Open access
Technical Notes
May 23, 2019

Production of High Compressive Strength Geopolymers Considering Fly Ash or Slag Chemical Composition

Publication: Journal of Materials in Civil Engineering
Volume 31, Issue 8

Abstract

More than 60  billionkg (60  milliont) of industrial by-products such as fly ash, garbage melting furnace slag, and steel slag are generated every year in Japan. Almost all of them seem to have been effectively used in mixtures with cement according to their chemical and mechanical characteristics in the field of civil work. Fly ash and slags can be used to create geopolymers in a process that emits less carbon dioxide than does the cement-making process. This reduction in CO2 emission is important because CO2 is one of the substances known to contribute to global warming. In the future, further uses of these fly ash and slags must be explored. The chemical mechanism for hardening composite materials by mixing aluminosilicate binders such as fly ash and slags with alkaline activators such as liquid NaOH and sodium silicate is known as a geopolymeric reaction, and the hardened composite material is called geopolymer. Geopolymers are produced by mixing two components of solid binders, such as fly ash, slags, and so on, with liquid activators such as NaOH, KOH, sodium silicate, and so on. Geopolymers have recently been developed to be used as a replacement for portland cement concrete. Research has been done to improve the compressive strength characteristics of both geopolymers and the combination of soft soil and fly ash–based geopolymer. The development of high compressive strength geopolymer using fly ash and slags will strongly contribute to the fields of construction, geotechnical engineering, and architecture. This paper describes the method for production and strength diagnosis of geopolymers considering the chemical composition of fly ash or slags.

Introduction

In 1979, Davidovits discovered the chemical mechanism for hardening composite materials resulting from geopolymeric reactions by mixing aluminosilicate binders with alkaline activators (Davidovits 1991, 2013). Geopolymers have recently been developed to use as a replacement for portland cement concrete, the cement of which discharges 1,000 kg (1.0 t) CO2 to produce 1,000 kg (1 t) cement (Ikeda and Mikuni 2006). Although geopolymers typically outperform portland cement concrete in compressive strength, the high cost of the alkaline activator and underdeveloped quality assurance (QA) and quality control (QC) have been the primary barriers that have prevented geopolymers from being used in construction and other industries. Buchwald (2006) defined geopolymers as being produced by mixing two components: a solid binder, such as fly ash, slags, and so on; and a liquid activator, such as NaOH, KOH, sodium silicate, and so on.
According to many studies, factors affecting the compressive strength qu include the chemical composition of solids; fineness of materials; w, the ratio of solution (NaOH + sodium silicate) to solid; η, the ratio of NaOH to sodium silicate; the curing time; the curing temperature; the molarity of NaOH (Suresh and Manojkumar 2013; Budh and Wahhade 2014); and so on. In addition to w and η, the composition of the fly ash and slags is another important variable.
This paper describes the method of production, strength diagnosis, and strength improvement of geopolymers made from binders of fly ash and slags, which are then mixed with activators (solutions of NaOH and sodium silicate). The chemical compositions of the binders are taken into account to find the optimum values of w and η which yields the maximum compressive strength qumax that is greater than 70  N/mm2 (Pawan Kumar and Surendra 2016).

Materials

Preparation of Geopolymer Materials

Steelmaking slags and garbage melting furnace slags are provided as granular materials, unlike fly ash. The particle size d50 of materials seems to affect the compressive strength of geopolymers qu (Fig. 1). The finer the particle size, the greater is the compressive strength of geopolymers. In the present tests, all slags were ground after being air dried to a maximum particle size of 0.106 mm for effective chemical reaction.
Fig. 1. Effect of slag particle size (d50) on compressive strength.

Chemical Compositions of Geopolymer Materials

The main chemical compositions of the geopolymer materials tested are listed in Table 1. The six kinds of geopolymer materials, two each of Fly ash, Slag 1 (steel factory slags), Slag 2 (garbage melting furnace slags), were the starting geopolymer materials, and ten mixtures of two each of fly ash and fly ash, fly ash and Slag 1, fly ash and Slag 2, and Slag 1 and Slag 2, were prepared as materials with a wide range of chemical compositions.
Table 1. Chemical composition of tested geopolymer materials
Geopolymer materialGeopolymer sampleChemical composition (%)
SiO2Al2O3CaOFe2O3MgOSO3
Fly ashReihoku55.021.19.15.31.10.9
Karita38.824.319.51.60.56.6
Slag 1Koro34.614.842.70.45.70.0
Stainless26.75.348.21.05.50.4
Slag 2Kazusa34.213.242.02.61.90.7
Narashino34.213.939.33.71.80.6
Fly ash and fly ashKariRei46.922.714.33.50.83.8
KariRei249.622.212.64.10.92.8
Fly ash and Slag 1ReiKoro44.818.025.92.93.40.5
ReiKoro241.416.931.52.04.20.3
KariKoro36.719.531.11.03.13.3
KariKoro236.018.035.00.84.02.2
Fly ash and Slag 2ReiKazu44.617.225.64.01.50.8
ReiNara44.617.524.24.51.40.8
Slag 1 and Slag 2KazuKoro34.414.042.41.53.80.4
KazuKoro234.514.342.51.14.40.2

Note: Koro = ground blast furnace slag; Stainless = ground stainless steel–making slag. Mixture ratio of geopolymer samples: KariRei: KaritaReihoku=11 by weight; KariRei2Karita: Reihoku = 12 by weight; ReiKoro: ReihokuKoro=11 by weight; ReiKoro2: ReihokuKoro=12 by weight; KariKoro: KaritaKoro=11 by weight; KariKoro2: KaritaKoro=12 by weight; ReiKazu: ReihokuKazusa=11 by weight; Rebyara: ReihokuNarashbyo=11 by weight; KazuKoro: KazusaKoro=11 by weight; and KazuKoro2: KazusaKoro=12 by weight.

Making Geopolymer Samples

Although several kinds of liquid sodium hydroxide and sodium silicate are available as activators, for the purpose of this research 48% NaOH (18  mol/L) and sodium silicate (Na2·nSiO2, n=3.2) were chosen due to their commercial availability and high concentration, with the expectation to yield high compressive strength geopolymers. The compressive strength of geopolymers, qu, is generally considered to be a function of the weight ratio of activator to binder, w, and the weight ratio of NaOH to sodium silicate, η. In this research, initially, geopolymer samples were made for wide range of ratios of w (0.3–0.6) and η (0.0–1.0), depending on geopolymer materials for fly ash, Slag1 and Slag2.
For mixtures of geopolymer materials, geopolymer samples were made for variable η values at a constant value of w=0.4, according to the aforementioned compression test results. After mixing the activator and binder, the geopolymer samples were placed in plastic molds of diameter D=50 mm and height H=100 mm, and vibrated slightly. The geopolymer samples were removed from the molds after 1 or 2 days and cured at room temperature under dry condition for 28 days.

Tests and Results

Before compression tests, physical properties of geopolymer samples such as diameter d, height h, and weight W were measured to obtain characteristics of shrinkage and density.
Compression tests of geopolymer samples were carried out at the Saga Construction Technology Support Organization (SCTSO) using the concrete testing apparatus in the same test method for concrete samples (sample diameter d=50  mm and height h=100  mm, loading rate=0.6±0.4  N/mm2, and loading plate φ=300  mm).
Compression test results are shown in Fig. 2 for key geopolymer samples and Fig. 3 for mixture geopolymer samples. Fig. 2 shows the relationship between the compressive strength qu and η value for each w value. Fig. 3 shows the relationship between qu and η only for w=wopt=0.4, which was determined from Fig. 2. Characteristics of the volume shrinkage ratio ΔV/V at qumax are summarized in Table 2.
Fig. 2. Compression test results for fly ash and slag geopolymers: (a) Reihoku; (b) Karita; (c) Koro; (d) Stainless; (e) Kazusa; and (f) Narashino.
Fig. 3. Compression test results for mixture materials geopolymers.
Table 2. Shrinkage characteristics of geopolymer samples at qumax
Geopolymer samplequmax (N/mm2)d (mm)Δd (mm)h (mm)Δh (mm)ΔV (mm3)ΔV/V (%)
Reihoku45.250.00.099.80.2392.70.20
Karita104.350.00.099.80.2392.70.20
Koro168.049.80.299.60.42,346.81.20
Stainless48.148.11.995.84.222,270.911.34
Kazusa110.049.70.399.70.32,931.11.49
Narashino86.449.90.199.60.41,566.90.80
KariRei70.149.90.199.60.41,566.90.80
KariRei240.949.80.299.80.21,957.21.00
ReiKoro79.349.90.199.70.31,371.30.70
ReiKoro2118.049.80.299.60.42,346.81.20
KariKoro117.049.90.199.50.51,762.40.90
KariKoro2159.049.90.199.70.31,371.30.70
ReiKazu102.049.80.299.40.62,736.31.39
ReiNara94.349.70.399.30.73,707.11.89
KazuKoro132.049.80.299.50.52,541.61.29
KazuKoro2166.049.80.299.30.72,931.11.49

Note: d = diameter, h height; V = initial volume of geopolymer sample; Δd, Δh, and ΔV = amount of shrinkage of d, h, and V, respectively; and ΔV/V = volume shrinkage ratio.

Discussions

Determination of wopt, ηopt, and qumax

In Fig. 2 qu generally increases with a decrease in η for each w value and reach the maximum value qumax at a certain η value for the case of w=0.4. The w value which yields the qumax is defined as the optimum value wopt; wopt is a constant of 0.4 for all key geopolymers.
Olivia and Nikraz (2012) and Prakash and Urmil (2012) evaluated how the ratio of alkaline liquid to fly ash affected the compressive strength of geopolymer. Olivia and Nikraz (2012) mentioned that the ratio, w, in the range 0.30–0.45 improved the strength to a certain extent, and Prakash and Urmil (2012) mentioned that the ratio, w, in the range of 0.30–0.40 did not much improve the strength. The values of qumax were produced at wopt=0.4, similarly to Olivia and Nikraz (2012) and Prakash and Urmil (2012).
The maximum value qumax occurs at a certain η value, which is defined as the optimum value ηopt (Fig. 2). From Fig. 3 which shows the relationship between qu and η for mixture material geopolymer for a constant value of w=wopt=0.4, the values of both qumax and ηopt were obtained.
The values of wopt, ηopt, and qumax from Fig. 2 and the values of ηopt and qumax from Fig. 3 are summarized in Table 3. Table 3 also includes the values of the density ρt of geopolymer samples at the qumax and the volume shrinkage ratio ΔV/V.
Table 3. Compression test results of geopolymer samples
Geopolymer materialsGeopolymer samplesCompression test results
qumax (N/mm2)woptηoptρt (kg/m3)ΔV/V (%)
Fly ashReihoku45.20.400.801,8680.20
Karita104.30.400.501,9850.20
Slag 1Koro168.00.400.402,2271.20
Stainless48.10.400.002,26911.34
Slag 2Kazusa110.00.400.302,2021.49
Narashino86.40.400.502,0430.80
Fly ash and fly ashKariRei70.10.400.651,9710.80
KariRei240.90.400.701,9861.00
Fly ash and Slag 1ReiKoro79.30.400.602,0980.70
ReiKoro2118.00.400.602,1441.20
KariKoro117.00.400.452,1220.90
KariKoro2159.00.400.432,1490.70
Fly ash and Slag 2ReiKazu102.00.400.552,0811.39
ReiNara94.30.400.652,0241.89
Slag 1 and Slag 2KazuKoro132.00.400.352,2191.29
KazuKoro2166.00.400.362,2181.49

Note: qumax = maximum value of compressive strength qu; wopt = optimum value of w yielding qumax; ηopt = optimum value of η yielding qumax; ρt = density of geopolymer sample at qumax; and ΔV/V = volume shrinkage ratio of geopolymer sample.

Correlation between ηopt and Chemical Compositions of Binders

Because geopolymerization is a chemical reaction, specifically an aluminosilicate reaction, the amount of NaOH which contributes to ionize metals contained in the binder increases with the increase of the amount of Al2O3 and SiO2. Accordingly, η, which is the weight ratio of NaOH to sodium silicate, is considered as the function of Cas(=Al2O3+SiO2). Fig. 4 shows the correlation between ηopt and Cas. The relationship between ηopt and Cas is well correlated and expressed by the straight line with a high correlation coefficient of r=0.949 as
ηopt=0.0157Cas0.414
(1)
Fig. 4. Correlation between ηopt and Cas.
Eq. (1) is used to calculate the ηopt value from the value of Cas of binders to effectively produce high compressive strength geopolymers.

Correlation between qumax and Chemical Compositions of Binders

The value of qumax generally decreases with the increase of the amount of each SiO2, Al2O3, and Fe2O3 (Tables 1 and 3). In contrast, qumax is in proportion to the content of CaO. Because the total contents of chemical compositions are constant, binders which have high content of SiO2 or Al2O3 are said to be binders which have less content of CaO. Using the factor of CaO/Cas, the mutual chemical characteristics of qumax can be shown. Fig. 5 shows the relationship between qumax and CaO/Cas. Binders with CaO/Cas in the range 0.3–0.9 yielded high compressive strength geopolymers with qumax = 80168  N/mm2 (Fig. 5).
Fig. 5. Relationship between qumax and CaO/Cas.

Relationship between qumax and ηopt

Fig. 6 shows the relationship between the qumax and ηopt. Test data obtained by Prakash and Urmil (2012) (qumax=40  N/mm2 for ηopt=0.5), Pawan Kumar and Surendra (2016) (qumax=70.6  N/mm2 for ηopt=0.83), and Mustafa Al Baki et al. (2012) (qumax=57  N/mm2 for ηopt=0.4) are plotted; they all lie in the lower range of the present qumax. Some reasons for their lower compressive strengths are the difference of the chemical compositions of binders, the value of w, curing conditions, and so on. In Fig. 6, the qumax becomes greater than 100  N/mm2 for ηopt in the range 0.3–0.5.
Fig. 6. Relationship between qumax and ηopt.

Shrinkage of Geopolymer Samples

Fig. 7 shows the relationship between the volume shrinkage ratio of geopolymer samples ΔV/V and CaO/Cas. The values of ΔV/V were generally less than 2%. These results will be helpful for the design of geopolymer construction.
Fig. 7. Relationship between ΔV/V and CaO/Cas.

Conclusions

The methods for production and strength diagnosis of geopolymers to determine the optimum values of wopt and ηopt yielding the maximum compressive strength qumax, and to correlate ηopt with the appropriate chemical composition of binders to produce effectively the high compressive strength geopolymers, are as follows:
1.
to produce high compressive strength geopolymer, slags have to be ground as fine as possible;
2.
the value of qu generally becomes the maximum value (qumax) when the weight ratio, w, of the mixed solutions of NaOH and sodium silicate to the binder is 0.4 (wopt, which is the optimum value of w yielding qumax), regardless of the type of the binder;
3.
there is an optimum value, ηopt (the weight ratio of NaOH to sodium silicate), yielding qumax for each binder;
4.
the value of ηopt is well correlated with the factor Cas [Eq. (1)];
5.
The value of ηopt for an arbitrary binder necessary to manufacture high compressive strength geopolymer is calculated by Eq. (1);
6.
binders with CaO/Cas in the range 0.3–0.9 (ηopt=0.30.5) yield high compressive strength geopolymers with qumax=80168  N/mm2; and
7.
the volume shrinkage ratio ΔV/V is generally less than 2%.

Notation

The following symbols are used in this paper:
Cas
Al2O3+SiO2 (%);
CaO/Cas
factor;
D
diameter of mold (mm);
d
diameter of geopolymer sample (mm);
d50
particle size in 50% of particles by mass (mm);
H
height of mold (mm);
h
height of geopolymer sample (mm);
qu
compressive strength (N/mm2);
qumax
maximum compressive strength (N/mm2);
r
correlation coefficient;
V
volume of mold (mm3);
w
weight ratio of activator solution of sodium hydroxide + sodium silicate) to binder;
wopt
optimum value of w at yielding qumax;
ΔV
amount of shrinkage of geopolymer sample (mm3);
ΔV/V
volume shrinkage ratio of geopolymer sample (%);
η
weight ratio of solution of sodium hydroxide to sodium silicate;
ηopt
optimum value of η at yielding qumax; and
ρt
density of geopolymers at qumax (kg/m3).

Acknowledgments

The author is grateful to Maruwa Giken for preparing ground solid waste incinerator slags. The author is also grateful to Kyuden Sangyo, Shunan Works, Nisshin-Steel, and Nippon Steel & Sumikin Slag Products for contributing coal-fired power plant ash, stainless manufactured slag, and ground granulated blast slag, respectively.

References

Buchwald, A. 2006. “What are geopolymers? Current state of research and technology, the opportunities they offer, and their significance for precast industry.” Betonwerk und Fertigteil-Technik. 72 (7): 42–49.
Budh, C. D., and N. R. Wahhade. 2014. “Effect of molarity on compressive strength of geopolymer mortar.” Int. J. Civ. Eng. Res. 5 (1): 83–86.
Davidovits, J. 1991. “Geopolymers inorganic polymeric new materials.” J. Therm. Anal. 37 (8): 1633–1656. https://doi.org/10.1007/BF01912193.
Davidovits, J. 2013. Geopolymer cement: A review. Saint-Quentin, France: Institute Geopolymer.
Ikeda, K., and A., Mikuni. 2006. “Recent development of geopolymers from viewpoint of carbon dioxide emission and waste management problems.” Trans. MRS-J. 31: 319–324.
Mustafa Al Baki, A. M., H. Kamarudin, M. Bnhussain, A. R. Rafiza, and Y. Zarina. 2012. “Effect of Na2SiO3/NaOH ratios and NaOH molarities on compressive strength of fly-ash-based geopolymer.” Mater. J. 109 (5): 503–508.
Olivia, M., and H. R. Nikraz. 2012. “Strength and water penetrability of fly ash geopolymer concrete.” ARPN J. Eng. Appl. Sci. 6 (7): 70–78.
Pawan Kumar, K. R., and B. V. Surendra. 2016. “Study on strength of geopolymer concrete with ambient temperature curing and low alkali content.” Int. Res. J. Eng. Technol. 3 (5): 1073–1077.
Prakash, R. V., and V. D. Urmil. 2012. “Parametric studies on compressive strength of geopolymer concrete.” Procedia Eng. 51: 210–219. https://doi.org/10.1016/j.proeng.2013.01.030.
Suresh, G. P., and Manojkumar 2013. “Factors influencing compressive strength of geopolymer concrete.” Int. J. Res. Eng. Technol. 372–375. https://doi.org/10.15623/ijret.2013.0213070.

Information & Authors

Information

Published In

Go to Journal of Materials in Civil Engineering
Journal of Materials in Civil Engineering
Volume 31Issue 8August 2019

History

Received: Mar 10, 2018
Accepted: Feb 4, 2019
Published online: May 23, 2019
Published in print: Aug 1, 2019
Discussion open until: Oct 23, 2019

Authors

Affiliations

Professor Emeritus, Dept. of Agricultural Sciences, Saga Univ., 1 Honjo-machi, Saga 840-8502, Japan; Director, Geopolymer Research Laboratory, 3-3-15 Takakise-nishi, Saga 849-0921, Japan. ORCID: https://orcid.org/0000-0001-6889-5072. Email: [email protected]

Metrics & Citations

Metrics

Citations

Download citation

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by

View Options

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share with email

Email a colleague

Share