4144
Langmuir 2001, 17, 4144-4147
First Direct Visualization of Oriented Mesostructures in Clay Gels by Synchrotron-Based X-ray Fluorescence Microscopy Isabelle Bihannic,† Laurent J. Michot,*,† Bruno S. Lartiges,† Delphine Vantelon,† Jeroˆme Labille,† Fabien Thomas,† Jean Susini,‡ Murielle Salome´,‡ and Barbara Fayard‡ Laboratoire Environnement et Mine´ ralurgie, INPL-ENSG-CNRS UMR 7569, 15 Avenue du Charmois, BP 40, 54501 Vandoeuvre Cedex, France, and European Synchrotron Radiation Facility, BP 220, 38043 Grenoble Cedex, France Received January 29, 2001. In Final Form: March 29, 2001 A direct in-situ visualization of montmorillonite gels at 50 g/L has been obtained for the first time, using silicon fluorescence yield imaging at the ESRF ID21 X-ray microscopy line. An unexpected superstructure formed by alternating clay-rich and clay-poor domains was evidenced with mesoscopic orientational order. The order and size of the oriented entities extends over distances at least 2 orders of magnitude larger than the dimensions of individual clay platelets. Such long-range organization must certainly be considered for assessing the formation mechanisms and properties of gels in systems of charged colloidal platelets.
Introduction Phase transitions in colloidal suspensions have attracted considerable attention in recent years due to the fundamental and applied implications of such systems. In the case of spherical particles, which have been investigated in depth both theoretically and experimentally, the phase diagrams are now fairly well understood for both charged and uncharged spheres. The phase behavior of anisotropic colloids (relevant to most natural systems) is more complex and richer due to possible orientational ordering which can lead to inorganic liquid crystals.1 The case of suspensions of rodlike particles was theoretically described in the 1930s and 1940s by Onsager,2 who, by considering competition between orientational entropy and excluded volume entropy, explained the isotropic/nematic phase transition observed experimentally in various systems.3-5 Onsager’s theory also applies to the case of uncharged platelike colloids. Indeed, recent experiments on monodisperse or slightly polydisperse platelets revealed two clear phase transitions, with the system evolving subsequently from isotropic to nematic and from nematic to columnar with increasing volume fraction.6-8 The situation is much less clear in the case of charged colloidal plates such as natural swelling clays. The structure of these anisotropic minerals (typical aspect ratios between 25 and 1000) consists of two tetrahedral sheets (silica) sandwiching an octahedral sheet (diocta* To whom correspondence should be addressed. E-mail:
[email protected]. † Laboratoire Environnement et Mine ´ ralurgie. ‡ European Synchrotron Radiation Facility. (1) Gabriel, J.-C. P.; Davidson, P. Adv. Mater. 2000, 12, 9. (2) Onsager, L. Ann. N. Y. Acad. Sci. 1949, 51, 627. (3) Langmuir, I. J. Chem. Phys. 1938, 6, 838. (4) Van Bruggen, M. P. B.; Dhont, J. K. G.; Lekkerkerker, H. N. W. Macromolecules 1999, 32, 2256. (5) Bernal, J. D.; Fankuchen, I. J. Gen. Physiol. 1941, 25, 111. (6) Van der Kooij, F. M. Thesis, Utrecht University, The Netherlands, 2000. (7) Van der Kooij, F. M.; Lekkerkerker, H. N. W. J. Phys. Chem. B 1998, 102, 7829. (8) Van der Kooij, F. M.; Kassapidou, E.; Lekkerkerker, H. N. W. Nature 2000, 406, 868.
hedral aluminum hydroxide or trioctahedral magnesium hydroxide). The chemical composition of the sheets includes isomorphic substitutions by less charged cations. This generates a net negative layer charge compensated by interlayer exchangeable cations whose valency and hydration properties control both swelling and colloidal behavior. Clay mineral suspensions do not9 exhibit a clear isotropic/nematic phase transition with phase separation, but instead, at very low volume fractions (as low as 0.5 wt %10) a sol-gel transition turns out to be ubiquitous. The structure of the gel formed is still under debate. Two main models were proposed in the 1950s on the basis of either the formation of a tridimensional network governed by electrostatic attraction between platelets11 or the formation of an oriented network stabilized by repulsive forces caused by interacting double layers.12,13 Regarding the importance of clay-based systems in industry (thickeners, drilling fluids, paints, ...) and in the environment (soil stability, geocycling, ...), understanding the phase behavior in such systems represents obviously a major issue for colloid science. One of the open questions in clay suspensions is related to the presence of long-range order in the gel phase and to its relationship with the gel microstructure. Static light scattering,14,15 ultrasmall- and small-angle X-ray scatter(9) In his 1938 paper, Langmuir observed after several hundred hours a phase separation for his bentonite samples at concentrations between 2.0 and 2.2 wt % that he assigned to an isotropic/nematic phase transition. The chemical formula of the bentonite he used (written as SiO2, 39.2%; Al2O3, 0.11%; Fe2O3, 0.15%; CaO, 15.1%; MgO, 19.2%) reveals that the clay he used must have been a trioctahedral mineral (very low content in aluminum), very likely hectorite (the natural equivalent of Laponite). Such a mineral is classically associated with calcium carbonate impurities, which explains the high content in CaO. In view of this high amount of impurity and considering the known chemical instability of hectorite, the assignment of the observed phase transition could be questionable. (10) Lecolier, E. Thesis, Universite´ d'Orle´ans, France, 1998. (11) Van Olphen, H. Discuss. Faraday Soc. 1951, 11, 82. (12) Norrish, K. Discuss. Faraday Soc. 1954, 18, 120. (13) Callaghan, I. C.; Ottewill, R. Faraday Discuss. Chem. Soc. 1974, 57, 110. (14) Rosta, L.; Von Gunten, H. R. J. Colloid Interface Sci. 1996, 134, 397. (15) Pignon, F.; Magnin, A.; Piau, J.-M.; Cabane, B.; Lindner, P.; Diat, O. Phys. Rev. E 1997, 56, 3281.
10.1021/la0101494 CCC: $20.00 © 2001 American Chemical Society Published on Web 06/02/2001
Letters
Figure 1. Silicon fluorescent yield mapping of a montmorillonite (Swy-2) clay gel at 50 g/L (resolution, 1 µm; dwell time, 400 ms). A, B, and C correspond to the three profiles analyzed in Figure 4.
ing,10,15-18 and small-angle neutron scattering measurements19,20 on Laponite and montmorillonite gels revealed the presence of density fluctuations on distance scales reaching 1 µm, that is, the lowest q-range reachable using such techniques. Optical birefringence measurements confirmed this long-range ordering and even showed the presence of orientional order extending up to the centimetric range.21 Furthermore, some of the birefringence images observed revealed the presence of Schlieren’s textures that were assigned to topological defects usually encountered in nematic crystal liquid phases.21 New insights on clay gels could therefore certainly be obtained by carrying out experiments that allow a direct visualization of gels. Some preliminary attempts using solvent-exchange methods or fast-freezing techniques10,22,23 have been performed, but the images obtained can never be considered as truly artifact-free. In that context, X-ray microscopy appears as a most promising technique, since it enables the investigation of hydrated samples without any pretreatment.24 The potential of this technique has been illustrated by recent measurements on suspensions of soil aggregates using soft X-rays of energies around 500 eV in absorption contrast.25,26 This requires very thin (16) Mourchid, A.; Delville, A.; Lambard., J.; Le´colier, E.; Levitz, P. Langmuir 1995, 11, 1942. (17) Mourchid, A.; Le´colier, E.; Van Damme, H.; Levitz, P. Langmuir 1998, 14, 4718. (18) Morvan, M.; Espinat, D.; Lambard, J.; Zemb, T. Colloids Surf., A 1994, 82, 193. (19) Ramsay, J. D. F.; Swanton, S.; Bunce, J. J. J. Chem. Soc., Faraday Trans. 1990, 86, 3919. (20) Ramsay, J. D. F.; Lindner, P. J. Chem. Soc., Faraday Trans. 1993, 89, 4207. (21) Gabriel, J.-C. P.; Sanchez, C.; Davidson, P. J. Phys. Chem. 1996, 100, 11139. (22) Gu, B.; Doner, H. E. Clays Clay Miner. 1992, 40, 246. (23) Vali, H.; Hesse, R. Clays Clay Miner. 1992, 40, 620. (24) Preis, T.; Thieme, J. Langmuir 1996, 12, 1105. (25) Neuhausler, U.; Abend, S.; Jacobsen, C.; Lagaly, G. Colloid Polym. Sci. 1999, 277 (8), 719. (26) Thieme, J.; Niemeyer, J. Geol. Rundsch. 1996, 85, 852.
Langmuir, Vol. 17, No. 14, 2001 4145
Figure 2. TEM image of Swy-2 montmorillonite. Rounded black dots correspond to the molecular organic molecule Masquol used to disperse the clay layers. The images were obtained at 80 kV in order to improve the contrast of the clay layers.
Figure 3. Silicon fluorescent yield mapping of a montmorillonite (Swy-2) clay gel at 50 g/L (resolution, 1 µm; dwell time, 400 ms). D, E, and F correspond to the three profiles analyzed in Figure 4. The arrows show the various types of orientational defects.
samples (typically < 1 µm), which is inappropriate for studying gels, as the presence of wall effects in such setups can never be safely written out. In contrast, multikiloelectronvolt X-ray microscopy attributes several advantages to the study of gels: (i) due to the higher focal length, wet cells can be used, (ii) thick samples can be studied thanks to a longer focus depth, and (iii) in the high-energy range, fluorescence yield becomes high enough to be mapped with significant contrast. The aim of this Letter is to prove that X-ray microscopy can reveal new features related to the texture of montmorillonite gels.
4146
Langmuir, Vol. 17, No. 14, 2001
Letters
Figure 4. Silicon concentration profiles perpendicular to the lamellae. A, B, and C (respectively, D, E, and F) refer to profiles labeled in Figure 1 (respectively, Figure 3). The vertical lines indicate domains selected to perform Fourier transforms. Part A′ presents a typical decomposition of the silicon fluorescence profiles after Fourier analysis where the dotted curve represents the experimental points and the plain line curve represents the corresponding smoothed curve. The two plain curves at the bottom of part A′ are the high- and low-frequency components extracted after Fourier analysis.
Experimental Section Montmorillonite gels were examined using the X-ray microscopy beamline, ID21, at the European Synchrotron Radiation Facility (Grenoble, France). The X-ray beam was focused down to a microprobe of 0.5 × 0.5 µm2 by mean of a Fresnel zone plate. Two Si 〈111〉 crystals provided a monochromatic beam (∆E/E ) 10-4).27 The fluorescence signal was analyzed using a high-energy resolution germanium solid-state detector. The beam energy was fixed at 2500 eV to ensure a good fluorescence yield for silicon. The investigated depth was estimated to be around 50 µm. The sample investigated was a Wyoming montmorillonite provided by the Clay Minerals Society (Sample Swy-2). Prior to use, the clay was purified and sodium saturated by washing and centrifugation. The final suspension was air-dried. The powder was then redispersed at a concentration of 50 g‚L-1 (∼2% volume) in bidistilled water under high shear to obtain a gel with a yield stress around 300 Pa. A few milligrams of gel were placed in a vacuum-tight cell with two Kapton windows 12.5 µm thick.
Results and Discussion Figure 1 presents the silicon fluorescence yield image obtained on a first montmorillonite gel. It has been rebuilt from four different scans (100 µm × 100 µm with a (27) Further information on the experimental setup can be found on the web at the address http://www.esrf.fr/exp_facilities/ID21/.
resolution of 1 µm and a dwell time of 400 ms) after displacement of the sample holder by high-precision step by step motors. The most striking feature of this image is the presence of long-range orientational order ()200 µm) with aligned domains richer in silicon (3-8 µm wide) alternating with Si-poor zones (5-30 µm wide) corresponding to water domains. Some extra water pockets with a nonlamellar morphology can be observed in zones where different clay-rich lamellae end. The silicon-rich objects are at least 2 orders of magnitude larger than the size of the elementary clay layers (Figure 2) which are polydisperse between 0.1 and 0.8 µm. This image therefore represents the first direct experimental evidence for the existence of a superstructure in montmorillonite gels. Such superstructure accounts for the long-range order revealed by optical birefringence measurements.21 However, it must be stressed that this orientational order is not directly due to individual clay platelets but to “packs” of clay layers, the structure of which remains to be determined. A second image (Figure 3) obtained on a second montmorillonite gel with the same concentration, using identical acquisition parameters to those for the previous image, reveals similar oriented domains of silicon-rich lamellae alternating with water domains. However, in
Letters
Langmuir, Vol. 17, No. 14, 2001 4147
correspond either to the existence of a substructure or to experimental effects, and demands further investigation. In contrast, the period of the low-frequency wave varies between 15 and 21.5 µm. As shown in Figure 5, it displays an inverse linear correlation with the mean fluorescence yield of the profile. This suggests metastable or unstable processes which could explain the long-term evolution of clay gels observed by various authors.10,28 Conclusion and Perspectives
Figure 5. Period variation of the extracted components (measured on each profile of Figure 4) as a function of the local mean fluorescent yield: open circles, low-frequency component; filled circles, high-frequency component.
this case, orientational defects can clearly be observed. Two main types of defects can be distinguished: (i) bundles of lamellae secant to other bundles with an angle of around 110° (arrows 1 and 2) and (ii) kink-folded structures affecting silicon-rich lamellae (arrows 3 and 4). However, the question arises as to whether such kinks are intrinsic or due to the shearing imposed to the gel upon its introduction in the analysis cell. In our opinion, we would tend to favor the second assumption, as the kinks observed are typical of sheared structures observed in sedimentary and metamorphic rocks. Still, it must be noted that those patterns were observed more than 4 h after the sample was introduced in the cell, which shows that even if such structures are shear-induced, they must be taken into account to understand the rheological properties of claybased gels. Such time-persistent shear induced structures were already observed in SANS experiments,20 which suggests that wall effects in the cell are not dominant. To analyze the superstructure, concentration profiles were performed perpendicular to the lamellae (Figure 4). A periodical evolution can be evidenced on all the sections. This clearly reveals ordered microsegregation of clay-rich domains the origin of which remains to be investigated (isotropic/nematic transition frustrated by charge and/or entanglement, demixtion, spinodal decomposition). A Fourier analysis was also carried out on selected regions of the profiles noted by vertical lines in Figure 4. Large water pockets were then excluded from the analysis. Profile C was split into two parts, as there is a strong concentration variation on the whole section. In all cases, the fluorescence signal can systematically be interpreted as the sum of two periodical functions (Figure 4A′). The period of the high-frequency wave is nearly constant around 6 µm for all the profiles (Figure 5). It could
Our preliminary X-ray microscopy study of montmorillonite gels unambiguously reveals for the first time the existence of large-scale structures much larger than individual clay platelets. At the same time, it raises numerous questions about the fundamental physical mechanisms underlying the formation of such entities. First attempts to observe structures at a lower scale, within the clay-rich lamellae, did not give any convincing visual evidence for local ordering. However, this may be due to insufficient resolution and too large a penetration depth, and should be investigated thoroughly with better adapted operating conditions. Furthermore, if, as suggested by our preliminary results, the metastable nature of clay gels can be clearly proven, it will be necessary to study in detail the influence of gel preparation conditions (solid dispersion or suspension concentration by osmotic stress) on the final organization in the gels. As shown by these preliminary results, X-ray microscopy can provide much information about inorganic gels, especially if coupled with optical measurements. In particular, the full phase diagram concentration/ionic strength should be explored using both techniques in parallel for a given clay sample. The influence of particle morphology (size and anisotropy) and polydispersity on the existence and formation of superstructures should also be investigated. The case of synthetic Laponite clay samples, where gel formation at low ionic strength can be interpreted as a glass transition,29-31 should also be analyzed for trying to understand the differences observed between the colloidal behaviors of Laponite and montmorillonite. Acknowledgment. We would like to acknowledge Dr. Pierre Levitz (CRMD) for very fruitful discussions and helpful comments. LA0101494 (28) Willenbacher, N. J. Colloid Interface Sci. 1996, 182, 501. (29) Avery, R. G.; Ramsay, J. D. F. J. Colloid Interface Sci. 1986, 109, 448. (30) Bonn, D.; Kellay, H.; Tanaka, H.; Wegdam, G.; Meunier, J. Langmuir 1999, 5, 7534. (31) Levitz, P.; Le´colier, E.; Mourchid, A.; Delville, A.; Lyonnard, S. Europhys. Lett. 2000, 49, 672.
