Langmuir 2008, 24, 14215-14221
14215
Aminopropyl-Modified Magnesium-Phyllosilicates: Layered Solids with Tailored Interlayer Access and Reactivity Ricardo B. Ferreira, Ce´sar R. da Silva, and Heloise O. Pastore* Instituto de Quı´mica, CP 6154, UniVersidade Estadual de Campinas, UNICAMP, 13083-970 Campinas, SP, Brasil ReceiVed July 7, 2008. ReVised Manuscript ReceiVed October 2, 2008 Despite its wide application, the synthesis of aminopropyl-modified magnesium-phyllosilicates was known only in the case where every silicon atom bore an organic pending group. This paper shows the preparation of aminopropylmodified talc where tailored amounts of silicon atoms are bound to an aminopropyl group. The decrease in the concentration of the organoamino group leaves a proportional concentration of interlayer SiOH groups that can be used to react with other silylation agents. The amino group reacts with CO2, forming a carbamate functionality; it seems that the presence of this group avoids delamination in water as performed for the parent compound. Bearing in mind that the aminopropyl group can be changed by other groups, the present synthesis strategy demonstrates ways to produce solids with controlled surface properties with interlayer amino and SiOH groups in variable concentrations, allowing formation of several other interlayer functionalities.
Introduction Layered silicates have found ingenious applications in the past decade. If their initial appeal was the ion-exchange capacity and possibility of intercalation of organic species, studies in the area have broadened quite a lot in fields where they can be of use. Smectites, in particular, are good examples of the attractive characteristics of layered inorganic solids. They display interesting adsorptive and ion-exchange properties, thermal and chemical stability, swellability in polar solvents, and ability to orient to form films. Use of natural smectites is however strongly hindered by the presence of impurities. To overcome this problem synthetic smectites have been prepared by hydrothermal methods that had long duration. Although direct synthesis of covalently linked organoclays is highly desirable because the products should have wide applicability, some problems have been associated with the preparation of organic smectites. The first one is the synthesis conditions that could prevent the use of organic modifiers needed to keep the swellability in low-polarity solvents due to their duration in time.1 Besides these inconveniences, Fukushima and Tani2 also commented that a silicon trialkoxide containing an organic chain bound to a silicon atom would not be able to generate a phyllosilicate material due to an inadequacy of bonding sites around the Si atom.2 The authors however prepared what they called a polymer with a smectite-type structure using 3-methacryloxypropyltrimethoxysilane as the silicon source. In this case they proposed that Si-C bonds rendered formation of SiO4 tetrahedral sheets very difficult. The polymer had well-ordered octahedral sheets and slightly disordered -C-SiO3 tetrahedral sheets. They concluded that formation of such a smectite-like polymer was mainly dominated by formation and growth of the octahedral sheet; formation of the tetrahedral sheets was not essential for synthesis of the phyllosilicate. The suitable arrangement of the organic groups caused by hydrophobic * To whom correspondence should be addressed. E-mail: gpmmm@ iqm.unicamp.br. (1) Ukrainczyk, L.; Bellman, R. A.; Anderson, A. B. J. Phys. Chem. B 1997, 101, 531–539. (2) Fukushima, Y.; Tani, M. J. Chem. Soc., Chem. Commun. 1995, 241–242.
interactions might also promote formation of a two-dimensional arrangement of the lamella.2 Carrado3 proposed a similar set of interactions as the driving force for formation of hectorite (a 2:1 trioctahedral phyllosilicate). Burkett et al.4 elaborated a little more on Fukushima’s and Carrado’s arguments and proposed that the limited coordination of the organosiloxane units, the hydrophobicity of the organic moieties, and microphase segregation all facilitated formation of the lamellar structure. Many other studies reported on the changes of the organic groups bound to the Si atom of the trialkoxyalkylsilane reactant and a few others on chemical modification of the organic groups in order to make them more prone to reactions in the interlayer space. A representative and elegant work was reported by Lagadic5 on formation of the Schiff base by reaction of an aminofunctionalized magnesium-layered phyllosilicate with aldehydes and a ketone. From these organomodified magnesium phyllosilicates complexes with copper, nickel, and cobalt were prepared. It was observed that formation of the Schiff base or complexes did not disturb the layered arrangement of the solid. These organic-inorganic hybrids can be used in heterogeneous catalysis as such or after complexation with transition metals.6 One specific organomodified magnesium phyllosilicate that has been widely used either as a model or in real applications is the aminopropyl-functionalized magnesium phyllosilicates. Mann and co-workers7,8 have shown that aminopropyl-functionalized magnesium phyllosilicates could be exfoliated by a simple ultrasound treatment in water since the RNH2 groups were protonated in these conditions and created a number of sites for ion exchange within the interlayer space that, upon exfoliation, became essentially surface groups on the lamella. The common feature of all these works is that the organomodified magnesium phyllosilicate has an organic propylamine (3) Carrado, K. A. Ind. Eng. Chem. Res. 1992, 31, 1654–1659. (4) Burkett, S. L.; Press, A.; Mann, S. Chem. Mater. 1997, 9, 1071–1073. (5) Lagadic, I. L. Microporous Mesoporous Mater. 2006, 95, 226–233. (6) Sharma, S. K.; Patel, H. A.; Jasra, R. V. J. Mol. Catal. A: Chem. 2008, 280, 61–67. (7) Patil, A. J.; Muthusamy, E.; Mann, S. Angew. Chem., Int. Ed. 2004, 43, 4928–4933. (8) Patil, A. J.; Muthusamy, E.; Mann, S. J. Mater. Chem. 2005, 15, 3838– 3843.
10.1021/la802142s CCC: $40.75 2008 American Chemical Society Published on Web 11/14/2008
14216 Langmuir, Vol. 24, No. 24, 2008
group pending on every silicon atom in each tetrahedron layer. When this material is exfoliated in water the amino groups are protonated and thus the density of bonding sites per silicon atom is the highest possible. When dealing with electrostatic interactions only this can be the ideal situation.7-9 This paper aims at describing the preparation of an aminopropyl-modified magnesium phyllosilicate where a controlled amount of aminopropyl groups is bound to the surface silicon atoms. Characterization of this material showed not only that the talc-like structure was obtained but also that in diminishing the concentration of organic groups Si-OH groups began to populate the surface of the lamella and that these are available for further silylation. The products and synthesis strategies here described demonstrate ways to create solids with controlled surface properties, in this case with interlayer amino groups in variable concentrations, and SiOH, amenable for grafting other functionalities.
Experimental Section Materials. The reagents were used as received: 3-aminopropyltriethoxysilane (AMPTS, Acros), tetraethoxyorthosilane (TEOS, Acros), magnesium nitrate hexahydrate (Ecibra), sodium hydroxide (Merck), n-octadecyltrimethoxysilane (Fluka), absolute ethyl alcohol (J. T. Baker), toluene (Vetec), 2-propanol (J. T. Baker), and distilled water. Aminopropyl-Functionalized Magnesium Phyllosilicate. Magnesium nitrate hexahydrate (0.012 mol) was first dissolved in 100 mL of distilled water, and the solution was stirred magnetically at room temperature. Then, the silicon sources (AMPTS and TEOS) were added according the proportions shown in the Table 1, maintaining the Si/Mg molar ratio at 4/3, which is the molar ratio found in the natural talc. This mixture formed a white suspension. In this mixture, 48 mL of 0.5 mol · L-1 aqueous sodium hydroxide solution was added dropwise under magnetic stirring. The resultant suspension was aged for 4 h at 50 °C. Then, the suspension was submitted to hydrothermal treatment for 48 h at 100 °C. The products were centrifuged, washed three times with ca. 45 mL of ethyl alcohol, and dried at room temperature in desiccators over silica. The white solid formed was ground and sieved at 0.106 mm to produce a powdered material. Reaction of n-Propylamine-Modified Phyllosilicate with n-Octadecyltrimethoxysilane. S100 and S025 samples were dried under vacuum for 4 h at 80 °C. Then, 10 mL of dry toluene was added with magnetic stirring under argon flow. To this suspension 2 mL of n-octadecyltrimethoxysilane was added. The system was maintained under argon at 70 °C for 48 h. The resulting product was centrifuged and washed with dry toluene, 2-propanol, and finally ethanol. The samples produced were called S100C and S025C. Delamination Procedure.10 Approximately 200 mg of the samples S100, S067, S050, and S025 was dispersed in 10 mL of distilled water followed by sonication for 5 min. The resultant mixtures were centrifuged for 30 min at 120 rpm and dried at room temperature in desiccators over silica. The materials were then ground and sieved at 0.106 mm. X-ray diffraction analyses were performed with known masses of each sample before and after the delamination procedure; the same sample holder was used for the samples of each AMPTS proportion. Characterization. X-ray diffraction patterns were obtained with Cu KR radiation (40 kV, 30 mA) on a Shimadzu model XRD 7000 diffractometer at room temperature using 0.5°, 0.5°, and 0.3 mm slits for entrance, scattering, and exit. Infrared spectra were obtained on a Nicolet model 6700 FTIR spectrophotometer using 0.05% KBr pressed samples. Thirty-two scans at 4 cm-1 resolution were accumulated. Nuclear magnetic resonance spectra of the solid materials were obtained on a Bruker Avance II+ 400 at room temperature. The (9) Fujii, K.; Hayashi, S. Appl. Clay Sci. 2005, 29, 235–248. (10) Lebeau, B.; Brendle, J.; Marichal, C.; Patil, A. J.; Muthusamy, E.; Mann, S. J. Nanosci. Nanotechnol. 2006, 6, 325–359.
Ferreira et al. measurements were made at a resonance frequency of 79.5 MHz for 29Si and 100.6 MHz for 13C. For the 29Si spectra the HPDEC technique with a pulse repetition time of 60 s and a pulse angle of 90° was employed. For the 13C spectra the CPMAS technique was used with a pulse repetition time of 3 s and a contact time of 0.003 s. Thermal analyses were performed using a TA Instruments model 2950. The samples were analyzed under argon and air flow from room temperature to 990 °C at a heating rate of 10 °C · min-1. N2 adsorption experiments were performed at the liquid nitrogen temperature on a Quantachrome Instruments model Autosorb-1MP. The samples were treated at 120 °C for 12 h prior to measurements.
Results and Discussion Synthesis and Structure of Materials. Hydrothermal treatment was introduced in the synthesis procedures described here because it had already been shown in the literature that at room temperature layered magnesium silicate structures were not obtained in the absence of hydrophobic organosilanes, the situation found when triethoxysilane (SiH(OC2H5)3) or tetraethoxysilane (Si(OC2H5)4) was used.4 As a complete range of (AMPTS)100-X:(TEOS)X should be used in the synthesis, it was necessary to guarantee that use of pure TEOS in the synthesis would afford a solid for comparison. Another point that should be highlighted is that, contrary to the majority of the studies in the synthesis of the organomodified magnesium silicate, magnesium salt was dissolved in water instead of methanol or ethanol. The mixture of the two silicon alkoxides was added to the aqueous solution in the proportion displayed in Table 1. Figure 1 shows the powder X-ray diffractograms of the samples prepared in this work. The sample prepared only AMPTS as silicon source; Figure 1a shows signals at 5.3° and 10.4° 2θ corresponding to the 001 and 002 reflections, narrower than the ones at ca. 20°, 35°, and 60° 2θ assigned to (020,110), (130,200), and (060,330) diffractions.11 It is the presence of these last reflections that ensure that trioctahedral clay had been prepared, even if the signals are large. Curve b in Figure 1 displays essentially the same signals as discussed for curve a. The only difference is a significant decrease in the 002 diffraction at approximately 10° 2θ. The effect of the decrease in the amount of AMPTS is a reflection of the diminished organization in the material brought about by the smaller hydrophobic effect. When 50% of AMPTS is substituted by TEOS and from there on toward lower values of AMPTS concentration the 002 signal completely disappears and the rest of the signals become progressively weaker. The (060,330) reflection however is present in the diffractograms, confirming the structure of the materials. Pending Organic Groups. Concentration. Figure 2 displays the thermogravimetry and its derivative measured under air. The profiles are composed of mass losses in two regions: below 200 °C corresponding to water loss and above 200 °C due to organics decomposition/combustion under air. The decrease in the amount of AMPTS in the synthesis gel causes a decrease of mass loss in the second range of temperatures. The percentage mass loss under air of the organic part (% organics) and the concentration of pending groups per gram of solid material (CAMP) are in Table 2. Elemental analysis of carbon and nitrogen allowed calculation of the C/N molar ratio, also shown in Table 2. The values of the C/N molar ratio are always higher than 3; this aspect is going to be further discussed below. Nature of Bonding and Control of the Access to the Interlayer Spaces. The tetrahedra of Si atoms in these materials are of two main types: silicon atoms with one Si-C bond and three Si-O bonds and silicon atoms with no Si-C bonds, only Si-O ones.
Aminopropyl-Modified Magnesium-Phyllosilicates
Langmuir, Vol. 24, No. 24, 2008 14217
Figure 1. Powder X-ray diffraction of (a) S100, (b) S067, (c) S050, (d) S025, and (e) S000. Table 1. Molar Proportions and Volumes of AMPTS and TEOS Used in the Synthesisa sample
molar proportion/%
AMPTS (mL)
TEOS (mL)
S100 S067 S050 S025 S000
100 AMPTS 67 AMPTS-33 TEOS 50 AMPTS-50 TEOS 25 AMPTS-75 TEOS 100 TEOS
3.76 2.51 1.88 0.94 0
0 1.18 1.78 2.68 3.57
a
The figure in the code of the sample is the percentage of AMPTS.
The first group is represented by Si atoms of Tn type, and their chemical shifts in 29Si-MAS-RMN appear in the region between -40 and -75 ppm (in relation to TMS). They can be mainly of three types: RSi(OH)2(OMg), RSi(OSi)(OH)(OMg), and RSi(OSi)2(OMg), respectively T1, T2 and T3.12 The other group of silicon atoms is comprised of four types of tetrahedra: MgOSi(OH)3, MgOSi(OSi)(OH)2, MgOSi(OSi)2 (OH), and MgOSi(OSi)3, which are known as Q0, Q1, Q2, and Q3, respectively, appearing between -75 and -120 ppm from TMS. Figure 3 presents the 29Si-MAS NMR of the samples prepared in this work. Curve a, corresponding to the sample where all silicon atoms are of the T type, has two peaks at -50 and -54 ppm, assigned to T1 and T2 silicon atoms, respectively.12 The progressive substitution of AMPTS by TEOS causes the appearance of a group of signals at -78, -86, and -92 ppm corresponding to the Q0, Q1, and Q2 types of silicon groups, respectively. As the amount of TEOS increases in the gel the overall intensity of these signals increases as well until they are the only ones in the spectrum as it observed in Figure 3e. Integration of signals in the T and Q regions gives the ratio between silicon atoms that are bound to an aminopropyl group and those that are not bound to an organic group (SiT/SiQ). These values are given in Table 3. It is easily confirmed that the proportion of silanes used in the gel is nicely reproduced in the solid sample, confirming the efficiency of this simple synthesis method. Access to the interlamelar space is also modified by variation of the concentration of pending organic groups. Figure 4 displays (11) Kumarazwamy, G.; Deshmukh, Y.; Agrawal, V. K.; Rajmohanan, P. J. Phys. Chem. B 2005, 109, 16034–16039. (12) Whilton, N. T.; Burkett, S. L.; Mann, S. J. Mater. Chem. 1998, 8, 1927– 1932. (13) Brunauer, S.; Emmett, P. H.; Teller, E. J. Am. Chem. Soc. 1938, 60, 309–319.
Figure 2. Thermogravimetric curves and their derivatives in air of the materials: (a) S000, (b) S025, (c) S050, (d) S067, and (e) S000. Table 2. Results of the Thermogravimetry under Air and Elemental Analyses sample
% organics
S100 S067 S050 S025 S000
31.3 29.5 26.6 20.8 a
CAMP (mmol · g-1)
%Si (pending groups/ Si(total)
C/N
5.39 5.07 4.59 3.58
86 76 65 43
3.24 3.19 3.06 3.64
a No AMPTS was used in the synthesis of S000; however, the mass loss was found in this region on the order of 10.6%.
the isotherms of N2 physisorption at 77 K; values of the specific surface area using the BET model13 and total volume of pores are given in Table 4. The isotherm corresponding to the solid with 100% AMPTS is of type II according to IUPAC classification, which is characteristic of nonporous materials.14 The other materials have a type IV isotherm characteristic of mesoporous solids with a type H2 hysteresis, typical of disordered materials with a pore size distribution poorly defined.14 Ink bottle pores do not seem to be present in these materials by the way in which the hysteresis loop closes. As shown in Table 4 use of a smaller amount of AMPTS in the synthesis of the magnesium phyllosilicate causes an increase in the surface area and total pore volume, better envisaged in Figure 5. These effects of organics reduction were in fact the type of properties the present studies were aiming at producing. ReactiVity of Pending Groups. The infrared spectra with Fourier transform of the samples prepared in this work are shown in Figure 6. Bands typical of magnesium phyllosilicates are observed in the region from 1186 to 1000 cm-1 with the most intense one
14218 Langmuir, Vol. 24, No. 24, 2008
Ferreira et al.
Figure 5. Surface area (0) and total pore volume (b) as a function of AMPTS concentration in the synthesis.
Figure 3. 29Si solid-state HPDEC nuclear magnetic resonance spectra from the materials: (a) S100, (b) S067, (c) S050, (d) S025, and (e) S000. Table 3. Results of the Integration of Signals in the 29 Si-MAS-NMR area of signals (109) sample
T-type signals
S100 S067 S050 S025 S000
3.48 2.75 3.70 1.45
Q-type signals
SiT/SiQ (%)
1.31 3.61 4.34 4.21
100.0 67.7 50.6 25.1 0.0
Table 4. Specific Surface Area (SBET) and Total Pore Volume (VT) of the Samples sample
SBET (m2 · g-1)
VT (cm3 · g-1)
S100 S067 S050 S025 S000
2.2 160 191 393 395
0.02 0.13 0.13 0.27 0.28
Figure 6. Infrared spectra from the materials synthesized: (a) S100, (b) S067, (c) S050, (d) S025, and (e) S000. Scheme 1
at 1013 cm-1 in sample S100 and at 1021 cm-1 in sample S000. These bands are assigned to Si-O-Si asymmetric stretching, while the bending is found at 457 cm-1 in sample S000 and is
Figure 4. N2 adsorption isotherms of (a) S000, (b) S025, (c) S050, (d) S067, and (e) S100.
progressively displaced to lower wavenumbers as the amount of AMPTS increases in the sample until it is displaced to wavenumbers lower than 400 cm-1 in sample S100. The bands near 550 cm-1 correspond to the stretching of Mg-O in the octahedral layer. Between 3000 and 2750 cm-1 the bands corresponding to symmetric and asymmetric stretchings and deformations of -CH2- groups of the 3-aminopropyl groups pending in the interlayer space are observed. These bands become progressively less intense as the amount of AMPTS diminished in the reaction mixture until their complete absence in sample S000. The same behavior was observed with the bands in the range between 1600 and 1400 cm-1 due to the deformation modes of the organic chain. The decrease in the bands assigned to the organic pending groups is accompanied by the appearance and increase of a band at 900 cm-1 due to the SiO-H groups
Aminopropyl-Modified Magnesium-Phyllosilicates
Figure 7. 13C-MAS NMR of S100 sample showing the presence of the carbamate carboxyl carbon atom at 165 ppm.
that substitute the organic groups in the surface of the tetrahedral silicon layers. A very narrow band appears in all samples with varied intensity at 1381 cm-1 assigned to formation of the ammine carbamate, according to the reaction displayed in Scheme 1, by reaction of CO2 from air. The identity of this anchored compound was confirmed by 13C-CP-MAS NMR. Figure 7 shows the results for sample S100. The peaks labeled 1, 2, and 3 at approximately 16, 27, and 45 ppm correspond, respectively, to the carbon of the methylene group directly bound to the silicon atom, the carbon atom of the central methylene group, and the carbon directly bound to the nitrogen atom. Besides these a new peak appears that is not correlated with the organic chain. The signal at 165 ppm is assigned to the carbonyl atom of the carbamate.15The presence of CO2 bound to the solid is responsible for the increased C/N molar ratios found in the CHN elemental analysis displayed in the last columm of Table 2. In fact, one would expect that the difference between the C/N molar ratio for each sample to the value 3 for propylamine would be very close; however, it goes through a maximum at S025 where the best compromise between the best concentration of sites for anchoring and the highest access to these sites is achieved.
Langmuir, Vol. 24, No. 24, 2008 14219
Delamination of the aminopropyl-modified clay was performed as indicated in the literature10 by sonication in water. The process did not effect changes in the XRD profiles; in fact, they are essentially the same as before the delamination procedure. This is probably the result of having the carbamate in the interlayer space; these groups are not as basic toward water as the free aminopropyl group; creation of protonated amino groups was not observed, and the delamination did not occur. The availability of interlayer SiOH in S025 as a response to the smaller amount of AMPTS used in the synthesis was evaluated by reaction with n-octadecyltrimethoxysilane, ODTMS. The same reaction was performed with S100 as a blank experiment. Figure 8 shows the XRD profiles of samples S025 and S100 before reaction and S025C and S100C after silylation with ODTMS. The general observation is that the reaction does not change the structure of the material, not even the interlayer space is changed by introduction of a larger alkylsilyl group. This should not be a surprise since the C18 alkyl chain is not necessarily straight in the interlayer space. One aspect should be remarked on: the signal corresponding to the (020, 110) diffraction at approximately 20° 2θ, very clear in Figure 8a and 8b, of a more organized material, reappears in sample S025C (compare Figure 8c and 8d), that is, after reaction with ODTMS, indicating a more organized situation after introduction of a C18 chain in the interlayer space. Reaction of free SiOH groups with ODTMS would transform SiOH into Si-O-Si and create new Si-C groups, as shown in Scheme 2. The problem with this reaction is that despite the fact that the number of SiOH groups is known by 29Si-MAS NMR their neighboring conditions cannot be precisely known: one ODTMS molecule may react with one, two, or three surface SiOH produced by hydrolysis of -OCH3 groups, two, one, or zero new SiOH groups. Thus, monitoring the disappearance of SiOH groups will not give information about the extent of the reaction between samples S025 and S100 with ODTMS. On the other hand, reaction of O2(MgSi(OSi)2(OH)) with AMPTS would produce O3(MgOSi(OSi)3) and should therefore cause the increase in intensity in the signals at more negative values of chemical shift. This was seen in the spectrum of Figure 9d in relation to Figure 9c: the ratio Q4(-94.8 ppm)/Q3(-87.9 ppm) increases after reaction of S025 (Figure 9c) with ODTMS to produce S025C (Figure 9d). This effect seems to be accompanied by a slight increase in the signal at -55.4 ppm in S025C (Figure 9d) in comparison to S025 (Figure 9c), indicating a larger amount of Si-C bonds in the solid after reaction with
Figure 8. X-ray diffraction pattern from the materials: (a) S100, (b) S100C, (c) S025, and (d) S025C.
14220 Langmuir, Vol. 24, No. 24, 2008
Ferreira et al.
Figure 9. Decoupled 29Si solid-state nuclear magnetic resonance spectra from the materials: (a) S100, (b) S100C, (c) S025, and (d) S025C. Scheme 2
to the carbon atoms of methylene groups bound to the Si atom (carbon 1 in Figure 10), the central methylene group (carbon 2 in Figure 10), and the methylene group bound to NH2 (carbon 3 in Figure 10), respectively.16 Reaction of S100 with OTDMS as discussed should have occurred only at the free external surface of the modified clay. Thus, besides the signals already observed for the n-propylamine group (marked with asterisks), a methyl group should appear and methylene groups not very different from the ones already existing in the reactant clay. Central chain methylene groups appear as very narrow signals at 24.2, 31.4, and 34.0 ppm. These signals are present in the spectrum of Figure 10c with larger intensity. In this case the methyl group marked 18′ in Figure 10 appears together with carbon 1′ as a large band at lower values of chemical shift than the band at 15.4 ppm.
Conclusions
Figure 10. 13C solid-state CPMAS nuclear magnetic resonance spectra from the materials: (a) S100, (b) S100C, and (c) S025C.
ODTMS. These changes are completely absent when sample S100 (Figure 9a) reacts with ODTMS to produce sample S100C (Figure 9b). Taking into consideration that the surface area of S100 is very low, easy acess of ODTMS to the interlayer space should not be expected. Therefore, only external sites would be modified and would not clearly appear in the NMR experiment. 13C-CP-MAS NMR provides strong evidence of ODTMS reaction with aminopropyl-modified phyllosilicate. Figure 10 displays the 13C-CP-MAS NMR curves of S100 (curve a), S100C (curve b), and S25C (curve c). The three peaks for sample S100 at 15.4, 30.0, and 43.6 ppm (marked with asterisks) are assigned
Simple modifications in a traditional synthesis procedure allowed preparation of aminopropyl-modified phyllosilicates with controlled and variable amounts of organic pending groups in the interlayer space. The changes introduced in the solid profoundly changed the physicochemical characteristics of these clay-type material in relation to the ones of the samples prepared with only AMPTS as the silicon source. The diminished amount of organic pendant groups in the interlayer space increases access to the interlayer sites, surface area, and pores volume. Thus, aminopropyl groups are available and (14) Lowell, S.; Shield, J. E.; Thomas, M. A.; Thommes, M. Characterization of Porous Solids and Powders: Surface Area, Pore Size and Density, 2nd ed.; Particle Technology Series; Springer: Dordrecht, 2006; Chapter 3. (15) Silverstein, R. M.; Webster, F. X.; Kiemle, D. J. Identificac¸a˜o Espectrome´trica de Compostos Orgaˆnicos, 7th ed.; LTC: Rio de Janeiro, 2006; p 236. (16) Breitmaier, E.; Voelter, W. Carbon-13 NMR Spectroscopy, 3rd ed.; VCH: Weinheim, 1987; pp 183-185.
Aminopropyl-Modified Magnesium-Phyllosilicates
relatively accessible in the interlayer space as a lower amount of AMPTS is used in the synthesis of the aminopropyl-modified phyllosilicate. They react with CO2 from the air, producing the aminopropylcarbamate-modified clay that does not delaminate upon treatment with water. In support, interlayer SiOH groups produced by the absence of AMPTS in the synthesis, react with octadecyltrimethoxysilane, and cause the appearance of the long chain methylene groups in the 13C-CP-MAS NMR. The type of modification introduced in the organo-phyllosilicate allows it to act as a nanoreactor since the interlayer space became accessible. Anchoring of molecules can be
Langmuir, Vol. 24, No. 24, 2008 14221
designed according to the number of sites available, introducing larger control over steric and electronic effects on the reactions occurring therein. Acknowledgment. The authors are deeply indebted to the Fundacão de Amparo no Estado de São Paulo (FAPESP) for financial support to this work. FAPESP and Conselho Nacional de Desenvolvimento Cientifico e Tecnológico (CNPq) are acknowledged for fellowships (RBF and HOP, respectively). LA802142S
