pubs.acs.org/Langmuir © 2009 American Chemical Society
Azoic Dye Hosted in Layered Double Hydroxide: Physicochemical Characterization of the Intercalated Materials Sujata Mandal,†,‡ Didier Tichit,*,† Dan A. Lerner,† and Nathalie Marcotte† †
Institut Charles Gerhardt UMR 5253 CNRS/UM2/ENSCM/UM1, Mat eriaux Avanc es pour la Catalyse et la Sant e, 8 rue Ecole Normale, 34296 Montpellier Cedex 5, France, and ‡National Chemical Laboratory, Dr. Homi Bhabha Road, Pune 411008, Maharashtra, India Received April 6, 2009. Revised Manuscript Received June 24, 2009 Intercalation compounds were obtained by introduction of guest methyl orange (MO) into the interlayer space of host Mg/Al and Ni/Al layered double hydroxides (LDHs). Three synthesis methods of organic anion-LDH intercalation compounds, i.e., coprecipitation, reconstruction of the MII(Al)O mixed oxides, and anion exchange of LDH, were compared. The former method gives rise to a highly organized MO-intercalated Mg/Al LDH with an interlayer spacing of 2.43 nm and up to seven (00l) reflection orders. Reconstruction of the mixed oxide by intercalation with MO in the restored LDH was only achieved with Mg(Al)O. In this case, a competitive adsorption of MO on the external surface of the crystals was also seen. On the other hand, intercalation compounds exhibiting interlayer spacing of 2.43 nm were obtained with both Mg- and Ni-containing LDH using the anionic exchange method. The equilibrium and kinetic adsorption properties of the compounds were analyzed by UV-visible spectroscopy in anionic exchange experiments. According to the pseudo-second-order adsorption model, the amounts of adsorbed MO reach 3.82 and 2.83 mequiv/g for Mg- and Ni-containing LDHs, respectively, which are close to their respective anionic exchange capacity. The adsorption rates are on the same order of magnitude for the two LDHs (0.10-0.44 g mmol-1 min-1), the equilibrium being reached in less than 60 min. The decomposition of MO by combustion of the organic moieties under an oxidizing atmosphere is delayed in Mg-containing MO-LDH hybrids when compared to the free MO molecule, showing that the thermal stability of MO species is enhanced after intercalation. In Ni-containing LDH, the main decomposition step of MO occurs 300 °C below that of Mg-containing LDH. This was rationalized in terms of a catalysis by the Ni-containing oxides formed during the thermal treatment. So these materials exhibit several advantages useful for the development of eco-friendly processes for the removal of dyes from effluents of textile, plastic, and paper industries.
1. Introduction Layered double hydroxides (LDHs), also known as anionic clays, have received much attention as inorganic host structures for adsorbing large organic molecules, likely by intercalation, leading to organic-inorganic hybrid materials. LDHs consist of a flat two-dimensional structural network composed of brucite-like layers [Mg(OH)2] containing bivalent and trivalent metal cations. Their general formula is [M1-x2þMx3þ(OH)2]xþAx/nn- 3 mH2O, where M2þ is Mg2þ, Ni2þ, Zn2þ, Cu2þ, Co2þ, etc., M3þ is Al3þ, Ga3þ, Fe3þ, Cr3þ, etc., An- is an exchangeable anion, and the value of x ranges between 0.2 and 0.33. The partial replacement of bivalent cations with trivalent cations generates an excess of positive charge in the layers that is counterbalanced by anions located in the interlayer domains. A relevant feature with regard to the hybrid materials obtained from host LDHs is that a large variety of guest organic molecules can be incorporated into the interlayer space, e.g., aliphatic and aromatic carboxylates, sulfonates, and phosphonates,1-4 alkyl sulfate anions,5 porphyrines and phthalocyanine derivatives,6 anionic drugs,7,8 and *To whom correspondence should be addressed. Telephone: 00 (33)4 67 16 34 77. E-mail:
[email protected]. (1) Meyn, M.; Beneke, K.; Lagaly, G. Inorg. Chem. 1990, 29, 5201–5207. (2) Carlino, S. Solid State Ionics 1997, 98, 73–84. (3) Khan, A. I.; O’Hare, D. J. Mater. Chem. 2002, 12, 3191–3198. (4) Rives, V.; Ulibarri, M. A. Coord. Chem. Rev. 1999, 181, 61–120. (5) Kopka, H.; Beneke, K.; Lagaly, G. J. Colloid Interface Sci. 1988, 123, 427– 436. (6) Newman, S. P.; Jones, W. New J. Chem. 1998, 105–115. (7) Del Hoyo, C. Appl. Clay Sci. 2006, 36, 103–121. (8) Palinko, I. Nanopages 2006, 1, 295–314. (9) Aloisi, G. G.; Costantino, U.; Elisei, F.; Latterini, L.; Natali, C.; Nocchetti, M. J. Mater. Chem. 2002, 12, 3316–3323.
10980 DOI: 10.1021/la901201s
organic dyes.9,10 The incorporation of well-chosen organic guests allows one to tailor the functionalities of the hybrids, which find applications as pigments,10,11 optical devices,12,13 nano- and macrofillers in polymer composites,14 drug carriers,7 and catalysts.15,16 Several main structural aspects concerning these hybrid assemblies have been pointed out, particularly the large expansion of the interlayer space upon anion exchange,4 the different orientations, the reorientation or the grafting of the guest entities to the hydroxide layers,2,17-19 the formation of second-stage intermediate phases,20-22 and the interaction between neighboring guest species.23 The thermal stability of the intercalated species is (10) Bauer, J.; Behrens, P.; Speckbacher, M.; Langhals, H. Adv. Funct. Mater. 2003, 13, 241–248. (11) Laguna, H.; Loera, S.; Ibarra, I. A.; Lima, E.; Vera, M. A.; Lara, V. Microporous Mesoporous Mater. 2007, 98, 234–241. (12) Zlatanova, K.; Markovsky, P.; Spassova, I.; Danev, G. Opt. Mater. 1996, 5, 279–283. (13) Ogawa, M.; Kuroda, K. Chem. Rev. 1995, 95, 399–438. (14) Leroux, F.; Taviot-Gueho, C. J. Mater. Chem. 2005, 15, 3628–3642. (15) Figueras, F.; Kantam, M. L.; Choudary, B. M. Curr. Org. Chem. 2006, 10, 1627–1637. (16) Choudary, B. M.; Kantam, M. L.; Kavita, B.; Reddy, C. V.; Rao, K. K.; Figueras, F. Tetrahedron Lett. 1998, 39, 3555–3558. (17) Fogg, A. M.; Rohl, A. L.; Parkinson, G. M.; O’Hare, D. Chem. Mater. 1999, 11, 1194–1200. (18) Prev^ot, V.; Forano, C.; Besse, J. P. Inorg. Chem. 1998, 37, 4293–4301. (19) Li, F.; Zhang, L.; Evans, D. G.; Forano, C.; Duan, X. Thermochim. Acta 2004, 424, 15–23. (20) Fogg, A. M.; Dunn, J. S.; O’Hare, D. Chem. Mater. 1998, 10, 356–360. (21) Williams, G. R.; O’Hare, D. Chem. Mater. 2005, 17, 2632–2640. (22) Pisson, J.; Taviot-Gueho, C.; Israeli, Y.; Leroux, F.; Munsch, P.; Itie, J. P.; Briois, V.; Morel-Desrosiers, N.; Besse, J. P. J. Phys. Chem. B 2003, 107, 9243– 9248. (23) Costantino, U.; Coletti, N.; Nocchetti, M.; Aloisi, G. G.; Elisei, F. Langmuir 1999, 15, 4454–4460.
Published on Web 07/09/2009
Langmuir 2009, 25(18), 10980–10986
Mandal et al.
Article
generally enhanced by electrostatic interactions and/or hydrogen bonding with the host layers, and its confinement in the twodimensional interlayer space.11,18 Several methods for synthesizing organic molecule-LDH assemblies are available (coprecipitation, direct anionic exchange, and reconstruction) as recently reviewed by several authors.2,3,6,14 They allow a tuning of their properties. The formation of intercalation compounds is particularly relevant for immobilization or adsorption of organic dyes. The intercalation compounds are indeed useful for improving the thermal stability and photostability of pigments, such as the azo dye HSAB [2-hydroxy5-(4-sulfophenylazo)benzoate disodium salt].24 A promising application is to use LDHs as an adsorbent to remove dyes from effluents of textile, plastic, and paper industries, with the aim of developing eco-friendly processes. Indeed, dyes color the water, hindering the penetration of light that is essential for aquatic ecosystems. Methyl orange (MO) {4-[40 -(dimethylamino)phenylazo]benzesulfonate} is an interesting model of dyes encountered in effluents of the textile industry. Previous studies have reported on the intercalation of MO by anionic exchange into Zn/Al-Cl LDHs.23 Studies on the adsorption of MO by reconstruction, using the so-called “memory effect” of mixed oxides MII(Al)O (MII is Mg or Zn) obtained by calcination of LDH precursors, have also been reported.11,25 They showed that intercalation of MO was successfully achieved by reconstruction of Zn(Al)O25 and failed with Mg(Al)O.11 These results and the strong interest in the adsorption of dyes for water treatment prompted us to study the intercalation of MO into Mg/Al and Ni/Al LDH hosts. The anionic exchange method, generally highly efficient for the intercalation of large organic molecules, was used in both LDHs.1 The direct coprecipitation method, generally more versatile, was also used for the Mg/Al LDH.2,6 To the best of our knowledge, these methods have not been investigated until now for the intercalation of MO in these LDHs. Moreover, the adsorption properties of MO by the two MII(Al)O (MII is Mg or Ni) mixed oxides are also reported. These two LDHs were chosen because of their well-known difference in thermal stability and reconstruction ability.26-29 Great differences were expected, in particular, in the decomposition processes of the intercalated dyes because of the nature of the divalent cation. The structure of the intercalated compounds and the host-guest interactions have been characterized by means of XRD as well as DRUV and DRIFT spectroscopy. The thermal decomposition of the different materials under an inert or oxidizing atmosphere has been examined by TG analysis.
2. Materials and Methods 2.1. Preparation of LDH Structures. LDHs with a M2þ/
3þ
Al (M2þ is Mg2þ or Ni2þ) molar ratio of 2 have been synthesized by coprecipitation following a classical method.29 Coprecipitation of an aqueous solution containing the required amounts of M2þ (Mg2þ and Ni2þ) and Al3þ nitrates (MAN and NAN samples, respectively) or chloride salts (MACl sample), with NaOH or with a NaOH/Na2CO3 (1/1) solution (MAC sample), (24) Tang, P.; Xu, X.; Lin, Y. L.; Li, D. Ind. Eng. Chem. Res. 2008, 47, 2478– 2483. (25) Ni, Z. M.; Xia, S. J.; Wang, L. G.; Xing, F. F.; Pan, G. X. J. Colloid Interface Sci. 2007, 316, 284–291. (26) Perez-Ramı´ rez, J.; Abello, S.; Van der Pers, N. M. J. Phys. Chem. C 2007, 111, 3642–3650. (27) Sato, T.; Fujita, H.; Endo, T.; Shimada, M.; Tsunashima, A. React. Solids 1988, 5, 219–228. (28) Rebours, B.; D’Espinose de la Caillerie, J. B.; Clause, O. J. Am. Chem. Soc. 1994, 116, 1707–1717. (29) Prinetto, F.; Tichit, D.; Teissier, R.; Coq, B. Catal. Today 2000, 55, 103– 116.
Langmuir 2009, 25(18), 10980–10986
was performed at a constant pH (9.5 ( 0.2) under air. The resulting suspension was stirred at 80 °C for 17 h. The solid was recovered by centrifugation and dried overnight at 80 °C. 2.2. Intercalation of MO. The samples prepared following the three classical methods available to achieve the intercalation of large molecules within the interlayer space of LDHs are (i) coprecipitation, (ii) reconstruction of the MII(Al)O mixed oxides, and (iii) anionic exchange which are described below. (i) For coprecipitation of MO-LDH molecules, 100 mL of a solution containing 40 mmol of Mg(NO3)2 3 6H2O and 20 mmol of Al(NO3)3 3 9H2O (Mg/Al molar ratio of 2) and 100 mL of an aqueous solution of methyl orange sodium salt (MO) at 0.20 mol/L (Acros Organics) were delivered simultaneously with two peristaltic pumps into a beaker and were coprecipitated by an aqueous solution of NaOH (1 M). The base solution was added using a pH-Stat apparatus to maintain a constant pH of 9.5 ( 0.2. The precipitate was aged for 17 h at 70 °C, followed by separation through centrifugation. The solid thus obtained was washed thoroughly with distilled water until the washing was neutral and dried at 50 °C in an air oven. This sample was hereafter denoted as MO-Cp-MAN. (ii) For reconstruction of mixed oxides with MO, the first step was the preparation of the MII(Al)O (MII is Mg or Ni) mixed oxides by calcination of the LDH precursors at 450 °C for 4 h under an air flow (100 cm3/min), the temperature ramp being 1 °C/min from room temperature. After being cooled to room temperature, the samples were kept in a desiccator to avoid any adsorption of moisture from air. They were hereafter denoted as MAN-cal and NAN-cal. Reconstruction experiments were conducted in the liquid phase by dispersing 0.5 g of the calcinated mixed oxides in 200 mL of an aqueous solution of MO (10 mmol/L) under vigorous mechanical stirring for 2 h, at room temperature and neutral pH. The materials were recovered by centrifugation. The samples are hereafter denoted MO-MANcal and MO-NAN-cal, respectively. (iii) Anion exchange experiments with LDHs with MO were performed by mixing 0.5 g of each of the LDH precursors (MAN and NAN) with 200 mL of an aqueous solution of MO (10 mmol/L) at 25 °C for 2 h. The solids recovered by centrifugation were washed thoroughly with distilled water and dried at 60 °C. The MO-LDH hybrids thus obtained from MAN and NAN are hereafter denoted MO-ex-MAN and MO-ex-NAN, respectively. 2.3. Adsorption Experiments. Dye uptake curves for the different LDHs (MAN, NAN, MACl, and MAC) and MII(Al)O mixed oxides (MAN-cal and NAN-cal) have been determined by mixing 0.5 g of the solids with various volumes (20-500 mL) of an aqueous solution of MO (10 mmol/L) at 25 °C for 2 h to yield a concentration of MO ranging from 0.4 to 10 mmol/g of solid. The solids were then recovered by centrifugation. The concentration of MO incorporated in the LDHs was calculated from the difference in the UV-visible absorption spectra of MO before and after anionic exchange using Beer-Lambert’s law at 466 nm (ε = 12860 L mol-1 cm-1). The solids were washed thoroughly with distilled water, dried at 50 °C, and kept in a desiccator for characterization studies. 2.4. Characterization Techniques. The chemical composition and physical characteristics of the as-synthesized and intercalated LDH materials were determined using different techniques. Percentages of Mg, Ni, and Al in the as-synthesized LDHs were determined using an atomic absorption spectrometer (Varian SpectrAA 220). The C and N content in the LDHs was determined using a CHNS (Carlo-Erba, Italy) analyzer. The Cl content was determined titrimetrically with silver nitrate using K2CrO4 as an indicator. XRD patterns were recorded on a Bruker D8 Advance X-ray diffractometer using Cu KR1 radiation (λ= 1.542 A˚, 40 kV and 50 mA). Data were collected at 2θ angles between 2° and 70°, with a step size of 0.02° and a counting time of 1 s/step. The concentration of MO in solution was determined using UV-visible absorption spectroscopy (Uvikon XL from Bio-Tek Instruments) using Beer-Lambert’s law. Diffuse DOI: 10.1021/la901201s
10981
Article
Mandal et al.
reflectance UV-visible (DRUV) spectra of the solid samples were recorded on a Perkin-Elmer Lambda 14 spectrophotometer equipped with an integrating sphere (Labsphere) using BaSO4 as the reference. The spectra were displayed as their KubelkaMunk transform [F(R)]. N2 sorption experiments at 77 K were conducted with samples previously calcined at 450 °C for 5 h and outgassed at 250 °C (10-4 Pa) with a Micromeritics ASAP 2000 instrument. Specific surface areas were calculated using the BET method. TG-DTG experiments were conducted using a Netzch TG 209 C thermogravimetric analyzer at a typical heating rate of 5 °C/min from room temperature to 900 °C in an O2/N2 stream (5/95) or nitrogen at a flow rate of 20 mL/min. Diffuse reflectance infrared Fourier transform (DRIFT) spectra of the samples were recorded on a Bruker Vector 22 spectrometer in the 4000600 cm-1 wavenumber range.
3. Results and Discussion 3.1. Chemical Composition and XRD of the Host LDHs. All the as-synthesized samples exhibit the typical XRD pattern corresponding to LDHs (Figure 1). The crystallinity is higher for the Mg-containing than for the Ni-containing LDH. The reflections were indexed in a hexagonal lattice with R-3m rhombohedral symmetry. The unit cell parameters, c and a, can be estimated from the positions of the (003) (c=3d003) and (110) (a=2d110) reflections. As expected, the c parameter is similar (c=2.640 nm) for the two MAN and NAN nitrate-containing LDHs (Figure 1). Lattice a parameters of 0.304 and 0.302 nm were calculated for MAN and NAN, respectively. These values agree with those reported in the literature for Mg- and Ni-containing LDHs having a M2þ/Al3þ molar ratio close to 2 in the brucite-like layers.30 The (003) Bragg reflections for Mg/Al LDHs intercalated with chloride and carbonate anions give interlayer distances of 0.780 and 0.747 nm, respectively, instead of the value of 0.880 nm found with the nitrate compensating anion. These values are in good agreement with those generally reported in the literature.31 The elemental analyses of the as-synthesized LDHs reported in Table 1 show that all of them exhibit a M2þ/Al3þ molar ratio close to 2, similar to the precursor ratios in solution. The proposed structural formulas are given in Table 1. For the carbonate-containing sample, the FTIR spectrum (not reported) shows that both carbonate and bicarbonate anions are present as generally observed.32 On the basis of the formula reported in Table 1, the anionic exchange capacities (AEC) of MAN and NAN are 3.60 and 2.80 mequiv/g, respectively. 3.2. Characterization of MO-Containing Samples. 3.2.1. Coprecipitation Method. The XRD pattern of MO-Cp-MAN exhibits sharp and intense lines in the 2θ range between 2° and 30° which correspond to diffraction by (00l) planes (Figure 1). The d003 interlayer spacing of 2.43 nm is in agreement with that reported by Costantino et al.23 for Zn/Al LDH intercalated with MO by anionic exchange. Noteworthy is the fact that up to seven (00l) harmonics are observed in relation to the large size of intercalated MO species. Therefore, we can conclude that the spontaneous self-assembly method, although poorly attractive for the removal of dyes from effluents, allows Mg/Al LDH intercalated with MO as the unique compensating anion to be obtained. These MO-intercalated LDHs are highly crystallized.
(30) Drits, V. A.; Bookin, A. S. In Layered Double Hydroxides: Present and Future; Rives, V., Ed.; Nova Science Publishers, Inc.: New York, 2001; p 39. (31) Suzuki, E.; Idemura, S.; Ono, Y. Clays Clay Miner. 1989, 37, 173–178. (32) Rives, V. In Layered Double Hydroxides: Present and Future; Rives, V., Ed.; Nova Science Publishers, Inc.: New York, 2001; p 115.
10982 DOI: 10.1021/la901201s
Figure 1. XRD patterns of MAN (a), NAN (b), MO-cp-MAN (c), MAN-cal (d), MO-MAN-cal (e), NAN-cal (f), MO-NAN-cal (g), MO-ex-MAN (h), MO-ex-NAN (i), and pristine MO (j).
3.2.2. Reconstruction Method. The experiments are based on the reconstruction ability of the mixed oxides, which is known to be very different for Mg- and Ni-containing samples.26-29 Previous studies have shown that the thermal decomposition of Mg/Al and Ni/Al LDHs proceeds topotactically33 but that the composition of the MII(Al)O mixed oxides varies locally. Indeed, a segregation of the spinel-like phase occurs in calcinated Ni/Al LDH, which is responsible for the higher thermal stability of the Ni(Al)O mixed oxide compared to that of NiO.34 This feature and the greater difficulty in rehydrating NiO rather than MgO into their respective hydroxides34 explain why reconstruction of calcinated Ni/Al LDH cannot be achieved in presence of water at ambient temperature under atmospheric pressure, in contrast to calcinated Mg/Al LDH. Partial reconstruction of Ni(Al)O (∼80%) has been achieved by hydrothermal treatment at 160 °C in ammonia.29 We can therefore expect that reconstruction in an aqueous solution of MO would occur with calcinated MAN but fail with NAN-cal. The XRD patterns of the Mg- and Ni-containing mixed oxides (MAN-cal and NAN-cal, respectively) calcinated at 450 °C (Figure 1) show broad lines at 2θ angles of ∼35-38°, ∼43°, and ∼62-63° corresponding to the rock-salt MgO (periclase) and NiO (bunsenite) structures (ICDD powder diffraction files 45-0946 and 47-1049, respectively). The crystallinity of NANcal is higher than that of MAN-cal according to their specific surface areas, reaching 61.5 m2/g for NAN-cal and 123 m2/g for MAN-cal. Each of the calcinated MII(Al)O mixed oxides (0.5 g) has been mixed with 2 mmol of MO to achieve reconstruction. Assuming that reconstruction of the mixed oxides in the lamellar structure is achieved with MO as a unique compensating anion, the quantity of MO that can be theoretically introduced into MAN and NAN samples should be 7.8 and 5.1 mmol/g of mixed oxides, respectively. The amount of MO mixed with MAN-cal and NAN-cal in our experiments thus corresponds to ∼50 and 80%, respectively, of the amount necessary for total restoration of the lamellar (33) Reichle, W. T.; Kang, S. Y.; Everhardt, D. S. J. Catal. 1986, 101, 352–359. (34) Clause, O.; Rebours, B.; Merlen, E.; Trifiro, F.; Vaccari, A. J. Catal. 1992, 133, 231–246.
Langmuir 2009, 25(18), 10980–10986
Mandal et al.
Article Table 1. Chemical Compositions and Proposed Formulas of the As-Synthesized LDHs chemical composition (wt %)
sample MAN NAN MAC MACl
2þ
3þ
M
Al
18.34 33.40 19.86 20.20
9.71 7.65 11.18 11.20
C
N
Cl
formula
0.035 0.035 3.00 0.24
4.88 3.90 0.00 0.00
0.00 0.00 0.00 13.83
[Mg0.68Al0.32(OH)2][(CO3)2.58 10-3(NO3)0.31] 3 0.59H2O [Ni0.67Al0.33(OH)2][(NO3)0.33(CO3)0.0034] 3 0.85H2O [Mg0.66Al0.33(OH)2][(CO3)0.13(HCO3)0.07] 3 0.55H2O [Mg0.67Al0.33(OH)2][(CO3)0.016Cl0.31] 3 0.48H2O
Table 2. Kinetic Parameters, Correlation Coefficients (r2), and Reduced χ2 Values of Three Kinetic Models First-Order sample
qe
k1
r2
χ2
MAN NAN MAN-cal
3.63 2.76 2.22
0.208 0.425 0.016
0.997 0.995 0.983
0.0049 0.0055 0.0122
Second-Order sample
qe
k2
r2
χ2
MAN NAN MAN-cal
3.82 2.83 2.63
0.102 0.443 0.007
0.996 0.999 0.991
0.0069 0.0007 0.0060
Elovich sample
R
β
r2
χ2
MAN NAN MAN-cal
35138 9.9 1012 0.573
0.265 0.0087 0.483
0.963 0.987 0.971
0.0703 0.0029 0.0205
structures. The MO equilibrium sorption capacity (qe) of MANcal, calculated from adsorption kinetics data (see Table 2), is 2.63 mmol/g which corresponds to ca. 34% of the theoretical amount necessary to restore the lamellar structure. On the other hand, a value of 0.24 mmol/g was obtained for NAN-cal after 2 mmol of MO had been stirred with 0.5 g of the mixed oxide. This corresponds to only 4.7% of the theoretical amount necessary to restore the lamellar structure and suggests that reconstruction can occur for MO-MAN-cal and, on the other hand, fail for Ni-containing LDH. This is consistent with the structural evolutions observed by XRD (Figure 1). After MAN-cal had been mixed with an aqueous solution of MO, significant changes are observed: two new and low-intensity peaks appear at 2θ angles lower than 10° and correspond to diffraction by planes (003) and (006), respectively, of the layered structure. They show that a partial reconstruction takes place. The interlayer distance d003 of 2.43 nm is similar to that of MO-cp-MAN prepared by coprecipitation. Besides, the presence of reflections that can be assigned to the periclase-like phase at 2θ angles of ∼43° and ∼62.3° in the XRD pattern of MO-MAN-cal, corresponding to distances of 2.10 and 1.49 nm, respectively, shows that the reconstruction is not complete. This is in agreement with the amount of MO retained (2.63 mmol/g) that represents only 34% of the theoretical value. Several intense diffraction peaks in the 2θ angle range below 30° are seen at 3.95°, 4.95°, 20.20°, and 24.90°, which correspond to distances of 22.34, 17.83, 4.39, and 3.57 nm, respectively. They are due to the presence of MO adsorbed on the external surface of the crystals. This attribution is confirmed by the XRD pattern of the crystallized pristine MO dye (Figure 1) that exhibits the previous reflections, but with different relative intensities. This suggests adsorption of the dye with a preferential orientation on the crystal surface, as previously reported by Laguna et al.11 using Langmuir 2009, 25(18), 10980–10986
13
C CP/MAS NMR spectroscopy. However, in their work, the observed interlayer distance corresponds to intercalation of carbonates rather than that of MO. The lack of MO intercalation they observed is not surprising since the mixed oxide was mixed with an amount of MO corresponding to ca. 9% of the AEC of the restored lamellar structure. Reconstruction is therefore better performed with carbonates present in the ambient atmosphere rather than with MO. As opposed to MO-MAN-cal, MO-NANcal exhibits the XRD patterns of the mixed oxide only (Figure 1). This shows that the reconstruction has failed, as supported by the calculated equilibrium sorption capacities. In contrast to MOMAN-cal, the lamellar structure is not recovered, though NANcal has been mixed with a larger amount of MO relative to its AEC. This is due to the particularly high stability of this mixed oxide.26,29 To conclude, intercalation of MO in Mg-containing LDH has been achieved by reconstruction of the Mg(Al)O mixed oxide. Additional adsorption of MO on the external surface of the crystals was observed. On the other hand, reconstruction was not successful for Ni-containing mixed oxides, whose high stability obviously impedes intercalation of MO. 3.2.3. Anion Exchange Method. The XRD patterns of MOex-MAN and MO-ex-NAN obtained by exchange of 0.5 g of MAN and NAN with 200 mL of an aqueous solution of MO (10 mmol/L) are reported in Figure 1. The amount of MO used (4 mmol/g of LDH) represents between ∼110% (MAN) and 140% (NAN) of the theoretical AEC of the original LDHs and is therefore able to achieve complete exchange. The XRD patterns of the two MO-exchanged LDHs exhibit the same general shape, with a remarkable increase in crystallinity in comparison to those of MAN and NAN (Figure 1). A shift of the intense (003) reflections toward lower 2θ values is also seen. This is in agreement with the intercalation of MO giving an interlayer spacing of 2.43 nm. Several (00l) reflections are observed in relation to the large size of the intercalated MO species. The presence of the very weak (006) reflection at a d006 of 0.443 nm assigned to the NO3phase shows that NO3- is not totally exchanged by MO. We showed that the partially reconstructed MO-MAN-cal (Figure 1) exhibits the presence of an additional MO phase at a 2θ angle of Cl- > (CO32- þ HCO3-). NO3- and Cl- are almost totally exchanged with MO. This must also be the case for the monovalent HCO3- anions present in the MAC sample. Therefore, in the latter material, ∼30% of the initially present CO32- is not exchanged by MO. Unlike the sorption capacity, the adsorption rate decreases as follows: NO3- > (HCO3- þ CO32-) > Cl-. The MO uptake at equilibrium (qe) was reached after 0.5 g of solid and various amounts of MO (from 0.4 to 10 mmol) were mixed for 120 min. Figure 4 depicts the room-temperature equilibrium adsorption of MO for MAN and NAN. Three models were used in an attempt to fit the equilibrium isotherms: (i) the Langmuir (eq 5), (ii) the Freundlich (eq 6), and (iii) the Langmuir-Freundlich (eq 7) models: qe ¼
q m KL Ce 1 þ KL Ce
qe ¼ KF Ce n qe ¼
q m KL Ce n 1 þ KL Ce n
ð5Þ ð6Þ ð7Þ
where qe is the equilibrium sorption capacity (millimoles per gram), qm is the theoretical maximum monolayer adsorption capacity (millimoles per gram), Ce is the concentration of MO in solution at equilibrium (millimoles per liter), and KL, KF, and n are empirical constants. The experimental equilibrium MO uptake (qe) was calculated using eq 1, in which t is the time required to reach equilibrium (120 min). On the basis of r2 and χ2 statistical parameters, good fits were obtained with the Langmuir and Langmuir-Freundlich models for all the samples (Table 3). For the two LDHs, two adsorption domains can be distinguished (Figure 4). In the first domain, corresponding to MO uptakes of up to ∼3.0 mmol/g for MAN and ∼2.2 mmol/g for NAN, all the MO in solution is taken up by the solid. It is noteworthy that these limit values are close to the theoretical AEC. In the second domain ([MO]uptake > 3.0 mmol/g for MAN and 2.2 mmol/g for NAN), the MO uptake is always lower than that expected from the MO concentration in solution. This could Langmuir 2009, 25(18), 10980–10986
Figure 4. MO uptake at equilibrium by MAN (9) and NAN (O) as a function of MO concentration in solution. Solid lines represent the fit by the Langmuir model. Table 3. Langmuir, Freundlich, and Langmuir-Freundlich Isotherm Constants for Absorption of MO by LDHs Langmuir sample
qm
KL
r2
χ2
MAN NAN
1.3 106 3634.6
3.9 10-7 0.00032
0.475 0.599
2.4196 3.8708
Freundlich sample
n
KF
r2
χ2
MAN NAN
6.22 4.21
0.00016 0.022
0.999 0.988
0.0030 0.1369
Langmuir-Freundlich sample
qm
KL
n
r2
χ2
MAN NAN
5.43 6.83
1.4 10-8 7.1 10-12
12.11 24.88
1.000 1.000
5.0 10-6 6.3 10-7
reveal an external adsorption process, which would be consistent with the calculated maximum monolayer adsorption capacity (qm) values that are higher than the theoretical AEC values. 3.4. TG-DTG Analyses. To gain insight into the influence of the cationic composition of the layers on the thermal stability of the MO-intercalated LDHs, thermogravimetric analyses were performed under oxidizing and inert atmospheres. The solids were obtained by exchange of the host LDHs with a MO concentration in solution corresponding to 4 mmol/g of solid. The anionic exchange results showed that this concentration produces highly exchanged LDHs. The TG-DTG curves of MAN and NAN obtained under air flow display the typical profile generally reported for LDHs.32 A first mass loss event recorded between room temperature and 180-200 °C corresponds to the removal of adsorbed and interlayer water molecules, and a second one, between 200 and 600 °C, corresponds to the dehydroxylation of the brucite-like layers and decomposition of the anions, giving an intense DTG peak at 423 °C (MAN) or 312 °C (NAN). The total mass losses reach ca. 50.5 and 41.5% for MAN and NAN, respectively. The TG-DTG curves of the MO-exchanged LDHs recorded under air flow are shown in Figure 5. Compared to those of the parent nitrate-containing LDHs (MAN and NAN), several relevant features deserve comment. (i) For MO-ex-MAN, the second mass loss event observed above 200 °C is larger and occurs DOI: 10.1021/la901201s
10985
Article
Mandal et al.
MO decomposition. Perez-Ramı´ rez et al. have already reported a lower thermal stability of Ni/Al LDH as compared to Mg/Al LDH, with a CO32- decomposition at 300 °C instead of 380 °C.26 Comparison between the air flow TG-DTG results of MO-ex-NAN and an experiment performed under a nitrogen flow (Figure 5b,c) confirms the assumption of a Ni-catalyzed decomposition of MO. Indeed, decomposition under nitrogen reveals the presence of a broad DTG peak at 472 °C, therefore 70 °C above the decomposition observed under air. This MO decomposition corresponds to a mass loss of ca. 32.5%.
Figure 5. TG-DTG curves of MO-ex-MAN under air flow (a), MO-ex-NAN under air flow (b), MO-ex-NAN under N2 flow (c), and MO under air flow (d).
in four steps appearing as distinct DTG peaks at 350, 400, 475, and 675 °C. A single peak at 405 °C instead of 312 °C is observed for MO-ex-NAN. (ii) An additional mass loss accounting for 18% in MO-ex-MAN is observed between 550 and 700 °C. This loss accounts for only 6% for MO-ex-NAN and is shifted above 700 °C. (iii) The total mass losses increase to 69 and 62.5% for MO-ex-MAN and MO-ex-NAN, respectively. The behavior observed for MO-ex-MAN accounts for the concurrent dehydroxylation of the layers and decomposition phenomena of the intercalated MO organic moieties. The latter occur in several stages corresponding to successive partial decomposition steps. Indeed, the TG-DTG curve of pristine MO depicted in Figure 5 shows decomposition in five distinct oxidation steps between 180 and 600 °C. The mass loss above 550 °C in MO-ex-MAN, which is not observed in the nitratecontaining Mg/Al LDH (MAN), can then be related to the decomposition process of intercalated MO species. As the last DTG peak is observed at ∼600 °C in pristine MO and at 675 °C in MO-ex-MAN, we can conclude that the thermal stability of MO is enhanced after intercalation. Considering that the fully exchanged MO-LDH hybrids are decomposed into oxides, total mass losses only 6% lower than the expected values of ca. 74.8 and 67% can be estimated for the Mg- and Ni-containing samples, respectively. For MO-ex-NAN, the one-step mass loss event between 180 and 500 °C shows that dehydroxylation of the layers and decomposition of MO occur almost simultaneously. The DTG peak that shifts from 312 to 405 °C when nitrate has been exchanged with MO shows that the dehydroxylation is delayed. A residual mass loss (∼6%) is observed between 700 and 900 °C. It is noteworthy that the oxidative decomposition of MO is achieved at a lower temperature in the Ni-containing sample than in pristine MO and the Mg-containing sample. This underlines the fact that the presence of Ni oxides catalyzes
10986 DOI: 10.1021/la901201s
4. Conclusions The intercalation of MO into Mg/Al and Ni/Al LDHs to yield MO-LDH hybrid materials has been investigated using several well-known preparation methods (coprecipitation, reconstruction, and anionic exchange). Both the spontaneous self-assembly method of coprecipitation and the anionic exchange of the nitrate-containing LDHs allow successful incorporation of MO in its anionic form into highly ordered lamellar structures. The intercalation expands the interlayer space of the nitrate-containing LDHs from 0.88 to 2.43 nm. This distance corresponds to a perpendicular orientation of the MO species between the layers. During the anionic exchange, NO3- ions are almost completely displaced, so the amounts of anionic species retained by the different host LDHs depend on their AEC. They increase from 2.83 mequiv/g for Ni-containing LDH to 3.82 mequiv/g for Mgcontaining LDH, values that are close to their theoretical AEC (2.80 and 3.60 mequiv/g, respectively). The kinetics of MO uptake during exchange depend on the nature of the interlayer anion present in the LDH and are ordered as follows: NO3- > (CO32- þ HCO3-) > Cl-. This confirms that carbonate- and chloridecontaining LDHs are less suitable for anionic exchange than nitrate LDHs. The formation of MO-LDH intercalated compounds by reconstruction of the layered structure from their corresponding mixed oxides was possible only in the case of Mg/ Al LDH. TG-DTG analyses revealed that nickel present in the host LDH catalyzes the oxidative decomposition of the MO species, which occurs ∼270 °C below that of the Mg-containing MO-LDH. This lowering of the dye decomposition temperature induced by a selected metal ion, associated with an observed efficient adsorption of dye from polluted waters, lends itself to the development of an elimination process with reduced energy consumption. Therefore, this work shows further that the potential of LDH-based processes in environmental remediation has still not been totally exploited. Acknowledgment. Sujata Mandal thanks “The French Embassy in India” for financial support. We express our gratitude to Geraldine Layrac for assistance with measurements. Supporting Information Available: DRUV spectra of pure MO, MO physically mixed with MAN, and exchanged MOex-MAN. This material is available free of charge via the Internet at http://pubs.acs.org.
Langmuir 2009, 25(18), 10980–10986
