Expansion volume of alkali activated foamed cork composites from tungsten mining mud waste with aluminium

Several mixes of alkali activated foams from tungsten mud waste (MW), grounded glass (GG) and metakaolin (MK) were developed incorporating expanded granulated cork (EGC). This study presents preliminary results of the expansion process obtained with the addition of aluminium (AL) powder as a foaming agent. 0.3 wt.%, 0.4 wt.% to 0.5wt.% of AL powder were added to the alkali activated matrix. The physical and mechanical properties of the obtained foams, the effects of the type and amount of the foaming agent added are presented and discussed. Highly porous structures were obtained, with overall expansion up to 68.2% when the AL powder was added. The size and distribution of pores are shown. The compressive strength of foams in the case of highly porous structures achieved of 1.2 MPa for the samples containing 0.5 wt.% of AL powder. Mechanical and thermal properties of the cured structure are good and can therefore be used for applications in acoustic panels and lightweight prefabricated components for thermal insulation purposes.

A lot of research is currently being carried out to produce alkali activated material from mining waste [1], and the focus is to reuse the waste materials by integrating them into the manufacture of new materials [2], the EU and other countries have committed to the reutilization of the mineral wastes as precursor materials for alkali activated binders from an environmental, technical and economic point of view [3] [1].
The Panasqueira tungsten mining waste in Covilhã, Castelo Branco District, Portugal (figure 1) present a very good reactivity with alkaline activators such as calcium hydroxide, for high alkali concentration and the mechanical strength thru curing at room temperature was improved later on by mixing the mud waste (MW) with different source of silica (river sand and amorphous ground waste glass) and cured at moderate temperatures [4] [5].
Recently, alkali activated foamed materials have been proposed as a novel idea which involves the production of the lightweight materials combining the performance and the benefits of energy saving (carbon footprint) with the cradle-to-gate emission reductions obtained [6] [7].
In the last few years, several examples of alkali activated foams have been obtained from different raw materials [8] [9]. One of the used methods to produce alkali activated materials is the thermal expansion of (Na, K)-poly (silicate-multisiloxo) [10].
Various chemical foaming agents can be used in the foaming process of inorganic polymers. For example, AL powder has been used to produce foams of inorganic polymers [11] [12] as a chemical foaming agent. The reaction between the AL powder and the alkaline activator takes place very quickly, as shown in equation (1); [13].
2Al + 2NaOH + 2H2O → 2NaAlO2 + 3H2. (1) Currently, cork has found a wide range of application in the building industry as a solution for lightweight, thermal insulation and shows good environmental advantages [14]. Karad [15] studied the influence of cork granules for the manufacture compatibility with cement and the hydration test results showed that cork granules (both natural and expanded) are compatible with cement and can be added up to 30%.
The aim of this study is to investigate the expansion process (mixing procedure) and parameters involved (time of the reaction and expansion volume) to produce foamed lightweight materials using different combinations of mud waste (MW), grounded glass (GG), metakaolin (MK) and expanded granulated cork (EGC) mixed with sodium silicate and sodium hydroxide, while using AL powder as a foaming agent. The materials used in the study reported were mud waste (MW), grounded glass (GG), metakaolin (MK) and expanded granulated cork (EGC). The size of all powders was under 300 µm and cork particles sizes range between 2 and 4 mm. The mud waste (MW) obtained from Panasqueira tungsten mine was used after the preparation and passing through a 300 µm sieve as aluminosilicate rich precursor. Waste glass (GG) from glass bottles was also crushed and prepared at UBI's laboratory. Metakaolin (MK) provided by BASF with Specific Gravity of 2.50 (g/cm3) and pH = 6 (28% solids). Aluminium (AL) powder used was provided by ACROS Organics with purity and molar mass of 99 % and 26.98 Al respectively. The expanded granulated cork (EGC) used was provided by (SOFALCA Sociedade Central de Produtos de Cortiça, Lda, Abrantes, Portugal).    The precursor was prepared by mixing dry powders of mud waste (MW), grounded glass (GG), metakaolin (MK) and cork as shown in figure 3. Different percentage of AL powder was added to the precursor which was then mixed with all dry components by hand for 1 minute. After the magnetic stirring at a rate of 400 r.p.m for about 5 minutes of (10M) sodium hydroxide (SH) with sodium silicate (SS), all components of the formulation were then mixed together for about 2 minutes at a medium speed. Several alkalis activated mixes were manufactured with the same ratio of precursor/activator, P; A = 2.75 and same activators ratio of SS; SH = 3.1 to study the influence of adding different percentage of AL powder on the expansion process as shown in table 3. Various mixes were produced with different percentage content of AL powder; ranging between 0.3 wt.% to 0.5 wt.%. The mixes were poured into 200*200*40 mm 3 molds and were placed firstly in the oven at 60 °C for 24 hours for curing to speed up the alkali activation. After curing in the oven, the samples were demoulded and left to continue the curing in laboratory conditions for the following days until testing of the compressive strength after 28 days. The compressive strength test for the alkali activated foam was made with a very precise machine (AUTOGRAPH AGS-X, 10 KN SHIMADZU, at the mechanics of materials and mechanical technology laboratory, DECA UBI). All reported results are an average of 3 different measurements.
As it was expected, the mechanical, bending and the compressive strength is strongly affected by the porosity and cork particles content in the alkali activated foamed cork composites. As shown in figure 4   The density of the samples was measured by dividing the dry mass by the volume (i.e. so-called geometrical density). Cubic specimens (40*40*40 mm 3 ) were used for dry density measurement's. In the case when cork particles were added the lowest foam composite density was achieved at about 542.02 Kg/m 3 with 50 vol.% of expanded granulated cork (M5 and M6). The dry density of the composites is reported in figure 7. It can be observed that the increase of the AL powder enlarges the pore size of the samples. In addition, when adding large amount of foaming agent, the reaction between the AL metal powder and the alkaline activator (NaOH) takes place very quickly, therefore the expansion is very fast and violent. The expansion volume of the mixes was about 68.2 % for all samples which didn't contain cork particles. Samples with 25 vol.% and 50 vol.% of cork particles were achieved an expansion of 17.10% and 13.64 respectively. It can be clearly observed that the cork particles affect the expansion volume during the foaming process, it was also observed that the reaction time decrease with increased foaming agent content due to the faster reaction between AL powder and the catalyser (NaOH) (10 min with 0.3wt.% and 3min with 0.5 wt.%).

Fig. 7. The relation between cork content and density.
Images of the central foam section show a macrostructure with irregular spherical ultramacrospores derived from the foaming process. The degree of reactivity of the blowing agent and the homogeneity of the precursor affect the morphology of the pores (shape and diameter) and their distribution (regularity). The pore distribution of the alkali activated foam composites was presented in figure 8. The pore size is directly a result of the amount of foaming agent added, the analysis of particle size and its distribution of the foams composites was determined by "Image J" software that could characterize individual particles (size, shape and surface) of the convex particles. The Feret diameter averaged over all directions (〈F〉) is equal to the ratio of the object perimeter (P) and the number pi (constant of Archimedes), i.e., 〈F〉 = P/π. There is no such relation between 〈F〉 and P for a concave object. Figure 9 shows the histograms of the pore size of the foams composites.