A lkali activated binder, commonly known as geopolymer cement, has replaced Portland cement in the production of mortar and concrete globally over the past few years. The density, particle size distribution, and specific surface area (SSA) are important physical parameters affecting strength and durability of alkali activated binders. This study carried out tests for physical and chemical properties of the natural pozzolan and calcium hydroxide and then determines the influence of alkali solution (sodium silicate and sodium hydroxide) on strength development of natural pozzolan calcium hydroxide binders. The particle size distribution (PSD), relative densities (RD), and specific surface areas (SSA) of powder natural pozzolan and calcium hydroxide materials and for the mixture of natural pozzolan and calcium hydroxide were determined by using Blaine air permeability apparatus. The optimum proportion of 75% natural pozzolan and 25% calcium hydroxide was obtained which produces the compressive strength of 7.5 MPa at 28 days cured paste. The mixture of natural pozzolan and calcium hydroxide were further grinded at three different fineness es and the particle size gradation, specific densities, specific surface areas and mean particles sizes for the mixture were determined. The compressive strength of alkali activated binders increased with increasing curing period and fineness. The maximum compressive strength for 28 days cured specimens was 26.1 MPa which was obtained at a solution of 8 moles sodium hydroxide concentration. The test results showed that natural pozzolan materials can be used to make geopolymer binders for mortars and concretes. The geopolymer binders for mortars and concretes reduce green gas emission from cement factory but also it can be used to produce durable mortar and concrete with comparable strengths with mortars and concrete made from conventional Portland cement.
Pozzolan materials (artificial and natural) are extensively used as supplementary cementitious materials for mixing with industrial cements for producing mortars and concretes (Moon et al., 2014). In the recent years the use of pozzolanic materials has been extended for making geopolymer mortars and concrete through activation with alkali solution. The use of alkali activators on making geopolymer binders for concrete and mortar mixes has found to increase durability and reduce green gases emission from cement factories (Alp et al., 2009). Alkali activated cement (AAC) which is known as geopolymer cement has been extensively studied in terms of its properties and characteristics. However, few researches have been conducted on natural pozzolan-based alkali activated binders (Alraddadi & Assaedi, 2020).
Some studies conducted to activate pozzolanic materials using alkali solutions having substantial amount of silicates and aluminates have indicated promising results on strength development and durability (Zheng et al., 2016). Pozzolan are siliceous and aluminous materials which themselves possess little or no cementitious values but, when finely grounded chemically react with calcium hydroxide in the presence of water at ordinary temperature to form calcium silicates hydrates (CSH) and calcium aluminates hydrates (CAH) (Robayo-Salazar & De Gutiérrez, 2018). When calcium reacts with silicate and aluminum from pozzolanic materials the process is called pozzolanic reaction and when the mixture of pozzolan and calcium oxide or hydroxide is activated by alkaline elements the process is called geopolymer reaction.
Pozzolanic reactions occur over long time scales (months to years). The main mechanism involves the transportation of calcium hydroxide via water to combine with silicate and/or aluminate in pozzolan minerals (Duxson et al., 2007) The pozzolan, which in the presence of water having alkaline environment (e.g. presence of calcium) produce cementitious materials, comprise calcium silicates and calcium aluminate hydrates as the main pozzolanic compounds produced (Zheng et al., 2016). The dissolved Ca2+ ions react with any dissolved SiO2 and Al2O3 on pozzolan to produce hydrated gels of C-S-H and C-A-H. Under normal curing environment of pozzolanic binders, the following reaction process and cementitious products may occur (Alp et al., 2009).
CaO + H 2 O → Ca ( OH ) 2 ( CH ) (1)
2 S + 3 CH → C 3 S 2 H 3 ( CSH ) (2)
A + 4 CH + 9 H → C 4 AH 13 ( CAH ) (3)
However, in the pozzolan-lime system, other products of pozzolanic reaction may develop such as C2SH2, CAH10, C3AH6 (katoite) and C2AH8 (mayenite), gehlenite (C2ASH8) and hydrogarnet (C3ASH6) (Juenger et al., 2011).
The pozzolanic reaction are greatly affected by physical, chemical, morphology and mineralogical composition (Chengula, Msambichaka, & Middendorf, 2018; Arvaniti et al., 2015a). The density, particle size distribution, and specific surface area (SSA) are important physical parameters affecting pozzolanic reaction and geopolymer reaction (Sharma & Khan, 2016). Adequate physical characterization of binder materials is important to better predict their performance and optimize their use in mortar and concrete production (Juenger et al., 2011). Fineness is one of the most essential physical parameters determining binder materials’ reactivity (Mboya et al., 2017). The rate of dissolution is increased by reducing the average particle size, which improves pozzolanic reactivity and therefore makes for greater strength (Walker & Pavía, 2011). The effects of density, specific surface area and particle size distribution were discussed in detail by previous researcher (Chengula, Msambichaka, & Middendorf, 2018). The study encouraged the materials with variances in densities to be proportioned by volume (Chengula, Msambichaka, & Middendorf, 2018). It was noted that for large specific surface area the rheological, compressive strength and durability were improved (Arvaniti et al., 2015a). The flow properties of fresh paste, mortars and concrete are improved by uniformly graded particle sizes. The packing density and compressive strength development are also improved (Arvaniti et al., 2015b).
The pozzolanic reaction which occurs in the setting of the mixture of slaked lime and pozzolanic materials has also features similar to the alkali-silica reaction (geopolymer reaction), mainly the formation of calcium silicate hydrate (C-S-H) (Ibrahim et al., 2019).
The geopolymer reaction is the chemical reaction involving alkaline (base) elements in solution with acidic elements (Gharzouni et al., 2015). The common alkaline elements considered for geopolymer reactions are sodium, potassium and manganese and common acidic elements considered for geopolymer reactions are silica and alumina (Djobo et al., 2016). The alkali-acidic reactions can take place in cement and pozzolanic materials provided there are presence of reactive chemicals and suitable environment (Gharzouni et al., 2015). The literature demonstrates that, when alkali reacts with silica in a hardened concretes and mortars is called alkali-silica reactions and it is harmful to the concrete and mortars because it causes severe cracks and deterioration of concrete and mortars due to volume increase.
The alkali-silica gel formed during early age of pozzolanic and geopolymer reactions are harmless to the concrete or mortar (Akbari, Mensah-Biney, & Simms, 2015). This is because the stresses induced pressures by alkali-silica gels formation are absorbed in the fresh concrete or mortar. The cracks formed due to extra stress in concrete or mortar are filled immediately by pozzolanic gels such as CSH and CAH and geopolymer chain gels such as NSH and NAH during the process of hardening which results into high strength development (Irshidat, Abdel-Jawad, & Al-Sughayer, 2018; Walker & Pavía, 2011). Increase in volume due to alkali-silica formation at early age of concrete or mortar is limited due to confinement in formworks or moulds. Apart from alkali-silica reaction in pozzolanic binders but also alkali-aluminum reaction can take place in the same system provided there are reactive alkali and aluminum elements in the solution (Chen et al., 2021). There have been no reports of concrete and mortar being destroyed because alkali-aluminum gels formed during the early and late days of curing.
The simplest scheme of geopolymer reactions in pozzolanic binders under normal environment of pressure, temperature, humidity and alkaline concentration can be as follows:
2 NaOH + SiO 2 + H 2 O → Na 2 SiO 3 ⋅ 2 H 2 O ( NSH ) (4)
2 NaOH + Al 2 O 3 + H 2 O → Na 2 Al 2 O 4 ⋅ 2 H 2 O ( NAH ) (5)
Polymerization process involves a substantially fast chemical reaction under alkaline conditions on silicon-aluminum minerals that results in a three-dimensional polymeric chain and ring structure (Davidovits, 2017). The reaction takes three stages; stage 1 is dissolution of Si-O-Si, Al-O-Al, and Si-O-Al bonds provided by the precursor (often referred as nucleation stage), stage 2 is coagulation or polycondensation in which a coagulated structure between the disbanded/disaggregated composition from precursor takes place and stage 3 is crystallization, in which crystals begin to develop and shape an inorganic hardened and 3-dimensional polymer structure (Allahverdi & Ghorbani, 2006).
The geopolymer chain product is an amorphous three-dimensional network polymer, formed by the condensation of silicon oxygen tetrahedrons and aluminum oxygen tetrahedrons. The general formula of alumino-silicate materials that provide polymeric Si-O-Al bond is given as follows (Bondar et al., 2012):
Mn[(SiO2)zAl2O3]n·wH2O (6)
where:
M—is an alkali cation (such as Na+, K+ or Ca2+),
n—is the polymerization degree,
n—is 1, 2, to 3 and
w—is the number of moles of water.
The main parameter which influences the properties of geopolymer binder is alumino-silicate source (Oyebisi et al., 2018). The common alkaline activator used is sodium hydroxide and sodium silicate or potassium hydroxide and potassium silicate (Hajimohammadi & van Deventer, 2016).
The binders made from pozzolan and calcium hydroxide which have been activated by alkali solution both products C-(A)-S-H and N(K)-A-S-H might co-exist (Davidovits, 2017). But also there might be intermixing of the gels due to modification process taking place within the systems leading to a partial or total replacement of sodium with calcium to form a giant geopolymer chain (N,C)-A-S-H gels (Kropyvnytska et al., 2019).
The precursor materials used for this study were locally available volcanic ashes and the activators of the geopolymer reactions were industrial alkali solution. Materials characterizations, mixing process of binders and tests for compressive strengths of cured specimens were conducted under laboratory condition.
The materials used for mixing geopolymer binders are natural pozzolan from Songwe region in Tanzania, industrial processed calcium hydroxide from local suppliers. In order to improve reactions process of pozzolanic binders and improve strengths of cured specimen then alkali solutions were used as activators to mix the pastes. The alkali activators used are sodium silicate solution and sodium hydroxide pellets obtained from local supplier in Mbeya region.
Water was added into solid sodium hydroxide pellets (NaOH) at three different concentrations of 6 moles, 8 moles and 10 moles to make sodium hydroxide solution. The two alkali solutions were then mixed together for mixing pozzolan calcium hydroxide paste. The sodium hydroxide pellets (NaOH) had a purity of 97% with proportions of 77.53% of Na2O and 22.47% of H2O by mass. The Sodium silicate (Na2SiO3) in solution form had a purity of 98% with proportions of 34.78% of SiO2, 16.22% of Na2O and 49.00% of H2O by mass.
Natural pozzolan from Songwe region and industrial calcium hydroxide from supplier were used to prepare paste for determination of compressive strengths of cured specimens. The natural pozzolan were air dried and grinded into powder form using disc grinder machine while calcium oxide were obtained in powder form.
Physical and chemical properties of natural pozzolan and calcium hydroxide were determined. The chemical compositions of the materials were determined using XRF equipment conducted at Government Chemistry Laboratory located in Mbeya Region.
The physical properties of materials were determined by using Blaine air percolation apparatus, the method described by Chengula, Msambichaka and Middendorf (2018). The properties determined are specific densities, specific surface areas, particle size distribution, mean particle sizes, median, mode, maximum and minimum particle sizes. The specific densities and specific surface areas of powder natural pozzolan and calcium hydroxide were determining using Equation (7) and Equation (8) respectively (Chengula, Msambichaka, & Middendorf, 2018). However Equation (9) and (10) were used to determine particle size distribution of powder materials (Juimo Tchamdjou et al., 2018). The physical properties of the raw natural pozzolan and calcium hydroxide powder materials are summarized in
| Chemicals | SiO2 | Al2O3 | Fe2O3 | CaO | MgO | SO3 | Na2O | K2O | P2O5 | LOI | Total |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Natural pozzolan | 55.84 | 19.5 | 8.88 | 1.5 | 0.4 | 0.02 | 1.44 | 3.54 | 0.05 | 8.83 | 100 |
| Calcium hydroxide | 0.4 | 0.6 | 0.6 | 74.7 | 0.8 | 0.2 | 0.6 | 0.3 | 0.05 | 21.75 | 100 |
| Properties | Specific density (gm/cm3) | Specific surface area (cm2/gm) | Mean particle size (μm) |
|---|---|---|---|
| Pozzolan | 2.496 | 5560 | 1.745 |
| Calcium hydroxide | 2.269 | 7080 | 1.444 |
ρ = 2 M V (7)
where: M—Weight of sample at turning point,
V—Volume of Blaine tin (1.628 cm3).
S S A = K m i M i ρ s i (8)
where: SSA—Specific surface area (cm2/g),
m—Curve slopes of time air flow versus weight of sample (sec/g),
M—Mass of sample in Blaine tin at turning points,
ρs—Densities of materials at turning points,
K—Apparatus and measurement constant (1050).
G p s = K [ 1 η a ( m 1 + m 2 ) ] × [ M i − α M o M f − M o ] (9)
% C P = 100 T i T f (10)
where: Gps—Grain particle size (g),
K —Constant of equation (13,650),
ηa—Dynamic viscosity of air at room temperature,
m1 & m2—gradient at lower and middle curve,
Mo—Weight of sample at zero time,
Mf—Weight of sample at second turning,
α—factor for sample weight at zero air flow (0.99),
Ti—Time air flow for each measurement,
Tf—Time air flow at second turning.
The paste specimens were prepared by mixing natural pozzolan and calcium hydroxide powder. Different mixtures were prepared by varying the proportions of pozzolan contents from 20% to 90% as indicated in
Three specimens for each proportion of natural pozzolan and calcium hydroxide mix of the sizes 50 mm cubes were casted and cured for 28 days.
The variation of pozzolan and calcium hydroxide on paste specimens were done in order to determine optimum proportions of pozzolan and calcium hydroxide to which the maximum compressive strength of paste specimens can be obtained.
The powdered natural pozzolan and calcium hydroxide were mixed together at mix proportions of 75% natural pozzolan and 25% calcium hydroxide to prepare geopolymer paste specimens. In order to increase reactivity, the mixture of natural pozzolan and calcium hydroxide materials were further grinded on which three different fineness were obtained. The physical properties of the mixture of natural pozzolan calcium hydroxide binders are summarized in
The pellets of sodium hydroxide were hydrated with water to form solution at different concentrations. The solution of sodium hydroxide was prepared at concentration of 6 moles, 8 moles and 10 moles of sodium hydroxide pellets.
Three specimens of the sizes 50 mm cubes were casted and cured for 7 days, 14 days, 21 days and 28 days.
Geopolymer paste specimens were prepared at solution to binder ratio of 0.52. The solution was made from the mixture of sodium silicate solution and sodium hydroxide solution. The sodium hydroxide solution was prepared 24 hours prior to use. The sodium hydroxide solution thus prepared was mixed well with sodium silicate solution to get desired alkaline solution. The ratio of sodium silicate solution and sodium hydroxide solution used for this study was 2.5. The amount of alkali solution (AS) used for mixing paste specimens were determined from 2000 gm of binders as indicated in
| S/N | Sample ID | Pozzolan (gm) | Lime (gm) | Water (gm) |
|---|---|---|---|---|
| 1 | 20P80L | 400 | 1600 | 960 |
| 2 | 40P60L | 800 | 1200 | 960 |
| 3 | 60P40L | 1200 | 800 | 960 |
| 4 | 80P20L | 1600 | 400 | 960 |
| 5 | 90P10L | 1800 | 200 | 960 |
| Description | Specific density (gm/cm3) | Specific surface area (cm2/gm) | Mean particle size (μm) |
|---|---|---|---|
| Fineness No. 1 | 2.463 | 6059 | 1.681 |
| Fineness No. 2 | 2.460 | 6148 | 1.674 |
| Fineness No. 3 | 2.462 | 6642 | 1.432 |
| Materials | Binder | Total solution | Pozzolan | Calcium hydroxide | Sodium silicate solution | Sodium hydroxide solution |
|---|---|---|---|---|---|---|
| Amount | 2000 | 1040 | 1500 | 500 | 743 | 297 |
Solution = Binder × 0.52 = 2000 × 0.52 = 1040 gm,
Pozzolan = 0.75 × 2000 = 1500 gm; Calcium oxide = 2000 × 0.25 = 500 gm,
Sodium silicate solution = 2.5 × 1040/3.5 = 743 gm; Sodium hydroxide solution = 1040/3.5 = 297 gm.
The amount of sodium and silicon dosage in the paste for each concentration of sodium hydroxide were determined from molar weights and compositions in the solution:
The compositions of the sodium silicate solution from the manufacturer are SiO2 = 34.78%, Na2O = 16.22% and H2O = 49.00% with purity of 98%. The molecular weights are 60 gm for SiO2, 62 gm for Na2O and 18 gm for H2O.
The amount of sodium and silicon in sodium silicate solution are determined as follows:
Silicon = 28 60 × 0.3478 × 0.98 × 743 = 118.18 gm
Sodium = 46 62 × 0.1622 × 0.98 × 743 = 87.63 gm
The molecular weights are 40 gm for NaOH and 18 gm for H2O. The compositions of the sodium hydroxide solution for 1mole concentration of NaOH are 3.85% for NaOH and 96.15% for H2O with purity of 97%.
The amount of sodium in sodium hydroxide solution for 6 moles, 8 moles and 10 moles are determined as follows:
1) For 6 moles of sodium hydroxide in a litter of water
Sodium = 240 1240 × 23 40 × 0.97 × 297 = 32.06 gm
2) For 8 moles of sodium hydroxide in a liter of water
Sodium = 320 1320 × 23 40 × 0.97 × 297 = 40.16 gm
3) For 10 moles of sodium hydroxide in a liter of water
Sodium = 400 1400 × 23 40 × 0.97 × 297 = 47.33 gm
Therefore the amount of sodium, silicon and water for each concentration of sodium hydroxide pellets in the geopolymer paste specimens are given in
The characterization of raw materials which are natural pozzolan and calcium hydroxide to determine chemical composition and physical properties were conducted. The physical properties determined for natural pozzolan and calcium hydroxide are specific densities, specific surface areas, mean particle sizes and particle size distribution curves. The physical properties of the blended natural pozzolan calcium hydroxide were also determined. The compressive strengths and densities of cured specimens made from natural pozzolan calcium hydroxide mixtures were determined under laboratory condition. In order to increase the compressive strengths of natural pozzolan calcium hydroxide mixtures, the materials were activated using alkali solution of sodium silicate and sodium hydroxide at different concentration.
| Alkali | Sodium | Silicon | Water | Na/Si | Na/Water | Si/Water |
|---|---|---|---|---|---|---|
| 6 moles of NaOH | 119.69 | 118.18 | 802.13 | 1.013 | 0.149 | 0.147 |
| 8 moles of NaOH | 127.79 | 118.18 | 794.03 | 1.081 | 0.161 | 0.149 |
| 10 moles of NaOH | 134.96 | 118.18 | 786.86 | 1.142 | 0.172 | 0.150 |
The chemical compositions of the raw materials for this study were determined using XRF technique.
The results indicate that the chemical index of natural pozzolan materials is 84.14% which is greater than 70% minimum required. Therefore the natural pozzolan materials can be used to make binders for construction purposes. Densities, specific surface areas and particle size distribution cures of natural pozzolan and calcium hydroxide were determined.
The physical compositions determined for raw natural pozzolan and calcium hydroxide are specific densities, specific surface areas, particle size distribution and mean particle sizes.
The results shown in
pozzolan of 1.8 kg/m3 to 2.9 kg/m3 determined by other researchers (Chengula, Msambichaka, & Middendorf, 2018). The density of calcium hydroxide determined (
The specific surface area of calcium hydroxide is higher than specific area of natural pozzolan which indicates that mean particles size of calcium hydroxide is less compared to mean particle size of natural pozzolan. From
Physical properties of natural pozzolan calcium hydroxide mixtures were determined. The pozzolan were mixed with calcium hydroxide at a proportion of 75% natural pozzolan and 25% calcium hydroxide. The mixtures were grinded together and three different fineness were obtained. The physical properties determined are specific densities, specific surface areas, particle size distribution and mean particle sizes.
The same phenomenon of “S” curve has been depicted for pozzolan calcium hydroxide mixtures similar to individual raw materials (refer
Results of specific densities, specific fineness and mean particle sizes of the natural pozzolan calcium hydroxide mixed at a proportion of 75% natural pozzolan and 25% calcium hydroxide grinded at different fineness are indicated in
The compressive strengths and densities of cured specimens made from the mixtures of natural pozzolan calcium hydroxide mixed with water and other specimen with alkali solution were determined.
Paste specimens were made from the mixtures of pozzolan and calcium hydroxide at varying mix proportions of 20%, 40%, 60%, 80% and 90% natural pozzolan content as indicated in
The mixture of natural pozzolan and calcium hydroxide at a proportion of 75P25CH were grinded together at different grinding effort and time and three different fineness were obtained (refer
solution were mixed with solution of sodium silicate for mixing of paste specimens. The sodium hydroxide solutions were prepared at three different concentrations of 6 moles, 8 moles and 10 moles. For each moles of sodium hydroxide solution was mixed with sodium silicate solution at a mix ratio of 1:2.5.
The values of compressive strengths for 28 days curing periods for fineness No. 01 are 20.4 MPa, 22.4 MPa and 19.4 MPa for 6 moles, 8 moles and 10 moles of NaOH concentration respectively. The compressive strengths for 28 days curing periods for fineness No. 02 are 20.8 MPa, 22.9 MPa and 19.6 MPa for 6 moles, 8 moles and 10 moles of NaOH concentration respectively. The compressive strengths for 28 days curing periods for fineness No. 03 are 24.8 MPa, 26.1 MPa and 21.5 MPa for 6 moles, 8 moles and 10 moles of NaOH concentration respectively.
The values of densities for 28 days curing periods for fineness No. 01 are 1833 kg/m3, 1832 kg/m3 and 1901 kg/m3 for 6 moles, 8 moles and 10 moles of NaOH concentration respectively. The densities for 28 days curing periods for fineness No. 02 are 1846 kg/m3, 1928 kg/m3 and 1887 kg/m3 for 6 moles, 8 moles and 10 moles of NaOH concentration respectively. The densities for 28 days curing periods for fineness No. 03 are 1887 kg/m3, 1955 kg/m3 and 1911 kg/m3 for 6 moles, 8 moles and 10 moles of NaOH concentration respectively.
The use of alkali solution as activator of natural pozzolan calcium hydroxide binders for this study have increased compressive strength of 28 days cured specimens from 7.5 MPA to 26.1 MPa at 8 moles sodium hydroxide concentration. The use of locally available natural pozzolan to make geopolymer binders for mortar and concretes mixed reduces green gases emissions from cement factory which also reduces environment pollution.
For this study geopolymer binder was made by activating natural pozzolan calcium hydroxide mixtures with alkali solution. The alkali activator solution used was sodium silicate and sodium hydroxide at a varying concentration of sodium hydroxide at 6 moles, 8 moles and 10 moles. The particle size distribution (PSD), relative densities (RD), and specific surface areas (SSA) of powder materials were determined by using Blaine air permeability apparatus.
The specific densities of 2.496 gm/cm3 and 2.269 gm/cm3, specific surface areas of 5560 cm2/gm and 7080 cm2/gm and mean particle sizes of 1.745 μm and 1.444 μm for natural pozzolan and calcium hydroxide respectively were determined. The natural pozzolan and calcium hydroxide powders were mixed at varying proportions to make paste specimens and cured for 28 days. The optimum proportion of natural pozzolan calcium hydroxide mixtures was determined to be 75% natural pozzolan and 25% calcium hydroxide and at this proportion the maximum compressive strength of cured paste specimen was 7.5 MPa. The optimum proportion of natural pozzolan and calcium hydroxide was milled at three different finenesses. It was observed that specific surface area increased as the grinding effort and time increased, the mean particle size decreased with increasing specific surface areas while the specific density does not change with the change of specific surface areas. The results indicated that, specific surface areas increased from 6059 cm2/gm to 6642 cm2/gm and mean particle size decreased from 1.681 µm to 1.432 µm.
The optimum proportion of 75% natural pozzolan and 25% calcium hydroxide binder was activated by the mixtures of alkali solution of sodium silicate and sodium hydroxide at varying concentration of sodium hydroxide of 6 moles, 8 moles and 10 moles. The compressive strength of 28 days cured geopolymer binders reached 26.1 MPa at 8 moles sodium hydroxide concentration. It is therefore important to use the available natural pozzolan materials to make geopolymer binders for mortar and concretes which will then reduce green gases emission and environmental pollution.
The present study focused on the influence on physical properties of geopolymer binder, more studies are recommended on the influence to micro structure, mineralogical composition and durability.
The authors declare no conflicts of interest regarding the publication of this paper.
Malisa, A. S., & Chengula, D. H. (2022). Study on the Influence of Alkali Activator Solutions on Strength Improvement of Pozzolan Calcium Hydroxide Binders. Journal of Geoscience and Environment Protection, 10, 313-330. https://doi.org/10.4236/gep.2022.1012018