Montmorillonite-supported Pd0, Fe0, Cu0 and Ag0 nanoparticles ...

Applied Surface Science 402 (2017) 314–322

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Full Length Article

Montmorillonite-supported Pd0 , Fe0 , Cu0 and Ag0 nanoparticles: Properties and affinity towards CO2 Nabil Bouazizi a,b,∗ , Diana Barrimo a , Saadia Nousir a , Romdhane Ben Slama b , René Roy a , Abdelkrim Azzouz a,∗∗ a b

Nanoqam, Department of Chemistry, University of Quebec at Montreal, QC, H3C 3P8 Canada Research unit: Environment, Catalyzes and Process Analysis, ENIG University of Gabes, Tunisia

a r t i c l e

i n f o

Article history: Received 19 October 2016 Received in revised form 26 December 2016 Accepted 1 January 2017 Available online 3 January 2017 Keywords: Montmorillonite MNPs CO2 adsorption Surface properties

a b s t r a c t This study reports the carbon dioxide (CO2 ) adsorption on montmorillonite (NaMt) incorporating Cu0 , Fe0 , Pd0 and Ag0 as metallic nanoparticles (MNPs). The changes in structural, textural, morphological and adsorption properties of the resulting materials (NaMt-MNPs) were investigated. Electron microscopy and X-ray diffraction showed that dispersion of fine MNPs occurs mainly within the interlayer space of NaMt, producing a slight structure expansion. This was accompanied by a visible enhancement of the affinity towards CO2 , as supported by thermal programmed desorption measurements. NaMt-MNPs displayed high CO2 retention capacity (CRC) of ca. 657 ␮mol/g for NaMt-Cu as compared to NaMt. This was explained in terms of increased number of available adsorption sites due to enlarged interlayer spaces caused by MNP insertion. The differences in CO2 adsorption capacities clearly demonstrate the key role of MNPs in improving the surface properties and adsorption capacity. The results reported herein open new prospects for clay supported metal nanoparticles as efficient adsorbents for CO2 . © 2017 Elsevier B.V. All rights reserved.

1. Introduction In recent years, the increased CO2 concentration in the atmosphere has drawn great attention being the main anthropogenic contributor to the global warming [1–3]. The huge energy demand using fossil fuels has become a major challenge to be addressed. Therefore, CO2 adsorption has become an interesting research topic to be explored for designing some means which minimize the CO2 amount in the atmosphere and providing new CO2 -based energy sources. Several materials like zeolites, silica, activated carbons and alumina have emerged as an efficient materials for CO2 capture by adsorption. The choice of these materials is due to the simplicity of their low energy-consuming preparation procedures [4]. Recently, organic matrices and amine functionalized adsorbents have been studied for CO2 separation [5]. In order to improve their retention capacity and selectivity for CO2 , amine functionalization on the existing materials was used, but this produces a series of draw-

∗ Corresponding author at: Nanoqam, Department of Chemistry, University of Quebec at Montreal, QC, H3C 3P8 Canada. ∗∗ Corresponding author. E-mail addresses: [email protected] (N. Bouazizi), [email protected] (A. Azzouz). http://dx.doi.org/10.1016/j.apsusc.2017.01.004 0169-4332/© 2017 Elsevier B.V. All rights reserved.

backs [6]. Additionally, these materials cannot be studied for CO2 adsorption–desorption at high temperatures because of amine loss and degradatation or under normal conditions because of the formation of strong carbamate groups. In most applications, a key factor is devoted to the size distribution and metallic nanoparticles (MNPs) dispersion. The latter strongly depend on the nature and synthesis process of the immobilization agent, which prevents both metal particle aggregation and decrease in the contact surface and of most properties of the MNPs [7]. Insertion of MNPs with OH-dendrimers turned out to produce efficient host matrices due to the chelating interaction of the electron pair of the oxygen atom and the sequestrating effect of their branch entanglement [8]. Clay mineral can be suitable materials for certain adsorption purposes in spite of their small distance between the layers. Among these, montmorillonite (Mt) is a clay mineral belonging to the swelling group [9]. The interlayer distance in Na+ exchanged Mt (NaMt) is about 0.1–0.3 nm [10–12], which is smaller than the size of H2 molecule (ca. 0.38 nm) [13,14], but the clay swelling capacity may result in a large distance between layers improving the adsorption effectiveness. In this paper, the dispersion of high number of small metallic nanoparticles is supposed to functionalize the surface of NaMt and to confer adsorption properties to the starting materials. This process is simple, low cost, and can consequently be used

N. Bouazizi et al. / Applied Surface Science 402 (2017) 314–322

315

for the large-scale production of nanocomposite materials. This is regarded as the main objective of the present work. For this purpose, and because of the very few studies reported on this subject, CO2 was adsorbed on NaMt samples loaded with various metal nanoparticles. The changes in structural, morphological, and adsorptive properties of modified NaMt were investigated by using some analysis techniques such as IR, TEM, SEM, XRD, XRF, BET and CO2 -TPD. The effect of the previous saturation of the adsorbents with hydrogen on the affinity towards CO2 was also investigated. 2. Experimental 2.1. Montmorillonite purification and modification A crude bentonite, supplied by Aldrich with 51:32.3 silica to alumina weight ratio, was used as raw material. The latter was prone to a thorough purification into a Montmorillonite-rich clay sample (NaMt), using an effective procedure fully described elsewhere [15]. Briefly, NaMt was obtained after numerous extensive ion exchanges of bentonite using NaCl solutions at 80 ◦ C for several hours in distilled water. NaMt was then washed, centrifuged, and repeatedly dialyzed with distilled water in cellulose bags to remove the residual NaCl. NaMt was centrifuged and dried under at 80 ◦ C overnight. The modification of NaMt was achieved by dispersion of metallic nanoparticles (MNPs) using Pd(II)(OAc); Fe (NO3 )2 , 9H2 O; Cu(NO3)2 , 2·5H2 O and Ag(NO3 )2 as precursors in toluene (99.5%, d = 0.865 gmL−1 ) in the presence of NaBH4 as the reducing agent. The resulting mixture turned black after 5 h of stirring at room temperature, indicating the formation of MNPs within NaMt interlamellar structure. The resulting NaMt-MNPs were dried at 80 ◦ C for 24 h and then stored in sealed enclosure containing dry and oxygen-free nitrogen. 2.2. Materials characterization Insights in the Montmorillonite dispersion with M-NPs were achieved through X-rays diffraction (Siemens D5000 instrument, Co K␣ at 1.7890 Å), in the 2␪ range (10–50◦ ). Besides, infrared spectra were recorded using a KBr IR cell and Fourier Transform Infrared spectroscopy equipment (Model IR 550, Magna Nicolet) at wavenumbers ranging from 4000 to 400 cm−1 . The chemical compositions of the raw clay, unmodified and modified montmorillonite samples were assessed on an X-ray fluorescence spectrometer (XRF) made by ARL ADVANT’X (Intelli Power TM 4200, Thermo Fisher). BET and BJH measurements of the specific surface area, porosity and pore size distribution were performed on a Quantachrome device (Sorptomatic 1990 Thermo Finningen), with an Autosorb automated gas sorption system control. Thus, samples of 100–150 mg were previously dried at 80 ◦ C for 24 h, degassed at 80 ◦ C for 4 h under a 10−4 Torr vacuum and then the nitrogen adsorption was made at −195.7 ◦ C. A JEOL 2100 Transmission Electron Microscope (TEM) and Scanning Electron Microscopy (SEM) (HITACHI S-4300 SE/N-VP Emission Scanning Electron Microscope) were also employed to investigate the dispersion of metal nanoparticles. 2.3. Thermal desorption measurements The CO2 retention capacity (CRC) of NaMt and NaMT-MNPs adsorbents was assessed by thermal programmed desorption of carbon dioxide (CO2 -TPD) according to a procedure fully described elsewhere [16]. Measurements were achieved under dry nitrogen stream (15 mLmin−1 ) using a TPD tubular oven coupled to a CO2 -detector (Li-840A CO2 /H2 O Gas Analyzer). Prior to TPD inves-

Fig. 1. IR spectra of 1: bentonite and 2: NaMt.

tigations, special care was taken in order to avoid affecting the CRC value by forced convection under strong nitrogen stream [17]. 3. Results and discussion 3.1. Properties of purified montmorillonite We report, herein, the effect of the purification procedure of bentonite into montmorillonite for enhancing the surface properties. The purification method using NaCl produced a visible reduction of potassium, calcium and sulfur, indicating that interlayer cations were successfully exchanged. This was demonstrated by XRF analysis, as supported in (Table 1). A confirmation of the clay structure was established by IR spectra, which showed the presence of all bands characterizing montmorillonite. Indeed, the sharp bands at 3628 and 1618 cm−1 were attributed to the bending and stretching modes of adsorbed water (Fig. 1). The bands observed around 1200 cm−1 was associated to the stretching vibration of Si–O and Si–O–Si for NaMt [18,19]. Additionally, the X-ray diffraction spectra revealed that the dry purified clay sample contains mainly montmorillonite (Fig. 2). These results were in good agreement with previous works [20,21]. 3.2. Properties of modified montmorillonite Incorporation of the metallic nanoparticles produced an increase in the interlayer basal spacing (d001 ) from 0.96 nm to 1.30 nm as supported by XRD analysis. This increase of 0.34 nm suggests the formation of either subnanoparticles or bulkier but flattened metal particles in laying position on the clay mineral surface. Sharp 001 XRD lines provide clear evidence of the parallel arrangement of the clay sheets and uniform insertion of MNPs (Fig. 3). Therefore, it can be concluded that metal-loaded NaMt presents an ordered lamellar structure consisting of a face-to-face arrangement with parallel clay sheets. MNPs dispersion was found to improve the microporosity (pore size ≤ 30 Å). This was accompanied by a decrease in the specific surface area from 64 m2 .g−1 for NaMt to 20.15 m2 .g−1 for NaMt-Cu. The pore volume was almost doubled and thorough changes in the pore size distribution were also noticed (Table 2). Here, this result indicates sufficient clay exfoliation and MNPs dispersion, producing an expansion of the clay sheet stacks and favoring the accessibility to the interlayer space between terminal hydroxyl groups. The nitrogen adsorption–desorption isotherms on NaMt, NaMtAg, NaMt-Pd and NaMt-Cu (Fig. 4) revealed typical type III patterns,

316


Table 1 XRF analysis of NaMt, NaMt-Pd, NaMt-Cu, NaMt-Fe and NaMt-Ag samples. Sample/components

NaMt

NaMt-Pd

NaMt-Cu

NaMt-Fe

NaMt-Ag

O Na Mg Al Si S K Ca Ti Mn Fe Zn Sr C Cu Pd Ag

48.94% 1.77% 1.43% 12.37% 31.17% 287 PPM 0.12% 0.64% 965 PPM – 2.75% – – – – – –

49.26% 4.10% 1.37% 11.89% 29.69% 255 PPM 0.11% 0.55% 848 PPM – 2.24% – – 0.47% 100 PPM 0.21% –

48.84% 4.66% 1.33% 11.69% 28.76% – 902 PPM 0.49% 876 PPM – 2.52% – – 0.60% 0.94% – –

49.40% 4.39% 1.34% 11.72% 28.77% – 0.20% 0.51% 870 PPM – 2.93% – – 0.64% 1.20% – –

49.44% 4.55% 1.33% 11.77% 28.79% 101PPM 0.15% 0.48% 866PPM 2.50% – – 0.62% 100PPM – 0.25%

Fig. 2. XRD patterns of 1: bentonite and 2: NaMt.

Table 2 Textural properties of the prepared adsorbents. Specific surface area (BET, m2 g−1 )

Sample

a

NaMt NaMt-Pd NaMt-Cu NaMt-Fe NaMt-Ag

64.00 22.85 20.15 24.60 42.36

a b

b

Pore width (Å)

8.90 14.70 36.20 17.70 13.90

Specific surface area as calculated by the BET method. Pore width determined by applying the BJH method.

indicating the formation of adsorbate multilayers and weak adsorbate-adsorbent interactions [22,23]. The different shapes of the adsorption–desorption isotherms must probably be due to the composition and size distribution of the metallic nanoparticles. 3.3. Interaction between metallic nanoparticles and NaMt Comparison of the IR spectra for all adsorbents is illustrated by the intensity depletion for the OH vibration band at 3445 cm−1 (Fig. 5). In this region, MNPs can interact with terminal hydroxyl groups belonging to NaMt, which, in turn, can also interact with water molecules adsorbed on both the MNPs and NaMt surfaces

[25]. In addition, the intensity of all primary bands associated to the functional groups shifted to low wavenumbers. Here, this shift is a key factor, which indicated the success of MNPs dispersion by hydroxyl groups. The latter may be either directly or indirectly involved in the metal stabilization through interactions with their oxygen atoms [26]. The strong band observed around 1000 cm−1 was attributed to the Si O stretching. Furthermore, the band at 987 cm−1 showed not only a little shift but also an intensity increase band indicates the formation of bridging associations between MNPs and Si O Si groups. The bands below 800 cm−1 can be attributed to Si-OH. A wide band centered at 770 cm−1 was assigned to such M O associations, whereas the observed vibration in lower spectral region at 700 cm−1 was assigned to the M O stretching [27–29]. In summary, all IR data confirmed the changes in the interlayer distances of montmorillonite by MNPs dispersion. A visible change in the space between clay layers was registered for NaMt-Cu, NaMt-Fe and NaMt-Pd as supported by the increases of all (001) line completely dehydrated structure [24]. As previously stated, the different interlayer distances of the modified NaMt structure indicates a strong dependence on the nature of MNPs dispersed and corresponding moisture capacity.


317

Fig. 3. XRD patterns of 1: NaMt, 2: NaMt-Fe, 3: NaMt-Pd and 4: NaMt-Cu.

Fig. 4. Nitrogen adsorption-desorption isotherms at 77 K for NaMt, NaMt-Cu, NaMt-Pd and NaMt-Ag.

3.4. Dispersion and immobilization of MNPs Observations by SEM (Fig. 6) revealed that incorporation of MNPs results in dramatic change on the morphology to spindle shape, as well as change in the fine structure clay sheetclusters. As a general tendency, microporosity significantly increased for each adsorbent after MNPs incorporation (Table 2). This increase

can be explained by a structure expansion due to the presence of MNPs strongly immobilized between terminal functional groups, i.e. through strong OH:MNPs interactions. TEM observations clearly show MNPs dispersion on all support surfaces. Metallic nanoparticles appear as dark and flattened stains with a uniform distribution (Fig. 7). The low number of MNPs near

318


Fig. 5. IR spectra of 1: NaMt, 2: NaMt-Pd, 3: NaMt-Cu, and 4: NaMt-Fe.

Fig. 6. SEM images of 1: NaMt-Cu and 2: NaMt-Pd.

the edge of the montmorillonite sheets demonstrates their incorporation within the interlayer spaces. Comparison of the X-ray diffraction patterns of NaMt and NaMtMNPs showed that all characteristic peaks (10◦ < 2 theta < 50◦ ) of the clay are present. The interlamellar distances between the clay sheets as estimated by TEM are of 0.71 and 0.73 nm, respectively.These values were in good agreement with those assessed by XRD analysis. These different d001 values correspond to changes in the interlayer distances, as previously indicated in the literature [30]. Another confirmation was provided by the depletion and

shift of all bands observed by IR analysis, due to the incorporation of MNPs.

3.5. CO2 adsorption CO2 -TPD measurements provided visible changes in NaMt behavior before and after MNP incorporation (Figs. 8 and 9). All TPD patterns showed similar shape in the temperature range 20–80 ◦ C. NaMt showed low CO2 adsorption capacity (CRC) as compared to


319

Fig. 7. Dispersion of MNPs within the interlayer space and on the outer surface of NaMt.

Fig. 8. Different TPD cycles of desorbed CO2 for NaMt-MNPs between 20 and 400 ◦ C after dynamic impregnation of CO2 at 20 ◦ C in 15 mL min−1 dry nitrogen stream, followed by a purge and TPD at 400 ◦ C (200 mL CO2 ) for 1: NaMt,2: NaMt-Fe, 3: NaMt-Pd, 4: NaMt-Cu, and 5: NaMt-Ag.

320


Fig. 9. Different TPD cycles of desorbed H2 O for NaMt-MNPs between 20 and 400 ◦ C after dynamic impregnation of CO2 at 20 ◦ C in 15 mL min−1 dry nitrogen stream, followed by a purge and TPD at 400 ◦ C (200 mL CO2 ) for 1: NaMt, 2: NaMt-Ag, 3: NaMt-Cu, 4: NaMt-Pd and 5: NaMt-Fe. Table 3 WRC and CRC (␮mol g−1 ) between 20 and 400 ◦ C in 15 mLmin−1 dry nitrogen stream of the different NaMt-MNP samples. VCO2 (mL)

CRC (␮mol g−1 ) NaMt

1.5 50 200 VCO2 (mL)

1.5 50 200

NaMt-Pd

NaMt-Fe

NaMt-Ag

123.79 978.29 178.55 315.09 326.55 469.78 WRC (␮mol g−1 )

NaMt-Cu

261.00 255.23 457.00

191.83 363.23 591.02

186.60 352.91 602.13

NaMt

NaMt-Cu

NaMt-Pd

NaMt-Fe

NaMt-Ag

4.29 3.45 4.26

5.54 4.06 2.33

3.52 3.23 3.54

4.57 3.10 2.69

2.54 2.70 4.46

its MNPs-loaded counterparts, presumably due to the its lower porosity (Table 2). Interestingly, MNPs incorporation induced an improvement of the affinity towards CO2 , as supported by the CRC increases for all the amounts of impregnating CO2 employed for each adsorbent sample (Table 3). The desorption peak appeared around 80 ◦ C indicates that MNPs insertion within the NaMt layers generates additional basicity but of weaker strength. The most probable explanation in this regard resides rather in acidity attenuation as a result of depletion of the number of available terminal –OH groups. No significant changes in the hydrophilic character were observed after MNPs incorporation. The slight increase of the WRC for NaMtMNPs must be rather due to a release of residual moisture which persists even upon drying. This release can be explained in terms of an attenuation of water retention strength through H-bridges with the lattice oxygen of the silica surface, most likely due to a competitive effect of MNPs, which should act as oxygen electron pair scavengers.

Unlike in OH-enriched organoclays, here the presence of water was found to reduce the amount of captured CO2 by the different adsorbents. This is presumably due to a competitive water association with the available silanol and aluminol groups and with lattice oxygen atoms. On dry media, MNPs dispersion in NaMt makes more efficient CO2 adsorption, most likely due to a strong reduction of acidic out-of-plane silanol groups quantity through interactions with MNPs. In other words, interaction between MNPs and OH groups are expected to attenuate the surface acidity, increasing the CO2 density on the solid surface [31,32].

3.6. Competitive adsorption between H2 O and CO2 Unlike what might be expected, it appears that water adsorption is enhanced by the presence of MNPs. In this regard, the NaMt-MNP samples were studied in consecutive adsorption-desorption cycles with several amounts of injected CO2 (Table 3). The experiment results indicated a visible reduction on both WRC and CRC. Here, this behavior was in good agreement with previous data, which provided an explanation about the presence of water on NaMt that affects the CO2 capture by reducing the porosity [17]. In other words, adsorption of water molecules within the interlayer space hinders CO2 diffusion between the clay sheets. Additional attempts in this regard were achieved by previous saturation of all adsorbents by dry and O2 -free hydrogen overnight (Fig. 10). As result, TPD-measurements for these materials led to further increases in CRC. This must be due to the appearance of competitive Metal-H2 interaction and the formation of a thin outer layer of metal hydride that weakens metal-:OH-interaction. This is expected to release more OH groups, thereby enhancing the affinity towards CO2 . The reverse phenomenon is expected to occur with O2 , which should affect the CO2 adsorption by competitive adsorption [33–35].


321

Fig. 10. Different TPD cycles of desorbed CO2 (a) and water (b) for 1: NaMt, 2: NaMt (saturated by H2 ), 3: NaMt-Pd, 4: NaMt-Pd (saturated byH2 ), between 20 and 400 ◦ C after dynamic impregnation of 50 mL CO2 at 20 ◦ C in 15 mL min−1 dry nitrogen stream, followed by a purge and TPD at 400 ◦ C. Table 4 WRC and CRC (␮mol g−1 ) before and after saturation hydrogen, between 20 and 400 ◦ C in 15 mL min−1 dry nitrogen stream of NaMt, NaMt-Cu, NaMt-Fe, NaMt-Ag, and NaMt-Pd. Sample

CRC

WRC

NaMt NaMt-saturated by H2 NaMt-Pd NaMt-Pd-saturated by H2 NaMt-Fe NaMt-Fe-saturated by H2 NaMt-Fe NaMt-Fe-saturated by H2 NaMt-Cu NaMt-Cu-saturated by H2

195.90 376.36 248.49 656.55 363.23 431.90 363.23 521.02 315.09 470.50

3.35 5.06 3.40 3.34 3.10 9.37 3.10 3.92 4.06 4.66

dioxide. The appreciable amounts of metal particles dispersed was attributed to an attenuation of the acid character of the NaMt surface via MNPs with both the lattice oxygen atoms and OH groups. This effect promotes affinity towards CO2 and found to produce a slight structure compaction, due to the strong MNP interaction with the oxygen atoms of the clay sheet surface. The use of metallic nanoparticles for designing such materials is a novelty and the above properties make NaMt-MNPs, obtained by a simple, lowcost and eco-frendly procedure to find applications not only in the adsorption of CO2 but also hydrogen storage. Acknowledgement This work was supported by grants from FODAR-UQ 2015 (QC, Canada) to A.A. and R.R.

On the other hand, H2 saturation produced a visible reduction of the WRC, indicating that H2 O was in competitive adsorption with CO2 . This effect was slightly attenuated in the coexistence of H2 O, H2 and CO2 , due to the reduced quantity of the free OH groups. Therefore, water and CO2 come into competitive interactions with OH groups, and it is clear that H2 saturation increases the CRC and decreases the WRC of all materials (Table 4). 4. Conclusions This work clearly demonstrates that interactions between MNPs and OH groups of a clay surface induce changes in the structural and textural properties, generating new stable materials. Incorporation of metallic nanoparticles between layers and on the surface of montmorillonite turns out to be an interesting route to prepare potential materials with improved affinity towards carbon

References [1] T. Takeshita, Appl. Energy 86 (2009) 222–232. [2] F. Akhtar, L. Andersson, N. Keshavarzi, L. Bergstrom, Appl. Energy 97 (2012) 289–296. [3] H. Li, J.P. Jakobsen, O. Wilhelmsen, J. Yan, Appl. Energy 88 (2011) 3567–3579. [4] S. Sengupta, V. Amte, R. Dongara, A.K. Das, H. Bhunia, P.K. Bajpai, Energy Fuels 29 (2015) 287–297. [5] J. Liu, P.K. Thallapally, B.P. McGrail, D.R. Brown, J. Liu, Progress in adsorption-based CO2 capture by metal-organic frameworks, Chem. Soc. Rev. 41 (2012) 2308–2322. [6] X. Xu, C. Song, J.M. Andresen, B.G. Miller, A.W. Scaroni, Int. J. Environ. Technol. Manage. 4 (2004) 32–52. [7] E. Groppo, W. Liu, O. Zavorotynska, G. Agostini, G. Spoto, S. Bordiga, C. Lamberti, A. Zecchina, Chem. Mater. 22 (2010) 2297–2308. [8] A. Azzouz, S. Nousir, N. Bouazizi, R. Roy, ChemSusChem 8 (2015) 800–803. [9] D.J.K. Ross, R.M. Bustin, Can. Petrol. Geol. 55 (2007) 51–75. [10] L. Aylmore, I. Sills, J. Quirk, Clay. Clay. Miner. 18 (1970) 91–96. [11] L. Michot, F. Villieras, Dev. Clay Sci. Elsevier Amsterdam 965 (2006) 972.

322 [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23]

N. Bouazizi et al. / Applied Surface Science 402 (2017) 314–322 L. Heller-kallai, Dev. Clay Sci. Elsevier Amsterdam 289 (2006) 296. C. Volzone, J. Ortiga, Process. Saf. Environ. 82 (2004) 170–174. I. Suzuki, H. Kiuchi, K. Saitou, J. Catal. 155 (1995) 163–165. A. Azzouz, N. Platon, S. Nousir, K. Ghomari, D. Nistor, T.C. Shiao, R. Roy, Sep. Purif. Technol 108 (2013) 181–188. A. Azzouz, D. Nistor, D. Miron, A.V. Ursu, T. Sajin, F. Monette, P. Niquette, R. Hausler, Thermochim. Acta 449 (2006) 27–34. S. Nousir, N. Platon, K. Ghomari, A.S. Sergentu, T.C. Shiao, G. Hersant, J.Y. Bergeron, R. Roy, A. Azzouz, J. Colloid Interf. Sci. 10 (2013) 03–050. Y.Y. Choi, E.H. Lee, S.H. Ryu, Polym. Bull. 63 (2009) 47–55. A.M.F. Guimarães, V.S.T. Ciminelli, W.L. Vasconcelos, Appl. Clay Sci. 42 (2009) 410–414. L. Ukrainczyk, K.A. Smith, Environ. Sci. Technol. 30 (1996) 3167–3176. Y. Zheng, A. Zaoui, I. Shahrour, Am. Mineral. 95 (2010) 1493–1499. A.M. Shanmugharaj, K.Y. Rhee, S.H. Ryu, J. Colloid Interf. Sci. 298 (2006) 854–859. M. Khalfaoui, S. Knani, M.A. Hachicha, A.B. Lamine, J Colloid Interf. Sci 263 (2003) 350–356.

[24] Y. Peng, L. Hongmei, L. Dong, T. Daoyong, Y. Wenchang, H. Hongping, Appl. Clay Sci. 75 (2013) 82–91. [25] D.S. Tong, H.S. Xia, C.H. Zhou, Chin. J. Catal. 30 (2009) 1170–1187. [26] G. Cao, Nanostructures and Nanomaterials, Imperial College Press, 2004. [27] M. Del Arco, R. Trujillano, V. Rives, J. Mater. Chem. 8 (1998) 761. [28] E. Uzunova, D. Klissurski, I. Mitov, P. Stefanov, Chem. Mater. 5 (1993) 576. [29] H.C.B. Hansen, C.B. Koch, R.M. Taylor, J. Solid State Chem. 113 (1994) 46. [30] S. Park, B.J. Kim, D. Seo, K.Y. Rhee, Y.Y. Lyu, Mater. Sci. Eng. A 526 (2009) 74–78. [31] C.M. Tenney, C.M. Lastoskie, Environ. Prog. 25 (2006) 343–354. [32] X.C. Lu, F.C. Li, A.T. Watson, Fuel 74 (1995) 599–603. [33] D.J.K. Ross, R.M. Bustin, Mar. Petrol. Geol. 26 (2009) 916–927. [34] J.R. Li, R.J. Kuppler, H.C. Zhour, Chem. Soc. Rev. 38 (2009) 1477–1504. [35] G. Li, P. Xiao, P.A. Webley, J. Zhang, R. Singh, Energy Procedia 1 (2009) 1123–1130.