Self-setting particle-stabilized foams with hierarchical pore structures

Materials Letters 64 (2010) 1468–1470 Contents lists available at ScienceDirect Materials Letters j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / m a t l e t Self-setting particle-stabilized foams with hierarchical pore structures Franziska Krauss Juillerat a,⁎, Urs T. Gonzenbach a, André R. Studart b, Ludwig J. Gauckler a a b Nonmetallic Inorganic Materials, Department of Materials, ETH-Zurich, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland Complex Materials, Department of Materials, ETH-Zurich, Wolfgang-Pauli-Strasse 10, CH-8093 Zurich, Switzerland a r t i c l e i n f o Article history: Received 7 January 2010 Accepted 25 March 2010 Available online 1 April 2010 Keywords: Ceramics Microstructure Porosity Self-setting Structural applications a b s t r a c t This study presents a novel method to produce self-setting inorganic foams with unique hierarchical pore structures. The combination of particle-stabilized alumina foams with calcium aluminate cement leads to a macroporous ceramic material which can be shaped and consolidated at room temperature, bypassing the challenging and sometimes extensive drying and sintering steps. Due to the water consuming cement hydration reaction, no macroscopic shrinkage was observed and crack and cavity formation was prevented throughout the entire specimens. The final microstructure features a porosity of 76 vol.% and a unique hierarchical pore structure with interconnecting pores of 200 µm diameter, separated by mesoporous pore walls. The authors believe that this newly developed method opens a door to novel applications where so far the drying and sintering step and its inherent shrinkage of the ceramic foams were the limiting factors. © 2010 Elsevier B.V. All rights reserved. 1. Introduction 2. Experimental Porous ceramic materials are used in numerous fields such as filtration, catalysis, and refractory insulation [1]. Most of the fabrication routes make use of a drying and sintering step to consolidate the foams and give mechanical strength to the final material [2]. For the present study, we apply a recently developed direct foaming method [3] which has the potential for industrial applications due to its versatility [4] and straight forward processing [5]. This direct foaming method is based on using partially hydrophobized particles to stabilize the air–water interface of freshly generated air bubbles [4]. The resulting wet foams are stable against bubble coarsening and drainage, and bubble size and air content can be accurately tailored in a wide range by the processing and compositional parameters [6]. Up to now, these foams were made from various inert materials and were therefore dependant on a drying and sintering step to reach their excellent mechanical properties after heat treatment [3,4]. Drying and sintering are usually accompanied by shrinkage that might cause cracks and defects. In order to avoid shrinkage, to bypass these often delicate and costly drying and sintering steps and to widen the application range we chose calcium aluminate cement as reactive phase to consolidate the alumina foams without the need of a heat treatment. As a result an alumina–calcium aluminate composite foam is produced. 2.1. Materials ⁎ Corresponding author. Tel.: + 41 44 633 6834; fax: + 41 44 632 1132. E-mail address: [email protected] (F. Krauss Juillerat). 0167-577X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2010.03.062 The materials used were α-Al2O3 powder (Ceralox, USA), propyl gallate, lithium carbonate (both Fluka AG, Switzerland), hydrochloric acid, sodium hydroxide (Titrisol, Merck, Germany), deionized water (18.2 MΩ/cm) and calcium aluminate cement (CA-270, Almatis GmbH, Germany). 2.2. Methods The suspension was prepared by adding 50 vol.%. alumina powder to an aqueous solution containing 27 mmol/l NaOH and 22 mmol/l propyl gallate. After ball-milling for 18 h, the solids loading was adjusted to 20 vol.% and the propyl gallate concentration was set to 2 wt.% (to alumina particles). Directly before foaming (or casting), cement powder (200 wt.% to alumina) and the setting accelerator lithium carbonate (5 wt.% to cement) were added. The early setting of the cast samples was monitored using a stress controlled rheometer (model CS-50, Bohlin Instruments, U.K.) with profiled parallel plates (diameter: 25 mm, gap: 1 mm). The advanced setting behavior was determined by placing stainless steel balls (weight: 67 g, diameter: 2.54 cm) on the cast samples and measuring the indent depth (Fig. 1). Foaming and foam characterization were carried out as described elsewhere [3]. In addition, to analyze the set foams, SEM micrographs were used (LEO 1530, Zeiss, Germany). The compressive strength measurements were performed on a universal testing machine (Instron 8562, model A1477-1003, Norwood, USA) on cylindrical samples (diameter: 10 mm, length: 20 mm, 10 samples per data point), tested with a compression speed of 0.5 mm/min. The F. Krauss Juillerat et al. / Materials Letters 64 (2010) 1468–1470 Fig. 1. a) Early setting behavior of a cast sample measured by means of rheology; b) setting behavior of the cast sample measured with an indentation method as shown in c: ( ) liquid phase, the steel ball sinks to the ground; ( ) initial setting phase, the steel ball is supported by the consolidating mass but leaves an indent larger than 5 mm in diameter; (■) advanced setting phase, the steel ball leaves an indent of 5 mm in diameter or smaller. The final setting time is defined when no indent is visible anymore; c) different setting phases of the indentation method according to b. Ball diameter: 2.54 cm. samples were stored in a water-saturated atmosphere at room temperature. 3. Results and discussion In a first step, the alumina suspension was modified with propyl gallate in order to induce hydrophobicity to the particles and make the suspension foamable [4]. Calcium aluminate cement was added to this suspension and the setting behavior of this un-foamed mixture was investigated. It was found that the suspension solidified, however the setting time was 5 days. This is rather long compared to the setting time of a pure cement slurry which takes about 6–8 h (suppliers specification). We believe that the propyl gallate molecules are partially adsorbed onto the cement particles and inhibit the dissolution of the cement and/or poison the nuclei of the freshly precipitated hydrated cement phase. Therefore the setting accelerator lithium carbonate was added to the suspension mixture, acting as a nucleation agent for the hydrated cement phase [7]. The early setting behavior of this cast but not foamed suspension mixture is displayed in Fig. 1a. After an induction period of about 10 min, the setting reaction begins and the suspension becomes gradually stiffer. Before the viscous and elastic moduli were reaching 106 Pa, the experiment was stopped to avoid damage to the rheometer-equipment. To obtain time resolved data of the final setting phase, a simple indentation method (Fig. 1c) was used. As shown in Fig. 1b, the final setting of the cast suspension using lithium carbonate as setting accelerator is already achieved after 115 min. After having adjusted the setting time of the cement enriched alumina suspension, foaming experiments were conducted. The mechanical frothing of the modified suspension produced wet 1469 foams with air contents of 75 vol.% and an average bubble size of 85 µm, showing no drainage prior to setting. During the setting process, the samples were kept from drying and did not exhibit any shrinkage. As a result, the consolidated foam did not peel away from the borders of a Petri dish and could not be removed from it, as shown in Fig. 2a. The consolidated foams were free of cracks or cavities throughout their entire volume and could be machined into various shapes (Fig. 2b). Their porosity was 76 vol.% and similar to the wet foam air content, which implies that most of the water in the wet foam is used up during the cement setting reaction. The consolidated foam microstructure (Fig. 2c) features an average pore size of about 200 µm. This is much larger compared to the wet foams bubble size of 85 µm, thus bubble growth must have occurred prior to setting. We believe this partial destabilization of the wet foam to be a result of an adsorption competition of the propyl gallate molecules to the alumina as well as to the cement particles. This is rendering the alumina particles less hydrophobic compared to the system without cement and hence, the tendency of the particles to adsorb to the air–water interface is reduced. As a result, particles are less strongly adsorbed and bubble coarsening occurs. Similar effects have already been observed in earlier studies where a reduction in surface modifier concentration leads to reduced foam stability [6]. Furthermore, the pores in the micrograph exhibit small pore openings — a new feature for such type of particle-stabilized foams and most likely resulting from partial coalescence and setting of the wet foam. This effect will be addressed in continuative experiments. A remarkable feature of this porous material is its hierarchical pore structure that spans over several orders of magnitude. While in sintered materials the macropores are usually separated by dense pore walls [3], they contain mesopores in this self-setting foam. Fig. 2d shows a single foam lamella consisting of alumina grains that are embedded into newly formed cement crystals which are filling the voids between the inert particles and form pores smaller than 50 nm. This feature might be of special interest in the area of thermal insulation. The mechanical strength of foam cylinders was tested after 1 day, 3 weeks as well as 2 and 3 months. The experiments confirmed that the compressive strength of the foam strongly depends on the progression of the cement reaction and is therefore time dependant (Fig. 3). After 1 day, the cement hydration reaction has not completed and a maximum strength of 0.92 MPa is reached after 3 weeks before it decreases to 0.57 MPa after 3 months. This decrease in strength is due to the continuing cement reaction, the conversion of the hexagonal cement phases into the cubic phase, causing a localized contraction in volume and hence a formation of local microstructural defects which leads to a reduction in strength [8]. 4. Conclusion In conclusion, we have shown that porous self-setting materials can be obtained with the particle-stabilized foaming method. Using a calcium aluminate cement bypasses the often delicate and costly drying and sintering process and zero-shrinkage, crack-free samples were produced. Adding the setting accelerator lithium carbonate reduces the setting time from 5 days to 115 min. The microstructure of the final material shows a unique hierarchical pore structure, consisting of interconnected macropores which are separated by mesoporous walls. The sample features a porosity of 76 vol.% and the average macropore is 200 µm in size. From these first results, we believe that this method has a large potential for tailoring the microstructure of porous self-setting materials in a much wider range compared to already existing technologies [9–11]. This will allow the fabrication of a variety of materials with application-specific properties for example in the area of construction or for biomedical applications. 1470 F. Krauss Juillerat et al. / Materials Letters 64 (2010) 1468–1470 Fig. 2. a) Petri dish containing consolidated foam: no shrinkage or crack formation is observed; b) machined specimens; c) SEM image of the set foam microstructure showing interconnected pores; and (d) foam lamella showing alumina particles (∼ 200 nm) embedded in a mesoporous calcium aluminate phase. financial support and Almatis GmbH which kindly provided us with samples of calcium aluminate cement. References Fig. 3. Time dependant compressive strength measured for 10 foam cylinders at each data point. Acknowledgements The authors would like to thank Pierre Elser, Roman Kontic and Mario Mücklich for their contributing work, CCMX and SPERU for [1] Scheffler M, Colombo P. Cellular ceramics: structure, manufacturing, properties and application. Weinheim: Wiley-VCH; 2005. [2] Studart AR, Gonzenbach UT, Tervoort E, Gauckler LJ. Processing routes to macroporous ceramics: a review. J Am Ceram Soc 2006;89:1771–89. [3] Gonzenbach UT, Studart AR, Tervoort E, Gauckler LJ. Macroporous ceramics from particle-stabilized wet foams. J Am Ceram Soc 2007;90:16–22. [4] Gonzenbach UT, Studart AR, Tervoort E, Gauckler LJ. Stabilization of foams with inorganic colloidal particles. Langmuir 2006;22:10,983–8. [5] Gonzenbach UT, Studart AR, Steinlin D, Tervoort E, Gauckler LJ. Processing of particle-stabilized wet foams into porous ceramics. J Am Ceram Soc 2007;90: 3407–14. [6] Gonzenbach UT, Studart AR, Tervoort E, Gauckler LJ. Tailoring the microstructure of particle-stabilized wet foams. Langmuir 2007;23:1025–32. [7] Rodger SA, Double DD. The chemistry of hydration of high alumina cement in the presence of accelerating and retarding admixtures. Cem Concr Res 1984;14: 73–82. [8] Scrivener KL. Calcium aluminate cements. In: Newman J, Choo BS, editors. Advanced concrete technology. Oxford: Elsevier Ltd; 2003. p. 2/1–2/30. [9] Takagi S, Chow LC. Formation of macropores in calcium phosphate cement implants. J Mater Sci Mater Med 2001;12:135–9. [10] Sugama T, Wetzel E. Microsphere-filled lightweight calcium-phosphate cements. J Mater Sci 1994;29:5165–76. [11] Tonyan TD, Gibson LJ. Structure and mechanics of cement foams. J Mater Sci 1992;27:6371–8.