Abstract

This study focuses on the preparation of porous geopolymer cements for thermal insulation applications using commercial calcium carbonate and low-value calcium carbonate-rich waste such as eggshell powders as pore-forming agents. The control and porous geopolymer cements were prepared by adding phosphoric acid (4M) as a chemical ingredient to metakaolin containing 0 and 10 wt% of the foaming agent. Results showed that Sangare clay is sandy with a few silts and traces of gravel. Diffractogram of the clay indicates the presence of: Kaolinite (Kao); Illite (Ill); Quartz (Qz) and Potassium Feldspars (Fds). This was confirmed by the Infrared spectrum of the material where bands characteristic of the presence of kaolinite (3689-3649 cm ^-1) and of other minerals have been found. Characterization of eggshell powder shows that the source of calcium consists mainly of CaCO₃. The apparent densities of geopolymer cement obtained decreased when the aluminosilicate materials (metakaolin) are replaced by proportions of eggshell powders and commercial calcium carbonate (0, 1, 3, 5, 7 and 10%). It was also observed that the compressive strengths of geopolymer cements based on eggshells are greater than those based on commercial calcium carbonate and the thermal conductivity decreases with the increase of the porogen. These results corroborates with that of the apparent density and compressive strength which decreases with the increase of the porogen. Results shows that geopolymer cements obtained have accumulated high pores in their structures. The values of thermal conductivity of the control and porous geopolymer cement from eggshell ranges between 0.10 and 0.17 W/mK, respectively. It was concluded that the low-value calcium carbonate-rich wastes (egg shell) could be used for producing porous geopolymer cements which could be utilized for thermal insulation applications.

1. Introduction

The term geopolymer was first introduced in 1979 by Professor Joseph Davidovits. He defines it as a class of semi-crystalline aluminosilicate material with a three-dimensional network, obtained by mixing an aluminosilicate powder with a hardener ¹. The hardener generally used geopolymers are solutions of sodium or potassium silicate and a solution of phosphoric acid ^{1, 2, 3, 4}. The work carried out on geopolymers for more than two decades is oriented in the construction of large buildings. Little work has been done on the synthesis of porous geopolymer cements, although the latter find applications in eco-construction ⁵, adsorption ^{6, 7}, catalysis ⁸, water absorption ^{9, 10}. The porous geopolymer materials are synthesized in the same way as those which do not contain pores with the only difference that in addition to the hardener, a pore-forming agent which can be a powder or a solution is added to the aluminosilicate material. The pore-forming agents generally used for the synthesis of porous geopolymer materials in an alkaline medium are aluminum powder ^{11, 12}, silica fume ¹³, hydrogen peroxide ^{1, 14, 15, 16}, sodium perborate ¹⁷. These pore-forming agents used by these authors are expensive. It is important to note that these authors used metakaolin as a source of aluminosilicate for the production of porous materials. Previous work has shown that the non-porous geopolymeric materials synthesized from metakaolin have properties far superior to those obtained from volcanic slag ¹⁸. This is due to the low reactivity of volcanic slag compared to that of metakaolin. This low reactivity is linked to a low specific surface area of volcanic slag (2.3 or 15.7 m ² /g) compared to that of metakaolin (20.5m² /g) ¹⁸. Recycling of natural waste such as eggshells, oyster shells, snail shells, rice husk ash, wheat ash, etc. offers many economic and environmental benefits, as it reduces disposal costs in landfills and maintains a healthy environment. Eggshells have a high content of calcium carbonate and until today, this waste is not yet used as a pore-forming agent for the production of porous geopolymer materials in acidic and/or alkaline media. The main goal of this work is to use eggshell powders as a pore-forming agent and garoua sangaré clay for the preparation of porous geopolymer cements and then compared the properties of these materials to those obtained using calcium commercial carbonate.

2. Literature Review

Some researchers ^{19, 20} have concluded that geopolymeric materials are characterized by their low thermal conductivity, low compressive strength and low bulk density. The blowing agents commonly used for the production of porous geopolymer cements are aluminum powder ^{19, 20, 21}, hydrogen peroxide ^{1, 22}, silica fume ²³, the sodium perborate ^{17, 24} used glass powders to prepare porous geopolymers and eggshell powders were used as blowing agents ²⁵ used eggshells ²⁶ prepared porous geopolymeric cements based on phosphoric acid using limestone as a pore-forming agent. It is important to indicate that these authors studied the thermal behavior of porous geopolymer cements up to 1000°C.

3. Material and methods

During the construction of a well for access to underground water, a strange observation was made. On a lateritic horizon, at a depth of about 3 m, we notice the appearance of a layer of soil of a whitish and greyish color, layer of soil to which we pay all the interest of this work. The objective of this part is to present the major features of the studied area. Regarding soils, the parent material is mainly composed of clay and sand.

Geologically, Garoua is based on sedimentary formations mainly sandstone which is a detrital rock formed in the Cretaceous and resulting from the aggregation of grains of sand ²⁷. There is also the presence of massifs of gneiss which is a metamorphic rock composed of quartz, mica and feldspars. Sudanese tropical climate of the region combined with the presence of a river leads to the alteration of parent rocks and the formation of lacustrine clays ²⁸ dating from the Quaternary period which sediment in alluvium, eluvium and are sometimes covered with a humic horizon and lateritic.

On the whitish and grayish soil layer extracted from the well, a sampling targeted at the point of coordinates in decimal degrees, Latitude 9.285126° and Longitude 13.455767° was carried out. More attention was paid to the compact and gray masses of uncontaminated soil which crumbled to the touch because the other fraction of completely whitish soil was macroscopically only sand. Documentary studies ²⁷ suggested the clayey nature of this greyish material (Figure 1). The samples taken are shown in Figure 2.

The particle size analysis determines the dimensional distribution by weight of the grains constituting the sample.

They include X-Ray Diffractometry (XRD) and Infrared Spectrometry (IR).

The sources of calcium carbonate used are commercial calcium carbonate and chicken eggshells. For this study, the shells were collected in municipal landfills in Maroua. These shells, are washed with tap water, air dried and then broken by hand into small particles (about 2-4 mm). Before use, the organic material was removed and the shells devoid of this organic material were air-dried, then calcined at 500°C for 2 hours in a programmable electric oven (Nabertherm, Mod. LH 60/14), according to a heating rate of 10°C/ min. Then, it was ground using an AMS brand ball mill to obtain the powder. The powder was sieved until full passage through a 80 µm mesh sieve. Commercial calcium carbonate was supplied by Sigma-Aldrich.

4. Experimental Methods

The hardener is obtained by diluting the commercial solution of phosphoric acid (H₃ PO_4, 85% purity) in distilled water in order to obtain a solution of molar concentration 4 M. The hardener thus prepared is left at room temperature for at least 24 hours before use.

The fresh geopolymer cement paste is obtained by gradually adding the hardener to the previously calcined clay (metakaolin) containing 0 and 15% by mass of each source of calcium in a porcelain mortar. The liquid / solid mass ratio is maintained at 0.83. The different formulations are then mixed manually for 5 minutes. The geopolymer cement pastes obtained by each formulation are molded in cubic (40 x 40 x 40 mm) and rectangular (40 x 40 x 15 mm plexiglass and cylindrical (63 x 30 mm) molds. The test pieces are then placed in a Genlab. Prime brand oven at 65°C for 24 h in order to accelerate the polycondensation process. The geopolymer cement bricks obtained are demolded, then sealed in plastics and left at room temperature (25±1) °C with a humidity of (55 ± 5) %, for 28 days.

The chemical composition of Sangare clay, egg shell powder and geopolymer cement was analyzed using a wavelength dispersive X-ray fluorescence apparatus (Shimadzu XRF-1800). In addition, X-ray diffraction (D RX) analysis was performed to test the mineralogical category using a Philips X'Pert PRO diffractometer equipped with a Ni- filtered Cu-Ka ( λ = 1.542Å) radiation. Thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) analyzes were performed using a 2960 TA Instrument under argon from room temperature up to 1000°C at a heating rate of 5°C / min. The microstructures of the cement were analyzed by scanning electron microscopy (SEM) (Hitachi S-4500). The particle size distributions of the various obtained geopolymer powders were measured using a laser diffraction particle size analyzer (Malvern Mastersizer 2000). Infrared spectroscopy (FTIR, iS50 RAMAN) was used to identify some surface groups.

This measurement was made by the Archimedes thrust method using an automatic density meter (Ceramic Instruments mod DDA/2). This analysis was carried out in the Civil Engineering Laboratory of Cameroon.

Compressive strengths were measured by crushingcubic molds (40 x 40 x 40 mm) of cured geopolymer materials, maintained at room temperature in the laboratory for 28 days. The test consists of subjecting the specimen between two opposing axial forces, located between the plates of the electrohydraulic press (IMPACT TEST Limited, Building 21, Stevenston Industrial Estate, Stevenston, Ayrshire, Scotland, UK KA2 0 3LR). Subsequently, the sample will be subjected to a continuous and progressive load at the average speed of 0.500 MPa/s until crushing. This analysis was carried out in the Civil Engineering Laboratory of Cameroon.

The fragments of different géolpolymèr samples were used for light microscopy. It was made using an optical microscope equipped with a stereo microscope (Ceramic Instruments, Model 101T-M7), a binocular head and a tablet 7 with integrated micro camera. This analysis was carried out in the Civil Engineering Laboratory of Cameroon.

Mercury intrusion porosimetry (MIP) was performed on fragments of samples of approximately 1cm³ in volume using an Autopore IV MIP 9500, 33,000 Psia (228MPa) covering the pore diameter range of about 0.006 to 350 µm having two low pressure ports and one high pressure chamber.

Thermal conductivity is an analysis that studies the diffusion of energy within a material. The method is based on the use of a transient plane sensor, the most common adaptation of which is the Hot Disk thermal properties analyzer. The specimens produced for this analysis are cylindrical specimens of dimensions (63 x 30 mm).

5. Results and Discussion

These results concern the geotechnical, physical, mineralogical characterization and global evaluation of the properties of the Sangaré clay material once transformed into ceramics.

The granulometric analysis data are grouped in Table 1. The clay material consists of 0.53℅ gravel, 41.15% sand, and 12.28℅ silt and 46.04℅ clay. It is apparent from the exploitation of the results that the Sangare clay material has a very wide granulometry because its uniformity coefficient is greater than 200. It is sandy clay with a few silts and traces of gravel.

Mineralogical analyzes

i) X-ray diffractometry

Examination of the total powder diffractogram obtained for the sample tested shows the presence of the following minerals with the inter-reticular distances of their main lines: Kaolinite (Kao): reflection at 7.17 Å, 1.48 Å, 3.576 Å; Illite (Ill): reflection at 9.995 Å, 4.996 Å, 4.45 Å; Quartz (Qz): reflection at 4.255 Å, 3.343 Å, 1.81 Å; Potassium Feldspars (Fds): reflection at 3.240 Å, 2.753 Å, 2.159 Å. We also note the presence of the peak corresponding to the total clay inter-reticular distance at 4.45 Å. At this peak, all clay minerals were found.

Figure 3 shows the oriented blade curves obtained from the clay fraction less than 2 microns.

The diffractogram obtained on the normal powder shows the presence of kaolinite and illite.

In order not to make an identification error in the event that there would be clay minerals in the neighboring interfollial space in the normal powder, a diffractometry was also carried out on an oriented slide after heating to 500°C (SP1-500°C). The obtained diffractogram shows that the peak of kaolinite detected on the diffractogram of the normal oriented plate disappears, which confirms that it is indeed kaolinite because it is very sensitive to heat. Illite is present with a significant peak of inter-reticular distance 9.97 Å. The ethylene glycol (SP1-EG) saturation of the clay fraction on an oriented slide shows the absence of interbedded clay minerals. The calcination of the kaolin will make it possible to obtain metakaolin which will be used subsequently for the production of cement.

ii) Infrared spectrometry

In the clay material, bands characteristic of the presence of kaolinite (3689-3649 cm^-1) have been found because the valence vibrations of four OH groups composing the elementary mesh of kaolinite result in four bands centered at 3695, 3667, 3651, and 3620 cm^{-1 29}. The 3619 cm^-1 band reflects the OH bond deformation and materializes the presence of kaolinite ²⁹. The 1114 cm^-1 band and the 1004 cm^-1 band correspond to the elongation vibration of the Si-O bond of kaolinite. The bands observed around 914 and 916 cm^-1 correspond to the deformation vibrations of the Al-O bond of the clay minerals ³⁰. The presence of Quartz is indicated by bands between 600 and 800 cm^{-1 31}. Figure 5 crosses the bands between 3500 and 4000 cm^-1 of the infrared spectrum and corresponding to the main vibrations characteristic of kaolinite.

Chemical compositions (mass/100g)

The chemical compositions expressed in percentages by weight of oxides of the raw materials used such as the hen egg shells (ES) is recorded in Table 3. It emerges from this Table that the chemical composition of snail shells (SS) and commercial calcium carbonate (CC) has a high content of CaCO₃ (98.39%). The results show that these sources of calcium consist mainly of CaCO₃.

It emerges from this table that the chemical composition of commercial calcium carbonate (CC) has a high CaCO₃ content of 99.35%. The results show that these calcium sources are mainly made of CaCO₃. The loss on ignition is 12.36 % by mass in the metakaolin obtained from the kaolin.

It emerges from this table that the chemical composition of commercial calcium carbonate (CC) has a high CaCO₃ content of 99.35%. The results show that these calcium sources are mainly made of CaCO₃. The loss on ignition is 12.36 % by mass in the metakaolin obtained from the kaolin.

ii. X-ray diffractograms

Figure 6 shows the X-ray diffractogram of eggshell powders (ES). This diffractogram shows the characteristic peaks of calcite (CaCO₃).

iii. Thermal analyzes (TGA / DSC) of eggshell powders

Figure 7 shows the DSC and ATG curves of the eggshell powders. We observe on the DSC curve, an endothermic peak around 96 ° C accompanied by a loss of mass on the ATG curve corresponding to the evaporation of adsorbed water ³². The DSC curve shows an exothermic peak only around 375°C which is accompanied by a low mass loss over that of A TG. This peak is attributable to the decomposition of organic matter ^{32, 33}. The endothermic peak around 622°C on the camber e DSC but not accompanied by any mass loss on the TGA curve suggests that c and accident is not due to a dehydroxylation but the décomposition aragonite in an intermediate amorphous phase ³². The endothermic peak on the DSC curve at about 906°C is accompanied by a loss of mass on the ATG curve. This phenomenon is attributed to the decomposition of calcium carbonate according to the equation:

iv. Infrared spectroscopy

Figure 7 shows the infrared (IR) spectrum of eggshell powders. This spectrum shows the presence of a weak absorption band at 3411 and 1662 cm^-1. This is attributable respectively to the OH and H-O-H bonds of the water of hydration molecules. The bands at 780, 2516, 2876, 2983 cm^-1 are assigned to the modes of vibration of the C = O bonds of carbonate groups ³⁴. Those at 713, 875 and 1419 cm^-1 are attributed to the vibrations of elongation of the CO bonds of calcite. The bands at 1156 and 672cm^-1 are attributable to the CO₃^2- ions of aragonite ³⁵. The band at 594 cm ^-1 is assigned to the O-Ca-O and Ca-O bonds ^{34, 36}.

i. Apparent density

It emerges from these values that the apparent densities of geopolymer cements decrease when the aluminosilicate materials (metakaolin) are replaced by proportions of eggshell powders and commercial calcium carbonate (0, 1, 3, 5, 7 and 10 %). This decrease is due to the ion size of the pores in geopolymeric cement structures and the pore diameters could increase as more eggshell powders are added. Although these values decrease, the values of the bulk densities of the geopolymer cements synthesized are much higher than those obtained by ^{37, 38}.

ii. Compressive strengths

It is observed that the compressive strengths of geopolymer cements based on eggshells are greater than those based on commercial calcium carbonate. This would be due to a weak reactivity of calcium carbonate causing rather a precipitation reaction between the dissolved species during the depolymerization of calcium carbonate favoring the formation of the crystalline phases which can consume a large quantity of the PO₄^3- ions coming from the 'Phosphoric acid. The increase in compressive strengths when the percentage of eggshell powders increases, would be due to the formation of a large amount of crystalline phases such as brushite, monetite, newberyte and tricalcium phosphate in the structures of geopolymer cements with 7% incorporation and which could play the role of filler. This result clearly corroborates that of the apparent densities (Table 4) which presents the same phenomenon.

iii. Thermal conductivity

According to the results of the thermal conductivities obtained in Table 6, the thermal conductivity decreases with the increase of the porogen. The decrease in thermal conductivity after bulk addition of the calcium sources is attributed to the formation of pores in these materials due to the emission of CO₂. This result corroborates with that of the apparent density and that of the compressive strength which decreases with the increase of the porogen. We can conclude that the thermal conductivity of porous geopolymer cements obtained from these additives rich in calcium has properties suitable for thermal insulation applications ³⁹ combined fly ash with sodium carbonate and sodium silicate for making of porous geopolymer having a thermal conductivity between 0.29 and 0,34W / mK. While ³⁷ used sawdust to prepare porous geopolymer cements and obtained a thermal conductivity between 0.15 and 0.24W/mK ⁴⁰ studied the mechanical and thermal properties of a geopolymer mortar lightweight by incorporating granular rubber. They indicated that the thermal conductivities were between 0.237 and 0.298 W/mK. It is important to note that all of these authors used sodium silicate as a hardener. The results obtained in this work are comparable to those obtained by ³⁷.

iv. Microstructures

Micro spellings

It can be seen in these Figure 9 and Figure 10 that the MKS0 geopolymer cements based on commercial calcium carbonate are homogeneous, compact and dense, whereas those of MKS0 based on eggshell are heterogeneous and the particles poly (phospho-siloxo) are not well connected to each other. It emerges from these micrographs that the pores are more accentuated on the micrographs of geopolymer cements based on eggshell. It is important to note that during the first stage of geopolymerization in an acidic medium, there is a binding of the proton H from H₃PO₄ to the pair of electrons free of the siloxane or siloxo bond, Si-O-Si ¹ according to the following equation:

This difference would be due to the fact that the H ⁺ proton, which fixes the free electron pair of the -Si-O-Si- bond rather reacts with the calcite contained in the eggshell powders. This is causing the formation of pores in the network of geopolymer cements. This was followed by the release of carbon dioxide, oxides such as CaO and MgO contained in commercial calcium carbonate (Table 1) promoting the precipitation of CaHPO₄, Ca₃(PO4)₂ and MgHPO₄ .3H₂O. This was contributed to the consumption of phosphate ions thus hampering the depolymerization process on the one hand and polycondensation on the other hand of the Si-OH and Si-OPO (OH)₂ species.

v. Mercury intrusion porosimetry (MIP)

According to data reported in Table 7 and Table 8, the volumes of cumulative pore geopolymer cements show that geopolymer cements obtained have accumulated high pores in their structures. These results are in agreement with those of compressive strengths. But, the values of the cumulative pore volumes obtained in this study are higher than those obtained by ⁴¹ (between 116 and 179.5 mm³ /g). It is important to note that these authors replaced up to 40 wt. % of metakaolin with rice husks (pore-forming agent). This difference proves that the nature and quantity of the added pore-forming agent plays a crucial role in the formation of the pores (dimensions, surface area and total volume of the pores) ³⁷. Replaced metakaolin by volume percentages of sawdust (0, 50, 75, 100, 125, 150, 175 and 200 % by volume) as a pore-forming agent for the synthesis of porous geopolymer cements using a solution of sodium silicate as a hardener. They reported that the cumulative pore volumes were between 180 and 700mm³/g. This shows that the type of hardener, its concentration and the blowing agent used could significantly influence the pore formation mechanism during the synthesis of porous geopolymer cements.

The values of the total porosities calculated with those of the apparent and absolute densities measured using a pycnometer of geopolymer cements are presented in Table 7 and Table 8. Using the bulk density and skeletal density obtained by measuring mercury intrusion porosimetry and absolute density from the helium pycnometer, the open and closed porosities were determined using the formula:

From the values recorded in Table 7 and Table 8, it is clear that the total porosity calculated in this case is greater than that obtained by the Porosity intrusion mercury, we can conclude that powders from porous poly(phospho-siloxo) networks could be used for thermal insulation applications.

6. Conclusion

The metakaolin particles in the structure of the final products involve the formation of pores in the network which is confirmed by the high value of the mean pore diameters and the low values of the compressive strengths. The thermal conductivity values obtained suggest that these calcium sources could be used to produce porous geopolymer cements. The values of thermal conductivity, the cumulative volumes of the pores and the total porosities of the porous geopolymer cement obtained by using commercial calcium carbonate as a pore-forming agent are slightly higher than those obtained by using the powders of eggshells. The results show that eggshells can be used as pore-forming agents during the synthesis of porous geopolymer cements that can be used as thermal insulator. The properties of the porous geopolymer cements could be improved for the rest of our work by increasing the concentration of the phosphoric acid solution.

References