.
Figure 1. Synthesis of dithiocarbamate-modified starch (DTCS).
Figure 2. Effect of pH value on adsorption of anionic dyes onto DTCSCu.
Figure 3. Leachability of copper(II) ions in various solutions at equilibrium under different pH values.
solution (mmol L-1). The Langmuir isotherm parameters are a and b. The capacity of the adsorbent can be evaluated by a, and the parameter b includes various physical constants.20 Freundlich Isotherm. Equation 3 is the well-known Freundlich isotherm, which describes heterogeneous systems,21 i.e., surfaces with nonenergetically equivalent sites.22 It is an empirical equation and can be written as follows: qe ¼ Kf ce 1=n
ð3Þ
where Kf is the Freundlich constant, which is indicative of the extent of adsorption, and 1/n is the heterogeneity factor, an indicator of adsorption effectiveness. Langmuir-Freundlich Isotherm. Another useful equation is the Langmuir-Freundlich isotherm,23 which includes three parameters. This isotherm is based on the generalized Langmuir and generalized exponential isotherms24 and is the most promising extension of the Langmuir and Freundlich isotherms. The Langmuir-Freundlich isotherm is expressed as qe ¼
qm ðKlf ce Þv 1 þ ðKlf ce Þv
Freundlich isotherm is essentially the Freundlich isotherm approaching a maximum at high concentrations.
3. Results and Discussion ð4Þ
where qm is the maximum adsorption (mmol g-1), Klf is the Langmuir-Freundlich constant (L mmol-1)1/v, and v is the Langmuir-Freundlich heterogeneity constant. The Langmuir(20) Giles, C. H.; Smith, D.; Huitson, A. J. Colloid Interface Sci. 1974, 47, 755– 765. (21) Freundlich, H. M. F. Z. Phys. Chem. (Munich) 1906, 57, 385–470. (22) Reddad, Z.; Gerente, C.; Andres, Y.; Ralet, M. C.; Ghibault, J. F.; Le Cloriec, P. Carbohydr. Polym. 2002, 49, 23–31. (23) Marczewski, A. W.; Derylo-Marczewska, A.; Jaroniec, M. J. Chem. Soc., Faraday Trans. 1988, 84, 2951–2957. (24) Lazaridis, N. K.; Kyzas, G. Z.; Vassiliou, A. A.; Bikiaris, D. N. Langmuir 2007, 23, 7634–7643.
754 DOI: 10.1021/la9039489
Figure 4. Adsorption isotherms fitted to Langmuir model (the real lines) and Freundlich equation (the dash-dotted lines).
3.1. Influence of pH. The pH of the aqueous solution is an important controlling parameter in the dye adsorption process. It significantly affects the sorption mechanism. The influence of solution pH was observed in a pH range of 3-8. It can be seen from Figure 2 that the capacity of DTCSCu for each dye was pH dependent and that the maximum capacity for each dye was achieved under acidic conditions. With increasing pH, the adsorption capacity for each dye decreased. This pH-dependent trend has also been observed for the adsorption of similar anionic dyes onto ZnO25 and activated carbon modified with Y3þ ions,26 (25) Bauer, C.; Jacques, P.; Kalt, A. Chem. Phys. Lett. 1999, 307, 397–406. (26) Tamai, H.; Yoshida, T.; Sasaki, M.; Yasuda, H. Carbon 1999, 37, 983–989.
Langmuir 2010, 26(2), 752–758
Cheng et al.
Article Table 2. Parameters of Langmuir and Freundlich Isotherms for Anionic Dyes Adsorption onto DTCSCu Langmuir isotherm -1
-1
Freundlich isotherm 2
dye
a (mmol g )
b (L mmol )
R
AO7 AO10 AG25 AR18
0.316 0.118 0.181 0.164
103.4 277.9 614.7 3953.4
0.977 0.987 0.859 0.830
Figure 5. Adsorption isotherms fitted to Langmuir-Freundlich equation (the real lines).
indicating the interaction of the anionic dyes with metal centers, which direct the adsorption process. As shown in Figure 3, the Cu2þ concentrations in the individual dye solutions were markedly higher than those in the deionized aqueous solutions at the same pH. These data suggest that the dyes strongly interact with the Cu(II) ions on DTCSCu. When DTCSCu was added to the aqueous dye solution, the free hydroxyl ions and large numbers of water molecules approached the Cu2þ center because of its positive charge and coordination properties. At the same time, the dye molecules, which contain sulfonate groups with negative charges, changed places with the water molecules or hydroxyl ions to chelate the Cu2þ ions, resulting in their stronger adsorption. This process can be achieved easily in weak acidic solutions and the capacity of DTCSCu for each dye reached a maximum at pH 4.0. When the pH was increased, more free hydroxyl ions formed and interacted strongly with the copper ions, preventing the dyes from chelating the Cu2þ on the polymer surface. When the solution pH was below 4.0, the Cu2þ tended to be released from the DTCSCu and the capacity of DTCSCu for each dye decreased. Consequently, the optimal pH 4.0 was selected for the equilibrium studies. 3.2. Adsorption Equilibrium and Molecular Mechanism. Adsorption isotherms describe how adsorbates interact with adsorbents. All the equilibrium studies were carried out at the optimal solution pH of 4.0. Three isotherms (Langmuir, Freundlich, and Langmuir-Freundlich isotherms) were used to fit the experimental data. The results of the experimental data fitted to the Langmuir and Freundlich isotherms are shown in Figure 4 and Table 2. In the Langmuir model, the coefficient (R2) of adsorption for the smaller dyes AO7 and AO10 are satisfied, but those for AG25 and AR18 are markedly low. When the Freundlich isotherm is applied, there is particularly good agreement with the equilibrium data for the adsorption of the largest molecule, AG25. Incorporating the Langmuir equation and Freundlich equation, the Langmuir-Freundlich isotherm was used to fit the experimental data. The adsorption isotherms based on this model are shown in Figure 5, and the parameters are listed in Table 3. The Langmuir-Freundlich isotherm clearly provides satisfactory fits for the four dyes. The capacity of DTCSCu for each dye Langmuir 2010, 26(2), 752–758
1-1/n
Kf (mmol
0.320 0.118 0.184 0.175
1/n
L
g-1)
1/n
R2
0.102 0.028 0.030 0.049
0.841 0.828 0.968 0.893
follows the sequence AO7 > AG25 > AR18 > AO10. It can be seen that the capacities do not show a direct molar mass dependence. The amount of negative charge for AO7, AO10, and AR18 is in the order AR18 > AO10 > AO7. However, the capacities for the three dyes follow the sequence AO7 > AR18 > AO10. AO10 and AG25 have the same amount of sulfonate groups, but the capacities for them differ. Therefore, the capacity is not controlled by the number of negative charge on the dye molecule. To a certain extent, the capacity was affected by structures of the dyes. The anionic dyes with sulfonate groups on benzene are better adsorbed than are those with sulfonate groups on naphthalene. Although the molecular size of AO10 is close to that of AO7, the capacity for AO10 is lower, which is attributed to the sulfonate groups located on the naphthalene ring. The findings are very similar to the behaviors of AO7 and AO10 adsorption onto the surfaces of TiO2, Al2O3, and FeOOH.27 Another research on activated bleaching earth demonstrated the fact that the capacity for AO10 was much lower than those of anionic dyes containing sulfonate groups on benzene rings.28 This trend is also apparent for AG25 (with sulfonate groups on benzene rings) and AR18 (with sulfonate groups on naphthalene rings), which indicates that DTCSCu has a higher capacity for AG25 than for AR18. Moreover, the capacity of DTCSCu for AR18 is higher than that for AO10, even though all their sulfonate groups are located on the naphthalene ring. The difference between these two dyes is attributed to the conjugated effect. Another naphthalene ring adjacent to the disulfonate naphthalene moiety of AR18 produces a greater conjugated effect than that observed for AO10.29 The stronger conjugated effect stabilizes the sulfonate anion and enhances its acidity. Then AR18 is relatively hard to combine with free Hþ ions in aqueous solution and has a much stronger capacity to chelate metal ions than does AO10. Despite the difference in maximum capacity, the molar n(dye): n(Cu) ratio is certain to some extent. When the maximum capacity for each dye is achieved at pH 4, it can be seen from Table 3 that the molar n(dye):n(Cu) ratio is approximately 1:2, except for AO7 (with a molar ratio of 1:1). A molar n(dye):n(Cu) ratio of about 1:2 for the three dyes indicates that one dye molecule binds to two copper ions. If a dicopper core is involved, the n(Cu2):n(dye) ratio will be approximately 1:1 for the three dyes, suggesting the presence of a dinuclear core in DTCSCu. The dye molecules predominantly interact with the dinuclear copper core. The saturation of the dye chelation of the dicopper center greatly contributes to the maximum adsorption capacity of DTCSCu for the dye. AO7 is the smallest molecule examined, and more molecules may interact with the dicopper core to produce an n(Cu2):n(dye) ratio of 1:2. This implies that each copper ion chelates one AO7 molecule, showing the typical property of monolayer adsorption consistent with the Langmuir isotherm (27) Bandara, J.; Mielczarski, J. A.; Kiwi, J. Langmuir 1999, 15, 7670–7679. (28) Tsai, W. T.; Chang, C. Y.; Ing, C. H.; Chang, C. F. J. Colloid Interface Sci. 2004, 275, 72–78. (29) Tong, B. Y.; Xu, H. Y.; Yu, Q.; Li, C.; Guang, S. Y. Chin. J. Synth. Chem. 2004, 12, 210–212.
DOI: 10.1021/la9039489
755
Article
Cheng et al. Table 3. Parameters of Langmuir-Freundlich Isotherm for Dyes Adsorption onto DTCSCu molar ratio -1
dye
qm (mmol g )
AO7 AO10 AG25 AR18
0.327 0.119 0.196 0.177
1/v
Klf (L
-1/v
mmol
)
108.6 507.7 17876.7 8240.2
2
v
R
n(dye):n(Cu)
n(dye):n(Cu2)
0.741 0.743 0.272 0.350
0.995 0.997 0.995 0.989
1.01 0.37 0.61 0.55
2.02 0.74 1.22 1.10
Figure 6. Powder ESR spectra of DTCSCu before and after adsorption of dyes.
(30) Singh, V.; Sharma, A. K.; Sanghi, R. J. Hazard. Mater. 2009, 166, 327–335. (31) Geetha, K.; Nethaji, M.; Chakravarty, A. R.; Vasanthacharya, N. Y. Inorg. Chem. 1996, 35, 7666–7670. (32) Bernhardt, P. V. Inorg. Chem. 2001, 40, 1086–1092. (33) Slichter, C. P. Phys. Rev. 1995, 99, 479–480. (34) Long, R. C.; Hendrickson, D. N. J. Am. Chem. Soc. 1983, 105, 1513–1521. (35) Gagne, R. R.; Koval, C. A.; Smith, T. J.; Cimolino, M. C. J. Am. Chem. Soc. 1979, 101, 4571–4580.
756 DOI: 10.1021/la9039489
the g (2.270) splits into four peaks with an average A=155 G. Such complicated hyperfine patterns of dinuclear copper(II) complexes have been demonstrated in many small molecular copper(II) complexes and confirmed by single-crystal X-ray diffraction studies.36,37 It is well-known that the Cu(dtc)2 complex (dtc is diethyldithiocarbamate) contains a sulfur-bridging dinuclear center, producing being Cu2(dtc)4.38,39 The dialkylcarbamate complexes of copper(II) also have a strong tendency to form dinuclear complexes in the solid state.40-42 After the adsorption of the anionic dyes onto DTCSCu by their interaction with the dinuclear copper center, the ESR lines of the half-field transition narrowed. The splits at about 2900 G were weakened, and a new peak at 2842 G appeared. The facts indicate that the dye chelates to the dinuclear copper of DTCSCu, resulting in formation of a ternary dinuclear copper complex. FT-IR spectra were also used to study the dye chelating to DTCSCu. The different FT-IR spectra of DTCSCu before and after the adsorption of the dyes are shown in Figure 7 in the region of 1750-750 cm-1. The peaks for DTCSCu at 1640 and 1502 cm-1 are attributed to the v(S-CdN) and ν(SdC-N) vibrations of dithiocarbamate.43,44 The νC-N (-N-CS2) vibration45,46 is seen at 1466 cm-1, whereas the νC-N (CHN-CH2CH2) vibration47,48 appears at 1157 cm-1. The weak and broad peaks at 1405, 1381, and 1208 cm-1 are attributed to the coupling of the δO-H and δC-H vibrations. The strong and )
(Table 2). Interestingly, the ambient temperature has much less effect on the chelating adsorption. It is often observed that a better interaction between the dye and the organic polymer (such as chitosan) takes place at higher temperature due to increased collisions between them as the viscosity of the medium decreases.30 Such interactions are often controlled by hydrogen bonding and hydrophobic force. However, in our case the adsorption is governed by chelating mechanism, and the available copper centers are definite. Therefore, the capacity for the anionic dye increased very slightly with the increase of temperature. When the adsorption was performed at 318 K, the capacity increased no more than 1.5% (see Supporting Information). The changes affected the molar n(dye):n(Cu) ratio indistinctively. In order to probe further into the mechanistic aspects of the anionic dye adsorption onto DTCSCu, desorption studies were investigated (see Supporting Information Figure S5). The use of water and ethanol solutions for the dye desorption is ineffective. Very low desorption with these solutions suggested that metal-dye complexes were formed on DTCSCu. The high basic solution (above 0.01 mol L-1) did help to breaking this chelating interaction, in which the hydroxyls replaced the dyes to bind the DTCSCu. To avoid release of large amount of copper ions, the high acidic solution was not used in desorption studies. The chelating mechanism that the anionic dye interacted with dinuclear copper center was confirmed by spectroscopic studies. The ESR spectra of copper complexes can usually provide meaningful information. As shown in Figure 6a, the appearance of a weak forbidden half-field transition ΔMs = (2 between 1200 and 1800 G is consistent with a dinuclear copper(II) complex.31 The easily recognizable nine-line hyperfine splitting pattern (Figure 6b,c) in the range of 2700-3600 G suggests an anisotropic exchange contribution of two copper(II) ions,32 differing from the four- or five-line hyperfine patterns of mononuclear copper(II) complexes.33-35 The peak center at 3355 G splits into five peaks, with an average A = 45 G, showing the g^ (2.011). Meanwhile,
(36) Rammal, W.; Belle, C.; Beguin, C.; Duboc, C.; Philouze, C.; Pierre, J. L.; Le Pape, L.; Bertaina, S.; Saint-Aman, E.; Torelli, S. Inorg. Chem. 2006, 45, 10355– 10362. (37) Comba, P.; Gavrish, S. P.; Hay, R. W.; Hilfenhaus, P.; Lampeka, Y. D.; Lightfoot, P.; Peters, A. Inorg. Chem. 1999, 38, 1416–1421. (38) Cowsik, R. K.; Srinivasan, R. Pramana 1973, 1, 177–187. (39) Bonamico, M.; Dessy, G.; Mugnoli, A.; Vaciago, A.; Zambonelli, L. Acta Crystallogr. 1965, 19, 886–897. (40) Coucouvanis, D. J. Am. Chem. Soc. 1970, 92, 707–709. (41) Weeks, M. J.; Fackler, J. P. Inorg. Chem. 1968, 7, 2548–2553. (42) Rajasekharan, M. V.; Sethulakshmi, C. N.; Manoharan, P. T.; Gudel, H. Inorg. Chem. 1976, 15, 2657–2662. (43) Nakamoto, K.; Fujita, J.; Condrate, R. A.; Morimoto, Y. J. Chem. Phys. 1963, 39, 423–427. (44) Kaul, B. B.; Pandeya, K. B. J. Inorg. Nucl. Chem. 1978, 40, 229–233. (45) Macı´ as, B.; Villa, M. V.; Rodriguez-Gallego, M. R. Transition Met. Chem. (Dordrecht, Neth.) 1995, 20, 347–350. (46) Jian, F.; Wang, Z.; Bai, Z.; You, X.; Fun, H. K.; Chinnakali, K.; Razak, I. A. Polyhedron 1999, 18, 3401–3406. (47) Brown, T. M.; Smith, J. N. J. Chem. Soc., Dalton Trans. 1972, 1614–1616. (48) Bhat, A. N.; Fay, R. C.; Lewis, D. F.; Lindmark, A. F.; Strauss, S. H. Inorg. Chem. 1974, 13, 886–892.
Langmuir 2010, 26(2), 752–758
Cheng et al.
Article
Figure 7. FT-IR spectra of DTCSCu before and after adsorption of dyestuffs: (a) DTCSCu; (b) AO7 on DTCSCu; (c) AO10 on DTCSCu; (d) AG25 on DTCSCu; (e) AR18 on DTCSCu.
Figure 8. TGA curves of DTCSCu and dye loaded DTCSCu.
Figure 9. DSC curves of DTCSCu and dye loaded DTCSCu: (a) AO10 loaded DTCSCu; (b) AR18 loaded DTCSCu; (c) DTCSCu; (d) AO7 loaded DTCSCu; (e) AG25 loaded DTCSCu.
broad peaks at 1108, 1082, and 1023 cm-1 are caused by the coupling vibration of νC-O and νC-O-C. They are always split into multiple peaks because various hydrogen bonds are formed. After the adsorption of the dyes, the bands of the four characteristic vibrations of the aromatic rings are observed at about 1508, 1423, 1331, and 1266 cm-1. The peaks at 1108, 1082, and 1023 cm-1 caused by the coupling of νC-O and vC-O-C become much broader with the incorporation of the significant νsSO2 vibration.49 The peak at 1208 cm-1 also broadens with the incorporation of the νasSO2 vibration.50 It has been reported that the peaks of the νSO2 vibration can broaden significantly after chelating to metal ions.51 In particular, the δNH peaks for AG25 at 1593 and 1570 cm-1 are clearly identifiable. These data indicate (49) Vongchan, P.; Sajomsang, W.; Subye, D.; Kongtawelert, P. Carbohydr. Res. 2002, 337, 1233–1236. (50) Bellamy, L. J.; Willams, R. L. J. Chem. Soc. 1957, 863–868. (51) Zhao, Y.; Sun, B.; Xu, Y.; Wang, D.; Weng, S.; Wu, J.; Xu, D.; Xu, G. J. Mol. Struct. 2001, 560, 115–120.
Langmuir 2010, 26(2), 752–758
Figure 10. Proposed interaction modes between the dyes and DTCSCu.
that DTCSCu effectively adsorbs the dyes and the sulfonates interact with copper ions. After the dyes chelating to the copper ions, the thermal stability of DTCSCu enhanced. As shown in Figure 8, four weight loss stages can be distinguished from the TGA curves. For DTCSCu, the weight loss in the 50-160 C range can be confidently attributed to the release of adsorbed water. The second weight loss occurring in the 210-300 C range is due to the breakdown of the dithiocarbamate and partial hydroxyl goups.52,53 These processes are followed by glucose ring scissions (stage III) and carbonization of the material (stage IV).54 The weight loss is (52) Patel, A.; Mequanint, K. Polymer 2009, 50, 4464–4470. (53) Athawale, V. D.; Lele, V. Starch 2000, 52, 205–213. (54) Zhang, X. Q.; Golding, J.; Burgar, I. Polymer 2002, 43, 5791–5796.
DOI: 10.1021/la9039489
757
Article
Cheng et al.
continuous in the last three stages. After chelating adsorption occurred, complex of the anionic dye with DTCSCu (DTCSCudye) was formed. The DTCSCu-dye shows a lower decomposition rate than that of DTCSCu, indicating formation of a more stable complex. All the DTCS-dye curves appear similar in shape, and four weight loss stages can also be identified. From DSC curves (Figure 9), the four decomposition stages are observed clearly. It can be seen that the endothermic peaks at 80 C correspond to the loss of bound water. The first exothermic peak of DTCSCu appears at 215 C for decomposition of dithiocarbamate group in DTCSCu, whereas the peak exhibits at higher temperature (225-255 C) for DTCSCu-dye. It demonstrates that the anionic dye strongly binds to the copper complex of dithiocarbamate modified starch. The second and third exothermic peaks are wide and assigned to strong decomposition of starch and the anionic dye.55,56 According to the above facts, the possible interaction mode between the dye and DTCSCu was illustrated in Figure 10.
4. Conclusion In this study, we investigated the interaction of anionic dyes (acid orange 7, acid orange 10, acid green 25, and acid red 18) with the copper(II) complex of dithiocarbamate-modified starch (DTCSCu). The sorption studies demonstrated that the structure of the dye affected its capacity for adsorption and its sorption (55) Liu, X.; Yu, L.; Liu, H.; Chen, L.; Li, L. Polym. Degrad. Stab. 2008, 93, 260–262. (56) Vasconcelos, H. L.; Guibal, E.; Laus, R.; Vitali, L.; Favere, V. T. Mater. Sci. Eng., C 2009, 29, 613–618.
758 DOI: 10.1021/la9039489
isotherm. The Langmuir-Freundlich isotherm best fitted the experimental data for the adsorption of all four dyes. The capacities for different dyes followed the sequence AO7 > AG25 > AR18 > AO10, which was dictated by their structures. The dyes with sulfonate groups on benzene were better adsorbed than those with sulfonate groups on naphthalene. When the maximum capacity of DTCSCu was achieved at the optimal pH of 4, the molar n(Cu):n(dye) ratio for the smallest dye, AO7, was 1:1. For the larger dyes, the molar n(Cu):n(dye) ratio was approximately 2:1, suggesting the interaction of one dye molecule with two close copper ions. The spectrometric studies confirmed the chelating adsorption that the anionic dye bound to the dinuclear copper core on DTCSCu. The dye adsorption efficiency was largely related to the dinuclear copper cores available on the polymer. The thermogravimetric analyses indicate that the decomposition rates of complexes of DTCSCu with each dye are smaller than that of DTCSCu, suggesting strong chelating interactions between anionic dyes and DTCSCu. Acknowledgment. We thank Dr. Weikuan Li for useful discussions. This work is supported by the National Natural Science Foundation of China (No. 20577034). Supporting Information Available: Parameters of Langmuir-Freundlich isotherm for dyes adsorption onto DTCSCu at different temperatures (Table S1); sorption of AO7, AO10, AG25, and AR18 at different temperatures (Figures S1-S4); dye desorption at various solutions (Figure S5). This material is available free of charge via the Internet at http://pubs.acs.org.
Langmuir 2010, 26(2), 752–758
